Koterski, Joseph W. - Oppy, Graham - Theism and Atheism - Opposing Arguments in Philosophy-Macmillan Reference USA (2019)

Theism and Atheism
Opposing Arguments in Philosophy
EDITORS IN CHIEF
Joseph W. Koterski, S.J. (Theism)
Graham Oppy (Atheism)
COPYRIGHT 2019 Macmillan Reference USA, a part of Gale, a Cengage Company WCN 02-200-210
Theism and Atheism
Theism and Atheism
Joseph W. Koterski, S.J.

EDITOR IN CHIEF
Graham Oppy
EDITOR IN CHIEF
Theism and Atheism: Opposing Arguments © 2019 Macmillan Reference USA, a part of Gale, a Cengage Company.
in Philosophy
ALL RIGHTS RESERVED. No part of this work covered by the copyright herein
Joseph W. Koterski, S.J. and Graham Oppy,
may be reproduced or distributed in any form or by any means, except as
Editors
permitted by US copyright law, without the prior written permission of the
copyright owner.
Project Editor: Jonathan Vereecke
Contributing Editor: Rebecca Parks
For product information and technology assistance, contact us at
Product Design: Kristine Julien Gale Customer Support, 1-800-877-4253.
Acquisitions Editor: Laurie Malashanko For permission to use material from this text or product,
submit all requests online at www.cengage.com/permissions.
Further permissions questions can be emailed to
permissionrequest@cengage.com
Cover photograph: Christopher Smith / EyeEm / Getty Images.
While every effort has been made to ensure the reliability of the
information presented in this publication, Gale, a Cengage Company, does
not guarantee the accuracy of the data contained herein. Gale accepts no
payment for listing, and inclusion in the publication of any organization,
agency, institution, publication, service, or individual does not imply
endorsement of the editors or publisher. Errors brought to the attention of
the publisher and verified to the satisfaction of the publisher will be
corrected in future editions.
LIBRARY OF CONGRESS CATALOGING-IN-PUBLICATION DATA
Names: Koterski, Joseph W., editor.

Title: Theism and atheism : opposing arguments in philosophy / Joseph W.
Koterski, S.J. and Graham Oppy, editors.
Description: Farmington Hills, Mich. : Macmillan Reference USA, a part of Gale, a
Cengage Company, 2019. | Includes bibliographical references and index.
Identifiers: LCCN 2018041749| ISBN 9780028664453 (hardcover) | ISBN
9780028664460 (ebook)
Subjects: LCSH: Theism. | Atheism.
Classification: LCC BL200 .T44 2019 | DDC 211/.3–dc23 LC record available at https://
lccn.loc.gov/2018041749
Gale, a Cengage Company

27500 Drake Rd.
Farmington Hills, MI 48331-3535
ISBN 978-0-02-866445-3 (this volume)
This title is also available as an e-book.

ISBN: 978-0-02-866446-0
Contact your Gale sales representative for ordering information.
Printed in Mexico
1 2 3 4 5 6 7 23 22 21 20 19
Editorial Board
THEISM
EDITOR IN CHIEF
Associate Professor of Philosophy
Fordham University
ASSOCIATE EDITORS
Mashhad Al-Allaf
American University of Ras Al Khaimah
Robert Fastiggi
Professor of Systematic Theology
Sacred Heart Major Seminary
David Shatz
Ronald P. Stanton University Professor of Philosophy,
Ethics, and Religious Thought
Yeshiva University
ATHEISM
EDITOR IN CHIEF
Graham Oppy
Professor of Philosophy
Monash University
ASSOCIATE EDITORS
Gregory Dawes
Professor, Philosophy and Religion
University of Otago
Evan Fales
Emeritus
University of Iowa
Contents
Preface xv
Editor’s Note: Theism xvii
Editor’s Note: Atheism xix
Contributors xxi
TOPIC 1: DEFINITION
Definition: Theism ..............................................................................................................................1

David B. Twetten, Associate Professor, Marquette University
Brian Carl, Assistant Professor of Philosophy, Dominican House of Studies
Mark Johnson, Associate Professor, Marquette University
Francisco Romero Carrasquillo, Associate Professor, Universidad Panamericana
This chapter discusses how theism and atheism are best defined, primarily as a function of what
“God” or “god” means. It sets out the range of meanings that “god” and “God” can have in
different argumentative contexts, suggesting that proofs for God’s existence need not immediately
conclude to the existence of a being with personal or omni-attributes. It also explores the notions of
analogy and apophaticism, given that these themes condition the way many theists understand the
affirmation that God exists. The chapter thereby offers principles that should govern the proper use
of words in theism versus atheism debates.
Definition: Atheism ...........................................................................................................................19
Robert Nola, Emeritus Professor, The University of Auckland
This chapter discusses atheism, theism, and agnosticism and the nature of some of the definitions
of these doctrines. Rather than address arguments for or against god (or God), the chapter also
considers definitions that can clarify what these doctrines mean. It also discusses concepts of god as
spelled out in various kinds of definition.
TOPIC 2: METHOD
Method: Theism ................................................................................................................................35

Robert Audi, Professor, University of Notre Dame
This chapter presents many of the main issues that must be understood to arrive at an appropriate
method for appraising the rationality of theistic worldviews. It outlines several conceptions of
theism; it explores the kinds of evidences possible for it and compares those with the kinds
appropriate to confirming scientific theories; and it specifies a range of positive attitudes, such as
faith and hope, that theists may have regarding the existence of God. The chapter considers both
the kind and degree of rationality of theistic attitudes and the need for rationality in actions based
on those attitudes. Both theistic attitudes and certain actions based on them are shown to be
important, and their rationality is also shown to be both a highly complex matter and a status that
is not ruled out on methodological grounds.
Method: Atheism ...............................................................................................................................49
Graham Wood, Lecturer, University of Tasmania
vii
CONTENTS
Suppose that you are given two worldviews, one championed by a theist and one championed by an
atheist. What method or methods should be used in attempts to assess the comparative merits of
these worldviews? What kinds of considerations should feed into these methods? This chapter
begins with a discussion of themes central to answering these two questions, including specifying
two worldviews to allow for meaningful comparison. Then a series of topics is addressed for the
benefit of a person, identified as the “undecided person,” who does not yet endorse either
worldview, in order to establish if there are reasons to prefer the atheistic worldview presented here.
TOPIC 3: LOGIC
Logic: Theism ....................................................................................................................................65

Mashhad Al-Allaf, American University of Ras Al Khaimah
This chapter considers the plausibility of theism by giving attention to the logic of argumentation
offered for the existence of God. It also discusses the various attributes of God, including
omnipotence, omniscience, and impassibility.
Logic: Atheism ...................................................................................................................................81
Peter Millican, Gilbert Ryle Fellow and Professor of Philosophy, Hertford College
This chapter discusses whether theism can be either established or refuted on broadly logical
grounds, and considers the role of logic in theistic and anti-theistic argument. In the first section, I
briefly review some systems of formal logic, highlighting some issues about logical proof and
providing background to the subsequent discussion. In the next section, I make some important
general points about logic and its limits: what we can reasonably expect to achieve by logical
argument. I then turn to the question of whether theism can be refuted a priori on the basis of
internal inconsistency, discuss how it might be defined so as to evade such refutation, and briefly
consider arguments for and against theism that are aprioristic in the sense of being based on
minimal empirical data (such as the existence of contingent things, or the existence of evil). The last
main section examines the ontological argument, which purports to establish theism purely a priori.
TOPIC 4: DOXASTIC FOUNDATIONS
Doxastic Foundations: Theism .........................................................................................................103

Paul K. Moser, Professor of Philosophy, Loyola University Chicago
This chapter focuses on some epistemic concepts and their bearing on theism. It considers the
nature of belief both as assent and as a disposition involving trust. It also characterizes foundational
evidence of God’s reality in terms of divine self-manifestation in human moral conscience, whereby
a unique kind of agapē-conviction can arise.
Doxastic Foundations: Atheism ........................................................................................................119
Ali Hasan, Associate Professor, University of Iowa
Richard Carrier, Educator (PhD), The Secular Academy
This chapter centers around the question of whether theism is rational. We begin by discussing
different theories of rationality, and introducing some importantly related epistemic concepts and
controversies. We then consider the possible sources of rational belief in God and argue that even if
these provide some positive support, the fact of religious disagreement defeats the rationality of theism.
TOPIC 5: RELIGIOUS EXPERIENCE
Religious Experience: Theism ...........................................................................................................135

Samuel Lebens, Senior Research Fellow, University of Haifa
What is the philosophical significance of religious experience? Could a religious experience give you
reason to believe in God? If so, what sort of experience? If not, why not? And could the religious
experiences of others give you reason to believe in God even if you’ve never had such an experience
yourself? In this chapter, we explore these questions.
Religious Experience: Atheism ..........................................................................................................147
John R. Shook, Lecturer in Philosophy, Bowie State University
viii THEISM AND ATHEISM: OPPOSING ARGUMENTS IN PHILOSOPHY
CONTENTS
The aim of this chapter is to clarify what is meant by “religious experience,” discuss the ways in
which such experiences can be explained, and assess what kind of evidential support they offer to
either theism or atheism. Atheism uses a variety of arguments concluding that religious experiences
cannot justify the idea that a God is involved.
TOPIC 6: FAITH AND REVELATION
Faith and Revelation: Theism ..........................................................................................................163

Robert Fastiggi, Professor of Systematic Theology, Sacred Heart Major Seminary
This chapter defines revelation and its relation to scripture. The discussion covers topics such as the
qualities of divine revelation, nonreligious explanations, assessing claims and who decides whether
some claim is true or false, how claims are recorded, how reason is related to faith, and how
considerations about faith and revelation might support theism.
Faith and Revelation: Atheism .........................................................................................................179
Evan Fales, Emeritus, University of Iowa
Long tradition connects revelation with faith, understood as trusting the truth of revelation, the divine
source of which is assured by miracles. Recent alternative conceptions of faith and revelation (Søren
Kierkegaard, William Clifford, William James) are critically examined, and difficulties in distinguishing
genuine revelation from imposture and deception are evaluated, including many-contenders objections
and the evidential weight of religious experience. Finally, we examine the difficulties that attend the
reception and interpretation of putative revelations, even assuming divine provenance.
TOPIC 7: MIRACLES
Miracles: Theism .............................................................................................................................195

Ira M. Schnall, Lecturer (retired), Bar-Ilan University
In this chapter, we first examine what a miracle is supposed to be from a theistic point of view.
Then we consider whether, or to what extent, reports of miracles are to be believed. Finally, we deal
with the role of miracles in theistic religions, and in particular, whether miracles can establish the
truth of theism.
Miracles: Atheism ............................................................................................................................211
Arif Ahmed, University Reader in Philosophy, University of Cambridge
This chapter distinguishes three main conceptions of miracles: extraordinary events, violations of
the laws of nature, and divine interventions. Further discussion looks at whether miracles of any
type are possible. The chapter considers David Hume’s argument that we have no reason to think
that any events considered miracles are actual. Finally, the chapter asks whether we should regard
talk of miracles not as a description of anything that happened but rather as an interpretation of
events that are agreed on all hands.
TOPIC 8: RELIGIOUS DIVERSITY
Religious Diversity: Theism .............................................................................................................227

Daniel Rynhold, Professor of Jewish Philosophy, Bernard Revel Graduate School of
Jewish Studies, Yeshiva University
This chapter considers religious diversity and its philosophical implications. Topics discussed
include the purported challenge to theism posed by the links between religious diversity and
geographical and biographical contingencies; theistic approaches to diversity including exclusivism,
inclusivism, pluralism, and relativism; and the extent to which religious diversity supports theism.
Religious Diversity: Atheism ............................................................................................................243
Tiddy Smith, Professor, University of Otago
This chapter explores the nature of religious diversity, the various theological responses to it, and
how the existence of such diversity serves to undermine theism.
THEISM AND ATHEISM: OPPOSING ARGUMENTS IN PHILOSOPHY ix

CONTENTS
TOPIC 9: CAUSATION AND SUFFICIENT REASON
Causation and Sufficient Reason: Theism ........................................................................................259

Victor M. Salas, Associate Professor of Philosophy, Sacred Heart Major Seminary
This chapter discusses the principles of causality and sufficient reason and their relation to theism.
Causation and Sufficient Reason: Atheism .......................................................................................281
Felipe Leon, Professor, El Camino College
This chapter discusses the nature of causation and its fundamental role in the debate between
theism and atheism. The discussion covers cosmological arguments that deploy causal or
explanatory principles to prove God’s existence.
TOPIC 10: A PRIORI
A Priori: Theism..............................................................................................................................301
Aaron Segal, Lecturer, Hebrew University of Jerusalem
This chapter discusses whether the existence of a priori knowledge bears positively on theism, and
argues tentatively that it does.
A Priori: Atheism.............................................................................................................................313
Felipe Leon, Professor, El Camino College
The primary aim of this chapter is to explore whether considerations about a priori domains and
abstract objects favor atheism over theism.
TOPIC 11: OUR UNIVERSE
Our Universe: Theism .....................................................................................................................331

Robert J. Spitzer, S.J., President, Magis Center of Reason and Faith
James Sinclair, Senior Physicist, United States Navy
This chapter will discuss the arguments that have been made in favor of theism on the basis of the
finetuning of our universe for life. It will discuss the objective basis of fine-tuning, six instances of it
in our universe, and consider six candidates as an ultimate explanation for it—a theory of
everything, inflationary cosmology (and its variants), cyclic cosmologies (including Penrose’s
conformal cyclic model), a multiverse, Tegmark’s Level IV multiverse, and transcendent
intelligence. Using a general abductive argument, it concludes that the most likely ultimate
explanation is transcendent intelligence.
Our Universe: Atheism ....................................................................................................................359
Neil A. Manson, Professor of Philosophy, University of Mississippi
Sahotra Sarkar, Professor, University of Texas at Austin
Cory Juhl, Professor, University of Texas at Austin
This chapter discusses putative scientific evidence for the existence of God from biology and from
physics and cosmology. After presenting some of those items of evidence and articulating the
arguments based on them, the authors explain why that evidence and those arguments either
disconfirm or do not support theism over atheism.
TOPIC 12: HUMAN HISTORY
Human History: Theism ..................................................................................................................377

Brendan Sweetman, Professor, Rockhurst University
This essay discusses the arguments in favor of theism and atheism through a consideration of
human evolutionary history, including the relationship between evolution and scriptural revelation,
the role of chance in evolution and science, as well as the problem of evil and suffering in the
universe. The question of design in nature, along with the development of human civilizations and
the demographics of theistic belief, is also a focus.
x THEISM AND ATHEISM: OPPOSING ARGUMENTS IN PHILOSOPHY
CONTENTS
Human History: Atheism .................................................................................................................387

Michael Ruse, Lucyle T. Werkmeister Professor and Director of HPS Program, Florida State University
Susana Nuccetelli, Professor, St. Cloud State University
Keith Parsons, Professor of Philosophy, University of Houston–Clear Lake
Gregory Paul, Independent Scholar, Baltimore
Matthew Wade Ferguson, Doctoral Candidate in Classics, University of California, Irvine
Theism, the belief that the God of Abrahamic religions objectively exists, faces a challenge from
evolutionary accounts of belief in supernatural agencies. It does put the burden of argument on
theists, who must argue persuasively for either the epistemic justification of their fundamental belief
or for the inaccuracy of the evolutionary hypothesis.
TOPIC 13: HUMAN BEINGS
Human Beings: Theism ...................................................................................................................407

Brendan Sweetman, Professor, Rockhurst University
This chapter defends the view that theism is a better explanation than naturalism of the remarkable
phenomenon of the human mind and its activities. It discusses arguments concerning dualism and
materialism, consciousness and intentionality, personal identity, and free will and moral agency.
Related issues such as determinism, God and freedom, true beliefs and knowledge, and the
significance of “God of the gaps” objections are also considered.
Human Beings: Atheism ..................................................................................................................429
Kenneth Williford, Associate Professor and Chair, University of Texas at Arlington
Konrad Talmont-Kaminski, The Head of the Society and Cognition Unit, University of Bialystok
Diane Proudfoot, Professor, University of Canterbury
Mariam Thalos, Professor of Philosophy, University of Tennessee
This chapter observes the way theists have supposed that metaphysical considerations about human
beings have supported theism over atheism. The chapter further discusses questions about
consciousness and intentionality, reason, personal identity, and freedom. It is argued that the
incompleteness of current neuroscientific accounts of these phenomena does not lend any
significant support to a theistic account of them.
TOPIC 14: ETHICS
Ethics: Theism .................................................................................................................................449

Michael J. Harris, Faculty of Divinity, University of Cambridge
This chapter first discusses whether the existence of moral norms constitutes proof of the existence
of God. It then evaluates various forms of the divine command theory of ethics. Finally, the chapter
considers certain concepts such as human rights, conscience, and virtue as supporting theism,
atheism, or neither.
Ethics: Atheism ................................................................................................................................467
Jason Thibodeau, Assistant Professor of Philosophy, Cypress College
Thaddeus Metz, Professor, University of Johannesburg
Bruce Russell, Professor, Wayne State University
David Neil, Lecturer in Philosophy, University of Wollongong
The aim of this chapter is not to consider particular ethical questions; it is to ask whether there are
general facts about morality and our ability to make moral judgments that count in favor of either
theism or atheism.
TOPIC 15: MEANING
Meaning: Theism.............................................................................................................................491
Mirela Oliva, Associate Professor, University of St. Thomas
THEISM AND ATHEISM: OPPOSING ARGUMENTS IN PHILOSOPHY xi

CONTENTS
This chapter discusses the ways in which theism provides answers to the question of the meaning of
life. Theists read this modern question as a quest for the ultimate meaning of our life that entails
the understanding of our existence in the great scheme of things created by God. Theistic positions
fall under two main categories, namely pluralism and singularism. This latter is in turn specified in
four classes: metaphysical, experiential, narrative, and subjective.
Meaning: Atheism............................................................................................................................507
Thaddeus Metz, Professor, University of Johannesburg
This chapter explores what it means to live a meaningful life. It also evaluates the idea that belief in
God is required for such a life.
TOPIC 16: SUFFERING
Suffering: Theism ............................................................................................................................523

Siobhan Nash-Marshall, Mary T. Clark Chair of Christian Philosophy, Manhattanville College
This chapter discusses what evils exist in the world with a specific focus on suffering. Natural evil
and moral evil are explained in this chapter as well as the problem of evil.
Suffering: Atheism ...........................................................................................................................539
Bruce Russell, Professor, Wayne State University
Daniel Linford, Graduate Student, Purdue University
This chapter discusses the existence of suffering as it pertains, in several dimensions, to traditional
theism. A central component of this chapter concerns the conflict between theism and the premise
that excessive gratuitous evil exists.
TOPIC 17: SCIENCE
Science: Theism ...............................................................................................................................561

Guy Consolmagno, Director of the Vatican Observatory, President of the Vatican
Observatory Foundation
This chapter discusses the relation between religion and science, providing an overview of
contemporary science and the very nature of scientific knowledge. Predominant methods of science
are detailed as well as its proper subject matter and goals, which are essential for addressing questions
about whether scientific knowledge seems to support or to conflict with the sacred texts of theism.
Science: Atheism ..............................................................................................................................583
Herman Philipse, Distinguished Professor of Philosophy, Utrecht University
Kenneth Williford, Associate Professor and Chair, University of Texas at Arlington
Keith Augustine, Executive Director and Editor-in-Chief, Internet Infidels
Taner Edis, Professor, Truman State University
A proper assessment of the bearing of scientific inquiry on theistic religion requires recognition that
conflict, mutual consistency, independence, or concilience are possible, but depends on the
methods accepted and the claims made in each domain at particular times and places—which can,
and have, varied. This chapter, therefore, focuses on questions relating to whether the best
methods, findings, and theories in contemporary scientific disciplines support, cohere with, or
conflict with commitments made by theistic theologies.
TOPIC 18: THEORIES OF RELIGION
Theories of Religion: Theism ...........................................................................................................605

Margaret I. Hughes, Tutor, Thomas Aquinas College
After giving a brief overview of the history of the development of theories of religion, this chapter
delves into scientism, which is the foundation on which many contemporary theories rest. It
considers the way in which scientism has come to dominate a certain approach to studying religion,
xii THEISM AND ATHEISM: OPPOSING ARGUMENTS IN PHILOSOPHY
CONTENTS
and then offers a critique of this approach. It concludes that faith is necessary in order to come to a
fuller understanding of the phenomenon of religion because, unlike an approach based in
scientism, faith allows for an openness to finding what is true in any religion.
Theories of Religion: Atheism ..........................................................................................................619
Konrad Talmont-Kaminski, Head of the Society and Cognition Unit, University of Bialystok
Todd Tremlin, Lecturer, Central Michigan University
Gregory Dawes, Professor, University of Otago
This chapter discusses whether there are satisfactory natural theories of religion. Furthermore, the
authors consider whether these natural theories favor atheism.
TOPIC 19: PRUDENTIAL/PRAGMATIC ARGUMENTS
Prudential/Pragmatic Arguments: Theism ........................................................................................637

Joshua Golding, Professor of Philosophy, Bellarmine University
This chapter discusses pragmatic arguments for religious commitment, such as the wager-approach,
the will-to-believe, and an argument for pragmatic faith.
Prudential/Pragmatic Arguments: Atheism .......................................................................................645
Richard Feldman, Professor of Philosophy, University of Rochester
Malcolm Murray, Professor, University of Prince Edward Island
Charles Pigden, Associate Professor, University of Otago
The aim of this chapter is to decide whether, in the absence of adequate evidence of the (probable)
truth or falsity of theism, we may be justified in making a religious commitment on prudential or
pragmatic grounds.
TOPIC 20: FINAL RECKONINGS
Final Reckonings: Theism ................................................................................................................665

Joseph W. Koterski, S.J., Associate Professor of Philosophy, Fordham University
The first half of this chapter examines Augustine of Hippo’s use of philosophical distinctions to
clear away various difficulties that stand in the way of pursuing such questions as the existence of
God and the relations between some of the attributes normally attributed to God (omnipotence,
omniscience, omnibenevolence) and the problems presented by evil and freedom. The second half
of the chapter examines natural law theory to make a case that morality requires the existence of
God.
Final Reckonings: Atheism ...............................................................................................................679
Graham Oppy, Professor of Philosophy, Monash University
This chapter explains how the sum of the considerations in the previous chapters fit together in a
comprehensive case for preferring atheism to theism.
Index .............................................................................................................................................695
THEISM AND ATHEISM: OPPOSING ARGUMENTS IN PHILOSOPHY xiii

Preface
Theism and Atheism: Opposing Arguments in Philosophy is a coedited publication, with each of us
heading up separate editorial boards representing theism and atheism. The format of the work is such
that an essay on a topic (e.g., Religious Experience, Miracles, Our Universe, Human Beings, Science)
by an author taking the side of theism is followed by an essay on the same topic by an author taking
the side of atheism.
Reading through the essays on both sides has proved a great education for us as editors, and we
hope that those who use this volume will similarly profit from the thoughtful reflections offered here.
It became quite clear in the process of editing this volume—if it was not already clear—that
there is not a single position called “theism” nor just one position called “atheism” but rather many
forms of both. In our separate introductory notes, we each have more to say about how we dealt with
the diversity of approaches and positions on our side.
For the purposes of this volume, the basic question that we asked our essayists to answer is
whether the material assigned to them to discuss provides good reasons to prefer theism to atheism, or
atheism to theism, or neither one to the other. Naturally, our standing expectation was (a) that
authors on the theist side would argue that the material assigned to them provided good reasons to
prefer theism to atheism or, at worst, provided no reason to prefer atheism to theism; and (b) that
authors on the atheist side would argue that the material assigned to them provided good reasons to
prefer atheism to theism or, at worst, provided no reason to prefer theism to atheism. (Of course, it is
open to theists to suppose that there are some considerations—overall outweighed—that favor
atheism over theism; and it is open to atheists to suppose that there are some considerations—overall
outweighed—that favor theism over atheism. Moreover, it is not, at least initially, ruled out that there
are some, perhaps even many, cases where there is agreement across the two sides that neither is
favored by the advanced considerations.)
To make for an orderly program of investigation, the theist and atheist editorial teams agreed on
a list of twenty topics—some more theoretical, some more practical. In at least some cases, discussion
of what might be viewed as a single topic is distributed over several chapters; often, though perhaps
not always, these cases are marked with internal cross-references. In some cases, the theist position or
the atheist position has a single author to cover a given topic. In other cases, multiple authors have
joined together to serve as coauthors on a given topic. In yet other cases, at least one side chose to
distribute the authorship of an essay into segments in order to provide for better coverage.
It is our common conviction that this work is profoundly important for a number of reasons.
Among them, we feel that there is a need for those who disagree on important questions to put
forward their best articulations of their views and to deal with the best articulations of views on the
opposing side while avoiding any sort of caricature. Proceeding in this way not only encourages
authors to aim for respect, clarity, and cogency but helps to provide readers with essays that can assist
them in making up their own minds on these questions.
It goes without saying that the central question of this volume—the merits of the case for theism
and the merits of the case for atheism—has enormous importance for our lives. If one supposes there
are compelling reasons to hold that God exists, that has profound implications for how one lives.
Likewise, if one supposes there are compelling reasons to hold that no god (or God) exists, that, too,
has profound implications for how one lives. Each of us—no matter what we are currently inclined to
believe—has a vital interest in learning whether there are compelling arguments on either side in this
xv
PREFACE
dispute, that is, whether there are reasons that ought to compel all of us to agree in our judgment
about the existence of God (or gods).
A work of this size is a significant undertaking. In addition to our own acknowledgments in our
separate editors’ notes, we wish jointly to offer recognition and hearty thanks to Laurie Malashanko at
Gale Cengage for conceiving of and proposing this publication to us, and to Alan Hedblad, Jonathan
Vereecke, and Rebecca Parks, also at Gale Cengage, for providing such able assistance in its
production. We offer our gratitude to the associate editors—David Shatz, Robert Fastiggi, and
Mashhad Al-Allaf on the theist team, and Gregory Dawes and Evan Fales on the atheist team—for
their enormous contributions to the completion of this project. Our appreciation also extends to all
of the authors who agreed to contribute to this work.
Graham Oppy
xvi THEISM AND ATHEISM: OPPOSING ARGUMENTS IN PHILOSOPHY
Editor’s Note: Theism
Philosophical integrity always requires a genuine respect for the positions with which one disagrees
and for their proponents, as well as a sustained readiness to explain one’s position and to defend it
against objections brought to bear against it. As the theist editor of this publication, let me express my
thanks to the atheist editor, Graham Oppy, to his editorial team, and to the authors for their
contributions. I learned much from their work, not least in clarifying the problems and challenges
that theists must address.
I am deeply grateful to my fellow theist editors for their hard work and great wisdom: Prof.
David Shatz, Chair of the Philosophy Department at Yeshiva University in New York, Prof.
Mashhad Al-Allaf, Chair of the Humanities and Social Sciences Department at the American
University of Ras Al Khaimah in the United Arab Emirates, and Prof. Robert Fastiggi, Professor of
Systematic Theology at the Sacred Heart Major Seminary in Detroit. Their advice and dedication
were invaluable in selecting the topics, finding appropriate authors, and reviewing the chapters
submitted.
Working together with authors of various religious traditions proved a delightful experience of
interreligious cooperation and dialogue. By common agreement, the chapters presented here make
reference not only to the religious beliefs and practices of their authors but to those of monotheistic
religion considered more generally. In writing and editing these chapters, we were mindful that there
are brands of theism that are not well represented here, including pantheism, panentheism, and
polytheism, but attempting to treat those positions proved to be unfeasible. The authors were free to
employ whatever philosophical approaches they deemed most appropriate for their assigned topics.
What we were most eager to do was to show that taking theistic positions, in keeping with the
embrace of religious faith, is not (as sometimes alleged) merely the result of arbitrary choices but a
stance that is rationally defensible.
The essays on the theist side are composed by believers who belong to some of the world’s great
monotheistic traditions—some are Christian, some Jewish, some Muslim. By prior agreement, we
have not here tried to address the question of the relative merits of a particular tradition of theism.
Nonetheless, sometimes the authors understandably use examples from the traditions of belief that
they know best. But we encouraged them also to take into consideration traditions other than their
own, for each essay in this volume is intended to provide an examination of the general question of
whether the specific topic under consideration lends itself to providing support for belief in God or,
on the contrary, support for atheism, or for neither side. The reader will also find that the authors of
these essays come from diverse philosophical schools, and this too makes a difference with regard to
the type of case they make.
One additional point about the theist position that deserves special mention here is this:
Sometimes people get the impression that belief in God is simply a matter of following the
traditions and practices of one’s family or one’s land, or that it is an arbitrary act of will or an
idiosyncratic decision, as if no reasons for the position could be given. Admittedly, some believers
do hold such views. But it is by no means the prevailing opinion among theists, and the theistic
contributions to this volume try to make as clear as possible the reasoning that can be provided for
religious belief. The charge given to our essayists was to bring philosophical scrutiny to these
patterns of reasoning.
xvii
EDITOR ’ S NOTE : THEISM
We hope that the resulting volume will assist readers in their own quest to explore these
important issues.
Philosophy Department
Fordham University, New York
xviii THEISM AND ATHEISM: OPPOSING ARGUMENTS IN PHILOSOPHY
Editor’s Note: Atheism
Atheists are a diverse lot. Some who self-identify as atheists claim that atheists believe that there are
no gods. Some who self-identify as atheists claim that atheists merely lack the belief that there are
gods. Others who self-identify as atheists endorse a wide range of further claims about the proper
characterisation of atheism. Given that those who self-identify as atheists argue among themselves
about what atheism is, perhaps the only thing that unites atheists is their shared identification as
atheists.
Given the diversity of atheists, an argument for representation of diverse approaches to atheism
in the chapters of this book won out. In assigning chapters—and parts of chapters—to particular
authors, no consideration has been given to the consistency of what is said across chapters or between
the different parts of individual chapters. Moreover, there is no doubt of a significant inconsistency
across chapters and between the different parts of individual chapters. (But, no matter where your
sympathies lie in the debate between atheists and theists, you should not regard the presence of
differing opinions among the atheists as a weakness of the atheist side.)
How, then, were the authors of chapters and parts of chapters selected? The primary
consideration—apart from self-identification as an atheist—was that chosen authors should be
recognized experts on the material assigned to them. In some of the atheist chapters, there are four or
five different authors, because each of those atheist authors brings special expertise to a particular
aspect of the topic under discussion. I would like here to record my gratitude to all of the authors who
agreed to contribute material to the atheist side of the book; in each case, the whole would be
diminished by the removal of what the author in question contributed.
I am deeply indebted to Gregory Dawes and Evan Fales, the associate editors on the atheist side.
Greg and Evan were involved in all areas of the development of this project, including identification
of the topics that became the subject matter of the twenty chapters, mapping of content to be covered
in each of those chapters, assignment of authors to chapters and parts of chapters, reading and
commenting on drafts of chapters and parts of chapters, and contributing some of the material that
makes up the twenty chapters. Without their patience, wisdom, and sage advice, the atheist
contribution to the work would have nowhere near the quality that it in fact has.
I am similarly indebted to Joseph W. Koterski for his editorial work on the theist side of this
book. The claims that I made above about the diversity of atheists apply in equal measure to theists.
He has done a wonderful job of incorporating diverse theistic perspectives into the theist chapters.
Moreover, it should be said that the identification of the topics that became the subject matter of the
twenty chapters emerged from cooperative discussion between the atheists and theists, and that this
discussion proceeded fruitfully and amicably. During the production of the work, I have enjoyed
reading drafts of the theist material, noting the many points of engagement between paired chapters.
I would also like to acknowledge the work of many people at Gale Cengage. This includes Laurie
Malashanko, who, as acquisitions editor, oversaw the first part of the project, in concert with senior
development editors Alan Hedblad, Rebecca Parks, and Jonathan Vereecke, who oversaw the
subsequent stages.
Finally, I must thank the many other people who have supported me during the production of
this work. Most prominently, I am grateful to my colleagues at Monash University—in the
Department of Philosophy, in the Faculty of Arts, and in the wider university community—who
have supported my work in philosophy of religion throughout the many years that I have been
xix
EDITOR ’ S NOTE : ATHEISM
fortunate enough to have them as my colleagues. I am also indebted to my colleagues in Australia,

New Zealand, and Singapore who work in the discipline of philosophy, and who have supported my
philosophical enterprises, particularly through the Australasian Association of Philosophy. Then there
are the many friends who have supported me in enterprises outside academic philosophy, including,
in particular, friends at the Glen Waverley Cougars Cricket Club and the Monash Blues Football
Club. Last, but certainly not least, I acknowledge the support of family and friends, in particular my
wife, Camille, and my sons, Gilbert, Calvin, and Alfred. When one name goes on the cover of a
book, there is a multitude that stands behind.
Graham Oppy
School of Philosophical, Historical and International Studies
Monash University, Melbourne, Australia
xx THEISM AND ATHEISM: OPPOSING ARGUMENTS IN PHILOSOPHY
Contributors
Arif Ahmed Evan Fales Joseph W. Koterski, S.J.

University of Cambridge, UK University of Iowa Fordham University, NY
MIRACLES : ATHEISM FAITH AND REVELATION : ATHEISM FINAL RECKONINGS : THEISM
PRUDENTIAL / PRAGMATIC ARGUMENTS :

Mashhad Al-Allaf ATHEISM Samuel Lebens
American University of Ras Al Khaimah, University of Haifa, Israel
THEORIES OF RELIGION : ATHEISM
United Arab Emirates RELIGIOUS EXPERIENCE : THEISM
LOGIC : THEISM
Robert Fastiggi Felipe Leon
Robert Audi Sacred Heart Major Seminary, MI El Camino College, Torrance, CA
FAITH AND REVELATION : THEISM
University of Notre Dame, IN A PRIORI : ATHEISM
METHOD : THEISM
Richard Feldman CAUSATION AND SUFFICIENT REASON :
University of Rochester, NY ATHEISM
Keith Augustine
PRUDENTIAL / PRAGMATIC ARGUMENTS :
Internet Infidels, Carson City, NV
ATHEISM Daniel Linford
SCIENCE : ATHEISM
Purdue University, IN
Matthew Wade Ferguson SUFFERING : ATHEISM
Brian Carl University of California, Irvine
Dominican House of Studies, HUMAN HISTORY : ATHEISM Neil A. Manson
Washington, DC
University of Mississippi
DEFINITION : THEISM
Joshua Golding OUR UNIVERSE : ATHEISM
Bellarmine University, Louisville, KY
Francisco Romero Carrasquillo PRUDENTIAL / PRAGMATIC ARGUMENTS : Thaddeus Metz
Universidad Panamericana, Mexico THEISM University of Johannesburg, South Africa
DEFINITION : THEISM
ETHICS : ATHEISM
Michael J. Harris
Richard Carrier MEANING : ATHEISM
University of Cambridge, UK
The Secular Academy, Washington, DC ETHICS : THEISM
DOXASTIC FOUNDATIONS : ATHEISM Peter Millican
Ali Hasan Hertford College, Oxford University, UK
MIRACLES : ATHEISM
University of Iowa LOGIC : ATHEISM
SCIENCE : ATHEISM DOXASTIC FOUNDATIONS : ATHEISM
Paul K. Moser
Guy Consolmagno Margaret I. Hughes Loyola University Chicago, IL
Vatican Observatory Foundation, Thomas Aquinas College, Santa Paula, DOXASTIC FOUNDATIONS : THEISM
Tucson, AZ CA
SCIENCE : THEISM THEORIES OF RELIGION : THEISM Malcolm Murray
University of Prince Edward Island,
Gregory Dawes Mark Johnson Canada
University of Otago, New Zealand Marquette University, Milwaukee, WI PRUDENTIAL / PRAGMATIC ARGUMENTS :
THEORIES OF RELIGION : ATHEISM DEFINITION : THEISM ATHEISM
Taner Edis Cory Juhl Siobhan Nash-Marshall

Truman State University, Kirksville, MO University of Texas at Austin Manhattanville College, Purchase, NY
SCIENCE : ATHEISM OUR UNIVERSE : ATHEISM SUFFERING : THEISM
xxi
CONTRIBUTORS
David Neil Michael Ruse Robert J. Spitzer, S.J.

University of Wollongong, Australia Florida State University Magis Center of Reason and Faith,
ETHICS : ATHEISM HUMAN HISTORY : ATHEISM Garden Grove, CA
OUR UNIVERSE : THEISM
Robert Nola Bruce Russell

The University of Auckland, New Wayne State University, MI Brendan Sweetman
Zealand ETHICS : ATHEISM Rockhurst University, Kansas City, MO
DEFINITION : ATHEISM SUFFERING : ATHEISM HUMAN BEINGS : THEISM
HUMAN HISTORY : THEISM

Susana Nuccetelli Daniel Rynhold
St. Cloud State University, MN Bernard Revel Graduate School of Jewish
Konrad Talmont-Kaminski
HUMAN HISTORY : ATHEISM Studies, Yeshiva University, NY
University of Bialystok, Poland
RELIGIOUS DIVERSITY : THEISM
HUMAN BEINGS : ATHEISM
Mirela Oliva
Victor M. Salas THEORIES OF RELIGION : ATHEISM
University of St. Thomas, Houston, TX
MEANING : THEISM
Sacred Heart Major Seminary, Detroit, MI
CAUSATION AND SUFFICIENT REASON : Mariam Thalos
THEISM University of Tennessee, Knoxville
Graham Oppy
HUMAN BEINGS : ATHEISM
Monash University, Australia Sahotra Sarkar
FINAL RECKONINGS : ATHEISM
University of Texas at Austin
Jason Thibodeau
OUR UNIVERSE : ATHEISM
Keith Parsons Cypress College, Cypress, CA
ETHICS : ATHEISM
University of Houston–Clear Lake, TX Ira M. Schnall
HUMAN HISTORY : ATHEISM Bar-Ilan University, Ramat-Gan, Israel
MIRACLES : THEISM Todd Tremlin
Central Michigan University, Mount
Gregory Paul
Aaron Segal Pleasant
Independent Scholar, Baltimore, MD
Hebrew University of Jerusalem, Israel THEORIES OF RELIGION : ATHEISM
HUMAN HISTORY : ATHEISM
A PRIORI : THEISM
David B. Twetten
Herman Philipse
John R. Shook Marquette University, Milwaukee, WI
Utrecht University, Netherlands
Bowie State University, MD DEFINITION : THEISM
SCIENCE : ATHEISM
RELIGIOUS EXPERIENCE : ATHEISM
Kenneth Williford
Charles Pigden James Sinclair University of Texas at Arlington
University of Otago, New Zealand United States Navy, Washington, DC HUMAN BEINGS : ATHEISM
PRUDENTIAL / PRAGMATIC ARGUMENTS : OUR UNIVERSE : THEISM
ATHEISM SCIENCE : ATHEISM
Tiddy Smith
Diane Proudfoot University of Otago, Dunedin, New Graham Wood
University of Canterbury, New Zealand Zealand University of Tasmania, Australia
HUMAN BEINGS : ATHEISM RELIGIOUS DIVERSITY : ATHEISM METHOD : ATHEISM
xxii THEISM AND ATHEISM: OPPOSING ARGUMENTS IN PHILOSOPHY
TOPIC 1
Definition: Theism
David B. Twetten
Associate Professor, Department of Philosophy
Marquette University, Milwaukee, WI
Brian Carl
Assistant Professor of Philosophy
Dominican House of Studies, Washington, DC
Mark Johnson
Associate Professor, Department of Theology
Marquette University, Milwaukee, WI
Francisco Romero Carrasquillo
Associate Professor, Departamento de Humanidades
Universidad Panamericana, Mexico
This chapter discusses how theism and atheism are best defined, primarily as a function of what “God” or
“god” means. It sets out the range of meanings that “god” and “God” can have in different argumentative
contexts, suggesting that proofs for God’s existence need not immediately conclude to the existence of a being
with personal or omni-attributes. It also explores the notions of analogy and apophaticism, given that these
themes condition the way many theists understand the affirmation that God exists. The chapter thereby
offers principles that should govern the proper use of words in theism versus atheism debates.
WHAT THEISM, WHICH GOD?
“Theism” enters the English language on analogy with the older terms “atheism” and “polytheism”
(based on Greek atheos and polutheos). If atheism is the denial that God exists or disbelief in God,
theism should be the acceptance that God exists or belief in God. However, much turns on what
the word “God” means. The oldest entry in the Oxford English Dictionary (Cudworth 1678) takes
“theism” to include polytheism: the affirmation of many gods. In fact, theos in Greek, as originally
used, not unlike its oldest Hebrew and Latin equivalents, ’el and deus, respectively, is best captured
by the lowercase term “god.” The distinction between “god” and “God” is an orthographic
convention that has become standard in Romance languages since the advent of the printing press.
Nevertheless, we should not take for granted that we understand the import of the convention. The
convention is not even possible in languages that lack capital letters or that capitalize all nouns, as in
German, although context can still convey the distinction intended. The conceptual difference
reflected in the orthography of the Romance languages precedes Johannes Gutenberg, but what
the difference amounts to needs to be explained.
The word “theist” as used today varies in its semantic range. (1) In the widest sense, a theist is
one who accepts gods and/or God. Presumably an atheist disavows both. However, other possibilities
remain: the Greeks would call a person atheos who refused worship to the gods of the local religion.
(2) In a second sense, a theist includes a “deist” or a “pantheist,” but excludes someone who believes
1
Definition: Theism
only in gods and not God. As usually understood, a deist

KEY CONCEPTS believes in God in some sense of the term but rejects many or
most of the properties of “classical theism,” as it has been
dubbed (Owen 1971). A deist typically denies omni-properties,
Analogy such as omniscience or omnipotence, and perhaps also denies
Apophatic (Negative) Theology God’s awareness of human affairs. A deist might even reject
Classical Theism personal properties altogether and affirm God to be a noncog-
nizant cosmic force. A pantheist takes God in some way to be
Classical Theist Kind-Name Theory of God one with the universe. (3) In a common, narrow sense of the
Creation Theology term, a theist, as opposed to a deist, affirms all the properties
Definiendum of classical theism. But what are God’s properties according to
classical theism, and why those properties rather than others?
Differentia Undoubtedly the properties are thought to coincide with those
Equivocation of the God of the great monotheist religions, Judaism, Chris-
Identity Stage of Arguments for God’s tianity, Islam, and Sikhism, to say nothing of certain forms of
Existence Hinduism, and so on, as grounded in such authorities as their
scriptures. Classical theism, then, appears to be a term of art,
Kataphatic (Affirmative) Theology describing a family resemblance. Nevertheless, in addition to
Minimal Kind-Name Theory of God intra- and interreligious disagreement over such properties as
New Theist God’s spirit or God’s immanence in all things, in-house dis-
putes within classical theism remain. A “process theist,” for
Omni-properties of God example, finds ways of saying that God can change without
Open Theist being imperfect; an “open theist” or a “new theist” denies cer-
Perfect Being Theology tain standard versions of God’s providence over freewill actions;
and other disagreements abound as well.
Problem of the Gap
The range of possible meanings for “theism” is thus
Process Theist largely founded on a corresponding range in the meanings of
Pros Hen Equivocation “god” and “God.” Philosophers advance their preferred ver-
sion of theism largely in the course of defending some account
Theistic Personalism
of these terms as the most satisfactory. It is imperative, then,
Univocity for interlocutors discussing theism and atheism to clarify first
what they mean by “god” or “God.” In what follows, we
review the major alternative theories of what the term “God”
means, first in the abstract, then as they are applied in the
two contexts in which the term is used: proofs of God’s existence and discussions of God’s nature.
Distinguishing these contexts is indispensable for adjudicating among the theories.
IS “GOD” A PROPER NOUN OR THE NAME OF A KIND?
In the lowercase sense, “god” ranges as widely as does theos in classical Greek, which minimally sig-
nifies something immortal or superhumanly powerful (François 1957, 305–316). A precise defini-
tion can be given after first defining the uppercase sense: a god is not “God” or “a God.” The
uppercase term “God” is used in two ways: as a singular noun or name of address, like a proper
name, or as the name of a kind (a “kind name”). Interpretations of the term vary accordingly.
Contemporary philosophers of religion commonly hold that “God” is exclusively a proper
name. Like a proper name, “God” is a capitalized term often used in a direct form of address,
and only one thing is named by it (see den Brouw 1994, 18). Since “God” is not used in the plu-
ral, it is argued, it cannot be a count noun (Ziff 1961). Debate accordingly has focused on
whether the name “God” is best explained by a descriptivist or a referentialist account of proper
names. According to simple descriptivist accounts, a proper name is an abbreviated definite
description (unless, on one account, it is a logically proper name, like “this” used of an individual).
According to a referentialist account of proper names, however, a proper name is a mere mark (as
it is for John Stuart Mill), and descriptions are used only to fix the reference of the name, that is,
2 THEISM AND ATHEISM: OPPOSING ARGUMENTS IN PHILOSOPHY
Definition: Theism
to determine to whom or what the name refers. The reference to God might be secured, after hav-
ing been handed down through a causal series, by, for example, “that to which the founder of my
religious community referred” (Alston 1988, 119). One can, of course, disagree over which
description best singles out that to which the proper name “God” refers. Richard Swinburne,
for example, initially prefers “the personal ground of being” (Swinburne 1977, 235–239), though
he subsequently opts for “a person without a body (i.e., a spirit) who necessarily is eternal, per-
fectly free, omnipotent, omniscient, perfectly good, and the creator of all things” (Swinburne
2004, 7); still later he replaces it with “a person, omnipotent, omniscient, and perfectly free,”
since from these three properties all of the other aforementioned “essential properties” of God
can be derived (Swinburne 2010, 8–10).
That said, a number of arguments suggest that “God” is not a proper name at all. (1) Bertrand
Russell argues that “God” is not a (proper) name but a (disguised) description; otherwise, one could
not (logically) affirm “There exists God” (Russell 1918, 175–176). (2) If “God” were a proper
name, we could not meaningfully ask, “Is there only one God, or are there many Gods?” (Geach
1961, 109) (3) The missionary and the pagan could agree on the meaning of “God” in saying,
respectively, “YHWH is God” and “Woden is God.” Their disagreement, intuitively, is not resolved
by consulting a directory of names and applying the principle of transitivity. (4) “God,” unlike a
proper name, is translated, not transliterated, from one language to another (Geach 1969, 57–58,
108–109, 115). With Russell, Peter Geach holds “God” to be a definite description, like “Mr.
Prime Minister”: essentially, “God” means “the one and only God,” or perhaps “the one and only
bearer of divine attributes.” An alternative is Thomas Aquinas’s claim, as Geach acknowledges and
as we shall discuss: “God” is not a proper name but a common noun or kind name (nomen naturae).
In addition, some philosophers even maintain that “God” is both a kind name and a proper name
(Attfield 1971, 22–23), depending on the context, whereas others argue that such a term is impos-
sible (Durrant 1973a).
The major contemporary alternative to taking “God” as an individual name is the classical
theist kind-name theory (Peterson et al. 2009, 9–10). In Meditations on First Philosophy 3, René
Descartes writes: “By the name ‘God’ I understand a certain substance that is infinite, independent,
supremely intelligent, supremely powerful, and by which both I myself and everything else whatso-
ever, if anything else does exist, have been created” (Descartes 1904, 45). One interpretation of clas-
sical theist kind names is highly prescriptive: if any one attribute of classical theism is missing, there
is a reference failure in the use of “God” (Kretzmann 1997, 112–113, 169; cf. Geach 1969, 114).
Alternatively, one might hold that different kind names serve different argument strategies in
discussing God. “The greatest possible being” grounds “perfect being” theology, whereas “creation
theology” offers as a rival definition of “God” “the font of all possible existence distinct from
himself” (Morris 1991, 278–280). Below, we analyze “God” differently according to the various
contexts in which the term is used by philosophers.
Adjudicating contemporary disagreements over the term “God” will ultimately require addres-
sing many philosophical issues. The distinction between (logically) proper names and descriptions
arises in a post-Fregean semantics that requires univocal terms and the separate treatment of univer-
sal content and individual objects (Sullivan 2003, 16–82). By contrast, Aristotelian semiotics typi-
cally presuppose that conceptual content is somehow in objects and that, at least prior to William
of Ockham, kind terms signify the world through universal conceptual contents in the mind drawn
from the world (Klima 1991). The philosopher must also, in this case, yield to the evidence of the
grammarian and philologist. It is instructive to contrast the emergence of the capital “G” sense of
“God” in Hebrew, Greek, and Latin thought.
THE EMERGENCE OF THE UPPERCASE “GOD”
Contemporary use of “God” reflects the practice of linguistic communities that have taken seriously
such injunctions as: “Thou shalt have no other gods (’elohim) before me,” and “Thou shalt not take
the name of YHWH, your god (’eloheka), in vain” (Exodus 20:3–7). Nevertheless, the Hebrew
THEISM AND ATHEISM: OPPOSING ARGUMENTS IN PHILOSOPHY 3

Definition: Theism
scriptures, taken broadly, are themselves perhaps better called “henotheist” than “monotheist”: they
put worship of one god (monolatry) before all other gods. On a standard scholarly view, the wide-
spread recognition of “monotheism” appears to be gradual, namely, the recognition that there is
only one “God,” and that all others are merely “gods” or angels (an “inclusive monotheist” holds
also the existence of “gods”).
On this view, monotheism becomes dominant starting from the exilic period, though it is
certainly rooted in such previous beliefs as are reflected, for example, in the creation accounts of
Genesis (Smith 2002, 195–199). Isaiah 44:6–7, 24 reads: “I am the first and I am the last; there
is no god apart from me.… I am YHWH, who made everything.” At the same time, the pairing
of “monotheism” vs. “polytheism” fits the propositional assessments of religions favored during
the Enlightenment. As such, they are inadequate to the data explained (MacDonald 2003, 5–53,
211–221). “Polytheism” today is normally the claim not that there are many Gods but that there
are many gods and no God (in this sense, only some forms of Hinduism are polytheist). But how
many ancients explicitly held this belief? Evidently, monotheism “presupposes a rather elevated con-
ception of God,” such that “if one believed in this sort of god, there would be no point in believing
in any other gods besides this god” (Frede 2010, 58–59). Why was this conception dominant by
the time of the Enlightenment? Its development helps us understand the term.
The two Hebrew words for “God,” ’el and ’elohim, each signify: (1) god or gods (including ’elo-
him as a true plural), but are also used of human leaders and angels; (2) an image of these gods, or
an idol; (3) “God” (whether with or without the definite article), including as used by YHWH of
himself (Koehler, Baumgartner, and Stamm 1994, 49, 53). Both terms can also be used as a form
of address, by themselves or in compounds: “O God, …” (not symbolized in post-Fregean seman-
tics) (Vorländer 1975). By contrast, “YHWH” is always taken to be a proper name for God (usually
replaced in English translations, on the model of the Septuagint’s kyrios, by “LORD” all caps). As
God’s proper name, its pronunciation was prohibited, and its vocalization was subsequently forgot-
ten. Nevertheless, the name had a signification for its original audience (Toorn 1999, 611). God
himself, when he tells Moses the name by which his prophecy may be authenticated (Exodus
3:12–15), connects it to “I will be ‘I will be’” or “I am ‘I am’” (’ehyeh ’asher ’ehyeh), or simply to
“I am” (the translation is disputed, but the Septuagint gives ho on, the one who is; see Cronin
2018). The most likely interpretation of “YHWH” remains that it is a verbal phrase used as a
proper name: “he who is” or “he who causes being” (Toorn, Becking, and van der Horst 1999,
911–914). There is evidence that YHWH was a theonym for peoples prior to Israel, but Exodus
6:2–6 suggests that Israel arrives at a new understanding of this name—a fact that apparently holds
true for every previously used Hebrew theonym (Vacant, Mangenot, and Amann 1939, 956–959;
Jenni and Westermann 1997, 184). Thus, even the proper name of God in the scriptures has a con-
notation. In contrast, ’el and ’elohim, like theos and deus, are considered kind names that can also be
used as names of address, and so can be confused with proper names.
Under the influence of Jewish and Christian thought, the term “God” undergoes a transition in
the Greek and Latin languages, especially during the late Roman Empire, so that the term is increas-
ingly used in the capital “G” sense, whether or not it is actually capitalized (Lampe 1961, 632–635).
In this context, “gods” can still designate deities such as Hermes and Apollo, as it does in the
Christian scriptures. It has been argued that even educated pagans during this period were increas-
ingly “monotheist” in their affirming a single first principle that governs the world, even if not in
their use of theos (Frede 1999, 43–50). Aristotle himself awarded one particular divine being a status
unique enough to be called “the God.” Indeed, most of the Greek philosophers affirm one divine,
first principle of the universe. Plato, Aristotle, and Plotinus offer arguments for the existence of a
first principle and discuss how attributes such as knowledge belong to it. One can see the blending
of such philosophizing with Egyptian and Hebrew thought, for example, in the pagan thinker Plu-
tarch (Brenk 2014). Such practices lead to reflection on the meaning of God in the capital “G”
sense that is uncharacteristic of the Hebrew scriptures. In short, one cannot account for the transfor-
mation of the term in a Judeo-Christian context without taking into account the influence of
philosophical thinking on the likes of Aristobulos and Philo, as well as later on Origen, Augustine,
Philoponus, and others.
Definition: Theism
Philological evidence, then, suggests that “god” and “God” name a kind of thing, like “dog” or
“helium.” The capitalized term “God” in the Romance languages expresses a “conceptual content,”
serving as a kind name for something that, as a kind, is not plural in number. Consequently, it is
also used as a singular noun and as a form of address, just as is “Thomas Jefferson.” Something sim-
ilar occurs with “Sun.” We now recognize that there are many “suns”; yet, we give the kind name a
capital in one special case. We can also use “Sun,” like “Love,” as a form of address (Clarke 1961,
234–235). So, some kind names can also be used as singular nouns. We might also give the Sun
a proper name: “Kaga.” Similarly, “God” names a kind of thing that, it will be argued, cannot as
such be pluralized, unlike “Sun.” The point is well expressed by Mill in A System of Logic:
But there is another kind of names, which, although they are individual names, that is,
predicable only of one object, are really connotative. For, though we may give to an individual
a name utterly unmeaning, which we call a proper name, a word which answers the purpose of
showing what thing it is we are talking about, but not of telling anything about it; yet a name
peculiar to an individual is not necessarily of this description. It may be significant of some
attribute, or some union of attributes, which, being possessed by no object but one,
determines the name exclusively to that individual. “The sun” is a name of this description;
“God,” when used by a monotheist, is another. These, however, are scarcely examples of what
we are now attempting to illustrate, being, in strictness of language, general, not individual
names: for, however they may be in fact predicable only of one object, there is nothing in the
meaning of the words themselves which implies this: and, accordingly, when we are imagining
and not affirming, we may speak of many suns; and the majority of mankind have believed,
and still believe, that there are many gods. (Mill 1974 [1843], 33)
Gottlob Frege himself routinely analyzes “God” as a first order “concept term,” but in his Letter to
Liebmann (25 August 1900), he acknowledges the same situation: “As regards language, proper
names correspond to objects, Concept-words (nomina appellationis) to Concepts. Nevertheless, the
sharpness of this distinction is somewhat blurred in language, given that what originally were proper
names (e.g., ‘Moon’) can become Concept-words, and what originally were Concept-words (e.g.,
‘God’) can become proper names” (Frege 1980, 28).
Granted that “God” is a kind name of this sort, nonetheless style guides list many capitalized
theonyms that can also be terms of address: to take a prominent example, the Chicago Manual of Style
lists Adonai, the Almighty, the Lord, Providence, the Supreme Being, and so on. What do “God” or
“the Deity” name as distinct from these? Philosophers have addressed this question in two important
contexts: (1) proofs for the existence of God; and (2) discussions of the nature of God.
“GOD” AS USED IN PROOFS OF GOD’S EXISTENCE
As we have seen, contemporary philosophers of religion frequently resort to “proper name theory”
to explain the term “God.” This strategy has implications for mounting proofs of God’s existence.
A major problem has arisen because of verificationist theories of meaning. Is it even meaningful
to refer to an individual who is in no way given empirically? (Crombie 1957, 39–43; Shepherd
1974). Alternatively, suppose one grants that “the one and only God of worship” could have a ref-
erent. Presumably an argument for God’s existence has a stage in which what is concluded through
a series of premises is identified as “God.” This is the “identification stage” (O’Connor 2004). Yet
how could one know that what is reached by the premises is numerically the same as the individual
named “God”? This can be called “the problem of the gap.” Indeed, few arguments for God’s exis-
tence conclude to the existence of an individual as individual. The class of arguments that are excep-
tions to this general claim have come to be regarded as the preferred approach, namely, arguments
from religious experience of the individual named “God.” Still, how could one know that the indi-
vidual whom one has experienced is the same as the God of worship, let alone as “the one and only
object of worship of Jews, Christians, and Moslems?” (Harris 1991, 83–85) If there is no shared
conception of God even among religious communities, proof or disproof of God’s existence can
come to appear meaningless (Gastwirth 1972, 151–152). A standard response is that no one knows
what God is. But if not, must we not affirm atheism or agnosticism? (Englebretsen 1974). For

Definition: Theism
similar reasons, establishing God’s existence at the ground level of the philosophy of religion has
fallen into disfavor, apart from perennial interest in ontological arguments that use only a priori
concepts.
One reaction has been to appeal to the classical theist kind-name theory of “God,” and thus to
target a set of divine properties from the start of a proof. After all, as has been well said by Graham
Oppy (2014, 92), “argument about the existence of God depends upon prior agreement about what
God would be if God were to exist.” Many such sets of properties have been proposed as objects of
proof, sometimes with extensive justification: “a personal cause of continuance in existence” (Braine
1988, 168–173, 266); “provident, creator of the universe” (Steenberghen 1961, 27–44); “subsistent
Existence, unique, and the cause of the universe” (Miller 1992, 137, 153); “first efficient, exemplar
and final cause that is infinite and unique” (Duns Scotus 1974, 32–130). Still, where the choice of
properties is not arbitrary, it can seem ad hoc, that is, tailored to suit the author’s preferred reason-
ing, no matter how complicated that reasoning needs to be in order to attain its target.
Another approach, the minimal kind-name theory of “God,” sets no single preordained target
and thereby embraces at once a plurality of possible arguments. Motivating this approach is the idea
that at ground zero in philosophical theology, an argument for God’s existence seeks merely to “get
the ball rolling” (Davies 1993, 26). An argument need not conclude to a being with all of the prop-
erties of classical theism. Instead, a minimalist strategy merely seeks to establish as existing some-
thing whose nature will subsequently be explored. The term “God” names that nature with the help
of whatever properties can serve to distinguish that nature from all else.
Consider, as a limited analogy, the case of nonempirical explanatory entities like black holes. We do
not know the nature of black holes as such, and in fact we may doubt that we have penetrating vision into
the transparent essence of anything. Nevertheless, we can at least work through the “all-and-only proper-
ties” that distinguish something—black holes, god—from other kinds (Twetten 2017, 67, 76–78).
Instead of such empirically based properties, minimal kind-name theory uses, in the case of “God,” rela-
tion to or negation of all other things as the distinguishing properties of that kind. Or, better, relation to
and negation of empirical traits serve as the empirically based properties in the case of this nonempirical
nature. “God” names the nature of something beyond or more excellent than all others, or that is differ-
ent from all others to which it is causally related (cf. Kenny 2004, 12–13).
One can see immediately that such a concept picks out one hypothetical kind from every other
kind. For that reason, it makes sense that we use a capital “G” rather than a lowercase “g” for the
term. Aristotle and Isaiah each sought out what is beyond all other things while being causally
related to them, even though they did not identify this clearly with an exclusive kind name. The
Romance languages, inheriting orthographic developments in late Latin, have assigned to this kind
the capital “G” term “God” (some philosophers use instead the lowercase “god” to refer to this min-
imal kind, distinguishing this use of “a god” from the weak sense of “god” noted above, namely, as
merely a superhuman or immortal being).
One can see why, in an existential proof, the minimal kind-name theory of “God” at once sets
a well-defined target, yet its minimalist character makes it minimally restrictive, and so applicable
to many argumentative approaches. Assume the categorization of proofs of God into ontological
(“perfect being” theologizing from nonempirical premises) and cosmological (arguments that
employ empirical premises and that generally argue from effects to causes) (Morris 1991,
278–280). Then, even independent of considerations of cogency, minimal kind-name theory is
applicable to both sorts of proof. Anselm’s characterization of God as “that than which nothing
greater can be thought” (Anselm of Canterbury 1946) names something beyond all other things,
as do such phrasings as “provident cause of all other things” and “a per se necessary being,” whereas
“qualitatively infinite simplicity” names something different from every other sort of thing. An argu-
ment that cogently concludes to anything that has these properties concludes to “God” under mini-
mal kind-name theory. By contrast, if “prime mover,” “designer of the universe,” or “source of
moral obligation” can name something that is less than the First Being, such as an angel or a god,
then a proof that concludes to such a thing has not thereby arrived at “God” with a capital “G.”
A complete defense of an argument for God should include indications why what is concluded
cannot be a subordinate being.
Definition: Theism
Notice that on the minimal kind-name theory of “God,” no one attribute is prescribed for a
proof, including “creator” or “only one such.” Such attributes may be entailed but need not be
expressly proved at the initial stage, as long as exclusive “beyondness” or “otherness” is proved.
Oppy captures the strategy as follows: “On a natural ‘minimal’ conception, God is the source, or
ground, or originating cause of everything that can have a source, or ground, or originating cause”
(Oppy 2013, 5); and “this claim,” he says, “is all that Minimal Theism is committed to” (2013,
19). Strictly speaking, on the minimal kind-name theory here sketched, a causal relation is also
not required (and it is in fact absent in mere “perfect being” arguments). The quotations from
Oppy simply reflect the fact that, in many circles, only cosmological arguments are considered to
be genuine contenders for proofs of God’s existence. God, it is often thought, must be arrived at
only as “the ultimate explainer of all explained existence” (Leftow 2012, 5–6).
An advantage of the minimal kind-name theory of “God” is that it is grounded in the method-
ology of existential reasoning found in classical logic. An extensive investigation of what is a black
hole is pointless unless one is sure that a black hole exists. In fact, the proof that it exists is the best
and only ground for understanding what it is. But it is not possible to establish that a black hole
exists unless one has a working understanding of what we mean when we speak of it, that is, a defi-
nition that singles out, with all-and-only characteristics, that putative kind, black hole, from all
other kinds. All definitions of God, like all definitions of a black hole, are only nominal definitions.
As nonempirical explanatory entities, they provide us no access to anything that we can claim to be
a real definition, a formula for what the thing is in itself, a formula that states its genuine essence in
contrast to its properties or the necessary and sufficient conditions for applying the corresponding
term correctly. Real definitions typically put something in a class that includes other things (e.g.,
in the class of plane figures) and identify the differentia, a property that is primary and coextensive,
and that therefore distinguishes the definiendum from all else in the class (e.g., three-sided). The
notion of God is a special case of a putative kind claimed to be unlike every other kind. We cannot
put what is beyond all other things into a category shared with the rest. Nevertheless, we can use
formulas, nominal definitions, that pick God out as distinct from all others.
This methodology for existential proof through nominal definition was part of an academic
training that became standardized around the second century CE (Ebbesen 1981, 52–62) and that
influenced Gottfried Wilhelm Leibniz and John Locke. Aristotle asks the question, “Does the god
exist?” (Posterior Analytics 2.1), and he subsequently lays out canons for how to address such ques-
tions (the expression “the god” will subsequently be used in contexts where philosophers may be
thinking of “a god,” “a God” or both, and it is unclear which). Thinkers of the Hellenistic period
through Philo, even when they denied or argued against the existence of “the god,” distinguished
the discussion of two questions: first, whether the god exists, and second, what the god is (Runia
2002, 281–282, 299–303). In the fourth century CE, Themistius, applying Aristotle’s canons,
observes that the proof of the god’s existence is through nominal, not real, definitions: through
“something of” the god, that is, through effects such as healing, fulfilled prophecy, or everlasting
motion (Themistius 1900, 43–44, 48–51). Rarely did thinkers in a monotheist context draw
explicit attention to the relation between the nominal definitions used in the identity stage of argu-
ments for God’s existence and the methodology that lay behind existential proof (Twetten 2007,
54–64). Even more rarely did they formulate a minimal kind-name definition grounded in this
methodology that could serve as a target for many possible arguments. Aquinas’s version is based
on Pseudo-Dionysius’s supposed threefold way of knowing God: through causality, negation, and
eminence (Twetten 2005). He gives this formula for “God”: “something existing above all things
that is the principle of all things and is removed from all things” (Summa theologiae I.13.8 ad 2).
Something similar has been expressed in a Wittgensteinian idiom: “God is not a god”
but rather a “supreme being,” something “wholly other,” not in any genus and species, yet that is
a “formal concept” with its own “internal qualities” or “formal properties” and “grammar” (Durrant
1973b). We might express the minimal kind name “God” as follows:
Godm df the nature G such that for any nature F, if G is other than F, G is superior to F
Nevertheless, it would be contrary to the spirit of the minimal kind-name theory of “God” to take a
given formula as by itself rigorously prescribing the target of a proof for God. Instead, the

Definition: Theism
aforementioned formulas are suggestive of the minimalist capital “G” target of all of the classical
arguments for God’s existence (independent of the question whether these arguments would satisfy
us today). Minimal kind-name theory is an umbrella under which one can locate nearly every proof
strategy prior to the twentieth century, and many such subsequent strategies.
In the process of guaranteeing a univocity adequate to mathematical reasoning, Fregean seman-
tics, taken by itself, erases the distinction between real and nominal definitions, and it forces us to
treat “God” either as a predicate “Concept” or as a conceptually contentless particular “object.” It
appears intuitively false to treat “God” as a Concept-term or kind name and to say that “x gods,”
as is required on a post-Fregean analysis (Smart and Haldane 2002, 33–34) even though to say that
“x frogs,” though correct, sounds equally false. As a result, “God” usually comes to be thought of
not as a Concept-term or kind name but as the name for an individual, a proper noun or a disguised
definite description. Thus, we struggle to make sense out of what an existential proof would amount
to. Most philosophers would not agree with Willard Van Orman Quine (1953, 8–12) in saying that
“Socrates exists” should be formulated (9x)(Sx). But the alternative logical translation, (9x)(xgI)—g
stands for “God,” xyI for the identity relation—expresses the identity stage only of an argument that
concludes to a fact about an individual as individual, such as occurs in an argument for God from
religious experience. It becomes difficult to see the point of a proof of God’s existence when it is
construed as a proof of an individual’s existence. Does one use arguments to become acquainted
with an individual? Either that individual exists or it doesn’t, and experience alone can tell us
which. The project of a proof of God’s existence thus ironically comes to appear meaningless to
contemporary philosophers of religion. The suggestion here is that they need not resort to classical
theist kind-name theory alone in order to make sense out of the apparently absurd.
There is a further advantage of the minimal kind-name theory of “God” for theism, in which
“God” names the nature of a thing that is beyond all others, and “god” names the nature of a thing
that is not beyond all others. A nominal definition, given its all-and-only properties and their entail-
ment relations, does permit one to understand how the sense of the term “theism” can be systemat-
ically, and not merely arbitrarily, narrowed, and does allow certain positions to be excluded from
“theism” in the narrow sense. Suppose, for example, one says that God is beyond all things and that
the universe is God. There are interpretations of this position that are consistent with theism, such
as Parmenides’s view that the universe is, strictly speaking, not real but merely a product of poetic
license. But any interpretation of pantheism that allows for something in some sense other than
God would be ruled out from capital “G” theism. One might say the same for polytheism: it is
not a theism in the narrow sense if it excludes “God” as beyond all other gods. Similarly, if a deistic
God is an unintelligent cosmic force, is it truly superior to human scientists? At the same time,
“monotheism” might also be systematically widened. If Hinduism, for example, affirms Brahman
as the supreme principle over all other things, including other gods, it is thus far monotheist. Simi-
larly, minimal kind-name theory provides a basis for further discussion of what God is, a discussion
that becomes engaging if minimal theism has established a sense of “God” under which it is reason-
able to affirm that God exists.
“GOD” AS FOUND IN THE PROJECT OF EXPLAINING GOD’S NATURE
The minimal kind-name theory of “God” offers certain advantages when it comes to understanding
the role of the term “God” in proofs for the existence of God. By contrast, the classical theist kind-
name theory, in which “God” designates a kind possessing a certain set of properties (such as “omni-
properties”), is important to invoke in reflecting on what philosophers say about what God is.
Throughout this latter inquiry, “God” is the name given to the nature that is being progressively
explored, a nature shown by argumentation to possess properties such as uniqueness, omnibenevo-
lence, omniscience, and omnipotence. In what follows, the nature designated by “God” in this sense
will also be referred to simply as the “divine nature” (lower case).
Arguments that the divine nature possesses the properties of classical theism are ultimately
founded on the contents of whatever nominal definition of the capital “G” sense of “God” was
Definition: Theism
operative in a prior proof of God’s existence. In this way, Anselm’s argumentation concerning the
divine attributes is grounded in his understanding of God as “that than which none greater can
be thought,” and Aquinas’s argumentation is grounded especially in his understanding of God as
the supereminent causal source of being. Both Anselm and Aquinas, to take examples from “perfect
being theology” and “creation theology,” respectively, put forward an account of the divine nature
that falls under the label classical theism, but they arrive at their conclusions about the properties
in question by different paths.
After Anselm argues in the Proslogion that “that than which none greater can be thought”
must exist, he proceeds to argue that God, understood thus, must be uncaused and must be the
causal source of all other things. For, “whatever is not this is less than [that than which none
greater] can be thought” (Proslogion 5). That God must be the first cause of all other things is
for Anselm a further conclusion beyond the proof for God’s existence. This argument for God
as the first cause also exemplifies Anselm’s procedure throughout the Proslogion, in which he
argues for the attribution of various properties to God insofar as it is absolutely better to have that
property than not to have it. In this way Anselm proceeds to argue that God must have knowledge
(albeit not in the way that animals with sensation do), that he must be omnipotent, that he must
be both merciful and impassible, that he is living, wise, good, happy, eternal, “and whatever it is
better to be than not to be” (Proslogion 11). Having defended each of these attributions, Anselm
proceeds to explain that God must be beyond the human power for knowing, because to be
“greater than can be thought” is in fact also true of “that than which nothing greater can be
thought” (Proslogion 15). Finally, after further reflections on the ways in which God is beyond
human knowledge, Anselm attributes absolute unity and simplicity to God, arguing that God is
without parts or composition, because plurality and conceptual divisibility are foreign to that than
which none greater can be thought (Proslogion 18, 21). Along the way in this accounting of divine
attributes, Anselm has concluded that God is unique and possesses the omni-properties of classical
theism, but he has grounded them all in his initial definition of God as that than which none
greater can be thought.
Aquinas, in his Summa theologiae, takes as his primary point of departure for exploring the
divine nature a notion common to several of the nominal definitions of God operative in his
five ways, namely, that God is the First Being, the preeminent, uncaused causal source of being.
Actually, we find something very similar in the approaches of the great Muslim philosophers Abū
Nasr al-Fārābī and Avicenna, and many features of Aquinas’s order of procedure in the first
questions of the Summa theologiae reflect Avicenna’s influence (Houser 2011, 368–369). From the
identification of God as the First Being, Aquinas proceeds to argue that God is pure actuality,
and consequently that he is absolutely simple, eliminating every mode of composition found within
caused substances. A compound being would have one part actualizing another in some way. The
First Being must be purely actual, as Aristotle had already held, because in anything that is not pure
actuality, whether it has unfulfilled potentiality or is fully actualized potentiality, there must be a
prior cause of the actualization of its potentiality.
From God as First Being and from His absolute simplicity, Aquinas argues that God is perfect,
in two senses. First, he argues that God is intrinsically perfect, lacking nothing of what befits him
according to his nature. This follows, he argues, from God’s being a first efficient cause that is
supremely and purely actual: there is no possibility that God should lack anything, if he is pure
actuality. Second, Aquinas argues that God is universally perfect, possessing every unalloyed excel-
lence found in his effects in a more eminent mode. He argues for this conclusion from God’s being
the first efficient cause of all things, the source of every excellence or perfection found in creatures,
claiming that the first cause must in some way possess the excellences it causes, albeit in a more
eminent mode. The characterization of God as universally perfect serves as an important turning
point in Aquinas’s argument toward the attribution of perfections such as knowledge and will to
God, although he is less dependent on deductive arguments from God’s universal perfection than
is sometimes supposed (cf. Kretzmann 1997, 141–142).
Comparing Anselm and Aquinas allows us to contrast perfect being theology with effect-to-
cause reasoning. Anselm is explicit in his Proslogion about attempting to ground everything said

Definition: Theism
about the divine nature in the notion of God that is operative in his proof for God’s existence: the
argumentation concerning the divine nature is thus radically unified in Anselm’s presentation.
Anselm is also content to speak of “that than which nothing greater can be thought” as a knowledge
of what God is, even if he also acknowledges that, in the end, this God is “greater than can be
thought.” Aquinas, by contrast, prefaces his discussion of the divine nature by denying that the
human mind can, in this life, arrive at a knowledge of what God is in himself (Summa theologiae
I.3 prologue). This is why Aquinas begins his account of the divine attributes with simplicity rather
than with perfection. It is necessary to begin by saying “how God is not” so as to distinguish the
divine nature from any creature’s nature. Finally, although Aquinas privileges the nominal under-
standing of God as First Being at the beginning of his derivation of the divine attributes, he exhibits
less concern for radically unifying his account of the divine nature in any one nominal definition of
God. We find some fifteen different arguments for God’s existence in Aquinas’s works (Baisnée
1952). At the same time, simplicity and perfection taken together form, as it were, a prism through
which the other properties in Aquinas’s account emerge (Velde 2006, 78–83).
The distinction between the minimal kind-name theory of “God” operative in classical proofs
for God’s existence and the more robust kind-name theory at issue in discussion of the divine nature
helps us distinguish some of the varieties of theism. Two interlocutors might agree that God exists,
where “God” is understood by a minimal kind-nominal definition, but then proceed to disagree
about the properties that characterize the divine nature. It is at this level that deism, process theism,
open theism, and so on can be contrasted with standard classical theism: these are various positions
concerning the divine nature. Two interlocutors who both accept the existence of God, understood
in terms of a minimal kind name, can each nevertheless truthfully say that they do not accept the
existence of their interlocutor’s God, understood in terms of a more robust kind name. Neverthe-
less, the minimal kind-name theory provides resources for their seeing their agreement and poten-
tially resolving their points of disagreement.
Whether theism’s “God” is taken in terms of proper name theory or some variety of kind-name
theory, to be a theist is to assert the existence of something. How one understands theism, along
with its varieties, must then be in part a function of how one understands this initial claim. Further-
more, insofar as theism in a robust sense is the assertion of the existence of a being with the divine
nature, elaborated beyond the contents of a minimal kind, theism involves the attribution of various
properties or perfections to God. One’s understanding of theism in this sense will be a function of
how one understands the force of these various attributions, of this “God talk.”
So far, the discussion of the distinction between proper-name and kind-name theories of the
meaning of “God” has presupposed some sort of continuity with the way that we think and speak
about the existence and properties of ordinary physical objects. But God, as classically conceived,
is far from any ordinary physical object. Philosophers influenced by the Abrahamic traditions of
Judaism, Christianity, and Islam, including “pagan” thinkers, have often asserted strict limits on
the extent to which truthful speech about God expresses any positive understanding of what God
is. Consequently, they have also set limits on the extent to which the meaning of truthful speech
about God shares anything in common with the meaning of truthful speech concerning caused
things.
We can frame questions about the meaningfulness and the limits of speech about God with
respect to the distinctions (1) between univocity and analogy and (2) between affirmative (kataphatic)
and negative (apophatic) theology.
UNIVOCITY, EQUIVOCATION, ANALOGY
The distinction between univocity and analogy rests on the more fundamental distinction between
univocity and equivocation, which are two different ways in which the same term can be predicated
of distinct subjects. One speaks univocally when one uses the same word with the same meaning or
conceptual content in speaking of distinct subjects. For example, Fido is a dog, and Rover is a dog.
Here “dog” is used with the same meaning, definition, or conceptual content. The indication that
Definition: Theism
there is univocity is that one will give precisely the same answer in each case to the question, “What
do you mean by saying that x is a dog?” However well or poorly one understands what a dog is—
whether or not one’s concept of a dog rises to the level of a real definition, as this notion was dis-
tinguished above—still one will explain one’s meaning in the same way for both Fido and Rover.
Equivocation occurs when one uses the same word with diverse meanings, definitions, or con-
ceptual contents. For example, First National is a bank, and the boundary of the Mississippi River is
a bank. Here the word “bank” is predicated of these distinct subjects with diverse meanings. The
indication that one has spoken equivocally is that one will give a different answer in each case to
the question, “What do you mean by saying that x is a bank?” Equivocation, broadly, obtains when-
ever there is any difference in meaning or conceptual content among several uses of the same word.
Again, however well or poorly one understands the meaning of “bank” in a given case, one speaks
equivocally when one would express one’s meaning differently in distinct cases.
Not all cases of equivocation, however, involve radically diverse, entirely unrelated meanings.
Aristotle distinguishes what he calls equivocation pros hen (with respect to one), which scholars
now often call associated meaning, focal meaning, or referential multivocity. Aristotle famously
offers as an example of pros hen equivocation the term “healthy” as predicated of an animal, of food,
and of urine. When we say that food is healthy or that urine is healthy, we would explain our mean-
ing by making reference to the meaning of “healthy” as it is predicated of an animal: food is healthy
insofar as it causes the health of an animal, and urine is healthy when it is a sign of the health of an
animal. Again, however well or poorly we understand what health is in an animal, still this is the
primary meaning of the term “healthy.” It is important to note that the example of health does
not involve any formal likeness between the “healthiness” of food or of urine and the health of
the animal: food and urine are called healthy not because they are similar to the health of the ani-
mal, but only insofar as they are respectively a cause and a sign of it.
Aristotle also employs the notion of analogy, by which he means a proportional likeness
between two things. The notion of analogy has its origin in mathematics, in the four-term compar-
isons made between numbers or figures: x:y :: a:b, such that x is like a in that it stands to y as a does
to b. Sometimes analogies are employed simply to observe likenesses in nature, as when Aristotle
compares the feathers of birds to the scales of fish. But one can employ a four-term analogy in order
to explain the meaning of a term’s extended usage. For example, if one says that the mind “sees,”
this means that the mind is related as a cognitive power to its object in a manner similar to how
the visual power is related to its object. The use of proportional likeness as an explanation of mean-
ing, however, may amount in a given case—or, some would hold, in every case—only to metaphor.
By the time thirteenth-century Latin authors received the notion of analogy, its meaning had
come to include what Aristotle had called pros hen equivocation, in addition to cases of four-term
proportional likeness (Hochschild 2010, 1–10). Aquinas routinely offers the case of “healthy” as
an instance of analogous naming, and he also characterizes analogous naming as a mean between
univocity and pure equivocation. Aquinas does successfully distinguish between analogy as propor-
tional likeness (its original meaning in Aristotle) and analogy as associated meaning. Nevertheless,
he varies over the course of his career in his choice of which sort of analogy he employs in explain-
ing the meaning of “God talk” (Montagnes 1963; Pini 2011, 499–501).
Various philosophers have accounted for the commonality between “God talk” and “creature
talk” in terms of univocity, equivocation, or analogy. John Duns Scotus is the most famous histori-
cal advocate for univocity in the attribution of being and other properties to God, although his posi-
tion rests on a distinctive account of what univocity means, and he also incorporates analogy into
his account (Dumont 1998). Anselm has been interpreted as presupposing univocity in some ele-
ments of his theological discourse (Visser and Williams 2009, 114–117). “Theistic personalism” is
a label used of contemporary philosophers within classical theism who take seriously the univocity
of the personal attributes of God, given the univocity employed in post-Fregean semantics. God is
the same kind of person as any other person, with the same kind of mental states that any person
has. Adherents to such univocity in God talk are not as concerned with divine transcendence,
immutability, and perfection, and they often give up on simplicity (Plantinga 1980), timelessness
(Swinburne 2016, 230–244), or both (Craig 2001, 240–241). Maimonides is the most famous

Definition: Theism
advocate for an account that classifies the meaning of affirmative God talk as purely equivocal
(Guide for the Perplexed 1.52). He holds that positive predications common to God and creatures
amount only to negations and metaphorical comparisons, and they affirm nothing truly common
to God and creatures. For example, to say that God is wise is to say that God is not foolish (Guide
for the Perplexed 1.58); to say that God is merciful is to say only that he produces effects comparable
to those produced by merciful human beings (Guide for the Perplexed 1.54). Finally, Aquinas is
prominent among the advocates for the analogical character of positive speech concerning God,
standing within a lengthy tradition that draws on middle Platonists, Christian Neoplatonists such
as Pseudo-Dionysius, and Arabic philosophers in their accounts of knowing and naming the first
principle (Rocca 2004, 89–90).
If the meaning of God talk—of the attribution of being and of other properties or perfections
to God—is anything other than strictly univocal, then one’s understanding of theism will be signif-
icantly affected. For, any case of pros hen equivocal or analogous naming, in which one term is pred-
icated of more than one subject without univocity, is logically consistent with the denial of that
predicate of each subject according to the meaning proper to another subject. The boundary of a
river is not a bank (understood as a place for depositing money). Urine is not healthy (understood
in the sense in which the animal is healthy). And if goodness and wisdom are analogically common
to God and creatures, however one might understand or explain these analogies, then there is also a
sense in which God does not have goodness and is not wise (understood in the senses proper to cre-
ated goodness and to human wisdom). It is for this reason that the assertion of analogy in speech
about God goes hand in hand with some degree of apophaticism.
THEISM IN LIGHT OF APOPHATIC AND KATAPHATIC THEOLOGY
The distinction between what we affirm or positively say concerning God and what we deny or say
negatively concerning God is of fundamental importance. Apophatic or negative theology empha-
sizes the denial of whatever is unfittingly attributed to God, whereas kataphatic or affirmative theol-
ogy attends to what can be affirmed of God, whether symbolically/metaphorically or analogically/
properly.
PSEUDO-DIONYSIUS: A RADICAL APOPHATICISM

There are a number of varieties of negative/apophatic theology. One can see the range of apophatic
theology by considering one of its most famous proponents, Pseudo-Dionysius, a late-fifth-century
Christian student of Proclus (Perczel 2000). Within Dionysius’s writings, two sorts of negative the-
ology may be broadly distinguished (Jones 1980, 15–26, 89–103; Rocca 2004, 21). First, there is a
radical sort of apophatic theology, such as the one articulated and explored in the Mystical Theology.
As presented in its opening chapters, the work of negative theology is to purify one’s mind of every-
thing that God is not, so as to prepare for a union with God initiated by God himself. At the outset,
Dionysius indicates that the purpose of the work is to function as an exercise that helps one to rise
toward a union with the One that transcends all being and knowledge.
Dionysius thus explains that a negative theology ordered toward mystical contemplation pro-
ceeds by denying about God everything that he is not, following a certain order, proceeding from
the lowest and most imperfect among sensible things—the things farthest from God, as it were—
all the way to the denial of even the highest and most perfect things intellectually known to
us among creatures. In the actual execution of this negative theology in the later chapters of the
Mystical Theology, Dionysius begins by denying that God is a body, that he has any accidents, that
he is sensible, that he is subject to any change, passion, corruption, or defect. He then proceeds to
deny that God is soul or intellect, that he is conceivable, that he is great or small, equal or unequal,
like or unlike, moving or at rest, powerful or impotent. He denies that he is light, that he is alive or
life, that he is essence, or eternity or time, that he is subject to contact by intellect, that he is knowl-
edge or truth or wisdom, that he is one or unity, that he is divinity, that he is goodness, that he is
anything known to us or any other being among the things that are or are not, that anything that is
Definition: Theism
knows him as he is, that reason can attain to him, or name him, or know him, that he is darkness or
light, that he is false or true, or that any affirmation or negation can be applied to him.
The ultimate conclusion of this “ascent” through negations is a recognition that God is beyond
all affirmations and negations, because God, as the first cause of all things, is free of all limitation.
Again, Dionysius presents this radically negative theology as a sort of spiritual exercise ordered
toward spiritual union with God.
We find a second, attenuated sort of negative theology within the context of the affirmative
theology of Dionysius’s On the Divine Names. This work is concerned with what Dionysius calls
the intelligible perfections flowing from God to creatures, such as goodness, being, wisdom, and
power. Dionysius insists that although it is possible to affirm truly such positive names of God,
nevertheless these affirmations are “unfit to the mystery,” while negations concerning God are true
in a more unqualified way (Celestial Hierarchy 2.3). Every affirmation of an intelligible perfection
concerning God is made insofar as God is the supereminent cause of that perfection, such that
God as cause is understood to be beyond the perfection he causes, while the created way of posses-
sing that perfection is denied of God. There is thus no contradiction between the attenuated nega-
tive theology of the Divine Names and the more radical negative theology of the Mystical Theology:
the fact that no created intelligible perfection belongs to God in the same mode as it belongs to
creatures is why the radically negative path of the Mystical Theology is possible. Indeed, the Mystical
Theology both begins and ends with the characterization of God as the all-perfect cause above all
creatures. Even the most radical negative theology, it has been observed, is ultimately grounded
on the theological affirmation of God as the cause of all creatures (Johnson 1998). Furthermore,
as Dionysius insists, radical negative theology denies created perfections of God not because he is
deficient or less than creatures, but because he exceeds them.
Now that we have distinguished in Dionysius a radical apophaticism from a somewhat attenu-
ated apophaticism functioning in conjunction with his kataphatic theology, we can also distinguish
kataphatic theology that is unrestrained and unqualifiedly univocal from kataphatic theology that is
restrained by attenuated apophaticism. In recent philosophy of religion, theistic personalism often
advances an unrestrained univocal kataphaticism. Arguably the two greatest exponents of a katapha-
ticism restrained by an attenuated apophaticism are Aquinas and Duns Scotus. Comparing them
can give us a sense of high points in the classical discussion.
RESTRAINED KATAPHATICISM: AQUINAS AND DUNS SCOTUS

Duns Scotus insists that the most general concepts, such as “a being” and “wise,” must be univocal
as predicated of both God and creatures if there is any sense in which we can know that God has
these attributes (Duns Scotus 1954, 18–27). Whereas Duns Scotus attenuates this strong kataphati-
cism with finite versus infinite modality, Aquinas sets out a weaker kataphaticism of analogous nam-
ing. Nevertheless, Aquinas agrees with Duns Scotus that there is a commonality involved when we
predicate “wise,” for example, both of God and of creatures (Gilson 1951, 88). That commonality,
however, is not at the level of a predicate or a concept but at a sublevel in the semantic story (McI-
nerny 1986). As indicated above, for Aquinas, there is analogy in the predication of a term wherever
one explains one’s meaning differently in different predications. If one adequately explains the
meaning of “wise” in “Socrates is wise” by offering a definition of wisdom articulated in terms of
genus and difference, this definition will fail to explain the meaning of “wise” in “God is wise.”
The definition of “wise” as predicated of Socrates is for Aquinas something like “an intellectual
habit (genus) according to which one understands the highest things (difference).” Because an intel-
lectual habit is a quality and an acquired capacity, Aquinas denies that the genus of human wisdom
can be attributed to God, who has no accidents and is pure actuality. When one says that God is
wise, then, one affirms this predicate only according to its differentia, not according to its genus.
It is the differentia—the perfection that is “understanding the highest things”—that is attributed
to God, and we understand the very perfection that is found as a quality in Socrates to be, in
God, the subsisting wisdom that is extramentally the same as the subsisting justice that is also
God. Aquinas holds that the perfection wisdom is in fact found in a primary way in God and only
secondarily in creatures. Consequently, the perfection signified by the term “wise” as it is in God

Definition: Theism
belongs to the definition of “wise” as found in creatures, without altering the common differentia
in itself (Summa theologiae I.13.6). Crucial to Aquinas’s account is his appeal to contemporary
logicians’ distinction between a perfection signified as it is in the mind and as it is found in things
(Ashworth 1991, 52–53).
Does this comparison to Duns Scotus amount to affirming a latent univocal core within Aqui-
nas’s account of the analogy of divine names? (Rocca 2004, 334) Again, for Aquinas the needed
commonality is below the level of terms and concepts. In brief, on Aquinas’s account there fails
to be univocity simply because, whereas we do adequately grasp human wisdom through a finished
concept using genus and differentia, we fail to grasp or “circumscribe” conceptually the divine wis-
dom. In addition to the fact that parts of the definitions of created perfections like wisdom cannot
be attributed to God, Aquinas expresses two grounds for denying all positive perfections of God: (1)
All names signify their conceptual content in either an abstract way, such as through “wisdom” or
“to be,” or in a concrete way, through “wise” or “a being.” Each of these terms, when the abstract
or concrete “mode of signifying” that is part of the term is considered, must be denied: God is
not abstract wisdom but something that has, without extramental distinction, wisdom; also, God
is not “a being” but subsistent “to be itself,” and so forth (McInerny 1978). (2) When one names
God “wise,” one does not intend to signify anything extramentally distinct in God, namely, “the
wisdom part,” as one does when one calls Socrates “wise.” Because God, with respect to the divine
nature, is absolutely simple, all of the perfections that belong to God are identical in extramental
reality. Given God’s maximal simplicity, says Augustine, to be and the act of wisdom are in this case
the same (De trinitate 7.1.2). Therefore, says Aquinas in Dionysian language, each positive name of
a divine perfection does not “circumscribe” that perfection as it is found in God, in whom the per-
fection signified by these terms is not ontologically distinct, as it is in Socrates (Dewan 1980, 24–
32). One cannot adequately grasp the perfections as they are found in God uncircumscribed, and
so one resorts, instead, to a multiplicity of positive names, intending each to be predicated in an
uncircumscribed way.
Hence, for Aquinas the project of knowing what God is becomes instead a project of naming him
kataphatically and apophatically, knowing that our positive names do signify a perfection in him,
without our completely comprehending that perfection as it is in him. Following Aristotle, Aquinas
thinks of perfect knowledge as involving a formal identity between the knower and the known. We
do not achieve such cognitive identity (in this life) with any of the divine perfections as they are found
in God, or with the divine nature with which these perfections are extramentally identical. In this
sense one knows only, at best, what God is not: we do not have cognitive identity with God’s nature
or perfections. Nevertheless, one can know that the perfections we do understand belong to him first,
and one names him accordingly, making a number of qualifications as one proceeds.
Given Henry of Ghent’s distinctive account of analogy, Duns Scotus is legitimately concerned
about our ability to defend affirmative statements about God (Dumont 1998). Accordingly, Duns
Scotus insists on a certain degree of cognitive identity with the divine perfections even as they are
found in God, although he will of course deny that one has natural cognitive identity in this life
with the divine essence. Would Duns Scotus say, then, that one completely grasps God as “a being”
or as “good?” Absolutely not, because these perfections belong to God in a mode that is infinite.
Thus, the mode of infinity plays a role in his thought similar to the role that the “uncircumscribed”
mode plays in Aquinas’s thought, thanks to divine simplicity. There is a problem, however. Thinkers
from Dionysius through Avicenna and Maimonides, to the extent that they are in the Platonic tradi-
tion, are inclined to emphasize apophaticism about divine attributes rather than kataphaticism. In the
Platonic tradition, form signifies a limit that is grasped by the human mind and that is therefore finite.
By contrast, the capital “G” God or the First must be beyond all limits, and therefore, apparently,
beyond all of our positive names. On the whole, scholastic thinkers stood apart from this tradition
by insisting (1) that form is not in itself limited; and (2) that the pure perfections can be really iden-
tical in their source although they are conceptually distinct from each other (and hence known and
named by us through a plurality of concepts and names).
The problem for Duns Scotus is that he appears not to be able to use the infinite mode of the
divine perfections so as to make the perfections extramentally identical in God. The “formal
Definition: Theism
distinction” of these perfections from each other in God, Duns Scotus must hold, is an extramental
distinction, unlike what has been previously thought (Dumont 1998). Perhaps the solution is that,
in the infinite mode, the perfections are merely “formally nonidentical.” Duns Scotus himself
certainly possessed the resources to answer this threat to divine simplicity, but at the very least there
appears to be significant tension between maintaining both univocity and simplicity. Duns Scotus
would respond, with good reason, that the defenders of analogous naming have not clearly articulated
where, semantically, there is sufficient commonality to justify the ascription of “wise” to both God
and humans. Alternatively, Duns Scotus has located God and creatures under overarching terms
“wise” and “being” that “transcend” their finite and infinite modes, a move systematically excluded
in Aquinas’s attenuated kataphaticism and account of “transcendental terms” (Carl 2018, 228–232).
What is the import of apophatic theology for the meaning of theism and of its varieties? In its
most radical form, apophatic theology denies that God is anything conceptually grasped in any way
by the human mind, and it affirms that the most authentic religious experience of union with God
is beyond any knowledge that can be articulated. Brian Davies has highlighted the extent to which
the appropriately attenuated form of kataphatic theology must affect one’s understanding of theism,
arguing that one who takes apophatic theology seriously can be truthfully described as an a-theist, as
one who would deny that God is “a being” (Davies 2007). The being of God and all attributions of
perfection to God are affirmed only in such a way that one recognizes that these perfections are also
denied of God in their created mode.
It is important to distinguish between those advocates for theism who understand the attribu-
tion of being and other perfections to God in unqualifiedly univocal terms and those who regard
such attributions as always falling short—infinitely short—of the reality of God. There may be as
much disagreement between theists about the meaning of theism as there is between atheists and
theists about the truth of theism.
In sum, “Theism” is an analogous term, that is, a term used in several different but related
senses. In the usual sense, “theism” affirms a God with a capital “G.” Minimally, “God” in this
sense refers to a supreme being that is beyond all other things, including other “gods.” Accordingly,
the capital “G” term is also used in the singular and as a term of address in the way that a
proper name is used. This use can, potentially, be further justified by arguments affirming that there
can be only one supreme being. Understanding “God” in this sense allows one to adjudicate dis-
putes within “theism” over divine properties, since some properties are inconsistent with “what is
beyond all other things.” By contrast, “theism” in the broad sense affirms “God” or “gods,” namely,
subordinate beings above mere humans. Under these notions of “God” and “god,” a “capital G”
monotheist might also affirm lowercase “gods.” To the extent that apophaticism and strategies to
countenance divine simplicity and infinity inform one’s account of “God” and of the properties
attributed to the divine nature, one’s understanding of theism as the affirmation that God exists
will be significantly altered, and the properties of classical theism can be taken seriously and given
individual justifications.
Bibliography
Alston, William P. “Referring to God.” International Journal for Philosophy of Attfield, Robin. “The Individuality of God.” Sophia 10, no. 1 (1971): 20–
Religion 24 (1988): 113–128. 27.
Alston, William P. Perceiving God. Ithaca, NY: Cornell University Press, 1991. Baisnée, Jules. “St. Thomas Aquinas’s Proofs of the Existence of God
Anselm of Canterbury. Proslogion. In Anselmi Opera Omnia, edited by Presented in Their Chronological Order.” In Philosophical Studies in
Franciscus Salesius Schmitt, vol. 1, 97–124. Edinburgh: Thomas Nelson Honor of the Very Reverend Ignatius Smith, O.P., edited by John K. Ryan,
and Sons, 1946. 29–64. Westminster, MD: Newman Press, 1952.
Aristotle. Analytica priora et posteriora. Edited by W. D. Ross and Lorenzo Braine, David. The Reality of Time and the Existence of God. Oxford:
Minio-Paluello. Oxford: Clarendon, 1964. Clarendon, 1988.
Ashworth, E. Jennifer. “Signification and Mode of Signification in Brenk, Frederick E. “Plutarch and ‘Pagan Monotheism.’” In Plutarch in the
Thirteenth Century Logic: A Preface to Aquinas on Analogy.” Medieval Religious and Philosophical Discourse of Late Antiquity, edited by L. Roig
Philosophy and Theology 1 (1991): 39–67. Lanzillotta and I. Muñoz Gallarte, 73–84. Leiden, Netherlands: Brill, 2012.

Definition: Theism
Carl, Brian T. “The Transcendentals and the Divine Names in Thomas Empire, edited by Stephen Mitchell and Peter van Nuffelen, 53–81.
Aquinas.” American Catholic Philosophical Quarterly 92, no. 2 (2018): Cambridge, UK: Cambridge University Press, 2010.
225–247.
Frege, Gottlob. Gottlob Freges Briefwechsel mit D. Hilbert, E. Husserl, B.
Clarke, W. Norris. “On Professors Ziff, Niebuhr, and Tillich.” In Religious Russell, sowie ausgewählte Einzelbriefe Freges, mit Einleitungen, Anmer-
Experience and Truth: A Symposium, edited by Sidney Hook, 224–230. kungen und Register herausgegeben von Gottfried Gabriel, Friedrich
New York: New York University Press, 1961. Kambartel, Christian Thiel. Hamburg: Meiner, 1980.
Craig, William Lane. Time and Eternity: Exploring God’s Relationship to Gastwirth, Paul. “Concepts of God.” Religious Studies 8, no. 2 (1972):
Time. Wheaton, IL: Crossway, 2001. 147–152.
Crombie, Ian M. “The Possibility of Theological Statements.” In Faith and Geach, Peter T. “Aquinas.” In Three Philosophers: Aristotle, Aquinas, Frege, by
Logic: Oxford Essays in Philosophical Theology, edited by Basil Mitchell, G. E. M. Anscombe and Peter T. Geach, 67–125. Oxford: Blackwell;
31–83. London: Allen and Unwin, 1957. Ithaca, NY: Cornell University Press, 1961.
Cronin, K. J. “The Name of God as Revealed in Exodus 3.14: An Geach, Peter T. God and the Soul. London: Routledge and Kegan Paul,
Explanation of Its Meaning.” June 20, 2018. http://www.exodus-314 1969.
.com. Gilson, Etienne. Jean Duns Scot: Introduction à ses positions fondamentales.
Cudworth, Ralph. The True Intellectual System of the Universe: Wherein All Paris: Vrin, 1952.
the Reason and Philosophy of Atheism Is Confuted, and Its Impossibility Harris, James F. “The Causal Theory of Reference and Religious Language.”
Demonstrated. Vol. 1. London: Richard Royston, 1678. International Journal for Philosophy of Religion 29, no. 2 (1991): 75–86.
Davies, Brian. The Thought of Thomas Aquinas. Oxford: Clarendon; New Hochschild, Joshua P. The Semantics of Analogy: Rereading Cajetan’s De nominum
York: Oxford University Press, 1993. analogía. Notre Dame, IN: University of Notre Dame Press, 2010.
Davies, Brian. “Aquinas Proves Atheists Are Closer to God Than They Houser, Rollen Edward. “Aristotle and Two Medieval Aristotelians on the
Think.” Times (London), 13 January 2007. Nature of God.” International Philosophical Quarterly 51, no. 3 (2011):
den Brouw, Rien op. “The Problem of the Missing Article in the Use of 355–375.
‘God.’” Religious Studies 30, no. 1 (1994): 17–27. Jenni, Ernst, and Claus Westermann, eds. Theological Lexicon of the Old
Descartes, René. Meditationes de prima philosophia. In Oeuvres de Descartes, Testament. Vol. 1. Translated by Mark E. Biddle. Peabody, MA:
edited by Charles Adam and Paul Tannery, vol. 7. Paris: Cerf, 1904. Hendrickson, 1997.
Dewan, Lawrence. “St. Thomas and the Divine Names.” Science et Esprit 32 Johnson, Mark. “Apophatic Theology’s Cataphatic Dependencies.” Thomist
(1980): 19–33. 62, no. 4 (1998): 519–531.
Dumont, Stephen D. “Henry of Ghent and Duns Scotus.” In Routledge Jones, John D. “Introduction.” In The Divine Names, and The Mystical
History of Philosophy. Vol. 3: Medieval Philosophy, edited by John Theology, by Pseudo-Dionysius the Areopagite, translated by John D.
Marenbon, 291–328. London: Routledge, 1998. Jones. Milwaukee, WI: Marquette University Press, 1980.
Duns Scotus, John. Opera omnia. Vol. 3: Ordinatio I, dist. 3, edited by Kenny, Anthony. “The Ineffable Godhead.” In Unknown God: Agnostic
Carolus Balić et alia. Vatican City: Typis Polyglottis Vaticanis, 1954. Essays, 7–24. London: Continuum, 2004.
Duns Scotus, John. Tractatus de primo principio—Abhandlung über das erste Klima, Gyula. “Ontological Alternatives vs. Alternative Semantics in
Prinzip. Edited and translated by Wolfgang Kluxen. Darmstadt, Germany: Medieval Philosophy.” S: European Journal for Semiotic Studies 3,
Wissenschaftliche Buchgesellschaft, 1974. no. 4 (1991): 587–618.
Durrant, Michael. The Logical Status of “God” and the Function of Theological Koehler, Ludwig, Walter Baumgartner, and Johann Jakob Stamm, eds. The
Sentences. London: Macmillan; New York: St. Martin’s, 1973a. Hebrew and Aramaic Lexicon of the Old Testament. Vol. 1. Translated
and edited under the supervision of M. E. J. Richardson. Leiden,
Durrant, Michael. “Appendix B: The Term ‘God’ in Christian Usage.” In Netherlands: Brill, 1994.
Theology and Intelligibility, 75–86. London: Routledge and Kegan Paul,
1973b. Kretzmann, Norman. The Metaphysics of Theism: Aquinas’s Natural Theology
in Summa contra gentiles I. Oxford: Oxford University Press, 1997.
Ebbesen, Sten. Commentators and Commentaries on Aristotle’s Sophistici
Elenchi: A Study of Post-Aristotelian Ancient and Medieval Writings on Lampe, Geoffrey W. H., ed. A Patristic Greek Lexicon. Oxford: Clarendon,
Fallacies. Vol. 1. Leiden, Netherlands: Brill, 1981. 1961.
Englebretsen, George. “Durrant on God.” New Scholasticism 48 (1974): Leftow, Brian. God and Necessity. Oxford: Oxford University Press, 2012.
251–252. MacDonald, Nathan. Deuteronomy and the Meaning of “Monotheism.”
Tübingen, Germany: Mohr Siebeck, 2003.
François, Gilbert. Le polythéisme et l’emploi au singulier des mots ΘΕOΣ,
ΑΙΜΩN dans la littérature grecque d’Homère à Platon. Paris: Les Belles McInerny, Ralph. “Can God Be Named by Us? Prolegomena to Thomistic
Lettres, 1957. Philosophy of Religion.” Review of Metaphysics 32, no. 1 (1978): 53–73.
Frede, Michael. “Monotheism and Pagan Philosophy in Later Antiquity.” In McInerny, Ralph. “Scotus and Univocity.” In Being and Predication,
Pagan Monotheism in Late Antiquity, edited by Polymnia Athanassiadi 159–164. Washington, DC: Catholic University of America Press, 1986.
and Michael Frede, 41–67. Oxford: Clarendon, 1999.
Mill, John Stuart. A System of Logic Ratiocinative and Inductive, Part I (The
Frede, Michael. “The Case for Pagan Monotheism in Greek and Graeco- Collected Works of John Stuart Mill, edited by J. M. Robson, vol. 7).
Roman Antiquity.” In One God: Pagan Monotheism in the Roman Toronto: Kegan Paul, 1974. First published 1843.
Definition: Theism
Miller, Barry. From Existence to God: A Contemporary Philosophical Argument. Smith, Mark S. The Early History of God: Yahweh and the Other Deities in
London: Routledge, 1992. Ancient Israel. 2nd ed. Grand Rapids, MI: Eerdmans, 2002.
Montagnes, Bernard. La doctrine de l’analogie de l’être d’après Saint Thomas Steenberghen, Fernand van. Dieu caché: Comment savons-nous que Dieu
d’Aquin. Louvain, Belgium: Publications Universitaires, 1963. existe? Louvain, Belgium: Publications Universitaires de Louvain; Paris:
Morris, Thomas V. “Defining the Divine.” In Definitions and Definability: Nauwelaerts, 1961.
Philosophical Perspectives, edited by James H. Fetzer, David Shatz, and George Sullivan, Arthur. Logicism and the Philosophy of Language: Selections from
N. Schlesinger, 269–283. Dordrecht, Boston, and London: Kluwer, 1991. Frege and Russell. Peterborough, ON: Broadview Press, 2003.
O’Connor, Timothy. “And This All Men Call God.” Faith and Philosophy Swinburne, Richard. The Existence of God. 2nd ed. Oxford: Clarendon; New
21, no. 4 (2004): 417–435. York: Oxford University Press, 2004. First published 1979.
Oppy, Graham. The Best Argument against God. Basingstoke, UK: Palgrave Swinburne, Richard. Is There a God? Rev. ed. Oxford: Oxford University
Macmillan, 2013.
Press, 2010. First published 1996.
Oppy, Graham. Reinventing Philosophy of Religion: An Opinionated
Swinburne, Richard. Coherence of Theism. 2nd ed. Oxford: Oxford
Introduction. Basingstoke, UK: Palgrave Macmillan, 2014.
University Press, 2016. First published 1977.
Owen, Huw Parri. Concepts of Deity. London: Macmillan, 1971.
Themistius. Analyticorum posteriorum paraphrasis. Edited by Maximilianus
Perczel, István. “Pseudo-Dionysius and the Platonic Theology: A Preliminary Wallies. Berlin: Reimer, 1900.
Study.” In Proclus et la theologie platonicienne, edited by Alain-Philippe
Segonds and Carlos Steel, 491–532. Louvain, Belgium: University Press, Toorn, Karel van der, Bob Becking, and Pieter W. van der Horst, eds.
2000. Dictionary of Deities and Demons in the Bible. 2nd ed. Leiden,
Netherlands: Brill, 1999.
Peterson, Michael L., William Hasker, Bruce Reichenbach, and David
Basinger. Reason and Religious Belief: An Introduction to the Philosophy of Twetten, David B. “Aquinas’s Aristotelian and Dionysian Definition of
Religion. 4th ed. Oxford: Oxford University Press, 2009. God.” Thomist 69, no. 2 (2005): 203–250.
Pini, Giorgio. “The Development of Aquinas’s Thought.” In The Oxford Twetten, David B. “To Which ‘God’ Must a Proof of God’s Existence
Handbook of Aquinas, edited by Brian Davies and Eleonore Stump, 491– Conclude for Aquinas?” In Laudemus Viros Gloriosos: Essays in Honor of
510. Oxford: Oxford University Press, 2011. Armand Maurer, CSB, edited by Rollen E. Houser, 146–183. Notre
Plantinga, Alvin. Does God Have a Nature? Milwaukee, WI: Marquette Dame, IN: University of Notre Dame Press, 2007.
University Press, 1980. Twetten, David B. “¿Un Dios para no-teístas? La definición pluralista del
Pseudo-Dionysius the Areopagite. Corpus Dionysiacum I: De divinis nombre ‘Dios’ en Tomás de Aquino.” In La sabiduría en Tomás de
nominibus. Edited by Beate Regina Suchla. Berlin: De Gruyter, 1990. Aquino. Inspiración y reflexión: Perspectivas filosóficas y teológicas, edited
by Liliana B. Irizar, 54–85. Bogotá: Universidad Sergio Arboleda,
Quine, Willard Van Orman. “On What There Is.” In From a Logical Point of
2017.
View, 1–19. Cambridge, MA: Harvard University, 1953.
Rocca, Gregory P. Speaking the Incomprehensible God: Thomas Aquinas on the Vacant, Alfred, Eugène Mangenot, and Emile Amann, eds. Dictionnaire de
Interplay of Positive and Negative Theology. Washington, DC: Catholic théologie catholique. Vol. 4. Paris: Letouzey et Ané, 1939.
University of America Press, 2004. Velde, Rudi te. Aquinas on God: The “Divine Science” of the Summa
Runia, David T. “The Beginnings of the End: Philo of Alexandria and theologiae. Aldershot, UK: Ashgate, 2006.
Hellenistic Theology.” In Traditions of Theology: Studies in Hellenistic Visser, Sandra, and Thomas Williams. Anselm. Oxford: Oxford University
Theology, Its Background and Aftermath, edited by Dorothea Frede and Press, 2009.
André Laks, 281–316. Leiden, Netherlands: Brill, 2002.
Vorländer, Hermann. Mein Gott: die Vorstellungen vom persönlichen Gott im
Russell, Bertrand. “The Relation of Sense-Data to Physics.” In Mysticism and Alten Orient und im Alten Testament. Kevelaer, Germany: Butzon and
Logic and Other Essays, 145–179. London: Longmans, 1918. Bercker, 1975.
Shepherd, John J. “Referring to God.” Religious Studies 10 (1974): 67–80. Ziff, Paul. “About ‘God.’” In Religious Experience and Truth: A Symposium,
Smart, J. J. C., and John J. Haldane. Atheism and Theism. 2nd ed. Oxford: edited by Sidney Hook, 195–202. New York: New York University
Blackwell, 2002. Press, 1961.

TOPIC 1
Definition: Atheism
Robert Nola
Emeritus Professor
The University of Auckland, New Zealand
This chapter discusses atheism, theism, and agnosticism and the nature of some of the definitions of these
doctrines. Rather than address arguments for or against god (or God), the chapter also considers definitions
that can clarify what these doctrines mean. It also discusses concepts of god as spelled out in various kinds of
definition.
SECTION 1: THEISM AND ATHEISM: DEFINING

ONTOLOGICAL ASPECTS
The doctrine of theism can be broadly understood to claim: there exists at least one or more gods.
This yields two varieties of theism. The first is polytheism, the doctrine that there are at least two or
more gods (in the case of Hinduism some say that there are millions of gods). Polytheisms often
have a pantheon of deities, spirits, ancestor spirits, divinities, ghosts, devils, demons, “forces”, “cos-
mic minds”, and the like; whether all of these are also gods is a matter for polytheistic theologians to
sort out. The second variety of theism is monotheism, the doctrine that there is just one and only
one unique god (often with a capital ‘G’). Monotheists pride themselves in having replaced the
polytheistic pantheon, or in suppressing belief in it. Thus Christianity replaced the polytheisms of
ancient Greece and ancient Rome. Pre-European New Zealand Māori polytheism, once it came into
contact with Christianity, had to find room for a single supreme god for which an old word already
in use in their “religion”, ‘atua’, was pressed into service as that deity’s name. Finally, atheism is the
zero option—namely, there are no gods; or, there is not even one unique god.
The three doctrines (polytheism, monotheism, and atheism) exhaust the space of possibilities
about the number of gods that are said to exist (many, one, or none); they are also exclusive of
one another and only one can be true. For clarity these three doctrines are set out as explicit defini-
tions. But the definitions say nothing about any properties god(s) might have, and only say whether
god(s) exist or not. Some properties commonly attributed to god(s) are discussed later.
The word ‘atheism’ came into English in the seventeenth century, following the French
‘athéisme’ of the previous century. The term ‘theism’ comes from the ancient Greek for god, ‘theos’;
ancient Greek also gives us the prefixes ‘poly-’ (many) and ‘mono-’ (one, single). Without going
into the niceties of the ancient Greek language (not strictly relevant to our modern English usage)
the privative ‘a’ can be used to indicate negation; hence a-theism, no god (see Whitmarsh 2015,
116–117). Ancient Greek provides English with a number of pairs of words that parallel the athe-
ism/theism distinction. Thus in the case of wounds we say that the word ‘aseptic’ means ‘no infec-
tion’, while the word ‘septic’ means the presence of infection.
Each statement of the three doctrines employs the word ‘exist’, to be understood in its broad-
est, most inclusive, sense. Here philosophers introduce a technical term and talk about ontology—
the study of what exists, or does not exist, in the total world. The god(s) that are said to exist are
taken to exist independently of us humans and/or our cognitive powers. This is a realist view to
19
Definition: Atheism
be contrasted with the antirealism of a “constructivist”, or “pro-

KEY CONCEPTS jectionist”, view about god(s), in which their existence depends
in some way on what we think or believe. The three existence
doctrines are strictly ontological in character. They are not
Antitheism
doctrines about what we know, believe, think, or suppose, or
Bayesianism doctrines for which we have evidence for or against; these are
Cluster Definition independent matters about our cognitive or evidential stances
toward each of the three doctrines. Much confusion is
Comprehensive Atheism
introduced if cognitive matters are muddled with matters of
Constructivist View about God(s) ontology—as is noted in the next section.
Deism In the doctrines of monotheism and atheism, what god is
Empiricism being talked about? When it comes to defining the term
‘god’, there is a large number of different definitions, or “con-
Evidential Agnostics cepts” arising from various religions, or from various philoso-
Immanence (of God) phies of religion. Attempts can be made to set out each concept
Ironic Agnosticism as an explicit definition that employs other well-understood
properties in what does the defining. One example is the classi-
Materialism cal, traditional definition, which employs the set of properties:
Misotheism {Omnipotence, Omniscience, Omnibenevolence}. For short
Ontological Atheism we can say that this gives us the concept of the “OOO-god”.
The monotheist claim that a unique god exists is then taken
Ontology
to be definitionally equivalent to the following conjunction:
Panentheism there is a unique item x such that (x is omnipotent & is omni-
Pantheism scient & is omnibenevolent). But is there a unique entity that
satisfies these properties? Are atheists right in claiming that no
Positivism
such god exists, that is, no (unique) entity satisfies this conjunc-
Rational Degree of Belief tion of properties? Further discussion of this, along with other
Strong Agnosticism “conceptions” of god, are deferred until the final section.
Transcendence (of God) Definitions other than explicit definitions of “concepts” can
be employed. An important alternative is the “family-resemblance”
Unknowability Agnostics style of definition proposed by Ludwig Wittgenstein. Each “con-
Weak Agnosticism cept” is taken to be a cluster of properties, the cluster being under-
stood to be a conjunction of at least a minimal subset of properties
drawn from a larger set of properties. Different clusters of proper-
ties (drawn from the larger set) can be employed for different
“concepts”. Here the different “concepts” of god can have a family resemblance to one another. But
there is no set of necessary and sufficient properties for being a god, and there is no set of properties that
gives us the “nature” or “essence” of god. Importantly, whether or not god exists can depend crucially on
what cluster of properties is associated with the “concept” of that god. A further style of definition is dis-
cussed in section 5.
Having introduced the above three doctrines we can ask: under which are pantheism and
panentheism to be placed? The term ‘pantheism’ comes from a Latinized version of the Greek
‘pan’ (‘all’) combined with ‘theos’. This suggests, when expressed succinctly, that god is everything
and everything is god. Putting this another way, one might claim that pantheism is the doctrine that
god is identical with the total (natural) world. On the face of it this makes pantheism a monothe-
ism. If we allow that the total (natural) world exists and there is just one such world (obviously),
and god is said to be identical with it, then there is just one god.
At this point we need to introduce a pair of contrasting properties that have been used to
“define” different conceptions of the traditional god (these properties might also be part of a Witt-
genstein family-resemblance definition). It is claimed either that god is transcendent (existing
beyond the total world) or that god is immanent (somehow only pervading the total world). If some
insist against pantheism that god cannot be identical to the total world and in some sense must tran-
scend it, then pantheism, in denying a transcendent god, is a version of atheism (and is also akin to
Definition: Atheism
naturalism). Historically, pantheism was often regarded as a form of atheism. Pantheism makes
it difficult to account for the traditional claim that an independent god created the total world.
Arthur Schopenhauer objected to the doctrine of pantheism, saying that it merely adds another
word for ‘atheism’ to our vocabulary. He added that, “to call the world God is not to explain it;
it is only to enrich our language with a superfluous synonym for the word ‘world’” (Schopenhauer
1974 [1851], Vol. 2, 99; single quotation marks around the last word have been added). It is not
our task here to enter into the various arguments about these matters. But panentheism attempts to
avoid these problems with pantheism. It maintains that even though god “interpenetrates” the total
world, god is not identical with it; god is said to be greater than, and independent of, the total world.
However, this is to be understood, it appears to be a version of monotheism rather than atheism.
We can also ask where, in the tripartite distinction, some of the various religions are to be
placed. For example, there is a scholarly argument about whether or not Theravada Buddhism or
Confucianism are atheistic (Martin 2007a). Clearly the Christian god and the god of Islam are
monotheistic; and even though the early Judaic religion might be claimed to be polytheistic, later
Judaism is firmly monotheistic. These matters are different from the question whether, in the three
Abrahamic religions, the three gods are distinct or the same (for more on this, see section 3).
Historically, an important distinction was drawn between theism and deism (from the Latin for
god, ‘deus’). Both accounts of god are monotheist, but each draws on different sets of defining
properties in its respective cluster definitions. The idea of deism is of a more remote, noninterven-
tionist god who only creates the universe along with the laws that govern it; then the deistic god
leaves the universe to evolve through its history according to those laws without any further inter-
vention. Contrary to common suppositions about a theistic god, the deistic god does not intervene
in the world in the following ways: (a) to create human beings and other creatures, (b) to deliver a
holy book for humans to read and obey, (c) to intervene by causing miracles, (d) to issue divine rev-
elations and/or make prophesies, (e) to make prayer efficacious, and so on. Humans have come to
exist with powers to reason and to make observations, both of which are important for any science.
The deistic god’s relation to the world is conducive to a scientific approach to the world. Apart from
the initial act of creation, there is a law-governed world, independent of any further action by a deis-
tic god, that is open to scientific investigation. Hence the prevalence of the idea of a deistic rather
than a theistic god during the early periods of the scientific revolution. Both deism and theism
are monotheistic; but they do differ considerably over what properties are to be associated with their
respective “concepts” of god.
Some might object that the definition of ontological atheism given at the beginning of this sec-
tion is too thin; there is a much more common, “thicker” concept, which can be called ‘comprehen-
sive atheism’. Its “definition” makes explicit the kind of impact the atheistic denial of god’s existence
has on religions such as Christianity. Such an atheism is taken to provide a comprehensive picture of
the world that rivals the various comprehensive pictures provided by a religion. Thus a comprehensive
atheism, like deism, not only rejects the items above (a) to (e), it also rejects the following: (f) the deis-
tic idea that a god might have created the world with its laws, (g) the idea that there exists a deistic, or
a theistic, god, (h) any role for Jesus Christ such as his crucifixion saving us from original sin, (i) the
idea that Jesus Christ might have existed, (j) the idea that there is life after death, (k) any kind of reli-
gious eschatology in which there is a final end-story to be told about life and its meaning as well as an
end-goal of history, such as establishing the Kingdom of God on Earth, (l) claims about religious spir-
ituality, (m) the idea that anything is sacred, (n) the idea that morality can be founded on the com-
mands of a divine god, and so on. Comprehensive atheism can also have positive aspects. On the
whole it accepts the following: a secular society and education system, freedom of speech, tolerance,
a naturalistic worldview, and so on. There can be different versions of comprehensive atheism depend-
ing on which of the above rejected and/or accepted properties are part of any associated cluster defini-
tion. What we can now envisage is a range of conceptions of atheism, from bare ontological atheism
to some form of comprehensive atheism (analyzed as a cluster definition). However, in what follows
the core idea of ontological atheism will be the main focus.
Though some atheists might be naturalists (or materialists), the doctrine of atheism does not
entail naturalism. In ontology there are many categories of what exists. Here are four. One quick

Definition: Atheism
way of defining naturalism is to say that it comprises a causal nexus of all the items to be found in
the space-time framework. Some of these entities are part of our commonsense view of the world,
but many of them are the result of scientific discoveries. Distinct from this is a second category of
what might be called “abstract entities”, such as properties, universals, sets, numbers, geometrical
objects, meanings, possible worlds, and the like. A third category can be envisaged in which there
exist minds and souls as substances (such as mentioned in the Sāmkhya doctrine of atheism—see
section 3). These can exist without there being a god. A fourth category could then be reserved
for supernatural entities such as god(s). Not all gods are supernatural, as pantheists would
claim. But it is clear from the fourfold division of ontological categories that it would be wrong to
infer that atheists, who deny the category of the supernatural, must thereby be naturalists. Anti-
supernaturalism would be tantamount to naturalism for those atheists who also deny the second
and third categories. But atheists can still be nonnaturalists in admitting that there are abstract enti-
ties and/or in being dualists about the mind without admitting that there are supernatural entities.
SECTION 2: THEISM AND ATHEISM: COGNITIVE ASPECTS
People can adopt a number of different cognitive attitudes to each of the three doctrines of mono-
theism, polytheism, and atheism. In what follows, let letters like ‘x’, ‘y’, and so on stand for persons;
let ‘C’ be some cognitive relation of, say, believing, thinking, supposing, hoping, wondering
whether, and so on; and let ‘p’ stand for any proposition to which a person can be cognitively
related, in particular any one of the above three doctrines. Generally cognitive attitudes will be of
the form ‘xCp’. For example, a polytheist x believes (or, if you like, thinks, supposes, accepts, etc.)
that there exist many gods (and can so believe with or without evidence). What cognitive attitude
should polytheists then take toward the other two rival doctrines? If the person is minimally ratio-
nal, person x ought to have a cognitive attitude of not believing the two rival doctrines, namely,
it is not the case that x believes that there is just one god, and it is not the case that x believes there
is no god. We can take people to be rational in the narrow sense that they do not hold inconsistent
beliefs (though, as is well known, humans are prone to this cognitive defect).
Similarly, ‘x is a monotheist’ can be defined as: ‘x believes that there is just one and only one god
(with or without evidence)’. If we suppose that x is minimally rational in not entertaining inconsistent
beliefs, then it is not the case that x believes that there is no god, and it is not the case that x believes
that there exist several gods. Finally, a contextual definition of what it is to be an atheist can be stated
as: ‘x is an atheist = (by explicit definition) x believes that there is no god (or no gods).’ If x is mini-
mally rational then it is not the case that x also believes the doctrines of monotheism and polytheism.
An inquiring person x can have different cognitive attitudes, such as wondering whether atheism,
or theism, or polytheism, is true. Others might hope that one of these doctrines is worthy of enter-
taining. Still others merely going along with their community might only accept that there is one
and only one god (but not, say, accept that there are many or no gods); they do not have a stronger
cognitive stance of, say, believing. Again, having faith can be taken to be a cognitive attitude; an
atheist can have faith that there is no god just as a theist can have faith that a unique god does exist.
Looking at matters this way enables a number of fine distinctions about different kinds of cog-
nitive attitude. Consider the cognitive attitude of believing (where ‘B’ stands for this cognitive rela-
tion, and ‘p’ is the “propositional object” of cognition, say, the doctrine that god exists). Suppose
person x believes that god exists: this has the form xBp. The denial of this has two forms, since
the negation operator can occur in two different places. The first is in front of the entire expression:
not-(xBp). This says that it is not the case that x believes that god exists. Sometimes this is called
“negative atheism”, that is, the absence or lack of a belief in god. But more correctly this ought to
be called something like ‘the negative atheistic cognitive attitude’ so as not to suppose that the
phrase ‘negative atheism’ has anything to do with the ontological doctrine of atheism.
In the second form the negation sign is inside the expression and operates on the proposition
‘p’, so that we get ‘not-p’, that is, xB(not-p). That is, x believes that it is not the case that god exists
(xB(not-god exists)). Sometimes this is said to express “positive atheism”, that is, the positive belief
that god does not exist (see Martin 2007b, 1–2). But more correctly this ought to be called ‘the
Definition: Atheism
positive atheistic cognitive attitude’ so as not to suppose that the phrase ‘positive atheism’ has any-
thing to do with the ontological doctrine of atheism. The difference between the two atheistic
stances is that between merely lacking a belief versus having a positive disbelief. We commonly
use the expression ‘x disbelieves p’ but it can harbor an ambiguity. It is not always clear whether dis-
belief is of the form ‘not-(xBp)’ or of the form ‘xB(not-p)’. However, the context of use of the term
‘disbelieve’ might help to eliminate ambiguity.
Taking note of these points now enables us to express the idea of suspension of belief
(sometimes adopted by agnostics), that is, it is not the case that person x believes that god
exists, and it is not the case that x believes that god does not exist. Looked at formally, this says:
not-(xBp) and not-(xB(not-p)). Clearly, suspension of belief is not the same as being an atheist,
which just says ‘xB(not-p)’. Keeping in mind the syntax of these expressions and the way logical
operations occur, such as negation, enables one to express many kinds of cognitive relation that
humans can have in a religious context, and the different propositions to which x can have a
cognitive attitude.
Antitheism is the expression of an opposition to god (from the Greek anti, against), an extreme
version of this is misotheism (from the Greek miso, hatred), a hatred of god or of god’s existence
(which can occur even when god exists). Antitheism (the word dates from late-18th-century
English) can come in a number of different strengths and varieties of opposition and can be directed
on different kinds of religious objects, such as: god, belief in god’s existence, god’s works, and reli-
gion in general. Thus Christopher Hitchens adopts an anti-theistic stance to religion as a whole
when he says:
I’m not even an atheist so much as I am an antitheist; I not only maintain that all religions are
versions of the same untruth, but I hold that the influence of churches, and the effect of
religious belief, is positively harmful.… I do not envy believers their faith. I am relieved to
think that the whole story is a sinister fairy tale; life would be miserable if what the faithful
affirmed was actually the case. (Hitchens 2001, 55)
Note that the object of Hitchens’s anti-theism shifts from god to institutional religion and its
alleged harms.
The philosopher Thomas Nagel gives a different version of anti-theism when he talks of the
cognitive attitude of wanting atheism to be true:
I want atheism to be true and I am made uneasy by the fact that some of the most intelligent
and well-informed people I know are religious believers. It isn’t just that I don’t believe in God
and, naturally, hope that I’m right in my belief. It’s that I hope there is no God! I don’t want
there to be a God; I don’t want the universe to be like that. (Nagel 1997, 130)
Finally, Wittgenstein expressed a version of anti-theism when he said to a friend: “if I thought of
God as another being like myself, outside myself, only infinitely more powerful, then I would regard
it as my duty to defy him” (Rhees 1984, 108).
Can a person be an atheist without being anti-theistic, and conversely? Suppose a person is an
atheist. Since they believe that god does not exist, then they also believe that there is nothing to
oppose in the case of such a nonexistent. Their anti-theism concerning god lacks any actual opposi-
tion. But anti-theists like Nagel or Wittgenstein need not always be atheists. Nagel hopes that athe-
ism is correct but recognizes that he could be wrong. If Nagel’s hopes are realized, then there is no
god. But one can envisage a possible world (ours might be like this) in which god exists and in
which putative atheists have their hopes dashed about the nonexistence of god. However, the person
still actively opposes god and god’s works while perhaps being in a state of anguish or anger about
their situation.
In the above, a cognitive relation is directed upon a “propositional object”. But cognitive rela-
tions can also be directed upon ordinary objects or putative objects (since they might not exist).
Thus a person might worship the queen of England (she does exist) and also worship the god
Wotan (which does not exist) or worship Santa Claus (an agreed nonexistent). Worshipping is a
cognitive stance that people can take toward the putative object god. Clearly if a person x believes
that god does not exist, then x cannot sensibly be said to worship god. In order for monotheist y

Definition: Atheism
to worship their god, then y has to at least believe that god exists (even when they are mistaken
about what exists). From an atheist point of view, y might worship a nonexistent, but y does not
believe that they do so.
Do polytheists worship all the gods they think exist in their pantheon? Maybe not, as some
might include devils and evil spirits in their pantheon that one would not normally worship. At
best, only some of the gods in the pantheon are to be worshipped. The term ‘henotheism’ was
introduced by scholars in the nineteenth century to mean ‘one god’ (henos, ‘one,’ + theos). They
recognized that polytheists might believe that there are many gods but worship only one of them.
Such a possibility might indicate a change in polytheism in the direction of monotheism. But not
always. Suppose the various tribes of a broader nation each have a different local tribal god.
Members of (say) Tribe 1 worship god1 but members of Tribe 2 worship a different god, god2.
However, the members of each tribe recognize that other tribes have their own god; each worships
their own god while members of other tribes do not worship that god (even though they think
that that god still exists). Outside the context of polytheism, a version of henotheism might be
envisaged among the adherents to each of the three Abrahamic faiths, Judaism, Islam, and Chris-
tianity. Here there appear to be three gods (we set aside theological considerations about whether
Yahweh, Allah, and the Christian God are the same and treat them as different). A person might
be unorthodox and be somewhat polytheistic in admitting that all three gods exist, but they wor-
ship only one of them. More likely they would be theistic toward one of these gods and atheistic
toward the other two rival gods.
It is important that ontological matters concerning the existence of god(s) not be confused with
cognitive attitudes toward god(s), such as belief in their existence or nonexistence. But alas this con-
fusion is made by theists and atheists alike. Thus Julian Baggini says: “Atheism is in fact extremely
simple to define: it is the belief that there is no God or gods” (2003, 3; italics added). Not so, since
the word ‘belief’ makes this more a definition of what it is for a person to be an atheist rather than
the ontological doctrine of atheism itself. Tim Crane tells us: “Atheism may be quite simply defined
as disbelief in god” (2017, 18). A similar position is adopted by Steven Pinker when he says that
atheism is “the absence of a belief in God” (2018, 430). If these two accounts involve either a posi-
tive or negative atheistic cognitive attitude (as defined above), then they are to be distinguished from
the ontological doctrine of atheism. Finally, Stephen Bullivant (2013) notes the ambiguities
involved in attempting to define ‘atheism’ and ‘atheist’ but prefers to live with them rather than
eliminate them. This is in contrast with the line adopted here in which ontological matters are
not to be confused with cognitive matters; and the definitions proposed take into account the ontol-
ogy/cognitive distinction.
SECTION 3: SOME HISTORY OF ATHEISM, AND OF ATHEISM TOWARD

THE GODS OF OTHERS
Atheism is a doctrine as old as many religions. There are strains of atheism in Indian “religious”
movements such as Jainism (in which there are gods but no supreme God) and Buddhism (which
tends to deny a personal god; but some Buddhist traditions are not averse to a pantheon of divini-
ties). The most explicit and oldest form of atheism is Cārvāka (sixth–fifth century BCE). It is also
known as Lokāyata because it emphasizes that only this (material) world (loka) exists and no other
(Chattopadhyaya 1973, 3). It can be traced back to the Vedic religions and is mentioned in some
dialogues of the Buddha. It adopts the following positions: a materialism that denies that there is
any soul or mind distinct from the body; it denies that there are any gods and rejects the idea of
reincarnation and an afterlife; and it tells us not to bother with religious rituals. In a slightly differ-
ent vein, the earlier doctrine of Sāmkhya rejects the materialism of Cārvāka in favor of a dualism of
matter and spirit; but like Cārvāka it is atheistic in rejecting god. What is important about these
ancient Indian doctrines is that both monistic and dualistic ontologies can be maintained without
any commitment to a further ontological category of god. (See Sen 2006, chap. 1; Chattopadhyaya
1973, 1980.)
Definition: Atheism
There is an ancient Greek tradition of providing lists of those who were said to be atheists. On
most lists is Diagoras of Melos (later fifth century BCE), often said to be the first Greek atheist. A
shadowy figure about which little is known today, he is rumored to have written a book (now lost
and content unknown) with an obscure title that could be understood to mean “destructive consid-
erations” against god(s) (see Drachmann 1922, chap. 3; and Whitmarsh 2015, chap. 8). He may
have attracted attention as the “first” atheist because, in about 415 BCE, he was charged with impi-
ety and driven out of Athens. Since he was influenced by atomists such as Leucippus and Democri-
tus, who also espoused atheism, perhaps it is misleading to think of him as the “first” atheist; rather,
he was a member of a small but growing group of philosophical materialists and/or religious skeptics
of his day. In fact, given that the Greek word ‘atheos’ was in use long before Diagoras was said to be
one, there must have been a longer tradition of atheism. Ancient Greek materialism, like the mate-
rialism of Cārvāka, had no room for god(s) in its ontology. In light of this we should not include
the later Epicurus (c. 300 BCE) or the Roman Lucretius (first century CE) among atheists; even
though they were atomists they claimed that gods exist but such gods were entirely uninterested
in human affairs.
The word ‘atheism’ has not always been used to indicate an ontological doctrine. In ancient
Greece and medieval Europe, it was used to talk of those who worshipped the wrong gods or
god, that is, those not recognized by the state or by the dominant religious authorities. Thus
Socrates was accused at his trial of both teaching atheism and being an atheist (atheos) (Apology
23D, 26B–C). But he found the accusation of being an atheist in a polytheistic society ambiguous
between disbelief (in all gods) and wrong belief (believing in the wrong bunch of gods—a stance
that could also get one in trouble). Socrates asks his accuser Meletus to clear up the ambiguity:
Is it that I teach people to believe in some gods (which implies that I myself believe in gods
and am not a compete atheist, and so not guilty on that score), but in different gods from
those recognized by the state, so that your accusation rests upon the fact that they are
different? Or do you assert that I believe in no gods at all, and teach others to do the same?
(Plato 1993, Apology 26C)
Meletus replies that the charge is “that you disbelieve in gods altogether,” something Socrates
denies. In the polytheistic context one is a “complete atheist” in the sense that one believes in no
gods at all; or one is (let us say) a “quasi-atheist” in the sense that others, or the state, do not recog-
nize the gods in which one believes. The issues raised are in the context of polytheism, but they also
resonate with monotheism. In that context, too, a “complete” atheist is one who believes that there
is no god at all. Quasi-atheism, when applied to monotheisms, becomes the view that each mono-
theist does not recognize the god of other monotheists. Early Christians were often regarded as athe-
ists. Their insistent monotheism would have prevented them from recognizing the approved gods of
the Roman state; and the Roman state did not acknowledge their Christian god. From the stance of
the state they are “quasi-atheists” (in the above terminology).
Similar issues arise in the account of atheism given by the former Archbishop of Canterbury,
Rowan Williams, in his “Analysing Atheism: Unbelief and the World of Faiths.” He tells us that
“people often talk” of atheism as a view resting on a “central conviction—that there is no tran-
scendent creative power independent of the universe we experience” (Williams 2012, 281). But
if the cognitive attitude of a conviction is included in the definition then this might tell us what
an atheist is, but not atheism. So take atheism to be the propositional content of the conviction,
namely, “there is no transcendent creative power independent of the universe we experience”.
This builds quite a lot into what god is supposed to be: transcendent, independent of the word
we experience, creative, a power. Some theists might believe some of this but not all, for example,
pantheists. In the ontological definition of theism given so far the word ‘god’ is used in a thin and
neutral way; nothing is specified about the properties god might or might not have (see more on
this in the final section).
The main thrust of Williams’s essay is indicated in its subtitle, “Unbelief and the World of
Faiths.” What Williams wrestles with is the fact that adherents of each of the various current reli-
gions, especially the three Abrahamic religions, maintain claims about their own religion’s god while
other religionists disbelieve these claims and instead make quite different claims about their

Definition: Atheism
respective god. Williams lists many such contested properties (2012, 285–289). His position can be
summarized thus: Christians tend to hold that God is a trinity while Muslims reject that Allah is a
trinity; Christians accept a God who has personal agency while Buddhists tend to reject this; the
Jewish Yahweh has a special covenant with his people, but Muslims do not believe this of Allah
and Christians do not believe this of their god; and so on. So we can now ask: for the monotheists
of the Abrahamic religions, are there one, two, or three gods?
One answer might be a kind of syncretism (as in the Bahá’í religion) in which the differences
are stripped away and one and the same god exists for all the different religionists. Another answer
might be to adopt an atheistic stance toward the god of other monotheists. Other gods would then
be on a par with the more than 170 divinities the satirist H. L. Mencken lists in his “Memorial
Service” (see Hitchens 2007, 143–146), such as Baal, Zeus, Wotan, Mithra, Huitzilopochtli
(a Mexican god to whom thousands of the reverent were sacrificed), and so on. Even those who are
monotheistic with respect to their own god are resolutely complete atheists when it comes to these
gods; in particular, each monotheist is atheistic toward the god of the other Abrahamic religions.
This leads to a problem for different monotheists that does not arise for polytheists with their
many gods. Each monotheist believes there are properties of their god that other monotheists do
not believe hold true of their respective god. Putting this in another way, one is a theist toward
one’s own god (obviously) but somewhat unaccepting or “atheistic” toward the god of other reli-
gions because that god lacks some important property of one’s own god, or has some undesirable
property one’s god does not possess. This is akin to Socrates’s “quasi-atheism” but in a slightly
new guise. It is not, as in the case of polytheism, that one believes in gods that others do not believe
in; rather, in the case of monotheisms each monotheist predicates of god properties that other
monotheists deny. One might ask whether the differences between the predications could be so
great that it cannot be said that all monotheists believe in the same god! This is highly likely, but
it is not an issue addressed by Williams.
From this diversity of belief Williams concludes: “There is no such thing as a global system of
‘atheism’: there are denials of specific doctrines on varying grounds” (2012, 290). This is wrong in
one way and right in another. It is wrong if Williams thinks that there cannot be any global argument
against all theisms (see Draper 2017 on this). But he is right if he means that, insofar as atheists enter
into a dialectic with theists, there are many different arguments to advance against each of the different
conceptions of god, of which the OOO-god is just one. For more on this see the end of section 5.
SECTION 4: VARIETIES OF AGNOSTICISM
Agnosticism is not an ontological doctrine; it is an epistemic doctrine about what limits there might
be to our knowledge of god’s existence. In a minimal sense agnosticism is an absence of knowledge
about whether god exists or not. Agnosticism is an old doctrine. The ancient Greek sophist Pro-
tagoras is said to have begun one of his (lost) books, titled On the Gods, by saying: “Concerning
the gods I am unable to discover whether they exist or not, or what they are like in form; for there
are many hindrances to knowledge, the obscurity of the subject and the brevity of human life”
(Guthrie 1971, 234). Though the doctrine is old, the name for it is much more recent. It was intro-
duced in 1869 by Thomas Henry Huxley to name his uncertainty about god’s existence or non-
existence. He derived the word ‘agnosticism’ from the Greek privative ‘a’ combined with the Greek
‘gnosis’, meaning ‘knowledge’, particularly divine or spiritual knowledge. Agnosticism comes in
several forms, a few of which are mentioned here.
Strong agnosticism denies that it is possible to complete any inquiry into whether or not god
exists. These are unknowability agnostics (see Drange 1998). The truth or falsity of the claim that
god exists is said to transcend our epistemic ability to ever have sufficient evidence to determine
which truth value holds. On this view the absence of knowledge concerning god’s existence is not
a contingent matter that could be rectified; the lack of knowledge arises necessarily from the nature
of both the subject matter and the extent of what we humans can and cannot know about god’s
existence.
Definition: Atheism
From a realist stance, either theism holds and god is in the totality of the world, or atheism
holds and god is not in it. But the stance of unknowability agnosticism says that we can never tell
which of theism or atheism is correct. Another way to put the matter is that both theism and athe-
ism lack testability conditions; it is impossible for us to test (confirm or disconfirm on the basis of
evidence or reasons) either theism or atheism. There is a tradition in philosophy, called among other
things ‘empiricism’ or ‘positivism’, which says that if it is impossible to test a pair of contradictory
claims, such as theism and atheism, then they are, in a certain sense, meaningless; they lack all
cognitive significance. If we grant this thesis of meaning, then another form of agnosticism can be
distinguished: ironic agnosticism. Such agnostics do not say that one of theism or atheism is correct;
rather they say “a plague on both -isms.” On this view, too, the question “does god exist?” has no
answer one way or the other and so is said to lack cognitive significance. Perhaps such an agnosti-
cism can be attributed to the philosopher David Hume (see Blackburn 2008, chap. 9, “Natural
Religion”; and Blackburn 2009). Those who adopt such a Humean stance do not join the the-
ism/atheism debate; they simply gaze, with detached irony, at the pointless fuss over something that
is intrinsically meaningless, to the annoyance of both theists and atheists. Some atheists are attracted
to the stance of meaninglessness about divine matters. Thus Bertrand Russell makes the supposition
of an invisible teapot orbiting the sun: “If I were to suggest that between the Earth and Mars there
is a china teapot revolving about the sun in an elliptical orbit, nobody would be able to disprove my
assertion provided I were careful to add that the teapot is too small to be revealed even by our most
powerful telescopes” (Russell [1952] 1997, 547–548). This claim, like those about god’s existence,
is untestable.
Some theists, most atheists, and what will be called evidential agnostics, reject unknowability
agnosticism, claiming that it is in principle possible to come to know the truth value of the claim
‘god exists’. Evidential agnostics say that, given current evidence, we cannot know which truth value
it has (though it will have one). But it remains an open possibility that further evidence will favor
the atheists’ verdict that it is false or the theists’ verdict that it is true.
In section 2 the cognitive stance of belief was introduced. But belief is not an “all-or-nothing”
cognitive state. In Bayesianism, belief can come in degrees and when interpreted as a probability can
be understood as a rational degree of belief (see Bradley 2015). This puts talk of agnosticism in a new
light, since some of the above positions can be usefully reformulated using the notion of a rational
degree of belief rather than that of “knowledge” (which, at best, becomes a high degree of rational
belief). Thus if x is a person and ‘RB’ stands for rational degree of belief and ‘G’ is some proposition
(or hypothesis) believed, such as that god exists, then we can express this as xRBt(G) = r, where ‘r’ is
a number in the interval [0, 1], and ‘t’ is a time at which the belief is held by x (note that rational
degree of belief can vary over time with evidence). On the Bayesian view that these are probabilities,
we can write this as Probx, t(G) = r, where the probability is that for person x at time t. (If these are
understood then we can drop the indexing to x and t and write this more simply as Prob(G)).
Understanding rational degrees of belief to obey the probability calculus enables us to give them
numerical values on the scale [0, 1]. Also it enables us to consider both degrees of rational belief
without evidence (as expressed above) and with evidence, such as xRBt(G, E) = r (where G is some
hypothesis, such as that god exists, and E is evidence). Understood as a probability this becomes
Prob(G, E) = r (the indexing to x and t being understood). These considerations are of a piece with
the powerful Bayesian theory of rational degrees of belief, which is widely employed in many con-
texts in philosophy and can be extended to claims in the philosophy of religion.
Granted this understanding, we can express the positions of atheists and theists, not in terms of
“knowledge”, but in terms of numerical degrees of rational belief. In the case of atheists we can say
that their rational degree of belief (i.e., probability) that god exists, given evidence E, is 0 or close to
0. For less hard-line atheists the probability that god exists, given E, can range anywhere within the
scale from 0 to 0.5 but must be less than 0.5. Theists will have a reasonable degree of belief that god
exists given E, which is close to or exactly 1. In general, for theists their reasonable degree of belief
given E can range from just above 0.5 to 1. The position of weak agnosticism can now be character-
ized as falling in the middle of the scale: in light of their current evidence, the weak agnostics’ ratio-
nal degree of belief that god exists given E is the same as their rational degree of belief that god does

Definition: Atheism
not exist given E; both are 0.5, or about 0.5. For weak agnostics, current evidence E is not decisive
either way; but it is open for the evidential situation to change so that the reasonable degree of belief
will change either up or down.
If numerical values are not available (as is common), one can consider comparative probabili-
ties; that is, we may consider whether Prob(G, E) is less than, equal to, or greater than Prob(not-
G, E). Or as evidence changes from E to E* we can ask which of the following hold: Prob(G, E)
is less than, equal to, or greater than Prob(G, E*). And so on for other important relationships
within not just agnosticism but more generally within the theory of rational degrees of belief applied
in the context of religious matters.
It is important to note that weak agnosticism is quite different from strong agnosticism, which
is not open to evidential change. Weak agnosticism says that we do not now have sufficient evidence
one way or the other about god’s existence, but this could change; it does not make strong agnosti-
cism’s claim that we could never have such evidence. The ontological doctrine of atheism and the
epistemic doctrine of agnosticism are logically independent of one another; one does not entail
the other. For example, it is possible for both to obtain together; the position of (weak) agnosticism,
namely, the degree of rational belief is middling, is consistent with ontological atheism, namely, god
does not exist. Importantly, an atheist’s degree of belief in theism can be low (but not necessarily 0)
while atheism is true.
SECTION 5: IS THE WORD ‘GOD’ ABOUT SOMETHING, OR NOTHING?
In the definitions of monotheism and atheism, the word ‘god’ is used. But what does the word
mean and does it refer to anything? For the atheist, whatever the word means, it does not refer to
anything (any referent is fictional). In contrast, the monotheist must suppose that the word refers
to a unique, actual “something”; but often theists are hard pressed to say what the “something” is
even though they claim that the unique “something” exists. Sometimes monotheists of a mystical
bent say that it is a bare existent about which we can say nothing beyond the fact that it exists;
so we cannot properly speak of god (see Williams 2012, 284). But this is an extreme position that
is too paradoxical, for one can ask: what does the last word ‘god’ of the previous sentence refer to (if
anything)? It is not that the word is not about anything—that is atheism. Rather there is a “some-
thing”, but we are not in a position to use a word to refer to it or say anything else about it (for
whatever reason). Most monotheists do not adopt such an extreme position and assume that the
word ‘god’ does refer to something and has a meaning, though it might be very hard to specify
either the meaning or the referent. In this section we consider some of the ways in which the word
‘god’ might have a referent attached to it by means of associated descriptions.
Philosophers of religion often talk about “the concept of god”, which can be spelled out by
means of an explicit, or a cluster, definition employing a number of “defining” properties. As
already suggested in section 1, a common, traditional conception of god might be specified by
an explicit definition of ‘god’ in terms of the conjunction of the three properties of being omni-
scient, being omnipotent, and being omnibenevolent—the OOO-properties. Some contend that
there can be concepts of god other than the concept of the OOO-god. Here one might suppose
that there is one and only one god, but that we might have different concepts of it in much the
same way that there is one and only one planet Venus, but we have different uniquely identifying
descriptions of it such as ‘the morning star’ and ‘the evening star’. It was an empirical discovery
that the so-called “stars” were not really two but one. But those who talk of “different concepts
of god” do not always tell us how the different concepts manage to pick out the one and same
god. So sometimes we are left with differing “concepts” with differing items that might uniquely
fall under each of these concepts.
For those who adopt a realist stance about their understanding of god this is unfortunate. The
game has been given away to those who are “constructivists” about god; we humans supply the con-
cepts but reality has nothing to do with providing an actually existing item to fall under each con-
cept. Perhaps we could say that we humans construct fictional objects to satisfy each concept. Such
Definition: Atheism
an antirealist account of god might suit those who think that god is merely a human projection onto
the world (Ludwig Feuerbach is such a projectionist and Hume is sometimes said to be another).
Such a position would be included under atheism, as no god exists independently of its human pro-
jection; there is just the human-dependent projection, whatever that may be. The following discus-
sion sets aside such theistic antirealism.
Returning to the traditional concept of god just mentioned, a number of requirements have to
be met. Theists would have to: (1) define carefully what are the OOO-properties mentioned in the
definition; (2) show that some item does fall under the conjunction of OOO-properties; (3) show
that one and only one item falls under the conjunction. Some have argued that this project cannot
be completed; there can be no item that falls under the conjunction of three properties. Anthony
Kenny (1979) makes the case that the two properties of omniscience and omnipotence are not com-
patible with one another, and are not consistent with other properties we also ascribe to god, such as
timelessness and immutability. If Kenny is right, no unique item has been picked out in the explicit
definition. Theists might then become atheists; or they could explore alternative “concepts” of god
(e.g., god as a creator of the universe, or god, in some sense, being identical to love, etc.). It would
then be up to atheists to show that they can remain atheistic toward each of these alternative con-
ceptions of god.
In what follows we will set aside the idea of explicit definitions of ‘god’ in terms of defining
properties and consider a quite different style of “definition”. This is a semantic model for introduc-
ing a name by means of a set of descriptive properties (see Lewis 1983). As an example, we might
ask, what does the term ‘Santa Claus’ refer to? One way of fixing a referent for it is via an associated
set of descriptions:
The referent for the name ‘Santa Claus’ = the (one and only) man with a big white beard
who dresses in red and each Christmas comes down chimneys to deliver presents.
Given such a description, does it actually pick out anyone as a referent of the name? If a person
really does these things, then a referent has been fixed; Santa Claus refers to the “something” that
is picked out by the description. Observing such a person is one way of telling if there is a referent
or not; but this is an epistemic matter and not an ontological/semantic matter of whether or not
some item actually satisfies the description, whether we know this or not. If no such person ever
does these things then the description picks out nobody; consequently no referent for the name
‘Santa Claus’ is specified. Finally, if two or more persons satisfy these descriptive properties, then
no unique person has been picked out. Perhaps we might allow for multiple referents here, but
clearly the uniqueness requirement has not been satisfied. Satisfying the uniqueness requirement is
important for the religious context of monotheism in which the term ‘god’ is being introduced.
Let us now consider some religious cases modeled on this example.
Already mentioned is the common, traditional conception of an OOO-god. A reference-fixing
definition of the following sort can be provided:
(R) The referent of the word ‘god’ = the unique item x such that [x is omniscient &
x is omnipotent & x is omnibenevolent].
Importantly, we have to add that one and only one “item” satisfies this conjunction. We say “item”
because we have no account of the essence or nature of god at this point. It is left open as to what
category god falls under, for example, person, substance, “thing”, creative power (e.g., Williams as
referenced in section 3), process, supernatural entity, and the like. But hopefully the conjunction
of three properties will give us a “handle” on how we can get to talk about god and learn more
about him and his nature (if he exists). This marks an important difference between: (a) explicit
definitions, which somehow are meant to tell us about the “nature” or “essence” of what is being
defined, or which specify necessary and sufficient conditions; (b) reference-fixing “definitions”,
which only attempt to put us referentially in touch with some item and may not necessarily tell
us anything about the “essence” or “nature” of what is picked out.
(R) fixes a referent for the term ‘god’ if there is such a unique x. If there is no x, or no unique
x, then no reference has been fixed. If the arguments given in Kenny (1979) are right, then no

Definition: Atheism
referent has been fixed for ‘god’ by (R). Whether there is such a referent is an ontological-semantic
matter; it is not an epistemic matter, though epistemic considerations are involved in our coming to
know whether there is a referent or not.
Saint Thomas Aquinas is famous for giving us Five Ways to prove that god exists (Thomas
Aquinas 1964, 13–17). His proofs attempt to show that some entity exists; and he ends his proofs
by saying something like “and this we call ‘God.’” Aquinas is sensitive to semantic issues concerning
what the word ‘god’ is about or what it refers to. Whether or not Aquinas adopted the semantic
model adopted here, we will take it that the word ‘god’ functions like a name in our language; it
purports to name something but may fail to do so, like ‘Santa Claus’, ‘Atlantis’, ‘phlogiston’, or
‘electromagnetic ether’ (on the reference of names see Sobel 2004, chap. 1, and Johnston 2009,
chap. 1). Each of Aquinas’s Five Ways can be understood to provide a proof that purports to show
that there exists a unique entity that satisfies a given description. So five different reference fixers are
given for the word ‘god’ (though we might well wonder whether all five fix a reference to the same
item). What are some of Aquinas’s ways of reference fixing?
Aquinas’s First Way is an argument based on the obvious fact that there is change. Along with this
fact and other premises about the nature of change, the conclusion of his argument is that there exists a
“prime mover”: “Hence one is bound to arrive at some first cause of change not itself being changed by
anything, and this is what everybody understands by God” (Thomas Aquinas 1964, 15). The main
focus here is on how a referent is given to the word ‘god’. We can set this out as follows:
(1): Referent for the word ‘god’ = the unique something that is the first cause of change not in
itself being changed by anything (viz., a prime mover).
This meets the formal requirements for providing a referent for ‘god’. And it is done by means of a
description that specifies a property that god allegedly possesses, namely, being a prime mover. But,
alas, the proof of the existence of such a prime mover is flawed. No such item, the prime mover, has
been shown to exist; so no referent is given for ‘god’. (There are many criticisms of Aquinas’s argu-
ment here, in particular Kenny 1969, chap. 2.)
Aquinas has four other “proofs” that there is a god, the results of which can be construed along
the lines of the semantic model given above. For brevity we can state the reference-fixing definitions
as follows:
(2): Referent for the word ‘god’ = there is a unique x such that x is the first (efficient) cause of
the ordered series of causes. (This is based on a first-cause argument.)
(3): Referent for ‘god’ = the unique something that must be, and owes this to no other thing
than itself. (That is, there is a unique necessary being.)
(4): Referent for ‘god’ = there is a unique something that causes in all other things their being,
their goodness, and whatever other perfection they have. (That is, there is a unique,
sustaining cause.)
(5): Referent for ‘god’ = the unique something that has understanding (intelligens) and directs
everything in nature to its goal. (This turns on a version of the argument from design.)
The five reference fixers indicate how we might attach a referent to the name ‘god’. Supposing
that they are successful reference fixers, can we be assured that the same item is picked out in each
case? Aquinas offers no proof of this and seems to assume it. After all, it is quite possible that five
(or four, or three, or two) different entities are picked out. The traditional assumption is that there
is not a committee of five things, one as prime mover, another as the unique necessary being, and so
on, that are picked out. Rather the one and only one item, god, is picked out by all five reference
fixers.
At the completion of his proofs, Aquinas says that though we have “recognized that a certain
thing exists, we have still to investigate the way in which it exists” (Thomas Aquinas 1964, 19).
On the first part of this, the consensus is that no “certain thing” has been proved to exist (again,
see Kenny 1969, which devotes a chapter to each of the Five Ways and what is wrong with each).
Definition: Atheism
Much of the rest of the second volume of Summa theologiae is devoted to showing in what way
god exists, for example, god has the properties of being simple, perfect, good, limitless, existing in all
things, unchangeable, eternal, one (has a unity), and supreme. How well he is able to establish these
claims is not a matter we need go into here. But note that they are not employed in any referent-
fixer but are shown later to be part of the “nature” of god—and we have fixed a reference to god.
This goes well beyond the bare existence claims that are supposed to be established in the Five
Ways. So, unlike the more mystical religious believers, for Aquinas two things are established:
(i) god exists; and (ii) we can say something about the properties he possesses (even though we do
not have the full story). But if the proofs in the Five Ways fail, as most critics recognize, then no
item is picked out and the project to establish (i) fails.
Of course Aquinas’s Five Ways are not the only ways in which attempts have been made to
attach a referent to the word ‘god’. The history of theology is replete with proposals, a few of which
can be set out in the following reference-fixing definitions.
Saint Anselm gave us what is commonly called “the ontological argument for god’s existence”.
We can use this to fix a referent in terms of what we are able to conceive (a matter that would be
extrinsic to god):
(6) Referent for ‘god’ = the unique x such that x is something than which a greater cannot be
conceived by us.
Terry Eagleton considers two strands of traditional theology, one from the “something-rather-
than-nothing” issue and the other from a version of the cosmological argument about god’s love
providing sustaining causes for existents (2009, 7). (This is akin to Aquinas’s reference fixer [4]).
From this Eagleton provides a “reference fixer” such as:
(7) Referent for ‘god’ = the unique thing that provides a reason why there is something rather
than nothing, and that sustains, through love, the condition of the possibility of any entity
whatever.
Finally, Mark Johnston tells about the “the Highest One,” which yields the following kind
of reference fixer for the word ‘god’ in which “the familiar conceptions of God are purged of their
idolatrous elements” (2009, 158):
(8) Referent for ‘god’ = “the unique item which is such there is outpouring of Being by way of
its exemplification in ordinary existents for the sake of the self-disclosure of Being.”
(Johnston 2009, 158)
Note that these are referent-fixing devices; none of them needs necessarily be taken to be an expli-
cation of concepts of god (though some can be transformed into that).
The last reference fixer, like much in twentieth-century theology, takes us well away from tra-
ditional conceptions of god. Commonly, when atheists and theists focus on arguments for or against
the existence of god, a lot of effort is expended on just the traditional concept of the OOO-god.
But how many other arguments, or concepts, might there be to consider? A satirical novel by
Rebecca Goldstein (2010) is titled 36 Arguments for the Existence of God: A Work of Fiction. In an
appendix, 36 arguments for the existence of god are set out (more can be found elsewhere!). If so,
atheists have a harder job to defend atheism than theists have to defend theism. They have to show
that each of the thirty-six theistic arguments for the existence of god fails—unless, of course, they
can come up with a “global” argument that sweeps away all theistic considerations (see Diller
2016 and Draper 2017). In contrast, theists, to defeat atheism, need only establish one agreed-on
successful argument among the thirty-six.
What about considerations based on different conceptions of god? To simplify matters for the
sake of the argument, let us suppose there are nine different extant conceptions of god with which
to contend. (There may be no exact number of concepts of god; the number may be open-ended with
some as yet unrealized concepts.) Let Ci stand for a conception of god (i = 1, 2, …, 9), and ‘(9!x)’ is
the uniqueness operator ‘there is a unique x such that …’. Following Jeanine Diller (2016), we can

Definition: Atheism
speak of a local theism in which some concept C of god is said to be exemplified, (9!x)Cx, and a local
atheism in which that concept is denied to have any exemplification at all, not-(9!x)Cx. Granted that
there are several extant conceptions of god (say nine, as supposed here), then global theism is the fol-
lowing doctrine in which at least one of the various concepts is satisfied by an actual god:
Global Theism: Either (9!x)C1x or (9!x)C2x or … or (9!x)C9x.
Now, if there is a god that satisfies one of these concepts, say C5, then the doctrine of theism holds.
But note that a theist must then become an atheist with respect to all the other eight concepts; these
are not satisfied by any god at all. The doctrine of global atheism can be expressed as:
Global Atheism = not-(Global Theism) = not-{(9!x)C1x or (9!x)C2x or …
or (9!x)C9x} {not-(9!x)C1x & not-(9!x)C2x & … & not-(9!x)C9x}.
Here again the work of the atheist is cut out in having to show that all the different concepts of god
are empty—unless, of course, there is some global argument for atheism that an atheist can employ to
sweep away all the different conceptions at once. But the theist also has a job of work to do, nearly as
big as the atheist’s. If there is just one conception of god that is correct, say C5, then the theist will
have to become an atheist with respect to all the other eight conceptions of god. For example, they
might have initially thought that the traditional OOO-god was satisfactory until they accepted argu-
ments against it and adopted some other conception of god. But this is tantamount to becoming an
atheist with respect to the OOO-god. And so on for other concepts that are rejected—except the
one remaining accepted concept. The task for the atheist is now only slightly larger; they have to show
that the one remaining concept the theist accepts is also to be rejected.
There is a useful parallel with the problem noted by Williams at the end of section 3. The issue
there, however, concerned current alternative conceptions of god in the various extant world reli-
gions. Each believer in each religion is theistic toward the god of their religion (of course), but each
believer is atheistic toward the god(s) of other religions. The theist is an atheist toward rival gods;
for the atheist, all such gods must go.
Bibliography
Baggini, Julian. Atheism: A Very Short Introduction. Oxford: Oxford University Drange, Theodore M. “Atheism, Agnosticism, Noncognitivism.” The Secular
Press, 2003. Web. 1998. http://www.infidels.org/library/modern/theodore_drange
/definition.html.
Blackburn, Simon. How to Read Hume. London: Granta, 2008.
Draper, Paul. “Atheism and Agnosticism.” In Stanford Encyclopedia of
Blackburn, Simon. “Divine Irony.” Times Higher Education, 5 March Philosophy, edited by Edward N. Zalta. Fall 2017. https://plato
2009. .stanford.edu/archives/fall2017/entries/atheism-agnosticism/.
Bradley, Darren. A Critical Introduction to Formal Epistemology. London: Eagleton, Terry. Reason, Faith, and Revolution: Reflections on the God Debate.
Bloomsbury, 2015. New Haven, CT: Yale University Press, 2009.
Bullivant, Stephen. “Defining ‘Atheism.’” In The Oxford Handbook of Goldstein, Rebecca Newberger. 36 Arguments for the Existence of God: A
Atheism, edited by Stephen Bullivant and Michael Ruse, 11–20. Oxford: Work of Fiction. New York: Pantheon, 2010.
Oxford University Press, 2013.
Guthrie, W. K. C. The Sophists. London: Cambridge University Press, 1971.
Chattopadhyaya, Debiprasad. Lokāyata: A Study in Ancient Indian
Hitchens, Christopher. Letters to a Young Contrarian. New York: Basic
Materialism. 3rd ed. New Delhi: People’s Publishing House, 1973.
Books, 2001.
Chattopadhyaya, Debiprasad. Indian Atheism. New Delhi: People’s Hitchens, Christopher, ed. The Portable Atheist: Essential Readings for the
Publishing House, 1980. Nonbeliever. Philadelphia: Da Capo Press, 2007.
Crane, Tim. The Meaning of Belief: Religion from an Atheist’s Point of View. Johnston, Mark. Saving God: Religion after Idolatry. Princeton, NJ: Princeton
Cambridge, MA: Harvard University Press, 2017. University Press, 2009.
Diller, Jeanine. “Global and Local Atheisms.” International Journal for the Kenny, Anthony. The Five Ways: St. Thomas Aquinas’ Proofs of God’s
Philosophy of Religion 79, no. 1 (2016): 7–18. Existence. London: Routledge and Kegan Paul, 1969.
Drachmann, Anders Björn. Atheism in Pagan Antiquity. London: Gyldendal, Kenny, Anthony. The God of the Philosophers. Oxford: Clarendon; New
1922. York: Oxford University Press, 1979.
Definition: Atheism
Lewis, David. “How to Define Theoretical Terms.” In Philosophical Russell, Bertrand. “Is There a God?” (1952.) In The Collected Papers
Papers. Vol. 1, 78–95. New York: Oxford University Press, 1983. of Bertrand Russell. Vol. 11: Last Philosophical Testament 1943–68,
edited by John G. Slater and Peter Köllner. London: Routledge, 1997.
Martin, Michael. “Atheism and Religion.” In The Cambridge Companion to
Atheism, edited by Michael Martin, 217–232. Cambridge, UK: Schopenhauer, Arthur. Parerga and Paralipomena: Short Philosophical
Cambridge University Press, 2007a. Essays. (1851.) 2 vols. Translated by E. F. J. Payne. Oxford: Clarendon,
1974.
Martin, Michael, ed. The Cambridge Companion to Atheism. Cambridge,
Sen, Amartya. The Argumentative Indian. London: Penguin, 2006.
UK: Cambridge University Press, 2007b.
Sobel, Jordan Howard. Logic and Theism: Arguments for and against Beliefs in
Nagel, Thomas. Last Word. New York: Oxford University Press, 1997. God. Cambridge, UK: Cambridge University Press, 2004.
Pinker, Steven. Enlightenment Now: The Case for Reason, Science, Humanism, Thomas Aquinas. Summa theologiae. Vol. 2: Existence and Nature of God
and Progress. New York: Viking, 2018. (1a.2–11). Translated by Timothy McDermott. London: Blackfriars and
Eyre and Spottiswoode, 1964.
Plato. The Last Days of Socrates. Translated by Hugh Tredennick and Harold
Whitmarsh, Tim. Battling the Gods: Atheism in the Ancient World. New York:
Tarrant. London: Penguin, 1993.
Knopf, 2015.
Rhees, Rush, ed. Recollections of Wittgenstein. Oxford: Oxford University Williams, Rowan. “Analysing Atheism: Unbelief and the World of Faiths.”
Press, 1984. Chap. 23 in Faith in the Public Square. London: Bloomsbury, 2012.

TOPIC 2
Method: Theism
Robert Audi
Professor, Department of Philosophy
University of Notre Dame, IN
This chapter presents many of the main issues that must be understood to arrive at an appropriate method
for appraising the rationality of theistic worldviews. It outlines several conceptions of theism; it explores the
kinds of evidences possible for it and compares those with the kinds appropriate to confirming scientific
theories; and it specifies a range of positive attitudes, such as faith and hope, that theists may have regarding
the existence of God. The chapter considers both the kind and degree of rationality of theistic attitudes and
the need for rationality in actions based on those attitudes. Both theistic attitudes and certain actions based
on them are shown to be important, and their rationality is also shown to be both a highly complex matter
and a status that is not ruled out on methodological grounds.
THICK THEISM AS THE FOCUS OF CONCERN
Both theists and nontheists care about whether theism is rational. When the question of its rational-
ity arises, it is usually directed toward a set of theological propositions, notably including that God
exists. It is not obvious just what propositions belong under theism. Moreover, theism, when part
of a religion, is commonly a life stance, a position determined both by cognition, typically belief,
and by practical elements, typically motivated behavior. Intellectual theism may be just a matter
of accepting certain propositions. Call this thin theism. By contrast, global theism is a worldview
and is by comparison full-blooded. These terms need much clarification if we are to see how to
describe a reasonable methodology for appraising theism. My procedure will be to begin with defin-
ing roughly what each kind of theism is and to discuss its rational appraisal, first in relation to the
intellectual case, and later in relation to the full-blooded, global case.
THE DEFINITION PROBLEM
Theism may be variously defined. Commonly, religious people consider themselves theists, and
many others conceive theists as having a religion. But we must distinguish between theism and reli-
gion. One could be a theist on intellectual grounds and not religious, certainly not in the sense of
having a theistic religion. There are also nontheistic religions, such as some forms of Buddhism.
Being religious, then, does not entail being a theist.
SOME KINDS OF THEISM

For most discussions of theism, it is desirable to think not only about versions usually accepted by reli-
gious people but also about the versions most often discussed by philosophers, theologians, and others
concerned with intellectual matters and related questions about education, democratic politics, and the
“ethics of belief”—the normative standards bearing on the critical appraisal and, so far as is in our
power, the regulation of our beliefs. Here the paradigm of theism is perfect being theology: the position
35
Method: Theism
that God is omniscient, omnipotent, and omnibenevolent—

KEY CONCEPTS omnicompetent, for short. Omniscience entails knowing all know-
able truths (some philosophers allow for unknowable truths, at
least where future contingency is in question; see Audi 2011,
Evidence
chap. 9). Omnipotence entails ability to do anything metaphysi-
Existence of God cally possible (loosely, possible within the bounds of logic).
Faith Omnibenevolence entails being perfectly good.
Methodological Naturalism Even quite rich forms of theism need not embody perfect
being theology. For process theologians, God is not omnipo-
Open Theism tent. Process theology might be combined with a version of
Perfect Being Theology open theism, which may take omniscience not to extend to what
Practical Principles free beings would do under all circumstances or even to their
actual future free actions. Neither qualification of perfect being
Process Theology theology necessarily undermines the view that God created the
Rationality universe, nor even the view that God created it “from nothing”
Scientific Method (ex nihilo).
A global theistic view, such as that of Christianity, Judaism,
or Islam, will have one or another kind of creation narrative. A
creation narrative may not be as elaborate as Genesis is. But it
is characteristic of theistic positions to take God to have not only power of the magnitude required
for creation of the universe but also sovereignty over it. This entails the ability to alter it in any possi-
ble way and, presumably, to destroy it, at least if that is compatible with the divine nature. Theists do
not take omniscience to imply God’s having ability to act contrary to the divine nature. Notice, too,
that (divine) sovereignty does not entail that divine creation is “from nothing.” Sovereignty entails
the universe’s dependence on God; but negative dependence, as implying nonexistence without God’s
“permission”—does not imply positive dependence. The latter would entail that God must do
something to prevent the universe’s ceasing to exist. René Descartes and others have affirmed such a
dependence: God must realize continuous creation: a sustenance of the universe without which it would
run out of ontological momentum and cease existing.
The distinction between negative and positive dependence on God is important for understand-
ing the theist’s intellectual options. Suppose one takes continuous creation to be the sustaining rela-
tion of God to the universe and considers it an element in sovereignty. One might then hold that
the universe has always fully depended on God (positively and negatively), but was never created in
time, much less from nothing. Since creation from nothing is widely felt to be mysterious, positive
dependence provides an apparently less problematic option for theists. To maintain it, one must give
up what could be called a Newtonian metaphysical paradigm. There, just as we need not explain why
a body at rest unaffected by another body remains at rest, we may take the universe to have a coun-
terpart kind of ontological inertia: it continues existing unless destroyed by an outside force. By con-
trast, one might take the passage of time to be destructive, if only by progressive corrosion; and, unlike
a body in motion that continues if not retarded by any external force, it ceases to exist if not sustained.
The Newtonian paradigm is more attractive to modern thinkers, but the Cartesian idea remains an
intellectual option for certain kinds of theism. Proponents could affirm divine sovereignty as including
both positive and total negative dependence of the universe on God and avoid positing creation from
nothing (though that is compatible with continuous positive dependence).
NORMATIVE ASPECTS OF THEISM

So far, nothing has been said about whether theism has normative implications, for example, impli-
cations concerning intrinsic value. But if God is perfectly good, it is highly plausible to hold that
God would not create beings with the human capacity for joy and suffering without caring about
their well-being (a normative notion) and indeed about how they treat one another, thus about
the right. Reflective theists see this and, in addition, hold that the notion of the goodness of God
is itself normative—the reference is to intrinsic goodness, as opposed to instrumental goodness,
which is only derivatively normative: what is instrumentally good is a means to something further
Method: Theism
(and ultimately, if vicious infinite regress is to be avoided, to something intrinsically good). The
intrinsically good is a source of normative reasons for action such as reasons to preserve it, bring
it about, or honor it. Moreover, the concept of a good person, including an omnibenevolent God
(at least conceived agentially, as in the Bible), implies tendencies to respond to such reasons.
Here theists have at least two major options. Following Thomas Aquinas and, more broadly,
the later rationalist tradition, they may hold that basic moral principles are necessary and knowable
a priori—roughly, through an adequate understanding of their content. Since, for Aquinas and
others, necessary truths are not alterable but also no limitation on omnipotence, reverence for
God allows taking knowledge of such truths to constitute knowledge of elements in the divine
mind, which of course clearly apprehends such truths. Furthermore, omnibenevolence implies that
God would want human agents to live in accord with true principles of obligation, such as the prin-
ciples requiring noninjury of others, fidelity to promises, and veracity (see Audi 2011, chap. 9).
A different option for theists is a divine command ethics on which sound moral principles are based
on divine commands. Illustrative principles might be versions of the ethical content of the Ten
Commandments. On what seems the best-developed version of a divine command ethics, however,
the commands that ground morality are viewed as those of a loving God. The notion of being loving
implies the applicability of such normative notions as caring about the good of the loved beings.
That implication, then, may enable proponents of the view to explain how malice and rape, for
example, cannot be divinely commanded (R. Adams 2000). On either approach, a global theism
need not be objectionable as implying normative falsehoods.
One further point is needed before we consider other evidential matters. In characterizing the-
ism intellectually, one might miss something central for theism as embodied in a religion—it might
be called worshipworthiness or, perhaps, reverenceworthiness. Theism characteristically includes this
element. What I have said so far makes room for this characteristic of God but also provides for a
conception of theism that allows for avoiding intellectual problems that, left unsolved, may prevent
those conscious of them from even seriously considering the possible benefits of adopting (or retain-
ing) a theistic or, especially, a religiously theistic worldview. Here creation narratives, parables,
stories illustrating religious commitments in their best forms, should be borne in mind. These are
generally not by themselves, however, evidence for God’s existence.
EVIDENTIAL DIMENSIONS OF THEISTIC APPRAISAL
The question what constitutes evidence is much more difficult than one might think. Evidence may
be constituted by such things as chemical traces, testimony as assertive speech, propositions in the
abstract, and such experiences as visual or auditory representations of our environment—a kind of
psychological condition compatible with illusion or even hallucination (Conee and Feldman 2004;
Plantinga 2000). We can simplify our discussion of evidence if we take evidence to be expressible
in propositions that describe objects, actions, or experiences presumptively supportive of what they
are represented as evidence for, say a miracle. Testimonial evidence for a proposition, p, then, is one
or more propositions attested to as supporting p. The evidence of the senses, for example, visually
experiencing a person’s walk on water, is expressible in a proposition reporting the experience, for
instance that it looked to me exactly as if he walked on the sea.
It is generally assumed that only meaningful claims are even candidates for being evidenced or
true. Many have thought that since theism is a view about reality, it should meet the criteria for mean-
ingfulness and testability that claims to scientific knowledge must meet. This is an important matter,
and we should explore it with a view to seeing whether a scientific habit of mind—which many theists
claim is possible for them—is compatible with theistic commitments. If so, this supports viewing
theism as an intellectual option in the contemporary world. If not, we should see why not.
TESTABILITY
First, it is widely believed that scientific hypotheses (indeed any statement belonging to science)
must be observationally testable, in the sense that in principle some observations could confirm or

Method: Theism
disconfirm it. Scientists tend to demand that proposed hypotheses be testable at least in that those
competent in the relevant field can determine what kinds of empirically possible observations would
tend to confirm or disconfirm it.
Testability might seem to entail something further: falsifiability. There is such a thing as falsifi-
cation of a hypothesis to the satisfaction of competent judges; but falsifiability is not equivalent to
testability. Indeed, apart from special exceptions unlikely to be important for understanding scien-
tific method, existential claims (those asserting the existence of something) can be verified and dis-
confirmed (hence tested), but not falsified. Consider the proposition that human beings exist else-
where in our galaxy. This is in principle verifiable but is not falsifiable. Experiments could
disconfirm it, and well-confirmed theoretical grounds could render it improbable, but those are dif-
ferent points. Similarly, logically speaking, universal propositions (of a scientific kind), for example,
that all metals conduct electricity, can be falsified (by counterexample) but not verified—in the
strong sense entailing demonstration. Theists have held that theism is not falsifiable but is verifiable
by resurrectional experiences (Hick 1990).
Consider also what counts as observability. Take a psychological hypothesis: that if I mentally
recite certain lines from Macbeth, I will next have a mental image of the dagger Macbeth halluci-
nated. I could test this mentally. Contrary to the usual assumption that observability must be
through the five senses, here we have observability in the sense of (replicable) experienceability. Are
such tests unscientific? I doubt that. In any case, there is something essential for understanding sci-
entific inquiry that the example does not challenge, namely, that the confirmatory relevance of expe-
rience is crucial for scientific status. As we shall see, theists may hold that the relevant standard is
met by theism.
PUBLIC OBSERVABILITY
I have assumed both that mental phenomena are of scientific interest and that our mental lives are
in a way private. To see how the latter assumption squares with the idea that science is “public,” we
might generalize our psychological hypothesis: Anyone (with my background) who has the relevant
sequence of thoughts will have the same or a similar image. This generalization meets a publicity test:
others could run the experiment and give publicly observable testimony as to its outcome. Public
observability is in fact the second element in a standard conception of scientific method. In part,
this requirement is needed both to rule out epistemically prejudicial idiosyncrasy and to accommo-
date the cooperative character of science. If I discover a connection between silent utterance of cer-
tain words and forming certain mental images, this should be a publicly testable result even if each
person seeking confirmation must ascertain it internally.
It should be granted that if reports of the results could not be publicly produced and made acces-
sible to competent observers, we would not regard the experiment as meeting the publicity condi-
tion. Some might not even take it to provide evidence. It may be true that we cannot directly verify
someone’s report of an inner experience, such as imaging a person; but even to check on someone’s
perceptions of what is experienced in laboratories (say, the color of precipitates) we rely on other
perceptions, if only perceptions of testimony. Here, to determine properties of what is external we
need “private” sensory experiences no less interior than mental images. Publicity implies a kind of
intersubjectivity. But intersubjectivity does not require exclusive reliance on nonmental phenomena.
Religious experiences, then, are not ruled out as evidence by the public observability criterion.
EMPIRICALITY
A third element important in understanding scientific method is its restriction to empirical matters.
Arguably, this restriction follows from testability. Perhaps, but it does not obviously follow. The
idea underlying the restriction is, in part, to set aside logic and pure mathematics. These are non-
empirical but not unscientific. However competent in these fields some scientists may have to be,
the claims of logic and mathematics are not usually considered the kinds of things that science
investigates (Audi 2000, chap. 4). A major contrast here is between the empirical and the a priori.
At one time many would have said, following Immanuel Kant, that the a priori is equivalent to
the necessary and hence science must be concerned with what is both empirical and contingent.
Method: Theism
It seems clear, however, that what is necessary—say, that water is H2O—need not be a priori
(Kripke 1980). A necessary truth, then, can be discovered empirically, even if the modal fact that
it is necessary is not itself empirical.
Suppose that necessary propositions are not beyond scientific confirmation. Why might certain
necessary theological propositions not be scientifically confirmable? That God’s existence is empiri-
cally confirmable has been held by proponents of the argument from design (roughly, the argument
from the order in nature to the existence of God as needed to explain it—or at least as what best
explains it) but may also be considered confirmable by what it might predict about future discover-
ies concerning fossils or other evidences from the past. To be sure, theism as such does not entail
that the existence of God is necessary. In that case, an empiricality requirement alone would be
plainly compatible with the scientific confirmability of theism.
FALLIBILISM AND THE DEMARCATION PROBLEM

In addition to testability, public observability, and empiricality, fallibilism as a procedural attitude
has also been thought essential to scientific method. I call it procedural because the point is about
scientific practice and the epistemic claims made in assessing it. The actual psychological disposition
of scientists engaging in the practice is of secondary importance. In my view, procedural fallibilism
is better taken as a requirement for the proper use of scientific method than as definitory of scientific
method itself. A dogmatic person could scientifically confirm hypotheses. But, given the fallibility of
both the procedures and their users, the person would have an inappropriate attitude. If fallibilism
as a procedural attitude is above all (roughly) an openness to the possibility of one’s being mistaken
even in what seems obvious or established, it is indeed necessary for a scientific attitude. But this
kind of fallibilism is also appropriate in theology and other fields.
If we now note that religious experiences can be both predicted from theological assertions and
also experientially confirmed, we can see that the kinds of demarcation criteria we have been consid-
ering do not strictly show that no theological hypothesis is experimentally, or at least observation-
ally, testable. To be sure, theologians and lay religious people rarely stake their faith on predictions.
But consider a religiously significant prediction: ardent believers who, when feeling anxious, pray
wholeheartedly will more often than not feel better immediately afterward. This can actually be con-
firmed or disconfirmed. Suppose it is true. Skeptics will likely say that the explanation of the pattern
lies in secular psychology. Suppose this is so. It should be balanced by the point that it is common
for clearly scientific data (and hypotheses) to be explainable in more than one way; so, just as differ-
ent scientists may retain conflicting theories to explain data they agree on, theists may claim that the
confirmation of this hypothesis provides some evidence for divine action in the world, and that a
multitude of such evidences can yield a cumulative case presenting strong evidence for theism.
Suppose we set aside these difficulties of analysis and conceive scientific method in relation to
the three criteria of testability, public observability, and empiricality. We might then partly charac-
terize a scientific claim (a hypothesis or a full-scale theory) as one that can in principle be con-
firmed or disconfirmed using this method. The problem—the demarcation problem—is to distin-
guish scientific claims, in the broad sense of claims appropriately assessable by using scientific
method, from others, such as those of logic on one side and theology on another side. A great deal
has been written on this problem (George 2005). All I can do here is further indicate the difficulty
of ruling out certain kinds of claims, including some by creationists or intelligent design theorists,
as nonscientific.
It is commonly thought that theological claims are intrinsically untestable empirically (Hick
1990). But consider the claim that God will end a drought at noon. This cannot be established obser-
vationally, since the occurrence would only confirm the claim and leave open what or who caused the
rain. Yet it is empirically confirmable and, obviously, empirically disconfirmable. We might also ask
about human origins. Could someone confirm “God created the earth in 4004 BCE” if this implies
that we will find errors in, say, carbon dating and develop a new dating system that confirms this time
of origin? I doubt it, but the scenario is not inconceivable (Hick 1990).
The intelligent design view is more modest than creationism. In principle some version might
yield predictions about how human brains are constructed or how human beings would adjust to

Method: Theism
certain changes, predictions that could be confirmed or disconfirmed in, say, a lifetime (Dembski
and Ruse 2004). Suppose that such predictions are offered by intelligent design theorists. Some
critics would hold that the designer positions are not even clear enough for predictions of these sorts
to be said to follow from them (even with high probability), in which case their observationally
bearing out is not confirmatory. Even if this is so, there remains the idea that only what best (or
well) explains observable phenomena is thereby confirmed. This idea raises more questions than
can be pursued here. One is what counts as a good explanation; another is whether predictive power
is a condition for qualifying as one.
METHODOLOGICAL VERSUS METAPHYSICAL QUESTIONS

The elements of scientific method so far emphasized are methodological, not ontological. Neither
the testability nor the public observability nor the empirical character of scientific hypotheses entails
any substantive ontological conclusion, such as that only natural events can be causes or that there
cannot be an infinite chain of causes extending into the past. But it is commonly and plausibly held
that a commitment to scientific method presupposes a commitment to methodological naturalism:
roughly, the view that causes and explanations of natural phenomena should be sought in the natu-
ral world, paradigmatically in terms of what meets the other three criteria characterizing scientific
method (testability, public observability, and empiricality) and does not presuppose supernatural
agency. This view is no clearer than the notion of the natural, but a standard assumption is that
physical, chemical, and biological phenomena are natural and that God as understood in monotheis-
tic religions is supernatural.
Methodological naturalism neither affirms nor denies theism. In this, it is like liberal democracy,
which, as a sociopolitical framework, provides the cultural context of our discussion. Methodologi-
cal naturalism constrains scientific inquiry but implies no substantive conclusions in either science
or, especially, metaphysics. My formulation of methodological naturalism allows that scientific
inquiry might posit causal elements, such as the big bang, which theists may argue require super-
natural causes. The point is that, methodologically and epistemically, one would need no evidence
of any supernatural proposition in order to conduct scientific inquiry and confirm scientific
hypotheses. One might, then, qualify as using scientific method to understand the world even if
one takes the big bang to have a supernatural cause, provided one does not treat any supernatural
factor as essential in the content or confirmation of one’s scientific work. A methodological natu-
ralist might even hold that every natural event, whether mental or physical, has a natural cause
and even a sufficient naturalistic—indeed perhaps physicalistic—explanation, as well as a theistic
explanation, provided that the latter, supernaturalist claim is no part of one’s scientific account
(Kim 2005).
Given this conception of methodological naturalism, the best response to the demarcation
problem for purposes of this chapter is at least threefold. First, there is apparently no way to show
that theological hypotheses are intrinsically untestable by observation or experience. Second, suppos-
ing they are observationally testable, it does not follow that they constitute scientific hypotheses.
Third, we have seen no reason to think that theism has no testable implications or that some
hypotheses or claims about God are unfalsifiable. None of this shows, however, that theism is prop-
erly understood as a kind of scientific view. I have not explicated that difficult notion but have
instead argued, with scientific hypotheses in mind, that certain theistic propositions can in principle
also meet rather demanding criteria for meaningfulness, testability, and even potential acceptance on
the basis of evidence. This is an important result. It does, however, leave open many questions
about kinds and degrees of support for theism.
THEISTIC ARGUMENTATION
Suppose, then, that we imagine sophisticated theists expressing all their grounds for their theism in
propositional form. Might they then have good arguments? The notion of a good argument needs
comment. Here are a few points of basic logic.
Method: Theism
SOME TYPES OF ARGUMENTATION

Call an argument valid when there is no logical possibility of its premise(s) being true and its
conclusion false: the conclusion follows from the premise(s). Some would put it thus: there would
be a logical inconsistency in affirming the premises and denying the conclusion, for example,
granting that all As are Bs and all Bs are Cs, yet denying that all As are Cs. A sound argument
is valid and has true premises, hence a true conclusion. Since sound arguments may be circular,
we should not automatically consider them cogent, that is, such that the premises provide strong
reason for the conclusion. Proof is much harder to characterize. Usually, proofs are taken to be
sound arguments with premises that are at least beyond reasonable doubt. I have already sug-
gested that proof in this sense is not claimed by philosophers of science or (appropriately
informed) scientists even for well-established scientific hypotheses. What kinds of arguments
are appropriate for theism?
From the history of philosophical theology, one might think that proof of theism is generally
taken by theistic philosophers to be possible. This impression is created by Saint Anselm in present-
ing his ontological argument and by Aquinas, Descartes, and later philosophers in presenting their
arguments for the existence of God. But, particularly since Kant’s widely known case against the
ontological argument and his influential statement that he had to deny knowledge of God in order
to make room for faith, philosophers and theologians have taken the main evidential grounds for
theism to be experiential and probabilistic. Let me explain.
Here we need the notion of inductive arguments. These are, in broad terms, not “aimed” at
achieving undeniable certainty but (properly presented) at providing a high degree of probability
for their conclusion. We also need here the idea that religious experience might provide good induc-
tive grounds (not proof) of God’s existence. The point is not that such experience cannot be given
some kind of propositional expression; but its function, so far as grounding theism is concerned,
is to yield theistic belief and other positive theistic attitudes directly rather than by inference from
premises. A prominent case is perception: readers of this chapter do not infer that there is print
before them because they can argue from their visual experience to that point; they believe
this directly through visual experience. Might there be a case for “perceiving God”? (Alston 1991;
Plantinga 1993).
A second and prominent kind of inductive case for theism comes from the realm of abduction,
roughly, inference to the best explanation of what is seen as true—and, especially, as obvious. The
intuitive idea is that what can well explain facts is thereby confirmed. Examples are “fine-tuning”
arguments, on which what best explains certain empirical patterns that would be unexpected under
nontheistic assumptions is that they are divinely established.
One may also note that there could, logically speaking, have been nothing, and it may be
argued that we cannot explain why there is a universe without taking its existence to derive from
God (either by creation from nothing or at least by divine sovereignty). If it seems that the big bang
might serve here, note that it is a natural event; why it occurred is as much in need of explanation as
why there is anything at all. If God is a being that could not have failed to exist, this cosmological
why-question is considered answered by positing divine action as explaining why there is something
at all (Swinburne 2004). It should not be thought that theistic philosophers who use inductive
arguments consider abduction the only source of support for theism. They may also consider other
arguments cogent.
This last point raises the question whether the best case for theism might be through cumula-
tive case argumentation. One might think, for example, that the best explanation of why there are
so many plausible if individually inconclusive evidences for theism is that it is true in a form in
which the conception of God makes clear why God might have created a universe in which that
battery of confirming considerations is available: from deductive arguments to abductive ones to
religious experience. Cumulative case arguments may seem like appeals to coherence as a standard
of justification, but they are not. A pure coherence view—requiring joint consistency among the ele-
ments and some degree of mutual support—would be very permissive: one could hold any consis-
tent view with a high degree of internal “fittingness” among its elements, at least if it is not bettered
by a competing more coherent one.

Method: Theism
It should be granted that coherence is a justificatory standard that non-coherentist views respect
in evaluating a view given elements to start with that are both credible in some non-coherentist way
and support the view. Such starting points seem essential. A sheer fabrication, after all, can be mar-
velously coherent. Suppose, however, that the evidences of the senses and logical intuition are inputs
to a view in the first place. Then, with those grounded rungs of the ladder already ascended, the
overall coherence of that view may give it an advantage over competing views. To be sure, if coher-
ence is in good part a matter of “fit with the data” (say, of beliefs about our environment fitting our
sensory evidence), then the idea of confirmation by coherence is absorbed into what is usually con-
sidered a position opposing coherentism—the “foundationalist” position on which some evidences
(typically kinds that, like perceptions, seem psychologically inevitable) are properly basic, a kind of
foundation not needing defense by argument (Audi 2010, chap. 9).
SIMPLICITY AS A STANDARD OF APPRAISAL

One may wonder whether considerations of simplicity may support theism in some forms, in the
way it is usually taken to do regarding scientific theories. In this context theism should not be
viewed as simply the proposition that God exists, but as at least that together with an answer to
the cosmological question why there is something rather than nothing. Suppose now that theism
is not ruled out or rendered highly implausible by any considerations drawn from science—with
which in my view it does not compete—or from metaphysics. Isn’t some nontheistic position, such
as philosophical naturalism, a simpler worldview? Answering this requires some account of simplic-
ity, an elusive notion I cannot fully explicate here (Swinburne 2004; Swinburne 2001). Suppose,
however, as many writers on simplicity have, that the most important single factor determining
the degree of simplicity of a worldview is the number of basic assumptions and irreducible kinds
of entities it must posit. Then we might suppose for the sake of argument that there is some reason
to think that the naturalistic worldview is simpler. Assume that the important comparison—as most
likely to engage open-minded inquirers in the present age—is between a scientifically oriented
worldview that is theistic and a scientifically oriented one that is not. Still, in any plausible episte-
mology, simplicity is only one criterion of acceptability.
Suppose, then, that the naturalistic worldview is simpler than any plausible theism. It still pro-
vides no answer to the classical question why there is anything at all. And, for many lives, it does
not accommodate the most plausible explanations of certain religious experiences or, indeed, of
some people’s long-standing patterns of religious experience. Either of these points may well suffice
to outweigh even differences in simplicity greater than those that may exist between a classical theis-
tic worldview and naturalism. Methodological naturalism is not at issue here, since it is compatible
with theism.
Even on the intellectual side, then, there is reason to doubt that Ockham’s razor—understood
as the principle that in framing and adopting explanations and theories, we should prefer the sim-
pler, other things equal—cuts deeply against a classical theistic worldview as such. It certainly does
not justify the conclusion that such a worldview is not rational. For one thing, the principle does
not say that it is irrational to hold a view while aware that an alternative that is otherwise as plausi-
ble is simpler. Mere rational preferability of one view over another does not imply the irrationality,
as opposed to the lesser rational desirability, of holding the second.
So far, I have spoken as if it were clear what the status of the razor is (Collins 2006). But it is not.
Once this is seen, the bearing of the principle on the rationality of theism can be clarified. The prin-
ciple is generally considered epistemic, in the broad sense that it provides an evidential standard rele-
vant to assessing candidacy of a belief or even an entire worldview for justification or knowledge or
both. To assess this principle, we should distinguish internal from external epistemic considerations:
roughly, between justificatory elements—evidences, in a sense—that are accessible to reflection or con-
sciousness and, on the other hand, objectively truth-conducive factors that are not thus accessible. On
internal grounds, even a hallucination, as qualitatively like a genuine perception, can—despite its
unreliability—justify certain beliefs, for example, Macbeth’s belief that there is a dagger before him.
On external grounds, Macbeth could be justified in believing this by a kind of reliable blindsight,
despite its providing no accessible grounds, such as a visual experience, in consciousness.
Method: Theism
Is the simpler view better justified on internal grounds (other things equal)? This is far from
obvious. What may be obvious is the quite different principle that, other things equal, it is prefera-
ble to work with a simpler view in constructing a theory or in framing or retaining a worldview.
That, however, is a practical principle and could hold even if there were no difference in evidential
status between more and less simple views. Call it the principle of optimal efficiency to emphasize that
it opposes expending more effort than needed for one’s purpose. This principle may be self-evident,
but it does not entail, and I doubt that it even implies, the razor principle as stated epistemically.
A hypothesis that is easier to work with than a competitor need be no better evidenced.
Do we have reason to think that from an external perspective simpler views are more likely to
be true? Perhaps, for some plausible notion of simplicity, we do, given common experience and
the track record of scientific hypotheses. The intuitively simpler hypotheses have apparently tended
to be better confirmed (though this might be at best hard to show). I do not see how to demon-
strate this. We might entertain an evolutionary explanation—say, that since we do in fact prefer
simpler hypotheses, we probably would not have survived, at least in the way we have, if the truths
of nature were not better evidenced by simpler hypotheses. This point provides at best limited sup-
port to the razor principle, however, and it presupposes that we have perceptual knowledge or at
least true perceptual beliefs in the first place. In that case, the question whether simplicity enters
into those notions would have to be addressed before that justification could be considered suffi-
ciently independent of simplicity to support the razor without a vitiating circularity. Such inductive
reasoning in favor of the razor taken epistemically also provides only an empirical justification. I
propose simply to assume that nature is (other things equal) better described by simpler hypotheses
and thus to grant that the razor may have at least external evidential force. How might we explain
why—why nature’s ways are better evidenced by simpler hypotheses?
In the context of assessing the rationality of theism, impartial appraisers should not overlook
the possibility of a theistic explanation. Nature could be designed to be understood and predicted
by beings like us who prefer simpler hypotheses (Koons 2000, chap. 17). Theists need not claim
to know that this explanation is correct, but it cannot be simply ruled out of hand. A further point
is that there need be no conflict between a theistic worldview and a deeply scientific habit of mind.
Nature can be studied with as much scientific rigor by those who believe it is God’s creation as by
those who think it a fortuitous collection of particles. Some theists, moreover, might think it is reli-
giously fitting or even obligatory to seek a scientific understanding of the creation.
If, as some theistic philosophers hold, theism is simpler than (metaphysical) naturalism,
might it also have greater explanatory breadth and depth? I take the explanatory breadth of a given
proposition or theory to be roughly how much it explains; its explanatory depth is a matter of
how good its explanations are in providing understanding. Depth is metaphorical, but here models
come in. The kinetic-molecular theory of gases is often cited as giving a deeper and better expla-
nation of the behavior of heated gases than we get from simply noting that heating gases expands
them and using that law to explain why heated gases raise a piston in a closed cylinder. Moreover,
we human beings understand phenomena produced by intentional agency best of all—or we tend
to think we do.
It may be claimed, then, that divine ordinance of the laws of nature provides both a broad and
a deep explanation—at least given the idea that God as omnibenevolent wishes to create a world at
once intelligible to us and in principle hospitable to us. Here, however, debate continues between
those who find divine agential explanations in such cases satisfying and those who, perhaps pointing
to the absence of any “mechanism,” find them wanting. Do they really produce understanding in a
sense relevant to abduction or simplicity? And how, it may be asked, do we assign probabilities to
cosmic hypotheses, where we have neither direct evidence nor relative frequencies to justify the
assignments? Understanding is not a psychological matter—or not just that; it must be distin-
guished even from the natural sense of no further need to ask why.
This is an appropriate point to recall that there are degrees of credence appropriate in different
circumstances and roughly proportional to the strength of the evidence. How high need our cre-
dence be for theism to warrant accepting it? Here we find that theism may be held with different
degrees of confidence and, in that way as with scientific hypotheses, considered well worth working

Method: Theism
with over time even without decisive evidence, much less proof. This point suffices to refute the ste-
reotype of theism as simply a matter of faith conceived as unresponsive to negative and positive evi-
dences (Buchak 2012).
THE INTELLECTUAL RESPONSIBILITY OR “ETHICS OF BELIEF” QUESTION
If theism cannot be decisively either ruled out or established, and if that is or can be true also of
scientific hypotheses, one might think that the question whether theism can be rationally held is set-
tled. But parity does not follow: that theism cannot be ruled out as meaningless or beyond confir-
mation or rational argument does not suffice to show it to be a candidate for a rational posi-
tion—even at a low level of credence. Moreover, because theism is generally important both for
those who hold it and for others—many of whom it profoundly affects—it should not be held with
indifference to grounds for it of any kind. Many would go further and say that it is subject to a bur-
den of proof. If “proof” is taken seriously, this is an unreasonable demand for both theism and scien-
tific hypotheses. But where, for example, theists call for exceptions to legal obligations such as mili-
tary conscription, nontheists understandably tend to feel that there is a burden of justification.
Many theistic philosophers and theologians would not claim to have even sound deductive argu-
ments for theism, much less proof. Nor would many of them claim that religious experience alone
justifies a full-scale theistic view—one positing an omnicompetent God or anything approximating
that high ideal. Some few may appeal to the idea that there are practical reasons for theism—reasons
such as that being a theist makes life happier or even, as Blaise Pascal thought, makes eternal blessed-
ness possible. Here I can say only that the rationality of causing oneself to believe or continue believing
something does not imply that the belief is rational. There is disagreement about this, but it need not
be detailed here given that theists have not in general resorted to such arguments for theism (Audi
2011, chap. 5).
Understanding the intellectual responsibility question is best pursued by considering not just
whether theism is rationally acceptable as a body of propositions but also whether, in general terms,
a globally rational person can accept it and live by a global theism. A rational person must not only
have a belief system of a certain kind—for instance, reflecting responsiveness to evidence—but must
also meet certain practical standards. Given my life experience, I could not rationally fail to believe fire
will burn me; but if, believing that, I do not try to avoid its excruciating effects, then the rationality of
my belief will not suffice for my overall rationality as a person. I will often need protection from my
own practical irrationality. The point that rationality has both cognitive and behavioral sides is crucial
here because one reason for the great importance of the question whether theism can be rational—or
as rational as atheism—is that accepting theism tends to affect behavior in major ways.
In appraising a worldview or a body of propositions (or “beliefs”) constituting a significant part
of a worldview, we should ask both what supporting grounds there are for the truth of the proposi-
tion(s) and whether the actions we would tend to take if we fully accept the proposition(s) are ratio-
nal. These are different but related questions. We should not take the fact that holding a view will
lead us to do good things as evidence for its truth, but we should certainly take the fact that holding
it will lead us to do bad things as a reason to scrutinize purported evidential grounds for it and,
especially if it is insufficiently justified, then to resist doing bad things on the basis of it. History
shows that terrible things have been done in the name of certain theistic beliefs, and that partly
explains why the question whether there are adequate grounds for theism is widely considered
important. In those cases, of course, theism is part of a religion, as it need not be. But given how
natural it is for accepting theism to be combined with religious practice, we should widen our focus
to include the rationality of theism as embodied in commitment to a religion. Here I must be quite
brief, but the aim is only to outline the issue to be resolved, not to resolve it. My aim will be simply
to indicate—in a way that is methodologically acceptable to theists and atheists alike—how we
might approach the question whether theism may be rationally held by persons who are globally
rational both in their cognitive lives and in their behavioral practices, especially as affecting others
(Audi 2011).
Method: Theism
THEISM AND MORAL RESPONSIBILITY FOR ACTION
Whether we are morally responsible for an action depends in part on the rationality of our belief(s)
underlying it. Thus, when it comes to the moral implications of a commitment to theism as
engendering action, the degree of rational grounding a person may have for theism is important.
In exploring this, we should consider not just perfect being theology but also qualified forms of
theism for which the problem of evil—widely taken to be a serious objection to theism—is less
serious (M. Adams 2000; R. Adams 2000; Adams and Adams 1990; Wainwright 2005; Audi
2011, chap. 9).
RATIONALITY AS RELATIVE TO GROUNDS

Different people or even the same people at different times have differing degrees of evidence (or
counterevidence) for theism, in the widest sense of “evidence,” in which it includes religious experi-
ences in which a supernatural being is felt to be communicating to one. Some theistic philosophers
have both religious experiences and a mastery of the best available arguments for God’s existence.
Other theists simply accept the theism of their religion without question. Many are in between
these two categories. We cannot here pursue the rationality status of testimony-based beliefs, but I
would argue, as have many, that it can be rational to believe a proposition on testimony without
independent grounds for it and that even a long testimonial chain—such as one tracing long
into the past (as in some scriptural cases)—need not be weaker than its weakest link (Audi 2010,
chap, 7).
May we say, then, that some theists rationally hold their view, even when they conceive God as
omnicompetent? With relativity to grounds in mind, I find it plausible to say that some theists do
rationally hold the view, and other theists do not. But these points need refinement. First, the rela-
tivity to grounds is situational, not epistemic: what grounds we have, in science as well as in religion,
depends on our experience, which varies with environments and times, but the support that
grounds give to the propositions they support—say, in terms of the degree of probability they con-
fer on those propositions—is not relative in this way and is objective. The second point is implicit.
Different degrees of support, whether from experience “directly” or from inductive or deductive
arguments, or, as is common, from combinations of these sources, justify different attitudes, such
as barely accepting a proposition versus holding it confidently. (Some would here speak of degrees
of belief, and that should be kept in mind as an alternative.)
THREE TYPES OF THEISTIC RESPONSES TO GROUNDS

In the light of these points, we can refine the issue: the question here—and in many theism-atheism
debates—is not just what possible grounds there are for theism in the abstract. The question is what
grounds a person may have for a particular theistic view and what theistic attitude theists may prop-
erly have given their grounds. Let me note three attitudes, taking theism as simply the proposition
that an omnicompetent God exists. Most common here is doxastic theism: belief that God exists. But
faith that God exists does not entail belief that this is so—though it is incompatible with believing it
false; non-doxastic faith is like trusting, for instance, that a friend will survive surgery where the
grounds are substantial but not conclusive (Howard-Snyder 2013; McKaughan 2013). Call this
fiduciary theism. Suppose, however, that one’s grounds for theism are weak. Even with very weak
grounds one may still rationally hope that God exists. This may be called aspirational theism, at least
where the hope, as is possible, is combined with attitudes and practices that give theism a place in
thought and action.
For sophisticated theistic philosophers, doxastic theism may be rational; for skeptically minded
religious people who lack a mastery of the issues surrounding the arguments for God’s existence and
the problem of evil, aspirational theism may be as strong a position as they may rationally take. For
them, the rationality of doxastic theism might be defeated by prima facie justification for believing
that the argument from evil renders God’s existence highly improbable. Granted, theistic hope com-
bined with this pessimistic probability belief would not be a normal position for anyone plausibly
called a theist. But although hope is a less strong attitude than faith—which, unlike hope, implies

Method: Theism
some confidence or at least trust—hope can be free of such negativity. It can also combine with
commitments to religious practices in a way that allows the person to lead a religious life.
The main reason to emphasize that even just hoping that God exists can be central in an
authentically religious life is to meet the objection that aspirational theism is trivial—a mere intellec-
tual acknowledgment that when one finds a view attractive and it has some rational support, one
may aspirationally hold it. Again, we should compare the scientific case. One might quite rationally
test a hypothesis that one only hopes is true, especially if acting in accordance with it is important.
An aspirational theist is likely to agree to something similar and may seek experiences or intellectual
considerations giving support to theism. A given person may or may not think these have been
found. Suppose they are not.
Here we come to a crucial point about the rationality of theism. Provided that theists do not
have attitudes incompatible with the evidences that exist in their own lives, may they hold their the-
ism rationally? If aspirational and fiduciary theism are countenanced, the answer seems to be yes, at
least for the aspirational theist whose hope coexists, as it well may, with an appropriate responsive-
ness to evidence. This point cannot be fully appreciated apart from an awareness of the behavioral
side of rationality. Objectors to theism would be much less concerned to undermine it if they
thought that holding a theistic views is a matter simply of isolated beliefs having no interpersonal
implications. But given that theism is normally embodied in certain religious commitments and
practices, the question of its rationality has moral and indeed political importance. Is there, then,
good reason to seek all the counterevidence one can find as a way of undermining a harmful doc-
trine?
On this matter, I would stress that theistic commitments—as distinct from the tenets of any
particular religion—need not lead to a tendency to be violent or to dominate others or even to be
undemocratic. I have already said that piety allows—as is also theologically plausible—taking certain
moral standards to apply to everyone and governing oneself by those in interpersonal relations. Sup-
pose I am a biomedical scientist and want to test a dangerous drug that I hope will work, even have
faith that it will work. I may rationally try it out on myself even with inconclusive evidence, but I
may not conscript others to try it, even though my hope of its success and my benevolent desires
are rational. Similarly, if my theism requires me to observe certain practices and make certain sacri-
fices for the common good, this fact may rationally support my doing that; but as a rational person I
should see that requiring the same behavior of others outside my theistic worldview is not justified
by my own commitments. For the coercion of others there must be reasons of a certain universally
recognizable kind (Audi 2000, 2013).
By way of conclusion, I would emphasize that there is not just one form of theism, and theists
and atheists alike should therefore carefully specify the version under discussion. Perfect being the-
ology is an attractive and arguably quite simple ideal, but it is not a philosophical straitjacket.
Nor is there just one body of evidence relevant to appraising theism. Competent inquirers can find
evidences, counterevidences, and responses to both. There is also the possibility of private experi-
ences having some confirming weight for some versions of theism. This point is widely realized but
also widely underemphasized. What is also underemphasized is the range of attitudes, both cognitive
and motivational, that theists may have in the light of their overall experience and reflections. These
attitudes may be as modest as hope or as strong as certainty. Rationality is much more readily
achieved in the former case than in the latter. Rationality, moreover, admits of degrees. This applies
to motivation and action as well as to cognition. A theistic commitment may or may not be cogni-
tively strong in the attitudinal dimensions ranging from hope to certainty, and it may or may not
tend to yield rational desires; but what counts as rational desire and indeed as rational action differs
considerably in the domain of religious observances by contrast with the public domain of interper-
sonal action. In that domain, rational theists may adhere to ethical standards of universal validity
and may, in their interpersonal actions, do little if anything that, quite apart from their theism, can-
not be justified by the kinds of reasons any rational person can respect. If there are objections to
theism in the light of which no version of it is rational, this will have to be shown on substantive
grounds; theism is not untenable on methodological grounds.
Method: Theism
Acknowledgments. This chapter was written initially under the title “Methodological Aspects of The-
istic Appraisal” and had numerous footnotes (containing qualifications or comments) not included here.
For helpful comments on earlier versions I heartily thank Nevin Climenhaga and René van Woudenberg.
Bibliography
Adams, Marilyn McCord. Horrendous Evils and the Goodness of God. Ithaca, Koons, Robert. Realism Regained: An Exact Theory of Causation, Teleology,
NY: Cornell University Press, 2000. and the Mind. Oxford: Oxford University Press, 2000.
Adams, Marilyn McCord, and Robert Merrihew Adams. The Problem of Evil. Kretzmann, Norman. “Evidence and Religious Belief.” In Philosophy of
Oxford: Oxford University Press, 1990. Religion: A Guide and Anthology, edited by Brian Davies, 95–107.
Oxford: Blackwell, 2000.
Adams, Robert Merrihew. Finite and Infinite Goods. Oxford: Oxford
University Press, 2000. Kripke, Saul A. Naming and Necessity. Cambridge, MA: Harvard University
Press, 1980.
Alston, William P. Perceiving God: The Epistemology of Religious Experience.
Ithaca, NY: Cornell University Press, 1991. McKaughan, Daniel. “Authentic Faith and Acknowledged Risk: Dissol-
ving the Problem of Faith and Reason.” Religious Studies 49 (2013):
Audi, Robert. Religious Commitment and Secular Reason. Cambridge, UK:
101–124.
Cambridge University Press, 2000.
Murphy, Mark. Natural Law and Practical Rationality. Cambridge, UK:
Audi, Robert. Epistemology. 3rd ed. London: Routledge, 2010.
Audi, Robert. Rationality and Religious Commitment. Oxford: Clarendon
Murray, Michael J., and Michael Rea. An Introduction to the Philosophy of
Press, 2011.
Religion. Cambridge, UK: Cambridge University Press, 2008.
Audi, Robert. Democratic Authority and the Separation of Church and State.
Peterson, Michael, and Raymond VanArragon, eds. Contemporary Debates in
Oxford: Oxford University Press, 2013.
Philosophy of Religion. Oxford: Blackwell, 2004.
Baker, Lynne Rudder. “Christians Should Reject Mind-Body Dualism.” In
Plantinga, Alvin. Warrant and Proper Function. Oxford: Oxford University
Contemporary Debates in Philosophy of Religion, edited by Michael Peterson
Press, 1993.
and Raymond VanArragon, 327–338. Oxford: Blackwell, 2004.
Plantinga, Alvin. Warranted Christian Belief. Oxford: Oxford University
Buchak, Lara. “Can It Be Rational to Have Faith?” In Probability in the
Press, 2000.
Philosophy of Religion, edited by Jake Chandler and Victoria Harrison,
225–246. Oxford: Oxford University Press, 2012. Rowe, William L. Philosophy of Religion: An Introduction. 3rd ed. Belmont,
CA: Wadsworth, 2001.
Collins, Robin. “The Design Argument: Between Science and Metaphysics.”
In Analytic Philosophy without Naturalism, edited by Antonella Corradini Rowe, William L., Michael Peterson, Bruce Reichenbach, and David
and Sergio Galvan, 140–152. London: Routledge, 2006. Basinger, eds. Reason and Religious Belief. 5th ed. Oxford: Oxford
University Press, 2012.
Conee, Earl, and Richard Feldman. Evidentialism. Oxford: Oxford
University Press, 2004. Stump, Eleonore, ed. Reasoned Faith. Ithaca, NY: Cornell University Press,
1993.
De Caro, Mario, and David Macarthur, eds. Naturalism in Question.
Cambridge, MA: Harvard University Press, 2004. Stump, Eleonore. Wandering in Darkness: Narrative and the Problem of
Suffering. Oxford: Oxford University Press, 2010.
Dembski, William A., and Michael Ruse, eds. Debating Design: From
Darwin to DNA. Cambridge, UK: Cambridge University Press, 2004. Swinburne, Richard. Providence and the Problem of Evil. Oxford: Oxford
George, Alexander. “What’s Wrong with Intelligent Design, and with Its
Critics.” Christian Science Monitor, 22 December 2005. https://www Swinburne, Richard. Epistemic Justification. Oxford: Oxford University
.csmonitor.com/2005/1222/p09s02-coop.html. Press, 2001.
Hasker, William. The Emergent Self. Ithaca, NY: Cornell University Press, Swinburne, Richard. The Existence of God. 2nd ed. Oxford: Oxford
1999. University Press, 2004.
Hick, John. Philosophy of Religion. 4th ed. Englewood Cliffs, NJ: Prentice- Taliaferro, Charles. Consciousness and the Mind of God. Cambridge, UK:
Hall, 1990. Cambridge University Press, 1994.
Howard-Snyder, Daniel. “Propositional Faith: What It Is and What It Is Wainwright, William. Reason and the Heart. Ithaca, NY: Cornell University
Not.” American Philosophical Quarterly 50, no. 4 (2013): 357–372. Press, 1995.
Kim, Jaegwon. Physicalism or Something Near Enough. Princeton, NJ: Wainwright, William, ed. Oxford Handbook of Philosophy of Religion.
Princeton University Press, 2005. Oxford: Oxford University Press, 2005.

TOPIC 2
Method: Atheism
Graham Wood
Lecturer, School of Humanities
University of Tasmania, Australia
Suppose that you are given two worldviews, one championed by a theist and one championed by an atheist.
What method or methods should be used in attempts to assess the comparative merits of these worldviews?
What kinds of considerations should feed into these methods? This chapter begins with a discussion of
themes central to answering these two questions, including specifying two worldviews to allow for
meaningful comparison. Then a series of topics is addressed for the benefit of a person, identified as the
“undecided person,” who does not yet endorse either worldview, in order to establish if there are reasons to
prefer the atheistic worldview presented here.
DISTINGUISHING TWO USES OF THE WORD WORLDVIEW
For now we see through a glass, darkly.

1 COR. 13:12
Consider what is fundamental, what there fundamentally is, what it is fundamentally like, what
it fundamentally does, what fundamental norms and values apply. This is a worldview. But there are
two ways in which the word worldview might be understood.
First, it can be understood as an account of what is fundamental, and second, as what is fun-
damental (i.e., the world itself). Assuming that the account is not the world itself, an issue arises.
Aspects of the account may be artifacts of the account and not aspects of the world. Consider
Immanuel Kant’s ([1781] 1970) distinction between the noumenal (i.e., what is fundamental)
and the phenomenal (the world of experience). Kant asserts that accurate accounts of the nou-
menal are not possible. Thus a question emerges: what aspects of the account are artifacts of
the account and not aspects of the world itself? Answers to this question bear directly on what
methods might be used to assess the comparative merits of worldviews (understood in both senses
of the word).
GETTING AT THE ANSWER TO THE QUESTION:

WHAT IS FUNDAMENTAL?
Ideally a worldview (understood as an account) should refer only to aspects of the world itself and
should avoid (if possible) distortions that are artifacts of the account. But the problem is knowing
what are, and what are not, artifacts of the account. As the commonplace observation teaches, if
one is looking at the world through colored glass, one should not assume the world is fundamentally
all shades of that color.
Consider, for example, “comprehensibility” as a possible artifact of the account. Imagine a
totally accurate account of the world itself that is incomprehensible to humans. It will have no
49
Method: Atheism
merit to humans. Perhaps any comprehensible account of the

KEY CONCEPTS world itself is not accurate, but nonetheless it will have merit
because it is comprehensible. For example, contemporary
physics asserts that fundamentally there are no midsized
Cartesian Circle
physical objects (because “fundamentally” they are consti-
Closed Naturalism tuted by arrangements of subatomic particles). But perhaps
Coherence Theory of Truth because humans “naturally” think in terms of physical
Correspondence Theory of Truth objects, then accounts of the world involving physical
objects have merit, even though those accounts are not accu-
Criterion of Ontological Commitment rate. Thus, perhaps, all that humans can do is construct
Euthyphro Problem accounts of the world itself that are comprehensible (or
Forms, the (Plato); Realm of the Forms expressible). That which is not comprehensible (or express-
ible) will not be included in such accounts. This is the
Logical Positivism
essence of Ludwig Wittgenstein’s observation, “Whereof
Logical Possibility/Impossibility one cannot speak, thereof one must remain silent” (1922,
Manifest Image/Scientific Image Proposition 7). A related issue is whether human judgments
Methodological Naturalism about possibility (and in particular, logical possibility) are
about metaphysical states of affairs, or merely reflective of
Naturalism our representations of the world. (For more on this, see
Ontological Naturalism Wood, 2005.)
Ontological Possibility
Open Naturalism DID MATTER EMERGE OUT OF MIND OR
Pragmatic Theory of Truth MIND EMERGE OUT OF MATTER?
Problem of Evil Argument
Problem of Induction The debate between theism and atheism can be understood as
one manifestation of a broader contrast that is represented (if
only imperfectly) by the question: did matter emerge from
mind or did mind emerge from matter? Consider the concept
of mind broadly to include any type of abstract ground of reason or rationality. Consider
the concept of matter very broadly to include any (nonmental) concrete particulars (i.e., bits of
matter and energy). Now consider: which is fundamental?
A number of philosophical positions share the assumption that matter emerged out of mind.
Consider Plato’s (1993) Realm of the Forms, or the Logos (understood as “eternal reason”) of the
Stoics as described by Diogenes Laërtius (Long and Sedley 1987), or the “pure reason” of Kant
(1970), or indeed theism, where God is understood as a divine mind creating the cosmos. All of
these share the basic idea that comprehensible order is the fundamental foundation of existence.
In contrast, consider the brute existence of the universe as it is described by the Epicureans or the
brute matters of fact as described by David Hume ([1779] 1993). These positions assume that mind
(understood as comprehended order) has arisen out of matter.
Whether mind arises out of matter or matter arises out of mind is the fundamental distinc-
tion. It may be the case that there exists a God, understood essentially as a mind, and that this
mind is the origin of the material world. But now recall the possibility of looking at the world
through colored glass and assuming that the world is all shades of that color. Humans “look”
at the world through a mind, and thus humans should always be aware of the possibility that
they may be “seeing” a mind (or “seeing” comprehended order, perhaps understood as the con-
straints of logical possibility) at the very foundation of the world but that that mind is not a fun-
damental aspect of the world. Perhaps humans are seeing this only because they are looking at
the world through a mind. On this reading, Platonism, Stoicism, and theism are in fact all in
the same broad category. Proponents of these positions are all “seeing” the Forms, or Rationality,
or the mind of God as if one of these were the foundation of reality, when in fact they are seeing
their own minds as they look through those minds out into the world. Kant (1970), in contrast,
Method: Atheism
recognizes that the mind imposes structure on itself, for example, in his account of the pure
forms of intuition, such that we experience objects in space and time.
STIPULATING TWO WORLDVIEWS FOR COMPARISON
This book considers the opposing viewpoints of theism (the claim that there is at least one god) and
atheism (the claim that there are no gods). But comparing theism with atheism is not a manageable
task, because these terms could refer to many distinct worldviews. So, to make the task manageable
two more precisely defined worldviews will be compared.
In this chapter significant aspects of both views will be assumed to be held in common. These
common aspects will be identified first, followed by discussion of the aspects of the worldviews
not held in common. Along the way, consideration will be given to whether or not the particular
aspects under consideration are fundamental. Much depends on what is, or is not, taken to be
fundamental, because this will bear directly on what methods can be used to determine the merits
of each worldview.
Common aspects of both views are as follows. There is a universe that includes matter and
energy understood as existing within space-time. The arrangements of matter and energy located
in space change over time. The changes in the arrangements of matter and energy can be under-
stood with reference to laws of nature (that may be deterministic or indeterministic). Living organ-
isms, as arrangements of matter and energy, exist and came into existence via a process of biological
evolution, subject to the laws of nature. Some living organisms experience conscious awareness and
are able to think about their own existence and the existence of the universe. Finally, they are able
to form questions in their minds about what might exist beyond the universe.
Now consider the atheistic worldview, to be considered here as a form of naturalism. Call it
open naturalism. When characterizing naturalism, a distinction is often drawn between ontological
naturalism and methodological naturalism. Ontological naturalism assumes that only natural
objects, events, and states of affairs exist. Ontological naturalism, so characterized, might be called
closed ontological naturalism, in comparison to open (ontological) naturalism. Open naturalism
assumes that natural objects, events, and states of affairs do exist, but it does not necessarily preclude
the existence of nonnatural objects, events, or states of affairs.
How are the natural objects, events, and states of affairs of open naturalism to be identified?
The “short answer” (not filled out in this chapter) is: by using the methods of science. This short
answer is the essence of methodological naturalism. But, just as open naturalism has an openness
to ontological possibilities beyond the “natural,” it also has an openness to methodologies beyond
those uses within the sciences. Open naturalism assumes the existence of objects, events, and states
of affairs identified by science, but it does not necessarily reject the assumed existence of objects,
events, and states of affairs identified by other methods. As such, open naturalism (Wood 2017)
aligns with Willard Van Orman Quine’s (1969) “naturalized epistemology.”
What is fundamental in open naturalism? The approach to answering this question will pro-
gress as follows. First, a minimal answer will be offered, and then additional features will be consid-
ered as possible additions to the list of that which is fundamental.
The brute existence of the universe is fundamental in open naturalism. Open naturalism takes
the following to be fundamental. There is a universe that includes matter and energy understood as
existing within space-time. The arrangements of matter and energy located in space change over
time. The changes in the arrangements of matter and energy can be understood with reference to
laws of nature (that may be deterministic or indeterministic). Note at this point that “laws of
nature” are not fundamental. This, traditionally, is taken to be Hume’s position. Hume (1975), it
is said, thought that there were simply brute matters of fact and that humans apprehended certain
regularities in the arrangements of matter and energy, formalizing these regularities as laws of
nature. Now, adding to the “what is fundamental” list, and contra the traditional Humean position,
it may be that laws of nature are fundamental such that the change that matter and energy undergo
is due to laws of nature.

Method: Atheism
No more of the commonly shared view is fundamental to open naturalism. Thus the existence
of life, including the existence of conscious life (and all that conscious experience and conscious
reflection involves) is not fundamental. So, it follows that there are no fundamental norms or values
in open naturalism. There are, of course, nonfundamental norms and values, but they exist as pro-
ducts of the biology, psychology, and society of humans (and other conscious organisms).
Now we turn to theism, which here is broadly understood as classical theism. Classical theism
shares all the common aspects of both views specified above. But it also includes the existence of
God, understood as follows: “there exists a person without a body (i.e., a spirit) who is eternal, is per-
fectly free, omnipotent, omniscient, perfectly good, and the creator of all things” (Swinburne 1991, 8).
So, what is fundamental in classical theism? Minimally, God. It is important to note the signif-
icance of this minimal position. Perhaps nothing but God is fundamental in classical theism.
Let’s move now to the common features of both views. The minimal position, considered
above, assumed that nothing but God is fundamental. But having left the minimal position behind,
once created by God, many of the features common to both views might be taken to be fundamen-
tal. Matter, energy, and how their arrangements change through space and time, and even laws of
nature (understood either deterministically or probabilistically), can all be taken to be fundamental.
Furthermore, according to one sophisticated theist account, offered by Richard Swinburne (1991),
God created the universe, and its laws and initial conditions, such that it would give rise via evolu-
tion to the existence of intentional agents, namely human beings.
Importantly, both for Swinburne’s account and classical theism more generally, “persons” are also
taken to be fundamental features of the theistic worldview. While in open naturalism the existence of
humans and their capacities of reflection are contingent, and thus nonfundamental features of the
worldview, the place of “persons” is fundamental in theism. The distinction between “person” and
“human” is drawn here to allow for the dualism that is a central feature of classical theism. To put
the point simply, God is said to have a loving relationship with persons (perhaps understood as souls),
and God creates human bodies as the place of the spatiotemporal embodiment of these persons.
Finally, norms and values are also fundamental to the worldview of classical theism. The details
of these norms and values is not considered here, but presumably they are the norms and values of
God. One way to understand the implications for humanity is to reflect on a characterization,
offered by Peter van Inwagen, of the purpose of a person’s life as to “love and serve” God forever
(2002, 4).
So the contrast at the fundamental level between classical theism and open naturalism is pro-
found. The theistic worldview is infused with fundamental meaning, purpose, norms, and values
that originate in the mind of God and can be recognized and endorsed in the minds of humans.
But any meaning, purpose, norms, or values that exist in open naturalism are not fundamental.
They are contingent and exist only in the minds of conscious organisms. It’s important to note,
though, that they can be projected onto the world in the way Hume suggested with his observation
that “the mind has a great propensity to spread itself on external objects” ([1739] 1984, 217). Thus
humans may think that norms and values are fundamental, but they are not.
WHAT IS A WORLDVIEW FOR?
In order to consider the merits of any worldview, it is necessary to ask: what is the worldview for?
Without an answer to this question it is impossible to make any progress considering the merits
of that worldview. The world itself, according to open naturalism, is not “for” anything. The exis-
tence of the world itself, according to classical theism, presumably aligns with the purposes of
God. So presumably God could provide an answer to the question: what is the universe for? But
what is a worldview for (when understood as an account of the world)? Theists would presumably
reply that (perhaps among other things) a worldview is to allow humans to have a relationship with
God. And notice the close alignment between the purpose of the world itself and the purpose of the
theistic worldview. But in the case of open naturalism the answer to the question of what a world-
view is for is more complex.
Method: Atheism
One possibility is that a worldview attempts to provide a true account of the world itself. But
there are a number of theories of “truth.” The correspondence theory understands a true worldview
to correspond to the world itself. The coherence theory understands a true worldview to be an
account of the world that is holistically coherent (i.e., internally consistent). The pragmatic theory
of truth understands the true with reference to purpose. To understand this theory, consider
William James on truth: “The true is the name of whatever proves itself to be good in the way of
belief, and good, too, for definite assignable reasons” ([1909]1975, 42). Another way to understand
what worldviews are for is to consider conceptual schemes and ask: what are conceptual schemes
for? A conceptual scheme is a set of concepts used to understand the world and to act in that world.
A human individual may employ several different conceptual schemes (that may not be consistent
across their different content), but if they have one scheme, it could be understood as a worldview.
Here is Quine on the conceptual scheme of science:
As an empiricist I continue to think of the conceptual scheme of science as a tool, ultimately, for
predicting future experience in the light of past experience. Physical objects are conceptually
imported into the situation as convenient intermediaries—not by definition in terms of
experience, but simply as irreducible posits comparable, epistemologically, to the gods of Homer.
… Both sorts of entities enter our conception only as cultural posits. The myth of physical
objects is epistemologically superior to most in that it has proved more efficacious than other
myths as a device for working a manageable structure into the flux of experience. (1961b, 44)
This approach can be applied to other conceptual schemes, and if a worldview is understood as a
single conceptual scheme, then it can be applied to worldviews. Essentially, it involves asking the
question: what is a worldview for?
Does a worldview aim at truth? It might be assumed by the undecided person that truth
(understood within something like a correspondence theory) should be the aim of the two world-
views under consideration. But, as has been noted, Quine did not identify the pursuit of truth as
the purpose of the conceptual scheme of science; rather he identified prediction.
Different worldviews may serve different purposes, so it is important to identify the purpose of
a worldview. Furthermore, the “merits” of a worldview must be considered with reference to the
purpose of that worldview. Thus the question “what is a worldview for?” is relevant throughout this
chapter. If the purpose of a worldview is to aim at truth, then the merits of that worldview must be
measured in terms of that purpose. But if the purpose of a worldview is not to aim at truth, then
the purpose of the worldview must be established, and then the merits of that worldview can be
measured against that purpose.
The existence of multiple conceptual schemes (or worldviews) is highlighted by the distinction
between the manifest and scientific images drawn by Wilfrid Sellars (1964, 57). One of the most
insightful applications of this distinction has been made by Daniel Dennett, in reference to cogni-
tive systems (both producing forms of manifest image) that generate the conceptual schemes of both
“folk physics” and “folk psychology”:
It is no accident that we have the manifest image that we do; our nervous systems were
designed to make the distinctions we need swiftly and reliably, to bring under single sensory
rubrics the relevant common features of our environment, and to ignore what we can usually
get away with ignoring.… Thanks to folk physics we stay warm and well fed and avoid
collisions, and thanks to folk psychology we cooperate on multiperson projects, learn from
each other, and enjoy periods of local peace. These benefits would be unattainable without our
extraordinarily efficient and reliable systems of expectation-generation. (1987, 11)
This passage is not only a powerful example of the view that mind arises out of matter but also a
good illustration of the fact that conceptual schemes, or more broadly worldviews, can be under-
stood instrumentally, and, most importantly, different conceptual schemes (or worldviews) may
serve different purposes. Essentially, from the perspective of open naturalism, worldviews are for
helping humans make their way in the world.
Thus the task here is to continually ask what the purpose of a worldview is and then to judge
the merits of a worldview in the light of its purpose.

Method: Atheism
THE MEANINGFULNESS OF THE LANGUAGE IN WHICH THE WORLDVIEW

IS EXPRESSED
Imagine an undecided person who currently endorses neither open naturalism nor classical theism
comparing the merits of the two worldviews. If one worldview is expressed in meaningful language
(a comprehensible worldview) and the other is expressed in meaningless language (an incomprehen-
sible worldview) that would be one reason to prefer the worldview expressed in meaningful
language. But is it the case that one worldview is really meaningless? Advocates on both sides assert
the meaningfulness of their worldviews. In the spirit of engagement that motivates this book, every
effort must be made to understand the meaning that is asserted by the advocates of the different
worldviews.
In particular, it will be taken to be contrary to this spirit of engagement for advocates of one
worldview to simply deem essential features of the other worldview to be meaningless. To illustrate
one such disengaged approach, consider logical positivism. The logical positivists considered science
to be a trusted method in the pursuit of knowledge of reality and sought to clearly separate “science”
from “metaphysics” (which includes claims about God). For example, A. J. Ayer (1952) performed
this separation by applying the principle of verification as a criterion of meaning. For a sentence to
be meaningful, either it must be analytically true (true by virtue of the meanings of the words in the
sentence) or there must be a method by which the truth of the sentence could be empirically veri-
fied. If there is no way for the claim of the sentence to be empirically verified (or refuted), the claim
of the sentence is meaningless.
But recall the undecided person, comparing the merits of the two worldviews. Let it be
assumed, for argument’s sake, that the person takes the claim “there exists a person without a body
(i.e., a spirit) who is eternal, is perfectly free, omnipotent, omniscient, perfectly good, and the crea-
tor of all things” (Swinburne 1991, 8) to be meaningful (independently from whether it is true or
false). After all, she has the experience of her own mental life as a person. So, it will not be sufficient
to simply define certain essential features of an opposing worldview as meaningless.
However, there is another way to engage with the question of the meaningfulness of language,
and that is by attending to the different ways language can be used. First, consider Wittgenstein’s
distinction between language used to set up a framework and language used to work within that
framework. To illustrate, he observes that: “It is one thing to describe methods of measurement,
and another to obtain and state results of measurement” (1953, sec. 242).
Second, consider Rudolf Carnap’s distinction between “internal” and “external” questions, with
reference to conceptual schemes. Internal questions relate to “testing the conformity of a given
remark or doctrine to a prior ontological standard” (Quine 1961a, 15). External questions relate
to which conceptual scheme to adopt. Quine addresses the external question from “the epistemolog-
ical point of view.” But he notes that this point of view “is one among various, corresponding to
one among our various interests and purposes” (1961a, 19).
Now return to the undecided person. This person can, first, attend to the different ways lan-
guage can be used, and second, use the distinction between “internal” and “external” questions relat-
ing to conceptual frameworks, and third, accept the further insight offered by Wittgenstein that “the
speaking of language is part of an activity, or of a form of life” (1953, sec. 23). The person can now
understand that, according to open naturalism, the claim “God exists” (Swinburne 1991, 8) is not a
claim internal to a worldview but rather is a framing proposition of a worldview, of an entire form
of life. Thus the person can choose either to engage or not to engage in that form of life, based on
its merits.
FALSIFIABILITY OF SIGNIFICANT FEATURES OF THE WORLDVIEW
Continue to imagine an undecided person. Now reflect on the question (considered above): when a
worldview is understood as an account of the world, and not the world itself, what is a worldview
for? One answer might be to give an accurate account of the world. But some might consider this
Method: Atheism
answer, upon reflection, to be intellectually idle. After all, why do humans need an accurate account
of the world itself? One answer is to be able to predict the future. As mentioned above, Quine
thinks of science as a tool for predicting the future. So perhaps a worldview is for prediction. If
so, how can the success of predictions generated by a worldview be tested?
This was a question addressed by Karl Popper with reference to the claims of science in the fifth
edition of his book Conjectures and Refutations (1974). Popper provides an account of the scientific
method based on a two-step process. The first step is to make a claim (the process behind this step
is not centrally relevant here). The second step is to attempt to refute the claim. So, perhaps
counterintuitively, Popper argues that the strength of science is in the fact that good science involves
specific claims that are highly falsifiable, and yet not falsified. He endorses more specific claims
because they are more falsifiable than less specific claims. Furthermore, he stresses that changes
made to the claims of science that only serve to avoid the falsification of those claims is bad science.
If a worldview is in the business of prediction, and furthermore, if its claims are highly
falsifiable and yet not falsified, this would count in its favor. So now return to the undecided per-
son. If this person assumes that both classical theism and open naturalism are in the business of pre-
diction, then: is there a difference in prediction that could provide reason to prefer one over the
other? As stipulated above, the laws of science are held in common across both worldviews, so it
is not predictions based on science that can give such a reason. It is only predictions generated by
the features of the worldview, then, that are not held in common that are relevant. Open naturalism
only provides predictions based on science. Classical theism can be understood to make predictions,
such as the survival of a person’s soul after the death of the body. And by Popper’s lights this is
unfalsifiable (at least before death). So, if classical theism is in the business of predictions (like this
one), then lack of falsifiability counts against it.
THE (ALLEGED) EXISTENCE OF PROOFS OF THE WORLDVIEW
Alleged proofs offered by theists include the ontological argument and the cosmological argument,
and alleged proofs offered by atheists include the problem of evil argument. If any proofs of a world-
view were available, these would be reasons to prefer that worldview over another that lacked such
proofs. But the question is: are there any such proofs available?
Proofs in this context are to be understood as valid deductive arguments with true premises.
Arguments in this context are to be understood as based on inferential relations that hold between dif-
ferent propositions, where those propositions are identified as either premises or a conclusion. A valid
deductive argument has a form such that if the premises are true the conclusion must also be true.
Of course, the stability of the meaning of terms used within the argument is important. If the
meaning of any term changes between instances of that term in the argument, then this calls the
argument into question.
Consider again the undecided person. For argument’s sake, assume that the meanings of all the
terms are stable across all instances of each term in the argument. In such circumstances, whether or
not a proof exists will now depend on the truth of the premises.
A central problem here is when an argument assumes that which it sets out to prove (called
begging the question). If the purpose of the argument is to prove that God does (or does not) exist,
it is not appropriate to assume either of those positions in one or more of the premises. Arguing to
the conclusion that God exists (or does not exist) from premises assuming that God exists (or does
not exist) will not convince an undecided person.
If it is to work for an undecided person, the method of proof must start from a position that is
accepted by that undecided person. For example, assuming that the undecided person accepts the
claim that evil exists, an atheist may argue that the presence of evil is logically inconsistent with a
God that is perfectly good and omnipotent and omniscient, as is argued by, for instance, J. L.
Mackie (1982).
From the perspective of open naturalism, there are some issues with proof as a method. First,
for the open naturalist, worldviews are understood instrumentally. Thus, the question “what is this

Method: Atheism
worldview for?” must be addressed. For example, following Quine, the conceptual scheme of science
is for prediction. Second, a particular subset of propositions associated with a worldview is that of
framing propositions, and these need to be accurately identified as such. For example, “God exists”
is taken to be a framing proposition for the theistic worldview. Once these two issues are addressed,
things become clearer. From the perspective of open naturalism, the conceptual scheme of science
(adopted for the purpose of prediction) and theistic conceptual schemes that include the framing
proposition “God exists” are not adopted for the same purpose. (Open naturalism assumes theistic
worldviews are adopted to further other biological, psychological, and social purposes, such as to
facilitate cooperation.)
THE SELF-EVIDENT TRUTH (OR FALSITY) OF THE WORLDVIEW
The world itself is not “true” or “false” because it is not the type of thing that can be true or false; it
is not a worldview. But, setting that aside, can a worldview be true or false, and can a worldview be
self-evidently true or false? These questions prompt a number of observations.
First, recall Quine on conceptual frameworks. He would counsel that to ask whether a frame-
work was true or false was to ask the wrong question. The appropriate question to ask is whether
the framework is suitable for its purpose. So perhaps the appropriate questions to ask are: (1) is
there a self-evidently most obvious purpose to be pursued? And (2) is there a self-evidently
most suitable worldview to adopt in the pursuit of that purpose? Those who claim that there is a
self-evidently true worldview presumably have already taken a position on these two questions,
and perhaps taken these positions without seeing any other positions that could displace the “self-
evident” positions they hold. Or perhaps they have mistaken the apparent self-evidence of some fea-
ture of the world for an assumption they themselves have made about the world.
Second, these questions presume that a worldview as a whole can be considered self-evidently
true or self-evidently false. Recall the question considered above: does matter emerge from mind,
or does mind emerge from matter? This question is a way to understand how a worldview as a
whole might appear self-evidently true. As discussed above, some people may take it as self-evidently
true that fundamental reality must be grounded in the Forms (Plato), or eternal reason (the Stoics),
or the divine mind (theism). But, as observed above, the self-evidence of this position may be an
artifact produced by the fact that these people are “looking” at the world through a mind. Perhaps
they are mistaking a cognitive structure they are using to represent the world for a structure in the
world itself. Recall Kant’s (1970) position that representations of space and time are pure forms of
intuition.
An alternate approach is not to consider the self-evidence of the worldview as a whole, but
rather the self-evidence of a distinct element of a worldview. What does it mean for a distinct ele-
ment within a worldview to be self-evident? For present purposes, take “self-evident” to mean that
there is no consciously accessible chain of mental processes that led to the apprehension of the
apparently self-evident element of a worldview. And here strong dispositions to believe, or strong
intellectual seemings, will also be considered self-evident.
For something to be self-evident it must be apprehended by some human faculty, so questions
can be asked about the constraints being imposed by that faculty (recall the person looking through
colored glass). It could be self-evident that God exists as an entity with beliefs and desires, or self-
evident that rationality has a certain authority, but these might be artifacts generated by considering
these matters from the perspective of a rational mind with beliefs and desires.
Consider the fact that humans think of the world in terms of physical objects, or that humans
think of themselves as having beliefs and desires, or that humans think of themselves as being
to some extent rational. While all of these may appear, from a first-person perspective, to be “self-
evident,” accounts of the mind are emerging that call into question how these aspects of conscious
experience should be understood. The contrast between the manifest image and the scientific image
has already been introduced. There are a number of ways to understand this contrast, but here, con-
sider the work of Elizabeth Spelke and Katherine Kinzler. They claim that humans have four “core
Method: Atheism
knowledge systems” that represent “[1] inanimate objects and their mechanical interactions, [2]
agents and their goal-directed actions, [3] sets and their numerical relationships of ordering, addi-
tion and subtraction and [4] places in the spatial layout and their geometric relationships” (2007,
89). Notice that two of these core knowledge systems could correspond to what has been called
“folk physics” and “folk psychology” (Dennett 1987).
It may be the case that the “self-evident” existence of physical objects, or the “self-evidence” of
the fact that humans have beliefs and desires, may need to be reexamined in the light of the exis-
tence of the scientific image as a rival to the manifest image. This is not to say that thinking in
terms of physical objects or in terms of human beliefs and desires should be abandoned (that,
presumably, would be impractical). Rather, thinking in these terms should be understood as an
activity undertaken from within a particular conceptual scheme and for a particular purpose. Put
another way (following Wittgenstein), thinking in these terms should be understood as a particular
“language game” within a particular “form of life.”
With particular reference to the debate between theism and atheism, research is emerging to
suggest that humans are “intuitive dualists” (Bloom 2004) and “intuitive theists” (Kelemen 2004).
Again, this research should be understood with reference to the question: what is the purpose or
purposes of the conceptual schemes within which such intuitive or self-evident thoughts emerge?
Open naturalism would suggest evolutionary purposes related to cooperation and communication.
With all this said, a number of concerns remain. Return again to the undecided person. What
should such a person make of the apparent fact that there is variability among other people about
what is self-evident? What should the person make of that which he himself takes to be self-evident?
What should be made of the possibility that what appears self-evident may be an artifact of a per-
son’s cognitive system, conceptual scheme, form of life, or worldview? From the perspective of open
naturalism, the apparent “self-evidence” of the truth or falsity of a worldview should be treated with
caution.
THE BURDEN OF PROOF BORNE BY THE WORLDVIEW
A burden of proof can be understood as a responsibility to provide a justification that rests with a
person advocating some change in her own (or others’) beliefs. This justification might be
demanded by the person herself or by others.
Consider the following situations. The first is one in which a departure from a commonly held
belief is advocated. Most often in life people receive a richly detailed worldview via enculturation;
that worldview is taken (perhaps unquestioningly) to be “true,” and then it is departures from that
worldview that requires justification. So, if a culture already holds a belief in God, the burden of
proof might lie with the members of that culture who are proposing that God does not exist. Alter-
natively, if the commonly held belief is that God does not exist, then the burden of proof might lie
with those who are proposing that God does exist. But the undecided person central to these delib-
erations neither actively believes nor disbelieves in God, so this first situation is not centrally rele-
vant here.
The second situation is one in which a belief replaces the absence of a belief. In his Meditations
(1931), René Descartes seeks a firm foundation of belief. He begins by rejecting all beliefs that are
possible to doubt. And he is left holding one belief that he claims is impossible to doubt, namely, “I
think, therefore I am.” But what justification can Descartes then offer himself to accept beliefs
beyond this belief? The justification he offers has been called the Cartesian circle. Descartes uses
the clear and distinct idea of a perfect God to conclude that it is impossible that a perfect God
would deceive him, thus allowing him access to the clear and distinct idea he originally used to
establish the existence of God (1931, 171–172). And he goes on from there. Many see a vicious cir-
cle of reasoning here. But, regardless of whether or not it is convincing, Descartes recognizes he has
a burden of proof to justify his move beyond the lonely thought “I think, therefore I am.”
Another way of understanding this second situation is to consider when it is justifiable to add
new beliefs to a set of existing beliefs. Say the undecided person accepts all that is common to the

Method: Atheism
two views under consideration in this chapter. What method could this person use to add beliefs?
One method is suggested by Quine. The following presents Quine’s own application of this
method, and then generalizes from that method.
Consider the conceptual scheme of science as a set of beliefs. What criteria should be used to
decide whether to add a new belief? Quine answers with the criterion of ontological commitment,
the essence of which is expressed by the slogan “to be is to be the value of a bound variable.” This
means that the entities to be taken seriously are the entities that have an essential place in our best
conceptual scheme of science. Here is how Quine expresses it: “a theory is committed to those and
only those entities to which the bound variables of the theory must be capable of referring in order
that the affirmations made in the theory be true” (1961a, 13–14).
But now recall that Quine admits that the conceptual scheme of science is only one possible
conceptual scheme. Other conceptual schemes may exist. Thus there may exist different conceptual
schemes (and different worldviews) with different purposes, such as cooperation and/or communica-
tion. Adapting Quine’s criterion of ontological commitment to a worldview with a purpose other
than prediction yields a different criterion of commitment. This criterion will identify the entities
that should be taken seriously because they have an essential place in that conceptual scheme. For
example, Kant ([1788] 1949) effectively uses a criterion of ontological commitment to assume the
existence of God in his moral argument for the existence of God (an argument considered in more
detail below). This is because God is an essential element of the conceptual scheme. Such a scheme
is employed, according to open naturalism, to further the purpose of cooperation, although Kant
presents his argument in terms of an alignment of happiness and virtue.
Now the question of the burden of proof can be understood differently. If it is accepted that
conceptual schemes, or worldviews, or forms of life are adopted to serve particular purposes, then
the questions of the burden of proof should be understood as requests to justify (1) why a particular
conceptual scheme, worldview, or form of life has been chosen, and then to justify (2) why particu-
lar entities understood within that context should be taken seriously. Thus the undecided person
should seek such justifications. Furthermore, the burden of proof can now be understood in light
of the fact that in order to make sense of the world using a conceptual scheme, some things must
be assumed simply in order to get the conceptual scheme up and running. This is the sense in
which the Cartesian circle can be understood as a virtuous (or more accurately a pragmatic) circle
rather than a vicious circle.
THE (ALLEGED) POSSESSION OF “PRIVATE EVIDENCE” THAT FAVORS

THE WORLDVIEW
What is the undecided person to make of the claim that there exists “private evidence” that favors
one worldview over the other? First, consider the possibility that although the undecided person is
undecided about which worldview to adopt, he takes himself to have access to this private evidence.
If so, presumably, he will consider that evidence in his own deliberations. Second, consider the sit-
uation in which the undecided person does not have personal experience of the private evidence. He
could choose to believe in the existence of the private evidence, based solely on the testimony of
others, without further reflection. Third, the undecided person could seek further information
about the nature of this private evidence. Then the person might be in a position to understand
the nature of private evidence and to recognize whether or not he also has access to it.
One relevant form of private evidence is “religious experience.” Reading James’s book The Vari-
eties of Religious Experience (1903) would give the undecided person much insight into the nature of
religious experience as “private evidence.” Theists have drawn on religious experience as evidence for
the existence of God, but atheists have replied that religious experience does not meet a certain stan-
dard of evidence suitable to justify adoption of a belief (a standard associated with the pursuit of
what is sometimes called the doxastic ideal). Some theists who advocate “reformed epistemology”
(Plantinga and Wolterstorff 1983) argue in reply that several other aspects of the common epistemic
practices of humans, including belief in the trustworthiness of induction, or belief in the testimony
Method: Atheism
of others, don’t meet the same standard either. So, they argue, it would be inconsistent to exclude
one (i.e., religious experience) and not the other (e.g., confidence in induction).
The undecided person can learn from this exchange. Consider “faith in” induction. First, consider
induction from the perspective of common sense. It is based on the assumption that the unobserved
(say events in the future) will be similar to the observed (say events in the past). This is one of the
most fundamental assumptions humans hold. Furthermore, notice that the concept of prediction
depends on it. Now consider “faith in” induction within the context of the conceptual scheme of sci-
ence drawing on the insights of Quine. As has been mentioned, conceptual schemes are for some pur-
pose. For Quine, science is for prediction, and as Quine (and others) points out, a conceptual scheme
requires framing concepts. Open naturalism understands “faith in” induction as one of the framing
concepts of prediction, and hence a framing concept of the conceptual scheme of science.
As an aside, this explains why there is a so-called problem of induction. Many philosophers
(including naturalistically inclined philosophers) have addressed the so-called problem of induction
by questioning the rational justification of induction. The solution to the problem of induction is
to recognize a pragmatic rather than a rational justification. Induction is a framing concept of at least
two conceptual schemes, one associated with common sense (the manifest image) and another with
science (the scientific image). Conceptual schemes are adopted for pragmatic reasons, so there won’t
be a “solution” to the problem of induction involving a rational justification. The solution to the
problem is to understand that conceptual schemes are adopted for pragmatic reasons (and some
form of conceptual scheme involving induction was adopted deep in our evolutionary past).
Notice that one cannot engage with much of the practice that is central to science without
“faith in” induction. So the undecided person, when attempting to understand the private evidence
that is “religious experience,” can learn from “faith in” induction. Faith issuing from “religious expe-
rience” is not the same as “faith” in induction, but the two forms of “faith” perform the same func-
tional role within two different conceptual schemes. “Faith in” induction is the foundation of the
conceptual scheme adopted for the purpose of prediction, and faith issuing from religious experience
is the foundation of the conceptual scheme for some other purpose (perhaps furthering cooperation)
related to a religious form of life.
Those who are committed to the “form of life” that involves the conceptual scheme of science
can reflect on their “faith in” induction, and thereby have some insight into the “private evidence”
that those committed to a religious forms of life call “religious experience.” The undecided person
might be surprised by this and ask: does a religious person really understand religious faith in the
same way as any person understands “faith in” induction? The answer, drawing on the insights of
open naturalism, is a qualified yes. Functionally they are the same type of belief. These two “faiths”
perform the same function (albeit in different conceptual schemes). They are the framing concepts
of two different conceptual schemes, or “forms of life.”
But, importantly, among the variety of conceptual schemes some are more or less deeply
embedded in human cognitive practices than others. For example, perhaps all conscious organisms
have something equivalent to “faith in” induction. But perhaps only some humans have faith based
on religious experience. So, while the two “faiths” perform the same function they are distinct and
may be understood as more or less “self-evident” depending on how deeply the conceptual scheme
is embedded in any individual’s cognitive practices. The issue of how deeply different conceptual
schemes are embedded in human cognitive practices is implicit in J. C. A. Gaskin’s (1988, 116)
analysis of Hume’s discussion of “natural instincts.” Finally, a point about why “religious experi-
ence” might tend to be understood in relation to the existence of a “God”: the answer, from the per-
spective of open naturalism, is that humans look at the world through a mind, and thus it is not
surprising that humans “see” a mind looking back.
THE NORMATIVE AND EVALUATIVE CLAIMS OF THE WORLDVIEW
Classical theism incorporates fundamental norms and values at the heart of the worldview. God is
characterized as perfectly good, has a personal interest in each human being, and wants the best

Method: Atheism
for each person. Open naturalism has no fundamental norms or values. Humans are contingent
arrangements of matter and energy produced via biological evolution, understood as operating
within the laws of nature, lacking any fundamental purpose or value. Humans articulate and may
endorse norms and values with respect to such things as ethical and epistemological matters, but
such articulations and endorsements are produced by various biological, psychological, and social
factors. Humans are organisms that possess cognitive systems (adopting various conceptual schemes)
that have emerged through a purposeless process of variation and selection. That which first arose
via variation and was not selected out remains; that is all. Thus there is a profound difference
between the two worldviews.
Now turn to the undecided person considering the comparative merits of the two worldviews and
considering whether she might have reason to prefer one to the other. Assuming that the person,
though undecided between classical theism and open naturalism, has nonetheless come to hold various
norms and values (as a result of biological, psychological, and social factors), she might as a conse-
quence of those norms and values find open naturalism unattractive. According to open naturalism,
when Friedrich Nietzsche asserted that “God is dead. God remains dead. And we have killed him”
he was describing the abandonment of a conceptual scheme, or form of life ([1882] 1974, 181).
Furthermore, Nietzsche (1911) claimed, humans are free to undertake a “revaluation of all
values,” and perhaps values that may emerge after such a process (e.g., “might makes right”) will
be repellent to our current normative and evaluative conceptual schemes. The undecided person
may prefer classical theism. Kant grounded morality in rationality, but on Kant’s account there
was no guarantee that the good person would be happy. Thus he offered an argument (discussed
above) for the existence of God based on the assumption that God’s existence must be presupposed
in order to provide an alignment between happiness and virtue (Kant 1949). Thus Kant can
be understood to be endorsing the very conceptual scheme that Nietzsche claims “we” have
abandoned.
Now consider the place of truth in apportioning merit to worldviews. Assume truth is under-
stood pragmatically as James characterizes it: “The true is the name of whatever proves itself to be
good in the way of belief, and good, too, for definite assignable reasons” (1975, 42). If there are def-
inite assignable reasons to prefer a worldview with fundamental values and norms, then there are
reasons to prefer classical theism. But if truth is understood as the correspondence theory supposes,
in that true propositions correspond accurately to the world itself, then the undecided person cannot
accept classical theism, unless the person comes to decide that classical theism accurately corre-
sponds to what is fundamental. The possibility of coming to the view that classical theism is true,
by simply starting to live as if it is true, is explored by Blaise Pascal in his Wager (1910 [1670]).
Assuming there are no fundamental values is not to assume there are no values. But the values
that do emerge will be contingent on biological, psychological, and social factors. Consider Plato’s
Euthyphro problem: Do the gods love “the good” because it is good, or is “the good” good because
the gods love it? Plato proposed that the good derives its goodness not from the gods but from the
Form of the Good that exists in the abstract realm of the Forms. But open naturalism does not
appeal to the realm of the Forms (or any other abstract foundation of norms and values, e.g., Kant’s
rational foundation for morality), and given that open naturalism does not appeal to God either, it
is left with the perhaps unsettling conclusion that the good is good contingently (the good is good
because the gods love it). But this is not as unsettling as it may first appear, simply because the con-
stitution of humanity is (contingently) filtered through evolution and, perhaps mercifully, there is
generally an alignment between normal human psychology and normal human ethics.
Unfortunately, at least in the minds of some, the normative and evaluative foundations of epis-
temology are also the result of evolutionary processes. The classical theist can rely on God to ensure
the authority of the outcomes of suitably rigorous epistemic inquiries, as Descartes (1931) assumed.
But the open naturalist has no such guarantees, as has been argued by Alvin Plantinga (2000). The
cognitive capacity to identify the truth and nothing but the truth (when that is understood as a cor-
respondence with the world itself) does not appear to be a cognitive capacity that can be selected
directly by evolution. On the other hand, the ability to predict the world is a cognitive capacity
directly relevant to survival, and can be selected directly by evolution. But there is no obvious
Method: Atheism
way for natural selection to select between (1) cognitive systems that successfully generate predic-
tions and (2) cognitive systems that successfully generate predictions and “truth” (Wood 2013).
Furthermore, while conceptual schemes that aim at truth may, hypothetically, do a good job of
prediction, that may not actually be the case. If, in order to generate a prediction, a conceptual
scheme requires more computational capacity than is available to an organism, then the organism
will not be able to complete the computation and no prediction will be generated. Finally, concep-
tual schemes that are not truth-tracking may still do a good job of prediction (Gigerenzer and Selten
2001; Kahneman 2011).
Some have attempted to resist Plantinga’s arguments by contesting the argument itself (see
Everitt 2004), or by seeking an independent authority for human epistemic practice, drawing on,
for example, Kantian accounts of rationality (Korsgaard 1996). But, according to open naturalism,
Plantinga’s argument has force, hence the endorsement of a pragmatic theory of truth rather than
a correspondence theory of truth.
Thus, here it will be assumed that, while advocates of worldviews, conceptual schemes, or
forms of life may deem truth (as correspondence) to be the aim of these schemes, it need not be
the deemed aim. Here worldviews, conceptual schemes, and forms of life are understood with refer-
ence to their purpose. And once worldviews are understood with reference to purpose, this aligns
with a pragmatic account of truth. Recall James: “The true is the name of whatever proves itself
to be good in the way of belief, and good, too, for definite assignable reasons” (1975, 42). Using
James’s phrasing, then, the definite assignable reasons are the purposes of the worldview under con-
sideration, and then the true proves itself to be good by furthering those purposes. Thus the task of
the undecided person is to decide on a purpose (or purposes) and then consider which worldview
best serves that purpose (or purposes). Then “truths” are those propositions that best serve the
purpose of the worldview.
THE COMPORTMENT OF THE WORLDVIEW WITH WHAT WE WISH

OR WANT TO BE THE CASE
What might the undecided person wish or want to be the case? He might want the world itself to be
comprehensible, or the worldview he comes to hold to be a comprehensible and accurate account of
what is fundamental in the world itself, or the world itself to affirm certain norms and values. But,
as the saying goes, wishing won’t make it so.
The discussion of ethical and epistemological norms and values in the preceding section is again
relevant here. But putting the point more broadly, humans may simply want the world to be cogni-
tively manageable. This may have been the motivation for the Stoics to claim that, ultimately, exis-
tence was subject to a perfectly rational order. Of course, the undecided person could simply assume
that human rationality is authoritative. But this could be interpreted by others as the undecided
person simply wishing to make it so.
AN OVERALL VERDICT?
This chapter considers two worldviews. Worldviews (understood as accounts of the world) may or
may not be accurate accounts of the world itself. Worldviews may be presented by their advocates
as true, where truth is understood as correspondence with the world itself, or where truth is associ-
ated with a set of beliefs such that the set taken as a whole is maximally coherent. Alternatively,
worldviews may be presented by their advocates as tools that serve certain purposes, and if they
are characterized as “true” then that “truth” can be understood instrumentally, according to how
well the worldview serves some purpose.
The concept of merit is central to addressing the two questions posed in this chapter: the ques-
tion of what method or methods should be used in attempts to assess the comparative merits of
these worldviews, and the question of what kinds of considerations should feed into these methods.

Method: Atheism
What are the merits of a worldview? Is being true the only merit? And what understanding of truth
is in play? If being true, in the relevant sense, is not the only merit, what other merits are relevant to
the current deliberations?
The two worldviews considered here are classical theism and open naturalism. Let it be
assumed that advocates of classical theism assert that, as a worldview, it is an accurate account
of the world itself, and as such it is true, where true is understood as correspondence with the
world itself. Let it also be assumed that advocates of classical theism assume that a worldview
can be a unified and coherent whole (which corresponds to a unified and internally coherent
world in itself).
According to open naturalism, worldviews, conceptual schemes, or forms of life are all to be
understood instrumentally. They are tools for specific purposes, tools to help humans make their
way in the world. Thus worldviews are not necessarily accurate accounts of the world itself. The
truth they offer is to be understood as a pragmatic truth, directed at the purposes the worldviews
serve. This attitude toward truth can be understood by substituting “truth” for the word “happi-
ness” in a popular and insightful saying (attributed to J. Richard Lessor): “Truth is like a butterfly:
the more you chase it, the more it will elude you. But if you turn your attention to other things, it
comes and sits softly on your shoulder.”
Open naturalism does not assume humans have any particularly authoritative epistemic access
to the nature of the world itself. Thus it may not be possible to unify all of the conceptual schemes
(that are useful in pursuit of the various human purposes) into one holistic (unified and coherent)
conceptual scheme, or worldview. This position is in some tension with the understandable hope
for a single coherent worldview.
To understand fundamental reality is a challenge faced by both classical theism and open natu-
ralism. Theists can claim that God exists necessarily, and “God’s will” then explains the rest of fun-
damental reality. But this may not convince the undecided person. Hume ([1779] 1993) observed
that there is no advantage in moving beyond the brute unexplained existence of the universe to
the brute unexplained existence of God, and the undecided person might learn from Hume and
choose not to make such a move.
Classical theism and open naturalism understand the concept “worldview” differently, and thus
when addressing the question of which methods to use to assess the comparative merits of the
worldviews, the undecided person should reflect on (1) the nature and purpose of worldviews,
(2) the merits of worldviews for humans, and (3) the nature of humans. But what answers would
emerge from such reflection? Perhaps, when one is considering what is ontologically fundamental,
in order to say anything at all one must begin by begging the very question one seeks to answer.
Acknowledgments. I thank James Chase for his intellectual generosity in offering invaluable advice
that significantly enriched this chapter. I also thank Kathryn Gillams for reading the chapter closely
and offering suggestions to improve the clarity of expression of ideas.
Bibliography
Ayer, A. J. Language, Truth, and Logic. New York: Dover, 1952. Evans, Jonathan St. B. T., and Keith Frankish, eds. In Two Minds: Dual
Processes and Beyond. Oxford: Oxford University Press, 2009.
Bloom, Paul. Descartes’ Baby: How the Science of Child Development Explains
What Makes Us Human. New York: Basic Books, 2004. Everitt, Nicholas. The Non-existence of God. London: Routledge, 2004.
Dennett, Daniel. The Intentional Stance. Cambridge, MA: MIT Press, 1987. Gaskin, J. C. A. Hume’s Philosophy of Religion. 2nd ed. London: Macmillan,
1988.
Descartes, René. Philosophical Works of Descartes. Vol. 1. Translated by
Elizabeth S. Haldane and G. R. T. Ross. Cambridge, UK: Cambridge Gigerenzer, Gerd, and Reinhard Selten, eds. Bounded Rationality: The
University Press, 1931. Adaptive Toolbox. Cambridge, MA: MIT Press, 2001.
Descartes, René. “Letter to Mersenne, 15 April 1630.” In Philosophical Hume, David. Enquiries concerning Human Understanding and concerning
Writings, edited and translated by Elizabeth Anscombe and Peter the Principles of Morals. Oxford: Clarendon, 1975. First published
Thomas Geach. London: Open University, 1970. 1751.
Method: Atheism
Hume, David. A Treatise of Human Nature. London: Penguin, 1984. First Plato. Republic. Translated by Robin Waterfield. Oxford: Oxford University
published 1739. Press, 1993.
Hume, David. Dialogues and Natural History of Religion. Edited by J. C. A. Popper, Karl. Conjectures and Refutations: The Growth of Scientific Knowledge.
Gaskin. Oxford: Oxford University Press, 1993. First published 1779. 5th ed. London: Routledge and Kegan Paul, 1974.
James, William. The Varieties of Religious Experience: A Study in Human Quine, W. V. “On What There Is.” In From a Logical Point of View: 9
Nature. London: Longmans, Green, 1903. Logico-Philosophical Essays. 2nd ed. Cambridge, MA: Harvard University
James, William. Pragmatism and The Meaning of Truth. Cambridge, MA: Press, 1961a.
Harvard University Press, 1975. First published 1909. Quine, W. V. “Two Dogmas of Empiricism.” In From a Logical Point of
Kahneman, Daniel. Thinking, Fast and Slow. London: Penguin, 2011. View: 9 Logico-Philosophical Essays. 2nd ed. Cambridge, MA: Harvard
University Press, 1961b.
Kant, Immanuel. Critique of Practical Reason and Other Writings in Moral
Philosophy. Translated by Lewis White Beck. Chicago: University of Quine, W. V. “Epistemology Naturalized.” In Ontological Relativity and
Chicago Press, 1949. First published 1788. Other Essays. New York: Columbia University Press, 1969.
Kant, Immanuel. Critique of Pure Reason. Translated by Norman Kemp Sellars, Wilfrid. “Philosophy and the Scientific Image of Man.” In Frontiers
Smith. London: Macmillan, 1970. First published 1781. of Science and Philosophy, edited by Robert G. Colodny, 35–78. London:
Kelemen, Deborah. “Are Children ‘Intuitive Theists’?” Psychological Science Allen and Unwin, 1964.
15, no. 5 (2004): 295–301. Spelke, Elizabeth S., and Katherine D. Kinzler. “Core Knowledge.”
Korsgaard, Christine M., with G. A. Cohen, et al. The Sources of Normativity. Developmental Science 10, no. 1 (2007): 89–96.
Edited by Onora O’Neill. Cambridge, UK: Cambridge University Press, Swinburne, Richard. The Existence of God. Rev. ed. Oxford: Clarendon; New
Long, A. A., and D. N. Sedley. The Hellenistic Philosophers. 2 vols. Van Inwagen, Peter. Metaphysics. Edited by Norman Daniels and Keith
Cambridge, UK: Cambridge University Press, 1987. Lehrer. Boulder, CO: Westview Press, 2002.
Mackie, J. L. The Miracle of Theism. Oxford: Clarendon, 1982. Wittgenstein, Ludwig. Tractatus Logico-Philosophicus. Translated by C. K.
Nietzsche, Friedrich. The Twilight of the Idols; or, How to Philosophise with the Ogden. London: Routledge and Kegan Paul, 1922.
Hammer; The Antichrist; Notes on Zarathustra; and Eternal Recurrence. Vol. Wittgenstein, Ludwig. Philosophical Investigations. Translated by G. E. M.
16 of The Complete Works of Friedrich Nietzsche. London: Foulis, 1911. Anscombe. Oxford: Blackwell, 1953.
Nietzsche, Friedrich. The Gay Science. Translated by Walter Kaufmann. New Wood, Graham. The Fine-Tuning of the Universe: A Philosophical Analysis.
York: Random House, 1974. First published 1882. PhD Thesis, University of Tasmania, 2005. https://eprints.utas.edu.au
Pascal, Blaise. Pensées. Translated by W. F. Trotter. London: Dent, 1910. /17430/.
First published 1670. Wood, Graham. “The Rationality of Heuristic Religious Belief.” In A New
Plantinga, Alvin. Warranted Christian Belief. Oxford: Oxford University Science of Religion, edited by Gregory W. Dawes and James Maclaurin,
Press, 2000. 189–204. New York: Routledge, 2013.
Plantinga, Alvin, and Nicholas Wolterstorff, eds. Faith and Rationality: Reason Wood, Graham. “Do Religious Beliefs Have a Place within an ‘Epistemically
and Belief in God. Notre Dame, IN: Notre Dame University Press, 1983. Naturalized’ Cognitive System?” Sophia 56, no. 4 (2017): 539–556.

TOPIC 3
Logic: Theism
Mashhad Al-Allaf
American University of Ras Al Khaimah, United Arab Emirates
This chapter considers the plausibility of theism by giving attention to the logic of argumentation offered for
the existence of God. It also discusses the various attributes of God, including omnipotence, omniscience,
and impassibility.
THE RATIONAL FOUNDATIONS OF FAITH
Is it possible to prove that God exists? This chapter considers the relation of logic and theism by
examining the reasoning used in two types of philosophical argumentation: a priori arguments
(those based on premises that are prior to and independent of experience) and a posteriori arguments
(those based on a premise known by experience). The chapter shows that different systems of logic
are sometimes operative, for some make use of a first-order predicate calculus while others employ a
higher-order calculus and modal logic.
THE ONTOLOGICAL ARGUMENT
The logic behind the ontological argument for God is purely a priori. It uses the idea of the perfect
being when it defines God as a being than which none greater could be conceived and then works
to show that we must grant that such a being exists, for if it lacked existence, then we would not
have been discussing what we claimed to be discussing, namely, that than which nothing greater
can be conceived. A classic version of this argument comes from Saint Anselm, who says in his
Proslogion (1926, chapter II):
AND so, Lord, do you, who do give understanding to faith, give me, so far as you knowest it to
be profitable, to understand that you are as we believe; and that you are that which we believe.
And indeed, we believe that you are a being than which nothing greater can be conceived. …
Therefore, if that, than which nothing greater can be conceived, exists in the understanding alone,
the very being, than which nothing greater can be conceived, is one, than which a greater can be
conceived. But obviously this is impossible. Hence, there is doubt that there exists a being, than
which nothing greater can be conceived, and it exists both in the understanding and in reality.
In chapter 3, Anselm concludes:
For, it is possible to conceive of a being that cannot be conceived not to exist; and this is
greater than one that can be conceived not to exist. Hence, if that than which nothing greater
can be conceived, can be conceived not to exist, it is not that than which nothing greater can
be conceived. But this is an irreconcilable contradiction. There is, then, so truly a being than
which nothing greater can be conceived to exist, that it cannot even be conceived not to exist;
and this being you are, O Lord, our God.
The argument is based on the very definition of God as utterly perfect. To avoid the fallacy of
assuming what he intends to prove, Anselm uses a definition of God that even someone who denies
65
Logic: Theism
God’s existence understands and accepts, for even one who

KEY CONCEPTS denies that God exists knows what he is denying, namely, that
there is a perfect being. It is important here to clarify his point:
the claim is not merely that this being is the greatest being
A Posteriori Arguments
within some set but that than which nothing greater can in
A Priori Arguments principle be conceived. As an a priori argument, what Anselm’s
Cosmological Argument for God (also analysis does is to show that the nonexistence of such a being is
called contingency argument or first cause inconceivable. The argument is not based on religious experi-
argument) ence or on what religious people might wish to think but on
a priori analysis of a concept independent of any experience.
Descriptive Statement
In a way, Anselm’s argument gives us reason in advance to
Disjunctive Syllogism dismiss J. N. Findlay’s effort to discredit this reasoning by sug-
Divine Attributes gesting that we take the idea of God as “an adequate object of
Kalām argument religious attitudes” (1948, 52). For Findlay, religious feeling
presumes a surpassing greatness in some object in order to
Nous (Greek) make it worthy of worship. The unsurpassable supremacy of
Ontological Argument for God such an object presumably makes all actual and possible reality
Rigid Designator depend on it, such that the existence of all things is inconceiv-
able without it. Even its own nonexistence is unthinkable, since
Tautological Statement various things of our experience are already clearly in existence.
Teleological Argument for God (also called Findlay’s argument, however, confuses the a priori idea of
argument from design) God that is employed in the ontological argument with the idea
of God that is based on religious attitudes and feelings. In fact,
the ontological argument is based on the very logic of definition
by its careful use of the phrase “that than which nothing greater
can be conceived.” It is not the experience of anything that grounds the idea of perfection implied
by this definition but simply the rules of logic, for something lacking existence cannot be that than
which nothing greater can be conceived. The truth value of the statement that “we must grant that
God understood as that than which nothing greater can be conceived exists” is determined by the
meaning of its terms. The predicate adds nothing to the subject. To deny that statement would
be self-contradictory, for one would be saying: “the being than which none greater than can be con-
ceived exists and does not exist.” Thus the conclusion that God exists becomes necessarily true
because it is logically impossible for it to be other than that.
Norman Malcolm raises a different sort of criticism of the ontological argument, writing:
“Anselm’s ontological proof of Proslogion 2 is fallacious because it rests on the false doctrine that
existence is a perfection (and therefore that ‘existence’ is a ‘real predicate’)” (1960, 41–62). But
Malcolm’s critique misses the point, for Anselm’s argument is not about “existence” as a “property.”
Rather, it is about what needs to be predicated of God once one has granted the idea of a being
“than which nothing greater can be conceived,” for this formulation is utterly comprehensive. To
use this definition is not to hold that “existence” in the ontological argument is a “property” that
might or might not belong to something but to hold that, once we understand the analysis of the
definition, we are obligated to grant that there is such a being.
The nature of this argument is clearly a priori and analytic: it is based on a definition formu-
lated in such a way that it must be granted even by one who wants to deny the existence of God,
for even one who makes such a denial must know what he intends to deny.
Even from the time of Anselm, the argument was often misunderstood. His contemporary,
Gaunilo of Marmoutiers, criticized the claim made in its first premise about the conceivability of
the idea of God as Anselm defined it, but he seems to have confused Anselm’s careful definition
with the notion of the greatest possible being within a given set (e.g., the idea of an island than
which no greater island can be conceived) and then to have denied that such an island might exist.
At the heart of Gaunilo’s case is the assumption that conceivability is only possible for concepts that
are finite and within our own experience. For Gaunilo, the concept proposed by Anselm is not even
a meaningful term for us. As a fellow believer, Gaunilo is not casting doubt on the existence of God
Logic: Theism
by his objection but only on Anselm’s argument for that claim. In support of his objection, Gaunilo
uses a reductio ad absurdum to show that Anselm’s argument is not sound. One can, he says, imag-
ine an island than which none more beautiful could be conceived, but there is no reason to think
that it exists.
In response to this objection Anselm concedes the point about the idea of any finite being like an
island but argues that the case of God is different, for in this case we are dealing not with the idea of
some finite being that we can imaginatively conceive but with the notion of a being that is limited in
no respect and that we can name by the word “God” even though what we are naming is something
that is in principle greater than anything that we can conceive. He then insists that if we were to deny
the existence of God when properly defined as “that than which nothing greater can be conceived,” we
would not have been considering “that than which nothing greater can be conceived,” for a such a
being that really exists would be greater than one that lacked existence. Hence, for Anselm, it is the
very definition of God properly understood that allows us to assert that it cannot be thought not to
exist. Yet Anselm thought Gaunilo’s “On Behalf of the Fool” so important that he insisted that it
and his response to it always be included in any future edition of his Proslogion.
Many subsequent thinkers have addressed the question of the validity of Anselm’s argument.
Thomas Aquinas, for instance, not only presents his own arguments for the existence of God by rea-
soning from observed effects back to the cause of those effects but also critiques the argument
offered by Anselm. To point out what he considers to be a fatal flaw, Aquinas makes a distinction
between two ways in which a proposition can be self-evident. He uses this distinction to argue that
Anselm’s claim about the existence of God being self-evident is in principle outside the possibility of
ever being known by us because we need to understand the essence of God in order to see that it
implies existence, but we simply do not know the essence of God:
It must be said that a thing can be called “self-evident” in two- ways, in itself and in relation to
us. A proposition is self-evident when its predicate is included in the definition of its subject.
[For example, in the proposition “man is an animal,” the idea of “animal” is included in the
definition of “man.” Thus if everyone knows the definitions of both subject and predicate, the
proposition will be self-evident to all, as is the case with the first principles of demonstration, the terms
of which are so common that no one is ignorant of them, such as “being” and “nonbeing,” “whole” and
“part,” etc. ] If, the proposition may be self-evident in itself, but not to them. Thus it happens, as
Boethius says, that some things are common conceptions of the mind” and are self-evident
“among the learned only, such as that incorporeal beings do not occupy a place.”
I say, therefore, that this proposition, “God exists,” is self-evident in itself, since the
predicate is the same as the subject. For God is his own existence, as will be seen later.
Nevertheless, because we do not know what is involved in being God, the proposition is not
self-evident to us, but needs to be demonstrated through those things that are more evident to
us though less evident to themselves, namely God’s effects. (Summa theologiae 1996, 1.2.1)
It is thus precisely because of the abstract formalism of Anselm’s definition of God as “that than
which nothing greater can be conceived” that Aquinas charges Anselm with begging the question:
“He thinks that we cannot argue that what is named by the term ‘God’ is actually exist, unless it
is somehow acknowledged that there actually exists something than which nothing greater can be
thought; but this is not admitted by those who think that God does not exist. Thus, the argument
begs the question.”
Without mentioning Anselm by name, René Descartes echoes his approach by arguing from
the idea of God as perfect (and thus not something that imperfect minds like our own could have
ever produced). From this Descartes concludes that we have to grant that such a being exists, or else
we will stand in a performatory contradiction, namely, that we would be simultaneously considering
and not considering the idea of a perfect being if we denied that existence is necessarily part of per-
fect being, much as having three angles is a necessary part of a triangle (Descartes 1984).
In his Critique of Pure Reason ([1787] 1933), Immanuel Kant voices an important criticism of
the a priori approach. He argues that “existence” does not operate as a real predicate, for it does not
add anything to the subject. Rather, to say that anything exists is simply to say that the concept of
that thing is exemplified in experience. If one were able to give a complete description of all the

Logic: Theism
properties of a given sort of thing, to say that the thing exists would not add anything qualitatively
different. From this consideration Kant concludes that ontological arguments fail, for there is no
qualitative difference between an existing “that than which nothing greater can be conceived” and
one that does not exist.
Despite widespread acceptance of Kant’s critique among professional philosophers today, an effec-
tive response to Kant can be mounted by considering the difference between any real object and the
mere thought of it: it makes all the difference whether there is, for example, a real virus than which
none greater can be conceived (e.g., one that can resist every possible remedy) and the mere thought
of one. In the case of God, the significance of the predication of existence is equally relevant.
Historically, this approach by a consideration of “existence” as a predicate in propositional logic
was first discussed centuries before Kant by the Muslim philosopher Abū Nasr al-Fārābī. To the
question “Does the proposition ‘Man exists’ have a predicate, or not?” Al-Fārābī explains that it is
possible to say yes and to say no, depending on the perspective under discussion:
To my mind, both of these judgments are in a way correct, each in its own way. This is so
because when a natural scientist who investigates perishable things considers this sentence (and
similar ones) it has no predicate, for the existence of a thing is nothing other than the thing itself,
and [for the scientist] a predicate must furnish information about what exists and what is excluded
from being. (Arabic 2012, 300, English translation from Rescher 1967 or 1968, 71–72)
To say, for instance, that “an apple is red, ripe, and exists,” the word “exists” is not giving new
information about the very nature or māhiyyah of the natural object called apple, while words such
as “red” and “ripe” do add information. But logicians might look at the same statement from the
perspective of its logical-grammatical structure. For this reason, Al-Fārābī said:
When a logician investigates this proposition [e.g., “Man exists”], he will treat it as composed
of two expressions, each forming a [constructive] part of it, and it [i.e., the composite
proposition] is liable to truth and falsehood. So, it does have a predicate from this point of
view. Therefore, the assertions are both together correct, but each of them only in a certain
way. (Arabic 2012, 300, English translation from Rescher 1967 or 1968, 72.)
Nicholas Rescher invokes al-Fārābī in his critique of Kant: “Thus due to close parallelism between
the logical and the grammatical relations (especially in Arabic) al-Fārābī unhesitatingly classes ‘exists’
as a legitimate grammatical (or logical) predicate. Even Kant agrees with this, affirming that: ‘zum
logischen Prädicate kann alles dienen, was man will.’” (1967 or 1968, 72). In regard to “essence”
and “existence,” the logical-grammatical structure indicates that the word “exists” operates (structur-
ally) as predicating “something” of the given subject “apple.” Thus the real question to be raised
here is this: what is that “something” that the word “exists” is predicating?
Al-Fārābī presumably agrees that “existence” does not add new information to the subject and
reminds us that the essence of anything is distinct from its existence. While essence makes things
“things” in the first place, existence operates as a mere attribute and as such cannot be a predicate
that attributes a property to its essence. On the other hand, “exists” as a predicate does not predicate
anything to “the necessary” since the necessary is necessary by essence and not by existence. In both
cases I read al-Fārābī as saying that the word “exists” cannot be considered as a predicate in the
sense that it does not attribute any new information to the essence. (See al-Fārābī’s contingency
argument in Sharh Risālat Zenon, 1349, 4–5, my translation. Also see Alfarabi’s philosophische,
1890, 57; and in Al-Allaf 2006, 109.)
GÖDEL ON THE ONTOLOGICAL PROOF

Kurt Gödel treats essential properties as positive properties. Accordingly, any being that is God or
“Godlike” should have essential properties that are “positive.” He concludes that we may say that
such a being necessarily exists if and only if every essential property for that being is necessarily
exemplified. By a “positive property” Gödel meant a property of no defect, and so it is an essential
property in making that which is under consideration. For example, “necessary existence” and “per-
fection” are positive properties in this sense. If a being has essential properties as positive properties,
Logic: Theism
then that being is God or Godlike. If this unique being necessarily exists in any possible world, then
it will necessarily exist in the actual world.
A detailed presentation and discussion of Gödel’s proof by Graham Oppy goes as follows:
Definition 1: x is God-like if and only if x has as essential properties those and only those
properties that are positive
Definition 2: A is an essence of x if and only if for every property B, x has B necessarily if and
only if A entails B
Definition 3: x necessarily exists if and only if every essence of x is necessarily exemplified
Axiom 1: If a property is positive, then its negation is not positive.
Axiom 2: Any property entailed by—i.e., strictly implied by—a positive property is positive.
Axiom 3: The property of being God-like is positive.
Axiom 4: If a property is positive, then it is necessarily positive.
Axiom 5: Necessary existence is positive.
Axiom 6: For any property P, if P is positive, then being necessarily P is positive.
Theorem 1: If a property is positive, then it is consistent, i.e., possibly exemplified.
Corollary 1: The property of being God-like is consistent.
Theorem 2: If something is God-like, then the property of being God-like is an essence of
that thing.
Theorem 3: Necessarily, the property of being God-like is exemplified. (Oppy 2017)
Oppy criticizes Gödel’s proof, saying that Gödel did not furnish a definition of what he calls
“positive property” (see also Maydole 2009, 574). I think we can understand the term “positive
property” in the context of talking about God and Godlike as a property of the greatest possible
thing or of perfection. Oppy suggests that if the positive properties are presented as a set, then there
is no sufficient basis to believe that such a set exists (Oppy 2017). In fact, Oppy’s point is similar to
the example that the conceiving of the most perfect island does not necessarily imply its existence;
but as we discussed with Anselm, the case with God is different.
Gödel’s argument seems to me to be a mathematical presentation of the ontological argument.
In my view, there is a problem with the way in which Gödel is proceeding here. If we think about
the positive, essential, and necessary properties of God or what is Godlike to be a set in the mathe-
matical sense, then one would need to talk of an infinite set of positive properties and of all possible
properties that are possibly exemplified, for by definition God is infinite. But no infinite set of pos-
itive properties exists in actuality; therefore, Gödel’s argument in regard to presenting positive prop-
erties in a set will not work, for the set must be infinite.
I should also mention that some of Gödel’s axioms are not self-evident. For example, “necessary
existence is positive” is not axiomatic unless in a very specified realm of being and a special context
of discourse. Yet self-evident statements should be axiomatic to all without limitation. Thus it is axi-
omatic only in the realm of God and his divine attributes.
Axiom 3 (that the property of being Godlike is positive) is not clearly axiomatic. If that posi-
tive property is not existence and if not existence alone, then the whole argument begs the ques-
tion, because that Godlike being must first exist in order to have the property X that is a positive
property. Thus, in the hierarchy of positive properties of such a being (God), existence must be
established first in order for all other positive properties to exist, but that is exactly what the argu-
ment is meant to prove. Thus to assume to be true what is needed to be proven to be true begs
the question.
PLANTINGA’S ARGUMENT
The ontological arguments can be better presented by using modal logic. Alvin Plantinga appeals to
concepts of possibility and necessity in order to prove that God exists. He uses terms such as

Logic: Theism
“maximal greatness” to designate a being that is possible, and, if possible at all, then in all possible
worlds. In his 1974 book The Nature of Necessity, Plantinga states his argument in the following
way:
1. A being has maximal excellence in a given possible world W if and only if it is omnipotent,
omniscient and wholly good in W; and
2. A being has maximal greatness if it has maximal excellence in every possible world.
3. It is possible that there is a being that has maximal greatness. (Premise)
4. Therefore, possibly, it is necessarily true that an omniscient, omnipotent, and perfectly good
being exists.
5. Therefore (by axiom S5), it is necessarily true that an omniscient, omnipotent and perfectly good
being exists.
6. Therefore, an omniscient, omnipotent and perfectly good being exists. (1974, Ch. 10)
The weight of the argument is carried by the modal statement that if it is possible that God exists,
then it would be necessary that God exists, because if it is possible for God to exist, then it will be
possible for God to exist in all possible worlds, and then of course it must be possible for God to
exist in the actual world, and therefore God exists (see also Oppy 2017; Pruss 2010). Plantinga thus
argues that if it is possible for that being to have the maximal greatness, then it is necessarily true
that an omniscient, omnipotent, and perfectly good being exists in at least one possible world and
thus it is necessary that such a being exists in all possible worlds.
All versions of the ontological arguments revolve around the relationship between “existence”
and the very definition of “God.” I think that the ontological argument might be further clarified
if the concept of God and his divine attributes would be logically presented in relation to unique
quantification. In the following section I present my own understanding of the divine attributes
as unique quantification to God.
THE DIVINE ATTRIBUTES AND UNIQUE QUANTIFICATION

Let us now turn to a related topic, the logic employed in discussions of the divine attributes. In the
Islamic tradition, as in other monotheistic religions, the attributes are understood to be identical
with the essence of God. For this reason they operate as predicates in propositional logic, for they
do not add anything new to the subject but rather refer to the richness of the divine essence.
For example, from the perspective of Gottlob Frege or Bertrand Russell, the statement “the
Most Merciful (al-Rahman) is God (Allah)” would appear to be one in which A=B. Passages like
the following one from the Qurʾan make clear that the words used for the divine names and attri-
butes ought to be interpreted as taking these words to be the same in reference but different in
meaning: “Say, ‘Call upon Allah or call upon the Most Merciful. Whichever [name] you call—to
Him belong the best names.’ And do not recite [too] loudly in your prayer or [too] quietly but seek
between that an [intermediate] way” (Qur ʾan 17:110). Thus “the Merciful” (al-Rahman) and
“God” (Allah) have different meanings but the same reference. Logically, the relationship is like
Frege’s famous example about the morning star (A) and the evening star (B). Both A and B refer
to the same star, but under different descriptions. To say, “the morning star is the morning star”
(A=A) is a clear statement of identity in which the words are used with the same sense and same ref-
erence. To say “the morning star is the evening star” (A=B) is a different sort of identity, for the
words both refer to the same body (Venus) but use these words in different senses. Put schemati-
cally, Frege’s symbolism can be helpful in understanding the logic of the divine names:
A=A is identical in sense and reference,

A=B is identical in reference but different in sense,
Then B is just another name of the same thing A, and
therefore, by the rule of substitution: (A=A) = (A=B).
Logic: Theism
THE SET OF ONE MEMBER

Philosophically considered, monotheism can hold that all the words used for the divine names and
attributes should be treated logically as belonging to a set with one member and only one member.
As predicates, these words have diverse meanings, but they exclusively refer to that member and that
member alone. For example, in the set of eternal beings, there is only one member, and that unique
member is God.
It is possible to represent this understanding of the words used for the divine names and attri-
butes in the following way:
A = the negation of all that is not A, and A = nothing but A, denoted in logical
symbols by the universal quantification (8) as a negation of all that is not a:
∼8 (A), it is not the case that in any given A.
Using the existential quantifier, one can, for example, express a related point in this way:
∼ (9x) HBx, human beings do not exist with some specific divine attribute x.
Or:
∼ (9x) (HBx ^ MMx), not a single human being exists with the attribute
x such as that x is most merciful.
Thus to say “God is Eternal Being” will be read by Frege as A=B, but what is intended by mono-
theists is that there is no other “being” that is eternal except God, for God is the unique member
of a set with one and only one member, and the words “Eternal Being” (B) have a different meaning
but the same reference as (A). Therefore, (A=A) = (A=B) by the rule of substitution.
TAUTOLOGY, DESCRIPTION, AND UNIQUE QUANTIFICATION

According to Russell, a statement of the A=A sort is essentially different from one of the A=B type.
The first one is tautological (such as “John is John”), while the second is descriptive. Descriptive
statements can be quite deceptive, according to Russell, because they can be used to refer to some-
thing that does not exist, such as “the unicorn is white.” But the statement “God is God” is not a
mere tautology of the A=A sort, for there is no other thing that even resembles him and nothing
that can be added to him. This follows from understanding that God is the perfect being (that than
which nothing greater can be conceived). Furthermore, what we say about God when we use words
as divine names and attributes is not merely adding descriptive terms that would be true if God
existed, as in the above example of calling the unicorn white. Nor is what we say an instance of
the descriptive statements that we use to record the ways in which individuals participate in realities
other than their own essence, as when we say that a certain human being has a certain quality or a
certain relationship, e.g., “John is smart” or “John is old.” Each of these descriptions describes him
by adding something to John in a way that makes the statement nontautological. In the case of
God, however, whatever we explicitly mention of the perfect being merely makes explicit what is
already part of God’s essence and thus is already included in his very definition.
Given the difficulties found in Russell’s account, monotheists often take a different logical route
for discussing the divine essence, namely, by an appeal to negation. We find this approach, for
instance, when the Qur ʾan says: “[He is] the Creator of the heavens and the earth: He has made
for you pairs from among yourselves, and pairs among cattle: by this means does He multiply
you: there is nothing whatever like unto Him, and He is the One that hears and sees (all things)”
(Qur ʾan 42:11; emphasis mine). In this case one can proceed as follows:
A= the negation of all and everything, and the affirmation of Him alone.
Using the logical symbol of uniqueness quantification (9!), the statement reads in this way:
9! a: P(a) meaning that there is exactly one (one and only one) a such that P(a) is true.
There is one and only one being (a) to whom the attribute eternity (P), for example, is true.

Logic: Theism
The above statement can also be presented by negation. Using the logical symbol for the exis-
tential quantifier (9), the statement reads as:
(9x)(∼Ax ^ ∼Bx ^ ∼Nx), there exists an x such that x is not an
A and not a B and not anything else that N might stands for.
The above formulation, however, is hard to use in definitions because in order to complete the def-
inition we would have to negate all the members of an infinite set, which is a process that can only
be suggested and can never be finished or actualized. Therefore, rather than persist in Russell’s
method of description, it is better to use a logical description for the divine attributes that recognizes
that all the names of God require unique quantification. In this way I move from Russell’s theory of
definite description (descriptive statement) to what I suggest as the theory of unique description (by
unique quantification) that works better in the area of the divine attributes.
In his book Naming and Necessity, Saul Kripke holds that a rigid designator and definite
description are not the same, and actually exclude each other. This distinction can help to provide
the needed alternative. Kripke explains: “Let’s call something a rigid designator if in every possible
world it designates the same object, a nonrigid or accidental designator if that is not the case”
(1972, 48; here and below, italics in the original). Kripke then argues that proper names are rigid
designators. To paraphrase his example: In 1970 Richard Nixon was president of the United States.
While he might not have been president then (someone else might have), it is not the case that
Nixon might not have been Nixon (though he might not have been called “Nixon”).
Kripke’s notion of a designator that is “strongly rigid” has helpful application in relation to
the divine attributes. He writes: “A rigid designator of a necessary existent can be called strongly
rigid ” (1972, 48). When, for instance, we call God the Necessary Being (‫)ﻭﺍﺟـﺐ ﺍﻟـﻮﺟـﻮﺩ‬, the name
is not only a rigid designator but one that is rigid in the “strongest” sense (what Ibn ‘Aqil calls
designation in the absolute sense) since there is no other possibility that God might not have been
eternal and no possibility for any other being to be of this sort. According to Ibn ‘Aqil (from
Baghdad) a proper name is the name that absolutely designates the thing named, without any
need for restriction by extra description. However, I disagree with both Kripke and Ibn ‘Aqil on
the definition of a proper name, because to me a proper name is a description of that which has
unique quantification.
UNIQUE QUANTIFICATION
Propositions of unique description (to which I will give the letter U) have not been given attention
by logicians. I define it as follows: A unique proposition (U) is a categorical proposition in which the
subject is the one and only unique member of a set and in which the predicates are uniquely related
to this subject and this subject alone. In other words, a U proposition uniquely describes its one and
only member by this specific description.
Thus a proposition of the form “A U B” can be read as holding that B is predicated to one
(unique) member and only one member and that member is A, and B is a specific description that
is only applicable to A. The following sentences provide some examples:
Mary is the virginal mother of Jesus.

Moses is the one (the prophet) who talked to God.
God is the most merciful.
The U proposition thus differs from statements that have words such as “the only,” for the expres-
sion “the only” can be translated into a universal affirmative proposition using the word “all.” The
statement “Students are the only ones who will receive a discount” can be translated as “All of those
who will receive discount are students.” But in U propositions there is no room for either “all” or
“some” because a U proposition designates one and only one unique member: there are no other
members in the set that even bear a resemblance to this unique member, and of course no other
identical member. Furthermore, a U proposition is not equivalent to Russell’s definite description
Logic: Theism
because it is about a unique description. (Compare this with a categorical proposition according to
Aristotle [1941, chap. 4, 17a].)
TWO PSEUDOPROBLEMS
There are two problems or puzzles that I should clarify and refute in this section. The first is the
stone puzzle: can God create a stone that he himself cannot carry? The second is the Euthyphro
problem, from Plato: is something right because God commanded it, or did God command it
because it is right? These pseudoproblems are related to the question of the divine attributes; there-
fore, let’s turn to the divine attributes first.
In the Islamic tradition, there are many names and divine attributes, but there are some very
essential attributes such as life, knowledge, will, power, hearing, and seeing, and there is nothing like
unto God. All divine attributes are consistent with each other and are logically hierarchical. In this
context we can understand that if there is something that is not performed by God, it does not
mean it is impossible on account of that specific attribute; rather, it means that it is inconsistent
with another divine attribute in the systematic rank. The specific attribute must be looked at in
the contextual hierarchy. Thus some puzzlement can be resolved based on the above understanding.
The Mistake of the Stone Puzzle. The stone puzzle concerns the power of God. If God can cre-
ate a stone that he himself cannot carry, then he is not omnipotent (all-powerful), and if he cannot
create such a stone, then he is also not omnipotent, therefore God is not omnipotent. By definition,
God is omnipotent (all-powerful); this definition means that he can do many things, and all things
are within his power. Not creating such a stone is not a limitation of his power, it is something that
is the case because of his knowledge: God is omniscient, and he will not create a stone like this
because he knows that such an action is nonsense and has no purpose in the scheme of creation
itself. God does not make nonsense things, and he knows that no purposeful actions are not fitting
with respect to his majesty:
“And We did not create the heaven and earth and that between them in play.” (Qur ʾan 21:16)
“Had We intended to take a diversion [a pastime], We could have taken it from [what is] with
Us—if at all We were (inclined) to do so.” (Qur ʾan 21:17)
“Verily, all things have We created in proportion and measure.” (Qurʾan 54:49)
In William Rowe’s discussion of the consistency of the divine attributes, he says that creating a
stone that is so heavy that God cannot lift it is an action of doing something that is not consistent
with one of God’s essential attributes, the attribute of omnipotence. As he sees it, the “proper solu-
tion to the puzzle is to say that God cannot create such a stone any more than he can do an evil
deed” (2007, 8). But this is not a proper solution, because the attribute omnipotence does not cre-
ate or bring anything into existence on its own. In fact, this attribute is only the fourth attribute in
the logical hierarchy of the divine attributes: first, life; then, this being is all-knowing; then, his
absolute will; then, based on knowledge and will, the power or the ability of God to do things.
Thus his action is not related directly to his being omnipotent, rather to him as omniscient being.
The Euthyphro Problem: Plato’s Mistake. Another puzzle with regard to the divine attributes is
one related to ethics and not to the ontological argument, but it is useful to this discussion in order
to show that some philosophers have confused thinking about God and his divine attributes. In Pla-
to’s Euthyphro, Socrates raises the issue about moral actions: is an action right because God com-
manded it, or did God command it because it is right? If it is right because God commanded it,
then everything that God commands will be right, but this is contradictory to the goodness of
God, and it would make the commands of God arbitrary. On the other hand, if God commanded
it because it is right, then it seems there is a list of moral standards that coexisted eternally and inde-
pendently from God and that God selects from it what is right and commands it. This is a theolog-
ical contradiction.

Logic: Theism
This issue stands as a clear example of the misunderstanding of divine attributes; the issue
revolves around the “goodness” of God. But God does not work based on his goodness alone. This
is only one attribute, and it must be harmonious with the rest of the divine attributes, especially
knowledge, because God is one and undivided into parts.
To solve the problem, let us consider this example: Imagine that a woman, K, is so interested in
wearing perfumes every day that she has bottles of many different brands. She is obsessed by smell-
ing good; having a fragrance is one of the essentials in her life. One day she notices that all her bot-
tles are empty except for one small bottle, but that small bottle holds bug spray. Now, do you think
that K will spray that on herself to have some fragrance, since fragrance is essential to her? Of course
not, because she knows that bug spray does not have a good fragrance. Thus, she is not operating on
the fragrance; rather, she is operating based on her knowledge that she knows what smells good and
what does not. Thus, knowledge and not fragrance is what is most essential here (having in mind
that K is an intelligent being).
The same explanation can be given to the issue raised by Plato; it is not the goodness of God
that is the issue; rather, it is the knowledge of God, and of course, his will and power after that.
God knows what is good, therefore God commands that specific thing on the basis of knowledge
and not on goodness. In fact, Plato’s conception of God is inadequate and is limited to his culture
and maybe to the images of the gods of Olympus. However, God as conceived in monotheistic reli-
gions is very different from God as conceived in Greek mythology. Rowe explains: “Since doing evil
is inconsistent with being perfectly good, and since being perfectly good is a basic attribute of God,
the fact that God cannot do evil will not conflict with the fact that he is omnipotent” (2007, 8).
Rowe’s explanation is still limited to the goodness of God as a basic attribute; but goodness is not
responsible for choosing or refraining from an evil action. Such an action is based on and related
to knowledge, will, and then power. It is not based on goodness.
Unfortunately, this issue was misplaced for about 2,400 years in the goodness of God, where it
does not belong.
COSMOLOGICAL ARGUMENTS
The ontological argument is only one way to prove the existence of God. This section discusses the
cosmological arguments, also called contingency arguments or first cause arguments.
Whatever exists in the world of our experience exists contingently—that is, it is the result of
causation and would not exist if it had not been caused. All versions of the cosmological argument
begin with the mention of some contingent reality and argue for the necessity of God as a first and
uncaused cause as the ultimate explanation for what is contingent. The argumentation invariably
makes use of a posteriori knowledge derived from experience of the world and its caused objects
in the course of leading to a conclusion that affirms the necessity of admitting the uncaused cause.
Not all versions of the cosmological principle have been successful, and this seems to be the result,
at least in part, of using theories once thought to have scientific cogency but that have since been
discarded.
Aristotle, for instance, presented a version of the contingency argument that was based on the
static physics of antiquity, which held that material objects are in a static state and do not move
unless there is a force or a cause that caused them to move. Since there is motion in the world,
therefore there must have been a cause for that motion, and yet the chain of causes and effects can-
not go back to infinity, or we would not have provided a sufficient explanation for the motion that
we presently observe. Therefore we must admit the reality of a first cause that is the cause of all
other motion and that must be the unmoved and uncaused first mover; otherwise, there will not
be an adequate explanation for the motion that we actually observe in the present.
Aristotle also provided a second argument that made use of the idea of final causality. In Book
Lambda of the Metaphysics he envisions the motion of an eternally existing world as the result of the
attraction of each thing to the one necessary being, namely, God conceived of as continuously
engaged in a kind of self-contemplative intellectual activity. This type of argumentation, however,
Logic: Theism
presumes the eternity of the world, an assumption that is incompatible with all forms of contempo-
rary science and its theory of the big bang.
Near the beginning of his Summa theologiae, Thomas Aquinas offers three versions of the cos-
mological argument among his classic set of the Five Ways. The first argument is Aristotelian and
based on the need to posit an unmoved mover to explain the motion that can be observed in
the world, so as to avoid an infinite regress of explanation. Similarly, the second argument states the
need to posit an uncaused cause to explain the observed reality of causes that are themselves the effects
of previous causes. The third argument is based on the contingency of observed reality and argues for
the need to posit God as the necessary being if the contingent existence of things in this world is to
be explained. Aquinas’s arguments are based not only on the thought of Aristotle but also on such
thinkers as Al-Fārābī, Ibn Sīnā (Avicenna), and Moses Maimonides.
THE KALĀM ARGUMENT: AL-GHAZĀLĪ ON THE EXISTENCE OF GOD

In Islamic philosophy the cosmological argument is made in various ways. Al-Ghazālī structured an
argument about the existence of God of the sort that later came to be known as the kalām argu-
ment. In his book Iḥyā’ culūm ad-dīn (The revivification of the religious sciences), he proposes an
argument of the following sort:
1. An originated phenomenon cannot come into existence without a cause.
2. The world is an originated phenomenon, and so
3. it cannot come into existence without a cause.
The conclusion of this argument speaks about the entire cosmos and reflects on the impossibility of any
effect’s existing without a cause that is mentioned in the first premise. This argument is formally valid.
Much of the critical discussion about this argument therefore centers on the second premise. How do
we know that the universe began to exist? Today we might consider the evidence for the big bang, but
even apart from that, al-Ghazālī thinks that the entire cosmos is not independent of originated phe-
nomena, for what is not independent of originated phenomena is itself originated. Once its foundation
in originated phenomena has been established, its need for an origination in time becomes a necessity.
An objection can easily be raised against the second premise of the argument (the world is an
originated phenomenon). To this al-Ghazālī answers in the following way:
1. If the universe never began to exist (has no beginning in time), then the universe would be
eternal.
2. If the universe were eternal, then a series of actual infinite events would have already elapsed.
3. But we know that an actual infinity is impossible.
4. Therefore, the universe must have had a beginning in time.
THE QUR’ANIC ARGUMENT: THE CREATION ARGUMENT (DALIL AL-KHALQ WAL EJAD)
Although the following argument is quoted from the revelation, it is purely logical and corresponds to
the form of a disjunctive syllogism. In fact, revelation goes further to challenge the mind. The Qurʾan
offers an argument about the way human beings came into existence. Humans are either eternal and
uncaused in their existence, or were caused to exist by something else. For the Qurʾan, the notion that
human beings are eternal is clearly contradictory to their very nature as temporal beings. God says in
the Qurʾan: “Were they created by nothing, or were they the creators [of themselves]? Or did they
create the heavens and the earth? Rather, they are not certain” (Qurʾan 52:35–36). These verses are
interrogative, but they are inquiring in a disaffirmative way: “were they created by nothing?” Is it pos-
sible for a temporal being such as a human being to come into existence without a cause?
• Human beings are either caused or uncaused (eternal)
• They cannot be uncaused
• Therefore, they were brought into existence by a cause superior to them.

Logic: Theism
This form of reasoning is a disjunctive syllogism in which the negation of one of its disjuncts makes
it a valid form:
• Either A (caused) or B (uncaused)
• Not B (cannot be uncaused)
• Therefore, A (human existence must have a cause)
The second part of the verse raises another question: since human beings were caused to be and
brought into existence, then is it that they (themselves) are the cause of their own existence: “were
they the creators [of themselves]?” In my view, it is impossible for a “thing” to bring itself into exis-
tence, because it will be in existence prior to its existence, which is contradictory.
• Since the first argument already established the valid conclusion that human beings were brought
into existence and caused by something else, then this part of the verse is logically establishing
another argument on that conclusion: either human beings are the cause of their own existence,
or they were caused by something else.
• We know that they are not and cannot be the cause of their own existence. (If they were, then
they could skip death and give themselves extended existence, or they could make for themselves
a perfect existence, which clearly they cannot.)
• Therefore, they were brought into existence by a cause other than themselves.
This form of reasoning is also a disjunctive syllogism in which the negation of one of its disjuncts
makes it a valid form:
• Either this world has a cause, or it is uncaused.
• It is impossible to be uncaused (a physical world cannot be actually infinite).
• Therefore, the world has a cause.
Both arguments above are valid disjunctive syllogisms that negate one of the disjuncts and have the
valid form:
Either P or Q
Not Q
Therefore, P
ESSENCE AND GENERATION ARGUMENT

Al-Kindī presents a proof for the existence of God based on essence and generation (al-māhiyya wa’l-
ḥudūth). He thinks that it is not possible for a thing to be the cause of the generation of its essence;
its becoming a being is either from something or from nothing. Thus, we have four possibilities:
First: a nonexistent thing and its essence is nonexistent
Second: a nonexistent thing and its essence is existent
Third: an existent thing and its essence is nonexistent
Fourth: an existent thing and its essence is existent
The first is impossible, because it is nothing and its essence is nothing. “Nothing” is neither a
cause nor an effect (cause and effect are predicated only of something). Therefore it is not the cause
of the generation of its essence.
The second is, of course, with a similar proof, impossible. It is also impossible by another proof
based on the law of identity. Al-Kindī says:
1. If it is nonexistent and its essence is existent, then its essence is different from the thing
itself.
Logic: Theism
2. Thus, its essence would not be it (the thing).

3. However, the essence of everything is that thing itself.
4. Therefore, a thing would not be itself and it would be itself, but this is an impossible
contradiction (according to the law of identity).
5. Therefore, a thing cannot be the cause of the generation of its essence.
The third is impossible for the reason above, according to the law of identity.
The fourth is also impossible, al-Kindī says:
1. If the thing is an existent thing and its essence is existent, and it were the cause of the generation
of its essence, then its essence would be its effect.
2. But the cause is different from the effect.
3. Therefore, it would be the cause of its essence while its essence would be its effect.
4. Thus, its essence would not be it (the thing). However, the essence of everything is that thing
itself (according to the law of identity).
5. But from this argument it follows that it would not be itself, and that it would be itself. This is
an impossible contradiction.
6. Therefore, a thing cannot be the cause of the generation of its essence.
Al-Kindī tries to show that matter, motion, and time are inseparable from each other. If one of
them were eternal, then the rest would be eternal as well. Al-Kindī then presents several arguments
to prove that the eternity of any of these three is “impossible” and, in fact, self-contradictory. The
idea of an eternal world causes contradiction when the concept of actual infinity is applied. (For
details, see Al-Allaf 2006).
THE CONTINGENCY ARGUMENT: “THE POSSIBLE” AND “THE NECESSARY”

Al-Fārābī’s main argument is based on two essential concepts: “the possible” (al-mumkin) and “the
necessary” (al-wajib). The possible is everything that has existence not from its essence. The exis-
tence of every existent is either necessary or possible. The necessary existent is either necessary by
itself and by its essence, or it is necessary by something else other than itself—if it is necessary by
something other than its essence, then it is called “possible”:
Premise 1: The possible, by definition, is only possible by something else; thus it must be
preceded by a cause in order to exist (a cause by which it came into existence).
Premise 2: It is unreasonable to trace the chain of causes and effects to infinity (without reaching
an end).
Therefore, there must be an existent being, whose existence is necessary, and this being is uncaused,
and most perfect (al-Fārābī 1890, 57).
Al-Fārābī clarifies the second premise that possible causes cannot regress to infinity because
cause and effect are connected in a relation or chain. Each of them is an intermediary being, caused
in regard to one aspect, and a cause in regard to another. Everything that is intermediary must have
a limit and a limit is an end. Possible things must depend on the existence of a necessary existent
that is unaffected by causes, whether material, formal, final, or efficient.
Al-Fārābī argues that the “possible” cannot be the cause of its essence by a method similar to
that of al-Kindī on essence and generation. He argues that the possible existent cannot be the cause
of its existence but adds another point, that causation is an asymmetrical relationship: if A is the
cause of B, then B cannot be the cause of A. Thus the possible cannot be the cause of itself. If B
is the cause of A and A is the cause of B, then B is the cause of itself, and that thing’s existence must
precede its existence, and this is unsound.
Ibn Sīnā thinks that “the necessary existent does not have a cause, while the possible does have
a cause. The necessary existent is necessary in respect to existence in all of its aspects. It is not

Logic: Theism
possible for this necessary, by itself, to be coequal with another existent, so that each of them is
equal with the other in respect to the necessity of existence; thus both cannot necessarily accompany
each other. It is also not possible for that necessary being to be composed of a multitude at all. It is
not possible for the true nature of the necessary being to be shared with something else in any way
whatsoever.” (Ibn Sīnā, trans. Hyman, 1960; Ibn Sīnā 1985, 262–263; and see Hyman and Walsh
1973, 241–255).
When the possible is considered in respect to itself, it has both its existence and nonexistence
from a cause. If it exists, then existence (as distinguished from nonexistence) has come to it, and,
if it does not exist, then the possible is in a state of nonexistence (as distinguished from existence).
Now, each of these two attributes (existence and nonexistence) must come to it (A) either from
something else, or (B) not from something else.
(A) If its existence comes from something else, then that other is the cause.
(B) However, it is a contradiction for it to be possible as coming from something else and not
coming from something else, because it is evident that everything that does not exist at first
and then exists is determined by something other than itself.
Also, the essence of the possible is either sufficient for this determination (for its existence or non-
existence) or the essence is not sufficient for it.
Now, if the essence is sufficient for one of the two attributes (existence and nonexistence),
then that thing is something whose essence is necessary through itself (necessary being). However,
it was assumed that it is not necessary. Therefore, this is a contradiction, and its essence cannot be
sufficient. If the existence of its essence is not sufficient for its existence but something else
bestows its existence upon it, then its existence proceeds from the existence of some other thing
different from it and this is necessarily its cause. Therefore, the possible has a cause that makes
one of these two attributes necessary for it, not through itself, but through a necessary existent
cause.
Ibn Sīnā also considers the argument on the regression of causes and effects to be impossible.
If the series of causes and effects would go on to infinity, then it would not be. (Ibn Sīnā meant
that its existence would not be determined because the chain of infinity would not come to an
end.) This is absurd because the regression to infinity is impossible and that through which it is
determined (the cause) would not yet exist. However, it was assumed that the cause exists. There-
fore, it is clear that everything whose existence is possible will not exist unless it is necessary in
relation to its cause.
THE TELEOLOGICAL ARGUMENT (ARGUMENT FROM DESIGN):

THE AESTHETIC ORDER
Still another important kind of argument for the existence of God is the teleological argument, also
called the argument from design. This argument is discussed briefly in this section.
The irreducible complexity of the way things are, the order of things, the functionality of the
parts within the whole, the meticulous design in the world on micro and macro levels, and the
end-directedness found in so many things all led various thinkers to the conclusion that there must
be something standing beyond nature that is responsible for it. From the time of Anaxagoras, phi-
losophers have formulated arguments for the reality of such a nous, that is, the intelligent source
of order within the cosmos. Aristotle includes a sophisticated account of teleology (final causality)
in his system.
The thrust of the teleological argument in its various forms challenges the idea that the order of
the world could be a matter of coincidence. In Islamic philosophy, most Muslim philosophers sup-
port their teleological arguments by citing verses from the Qur ʾan. In monotheistic religions such as
Islam, and in Islamic scholastic theology, God created everything for a purpose and all are well
designed to serve certain goals: “Verily, all things have We created in proportion and measure”
Logic: Theism
(Qur ʾan 54:49). “It is not allowable for the sun to reach the moon, nor does the night overtake the
day, but each, in an orbit, is swimming” (Qur ʾan 36:40). The Qurʾan talks about existent things as
an indication of the miraculous action of God, and then argues that deep reflection on the cosmos
and the human being will lead us to realize the existence of the Creator: “We will show them Our
Signs in the universe, and in their own selves, until it becomes manifest to them that this [the
Qur ʾan] is the truth. Is it not sufficient in regard to your Lord that He is a Witness over all things?”
(Qur ʾan 41:53).
One classic version of the teleological argument is found in the Fifth Way of Thomas Aquinas:
We see that some things lacking cognition, such as natural bodies, work toward an end, as is
seen from the fact that they always (or at least usually) act the same way and not accidentally,
but by design. Things without knowledge tend toward a goal, however, only if they are guided
in that direction by some knowing, understanding being, as is the case with an arrow and
archer. Therefore, there is some intelligent being by whom all natural things are ordered to
their end, and we call this being “God.” (Summa theologiae I. q.2 a.3)
The logic behind the argument is valid, negating the possibility of natural things’ moving toward an end
without the aid of an intelligent being, but things do move toward an end, therefore, such a being exists.
THE FINE-TUNING ARGUMENT

The version of the teleological argument that seems to be the strongest is the fine-tuning argument,
for it has much support from science. The physical constants that are necessary to bring or create life
are so precise that the assumption of the existence of an intelligent being behind that will stand as a
good explanation. The anthropic principles shed light on this theory. Stephen Hawking, in A Brief
History of Time, admits this point: “The laws of science, as we know them at present, contain many
fundamental numbers, like the size of the electric charge of the electron and the ratio of the masses
of the proton and the electron.… The remarkable fact is that the values of these numbers seem to
have been very finely adjusted to make possible the development of life” (1988, 125). (Among the
many constants, scientists can count the ratio of proton to electron mass, the ratio of the number of
protons to the number of electrons, the ratio of proton to electron charge, the expansion rate of the
universe, the mass density of the universe, carbon-12 to oxygen-16 nuclear energy level ratio, etc.
For a detailed discussion on this topic, see the chapter by Robert Spitzer and James Sinclair.)
Bibliography
Al-Allaf, Mashhad. The Essential Ideas of Islamic Philosophy: A Brief Survey. Ayer, A. J. Language, Truth, and Logic. 2nd ed. London: Gollancz, 1948.
Lewiston, NY: Edwin Mellen Press, 2006.
Descartes, René. Discourse on Method, and The Meditations. Translated by
Al-Allaf, Mashhad. “Unique Quantification: Beyond Russell and Kripke, F. E. Sutcliffe. Harmondsworth, UK: Penguin; New York: Viking
Islamic Philosophy of Naming.” The Journal of Humanities and Social Penguin, 1968.
Sciences Review (HSSR), Volume 06, Number 01, 2016. pp. 67–86.
Descartes, René. The Philosophical Writings of Descartes. Vols. 1–3. Edited and
Anselm, Saint. Proslogium; Monologium: An Appendix in Behalf of the Fool by translated by John Cottingham, Robert Stoothoff, Dugald Murdoch, and
Gaunilo; And Cur Deus Homo. Translated from Latin by Sidney Norton Anthony Kenny. Cambridge, UK: Cambridge University Press, 1984.
Deane. With an introduction, bibliography, and reprints of the opinions of
leading philosophers and writers on the ontological argument. Chicago: Descartes, René. Meditations on First Philosophy. Translated by John
Open Court, 1903, reprinted 1926. The above text is part of the Internet Veitch. Classical Library. 2001. http://www.classicallibrary.org/descartes
Medieval Sourcebook. The sourcebook is a collection of public domain and /meditations/. First published in Latin in 1641.
copy-permitted texts related to medieval and Byzantine history. Fārābī, Abū Nasr Al-. Sharh Risālat Zenon al-Kabir al-yunani (Commen-
Anselm, Saint. Proslogion. In St. Anselm’s Proslogion. Edited and translated by tary on the treatise of the great Zeno of the Greec), Hyderabad, Indea,
M. J. Charlesworth. Oxford: Clarendon, 1965. 1349 AH.
Anselm, Saint. Proslogion. Translated by David Burr. Internet Medieval Fārābī, Abū Nasr Al-. Alfarabi’s philosophische. Edited by Friederich Dieterici.
Sourcebook. Fordham University Center for Medieval Studies. 1996. Leiden, Netherlands: Brill, 1890.
https://sourcebooks.fordham.edu/source/anselm.asp.
Fārābī, Abū Nasr al-. Al-Thamara al-Mardhiyyah fi Ba’dh al Risalat
Aristotle. On Interpretation. In The Basic Works of Aristotle. Edited by Richard al-Fārābīyyah. Collected and edited by Friederich Dieterici, reedited by
McKeon; translated by E. M. Edghili. New York: Random House, 1941. Emad Nabil, Dar al-Fārābī. Beirut, 2012.

Logic: Theism
Findlay, J. N. “Can God’s Existence Be Disproved?” Mind 57, no. 226 Oppy, Graham. Ontological Arguments and Belief in God. Cambridge:
(1948): 52, 176–183. Cambridge University Press, 1995.
Frege, Gottlob. The Foundations of Arithmetic. (1787.) Translated by J. L. Oppy, Graham. “Gödelian Ontological Arguments.” Analysis 56, no. 4
Austin. 2nd ed. Oxford: Blackwell, 1974. Die Grundlagen der Arithmetik (1996): 226–230.
(original German text). http://www.marilia.unesp.br/Home/Instituicao
Oppy, Graham. “Response to Gettings.” Analysis 60, no. 4 (2000): 363–367.
/Docentes/RicardoTassinari/DGA.pdf. Site maintained by Alain Blachair,
Académie de Nancy-Metz. Oppy, Graham. Arguing about Gods. Cambridge: Cambridge University
Press, 2006.
Gaunilo. “On Behalf of the Fool.” In St. Anselm’s Proslogion. Edited and
translated by M. J. Charlesworth. Oxford: Clarendon, 1965. Oppy, Graham. “The Ontological Argument.” In Philosophy of Religion:
Classic and Contemporary Issues, edited by Paul Copan and Chad Meister,
Gaunilo. “How Someone Writing on Behalf of the Fool Might Reply to All
112–126. Malden, MA: Blackwell, 2007.
This.” Translated by David Burr. Internet Medieval Sourcebook.
Fordham University Center for Medieval Studies. 1996. https:// Oppy, Graham. “Ontological Arguments.” In Stanford Encyclopedia of
sourcebooks.fordham.edu/source/anselm.asp. Philosophy, edited by Edward N. Zalta. Summer 2017. https://plato.stanford
.edu/archives/sum2017/entries/ontological-arguments/.
Hawking, Stephen. A Brief History of Time: From the Big Bang to Black Holes.
New York: Bantam, 1988. Plantinga, Alvin. God and Other Minds: A Study of the Rational Justification of
Belief in God. Ithaca, NY: Cornell University Press, 1967.
Hyman, Arthur, and James J. Walsh, ed. Philosophy in the Middle Ages: The
Christian, Islamic, and Jewish Traditions. Indianapolis, IN: Hackett, 1973. Plantinga, Alvin. The Nature of Necessity. Oxford: Clarendon, 1974.
Ibn Sīnā. Ash-Shifā’. In ‘Ilāhiyyāt. Edited by G. C. Anawati and Sa’id Zayed. Pruss, Alexander R. “A Gödelian Ontological Argument Improved.”
Translated by Arthur Hyman. Cairo: Organisation générale des Religious Studies 45, no. 3 (2009): 347–353.
imprimeries gouvernementales, 1960.
Pruss, Alexander R. “The Ontological Argument and the Motivational
Ibn Sīnā. Kitāb an-najāt. [Book of safety.] Beirut: Dār al-Āfāq al-Jadīda, 1985. Centres of Lives.” Religious Studies 46, no. 2 (2010): 233–249.
Kant, Immanuel. Critique of Pure Reason. Translated by Norman Kemp- Rescher, Nicholas. “The Ontological Proof Revisited.” Australasian Journal
Smith. 2nd ed. London: Macmillan, 1933. First published 1787. of Philosophy 37, no. 2 (1959): 138–148.
Kripke, Saul A. Naming and Necessity. Cambridge, MA: Harvard University Rescher, Nicholas. Studies in Arabic Philosophy. Pittsburgh: University of
Press, 1972. Pittsburgh Press, 1967 or 1968.
Malcolm, Norman. “Anselm’s Ontological Arguments.” Philosophical Review Rowe, William. Philosophy of Religion: An Introduction. 4th ed. Belmont, CA:
69, no. 1 (1960): 41–62. Thomson Wadsworth, 2007.
Maydole, Robert E. “The Ontological Argument.” In The Blackwell Thomas Aquinas. Summa theologiae. Translated by David Burr. Internet
Companion to Natural Theology, edited by William Lane Craig and J. P. Medieval Sourcebook. Fordham University Center for Medieval Studies.
Moreland, 553–592. Oxford: Wiley-Blackwell, 2009. 1996. https://sourcebooks.fordham.edu/source/aquinas1.asp.
TOPIC 3
Logic: Atheism
Peter Millican
Gilbert Ryle Fellow and Professor of Philosophy
Hertford College, Oxford University, UK
This chapter discusses whether theism can be either established or refuted on broadly logical grounds, and
considers the role of logic in theistic and anti-theistic argument. In the first section, I briefly review some
systems of formal logic, highlighting some issues about logical proof and providing background to the
subsequent discussion. In the next section, I make some important general points about logic and its limits:
what we can reasonably expect to achieve by logical argument. I then turn to the question of whether theism
can be refuted a priori on the basis of internal inconsistency, discuss how it might be defined so as to evade
such refutation, and briefly consider arguments for and against theism that are aprioristic in the sense of
being based on minimal empirical data (such as the existence of contingent things, or the existence of evil).
The last main section examines the ontological argument, which purports to establish theism purely a priori.
FORMAL LOGICS
The word logic can be understood in various ways, but it is now most commonly thought of as a
formal discipline in which arguments are regimented, represented, and assessed using well-defined
methods (e.g., the rules of so-called natural deduction, using semantic tableaux, or by computa-
tional methods such as resolution). Traditionally, the main point of this exercise has been to codify
rules of right reasoning and to give an objective way of distinguishing good from bad
arguments—as, for example, in Aristotle’s identification of the valid forms of syllogism. Medieval
logicians such as Boethius, William of Ockham, and Jean Buridan (among many others) extended
Aristotle’s theory in various interesting ways, but many early modern philosophers, most influen-
tially René Descartes and John Locke, were disdainful of scholastic formal logic—criticizing, for
example, its lack of practical use for the discovery of novel inferences—and turned their attention
instead toward nonformal study of the human reasoning faculty. On the other side, however,
Thomas Hobbes gestured toward a computational theory of ratiocination, while Gottfried Wil-
helm Leibniz went considerably further, though his influence in this direction derived more from
his invention of calculating machines than his theoretical work, which was significant but unpub-
lished until much later.
Radical change came only in the nineteenth century with the development of symbolic logic,
most notably by George Boole and Gottlob Frege. Boole aimed to give a mathematical foundation
for Aristotelian theory (and his logic is correspondingly limited in scope), while Frege was seeking a
calculus of reasoning in order to provide a logicist foundation for mathematics. Boole’s work
directly inspired the invention of logical machines, notably William Jevons’s “logic piano.” Frege’s
logic was influentially taken up by Bertrand Russell, who with Alfred North Whitehead attempted
to complete the logicist project. Questions about the scope and power of this logic then became
central to research in the foundations of mathematics, particularly through the work of David Hil-
bert and Kurt Gödel. The particular question of whether Fregean predicate logic is decidable
required delimitation of what could count as effective computation, and it was Alan Turing’s
81
Logic: Atheism
KEY CONCEPTS
A Priori Reasoning Logical Form
Accessibility Relation Logical Problem of Evil
Ampliative Reasoning Modal Logic
Argument Structure Modus Ponens and Modus
Compossible/Incompossible Tollens
Conceivability Principle Natural Deduction
Conclusion Ontological Argument
Epistemological Persuasiveness Paradox of the Stone
Evidential Problem of Evil Predicate Logic
First-Order Logic Premises
Formal Validity Problem of Evil
Higher-Order Logic Propositional Connectives
Hume’s Fork Propositional Logic
Inference Problem of Evil S5 System of Modal Logic
Kalām Cosmological Validity
Argument World-Indexed Properties
achievement in doing this (through what we now call a Turing machine) that led to the develop-
ment of the modern computer. In this century the computer has utterly transformed the practical
potential of formal logic, making possible rigorous investigation and checking of vastly more com-
plex systems and arguments, but also realistic search, by which novel arguments and conclusions
can be discovered in reasonable timescales.
As applied within the philosophy of religion, these later developments have had little impact
(though we shall see a notable exception in the final section below). Here, the role of formal
logic has been mainly to regiment and represent arguments that were already familiar in less for-
mal guise, and it is doubtful whether formalization can provide any radically distinctive contri-
bution in support of either theism or atheism (e.g., by delivering what would otherwise be sur-
prising verdicts on potentially convincing arguments). It can, however, greatly help to clarify
the structure of arguments, especially when they are set within a relatively complex theoretical
structure (e.g., involving iterated modalities). These points will become clearer as we briefly
review some important logical systems, from simple propositional logic to higher-order and
modal predicate logic.
PROPOSITIONAL LOGIC
Study of formal logic now usually starts with propositional logic, in which arguments are analyzed
in terms of complete propositions, which (ignoring various complications) we can take to be the
content of meaningful, unambiguous, declarative sentences that are hence determinately either true
or false. Here meaningful rules out nonsense such as Lewis Carroll’s “’Twas brillig, and the slithy
toves did gyre and gimble in the wabe.” Declarative rules out other forms of speech such as ques-
tions and commands (whose purpose is something other than making a statement, and hence are
neither true nor false). Unambiguous rules out shifting meanings and also statements whose ref-
erence—and therefore what is stated—can vary from instance to instance (e.g., “He is at the
bank now”).
Logic: Atheism
An example of an argument that can reasonably be analyzed propositionally is the following

presentation of the problem of evil:
If God exists, then he is omnipotent and perfectly benevolent. If God is benevolent then
he will try to eliminate evil, while if he is omnipotent and tries to eliminate evil, he will
succeed. But evil nevertheless exists, therefore God does not.
To perform this analysis, we first identify the conclusion of the argument and spell it out unambig-
uously and independently of its local context:
God does not exist.
We then identify and number the various premises of the argument, that is, the stated assumptions it
starts from (and which it thus takes for granted rather than arguing for them):
1. If God exists, then God is omnipotent and perfectly benevolent.

2. If God is benevolent, then he will try to eliminate evil.
3. If God is omnipotent and tries to eliminate evil, he will succeed.
4. Evil exists.
If the argument is to be susceptible to analysis within propositional logic, then it must now be pos-
sible to identify atomic propositions in terms of which the premises and conclusion can be expressed,
sufficiently faithfully that the logic of the argument is still recognizably preserved. (Atomic proposi-
tions can have complex internal structure, but are atomic in the sense that their presumed role
within the argument is that of simple units, requiring no consideration of that internal structure.)
To achieve such expression, the atomic propositions can be combined using brackets and a few stan-
dard truth-functional propositional connectives, each of which is usually represented symbolically
and—at least in the commonest systems—are roughly equivalent to the English terms not (symbol
“:” or “”), and (“&” or “^”), or (“_”), implies (“!” or “”), if and only if (“ !” “”). These
are propositional connectives, in that they operate on (or connect) propositions to generate a new,
compound, proposition (e.g., “and” connects the two propositions “Joseph married Mary” and
“Mary became pregnant” to form the compound proposition “Joseph married Mary and Mary
became pregnant”). They are truth-functional, because they are interpreted in such a way that the
truth or falsehood of the compound proposition is dependent only on the truth or falsehood of
the constituent propositions. This deviates from everyday English in which, for example, “P and
Q” would often (as in the previous parenthesis) be taken to imply that Q followed P, whereas inter-
preted truth-functionally, “P and Q” is exactly equivalent to “Q and P,” meaning simply that both P
and Q are individually true. Awareness of such deviations is important, because they can lead to
misleading formal assessments of arguments in natural language. The connective “!” is most prob-
lematic in this respect, “P ! Q” being assessed as true unless both P is true and Q false; thus, for
instance, the obvious formalization of “if Mercury is larger than Venus then Mercury is larger than
Jupiter” comes out true.
Any argument that cannot faithfully be represented in this restricted vocabulary cannot be
usefully analyzed within propositional logic. But in the current case this can plausibly be
achieved in terms of the following atomic propositions, to each of which we assign a mnemonic
letter:
G: God exists
O: God is omnipotent
B: God is (perfectly) benevolent
T: God will try to eliminate evil
E: Evil exists

Logic: Atheism
Our argument, thus represented, now becomes:
1. G ! (O & B)
2. B!T
3. (O & T) ! :E
4. E
∴ :G
Note that in doing this sort of translation, we often have to simplify what is being said so as to shoe-
horn the argument into the appropriate logical form—here, for example, “God will succeed in elim-
inating evil” has been rendered as simply “Evil does not exist.” It is a matter of judgment, and may
be controversial, whether such simplifications preserve the essential argument structure, or on the
contrary seriously misrepresent it.
The propositional argument form above is provably valid, in that whatever the individual truth
values of the atomic propositions might be, there is no way that the four premises can all turn out to be
true while the conclusion is false. Hence if the premises are all true, the conclusion must be true also.
(A standard way to prove such validity in propositional logic is by means of a truth-table, in which
we tabulate all of the thirty-two possible combinations of truth values of the five atomic proposi-
tions, and determine the truth or falsity of the premises and the conclusion in each case. Since all
the connectives involved are presumed to be truth-functional, the truth value of any complex prop-
osition can be determined entirely from the truth values of its constituent atomic propositions.)
This formal result, moreover, corresponds well with our likely logical verdict on the original argu-
ment: it appears to be a valid argument in precisely this same sense. So we have here a nice example
of how analysis—even in the very simplistic terms of propositional logic—can be philosophically
useful, helping to back up our natural logical judgment. A contrasting example, potentially revealing
the fallibility of our natural judgment, would be, “If Jesus redeemed us, then our sins have been for-
given. But Jesus can have redeemed us only if he was the son of God. Therefore, unless Jesus was
the son of God, our sins have not been forgiven.” Arguments of this general form can easily seem
valid, but they are not: in the current case, for example, the premises leave open that God has for-
given our sins without Jesus playing any part in the matter.
FIRST-ORDER PREDICATE LOGIC

Predicate logic extends propositional logic, permitting the internal structure of many propositions to
be represented by introducing names for objects (e.g., a, b, c), variables ranging over objects (e.g., x,
y, z), and two quantifiers, the universal quantifier “8”, meaning “for all”, and the existential quanti-
fier “9”, meaning “there exists some”. This greatly increases the expressive power of the system and
enables us to demonstrate the validity of some arguments that, when represented within proposi-
tional logic, would appear to be invalid. Take, for example, this simple formulation of the kalām
cosmological argument:
Everything that began to exist had a prior cause.

The physical universe began to exist.
Therefore, the physical universe had a prior cause.
Neither the two premises nor the conclusion of this argument can be analyzed into compo-
nent propositions, so if we try to represent them in propositional logic, we can do no better
than treating each of them as a distinct atomic proposition, yielding a plainly invalid argu-
ment structure:
1. P Everything that began to exist had a prior cause.
2. Q The physical universe began to exist.
∴R The physical universe had a prior cause.
Logic: Atheism
In predicate logic we can do much better, by analyzing the argument as follows:

Bx x began to exist
Cx x had a prior cause
p The physical universe
1. 8x (Bx ! Cx) Everything that began to exist had a prior cause.

2. Bp The physical universe began to exist.
∴ Cp The physical universe had a prior cause.
Here B and C represent predicates rather than complete propositions, and the generality of the first
premise is captured by translating it as a universally quantified conditional: “For all x, if x began to
exist, then x had a prior cause.” (Thus the universal quantifier is standardly combined with “!”,
whereas the existential quantifier is standardly combined with “&”. As an example of the latter,
“Something that began to exist had a prior cause” would be translated “9x (Bx & Cx),” meaning
“There exists some x, such that x began to exist and x had a prior cause.”)
The predicate logic argument structure above is provably valid, because the first premise yields,
as an instance, the simple conditional “(Bp ! Cp),” which combines with the second premise (by
the rule modus ponens) to deliver the conclusion. The contrast between these propositional and pred-
icate logic structures—formalizing one and the same argument in two different ways—illustrates an
important general point: an argument may be translatable into a variety of “logical forms” that may
give different verdicts regarding its validity. And one important motivation for developing richer for-
mal systems is to be able to represent a wider range of arguments in such a way as to reveal the dee-
per structure on which their validity potentially depends.
For this reason, talk of “the logical form” of an argument in natural language, though fairly
common, can be very misleading. There is no uniquely privileged formal system that can legiti-
mately claim primacy for representing the form of propositions and arguments. And although the
tradition deriving from Frege and Russell has been hugely influential in establishing predicate logic
(and its various extensions) as a de facto standard in much of the literature, there is no good reason
to consider this a comprehensive mechanism capable of faithfully representing all semantic relation-
ships between the propositions that we express. Moreover, it should not be assumed that an argu-
ment in natural language can be valid (i.e., such that it is impossible for the premises to be true
while the conclusion is false) only if it is representable as formally valid (i.e., as a substitution
instance of some valid logical form) within some logical system. Take, for instance, the following
simple argument:
1. Mars is red and round.
∴ Some round thing is colored.
This argument is valid in the informal sense—the truth of its premise guarantees the truth of its
conclusion—but it cannot be represented as formally valid without introducing an additional prem-
ise to express the semantic implication from being red to being colored, and this would modify it
into a different (though closely related) argument.
None of this is to deny that it can be extremely useful to represent arguments formally, even at
the cost of modifying them. When arguments are complex, formalization can make their analysis far
more tractable (and their validity potentially mechanically computable). Even when they are rela-
tively simple, formal representation—notably in predicate logic—can be extremely useful in identi-
fying and highlighting some of the seductive ambiguities of natural language. One familiar pattern
involves a sentence containing two quantifiers, such as “Some cause is prior to every effect,” which
is subject to two readings:
(a) 8x (Ex ! 9y (Cy & Pyx))
For all x, if x is an effect then there exists some y such that y is a cause and y is prior to x

Logic: Atheism
(b) 9y (Cy & 8x (Ex ! Pyx))

There exists some y that is a cause and is such that for all x, if x is an effect then y is prior to x
The former is relatively modest, claiming only that every effect is preceded by some cause or
other. The latter is more ambitious, implying the existence of some particular “first cause” that
is prior to all effects whatever. Inferring (b) from (a) would commit the “quantifier shift fallacy”
(allegedly committed by quite a number of notable arguments through history; see Millican
2004, n. 45), and it is a merit of these predicate logic representations to lay bare how clearly dif-
ferent (b) is from (a).
HIGHER-ORDER PREDICATE LOGIC

Standard predicate logic permits only first-order predicates and quantifiers: that is, specific predicates
representing properties or relations applying to individual things, and quantifiers that range over
those individuals. Second-order predicate logic goes further, admitting quantification over properties
of objects (or sets of objects, if the properties are being treated extensionally, in terms of the objects
to which they apply, rather than intensionally, in terms of their meaning, definition, or necessary
and sufficient conditions). This enables us to express some things that cannot be expressed, even
indirectly, in first-order language, for example, the general principle of mathematical induction:
8P ((P(1) & 8x (P(x) ! P(x+1))) ! 8x (N(x) ! P(x)))
Here we extend the notation of predicate logic to include the standard symbols for the number 1
and the arithmetical function “+”, and introduce brackets after the predicates N and P. But most
importantly, whereas N(x) represents the first-order predicate “x is a natural number” (i.e., a positive
integer), P(x) is here acting very differently, as a variable ranging over first-order properties in gen-
eral. Thus the principle is saying that if P is any first-order property that (a) is true of the number
1 and (b) itself has the following second-order property, that if P is true of any x then P is also true
of x+1, then P is true of every natural number.
Further higher-order logics are possible, for example, by incorporating properties of properties
(and then even properties of properties of properties, etc.). These logics can become relevant to
the philosophy of religion when considering formalized arguments about God’s properties, notably
Gödel’s ontological argument, which defines God as a being that has all positive properties (dis-
cussed later). Among the properties traditionally ascribed to God, moreover, some crucially involve
modality—that is, necessity and possibility. For example, God is commonly supposed to be a necessary
being (in contrast with the contingent beings that constitute his creation), and to be omnipotent:
able to achieve anything at all that could possibly be achieved. To handle such modal properties,
we need to extend our logic in another way.
MODAL LOGIC
Modal logic introduces operators for necessity and possibility, usually represented as “□” and “hi,”
and commonly interpreted as truth in all, or some, possible worlds. This implies that the two are
interdefinable (for example, if “possibly P” says that P is true in some possible world, then that is
the same as denying that P is false in every possible world):
hiP (“possibly P”) is equivalent to :□(:P) (“not necessarily (not P)”)
□P (“necessarily P”) is equivalent to :hi(:P) (“not possibly (not P)”)

Things get more complicated when the operators are iterated, when we are interested, for example,
in whether P is necessarily possible and/or possibly necessary (formulated as □hiP and hi□P, respec-
tively), thus considering what is possible, or necessary, in other possible worlds. The simplest—and
most common—assumption to make here is that what is necessary or possible is absolute, not vary-
ing at all from world to world: if P is possible in any world, therefore, P is possible in every world. In
this case, “possibly (necessarily P)” is the same as “necessarily P,” “necessarily (possibly P)” is the
same as “possibly P,” and indeed all iterated sequences of modal operators “collapse” so that only
Logic: Atheism
the last is relevant (so, hihi□hihihi□P is equivalent to □P, and □hi□hi□□hiP to hiP).
The system of modal logic that incorporates this assumption is known as S5.
S5 is widely taken for granted when handling absolute or logical necessity, fitting with the tra-
ditional view that any proposition at all that can be coherently formulated without contradiction
is possible in an absolute sense. This view is enshrined in the conceivability principle—“whatever is
conceivable is possible”—which became especially prominent through the influence of David
Hume, and hugely informed later analytic philosophy (see Millican 2017 for detailed discussion,
esp. 33–42). In some contexts, however, it can be more appropriate to see the range of possibili-
ties as potentially being affected—perhaps enlarged in some respects, and restricted in others—by
contingent occurrences. If we apply modal logic to physical possibilities over time, for example,
then which locations it is physically possible for me to reach within the next five minutes depends
crucially on where I am currently situated: the physically possible worlds that are accessible given
my current position in Hertford College are different from those that would have been accessible
were I at home. Likewise, when a new baby enters the world, or an old person leaves it, the world
correspondingly acquires new possibilities, or loses old ones. Different applications of modal logic,
involving different types of modality, can therefore require correspondingly different assumptions
about the relevant accessibility relation between worlds, thus giving rise to a variety of modal
logics. In what follows, however, we can largely ignore such complications, since the modality
applicable to God—for obvious reasons—is standardly taken to be absolute and independent of
any worldly contingencies. A priori theistic arguments that rely on modal logic accordingly tend
to presuppose S5.
GENERAL CONSIDERATIONS, AND THE LIMITS OF LOGIC
Logical validity by itself is not all-important: an argument can be valid but useless; or invalid but
powerful. We now turn to discuss some of the other factors that contribute to making an argument
rationally persuasive.
VALIDITY, SOUNDNESS, AND EPISTEMOLOGICAL PERSUASIVENESS

It is one thing for an argument to be valid, and quite another for it to be sound—that is, both valid
and having true premises (hence also a true conclusion, since, as explained earlier, a valid argument
is one in which it is not possible for the premises to be true and the conclusion false). Even a sound
argument, moreover, need not be of any use as a proof, for example:
1. The physical universe began to exist.
∴ The physical universe began to exist.
This argument is clearly valid: if its single premise is true, then certainly its conclusion (which is
exactly the same proposition) must also be true. And it may be that the premise is in fact true, in
which case the argument is sound. But it is obviously hopeless as a means of proving its conclusion,
because nobody will accept the premise (and thus be potentially persuaded by the argument) unless
they already accept that conclusion.
It is not straightforward to formulate precise conditions under which a logical argument should
be considered epistemologically persuasive; such persuasiveness is likely to depend on a number of fac-
tors, some of which will be relative to the background assumptions and beliefs of the individual con-
cerned. But in the typical case, at least, an argument can be potentially epistemologically persuasive
to someone S only if S could reasonably accept its premises as true while doubting its conclusion;
then the role of the argument would be precisely to persuade S that the conclusion is true, by show-
ing that it follows from the premises. (One important nontypical case is where an argument is
intended to show that some set of assumptions leads to a contradiction, so-called reductio ad absur-
dum, in which the point is to undermine one or more of the premises, rather than to establish the
conclusion.) This condition for epistemological persuasiveness is clearly not satisfied by the trivial

Logic: Atheism
argument just discussed, and we might reasonably doubt whether it is satisfied in other simple cases,
such as the kalām argument discussed earlier:
1. Everything that began to exist had a prior cause.
2. The physical universe began to exist.
∴ The physical universe had a prior cause.
Elsewhere, I have referred to arguments with this crude logical structure as “hole in one” arguments.
Their form is valid and, as we have seen, can easily be represented within predicate logic:
1. 8x (Bx ! Cx) All Bs are Cs
2. Bp p is a B
∴ Cp p is a C
Such arguments, to my mind, do not deserve the prominence that they have in popular philosophi-
cal discussion. Several have often been deployed in debates over abortion (as discussed in Millican
1992, 167–169), for example:
1. Killing human beings is wrong.
2. The fetus is a human being.
∴ Killing the fetus is wrong.
Again we have a general premise followed by a particular instance—each of them individually per-
haps fairly plausible—and then a controversial conclusion validly drawn from those premises.
But these features are not enough to make the argument epistemologically persuasive, because
when faced with such an argument, any rational person who rejects the controversial conclusion is
overwhelmingly likely to reject one of the premises. Suppose, for example, that I believe killing human
beings is wrong, but, having considered the matter, do not object to very early abortion (e.g., by the
morning-after pill). These attitudes together imply, if I am rational, that I do not consider the concep-
tus to be a human being. A person opposing all abortion may now try to persuade me that the con-
ceptus is indeed a human being (e.g., by arguing that it is a live biological entity, and of the species
Homo sapiens). But even if she succeeds in persuading me to recategorize the conceptus in this way,
I am most unlikely to go on to draw the conclusion she intends; far more likely, given my considered
attitude to early abortion, is that I will simply withdraw my agreement to the general principle that
killing human beings is wrong, on the grounds that the conceptus is an exception to it. Likewise, with
the kalām argument above, an atheist might initially agree to the general principle that everything that
began to exist had a prior cause, but, if the entire universe is then proposed as an instance, quite rea-
sonably withdraw that assent: “Oh, when I agreed that everything that began to exist had a prior
cause, I wasn’t foreseeing that you would ask me to apply this principle to the entire universe—that,
of course, is an exception, because there would have been nothing to cause it.”
All this is not to say that “hole in one” arguments are useless: they can, for example, play a
helpful role in structuring a discussion. But even in this role they are potentially misleading if they
give the impression that the premises of the argument can be considered independently of each other
(when, for example, a philosopher devotes one section of a paper to arguing for the first premise,
then another section to arguing quite separately for the second premise, before triumphantly draw-
ing the desired conclusion). Usually, the premises cannot properly be assessed independently in this
way, because someone who is unpersuaded of the conclusion (“p is a C ”) will accept the general first
premise (“All Bs are Cs”) only on condition that the controversial instance in the second premise
(“p is a B”) is ruled out. Hence to provide an epistemologically persuasive discussion, that controver-
sial instance has to be explicitly considered when arguing for the general principle.
EPISTEMOLOGICAL PERSUASION AND FORMALIZATION

With more complex argument structures, it becomes more plausible that a thoughtful person could start
out accepting the premises and rejecting the conclusion, failing to realize that the argument is valid.
Logic: Atheism
In such a case, the argument might indeed be epistemologically persuasive, and formalization could play a
valuable role in demonstrating the argument’s validity. Even here, however, epistemological persuasion is
not necessarily the most likely outcome. Take, for example, the problem of evil argument discussed above:
If God exists, then he is omnipotent and perfectly benevolent. If God is benevolent then he
will try to eliminate evil, while if he is omnipotent and tries to eliminate evil, he will succeed.
But evil nevertheless exists, therefore God does not.
We have seen how this can be represented formally within propositional logic, and in a form that is
provably valid. Suppose, then, that some thoughtful theist, S, who initially accepts the premises of
the argument, is presented with this formalized argument and acknowledges its validity: is S likely
to be persuaded by it that God does not exist? Given the momentous metaphysical and existential
significance of that conclusion, I suggest that it is instead far more likely that S—being initially uncon-
vinced of the argument’s conclusion while accepting its premises—will either question the adequacy
of the formalization or, failing that, will come to doubt one of the premises thus formalized. In the
current case, for example, the traditional theist will most probably question the second premise:
If God is benevolent, then he will try to eliminate evil.
which has been formalized thus:
B!T
Though perhaps superficially plausible, this premise is extremely crude when thus interpreted: it is
treating God’s benevolence and his trying to eliminate evil as separable all-or-nothing conditions.
Most theists are far more likely to think of God’s benevolence as tending toward a creation that
is good on the whole, but which may tolerate some evil notwithstanding God’s ability to remove it
forcibly should he wish to do so. Such toleration of evil may be explained in the various ways famil-
iar from traditional theodicies: as necessary for greater goods, as resulting from the desirable opera-
tion of general laws, or from human (and/or angelic) free will, and so on. There is, of course, plenty
of scope for debate about the potential adequacy of these different responses to the problem of evil,
and some theists might reasonably be unsure how to judge them. But rational theists who accept
that there is evil in the world—as stated by the fourth premise—are virtually certain to consider that
there is something wrong with the second premise (or perhaps another). The argument is just too
crude to have a realistic chance of persuading reflective theists, for in maintaining their theism
despite the acknowledged evil in the world, they are implicitly repudiating its premises. The logical
relationships presented in the argument are so straightforward that it is implausible to imagine it
revealing hidden implications that could persuade the theist to abandon belief in God.
To sum up the discussion in these last two sections: First, we have seen that validity is not the
same as rational persuasiveness, and a valid argument from premises to conclusion may serve to
undermine the premises rather than to establish the conclusion, a point often expressed in the aph-
orism “One man’s modus ponens is another man’s modus tollens.” (Modus ponens is the valid rule that
moves from P and (P! Q) to infer Q; modus tollens is the equally valid rule that moves from :Q
and (P! Q) to infer :P. So the point of the aphorism is that two people can agree on the validity
of an argument from P to Q, and hence agree that (P! Q); but whereas one may take the argument
to have proved Q from P, the other may take it to have disproved P, on the grounds that the con-
clusion Q is false.) Second, the validity of an argument in natural language is not to be equated to
formal validity within any formal system, though formalization can be extremely useful for analysis
and assessment. Third, putting an argument formally often adds nothing to its rational persuasive-
ness, since philosophical questions about its premises and conclusion—and their true relation-
ships—are likely to remain even if the formal gloss serves to conceal them.
HUME’S FORK AND THE LIMITS OF A PRIORI REASONING

Modern orthodoxy on the limits of a priori reasoning largely follows Hume:
All the objects of human reason or enquiry may naturally be divided into two kinds, to wit,
Relations of Ideas, and Matters of Fact. Of the first kind are the sciences of Geometry, Algebra,
and Arithmetic; and in short, every affirmation, which is either intuitively or demonstratively

Logic: Atheism
certain. Propositions of this kind are discoverable by the mere operation of thought, without
dependence on what is any where existent in the universe.… Matters of fact, which are the
second objects of human reason, are not ascertained in the same manner.… The contrary of
every matter of fact is still possible; because it can never imply a contradiction.… We should
in vain, therefore, attempt to demonstrate its falsehood. (Hume 2007a, 4.1–2)
This dichotomy, commonly known as “Hume’s Fork,” implies a threefold coincidence between
semantic, epistemological, and metaphysical categories: put very crudely, that the terms analytic
(i.e., true in virtue of the meaning of the terms), a priori (i.e., knowable without recourse to experi-
ence), and necessary (i.e., true in all possible worlds) all apply to exactly the same propositions.
The implication, if Hume is right, is that a demonstration (i.e., valid deduction) from a priori pre-
mises is confined to abstract and immutable relations between our ideas, and could never prove the
existence of a substantial being such as God is supposed to be:
Matter[s] of fact and existence … are evidently incapable of demonstration. Whatever is may
not be.… The non-existence of any being, without exception, is as clear and distinct an idea as
its existence. The proposition, which affirms it not to be, however false, is no less conceivable
and intelligible, than that which affirms it to be … that Caesar, or the angel Gabriel, or any
being never existed, may be a false proposition, but still is perfectly conceivable, and implies
no contradiction. (Hume 2007a, 12.28)
This seems extremely plausible on general commonsense grounds. A deductively valid argument can
be used to reveal logical consequences that are implicit in its premises, but it cannot be ampliative,
demonstrating new substantial conclusions that are not thus implicit. So if P is some proposition
about the world whose denial is not itself somehow absurd or self-contradictory, then it looks as
though we cannot possibly prove P to be true unless we include within the argument some premise
that supplies relevant information about the world, from which P can then be deduced.
This line of thought famously woke Immanuel Kant from his dogmatic slumbers, but since he
felt certain that we do in fact have a priori knowledge of the world, he was prompted to develop a
sophisticated theory to circumvent Hume’s objection and justify the claim that some synthetic pro-
positions (such as universal causation and the conservation of momentum) could be thus known.
Kant’s position is ingenious but no longer convincing, building as it does on his somewhat exotic
metaphysics of transcendental idealism, that the world we experience is largely mind-constructed.
This is supposed to provide a potential route to a priori knowledge of that world, but it fails to
evade Hume’s Fork, since it relies implicitly on the dubious assumption that we can have a priori
knowledge of our mind’s constructive activity (see Millican 2017, 58–60). Quite apart from such
philosophical objections, Kant’s theory was devastated by twentieth-century physics, which showed
convincingly that some of his supposed a priori truths are false (or at best a posteriori).
Other challenges to Hume’s Fork have emerged, notably from the work of Willard Van Orman
Quine and Saul Kripke. Addressing these requires some refinement of the Humean position, but they
do not threaten his central denial that matters of fact—including all questions of substantial (as opposed
to abstract) existence—are incapable of a priori demonstration (see Millican 2017, 48–61). This puts a
very strong onus of proof on anyone claiming to have found an a priori demonstration of God’s exis-
tence: such a demonstration would constitute a revolution in logic and metaphysics, and demand at least
some account of how Hume’s very plausible bar on such demonstration has been circumvented.
THEISM AND CONSISTENCY
Although logic alone cannot prove the existence of any substantial entity, it can potentially prove
something’s nonexistence, by showing the relevant concept to be inconsistent or self-contradictory.
DIVINE PROPERTIES AND PARADOX

If we start from the minimal understanding of a god as, in Hume’s words, an invisible intelligent
power (2007b, 2.5), there is no obvious inconsistency in the claim:
There is a god.
Logic: Atheism
But inconsistency can potentially arise if we attribute impressive properties to the postulated god,
for example by claiming:
There is an omnipotent god.
This invites the well-known paradox of the stone, as raised by the question: “Can an omnipotent god
create a stone that is too heavy for him to lift?” This seems paradoxical because both available
answers seem to lead to the contradiction of a being who is omnipotent—and hence by definition
can do anything—but yet has limits on what he can do. For if the answer is “no,” and he cannot
create such a stone, then that is something he cannot do; but equally if the answer is “yes,” and
he can create such a stone, then again there is something he cannot do, namely, lift that stone.
There are also much simpler questions to be asked about omnipotence, for example: can an omni-
potent being change the past, or make 1 + 1 equal to 3, or bring it about that what we think of
logical relationships—for example, that (P and Q) entails P—do not hold? These seem to us to be
absolutely impossible, but if they are impossible for an omnipotent being too, then there are things
he cannot do, which again apparently contradicts the supposition that he can do anything.
As extra divine properties are added, so more opportunities arise for contradiction, both from the indi-
vidual properties themselves, and from their interactions, for example:
There is an omnipotent, omniscient god.
Some philosophical views would suggest internal problems with the very notion of omniscience. For
example, if future propositions (or at least those that are undetermined) are neither true nor false,
then even an omniscient being—who by definition is supposed to know everything—cannot know
them. But more significantly, the combination of omnipotence and omniscience introduces new
possibilities for paradox, as raised by the question: “Can an omnipotent being create an agent
who is able to act without his knowledge?” (This can also be considered an internal problem for
omnipotence, given that infinite power arguably implies infinite power to know, and hence omni-
science.) Yet more possibilities arise if we then add moral perfection to the divine properties:
There is an omnipotent, omniscient, perfectly good god.
Can such a being commit evil, or know what it is like to commit evil? Would he have the power to
create something that generates evil of which he remains ignorant? As the ascribed properties augment,
so the potential for contradiction multiplies. Moreover any contradictions that may have previously
arisen remain, because if two claims P and Q contradict each other, then adding another claim—R,
say—cannot remove that contradiction. And some of the properties traditionally ascribed to the so-
called God of the philosophers (for which it is standard to use the capitalized term “God”) do seem to
invite such contradiction. For example, immutability and simplicity seem hard to square with divine
action in a changing world, or divine knowledge about it; while Hume’s Fork would imply that nec-
essary existence is incoherent when ascribed to a substantial being with causal power.
DEFINING AWAY THE PARADOXES

There has been considerable discussion about the consistency of the supposed divine attributes (see, e.
g., Swinburne 1977, Kenny 1979, Rowe 2005, and the various essays in part 2 of Flint and Rea 2009,
in particular by Wierenga on omniscience and Leftow on omnipotence). But rather than attempt to
deal individually with the potential multitude of problems here, I restrict this brief discussion to mak-
ing some general points. First, regarding the question of whether an omnipotent God would be bound
by logical constraints, the standard answer—dating back at least to Thomas Aquinas (Summa theolo-
giae 1a, q. 25, a. 3) is that he would indeed be so bound, but that this would not contradict, or con-
stitute any real limit on, his omnipotence (for nothing that can coherently be ascribed to him is being
put beyond his power). Thus an omnipotent being should be understood as a being who is able to do
anything that it is logically possible for him to do (and likewise an omniscient being as a being who knows
everything that it is logically possible for him to know). This answer obviously removes some of the
threatened paradoxes, but it is also well-motivated independently. For if God is considered to be
“above logic” (as most notoriously suggested by Descartes’s doctrine of the creation of the “eternal
truths” [e.g., Descartes 1991, 25, 358–359]), then it becomes impossible to reason coherently about

Logic: Atheism
him: God might both exist and not exist at the same time, or be omniscient and yet also ignorant, or
perfectly good and yet utterly evil, or truthfully promise heavenly salvation while delivering eternal
hellfire. If our logic cannot be relied on as applying to God, then we may not be able to refute his
existence by appeal to that logic, but by the same token, we cannot rely on that logic to reason about
him in any other respect either, and all attempts at religious knowledge become futile. Theism remains
unrefuted only at the cost of becoming utterly useless.
Allowing logic to constrain omnipotence removes the most immediate threats of contradiction
from that source, but in their place comes a milder threat of indeterminacy. The paradox of the
stone, for example, could be resolved in either of two ways: by insisting that logically no stone
can be too heavy for God to lift, and hence omnipotence does not require him to be able to create
such a stone; or, alternatively, by allowing God to create such a stone but then arguing that his
inability to lift it would not limit his omnipotence, because his lifting it would be logically ruled
out. Here the former resolution seems more straightforward and attractive, but in other cases there
may be no clear verdict on where the logical constraint should be applied.
This problem of indeterminacy becomes even more apparent when we consider potential
inconsistencies between the different divine properties. Suppose, for example, we define God as a
being who is able to do anything that it is logically possible for him to do, and know anything that it
is logically possible for him to know. Does this mean that God can create a being who can act without
his knowledge (because it is no limit on his omniscience to be unable to know what his omnipo-
tence makes logically unknowable to him)? Or, on the other hand, does it mean that God cannot
create such a being (because it is no limit on his omnipotence to be unable to do what would logi-
cally contradict his omniscience)? I suggest that the best way of dealing with this problem is to
define the theistic hypothesis in terms of some explicit hierarchy of properties, for example:
God is to be understood as a being who is:
(a) Perfectly good.
(b) Omniscient, in the sense of knowing everything that it is logically possible for a being
defined by (a) and (b) to know.
(c) Omnipotent, in the sense of being able to do anything that it is logically possible for a being
defined by (a), (b), and (c) to do.
Such a definition helps to sidestep many of the previously threatened paradoxes. For example: God
can know what it is like to do evil only if it is logically possible for a perfectly good being to know
this; he can do evil only if it is logically possible for a perfectly good being to do so; and he cannot
create a being that acts without his knowledge.
Although there is not enough space here to discuss all of the traditional divine attributes and
the possible conflicts between them, I have offered a method which seems likely to be able to deliver
a definition of theism that recognizably retains those attributes while avoiding contradiction. That
this should be feasible is hardly surprising, for the believer is free to define her theistic hypothesis
in any way that she chooses, and nothing apparently prevents her from including any desired logical
constraints (e.g., avoidance of contradiction) explicitly within the definition. But we have also seen
that this comes at a cost: formulating the definition requires choices regarding the ordering of the
hierarchy, and even after that has been settled, there will be further choices to be made regarding
how particular threatened contradictions are to be resolved (e.g., can God create the stone, or
not?). There is a risk that at the end of all this our definition will seem somewhat arbitrary, when
so many different combinations of choices could have been made. And such a multitude of choices
also undermines the traditional claim that the idea of God has intrinsic harmony and simplicity,
making it theoretically attractive both metaphysically and scientifically (as argued most prominently
in recent times by Swinburne, e.g., 2004, 96–109).
There also seems to be something artificial—even bizarre—in this philosophical game of reso-
lutely seeking the most impressive definition of God that we can devise (bounded only by the require-
ment of consistency). When investigating what exists, we usually frame our expectations on the basis
of the supporting arguments and evidence rather than by appealing to our most ambitious and
Logic: Atheism
optimistic imaginings. In Hume’s words, the latter “savours more of flattery and panegyric, than of
just reasoning and sound philosophy” (2007a, 11.27); and indeed Hume considered that it was flat-
tery, rather than logic, that led believers toward the perfect being of traditional theism (2007b, 6.5–6).
APRIORISTIC ARGUMENTS, AND THE PROBLEM OF EVIL

The method of the previous section may well enable “God” to be defined in a way that recognizably
retains at least the most important of his traditional attributes (e.g., omnipotence, omniscience, and
moral perfection), while avoiding self-contradiction. But this in itself seems a modest achievement,
merely taking theism to the starting line of the competition among rival theories by providing a
coherent hypothesis for assessment. It still remains to be seen what theism might have going for it
in terms of either a priori considerations or empirical evidence. The former takes us into the terri-
tory of the ontological argument, discussed in the following sections. The latter—involving argu-
ments from such specific features of the world as biological form, natural beauty, miracles, or reli-
gious experiences—lies mainly beyond the scope of this chapter. But there is some blurring
between the a priori and the empirical when we consider far more general aspects of the empirical
world, such as that some things move, or cause others, or are contingent (respectively the starting
points of the first three of Aquinas’s famous Five Ways), or even the more basic fact that there is
something rather than nothing. So-called cosmological arguments for theism have been built on each
of these general premises, purporting to reach God without any dependence on other a posteriori
assumptions that are more specific and potentially debatable. Such aprioristic theistic arguments,
however, are highly implausible when viewed in the light of Hume’s Fork (and their details are
beyond our scope here). Far more potent is the most common argument on the atheistic side, the
notorious problem of evil, drawn from the evident existence within the empirical world of apparent
imperfection or evil (of various sorts). Again, detailed discussion of this is beyond our scope, but
there are some important points to be made about the overall dialectical situation that it poses,
nicely illustrating some important logical features of philosophical argument more generally.
The problem of evil is an umbrella term for a family of “problems” for theism—that is, argu-
ments or claims in favor of atheism—appealing to different empirical evidence, and urged with dif-
fering logical force. At one evidential extreme, the atheist case can appeal to the mere existence of
any evil at all within the world, or—more contestable by the theist—the existence of some unab-
sorbed evil (a useful term from Mackie 1982, 154–155 for evil that is not necessary for some greater
good). Alternatively, the focus might be on the existence of particular kinds of evil, such as so-called
natural evil (i.e., suffering of sentient creatures arising from natural causes such as diseases, droughts,
and earthquakes) or moral evil (i.e., bad behavior by morally responsible agents). Going yet further,
the atheist might argue from the overall quantity of evil, and the fact that there seems to be so much
of it (either suffering, or immorality, or both) within the world.
Regarding the supposed logical force of the argument from evil, it might be claimed that the
evil to which the argument appeals demonstratively refutes theism—the so-called logical problem of
evil—or, less ambitiously, that it at least constitutes significant evidence against theism. How signif-
icant any particular evil is in this respect can of course be debated, but probably the most popular
form of this evidential problem of evil involves the claim that the evil in question is sufficiently
bad to render theism improbable (i.e., more likely false than true). Less ambitious still is the claim
that evil, though perhaps not counting so much against theism itself, at least refutes any empirical
case for theism. This inference problem of evil is the main thrust of Hume’s famous discussion in
his Dialogues (1779), where he argues very persuasively that evil refutes the design argument.
These dimensions of variation within the problem of evil illustrate some common themes in
philosophical debate. Premises that make more substantial claims (e.g., that there is a great deal of
unabsorbed evil) will usually be more disputable than premises that are relatively minimal (e.g., that
some evil exists). Arguments that are more ambitious (e.g., with minimal premises and substantial
conclusions) will usually be less convincing than those that are more modest. The most ambitious
form of the problem of evil argues that even a minimal amount of evil in the world is enough to
demonstrate logically the nonexistence of God. But this takes on an onus of proof which, given
the limitations of human knowledge—and the possibility that God has reasons of which we know

Logic: Atheism
nothing—looks unrealistic. The most convincing form of the problem of evil, by contrast, is far more
modest, merely insisting that it is impossible to infer the existence of God from an empirical world
that exhibits such a huge quantity of so many different evils (and in particular natural evil, for which
blaming human freedom of will is implausible). This challenges the advocate of the design argument
to “tug the labouring oar” (Hume 1779, 10.36) and take on the onus of proof to overcome the obvi-
ous tension between the grim reality from which her inference ought to start and the infinite excel-
lence of her conclusion. But this inference problem will not touch the more modest theist who claims
mere consistency and the possibility of reasonably maintaining an already existing belief. Such a belief
is most effectively challenged by the evidential problem, citing widespread and apparently unabsorbed
evil as strong (though not demonstrative) evidence against theism. The atheist argument from evil—
like many others—is most persuasive when it does not claim to have demonstrative force.
THE ONTOLOGICAL ARGUMENT
Viewed against this background, advocacy of the ontological argument—building on no empirical pre-
mises whatever but aspiring to prove an infinitely perfect being—seems outrageously optimistic.
Indeed, for most philosophers, both theist and atheist, the argument is more a logical puzzle than a
plausible contribution to natural theology, the question of interest being where exactly the fallacy lies.
DESCARTES AND THE PARODY OBJECTION

Let us start with Descartes’s ontological argument (from his fifth Meditation), which can be crudely
represented as follows:
1. God is by definition a being that possesses all perfections.

2. Existence is a perfection.
∴ God has the property of existence—God exists.
This seems obviously “too good to be true,” implausibly attempting to define a substantial being
into existence. One way of challenging such reasoning is to offer obviously unacceptable parodies
of it, as for example J. L. Mackie’s (1982, 42–43):
1. The Remartian is by definition an existing intelligent Martian.
∴ The Remartian exists.
Descartes himself was challenged with parodies involving an existing lion (1984, 72) and a perfect
Pegasus (1984, 225–226); his response (1984, 83–84) was to insist that “God” is a unified nature,
whereas the concepts in the parodies are artificially and arbitrarily put together. But such a response
seems weak and ad hoc, because the alleged natural unity of the concept of God played no clear role
in Descartes’s argument as presented; hence it is hard to see how the artificiality of the relevant con-
cept could undermine this form of reasoning if it were, in fact, valid.
KANT’S OBJECTION AND THE INTERNAL/EXTERNAL DISTINCTION

The parody-type arguments strongly suggest that something is wrong with Descartes’s argument, but they
do not identify what is wrong. In the Critique of Pure Reason, Kant famously tried to do so, as follows:
“Being” is obviously not a real predicate; that is, it is not the concept of something which could
be added to the concept of a thing. It is merely the positing of a thing, or of certain
determinations, as existing in themselves. Logically, it is merely the copula of a judgment. [In]
“God is omnipotent” … the small word “is” adds no new predicate, but only serves to posit the
predicate in its relation to the subject. If, now, we take the subject (God) with all its predicates
(among which is omnipotence), and say “God is,” or “There is a God,” we attach no new
predicate to the concept of God, but only posit the subject in itself with all its predicates, and
indeed posit it as being an object that stands in relation to my concept. (2007, A598–9/B626–7)
Logic: Atheism
This Kantian doctrine is commonly expressed as the slogan “existence isn’t a predicate” (though
more accurate would be “existence isn’t a property” or “‘exists’ isn’t a predicate”). It is widely
thought to doom the ontological argument, though what exactly its impact is on the argument is
rarely clarified.
Kant can plausibly (and charitably) be understood as insisting on a distinction between the
properties that make up or define a concept—its “characteristic” or “internal” properties—and its
“external properties” such as whether there exists some object that matches it: “Whatever, therefore,
and however much, our concept of an object may contain, we must go outside it, if we are to
ascribe existence to the object” (2007, A601/B629). Kant’s talk of “ascribing existence to the
object,” however, is potentially misleading, because from this perspective existence is better seen,
as Frege insisted, not as a property of a specific object but, rather, as a quantifier: a sort of second-
order property of a concept, which the concept possesses if it is instantiated or exemplified. For the sake
of clarity, therefore, in what follows I shall generally talk of instantiation rather than existence, and use
square brackets to refer unambiguously to a concept (e.g., “[God]”) rather than to any object that might
instantiate or fall under that concept.
Instantiation is not the only possible external property of a concept. For example, a concept
might have the property of being sophisticated, or rare, or amusing, or admired, or much-discussed, or
currently contemplated by me. As noted, these properties contrast with the internal or characteristic
properties of the concept, which are properties that must apply to any object that instantiates (or falls
under) it. In some cases, a concept might possess one or more of the very properties that characterize
it: thus the concept [sophisticated] is itself a sophisticated concept, and the concept [expressible in
English] is itself expressible in English. But mostly, this will not be the case: the concept [rare] is
not rare, the concept [amusing] is not amusing, and the concept [God] is not omnipotent.
Let us allow, in general, that anyone has complete freedom to define whatever concepts they
wish, simply by providing an appropriate list of characteristic properties, for example:
[God] = [omnipotent, perfectly wise, perfectly good]
[Aurelius] = [absolute ruler of the civilized world, notable philosopher, wise, beneficent]
[Laika] = [first dog in space]
But once the concept has been defined in this way—thus settling its internal properties—it is gen-
erally a question of fact rather than definition what external properties the concept has. So in partic-
ular, although Descartes is free to define the concept:
[DGod] = [possesses all perfections],
he cannot bring it about by mere definition whether this concept is admired, much-discussed, or
instantiated. (And if he tries to do this by adding external properties to his definition: “By [DGod] I
understand a concept whose internal properties are all the perfections, and which is, besides, externally
instantiated,” then the atheist can simply deny the existence of a concept that actually has these
properties, precisely because they are supposed to be external to the definition.) So Kant’s criticism
of Descartes is broadly on target: whatever properties may be put into a concept, the question can still
be raised whether it is instantiated, by asking “Is there one?” (cf. Mackie 1982, 49).
It is worth noting that this criticism applies whether or not we include real existence as an internal
property (e.g., perhaps in order to distinguish conceptually between supposedly real and fictional gods,
such as the fantasy author Terry Pratchett’s Zephyrus). So we can here bypass theoretical debates about
whether such inclusion is philosophically legitimate, for it is clear that the main debate—between the-
ist and atheist—concerns whether there is something real corresponding to the definition, and hence
involves an external rather than internal property. We can also here allow necessary existence as an inter-
nal property, and still apply the same criticism. Suppose, for example, that we consider the concept:
[NRabbit] = [rabbit, necessarily existing]
Anything that falls under this concept would have to satisfy its internal properties, and hence be a
necessarily existing rabbit. But having defined the concept, the question of its external instantiation

Logic: Atheism
is a quite different matter, and since a rabbit (I take it) is not the sort of thing that can exist of
necessity, it follows that [NRabbit]—so far from being necessarily instantiated—is necessarily unin-
stantiated. It might initially seem paradoxical that adding necessary existence to a concept brings
necessary noninstantiation in its wake. But this should not really surprise us: as we saw in the case
of the problem of evil, the more ambitious we are with our concepts or conclusions, the less likely
we are to succeed in vindicating them.
ANSELM’S ONTOLOGICAL ARGUMENT

Anselm’s ontological argument is far more subtle and slippery than Descartes’s. It takes the form of
a reductio ad absurdum, in which “the Fool” of Psalms 14 and 53—who “says in his heart ‘there is
no god’”—is shown to contradict himself. Addressing God, Anselm in the Proslogion expresses his
belief “that You are something than which nothing greater can be thought,” but then raises the
question suggested by the Fool’s denial: “can it be that a thing of such a nature does not exist?”
The subsequent reductio proceeds as follows:
But surely, when this same Fool hears what I am speaking about, namely, “something-than-
which-nothing-greater-can-be-thought,” he understands what he hears, and what he
understands is in his mind, even if he does not understand that it actually exists.… Even
the Fool, then, is forced to agree that something-than-which-nothing-greater-can-be-thought
exists in the mind, since he understands this when he hears it, and whatever is understood is in
the mind. And surely that-than-which-a-greater-cannot-be-thought cannot exist in the mind
alone. For if it exists solely in the mind, it can be thought to exist in reality also, which is
greater. If then that-than-which-a-greater-cannot-be-thought exists in the mind alone, this
same that-than-which-a-greater-cannot-be-thought is that-than-which-a-greater-can-be-
thought. But this is obviously impossible. Therefore there is absolutely no doubt that
something-than-which-a-greater-cannot-be-thought exists both in the mind and in reality.
(Anselm 1998, 87–88)
Many philosophical questions can be raised about Anselm’s language here, potentially casting doubt
on his argument. What is “greatness”? What is it for something to exist “in the mind”? Can one and
the same thing exist both “in the mind” and in reality? Can items in these two realms of existence
legitimately be compared in “greatness”? And so on. I have discussed such problems extensively else-
where (Millican 2004 and 2018), and here simply focus on what becomes of Anselm’s argument if
we avoid technical nitpicking and resolve them charitably. For the reasons explained above, this
involves interpreting the argument in terms of concepts and their properties.
Anselmian greatness involves an impressive catalog of qualities—“whatever it is better to be than
not to be” (Anselm 1998, 89)—which we can here abbreviate to omnipotence, perfect wisdom, and
perfect goodness. And his argument crucially turns on the principle that something that exists in real-
ity is thereby greater than if it did not, so we must also include real existence or instantiation in the
mix. Applying our previous discussion, however, we can see that this permits two quite different
interpretations of the argument, depending on whether or not greatness is to be understood as a
purely internal property of the relevant concept. Understood internally, we have something like:
[EGod] = [omnipotent, perfectly wise, perfectly good, really existing]
But then the argument straightforwardly fails for the familiar Kantian reason: whatever internal
properties we put into any concept, it is a further, external, question whether or not that concept
is actually instantiated, and this applies even if really existing has been included as one of those inter-
nal properties.
Where Anselm’s argument goes significantly beyond Descartes’s, however, is that it bears inter-
pretation in a quite different way, treating greatness as an external property of a concept (albeit one
that depends to a significant extent on the relevant internal properties). Thus understood, the great-
ness of a concept is a measure of its impressiveness both internally and externally, and accordingly on
at least one plausible interpretation, the comparative greatness of concepts can be roughly elucidated
through the following two clauses:
Logic: Atheism
(g1) If concept [x] is instantiated and concept [y] is not, then [x] is greater than [y];
(g2) If concepts [x] and [y] are both instantiated, or both uninstantiated, then the greater of the
two is the one whose internal properties involve more power, wisdom, and goodness.
Clause (g1) tells us that on this interpretation the external property of instantiation “trumps” all
internal properties when assessing the relative greatness of concepts. (Whether Anselm himself
would accept this principle is debatable, but it crucially assists his argument in concluding that
the greatest concept must be instantiated.) By contrast, the relative significance of differences in
power, wisdom, and goodness is left undefined by clause (g2), but we can here ignore this com-
plication and grant to the theist that the concept [God]—whose internal properties are infinite
power, wisdom and goodness—will outscore any other concept as far as (g2) is concerned. Nev-
ertheless, this concept’s greatness will depend crucially—by clause (g1)—on whether or not it is
instantiated. If [God] is instantiated, then it will be as great as any concept can possibly be. If,
on the other hand, [God] is not instantiated, then any concept that is instantiated, for example
[Laika], will be greater. As far as the atheist is concerned, therefore, the greatest concept will be
not [God] but instead whichever concept involves the strongest instantiated combination of
power, wisdom, and goodness. Let us suppose, for the sake of argument, that this is [Aurelius],
thanks to the real existence of the absolute Roman emperor and notable Stoic writer Marcus
Aurelius.
Interpreting Anselm’s argument in this way, his formula “something-than-which-nothing-greater-
can-be-thought” is to be understood not as specifying the internal properties of some concept, but
instead as providing an external description. And if greatness conforms to clauses (g1) and (g2), it
looks plausible to claim that the concept thus described must be at least as great as any other concept
that “can be thought,” and hence must be both instantiated (in accordance with g1) and internally
impressive (in accordance with g2). This yields a form of ontological argument that cannot be dis-
missed merely on the basis of the Kantian distinction between internal and external properties.
There is, however, another objection, which comes to light if we ask exactly which concept, if
any, is correctly described by the Anselmian formula:
A concept than which no greater concept can be thought.
Here modal predicate logic, as introduced earlier, can help to clarify the issues and disambiguate the
formula (for less formal discussions of these ambiguities, see Millican 2004, 463–467, and 2018,
39–43). If we use G to represent the greater-than relation, T to represent the predicate thinkable,
allow our quantifiers to range over all concepts (whether thinkable or not), and make the existential
claim explicit, then we have a fairly natural way of interpreting the formula:
(1) 9x (Tx & 8y (Ty ! :Gyx))

There is some thinkable concept x such that for all thinkable concepts y, y is no greater than x
On this reading, however, the relevant concept x will simply be whichever concept is in fact the
greatest. The theist will take this to be [God] while the atheist will (we are assuming) take this to
be [Aurelius]. Both agree that the concept in question is instantiated (for if it were not, some other
concept would be greater), but clearly this fails to persuade the atheist of the existence of God.
Since focusing on the actual greatness of concepts is not sufficient for this ontological argument
to work, we might instead—on Anselm’s behalf—consider their possible greatness, that is, their
greatness in other possible worlds. This requires that we interpret “can be thought” not as a prop-
erty of concepts but as a modal operator, so that when we consider how great a concept can be thought
to be we are considering a concept’s possible greatness. Now the Anselmian formula, again with the
existential claim made explicit, may come out something like this:
(2) 9x 8y :hiGyx
There is some concept x such that for all concepts y, it is impossible that y should be
greater than x

Logic: Atheism
The problem here is that the atheist—who almost certainly considers every agent in the
universe to be a contingent existent—will simply reject the claim that there is some concept that
cannot possibly be exceeded in greatness. To accept that claim, indeed, would be to concede
immediately that the greatest possible concept is actually (and, indeed, necessarily) instantiated.
Why should an atheist—who takes [Aurelius] to be contingently the greatest concept—accept
that?
There is yet another fairly plausible reading of the Anselmian formula, as aiming to identify a
concept which in some possible world reaches a level of greatness that no concept could possibly
exceed in any possible world. This is more tricky to formalize, requiring quantification not only over
concepts, but also over possible worlds or levels of greatness (or both). But if we introduce m as a
name for that maximal level of greatness, and make the simplifying assumption that only one partic-
ular concept ever possibly reaches level m (as Anselm himself seems to assume), then formalization
becomes easier. Using the predicate Rxm to mean “x reaches (or exceeds) level m,” and allowing our-
selves the identity relation “=” (which is commonly added to predicate logic to yield first-order pred-
icate logic with identity), we get:
(3) 9x (hiRxm & 8y (:y=x ! :hiRym))

There is some concept x that possibly reaches level m of greatness, and no other concept y
possibly reaches that level
Or more elegantly:
(3’) 9x 8y (hiRym ! y=x)

There is some concept x such that any concept y possibly reaches level m of greatness if and
only if y is identical with x.
But this is no better for Anselm’s ontological argument, for even if the atheist accepts our simpli-
fying assumption, identification of [God] as the only concept that can possibly reach level m of
greatness is not sufficient to show that [God] reaches that level of greatness in the actual world.
Hence no contradiction can be derived from the Fool’s claim that [God] is not, in fact, instanti-
ated.
In short, then, Anselm’s ingenious formula is (at least) three-ways ambiguous. On the first
reading, it refers to the greatest instantiated concept (e.g., [Aurelius]); on the second, it fails to refer
unless there is a necessarily greatest concept; and on the third, it succeeds in referring to the concept
[God], but in a way that only requires that concept to be hypothetically instantiated. Hence there is
no reading on which the atheist is forced into contradiction, and Anselm’s argument fails.
PLANTINGA’S MODAL ARGUMENT

Alvin Plantinga’s ontological argument is explicitly couched within the language of S5 modal logic,
using two world-indexed properties, which he defines as follows (1974, 214). A being is maximally
excellent in some possible world if it is omnipotent, omniscient, and morally perfect in that same world.
A being is maximally great in some possible world if it is maximally excellent in every world (in other
words, if it is necessarily maximally excellent). Under S5, as we saw earlier, what is possible or nec-
essary does not vary from world to world, so a being who is maximally great in any possible world
must be both maximally excellent and maximally great in all of them. With this framework in place,
the only substantial premise that Plantinga needs to prove the necessary existence of a maximally
great being is that such a being is possible, for under S5, iterated modalities collapse, and therefore
possibly(necessarily(P)) is equivalent to necessarily(P).
Assuming that his framework is accepted, Plantinga’s argument is valid, and its overall
strategy—though cloaked in technical guise—is very simple. First, he defines a maximally great being
in such a way that it cannot possibly exist contingently, but must be either necessary or impossible.
Then he adds the single premise that such a being is possible, and these together immediately imply
that since it is not impossible, it must be necessary.
Logic: Atheism
Such a crude strategy, not surprisingly, is open to parody, and indeed can very easily be
used to argue for the necessary existence of almost anything at all. Take, for example, the con-
cept of a necessarily existing rabbit (mentioned earlier), whose actual instantiation, since this log-
ically cannot be contingent, must be either necessary or impossible. Then add the premise that
the concept’s instantiation is possible, which immediately implies that it must be necessary. So
there is indeed—and must be—a necessarily existing rabbit. It thus becomes clear that the
premise of Plantinga’s argument is a Trojan horse: though appearing in the guise of a modest
claim of mere possibility, it conceals a very immodest claim of absolute necessity. So even if
the argument were both valid and sound (i.e., with a true premise and conclusion), it should
not be epistemologically persuasive to any rational investigator. Given the almost trivial equiva-
lence—within Plantinga’s framework—between his premise and conclusion, if anyone is doubt-
ful whether his maximally great being is necessary, he should be equally doubtful whether it is
even possible.
But we can go further, since in such cases we have good reason to turn the modus ponens of
Plantinga’s argumentative strategy into a modus tollens (as discussed above). For suppose that X is
the concept of some nonabstract, causally active entity such as a rabbit or a god. Then the appropri-
ate attitude to the concept of a necessarily existing X is not even-handed doubt as to whether this
modal concept is necessarily instantiated or, on the contrary, impossible (since these are the only
two options). Rather, the clear lesson of Hume’s Fork—and the failed attempts to refute it—is that
nonabstract, causally active entities cannot be necessary existents. Thus if we come across a rabbitlike or
godlike concept that cannot be contingently instantiated, the reasonable presumption is that its
instantiation is impossible rather than necessary. The onus of proof here is very clearly on the propo-
nent of any would-be ontological argument for such a necessary being, and it is therefore unreason-
able to start an ontological argument from the opposite presumption.
LEIBNIZ AND GÖDEL

As we have seen, in Plantinga’s modal version of the ontological argument (and other variants in the
literature), God is defined in such a way that his possible existence entails his necessary existence.
Descartes’s argument, briefly discussed earlier, appeared to work very differently, but Leibniz viewed
it in a rather similar light, as (validly) inferring God’s necessity from his possibility. Leibniz’s objec-
tion to Descartes concerned the initial establishment of this possibility, and the absence of any proof
that the definition of God—as possessor of all perfections—was consistent. Without this, he
thought, a would-be proof of God’s existence from that definition could not be relied on: “we can-
not safely infer from definitions until we know that they are real or that they involve no contradic-
tion. The reason for this is that from concepts which involve a contradiction, contradictory conclu-
sions can be drawn simultaneously, and this is absurd” (Leibniz 1969, 293). Leibniz accordingly set
out to fix this lacuna, by proving that Descartes’s definition of God was indeed noncontradictory,
from which he felt able to conclude that the existence of God, so defined, was at least possible.
His method was to argue that a perfection is a simple quality that is “positive” and “absolute,”
and hence that the various perfections must indeed be consistent, for there is no way that a contra-
diction can be demonstrated from their conjunction.
Since Leibniz, like Descartes, treats existence as a perfection, his argument is equally sub-
ject to the Kant/Frege objection that it should instead be treated quantificationally: as involv-
ing instantiation of a concept rather than a property of an individual. But even if this could
be overcome, his case would be incomplete without clarification of his notion of simple, pos-
itive, absolute qualities, and proof that these include existence (and indeed the other standard
divine perfections). Another problem is Leibniz’s assumption that inability to demonstrate a
contradiction implies real possibility, for as we have seen in the previous section, this impli-
cation is liable to fail when considering modal definitions of God (since neither Plantinga’s
maximal greatness, nor its denial, seems to be demonstrably contradictory, and yet only one
of these can be possible).
Inspired by Leibniz, Gödel sketched a novel version of the ontological argument that was later
developed by both C. Anthony Anderson (1990) and Alexander Pruss (2009) within a formal

Logic: Atheism
framework of quantified modal logic. To give a flavor, let us briefly (and informally) discuss the ver-
sion of Pruss, which starts from the following axioms:
Axiom F1 If A is positive, then :A is not positive.
Axiom F2 If A is positive and A entails B, then B is positive.
Axiom N1 Necessary existence is positive.
Axiom N2 Each of the properties of essential omnipotence, essential omniscience, and essential
perfect goodness is positive.
Axioms F1 and F2 plausibly imply that one positive property cannot exclude another (since if P is
positive and excludes Q, then P entails :Q, so from F2 it follows that :Q is positive, which by
F1 implies that Q is not positive). So if P and Q are both positive, they must be mutually compos-
sible, which—given S5—implies that they must be mutually instantiated at some possible world.
But then if the properties P and Q both involve necessity, they must also be mutually instantiated
at every possible world (since as we noted with Plantinga, in S5 what is possible or necessary does
not vary from world to world). The upshot—given N1 and N2—is that “there exists an essentially
omnipotent being, and there exists an essentially omniscient being, and there exists an essentially
perfectly good necessary being” (Pruss 2009, 349). If we are permitted a further axiom implying
that the conjunction of the properties in N2 is itself positive, then we can conclude that there is a
single being that combines all of these remarkable properties, that is, God.
In the light of our previous discussion, this argument cannot plausibly be considered an epistemo-
logically persuasive piece of natural theology. Some objections are fairly obvious: it relies on an intui-
tive and unanalyzed notion of a “positive” property; its third axiom takes for granted that necessary
existence is a legitimate property; while its fourth axiom both makes controversial assumptions about
the coherence of essential properties, and ignores the potential conflicts between properties that we
have seen in the section “Divine Properties and Paradox” above. Another feature that should make
us suspicious is the potentially open-ended nature of the proposed axioms, especially the second: “If
A is positive and A entails B, then B is positive.” This has counterintuitive implications, for example,
that if the property of being omniscient is positive, then so is the property of being either omniscient or
utterly ignorant (since the one entails the other). And “entails” is rather vague here, especially when
combined with modal properties which, as we have seen with Plantinga, can turn out to be Trojan
horses when instantiated within S5 (where apparent mere acceptance of a possibility can so easily
smuggle in a necessity). Suppose, for example, the atheist modestly proposes that it is possible that
there is no necessarily existent essentially omnipotent being. As in Plantinga’s framework, this mere
possibility immediately implies the impossibility of such a being. But then necessary existence would
be incompossible with essential omnipotence, so that the former would (in at least one plausible sense)
entail the negation of the latter, forcing rejection of at least one of the axioms.
It is only to be expected, given a valid argument, that the negation of its conclusion should
imply falsehood among its premises: this by itself is not enough to undermine them. But the more
ambitious the conclusion is, the more plausible its rejection becomes, and thus a greater burden of
proof falls on the premises that would rule such rejection out. To side with the Gödelian argument,
we must accept four axioms, all apparently based on little more than intuitive appeal, and at least
three of them potentially objectionable for reasons we have briefly canvassed. To oppose the Göd-
elian argument, we need only agree with the atheist’s apparently modest claim that theism is possibly
false. I would suggest that, in this situation, the reasonable verdict is very clear.
Since Gödel came up with his original argument, deeper analysis has emphasized how little
confidence we can have in our “intuitive” acceptance of general axioms such as his. More than thirty
years ago, J. Howard Sobel (1987) showed that they lead to “modal collapse” whereby every truth
becomes necessary, with necessarily(P) being derivable from P (see also Sobel 2004, 132–135). This
stimulated Anderson (1990) and others to adapt the argument, but despite further discussion over
the intervening decades, it was only very recently that computational analysis revealed a hitherto
unsuspected formal inconsistency in Gödel’s axioms (Benzmüller and Woltzenlogel Paleo 2014,
2016). My point here is not to deny that Gödelian arguments can be ingeniously devised to avoid
any such formal refutations, but rather to suggest that—as in the case of defining God’s qualities
Logic: Atheism
to avoid another nest of potential paradoxes (as discussed above)—we end up playing an artificial
philosophical game instead of pursuing a serious epistemological inquiry. The game is of some inter-
est, and has provided an excellent testing ground for novel methods in computational logic, but it
has a negligible prospect of adding any support to the case for theism. As we have seen repeatedly
in this chapter, logical validity is one thing, and epistemological persuasiveness quite another. The
existence of an infinite, morally perfect spirit is a huge metaphysical claim, radically out of line with
our daily experience and apparently running counter to the obvious evidence of the world’s moral
imperfection (as discussed earlier). To establish such a claim, an argument needs not only to operate
with good logic but also to start from premises that are sufficiently convincing to outweigh the ini-
tial implausibility of its conclusion. Most contemporary philosophers deny that there is any such
argument available, but even those who are more sympathetic to theism are very unlikely to expect
such an argument to come from the domain of a priori logical systems and their axioms.
Bibliography
Anderson, C. Anthony. “Some Emendations of Gödel’s Ontological Proof.” Kripke, Saul. Naming and Necessity. Oxford: Blackwell, 1980.
Faith and Philosophy 7, no. 3 (1990): 291–303.
Leftow, Brian. “Omnipotence.” In The Oxford Handbook of Philosophical
Anselm. Proslogion. (1077–1078.) Translated by M. J. Charlesworth. In Theology, edited by Thomas P. Flint and Michael C. Rea, 167–198.
Anselm of Canterbury: The Major Works, edited by Brian Davies and Oxford: Oxford University Press, 2009.
G. R. Evans, 82–104. Oxford: Oxford University Press, 1998.
Leibniz, Gottfried Wilhelm. Philosophical Papers and Letters. Translated and
Aquinas, Thomas. Summa theologiae. c. 1269. edited by Leroy E. Loemker. Dordrecht, Netherlands: D. Reidel, 1969.
Benzmüller, Christoph, and Bruno Woltzenlogel Paleo. “Automating Gödel’s Mackie, J. L. The Miracle of Theism: Arguments for and against the Existence of
Ontological Proof of God’s Existence with Higher-Order Automated God. Oxford: Clarendon; New York: Oxford University Press, 1982.
Theorem Provers.” In ECAI 2014: Proceedings of the Twenty-First European
Millican, Peter. “The Complex Problem of Abortion.” In Ethics in
Conference on Artificial Intelligence. Amsterdam: IOS Press, 2014.
Reproductive Medicine, edited by David R. Bromham, Maureen E.
Benzmüller, Christoph, and Bruno Woltzenlogel Paleo. “The Inconsistency in Dalton, Jennifer C. Jackson, and Peter J. R. Millican, 161–188. London:
Gödel’s Ontological Argument: A Success Story for AI in Metaphysics.” In Springer-Verlag, 1992.
IJCAI 25: Proceedings of the Twenty-Fifth International Joint Conference on
Artificial Intelligence. New York: AAAI Press, 2016. Millican, Peter. “The One Fatal Flaw in Anselm’s Argument.” Mind 113,
no. 451 (2004): 437–476.
Descartes, René. The Philosophical Writings of Descartes. Vol. 2: Meditations,
Objections, and Replies. Edited by John Cottingham, Robert Stoothoff, and Millican, Peter. “Hume’s Fork, and His Theory of Relations.” Philosophy and
Dugald Murdoch. Cambridge, UK: Cambridge University Press, 1984. Phenomenological Research 95, no. 1 (2017): 3–65.
Descartes, René. The Philosophical Writings of Descartes. Vol. 3: Correspondence. Millican, Peter. “Anselm.” In Ontological Arguments, edited by Graham
Edited by John Cottingham, Robert Stoothoff, Dugald Murdoch, and Oppy, 19–43. Cambridge, UK: Cambridge University Press, 2018.
Anthony Kenny. Cambridge, UK: Cambridge University Press, 1991. Plantinga, Alvin. The Nature of Necessity. Oxford: Clarendon, 1974.
Flint, Thomas P., and Michael C. Rea, eds. The Oxford Handbook of Pruss, Alexander R. “A Gödelian Ontological Argument Improved.”
Philosophical Theology. Oxford: Oxford University Press, 2009. Religious Studies 45, no. 3 (2009): 347–353.
Hume, David. An Enquiry concerning Human Understanding. (1748, 1777.) Quine, W. V. “Two Dogmas of Empiricism.” Philosophical Review 60, no. 1
Edited by Peter Millican. Oxford: Oxford University Press, 2007. Also at (1951): 20–43.
https://www.davidhume.org/texts/e/. (References by section and para-
graph number.) Rowe, William. “Divine Power, Goodness, and Knowledge.” In The Oxford
Handbook of Philosophy of Religion, edited by William Wainwright,
Hume, David. A Dissertation on the Passions and The Natural History of 15–34. Oxford: Oxford University Press, 2005.
Religion: A Critical Edition. (1757, 1772.) Edited by Tom L.
Beauchamp. Oxford: Clarendon; New York: Oxford University Press, Sobel, J. Howard. “Gödel’s Ontological Proof.” In On Being and Saying:
2007b. Also at https://www.davidhume.org/texts/n/. (References to Essays for Richard Cartwright, edited by Judith Jarvis Thomson,
Natural History by section and paragraph number.) 241–261. Cambridge, MA: MIT Press, 1987.
Hume, David. Dialogues concerning Natural Religion. Edinburgh, 1779. Also Sobel, J. Howard. Logic and Theism: Arguments for and against Beliefs in God.
at https://davidhume.org/texts/d/. (References by section and paragraph Cambridge: Cambridge University Press, 2004.
number.) Swinburne, Richard. The Coherence of Theism. Oxford: Clarendon, 1977.
Kant, Immanuel. Critique of Pure Reason. Rev. ed. Translated by Norman Swinburne, Richard. The Existence of God. 2nd ed. Oxford: Clarendon;
Kemp Smith. Basingstoke, UK: Palgrave Macmillan, 2007. First New York: Oxford University Press, 2004.
published 1781. Wierenga, Edward R. “Omniscience.” In The Oxford Handbook of
Kenny, Anthony. The God of the Philosophers. Oxford: Clarendon; New Philosophical Theology, edited by Thomas P. Flint and Michael C.
York: Oxford University Press, 1979. Rea, 129–144. Oxford: Oxford University Press, 2009.

TOPIC 4
Doxastic Foundations: Theism

Paul K. Moser
Loyola University Chicago, IL
This chapter focuses on some epistemic concepts and their bearing on theism. It considers the nature of belief
both as assent and as a disposition involving trust. It also characterizes foundational evidence of God’s
reality in terms of divine self-manifestation in human moral conscience, whereby a unique kind of agapē-
conviction can arise.
BELIEF
Some people believe that God exists; others do not. Some people, in addition, believe in God,
owing to their favorable commitment with trust toward God; others do not. What accounts for
these differences? An answer to that question requires an answer to a prior question: What exactly
are these differences? They are real differences, and we shall begin with what underlies them.
Belief that something is so (for instance, that God exists) is a psychological state of a person
that is related to a judgment or proposition that something is the case (for instance, that God
exists). The psychological state in question is an affirmative attitude of a person toward a proposi-
tion. It is affirmative in a dispositional manner that does not require constant assent to the proposi-
tion, even if it begins with an episode of assent. It thus is akin to my dispositional tendency to
scream something when a nail pierces my skin. We do not lose our beliefs when we fall asleep
and temporarily stop our episodes of assent. In this regard, belief that something is so is not simply
episodic, but has a dispositional component. This component includes one’s being disposed to
affirm, or assent to, the relevant proposition under certain circumstances. So, if you believe that
God exists, you will be inclined to affirm under certain circumstances the proposition that God
exists. We need not pursue here the complicated matter of specifying those circumstances.
Belief in God is not reducible to or entailed by belief that God exists. It includes, in contrast
with belief that God exists, one’s being related to God by a favorable commitment with trust toward
God. One can believe that God exists without one’s having a favorable commitment with trust
toward God. Indeed, one can believe that God exists and fully distrust God. A controversial issue
is whether one can believe in God without believing that God exists. It seems that one could have
a favorable commitment with trust toward God without having a concept of God. For instance, one
could have such a commitment toward what guides one’s conscience while one fails to conceive of
that guide as God. Even though one believes in God, owing to a favorable commitment toward
God’s role in one’s conscience, one could fail to conceive of the being one believes in as God.
One then could have a favorable commitment with trust toward God, but not think of the object
of one’s commitment as being God. This seems to be a live option. So, I do not assume that belief
in God requires belief that God exists, even if it requires a less demanding belief that a distinctive
object (of one’s favorable commitment) exists. We thus can distinguish between implicit faith in
God (without endorsement that God exists) and explicit faith in God (including endorsement that
God exists).
103
Assent to a judgment or proposition differs from commit-

KEY CONCEPTS ment to a reality that is not a judgment or proposition. This
is a distinction between a de dicto object and a de re object.
Some objects of commitment are de re and not de dicto. God,
Abductivism
for instance, could be an object of favorable commitment with
Agapē trust de re for a person, but not be an object of commitment de
Belief dicto for that person. In this respect, God would be similar to
Commitment many ordinary objects of belief de re. I can believe, for instance,
that this thing exists, even though I lack a fitting specific con-
Conviction cept (such as that of a turbocharger). We should acknowledge
De Dicto a minimalist way of selecting, or at least focusing on, an object
De Re de re, such as with a simple demonstrative (for example, “this”
or “that”). The demonstratives “this” and “that,” of course, do
Doxastic not entail an actual concept of God, given its robust content.
Epistemic Abductivism A de re relation in a belief, regardless of whether a de dicto
Epistemology component accompanies it, is inherently causal, being affected
Evidence by an actual object to which it is related. My belief, for
instance, in my friend’s moral character is causally influenced
Imitatio Dei by that moral character. A de dicto belief, in contrast, need
Intentional not be causally related to a nonpropositional or nonconceptual
Reformed Epistemology object. For instance, my belief that black holes are supremely
dense need have no causally influential object. In the latter case,
Skepticism I could hold the belief without having any causal influence
Trust from an object that supplies evidence for (the truth of) that
Truth Indicator belief. If there is no causal influence from evidence, however,
a belief will lack causal grounding in evidence, and its evidential
base will be causally deficient.
EVIDENCE AND KNOWLEDGE
A belief that something is the case may or may not have support from evidence. When it does have
evidential support, it has a truth (or factuality) indicator, that is, an indicator that it is true or factual.
Evidence for a claim is a truth indicator for that claim, and it can be fallible (capable of supporting a
falsehood), misleading (regarding actual truth), and defeasible, or overridable, by other evidence. In
the absence of a truth indicator, any one belief is no better than any other belief from an evidential
point of view; it then suffers from an epistemically debilitating kind of arbitrariness. In addition,
evidence is person-relative, being relative to the psychological and experiential perspective of some
person or other. In this respect, evidence differs from truth or factuality. Actual evidence, in contrast
with merely available or potential evidence, is what indicates truth or factuality for a person at a
time. It thus is irreducible to anything devoid of psychological features. One might talk of merely
available or potential evidence in terms, for instance, of just logical relations between propositions,
but such evidence would not be actual in the relevant sense, owing to its failure to indicate truth
from the actual perspective of a person.
Evidence is required for knowledge of the conception and kind examined in Plato’s Theaetetus,
the seminal work for epistemology in Western philosophy. Theorists, of course, can introduce alter-
native conceptions of knowledge, and some have done so, including conceptions that lack an evi-
dence condition. (On the philosophical import of conceptual variability, see Moser 1993.) If, how-
ever, we have an interest, theoretical or otherwise, in the conception and kind of knowledge pursued
in the Theaetetus (namely, factual knowledge), we will want to attend to the kind of evidence
required by knowledge that something is so. Such knowledge will not be merely reliably formed
belief, but will require knowers to have some kind of truth indicator, or evidence, for what they
know. For instance, if we have an interest in the kind of knowledge of God’s reality that entails
evidence of God’s reality, we can proceed with the general conception of knowledge offered in the
Theaetetus and attend to the kind of evidence it requires.
It does not follow that we should characterize knowledge as simply justified true belief. Even if
knowledge entails true belief with adequate evidence, we can acknowledge that cases of the kind
identified by Edmund Gettier call for a fourth condition for knowledge, beyond the evidence, truth,
and belief conditions. (See Gettier 1963; cf. Shope 1983, 2002, where three-condition approaches
to Gettier-style cases are challenged.) The specification of the fourth condition has generated
extensive controversy in epistemology, and no consensus on it has emerged. Even so, problems from
Gettier-style cases have not threatened human knowledge of God. Their challenge instead bears on a
precise analysis of the concept or kind of knowledge that requires justified true belief. So, we need
not digress now to a solution to that conceptual challenge. (For a proposed solution, see Moser
1989.)
Like belief, evidence and knowledge can have de dicto and de re components, depending on what
a person who has evidence and knowledge is related to (a judgment or something else). Famously,
Bertrand Russell (1911) distinguished knowledge by description (being de dicto) and knowledge by
acquaintance (being de re). In doing so, he identified a kind of knowledge that cannot be reduced
to factual knowledge that something is the case. It is controversial among philosophers what the rele-
vant acquaintance in knowledge by acquaintance encompasses, but it is at least de re in being causally
related to an object that differs from a judgment or proposition. In a theological context, a prominent
approach to such knowledge emerges from Martin Buber’s perspective, inspired by Søren Kierke-
gaard’s pseudonymous Climacus, on an “I–Thou” relation between a human and God. (See Buber
1958 [1923] and Farmer 1942.) Since God is a personal “Thou” in this approach, the knowledge
in question would be not just de re but also de te, in relation to a “You.” The “object” of knowledge
then would be not just an object, but also a subject, that is, a personal agent with self-consciousness
and intentional self-determination. This would bear directly, we shall see, on the kind of evidence
to be expected for human knowledge of God. Despite misgivings from some philosophers, we have
no good reason to exclude evidence for a proposition from a nonconceptual basis. Such a basis can
exhibit properties in human experience that qualify as evidence for a person.
In the tradition of Plato and Aristotle, Russell thought of knowledge by acquaintance as “foun-
dational,” as noninferential in that its evidence component does not depend (for being evidence) on
inference from beliefs. In Russell’s account, the noninferential evidence comes from what one is
directly acquainted with, for instance, sensory content (such as a red circular image) that is not a
proposition or a concept and does not depend on inference from a belief. Some philosophers hold
that foundational beliefs are “certain” in some sense, but this is optional; a belief depending on non-
inferential evidence need not be certain. It also would be misleading to portray foundationally justi-
fied beliefs as being “self-justified,” because they receive their evidence from something other than
themselves. Their key feature is that they do not owe their evidence to inference from other beliefs.
Their evidence, we shall see, comes from a different basis.
A STARTING POINT
Some proponents of so-called reformed epistemology regarding knowledge of God acknowledge

noninferentially justified beliefs about God, and label them as “properly basic” (see Plantinga
1983; Wolterstorff 1983). However one understands the details for being “properly basic” (various
accounts have been offered), the motivation for reformed epistemology stems from opposition to
evidentialism, the view that a belief suited to knowledge requires support from evidence. Opposing
evidentialism regarding theistic belief, Alvin Plantinga asks: “Why isn’t it perfectly sensible to start
with belief in God? Why does belief in God have to be probable with respect to some other body
of evidence in order to be rationally defensible?” (1993, 104).
The short answer, neglected by Plantinga, is that human belief in God is not a “body of evi-
dence” (at least in any non-questionbegging sense of the phrases), and it does not supply its own
needed truth (or factuality) indicator as a safeguard against credulity or arbitrariness in belief. To

avoid credulity or arbitrariness, such belief needs a truth indicator other than itself. In particular,
it needs a truth indicator regarding God’s reality to which human belief in God is suitably related.
In needing such a truth indicator, the belief in question needs evidence (regarding God’s reality)
irreducible to human belief. W. K. Clifford offered a related, more general principle: “It is wrong
always, everywhere, and for anyone, to believe anything upon insufficient evidence” (1877). If we
limit the principle to what is epistemically wrong, it merits serious consideration, but we cannot
digress to details. Avoidance of credulity requires starting, not with human belief, but with the
relevant truth indicator, perhaps a truth indicator courtesy of divine self-manifestation to a
human. A truth indicator from a divine intervention in human experience would not be reducible
to a human belief, and it would be a safeguard against credulity. Such a truth indicator would
have cognitive priority over mere belief in God as a starting point for inquiry, if only because it
would figure in a cognitively responsible answer to the key question of whether one actually has
belief in God.
If a reformed epistemologist offers belief in God as de re, in the proposal to “start” with belief in
God, this question will arise: Do we really have belief in God, in contrast with our merely supposing
that we have belief in God? (A related question will confront a de dicto understanding of belief that
God exists.) So, starting with belief in God will beg or ignore a key question that needs a cogent
answer for responsible inquirers. The lesson is that a cognitively prior question bears on any claim
to belief in God: Is there evidence from truth indicators for the alleged belief in God? Cognitively
responsible inquiry can start with the latter question, but it will not start with belief in God itself.
It is one thing to say that belief in God can be epistemically reasonable without relying on any
argument. It is quite another to say that such belief can be epistemically reasonable without relying
on any evidence from truth indicators. If “starting with” belief in God forgoes the cognitive priority
of such evidence, it will invite a serious challenge regarding credulity or arbitrariness from the per-
spective of a truth indicator. In any case, we should not identify evidence with an argument or with
propositions, although many advocates of reformed epistemology (including Plantinga 1983, 79)
neglect this lesson. An experience that is neither an argument nor a proposition (or even a concept),
such as an experience of a divine self-manifestation, can be a truth indicator and thus evidence for a
belief (including belief in God).
The main idea of reformed epistemology concerning a “starting point” stems in modern times
from the early Karl Barth (1962 [1925]). (Plantinga 1983, 72, includes Barth among proponents of
the reformed view that belief in God is properly basic.) We can use Barth’s position in 1925 to illus-
trate the deficiency of reformed epistemology in its proposed “starting point.” Barth puts “the Word
of God” first, as his theological starting point, in his sharp contrast of it with evidence from human
experience. He remarks:
“How the Holy Spirit is given” … is precisely what man must not wish to show.
(Melanchthon made the ill-judged attempt long ago). We must proclaim the Word,
subordinate the cure of souls to the sermon and not the reverse. And through the Word of
God, God himself will let himself be revealed. ([1925] 1962, 263)
He puts his point in terms of Jesus, as follows: “In the power of Jesus we ‘apprehend’ God acting.
But this statement cannot be established on the basis of preceding experiences” ([1925] 1962, 264).
Barth thus takes exception to what he regards as the evidential road of human experience from
below upward toward God.
Barth’s starting point includes “the road from above downwards,” and this seems common in
reformed epistemology. He writes:
This truth [of God revealed in Christ] is the beginning, the basis, and the presupposition apart
from which Christian preaching and dogmatics cannot say one meaningful word concerning
Christ. Without that truth, they both remain undeniably stuck fast in history. Here no other
“way” whatever exists except the road from above downwards. ([1925] 1962, 265)
We thus see that Barth’s dissent from the cognitive priority of evidence for theology and belief in
God focuses on a starting point, or beginning, for theology and belief in God. This is characteristic
of reformed epistemology, in an effort to give God cognitive priority.
Barth’s central mistake, shared by various other proponents of reformed epistemology, is to rec-
ommend a theological starting point (and belief in God) while neglecting the cognitive priority of
the evidential perspective of human recipients of divine revelation. If a starting point is to be cogni-
tively relevant to humans, then recommending or endorsing it must not be altogether arbitrary or
credulous from their evidential perspective. Instead, recommending or endorsing it must be eviden-
tially significant for them, given support in their evidential perspective. Otherwise, nothing in their
evidential perspective would make it cognitively responsible for them to recommend or endorse
the starting point in question. In that case, the starting point on offer would have no claim on them
from their evidential perspective. Indeed, from the latter perspective, any other arbitrary starting
point would be equally good. A serious arbitrariness in starting point then would threaten that per-
spective, owing to neglect of the cognitive priority of an evidential basis for recommending or
endorsing the starting point. As a result, any recommendation or endorsement of the starting point
would be arbitrary from the evidential perspective of the people in question. So an implausible,
credulous dogmatism and fideism would be a risk for that perspective, given that it lacks indication
of the truth or even the commendability of the starting point on offer.
The mistake in Barth’s approach, and in reformed epistemology in general, stems from the
assumption that “the datum with which dogmatics has to begin … [is] God himself in his Word”
(1925, 260). Barth’s similar language for his proposed starting point for theology is “God giving
himself to be known through his Word” ([1925] 1962, 260). This language will mislead, because
we have an unanswered pressing question: Giving himself to whom? Thus far we have no specifica-
tion of “God giving himself” to anyone at all. In particular, we have no specification of God relating
to humans at all, such as in their experience. So Barth’s talk of a starting point via “God himself in
his Word” and “God giving himself” is misleadingly nonrelational concerning any humans and their
evidential perspective. It shares the defect of one’s speaking of an offer of friendship but failing to
mention to whom the offer was made. Such talk would be unacceptably indeterminate in content,
at least for any notion of an actual friendship.
Barth, as noted, claims that “through the Word of God, God himself will let himself be
revealed” ([1925] 1962, 263). We see the indeterminacy in this claim’s semantic content by noting
its failure to answer this pressing question: Revealed where, and to whom? Merely in the abstract,
and to no one at all? In that case, we lack (a claim to) determinate, actual revelation to actual
humans. We thus see that Barth has not offered a well-formed, semantically determinate approach
to a starting point for theology or belief in God. Once we answer the pressing question at hand,
for the sake of semantic determinacy, we introduce the evidential perspective of a specific human
being. We then involve a human evidential perspective in the starting point, and thereby a human
as a personal, psychological agent with experiences pertinent to evidence that bears on, and is cog-
nitively prior to, a proposed starting point for theology or belief in God.
In ignoring the cognitive priority of truth indicators in a human evidential perspective, Barth’s
approach neglects that divine–human revelation is inherently interpersonal in a manner that affects
a recipient’s personal, evidential perspective. Such revelation does not fit with any mechanical or
coercive approach that omits a human decision from a personal, evidential perspective. Otherwise,
divine revelation would become one-sided in a way that excludes human recipients as evidence-
sensitive agents who are to make cognitively responsible decisions about starting points for theology
and belief in God.
Reformed epistemology, in neglecting the cognitive priority of truth indicators as evidence,
shares the defect of Barth’s approach. We see this defect in Plantinga’s reformed epistemology,
according to which “we human beings, apart from God’s special and gracious activity, are sunk in
sin; … without some special activity on the part of the Lord, we wouldn’t believe” (2000, 269;
cf. 303). We are sunk so deeply in sin that our coming to have Christian faith must come “by
way of the work of the Holy Spirit, who gets us to accept, causes us to believe, these great truths
of the gospel” (2000, 245; cf. 260). Plantinga notes that God’s causing us to do something rules
out our doing it freely (2000, 462). Given his view, it follows that no one freely comes to have
Christian faith. This position is distinctively Calvinist, particularly in its suggestions of God’s selec-
tive irresistible grace in regeneration of the “favored” (2000, 254).

Plantinga looks for support in Jesus’s remark (John 6:44) that “no one can come to me unless
the Father who sent me draws him” (2000, 269). The proposed Calvinism, however, fails to con-
vince. John’s Gospel, in particular, resists a Calvinist reading of divinely selected and caused regen-
eration. The notion of “drawing” people in John’s Gospel suggests both universality of drawing and
human freedom to reject or to embrace the drawing. Accordingly, Jesus remarks that “when I am
lifted up from the earth, I will draw all (pantas) [people] to myself” (John 12:32). John’s Gospel
(3:16) teaches that God so loved the whole world (ton kosmon) that he gave his unique Son, so that
anyone who trusts in him will have everlasting life. John’s Gospel does not portray God as causing
some people but not others to believe. Instead, it assumes a typical free human response of receiving
or rejecting God’s gracious salvation intended for all people (see John 1:11–13, 3:19–21). Accord-
ing to John’s Gospel, in seeking genuine love from humans and thus respecting human freedom,
God did not cause Jesus to obey (see John 10:17–18); likewise, God does not cause the “favored”
to believe or to be faithful.
In the New Testament generally, God graciously gives humans the freedom to receive or to
reject God’s salvation (see Meadors 2006; cf. Robinson 1956). In contrast, given the Calvinist story
of God’s regenerating just some favored people in moral and spiritual bondage, God would show a
kind of selective love incompatible with genuine love of all people, including enemies. As a result,
God would lack moral perfection and thus worthiness of worship (see Moser 2019a). In addition,
God would be inconsistent in commanding humans to love all of their enemies. By contrast, in
offering a gift of faith and salvation to all (and not just the favored), God would want none to perish
but everyone to come to repentance (see 2 Peter 3:9; cf. Revevelation 3:20). This divine want would
be free of harmful deceit; otherwise, its source would not be worthy of worship.
We can put our correction of reformed epistemology in terms of the basis of faith (or belief) in
God. Barth proposes that “the miracle of the epiphany [of God in Christ] was not merely the con-
tent of faith but was also the basis of faith” ([1925] 1962, 266). This proposal is misleading given
Barth’s one-sided understanding of “the content of faith”; it would need to be reworked to fit with
a different, interpersonal understanding sensitive to a recipient’s evidential perspective. Human faith
in God, in the dominant biblical perspective, is irreducibly interpersonal in a way that accommo-
dates two kinds of personal agency and power: divine and human. Regarding the human side, the
perspective relates the basis of faith in God to a human evidential perspective that is not exhausted
by divine agency, causality, or power. It recognizes that a human person’s faith in God needs a basis
in something that figures in that person’s evidence, including that person’s experience, regarding
God. We thus might say that it acknowledges a de re evidential component in human faith in
God. If we omit this component, we undermine the option that we humans have grounded faith
in God. (For discussion, see Moser 2017, chaps. 2–3.)
Barth’s mistake appears to include a confusion between an ontic end point for a theology and a
cognitively relevant starting point for a theology. The claim that God intervenes first, prior to
human seeking, might emerge as an ontic result, or end point, of theological inquiry, but it would
not follow that this claim serves as a cognitively responsible starting point for such inquiry. Such a
starting point, to avoid the credulity of arbitrary dogmatism or fideism, must be evidentially relevant
to a recipient’s perspective of truth indicators for inquiry. One eventually might reach the ontic end
point in question, but this result will be cognitively responsible in inquiry only if one’s starting
point for inquiry is evidentially relevant and cognitively responsible. If we neglect the cognitive pri-
ority of such a starting point, we invite credulous, arbitrary dogmatism and fideism, and our inquiry
thus becomes overly tolerant toward options for theological commitment. An “anything goes”
approach then threatens. If I may reasonably begin with my assumed God (or my belief in God),
in my preferred sense, then you may reasonably begin with your opposing assumed God (or your
opposing belief in God). Our opposing commitments regarding our assumed Gods then will “rea-
sonably” proceed for us relative to our favored beliefs, in the absence of a cognitively prior starting
point in our truth indicators. We will be left, in that case, without aid from a cognitively responsible
means of adjudicating the opposing views before us.
If mere belief in God is the starting point for reformed epistemology, the cognitive priority of
an evidential perspective of truth indicators loses its needed role in cognitively responsible
theological inquiry and belief. An experience-oriented evidential approach would be more responsi-
ble, from an evidential and explanatory point of view, than any proposal to start with belief in God.
In addition, such an evidentialist approach is compatible with the dominant biblical teaching that
God initiates communication with humans and their redemption. Contrary to Barth and some
other reformed writers, it poses no threat to God’s priority in such areas, and it has the important
benefit of portraying divine–human revelation as genuinely interpersonal and anchored in evidence.
Plantinga shares Calvin’s view that God’s Spirit somehow conveys “circumstances” that make belief
in God reasonable. One would need to recommend or endorse those circumstances in responsible
inquiry, however, without begging or ignoring the question whether one has genuine belief in
God. A cognitively responsible starting point will accommodate this lesson in a way neglected by
reformed epistemology. We shall consider a needed alternative that has a Christian emphasis but
is extendable to a morally perfect God who intervenes and works across theistic and even religious
boundaries.
A SELF-EVIDENCING FOUNDATION
Foundational evidence for a belief in God, if there is such, would not be a belief; it certainly would
not be a belief in God. It would need to be nonpropositional and nonconceptual if it is to give ulti-
mate evidential support to propositional and conceptual commitments by humans, including beliefs
that have such commitments. An ideal candidate for such evidence would come from the experience
of divine direct self-manifestation to a person. God would directly self-manifest to a person by
directly self-revealing God’s personal character to that person in experience, such as the experience
of moral conscience. Only God would be able to do this, because only God would have control over
directly self-revealing God’s personal character.
If, by divine title, God is worthy of worship, then God would be inherently perfectly good, and
thus the self-presentation of God’s personal character would include the self-manifestation of God’s
perfectly good moral character. In using the term “God” as such a perfectionist title, we give it nor-
mative value without assuming that God exists. So we avoid begging a key question against people
doubtful of God’s existence. In addition, we maintain an assumption of Abrahamic monotheism
that God would be without moral defect, because worthiness of worship, requiring worthiness of
full, unrestricted commitment, would entail moral perfection. Someone might propose a morally
inferior notion of God, but that would be to change the subject from traditional monotheism,
and, more to the point, it would be premature now. One might consider that option if the morally
superior notion ultimately finds no actual titleholder, but we have not reached that place of failure.
As suggested, the God in question would not be limited in intervention or work to a Christian con-
text that excludes other versions of theism or religion. This God could work redemptively even
among polytheists and atheists.
Our adopted notion of God will bear directly on our epistemic assessment of whether God
exists, owing to the parameters for relevant evidence suited to that notion. Conversely, an approach
to evidence can set standards for evidence or justification that conflict with, or otherwise challenge,
an evidence-based commitment to God’s reality. For instance, Richard Dawkins (2006, 52, 59)
offers an approach to evidence requiring that belief in God meet standards for scientific evidence,
as if such belief would have to be a publicly verifiable scientific hypothesis. This includes the
requirement that the evidence for belief in God be publicly sharable, rather than private. Such an
approach is question begging against many proponents of theism, who deny that belief in God,
given the character of God, fits with the public evidence of a scientific hypothesis. The lesson, in
any case, is that one’s notion of God matters for epistemic assessment of belief in God, and the
same is true of one’s notion of evidence for reasonable belief. Cognitively responsible inquiry needs
to attend carefully to both considerations.
In self-manifesting a perfectly good moral character to a person, God would be self-evidencing
for that person. As an intentional agent, God could undertake the intentional action of showing a
living divine character to that person. In doing so, God would give that person direct evidence of

God’s reality and presence. Intentional agents have the capacity of being self-evidencing in their
action toward others; they thus differ from arguments, beliefs, or claims. The intentional power to
intervene by one’s self-revealing in another person’s experience is limited to intentional agents,
and such power plays a distinctive role in interpersonal evidence and knowledge of a personal
agent’s reality. This role is evidentially basic to any role played by arguments, beliefs, or claims in
such evidence and knowledge. It is a foundational evidential role, and a personal God could exercise
that role distinctively in self-presenting the divine moral character to humans.
The self-manifestation of God’s moral character to a person would feature something inherent
in that character: unselfish caring toward the person for that person’s moral good. Part of this caring
would include God’s offering an ongoing relationship of morally relevant interaction with that per-
son for that person’s moral benefit. This kind of caring includes righteous, morally good love, and it
is called agapē by the earliest Greek-speaking Christians. Such self-manifested agapē could appear in
human conscience, as one is challenged, convicted, and guided to redirect one’s life toward imitating
and cooperating with God’s morally perfect character. Even so, it would be rejectable by humans as
they suppress or ignore any such challenge in conscience. God would not coerce humans to accept a
divine challenge or offer, because God would seek to preserve human agency in response to a divine
offer of a relationship. Otherwise, God would not have genuine agents to whom God relates. The
aforementioned Calvinist story of divine causation neglects this lesson and excludes genuine human
agency in response to God. It thus omits a place for human responsibility in relating to God.
One might say, as many theologians do, that the divine self-presentation of agapē comes as an
unconditional gift to a person, by divine grace rather than by human earning. The reality of the sit-
uation, however, would be more complicated. A perfectly caring God would not offer a gift in a way
that the offer or the gift brings moral harm to potential recipients. An unconditional gift of God’s
caring presence could be abused by a person in a way that morally harms that person. For instance,
one could use the gift for selfish purposes that morally corrupt oneself and bring harm to others. In
such a case, human control over having a divine gift could oppose the purpose of the gift and empty
it of redemptive value. As a result, a perfectly caring God would withdraw the gift from a person, in
order not to bring moral harm to that person.
The divine gift in question would include agapē as a divine gift to humans, but ideally it would
not be merely an event or episode of agapē. Instead, it ideally would include a divine offer of an
ongoing relationship of agapē with humans for their moral benefit. A divine expectation, for the
good of humans, would be human reciprocity in a relationship of agapē toward God (and other
humans). So the mutuality of agapē in an ongoing relationship would be God’s goal in self-manifes-
tation toward humans. Mere human knowledge that God exists would fall short of this goal. Such
factual knowledge that God exists would serve only as a means to God’s more demanding goal of
an ongoing relationship of mutual agapē with humans. The latter goal would result from God’s
being perfectly good toward humans, and therefore seeking what is morally and spiritually best for
them. (On this goal, see Moser 2017, chaps. 2, 5.)
It would be morally counterproductive for God to offer an ongoing relationship of agapē to
humans if they would abuse it for their opposing selfish purposes. Such an offer also could
strengthen, rather than weaken, human opposition to God, in cases where humans are set on their
selfishness. So God would wait for an opportune time to offer the relationship to humans, given a
divine desire for human cooperation with it. This could include God’s withdrawing an offer until
humans are ready to cooperate with it. Such withdrawing, however, would not be to withdraw
divine agapē toward humans. Instead, it would stem from divine agapē for them, including a sincere
concern for what is morally best for them.
The divine offer of an ongoing relationship with humans, and the corresponding gift of grace in
relationship, would be sensitive to human receptivity as cooperation. So they would not qualify as
ordinary, irrevocable “gifts.” They would not be given by a perfectly good God without the prospect
of withdrawal or retraction. Instead, they would be divine entrustments or entrusted gifts subject to
God’s withdrawal or retraction upon (impending) human abuse. So it can be misleading to talk of
divine grace in an offered relationship as a “gift” without qualification. In addition, such talk can
have a negative effect on human attitudes toward God, leading people to take a divine gift of grace
for granted, as if it did not come with divine expectations or conditions for its maintenance by
them. The negative effect can lead to a neglect of the need for human perseverance in relation to
God (see Berkouwer 1958). As a result, human indifference toward divine grace can result, owing
to neglect of its expected rigors. This lesson about grace in relationship has been largely neglected
in theology and the philosophy of religion, to the detriment of our understanding the evidence
for God’s reality.
Divine self-manifestation as evidence for God’s reality, in keeping with divine grace in relation-
ship, would not be an unqualified gift, but instead would be an entrusted gift, or an entrustment, to
humans. Owing to a divine redemptive purpose, such evidence would come and go in a manner
sensitive to human receptivity as cooperation. As perfectly good, God would not be complicit in
human abuse of this evidence. Otherwise, God’s moral standing as worthy of worship, and thus per-
fectly caring for the good of humans, would be at risk. So God would be discerning about the
opportunities for divine self-manifestation to humans and would not be promiscuous in presenting
such evidence of the divine reality and presence. Nothing good would be gained by evidential pro-
miscuity in this area, and real harm could threaten if humans abuse, neglect, or trivialize the evi-
dence they need for their flourishing. So we cannot separate the subject of foundational evidence
for God’s reality from morally relevant considerations about its potential recipients, particularly their
morally relevant attitudes toward it and its value.
Divine self-manifestation that yields static and irrevocable evidence for God’s reality would not
accommodate God’s redemptive character or purposes toward humans. God thus would be selective
at times in self-manifestation to humans for redemptive purposes. For instance, in cases where some
people are not ready to handle divine self-manifestation in a cooperative manner, God would with-
hold self-manifestation for their benefit. This would be a kind of redemptive divine hiding, and it
would falsify the view that occurrent evidence for God’s reality is uniformly possessed by all people
or always possessed by all people. Any theory of evidence that excludes such divine hiding would
threaten God’s morally perfect character and thus be inadequate to that extent. (For discussion,
see Moser 2017, chap. 3.) Some proponents of traditional natural theology offer their arguments
as if they can offer good evidence for all reflective people. This perspective does not fit well with
the portrait of a redemptive God who hides at times for the good of some people. (For relevant mis-
givings, see Moser 2010, chap. 3; Moser 2019b.) We need to look more carefully at some key fea-
tures of divine self-evidence as a foundation for belief in God.
CONVICTIONAL EVIDENCE FOR FAITH IN GOD
A God worthy of worship would be a morally perfect agent and hence an intentional agent with
morally perfect goals. One of these intended goals would be to have the divine moral character
cooperatively welcomed and reflected in human lives, for the moral good of humans. This would
include an aim for imitatio Dei, the intentional imitation of God’s moral character by humans, in
attitude and in action. As responsible moral agents, humans would need to cooperate with God
in the process of their moral maturation, but only on the basis of supporting evidence. Otherwise,
their needed moral development would rest on something no more stable than wishful thinking.
What, however, would this evidence look like in its support for a goal of imitatio Dei?
If God, as perfectly good, desires to be Lord over human moral development, then God would
seek to lead or guide humans in imitatio Dei, with their cooperation. This would be central to being
divine Lord in this domain, given that a person’s Lord is one who leads or guides that person. The
absence of this moral goal would be a defect in a candidate for a God worthy of worship, because it
would entail a moral shortcoming. It is historically noteworthy that the apostle Paul’s conception of
God offered in the Christian New Testament agrees with this lesson. For instance, Paul links being
an adopted “child of God” with being led by God: “All who are led by the Spirit of God are chil-
dren of God” (Romans 8:14, New Revised Standard Version [NRSV]). Paul also identifies a central
feature of God’s intervention in human experience as follows: “Hope [in God] does not disappoint
us, because God’s love [agapē] has been poured into our hearts through the Holy Spirit that has

been given to us” (Romans 5:5, NRSV). Paul would say the same of faith in God, regarding the
basis of its not being (cognitively) disappointing.
Paul is offering at least a denial of cognitive or epistemic disappointment concerning faith in
God. In particular, he has in mind a distinctive kind of evidence offered by divine self-manifesta-
tion: in particular, the manifestation of the righteous agapē integral to God’s perfect moral character.
The self-presentation of such agapē by God to a person, perhaps in conscience, would show God’s
moral character in a way that offers unique evidence of God’s reality and character to that person.
The relevant idea of divine self-presentation emerges in an announcement Paul attributes to God
in Romans 10:20 (NRSV; citing Isaiah 65:1): “I have shown myself to those who did not ask for
me” (cf. John 14:21, 23). This would be a de re self-manifestation to humans, and it would be akin
to a case where a human shows his or her (character of) unselfish care to another person, perhaps
without uttering a word.
Paul is not the only New Testament writer to identify a role for divine guidance in human
experience. The author of John’s Gospel identifies such a role in connection with an experience
of human conviction by God. For instance, an experience of being convicted by God figures in
John 16:8: “When [the Spirit of God] comes, he will convict [elenchei] the world concerning
sin and righteousness and judgment” (RSV, using “convict” as in its alternative translation; cf.
John 16:13). It also figures in Revelation 3:19: “As many as I love, I convict [elenchō] and
instruct [paideuō]” (my translation). The conviction in question would be morally significant,
because it would nudge one, perhaps via conscience, toward imitating and cooperating with
God’s perfect moral character. The nudging, however, would not be coercive of one’s will. It
would allow for responsible human cooperation and thus for human opposition and rejection.
The writer of John’s Gospel (3:20) suggests that many people seek to avoid being convicted of
their wrongdoing by God. This would be a matter of their decision; it would not be caused by
God. So it would not include the divine suppression of human agency in a human response to
a divine challenge and offer.
David Forrest (1906, 107) has characterized the role of divine intervention in human experi-
ence as follows:
God verifies Himself not as an idea, but as a power “to kindle or restrain” in every impulse,
resolve, and aspiration. We are sure of Him, because of what He is to us, because of the place
He has in our life, ruling, rebuking, uplifting. It was this inward and indisputable knowledge
which Christ Himself had and which He strove to realise in others.
The “ruling, rebuking, uplifting” mentioned by Forrest would be part of the intended divine leading
or guiding of humans toward moral improvement, including toward a morally improved relation-
ship with God and other humans. Similarly, the relevant talk of being “convicted” by God connotes
intentional challenge by God rather than a nonpersonal causal process. This challenge would have
an intended goal or purpose that includes imitatio Dei, and humans would have to resolve to coop-
erate with it rather than to ignore or reject it.
An intentional feature of being convicted by God could emerge in human experience, as the
goal-directedness of being convicted emerges in conjunction with human cooperation with it. Aim-
ing for the moral good of humans, God would seek the perfecting of divine agapē in humans over
time, and this would call for human cooperation with divine conviction toward that end. Given
such cooperation, human experience would undergo an increase in loving others, even enemies, in
depth and in scope. This increase would stem from one’s increasingly being convicted by God
toward imitating and cooperating with God’s perfect moral goodness, including the love of one’s
enemies.
One’s increase in loving others, while one is increasingly convicted, would support improve-
ment in one’s reconciliation and fellowship with God. Because it is intentional, it also would mani-
fest the role of a morally perfect intentional agent in one’s moral improvement and thus indicate
that this process is not just physical or nonpersonal. Including love of one’s enemies, the process
would be at odds with a skeptical claim that it stems just from one’s ordinary desires. In fact, the
process would go against natural human desires. Divine agapē would exceed typical humans in a
way that requires them to be led into it by being convicted toward it by a morally perfect agent who
has it inherently. A perfectly good God would want humans to be cooperatively led, while being
convicted, toward a morally improved life in reconciliation and fellowship with God and other
humans, even their enemies.
One could be led by God, via being convicted in conscience, while being an atheist or an
agnostic regarding God’s existence. One’s being led de re by God would not require one’s accept-
ing or even formulating correct de dicto content regarding God’s reality. One might use just min-
imal demonstratives regarding what convicts one in conscience, while lacking a notion of God as
an intentional agent worthy of worship. This kind of case would not preclude one’s being con-
victed and led by God de re, even though one’s understanding of the influence on one’s being
led, and of its purpose, would be deficient. In any case, one’s being thus led does not require
one’s adequately understanding the process of one’s being led. Similarly, it does not require one’s
appreciating or recognizing one’s evidence that the process is intentional in being indicative of an
intentional agent behind the convicting and leading. This lesson is important, because it would
enable God to convict and lead people who come from a wide range of intellectual and religious
perspectives, including atheists and agnostics. Such doxastically diverse people could have some
experiences of God that are qualitatively the same. They would explain them differently, even
though their quality is common to them. (Explaining experiences differently, however, does not
entail doing so equally well.) Intellectual and religious diversity, then, would not preclude divine
guidance for humans.
The divine redemption of humans, as suggested, would aim for something more profound and
morally significant than intellectual acknowledgment of God’s reality. It would aim at least for the
cooperation of humans with God’s perfect moral character and will, even if they lack intellectual
acknowledgment of God’s reality. A process of being cooperatively convicted and led by God, if
de re, would contribute to that aim. It also would allow for considerable diversity and even disagree-
ment in human understanding of the redemptive process. This is important, because social influ-
ences on humans will often result in such diversity and disagreement, even when humans have
the best of intentions, including moral intentions. A perfectly good God would value good moral
intentions over correct intellectual content regarding God. We should expect the evidence of God’s
reality to follow suit, and thus to focus on a key role for God’s moral character and will, even when
humans lack an adequate de dicto understanding regarding God’s purpose and intervention in
human lives.
De re faith in God would include a favorable commitment with trust toward divine convicting
and leading, perhaps in one’s conscience. This would encompass a sympathetic commitment toward
cooperating with the divine convicting and leading in one’s experience. De re faith in God thus
would be irreducible to factual knowledge that God exists, because its commitment with trust
would exceed such knowledge. Its foundational evidential support would come from a self-manifes-
tation of divine moral character in one’s experience, and that manifestation would offer moral con-
viction and guidance to one, even when one has not categorized it as divine. Such faith, then, would
have an evidential basis that is cognitively prior to faith itself. The evidential basis in question could
have a causal influence on some human beliefs, but it would not be merely causal, given that it
would confer epistemic status on a commitment. In addition, the relevant evidence need not be
public, or socially shared, but could be private to a person. Similarly, it need not be reproducible
by a human, given that, when veridical, its source would be God, and it would not be controllable
by a human. God would be self-evidencing in presenting a living moral-character manifestation to
humans, and this could serve as foundational evidence for them.
We now see the epistemic significance of one’s conception of God, particularly of the moral
features it ascribes to God. Those features figure centrally in the self-evidence that God would give
to cooperative humans, and that evidence would be foundational for belief that God exists and for
belief in God. God’s moral character thus would be epistemically relevant for such belief. The rele-
vant conception of a morally perfect God would exceed a conception of God in mere theism
or deism, and, in doing so, would supply explanatory and evidential potential unknown to mere
theism or deism.

SKEPTICISM AND DEFEATERS
Any claim to epistemically justified belief in God will elicit challenges from skeptics of various
stripes. Skeptics can introduce a specific conception of epistemic justification that makes particular
demands on such justification and thereby blocks justified belief in God. For instance, if a concep-
tion of epistemic justification requires that justified beliefs get their semantic meaning from sensory
experience (as various proponents of Humean skepticism have suggested; see Watkins 1984), then
justified belief in God will be ruled out, owing to supposed defects in the very idea of God. Simi-
larly, if a conception of epistemic justification requires logical entailment for an (evidential) episte-
mic support relation, justified belief in God will be excluded, given the absence of such a demand-
ing support relation. A serious problem arises, however, for the skeptics in question. Nonskeptics are
not required by reason or evidence to accept the stringent conceptions of epistemic justification
offered by those skeptics. So far as reason and evidence go, they can proceed with alternative con-
ceptions of epistemic justification, and some of those conceptions do not rule out at the start justi-
fied belief in God.
Skeptics need not make their case on the basis of a disputed conception of epistemic justifica-
tion. Instead, they can offer (potential) defeaters of justification for belief in God, while working
within a widely accepted notion of epistemic justification. The defeaters on offer would undermine
justification for one’s belief in God when conjoined with one’s supporting evidence. Defeaters, how-
ever, are not mere beliefs (contrary to Plantinga 2000, 485). If they were, one could create defeaters
just by forming beliefs, but that would make defeat too easy in allowing for the defeat of any evi-
dence one prefers to defeat. Evidence, of course, does not function in that easy way.
We can, and should, ask whether faith that God exists is correct in its interpretation of human
experience. Correctness would entail fact-based truth and thus factuality, thereby going beyond jus-
tified belief about reality to an actual feature of reality. So, a correct interpretation must match the
relevant facts. Departure from such a realist approach to truth is philosophically self-refuting (see
Moser 1989, chap. 1), and nobody has explained how we can proceed coherently without such an
approach. In particular, faith that God exists cannot be properly understood apart from the question
of whether it is true or factual that God exists.
A compelling approach, from an epistemic point of view, to the question of whether faith that
God exists is true fits with a broader approach to epistemic reasonableness, the kind of reasonable-
ness suited to factual knowledge. This approach relies on abduction, an inference to a best available
explanation of the relevant evidence for a person, as a central factor in epistemically reasonable belief
(for details, see Moser 1989, 2008). A person need not always draw an inference to a best available
explanation to have an epistemically reasonable belief. Instead, the epistemic reasonableness of a
belief for a person can depend on there being available to that person an inference to a best available
explanation from the person’s evidence to the belief in question.
Foundational evidence for a person will consist of that person’s base of experience, and the
nonfoundational evidence for a person will consist of that person’s beliefs made epistemically rea-
sonable (for that person) on the basis of the foundational evidence. An epistemic connection
between the two tiers can be an available inference to a best available explanation. One’s belief that
God exists will be fully grounded when it is suitably based on the relevant foundational evidence
and on the epistemic connection between that evidence and the belief. The proposition that God
exists, however, could be epistemically supported for a person on the basis of experience, even
though that person does not base a belief that God exists on the supporting evidence or epistemic
connection.
An abductive approach to epistemic reasonableness relies on explanation-seeking why-questions
about foundational evidence in a person’s experience. A typical why-question is: Why am I now
having this experience, rather than no experience at all or some other experience? This question will
prompt better and worse answers relative to a person’s overall experience and evidence. My current
experience of apparently being morally convicted, by an intentional agent, in conscience of selfish-
ness invites the question of why I am now having this experience rather than no experience at all
or some other experience. A bad answer, from an abductive point of view, would be that I am
now eating raw onions. The latter answer would not suitably explain why I am having the experi-
ence in question. That answer’s content does not suitably elucidate my experience of being morally
convicted in conscience.
Some skeptics will suggest that my present experience of being morally convicted in conscience
could be just a dream or an illusion, and, on this basis, recommend that I withhold judgment on
the matter. This recommendation, however, would be premature. If “could” indicates logical possi-
bility, we have no reason to deny the possibility, because neither epistemic reasonableness nor
knowledge rules out the logical possibility of being mistaken. Knowledge, given its truth condition,
rules out the actuality of being mistaken, but that leaves intact the logical possibility of being mis-
taken. In contrast, epistemic reasonableness does not require truth, because there can be epistemi-
cally reasonable belief that is false. So skeptics using a modal notion of “possible mistake” would
need to clarify the relevant domain of modality and then explain its actual pertinence to my actual
evidence of moral conviction.
My experiential evidence of being morally convicted in conscience does not give me any indi-
cation that I am now undergoing a dream or an illusion. I have had numerous dreams and illusory
experiences, and, so far as my actual evidence goes, they always have been accompanied by indica-
tors of being dreams or illusory experiences. So I do not find myself wondering now, in genuine
doubt, whether any of my actual experiences is, or has been, a dream or an illusion. I am able to
tell, so far as my actual evidence goes, which experiences of mine were dreams or illusions, because
my dreams and illusions, given my evidence, have had telltale indicators. Some of the familiar indi-
cators are inconstancy, abruptness, indistinctness, and incongruity. I, among others, have been well
aware of such indicators in connection with (what I know to be) dreams and illusions.
My present experience of being morally convicted in conscience of selfishness does not include
any of the telltale indicators of a dream or an illusion. So my present experience does not include
any evidence (or indicator) of its being a dream or an illusion. I logically could be undergoing a
dream or an illusion, but that modal truth is not to the epistemic point now. Epistemic reasonable-
ness for a person is determined by that person’s actual truth indicators, and I do not have a truth
indicator in my evidence of a dream or an illusion in the present case. The burden is on skeptics
to identify an actual indicator of a dream or an illusion, and a mere modal claim will not discharge
this burden. Without an actual truth indicator, skeptics will not be able to discharge the burden,
and their skeptical challenge then will be neutralized.
We now can identify an important lesson about foundational evidence for faith that a worship-
worthy God of perfect agapē exists. When I confront a certain kind of agapē toward me that is chal-
lenging me in my experience, I should ask (at least from an epistemic point of view): Why am I hav-
ing this agapē experience rather than no experience or a different experience? A responsible answer
calls for close attention to my actual experience, and I may need to put myself in a suitable position
to get such an answer. I may need, for instance, to curb any bias I have against the value of unselfish
agapē, including such agapē directed toward me. I also may need to be willing to cooperate with
such agapē in order to comply with what God intends it to accomplish.
A perfectly good God would intend divine agapē to launch or to sustain a relationship of fellow-
ship or communion between God and a human that is in the best interest of that human. If a person
is opposed or indifferent to divine agapē, God may hide from that person to avoid a kind of rejection
that hardens opposition or indifference to God. If God seeks human cooperation with divine agapē,
God’s self-manifestation of such agapē to a person would not aim for human responses that encourage
further alienation from God. (On the relevant kind of communion, see Moser 2017, chap. 5.)
We should not assume that all humans would experience divine agapē in exactly the same way.
Instead, we should allow for some variability, given distinctiveness in persons and their circum-
stances. Even so, an important issue will be whether one’s experience of agapē includes an experi-
ence of personal traits suited to a God worthy of worship. Such traits would include the love’s being
intentional, directed, and nudging without coercion toward a person’s imitating and cooperating
with it. A key issue will be whether a superhuman agent manifests such intentionality, directedness,
and nudging in any human experience. This amounts to the question of whether the manifestation
of unselfish agapē in human experience is a personal disclosure that transcends a human source.

My experience of being presented with and guided toward divine agapē could be diachronic
rather than just synchronic. In other words, it could take place over time, and not just at a moment.
So I could consider it over time regarding its source, without making a judgment about just a
moment of experience. My experience over time could include my conscience being challenged,
and perhaps even convicted, by the standard of experienced agapē over time, including my experi-
enced failure to conform to that standard. In experiencing an attempt to lead me away from my self-
ishness, I could experience my falling short of the agapē directed toward me. This could create dis-
sonance in my experience, including in my conscience. It also could challenge my moral self-
understanding and thus create a moral crisis in my life. (For relevant discussion, see Farmer 1935,
chaps. 1, 11, and Niebuhr 1941, vol. 1, chap, 9.)
My experience of agapē could indicate my being intentionally challenged to replace my self-
ishness with unselfish love toward others, including my enemies. The apparently intentional chal-
lenge could oppose my own preferences and intentions, and it could come to me via my con-
science without influence from another human. So I could lack evidence that the agapē
presented to me, including its apparently intentional challenge to me, comes from me or another
human. In that situation, nothing in my experience would defeat my evidence that my experience
of the challenge of agapē to me arises from a personal source independent of me and other
humans. Given my actual experience, I seem to be intentionally led, via conviction in conscience
without coercion, toward unselfish love for others and away from my selfishness. This happens to
be true of my own experience and conscience, and it indicates the involvement of a personal agent
other than a human agent.
Given my overall evidence, I could reasonably decide that an intervention from a perfectly good
God figures in a best available explanation (for me) of the challenging and guiding agapē that trou-
bles my conscience. I could (and, I submit, do) lack the availability of a better or even an equally
good alternative explanation of this agapē presented to me. A skeptic could propose, nonetheless,
a purely natural explanation as a defeater of my evidence for a theological explanation of my experi-
ence. A skeptic might propose, for instance, that I myself am creating my experience of the challeng-
ing and guiding agapē in question. We can grant the logical possibility of my creating this, even
without our specifying a mechanism, but actual evidence, as noted, requires an actual truth indica-
tor, beyond mere possibility. So a skeptic needs somehow to base the critical proposal in my actual
evidence; otherwise, the proposal will not get epistemic support, at least for me. Without a basis in
my evidence, the skeptical proposal of self-creation will not figure in a best available explanation for
me, relative to my actual evidence. It thus will have an inadequate epistemic status for me, given my
evidence. The same holds for any other proposed merely natural explanation. The alleged natural
explainer will need to figure in my evidence to gain epistemic standing. My own relevant experi-
ence, as it happens, does not underwrite such an explainer.
Some skeptics will ask how the thesis of a morally perfect God can be established from an exter-
nal perspective that is neutral relative to theistic positions. They will be unmoved by an attempt
from an internal perspective that favors a theistic position. An adequate response to such skeptics
would begin with a clarification of what an external position includes, and it would distinguish
between one’s having adequate evidence for a position and one’s establishing that position. We also
should distinguish between being external to beliefs of some sort and being external to evidence of
some sort. The position on offer does not face a problem if we have in mind being external to the-
istic beliefs that entail God’s existence. We have not built such beliefs into the relevant evidence
from experience that supports belief in God. So we have not imposed an “internal” perspective on
this evidence as a starting point. On the contrary, we have taken exception to reformed epistemol-
ogy for endorsing such a defective starting point.
My experience of the challenging and guiding agapē presented to me would benefit, epistemi-
cally, from evidentially grounded reports from others. Such reports could come from well-grounded
parts of a written religious tradition, such as parts of the Bible, and from well-grounded testimony
from my contemporaries. The explanatory coherence of such well-grounded reports, relative to my
overall experience, can add to the positive epistemic status of the best available explanation for
me. In contrast, mere reports or testimonies, like mere beliefs, would not add to positive epistemic
status. They are epistemically arbitrary in a way that well-grounded reports are not. Evidential sup-
port is the key difference here, and such support can vary among people. Any testimonial report,
however, would need ultimate support in foundational evidence.
Defeaters as underminers of epistemically reasonable belief for one can come from any relevant
aspect or area of one’s experience. Similarly, increased explanatory coherence for a belief relative to
any aspect or area of one’s experience can increase the positive (defeasible) epistemic status of that
belief for one. Epistemic abductivism is thus evidentially holistic for a person, with regard to posi-
tive evidence and defeaters. The ultimate epistemic status of evidence for a person depends on all
of that person’s evidence and potential defeaters. Epistemic reasonableness, as suggested, is relative
to persons and their evidence in this way, and thus differs from truth and factuality. It does not
entail or otherwise guarantee truth, but it is the evidentially responsible, nonarbitrary way to aim
at acquiring truth and avoiding error. In particular, it is responsible to one’s overall indicators (or
evidence) of what is true or factual, even if those indicators are fallible and defeasible.
Epistemic abductivism accommodates the aforementioned truth that basic, foundational evi-
dence in experience is not an argument or even a belief. The fact that the explanatory value of a
belief relative to one’s experience confers epistemic status on that belief allows that the basic evi-
dence in experience is free of an argument or a belief. So evidence from divine agapē in experience
need not include an argument or a belief. It can anchor the epistemic reasonableness of a belief that
serves in the best available explanation of that experiential evidence for a person. So one’s ultimate
evidence for God’s existence need not be an argument or a belief.
De dicto faith can offer an interpretation of basic experiential evidence (for instance, by speci-
fying what it is), and that interpretation can figure crucially in a best available explanation of one’s
experience. When it does figure thus, in the absence of defeaters, such faith can be well-grounded
and hence epistemically reasonable for a person who has the relevant experience. Even when offer-
ing an interpretation and a best available explanation of experience, however, de dicto faith in God
will allow for incompleteness, perplexity, and mystery in explanation. Indeed, we should expect
these when the object of faith is a God who cannot be fully comprehended by humans. After
all, the God in question would transcend the created order and any understanding its members
enjoy.
Finally, regarding evil as a potential defeater, we should not expect to have a theodicy that fully
explains God’s purposes in allowing evil. For such a full theodicy, God would need to reveal divine
purposes in a way that bears on all cases of allowed evil, but our evidence does not indicate such
revealing, and we should not expect it to do so. The absence of a full theodicy, in any case, does
not entail a defeater of the kind of foundational evidence offered here (see Moser 2018). Even so,
the human limitations we face could serve a divine redemptive purpose, because they could encour-
age a felt need for dependence on God (rather than just on a theory), thereby countering human
pride and presumed self-sufficiency (see Niebuhr 1949).
A LESSON
Epistemic reasonableness, unlike truth, is person-variable, owing to variation in human experience

that lies at the foundation of evidence. This consideration applies directly to the epistemic reason-
ableness of belief in God and belief that God exists. As a result, it is possible that belief in God is
evidentially grounded for one person but not for another, given a difference in their overall experi-
ences. This is no defect in belief in God; it is a consequence of the nature of evidence and epistemic
reasonableness. This lesson does not challenge the integrity of evidentially grounded belief in God.
Instead, it highlights that such belief owes its epistemic status to the actual evidence and grounding
experience a person has. Actual experience thus matters, critically, in this domain, as in matters of
evidence generally.
Acknowledgments. Thanks for helpful comments to Joseph Koterski and his editorial board for this
book.

Bibliography
Barth, Karl. “The Principles of Dogmatics according to Wilhelm Moser, Paul K. “Theodicy, Christology, and Divine Hiding: Neutralizing
Herrmann.” (1925.) In Theology and Church: Shorter Writings 1920– the Problem of Evil.” Expository Times 129, no. 5 (2018): 191–200.
1928, by Karl Barth, translated by Louise Pettibone Smith, 238–271.
Moser, Paul K. “Pascal’s Wager and the Ethics for Inquiry about God.” In
London: SCM Press, 1962.
Pascal’s Wager, edited by Paul Bartha and Lawrence Pasternack, 64–83.
Berkouwer, G. C. Faith and Perseverance. Translated by Robert D. Knudsen. Cambridge, UK: Cambridge University Press, 2019a.
Grand Rapids, MI: Eerdmans, 1958.
Moser, Paul K. “Natural Theology: A Deflationary Approach.” In Natural
Buber, Martin. I and Thou. Translated by Ronald Gregor Smith. New York: Theology: Five Views, edited by James K. Dew and R. P. Campbell.
Charles Scribner’s Sons, 1958. First published in 1923. Grand Rapids, MI: Baker Academic, forthcoming 2019b.
Clifford, William Kingdon. “The Ethics of Belief.” In Lectures and Essays, by Niebuhr, Reinhold. The Nature and Destiny of Man. New York: Charles
William Kingdon Clifford, 339–363. London: Macmillan, 1877. Scribner’s Sons, 1941.
Dawkins, Richard. The God Delusion. Boston: Houghton Mifflin, 2006. Niebuhr, Reinhold. Faith and History. New York: Charles Scribner’s Sons,
1949.
Farmer, Herbert H. The World and God. London: Nisbet, 1935.
Plantinga, Alvin. “Reason and Belief in God.” In Faith and Rationality,
Farmer, Herbert H. “The I–Thou Relationship.” In The Servant of the Word,
edited by Alvin Plantinga and Nicholas Wolterstorff, 16–93. Notre
by Herbert H. Farmer, 35–55. New York: Charles Scribner’s Sons,
Dame, IN: University of Notre Dame Press, 1983.
1942.
Plantinga, Alvin. Warranted Christian Belief. New York: Oxford University
Forrest, David W. The Authority of Christ. Edinburgh: T&T Clark, 1906.
Press, 2000.
Gettier, Edmund. “Is Justified Truth Belief Knowledge?” Analysis 23, no. 6
Plantinga, Alvin. “Arguing for God.” (1993.) In Readings in the Philosophy of
(1963): 121–123. Reprinted in Empirical Knowledge, 2nd ed., edited by
Religion. 3rd ed., edited by Kelly James Clark, 96–106. Peterborough,
Paul K. Moser. Lanham, MD: Rowman and Littlefield, 1996.
ON: Broadview Press, 2017.
Herrmann, Wilhelm. The Communion of the Christian with God. Translated
Robinson, N. H. G. Christ and Conscience. London: Nisbet, 1956.
by J. Sandys Stanyon and R. W. Stewart; edited by Robert T. Voelkel.
Philadelphia: Fortress Press, 1971. First published 1903. Russell, Bertrand. “Knowledge by Acquaintance and Knowledge by
Description.” Proceedings of the Aristotelian Society 11, no. 1 (1911):
Meadors, Edward P. Idolatry and the Hardening of the Heart. London: T&T
108–128.
Clark, 2006.
Shope, Robert K. The Analysis of Knowing. Princeton, NJ: Princeton
Moser, Paul K. Knowledge and Evidence. Cambridge, UK: Cambridge
Shope, Robert K. “Conditions and Analyses of Knowing.” In Oxford
Moser, Paul K. Philosophy after Objectivity. New York: Oxford University
Handbook of Epistemology, edited by Paul K. Moser, 25–70. New York:
Press, 1993.
Moser, Paul K. The Elusive God: Reorienting Religious Epistemology.
Watkins, John W. N. Science and Scepticism. Princeton, NJ: Princeton
Cambridge, UK: Cambridge University Press, 2008.
Moser, Paul K. The Evidence for God: Religious Knowledge Reexamined.
Wolterstorff, Nicholas. “Can Belief in God Be Rational if It Has No
Foundations?” In Faith and Rationality, edited by Alvin Plantinga and
Moser, Paul K. The God Relationship: The Ethics for Inquiry about the Divine. Nicholas Wolterstorff, 135–186. Notre Dame, IN: University of Notre
Cambridge, UK: Cambridge University Press, 2017. Dame Press, 1983.
TOPIC 4
Doxastic Foundations: Atheism

IS THEISM RATIONAL? 119
Ali Hasan
Associate Professor
University of Iowa
DOES THETESTIMONY OF SACRED SCRIPTURES FROM THE RELIGIONS 127

OF THE WORLD FAVOR (MONO)THEISM OR ATHEISM?
Richard Carrier
Educator (PhD)
The Secular Academy, Washington, DC
This chapter centers around the question of whether theism is rational. We begin by discussing different
theories of rationality, and introducing some importantly related epistemic concepts and controversies. We
then consider the possible sources of rational belief in God and argue that even if these provide some positive
support, the fact of religious disagreement defeats the rationality of theism.
IS THEISM RATIONAL?
Ali Hasan, University of Iowa
Questions regarding the existence and nature of God (or divine reality) figure prominently in the
philosophy of religion. Corresponding questions arise in the epistemology of religion regarding the
epistemic status of beliefs about God. The two sorts of questions are certainly related, but our con-
cern here is more directly with the latter, epistemological questions. Is belief in theism a case of
knowledge? Does it at least qualify as rational or justified? I argue that it is not rational to believe
in God. I am not, however, going to argue that it is rational to disbelieve in God or accept atheism.
Perhaps it is, or perhaps agnosticism—withholding or suspending belief about the existence of
God—is the only rational attitude to hold regarding God’s existence. I will not try to settle which
of these nontheistic positions is rational.
The main argument is that any justification we might have for theism is defeated by evidence
available to us, especially evidence of the unreliability of religious belief-forming practices. This
defeats the rationality of not only traditional monotheism but any specific religious doctrine or spe-
cific belief in supernatural or divine reality. (The problem of evil might provide additional evidence
against traditional monotheism but is not discussed in this chapter.)
KNOWLEDGE AND RATIONAL BELIEF

Do we have knowledge of the existence of God? I can know that p only if p is true and I believe that
p. If p is not true, then, while I might think I know that p, I would be wrong. And if I don’t believe
that p, it can’t be knowledge for me, even if I could, perhaps quite easily, come to believe and know
it. But true belief is not sufficient for knowledge. Intuitively, knowledge requires in addition that
119
my belief have some further positive property—that it be in

KEY CONCEPTS some sense “justified,” “rational,” or “warranted” (and perhaps
even this is not sufficient; see Gettier 1963). Many epistemolo-
gists focus on questions about the nature and extent of our jus-
Access Internalism tification or rationality rather than on the question of knowl-
Agnosticism edge, because (a) assessments of rationality or justification are
Belief central to determining whether we have knowledge, and (b)
even if we fail to know certain things (perhaps because it turns
Coherence Theory
out that knowledge is very hard to get), it remains interesting to
Defeaters ask whether we are at least justified or rational in believing
Disagreement them. I will similarly focus on questions regarding the justifica-
tion or rationality of belief in God, although if I am right that
Epistemic Rationality
belief in God is not rational, it will follow that it does not
Epistemology count as knowledge.
Evidence I am not concerned here with whether it might be pruden-
Evidentialism tially rational or morally rational to believe in God—roughly,
whether it is rational to adopt the belief in God because this
Externalism would be an effective means to my own well-being or the
Foundationalism well-being of others or because it would likely improve my
Internalism moral character or help me discharge some of my moral obliga-
tions. We are concerned with epistemic rationality, and
Justification
although the distinction is difficult to make precise, for our
Knowledge purposes it will do to say that epistemic rationality of a belief
Reformed Epistemology is related in some intimate way to the truth (or probable truth)
of the proposition believed, whereas prudential and moral ratio-
Skepticism
nality are not.
Uncertainty
Warrant THE STRUCTURE AND SOURCES OF RATIONALITY
Epistemologists have defended many different views regard-
ing the nature, structure, fundamental sources, and extent
of knowledge and rational belief. We cannot discuss all these
issues here, but we should briefly introduce two central debates in contemporary epistemology
that will help us with the epistemology of religious belief: the foundationalist-coherentist debate
and the internalist-externalist debate.
Foundationalism is often motivated by reflecting on the fact that at least many of our beliefs
depend on some sort of inference from other beliefs we have. How might the supporting beliefs
be justified? Assuming that at least some of our beliefs are justified, there seem to be four options:
1. Every belief depends on (actual or available) inference from other beliefs, and each of those
beliefs on yet other beliefs, ad infinitum.
2. Every belief depends on other beliefs for its justification, and the latter ultimately depend on the
original belief, so that eventually the same belief appears as part of its own justification.
3. The regress terminates with beliefs that are not themselves justified.
4. The regress terminates with beliefs that are justified but that do not depend on any other beliefs
for their justification.
Foundationalists hold that only the last of these options is viable. According to foundational-
ism, there are some beliefs—let’s call them basic or foundationally justified beliefs—that are justified
but that do not depend for their justification on any other beliefs; and any nonbasic, justified beliefs
ultimately depend for their justification on inferential relations to other justified beliefs, and ulti-
mately on a “foundation” of basic beliefs. Option 1, infinitism, according to which the regress of
reasons never stops, has hardly any proponents. It is doubtful that finite beings like us could have
infinitely many distinct reasons for belief. But even granting that we could, infinitism has the
counterintuitive consequence that any belief whatsoever could be justified simply by appeal to
another belief that entails it, and another belief that entails the latter belief, and so on. Option 2
seems to be an endorsement of circular reasoning. It is surely no good to defend a belief by relying,
even in part, on that belief itself! Option 3 isn’t any better. One cannot justify a belief by simply
relying on another belief that is unjustified; justifying beliefs would be much too easy! Foundation-
alism thus seems to be the best option (see Hasan and Fumerton 2016).
Coherentists might claim that the above argument for foundationalism assumes a rather linear,
one-directional conception of relations of epistemic dependence. According to coherentism, every
belief depends for its justification on how well it “coheres” with the rest of one’s beliefs, where
coherence is determined by the web of logical, probabilistic, or explanatory relations between one’s
beliefs. The foundationalist-coherentist controversy remains. Many foundationalists complain that
mere coherence of our beliefs does not seem to give us any reason to think that they are likely to
be true and that coherentism fails to do justice to the role of experience in justification. Many
coherentists worry that the foundationalist accounts of basic belief fail to make sense of how such
beliefs could be justified without depending on other (doxastic) reasons.
Insofar as the regress argument assumes that at least some beliefs are justified, it may be best to
take the argument to be that if any beliefs are justified, then they must either be basic or ultimately
depend for their justification on inferential relations to basic beliefs—not as an argument for the
claim that there are any basic beliefs. Foundationalists sometimes defend the existence of basic
beliefs by appeal to examples. My belief that I have a headache, or that I feel thirsty and warm, is
not based on any belief but simply on an awareness of the experiences themselves. I don’t seem to
hold these beliefs on the basis of other beliefs. My belief that there are no round squares, or that
circles don’t have corners, is not based on other beliefs, but on something like a clear understanding
of the concepts that figure in the belief or a clear grasp of the nature of the properties the belief is
about.
These foundational beliefs seem to be based on introspection and reason respectively. Many
contemporary foundationalists expand the sources of foundational justification to include sensory
perception and memory. A very permissive sort of foundationalism, sometimes called phenomenal
conservatism, holds that the mere fact that it appears or seems to me that p would provide me with
some reason or justification to believe that p.
Although most contemporary epistemologists are foundationalists of some sort or other, virtu-
ally all epistemologists accept something roughly like a coherence constraint on justification, a “no
defeater” condition. In addition to having positive, inferential or non-inferential justification for
some belief that p, the subject must (a) not have good reasons to think that ∼p (sometimes called
a “rebutting defeater”) and (b) must not have good reasons to think that one’s belief was formed
in an unreliable way or on the basis of an unreliable source (sometimes called an “undercutting”
or “undermining” defeater). Suppose I believe that there are fish in the neighborhood pond, and
suppose I have a good reason or justification to think this (e.g., an honest neighbor’s testimony).
If I also have reason to think that there aren’t any fish in the pond (e.g., another honest neighbor’s
conflicting testimony), or a reason to think my evidence is misleading or my source is unreliable
(e.g., knowing that my neighbor’s testimony is unreliable), then the belief is not rational. Thus,
the satisfaction of positive conditions for non-inferential and inferential justification is not sufficient
for justified belief; such a belief is “defeasibly” or “prima facie” justified, where this just means that
it is justified in the absence of defeaters.
In addition to the foundationalist-coherentist debate, there is the internalist-externalist debate.
There are in fact multiple debates that fall under the latter label (see Conee and Feldman 2001),
but we will focus on the debate between access internalists and access externalists (henceforth simply
“internalism” and “externalism”). This debate can be understood as a disagreement regarding the
appropriate way to understand the relation between epistemic justification and truth. Internalists
take epistemic justification for believing that p to require having access to good reasons to think that
p is (probably) true. Externalists deny this. Externalists require that there be some “external” con-
nection to the truth or probability of the belief. For example, according to one of the most popular
forms of externalism, reliabilism, epistemic justification requires that one’s belief-forming methods

or processes be reliable—roughly, that they yield true beliefs more often than not. Externalists usu-
ally also require the absence of defeaters: having reasons to think p true is not required for justified
belief that p, but not having reasons to think the belief false or untrustworthy is.
Internalism is initially plausible and very intuitive. One way to motivate it is by appeal to hypo-
thetical cases like Keith Lehrer’s (1990) case of Truetemp, who, without knowing it, has a chip
implanted in his head that produces very precise and highly accurate beliefs about the ambient tem-
perature. Truetemp often gets temperature beliefs from the operation of the chip, but he has never
checked their correctness and has no reasons for, or against, his having such an ability. Truetemp’s
belief that the temperature is now exactly 47 degrees is produced by a highly reliable process, in the
absence of defeaters. But intuitively, it is not justified. Reflection on this and similar examples sug-
gests that internalism is right about what is missing: although these beliefs are objectively probable,
highly reliable, track the truth, or satisfy some such external requirement, there is nothing within
the subject’s perspective that indicates this, no reason for the subject to take the belief to be true.
Externalists themselves often admit that internalism is intuitive. But they worry that the intern-
alist faces a dilemma. If the internalist accepts that there must be a strong connection between epi-
stemic justification and truth (or probable truth), then the demand of access to that connection is
too high, with the consequence that few if any of our beliefs are justified. If, however, the internalist
says that no such connection is required, or only a weak one, then epistemic justification is not tied
to the truth in any significant, objective way.
These debates are complicated. There are proponents of internalism and externalism, founda-
tionalism and coherentism, on opposing sides of the theism/atheism debate, and, as we’ll soon
see, some defend the rationality of theism by relying on a particular epistemological theory. None
of these epistemological views obviously or trivially favors theism, atheism, or agnosticism. How-
ever, when we consider whether there are defeaters for theism, we’ll find a strong case for denying
that theism is rational on any plausible account.
EVIDENCE AND EVIDENTIALISM

The concept of evidence figures prominently in contemporary epistemology and is closely related to
other epistemic concepts. But how, exactly, is evidence related to rationality, and how does it figure
in the foundationalist-coherentist and internalist-externalist debates?
One very neat and intuitive way to connect talk of evidence and rationality is to accept eviden-
tialism, roughly, the view that rationality requires having beliefs that fit the evidence. “Proportion
your beliefs to the total evidence!” is a good evidentialist slogan. Roughly, E is evidence for p if
and only if E raises the probability of p. One’s “total evidence” might involve (a) evidence that is
neutral with respect to p; (b) evidence in favor of p or evidence that raises the probability of p; (c)
evidence in favor of ∼p or evidence that lowers the probability of p (defeaters).
In order for some evidence E to make a difference to the rationality of S ’s belief that p, S
must in some way possess or have access to E. Evidentialism is thus in line with some form of
internalism. However, it is plausible that S must also in some way grasp or understand the rele-
vance of E to the probability or truth of p; being aware of something that has an objective connec-
tion to the truth of p, but where one has no grasp of that connection whatsoever, seems no better
than not being aware of it, and so the mere requirement of access to E seems to not be in the
spirit of (access) internalism.
Many contemporary evidentialists are foundationalists who hold that we can have evidence by
having non-doxastic states like experiences (Conee and Feldman 2004). My belief that I feel thirsty
and warm is not based on any belief but simply on an awareness of the experiences themselves, and
it would be odd to say that I have no evidence that I feel this way.
In the epistemology of religion, and in particular the evidentialist-reformist debate, “evidential-
ism” is sometimes used to stand for the view that, roughly, rational belief in God requires having a
good argument for God’s existence or can be justified only inferentially if at all (e.g., Plantinga
1981). It is important not to confuse this view, which we might call “doxastic evidentialism about
belief in God,” with evidentialism as a general theory of rationality. Some might be evidentialists
(belief must fit the total evidence) but hold that belief in God can be basic by being based on non-
doxastic evidence like a direct perception of God, and some might deny evidentialism (deny that all
rational beliefs require evidence) but think that belief in God does not satisfy conditions for non-
inferential justification and so must, like scientific beliefs, be justified by some argument or infer-
ence.
A complication arises when we ask how much evidence is required for belief or how probable a
proposition must be given one’s total evidence in order for it to be rational to believe the proposition.
A probability of 1 (certainty) is surely too high, slightly above 0.5 is too low, and any particular prob-
abilistic measure in between these values seems arbitrary. A natural move is to appeal to the idea that
belief itself comes in degrees. Perhaps the distinction between believing, disbelieving, and withholding
belief is a useful but rather coarse-grained and potentially misleading simplification; the attitude of
belief can be put on a scale, with different degrees of confidence or “credences” varying from full belief
(an attitude of absolute confidence in the truth of the proposition), full disbelief (an attitude of abso-
lute confidence in the truth of the negation of the proposition), and degrees between these two
extremes. The evidentialist can then claim that one’s credence must fit the evidence. In what follows,
however, we can leave open the question of whether belief comes in degrees or not. Those who deny
that belief comes in degrees can understand talk of the rational degree of confidence to have in p in
terms of the degree of probability that it is rational to believe p to have.
Is it rational to believe that God exists? Some may hold that it is rational to have full belief, or
the highest possible confidence, that God exists. Some might deny that, but hold that it is rational
to have a significantly high degree of confidence, perhaps high enough that it more or less matches
the degree of confidence it is rational for us to have in commonsense beliefs and our best, most sta-
ble, and well-tested scientific theories. More modestly, some might say that it is at least rational to
have significantly greater confidence that God exists than not. I argue against taking theism to be
rational even in this minimal way.
BRACKETING SKEPTICISM
If an argument that belief in God is not rational depends on a particularly demanding theory or
standard of rationality, one with radically skeptical consequences, then the theist is likely to object
that the standard is too demanding or that theism has at least not been shown to be any less rational
than common sense or the best scientific theories. Let us accordingly bracket the issue of skepticism,
as David Hume did in his famous dialogues (2007); let us grant for the sake of argument that radi-
cal skepticism is false in assessing whether belief in God is rational.
Classical or traditional varieties of foundationalism tend to be demanding in at least one way, or
possibly two: (a) being very restrictive about the sources of basic belief (e.g., holding that reason and
introspection can be sources of basic belief, but perception and memory cannot); (b) requiring epi-
stemic certainty or infallibility for basic belief. Many worry that either requirement greatly limits the
basic beliefs and leads to radical skepticism. Classical foundationalists think the demands appropri-
ate, and some hold that radical skepticism can be avoided (Hasan and Fumerton 2016). But many
others are not convinced. Since we are bracketing skepticism, however, let us avoid appealing to any
requirements of rationality that seem to have radically skeptical consequences.
WHY THEISM IS NOT RATIONAL

Even if we bracket radical skepticism and assume the rationality of our ordinary, commonsense
beliefs regarding the external world, we may have good reasons to deny that the belief in God satis-
fies appropriate requirements of rationality. As we shall see, although there are some serious con-
cerns with theism’s satisfaction of positive requirements for prima facie justification, doubts regard-
ing the satisfaction of the negative, no-defeater requirement, have the widest and strongest impact.
INFERENTIAL JUSTIFICATION FOR THEISM?

When someone claims that belief in God is not rational because it is not supported by good argu-
ments, theists often respond that they do have good arguments. There are the traditional arguments

for the existence of God, including various versions of the cosmological, teleological, and ontological
arguments. We cannot examine the specific arguments here. However, it is worth noting that phi-
losophers who study these arguments for a living, even the theists among them, disagree regarding
which, if any, of these arguments are sound or cogent. This is no surprise, for, as any student of phi-
losophy of religion quickly discovers, the arguments are complex and abstract, and many of the key
premises are controversial or depend on further complex arguments.
Perhaps these traditional arguments are not the only way. Perhaps there are special signs of God
around us. But what signs might these be? Miracles? Confirmed prophecies? Speaking for myself, I
have not observed any miracles, at least not in the sense of something that clearly requires a super-
natural explanation, let alone the existence of God. And I have not observed any shockingly accurate
prophecy that could not be explained away. Moreover, while others might claim to have made such
observations or have some special expertise or source of religious knowledge, there is a great deal of
disagreement among them, and no good, independent way to determine which of them are reliable.
Since all the main accounts of rationality accept no-defeater requirements, there is no escape
from the problem that significant disagreement about these inferential sources undermines reliance
on them.
REFORMED EPISTEMOLOGY AND BASIC BELIEF IN GOD

When someone claims that belief in God is not rational because it is not supported by good argu-
ments, some reply in a different way, by denying that argument is needed at all. According to
“reformed epistemology,” belief in God can be basic, not just in the sense that it is not based on
any other belief, but in the foundationalist’s sense that it is non-inferentially justified without
depending on any other belief for its justification. (To signal this difference, some reformed episte-
mologists call the latter a properly basic belief.) Reformed epistemology can be understood as grant-
ing, at least for the sake of argument, that we do not have adequate evidence—at least not doxastic
or inferential evidence—but claiming that no such evidence is required for rationality.
Could belief in God be (properly) basic? Some beliefs are basic because they are self-evident,
where this means, roughly, that the truth of the belief is obvious to anyone who genuinely under-
stands it. It is self-evident to me that there are no round squares, and that 2 + 2 = 4. While we
can allow for some variation when it comes to what is self-evident to different people, we have very
good reasons to strongly doubt that belief in God could be self-evident to us in this way. I seem to
have no difficulty in understanding the thought that God exists, and yet I can still wonder, quite
intelligibly, whether it is true.
Some foundationalists hold that ordinary, sensory perception can be a source of non-inferential
justification. Perhaps, as held within the Reformed tradition in Christianity, we were created with
a special sort of perception, an inner sense or awareness of God, a sensus divinitatus, that is acti-
vated in certain circumstances, disposing us to believe in God. So perhaps religious experiences
of various sorts can justify my belief that God is present, is speaking to me, forgives me, or has
created all of this.
William Alston and Alvin Plantinga have each defended reformed epistemologist accounts of
basic belief in God. Alston argues that religious belief (and specifically, Christian belief) can be jus-
tified by religious experience or religious belief-forming practices, just as perceptual belief can be jus-
tified by perceptual experience or perceptual belief-forming practices. To provide justification, the
experiences or practices in question must be reliable and so constitute a good ground of sorts, and
the subject must lack defeaters, but the subject need not have justification for believing that it is
reliable (see Alston 1983, 1991). According to Plantinga, a basic belief must be produced by a prop-
erly functioning and reliable cognitive faculty, in an appropriate environment, the sort of environ-
ment for which the faculties were designed, whether by God or evolution. A belief in God that satis-
fies these conditions is properly basic (absent defeaters) (see Plantinga 1981; 1993). Some theists
accept even weaker requirements for basic belief. For example, some are phenomenal conservatives
who hold that its seeming or appearing to me that God exists is sufficient, absent defeaters, for jus-
tification (Tucker 2011).
Alston and Plantinga are both externalists, for they deny that one must have some reason to
think that one’s belief is reliably produced, the output of a properly functioning cognitive faculty,
or that one has in some other way a reason to believe it. As we discussed, externalism seems to have
strongly counterintuitive consequences. However, internalism has difficulties avoiding skepticism
while retaining a robust connection between justification and truth. In any case, externalism is pop-
ular among many epistemologists, including many atheists.
One might object that religious belief-forming practices are unlike perceptual ones in that they
are not widely shared. But the fact that a method is not widely practiced does not show that it is
unreliable. To require that any practice be like sense perception in being universal or more widely
shared is a version of what Alston calls imperialism—insisting that what holds of perceptual practice
must hold of religious or mystical practice, without good reason.
Another objection is that religious practice is not independently verifiable. There is no way to test
whether its deliverances are right. It seems that we need some prior reason to trust the practice, source,
or method, one that does not depend on the very practice, source, or method being assessed.
Reformed epistemologists would here join other externalists in raising the worry that accepting this
as a general requirement is too demanding. Suppose that we could not rely on perception, memory,
introspection, or reflection unless we already had reasons to believe that these sources are reliable.
The problem is that, given that we have a finite number of sources, any such reason must itself be
the output of at least one such source, and so the need to already have a reason to trust the source
arises once again. There is no noncircular way to defend the reliability of our sources of belief. To
hold that one does not require a noncircular justification of reliability for ordinary perceptual sources
but requires it for religious sources is an instance of what Alston calls a double standard.
There are, however, other more serious concerns. We have alternative explanations for why we
have these religious beliefs and the apparent experiences that support them (see Fales 1996). First,
prior beliefs might be the cause of the experiences rather than the other way around. Second, there
is no denying the comfort and psychological benefit that many believers derive from their faith, the
emotional attachment they are likely to have to the possibility of a relationship to God, and the dif-
ficulty of giving up beliefs one was raised to take for granted.
There is also the problem of conflicting results of religious or mystical experiences. If all reli-
gious or mystical experiences depend on the same source or process, then this source or process is
unreliable because it leads to conflicting results. If we distinguish between the sources or processes
in some way, then the question arises: Which of these sources or processes should we rely on? “Rely
on the reliable ones!” is not helpful when we have no independent way of distinguishing between
the reliable or unreliable ones. We ought, therefore, to refrain from trusting any of them.
These criticisms can grant, for the sake of argument, that believers in fact satisfy the external
conditions for prima facie justification proposed by reformed epistemologists like Alston and Plan-
tinga. However, they provide us with reasons to doubt that we satisfy the no-defeater condition,
for we have good reasons to think that any religious and mystical experiences we have are unlikely
to be reliable.
THE PROBLEM OF DISAGREEMENT GENERALIZED

Each of the two preceding sections ended by raising concerns regarding the reliability of (a) inferen-
tial and (b) non-inferential sources of justification. We are now in a good position to state the argu-
ment in a general way that affects any account of the rationality of theism, whether internalist or
externalist, foundationalist or coherentist, so long as the account accepts a no-defeater condition
on rationality.
I was raised Muslim, in a Muslim country, and immigrated to the United States in my early
teens. I recall riding a bus to school, looking around at all the other passengers and thinking to
myself: I am awfully lucky to have been born into the one true faith. People, by and large, tend
to hold the religious beliefs of their parents or culture. Most children of Muslims, Christians, Jews,
Hindus, Sikhs, and so on, tend to accept and remain believers of their parents’ faith, and these faiths
say different, incompatible things. If all these believers, even the intelligent and thoughtful among
them, share a religious belief-forming practice, that method is highly unreliable. If there are many

different methods of religious belief-forming practice, then perhaps my belief is reliable after all.
However, that would be awfully lucky, and continuing to believe in my religion would be arbitrary
and dogmatic. I can be justified only if I have a good reason to take my religious belief-forming
method to be better.
Surely it is possible to rationally disagree with someone! Yes, of course. If you tell me that there
are round squares, or that circles have corners, or that 2 + 2 = 17, I’m not going to suspend judg-
ment, or even lower my confidence in any significant way. I’m much more likely to think you are
playing some kind of prank on me, or trying to see how I would react, and sooner or later, if you
persist, I may begin to wonder about your sanity. This would all be rational for me. So we don’t,
and shouldn’t, always budge in the face of disagreement—at least not significantly. But belief in
God is not like belief that there are no round squares or that 2 + 2 = 4.
Disagreement about God is more like the following case (see Christensen 2007): Suppose we
have just finished a meal at the restaurant and decide to split the bill evenly. We each calculate
how much we owe. I come up with $23, and you $21. Suppose we each know that the other person
is generally just as good at such calculations, and nothing seems different in the present case. After
finding out that you get a different—incompatible—answer, it seems it would be epistemically irra-
tional of me to hold steadfastly to my answer and believe that you made an error and I didn’t, or
that it is significantly more likely that you are wrong. It seems rational for me to suspend judgment.
At this point, one might object: If the problem of disagreement is a genuine problem, doesn’t it
affect all sorts of beliefs, including ordinary and scientific beliefs, and so result in skepticism? No.
First, people by and large agree with respect to a great deal of ordinary and scientific belief. Second,
as just discussed, there are various beliefs we hold that are justified to such a high degree that we
would be justified in dismissing others’ disagreement, justified in taking them not to be in as good
an epistemic position as we are, even if we can’t initially see exactly why.
However, won’t such a view still have a strongly skeptical result at least where there is signifi-
cant disagreement, including in science and philosophy? Well, let’s think about these sorts of dis-
agreements, on such topics as the freedom of the will, the nature of time, or competing interpreta-
tions of quantum mechanics. When it comes to such open disagreements between philosophers or
scientists with similar abilities, who seem to have the same evidence or have closely examined the
same arguments, it makes sense to say that they are not (at least not yet) epistemically rational in
believing theirs is the correct theory. Does this mean we should stop doing physics or philosophy,
or stop taking our research seriously? Of course not. There may be many good reasons to continue
to develop and defend one particular theory over another, or take it as a “working hypothesis,”
including for the sake of furthering future knowledge and understanding in our field, and finding
important points of agreement. But this should not be confused with being epistemically rational
in believing one theory over another.
The reader might have a final worry: But won’t philosophers disagree with your conclusion and
how you came to it? And if so, shouldn’t you suspend belief about whether these are genuine defea-
ters, in which case the defeater is defeated? The first thing to notice is that this is a disagreement
that occurs one level up: it is a disagreement about whether I am rational in believing that theism
is defeated by significant disagreement about religious matters. That would not take away the fact
of disagreement at the first level. And that seems to be enough to generate the problem. Compare
the restaurant case again: Suppose you initially react to the different calculations of the bill by sus-
pending judgment. Suppose you then find out that there are philosophers who disagree about
whether that’s the rational attitude to hold. Does that suddenly make it epistemically rational for
you to believe that your calculation was correct and mine is not? Surely not!
Let us take stock. We began by considering possible inferential and non-inferential sources of
rational belief in God. In each case, we considered undercutting defeaters, reasons to doubt that
these sources were reliable. We spent some time discussing the strongest reason, the problem of dis-
agreement regarding the existence and nature of the divine, and the existence of competing belief-
forming practices. That there is this sort of disagreement and divergence in religious belief-forming
practice is available to most if not all ordinary adults today. We should conclude that the case
against the rationality of theism, while defeasible, is quite strong.
DOES THE TESTIMONY OF SACRED SCRIPTURES FROM THE

RELIGIONS OF THE WORLD FAVOR (MONO)THEISM OR ATHEISM?
Richard Carrier, The Secular Academy
The basic principle of proposing “scripture” as a special source of knowledge is the assumption that
there is something uniquely reliable about a holy text that cannot be claimed of other, merely
human-sourced texts. Often this is because of the proposition that the holy text was written by
someone inspired by superhuman intelligence, or who was literally dictating the speech of superhu-
man intelligences. Otherwise, it is because of the proposition that the author wrote under the influ-
ence of an inspiring force or in supernatural harmony with the universe and its truth, or wrote what
they received in some other state of supernatural access to knowledge.
For example, Muhammad claimed to be dictating the words of the angel Gabriel, who in turn
was merely giving voice to the words of God, acting as God’s instrument. As such, the text of the
Qurʾan is proposed to be literally, verbatim, the word of God; and as God is omniscient and infalli-
ble, the text thus conveyed is likewise more reliable than any human text could possibly be. Simi-
larly, the Old and New Testaments are variously believed to have been composed or dictated under
inspiration from God’s Holy Spirit or through the conveyance of angels acting as God’s instru-
ments, and are thus also more reliable than any merely human texts—regardless of whether it is
assumed to be literally true, as fundamentalism contends, or conveys its knowledge in some other
capacity, such as allegory. Joseph Smith likewise claimed the Holy Texts he “discovered” were writ-
ten by men inspired by God and vouchsafed as true by the angel Moroni and a corresponding mir-
acle of God. L. Ron Hubbard claimed his lectures and texts in Scientology conveyed the psychically
recovered memories of past lives and superhuman civilizations dating back tens of millions of years,
and consequently he claimed that they were more reliable than any other merely human writing.
The holy texts of Daoism are likewise regarded by adherents as superior in reliability because their
authors, when they were composing them, were in perfect harmony with the Dao governing the
universe, and therefore the text is “perfect,” and more accurate in describing reality, in some sense
that a merely human text cannot achieve.
BUT NO SUCH TEXTS EXIST

The first and most obvious reason to conclude none of this is true is that there are so many such
texts purported to be thus supernaturally derived that pervasively contradict each other in funda-
mental ways. Daoists do not “receive the same information” as Muslims nor Buddhists nor Catho-
lics nor Lutherans nor Jews nor Mormons nor Scientologists, and so on. This is exactly what we
should expect to observe if none of these claims is true and not at all what we should expect if there
were any true source of supernaturally received knowledge. Even on the supposition that all but one
of these claims are fake, we should expect this to be evident in some distinctive way. For example,
the true source—if any there were—should substantially exceed the fake ones in its access to unex-
pected yet verifiably true information. Yet none of these “scriptures” displays any knowledge of the
world not already accessible to ordinary, uninspired humans at the time they were written. In fact,
they all display exactly the same ignorance and commonplace false beliefs prevalent at the time of
their writing, thus proving they had no supernatural source. That observation is so extraordinarily
improbable on the theory that supernaturally informed texts exist, that we can be reasonably certain
none does. Which is the second reason to conclude against the existence of such texts.
We can far more easily explain the belief that such texts exist, precisely because so many cul-
tures develop this belief, yet always in texts containing so many substantially different truth-claims.
In practice, Christians believe their holy texts are supernaturally informed for the same reasons Mus-
lims and Daoists and Mormons and Scientologists do. A Christian, for example, has to explain how
so many can have such adamantly false beliefs in the supernatural reliability of what the Christian
must agree are not inspired scriptures. And yet that explanation will prove just as valid for the
Christian themselves, in every particular. There is simply no evidence available to argue that the
Christian is not operating under those same error-inducing influences. And in the absence of such

evidence, the prior probability prevails: the Christian most likely believes what they do about their
scriptures for the same invalid reasons they must agree that Muslims, Daoists, Mormons, and Scien-
tologists do (Loftus 2013).
ILLUSTRATING THE PROBLEM

This author’s own past affords an example illustrating the problem. Unlike many Western atheists, I
did not grow up in or even much aware of any fundamentalist Christian tradition. My first real faith
was in Daoism. I was persuaded of the supernatural perfection of the Dao De Jing by a religious
experience, and I became a believer. I studied and practiced the teachings contained in that scrip-
ture, and further religious experiences continued to confirm the truth of my faith. I have detailed
the nature of those experiences before (Carrier 2005, 9–17). They are much the same as reported
by believers and converts to every other religious tradition: incredible, prescient coincidences; inner
feelings akin to experiencing a “Holy Spirit”; visions; and miraculously obtained insights through
Daoist meditation and scripture that substantially contributed to my happiness, success, and spiri-
tual fulfillment.
A true believer of any other religion must of course explain how all of this could happen to me.
How could my experience of the Dao be just as confirming and rewarding as their experience of the
Holy Spirit? How could I be receiving visions confirming the truth of Daoism over all other reli-
gions? How could the Daoist scriptures have so improved my life, my serenity, my character?
How could those amazing coincidences have happened but for the supernatural influence of the
Dao? How could my Dao-informed intuition be so successful at obtaining knowledge and wisdom?
Supernatural explanations, such as that I was deceived in all this by the Devil or some other lying
spirit are self-defeating: a Christian or a Muslim, for example, has no means to demonstrate that
they are not the one deceived by the Devil or some other lying spirit. Natural explanations entail
the very same reasoning, such as what I myself eventually came to realize. The coincidences I expe-
rienced were actually quite statistically ordinary in global context; the teachings were perfectly ordi-
nary human insights that had survived thousands of years precisely because they proved helpful to
the lives of those who studied them, and I was selecting and interpreting them in precisely the
way most useful to my own situation; my visions and insights were produced by natural processes
(altered states of consciousness and heightened thoughtfulness generated by meditative trance
inducement and sleep deprivation), and they only communicated to me my own assumptions,
intuitions, and beliefs. The same things could explain every other religious adherent’s experience.
Because this is so readily possible and so commonly the case worldwide, we need to learn and
apply cognitive tools for ruling these things out, before we can conclude we are special, and not just
like everyone else who is totally convinced of the truth of a false religion. I found my explanations
in mathematics and science (understanding the nature of coincidence, delusion, and hallucination,
for example), and by admitting there were failings and contradictions in the scriptures, that they
were merely human, and as such flawed and incomplete, possessing no knowledge not humanly
available to the ordinary geniuses of ancient China who composed them. And yet when we apply
these same standards to all other scriptures, we get the same results. Therefore, there are no super-
naturally inspired scriptures.
WHY SCRIPTURE FAILS

Scriptural religions are simply not backed by reality (Carrier 2005, 2011; Loftus 2010), except in
ways wholly ordinary and thus not indicative of any supernatural source or special reliability. The
manuscripts conveying the text of every scripture also routinely vary and disagree in their readings,
and we almost never have the originals to test them by; this is not indicative of a supernaturally
endorsed text, but an ordinary human one. A supernaturally endorsed text should be supernaturally
preserved, but none is. And a great deal of scripture is fake, being forged or falsified (e.g., Exodus;
the book of Daniel; sections of Isaiah; the Pastoral Epistles), as well as arbitrarily selected, resulting
in contradictions and errors. Which is also not expected on a supernatural origin.
Scriptures also rely a lot on fiction to convey beliefs about society and the world and the human
condition that are often factually false or wholly indemonstrable as anything other than some poorly
informed human guess or opinion. This is all just as true of the Jewish and Christian Bibles (e.g.,
Carrier 2014; Ehrman 2009, 2013; Avalos 2007; Crossan 2012; Helms 2006, 1989; Finkelstein
and Silberman 2002) as of the Qurʾan (Warraq 2011). Scriptures are not only plagued by contradic-
tions and factual errors but unethical teachings as well (Avalos 2011, 2015; Warraq 2011), which
we have since repudiated as not conducive to human happiness nor in accord with how the world
really works (e.g., Shermer 2015; Loftus 2014). Such impractical and unethical teachings may have
made sense to the primitive and ignorant peoples who wrote those texts, but that is precisely what
indicates their nonsupernatural origin (e.g., Avalos 2005, 2011, 2015).
As an example, the Qurʾan’s and New Testament’s claims to supernatural authority both
require the Old Testament’s supernatural authority (as in each case the one declares and depends
on the other). Yet modern democracy is defined by literally outlawing the Old Testament’s com-
mandments so as to protect human happiness from its God’s ordained evils; for example, the Bill
of Rights, and the rest of the US Constitution, outlaws six of the Ten Commandments, and
almost all the other commandments of God issued beyond those ten (Loftus 2014, 180–205).
And the factual errors, contradictions, and failed prophecies of the Old Testament are likewise
legion (Wells 2012; McKinsey 2004). It therefore cannot be supernaturally reliable. And if it is
not, then neither can the New Testament be, nor the Qurʾan. Which eliminates the scriptures
of the three most popular religions of the world. Which also demonstrates that popularity does
not correspond with truth.
The problem is basically laid out by Theodore M. Drange (2006): scriptures supporting
theism always contradict themselves or each other; are often too unclear in their meaning to be
useful; frequently contain factual errors, including unfulfilled prophecies, and ethical defects
(“such as God committing or ordering atrocities” and endorsing slavery); they frequently contain
interpolations or forgeries; manuscripts of them vary from each other, and the originals no longer
exist; and no objective means exists to identify which texts are supernaturally inspired (as what is
“canon” varies from religion to religion, and sect to sect). Despite supposedly having the desire
that we understand and enact his will, God provides us with no means to resolve any of these
issues with his supposed scriptures. This is not what a supernaturally inspired text should look
like. As Drange concludes, “Since those facts would be unexpected on the assumption that the
God” of any particular religion “exists, they constitute evidence for the nonexistence of that deity”
(2006, 374).
DEFENDING SCRIPTURE ANYWAY

Apologists for the various religions that rely on scriptural authority attempt to evade these conclu-
sions by insisting their scriptures are not errant and are superhumanly prescient. But the evidence
is simply overwhelmingly against both claims (Murray 2010, 46–54). Only with enormous contor-
tions of added suppositions can any scriptures be made to appear inerrant; yet inerrant texts should
not require such elaborate devices to make them so. As those machinations are fallaciously circular;
they assume the inerrancy of the texts as a premise in order to make them inerrant after the fact;
they require already knowing what’s true from information not in the scriptures in order to “reinter-
pret” those scriptures to fit it after the fact. But if all that has to be done, it is self-evident the scrip-
tures were not miraculously written to convey such knowledge. As otherwise they would have. No
devices would be necessary.
Supernatural prophecy has also never been established. Texts written after the fact and retro-
fitted to prior prophecies do not verify prophetic power, for example. Nor are vague predictions that
can be made to fit any outcome. Nor outcomes that are commonly expected anyway. Nor are out-
comes caused by humans deliberately aiming to fulfill a prophecy. And yet apart from those, there
are no fulfilled prophecies in any scripture. Scriptures claimed to be prophetic sometimes even con-
tain failed prophecies, which is evidence directly against their supernatural origin. For example, Eze-
kiel predicted that the city of Tyre would be entirely covered by the sea (26:19) and would never be
found (26:21). Yet neither has occurred—not to the ancient city, nor the modern. And Jesus pre-
dicted he would fly down from outer space with an army of angels and end the world within the
very lifetime of his peers (Matthew 24:14–34). It’s now been two thousand years and no such thing

has occurred. By contrast, if any scripture had predicted the precise date and location of a falling
meteor or tsunami, or indeed a universally observed flight of angels or any other extraordinary
event—and especially if it successfully made several such predictions—then we’d have a plausible
claim of superhuman information. Yet genuinely impressive prophecies like that are never to be
found. Which is suspicious … unless no scriptural theism is true.
Similarly, claims that scriptures miraculously predicted modern scientific facts are not credible
either. There are no real examples of any scripture doing so. But there are plenty of examples of
scriptures failing to. No scriptures correctly predicted that animals existed millions of years before
trees and flowers, for example, or that the universe is billions of years old or that the earth was
inhabited for billions of years by single-celled organisms before they started assembling themselves
into animals and plants. Such statements would have been incredible; yet nothing like that is to
be found. No scriptures contain, in fact, any knowledge not already available to people when they
were written. For example, the Qurʾan contains no information about embryology not already dis-
cernible from prior Greco-Roman scientific treatises. And this includes information that was incor-
rect. For instance, the Qurʾan claims stars were formed after the planet Earth (41:9–12). Genesis
claims the sun was formed after plants and trees (1:11–19); in the Gospel of Mark, Jesus is ignorant
of the existence of germs, and thus fails to inform the public of the need and value of careful
hygiene and even denounces it (7:1–15). Such failures conclusively demonstrate these texts were
written by ignorant humans with no access to supernatural knowledge.
That apologists have to evade these observations with dubious and feeble defenses of the super-
natural reliability of their scriptures demonstrates the absence of any supernaturally reliable scrip-
tures. If any existed, they should be far more impressive than this, and easily proven so. That they
never are, is extremely improbable on the thesis that any supernaturally reliable texts exist but
exactly what we should expect if none does.
WHY DO SO MANY CULTURES DEVELOP THESE FALSE BELIEFS?

This leaves the question of why so many cultures develop false beliefs in the supernatural authority of
certain texts. We do not actually have to answer this question, as we already know they do. The Chris-
tian must admit this of the Muslim, Daoist, Mormon, and Scientologist as surely as the Muslim, Dao-
ist, Mormon, and Scientologist must admit this of the Christian and each other. That it commonly
happens is a fact. Why it happens hardly matters. But apologists often aim to prove their holy texts
are somehow “superior” to their counterparts and that they are somehow “immune” to what causes
everyone else to have a false belief in “scriptures.” The apologist wants to claim their scriptures must
therefore be special. This does not actually work, as their scriptures still fail every other test. For exam-
ple, they contain no identifiably superhuman knowledge of anything, particularly given the context of
when they were written. But nevertheless, this is a consideration to examine.
The most likely explanation comes from cognitive science (Shermer 2011; Loftus 2010, 47–80):
the innate human characteristics of ambiguity intolerance, intolerance of uncertainty, and avoidance
coping (Furnham and Marks 2013) all correlate with a strong need for absolute, unquestionable, infal-
lible sources (examples and bibliography for the Christian West in Carrier 2017, 500–526; and Car-
rier 2009). This is in part because people feel uncomfortable with ambiguity, complexity, and uncer-
tainty, and yet most knowledge earned the hard way comes with those attributes. People also
biologically vary in how comfortable they can get with that fact, so there is often a correlation between
the degree of intolerance of ambiguity and uncertainty and the degree of belief in the supernatural reli-
ability of holy texts—or some substitute, like a belief in one’s own superhuman access to knowledge,
or that of a religious authority, although the latter tend to fare more poorly in cultural success, because
immutable texts can be more convincingly sold as superhuman. Likewise, unlike individuals, texts can
more readily accumulate the support of cultural reinforcement; and unlike religious leaders, they can
long outlive even a culture, much less an individual or institution.
The science of anthropology thus completes the explanation (Eller 2014; Loftus 2010, 25–46):
people have a biological tendency to go along with what is reinforced by their culture; cultures that
acquire a pervading belief in holy texts will reinforce that mode of resolving individual intolerance of
ambiguity and uncertainty; and cultures that choose texts to so revere will tend to outlive, and thus
out-propagate, cultures that choose individuals to so revere (because texts, unlike individuals, are
potentially immortal). No believer in “supernatural scriptures” can show themselves immune to
these two causes: psychological intolerances that are comforted by them and a happenstance cultural
tradition and reinforcement of them.
In short, people need to believe in supernaturally informed texts, and so they tend to create
texts to believe that of. And once created, they can’t be altered, lest it be admitted they were fallible.
This explains all beliefs the world over in “scriptures” as supernaturally reliable sources of knowledge
(Murray 2010, 218–226). No other explanation fits all examples, nor does any other explanation
make any successful predictions. No “scriptures” display any superhuman knowledge of anything;
they are, rather, always found to be ignorant and errant on many subjects. Yet belief in them
remains as strong in every religion that has any.
WHAT ABOUT THE TESTIMONY OF RELIGIOUS TRADITION AND AUTHORITY?

Of course, “scriptures” have to be “interpreted.” This interpretation derives from some form of “tra-
dition and authority.” Religions without scriptures already depend on tradition and authority. But
scriptural religions also depend on it for understanding what is written in their holy texts. The prin-
ciples and methods used to interpret a scripture fall under the category of hermeneutics. Some are
the same principles and methods employed to understand ordinary texts, such as understanding
the context of a passage (textually, culturally, historically, politically, etc.) and verifying hypotheses
about its author’s knowledge and intentions (Carrier 2012). But the assumption that scripture is
supernaturally informed—and by a specific divine being—entails other hermeneutic principles
unique to that condition. Are they sound? Probably not.
Here we will query scriptural hermeneutics specifically, which typically operates from faith-
specific dogmas rather than objectively verifiable principles. That already renders this “special” kind
of hermeneutics suspect as fallaciously circular—if, for example, you approach a text assuming cer-
tain things (such as that the Bible is supernaturally inspired), then use the text to verify those
assumptions (such as that the Bible is supernaturally inspired). Even when this error is carefully
avoided, the assumptions on which special hermeneutics must be built are not themselves supported
by evidence and thus are not credible. This explains why there is no consistently reliable or agreed
upon scriptural hermeneutics.
For example, scriptural hermeneutics is often believed to be supernaturally informed in and of
itself (as mentioned in the literature cited below). Hence a Christian is often expected to be guided
by “the Holy Spirit” in order to correctly apprehend the meaning of a scriptural text; and this, believ-
ers often assert, gives them some sort of supernatural power in understanding the text that is lacking in
other readers. But the existence of this power has never been demonstrated and is rendered implausi-
ble by all the same evidence calling scripture itself into question; for example, the vast contradictory
diversity of both axioms and results in biblical hermeneutics, even applied to exactly the same texts,
varying wildly by tradition, sect, religion, historical period, and even individual interpreter (e.g.,
Reventlow 2010; Porter and Stovell 2012; Cohen and Berlin 2016; Saeed 2006). This is exactly what
we should expect if biblical hermeneutics is just a human contrivance; it is not at all what we should
expect if it were coming from or informed by a universal supernatural source. A reliable method by
definition should get consistent results across all time periods, observers, and sects; therefore, that bib-
lical hermeneutics fails at this, and quite spectacularly, establishes it as unreliable.
Moreover, all scriptures are claimed to contain essential, lifesaving information the revered super-
natural powers want us to benefit from. But since to understand a lifesaving text correctly requires cer-
tain skills, any deity who expected their lifesaving text to be correctly understood would include in it
those very skills, plainly articulated, or otherwise ubiquitously and conspicuously provide a primer on
those skills by some other universal means, or, for instance, ensure that “the Holy Spirit” always cor-
rectly guided everyone to the same result (Drange 2006). Yet everyone is not guided by “the Holy
Spirit” to the same result, and until taught by secular philosophy, all scriptural societies lacked (and
often still lack) critical thinking tools; and none of their scriptures teaches or advocates such tools,
but they often teach inept thinking patterns or even attack necessary tools of critical thought (Carrier
2009, 329–368, 385–405). For example, curiosity, questioning, doubting and testing, attention to

logical reasoning, and requiring evidence rather than faith independent of evidence—all are funda-
mentally necessary to avoid error and false beliefs. Yet these are rarely promoted, and usually con-
demned, by sacred scriptures (Carrier 2017, 471–541). It is not expected that a supernaturally
informed text would come without the means to interpret it reliably, objectively, and consistently.
But that is what we would expect if none of these texts is supernaturally informed.
Instead, what has emerged is a bewildering and ever-changing divergence of interpretive princi-
ples. This phenomenon is self-refuting (Drange 2006; Avalos 2007; Murray 2010). A method that
produces consistently inconsistent results is by definition unreliable. A better explanation of this
observed fact is that “interpretation” is just substituting the interpreter’s own (or the interpreter’s
culture’s or sect’s) knowledge, opinions, and desires for the text being interpreted, which reduces
the reliability of any hermeneutic process to the reliability of the individual’s (or their culture’s or
sect’s) knowledge, opinions, and desires in accessing knowledge of the world. And that has never
been demonstrated to be remarkably better for anyone, religious or otherwise, much less from any
particular religion. Outside ordinary human experience (e.g., learned skills in predicting the envi-
ronment), in questions of empirical fact intuition has always been shown to be highly unreliable,
or its reliability unknowable and thus unusable, compared to controlled scientific methods and
other secular means of verifying claims to fact. Folk intuitions, even in the domains of logic and
mathematics, perform poorly; which is why formal procedures had to be invented to test them
by. Absent experience-based learning and well-tested formalisms, all intuition does is reinforce,
rather than defend against or control for, the litany of cognitive biases that plague our faculties. This
is what defined the scientific revolution and made tremendous advances in knowledge possible: the
realization that “intuition” is not a reliable source of knowing. And scriptural hermeneutics, when
not simply reduced to an already secular literary and historical method, is always just a variant of
intuitionism, further enslaved to arbitrarily selected dogmas; which is why every sect, every religion,
has its own hermeneutic principles, and thus its own completely different results; and even within
sects, there are many divergent hermeneutic traditions and disagreements, and no objective means
to verify any of them.
For example, in the balance of history, a reliance on scripture has led to incorrect scientific the-
ories (about the world and about human beings and their societies), and more often made the world
worse, producing inhuman harms, dangers, and injustices for thousands of years (see, e.g., Loftus
2014; Avalos 2005, 2011). We would be extremely wary of and no longer trust any human leader
who did the same; we would not defend them by saying, “Yes, they did not know about germs or
heliocentrism, and, yes, they commanded genocide and slavery, but they also condemned dishon-
esty, so we should revere them as supernaturally reliable and trust anything else they said is true
and good.” Nor can we defend any religion’s scriptures by saying some of what they contain is
good, which then influenced the world for good “despite” all the disaster and harm they caused as
well. Because there are no good teachings in any scripture not also found already endorsed in secular
literature without the bad accoutrements, “scriptures” can claim no special value but are inferior to
more judiciously constructed human texts. Either way, their inconsistency in results establishes their
unreliability.
To try to solve this problem by picking and choosing only the good and abandoning the bad
also requires a non-scriptural guideline, which as such does not need the scripture, whereas teaching
reverence for the scripture entails teaching reverence even for its errors and evils. A more humane
and less dangerous approach would be to treat all scriptures like all other literature: fallible human
products, of no supernatural origin, that we must approach critically, only taking from them what
we can independently verify is correct without them. That this is the case is unexpected on any pop-
ular version of theism. But it is entirely expected on atheism. It is even worse to suppose one can
solve this problem by choosing one’s faith tradition, then regarding its premises to be “properly
basic” knowledge (Plantinga 1993; Bolos and Scott 2015), requiring no evidence in their defense,
in the hopeful conviction that the God one chose would surely have created us with faculties suffi-
cient for discovering those premises. Because we must still always ask: what if there is no such God,
or some other God instead? This is precisely the same error that convinced this author of the
unquestionable truth of Daoism. If it can lead me to that error, it may have led the Christian or
the Muslim or the Mormon to the same error (Webb 2015). Allowing such reasoning is even dan-
gerous to human society (Peykani 2015). So we need some means of telling the difference. And
there is only one means known: independently verifiable evidence. And for any “scriptures” and
their “interpretation” being supernaturally informed, there simply is no evidence.
The most popular theologies entail the expectation that God would want us to have the knowl-
edge we need to ensure our continuing welfare. That no such knowledge has been communicated,
by any divine source, is therefore evidence against theism. This evidence can be “explained away”
by replacing theism simpliciter with a more convoluted theism burdened with ad hoc assumptions
for which we have no evidence, and thus we have no reason to believe any of them, apart from
our need to avoid the consequences of the evidence we actually have. The addition of such assump-
tions always reduces the probability of a hypothesis; it never raises it. Theism therefore cannot be
rescued by such a device, not in conformity with logic, at any rate.
This is all the more certain for popular theisms, which all maintain that God has communi-
cated such knowledge to us. Yet every religion claims a completely different set of communications,
and none can demonstrate any better access to the truth of the matter than any other. All contain
the same errors and ignorance possessed by the authors when they wrote them. None contains
any superhuman knowledge. This is simply improbable on the claim that they derive from a super-
human and not a merely human source, but this is exactly what we should expect if they are merely
human-derived and just believed to be divine, for reasons other than any objective demonstration in
the evidence. Indeed, from any such religion’s point of view, billions readily do believe false scrip-
tures divine, and with equal fervor. We therefore already have an explanation for why such beliefs
would exist, one that does not require any Ptolemaic epicycle like an actual deity communicating
with anyone: that all these beliefs come from human psychology, not gods.
The absence of any evidence that any of these disputed scriptures contains any supernaturally
derived knowledge at all only confirms the conclusion that none does. The fact that they are awash
with errors, ignorance, contradictions, and unethical teachings, and have all caused innumerable ills
to befall our civilizations for hundreds of years—so much so that even most believers have begun to
disregard most of their contents, and entire nations now define their progress by not only abandon-
ing but outlawing many dictates of these scriptures (Loftus 2014, 180–205)—only verifies that con-
clusion. The failure of any scriptures to be informative or defensible—other than as collections of
often fallible and perfectly ordinary human wisdom—is evidence against every theism those scrip-
tures maintain, and even, to a some extent, against theism generally.
This conclusion holds not only for all known scriptures but for all other special sources of
knowledge claimed for theism. By contrast, secular methods of advancing and improving human
knowledge (such as science, mathematics, logic, rational-empirical history, criminal forensics and
jurisprudence, and evidential and analytical philosophy) have all proven themselves more objectively
reliable than any competing alternatives, and have made continual and considerable progress in
helping and mastering both ourselves and the world. Theism therefore has no epistemic footing
capable of adding to or challenging the knowledge gained by these other methods. And that fact
is itself evidence against theism.
Bibliography
Alston, William P. “Christian Experience and Christian Belief.” In Faith and Avalos, Hector. The End of Biblical Studies. Amherst, NY: Prometheus
Rationality: Reason and Belief in God, edited by Alvin Plantinga and Books, 2007.
Nicholas Wolterstorff, 103–134. Notre Dame, IN: University of Notre
Avalos, Hector. Slavery, Abolitionism, and the Ethics of Biblical Scholarship.
Dame Press, 1983.
Sheffield, UK: Sheffield Phoenix Press, 2011.
Alston, William P. Perceiving God: The Epistemology of Religious Experience. Avalos, Hector. The Bad Jesus: The Ethics of New Testament Ethics. Sheffield,
Ithaca, NY: Cornell University Press, 1991. UK: Sheffield Phoenix Press, 2015.
Avalos, Hector. Fighting Words: The Origins of Religious Violence. Amherst, Bolos, Anthony, and Kyle Scott. “Reformed Epistemology.” In Internet
NY: Prometheus Books, 2005. Encyclopedia of Philosophy. 2015. https://www.iep.utm.edu/ref-epis.

Carrier, Richard. Sense and Goodness without God. Bloomington, IN: Helms, Randel. Gospel Fictions. Amherst, NY: Prometheus Books, 1989.
AuthorHouse, 2005.
Helms, Randel. The Bible against Itself: Why the Bible Seems to Contradict
Carrier, Richard. Not the Impossible Faith: Why Christianity Didn’t Need a Itself. Altadena, CA: Millennium Press, 2006.
Miracle to Succeed. Morrisville, NC: Lulu, 2009.
Hume, David. Dialogues concerning Natural Religion and Other Writings.
Carrier, Richard. “Christianity’s Success Was Not Incredible.” In The End of Cambridge, UK: Cambridge University Press, 2007.
Christianity, edited by John W. Loftus, 53–74. Amherst, NY:
Lehrer, Keith. Theory of Knowledge. London: Routledge, 1990.
Prometheus Books, 2011.
Loftus, John W., ed. The Christian Delusion. Amherst, NY: Prometheus
Carrier, Richard. Proving History. Amherst, NY: Prometheus Books, 2012.
Books, 2010.
Carrier, Richard. On the Historicity of Jesus. Sheffield, UK: Sheffield Phoenix
Loftus, John W. The Outsider Test for Faith: How to Know Which Religion Is
Press, 2014.
True. Amherst, NY: Prometheus Books, 2013.
Carrier, Richard. The Scientist in the Early Roman Empire. Durham, NC:
Loftus, John W., ed. Christianity Is Not Great. Amherst, NY: Prometheus
Pitchstone, 2017.
Books, 2014.
Christensen, David. “Epistemology of Disagreement: The Good News.”
McKinsey, C. Dennis. The Encyclopedia of Biblical Errancy. Amherst, NY:
Philosophical Review 116, no. 2 (2007): 187–217.
Cohen, Mordechai Z., and Adele Berlin. Interpreting Scriptures in Judaism,
Murray, Malcolm. The Atheist’s Primer. Peterborough, ON: Broadview Press,
Christianity, and Islam: Overlapping Inquiries. Cambridge, UK: Cam-
2010.
bridge University Press, 2016.
Conee, Earl, and Richard Feldman. “Internalism Defended.” American Peykani, Jalal. “Plantinga’s Reformed Epistemology and Religious Extrem-
Philosophical Quarterly 38, no. 1 (2001): 1–18. ism: Some Historical Evidences.” Philosophical Investigations 9, no. 17
(2015): 105–116.
Conee, Earl, and Richard Feldman. Evidentialism: Essays in Epistemology.
Oxford: Oxford University Press, 2004. Plantinga, Alvin. “Is Belief in God Properly Basic?” Noûs 15, no. 1 (1981):
41–51.
Crossan, John Dominic. The Power of Parable: How Fiction by Jesus Became
Fiction about Jesus. New York: HarperOne, 2012. Plantinga, Alvin. Warrant and Proper Function. New York: Oxford
Dougherty, Trent, and Justin P. McBrayer, eds. Skeptical Theism: New
Essays. Oxford: Oxford University Press, 2014. Porter, Stanley, and Beth Stovell, eds. Biblical Hermeneutics: Five Views.
Downers Grove, IL: InterVarsity Press Academic, 2012.
Drange, Theodore M. “The Arguments from Confusion and Biblical
Defects.” In The Improbability of God, edited by Michael Martin and Reventlow, Henning Graf. History of Biblical Interpretation. 4 vols. Boston:
Ricki Monnier, 369–379. Amherst, NY: Prometheus Books, 2006. Brill, 2010.
Ehrman, Bart. Jesus, Interrupted: Revealing the Hidden Contradictions in the Rowe, William L. “The Problem of Evil and Some Varieties of Atheism.”
Bible. New York: HarperOne, 2009. American Philosophical Quarterly 16, no. 4 (1979): 335–341.
Ehrman, Bart. Forgery and Counterforgery: The Use of Literary Deceit in Early Saeed, Abdullah. Interpreting the Qurʾān: Towards a Contemporary Approach.
Christian Polemics. New York: Oxford University Press, 2013. London: Routledge, 2006.
Eller, Jack David. Introducing Anthropology of Religion. London: Routledge, Shermer, Michael. The Believing Brain: From Ghosts and Gods to Politics and
2014. Conspiracies; How We Construct Beliefs and Reinforce Them as Truths.
New York: Times Books, 2011.
Fales, Evan. “Scientific Explanations of Mystical Experiences.” Religious
Studies 32, no. 3 (1996): 297–313. Shermer, Michael. The Moral Arc: How Science and Reason Lead
Humanity toward Truth, Justice, and Freedom. New York: Henry
Finkelstein, Israel, and Neil Asher Silberman. The Bible Unearthed:
Holt, 2015.
Archaeology’s New Vision of Ancient Israel and the Origin of Its Sacred
Texts. New York: Simon and Schuster, 2002. Tucker, Chris. “Phenomenal Conservatism and Evidentialism in Religious
Epistemology.” In Evidence and Religious Belief, edited by Kelly James Clark
Furnham, Adrian, and Joseph Marks. “Tolerance of Ambiguity: A Review of
and Raymond J. VanArragon, 52–73. Oxford: Oxford University Press,
the Recent Literature.” Psychology 4, no. 9 (2013): 717–728.
2011.
Gettier, Edmund. “Is Justified True Belief Knowledge?” Analysis 23, no. 6
Warraq, Ibn. Which Koran? Variants, Manuscripts, and the Influence of Pre-
(1963): 121–123.
Islamic Poetry. Amherst, NY: Prometheus Books, 2011.
Hasan, Ali, and Richard Fumerton. “Foundationalist Theories of Epistemic
Webb, Mark. Comparative Doxastic-Practice Epistemology of Religious
Justification.” In Stanford Encyclopedia of Philosophy, edited by Edward
Experience. Cham, Switzerland: Springer, 2015.
N. Zalta. Spring 2017. https://stanford.library.sydney.edu.au/archives/
spr2017/entries/justep-foundational/. Wells, Steve. The Skeptic’s Annotated Bible. Moscow, ID: SAB Books, 2012.
TOPIC 5
Religious Experience: Theism

Samuel Lebens
Senior Research Fellow, Philosophy Department
University of Haifa, Israel
What is the philosophical significance of religious experience? Could a religious experience give you reason to
believe in God? If so, what sort of experience? If not, why not? And could the religious experiences of others
give you reason to believe in God even if you’ve never had such an experience yourself? In this chapter, we
explore these questions.
DOES THIS BOOK EXIST?
You’re reading this book. Do you know that it exists? The skeptic says you don’t. You might be hal-
lucinating. I assume you’re not so radical a skeptic as that. You do believe that this book exists, that
you picked it up (or perhaps, downloaded it) without hallucinating.
We know this much: it’s impossible to have a genuine experience of a book if that book doesn’t
exist—you can think you’re seeing one; you can have an apparent experience of a book, but you
can’t really see a book if there’s nothing really there to see. Genuine experiences have to have
objects. That is to say, experiences are experiences of something. Even your emotional experi-
ences—mere feelings that don’t seem to be experiences of anything outside of you—still have
objects. For example, your experience of happiness has an object, namely, the fact that you are
happy or your happiness itself.
Indeed, you probably endorse this argument:
1. If a person has a genuine experience of this book, then this book exists.
2. Some people (in this instance, you) do have a genuine experience of this book, therefore:
3. This book exists.
Even the skeptic accepts premise 1. You can’t have a genuine experience of a nonexistent book.
The second premise is more controversial, but if you’re not a radical skeptic, it won’t trouble you.
The two premises combine to entail the conclusion.
A number of theists take themselves to have similarly good reason to believe in the existence of
God:
1. If a person has a genuine experience of God, then God exists.

2. Some people have a genuine experience of God, therefore:
3. God exists.
Again, the first premise is uncontroversial. The second carries all the weight. Even if you’re not
a skeptic about genuine book-experiences, you might be skeptical about God-experiences. Are they
genuine, or more like hallucinations?
135
If someone tells you that she (or he) has read a book, you
KEY CONCEPTS believe her. So why not believe her when she tells you that
she has had a God-experience? What makes book-experiences
so much more reliable? You can probably think of numerous
Belief
differences. We’ll try to enumerate and investigate them, as
God-Experiences we move on.
Monotheism We could go about the task of this chapter differently. We
Mysticism could begin with a definition of “religious experience.” That
would require a definition of “religion.” Some religions are the-
Religious Experience
istic. Some are not. Some religious experiences are theistic.
Some are not. We will discuss nontheistic mysticism in passing,
but we shouldn’t get distracted. The experiences that interest us
here, regardless of how we might want to define religiosity and religious experiences in general, are
the experiences that could power an argument for the existence of God. Let’s call them God-experi-
ences. What are God-experiences? How do they compare to our other experiences? Are they reliable,
and if not, why not?
Each of the following sections begins with a suggestion as to how one might distinguish the
reliability of book-experiences from the reliability of God-experiences.
AVAILABILITY
Suggestion 1: When you look in the direction of this book, with open eyes, adequate light, and no
obstacles, you’ll see it. God-experiences, by contrast, are hard to come by. They’re fleeting.
Moreover, one cannot reliably predict when, where, or to whom they will occur.
The Hebrew Bible states, in God’s name: “You cannot see my face, for no person shall see me
and live.… You will see my back, but my face shall not be seen” (Exodus 33:20–23). If finite
human beings are going to catch a glimpse of an infinite and transcendent reality, it shouldn’t sur-
prise us that the glimpses will be fleeting.
More damaging is the claim that religious experiences aren’t equally open to everyone. They
tend to happen only to religious people and, even then, not at the same time as one another. In
the right environment, book-experiences will invade your eyes whatever you believe about books.
God-experiences, by contrast, are generally the preserve of religious believers. Doesn’t that render
them suspect?
No. Perhaps God-experiences are genuine but require a certain expertise. Not everybody is
equally sighted. Perhaps not everybody is equally sensitive to God. Training might help to make
people susceptible to religious experience. Attendance at ashrams or meditation retreats purportedly
provides training for religious experience. Perhaps God-experiences are only open to experts. That
doesn’t make them unreliable.
Should we trust the experiences of someone trained to feel the presence of God? Perhaps they’re
simply the deluded victims of self-hypnosis. Wine tasting provides an analogy. Barry C. Smith notes
that the sense of taste is able to sense objective features in a wine. It can do so reliably, even for the
novice. For example, “My tasting experience can tell me: whether a wine has too much alcohol
because of the slight burning sensation at the back of my throat” (Smith 2007, 47–48).
But experts taste more. Utilizing retro-nasal breathing—allowing tasters to smell what’s in their
mouth—and with heightened skills of selective attention, experts have a more fine-grained experi-
ence than amateurs.
Not everything about the taste of a wine is surrendered at first, or is accessible without a
skillful search. A great bottle will … reveal more if we take our time and let our experience
develop like a photograph.… Are these further judgments and assessments open to the novice?
Yes, but not without training. (Smith 2007, 50–51)
Expert wine-tasters can disagree. Smith cites a famous dispute between experts about a particu-
lar wine. There was a disparity in the verdicts of these two experts. But that disparity “tended to
obscure the level of agreement there was between them about the actual characteristics of the wine”
(2007, 76). The very qualities singled out by one for praise were criticized by the other. Another
example: the novice listener hears a melody against a pleasant background. The expert is able to
identify all of the individual lines of orchestra. Religious experiences are fleeting and not open to
everyone. This doesn’t entail that they’re not genuine.
The analogy between wine tasting and religious experience might be very tight indeed. Smith
(2007) borrows the language of epiphany to describe the sudden awakening of a wine-taster’s sensi-
bilities (for more on this comparison, see Luhrmann 2012, 60–62).
Suggestion 1 presents another problem for theists. We’ve conceded that religious experiences
are often the province of experts. Some think that God-experiences, and more generally, belief in
God, wouldn’t be this hard to come by, if God really existed. J. L. Schellenberg (2015, 103) pre-
sents the argument:
1. If a perfectly-loving God exists, then he would always be open to a personal relationship with all
people.
2. If there were such a God, no person would suffer a state of nonbelief in relation to the
proposition that God exists, unless they actively resisted.
3. Therefore: If a perfectly-loving God exists, no person, non-resistantly, fails to believe that God
exists (from 1 and 2).
4. Some persons have non-resistantly failed to believe that God exists.
5. Therefore: No perfectly-loving God exists (from 3 and 4).
6. If no perfectly-loving God exists, then God does not exist.
7. Therefore: God does not exist (from 5 and 6).
This argument makes some assumptions. The first appears at line 1. In Schellenberg’s words:
Imagine your friend … describing his parents: “Wow, are they ever great.… Granted, they
don’t want anything to do with me. They’ve never been around. Sometimes I find myself
looking for them—once, I have to admit, I even called out for them when I was sick—but to
no avail.… But it’s so good that they love me as much and as beautifully as they do!” … You’d
think he was seriously confused. And you’d be right.… They could have set their son up in the
best house in town, with money and things galore. But their attitude toward him … doesn’t
amount to the most admirable love. (2015, 41–42)
But perhaps God wants to ensure that our nonresistance to belief is motivated by the right consid-
erations, rather than fear of punishment, or social pressures (see Howard-Snyder 1996, 2015). Perhaps
we’ll have a richer relationship if we first of all have to take the plunge. Accordingly, God-experiences
shouldn’t be too easy to come by; they should be earned through the hard work of self-refinement.
Moreover, perfect-love doesn’t necessarily translate into wanting a relationship. Is God’s per-
fect-love like that of a parent for his or her child, as Schellenberg assumes or more like the love dis-
played by a great philanthropist or like the care that a good surgeon would have for his or her
patients? (Rea 2015) Wanting a relationship might even be a selfish and human way to love. I
wouldn’t deign to say what God’s love for us must be like. Line 1 isn’t obviously true.
Line 2 is also an assumption. Some argue that a relationship with God is possible without believ-
ing that God exists. We could have a relationship with God without realizing as much. Or we could
have a rich, explicit, conscious, and reciprocal relationship with God on the basis of hope or faith or
some sort of partial belief that God exists (Poston and Doughtery 2007; Cullison 2010). Schellenberg
replies that these are insufficient grounds for a real relationship (2015, 58). Perhaps line 2 survives
scrutiny. But it’s shaky. Consider: Bob meets Brenda on an online dating site. He is concerned that
she isn’t a real person. He knows that the site is populated by numerous automated chat-bots. Never-
theless, he perseveres in chatting with her for a long time, he likes her, and their relationship blossoms,

despite his doubt that she really exists. Eventually, his doubt dissipates. They meet in person. They
marry. At what point did this rich, explicit, conscious, and reciprocal relationship begin? Can we really
rule out the suggestion that it began even before Bob believed that Brenda exists (see Cullison 2010)?
Is belief really necessary, at all stages, of a rich relationship?
Line 4 is the least controversial assumption. Some argue that everyone believes in God, some-
times unwittingly, and sometimes under another name (Wainwright 2002). But if there’s even a
single case, in all of human history, of nonresistant nonbelief—such as a tribal Amazonian who
never heard of monotheism and failed to be a monotheist, without any sort of resistance—then line
4 stands.
At line 6, Schellenberg stipulates that perfect-love is part of the definition of God. That’s because
God is a person, and a perfect person would be perfectly loving. The great Jewish theologian Moses
Maimonides didn’t conceive of God as a person. That would have completely undermined his nega-
tive theology (his view that God defies discursive description—see Maimonides 2000, I.59). Does that
mean that Maimonides was an atheist? In Schellenberg’s idiolect, that’s exactly what follows. But that’s
not how the intellectual tradition of Abrahamic monotheism has evolved.
Historically, Abrahamic monotheism contains many different schools of thought, and Mai-
monides has been a central figure in the transmission and evolution of Abrahamic monotheism.
One stretches the meaning of “atheism” too far if one is willing to put Maimonides and other great
torchbearers of the Abrahamic religions, under its scope. Maimonides, and other negative theolo-
gians (i.e., thinkers who deny that we can discursively describe God’s essence) simply won’t accept
line 6, or the definition of “God” on which it’s based.
However, negative theology—the claim that God cannot be discursively described—cannot be
true. Even the description that God’s essence defies description is a description of the essence of a
God whose essence defies description. Accordingly, I find such figures of speech important and illu-
minating, but literally false—just as metaphors can be important and illuminating, but literally false
(Lebens 2014). Unlike Maimonides (at least if we take his words literally), I wouldn’t want to deny
that God is a perfect person who loves us.
Elsewhere, I present my preferred response to Schellenberg’s problem (see Lebens forthcoming).
The basic idea is that God’s perfection cannot be completely manifest within God’s creation. This is
a complicated claim based on ideas that I draw from Jewish mysticism, and its view that God, given
God’s perfection, had to constrain himself in order to create a world beyond himself. It follows from
this claim, that if a perfect God existed, we should expect to find God hidden in various ways. I
can’t defend my response here. I would need to devote a longer work to it. I must constrain myself!
But we’ve already established that Schellenberg’s argument stands on a number of assumptions,
some of which are somewhat shaky to say the least.
CROSS-CHECKING
Suggestion 2: You see the book. You touch it. You knock on it, hearing the sound. You ask another
person to take a look. Book-experiences are open to cross-checking. God-experiences are not.
Cross-checking assures us that we’re not mistaken about the underlying causes of our experi-
ences (Fales 2004, 152–153). Henry-Samuel Levinson and Jonathan Malino suggest that a percep-
tual experience provides only a small amount of evidence, insufficient for forming a justified belief,
unless that experience can be augmented by confirmatory intersubjective tests (i.e., by other people
taking a look) or by the justified assumption that such tests would be passed if attempted (1999,
305). To the extent that cross-checking isn’t available for God-experiences, we should conclude that
God-experiences cannot deliver justified belief in God’s existence.
God-experiences do, indeed, need to be augmented with cross-checks, and with background
networks of confirming beliefs. But it doesn’t follow that the same procedures used to cross-check
experiences of physical objects should be adopted to cross-check experiences of nonphysical objects.
Tanya Luhrmann wrote a fascinating study of the religious lives and practices of the members of an
American Evangelical movement, the Vineyard Christians. She sumises that various practices help to
restructure the minds of these Evangelicals in such a way as to lead them to have experiences that
suggest to them that God is real (Luhrmann 2012, 40–41). You have a God-experience. Was it real?
Invite another person to train as you have. Does he (or she) have it too? This would be a form of
cross-checking.
Objection: spiritual cross-checks represent courses in self-delusion.
Response: How do you know that they are any more or less delusional than the cross-checks
you use for your book-experiences? If you’re a brain in a vat, then all of your cross-checking about
books merely serves to feed your delusion. Admittedly, if God doesn’t exist, then all God-experi-
ences and crosschecks are deluded too. But is there a salient difference here between book cross-
checks and God cross-checks?
Objection: if we assume that books exist, cross-checking will be useful, because it gives us some
way of distinguishing genuine book-experiences from merely apparent book-experiences. But even if
God exists, cross-checking doesn’t tell our genuine God-experiences apart from our apparent ones.
Response: we’ll come back to this issue, in response to suggestion 7.
Objection: sensory cross-checking is more immediate and convincing than spiritual cross-checking.
Response: Jerome Gellman notes that “ordinary physical-object beliefs are … overjustified.…
We have extremely luxurious constellations of confirming networks [regarding those beliefs]”
(2001, 27). Perhaps our book-experiences provide our book-beliefs with more justification than
our God-experiences can provide for our God-beliefs. But our book-beliefs are over-justified to
begin with. Hence, it doesn’t follow that our God-beliefs are under-justified.
INTERSUBJECTIVE AGREEMENT
Suggestion 3: Other people who see the book agree with your descriptions of it. God-experiences
don’t generate this sort of widespread, intersubjective agreement.
Most God-experiences, in present times, are of a monotheistic God. This is true even in Hindu
contexts, at least as those experiences are understood by the Hindu intelligentsia (Radhakrishnan
1927, chap. 1). In fact, monotheistic belief is particularly well adapted to our cognitive architecture,
and thus, once arrived at, monotheism tends to survive and thrive in many social environments
(Barrett 2004, 75–107). So, we can first point out that God-experiences tend toward a monotheistic
consensus rather than extreme disagreement.
Moreover, Keith Yandell (2002, 85) talks of a generic God concept that can be abstracted from
the overlap between otherwise conflicting monotheisms. (Yandell is acutely aware of the vast variety
among religious experiences [see Yandell 1994]. But despite the variety, there is also important over-
lap among theistic-experience.) According to this generic concept, God is necessarily ontologically
independent (i.e., it is logically impossible that God’s existence depends on anything); God is a
self-conscious person; God is transcendent; and God is the most valuable being. Yandell claims to
have abstracted this generic concept from the overlap between mainstream theological traditions
around the world (see also Gellman 2001, 8–10).
Imagine that everybody agreed that they could see a bright object in the sky. Some said that it
was square, some circular; some that it was red, some orange. Nevertheless, everyone agreed to its
location, time of appearance, and to the fact that it was a bright object in the sky. If you saw it
as red, you might be minded to suspend judgment. After all, other people claim that it was orange.
But you’d still have good grounds to think that it exists, whatever its color.
Similarly, there are a wide variety of God-experiences. Nevertheless, they’re all (at least appar-
ently) experiences of a necessarily existent, self-conscious, transcendent, and supremely valuable per-
son, despite disagreement about whether this God was incarnate in Jesus, or in Krishna, or in nei-
ther of them. These God-experiences, taken collectively, still give us reason to believe that the
God described by the generic God concept exists.

Not all religious experiences are God-experiences. Should we be worried that nontheistic mys-
tical experiences undermine the deliverances of God-experiences? Nontheistic mystical experiences
sometimes conflict with God-experiences. Sometimes, they suggest that a nonpersonal being is
supreme or that all of being is supreme. These claims stand in tension with the supposed supremecy
of God. Nontheistic mystical experiences, when they do conflict with the generic God concept, tend
to suggest that all things are one (see Stace 1961). According to this mystical conception of the
world, there is no God, nor is there a world apart from God. There is just one great Being. To
the extent that mystical experiences suggest that all things are one (rather than the claim that all
things depend on the one God), I tend to be skeptical that they could possibly be genuine/veridical.
The claim that everything is one tends to collapse into paradox and absurdity (see Lebens 2017).
Other nontheistic mystical experiences, such as the meditative experience of profound aspects of
one’s own self, needn’t conflict with the claims of theism.
VIVIDNESS
Suggestion 4: Book-experiences are vivid and well defined. God-experiences are inchoate and ineffa-
ble.
We shouldn’t expect the content of God-experiences to be as neatly packaged as the content of
our book-experiences. For one thing, books have spatial boundaries. Moreover, suggestion 4 ignores
the variety among religious experiences. Some religious experiences are visual, embodied, and vivid.
People literally hear God speak to them and see visions.
Vividness aside, William James noted that religious experiences have a “noetic quality” (James
1908). According to James, an experience has a noetic quality when its content seems more real
than anything else. Even when religious experiences lack sensory vividness, they compensate with
this noetic quality.
When Richard Swinburne comes to delineate varieties of religious experiences, he includes
everything from the pyrotechnic visions of a Sinai-like experience, through the supersensory tran-
scendent experiences of the mystic, all the way to a general sort of seeming. It just seems to some
people that God exists (Swinburne 2004, 300).
Even these most unremarkable experiences can play a role in justifying belief in God. You’re
looking at this book, but your visual data alone don’t seem sufficient to prove that your experience
is genuine. You know that you could be dreaming. You know that you could be under the spell of
an evil demon. Nevertheless, it simply seems to you that you’re seeing a book. It seems to you that
you’re having a genuine-book experience. If a general seeming is sufficient for you to believe that
you’re genuinely seeing a book, then why can’t a general seeming suffice for belief that God exists?.
Objection: Just because you feel that something is true, it doesn’t mean that it is true.
Response: Thankfully, the suggestion isn’t that what feels right must be right. Rather it’s the
view that something seeming to be true is, all things being equal, evidence. Something seeming to
be the case doesn’t obviate the duty to scrutinize it and look for counterevidence.
If it seems to you that God exists, and if that seeming stands up in the face of scrutiny, even in
the face of formidable riddles, such as the problem of evil, then the fact that it seems to you that
God exists may well justify your believing that God does exist. (For more on what seemings might
be, see Moretti 2015. For discussion of these issues as they relate to theism, see Tucker 2011.)
George Berkeley believed that the world was mental through and through (Berkeley 2009).
Even this “physical” book Berkeley would treat as a bundle of ideas in the mind of God. Bertrand
Russell couldn’t find independent reason to deny this view (Russell [1912] 1998, 4). To Russell,
it simply seemed that the world contained material objects. Russell related to this “instinctive
belief,” as evidence (12). But things seemed different to Berkeley. Perhaps he had a different set
of instinctive beliefs. Mere seemings only tend to have evidential weight (if they have any evidential
weight at all) for the people who experience them directly.
When I seem to see an X, I have some evidence that an X exists. When I tell you that I saw an
X, you have some evidence that X exists too, even though you didn’t see it for yourself. You trust
my testimony. When it seems to me that X exists, I may well have some evidence that X exists.
But when I tell you that it seems to me that X exists, and if it doesn’t seem that way to you too,
then my seeming doesn’t give you any new evidence that X exists. Seemings, unlike sightings, tend
to generate evidence that cannot be passed on by testimony.
Suggestion 4 ignores God-experiences that are vivid and well defined. Moreover, we should be
open-minded as to the sorts of experiences that can carry evidential weight. At least for the people
that have them, even a seeming seems to do some work.
NATURALISM
Suggestion 5: We have naturalistic explanations of book-experiences. Light bounces off books and
hits our retinas, causing signals that stimulate visual systems in our brains. Books are part of
what cause book-experiences. We also have naturalistic explanations of God-experiences. They
do not include God in their analysis.
Worse still, some have suggested that God-experiences are the product of naturally occurring
pathologies. Gellman reports that religious experiences have been explained in terms of “hypersug-
gestibility, severe deprivation, severe sexual frustration, intense fear of death, infantile regression,
pronounced maladjustment, and mental illness” ([2004] 2017).
Russell remarked that “we can make no distinction between the man who eats little and sees
heaven and the man who drinks much and sees snakes. Each is in an abnormal physical condition”
(Russell 1935, 188). If mystical experiences are pathological, we should conclude that they offer no
justification to the beliefs that they generate. C. D. Broad responds: “One might need to be slightly
‘cracked’ in order to have some peep-holes into the super-sensible world” (Broad 1939, 164). Per-
haps we should expect veridical mystical experiences to be somehow pathological.
It’s also far from clear that all, or even most, religious experiences stem from pathologies.
Luhrmann (2012) examines the claim that her subjects, who report hearing God speak to them,
are mentally ill. She found that they were, on the whole, very well adjusted, highly functioning,
socially adept, and cogent adults. There were no independent markers here of mental illness.
But even if not pathological, God-experiences can be explained without recourse to God.
Contemporary neurological theory, for example, associates experiences of a transcendent unity with
“variations … in various structures of the nervous system, and lesser religious experiences with
mild to moderate stimulation of circuits in the lateral hypothalamus” (Gellman [2004] 2017).
God-experiences are entirely in the head.
But that’s unfair. We shouldn’t be surprised that naturalistic accounts of God-experiences don’t
refer to God. God is supernatural. Scientists, be they theist or atheist, tend to adopt a methodologi-
cal naturalism. What is methodological naturalism?
Methodological naturalism claims that science should act as if God and other supernatural enti-
ties don’t exist (see Clark 2014, 42–44). One reason for adopting this policy stems from an appre-
ciation of how theistic and supernatural explanations have a long history of being science-stopping.
You needn’t investigate what thunder and lightning are, and why they happen when they happen, if
you simply put it all down to warring Gods throwing bolts of fire around. Only when you take
supernatural posits out of the picture is science forced to provide better explanations.
On the adoption of methodological naturalism, however, we shouldn’t be surprised that neurol-
ogists find no sign of God as the cause of God-experiences. Methodological naturalism is question-
begging in this investigation. If the question under discussion is the nature of religious experience,
and if you want to discover whether or not such experiences are genuine, then it would be dishonest
to assume from the outset that supernatural causes have no role to play in the phenomenon under
discussion. Of course, under the assumption of methodological naturalism, you’ll discover no

evidence of divine intervention. You’ll discover nothing more than certain neuropsychological phe-
nomena, none of which involves God in any direct way; but that’s because you’re ignoring from the
outset the very possibility that God is involved.
The failure of naturalistic accounts of God-experiences to involve God in their analysis should
be singularly unsurprising. It provides no grounds for skepticism. Many neurologists will readily
agree. Neurologists simply aren’t looking for God. They’re doing something else (see, for example,
Newberg 2018).
At the other extreme, to those who think of mystical experience as pathological, there are
researchers in the cognitive science of religion who think that susceptibility to religious belief is hard-
wired into our cognitive architecture (Barrett 2004). These cognitive processes are primed to interpret
experiences as God-experiences with very little prompting. On the basis of this research, you could
conclude either: (1) God-experiences are nothing more than an idiosyncratic feature of our evolved
cognitive architecture; or (2) the God who loves us and wants us to come to know him ensured
that humans would evolve a cognitive architecture receptive to theism. Can you find a non-question-
begging reason to opt for one of these options over the other? If not, we’re left—at least—with the
somewhat funny conclusion that if theism is true, then God-experiences are probably reliable.
SENSE ORGANS
Suggestion 6: Our sensory experiences are facilitated by specific sense-organs with a specific and reli-
able function. God-experiences, by contrast, are not mediated by specially calibrated sense-
organs for detecting the presence of God.
Gellman writes:
There may not be out-of-brain “God-receptors” in the body, analogous to those for sensory
perception, which might reinforce a suspicion that it’s all in the head. However, out-of-brain
receptors are neither to be expected nor required with non-physical stimuli.… God … does
not exist at a physical distance from the brain. ([2004] 2017)
We shouldn’t expect out-of-brain “God-receptors,” but if our God-experiences don’t have their own
neurological circuitry, so to speak, then we might worry that they are mere by-products of other
functions. Justin Barrett can argue all he likes that our cognitive architecture is particularly hospita-
ble to religious belief, but if that’s just a by-product of the evolution of other, more biologically
important modules in our brain, then religious belief can, perhaps, be explained away. If, however,
there exists an independent “God module,” then this module might play a role in blocking any such
debunking argument.
Brain scientists are increasingly suggesting that mystical experiences activate unique neural
pathways. (For citation of scientific research pointing in this direction, see Gellman 2001, 33.) Gell-
man concludes, “That brain physiologists are beginning to discover unique brain activities or forma-
tions associated with mystical episodes must count in favour of their validity” (ibid.). Research
points toward something like a set of specifically dedicated God functions. The force of suggestion
6 seems undermined.
THE MARK OF THE GENUINE
Suggestion 7: After ingesting certain sorts of mushrooms, we know not to trust all that we seem to see.
The fact that we’re able to identify situations in which we have reason to discount our sense
experiences can help us to distinguish between apparent and genuine book-experiences. By con-
trast, we have no way to tell genuine religious experiences apart from merely apparent ones.
Suggestion 7 isn’t true. It assumes that we have no way of ascertaining when a religious experi-
ence is fake. But we do. For example, investigative journalism was able to ascertain that the apparent
religious experiences of American “faith healer” Peter Popoff were feigned and fraudulent (Randi
1989, 139–182). So much for other people. What about our own God-experiences?
Jonathan Kvanvig, a professor of philosophy at Washington University in St. Louis, once put it
to me: if God came to him and told him to sacrifice his son, he’d know that it was either a test or a
hallucination. Why? Because we’ve already learned, from the Bible, that God doesn’t want child sac-
rifice. We have canons of metaphysical, theological, and ethical reason against which we can assess
the likely veracity of our putative religious experiences. We have to arrive at a balance between
the vividness of a given experience, and the relative strength of our background convictions.
More generally, William Wainwright presents a number of tests (1981, 86–88). One test has it
that the “consequences of the experience must be good” for the subject (Wainwright 1981, 86).
Gellman explains: “A person’s coming out of God-experiences propelled toward an evil, egocentric
life, would count strongly against the authenticity of the experience” (2001, 32). A second test, sug-
gested by Wainwright, is that a person’s God-experiences should prove “fruitful and edifying” for
others (1981, 86–87). A third test demands that the experience give rise to profundity. Gellman
explains: “The insignificance or inanity of what the mystic says counts against authenticity”
(2001, 32–33). As Gellman rightly points out, all three of Wainwright’s tests have been clearly
passed by some mystics and clearly failed by others.
A GENERAL ANALOGY
Classical foundationalism states that a belief can be justified only if it is entailed by other justified
beliefs or if it has a special status, a status that allows it to be a foundation in our system of beliefs,
a status such as infallibility, incorrigibility, or indubitability. Alvin Plantinga (2000) notes that clas-
sical foundationalism cannot license our ordinary beliefs about the external world. For example: our
book-beliefs are derived from our book-experiences, but our book-experiences are not infallible,
incorrigible, or indubitable, nor are our book-beliefs derived from any other firm foundation.
Instead, we tend to reject classical foundationalism. We allow that our sensory beliefs are properly
basic—that is, they are generally in need of no further justification.
Analogously, Plantinga suggests that we have a faculty—which John Calvin called the sensus
divinitatis—a faculty for sensing God. Our sense-experiences are mediated by sense-organs accord-
ing to a design plan that tends to be reliable. For example: the human eye tends to get things right,
providing that it finds itself in a hospitable environment. It runs according to a reliable design plan.
It gives rise to properly basic beliefs. If God exists, then we have every reason to believe that our sen-
sus divinitatis, whether or not it has its own neurological wiring, was designed by God, since God
might want us to know him. If God exists (granted: a big “if”), then our properly basic belief in
him is warranted; no more and no less than our book-beliefs.
William Alston (1991) develops a general theory of doxastic practice (i.e., belief-forming prac-
tices). Some of these practices can be justified in terms of other doxastic practices. The practice of
science, for example, can be justified in terms of the practices of sense-perception and reasoning.
But many doxastic practices cannot be justified without circularity. Famously, for instance, induc-
tion (i.e., reasoning from historical patterns to predictions about the future) can be justified only
on the basis of induction itself (since using induction generally worked well in the past, induction
advises us to predict that it will work well in the future).
According to Alston, our only justification for continuing to rely on these circular practices is
somewhat pragmatic. They have proven themselves to be well behaved, since they haven’t produced
massive, unavoidable contradictions on central matters, nor have they clashed in any major way with
other equally well established practices.
With this picture in place, Alston goes on to argue that the Christian practice of belief-forma-
tion (and, by extension, the doxastic practice of any religion that doesn’t give rise to massive contra-
dictions) has all of the relevant features of a justified doxastic practice. To undermine Alston’s argu-
ment, you would have to argue that his criteria for the rationality of a doxastic practice are too
permissive—but you’d better be careful not to make your ordinary book-beliefs irrational too.

POSSIBLE CONCLUSIONS
I offer a number of alternative conclusions from which to choose. Starting with the most ambitious,
I’ll dial it down notch by notch.
Conclusion 1: God, as described by the generic God concept, exists. God-experiences are a wide-
spread feature of human life, whether or not you’ve had one yourself. You believe that this
book exists on the basis of your book-experiences, and you would believe that it exists even
had you not seen it yourself, on the basis of another person’s testimony that he or she experi-
enced it. So, too, you should believe that God—as described by the generic concept—exists,
either on the basis of your own God-experiences (if you have any), or on the basis of other
people’s. We have found no good reason to make any relevant distinction between God-
experiences and book-experiences, so as to block this argument, unless we can find some
independent proof that God doesn’t exist.
Conclusion 2: We have found reason to be somewhat less trusting of God-experiences than we are
of book-experiences. The types of cross-checking available for God-experiences are less direct
and less compelling. Nevertheless, God-experiences—whether we’ve had them personally or
heard of them via testimony—still provide sufficient warrant for believing in God—as
described by the generic concept, less warrant than we have for believing in books, but suffi-
cient warrant notwithstanding.
Conclusion 3: God-experiences provide such a reduced level of justification for God-beliefs that
they cannot justify, on their own, belief that God exists. Nevertheless, God-experiences still
provide evidence that God—as described by the generic concept—exists, evidence that’s simply
insufficient for belief. God-experiences are a start. They need to be supplemented with further
evidence.
Conclusion 4: God-experiences have only the evidential weight of a seeming. If it seems to you that
God exists, and you can find no equally compelling evidence to suggest that God doesn’t, then
you might have good reason to believe that he does exist—at least as he is described by the
generic concept. But seemings carry weight only for the people that have them.
Conclusion 5: If God exists, then God-experiences are likely to be epistemically analogous to our
everyday sensory experiences. If, however, God doesn’t exist, then they’re certainly not reliable
and can easily be explained away in terms of some sort of pathology or evolutionary accident.
Conclusion 5—the weakest of the options—still concedes a lot to the theist. It grants that the
theist is warranted in believing that God exists—that theists act with due rationality when they give
weight to their religious experiences—so long, that is, as God exists. The theist likewise has to
accept that if God doesn’t exist, his or her God-experiences were delusional. But on the off chance
that God does exist, then the atheist will have to accept that he or she lacked a certain sensibility,
leaving him or her tone-deaf to the genuine music of religious experience.
Acknowledgments. The chapter was made possible through the support of a grant from Templeton
World Charity Foundation, Inc. The opinions expressed in this chapter are those of the author and do
not necessarily reflect the views of Templeton World Charity Foundation, Inc.
Bibliography
Alston, William P. Perceiving God: The Epistemology of Religious Experience. Berkeley, George. Principles of Human Knowledge and Three Dialogues.
Ithaca, NY: Cornell University Press, 1991. Edited by Howard Robinson. Oxford: Oxford University Press, 2009.
Barrett, Justin L. Why Would Anyone Believe in God? Walnut Creek, CA: Broad, C. D. “Arguments for the Existence of God, II.” Journal of Theological
AltaMira Press, 2004. Studies 40, no. 2 (1939): 156–167.
Clark, Kelly James. Religion and the Science of Origins: Historical Newberg, Andrew. Neurotheology: How Science Can Enlighten Us about
and Contemporary Discussions. New York: Palgrave Macmillan, Spirituality. New York: Columbia University Press, 2018.
2014.
Plantinga, Alvin. Warranted Christian Belief. Oxford: Oxford University
Cullison, Andrew. “Two Solutions to the Problem of Divine Hiddenness.” Press, 2000.
American Philosophical Quarterly 47, no. 2 (2010): 119–134.
Poston, Ted, and Trent Doughtery. “Divine Hiddenness and the Nature of
Fales, Evan. “Do Mystics See God?” In Contemporary Debates in the Belief.” Religious Studies 43, no. 2 (2007): 183–198.
Philosophy of Religion, edited by Michael L. Peterson and Raymond J.
Radhakrishnan, Sarvepalli. The Hindu View of Life. New York: Macmillan,
VanArragon, 145–148, 152–153. Oxford: Blackwell, 2004.
1927.
Gellman, Jerome. Mystical Experience of God: A Philosophical Enquiry.
Randi, James. The Faith Healers. Updated ed. Amherst, NY: Prometheus
Aldershot, UK: Ashgate, 2001.
Books, 1989.
Gellman, Jerome. “Mysticism.” In Stanford Encyclopedia of Philosophy, edited
Rea, Michael C. “Hiddenness and Transcendence.” In Hidden Divinity and
by Edward N. Zalta. Spring 2017. https://plato.stanford.edu/archives/
Religious Belief: New Perspectives, edited by Adam Green and Eleonore
spr2017/entries/mysticism/.
Stump, 210–225. New York: Cambridge University Press, 2015.
Howard-Snyder, Daniel. “The Argument from Divine Hiddenness.”
Russell, Bertrand. Religion and Science. London: Oxford University Press,
Canadian Journal of Philosophy 26, no. 3 (1996): 433–453.
1935.
Howard-Snyder, Daniel. “Divine Openness and Creaturely Nonresistant
Russell, Bertrand. Problems of Philosophy. Oxford: Oxford University Press,
Nonbelief.” In Hidden Divinity and Religious Belief: New Perspectives,
1998. First published 1912.
edited by Adam Green and Eleonore Stump, 126–138. New York:
Cambridge University Press, 2015. Schellenberg, J. L. The Hiddenness Argument: Philosophy’s New Challenge to
Belief in God. Oxford: Oxford University Press, 2015.
James, William. The Varieties of Religious Experience: A Study in Human
Nature. New York: Longmans, Green, 1908. Smith, Barry C. “The Objectivity of Tastes and Tasting.” In Questions of
Taste: The Philosophy of Wine, edited by Barry C. Smith, 41–78. Oxford:
Lebens, Samuel. “Why So Negative about Negative Theology? The Search
for a Plantinga-Proof Apophaticism.” International Journal for Philosophy
of Religion 76, no. 3 (2014): 259–275. Stace, W. T. Mysticism and Philosophy. London: Macmillan, 1961.
Lebens, Samuel. “Russell and Bradley: Revisiting the Creation Narrative of Swinburne, Richard. The Existence of God. 2nd ed. Oxford: Oxford
Analytic Philosophy.” Journal for the History of Analytical Philosophy 5, University Press, 2004.
no. 7 (2017): 1–20. Tucker, Chris. “Phenomenal Conservatism and Evidentialism in Religious
Lebens, Samuel. “Revelation through Concealment: Kabbalistic Responses Epistemology.” In Evidence and Religious Belief, edited by Kelly James
to God’s Hiddenness.” European Journal for Philosophy of Religion Clark and Raymond J. VanArragon, 52–73. Oxford: Oxford University
(forthcoming). Press, 2011.
Levinson, Henry-Samuel, and Jonathan W. Malino. “Who’s Afraid of a Wainwright, William J. Mysticism: A Study of Its Nature, Cognitive Value, and
BEE STING?” Iyyun: The Jerusalem Philosophical Quarterly 48 (1999): Moral Implications. Madison: University of Wisconsin Press, 1981.
293–326. Wainwright, William J. “Jonathan Edwards and the Hiddenness of God.” In
Luhrmann, Tanya M. When God Talks Back: Understanding the Divine Hiddenness: New Essays, edited by Daniel Howard-Snyder and
American Evangelical Relationship with God. New York: Vintage Paul K. Moser, 98–119. New York: Cambridge University Press, 2002.
Books, 2012. Yandell, Keith E. The Epistemology of Religious Experience. Cambridge, UK:
Maimonides, Moses. Guide for the Perplexed. Mineola, NY: Dover, 2000. Cambridge University Press, 1994.
Moretti, Luca. “Phenomenal Conservatism.” Analysis 75, no. 2 (2015): Yandell, Keith E. Philosophy of Religion: A Contemporary Introduction.
296–309. London: Routledge, 2002.

TOPIC 5
Religious Experience: Atheism

John R. Shook
Lecturer in Philosophy, College of Arts and Sciences
Bowie State University, MD
The aim of this chapter is to clarify what is meant by “religious experience,” discuss the ways in which such
experiences can be explained, and assess what kind of evidential support they offer to either theism or
atheism. Atheism uses a variety of arguments concluding that religious experiences cannot justify the idea
that a God is involved.
ATHEISM ON RELIGIOUS EXPERIENCE
Experiences about God seem objective enough to those convinced that God is present. There is
nothing about atheism requiring it to dismiss or distort reports of such experiences. This chapter
describes atheism’s position on religious experience, and defends the conclusion that features to reli-
gious experiences that seem unnatural do not justify the view that something supernatural is
involved.
Atheism would have little to say about religious experiences if theism never took them for
encounters with God. Where theism does so, the atheist position is that there is nothing supernat-
ural about religious experience. Atheism denies that anything from human experience is sufficient
for reasonably affirming the supernatural or God. There is no reason for atheism to deny experience,
however—it is experience, not gods, that humanity cannot live without. How inspirations from reli-
gious experience motivate people to change their lives and change the world is evident to atheism, as
it is to religion. If religious experience were entirely private, religion would not foster it, and theol-
ogy would not appeal to religious experience if nothing could be conveyed about it. In truth, reli-
gious adherents can find ways to disclose what their religiosity is like. Atheism raises its skeptical
objections when claims are made that God is experienced. This chapter’s justifications for that skep-
ticism apply to any theistic views, whether the label of “God” is used or not.
This chapter discusses religious experience as religions view it. Then atheist and theist perspec-
tives on religious experience are compared, focusing on implications for God’s reality as theism pro-
jects it. The difficulties attendant to the study of religious experience are considerable (Bush 2012),
but atheism can discuss the topic if theology can. Nothing approaching scientism or reductive nat-
uralism is presumed, although relevant scientific research has a small role. Religious experience can-
not be dismissed, and atheism can appreciate why it is taken seriously. To endure over time, a reli-
gion must lend its interpretations to ensure that special experiences have religious significance for
human followers. An actual deity’s involvement is not necessary.
OPENNESS TO RELIGIOUS EXPERIENCE
The gods are all unknown to atheism, so it can entertain all claims about divine encounters equally,
regardless of who has them. This atheist neutrality is not a disadvantage. Religions prejudge which
147
religious experiences are authentic and which are not. It would

KEY CONCEPTS be odd for atheism to display such partiality. Atheism never
presumes that only particular experiences about one or another
deity are plausible enough to warrant its skepticism. The com-
Abductive Arguments
plaint that atheism cannot be impartial toward religious experi-
Apophatic ence must be misdirected. It is characteristic of religions, not
Attributes of God atheism, to lend special credence to one or another mode of
religiosity.
Deductive Arguments
Atheism does not deny that religious experience has long
Direct Experiential Theism (DET)/ been prevalent and persuasive. Academic accounts of religion’s
Indirect Experiential Theism (INET) origin and proliferation often include religious experience to
Doxastic Practice explain how Homo sapiens developed its religious proclivities
Experiential Mysticism (EM) (Geertz 2013; Newberg 2017; Sidky 2017). Religions favor
their own origin narratives, so supernatural sagas are preferred
Inductive Arguments over natural accounts. Atheists have their own myth that sci-
Universal Consent Argument for God ence overcomes religion only with atheism’s aid. However,
fields such as cognitive science and cultural anthropology ably
uphold their methodologies and theories about religion (Pyy-
siäinen 2013; L. Martin and Wiebe 2017). The sciences rightly
set aside both atheism and theism while undertaking their independent investigations. Whether sci-
entific approaches can help demystify or debunk religious claims about deities is a separate matter
for philosophical and theological debate. Theology offers its antagonism or arbitration, either attack-
ing science’s competence to study religion, or integrating religious views with science. Atheism has
little to do with either option. Science can defend itself against conservative theology without atheist
cheerleading. As for speculative theology, finding ways to modify dogma to cohere with knowledge
is commendable, but creative theologies head in different directions.
No clear and consistent position on religious experience comes from theology. Theologians
disagree as vehemently with each other over religious experience as they do with whatever athe-
ism says about it. Those controversies often sound sharper within a religion than between differ-
ent religions. Whenever a theologian gripes that “atheism just won’t get religious experience
right,” we can hear the frustration of yet another minority view, unable to speak for a denomina-
tion much less an entire religion. There is no religious consensus that atheism has been blindly
ignoring. A theologian presumptively explaining to the world’s religions how their religious
experiences are essentially those of that theologian’s own faith is exhibiting close-mindedness,
not teaching atheism a lesson (e.g., Stace 1960, chap. 2; Davis 1989, 186–192). Atheism can
instead take an open-eyed perspective over all modes of religious experience in whatever manner
religions encourage them.
Openness to religious experience is not acquiescence. These experiences are not just a step or
two away from justifying theism. The reality of an experience of God is never equivalent—logically,
epistemically, or ontologically—to an experience of the reality of God. Because they are not equiva-
lent, successfully encountering God is not guaranteed by anyone’s experience about God. Four main
reasons for that nonequivalence are pertinent.
First, devout conviction is one thing, while reasonable belief is another. One’s certainty does
not guarantee correctness. The proposition that “person P holds belief B about God as securely as
any belief could be, and there is no God” does not state a contradiction or an unreasonable judg-
ment. The way that an individual will not come to doubt an experience, a belief from an experience,
or a belief about deity, can tell us something about human tenacity but nothing about godly acces-
sibility. Put another way, the test of reasonableness cannot be whether a person will admit being
unreasonable. Beliefs are usually acceptable until shown to be unreasonable, in principle. Yet this
charitable treatment of belief does not mean that someone’s conviction stays reasonable so long as
the believer stubbornly rejects all reasonable doubts. And there are reasonable doubts to be directed
at claims of encountering God. Religion understands that as well as atheism (Schellenberg 2007,
chap. 8). Each religion can be pointedly skeptical, as skeptical as atheism, toward encounters with
deities of other faiths, although a religion exempts its encounters from doubt. Atheism thinks that it
is even more reasonable to allow no exceptions.
Second, seeking a place where doubt cannot reach religious experience is a retreat into subjec-
tivity. A place beyond the reach of reason and religion would be a private space, where one is an
unimpeachable authority over one’s experiences. One could say, “I hold my belief about my God
as firmly as any belief could be, and my God is real.” Belief grounded in experience is compelling,
and faithfully expressing that belief sounds sensible to oneself since it is practically a tautology.
While stating one’s conviction in a deity, one would not admit its unreality. However, verbal tautol-
ogies cannot establish the independent reality of anything, so that leaves just a belief about subjec-
tive experience. That subjectivity is not compelling to either atheism or religion, but for different
reasons. The subjectivity of religious experience does not compel an admission from atheism that
an objectively real God is involved. As for religion, such subjectivity is inadequate, since theism
seeks an objectively real God, not just human experiences. Subjectivity has its own kind of author-
ity, but it cannot support theism.
Third, perceiving a deity could not be similar to perceiving the world, and religions have always
said so. Still, associating religious experience with perception might lend the appearance of reason-
ableness. Richard Swinburne says, “it is a principle of rationality that (in the absence of special con-
siderations) if it seems (epistemically) to a subject that x is present, then probably x is present: what
one seems to perceive is probably so” (1979, 254; consult Everitt 2004, chap. 8). How many deities
have appeared to credulous people around the world? Theism defends the reality of God, not all of
humanity’s innumerable gods. Swinburne would object that taking almost all phenomena to be real
is not what he means by reliable perception. As everyone knows, perception is often untrustworthy
without consulting common sense, and the object of perception should be familiar to anyone with
sound senses (M. Martin 1990, 174–177). Swinburne’s sensible objection derails his application of
his principle to theism. Where God’s presence is concerned, surely special considerations pertain, so
Swinburne’s principle cannot assist theism.
Fourth, the means of perceiving a god could turn out to be not so reliable. Perceiving objects
around us is easy for us. God would be regarded as just another worldly matter if perceiving God
was not difficult for much of humanity. A perception of ordinary matters cannot be godly; an expe-
rience of godly matters cannot be ordinary. Theologians never tire of scolding atheists for taking
God to be just another object. Yet Swinburne trusts the perception of objects and of God, and he
threatens general skepticism about the world unless his principle holds for both. Skepticism about
the external world does not follow from doubting alleged encounters with the supernatural world.
Everyone shares observations of evident objects around us, using bodily senses that work in under-
stood ways. If a perception of God bypasses bodily senses, how would extrasensory perception (an
ability with a poor reputation) permit a godly encounter? Theism suggests that divine powers are
at work, which begs the question. As for religions, they can only say that unpredictable gods reveal
themselves to whomever they please.
(For criticism of William Alston’s defense of the claim that a group can know that God has
contacted them, see Topic 4 “Doxastic Foundations: Atheism.”)
RELIGIOUS EXPERIENCES ARE COMMUNALLY OBJECTIVE
In the absence of theological consensus on religious experience, atheism concurs with science and
religion: religions are sustained where their practices enrich the experiences of committed partici-
pants (Boyer 2001, chap. 7; Bellah 2011, chap. 1; King 2017, chap. 1). Religions craft distinctive
practices referring to extraordinary beings, objects, and situations (Atran 2004, chap. 10), but such
matters are too heterogeneous to help define religion further. In any case, religions mainly focus on
their adherents. If religiosity is not central to engaged experiences of devout practitioners, there is
nowhere else for it to be.
The term “religious experience” came into use in the nineteenth century, but theologians do
not quibble over whether Augustine or Martin Luther had religious experiences (Gavrilyuk and

Coakley 2011), even if Zen meditation or indigenous hallucinogens complicate their classifications.
For anthropology, prehistoric humans were having religious experiences by many means, just as they
had aesthetic experiences with their art though “aesthetics” is a modern term. Theology gets preoc-
cupied with sorting out who has genuine religiosity, but those prejudgments distort empirical meth-
odology and detract from the search for religion’s origins (Shook 2017). A generous conception of
religious experience is more useful for the history of religion and the inquiries of science.
A religious experience is, at a generic minimum, an unusual experience of a person who ascribes
religious significance to it. Any number of atypical experiences can be candidates for that religious
interest. No list could be exhaustive. Nor would agreement converge on which experiences are
essentially religious in import. It is possible for two people to have the same sort of uncommon
experience but for only one of them to experience it religiously. It is a mistake to think otherwise.
Supposing that some experience E, so long as one person takes E as religious, necessitates that all
other people having E must have E in that same religious way, is a presumption that defies common
sense and contradicts the way that religions operate (consult Taves 2011, chap. 1).
Commonsensically, a person can grasp how an experience is quite unusual without simulta-
neously taking it to be religious in some sense or another, simply by attending to that experience’s
peculiar manifestations. The appearance of that experience is not, by itself, sufficient to guarantee its
significance as religious for each and every person who has it. Different people can have the same
sort of odd experience without agreeing on what its religious import must be, or even agreeing that
it has any religious significance. People can reflect back on an unusual experience to arrive at a real-
ization of its religious significance, a significance not occurring to them originally. People can also
assign one sort of religious significance to an odd experience at first, and then later change their
minds to assign a different religious meaning, or perhaps conclude that it actually lacks any religious
sense at all. These commonplace events would be impossible if an experience had to be religious in
the same way for anyone and everyone who has it, no matter their situation or preparation.
Religions agree with common sense here. Each religion offers guidance about selected kinds of
unusual experiences so they are understood with a specified religious significance. Suppose that
everyone who has a certain unusual experience (call it S) already automatically grasps it as reli-
gious—prior to any religious instruction or indoctrination—in one specific way (call it R). (Obvi-
ously, if one religion could dominate the rest to impose its singular religious interpretation on every-
one, this could be accomplished.) On this strange supposition, religions trying to instruct those
people in the religious import of S would be ineffective. People would either already agree with that
religion on R, so no instruction about S would be needed, or people would disagree, and religious
instruction could not change how people intuitively take S to be R. Either way, religions would
be unable to guide people toward preferred understandings of religious experiences. Yet this is not
the world of devout adherents and potential converts that religions expect to operate in. No religion
would accept a definition of religious experience that renders its religious guidance futile, including
theistic religions. A monotheistic religion about a unique God revealing himself to humanity may
expect that special sui generis religious experiences are unequivocally and irresistibly compelling to
anyone. Yet that religion will also reserve the right to confirm or deny reports of encounters with
its deity, checking for doctrinal consistency and offering its preferred interpretations. Religious con-
formity is a higher priority than experiential purity.
An unusual experience of a person, who regards it as having religious significance in some sense of
another, is a religious experience. A propensity to religiously appreciate an unusual experience differs
widely across humanity, and people may or may not be aware of their preparedness for assigning a
particular religious significance. Socialization supplies much implicit religious preparation. A person
native to Italy, Guatemala, or Alabama is more likely to report a religious encounter with Jesus or
Mother Mary than with a Hindu deity. The reverse is more likely for a person who grew up in
New Delhi or Kolkata. It is said that faith is needed to see. However, conviction is not always
required. Prior socialization where religion has influence, perhaps paired with religious guidance after
the experience, will suffice. Unusual experiences can happen to anyone, but religious experiences impli-
cate a group. A religion can recognize that someone has had a religious experience without checking
for church membership first, but churchly criteria are applied to that experience nonetheless.
Strictly speaking, no one has ever had a religious experience entirely on his own, even if one is
alone when it occurs. No denial of individual creativity is intended. Novel interpretations of reli-
gious experiences are offered by people modifying their ideas of religiosity. No difficulties need to
erupt over who had the “first” religious experience, either. Anthropology, psychology, and cognitive
science can agree that humans (and probably some hominids before Homo sapiens, such as Homo
erectus) have always been susceptible to an extremely wide variety of unusual experiences. Some of
them were eventually selected for relevance to religious ideas, as those ideas congealed and crystal-
lized into the cultural forerunners of what later became classed as religions. Religious experience is
not the only source of inspiration for religious ideas, so no paradox arises from supposing that reli-
gion shaped religious experience. No argument over “What was the first religious experience like?”
or “What was the first religious belief?” has to be resolved.
Individual religious experience could never have been the exclusive basis for religiosity. All sorts
of important events in Paleolithic existence, from the joys and dangers of childbirth to the exhilara-
tions and disasters of the hunt, and all of life’s stages and dramas right to the grave end, were mate-
rial for creative interpretation and idealization. The early religious life was lived out there, in the
wilds and shelters of nature, not within the inner theater of the mind. The peculiar significance of
trances and dreams arose from the way that cherished real-world things, from elders and big game
to mountaintops and full moons, could be encountered again by way of an inner journey through
memory and imagination. The subjectively internal side to religious experience always followed
the objectively external side, never the reverse. Even the blankest, most formless emptiness ascribed
to some mystical states can acquire religious import because those states lack familiar worldly
features.
Religious experience is basically objective in significance as well as in provenance. Only from
that basis could religious people help others interpret their new experiences as religious. Entirely
subjective experiences of two people can only be incomparable and noncommunicable. Religiosity’s
orientation of objectivity is evident in the way that veteran practitioners of a religious tradition can
assess whether another person had an authentic religious experience or not. Even with mysticism, no
one is treated as a genuine mystic by staying mute.
The objective standards of a tradition’s practitioners supply the tests for conformity. As the
novice joins the tradition, practices and ideas are provided to shape the resulting religious experi-
ence. A Christian can say, “Yes, that was your sense of guilty depravity, this is a welcome sign that
grace may come.” A Hindu can say, “Yes, that was the dropping away of all external sense, this is an
absorption into pure consciousness.” And so on. Religious communities provide that sense of mean-
ingful objectivity to their religious experiences. No gods are made real by this social process. That
communal objectivity inheres in the way that no individual religious person is responsible for assign-
ing or sustaining the religious meaning and significance of experiences promoted by a religious
community. Describing a religious person’s experiences within a religious community as simply
subjective is untrue to what those experiences are like and how those experiences are possible.
RESTRICTIONS ON CRITIQUES OF RELIGIOUS EXPERIENCE
Some observations about criticizing religious experience follow from the previous section’s points
about objectivity. First, an attempt to dismiss all religious experience by simply classifying it as
entirely subjective takes a mistaken view of religious experience. Second, an effort to discredit all
religious experience on the grounds that any religious experience is just a subjective manifestation
of a brain’s functioning, or a brain’s malfunctioning, will not succeed. Third, an argument trying
to invalidate all religious experience on the grounds that (a) religious people do not assign similar
religious significance to unusual experiences, or (b) people around the world do not assign the same
religious significance to the same experiences, is a mistaken tactic for the reasons explained at the
beginning of the previous section. Fourth, a tactic to dismiss all religious experience because reli-
gions prefer different kinds of religious experiences only misunderstands the bases to religious
experiences. That diversity is not equivalent to irrelevance, since communal objectivity at most

should be expected, not universal convergence. Uniformity and convergence are appropriate criteria
for evaluating whether a real God is present in religious experiences, or for judging the truth of pro-
positions based on religious experiences (such as revelatory messages), but diversity should be
expected in the experiential field of religious significance.
Atheism should accept these four observations on illegitimate criticisms of religious experi-
ence. So should any religious viewpoint—no religion can invalidate the religious experiences of
another religion with those four tactics. One religion can deny the legitimacy of another religion’s
experiences for failing to conform to its own preferences, because legitimacy in this context just
means “belonging to our religion.” But no religion has grounds to assert that other religions fail
to foster religious experiences. Communal objectivity is inherently hospitable to religious plural-
ism. If atheist critiques are based on isolating religious individuals or demanding experiential con-
sensus, religions rightly complain. Fortunately, nothing about atheism requires it to make those
mistakes, and theism should avoid them as well. The models for experiencing God offered by
Swinburne and Alston, discussed above, are failures not because atheism presumed subjectivity
or demanded unanimity, but instead because those models fail theism by collapsing into subjec-
tivities and disunities.
The second observation made above, that brain science cannot confirm or deny beliefs from
religious experience, requires expansion. Atheism cannot deny that religious experiences occur, but
it would expect any experiential episodes to correlate with ongoing neurological processes. (Atheism
does not entail reductive naturalism, but atheism endorses psychophysical parallelism to forbid
“souls” from supernaturally violating the conservation of energy.) The way that unusual brain states
occur when people report unusual experiences satisfies that scientific expectation. The atheist argu-
ment, “Religious experiences are just brain states so they are only subjective,” is fallacious because of
the ambiguity of “subjective.” Does “subjective” mean “internal” or “mental”? Pointing to a locali-
zation for experience within the brain cannot rule out an objective God’s involvement, since all
experiences are paired with brain events. If atheism expects brain science to prove that religious
experiences have only subjective content and significance, atheism misinterprets science and misun-
derstands the nature of religious experiences.
Theism must not misinterpret neuroscience either. The theist argument, “Religious experiences
are at least brain states so they are objective,” is also fallaciously ambiguous, in this case about
“objective.” Theism cannot render a real God’s involvement plausible by pointing out that religious
experiences are more than just mentally subjective. If theism expects brain science to agree that the
simpler explanation for unnatural-seeming experiences is the presence of a deity, theism will be dis-
appointed by neurological hypotheses that need no God. That disappointing result is delivered in
the section below on the abductive position for direct experiential theism.
Neuroscience is revealing how cognitive processes lend religious experiences their intriguing
features. Even near-death visions, feelings of a presence, and out-of-body experiences are yield-
ing to neurological explanations (Blanke and Dieguez 2009; Sharpless and Doghramji 2015;
Blackmore 2017). However, spontaneous religious experiences are not invalidated just because
similar experiences can be induced through artificial stimulations, of a pharmaceutical or electri-
cal sort, for example. (For skepticism about inducing religious experiences, see Andersen et al.
2014.) A determination that a religious experience of God is a brain malfunction or an experi-
mental artifact has to presuppose, and does not prove, that no deity is involved. To argue, “No
God is present while the subject reports experiencing God, so that experience is just odd brain
activity,” starts with a nonscientific premise and begs the question against theism (Murray
2009). To instead say, “While the subject reports experiencing God, a certain pattern of brain
activity was occurring,” is scientific but neutral about the subject’s belief that God is encoun-
tered. Again, theism is neither supported nor refuted by the occurrence of neurological pro-
cesses in the brain.
In summary, religious experiences are more than just an individual’s neurology or mentality due
to their communal significance, but that is not evidence for God, or against God. Since it is the
unusual features of the religious experience that appear to indicate God’s presence for theists, athe-
ism must directly deal with those features, and later sections explore that engagement.
RELIGIOUS EXPERIENCES ORIENTED TOWARD GOD
If atheism predefines what counts as “religious experience” with much specificity beyond this chap-
ter’s generalities, it is fairly accused of ignorance or partiality by defenders of one religion or another.
Fortunately, atheism can simply let theism speak for itself. What are theism’s objectives for religious
experience? Among the many roles that it can play in a religious life, theism looks to religious
experiences for opportunities to get acquainted with God. Atheism only has to dispute the theistic
view that they access God.
When theism says that believers have experiences about God, atheism has no right to interpret
them as really being about something else. Religions enjoy a privileged perspective, because they
control both sides of religious experience. Within a religious community, what religious experiences
are supposed to be like has been coordinated with what they are supposed to be about. Religions get
into disagreements among each other over religious experiences on those grounds. A religion can
claim that another religion’s followers have strange experiences lacking religious significance, or that
their experiences are not oriented toward the right objective (they worship the wrong god, for exam-
ple).
We have observed how religious experience is basically objective in significance and in prove-
nance, and now we see that it is also objective in orientation. What makes a religious experience
meaningful is independent from any individual (that is its objective significance); what provides
content to a religious experience is not originally subjective to the experiencers (its objective prove-
nance); and what a religious experience guides the experiencer toward is beyond just the communal
context (its objective orientation). A religion generally expects religious experiences, among their
manifold purposes, to direct the attention of followers toward something more that the religion
and its practices. Specifically, theistic religions expect some religious experiences to be oriented
toward a deity.
The threefold objectivity of significance, provenance, and orientation characterizes important
religious experiences for religions, and atheism should concur. Arguments of a theological nature
could produce actual or hypothetical exceptions to this general viewpoint. For example, a theological
stance could say that some equally important religious experiences are significant only to the indivi-
duals having them, with content entirely aroused within the psyche’s privacy, and oriented just
toward the mundane field of human activity. Atheism takes no position on possible exceptions,
although it notes that emphasizing subjectivism looks like a theological retreat from advancing skep-
ticism. In any case, atheism has to deal with triple-objective religious experiences so long as theism
claims that they reveal the supernatural.
Atheism denies that religious experiences are actual encounters with a god or anything with
godlike attributes, including apophatic attributes. An “attribute” is anything attributed to God, pos-
itively or negatively, and conceptually or nonconceptually. Nothing more by “attribute” is implied
here, such as meanings from the term’s usage in the history of metaphysics (see Oppy 2014). If the-
ology affirms that what it means by “God” or “the ultimate” or “the supreme” (etc.) is not an
object, or an entity, or a being (etc.), such views do not evade atheism’s scrutiny, since atheism
denies anything theism offers to distinguish itself from atheism. Theological views rendering
“God” indistinguishable from nothingness do not distinguish theism from atheism, even if some
mode of devout religiosity is still sustained. Atheism welcomes the good company of extreme nega-
tive theologies, radical atheologies, and the like (Shook 2018, chap. 3).
A theistic religion endorsing religious experiences about the supernatural typically ensures that
the significance, provenance, and orientation factors all refer to it. For example, a theistic religion
centered on a singularly supreme deity will provide (i) a consensus on features of religious experi-
ence appropriate for appreciating that deity; (ii) an account about contents of religious experience
crediting that deity’s involvement; and (iii) an orientation of attention through religious experience
toward the accessibility of that deity. By contrast, a retreat from full objectivity instead leaves a con-
sensus on religious features to congeniality or chance, lets ideas about God arise from individual
imagination, and allows adherents to focus on their private religiosity. Atheism need not reject that
pseudotheism, as it effectively reduces deity and divinity down to human dimensions. Where

seekers are embracing warm emotions or shining values with enthusiasm, thinking about God on
their familiar terms, and reveling in spirituality for its own sake, it is even easier for atheism to
see how “God” is made for human appreciation. Let God be Love, Hope, Mercy, or Righteousness
(or Justice, Truth, or Beauty, and so on). That identification sounds altogether natural from the
nonreligious standpoint. The capacity of religion to imaginatively project human hopes and worthy
ideals out into a supernatural realm is one of atheism’s explanations for god-belief.
Let us return to theisms that challenge atheism with objective accounts of religious experi-
ences about God. They presume that “God” is meaningful enough to religious adherents. To give
two illustrations, God is “our Father who art in heaven,” or God “is the source and ground of
being and the wellspring of all consciousness, but also therefore the final cause of all creation,
the end toward which all beings are moved, the power of infinite being that summons all things
into existence from nothingness and into union with itself; and God manifests himself as such
in the ecstasies of rational nature toward the absolute.” The first quotation comes from Jesus
(Matthew 6:9) and the second is from David Bentley Hart (2013, 286). Concrete or abstract
ideas of God can all be meaningful. Theism, not atheism, assigns attributes to God, so no atheist
list of godly attributes could be drawn up. Anything attributed to God, no matter how abstract or
apophatic, is an attribute of God where theism makes that attribution. Taking God to be the
inexpressibly infinite ground of all being is hence no less an attribute in this basic sense than tak-
ing God to be male, or all-powerful. Atheism holds that no theological account of religious expe-
rience (however formulated) reasonably supports the involvement of a supernatural deity (however
indicated).
Theology promotes two primary modes of theism based on religious experience oriented toward
God: (a) it is God within that experience, and (b) something due to God is experienced. The first
mode is supposed to access God directly, in one or more of God’s own essential attributes. Alterna-
tively, if a religious experience is a sign of God or a display of God’s might, or a manifestation of
God’s love or righteousness, those are classed with the second mode. Two theistic positions on
experiencing God have to be considered separately by atheism:
Direct experiential theism (DET): A specifiable feature F of a religious experience (RE) is the
presence of God.
Indirect experiential theism (INET): A specifiable feature F of a religious experience (RE) arises
due to God.
This chapter does not explore how religions also gesture at what they suspect could be godly:
Experiential mysticism (EM): Nonspecifiable aspects of an indefinable experience indicate an
encounter with unknowable mystery.
EM is not a third theistic view of RE—this mysticism does not claim to be about God but only
mystery. Some theologians controversially presume that mysticism is inherently oriented toward
God. When a theist claims that a mystical experience has feature(s) revealing God, then DET or
INET is pursued.
No major challenge to atheism arises from EM. Atheism contradicts theism, not mystical expe-
rience or pure mysticism. Atheism grants the reality of mystery. Labeling utter mystery as “God”
does not make theism reasonable (Shook 2018, 139–141). If the mysterious is entirely unknowable
and/or ineffable, one cannot surmise that it is God, or even supernatural. The number of people
able to agree to a list of mystical qualities is irrelevant, although theological interest in such lists is
a lingering ghost of the universal consent argument for God (Zagzebski 2011). Any list at most
points to a shared psychological sensitivity, not a list of confirmed godly attributes (Jones 2016).
If “supreme reality” is supposed to be so indescribable, why would mysticism be upheld as a way
to get acquainted with a deity? If the mystery is cognizable or distinctive enough to suspect God’s
presence, then theism is pursuing DET or INET. Mysteries showing up in human experience pres-
ent no problem for atheism. Scientism may find mystical experiences indigestible, but atheism does
not in itself entail either scientism or reductive naturalism.
DET and INET require closer atheological examination. To cover broad argumentative
ground, we shall analyze deductive, inductive, and abductive arguments for DET in the next sec-
tion. The following section analyzes inductive and abductive arguments for INET. All REs in ques-
tion are assumed to possess that triple-objective character, unless otherwise indicated.
DIRECT EXPERIENTIAL THEISM
For DET, there is a pairing of some RE features with some divine attributes in REs, indicating an
identity in reality: feature F is simultaneously an attribute A of God. For example, if one’s RE is rel-
evantly “blissful” then God is bliss. God need not only be bliss, or blissful to the same extent, or
essentially bliss, but God itself has to include bliss. As another example, if the relevant experiential
feature is “timelessness” then God is timeless. To appreciate God so directly, no veil of conscious-
ness or a representational medium can intervene: it is not just human experience that is experienced,
but something of God is directly present. Subsequent reflection or communication about that
encounter will resort to mediating signs and symbols, but DET never confuses memories, concepts,
or thoughts about God with God’s real presence. Theists focused on psychological intermediaries
turn to INET instead.
Atheism should not argue that DET’s kinds of alleged REs are just subjective, or conceptually
confused, or that experiencers are incompetent witnesses. Nor should atheism argue that religion
does not accurately recount what such REs are “really” like. These DET experiences are, phenome-
nally, precisely what a DET-type theism thinks they are—the privileged authority in this realm of
RE is theism, not atheism. However, whether DET experiences can demonstrate that a theistic
God is actually present is another matter. DET must not get overconfident on its home ground
by saying, “God has attribute A, one’s RE has F, and A = F, so one is present with God.” Begging
the question is unreasonable. Three kinds of arguments are available to DET, taking the form of
deductive, inductive, and abductive inferences.
The first argument has this deductive form: “One’s RE has F; F is an attribute A that God
would have; so one’s RE has A of God.” This effectively concludes that one is present with God,
without begging the question. It does assume that, if God is real, at least one of God’s attributes
must be present in some REs (the authentic REs). However, this argument does not limit Fs to only
attributes of a God (other things could display those Fs too), and there is no preset limit to the
number of gods displaying religiously significant Fs. Assuming that only one God is the aim of this
theistic argument, atheism must ask, how could any theism know that some A is truly among the
attributes of God? If theism says, “We define God as having A,” that appeal to a verbal tautology
cannot establish the reality of anything. Each theistic religion conceives of God in its own way,
but no religion is entitled to define God for all humanity, as that presumption far exceeds the triple
objectivity inherent to each religious community’s interest in particular REs.
If there were successful arguments for God from grounds other than REs—such as ontological,
cosmological, teleological, and moral arguments—theology would possess a list of legitimate godly
attributes and then coordinate the objectivity of REs with that kind of deity. However, no argu-
ments for God, individually or cumulatively, can make God reasonably plausible (Shook 2018).
Besides, the point of DET is to demonstrate God through REs, without assuming that other ways
to know God have succeeded.
DIRECT EXPERIENTIAL THEISM’S INDUCTIVE POSITION
We next turn to DET’s inductive argument, in which DET does not try to predefine God’s attri-
butes but instead looks to the common features of REs had by many people. The basic form is,
“Many REs have special Fs; those Fs would not be prevalent unless they are God’s attributes; so
some As of God are experienced.” Atheism, along with religions, must first ask how theism could
correctly identify the most common Fs across humanity. Theology understands how a search for

shared features of God’s presence is a perilous quest. Theologians have an easier time identifying
fairly common psychological qualities to REs, such as a loss of self-boundaries or a feeling of con-
tentment, but DET needs attributes of God, not just human mental states. Discerning two or three
near-universal qualities of mystical episodes cannot show that God is real.
The DET inductive argument proposes that a few features to REs could be attributes of God, if
those features appear to enough people. Theologians depict this procedure as empirically trustwor-
thy, but theists would be right to question the results. Among Christians having REs, if 30 percent
have experienced one attribute of God found on a list of ten proposed attributes, must God have
that essential attribute? We would rightly wonder if theism today would willingly agree with the
results from such polling among Christians two hundred years ago, or five hundred years ago. All
the same, theology suggests that several godly attributes can be discerned in this manner. Among
those contemporary Christians having REs, suppose that ten attributes from a large list are recog-
nized by at least 10 percent of those Christians—can it be reasonably concluded that the real God
has all ten attributes?
Perhaps theism should consult other religions, too. If there were four vaguely worded attributes
for God, each one recognized by at least 20 percent of people around the world having REs, is that
God’s reality reasonably confirmed? This method would lose credence among theologians when
some of those attributes strayed from traditional views of the Judeo-Christian deity. Of course, if
Christian theologians controlled the wording and meaning of a short list of key attributes garnering
some global recognition, a recognizable God could come into view, but this procedure is no longer
an unbiased inductive method. Ultimately, theists do not take DET’s inductive argument seriously
where quite different religions are concerned.
Atheism adds that RE aggregative cases must logically collapse from the heavy evidential burden
placed on REs. Because theology has no list of godly attributes confirmed from other arguments for
God, all candidates for godly attributes come from RE itself. But only authentic REs can count: to
first attain that triple objectivity to REs, theology has to be selective about which features are rele-
vant to godly attributes. Naturally, theology tries to select only those features that have a fair chance
of matching genuine godly attributes. Typically, theologians select features/attributes that already
seem sensible to their denominational audiences, or they speculate about which godly attributes
seem plausible from their own theological reasonings. Inspirational theology and speculative theol-
ogy are insufficient at this stage of the discussion, however. DET cannot appeal to RE for yielding
the right features, while simultaneously asking those features to serve as confirming evidence for
God’s presence. Arbitrarily picking what counts as evidence E for X, and then claiming that X is
confirmed by that E, is a fraudulent procedure and presents an insurmountable aggregate problem.
There is no way out of this dead end, unless theology is prepared to confirm innumerable different
gods (extreme polytheism) or theology locates gods entirely within human experiences (extreme sub-
jectivism).
DET’s interest in special RE features as godly attributes shifts theology in the direction of poly-
theism or subjectivism without much push from atheism. Popular theology has led the way, telling
laypeople “God is inside you” or “find God within yourself.” This counsel is not saying God’s real-
ity is entirely within humanity. But DET must be asked this question: Is God only what is appreci-
able through authentic RE, or does God inherently possess additional attributes never in REs? This
question is not about accidental or incidental attributes—if knowledge that God is present in special
REs is DET’s objective, only important or essential attributes of God could count. Pointedly asking
the question again, if DET thinks that God is more than everything present in human REs, how
could DET know this, without appealing to other arguments for God? DET is forced toward an
admission that it cannot prove that appreciable gods have any reality outside of human experience
(so they no longer could be real gods).
DET could argue that some experienced attributes necessarily involve others, which need not
also be present in REs. If two godly attributes are supposedly related but only one is ever experi-
enced, DET would explain why the other one is never experienced. (Divine hiddenness arguments
are not relevant here, since DET asserts that God is not so hidden.) DET could claim that the pres-
ence of a set of godly attributes necessarily implies a metaphysical unity behind them, which is
God’s reality. Otherwise, a set of contingently related attributes leaves God in a fragmentary condi-
tion. However, DET asserts that God’s important or essential attributes are really present within
REs, not just God’s secondary traits evident in relation to REs (that alternative is INET). If those
essential attributes within REs lack a wholeness or a hang-togetherness, then DET cannot arbitrarily
insert the needed unification. An aggregate does not automatically indicate a whole, without more
metaphysical justifications than DET can provide.
Theology cannot evade these difficulties by adding claims about special RE features that
somehow self-identify as genuine godly attributes. Expecting features of RE to display their self-
evident and self-verifying status (consult Yandell 1994, chap. 8) is theology’s wishful thinking
that it bears no extra burden of justifying why some RE features are authentically evidential. Even
if some sort of epistemological intuitionism or foundationalism has merit in other areas of knowl-
edge, that hope is fruitless here. For DET, all REs in question are already utterly convincing to
experiencers—people uncertain about God’s presence do not present candidates for relevant REs
in the first place. People perfectly certain about God’s presence, by whatever feature(s) so deeply
impressing them, have already appreciated those features as self-evident and self-verifying, even if
they lack philosophical terminology to label them so. And all these people cannot all be entirely
correct, unless humanity has encountered innumerable gods (the polytheism problem), or REs
are hosting gods entirely within experience (the subjectivism problem). Theism must reject both
destinations. Therefore, a DET theologian must also claim, “Many people have what they think
are direct REs with seemingly self-verifying features, but most of them are mistaken about what
their REs are like.” This is simply another unjustified claim. Atheism—along with rival theologies
and all religions—are right to require more justification, even if epistemic window dressing
impresses the unwary.
DIRECT EXPERIENTIAL THEISM’S ABDUCTIVE POSITION
Finally, DET offers an abductive argument: “One’s RE has feature F; F is otherworldly; if F = God’s
A then F is otherworldly, so God has A.” Candidates for otherworldly features are at least as numer-
ous as religious notions about gods, since “otherworldly” is left ambiguous by arguments of this
kind. That ambiguity lends plausibility to this argument among theists, who insert their implicit
understandings. DET theists accordingly point to all manner of features of REs, making them
sound unlike anything of this world. (Keep in mind that the absence of something can be a feature
of RE too.) The second and third premises should be challenged.
The second premise’s plausibility requires theism’s accurate identification of truly otherworldly
features. How could any F be known to be otherworldly? A feature’s reputation for otherworldliness
is not enough, nor is its popularity with mysticism. Knowing that F is otherworldly requires a pre-
established delimitation for what is worldly. DET would have to first know everything about the
world, including humanity. Theology lacks such knowledge, so it imitates a philosophical inventory
of the world, or borrows one from science. The first strategy is not enough to establish theism, since
defining X as not worldly does not show that X is real. The second strategy picks out Fs in experi-
ence missing from science’s inventories, and proposes that those Fs cannot be explained by this
world, so they must be otherworldly. This strategy is contradictory. It assumes (wrongly) that sci-
ence already completely understands the world, while presuming (hastily) that science will never
fully understand human experience. Theology is just defining certain Fs in terms contrary to natu-
ralistic terms to make them sound forever immune from worldly explanation. Atheism can admit
that some Fs in REs currently appear to be somewhat inexplicable to today’s science. Although
some things seem “unworldly” for being unlike natural matters, theism cannot arbitrarily deem
any of them to be otherworldly as essentially supernatural. Theism should not presume to know
what all future science will comprehend and what it will not.
Something that seems unworldly to some people is not automatically otherworldly. Like all
experiences, REs involve the nervous system, a matter that theists have emphasized (McNamara
2009). If theism were to deny that special Fs depend on the nervous system, or to presume that

awareness/experience is unnatural, that begs the question in favor of supernaturalism. Even if con-
sciousness has not yet been naturalized enough to suit theists, as Charles Taliaferro (2009) com-
plains, REs have not been liberated from brains. Unusual Fs in REs are probably due to modes of
brain functioning different from those for mundane awareness. That difference does not mean that
those Fs are not happening, or that experiencers are wrongly describing what those Fs are like. But it
does mean that “F is truly otherworldly” cannot automatically follow from “F is evidently
unworldly.” For example, a suspension of cognitive operations organizing inner and outer sensory
information will generate an experience of unlimitedness, wholeness, and/or unification. Brain mal-
functions or disorders need not be implicated; practitioners of meditation can enter such states. For
more examples, consider how a rush of endorphins can arouse intense bliss or other powerful emo-
tions, or the way that psychoactive substances can generate hallucinations and alterations to one’s
sense of self.
Theism cannot rule out all natural explanations for REs in advance. No RE by itself allows one
to know how neurology is or is not responsible (that is true for all of one’s experiences), and theism
cannot forbid all future scientific explanations for REs. No matter the particular feature F, the
experiencer cannot thereby know that something otherworldly is really present, and theism cannot
pretend to know anything more about this F without begging the question. Again, it is not neces-
sary for atheism to show that all REs are already scientifically explained. It is a fallacy to claim that
theism is well-justified just because science currently lacks explanations for some features of REs.
Even if special features were truly otherworldly, rather than just unworldly in appearance, the
third premise of the abductive argument is questionable, that “if F = God’s A then F is other-
worldly.” Is the presence of God the only explanation, or the best explanation, for otherworldly
encounters? Theism often proceeds as if discerning otherworldly matters is tantamount to confirm-
ing God. Exposing scientific naturalism as inadequate, even if that demonstration is possible, is not
enough to conclude that God is real. Nonreductive naturalisms embrace consciousness, the self, and
everything about experience (Shook 2011).
Setting aside capacious naturalisms, the aggregate problem still looms for DET. Detecting an
intriguing array of otherworldly occurrences is not the same as discerning a unified or singular
reality responsible for all of them. That logical gap opens wider for contrary or contradictory attri-
butes. A possible God might appear as both personal and impersonal to humans, for example, but
simpler explanations can omit God (contra Gellman 1997, 117–119). Abduction surely cannot
license a swift inference to a singular being possessing or displaying contradictory attributes. The-
ology may try to circumvent the aggregate problem by resorting to apophatic features. However,
negative attributes lack any obvious way to ensure that they bind together. How would “neither
finite nor infinite,” “not-being,” and “beyond suffering” (readers may substitute any set of their
own devising) manage to indicate by themselves how they (do not) characterize the same ultimate
reality?
INDIRECT EXPERIENTIAL THEISM
Indirect experiential theism (INET) in general asserts that one or more specifiable features F of a
religious experience (RE) arise due to God, while God’s reality does not inherently possess those
Fs. The Fs with significance for INET are taken to be indications, markers, signs (and so on) tightly
linking God with human experience. INET is a needed complement to DET. Taking REs to be
much like ordinary perception, reliably delivering veridical beliefs in a realistic manner, leaves
DET vulnerable to many objections. Besides, many types of REs, no less convincing in import,
are unlike those emphasized by DET.
Although INET starts from experiences indirectly about God, their impressive features are quite
immediate in those experiences themselves. The immediacy of meaning is not the same as the
directness of presence. As immediate, they do not seem to be mediated, and no inferential process
need be in conscious awareness. And there is no premise-conclusion reasoning of the sort familiar
to theologians and philosophers. For the religious person convinced by his or her experience, such
as hearing God, what is heard is something immediately experienced as God’s voice, not as a voice
that may or may not be God’s. A DET argument can also start from this experiential situation, if
God’s voice is taken for an essential attribute of God. However, what God sounds like to human
ears strikes most theologians as less essential, and more relational to humans. Other relational
situations supply further INET examples, such as being conscious of God’s presence, sensing godly
activity, or appreciating godliness in an object. Theological arguments that rely on explicit infer-
ences concluding that divine mercy or a heavenly destination (etc.) can explain REs are not this
chapter’s focus. Arguments for God from revelation, for example, are refuted elsewhere (Shook
2018, 154–158).
INET takes inductive or abductive forms. INET’s appeal to induction will not lead to reason-
able success. Any theism claiming that special REs reveal its deity must be asked how it knows that
those experiential episodes count as intermediary evidence. Should atheism take each religion’s word
for it, when most of the rest of the religious world will not? Furthermore, within a theistic religion,
adherents dispute RE episodes among each other. There appears to be no guarantee that REs from
the religious community are all about the same deity. A typical follower does not regard other
adherents as reliable witnesses, unless they are internally preapproved by religious authorities or their
accounts cohere with creedal standards.
Even among the most devout, some suspicion is understandable. The Old Testament reads,
“And the children of Israel again did that which was evil in the sight of Jehovah, and served the
Baalim, and the Ashtaroth, and the gods of Syria, and the gods of Sidon, and the gods of Moab,
and the gods of the children of Ammon, and the gods of the Philistines; and they forsook Jeho-
vah, and served him not” (Judges 10:6, American Standard Version). In the New Testament,
Jesus warned his disciples about human fallibility: “Then if any man shall say unto you, Lo, here
is the Christ, or, Here; believe it not. For there shall arise false Christs, and false prophets, and
shall show great signs and wonders; so as to lead astray, if possible, even the elect” (Matthew
24:23–24, American Standard Version). Jesus was just one of many religious figures to warn
against false prophets or unreal gods. If any duly convinced and consecrated devotee could be
fooled by REs, according to the religion itself, atheism can only add its agreement. No one is
infallible about REs. That is why religions impose creedal standards on all adherents to ensure
conformity. An “infallible” pope justifies a doctrine with scripture, not his fresh REs of Jesus. A
Christian offering deviant REs is starting a heretical sect or a different religion, not improving
Christianity’s knowledge of God.
Even if there were direct witnesses having REs involving Jesus, no one else’s REs are infalli-
ble, as Jesus pointed out. That is why a religion promulgates an approved view of its god(s) and
each religion’s theological efforts prop up that view as the worthy standard for all other REs.
But there are no good RE arguments for theism here. Christianity’s belief that witnesses had
direct REs of Jesus as a deity—not just observations of a man—is a claim that no DET argument
can reasonably support, as earlier sections explained. As for other Christians, REs about Jesus (a
voice, a vision, a message, etc.) are deemed authentic only if they are consistent with Christian
standards. However, this is not a procedure permitting independent judgment about whether
REs are evidence of anything. Religions already understand this—they do not appreciate adher-
ents, or infidels, passing their own judgments on evidential REs. Theologies already understand
this, too—they do not want their religions to be criticized by standards alien to internal commu-
nal standards. All theistic religions reliant on REs insulate themselves from critical scrutiny from
any neutral standpoint on REs. Atheism is positioned at that neutral standpoint, along with unbi-
ased historical, cultural, anthropological, sociological, and psychological studies of religion and
religiosity.
Theism is unwilling to acknowledge atheism’s neutrality on REs. However, theists place their
religious thumb on the abductive scale so that their own deity gets weighed as the best explanation
for features of REs, far better than “false” gods. Each religion’s theology asserts that, unless its own
creedal standard is granted legitimacy for evaluating REs as good evidence, external scrutiny is pre-
judiced against that religion. However, that sort of theological assertion defends a standard fitting a
deity already taken for real, so theism again begs the question.

INET can avoid begging the question only by running into a dilemma: either all gods are sup-
ported by ample evidence from REs (extreme polytheism), or no real gods receive any evidential sup-
port from REs (atheism’s neutral view). On the assumption that the point of INET is not to encour-
age extreme polytheism but only theism, INET collapses. INET tries to forestall that fate by insisting
that there must be better and worse evidence for a deity among features of REs. But INET cannot
justify any selection without either presuming an acquaintance with a specific deity (to pick relevant
features) or appealing to a religion’s traditions about authentic REs (to lend credibility to special fea-
tures). Neither begging the question nor arbitrary preference leaves INET in a reasonable condition.
SKEPTICISM ABOUT GODS
Atheism does not dispute the compelling features of REs, or the devout ways that people describe
what they are like. Religions guide those matters in the course of sustaining committed practi-
tioners. Theistic religions claim that a deity is involved. Atheism skeptically denies that anything
supernatural is really encountered in REs. The arguments of this chapter show why real experiences
of God are never equivalent to experiences of a real God. Theism subverts itself by claiming that
certainty and tenacity, at individual or group levels, is sufficient for justified belief, since those views
either collapse into mere subjectivity or inflate into extreme polytheism. Theistic religions need REs
to be objective only in significance, provenance, and orientation, so they can demarcate their own
god(s) from unreal gods of other religions. Atheism’s skepticism toward gods is no different in qual-
ity, just in quantity.
Although certain features of REs can strike people as quite unnatural, it is not possible to
deduce a supernatural presence. Perhaps there are unnatural-seeming features common to many
REs around the world. However, nothing demonstrably supernatural is evident in them, and no
god accrues from selecting one or another set of them. Those choices are invariably parochial and
pluralistic, just as religions must have them, and just as atheism would expect them to be.
Bibliography
Alston, William P. Perceiving God: The Epistemology of Religious Experience. Everitt, Nicholas. The Non-existence of God. London: Routledge, 2004.
Ithaca, NY: Cornell University Press, 1991.
Gavrilyuk, Paul L., and Sarah Coakley, eds. The Spiritual Senses: Perceiving
Andersen, Marc, Uffe Schjoedt, Kristoffer L. Nielbo, and Jesper Sørensen. God in Western Christianity. Cambridge, UK: Cambridge University
“Mystical Experience in the Lab.” Method and Theory in the Study of Press, 2011.
Religion 26, no. 3 (2014): 217–245.
Geertz, Armin W., ed. Origins of Religion, Cognition and Culture. Durham,
Atran, Scott. In Gods We Trust: The Evolutionary Landscape of Religion. UK: Acumen, 2013.
Gellman, Jerome I. Experience of God and the Rationality of Theistic Belief.
Bellah, Robert N. Religion in Human Evolution: From the Paleolithic to the Ithaca, NY: Cornell University Press, 1997.
Axial Age. Cambridge, MA: Belknap Press of Harvard University Press,
2011. Hart, David Bentley. The Experience of God: Being, Consciousness, Bliss. New
Haven, CT: Yale University Press, 2013.
Blackmore, Susan J. Seeing Myself: The New Science of Out-of-Body
Experiences. London: Robinson, 2017. Jones, Richard H. Philosophy of Mysticism: Raids on the Ineffable. Albany:
State University of New York Press, 2016.
Blanke, Olaf, and Sebastian Dieguez. “Leaving Body and Life Behind: Out-
of-Body and Near-Death Experience.” In The Neurology of Consciousness, King, Barbara J. Evolving God: A Provocative View on the Origins of Religion.
edited by Steven Laureys and Giulio Tononi, 303–325. Amsterdam: Chicago: University of Chicago Press, 2017.
Elsevier/Academic Press, 2009. Martin, Luther H., and Donald Wiebe, eds. Religion Explained? The
Boyer, Pascal. Religion Explained: The Evolutionary Origins of Religious Cognitive Science of Religion after Twenty-Five Years. London: Blooms-
Thought. New York: Basic Books, 2001. bury Academic, 2017.
Bush, Stephen S. “Are Religious Experiences Too Private to Study?” Journal Martin, Michael. Atheism: A Philosophical Justification. Philadelphia: Temple
of Religion 92, no. 2 (2012): 199–223. University Press, 1990.
Davis, Caroline Franks. The Evidential Force of Religious Experience. Oxford: McNamara, Patrick. The Neuroscience of Religious Experience. Cambridge,
Clarendon; New York: Oxford University Press, 1989. UK: Cambridge University Press, 2009.
Murray, Michael J. “Scientific Explanations of Religion and the Justification Shook, John R. “Are People Born to Be Believers, or Are Gods Born to Be
of Religious Belief.” In The Believing Primate: Scientific, Philosophical, Believed?” Method & Theory in the Study of Religion 29, nos. 4–5 (2017):
and Theological Reflections on the Origin of Religion, edited by Jeffrey 353–373.
Schloss and Michael J. Murray, 168–178. Oxford: Oxford University
Shook, John R. Systematic Atheology: Atheism’s Reasoning with Theology.
Press, 2009.
London: Routledge, 2018.
Newberg, Andrew. “The Spiritual Brain: Science and Religious Experience.” Sidky, Homayun. The Origins of Shamanism, Spirit Beliefs, and Religiosity: A
In The Physics of the Mind and Brain Disorders, edited by Ioan Opris and Cognitive Anthropological Perspective. Lanham, MD: Lexington Books,
Manuel F. Casanova, 649–670. Cham, Switzerland: Springer Interna- 2017.
tional, 2017.
Stace, W. T. Mysticism and Philosophy. London: Macmillan, 1960.
Oppy, Graham. Describing Gods: An Investigation of Divine Attributes.
Swinburne, Richard. The Existence of God. Oxford: Clarendon, 1979.
Taliaferro, Charles. “Explaining Religious Experience.” In The Believing
Pyysiäinen, Ilkka. “Cognitive Science of Religion: State-of-the-Art.” Journal Primate: Scientific, Philosophical, and Theological Reflections on the Origin
for the Cognitive Science of Religion 1, no. 1 (2013): 5–28. of Religion, edited by Jeffrey Schloss and Michael J. Murray, 200–214.
Schellenberg, J. L. The Wisdom to Doubt: A Justification of Religious
Skepticism. Ithaca, NY: Cornell University Press, 2007. Taves, Ann. Religious Experience Reconsidered: A Building-Block Approach to
the Study of Religion and Other Special Things. Princeton, NJ: Princeton
Sharpless, Brian A., and Karl Doghramji. Sleep Paralysis: Historical, University Press, 2011.
Psychological, and Medical Perspectives. Oxford: Oxford University Press,
2015. Yandell, Keith E. The Epistemology of Religious Experience. Cambridge, UK:
Shook, John R. The God Debates: A 21st Century Guide for Atheists and
Zagzebski, Linda. “Epistemic Self-Trust and the Consensus Gentium
Believers (and Everyone in Between). Malden, MA: Wiley-Blackwell, 2010.
Argument.” In Evidence and Religious Belief, edited by Kelly James Clark
Shook, John R. “Varieties of Twentieth Century American Naturalism.” and Raymond J. VanArragon, 22–36. Oxford: Oxford University Press,
Pluralist 6, no. 2 (2011): 1–17. 2011.

TOPIC 6
Faith and Revelation: Theism

Robert Fastiggi
Professor of Systematic Theology
Sacred Heart Major Seminary, MI
This chapter defines revelation and its relation to scripture. The discussion covers topics such as the qualities
of divine revelation, nonreligious explanations, assessing claims and who decides whether some claim is true
or false, how claims are recorded, how reason is related to faith, and how considerations about faith and
revelation might support theism.
WHAT IS MEANT BY FAITH AND REVELATION?
Faith and revelation are interrelated terms. When some truth or information is revealed, the recipi-
ent or recipients of the revelation must have faith in the source of the revelation and the message
revealed. The English word “revelation” comes from the Latin revelare, which means “to unveil or
disclose” (Dulles 2003, 193). The New Testament uses various Greek words for revelation such
as apokalupsis, epiphaneia, and kèrugma (Latourelle 1997, 46). Muslims identify the Qur ʾan as reve-
lation by the word wahi (Danner 1988, 243); and the Jewish tradition employs several biblical
words for “revelation,” including glh (uncover, unveil), yd’ (to proclaim), nggd (to report), and
dvr, which is used for “decisive communication on God’s part” (Deninger 1987, 359).
The word “faith” comes from the Latin fides, which means “trust” as well as “faithfulness” (Stel-
ten 1995, 102). The Greek term pistis (faith, faithfulness) is used in both the Septuagint and the
New Testament to translate corresponding Hebrew words such as he’ ĕmîn (to trust, to believe);
‘mn (firmness, certainty); and bātah, which means “security” or “trust” (Pickar 2003, 589–591).
In Islam the word for “faith” is īmān, which is complemented by the attitude of islām, which means
“submission to the Divine Will” (Danner 1988, 235–236).
Christian theology often makes a distinction between fides qua creditur—the faith by which one
believes, that is, the subjective dimension of faith—and fides quae creditur—the faith that is
believed, that is, “the content of faith as revealed by God” (Muller 1985, 117). This distinction goes
back to Saint Augustine (De Trinitate XIII, 2, 5), and it shows how faith is a subjective response
to an objective truth that is revealed. As Christianity developed, there was first the “rule of faith”
(regula fidei), a type of summary of the key articles of the deposit of faith. By the fourth century,
this “rule of faith” was expressed in more formal creeds (Wicks 2009, 71–73). For something to
be recognized as a revelation, therefore, it requires faith in what has been revealed and some means
of summarizing or preserving the content of the revelation.
NATURAL REVELATION VERSUS SUPERNATURAL REVELATION
The distinction between natural revelation and supernatural revelation is important for understand-
ing how the main theistic religions understand revelation. Natural revelation can be understood
within the domain of natural theology, which refers to truths about God that are “naturally, and
163
thus philosophically, knowable” (John Paul II 1998, no. 67).

KEY CONCEPTS Natural revelation refers to God’s self-disclosure of truths
through the created order that can be grasped by natural rea-
son. Natural theology, therefore, is actually a branch of philos-
Divine Revelation
ophy rather than theology since it appeals to what can be
Divine Source known about God by natural reason rather than by supernatu-
Mediation (pertaining to revelations ral revelation. Those who pursue natural theology might also
recorded in scripture) believe in supernatural revelation, but when doing natural the-
Model of Dialectical Presence ology they limit themselves to what can be known about God
via the natural order and human reason.
Natural Revelation In this regard, it’s important to note that natural theol-
Natural Theology ogy and natural revelation do not necessarily challenge the
Objective Revelation reality of supernatural revelation. Supernatural revelation
refers to God’s self-disclosure of truths that could not be
Supernatural Revelation
known by natural reason alone. There are, of course, rational-
ists who claim that the only secure knowledge of God is that
which is obtained by natural reason. This leads some think-
ers, such as John Locke and Immanuel Kant, to favor “religion within the bounds of reason
alone” (Fries 1996, 252). Radical fideists, such as Karl Barth in his early writings, stand in sharp
opposition to the rationalists. According to such fideists, only divine revelation provides authen-
tic truth about God. Natural theology, therefore, must be rejected as either useless or blasphe-
mous.
Most Christians do not reject natural revelation entirely. Some, however, distinguish between
natural revelation—“the knowledge of God to all people in and through the created order”—and
“special revelation,” which refers to “the special gift of saving knowledge in Christ and through
the prophets” (Muller 1985, 265). The First Vatican Council of the Catholic Church (Vatican I;
1870) affirmed natural revelation and natural theology as expressions of the order of natural reason.
In the council’s constitution, Dei Filius, we read:
There is a twofold order of knowledge, distinct not only in its principle but also in its object,
because in the one we know by natural reason, in the other by divine faith; in its object,
because apart from what natural reason can attain, there are proposed to our belief mysteries
that are hidden in God that can never be known unless they are revealed by God. (Denzinger-
Hünermann 2012, no. 3015)
These two distinct orders of knowledge, the natural and the supernatural, are seen as complemen-
tary rather than oppositional. In addition to recording various things that God explicitly says (super-
natural revelation), the Jewish and Christian scriptures, for instance, testify to the natural revelation
of God that comes through the order and beauty of the universe (Psalms 19:1–2; Wisdom of Solo-
mon 13:1–9; Acts 17:26–27; and Romans 1:19–20). The New Testament also points to an interior
recognition of universal moral norms written into the heart by God (Romans 2:14–15) as part of
natural revelation. The Bible, therefore, recognizes this twofold order of knowledge even if it is
not set forth systematically.
The Catholic scholastic tradition as exemplified by Saint Thomas Aquinas has always upheld
the fundamental harmony between the truths known by reason and those known by revelation
and faith. Pope John Paul II summarizes the contribution of Aquinas in this way:
In an age when Christian thinkers were rediscovering the treasures of ancient philosophy, and
more particularly of Aristotle, Thomas had the great merit of giving pride of place to the
harmony which exists between faith and reason. Both the light of reason and the light of faith
come from God, he argued; hence there can be no contradiction between them. More
radically, Thomas recognized that nature, philosophy’s proper concern, could contribute to
the understanding of divine Revelation. Faith therefore has no fear of reason, but seeks it out
and has trust in it. Just as grace builds on nature and brings it to fulfilment, so faith builds
upon and perfects reason. (John Paul II 1998, no. 43)
This understanding of a fundamental harmony between faith and reason, however, does not remove
the need to specify how the supernatural order of divine revelation differs from the order of natural
reason. The German theologian Matthias Joseph Scheeben notes that the supernatural is
a sphere which extends as much beyond nature’s cognitive faculty and principles of cognition as
it projects beyond nature itself; in a word, which is supra-rational to the degree and for the
reason that it is supernatural. This domain of truth is essentially distinct from that which reason
and philosophy can rule. Consequently it forms the object of a special science. Not only is it in
actual fact known by faith, on God’s authority; it cannot be known in any other way; and so it
constitutes a proper object of knowledge that is specific to faith. (Scheeben 1946, 736)
This distinction between the realms of faith and reason, however, does not mean there is opposition
between the two. Since both the order of reason and the order of faith come from God, there can-
not be any ultimate conflict between them. Nevertheless, the question must be raised: if the order of
supernatural revelation exceeds the capacities of reason and philosophy to pass judgment on the
truth of such claims, how is it that human beings are still capable of recognizing the truth and
meaning of what has been revealed? This is a question touched on later in this chapter.
RELIGIONS OF THE WORLD THAT CLAIM

SUPERNATURAL REVELATION
Most people are aware that Jews, Christians, and Muslims believe in the special revelations that are
recorded in the Torah, the New Testament, and the Qur ʾan. Other religions that claim special
divine revelation are Zoroastrianism and Hinduism (Deninger 1987, 362). The concept of “special
revelation” can be applied to Buddhism only in an extended sense. The enlightenment of the Bud-
dha is believed to disclose new insights about the human condition and the path to liberation from
pain. In this sense, the Buddha’s enlightenment may be considered a kind of “natural revelation.”
There is not, however, an implication of “a Supreme Being communicating some kind of divine
truth” (Brassard 2010, 438).
Philosophical insights are sometimes called “revelations,” but, as with Buddhism, there is not
usually the idea of a supreme being disclosing divine truths, and so these claims are presumably
about natural rather than special revelation. The exception might be Plato, who, in his doctrine of
ideas, “provides the philosophical presupposition for understanding everything finite as conditioned
and as sustained in being by the idea of God” (Deninger 1987, 357–358). Plato, though, never
claimed to be a prophet but a philosopher. The wisdom he received from the transcendent world
of ideas is theoretically open to all other people who seek true knowledge, and thus Platonic insights
would belong to the sphere of natural revelation rather than to that of special revelation.
Zoroastrianism, which originated in ancient Persia with the prophetical figure of Zarathustra, or
Zoroaster, in the seventh or sixth century BCE “saw revelation as having its source in the voluntary
action of a unique and personal God” (Deninger 1987, 361). Hinduism likewise makes a distinction
between the Vedas, or scriptures of divine origin, the Shruti—that which is heard—and the scriptures
of human origin, the Smirti—that which is recollected (Kramer 1986, 23–24). The recipients of the
Shruti are the ancient seers who heard or perceived the divine revelations from the gods (Deninger
1987, 361). Exactly how the ancient seers perceived these revelations is somewhat mysterious because
of the way in which the revelations behind the Vedas (which cover a period from roughly 2000 to 500
BCE) are described. According to Hindu tradition, the Vedas were “not produced by human agency”
because the wisdom behind the revelations is eternal. The seers, or rishis, are believed to have “seen,”
“heard,” or “discovered” the revelations that were later recorded in the Vedas (Dandekar 1987, 214),
and this would give reason to classify them as claims to special revelation.
Judaism understands the Torah (the first five books of the Bible) as the divinely revealed Word
of God. The great Jewish thinker Moses Maimonides, in The Event of Sinai, writes:
Remember, my brethren in this covenant, that this great, incomparable and unique historical
event is attested by the best of evidence. For never before or since, has a whole nation

witnessed a revelation or beheld His splendor. The purpose of all this was to confirm us in the
faith so that nothing can change it, and to reach a certainty which will sustain us in these
trying times of fierce persecution and absolute tyranny, as it is written, “for God is come to test
you” (Exod 20:17). Scripture means that God revealed Himself to you, thus in order to give
you strength to withstand all future trials. Now do not slip nor err, be steadfast in your
religion, and persevere in your faith and its duties. (Glatzer 1969, 394–395)
In this passage we can see that Maimonides believes that the revelation of God to the People of
Israel “is attested by the best of evidence.” This evidence serves to strengthen faith and lead to cer-
tainty. The relationship between revelation, faith, and certainty is discussed later in this chapter.
Judaism also attributes a revelatory status to prophetical and other writings, which combine to
form the Hebrew Bible, or the Tanakh, consisting of the Torah, the Prophets (Nevi’im) and the
Writings (Kethuvim). In addition to these biblical writings, Judaism also accepts the interpretative
authority of the Talmud, a collection of commentaries, legend, and law from Palestinian and Baby-
lonian rabbis around 500 CE (Fellows 1979, 288). The Talmud is not considered divine revelation
but a collection of rabbinical commentaries and applications of divine revelation.
Christians accept the divine character of the revelations given to the Jews as recorded in the
Tanakh, but they also believe that in Jesus, God has fulfilled the messianic hopes of Israel that are
among the contents of what God revealed. According to Christian belief, Jesus Christ is not only
the anointed one, the Messiah. He is also the Son of God, and the Word of God made flesh, the
Incarnation; for this reason Christians regard the words of Jesus as part of special revelation. Two
Christian scriptural passages in particular express this conviction that Jesus, the Word of God made
flesh, is the supreme revelation of God to the human race. The first is from the Gospel of John:
In the beginning was the Word, and the Word was with God, and the Word was God.… And
the Word became flesh and made his dwelling among us, and we saw his glory, the glory as of
the Father’s only Son, full of grace and truth.… From his fullness we have all received, grace in
place of grace, because while the law was given through Moses, grace and truth came through
Jesus Christ. No one has ever seen God. The only Son, God, who is at the Father’s side, has
revealed him. (John 1:1–18)
This passage points to Jesus as the revelation of God’s glory, truth, and grace as the Word made flesh.
The second important passage is from the beginning of the letter to the Hebrews:
In times past, God spoke in partial and various ways to our ancestors through the prophets; in
these last days, he spoke to us through a son, whom he made heir of all things and through
whom he created the universe, who is the very refulgence of his glory, the very imprint of his
being who sustains all things by his mighty word. (Hebrews 1:1–3)
Here we see an important insight into the nature of revelation as Christians understand it: it is a
form of divine speech. God spoke through the prophets in times past, but now, in Jesus, he has spo-
ken in a supreme and definitive way because Jesus is the very imprint of God’s being or essence.
The New Covenant in Jesus that is constitutive of Christianity is believed to fulfill the hopes of
the Old Covenant of Judaism. In a somewhat similar but not identical way, the Qurʾan is under-
stood by Muslims as the supreme revelation of God that supersedes and corrects the prior revela-
tions given through Moses (the Torah) and through Jesus (the Gospel, or Injil). Muslims generally
believe the Torah and the Gospel were divine revelations, but they believe that the texts that Jews
and Christians have now are corrupt forms of the original record of divine revelations (Esposito
1998, 17–18). The Qur ʾan, though, is seen as the book that God himself protects from corruption:
This is the Book in which there is no doubt, a guidance for the reverent, who believe in the
Unseen and perform the prayer and spend from that which We provided them, and who
believe in what was sent down unto thee, and what was sent down before thee, and who are
certain of the hereafter. It is they who act upon guidance from the Lord, and it they who shall
prosper. (Qur ʾan 2:1–4; Nasr 2015, 13–15)
Here we see that there is an internal testimony as to the reliability of the revelation of the Qur ʾan. It
is the book “in which there is no doubt.” This, of course, raises the question: how do the recipients
of a revelation have faith in its divine quality? This question is explored later in this chapter.
REVELATION AND SCRIPTURE
As can be seen from the example of the Qur ʾan, a revelation can be understood as a divinely
revealed book. Judaism looks upon the Torah in a similar way. The difference, though, is that the
Torah is more a collection of laws and a record of events (e.g., the Exodus) that reveal God’s saving
actions; it is thus not just a record of divine words. The Bible for Christians—which includes the
Old Testament and the New Testament—also is a record of deeds as well as words. For Christians,
it is not just the teachings of Jesus that are revelatory but also his actions, especially his death, res-
urrection, and ascension into heaven.
The question, of course, should be raised: is the Bible itself to be considered revelation or is the
Bible a record of revelatory deeds and words? From a Christian perspective, it would seem to be
both; for Christians the scripture itself is revered as the Word of God, and the reading of scripture
is central to Christian liturgical worship. Christians from Evangelical and Protestant traditions look
upon the Bible as the final authority and guide. The 1643 Westminster Confession of Faith, which
reflects a Calvinist perspective, looks upon holy scripture as the source of “our full persuasion and
assurance of the infallible truth and divine authority” (Bettenson and Maunder 2011, 306). The
Second Confession of the Baptists from 1677 also states: “The Holy Scripture is the only sufficient,
certain, and infallible rule of all saving knowledge, faith, and obedience” (Bettenson and Maunder
2011, 311).
The Eastern Orthodox Churches and the Catholic Church also look upon the Bible as the
revealed Word of God, but they do not assume the sola scriptura, or “scripture alone,” position typ-
ical of most Protestants. The Orthodox tend to regard the Bible as part of tradition, which includes
the books of the Bible, the creed, the decrees of the ecumenical councils, the teachings of the church
fathers, and “the whole system of doctrine, Church government, worship, spirituality and art which
orthodoxy has articulated over the ages” (Ware 1993, 196).
The Catholic Church, following the Council of Trent, “clearly perceives that the truth and rule
are contained in the written books and unwritten traditions that have come down to us, having
been received by the apostles from the mouth of Christ himself or from the apostles by the
dictation of the Holy Spirit, and have been transmitted, as it were, from hand to hand” (Denzinger-
Hünermann 2012, no. 1501). Here we see that the Catholic Church does not follow a “scripture
alone” approach. Vatican II, however, makes clear that there is a close connection between sacred
scripture, which is the Word of God “consigned to writing under the inspiration of the divine
Spirit,” and sacred tradition, which is “the Word of God entrusted by Christ the Lord and the Holy
Spirit to the apostles” (Vatican II, Dei Verbum, 1965, no. 9). Both sacred scripture and sacred tra-
dition flow from the same divine source and “form one sacred deposit of the Word of God, com-
mitted to the Church” (Vatican II, Dei Verbum, 1965, no. 10). Although the divine revelations
transmitted by sacred scripture and by sacred tradition ended with the death of the last apostle,
Catholicism holds that there is a further growth in understanding of this divine revelation that
can come about as the result of prayer and study. As René Latourelle, SJ, explains, “it is not the
apostolic Tradition itself, but rather our penetration of these inherited words and realities which
develops and grows” (Latourelle 1966, 477).
SCRIPTURAL REVELATION AND INTERPRETATION
When a revelation is given, there is a need for understanding of that revelation, and thus there is
usually need for the interpretation of the content of what has been revealed. The English Catholic
theologian John Henry Newman wrote: “A revelation is not given if there be no authority to decide
what it is that is given” (Newman 1989, 89). As a Catholic, Newman believed that there was the
need for a teaching authority, the magisterium of the pope and the bishops, to determine the proper
interpretation of revelation. This is also the position of Vatican II, which recognizes that the magis-
terium is not above the Word of God but serves it by giving proper interpretation to what has been
divinely revealed. From a Catholic perspective, “sacred tradition, Sacred Scripture, and the teaching

authority of the Church, in accord with God’s most wise design, are so linked and joined together
that one cannot stand without the others and that all together and each in its own way under the
action of the one Holy Spirit contribute effectively to the salvation of souls” (Vatican II, Dei Ver-
bum, 1965, no. 10).
Other religious traditions also recognize the need for some type of interpretive authority. In Islam
the Hadith, the record of Muhammad’s words and deeds, serves as “a source of divine instruction”
(Reynolds 2012, 87). For Shi’ite Muslims, the rightly guided imams like Muhammad “are a source
of revelation and a guide for Islamic law” (Reynolds 2012, 87). In classical Islam there are four sources
of the law (sharia) established by God: (1) the Qur ʾan, which always has priority; (2) the Hadith,
which records the sayings and deeds of Muhammad; (3) the analogical reasoning of the jurists or Mus-
lim legal scholars; and (4) the consensus of the community (Esposito 1998, 78–84).
In Judaism, the Talmud serves as a major source for interpreting and applying the teachings of
the Torah and the other books of the Hebrew Bible. The Talmud is a collection of the discussions
and elaborations of Jewish law given by the rabbis of Palestine and Babylonia. It dates from around
500 CE and is studied by rabbis up to the present day. There are, of course, different interpretations
and understandings of Jewish law according to the Orthodox, Conservative, and Reform traditions
(Fellows 1979, 306–309).
In Christianity, the Orthodox Church tends to understand divine revelation according to sacred
tradition, which includes not just the Bible but also the creed, the teachings of the church fathers
and ecumenical councils, along with the sacred liturgy. The instructions given by Orthodox bishops
and synods also carry authority. The Catholic Church, as we have seen, believes that the revelation
of sacred scripture requires the authoritative interpretation of the bishops in communion with the
pope. Vatican II states that “the way of interpreting Scripture is subject finally to the judgment of
the Church, which carries out the divine commission and ministry of guarding and interpreting
the Word of God” (Vatican II, Dei Verbum, 1965, no. 12).
Protestant and Evangelical Christians also recognize the need for proper interpretation of scrip-
ture. This is why they have their own confessional creeds. Nevertheless, the Bible is held up to be
the rule of faith and is not subject to any other norm, that is, it is the norma normans non normata.
The 1643 Westminster Confession of Faith believes that “the infallible rule of the interpretation of
Scripture is Scripture itself” (Bettenson and Maunder 2011, 306).
THE QUALITIES OF DIVINE REVELATION
Divine revelation is thus generally understood to include words and deeds that are typically recorded
in a divinely inspired scripture. The first quality of an authentic revelation is its divine source.
Human beings desire to know God and his will for them, and a genuine divine revelation commu-
nicates to them who God is and what he wants them to do. Revelations that are recorded in scrip-
ture, however, require mediation. The process of mediation varies according to the religious tradi-
tion. In traditional Judaism, Moses was considered to be the author of the Torah. God’s
revelation, therefore, came through the mediation of Moses. Even if Moses is not taken by a partic-
ular religious tradition as the author of the Torah, there would still need to be the mediation of the
authors who wrote and edited the biblical text.
Islam looks upon the Qur ʾan as God’s final revelation to the human race, and the Qur ʾan is
understood as superseding and correcting all prior scriptural traditions (e.g., the Torah and the Gos-
pel). The Qur ʾan, though, is mediated by Muhammad, who is the chosen instrument of God to
recite the words revealed to him (the word qur ʾan means “recitation”). The Qur ʾan “is said to have
been revealed to the Prophet [Muhammad] in piecemeal fashion over a period of some twenty-three
years” (Danner 1988, 238). Because Muhammad was illiterate, however, the Qur ʾanic revelation
required the further mediation of the scribes who recorded what Muhammad recited as the revela-
tions given to him. According to Muslim tradition, the third successor to Muhammad, the caliph
Uthman (r. 644–656), convened a council with Zayd, Muhammad’s personal scribe, and other
Muslim scholars. This council established the official text of the Qur ʾan and ordered all other
versions to be burned (Reynolds 2012, 100–101). We can see that the Qurʾan, which is believed by
Muslims to be God’s final revelation, involved several levels of mediation: first from God to the
angel, Gabriel; then to Muhammad; then to the scribes; and then to the council held under the
authority of Uthman, the third caliph.
According to Christians, the New Testament consists of twenty-seven books that combine with
the books of the Old Testament to form the Bible. Biblical revelation, therefore, involves the medi-
ation of multiple authors who are believed to be divinely inspired. The Second Vatican Council gives
this description of God’s use of these human authors:
Those divinely revealed realities which are contained and presented in Sacred Scripture have
been committed to writing under the inspiration of the Holy Spirit. For holy mother Church,
relying on the belief of the Apostles (see John 20:31; 2 Tim. 3:16; 2 Peter 1:19–20, 3:15–16),
holds that the books of both the Old and New Testaments in their entirety, with all their parts,
are sacred and canonical because written under the inspiration of the Holy Spirit, they have God
as their author and have been handed on as such to the Church herself. In composing the sacred
books, God chose men and while employed by Him they made use of their powers and abilities,
so that with Him acting in them and through them, they, as true authors, consigned to writing
everything and only those things which He wanted (Vatican II, Dei Verbum, 1965, no. 11).
This passage recognizes that God chose and employed certain men to write the words of sacred
scripture. These authors, therefore, mediate the revelation of God to human beings. In doing so,
however, they make use of their own powers and abilities, but God acts “in them and through
them” to communicate his revelation.
The mediation of revelation thus involves the use of human language, whether it be Hebrew, Greek,
or Arabic. The mediation of divine revelation likewise involves the use of various concepts and expres-
sions intelligible to the intended audience. The interpretation of a divinely revealed text necessarily
requires an understanding of the language and concepts used. Vatican II offers this explanation:
To search out the intention of the sacred writers, attention should be given, among other
things, to “literary forms.” For truth is set forth and expressed differently in texts which are
variously historical, prophetic, poetic, or of other forms of discourse. The interpreter must
investigate what meaning the sacred writer intended to express and actually expressed in
particular circumstances by using contemporary literary forms in accordance with the situation
of his own time and culture. For the correct understanding of what the sacred author wanted
to assert, due attention must be paid to the customary and characteristic styles of feeling,
speaking and narrating which prevailed at the time of the sacred writer, and to the patterns
men normally employed at that period in their everyday dealings with one another. (Vatican
II, Dei Verbum, 1965, no. 11)
Scriptural revelation, therefore, involves a divine source and the mediation of divinely inspired
messengers and authors who must use human language and concepts. How, though, is the revelation
communicated? The Bible itself shows that God can make use of such things as dreams (Genesis 20:6;
Matthew 1:20), visions (Ezekiel 2–3; Zechariah 1–6), prophetic communications (1 Samuel 9:15),
and symbolic images (Revelation 1:9–20). There are times, though, when God speaks directly in
words, as in Exodus 20:2–3: “I, the Lord, am your God, who brought you out of the land of Egypt,
that place of slavery. You shall not have other gods besides me.” Because Christians believe that Jesus
is the incarnate Word of God, the words of Jesus are understood as coming from God even though
they are mediated for us through the written text. In the Qurʾan there are many examples of direct
divine communications, as in 4:105: “Verily, We have sent down unto thee the Book in truth, that
thou might judge between men according to what God has shown thee” (Nasr 2015, 241).
MODELS OF REVELATION
Theologians such as Avery Cardinal Dulles, SJ, have proposed various models or approaches for
understanding revelation and its relation to faith. In his influential book Models of Revelation, Dulles
presents five basic models of divine revelation: (1) revelation as doctrine, (2) revelation as history,

(3) revelation as inner experience, (4) revelation as dialectical presence, and (5) revelation as new
awareness (Dulles 1983, 27–28).
Revelation as doctrine focuses on “clear propositional statements attributed to God as authori-
tative teacher” (Dulles 1983, 27). Following this model, Protestant Christians will highlight certain
key teachings of the Bible, especially those having to do with salvation. Catholics following this
model will tend to use the Bible as support for official church doctrine. In the Jewish tradition, rev-
elation as doctrine tends to focus less on theological doctrines and more on practice. The Jewish
scholar Samuel Belkin notes that Judaism was never overly concerned with theological doctrines:
“It desired, rather, to evolve a corpus of practices, a code of religious acts, which would establish a
mode of religious living. True, these acts and practices stem from basic theological and moral con-
cepts, but most significantly, these theological theories of Judaism remain invisible, apprehensible
only through the religious practices to which they gave birth” (Glatzer 1969, 580).
Islam resembles Judaism in this respect, that doctrine is expressed primarily in the observation
of certain religious obligations, namely, the five Muslim pillars: (1) the profession of faith (shahada),
(2) daily prayer (salat), (3) almsgiving (zakat), (4) the fast of Ramadan, and (5) the pilgrimage, or
hajj (Esposito 1998, 88–93). The pillars, though, are supported by the basic doctrines revealed in
the Qur ʾan such as the unity of God, the teachings of the prophets, the reality of angels and
demons, submission to the divine will, the day of judgment, and the final resurrection.
Revelation as history “maintains that God reveals himself in great deeds, especially those which
form the major themes of biblical history” (Dulles 1983, 27). This model tends to look upon sacred
scripture as a witness to the actions of God in history. There is a type of objectivity in this model in
that it highlights the importance of concrete deeds of God rather than simply of words. The biblical
texts, however, communicate not only deeds but also words of instruction, exhortation, command,
and so on. The deeds recorded in the Bible cannot be separated from the words that surround them.
Revelation as inner experience emphasizes “the self-manifestation of God by his intimate pres-
ence in the depths of the human spirit” (Dulles 1983, 115). The stress in this model is on the inte-
rior effect of the revelation on the human person. The scriptures might relate words and deeds of
God, but the real revelation is found in the inward experience of divine grace and communion with
God. Scripture, therefore, only serves as a vehicle for the real revelation of God’s love that is experi-
enced inwardly by each person. While this model is correct to highlight the interior effect of biblical
revelation, there is a danger of making revelation entirely subjective rather than objective.
The model of revelation as a dialectical presence tries to avoid the extremes of overly objective
and overly subjective views of revelation. The model of dialectical presence tries to see revelation as
an encounter between God and the individual subject or the community. Revelation, therefore, can
be understood as “God’s address to those whom he encounters with his word in Scripture and
Christian proclamation” (Dulles 1983, 115). God’s revelation is dynamic and interpersonal. This
model tries to harmonize the objective and subjective dimensions of revelation. Some, however,
might object that it is still overly subjective in that the revelation has value only when there is a
favorable subjective response.
The fifth model presented by Dulles is that of revelation as new awareness. Authentic revelation
occurs when “there is a breakthrough to a higher level of consciousness as humanity is drawn to a
fuller participation in divine creativity” (Dulles 1983, 115). According to this model, there must
be an elevating effect on those who receive a revelation. Revelation is not so much the disclosure
of new truths as it is the disclosure of a new awareness of the divine presence. This model is correct
to underscore the importance of a new awareness of God’s presence in revelation. By its emphasis
on an elevating experience, however, it tends to obscure the importance of the words and concepts
needed to express the content of the revelation. There must be words and concepts to mediate the
revelatory experience. In this regard, Dulles proposes “symbolic mediation” as the means for under-
standing how a revelation is communicated. As he explains: “Revelation never occurs in a purely
interior experience or an unmediated encounter with God. It is always mediated through sym-
bol—that is to say, through an externally perceived sign that works mysteriously on the human con-
sciousness so as to suggest more than it can clearly describe or define. Revelatory symbols are those
which express and mediate God’s self-communication” (Dulles 1983, 133). In the Christian
tradition, light can be understood as a revelatory symbol, especially because God is described as light
(1 John 1:5). The cross also is a revelatory symbol with a strong historical reality. Sacramental wor-
ship also involves many revelatory symbols such as water for the cleansing of baptism.
THE ROLE OF FAITH IN THE RECEPTION OF REVELATION
If revelation involves the mediation of symbols, words, and concepts, it also involves the reality of
faith. As noted earlier, faith involves a subjective response of trust in the objective content of what
has been revealed. According to Vatican I (1870), faith in divine revelation is “the full homage of
the intellect and the will to the God who reveals” (Denzinger-Hünermann 2012, no. 3008). Faith,
however, is not a merely human response. Instead, it is also “a supernatural virtue whereby, inspired
and assisted by the grace of God, we believe that what he has revealed is true, not because the
intrinsic truth of things is recognized by the natural light of reason, but because of the authority
of God himself who reveals them, who can neither err nor deceive” (Denzinger-Hünermann
2012, no. 3008). After Peter confesses his belief in Jesus as the Messiah and the Son of the living
God, Jesus says: “Blessed are you, Simon, son of Jonah; for flesh and blood has not revealed this
to you, but my heavenly Father” (Matthew 16:17). In other words, Peter’s confession of faith is
not something merely human but a response to a supernatural illumination and one that is possible
only because of that divine action.
In the Christian tradition, Thomas Aquinas believes the assent of faith to divine revelation
can have a twofold cause. The first cause is an external stimulus or inducement, as when someone
sees a miracle and is persuaded to embrace the faith. The second cause is God himself, who raises
human nature above itself and moves the human person “inwardly by grace” (Summa theologiae
2a2ae, q. 6, a. 1). The Jesuit theologian Francisco Suárez builds on this insight of Aquinas and
describes the infusion of faith as a “divine illumination” because faith is a type of light and the
baptized are said to be illuminated by the reception of faith (De fide, disput. III, sect. 3, no. 6,
in Opera omnia 1858, 47).
The Catechism of the Catholic Church (CCC) describes divine revelation as an act of loving
self-disclosure on the part of God who “addresses men as friends, and moves among them in order
to receive them into his own company” (CCC 1997, 142). The proper response to this invitation is
faith “by which man completely submits his intellect and will to God” and “with his whole being
gives his assent to God the revealer” (CCC 1997, 143). This human response is called “the obedi-
ence of faith” (Romans 1:5, 16:26).
FAITH, REASON, AND REVELATION
What is the role of reason in the assent of faith to revelation? There are different responses to this
question. One response is that of fideism, which tends “to undervalue the role of reason in examin-
ing religious claims” (O’Collins and Farrugia 1991, 78). Fideism in its more extreme form “dis-
counts the exercise of reason in religious matters, seeking to proceed by faith alone” (Feingold
2016, 132 n. 32).
Another response is that of rationalism, which refers to “any system which privileges reason in
the search for truth, including religious truth” (O’Collins and Farrugia 1991, 78). Those who favor
rationalism will only accept claims of revelation that correspond to their understanding of reason.
This rationalistic approach to revelation is favored by Enlightenment thinkers like Kant, the author
of Religion within the Limits of Reason Alone. Rationalism tends to reduce the range of what is
acceptable in the content of revelation to matters that correspond to reason. As a result, revelation
is often rendered superfluous because reason is thought to be able to discern the basic truths about
God and ethics apart from scripture or divine intervention. The Bible might be recommended for
its ethical and human insights, but the truths found in scripture are to be judged by the standards
of human reason.

The third response is to uphold the harmony of faith, reason, and revelation. The effort to har-
monize faith and reason with respect to scripture has a long history. Maimonides, in The Guide for
the Perplexed, notes that certain biblical terms and expressions must be understood as having a spiri-
tual rather than physical sense. With regard to the laws of the Torah, Maimonides believes that the
human mind can discern reasons for many of these laws (Hyman, Walsh, and Williams 2010, 362).
In some cases, though, the reasons for the laws are known only by God and believers must trust in
his wisdom for laying down these obligations. Maimonides, therefore, believes in the harmony of
faith and reason, but he recognizes that certain teachings of the Torah transcend the bounds of
human reason.
The Muslim philosopher Ibn Rushd, or Averroes, also tries to harmonize faith, reason, and
revelation in his work The Decisive Treatise. As a Muslim he accepts the revelation of the
Qur ʾan, and he maintains that “demonstrative study does not lead to [conclusions] conflicting
with what Scripture has given us, for truth does not oppose truth but accords with it and bears
witness to it” (Hyman, Walsh, and Williams 2010, 292). Ibn Rushd, however, also argues that
“if the apparent meaning of Scripture conflicts with demonstrative conclusions, it must
be interpreted allegorically, that is metaphorically” (Hyman, Walsh, and Williams 2010,
292). This seems to harmonize with Maimonides’s position that certain biblical terms and
expressions—especially those that depict God in an anthropomorphic way—should not be
understood physically but spiritually.
In spite of Ibn Rushd’s efforts to defend the integrity of philosophy and its harmony with
the Qur ʾan, most Muslim thinkers preferred the position of al-Ghazālī, who had authored The
Incoherence of the Philosophers. The philosophical approach of Ibn Rushd was often looked upon
with suspicion because of its perceived challenge to the authority of the Qurʾan. Some even viewed
philosophers such as Ibn Rushd “as rationalists and nonbelievers” (Esposito 1998, 74). This posi-
tion, however, does not mean that philosophy died out in Islam. It persisted in the philosophical
mysticism of thinkers such as al-Suhrawardī and Ibn al- ʿArabī, who nourished the Sufi mysticism
of divine illumination (Naeem and El-Ansary 2013, 815–816).
Although many Protestant and Evangelical Christians tend toward fideism, the Eastern Ortho-
dox and Catholic traditions affirm the fundamental harmony between faith, reason, and revelation.
The Orthodox generally follow the tradition of the spiritual and allegorical senses of scripture
favored by the church fathers. The Catholic tradition, influenced by Thomas Aquinas, teaches that
there can be no conflict between reason and revelation, provided that each remains within its own
proper domain. Vatican I, in its Dei Filius on the Catholic faith, teaches the following:
However, though faith is above reason, there can never be a real discrepancy between faith and
reason since the same God who reveals mysteries and infuses faith has bestowed the light of
reason on the human mind, and cannot deny himself, nor can truth contradict truth. The
deceptive appearance of such a contradiction is mainly due to the fact that either the dogmas
of faith have not been understood and expounded according to the mind of the Church or
fanciful conjectures are taken for verdicts of reason. (Denzinger-Hünermann 2012, no. 3017)
Here we see a recognition that both the light of reason and the light of faith come from God. Con-
flicts between faith and reason are only apparent and not real, because truth cannot contradict truth.
The Catholic Church also recognizes that there can be no conflict between the content of rev-
elation in scripture (properly understood) and physical science (when it makes methodologically
well-disciplined claims). Pope Leo XIII, in his 1893 encyclical Providentissimus Deus, makes it clear
that “there can never, indeed, be any real discrepancy between the theologian and the physicist, as
long as each confines himself within his own lines and both are careful, as Saint Augustine warns
us, ‘not to make rash assertions or to assert what is not known as known’” (Denzinger-Hünermann
2012, 3288). The same pope goes on to say that the authors of scripture “did not seek to penetrate
the secrets of nature, but rather described and dealt with things in more or less figurative language,
or in terms that were commonly used at the time and that in many instances are in daily use at this
day, even by the most eminent men of science” (Denzinger-Hünermann 2012, 3289). Christians of
many other denominations also agree with Leo XIII that there can be no real conflict among faith,
reason, science, and revelation, provided that each remains within its own methods.
MOTIVES OF CREDIBILITY IN DIVINE REVELATION
When believers accept the divine origin and truth of a revelation, what criteria should they use?
Why do they believe a certain revelation or scripture is a revelation from God? These questions
prompt theologians to examine their credibility. Since divine revelation comes through the
mediation of various prophets, messengers, and inspired authors, it is important to examine
the reasons why these intermediaries are thought to be credible. Lawrence Feingold explains it
this way:
Since God speaks to all through intermediaries, it must be possible for mankind to be able to
recognize these intermediaries and become certain of their divine commission. Otherwise it
would be extremely imprudent to believe, for one would run the risk of being deceived by any
charlatan (or even an honestly mistaken person) about God’s revelation concerning the most
important matters of human life. If God is to effectively reveal Himself, He must make it clear
who are His true spokesmen and give them some kind of trustworthy sign or badge of their
divine credentials. These signs can be called motives of credibility. Motives of credibility are
supernatural signs that manifest the miraculous action of God. Their purpose is to show that
an alleged revelation from God is truly His Word, not that of a false prophet. These motives
allow us to make the transition from human faith in the word of the prophet to divine faith in
God who speaks through the prophet. (Feingold 2016, 55–56)
The motives of credibility are here described as supernatural. In other words, they transcend the cri-
teria of natural reason even though they are not in opposition to sound reasoning. The CCC, draw-
ing on the teachings of Vatican I, offers this assessment of the motives of credibility:
What moves us to believe is not the fact that revealed truths appear as true and intelligible in
the light of our natural reason: we believe “because of the authority of God himself who reveals
them, who can neither deceive nor be deceived.” So “that the submission of our faith might
nevertheless be in accordance with reason, God willed that external proofs of his Revelation
should be joined to the internal helps of the Holy Spirit.” Thus the miracles of Christ and the
saints, prophecies, the Church’s growth and holiness, and her fruitfulness and stability “are the
most certain signs of divine Revelation, adapted to the intelligence of all”; they are “motives of
credibility” (motiva credibilitatis), which show that the assent of faith is “by no means a blind
impulse of the mind” (CCC 1997, 156).
Vatican I points to three main motives of credibility: miracles, the fulfillment of prophecies, and the
“marvelous propagation, eminent holiness, and inexhaustible fruitfulness in everything that is good
with her catholic unity and invincible stability” (Denzinger-Hünermann 2012, no. 3289). In addi-
tion to these motives, some Christians will add “the holiness of revealed doctrine, especially of the
person of Jesus Christ, and the correspondence of this Revelation with the ‘reasons of the heart’”
(Feingold 2016, 57).
Skeptics, of course, will challenge all of these motives of credibility. They will question whether
the miracles reported in scripture can be verified. They will also claim that the fulfillment of proph-
ecies recorded in the Bible is something created by the biblical authors. They will also point to the
corruption and sin found within the church as a reason for discrediting claims of holiness and fruit-
fulness. Finally, they will argue that the holiness of Jesus is nonverifiable because it is testified by
writings whose historical reliability can be challenged. Other skeptics might acknowledge the holi-
ness and goodness of Jesus, but they will claim these are natural rather than supernatural traits.
Some Christians like Thomas Aquinas will add other reasons for credibility such as the growth
of the church in spite of terrible persecution and the ongoing miracles within the church (Feingold
2016, 78–83). Nonbelievers and non-Christians, to be sure, will challenge these reasons for credibil-
ity. They will offer natural explanations for the growth and expansion of the early church as well as
for the claims of miracles. With respect to miracles, though, the question should be raised whether
some skeptics dismiss the possibility of miracles because of their own worldview. To claim that
miracles cannot happen might be more a reflection of a certain philosophical perspective than a rig-
orous examination of the alleged miraculous event.
There are other motives of credibility, such as “the reasons of the heart” mentioned by the
French philosopher and scientist Blaise Pascal. These reasons of the heart are not blind impulses

but deep-seated desires for meaning, purpose, and love. The message of revelation is believed
because it resonates with profound aspirations of the human heart. Sometimes, it is compared with
falling in love. A young man who falls in love with a young woman might not be able to explain
logically why he is in love with this woman and not another. The same would be true of a woman
who feels an impulse to love a certain man and not another. Children while growing up love and
trust their parents even though they cannot articulate their reasons for loving them with sophistica-
tion or logic. For Christians, the “reasons of the heart” move them to accept the Gospel as divine
revelation because it “confirms our noblest intuitions about human dignity and promises to fulfill
our wildest aspirations for unlimited love, goodness, beauty, and communion” (Feingold 2016, 78).
WHY BELIEVE IN THIS REVELATION AND NOT ANOTHER?
Jews accept the Torah and the other books of the Hebrew Bible as divine revelation. Christians
accept the Hebrew Bible (the Old Testament) as well as the writings of the New Testament as bib-
lical revelation. Muslims accept the Qur ʾan as the supreme revelation of God that supersedes and
corrects prior revelations that had been corrupted. How, though, do people decide to accept one
revelation and not others? This is a complex question because these different religions have their
own criteria for judging whether or not a revelation is from God. For Jews, the Christian claim that
Jesus is the promised messiah of Israel is questioned on the basis of biblical criteria for discerning
the fulfillment of messianic expectations. This is evident in the arguments of the Jewish scholar
Nahmanides in the famous debate he held in Barcelona in 1263 with Fra Paulo Christiani, a
Dominican friar who had converted from Judaism (Glatzer 1969, 475–484).
Christians accept the divine revelation of the Torah and the other books of the Hebrew Bible, but
they believe in Jesus as the supreme revelation of God because he is God incarnate. They also believe
Jesus fulfills the messianic hopes of Israel, and they see his miracles and his resurrection from the dead
as a confirmation of his divine status. Christians likewise believe that the moral teachings of Jesus res-
onate with the deepest aspirations of the human heart and provide direction, meaning, and hope.
Muslims accept the basic teachings of Judaism and Christianity because they recognize Moses and
Jesus as prophets and messengers of God. They believe, however, that the Jewish understanding of
God’s plan is too restricted and needs to be expanded to include all of mankind. Muslims revere Jesus
and his virgin Mother Mary, but they see Christian teachings of the Trinity and the divinity of Christ
as corruptions that challenge the unity of God. They also reject the death of Jesus as a redeeming sac-
rifice because they do not believe that Jesus was crucified. The Qurʾan says that the Jews “did not slay
him [Jesus]: nor did they crucify him, but it appeared so unto them.… But God raised him up to
Himself, and God is Mighty, Wise” (Qurʾan 4:157–158; Nasr 2015, 262–263).
In this chapter, we do not need to resolve these disputed claims between Jews, Christians, and
Muslims over the divinely revealed status of certain scriptures. Sometimes the adherence to one rev-
elation as opposed to others is determined by one’s membership in a community. Those who grow
up Jewish are given reasons why Christian claims are mistaken. Those who grow up Christian are
taught to accept Jesus as the Word of God made flesh and the supreme revelation of God to the
human race. Those who grow up Muslim are taught that Muhammad is the final messenger of
God and the Qur ʾan is the final revelation that corrects and supersedes all other revelations. All
three of these religions, however, accept the reality of divine revelation and have reasons for accept-
ing their own scriptures as divinely revealed. In this sense, they share a common faith in the reality
of divine revelation even though they differ on the divine origin of certain writings.
PUBLIC REVELATIONS VERSUS PRIVATE REVELATIONS
There is a difference between public revelation such as is found in the Bible and private revelations
said to have been given to certain visionaries or small groups. Throughout history there have been
people who claim to have received visions and messages from God or who have seen the Blessed
Virgin Mary, angels, or saints. These private revelations are distinguished from the public revelation
of the revealed Word of God. The CCC offers this assessment:
Throughout the ages, there have been so-called “private” revelations, some of which have been
recognized by the authority of the Church. They do not belong, however, to the deposit of
faith. It is not their role to improve or complete Christ’s definitive Revelation, but to help live
more fully by it in a certain period of history. Guided by the Magisterium of the Church, the
sensus fidelium knows how to discern and welcome in these revelations whatever constitutes an
authentic call of Christ or his saints to the Church.
Christian faith cannot accept “revelations” that claim to surpass or correct the Revelation
of which Christ is the fulfillment, as is the case in certain non-Christian religions and also in
certain recent sects which base themselves on such “revelations” (CCC 1997, 67).
The distinction between private and public revelation is ultimately one of content. The private rev-
elations can offer direction and insights for individuals or groups, but they do not disclose or reveal
new truths to the faithful. Instead they help people to live more fully the teachings of the public rev-
elation.
From a Catholic and general Christian perspective, it is not possible to accept “revelations” that
claim to surpass or correct the revelation of which Christ is the fulfillment. This conviction is
expressed by Vatican II’s Dei Verbum: “The Christian dispensation, therefore, as the new and defin-
itive covenant, will never pass away and we now await no further new public revelation before the
glorious manifestation of our Lord Jesus Christ (see 1 Timothy 6:14 and Titus 2:13)” (Vatican II,
Dei Verbum, 1965, no. 4). Judaism, of course, has a similar perspective reflected in the ninth article
of faith of Maimonides: “The Torah will never be changed and there will never be any other law of
God” (Fellows 1979, 296). Islam in a similar way believes that Muhammad is the “seal of the
prophets” and that “no Prophet will appear after his time until the Day of Judgment” (Danner
1988, 236). This explains why Islam resisted the Bahá’í movement of nineteenth-century Persia:
it was believed that Mirza Ali Muhammad, otherwise known as the Bāb, was claiming supernatural
revelations as a type of prophet (Smart 1991, 436–437).
In the Catholic tradition, the distinction between public and private revelation has allowed
the church to approve a number of Marian apparitions as worthy of belief. The most notable of
these are the apparitions of Our Lady of Guadalupe in Mexico (1531), Our Lady of Lourdes in
France (1858), and Our Lady of Fatima in Portugal (1917). Claims of prophecies likewise must
be discerned and tested by church authorities whose role is to “test everything” and “retain what
is good” (1 Thessalonians 5:19). Catholic mystics like Saint John of the Cross, however, warn of
“God’s displeasure” with those seeking revelations and visions. One reason for his caution was his
belief that the devil could deceive people “seeking the knowledge of things through supernatural
means” (Ascent of Mt. Carmel, Book II, chap. 21, no. 4). This is because the evil one, through
conjecture, could make “many reasonable manifestations turn out to be true” (Ascent of Mt. Car-
mel, Book II, chap. 21, no. 7).
In 1978 the Sacred Congregation for the Doctrine of the Faith (CDF) issued “Norms regarding
the Manner of Proceeding in the Discernment of Presumed Apparitions or Revelations.” These
norms provide some basic negative and positive criteria for judging the credibility of the presumed
apparitions or revelations. The positive criteria include:
a. Moral certitude, or at least great probability of the existence of the fact, acquired by means of a
serious investigation;
b. Particular circumstances relative to the existence and to the nature of the fact, that is to say:
1. Personal qualities of the subject or of the subjects (in particular, psychological equilibrium,
honesty and rectitude of moral life, sincerity and habitual docility toward Ecclesiastical
Authority, the capacity to return to a normal regimen of a life of faith, etc.);
2. As regards revelation: true theological and spiritual doctrine and immune from error;
3. Healthy devotion and abundant and constant spiritual fruit (for example, spirit of prayer,
conversion, testimonies of charity, etc.).

The negative criteria include:

a. Manifest error concerning the fact.
b. Doctrinal errors attributed to God himself, or to the Blessed Virgin Mary, or to some saint in
their manifestations, taking into account however the possibility that the subject might have
added, even unconsciously, purely human elements or some error of the natural order to an
authentic supernatural revelation (cf. Saint Ignatius, Exercises, no. 336).
c. Evidence of a search for profit or gain strictly connected to the fact.
d. Gravely immoral acts committed by the subject or his or her followers when the fact occurred
or in connection with it.
e. Psychological disorder or psychopathic tendencies in the subject, that with certainty influenced
on the presumed supernatural fact, or psychosis, collective hysteria or other things of this kind.
The CDF notes that “these criteria, be they positive or negative, are not peremptory but rather
indicative, and they should be applied cumulatively or with some mutual convergence.”
The usual authority to conduct the investigation into reported claims of apparitions or revela-
tions is the local bishop. The national conference of bishops, however, can intervene with the prior
consent of the local ordinary or bishop. The Holy See can also intervene if asked by the local bishop
or a qualified group of the faithful. If, after an investigation, the local bishop, the conference of
bishops, or the Holy See judge the apparitions to be worthy of belief as supernatural, they can issue
a statement or letter voicing this approval. They can also issue a statement saying that the appari-
tions are not supernatural or that the supernatural character has not been established. Even if
approval is given, the faithful are not bound to give their assent of faith to the apparition or revela-
tion in question. The local bishop or even the pope can, however, encourage devotion to a particu-
lar apparition or private revelation, as is the case with the approved feasts of Guadalupe, Lourdes,
and Fatima.
THE POSSIBILITY OF REVELATION
As we have seen above, various motives of credibility have been proposed for establishing the credi-
bility of divine revelations. These motives or criteria for credibility presuppose a transcendent order
and a personal God who freely chooses to reveal himself to human beings. The matter is usually
understood in terms of the possibility of revelation.
The possibility of revelation can be viewed from the side of God and from the side of man.
From the side of God, the possibility is evident. If God is understood as transcendent, personal,
and omnipotent, then, by definition, he is able to communicate himself to human beings through
his freely chosen means of revelation. The key here is an understanding of God as personal, which
would imply divine freedom and wisdom in choosing how to reveal himself and his will. For Chris-
tians, the supreme revelation is Christ, the Word of God incarnate. Vatican II states: “In his good-
ness and wisdom God chose to reveal Himself and to make known to us the hidden purpose of His
will (cf. Ephesians 1:9) by which through Christ, the Word made flesh, man might in the Holy
Spirit have access to the Father and come to share in the divine nature (cf. Ephesians 2:18, 2 Peter
1:4)” (Vatican II, Dei Verbum, 1965, no. 2).
Some theists, however, might not understand God as personal. Deism, for example, has “a
rationalistic conception of the Divinity, based on human reason,” and it often leads to “the system-
atic exclusion of divine revelation” (Parente et al. 1951, 72). For most deists, God is an impersonal
transcendent ordering power, and divine revelation is found in the laws of nature, human reason,
and the order of the universe.
In addition to the deists, some theists, like the Catholic Modernists of the early twentieth cen-
tury, are skeptical of claims of the miraculous and historical reliability of the Bible. For them, “rev-
elation is merely the inevitable development (immanent in history) of man’s religious needs” (Rah-
ner 1969, 349). The Holy Office in 1907, under Pope Pius X, condemned the Modernist thesis
that revelation is “nothing else than the consciousness acquired by man of his relation with
God” (Denzinger-Hünermann 2012, no. 3420). The Modernists tended to reject objective rev-
elation, and they understood dogmas of the faith as human interpretations of certain phenom-
ena based on human needs. Traditional Christianity, as well as Judaism and Islam, however,
believe that divine revelation is objective. It is the self-communication of God who chooses to
reveal himself through chosen intermediaries (prophets, messengers, inspired authors) and his-
torical events.
From the human side, revelation is possible only if human beings are oriented toward the
divine. From the biblical perspective, human beings are made in the image and likeness of
God (Genesis 1:27). This image of God makes them capable of receiving God (capax Dei)—as
Augustine (De Trinitate XI, 8) and Aquinas (Summa theologiae 1a2ae, q. 13, a. 10) make clear.
More recent theologians have tried to ground the possibility of revelation from the human side
in an anthropology of transcendence. The Jesuit theologian Karl Rahner used the term “supernat-
ural existential” to point to a fundamental orientation of human nature toward the transcendent.
This orientation means that human beings “are open to the supernatural in the sense of a
self-communication of God” (Fries 1996, 265). There is in human nature itself an “a priori
capacity for revelation and faith” that is “identical with the whole transcendental character of
man” (Rahner 1969, 352). This means that human beings are open to revelation by the very structure
of their existence.
CHALLENGES TO THE POSSIBILITY OF REVELATION
Atheism denies the possibility of divine revelation because there is no God to reveal himself. Other
philosophies, such as deism and Catholic Modernism, might accept the existence of God, but they
are skeptical about miracles and claims of supernatural revelation. Others, though, object to the
notion of divine revelation on epistemological or even moral grounds. From the epistemological
side, claims of supernatural revelation are viewed with skepticism, especially those that rely on the
testimony of witnesses from centuries past. The question is often raised whether there is any way
of verifying historical events such as the Exodus of the Jews from Egypt, the resurrection of Jesus,
or the miraculous revelation of the Qur ʾan. Some thinkers also claim that divine revelation is impos-
sible because historical events can ultimately be explained by natural causes rather than supernatural
intervention.
Others object to claims of revelation on moral grounds. They see “an appeal to revelation as a
shortcut to the answers of life’s problems” and a threat to human autonomy, which requires
“man’s taking responsibility for his own life” (Nichols 1991, 76). Others claim that human free-
dom “cannot be reconciled with the affirmation of a Lord who is the author and purpose of all
things” (Vatican II, Gaudium et Spes, 1965, no. 20). The notion of a transcendent God who
reveals his will and demands obedience is seen as a moral threat to human freedom and intelli-
gence. Those who believe in divine revelation see these reactions as unjustified. Because God is
good his revelations are motivated by love for human beings and not a desire to dominate them
and take away their freedom.
For those who believe in God, faith in divine revelation is fundamental. Human beings need to
know about God in order to love and follow him. Special divine revelation is needed to communi-
cate the identity and will of God beyond what natural reason can know. Faith is the response to the
God who reveals himself, and faith is only possible within a theistic worldview. Such a theistic
worldview includes also an understanding of human beings as open to God and capable of discern-
ing authentic divine revelation by various means. Faith, therefore, entails reason, because there must
be a rational inquiry into claims of revelation that is supported also by the “reasons of the heart.”
Those who deny the reality or even the possibility of divine revelation do so because they adhere
to worldviews that do not allow for the existence of a transcendent personal God who freely chooses
to reveal himself. The question, though, must be raised whether this rejection of divine revelation is
a priori or based on rational, objective investigation.

Bibliography
Bettenson, Henry, and Chris Maunder, eds. Documents of the Christian Latourelle, René. Theology of Revelation. Staten Island, NY: Alba House,
Church. 4th ed. Oxford: Oxford University Press, 2011. 1966.
Brassard, Francis. “Buddhism and Christianity.” In Catholic Engagement with Muller, Richard A. Dictionary of Latin and Greek Theological Terms. Grand
World Religions, edited by Karl J. Becker and Ilaria Morali, 438–443. Rapids, MI: Baker Academic, 1985.
Maryknoll, NY: Orbis Books, 2010.
Naeem, Fuad, and Waleed El-Ansary. “Islamic Philosophy.” In The New
Catechism of the Catholic Church. 2nd ed. Vatican City: Libreria Editrice Catholic Encyclopedia, Supplement 2012–2013: Ethics and Philosophy,
Vaticana, 1997. edited by Robert Fastiggi, 812–817. Detroit: Gale Cengage Learning,
Congregation for the Doctrine of the Faith. Norms regarding the Manner of 2013.
Proceeding in the Discernment of Presumed Apparitions or Revelations. 24 Nasr, Seyyed Hossein, ed. The Study Quran: A New Translation and
February 1978. http://www.vatican.va/roman_curia/congregations/cfaith Commentary. New York: HarperOne, 2015.
/documents/rc_con_cfaith_doc_19780225_norme-apparizioni_en.html.
Newman, John Henry. An Essay on the Development of Christian Doctrine.
Dandekar, R. N. “Vedas.” In The Encyclopedia of Religion. Vol. 15, edited by Notre Dame, IN: University of Notre Dame Press, 1989.
Mircea Eliade, 214–217. New York: Macmillan, 1987.
Nichols, Aidan. The Shape of Catholic Theology: An Introduction to Its Sources,
Danner, Victor. The Islamic Tradition: An Introduction. Amity, NY: Amity Principles, and History. Collegeville, MN: Liturgical Press, 1991.
House, 1988.
O’Collins, Gerald, and Edward G. Farrugia. A Concise Dictionary of Theology.
Deninger, Johannes. “Revelation.” In The Encyclopedia of Religion. Vol. 12, New York: Paulist Press, 1991.
edited by Mircea Eliade, 356–362. New York: Macmillan, 1987.
Parente, Pietro, et al. Dictionary of Dogmatic Theology, translated by
Denzinger, Heinrich, comp., and Peter Hünermann, ed. Compendium of Emmanuel Doronzo. Milwaukee, WI: Bruce, 1951.
Creeds, Definitions, and Declarations on Matters of Faith and Morals. 43rd
ed. English edition edited by Robert Fastiggi and Anne Englund Nash. Pickar, C. H. “Faith: In the Bible.” In New Catholic Encyclopedia, 2nd ed.,
San Francisco: Ignatius Press, 2012. vol. 5, edited by Berard L. Marthaler, 589–593. Detroit, MI: Thomson
Gale, 2003.
Dulles, Avery. Models of Revelation. Garden City, NY: Doubleday,
1983. Rahner, Karl. “Revelation.” In Sacramentum Mundi: An Encyclopedia of
Theology, vol. 5, edited by Karl Rahner. New York: Herder and Herder,
Dulles, Avery. “Revelation, Theology of.” In New Catholic Encyclopedia, 2nd 1969.
ed., vol. 12, edited by Berard L. Marthaler, 193–198. Detroit: Thomson
Gale, 2003. Reynolds, Gabriel Said. The Emergence of Islam: Classical Traditions in
Contemporary Perspective. Minneapolis, MN: Fortress Press, 2012.
Esposito, John L. Islam: The Straight Path. 3rd ed. New York: Oxford
University Press, 1998. Sacred Congregation for the Doctrine of the Faith. “Norms regarding the
Manner of Proceeding in the Discernment of Presumed Apparitions or
Feingold, Lawrence. Faith Comes from What Is Heard: An Introduction to
Revelations.” http://www.vatican.va/roman_curia/congregations/cfaith
Fundamental Theology. Steubenville, OH: Emmaus Academic, 2016.
/documents/rc_con_cfaith_doc_19780225_norme-apparizioni_en.html.
Fellows, Ward. Religions East and West. New York: Holt, Rinehart and
Scheeben, Matthias Joseph. The Mysteries of Christianity. Translated by Cyril
Winston, 1979.
Vollert. St. Louis: B. Herder, 1946.
Fries, Heinrich. Fundamental Theology. Translated by Robert J. Daly.
Smart, Ninian. The Religious Experience. 4th ed. New York: Macmillan,
Washington, DC: Catholic University of America Press, 1996.
1991.
Glatzer, Nahum N., ed. The Judaic Tradition. Boston: Beacon Press, 1969.
Stelten, Leo F. Dictionary of Ecclesiastical Latin. Peabody, MA: Hendrickson,
Hyman, Arthur, James J. Walsh, and Thomas Williams, eds. Philosophy in 1995.
the Middle Ages: The Christian, Islamic, and Jewish Traditions. 3rd ed.
Suárez, Francisco. Opera Omnia. Vol. 12. Edited by Carlo Berton. Paris: L.
Indianapolis, IN: Hackett, 2010.
Vivès, 1858.
John of the Cross, Saint The Ascent of Mount Carmel. In The Collected Works
Vatican II. Dei Verbum. [Dogmatic constitution on divine revelation.] 18
of St. John of the Cross, translated by Kieran Kavanaugh, and Otilio
November 1965. http://www.vatican.va/archive/hist_councils/ii_vatican
Rodriguez. Washington, DC: ICS, 1973.
_council/documents/vat-ii_const_19651118_dei-verbum_en.html.
John Paul II. Fides et Ratio. [Encyclical letter on the relationship between
Vatican II. Gaudium et Spes. [Pastoral constitution on the church in the
faith and reason.] 14 September 1998. http://w2.vatican.va/content
modern world.] 7 December 1965.
/john-paul-ii/en/encyclicals/documents/hf_jp-ii_enc_14091998_fides-et
-ratio.html. Ware, Timothy (Bishop Kallistos of Diokleia). The Orthodox Church. New
ed. London: Penguin, 1993.
Kramer, Kenneth. World Scriptures: An Introduction to Comparative Religions.
New York: Paulist Press, 1986. Wicks, Jared. Doing Theology. New York: Paulist Press, 2009.
TOPIC 6
Faith and Revelation: Atheism

Evan Fales
Emeritus
University of Iowa
Long tradition connects revelation with faith, understood as trusting the truth of revelation, the divine
source of which is assured by miracles. Recent alternative conceptions of faith and revelation (Søren
Kierkegaard, William Clifford, William James) are critically examined, and difficulties in distinguishing
genuine revelation from imposture and deception are evaluated, including many-contenders objections and
the evidential weight of religious experience. Finally, we examine the difficulties that attend the reception
and interpretation of putative revelations, even assuming divine provenance.
FAITH OF THE FATHERS
Now faith is the assurance of things hoped for, the conviction of things not seen. By faith we
understand that the world (κόσμος) was created by the word of God, so that what is seen was made
out of things which do not appear.
HEBREWS 11:1–3, REVISED STANDARD VERSION TRANSLATION
Reflection on the nature and value of faith goes back to the earliest strands of the Judeo-
Christian traditions, and surely antedates them considerably. At the head of this chapter is a famous
passage from Hebrews, singled out because it is for our purposes noteworthy in two ways. First, it
attests an intimate connection between faith and revelation, made more explicit in what follows
(e.g., “These all [cp. infra] died in faith, not having received what was promised, but having seen
it and greeted it from afar” Hebrews 11:13). Second, the unknown author of Hebrews (call this
author “Paul”) fleshes out his case for faith by rehearsing a long list of stories from the Hebrew
Scriptures about Jewish heroes who lived by faith and who were, or whose posterity was, richly
rewarded. They did not (literally) see what was promised, but they heard the divine commands
and promises, and dedicated themselves to the divine commission. And indeed, without fail, God’s
promises were fulfilled by him, even if only after they had died. “Therefore,” concludes “Paul,”
“since we are surrounded by such a cloud of witnesses, let us also lay aside every weight … and
let us run with perseverance the race that is set before us” (Hebrews 12:1). So, “Paul” insists, accept-
ing what we do not see, on the strength of God’s word, which is heard, is like money in the bank:
it’s a sure thing. This makes “Paul” look for all the world like an evidentialist; he is making an
inductive argument for the conclusion that God’s revelations are as good as gold, hence are to be
trusted. Faith, pretty clearly, is on “Paul’s” conception of it, a matter of commitment to some truth,
not on the strength of “direct” perceptual acquaintance with it, but on the strength of someone’s
testimony—in particular, God’s say-so, or the say-so of a prophet, to whom God has spoken.
The trust in question has two dimensions. First, it involves acceptance of the truth of what is
communicated in a revelation (or that a promise will be fulfilled); second, it requires acting in ways
that satisfy whatever is required of the recipients of a revelation by God. These themes appear
throughout the tradition but with variations, and, in particular, with the appearance of various
179
elaborations on the question of when revelation is a good bet

KEY CONCEPTS for belief and action. “Paul” concerns himself with the ques-
tion: how do we know that God is trustworthy? By the medie-
val period, this seems to have been settled: God, being necessar-
Apophaticism
ily morally impeccable and omniscient, will not traffic in
Basic Beliefs and Properly Basic Beliefs untruths. (One may note in passing that this is a shaky inference,
Doxastic Voluntarism as it remains to be shown that God could not have morally suf-
ficient reasons for lying to someone.) Focus thus shifted—for
Ecumenism
example, for Thomas Aquinas—to the demand for evidence that
Evidentialist a given prophetic tradition was to be trusted over others in pro-
Fideism viding insight into the content of the divine being’s revelatory
Higher Criticism; German Higher Criti- acts. One can trust what God says and wants, if only one can
cism be assured (in some non-question-begging way) of what it is that
he says and wants—and this is clearly a contested matter within
Internalist any given tradition. As Thomas Aquinas ([c. 1260] 1955) and
Many-Contenders Objection others saw it, faith is a matter of (rationally) believing the word
Pietism of God, accepting it because it is God’s word, and relying on
divine notarization of that word by means of miracles.
Reformed Epistemology
Aquinas appealed to miracles as providing a divine signa-
ture that could not be forged. The idea is that God—and only
God—can manufacture miracles, events that by their extraordi-
nary character are transparently beyond the powers of nature or other agents to produce and can be
recognized by human witnesses to be so. A second requirement is that these events be associated
with putative revelations in such a manner as to make evident their status as divine imprimaturs
for the revelations in question.
Neither of these epistemic conditions can easily be met. First, it is by no means easy to spec-
ify what it would be for an event to outstrip the powers of any natural provenience. Second, it
requires some imagination to specify scenarios such that, were they to occur under favorable con-
ditions, would be humanly recognizable as satisfying the first condition. In particular, deception,
magic, and the merely wondrous need to be eliminable as explanations. Third, it is no trivial mat-
ter to show that God—even an omnipotent God—can do what nature can’t; that is, to explain
what it is for divine agency to add to or cancel out the law-governed operations of nature. Fourth,
supposing that to be possible, the argument requires it to be shown that no being other than
God (in particular no mischievous or demonic being) could produce the putatively miraculous
occurrence.
All that is a fairly tall order; but there’s worse to come. David Hume famously pointed out that
most human knowledge of the miraculous comes, not from firsthand acquaintance with such events
but from the testimony of others—not God himself, but other human beings. In particular, the
miracles (and revelations) that matter most to Jews and Christians are ones that happened thousands
of years ago and can be known to us only via ancient accounts that appear to witness to them. What
are the conditions under which such accounts may properly be judged trustworthy? In “Of Mira-
cles” ([1758] 1955), Hume argued that, in the nature of the case, no evidence for the veracity of
such testimonial evidence can outstrip the reasons that point toward fraud or folly—that is, that
impugn the reliability of the miracle reports.
HUME’S LEGACY
These matters are discussed at some length in the chapter of this volume on miracles; it is necessary
to bring them in here because they have played a crucial role in the debate over the bona fides
of revelatory claims. That is not to say that there might not be other means by which a revelatory
experience can be certified: for example, it has been argued that, for a mystic, firsthand mystical
experiences can provide first-person rational certainty of a divine source. This, too, is highly
problematic—for one thing, because of the many-contenders objection. (The many-contenders

objection, which applies also to miracle-traditions, observes that there are competing revelation-
based religious systems that can’t all be true, and that there are no neutral criteria that permit one
to separate the goats from the sheep: the deceived from those, if any, to whom a genuine commu-
nion with God has been granted. The defender of tradition A will have to offer some debunking
explanation for the putative revelations/miracles appealed to in defense of competing tradition B,
and will have to show that A’s own appeals are immune to similar debunking.) Again, the reader
is referred to the discussions of religious experience in this volume for further elaboration and dis-
cussion of this line of defense for theism.
So it’s not smooth sailing for appeals to revelation as a means to knowing the divine and the
will of God. There are, however, some maneuvers in the arsenal of contemporary epistemology
that might provide a way of avoiding some of the rough seas. Take, for example, the thorny prob-
lem of demonic deception. Why couldn’t a demon produce wondrous miracles to give proof to
(and provide camouflage for) revelations that, innocent-seeming on the surface, hide some dark
purpose? One could reply that, though this might in principle be possible, God would never allow
it to happen. “For even the demons believe—and tremble [in fear of the power of God]” (James
2:19). And God, who loves every one of us, would never allow us to perish defenseless against the
machinations of the Devil. So one should rest assured that either God has deprived the demons of
miraculous powers, or else he has given us means of exposing miracles and revelations that are
tools of deception.
This line of thought bears some affinity with what has come to be known as reformed episte-
mology. The thought is that if (as theists believe) there is a God, then—as he is loving, perfect,
and all-powerful—he will not allow creatures with whom he desires a personal relationship to labor
under an epistemic cloud of doubt or deception that is beyond human remediation and thus infirms
knowledge of him. Thus one may rest confident in the assurance that, if there is a God, he will put
demonic powers on a short leash and indeed provide us with cognitive faculties that are cognitively
adequate to the task of gaining accurate knowledge of our world. That’s a big if, the skeptic
observes. But the reformed epistemologist has a reply. It is a big if, she concedes, but supposing
it’s true, then the believer can know many things about God, including that God exists. It is not
a condition of knowledge, according to her view, that we must antecedently know or have assurance
of the reliability of our cognitive faculties. Hyperbolic doubts about that can be ruled out of court.
This sort of view—to be sure, greatly oversimplified here—is representative of a range of epis-
temologies that try to evade the question begging that plagues traditional attempts to avoid skepti-
cism, by denying that we need to prove the veracity of our senses and other cognitive faculties to
have knowledge; it suffices that these faculties are in fact up to snuff. One way to put it is that
externalist epistemologies (as they are called) deny that, in order to know that some proposition P
is true, one must also know that one knows that P. Readers will have to judge for themselves
whether it isn’t (as internalists stoutly maintain) indeed begging the question at stake just to take
for granted that our cognitive faculties are functioning properly in a felicitous environment. Or,
to bring the matter back to our present discussion, whether one can safely just assume that there
is no demonic trickery we’re incapable of exposing by the use of our faculties (see Plantinga 1993,
chaps. 1 and 2).
KIERKEGAARD
An even more radical turn in Christian reflections on the nature of religious faith is marked by the
nineteenth-century writings of Søren Kierkegaard, whose thinking had a profound effect on subse-
quent Protestant thinking. In Fear and Trembling ([1843] 2006) and elsewhere, Kierkegaard consid-
ers the task of faith, and finds it overwhelming, indeed paradoxical. He agonizes over the story of
Abraham, the “Father of faith,” to whom God appears and commands that he take his son, Isaac,
the son whom he loves, the son of the Promise, and ascend Mount Moriah, there to build an altar
and sacrifice Isaac. What makes this command morally and religiously terrifying (to say nothing of

emotionally catastrophic) is not only that it contravenes the signal commandment of the Noahic
covenant (Genesis 9:6)—thou shalt not kill—and the fundamental duty of a father to his son,
but that God had previously promised Abraham, as a reward for his faith, that through Isaac he
would become the father of a great nation, of people as numerous as the stars (Genesis 15:5) with
whom God will “establish an everlasting covenant” (Genesis 18:19). In the story of the Akedah
(Genesis 22), God seems to behave toward Abraham like a moral monster. Yet Abraham dutifully
obeys. We are not told what his thoughts are as he takes his son to Mount Moriah; only that he
prepares to do God’s bidding, in spite of what Kierkegaard calls “the ethical.” So Kierkegaard (like
many others through the centuries) tries to reconstruct Abraham’s thoughts in some way that makes
sense of his relationship to God, and his steadfast faith in God. In the end Kierkegaard—or rather
his pseudonymous Johannes de Silentio—gives up: to have faith, to be a Knight of Faith, as Kierke-
gaard designates him, is a mystery beyond comprehension, the mystery of a state whose achievement
is unfathomable, accomplished only rarely, by a few who may outwardly seem to be quite ordinary
people. It is a mystery that embraces the paradox, embraces something not merely unlikely, but
even impossible: for Abraham, the paradox is that his young son should die on an altar, a sacrifice
to a morally incomprehensible but perfect God, and yet come to father a great nation as a result
of the Promise.
For Christians, in Kierkegaard’s view, the task of faith is to grasp the paradox of the Incarnation—
an event that Kierkegaard considers to be a logical impossibility. How, he wonders, is this act of belief
possible? And why should God demand of his children such a task? The answer he offers us is not sim-
ple. But at its core lies the idea that communion between God and man is, so far from being a churchly
affair, the most intimate and inescapably private of relationships, one involving a kind of trust that
escapes normal thought. In Kierkegaard’s view, the greater the risk one takes in performing an act of
trust, the greater the trust. Faith in God requires nothing less than the greatest of imaginable risks.
Now the greatness of a risk increases as the unlikelihood that the outcome of the act will be
successful, and the magnitude of what is at stake, increases. And, ordinarily, one measures such like-
lihoods by considering the empirical evidence one has in relation to whether a given act, in a given
circumstance, will succeed. On that showing, the risk of faith for a Christian is to be estimated, in
part, by sizing up the evidence that, among other things, the Incarnation did really occur in Jesus—
that he was, somehow, both God and man. Moreover, what is at stake is the eternal fate of one’s
soul, which, in Kierkegaard’s arresting image, hangs like someone suspended over 70,000 fathoms
of the sea. But what if the historical evidence for the Incarnation begins to appear dubious? What
if, on close inspection of the Bible and other antiquities, one finds that the evidence is, at best,
equivocal? (It is not irrelevant that Kierkegaard was writing in the shadow of contemporary develop-
ments in German higher criticism that began to cast significant doubt on the historical accuracy of
the Bible and on traditional understandings of biblical revelation.) For Kierkegaard, so much the bet-
ter: for then our risk goes up in proportion to the diminution of our evidence. But the greatest con-
ceivable counterevidence respecting the Incarnation is not the status of the historical evidence; it is
that an absurdity—a flat contradiction—should be true. That trumps any merely historical specula-
tions. So the Christian stakes eternal happiness or oblivion on the truth of an impossibility; greater
faith hath no one.
This is, clearly, a radical conception of faith—and a radical departure from long tradition. It
emphasizes the utter interiority and privacy of faith, reconfiguring, among other things, the relation-
ship between God and human beings into something radically personal and hence beyond the busi-
ness or judgment of merely human institutions. Because of this, there was little love lost between
Kierkegaard and his national Danish (Lutheran) Church; nevertheless, his writings exerted a pro-
found influence on subsequent Protestant theology (much less so on Catholicism).
On a more pedestrian level, one may suspect Kierkegaard to have given support to Christian
fideism and pietism. It also appears to open wide the floodgates (already well ajar) to idiosyncratic
interpretations of revelation (i.e., scripture), and, somewhat ironically, to free up the task of Bible
criticism from the shackles of orthodox doctrine. At its worst, one wonders whether Kierkegaard’s
thinking contributed to the corrupted conception of faith as an act that attests to the freedom to
believe whatever one wishes. (It’s a corruption because, on Kierkegaard’s view, simply believing
what one wishes is the mark of insanity. The Knight of Faith, rather—and here lies the paradox—
holds fast to both the verdicts delivered by the best empirical evidence and “interior” truth of the
absurd.) The pedestrian view, which seems to permeate certain quarters of modern culture, presup-
poses doxastic voluntarism—that is, the view that what one believes is something that can be freely
chosen. But it is easy to show that doxastic voluntarism is something of a psychological myth.
Here is straightforward experimental proof: Show a classroom full of students a $50 bill. Ask
them to close their eyes. Tell them that the bill goes to the first student who can bring him- or her-
self genuinely to believe that the lights in the room (which are on) are off. (Cartesian doubt doesn’t
count.) You will find (presuming student honesty) that your $50 are quite safe. This does not
address the complexities of belief formation, to be sure; it remains a challenge to explain how reli-
gious beliefs come to be formed and held firm. When faith is just a matter of leaping across gaps
in the evidence, it becomes the pedestrian version of faith as expressed sardonically by Mark Twain:
“Faith is believing what you know ain’t so” ([1897] 1989, epigraph to chap. 12). The genuine Kier-
kegaardian article is both more puzzling and more poignant. Ultimately, however, it holds the door
open to a kind of spiritual chaos. Once accepting contradictions becomes the hurdle one must clear
to have faith in its purest and most exalted form, it seems one has abandoned the precincts of debate
altogether. Let that which cannot be spoken not be spoken. And let us therefore return to more
classical conceptions of faith and of revelation.
UNETHICAL FAITH
First, it is worth stopping to consider an argument that was offered by the late-nineteenth-century
mathematician William Clifford (1999 [1877]), no doubt with the influence of Kierkegaard and
Blaise Pascal (see discussion of Pascal’s wager in the chapter on Prudential/Pragmatic Arguments in
this volume) in mind. Clifford maintains that belief in the absence of adequate evidence is not merely
foolhardy but immoral. His argument is predicated on the obvious fact that belief guides action, and
false belief opens one to the heightened hazard of performing actions with untoward, even cata-
strophic, consequences. It may be one thing to subject oneself to those hazards; it is another matter
when, as is pervasive in our lives, our actions have consequences for others. (Not that action guided
by true beliefs is proof against failure, but one can surely agree that the odds are better.) Even the
acceptance of unjustified claims that are inconsequential for action is harmful and therefore immoral,
Clifford argues, because this acceptance easily becomes habitual, and such habits are transmitted both
to consequential beliefs and to the general belief-forming practices of one’s culture. Thus belief not
supported by proper evidence is a malaise that can infect and corrupt an entire society.
In a famous reply to Clifford, William James ([1896] 1992) counters that there are conditions
under which it is legitimate to accept a theology even in the absence of dispositive evidence for its
truth. Unlike Kierkegaard, James is not suggesting that one accept a doctrine in the face of contrary
total evidence. Rather, he asks one to consider a belief regarding a question on which the evidence
leaves matters unsettled. Perhaps, then, it might be only on acceptance of that belief that one can
become cognizant of supporting evidence from which one would otherwise be shut off. (Think of
accepting friendship with someone who is a relative stranger; only a trustful reception will persuade
the stranger to reveal herself more fully.) Furthermore, while the strategy of skepticism respecting
unsupported propositions is a good way to avoid believing what is false, it sacrifices acceptance along
the way of many propositions that happen to be true. In contrast, if one adopts a more permissive
acceptance criterion for belief, one stands to gain many of these otherwise rejected true beliefs, even
if one risks increasing one’s stock of false ones. Perhaps that’s not such a bad trade-off.
In this, however, James is mistaken. Our doxastic attitudes toward propositions are, to begin
with, far more nuanced than only straightforward belief (or acceptance), suspension of belief, and
outright disbelief would suggest. We rank propositions (those we understand) in terms of degrees
of credibility, even if only roughly. If one does this properly, one does it, as Hume notes, in step
with one’s estimation of how strongly one’s total relevant evidence supports or disconfirms a claim.
Such estimates pervade our cognitive lives, and they are no luxury: they are critical to the ways we

judge and use cost-benefit analyses to guide our actions. (It might, for example, be well-advised to
risk an action, at poor odds of success, if the cost of failure is small enough and the reward for suc-
cess is large.) To ignore the range of epistemic probabilities that can be assigned to propositions, in
favor of either believing or not, and treating all believed propositions equally with respect to formu-
lating plans of action, would be catastrophic; it would make fools of us all. No intelligent person
would be indifferent to whether a proposition she believed to be true had strong evidential support
or weak. So James’s advice is not a recipe for practical wisdom.
WHEAT FROM CHAFF
If this is right, there is nothing for it but to return to the classical conception of faith as believing
(and/or trusting for the purposes of guidance in practical affairs) testimony on the strength of the
testifier’s say-so and good evidence of the testifier’s reliability. This requires us to consider the
criteria for judging the reliability of testifiers. It is not particularly difficult to discover the criteria
we ordinarily (and rightly) use if we are to do this rationally. These include, among others, evi-
dence respecting the competence of the testifier to make correct judgments with respect to the
propositions in question, estimates respecting the testifier’s honesty, probity, and lack of ulterior
motives for promulgating a lie, assessments of his own likely estimate of the chances of getting
caught in a lie. Beyond evidence regarding the credibility of the testifier, a judgment whether
the claims offered as revelation are true will rightly be influenced by any antecedent and inde-
pendent reason one has to believe that God exists, and that God has motivation for revealing
himself in the way the testimony presents. On the other hand, the richer the content of the
putative revelation, the smaller an antecedent probability one should assign to its being
(entirely) true.
So judging the veridicality of a putative revelation can be—indeed must be—the outcome of a
fairly complex line of reasoning; and that, unsurprisingly, is what one actually finds scholars of
religious texts (especially ancient ones) engaging in. Could there be a more direct way to reach a
conclusion as to the credentials of a putative revelation? Here again, reformed epistemology
might—indeed, does—offer an affirmative answer. The idea is that some of our beliefs are basic—
that is, formed “directly” (as opposed to resulting from a line of inference) in response to, for exam-
ple, sense experience, or intuition, or some other information-gathering faculty. A basic belief,
moreover, will be properly basic (that is, will prima facie have the standing of knowledge, and be a
legitimate foundation from which other known propositions may be inferred), provided that it is
formed as a result of a belief-forming process that is (roughly) well-designed for the purpose of true
belief-generation, functioning properly in an environment in which it is well suited to perform reli-
ably (see Plantinga 1993, chaps. 1 and 2). Once the notion of properly basic beliefs is in hand, the
reformed epistemologist is in a position to claim that, at least for those in whom certain capacities
to apprehend the divine are functioning well, many beliefs about God—for example, that a cer-
tain revelation is genuine and true—may be available as properly basic, hence not the result of
any inference from other, evidential beliefs but rather as “ground-level” knowledge that can be
had apart from any reasoning or appeal to evidence. After all, some beliefs must, so the argument
goes, form the basis for all our inferential knowledge, since inference must be from premises; and,
on pain of regress, one must allow that some ground-level propositions can stand “on their own
feet,” epistemically speaking.
Here the controversial claim is that, on the reformed view, it seems that just any proposition
might turn out to be properly basic for an epistemic agent. It will be so, provided that this agent
is (perhaps unknowingly) equipped with properly functioning, belief-generating capacities within
whose range the proposition lies. (Thus, for God, presumably every true proposition is properly
basic; God has no need of inference at all, as the divine light illuminates directly the truth of every
true proposition.)
But this is far too permissive. Suppose God endows someone (call her Knower, or K) with a
faculty by means of which beliefs spring unbidden and unanticipated into K’s mind. God ensures
the invariable truth of these beliefs, identifies the advent of each of them in K’s belief-box with
an audible “ding,” but doesn’t let K know that she’s in possession of this special faculty. K, of
course, also forms many other basic (undinged) beliefs, some of which are formed by defective
cognitive practices and are false. K realizes that some of her basic beliefs are dinged, and some
undinged, but, on reflection, cannot discern any other general difference between them. In partic-
ular, K can’t see any difference between the dinged and the undinged that bears on their epistemic
status.
Is one to say that her dinged beliefs are warranted, that they contribute to her store of knowl-
edge, while the undinged beliefs (because formed by unreliable means) are not? Most of us will
rebel—and rightly so—against this suggestion. If you agree, then you should bridle at the claim that
theistic beliefs can be properly basic, even if you cannot remember what experiences and chains of
reasoning led you to accept them. And hence you will find it implausible to suppose that the accep-
tance of a revelation as true, on the basis of another’s testimony, or even on the basis of your own
religious experience, can be properly basic, even if you cannot now reconstruct the (perhaps
implicit) lines of reasoning that led you from your evidence to that belief.
However, even if some theistic beliefs founded on revelatory experience are properly basic, it by
no means follows that they are warranted on full consideration. For they can claim such warrant
only if, in light of our total evidence, they remain undefeated by conflicting evidence. And, as it
happens, there are potent defeaters of revelatory claims close to hand. These are of several kinds.
Some allow the construction of many-contenders objections to revelatory claims. This sort of objec-
tion has already been encountered; it will be developed more fully in a form directed against revela-
tion as a source of knowledge. A second kind of defeater, which will also require discussion, is an
objection that invokes reasoning to the best explanation. The application of this kind of reasoning
to the present case comes in the form of a challenge to the evidential weight of revelatory experience
as support for beliefs that assume the truth of the revelation.
Here’s (roughly) how that goes: The theist, in appealing to revelation to provide knowledge of
God’s existence, his will, and so forth, implicitly takes the knowledge-conferring power of revelatory
experience to derive from the hypothesis that divine activity is the source of the experience. What
explains the fact that prophet P had revelatory experience R is that God intended to communicate
some information to P by means of causing P to have R, and hence proceeded to manufacture R
in P. Such an explanation is of exactly the same form as the kind of explanation that underlies
our confidence in the evidence of our senses. We take our visual experience to inform us, correctly,
that there is a Christmas cactus in bloom on the window sill because we take the best explanation of
our having certain red and green sense experiences to be the actual presence of a cactus on the sill
(together with other facts about our environment, such as the lights being on in the room). But, just
as a visual experience of cactus-like form could also be a dream experience or a hallucination, so,
too, a revelatory experience might be caused in a way that is deceptive. We (generally) rule out
dreaming and hallucination as explanations of our cactus experiences because, for the most part,
we judge on our total evidence that these are less plausible explanations than the presence of a phys-
ical cactus.
Such reasoning is not without its vulnerabilities; nor is the claim that one ordinarily goes,
explicitly and consciously, through such a line of thinking every time one judges one’s perceptual
experiences to convey knowledge of the surrounding world. The reasoning is implicit and forms a
kind of standing backdrop, originally acquired in infancy through considerable trial and error, to
our everyday realism about perceptual knowledge. It is, moreover, all too familiar how susceptible
such reasoning, once one tries to capture it formally, is to skeptical objections. But let us allow,
for the sake of argument, that the skeptical objections can be answered in defense of ordinary per-
ceptual realism. It by no means follows that parallel defenses can be mounted against challenges
to the veridicality of revelatory experiences. That is because there are salient differences between
ordinary sense experience and the religious experiences in question, and they tell against the reliabil-
ity of the latter as messengers carrying true information about the spiritual world. (For more on
Plantinga’s effort to immunize religious beliefs against defeat by contrary evidence, see Plantinga
2000, 473–484, and a critique by Dawes 2015.)

REVELATION ON TRIAL
Most apparent among these is the above-mentioned many-contenders objection, which begins with
the observation that as a matter of empirical fact, mystical and revelatory experiences can be found
worldwide, and across historical time, in a vast number of cultures and religious traditions. One
finds, perhaps not surprisingly, both some general commonalities among and many profound differ-
ences between the sorts of experiences people report across this range of occurrence. As for the com-
monalities, some of them concern the phenomenology of religious experiences, which can arguably
be classified into a relatively limited number of types in terms of “what it feels like” to undergo such
experiences. Cross-culturally, for example, one finds both ecstatic, uplifting experiences and ones
that are frightening, even terrifying. One finds experiences that seem to achieve some sort of union
between the subject and others, God, or the cosmos, and also ones that involve what feels like the
shrinking or even disappearance of the “self.” Many are life transforming (not always for the better)
(see Stace 1987; Zaehner 1961; Underhill [1911] 1974; Poulain [1901] 1978; James [1902] 1982).
On the other hand, the differences, which are quite dramatic, tend to be focused on the specific
content of the experiences and revelations that are communicated to the subject. These can be
highly stereotyped; but when they are, the stereotype is typically endemic to the traditions of the
subject’s “home” religion. Christians do not have intimate interactions with Vishnu; Hindus don’t
consort spiritually with the Virgin Mary. (Syncretistic religions do develop under certain conditions,
with parallel evolution of spiritual beings. Haitian Vodou, for instance, is an amalgam of Western
African [Yoruba] religious traditions carried to the new world by slaves, and Roman Catholicism.
And, mirabile dictu, the loa spirits that Vodou practitioners worship, and that possess their bodies
in trance states, have dual identities. Secretly, they are known by the initiate to be African gods;
publicly, they are identified as Catholic saints.)
At the moment, it is the differences that interest us. It is not just that there is enormous
variety—it is that the contents of the various experiences and associated revelations often come into
apparent conflict—and that the experiences themselves often appear to generate subjective certainty
that the revelations are true. On the face of it, however, these revelations cannot all be true, as they
are inconsistent. Qailertetang is venerated by the traditional Inuit, for whom Yahweh may at best be
alien and perhaps nonexistent (Inuit lack a supreme deity); conversely, Hildegard von Bingen, the
twelfth-century German Christian mystic, would surely have disavowed the allegiance-worthiness
of Qailertetang. Such conflicts both can and do arise intramurally as well: witness the fierce debates
over theological doctrine (often with appeals to divine illumination) between and within the various
Christian denominations.
On the face of it, we can’t all be right. What, then, are the options? One obvious response is
sectarian: namely, to argue, on the evidence, that a particular theological tradition has in fact got
it right. Such efforts have not, in the broad view, achieved much success. Disputes have sometimes
been settled but more often by force of arms or by cultural domination than by force of reasons.
And that is evidently because, while some theologies may boast greater theoretical virtues than
others, there is no decisive way to show the rights of the matter.
A second response is ecumenism. In its most general form, this approach rests on the claim that
“all” the religions (or all the ones in some favored subgroup—e.g., all “world” religions) are “at
heart” pursuing the same universal understanding but that each has achieved only a partial grasp
of the truth (Hick 1982). Here, the devil is in the details. How, faced with the immense welter
of wildly different spiritual ontologies, origin stories, religious values, ritual traditions, and the rest,
is one convincingly to tease out such a universal worldview?
A third option is extreme apophaticism: the view that the objects of religious worship are tran-
scendent in ways that preclude human understanding because the truths at issue seem utterly para-
doxical or unintelligible to us, for the divine lies beyond the grasp of finite minds. But that route
invites a kind of theological despair.
A fourth option is truth-relativism, the notion that there are no real contradictions—only
truths-for-religion-A and truths-for-religion-B. There is not enough space here to offer a thorough
analysis of this sort of position. Suffice it to say (a) that it collapses into radical skepticism, and
(b) that few theologians in any tradition, to my knowledge, have been willing consistently to
maintain such a position. Nor have they been willing to accept a fifth option, namely noncogni-
tivism, according to which theological claims are not “about” anything at all; they have the sur-
face appearance of assertions that are true or false, but are neither true nor false because they do
not assert anything. This last family of views was used by many naturalistic philosophers—in
particular, the logical positivists—to deny that religious discourse had any cognitive content.
But this view of the emptiness of religious assertions is now, and rightly, almost entirely a thing
of the past.
NATURALISTIC EXPLANATIONS
Are there other theistic options? No doubt—but the terrain does not leave much scope for opti-
mism. Instead, it provides support for the second kind of defeater of the claim that religious experi-
ence confers knowledge of the divine. That second defeater is an argument to the best explanation,
to the effect that religious and revelatory experiences can be better explained by naturalistic science
than by appeal to a spirit realm. Indeed, one has here a surfeit of naturalistic approaches. It is pos-
sible only to summarize briefly a few that seem promising enough to pose a strong challenge to the
spiritual explanations of religious experience. But first, one should note that there are spiritual expla-
nations that themselves offer a serious challenge to theistic ones. They are explanations that appeal
to other spirit-beings, in particular demonic ones, to debunk the religious bona fides of experiences
that seem to reveal the divine. Let us set these aside, though the theist cannot take them lightly.
One may begin by noting that mystical and revelatory experiences are thoroughly at home
among the clinically insane. Such experiences are common in certain forms of schizophrenia, para-
noia, bipolar disorders, and a variety of psychotic states, as well as with migraines and the use of cer-
tain drugs. They can be produced artificially as well, by electrostimulation of certain brain loci.
These facts directly suggest that certain sorts of brain activity are instrumental in the production
of such experiences. This is not to suggest that mystics are crazy; though, in fact, a good many
famous mystics and prophets have displayed (other) symptoms of mental disturbance: Hildegard
von Bingen, the mystics Teresa of Avila and Margery Kempe, Saint Francis of Assisi, the novelist
Hermann Hesse, and of course others—but it certainly encourages the thought that brain processes
mediate such experiences, and that investigation of brain physiology might tell us something about
their character. This has, indeed, become a productive line of research, with results that connect cer-
tain features of such experience with brain loci known to process and generate such experiences—
and with certain mental pathologies. Indeed, the association with pathology has been understood
for a long time; medieval mystics such as Teresa of Avila were quite concerned to discriminate
between pathology (Teresa called it “melancholia”), demonic influence, and the real article. But
we do know that brain activity, however caused, is quite sufficient to trigger such mystical episodes.
Divine contact is not only explanatorily superfluous, but in fact it complicates the task of explana-
tion without any gain in explanatory power, since the theist will have to postulate causes other than
divine communication for pathological, delusive (i.e., pagan), and artificially induced mysticism.
Physiological explanations can be supported by findings from psychology and anthropology
that show there to be definite social contexts that encourage and influence revelatory experiences.
Again, it is not clear how much of this the theist can concede without defanging claims of divine
inspiration. One cannot simply graft a supernaturalistic explanation onto the known natural factors
that are implicated in prophetic and mystical activity, since the theist will have to resort to special
pleading to set apart putative divine revelations, and will inevitably have recourse to naturalistic
explanations (or demonic activity?) to explain the rest.
It is worth noting that the problem of distinguishing bona fide revelatory experiences from
counterfeit has been with us for probably as long as such experiences have been a socially recognized
part of human culture. In the Hebrew Bible, for example, one finds Gideon asking God to perform
a miracle to confirm that it was really the divine will that Gideon defeat the Midianites and Amale-
kites when he heard the command to war against them (Judges 6); similarly, Elijah challenges the

prophets of Baal to a contest to see whose God can set alight a sacrifice laid on an altar atop Mount
Carmel. Such displays of divine power, had they occurred, would no doubt be quite convincing.
But, short of miracles, it is much harder to find signposts of a divine author of a revelation. The
Counter-Reformation nun Teresa of Avila offered the other nuns in her Carmelite convent a num-
ber of criteria for discerning the genuine from the demonic or deranged; but none of them carries
similar conviction. (See Teresa of Avila 1980, esp. the Life and The Interior Castle, through which
these criteria are scattered. For an enumeration and discussion, see Fales 1996.) Most of them are
either question begging or not shown to have any demonstrable force. (Thus it is question begging
to appeal, as Teresa does, to the conformity of an alleged revelation to scripture, or approval by
one’s father confessor, and no reason is given to think that the Devil cannot duplicate, as Teresa
avers, the speed with which a divine message is allegedly conveyed.)
PROPHETS AND PROBABILITIES
In this connection, it is worth singling out one type of revelatory experience that is directly subject
to empirical confirmation or disconfirmation: prophetic utterances—if they meet certain precondi-
tions. For a strong record of accuracy with prophecies that meet the requisite conditions will cer-
tainly provide not insubstantial evidence of supernatural assistance, even if the foretold events are
in themselves quite mundane and unremarkable. There are seven conditions that must be met:
1. The prophecy must be verifiably made prior to the occurrence of the events foretold.
2. It must be stated in unambiguous language, not susceptible to multiple plausible interpretations,
among which witnesses are, ex post facto, free to choose.
3. It must be specific and detailed enough to have quite narrow satisfaction conditions.
4. It must be beyond the powers of human beings accurately to guess at, with significant chance of
success, by using only normal human cognitive powers.
5. It must be beyond the powers of the prophet or the prophet’s allies or followers to fulfill by
ordinary means at their disposal.
6. An event that fulfills the prophecy must be demonstrably known to have occurred.
7. It must not involve commission of what might be called the crystal ball fallacy: namely, the
strategy of making many risky prophecies and then, ex post facto, cherry-picking some small
fraction of these that came to pass.
All of these conditions are a matter of common sense; none should require special defense.
Many, indeed, seem almost trivial. But a word is in order regarding some of them. The first condi-
tion, in particular, might seem superfluous. But it is important enough to have earned a special des-
ignation. A vaticinium ex eventu is a prophecy made after the event “foretold” has actually occurred,
and Bible scholars have shown that more than a few of the prophetic utterances in the Bible have
this characteristic. They obviously should be disqualified. Conditions 2–4 are designed, obviously
enough, to minimize the odds of a lucky guess. (Condition 2 prohibits commission of what might
be called the Nostradamus fallacy, after the supposed seer of the sixteenth century.) Condition 5 is
self-evident; it rules out, inter alia, self-fulfilling prophecies. It and condition 6 are, of course, not
innocent: they wreak havoc with the idea of prefiguration. It is not at all uncommon for prophecy-
fulfillment stories to be written about charismatic religious figures, especially leaders of new movements.
Condition 7 is meant to prevent even “lucky” guesses, if they come about as a result of a scattershot
approach to prophecy making.
Now, if a record of revelatory prophecy has a history of satisfaction of the above conditions,
along with a striking string of successes, it cannot be dismissed as lacking empirical support for
the claim of supernatural inspiration. One may next ask, naturally, whether there is any such record
of prophetic accuracy. Within the Judeo-Christian traditions, one of the best-known apologetical
gambits is to offer a long list of prophecies that were (supposedly) fulfilled; this strategy is especially
used in New Testament apologetics by way of claims about the ways in which Jesus fulfilled proph-
ecies issued by the prophets whose stories we have in the Hebrew Bible. What, then, should be said
about these? The perhaps surprising answer is that, to the best of my knowledge, the number of bib-
lical prophecies that satisfy the conditions listed above is zero. (Here is a challenge to theists: find an
exception.) A number of them fail the first condition, as noted. A fair number of them fail the sec-
ond. Some certainly fail the third condition (“You will hear of wars and rumors of wars,” Matthew
24:6). At least one—an alleged prediction by Jesus of the fall of the Temple—fails either condition
1 or condition 4, and arguably both. The fulfillments by Jesus of the Hebrew prophecies can be
accepted only if one has dispositive evidence that Jesus actually said or did these things, and further
evidence that, if he did them, they were not scripted performances, staged so as to present Jesus as
the Messiah.
It is also relevant, in pursuing this question, to inquire how many instances there are in the
Bible of failed prophetic proclamations. These, as it happens, are not particularly rare: examples in
the Hebrew Bible can be found at Isaiah 13:19–20, Ezekiel 26 (esp. verses 13–14, 21), Joel 3,
and, quite poignantly, Isaiah 62:8–9. Two that are especially noteworthy from the New Testament
are found at Matthew 12:38–42 and Matthew 16:27–28 (cf. Matthew 24:36 and 25:31–32, Mark
9:1 and 13:32–33, Luke 9:27). These are, arguably, the two prophecies most central to Christian
soteriology. But they are, on the face of it, well off the mark.
Matthew 12:38 is especially perplexing. In it, Jesus tells some scribes and Pharisees that “the
only sign” that will be given them is the sign of Jonah, for as that prophet spent three days and
three nights in the whale’s belly, so will the Son of man spend three days and three nights in the
heart of the earth (i.e., in Sheol; note that in Jewish lore, the whale, Leviathan, is a denizen of
the “deep” [tehom]—that is, the realm of death). This is the most chronologically explicit foretelling
by Jesus of his death and resurrection (in related sayings, Jesus uses the vaguer “[on] the third
(τῇ τρἰτῃ) day” or “after (μετα) the third day”), and it is Matthew’s Gospel that gives the most
chronologically explicit Passion narrative. On that narrative, Jesus dies on the cross at 3:00 p.m.
on a Friday, is buried around 6:00 p.m., and has emerged from the tomb at dawn (6:00 a.m.) on
Sunday—that is, pretty nearly a day and two nights, or half the predicted duration of his sojourn
in the realm of death. Not only is there a significant disparity here, but it is a disparity internal to
Matthew’s Gospel. (There are other, intra-gospel disparities, e.g., between the Synoptics’ account
on which Jesus dies on Passover Day, and the Gospel of John, which has it that he dies on the
Day of Preparation. These are, perhaps, less troubling.) Yet the author of Matthew seems quite
untroubled by this apparent difficulty.
Of course, there has been no dearth of attempts to harmonize Matthew’s Passion with Matthew
12, some quite ingenious, some far-fetched. Most look for evidence that the phrase “three days”
could, in ancient Judaism, denote a period shorter than seventy-two hours; others move back the
time of crucifixion. The former tend to look for recondite passages in the Jewish legal and prophetic
traditions that refer to three days in connection with a shorter period than seventy-two hours (e.g.,
Yerushalmi Shabbat 9.3, Hosea 6:2).
But on the face of it Jesus means precisely three days and nights: first, because the book of
Jonah speaks straightforwardly of that duration (Jonah 1:17), and it would be conversationally per-
verse for Jesus to describe to his listeners the “only sign” they will be given by means of a misdirec-
tion that refers rather to some obscure halachic ruling or remote prophetic passage, and second,
because the reference is made specifically to the book of Jonah, with its abundant use of the theme
of death and resurrection (Jonah’s prayer for rescue uses the same language as the death and resur-
rection psalms, e.g. Psalms 18, 28, 30–32, 69, and a good many others). More telling yet, perhaps,
is the fact that the proof texts to which appeal is made do not use the specific formulation “three
days and three nights” as Jonah does. Of course, this does not settle the matter, but perhaps one
should look elsewhere for a resolution. (In this case, and taking seriously the unlikelihood that Mat-
thew would allow such a blunder, there is a plausible way to resolve the contradiction—but not one
that would be congenial to modern-day orthodox Christians [Fales 2005]).
The other troublesome prophecy is given at Matthew 16:27–29 and cognate passages in Mark
and Luke. It has been the subject of a far greater amount of commentary and apologetic

maneuvering. But again, the meaning does seem quite clear, and there is plentiful evidence that the
early Christians indeed took the Second Coming—the Parousia—to be imminent. Paul evidently
did so (e.g., Romans 13:11–14; 1 Thessalonians 4:13–14); and when, after a few decades, the
prophecy had still not come to pass, one finds New Testament authors themselves backpedaling,
obviously in response to mounting expressions of impatience among the laity (e.g., 2 Peter 3:8–
10, probably from the early second century). Some have even argued, oddly, that the Transfigura-
tion scene (Matthew 17), which follows soon after the utterance of the prophecy, is the event to
which Jesus is alluding. But that can’t be right: what is being foretold is the Day of Judgment
(“For the Son of man is to come with his angels in the glory of his Father, and then he will repay
every man for what he has done,” Matthew 16:27; cf. Matthew 25:31–32).
However, this is not to suggest that these brief observations settle the matter of the Parousia
prophecy either. Readers who wish to pursue the matter will find a perplexing welter of arguments
for just about every conceivable understanding of this and the previous prophecy. But this serves
only to underline the crucial difficulty of satisfying the second and third constraints on evidential
prophecy. Here, even when one has what appears to be clear and precise language, the floodgates
for differing interpretations appear to be wide open. But this is not a difficulty confined to under-
standing prophecy: it applies to revelation generally.
INTERPRETATION
The problem is severe enough that recognition of it has led some thinkers, like the political theorist
and American revolutionary Thomas Paine, to argue that a genuine divine revelation would not be
given in a human language at all; for Paine, the divine will was, to the extent that it could be
known, to be gleaned from natural philosophy—that is, from observation of the creation, in partic-
ular the wonderful law-governed motions of the stars and planets.
But upon those who recognize a richer language of communication there falls the task of pro-
viding a general theory of interpretation up to the task of dissecting the myriad of tangles to which
linguistic communication falls prey. This is by no means a trivial undertaking, as the history of
Bible interpretation makes painfully clear. (As much can be said of almost any religious literary tra-
dition of significant age.) Our task is not made easier when the human author of a revelation is no
longer alive and able to answer questions. It is—obviously—made considerably more difficult when,
among others, the following problems must be faced: the revelation comes from the distant past,
from an alien culture, and requires translation from a foreign tongue; or was written at a time per-
haps unknown, by an unknown author or authors, or one about whom very little is known; or
comes to us in the form of documents, perhaps fragmentary, perhaps interpolated, which are, per-
haps, copies of copies of a speculative original text. As is well known, attempts to recover the mean-
ing of the biblical texts are beset by every single one of these vicissitudes. In consequence, and for
other reasons as well (e.g., archaeological evidence, pagan sources), for more than two hundred years
bible scholars have been debating how much of the material one finds in biblical narratives is histor-
ical. While a wide range of views can be found, it would be fair to say that the prevailing conclusion
among scholars who are not engaged in apologetics is that, despite the large number of verifiably
historical events, places, and individuals mentioned in the Bible, the greater part of the narratives
recount things that did not happen, cannot be independently verified, or differed dramatically from
their retelling in scripture.
This finding has long been resisted by conservative Christians and Jews, of course. Here a few
words about the doctrine of divine inspiration are in order, as it has played such a central role in
traditionalist reception of the canon. This doctrine itself has by no means been uniformly under-
stood. At one pole one finds the view that the Bible’s human authors were essentially God’s stenog-
raphers. They wrote down God’s words as he dictated them and whether or not they understood
them. So those very words, as they were originally given, carry infallible authority. A considerably
more modest suggestion is that the prophets heard the Word mediated through culture. Although
the message is timeless, it was cast into a form significantly determined by the prophets’ personal
history, cultural environment, and contemporary events. The exegete’s task is to identify these local
influences and to separate out the universal truths. A distinct, but related, doctrine is the view that
God himself tailors his message so as to enable his worshippers to best understand his will, in light
of the cultural and intellectual limitations that their historical circumstances impose. As human civ-
ilization progresses through different stages, from less to more advanced, God issues new marching
orders (though his ultimate aims are timeless) to shepherd human beings along the path to salvation.
This view has been pressed into service in order, among other things, to justify Christian revisions
of Jewish law, and to put some distance between the divine will and the morally repugnant com-
mandments to be found in various biblical texts (see, e.g., “Dispensationalism,” Theopedia).
However that might be, divine-inspiration views must face, in what we know about the forma-
tion and transmission of the canonical texts, a formidable challenge: they must reconcile doctrinal
purity with the vicissitudes of textual copying, and the deep political divisions and partisanship in
ancient Judaism and patristic Christianity over which texts, doctrines, and creedal formulations to
favor and promulgate and which to destroy—not to mention the role of sheer historical accident.
Given familiarity with this background, one cannot help but think it would be nothing less than
a miracle that the “true,” inspired message was preserved.
Prominent examples are the stories of the Exodus and blitzkrieg invasion of Canaan, as depicted
in Joshua, which archaeological findings have discredited, though they allow for much smaller
migrations of Semitic peoples between Palestine and Egypt and a gradual encroachment into
Canaan around the beginning of the first millennium BCE. In Exodus and Joshua—as in so many
other apparently historical narratives in the Bible—one evidently enters into the realm of myth.
This does not mean (as anthropologists now understand the function and reception of myth)
that myths are false stories, or—worse—just nonsense or children’s fables. Far from it. But it does
require a careful consideration of the interpretive strategies one needs to use to illuminate the pur-
pose, meaning, and significance of myths. That is itself a large enterprise, about which anthropology
has much to teach. Here it will have to suffice to say that, in general, it is most helpful to see the
primary function of myth to be deeply theoretical, to convey, typically in narrative form, profound
and often abstract understandings or specifications of the fundamental structuring principles of
social order. They are texts that try to answer, by means of illustrative stories, such basic questions
as what properly confers governing authority on a certain set of institutions, official roles, laws, cus-
toms, and the like. Every society needs conceptual tools for understanding the structure and proper
workings of social relationships, especially relationships that entail legitimate authority of some indi-
viduals over others. It is a large task, naturally, to show whether and how and in what respects the
narratives that constitute the canon in a given society are myths in this sense. Émile Durkheim
([1912] 2008) made fundamental progress in carrying out this task.
To my mind, the efforts of Durkheim, and of many anthropologists who have followed in his
footsteps, have been hamstrung by an inadequate account of social ontology. Durkheim himself
arrived at a key insight—that the gods are mental “projections” of society members’ experience of
the phenomena that create social solidarity and group identity among members of a clan or tribe.
But he proposed that the process was an unconscious one, and there are insufficient reasons for
thinking that. Why not consider the “natives,” even in “primitive” societies capable of intellectually
sophisticated theorizing about matters of vital importance to social integrity and stability? Why not
suppose that they knew that, for example, their clan totems were symbolic representations, or more
accurately, visible embodiments of social realities—the realities of clan and tribe as corporate entities?
If so, then might not the “stories” they tell about their totem animals be a way of encoding their
ways of understanding the foundations of their social order?
To some, this will seem far-fetched, especially as details are brought under analysis. But this
brings us, now, to a watershed issue that one must face if one is to engage in the interpretation of
sacred texts. The question is: regarding the creators of such texts and traditions, how charitable
should one be in one’s initial presumptions about the degree of intellectual sophistication, theoreti-
cal acumen, and level of commonsense contact with empirical reality? Of course, not everyone in a
given society is predisposed to engage in creative high-level theorizing, but everyone has an urgent
curiosity and desire to understand his or her environment. And, if anyone in a society is likely to

carry this curiosity to the point of sustained efforts at explanation, it would be those thinkers who
are the creators and transmitters of a society’s “charter”—its repository of legal traditions, customs,
arts (including storytelling), and judicial and moral wisdom. It is not hard to show that every learner
and user of a public language must possess a kit of fairly sophisticated tools of inference, as well as
perceptual capacities that are up to the task of discerning and individuating objects in a commonly
shared world of everyday objects. So all competent participants in a linguistic community must
share and wield antecedently given intellectual tools that form the bases of rational thought. Those
that are especially skilled in the use of language are, naturally, likely to considerably exceed the
threshold minimum of skills required for the business of mundane communication. All this leads
us to invoke a basic hermeneutical principle: a principle of charity, which mandates that, when a
text is subject to debated interpretations, preference should be given, other things being equal, to
an interpretation under which the text would make rational assertions, over interpretations on
which the author of the text appears to be spouting nonsense.
What, then—for example—of the Exodus story, once one surrenders one’s attachment to the
false belief that it narrates an actual history? Of course, the author(s) (probably sixth century
BCE) of the story might have thought they were writing a history, but on what rational evidence?
The alternative is to search for some purpose other than unvarnished history that might rationally
have engaged them, and in terms of which the construction of the story, in all its rich detail, can
be understood to reflect a serious and sustained effort to make sense of some aspects of the authors’
existence that human beings might naturally feel a need to understand. And here, one hypothesis
that deserves consideration is that the Exodus provides a kind of conceptual framework, in narrative
form, in terms of which Israel was to articulate its sense of identity, its legal traditions, and its mis-
sion as a nation: a political theory, in other words. If so, it would by no means have been a novel
undertaking in the ancient world. It is not difficult to argue, for example, that Virgil undertook
to write the Aeneid for the purpose of providing a similar scaffolding of legitimacy for the Roman
Empire (with its patron goddess Roma, clearly a personification of the Imperium) under the Cae-
sars. Indeed, it is an interesting guess that origin myths quite generally have, as a central piece of
business, the introduction and consecration of a system of relations in terms of which to deliberate
over what the legitimate demands of society on the individual are, and vice versa.
Here, then, is a challenge for theists who want to maintain the view that such ancient texts are
really primarily concerned with “spooky stuff”—gods, spirits, demons, immaterial souls, and the
like, that can intervene miraculously in the workings of nature. For social functions, corporate enti-
ties, social roles, social forces (constructive and destructive) are suspiciously like spooks. A nation is
not identical to its citizens, or its land, or its artifacts—in fact, it is not a physical thing at all, though
it can and must be embodied in items of the mentioned sorts. (Indeed, like the Body of Christ, it
can be individually embodied—in King Jesus or another king—and simultaneously or subsequently
multiply embodied in the members of the church or nation, and in the consecrated host or,
roughly, national flag.) The matter cannot be pursued further here but is presented only to suggest
one of the dimensions along which a nontheistic approach to understanding religious texts that are
on the face of it baffling might well be able to make better sense of the data than “traditional” the-
ology.
CONCLUDING THOUGHTS
It is appropriate to close with a final note about a claim, made by John Calvin, that there is, innate
in every human being, a sensus divinitatis, that is, that in addition to our innate and undeniable
rational faculties, each of us is born with a faculty whose natural function is to apprehend, however
dimly at first, the existence and love of God, creator of the universe—and that we have this ability
to sense or intuit God’s presence independently of any of the arguments from natural theology or
other sources of information. On the view in question, then, we can have, thanks to the sensus divi-
nitatis, properly basic beliefs, amounting to knowledge, of God apart from any evidence provided by
the ordinary senses.
For a theist, this might seem an attractive idea. But there is little evidence to support it. The
traditional home for innate ideas is a priori truths such as those of mathematics and logic. But
one distinctive feature of these truths, one that bears on their status as innate, is that they—at least
the simpler ones—command essentially universal agreement. A second distinctive feature is that
knowledge of at least the basic truths of logic and mathematics is a prerequisite for learning even
mundane facts about the world via sense perception. On its face, then, one can’t learn or be taught
such truths by empirical means—one needs already to have mastered them, at least implicitly, to
engage in that kind of learning at all. But none of this applies to the idea of God.
On the contrary, there are far fewer monotheistic religions than ones that recognize no gods at
all or recognize multiple gods, even a pantheon that includes no supreme deity. And that is hard to
square with the notion that all humans are endowed with a natural disposition to believe in a single
supreme being.
Bibliography
Alston, William. Perceiving God: The Epistemology of Religious Experience. Katz, Steven T., ed. “Language, Epistemology, and Mysticism.” In Mysticism
Ithaca, NY: Cornell University Press, 1991. and Philosophical Analysis, edited by Katz, 22–74. New York: Oxford
Clifford, William. “The Ethics of Belief.” (1877.) In The Ethics of Belief and
Other Essays, 70–96. Amherst, NY: Prometheus, 1999. Kierkegaard, Søren. Fear and Trembling. Edited by C. Stephen Evans and
Sylvia Walsh; translated by Sylvia Walsh. Cambridge, UK: Cambridge
Dawes, Gregory W. “Basic Beliefs and Christian Faith.” Religious Studies 51,
University Press, 2006. First published 1843.
no. 1 (2015): 61–74.
Lewis, I. M. Ecstatic Religion: A Study of Shamanism and Spirit Possession. 2nd
“Dispensationalism.” Theopedia. https://www.theopedia.com/dispensationalism.
ed. London: Routledge, 1989.
Durkheim, Émile. The Elementary Forms of the Religious Life. Translated
Locke, John. An Essay concerning Human Understanding. 2 vols. Edited by
by Joseph Ward Swain. Mineola, NY: Dover, 2008. First published
Alexander Campbell Fraser. Mineola, NY: Dover, 1959. See book 4,
1912.
chap. 8. First published 1689.
Ehrman, Bart. Misquoting Jesus: The Story behind Who Changed the Bible and
Plantinga, Alvin. Warrant and Proper Function. New York: Oxford
Why. San Francisco: HarperOne, 2005.
Fales, Evan. “Scientific Explanations of Mystical Experience, Part I: The
Plantinga, Alvin. Warranted Christian Belief. New York: Oxford University
Case of St. Teresa.” Religious Studies 32 (1996): 143–163.
Press, 2000.
Fales, Evan. “Taming the Tehom: the Sign of Jonah in Matthew.” In The Poulain, Augustin. The Graces of Interior Prayer: A Treatise on Mystical
Empty Tomb: Jesus beyond the Grave, edited by Robert M. Price and Theology. 6th ed. Translated by Leonora Yorke Smith. Westminster, VT:
Jeffery Jay Lowder, 307–348. Amherst, NY: Prometheus Books, 2005. Celtic Cross Books, 1978. First published 1901.
Forman, Robert K. C. Mysticism, Mind, Consciousness. Albany: State Stace, W. T. Mysticism and Philosophy. Los Angeles: Jeremy P. Tarcher,
University of New York Press, 1999. 1987.
Griaule, Marcel. Conversations with Ogotemmêli: An Introduction to Dogon Teresa of Avila. The Collected Works of St. Teresa of Avila. 2 vols. Translated
Religious Ideas. London: International African Institute/Oxford Univer- by Kieran Kavanaugh and Otilio Rodriguez. Washington, DC: Institute
sity Press, 1965. of Carmelite Studies, 1980.
Hick, John. God Has Many Names. Rev. ed. Philadelphia: Westminster Thomas Aquinas. On the Truth of the Catholic Faith: Summa Contra Gentiles.
Press, 1982. Translated by Anton C. Pegis. Garden City, NY: Image Books, 1955.
Hume, David. “Of Miracles.” Section 10 of An Inquiry into Human See chap. 3. First published c. 1260.
Understanding. (1758.) Edited by Charles W. Hendel. Indianapolis, IN: Twain, Mark. Following the Equator: A Journey around the World. (1897.)
Bobbs-Merrill, 1955. Mineola, NY: Dover, 1989.
James, William. The Varieties of Religious Experience: A Study in Human Underhill, Evelyn. Mysticism: A Study in the Nature and Development of
Nature. (1902.) Edited by Martin E. Marty. New York: Penguin, 1982. Man’s Spiritual Consciousness. New York: Penguin, 1974. First published
James, William. “The Will to Believe.” (1896.) In Readings in the Philosophy 1911.
of Religion, edited by Baruch A. Brody, 35–47. Englewood Cliffs, NJ: Zaehner, Robert C. Mysticism, Sacred and Profane: An Inquiry into Some Varieties
Prentice-Hall, 1992. of Praeternatural Experience. London: Oxford University Press, 1961.

TOPIC 7
Miracles: Theism
Ira M. Schnall
Lecturer (retired)
Bar-Ilan University, Ramat-Gan, Israel
In this chapter, we first examine what a miracle is supposed to be from a theistic point of view. Then we
consider whether, or to what extent, reports of miracles are to be believed. Finally, we deal with the role of
miracles in theistic religions, and in particular, whether miracles can establish the truth of theism.
WHAT IS A MIRACLE?
I suggest the following as a working definition: A miracle is an event that involves temporary sus-
pension of the natural order by a supernatural agent.
THE SUPERNATURAL AGENT

For most theists, God is the supernatural agent who performs miracles. God created the world
and is all-powerful as well as all-knowing and perfectly good. He is responsible for instituting
the natural order and has the power to abrogate it as He sees fit. God generally does not interrupt
the natural order, but He could do so at any time, and according to most theistic traditions, He
sometimes does so.
A human being—for example, a prophet or a priest—can be said to perform miracles but only
derivatively; in such cases it is God who is actually performing the miracles, either at the request of
or for the benefit of the human being. Thus, for example, the prophet Elijah did not actually bring
a dead child back to life, but rather God brought the child back to life at Elijah’s request (1 Kings
17:17–24); and Moses did not part the Red Sea, but rather God did so in order to save Moses and
the Israelites from the pursuing Egyptian army (Exodus 14:21). Generally, it seems more plausible,
from a theistic perspective, to understand biblical passages in which a prophet decrees that some-
thing should happen, and it does, not as the prophet’s directly making it happen but rather as God’s
making the prophet’s decree come true. (Actually, I find this issue quite baffling. For example, if I
were to say “Let that water turn to wine” and it did, would it feel any different to me whether God
turned the water to wine or He empowered me to do it?)
Some theists say that there are other supernatural beings, besides God, who can perform mira-
cles—namely, angels. Theists differ as to the nature and powers of angels. But most scriptural pas-
sages involving the activity of angels portray the angels as God’s messengers or agents; so even if
angels actually perform miracles, it is generally in God’s service that they do so. Thus we may say
that for theists, miracles are performed by God, either directly or through angels who serve as His
agents.
THE NATURAL ORDER

The concept of a miracle makes sense only against the background of a regular natural order. A mir-
acle is an exception to that order, an irregularity in an otherwise regular world. A miracle must be
something that goes against or beyond our normal, reasonable expectations; and in a disorderly,
195
Miracles: Theism
haphazard world, we could not have any reasonable expecta-

KEY CONCEPTS tions. A common way to express this point is to say that nature
operates according to laws, and a miracle is a violation of a law
of nature.
God
Theists differ with respect to God’s role in maintaining the
Miracles natural order. Some see the natural order as a system of causes
Naturalism and effects, originally created by God, but left for the most part
Supernatural Agent to operate on its own. That is, God, in creating the world,
endowed events of certain types with the power to bring about
events of certain other types, in a regular manner—for example,
He gave fire the power to burn, wounds the power to kill, and
eating bread the power to nourish. (These examples will serve. But there are perhaps better exam-
ples, ones that involve more direct and more uniform causality, perhaps on a more theoretical
level—e.g., God gave protons the power to attract electrons.)
Others assign to God a more active, ongoing role in maintaining the natural order. That is,
causes do not have in themselves the power (God-given or otherwise) to bring about their respective
effects; rather God is constantly making things happen in a regular manner—e.g., when there is fire,
God brings about burning; when a person is severely wounded, God brings about the person’s
death; when someone eats bread, God brings it about that he or she is nourished—and of course,
in these examples, God also brings about the fire, the wound, and the eating of the bread when
the usual antecedents of each these types of events occur, respectively. This view, usually associated
with the early modern philosopher Nicolas Malebranche, is called “occasionalism”—for it entails
that a natural event does not itself bring about its usual subsequent event, but rather the occurrence
of the first event (the “cause”) is merely the occasion on which God brings about the second event
(the “effect”).
A sort of compromise theistic view, associated with the early modern German philosopher
Gottfried Wilhelm Leibniz, is that the regularities in nature are due neither to any causal powers
in nature nor to God’s constant activity, but rather to God’s having arranged, from the time of
Creation, a “preestablished harmony” whereby events occur in a regular manner—that is, God
decreed (or so “programmed” the world), from the Beginning, that, for example, fire would gen-
erally be followed by burning, wounds would be followed by death, and eating bread would be
followed by nourishment.
SUSPENSION OF THE NATURAL ORDER

On the first view above, namely, that God endowed events in nature with causal powers, a miracle
would involve God’s intervening on a particular occasion either to neutralize some of those powers
or else simply to prevent the usual effect from occurring. God can also miraculously bring about an
effect in the absence of the usually necessary cause; for example, when God “spoke” to the Israelites
at Mount Sinai, what He did, on the view at hand, was create the sounds of an articulate voice
without the operation of any vocal apparatus of the sort that is generally causally necessary for the
production of such sounds. The idea is that God created the causal powers, and He has retained
the prerogative of suspending their operation as He sees fit.
On the second view, occasionalism, a miracle would involve God’s simply not acting, in a par-
ticular instance, in the way that He otherwise always acts in similar circumstances. For example, on
one occasion when a person is pierced through the heart, God may perform the miracle of not
bringing about that person’s death, though on all (or almost all) other occasions, before and after,
He brings about death whenever a person is pierced through the heart.
Finally, on the view that there is a preestablished harmony, a miracle involves God’s introduc-
ing a dissonance, so to speak, in the usual harmony—either by intervening then and there (e.g.,
though He decreed that whenever there is fire there is burning, He intervenes at a certain time
and place to overturn His decree just for that once) or by having preordained that this exception
to the usual order of things should occur at a certain time and place (e.g., He built into His original
Miracles: Theism
decree that whenever there is fire there is burning except at such-and-such a particular time and
place).
Several medieval philosophers, influenced by Aristotle, held that the world is eternal, and that
the natural causal order is necessary—which would render miracles, which involve departures
from the natural order, impossible. According to these philosophers, God did not create the nat-
ural order, and He does not have the power (nor does He have the desire) to abrogate it. Medieval
theists argued against this position. For example, the Islamic thinker al-Ghazālī argued that God,
in creating the world, instituted the causal order, and has the power to suspend it as He sees fit.
Furthermore, God is a person, who cares about His creatures and sometimes decides to intervene
in the course of events in order to benefit His creatures. The Jewish philosopher Moses Mai-
monides, though much more sympathetic to Aristotelianism than al-Ghazālī, nevertheless resisted
the doctrines of the eternity of the world and the necessity of causal laws. He argues that neither
the eternity of the world nor its creation in time can be demonstrated philosophically; therefore,
he is free to accept the belief, based on the Bible, that the world and its causal order were created
by God, and that therefore God has power over the causal order. He accepts this latter belief
because rejecting it would preclude belief in revelation, in miracles, and in God’s ability to keep
His promises to us, thus wreaking havoc with the foundations of Judaism (and, for that matter,
Islam and Christianity).
It should be noted that the mere fact that God created the natural order does not necessarily
entail that He has the power to control it (think of Dr. Frankenstein). Conversely, the mere fact
that the natural order is eternal does not necessarily preclude God’s power to control it. After all,
God is omnipotent, that is, able to do whatever is logically possible; and the necessity of natural
laws (at least as most modern thinkers conceive of it) is not logical necessity, so God should be able
to abrogate natural laws. Nevertheless, medieval thinkers seem to have assumed that creation implies
control and vice versa; so thinkers such as al-Ghazālī and Maimonides felt compelled to deny the
eternity of the world in order to make room for miracles.
ALTERNATIVE DEFINITIONS
COINCIDENCE MIRACLES
It has been argued that miracles need not involve the suspension of the natural order by a supernat-
ural agent. Some philosophers have suggested that we distinguish two different types of miracles—
“violation miracles” (which involve suspension of the natural order or violation of laws of nature)
and “coincidence miracles” (which do not). A coincidence miracle is a case in which two or more
causally unrelated sequences of events “converge,” or “coincide,” to yield an unexpected good result.
R. F. Holland gives the following example: a little boy is riding in his toy car, which gets stuck on a
railroad track; meanwhile a train is racing around the bend. The driver of the train, though he does
not see the boy on the track, faints (as a result of a causal chain of events that culminate in a blood
clot in his brain), and the brakes, which are set to engage when the driver ceases to exert pressure on
the controls, stop the train just a few feet from the boy. The boy’s mother, whose horror at the sight
of the train heading for her son was followed by relief and joy at his surviving unharmed, says, “It’s
a miracle”; and her saying so is appropriate, even though there was nothing supernatural involved in
the train stopping when it did, since it all happened in accordance with known laws of nature.
However, one might argue that in calling the event “a miracle,” the mother really does believe
that God intervened in the workings of nature to save her son, though she is mistaken. Alterna-
tively, one might suggest that the mother realizes that there was nothing supernatural going on,
and she is speaking figuratively—that is, she means that events unfolded as if God were intervening
in the natural order. So at least as far as this example shows, one can still maintain that real miracles
do involve God’s violating laws of nature, whereas surprising and beneficial coincidences can be
called “miracles” only in a figurative sense.
In any event, in what follows I shall limit my discussion to miracles involving supernatural sus-
pension of the natural order. This is because philosophers of religion are interested in miracles

Miracles: Theism
mainly insofar as miracles are supposed to support belief in God. Such support is strongest when it
seems that a particular miracle must be the work of a higher power; and this seems so most clearly
when the miracle involves violation of laws of nature.
RESULTS AND VALUES

In light of the above example, we may ask whether an event, in order to qualify as a miracle, must
have a good, or beneficial, outcome. The simple answer is yes. Theists attribute miracles to God,
who is perfectly good and just, and who loves us. So any miracle would be an act of benevolence,
justice, or love. If a given event E resulted in unmitigated evil, it could not be God’s doing, and
therefore could not be a miracle.
However, taking a cue from discussions of the theological problem of evil, a theist might reason
as follows: What seems to us to be evil may not really be so. We have faith that whatever God does is
good, though sometimes what He does seems to us to be evil. So if we were confronted with an
apparent violation of the laws of nature that results in apparent evil, we might conclude (perhaps
tentatively) that the event was indeed a miracle performed by God, and have faith that the results
were in fact good, attributing the appearance of evil to our limited understanding. Thus we might
recognize an event as a miracle, even if the outcome seems to us not to be good.
Nevertheless, the better the outcome, the more likely we are to view the event as a miracle. In
fact, the paradigm cases of miracles in the Bible and in theistic traditions have good outcomes, such
as saving people from harm, benefiting the righteous, or meting out punishment to those who
deserve it.
The idea of punishment raises some interesting issues. On some moral theories (e.g., utilitarian-
ism), suffering, even of the wicked, is in itself evil, and needs to be justified by its leading to some
good. One may interpret the Bible as working with a different sort of moral theory (perhaps a ver-
sion of Kantianism), according to which punishment of the wicked is intrinsically good, or else we
can assume that some justifying good always comes of God’s punishing the wicked. In any event,
according to theism, God’s motivation—whether in performing miracles or in performing any other
acts—must always be acknowledged to be morally beyond reproach. So again, we are more likely to
view an anomalous event as a miracle if its outcome seems to us to be morally good.
SUBJECTIVISM
The example of the child on the railroad track raises another possibility: Suppose we explain to the
child’s mother that what happened was just a lucky coincidence and that nothing supernatural was
involved, but she stubbornly insists that it was a miracle. One might suggest that we should simply
“agree to disagree” and say that to her it was a miracle, even though to us it was not. The suggestion
is that we make the concept of a miracle a subjective concept—a miracle, like beauty, is in the eyes
of the beholder.
A subjective concept is one whose application depends not only on the nature of the object to
which it is applied but also on the person, or subject, who is applying it. Aesthetic concepts are
often considered to be subjective: Thus if I say, of a particular painting, for example, that it is beau-
tiful, it seems as if I am attributing a quality to the painting, but I am really expressing my aesthetic
appreciation of the painting. If you say that the painting is ugly, then again it seems as if you are
attributing a quality to the painting, but you are really expressing your aesthetic displeasure on look-
ing at the painting. Though we have expressed contrary assessments, neither of us need be wrong;
we can just say that to me the painting is beautiful, and to you it is ugly.
Similarly, if we take the concept of a miracle to be subjective, then we could say, of one and
the same event, that to me it is a miracle, whereas to you it is not; and neither of us need be
wrong. For in calling the event a miracle, I am expressing my reaction to the event, that is, my
finding deep religious significance in it; and in denying that it is a miracle, you are expressing
your reaction (or lack thereof) to the event, that is, your being unmoved by it. Some may claim
that this subjective element needs to be added to our definition of “miracle”; that is, my calling
a particular event a miracle, in addition to describing it as being a suspension of the natural order
Miracles: Theism
by God, is also giving expression to the psychological effect that the event has had on me. Others
may make the more radical claim that the subjective element is all that there is to the concept of a
miracle—that is, all I am doing in calling an event a miracle is expressing how the event affected
me personally.
However, for reasons similar to those that I gave at the end of “Coincidence Miracles” above, I
prefer to use our original working definition, without the addition of any subjective element. Philo-
sophers of religion are generally more concerned with miracles insofar as they lead to belief in God
logically, rather than psychologically.
CAN LAWS OF NATURE BE VIOLATED?

It has been objected that the expression “violation of a law of nature,” as it appears in some definitions
of “miracle,” is a contradiction in terms, and therefore these definitions cannot be accepted. The idea
is that in order for something to be a miracle, the law it violates would have to be a universal law—
that is, of the form “All cases of A are cases of B,” or for short, “All A are B.” This is so because if
the law was not universal but stated only that most (or even almost all) A are B, then an A that is
not B would not necessarily be considered a violation of it and certainly would not need to be consid-
ered a miracle. (For example, it is a nonuniversal law of nature that most winter days are cold, but a
warm day in January in New York City would not be miraculous—only unusual.) But if a genuine
law of nature is universal, then there cannot possibly be an exception to it, or in other words, a viola-
tion of it. For if there is an exception to it—that is, an A that is not B—then the statement L that all
A are B is simply false, and a false statement does not qualify as a law of nature, so L is not a law of
nature after all. As John Stuart Mill put it ([1843] 2002, bk. 3, chap. 25, sec. 2), “We cannot admit a
proposition as a law of nature, and yet believe a fact in real contradiction to it. We must disbelieve the
alleged fact, or believe that we are mistaken in admitting the supposed law.”
But this does not seem correct. Rather it seems that L can still be considered a true, universal
law of nature, if the violation is due to supernatural intervention in the natural order (as in mira-
cles). That is, it may still be true that all A are B as long as events are allowed to follow their natural
course—and this truth would be sufficient for L to maintain its status as a universal law of nature,
despite the miraculous case of an A that was not B. Some have expressed this point by saying that
every law of nature needs to be qualified by a “God willing.”
Perhaps this point can be made most clearly in the context of the view that God endowed nat-
ural events with causal powers (the first of the three views of God’s relation to the natural order,
presented in “The Natural Order” in the first section, “What Is a Miracle?”). In this context, we
may view the system of laws of nature as a compendious and systematic description of the operation
of the causal powers. Those powers operate as the laws say they do, as long as God does not actively
intervene to disrupt their operation. If and when God does so intervene, the proper response would
be simply to acknowledge the violation and retain the violated laws; for those laws still correctly
describe the operation of the causal powers in nature. Any attempt to revise the laws so as to incor-
porate the intervention would lead to false scientific hypotheses and theories, appealing (implicitly,
at least) to natural causal powers that do not exist. Miracles are not to be subsumed under laws of
nature and explained by them; rather they are the direct outcome of God’s temporary suspension
of the natural order correctly described by the laws. Despite the miracle, the truth remains that
the causal powers in the world are such that L is law of nature—that is, all A must be B, unless
there is a special intervention by God.
HUME’S OBJECTION TO BELIEVING REPORTS OF MIRACLES
David Hume’s “Of Miracles” ([1748 et seq.] 2000, sec. 10) is a classic essay in which Hume offers a
powerful argument to the effect that it is always unreasonable to believe testimony that a miracle has
occurred.
Hume begins by noting that when witnesses testify to the occurrence of an event, the more
unusual or unexpected the event, the more reliable the witnesses must be in order for us to believe

Miracles: Theism
them. Thus, for example, if people tell us that they saw a dog bite a man, we can easily believe
them; but if they tell us that they saw a man bite a dog, we would be skeptical, unless or until
we know that the witnesses are reliable—that they are not joking, that they are not high on drugs,
that they have no personal interest in being quoted in the New York Times, and so forth. Or sup-
pose some people claim to have seen a UFO, from which aliens emerged and took samples of
earthly life-forms onto their spacecraft. We would be highly skeptical of their testimony, and we
would not believe it unless there were many witnesses whose testimonies are mutually consistent
and whose trustworthiness is beyond reproach. (And even then.…) The point is that a miracle, by
definition, is at the high end of the scale of unusualness and unexpectedness; and therefore the tes-
timony that a miracle occurred must be at the high end of the scale of reliability. Hume argues that
it is impossible, or at least highly unlikely, that any testimony could ever be so reliable that we
would accept it as establishing that a genuine miracle occurred.
Central to Hume’s argument is the definition of a miracle as a violation of a law of nature—
that is, as an exception to a law that has been found, on countless occasions until now, to hold with-
out exception. The argument is essentially as follows: Suppose we have a law of nature L, stating
that all A are B—for example, all cases of a bush’s burning are cases of the bush’s being consumed
(see Exodus 3:2). Now suppose that a number of people testify that on a certain occasion O, they
witnessed a miracle M—that is, an A that was not B. Should we believe that M really occurred
on O? As in all cases of judging whether an alleged event occurred, we must weigh the evidence that
it did occur against the evidence that it did not occur, and believe in accordance with the weightier
of the two. On the one hand, if L is true—that is, if all A are B—then M did not occur; and we
have extremely strong evidence that L is true (otherwise we would not consider L to be a law). That
is, we have extensive experience of As that are B, and no experience of an A that was not B. We may
add to Hume’s analysis that there may also be deeper scientific support for L; for example, presum-
ably chemistry tells us that on the molecular level, a bush’s burning and its being consumed come to
about the same thing. It follows that we have extremely strong evidence that any A would have been
B, so M did not occur.
On the other hand, we have the testimony of the witnesses that M did occur. How strong is
that evidence? That depends on the character of the witnesses and on the way they testify. Some
types of witnesses have proven to be more reliable than others. But it seems that no witnesses to
any event are ever so reliable that their testimony provides evidence that is as strong as the evidence
that typically serves to establish a law of nature. There are just too many ways that testimony of wit-
nesses, no matter how honest and intelligent, can be false. Even very intelligent people are suscepti-
ble to errors of perception or judgment, or to being fooled by an elaborate hoax; and even people
who are generally very honest might be tempted to lie, perhaps even to themselves, in support of
a cause that they deem sufficiently worthy—and many find the cause of religion sufficiently worthy.
It is difficult to imagine a type of testimony which has been found to be always true, in a multitude
of cases, and never false. So it seems that the evidence that M did not occur is stronger than, or out-
weighs, the evidence that M did occur. Thus it is always more reasonable to reject miracle reports as
false than to accept them as true.
Hume somewhat softens his negative conclusion. He admits that it is not out of the question
that there be testimony of a type that, as far as we know, is always true, or that must be true. Sup-
pose that very many people, each of impeccable judgment and honesty, all testify, in a serious man-
ner, that they clearly saw a burning bush that was not consumed. If there are enough witnesses of
sufficient integrity who claim to have observed the event in question under ideal perceptual circum-
stances, then perhaps we can say that our experience indicates that all testimony of this type must be
true. We might go so far as to say that it would be a miracle if the testimony of so many trustwor-
thy witnesses were false, that is, it would violate laws of human nature. In that case, we would be
faced with a choice between miracles—the miracle that a burning bush was not consumed versus
the miracle that the testimony of so many trustworthy witnesses was false. We must then ask which
would be the greater miracle and believe (tentatively) that the lesser miracle occurred. The bottom
line is that we should believe the testimony for the miracle that a burning bush was not consumed
if, but only if, the falsity of the testimony would be a bigger miracle.
Miracles: Theism
However, needless to say, for just about all miracle reports that we have heard and are likely to
hear, their falsity would be far from miraculous. It would be far more likely that those who report
miracles either are mistaken or are lying than that the miracle really occurred.
WITNESSING A MIRACLE FIRSTHAND
Hume’s argument concerns our evaluation of the testimony of others who claim to have witnessed a
miracle. But what if I myself were to have the experience of witnessing a miracle? Does Hume’s
argument apply to such a situation?
Of course, my witnessing an event counts for more with me than my hearing someone else tell
me that he or she witnessed it. (As the saying goes, seeing is believing.) The question is: Does it
count for enough to counter Hume’s argument—that is, does my experience of seeing what seems
to be a miracle provide me with enough evidence for the miracle’s actual occurrence to outweigh
the Humean evidence against the occurrence of such a miracle?
On the one hand, whereas in the case of testimony, the witnesses can be suspected either of
being mistaken or of lying, in the case of firsthand experience, lying doesn’t seem to be a worry.
There are senses in which a person can deceive himself or herself. However, one person’s con-
sciously and purposely lying to another in an attempt to deceive him or her does not seem to have
an analogue in self-deception, except perhaps in pathological cases.
On the other hand, I am just as susceptible to errors of perception and judgment as are other
people. Thus, my initial response to my miracle experience might very well be to ask myself: “Am I
dreaming? Am I hallucinating? Is this an illusion? Is it a hoax?” And in fact the most reasonable inter-
pretation of my experience might very well be that it is non-veridical (that is, mistaken or misleading)
in one of these ways. Furthermore, one’s perceptual experience is often influenced by one’s attitudes,
expectations, and desires; for example, a devout Christian might “see” a statue of a saint move
(“wishful seeing”?). Therefore, Hume’s conclusion (appropriately modified) applies: It is reasonable
to believe that a miracle M occurred only if the experience adduced as evidence for M is such that
its non-veridicality would be more miraculous than M itself. And it is arguable that due to our suscep-
tibility to error, the non-veridicality of my miracle experience would always be far from miraculous.
My perceptual experience may be so convincing that I cannot help feeling that it was veridical.
I could conceivably have a perceptual experience that leads me to believe so firmly that I have wit-
nessed a miracle that nothing and no one can persuade me otherwise. But that would be a psycho-
logical fact about me. Logically and epistemologically, at least according to Humean reasoning, I
should figure that it is more likely that my experience involved an error of one kind or another than
that I witnessed an actual miracle.
SOME CRITICISMS OF HUME’S ARGUMENT

THE CHARGE OF CIRCULARITY
A recurring objection to Hume’s argument is that it is question begging, or viciously circular—that
is, that Hume’s argument seems cogent only if we already believe its conclusion. (The paradigm
example of a question begging, or circular, argument is this: God exists, because the Bible says
He does; and what the Bible says is true, because God wrote it.) In its crudest form, the objection
is that Hume’s premise that “nature is uniform”—that is, that the laws of nature have so far been
found to hold without exception—is acceptable only if we already believe that there have never been
any miracles; so we cannot logically use this premise to conclude that miracles do not occur. A more
sophisticated version of the objection is this: An essential premise in Hume’s argument is that in our
experience laws of nature have been found to hold without exception. But our experience includes
hearing and reading tales of miracles, that is, reports of violations of laws of nature. So this premise
is acceptable only if we already disbelieve all those miracle reports; therefore, to use this premise to
prove that we should disbelieve miracle reports constitutes circular reasoning.

Miracles: Theism
I think the proper Humean response to these objections is as follows: First of all, by “in our
experience” Hume should be taken to mean one’s personal perceptual experience of the workings
of nature (including human nature), augmented by the content of reports from reliable people that
is consonant with one’s personal experience. For most people, miracles are not consonant with their
experience. Most of us have heard or read reports of miracles. Such reports are part of our experien-
tial data. But the miracles reported should not be included in the data, since they are not consonant
with the rest of our experience. Second, in order to accept Hume’s premise that the laws of nature
hold without exception, we need not antecedently believe that those reports are false. All we need to
do is withhold, or suspend, judgment about them. This suspension of judgment, for the purposes of
the argument, is only proper, given that the question of whether to accept such reports as true is
precisely what is at issue here; thus either believing them or disbelieving them would prejudge the
issue. Hume’s argument can proceed, while we maintain this suspension of judgment about past
miracle reports, until we reach the argument’s conclusion. Once we have reached the conclusion,
namely, that miracle reports generally are to be disbelieved, we can then legitimately apply it to
all those past tales of miracles that we have heard or read, and reject them on the grounds that it
is far more likely that the reports are false than that the miracles actually occurred.
PUTTING GOD IN THE PICTURE

One might be puzzled by Hume’s argument, for the following reason: He seems to have left out the
most important aspect of miracles—God. Of course, one might say, if you don’t believe in God,
you don’t believe in miracles—Hume has merely spelled out the logic behind this atheistic position.
But if we do believe in God, then his argument is beside the point; for God is omnipotent, and can,
by an act of will, cause any law of nature to be violated.
Hume briefly addresses this point. His response is, in effect, that even a theist must admit that
though God can intervene in the course of nature, nevertheless, to the best of our knowledge, He
never does so. All of our experience testifies to the fact that God allows (or causes, or has arranged
for) nature to follow its course without interference or aberration. In fact, it seems that something
like Hume’s argument against belief in miracles can be recast with God in the picture. To put it
briefly: In our experience, God never violates laws of nature; so if someone testifies that he wit-
nessed a miracle—that is, a violation of a law of nature, presumably by God—then it is much more
reasonable to believe that the testimony is false than to believe that God behaved in a way so
extremely out of character and that the miracle actually occurred (unless the falsity of the testimony
would be a bigger miracle than the one testified to—which is practically never the case). We may
call this argument “the noninterventionist induction.”
REVELATION AND TRADITION

I think that Hume is right in saying that his argument against belief in miracles holds even for the-
ists. However, I suggest that it holds primarily for a certain kind of theist—a natural theist, one
whose beliefs about God are based exclusively on evidence available to common sense and science.
If all we know about God and the way He operates is derived from our experience of the way the
world works, then yes, it seems that God always allows (or causes) events to occur in accordance
with the laws of nature, without exception. However, many theists subscribe to a revealed religion,
according to which God has revealed to us much more about Himself than what we can derive from
our perception of the world. According to some religious traditions, God has performed miracles in
the past, and He has revealed to us that He intends to perform miracles in the future. Given this
“inside information,” it may be reasonable, in certain circumstances, to believe a report of a miracle.
To illustrate this point, I shall take the liberty of altering a parable briefly presented by Bertrand
Russell in making a different, though related, point. In the parable, a turkey lives on a farm, and all
his life, every morning, the farmer has brought him food. One morning (on the holiday of Thanks-
giving) one of the chickens on the farm warns the turkey, saying “I saw the farmer coming, but
instead of the usual food, he is bringing an ax, and he intends to slaughter you and eat you for din-
ner.” Given the turkey’s experience of the farmer’s behavior, it would be reasonable for him to reject
the testimony of the chicken and believe that the farmer will do this morning what he has done
Miracles: Theism
every morning until now. (We know that the turkey might then be fatally mistaken, but that is
because we know more about farmers than the turkey does.) Now for the main alteration: Suppose
that on one of the first mornings when the farmer was feeding the turkey, he said to the turkey “Eat
hearty and fatten up, because one of these days I’m going to slaughter you and eat you for dinner.”
In this case, when the chicken testifies about the farmer’s anomalous behavior, it would perhaps be
reasonable for the turkey to accept her testimony and run for his life.
The natural theist is comparable to the turkey in Russell’s original story (actually in Russell’s
original example, it was a chicken). If someone tells a natural theist that God has suddenly behaved
differently from the way He has uniformly behaved on countless occasions until now—that is, that
God has performed a miracle—then it would be reasonable, as Hume argues, for him or her to
reject the report of the miracle as false. However, suppose that I adhere to a religious tradition that
tells me that God has promised to perform miracles. If someone testifies that he or she has witnessed
a miracle, then I would have much more reason to take that testimony seriously than would an
atheist, an agnostic, or a natural theist. If my religious tradition tells me further that God has prom-
ised to perform a specific sort of miracle at a specific time and place, then if someone reports to me
that he or she has witnessed just that sort of miracle at just that time and place, then it would not be
unreasonable for me to believe the report. The moral of the story is that if I subscribe to a revealed
religion, such as Judaism or Christianity or Islam, which includes promises of miracles, then I can
still reasonably believe in miracles, despite Hume’s argument.
HUME STRIKES BACK: REVELATION IS A MIRACLE

Of course, this is not the final word. (There are very few final words in philosophy.) For one thing,
revelation is itself a kind of miracle; after all, how many of us can say that God has revealed Himself
to us in an obvious way? Is my belief in a particular revelation—such as the one at Mount Sinai, as
reported in the Bible (Exodus 19–20)—itself reasonable? If I believe in miracles only because of that
revelation and its including promises of future miracles, then I cannot logically invoke God and His
promises, as I did above, to justify my belief in that miraculous revelation itself; for presumably
there were no promises before that revelation that could serve as a basis for believing in it—and if
there were, this would lead to an infinite regress (did God promise that He would promise that
He would promise … ?).
Some Jewish thinkers have argued, in effect, that the testimony that we have for the Revelation
at Mount Sinai is just the kind of testimony that Hume was talking about when he spoke of the
falsity of a miracle report being more of a miracle than the miracle being reported. The entire nation
of Israel heard God’s voice emanating from Mount Sinai proclaiming “I am the Lord thy God,” and
all of those people were charged with testifying about the miracles of the Exodus from Egypt and
the Revelation at Sinai to the next generation, and the next generation to the next, and so on, to
this very day. (Handing down this testimony is perhaps the most important aspect of the traditional
Jewish Passover seder). So it would be a miracle if all that testimony, handed down from generation
to generation, were false—a bigger miracle than the miracle of revelation itself.
Analyzing and evaluating this argument is beyond the scope of this article. However, I think
that, on the basis of what we have said so far, we can tentatively conclude that a believing traditional
Jew or Christian might reasonably believe a miracle report, despite Hume’s argument—pending fur-
ther discussion. That is, traditional religious Jews and Christians need not give up their belief in
miracles just because of Hume’s argument; in order to make it reasonable for them to give up belief
in miracles, further argument is needed to undermine some of their more fundamental religious
beliefs.
DOES MIRACLE TESTIMONY REALLY CONFLICT WITH EVIDENCE FOR LAWS?

Something that I said before (in “Can Laws of Nature Be Violated?” in the second section, “Alter-
native Definitions.”) is relevant to evaluating Hume’s argument: I suggested that a statement L may
qualify as a bona fide universal law of nature even if it was violated by the action of a supernatural
being. Hume’s argument is based on viewing testimony that a miracle occurred as conflicting with
L, or with the evidence for L. That is why he claimed that we must weigh the evidence for L against

Miracles: Theism
the evidence for the reliability of the testimony. But a theist might very well claim that there is no
conflict here. Every law of nature is supposed to hold only as long as events are allowed to follow
their natural course; and the evidence in favor of a universal law is evidence only that that law holds,
without exception, as long as nature follows its normal course, without supernatural interference. A
claim that there has been such supernatural interference—that L has been violated miraculously—
does not detract from L’s remaining a universal law of nature. In other words, every law of nature
L has an implicit proviso: All A are B, provided there is no supernatural intervention. A miracle
would simply be a “cashing in” of the proviso. Thus the claim that there has been a case of super-
natural intervention to bring about an A that is not B does not conflict with L, or with the evidence
for L, as properly understood.
A CUMULATIVE RESPONSE TO HUME’S ARGUMENT

Based on what has been said, above, in this section, we can build a cumulative case, from a theistic
perspective, against Hume’s argument. Consider the following points (which are not in the order in
which they were presented, above):
1. First of all, we have, I think, established that, from a theistic viewpoint, the statement that a
miracle occurred does not contradict the law of nature of which the miracle was supposed to be a
violation; rather it is in keeping with the implicit proviso that the law—since it is, after all, a law
of nature—holds unless there is supernatural interference. Therefore we need not weigh the
testimonial evidence that a miracle occurred against the overwhelming evidence that served to
establish the law in question.
2. So what is left of Hume’s argument is what we have called “the noninterventionist induction”:
the premise is that in our experience, as far as we know, there has been no supernatural
interference in the working of nature, no miracles—which is supposed to go against, and
defeat, any testimony that there has in fact been such interference, that is, that a miracle has
occurred.
3. The noninterventionist induction seems somewhat weaker than Hume’s original argument. In
particular, since the main premise states explicitly that we know of no miracles in the past, it
seems to invite, more than did the original argument, the charge of circularity—that is, that in
order to reach the conclusion that we are to disbelieve miracle reports, we must first disbelieve all
the miracle reports that we have heard or read until now.
4. We defended Hume against this charge of circularity, saying that his argument does not require
that we actually disbelieve (i.e., believe to be false) reports of past miracles; it requires only that
we suspend judgment about them. This means that we are to consider past miracle reports to be
possibly false, but also possibly true.
5. Our main response to Hume’s argument—and it is at least as apt a response to the non-
interventionist induction—was that a theist may subscribe to a revealed religion, such as
traditional Judaism, Christianity, or Islam, that includes the doctrine that God has promised to
perform miracles; and such a theist may, on that account, have some rational basis for believing a
miracle report. (Remember the turkey.)
6. The main problem with this response was that the revelation that is supposed to be the basis for
that doctrine would itself have been miraculous; so the theist is left with no similar justification
for believing that the revelation occurred (on pain of vicious infinite regress).
7. However, we may here appeal to point 4, above: since we are neither accepting nor rejecting
reports of past miracles, we can take seriously the possibility that those miracles occurred; in
particular, we can take seriously the possibility that God revealed to us that He intends to
perform miracles. Is this enough to justify us in believing a new miracle report?
8. To answer this question, consider the following alternative change to Russell’s parable: Instead of
hearing it from the farmer’s mouth, the turkey was told by a cow long ago that she had overheard
the farmer saying that he intends to slaughter the turkey someday. The turkey was hesitant to
believe the cow, especially since she seemed to be slandering the good farmer; and so the turkey
Miracles: Theism
has withheld judgment about it to this day. But then today the chicken makes her frightening
claim: “The farmer is coming to kill you.” What should the turkey think and do? I think he
should view the testimonies of the chicken and the cow as corroborating each other, and he
should run away.
Similarly, I suggest that the biblical accounts of God’s revealing to us that He will perform
miracles, even if we are suspending judgment as to their veracity, serve to lend some credence to
subsequent miracle reports. If I am right, then it may be reasonable for a traditional theist to believe
a report of a miracle, if the report is of a kind that is generally reliable, and if the miracle is of a kind
that is consonant with his or her religious tradition.
One might argue that I am being too hasty in applying what I said about the turkey in point 8,
above, to the adherent of a revealed theistic religion. After all, there is more at stake for the turkey
confronted with the chicken’s testimony than there is for a theist confronted with a miracle report;
in particular, we may assume that the theist’s accepting or rejecting the miracle report is not a mat-
ter of life and death. It may be prudent for the turkey to run away, just in case the chicken’s testi-
mony is true; but a theist deciding whether to believe a report of a miracle can afford to be more
detached and to weigh the evidence in a more disinterested manner, and perhaps he or she should
then conclude that it is indeed more likely that the report is false than that the miracle actually
occurred.
Nevertheless, I think that the considerations brought forward in points 1–8, above, serve at
least to weaken Hume’s argument. If Hume’s argument had the full force that he and others have
claimed for it, then by parity of reasoning, even the turkey (even in point 8) should simply dismiss
the chicken’s testimony and confidently await his usual daily meal from the farmer. The mere fact
that it would be prudent for the turkey to run away on the basis of the chicken’s (plus the cow’s)
testimony is sufficient to show that a traditional theist is entitled to remain open to accepting rea-
sonably reliable testimony to a miracle as true, and certainly to assessing firsthand perceptual experi-
ences of a miracle as veridical.
MIRACLE REPORTS IN DIFFERENT RELIGIOUS TRADITIONS

HUME’S ARGUMENT FROM DIFFERENT RELIGIONS
Hume, in “Of Miracles,” put forward several subsidiary arguments against believing in miracle
reports, in addition to his main argument, discussed above. Many are historical in nature, pointing
out the weakness or suspiciousness of most miracle reports in the past. Others are based on claims
about human nature—for example, the credulity of ordinary people, the desire of some people for
the fame that comes with founding a new religion, and enthusiasm on behalf of an existing religion.
But one argument seems to be more philosophical (though it does involve some historical claims).
We may call this “the argument from different religions.” Often miracles are reported as
having been performed within the framework of a particular religion and are taken to support that
particular religion. For example, any miracle attributed to Jesus, or to a contemporary Christian
clergyperson, is in the framework of Christianity, and can be taken as evidence for the truth of
the doctrines of Christianity, whereas a miracle attributed to Muhammad, or to a contemporary
imam, is in the framework of Islam, and can be taken as evidence for the truth of the doctrines
of Islam. Hume argues that different religions are mutually incompatible—that is, if one of them
is true, then all the others are false. Thus, for example, if Christianity is true, then Islam and Juda-
ism (as well as less closely related religions, such as Hinduism and Buddhism) are false. Hume con-
cludes that any miracle report that supports one religion also serves indirectly to undermine any
miracle report that supports any other religion; in other words, any claim that a miracle occurred
in the framework of one religion entails the claim that the reports of miracles in the framework of
other religions are false. The reasoning may be represented as follows:
Let M1 be a miracle that would support a particular religion R1; and let M2 be a different
miracle that would support a different religion R2.

Miracles: Theism
1. If M1 occurred, then R1 is true. (premise)

2. If R1 is true, then R2 is false. (premise)
3. If M2 occurred, then R2 is true. (premise)
4. If R2 is false, then M2 did not occur. (follows from 3)
Therefore:
5. If M1 occurred, then M2 did not occur. (follows from 1, 2, and 4 or 3)
And:
6. If M2 occurred, then M1 did not occur. (follows from 5, or from 1–4)
Virtually every religion includes miracle reports that support it. So any miracle report that supports
any one religion is undermined by the many miracle reports that support other religions. In other
words, for any assertion that a miracle occurred in the framework of one religion, there are many
(implicit) denials that that miracle occurred, in the form of assertions of miracles in the framework
of other religions. This consideration, if well-founded, would further undermine belief of any mira-
cle report.
RESPONSE: PLURALISM AND OPENNESS

However, some of the premises of Hume’s argument may be questioned. First of all, there is a lively
discussion nowadays as to whether different religions are really mutually incompatible (as is, in effect,
asserted in premise 2 of the above argument). A pluralist is one who thinks of religions as mutually
compatible; an exclusivist is one who thinks of religions as mutually incompatible, or who thinks that
his or her religion is the only true religion. Here is one pluralistic approach: The various world reli-
gions are all human responses to the Divine, and God recognizes many of these different religions
as valid forms of worship—even if the doctrines of any one of those religions are logically inconsistent
with those of other religions. That is, God, in His infinite wisdom and goodness, recognizes human
limitations and is prepared to accept various religions as valid forms of worship, even if they include
false theological, historical, or scientific claims. In performing a miracle in the framework of a given
religion, God is not necessarily endorsing the doctrines of that religion as true, and by implication
refuting the doctrines of other religions; rather He is merely endorsing that religion as one of the many
different religions that are valid. Such a pluralist approach effectively blunts Hume’s argument from
different religions. That is, since it is perfectly appropriate that God perform miracles endorsing differ-
ent valid religions, the assertion that a miracle occurred in the framework of one valid religion does
not constitute a denial that a miracle occurred in the framework of another valid religion.
One may also question premises 1 and 3. That is, one may question whether a miracle neces-
sarily serves to support any one religion over any other. There may be some miracles that are per-
formed specifically to support one religion and refute others. An example of this would be Elijah’s
“showdown” with the prophets of Baal on Mount Carmel (1 Kings 18:19–40), in which God made
fire descend miraculously from Heaven and consume Elijah’s offering, at his request, while the
prophets of Baal were unable to induce their god to do the same for them. But there may be mira-
cles in the framework of a particular religion that nevertheless do not necessarily support that partic-
ular religion. We may be (and perhaps should be) open to the possibility that God would perform
miracles on behalf of morally good people who adhere to religions different from ours. For example,
suppose a Catholic priest is in danger, and a miracle occurs to save him. Jews, Muslims, and Protes-
tants might recognize that the priest is a good man (though they think he is mistaken in religious
matters), and while still maintaining their belief in Judaism or Islam or Protestantism, maintain that
(their) God might very well have performed the miracle to save the good priest; thus such a miracle
need not be taken as supporting Catholicism over other religions.
Nevertheless, one could defend Hume by arguing against my version of religious pluralism,
noting that the doctrines of different religions are, in fact, mutually inconsistent, and claiming that
it is implausible that God would do anything to endorse a religion of false doctrines. And as for the
Miracles: Theism
idea that not every miracle serves to endorse the religion in the framework of which it was per-
formed, it should be noted that many religions include reports of miracles that are explicitly
intended to support or establish the truth of the doctrines of those respective religions.
So perhaps we should make Hume’s considerations a matter of degree: To the extent that M1
tends to establish R1, and M2 tends to establish R2, and to the extent that God would be reluctant
to endorse both R1 and R2, to that extent M1 is contrary to M2, and the assertion that M1
occurred counts as an implicit denial that M2 occurred, and vice versa.
MILL’S ARGUMENT THAT MIRACULOUS EVENTS MIGHT HAVE A

NATURAL EXPLANATION
Hume argued that when we are presented with testimony to an event involving violation of a uni-
versal law of nature, it is practically never reasonable to believe that the event actually occurred.
Mill, in contrast, deals with the question of what it is reasonable to do once we accept as fact that
such an event did occur. His answer is, essentially, that it is always more reasonable to search for
a naturalistic explanation of the event than to attribute it to the miraculous intervention of a super-
natural being or force.
Perhaps Mill’s position is better expressed as: What seems to be a violation of a law of nature
might not really be so. We might be able to explain the event in question by discovering that there
was a very unusual coincidence of natural circumstances and that the accepted laws of nature do
entail that in those circumstances just such an unusual event should occur. Or else, if we cannot dis-
cover such circumstances, we might admit that the generalization L, which we have accepted as a
law of nature, has in fact been violated; but then what is most likely is that L is not a bona fide
law of nature after all, and we need to replace it with a law that better fits the facts, or more accu-
rately depicts the relevant aspects of reality. It might turn out that L is an approximation of a bona
fide law of nature, or that L is a simplification of laws of nature, one that usually holds true. (E.g.,
the “law” that what goes up must come down holds generally but not for space vehicles, so it is not
a bona fide universal law of nature.) But if an event E has a natural explanation, then that means
that E accords with the laws and powers of nature, and therefore E is not a miracle.
What if we cannot find a naturalistic explanation of anomalous event E? Patrick Nowell-Smith
(1955) argued that we should take into account the fact that science progresses (or at least changes)
over time, and that therefore, though we are currently unable to explain E, we may very well be able
to explain E in the future. Maintaining our faith in science to be able eventually to provide an
explanation of E is more reasonable than trying to explain E as an act of God, since we do not know
enough about God to render such an appeal to His agency much of an explanation (e.g., we will be
faced with the unanswerable question of why God intervened to save this good person, yet allowed
other good people to perish). Thus it is never reasonable to believe, of any given event, that it was a
genuine miracle.
What shall we say about Mill’s and Nowell-Smith’s arguments? We might say that some
kinds of possible miracles lend themselves to naturalistic explanation more than others. For exam-
ple, suppose that we are in the Antarctic and, despite the cold, a piece of iron melts before our
eyes. Sounds miraculous, but maybe it was just a coincidence that the confluence of natural forces
created enough heat just there and then to melt the iron; or perhaps our laws governing the melt-
ing of metals need to be revised in light of this incident. But suppose there is more to the story:
while in Antarctica we are accosted by a terrorist brandishing a gun, and then, despite the cold,
his gun suddenly melts, allowing us to overpower him. These circumstances, and especially the
happy ending, would make us more inclined to ascribe the gun’s melting to a purposeful interven-
tion by God—that is, to consider the event a miracle. But still, one might “bite the bullet” and,
following Mill and Nowell-Smith, believe that there must be some natural explanation of our
“good luck.” But now consider this scenario: Someone comes along and claims that he is the
long-awaited Messiah. Suppose he says that he has (with God’s help) resurrected the dead, and
then presents to us our long-dead parents, grandparents, and other relatives, whom we see and

Miracles: Theism
feel before us and with whom we proceed to converse about old times. I would think it highly
unlikely that any discovery of hidden natural circumstances or any revision of our accepted scien-
tific laws and concepts could serve to explain that. So I think it would be reasonable, in such a
case, to believe that we have witnessed a genuine miracle.
One might quibble and say that even in this last example, I might just be dreaming. But I
would answer that even if I am dreaming, my dream rationally should include our judging that
we have witnessed a miracle. If I should suddenly wake up in my bed, with no deceased relatives
around, then I can revise that judgment.
WOULD THE OCCURRENCE OF MIRACLES BE GOOD EVIDENCE

THAT GOD EXISTS?
This is, of course, the big question. We must be careful in formulating the question. First of all, if
we define the term miracle as “an event involving the temporary suspension of the natural order by
God,” then if a miracle occurred, it follows logically that God exists. But then the “proof” of God’s
existence on the basis of the occurrence of a miracle would be question begging, for we would first
have to believe that God exists before we could believe that a miracle occurred. The interesting
question here is whether something could possibly happen that would make it reasonable for us
to conclude that a miracle occurred and therefore God exists. In other words, could we ever see
or hear or otherwise perceive something that would lead us rationally to conclude (1) that laws of
nature have been violated, (2) that the violation was due to the action of a supernatural agent,
and (3) that that supernatural agent was God?
We have examined Hume’s argument that it is practically never reasonable to believe that a law
of nature has been violated, as well as Mill’s argument that it is never reasonable to attribute (appar-
ent) violations of natural laws to supernatural agency. I believe that we have shown, in our discus-
sion of Hume’s argument, that a traditional (as opposed to a natural) theist, given his or her back-
ground beliefs, might very well be rationally justified in concluding, either from hearing testimony
of witnesses or from firsthand perceptual experience, that there occurred a violation of laws of
nature. We have also shown, in our discussion of Mill’s argument, that there are conceivable
sequences of events that defy any possible attempt to provide the kind of naturalistic explanation
that Mill says is preferable to invoking the supernatural. If traditional theists find themselves in a
situation in which it is reasonable for them to believe that there has been a violation of laws of
nature, and that the likeliest account of this violation is that it is the work of a supernatural agent,
then it would be natural for them to assume that the supernatural agent was the very God who
appears so prominently in the religious traditions in which they have been raised. The question is:
Would this assumption, though natural, also be rationally justified? After all, all that the miracle
tells us about the supernatural agent is that he or she somehow managed to violate those particular
laws of nature, not that he or she has all the attributes that characterize God.
Of course, if a miracle consisted in, or included, a booming (yet soothing) voice, seemingly
from Heaven, heard all over the world, proclaiming to each person in his or her own language, “I
am the Lord, your God, the omnipotent, omniscient, perfectly good Creator of the world,” that
would pretty much settle it, at least for me: God exists (see Hume 1980 [1779], part 3). But short
of that, even a bona fide violation of a law of nature, best accounted for as the work of a benevolent
supernatural agent, would fall short of proving that the theistic God exists.
A theist might argue that laws of nature can be violated only by the almighty Creator of the
world, for only the Creator of the natural order has the power to abrogate it; so a miracle would
indeed prove that the omnipotent Creator exists. However, as noted earlier (in “Suspension of the
Natural Order” in the first section, “What Is a Miracle?”), it is not so clear that a supernatural agent
cannot temporarily suspend the natural order unless he or she created the world and instituted that
order. Furthermore, it is not clear that in order to violate laws of nature, a supernatural agent needs
to be omnipotent; he or she may just be extremely powerful. Though the supernatural agent must
know something about what is going on in the world in order to intervene in an appropriate way,
Miracles: Theism
nevertheless, there is no basis for attributing to him or her omniscience. Finally, even if the miracle
has a good and just outcome—for example, the righteous are saved and the wicked are chastised—
nevertheless, there is no necessity to attribute to the supernatural agent responsible for the miracle
perfect goodness. Yet the theistic God is supposed to be the omnipotent, omniscient, and perfectly
good creator of the world; so evidence of His existence would have to be evidence that a Being with
all these attributes exists.
Nevertheless, first of all, perhaps proving that there is a supernatural being who is very power-
ful, intelligent, and benevolent is close enough to proving that God exists that theists should be sat-
isfied with it. Second, for all we can tell, the supernatural being who performed the miracle might
be not only powerful, intelligent, and benevolent but actually omnipotent, omniscient, and omnibe-
nevolent (or perfectly good). So we can understand a genuine miracle as strongly confirming theism
in part; and we can understand this strong confirmation as supporting, or at least weakly confirm-
ing, theism as a whole, complete with all the “omnis.” Finally, for many people steeped in a theistic
tradition, the only relevant hypotheses that appeal to them as even possibly true are (a) theism (per-
haps only their own particular theistic religion) and (b) naturalism (briefly, the view that there is
nothing supernatural, nothing beyond the world of common sense and science). If these are the
two competing hypotheses, then the occurrence of a miracle would refute naturalism and confirm
theism.
This last point involves a subjective element, for there are nontheistic religions, and adherents
of those religions will consider their set of nontheistic religious beliefs, rather than theism, to be
the hypothesis confirmed by the miracle. Nevertheless, one could argue that a certain amount of
subjectivity is unavoidable whenever anyone considers data as evidence supporting, or confirming,
some hypothesis. For in principle, there are indefinitely many alternative hypotheses that are equally
confirmed by any given evidence; so considering just one of these to be confirmed is, in a sense, a
subjective choice (even if we can’t think of any of the viable alternatives). Moreover, on some prom-
inent views about confirmation, the degree to which evidence confirms a hypothesis depends partly
on the initial, or prior, probability of the hypothesis; and this initial probability might well be a
subjective matter. But we are now getting into matters that are best left for a separate essay in the
philosophy of science. Suffice it to say here that it might well be reasonable for a theist to come
to believe that a miracle has occurred and to take the miracle as evidence for theism.
But so far, we have shown only that a theist (and only a traditional theist at that) can reason-
ably believe that a miracle occurred and take that miracle as evidence of God’s existence. However,
a theist already believes that God exists. Isn’t the important question whether a nontheist—for
example, an atheist or an agnostic—could ever reasonably believe that a miracle occurred and
thereby come to believe that God exists?
First of all, even traditional theists would, and should, welcome the kind of evidence of God’s
existence that the occurrence of a contemporary miracle could provide. For most theists, their faith
in God could always do with some strengthening, and their faith would reasonably be strengthened
if they came to believe that a miracle has recently occurred; that is, their belief in God would be cor-
roborated by the new evidence provided by the miracle.
Second, even a nontheist could view an occurrence as a miracle and thereby come to adopt the-
ism—provided that he or she has heard of theism and takes it somewhat seriously. Earlier, in
“A Cumulative Response to Hume’s Argument” in the fifth section, “Some Criticisms of Hume’s
Argument,” we argued, with regard to the claim that God promised to perform miracles, that even
if a person does not actually believe that the claim is true, but merely considers it possibly true, it
may be reasonable for him or her to believe a miracle report. Similarly, as long as someone considers
some form of traditional theism (which includes God’s having promised to perform miracles) to be
possibly true, it may be reasonable for him or her to believe a miracle report and furthermore
to consider the miracle to be evidence for theism. However, for someone who has never heard of
theism, or who has heard of it, but considers it not even possibly true, it may never be reasonable
to believe a miracle report or to consider anything to be evidence for theism.
In conclusion, I think we can say that it is possible that an event could occur that could reason-
ably be considered to be a miracle and that therefore would provide some evidence of God’s

Miracles: Theism
existence—at least for anyone who has heard of some form of traditional theism and who does not
simply dismiss it as absurd.
Some theists would prefer to be able to conclude that there are many possible scenarios in
which anyone would be rationally forced to believe that a miracle occurred and to recognize the
miracle as constituting proof that the theistic God exists. However, I am afraid that this is beyond
what our analysis yields. Personally, as a theist, I would be pretty much satisfied if our discussion
and its conclusion, such as it is, can be reasonably defended against rational criticism.
Bibliography
Basinger, David, and Randall Basinger. Philosophy and Miracle: The Lewis, C. S. Miracles. New York: Macmillan, 1947.
Contemporary Debate. Lewiston, NY: Edwin Mellen Press, 1986.
Maimonides, Moses. The Guide for the Perplexed. Translated by
Earman, John. Hume’s Abject Failure. Oxford: Oxford University Press, M. Friedlander. New York: Dover, 1956.
2000.
McGrew, Timothy. “Miracles.” In Stanford Encyclopedia of Philosophy, edited
Flew, Antony. Hume’s Philosophy of Belief, chap. 8. New York: Humanities by Edward N. Zalta. Spring 2017. https://stanford.library.sydney.edu.au
Press, 1961. /archives/spr2017/entries/miracles/. This entry contains a full bibliography.
Flew, Antony. “Miracles.” In Encyclopedia of Philosophy, 5:346–353. New Mill, John Stuart. A System of Logic. The most relevant selection appears in
York: Macmillan and Free Press, 1967. Philosophy and Faith: A Philosophy of Religion Reader, edited by David
Fogelin, Robert. A Defense of Hume on Miracles. Princeton, NJ: Princeton Shatz, reading 6.11, 409–411. New York: McGraw-Hill, 2002. First
University Press, 2003. published 1843.
Garrett, Don. Cognition and Commitment in Hume’s Philosophy, chap. 7. Millican, Peter. “Twenty Questions about Hume’s ‘Of Miracles.’” In
New York: Oxford University Press, 1997. Philosophy and Religion (Royal Institute of Philosophy Supplement 68),
edited by Anthony O’Hear, 151–192. Cambridge, UK: Cambridge
al-Ghazālī. The Incoherence of the Philosophers. Translated by S. A. Kamali. University Press, 2013.
Lahore, Pakistan, 1958.
Nowell-Smith, Patrick. “Miracles.” In New Essays in Philosophical Theology,
Holland, R. F. “The Miraculous.” American Philosophical Quarterly 2, no. 1 edited by Antony Flew and Alasdair MacIntyre, 243–253. New York:
(1965): 43–51.
Macmillan, 1955.
Hume, David. Dialogues concerning Natural Religion. (1779 et seq.) Edited
Russell, Bertrand. The Problems of Philosophy. New York: Oxford University
by Richard H. Popkin. Indianapolis, IN: Hackett, 1980.
Press, 1959. The parable appears on page 63. First published 1912.
Hume, David. An Enquiry concerning Human Understanding. (1748 et seq.)
Swinburne, Richard. The Concept of a Miracle. London: Macmillan, 1970.
Edited by Tom L. Beauchamp. New York: Oxford University Press,
2000. Swinburne, Richard, ed. Miracles. New York: Macmillan, 1989.
Johnson, David. Hume, Holism, and Miracles. Ithaca, NY: Cornell University Twelftree, Graham H., ed. The Cambridge Companion to Miracles.
Press, 1999. Cambridge, UK: Cambridge University Press, 2009.
TOPIC 7
Miracles: Atheism
MIRACLES IN MONOTHEISTIC RELIGION 211
Arif Ahmed
University Reader in Philosophy, Faculty of Philosophy
AT PRESENT, MIRACLE CLAIMS UNDERMINE THEISM 219

Richard Carrier
Educator (PhD)
This chapter distinguishes three main conceptions of miracles: extraordinary events, violations of the laws of
nature, and divine interventions. Further discussion looks at whether miracles of any type are possible. The
chapter considers David Hume’s argument that we have no reason to think that any events considered
miracles are actual. Finally, the chapter asks whether we should regard talk of miracles not as a description
of anything that happened but rather as an interpretation of events that are agreed on all hands.
MIRACLES IN MONOTHEISTIC RELIGION
Arif Ahmed, University of Cambridge
Many religious traditions assert the occurrence of extraordinary events known popularly as miracles.
The central cases allegedly occurred in the eastern Mediterranean between about 1,400 and 4,000
years ago and are central to the Jewish, Christian, and Islamic faiths. In the case of Judaism, these
include various massacres perpetrated on whoever happened to be the enemies of Israel at the time,
for instance: the smiting of all the firstborn of Egypt (Exodus 12:29–30), the slaughter of the Phi-
listines before the Ark (1 Samuel 5:1–12), and the consumption by fire of the captains of Ahaziah
(2 Kings 1:10–12). Christianity is centered on the miracle of the resurrection, but the New Testa-
ment mentions such others as the turning of water into wine (John 2:1–20), the raising of Lazarus
(John 11:1–44), and speaking in tongues (Acts 2:4). Muslims believe in the miraculous appearance
of Gabriel to Muhammad as reported in the Qur ʾan (e.g., Surah An-Najm 5–10), this being the
source of the Qur ʾanic text and the underpinning of its divine provenance; the Qur ʾan itself relates
sundry other miracles, such as God’s sending angels to defeat the pagans at Badr (Surah Al-Imran
124) (see Ahmed forthcoming).
This essay addresses the following questions: What are miracles? Is it reasonable to believe that
such events either could or did occur? And is belief in miracles really a matter of belief in this or that
event, or is it a way of interpreting the events?
WHAT IS A MIRACLE?
We can distinguish four senses of “miracle.” First, in what Voltaire calls the “fullest sense” of the
word, a miracle is just anything that is admirable (Voltaire 2004 [1764], 311). Similarly, Saint
211
Miracles: Atheism
Augustine in one place defines a miracle as “whatever appears

KEYWORDS that is difficult or unusual above the hope or power of them
who wonder” (Augustine 1887, s. 34). Voltaire says that in this
sense many things count as miraculous. “The prodigious order
Epistemic Evidence
of nature, the rotation of 100 million globes around a million
Eyewitnesses suns, the activity of light, the life of animals are perpetual mira-
Historical Testimony cles” (Voltaire 2004 [1764], 311).
Metaphysics In a second sense, perhaps more familiar from ordinary
usage, a miracle is just some extraordinary or extremely surpris-
Miracles ing event. In this sense it would be miraculous if, say, a ran-
Science domly shuffled pack of cards got dealt out in perfect order. Saint
Scriptures Augustine, and apparently also David Hume, used “prodigy” for
this type of event, each reserving “miracle” to convey a stronger
Skepticism sense (Augustine 1984, Book X, 16; Hume 1975 [1777], 111).
In that stronger sense, which Voltaire attributes to “conven-
tional usage,” a miracle is a violation of the laws of nature. “If
there is an eclipse of the sun at full moon, if a dead man walks
two leagues carrying his head in his arms, we call that a miracle” (Voltaire 2004 [1764], 311). This is
what Hume meant, and close enough to what Augustine seems to have meant, by “miracle” (Hume
1975 [1777], 114; Augustine 1984).
The fourth sense of “miracle” is that of Saint Thomas Aquinas, on which a miracle is an event
that has been brought about by the direct action of a divinity rather than by the productive forces
of nature. “Things that are at times divinely accomplished, apart from the generally established order
of things, are customarily called miracles.… God alone can work miracles” (Thomas Aquinas 1956,
3.101.1–3.102.1). Note that as far as Saint Thomas is concerned, an event may be miraculous and
yet not violate the laws of nature, so long as it was in fact brought about by God and not by natural
processes. “For example, a person may be cured by divine power from a fever which could be cured
naturally, and it may rain independently of the working of the principles of nature” (3.101.4; cf. Mark
5:25–29). However, “The highest rank among miracles is held by those events in which something is
done by God which nature never could do. For example, that two bodies should be coincident; that
the sun reverse its course or stand still; that the sea open up and offer a way through which people
may pass” (Thomas Aquinas 1956, 3.101.2; cf. Isaiah 38:8; Joshua 10:12–13; Exodus 14:21–22).
It would be pointless to dispute whether any of these definitions (or any other) captures what a
miracle really is. It makes more sense to keep track of all four definitions, for it may be that some
arguments for or against miracles are convincing when we take that term in one sense but not when
we take it in another. Nonetheless, nothing in what follows has any special application to miracles
in the first sense—not surprisingly, since as Voltaire says, in that sense very many quite familiar
things can count as a miracle. Accordingly, I’ll distinguish the three remaining senses by saying that
there are three kinds of miracles.
• An A-miracle is an extraordinary event.
• A B-miracle is a violation of the laws of nature.
• A C-miracle is an event that God has brought about independently of the general course of
nature.
As we’ll see, different arguments have been made, and different conclusions may be granted, con-
cerning each of these types.
ARE MIRACLES POSSIBLE?

It’s worth distinguishing two kinds of possibility. On the one hand, I’ll say that an event or state of
affairs is epistemically possible (e-possible) if we cannot rule out a priori, that is, without empirical
investigation, that it actually happens. For instance, it is epistemically possible that Queen Elizabeth
II was not the daughter of George VI, that Julius Caesar never visited Britain, or that dogs can talk.
Miracles: Atheism
These things are unlikely, but it takes empirical investigation to rule them out, whereas it is episte-
mically impossible that 2 + 2 = 5, that Wednesday ever comes one day after Thursday, and perhaps
also that nobody has ever had any thoughts. We can rule these things out from the armchair, that is,
without having to “look at the world.”
On the other hand, I’ll say that an event or state of affairs is metaphysically possible (m-possible)
if it could have occurred or obtained, that is, if it does occur or obtain in some alternative possible
world. It is metaphysically (and epistemically) possible that Caesar never visited Britain: he might
have been assassinated before he had a chance. It is metaphysically (but not epistemically) possible
that nobody ever had any thoughts: after all, sentient life might have failed to evolve. But it is
metaphysically (as well as epistemically) impossible that 2 + 2 = 5; and if Elizabeth II is the
daughter of George VI, then arguably it is metaphysically (but not epistemically) impossible that
she was not. There is no possible world at which that very person both existed and had different
parents (Kripke 1980, 112–113).
Plainly, A-miracles are possible in both senses. Extraordinary events happen all the time, and we
can expect an event of any given level of extraordinariness to happen if we wait long enough. Thus,
for example, if you keep dealing out bridge hands from a shuffled pack, it is practically certain that
eventually you will deal out a “perfect” hand (where each player has thirteen cards of the same suit).
Arguably, the same goes for C-miracles. C-miracles do not necessarily violate the laws of nature:
all that is needed to make an event a C-miracle is that it should be the product of divine agency,
independently of the laws of nature. For instance, it might be consistent with the laws of nature that
a person should spontaneously recover from an illness, but if this has been brought about by God,
then it is nonetheless a C-miracle. It seems that C-miracles are therefore both e- and m-possible: at
any rate, it seems that they are if God is. Of course, one might doubt whether God could influence
matter without violating fundamental conservation laws that happen to hold in our universe. But
that our universe conserves its total mass-energy, for example, is arguably neither knowable a priori
nor metaphysically necessary; nor, then, is the proposition that God never adjusts this quantity.
B-miracles violate the laws of nature, and one might think such violations at least e-impossible.
After all, if laws of nature are at least true (and so exceptionless) statements of general regularities,
then violations of the laws of nature never actually occur (as in, e.g., Lewis [1980] 1986a, 112ff.).
So B-miracles are e-impossible. Note that it is not e-impossible, on this view, that the sun should
stand still or go backward or that the Red Sea should part. What is e-impossible is that these events
should happen B-miraculously. But it takes empirical investigation to show that these events are B-
miracles. In any case, if laws of nature are all true generalities, then no B-miracle ever has occurred
or ever will occur. (One might object that some laws are defeasible in the sense of admitting of
objections, for instance because they involve ceteris paribus clauses. This conception of law, and
the attitude that it motivates toward the e- and m-possibility of B-miracles, will then be close to
J. L. Mackie’s view of them as briefly discussed below.)
This way of thinking about laws of nature leaves open that B-miracles are metaphysically pos-
sible, that is, that there is a possible alternative course of history, in which different laws hold and
where events occur that violate our actual laws. No contradiction is involved in supposing, for
example, that a man walks unsupported on water, although this would violate some actual law.
Indeed, our ordinary ways of thinking about counterfactual scenarios, and perhaps about causality
itself, may depend on the metaphysical possibility of such violations. This will be so if history is
deterministic, so that to conceive of a world in which history diverged from its actual course is to
conceive of this having been brought about by some unobtrusive miracle (Lewis [1979] 1986b;
for the connection with causality see Lewis 1973).
We can also argue, given a more demanding conception of laws, that B-miracles are neither e-
nor m-possible: they never actually happen, and they never could have happened. Benedict de Spi-
noza gave an obscure argument to this effect:
As nothing is necessarily true save by divine decree, it is plain that the universal laws of nature
are decrees of God following from the necessity and Perfection of the Divine nature. Hence,
any event happening in nature which contravened nature’s universal laws, would necessarily
also contravene the Divine decree, nature, and understanding; or if anyone asserted that God

Miracles: Atheism
acts in contravention to the laws of nature, he, ipso facto, would be compelled to assert that
God acted against His own nature—an evident absurdity. (Spinoza 1951, 83)
The argument seems to be that since the laws of nature are inevitable products of God’s nature, it is
impossible that they should be violated, and absurd that God should will that they be violated.
Notice that the m-impossibility of B-miracles follows from weaker assumptions than what appear
to be Spinoza’s. If laws of nature are metaphysically necessary, their violation is metaphysically
impossible. Hume explicitly rejected that assumption (Hume 1969, 1.3.6), but others have
endorsed it, or at least entertained it, on nontheistic grounds (as in, e.g., Shoemaker 1998).
However, and as one might expect, taking a more relaxed view of laws makes room for both the
e- and the m-possibility of B-miracles. Suppose that laws of nature are just true general descriptions
of the undisturbed course of nature, where “undisturbed” means undisturbed by any supernatural
agency (Mackie 1982, 20). This leaves open B-miracles as epistemic and metaphysical possibilities:
at any rate, it appears to be a coherent supposition, which we cannot rule out a priori, that God
should interfere with the course of nature in this way.
As with “miracle” itself, I don’t propose to lay down a single definition as capturing what pos-
sibility really is or what laws of nature really are. It is more informative and less tendentious simply
to keep track of their different senses, of what they imply for our questions, and of the significance
of these implications. They can, for now, be summarized as follows: A-miracles and C-miracles are
both possible in both senses; as for B-miracles, the position is as seen in Table 1:
Table 1. Modal Status of B-miracles
Laws of nature as:
True generalities Necessary generalities Interference-free generalities
(Hume, Lewis) (Spinoza, Shoemaker) (Mackie)
e-possible? No No Yes
m-possible? Yes No Yes
So much for whether miracles in some sense could occur. Should we think any did?
ARE MIRACLES ACTUAL?

We have seen Spinoza’s argument that B-miracles, being violations of God’s will, never could
(and so never do) occur. It is unclear how anyone could demonstrate the nonoccurrence of
A-miracles—after all, extraordinary things do happen. Even Hume’s essay “Of Miracles” (1975 [1777],
chap. 10), which I discuss in this section and which is perhaps the highlight of the entire literature,
doesn’t do that. But it does get us close. More precisely, what Hume’s argument seems to show is that
no testimony could make it rational to believe in a sufficiently extraordinary A-miracle. We can intro-
duce the argument by means of a thought experiment.
Suppose we are randomly screening for a very rare mutation X, present in one in every 10 mil-
lion people. The test is positive for 99 percent of carriers and for only 1 percent of noncarriers. Now
suppose that you take the test and it is positive. Doesn’t this make it highly likely that you have got
the mutation?
In every billion people, 100 are X-carriers. On average 99 percent of them (99 people) will test
positive. But 1 percent of noncarriers, about 10 million people, will also test positive. You should be
about 100,000 times more confident of being in the second group, because there are 100,000 times
more people in it. So you should be practically certain that your positive result is misleading.
Our present-day evidence for miracles derives entirely from (scriptural) testimony. How much
support does that testimony really give those beliefs? Practically none, said Hume, and for the same
reason that a positive test result gives little support to the conclusion that you are a carrier. After all,
which is more likely—that the Red Sea should part just long enough for the escape of an enslaved
people, or only that they should preserve such a myth? That a corpse was reanimated after three
days, or only that half the world should be deceived into believing this? Hume concludes with ven-
omous irony that “the Christian Religion not only was at first attended with miracles, but even at
Miracles: Atheism
this day cannot be believed by any reasonable person without one” (Hume 1975 [1777], 131).
(This paragraph and the preceding three are adapted from Ahmed 2015a.)
This argument, if successful, tells equally against the believability of testimony to any suffi-
ciently unlikely A-miracle. Although on Hume’s own definition a miracle violates a law of nature,
the argument is in fact indifferent as to whether the reported event violates laws of nature on any
view of laws of nature. All that matters is whether it is more likely (a) that the event should have
occurred or (b) that this report of it should have been mistaken. If (a) is more likely, then we have
a prima facie reason to believe that the miracle really took place. If—as will typically happen with
A-miracles—(b) is much likelier, then we don’t have any such reason, and the testimony can reason-
ably be disregarded.
This conclusion might seem a little strong. After all, someone might try to make a cumulative
case from all the testimony to different miracles in a given religious tradition. Just because testimony
in a single case is weak, it doesn’t follow that that testimony can be thereafter disregarded: it contri-
butes to the cumulative case. However, testimony to many miracles, even within a single religious
tradition, is likely to lend more support to the hypothesis that the religious tradition itself encour-
aged the generation and propagation of such tales than to that of an underlying supernatural truth
that the tradition reflects—unless, that is, one starts out with high confidence in the sort of theistic
hypothesis for which the miracles themselves were supposed to be providing an independent basis.
It is only slightly less obvious that the argument, if successful, tells equally against the believ-
ability of other forms of “testimony” to A-miracles, for instance the testimony of the senses. I seem
to see a man walking on water. What is more likely—that he really is walking on water or that my
eyes are deceiving me? Well, on the one hand, my eyes have certainly deceived me in the past—or at
least, people like me have been taken in by illusions; on the other hand, nobody has ever walked on
water—or at least, there is not a single undisputed case of this in all of history. It is therefore more
reasonable to suppose that my eyes are deceiving me on this occasion.
Hume’s argument has provoked an enormous range of responses since its publication in
1748. One very intuitive objection (Hambourger 1980, 591–592) is that if it is valid then we
should never believe testimony to any unlikely event. For instance, suppose I tell you that the
winning numbers in the UK National Lottery on 24 February 2018 are: 2, 11, 35, 40, 48, 55.
The odds of this happening are in the range of 14 million to 1, but people misreport sequences
of numbers much more often than that. Doesn’t Hume’s argument imply (absurdly) that you
shouldn’t believe me?
It does not. What we must weigh against one another for likelihood are (a) that the event took
place and (b) that this report of it both took place and was mistaken. It is much more likely that my
report of a lottery result is mistaken than that the winning numbers on 24 February are really 2, 11,
35, 40, 48, and 55. But that those were the winning numbers is more likely than that: they are not,
and yet I mistakenly reported those exact numbers. If any mistaken report is as likely as any other,
the probability that a mistaken report involves just those numbers is about 1 in 14 million. (And if,
as will almost always happen, many mistaken reports are much more likely than that one, then its
probability may be very much less.) If my reports are mistaken on average about 1 time in 10, this
means that you should in the face of my report have a confidence of at least about 90 percent that
the winning numbers are indeed 2, 11, 35, 40, 48, and 55.
The disanalogy between the lottery case and the miracles case is that in the miracles case, not all
misreports are equally unlikely—in fact, some are not unlikely at all. For instance, it is not especially
unlikely (a) that false reports of the events of Jesus’s life should include false reports of miraculous
healing, and (b) that false reports of the events following his death should include false reports of
resurrection—rather than of, say, alien abduction—because (a) there existed a Jewish tradition of
messianic prophecy that specifically mentions miraculous healing, for example, Isaiah 35:5–6, and
(b) the Old Testament appears to contain prophecies of bodily resurrection (e.g., Job 19:25–27; Isa-
iah 26:19: Daniel 12:2); it mentions some episodes of what may be resurrection (e.g., 1 Samuel
28:11–20; 1 Kings 17:17–22); and stories of resurrection were fairly common in the ancient Near
East (e.g., Dionysus, Osiris, and Salmoxis; for discussion see Mettinger 2001; Smith 2001, chap.
6). As David Friedrich Strauss remarks in connection with (a), “It is always incomparably more

Miracles: Atheism
probable that histories of cures of the lame and paralytic in accordance with messianic expectation,
should be formed by the legend, than that they should really have happened” (Strauss 1902, 457).
A second objection, originating with Charles Babbage (1838; see also Holder 1998, 53; Ear-
man 2000, 53–56), is that even if the report of one witness should not convince anyone of an A-
miracle, the concurrent testimony of enough independent witnesses certainly can. If, say, 100 wit-
nesses independently report seeing a man walking on water, then you might think that the proba-
bility that they are all mistaken is minute—far less than the probability that a man did walk on
water. And there are miracles to which we have extensive testimony—consider for instance the mir-
acle of Fátima of October 1917, in which at least 30,000 people claimed to have seen the sun mov-
ing erratically in the sky, or Saint Paul’s implication of more than 500 witnesses to the resurrection
in 1 Corinthians 15:3–9 (although he does not comment on their independence). So even if Hume
is right that one witness is never enough, this may be irrelevant when we have many.
The difficulty with this argument (Ahmed 2015b) is that the requisite kind of independence is
not easily achieved. It is not enough, for instance, that there is no collusion or other communication
between the witnesses. What is needed is that there be no remotely probable common cause of the
reports that might operate independently of any miracle. For instance, and in connection with the
Fátima case, suppose we cannot rule out the possibility that staring at the sun for long enough caused
the witnesses to suffer illusions of erratic solar motion. Then as long as this possibility is more likely
than that the sun itself spontaneously started to move erratically, it remains more likely even given
30,000 noncolluding witnesses, all testifying that that is what they saw. More generally, as more
and more witnesses testify to a miracle, we should be more and more confident of whatever natural
common causes of illusion, or delusion, seemed feasible at the outset. Since the likelihood of such a
natural common cause of mistaken testimony will always outstrip that of a miracle, even very extensive
noncollusive testimony to a miracle should ultimately be rejected on broadly Humean grounds.
It has been an assumption of the whole discussion so far that resurrections, or other events of
the sort usually called “miraculous,” really are highly unlikely, and certainly much more unlikely
than the hypothesis that this or that testimony to one is fraudulent or mistaken. The third objec-
tion, due to George I. Mavrodes, puts pressure on this premise. For after all, why do I think it
unlikely that this or that person was resurrected? Well, nothing in my experience confirms it, and
everything speaks against it. But my experience of the matter—or yours—is relatively narrow:
nobody has observed the postmortem activities of more than a tiny fraction of all persons whose
postmortem activities have been observed. But Hume himself appeals not only to my experience,
or yours, but to everyone’s experience, writing for instance that “it has never been observed in
any age or country” that a dead man should come to life (Hume 1975 [1777], 115). This, however,
is exactly what is at issue. Thus Mavrodes writes:
[Hume] says not merely that he has never observed a resurrection, but that a resurrection has
never been observed in any age or country. But he suggests absolutely no evidence at all for
this latter claim. But the fact is that some other people profess that they have personally
observed just such events. When it comes to resurrections, for example, there have apparently
been some people who claimed to have seen Jesus a few days after his execution, alive and well,
talking with them, eating with them, and so on. (Mavrodes 1998, 180)
It may perhaps be question begging to assume, as Hume does, that nobody has ever observed a
resurrection. But the present version of the Humean argument needn’t assume that much. All the
argument needs to suppose at the outset is that one is reasonable, prior to hearing any witness’s tes-
timony to it, to be extremely confident that this or that miracle did not occur. We can therefore get
by on the weaker assumption that the uniform experience of humanity has been that miracles occur
at most with vanishing rarity. And the assumption, that that has been the uniform experience of
humanity, is both dialectically legitimate and true.
A final objection puts pressure on the same point—that is, it too questions whether we can rea-
sonably start out by assuming the extreme unlikelihood of this or that miracle. For suppose that we
start out by giving significant credence to the supposition that God exists, that he takes an interest
in human affairs, and that one manifestation of this interest is his occasional miraculous interven-
tion in the course of history, for instance, by causing someone to return from the dead. In that case,
Miracles: Atheism
we will be reasonably open to the idea that this or that miracle occurred: we may consider it more
likely, for instance, that a person should have been raised from the dead than that detailed and
apparently sincere testimony to this, from an initially skeptical witness possessed of common sense
and integrity, should be so egregiously mistaken. But if we do think it more likely, then this per-
son’s testimony to that miracle should count as good evidence for it. Putting it another way and
more briefly: Hume’s entire argument proceeds on the assumption that one starts out with atheist
convictions and so can never appeal to anyone who isn’t already in that position.
It’s true that if one starts out thinking that resurrections and so on are reasonably likely, then
one will reasonably be open to being convinced of one by the right kind of testimony. But first,
it needn’t be true that the only way not to occupy this position is to be an atheist. After all, one
might take Spinoza’s position (as described in the second section, “Are Miracles Possible?”): God
almost certainly exists, and is omnipotent, but for that reason we should deny that there are any
B-miracles; insofar as the miracles in question count as B-miracles (and resurrections, etc., surely
do), we therefore have reason to think them most unlikely. More important, although there may
in principle be a rational route to belief in God other than via testimony or direct revelation, and
although philosophers and theologians have tried for millennia to find one, the fact remains that
for almost everyone in the history of the world, the only actual grounds for belief in (say) the
God of the Christian religion, is the testimony of long-dead witnesses concerning events that took
place in antiquity. So, although Hume’s argument may not move someone who is already suffi-
ciently convinced of the truth of that religion, it does still function as a sort of prophylactic. Anyone
who starts out skeptical of Christianity and is tempted to accept it after reading or hearing Saint
Paul’s and the Gospel writers’ accounts of the resurrection, should in fact remain skeptical for the
reasons that Hume gives. And what goes for Christianity also goes for Islam, Judaism, and any other
religion that is founded on miracles. It is therefore fair to describe Hume’s argument as (in his
words) “an everlasting check to all kinds of superstitious delusion” (1975 [1777], 110).
Finally, it is worth emphasizing the limitations of the Humean or neo-Humean argument pre-
sented here. The argument does not show that A-miracles never occur—extraordinary things do
happen often enough. Nor does it establish that no kind of evidence could make an initially skepti-
cal person highly confident that this or that miracle occurred. What it does establish is that neither
individual nor—as I suggested—collective testimonial evidence could do as much. But since, as
things stand, our evidence for the central A-miracles of the major religions is entirely testimonial,
this conclusion still has cosmic significance. We do not now have, and it is hard to see how anyone
ever could have, any good evidence that those events took place.
RELIGIOUS AND SCIENTIFIC “INTERPRETATIONS”

Up to this point we have assumed, in line with most thinking on the subject, that the status of
“miracle” is properly awarded to events themselves, and we have seen arguments over whether any
events do merit, or can reasonably be believed to merit, that status. But we may imagine that
whether an event counts as miraculous depends in part on the worldview or framework within
which one is considering it. The very same event may be miraculous relative to a religious frame-
work but not miraculous relative to a scientific framework.
This sort of relativity is what Ludwig Wittgenstein seems to have in mind in the following
passage:
Take the case that one of you [i.e., in the audience for this lecture] suddenly grew a lion’s head
and began to roar. Certainly that would be as extraordinary a thing as I could imagine. Now
whenever we should have recovered from our surprise, what I would suggest would be to fetch
a doctor and have the case scientifically investigated and if it were not for hurting him I would
have him vivisected. And where would the miracle have got to? For it is clear that when we
look at it in this way everything miraculous has disappeared; unless what we mean by this term
is merely that a fact has not yet been explained by science which again means that we have
hitherto failed to group this fact with others in a scientific system. This shows that it is absurd
to say “Science has proved that there are no miracles.” The truth is that the scientific way to
look at the fact is not the way to look at it as a miracle. (Wittgenstein 1993a [1930], 43)

Miracles: Atheism
Clearly the sense of “miracle” in which one can see something as miraculous is that of a B- or C-
miracle: a violation of the laws or a work of God. After all, everyone agrees that a man’s growing
a lion’s head is an extraordinary event: there is no relativity about that. What Wittgenstein has in
mind might be put like this: the same event could be seen by one person as simply being as-yet sci-
entifically unexplained and by another as a sign of God’s grace, and yet neither person is wrong
about the facts. It is rather that one person views reality through the grid of scientific explanation,
and the other views the same reality through the grid of religious or spiritual explanation; and reality
itself cannot settle which grid to impose on it.
This way of thinking about miracles reflects Wittgenstein’s attitude toward the analysis of alien
cultures that some are inclined to regard as “primitive.” For instance, he thinks that it is a mistake of
anthropologists like J. G. Frazer to think of primitive rituals, such as the rain dance, as manifesta-
tions of a mistaken scientific theory. According to Wittgenstein, the situation is not that they are
ignorant of the relevant facts—it is rather that they choose, or find it natural, to put a different
interpretation on them. “I believe that the characteristic feature of primitive man is that he does
not act from opinions (contrary to Frazer).… Frazer represents these people as if they had a
completely false (even insane) idea of the course of nature, whereas they only possess a peculiar
interpretation of the phenomena” (Wittgenstein 1993b, 137, 141; cf. Frazer 1966, 2).
What Wittgenstein leaves unclear, however, is what this “difference in interpretation” really
involves: in other words, by what criteria we can distinguish a difference in opinions from a difference
in interpretations. Why should we call the difference between “Western man” and the “primitive”
objects of Frazer’s study differences of interpretation and not of opinion? And similarly in the context
of miracles—what is the difference supposed to be between (a) believing that an event is a C-miracle
and (b) putting that interpretation on the event? It is hard to say; and to speak for myself I find it
much more attractive to accept a sort of gradualism, according to which there is no firm distinction
between matters of opinion and matters of interpretation, but rather a sort of shading-off from ques-
tions that have relatively specific empirical content (“Is there a rabbit in the room?”—although this
question may itself be “theory laden” to some extent) via more theoretical questions (“Did water turn
into wine at Cana?”), all the way to highly abstract questions (“Are there numbers?,” “Are there mira-
cles?”) that are still questions of fact but whose resolution depends, for any inquirer, on his or her
other beliefs and on a very wide range of more empirically focused findings. That line, which is in
essentials that of Willard Van Orman Quine (1980, sec. 6), would take far more space to elaborate
and to defend than is available here (but see Ahmed 2008, sec. 3.3, for further discussion).
In any case, it is worth questioning to what extent the facts support attributing the “interpreta-
tional” view of miracles to religious believers, even supposing that we can make good sense of it.
Saint Paul could hardly have been plainer when he said that Christianity stood or fell with the fac-
tuality of the resurrection (1 Corinthians 15:13–17): what is at stake is what happened, not whether
an agreed-upon course of events ought to be “interpreted” as miraculous. Perhaps we can say some-
thing similar about miracles in alien cultures. According to one of its own spiritualists, the Dagara
people of Burkina Faso
are well known throughout West Africa for beliefs and practices that outsiders find both
fascinating and frightening.… What in the West might be regarded as fiction, among the
Dagara is believed as fact, for we have seen it with our eyes, heard it with our ears, or felt it with
our own hands.… A Westerner will say, for example, that water always makes you wet, yet a
native healer who gets into a river and stays for hours doing what healers do might get out just
as dry as if he had been working in the Sahara Desert. (Somé 1999, 2, 11; my italics)
Why should we not take this person at his word? If we do, then whether such miracles as described
here really occur is, for the Dagara as much as for us, a question of fact not interpretation. So even if
that dichotomy makes sense, it is at least an open question whether belief in miracles falls on the
“interpretative” side; and if it falls on the factual side then all the traditional arguments of Hume,
Spinoza, and others are indeed as relevant to the question of miracles as the foregoing discussion
has presupposed.
We distinguished three main conceptions of miracles: (A) extraordinary events; (B) violations of
the laws of nature; (C) divine interventions. A-miracles are actual. Whether B-miracles are possible
Miracles: Atheism
depends on one’s conceptions of (i) possibility and (ii) laws of nature. C-miracles are possible if God
is possible, but they may not be actual even if God is actual. Hume’s arguments give strong reason
to reject individual or multiple testimony to a range of A-miracles; these include all the events
(which are also B- or C-miracles) on which the major monotheistic religions of the world are
founded. It may be possible to regard talk of miracles not as a description of anything that happened
but rather as an interpretation of events that everyone agrees on, but it is unclear both what this
means and whether it is compatible with what actual believers in religion or spirituality say and
think.
AT PRESENT, MIRACLE CLAIMS UNDERMINE THEISM
The miracle stories found in the texts produced by hundreds of religions in recorded history do not
increase the probability of theism. In fact, examined in the aggregate, they decrease the probability
of theism. All texts are historical documents, written and transmitted by fallible and unreliable
human beings, and as such subject to the same epistemological requirements before believing any
claims in them (see this volume’s chapter on “Doxastic Foundations: Atheism”). And in no known
case does any reported miracle become probable by any reliable historical method. Yet the ubiquity
of miracle claims across religions entails a state of contradiction for any theology that proposes any
miracle true. This is difficult for any theology to explain.
There is a difference, of course, between a miracle reported by an eyewitness, someone we
know and can interrogate personally, and an ancient text whose authors and editors remain ulti-
mately unknown and uninterrogated. There is likewise a difference between a miracle reported
recently and subject to a reliable investigation, and a miracle reported anciently that received no reli-
able investigation. And yet that difference is a gaping chasm. No miracle reported in the sacred text
of any religion derives from an eyewitness you or I know personally—nor anyone we can interro-
gate, nor any claim we can investigate; and no miracle reported anciently was ever investigated, by
any known means at all, much less one known to be reliable—other than to be exposed as dubious
(Carrier 2009). And yet lies, errors, delusions, distorted memory, and exaggerated retelling are all
commonly known origins of miracle tales, even today, the more so in premodern times. Posing an
additional problem is that it was routine in antiquity to compose miracle stories as educational sym-
bols and allegories—in the manner of parables—rather than as assertions of factual truth. This was
the function of myth in general (Carrier 2014, 389–395); for example, it seems fairly obvious that it
was the original function of the story of Jesus’s withering of the fig tree and capturing 153 fish (Car-
rier 2014, 433–435, 505–506).
THE VERDICT OF RATIONAL HISTORICAL EMPIRICISM

What we learned from the Enlightenment remains—David Hume having long since been improved
on (Fogelin 2003; Carrier 2012, 310). Apart from apologists who have rejected rational empiricism,
today all historians agree that the supernatural is always the least likely explanation of any tale of the
supernatural. “When we hear tales of talking dogs and flying wizards, we don’t take them seriously”
(Carrier 2012, 114). The Christian Acts of Peter, written in the second century, contained both, as
well as the resurrection of a cooked fish. That those things really happened is never entertained as a
hypothesis, because the status of a text as “sacred” to a given religion makes no demonstrable differ-
ence to its reliability. All texts are produced by human beings, who are all fallible, and just as prone
to dishonesty, gullibility, and error. If the author of the Acts of Peter could fake a talking dog, so
could the author of Exodus fake Aaron turning a stick into a snake; and we can demonstrate no
more reliability in reporting the fabulous in the Acts of Peter than we can in the Acts of the canon.
When we evaluate these claims objectively, the evidence needed to believe them is never found.
There are several sound reasons why modern historians have come to this conclusion (surveyed
in Carrier 2012). Religious texts never score well on the criteria of reliability that many secular texts

Miracles: Atheism
have achieved, criteria developed for precisely the purpose of increasing the reliability of a preserved
testimony, owing to the awful frequency with which preserved testimony is simply not reliable with-
out adequate controls. The fact that we needed to develop those procedures for increasing the reli-
ability of preserved testimony is alone enough to remind us that texts that didn’t employ them often
can’t be trusted. We have extensively confirmed by now that other explanations of a miracle claim’s
existence are always far more probable than the existence of the miracle itself; miraculous phenom-
ena, if they had ever been real, should have produced more and better evidence of their reality by
now. Instead, there is always a suspicious correlation between when and where miracle claims origi-
nate and the convenient unavailability of the actual means to test their veracity, whereas when the
actual means to test their veracity is present, miracle claims are always refuted, either being proved
false or to be an inaccurate description of nonmiraculous phenomena that would have occurred any-
way, on any theology or none.
Miracle claims have been a staple of every religious tradition. They are not peculiar to Chris-
tianity but are common in every other tradition known (pagan: Dodds 1951, Edelstein and Edel-
stein 1945, Cotter 1999; Muslim: Yazicioglu 2013; Jewish: Dennis 2015; and beyond: Woodward
2000). In the mid-second century CE, Lucian of Samosata wrote The Lover of Lies, a treatise on
how absurdly common claims of miracles and magic were, and how readily believed by the gullible.
He was not alone in noticing this (Remus 1983; Carrier 2005, 211–252). Such claims were central
to the Salem witch trials, and there attested by better evidence than any Christian miracle known
(McCormick 2012), yet surely no one of sense believes them today (the same can be said of modern
studies of magic: e.g., Evans-Pritchard 1991). Christian saints had miracle narratives showered on
them with abandon (e.g., McNamara, Halborg, and Whatley 1992). No rational historians credit
these claims as true today. Yet they are no different in their credibility than the claims made in
any scripture—such as for the resurrection of Jesus (Carrier 2010b).
THE REAL ORIGIN OF MIRACLE CLAIMS

Where miracle claims come from varies. Everyone agrees that some have been outright fabrications
(many a miracle having been exposed as a lie); some were deliberate cons (many a miracle having
been exposed as a trick); some, the misreported, misremembered, or exaggerated recounting of
wholly nonmiraculous events; and some were never even intended to be regarded as history but
crafted and told to convey a lesson (see, e.g., Nickell 2013 and Burns 1995; Crossan 2012 and John
2004; etc.). The question is not whether miracle claims are invented by all of these means. It is well
in evidence that they routinely are. The question is whether any of them are actually true or
whether all of them are a product of these other causes.
The first narrative about Jesus, the Gospel of Mark, for example, appears to have used the
model of the parable (as related by Jesus in Mark 4), crafting its miracle stories as allegories for
deeper truths (Carrier 2014, 387–506). For instance, its author cannot have expected anyone
but the gullible to believe his story of Jesus irrationally withering a fig tree for the inexplicable
crime of obeying the laws of nature, which deed would establish Jesus as insane. This tale is,
rather, an allegory for explaining God’s abandonment of the Jewish temple to destruction by
heathens (Mark 11:12–22; Carrier 2014, 433–436). Nor, likewise, could the author of Mark have
expected serious readers to believe that Jesus miraculously murdered thousands of pigs (Mark 5:2–
13; Carrier 2014, 415, 438) or that the sun would vanish for hours and this not be noticed in any
records of the era (Carrier 2012, 41–45, 54–60). The Gospels appear broadly to be allegorical
myth, even when we’ve lost what their original lesson must have been (Crossan 2012; Carrier
2014, 387–506), as was probably the case with most religious literature and its miracle tales (John
2004; Carrier 2014, 214–222). But that’s not always the case. By the time we get to the last nar-
rative about Jesus to be deemed sacred, the Gospel of John, this strategy had changed into one of
fabricating miracle claims to persuade readers to believe the message (“these are written, that you
may believe,” John 20:31), and not just as symbols of the message (Carrier 2014, 487–506). Mir-
acle claims can be fabricated to either purpose: as lessons or as “proofs.” The latter is simply lying.
The former, when taken as factual reporting, is simply misreading the author’s intent. Neither
supports theism.
Miracles: Atheism
In all ancient religions, forged and fake literature (histories and letters) were actually the norm
(Carrier 2014, 214–222). So we rarely have any basis for trusting them even on mundane asser-
tions, much less the miraculous. Given the evidence we have of past examples, we observe the prob-
ability that any tale is a fabrication is always greater than that the extraordinary occurred (Carrier
2012, 113–117). It is now commonly agreed among mainstream experts that the legends of Moses
in the Old Testament are all mythical and relate to no actual historical events (Carrier 2014, 215);
likewise, the miracles of his would-be successor, Jesus, are also considered mythical. Jesus could not
murder thousands of pigs with a wave of his hand because no one can. That simply isn’t a power
documented to exist in the world. There is no known mechanism for it. And there is no evidence
sufficient to believe an exception in his case. The same is so for Moses’s brother Aaron turning a
stick into a snake. Lest we forget, even the “heathen” sorcerers he was doing battle with could turn
their sticks into snakes (Exodus 7:8–13). The only way Aaron could prove that his magic was better,
was by having his snake eat theirs. It’s doubtful the author even intended this to be taken as some-
thing that happened, rather than as a symbol of how Yahweh’s might is the greater. In any event, we
have no reason to believe a word of it.
Fabrication and symbolism aren’t the only origins of miracle claims known to exist. Hallucina-
tion, misperception, and memory distortion are not only widely documented natural phenomena
that can cause any number of false tales but are peculiarly common in early religious movements
(Carrier 2014, 124–137). Another common cause is merely embellishing a natural phenomenon
into a miraculous one, either deliberately, carelessly, or erroneously. In the Christian tradition, for
example, the author of the canonical book of Acts narrates the Christian church engrossed in the
“miracle” of “speaking in tongues,” depicting it as Christians being able to speak in many foreign
languages they had never learned, such that pilgrims native to those languages could understand
them (Acts 2), implying this proves the truth of their theological beliefs. But from the only
actual eyewitness we have, the apostle Paul, we know that’s not what actually happened at such
events. In truth, “speaking in tongues” (glossolalia) meant speaking in unintelligible alien languages
(often presumed to be angelic), which required spiritually inspired “interpreters” to understand
(1 Corinthians 12–14). It was similar to pagan oracles, where the oracle spewed unintelligible utter-
ances, and a set of “interpreters” (prophets) standing by would decide what the utterances meant.
There is nothing miraculous in this. In fact, the phenomenon is widely attested across numerous
religions and has been studied scientifically (Carrier 2014, 124–125). What has happened, then,
is that the author of Acts has taken a mundane phenomenon, embellished it into something mirac-
ulous, and then used it as evidence of the truth of his religion, which combines a nonmiraculous
truth with a lie. There is no evidence that anyone ever actually did what Acts reports. No such
power has ever been witnessed in any verifiable way, not even by the apostles themselves.
Similarly, many miracle claims are reports or embellishments of natural psychosomatic phe-
nomena and thus not actually miraculous. Indeed, many a miracle claim can be both false and mun-
dane, since, just like glossolalia, the most typically observed “miracles” in the Christian experience
would be exorcisms and healings that were not actually miraculous; but because ancient witnesses
and storytellers did not know these things were nonmiraculous, as false tales of exorcisms and heal-
ings were created, what was widely believed to be miraculous came to be used as a model even for
fabricated miracles. Thus, in the Gospels, Jesus never performs any healing or exorcism that doesn’t
match known psychosomatic phenomena but for a few embellishments that are obviously fabri-
cated—such as expelled demons killing thousands of pigs or a severed ear actually being “healed,”
events for which we have no eyewitness source (indeed, with the ear, when you compare Luke
22:50–52 with his source, Mark 14:47–48, you find Luke has embellished the story with a healing
not found in Mark’s account).
That psychosomatic ailments are subject to psychosomatic cures is a natural phenomenon well
studied by science. And the ailments that arise and respond in this way will vary by culture (Shorter
1993; O’Sullivan 2015). Common ones are blindness and paralysis. Another is “demonic posses-
sion” (Caciola 2003; Goldish 2003). They are always ailments that can be produced entirely by
mental illness or delusion. Even death can respond this way, a condition so frequently misdiagnosed
as to readily produce the appearance of a resurrection (Carrier 2009, 86–89; Bondeson 2002). This

Miracles: Atheism
contrasts with ailments that involve documentable biological conditions, such as decapitation, lost
limbs and organs, broken bones, hemorrhagic fevers, uncontrollable vomiting, open wounds and
sores, which never get healed in any credible account, whereas a real (and perfectly natural) psycho-
somatic event can be observed and misremembered or misreported as more fabulous than it was.
Without the ability to verify the facts of the case, we can never know. And in “scriptures,” we are
never allowed to verify such claims.
MIRACLE CLAIMS ARE EVIDENCE FOR ATHEISM

The fact that ancient peoples did not know the distinction between psychosomatic and supernatural
healing is another clue to how we know they were never in the presence of the genuinely miracu-
lous. Jesus is alleged to have cured people whose conditions match psychosomatic illnesses that
would be naturally amenable to psychological cures, yet he never cures amputees or anyone for
whom psychological cures would be naturally ineffective (the incident with the ear being the later
embellishment of a tradition that began without it); and more important, for example, he never
cures cholera, by which I do not mean he never cures someone with cholera—though notably, he
never is depicted doing so, despite that and other lethal diseases being so rampant that half of all
children in his day did not survive childhood, and barely one in three who did survive childhood
lived beyond the age of fifty-five (Carrier 2014, 151). I mean, he never cures cholera.
Humans have the natural ability to thwart the spread of cholera on a mass scale now, and have
literally cured entire diseases nearly the world over (like smallpox, polio, and soon malaria). If
humans can naturally effect feats that vastly dwarf what a god was supposedly able to effect miracu-
lously, this does not lend any credence to one’s actually being a god. Of course, if you are going to
fabricate a truly divine miracle, such as eradicating smallpox (which killed tens of millions in antiq-
uity), you aren’t likely to succeed in claiming a disease has been miraculously eradicated that is
observably still around nor in claiming a disease once existed that no other sources attest. The only
miracles you could fabricate are miracles no one who would question them could verify. Which are
decidedly unimpressive, compared to what humans can naturally achieve without miracles at all.
That this feeble, unverifiable feat is precisely the only kind of miracle ever attributed to Jesus, and
likewise Muhammad, Buddha, and every other saint and sage in sacred traditions, confirms that
none of these figures ever really held supernatural powers (Carrier 2005, 211–252).
Today, explanations of miracle stories without the miracle being true are always more probable
than explanations requiring their truth, because we have a vast array of background evidence estab-
lishing that those things happen (the many and varied causes of false miracle claims just surveyed),
but no background evidence establishing that miracles happen (as opposed to merely reports of
them). So the question is not whether miracles occur. Like the question of whether magical spells
are real or urban legends are true, the evidence fails to support their reality. The question is rather,
what explains their existence and influence. Why do people believe them?
People once believed many false things, including alternative accounts of chemistry (e.g., phlo-
giston) or biology (e.g., vitalism) or psychology (e.g., spiritualism), and much else. Miracle asser-
tions likewise depend on anti-scientific assertions about how the world works. People believe many
false things that are even compatible with science—from Bigfoot and UFO fantasies (e.g., Loxton
and Prothero 2012; Pflock 2001) to an endless array of absurd conspiracy theories. But miracles
are by definition incompatible with any science that has not discovered supernatural causation.
Yet if such causes existed, they should have been discovered scientifically by now. Contrary to a
common misapprehension, scientific methods in no way even mention much less exclude discovery
of the supernatural (Carrier 2010a).
Confirming this conclusion, miracles always contradict each other in their root explanation. It
cannot be that miracles are “caused by the Catholic God,” for example, when they happen just as
often in Islam, Budhhism, Hinduism, other sects of Christianity, and every pagan tradition known.
That miracles are always used to affirm the truth of a theology that is contradicted by the ubiquity
of miracles in contrary theologies confirms that miracles are not being believed in because of their
reality but because of a need for their reality. Even if miracles were real, any sound “unified theory
of miracles” would entail that no theology is true, precisely because that is the only way to explain
Miracles: Atheism
how all theologies produce them. Which leaves us with the prospect of psionics: miracles as actually
the effects of human supernatural powers (or mischievous spirits unconcerned with what divinity we
credit their work to), which is then mistakenly attributed to a divinity. But after centuries of con-
certed failure by the field of parapsychology to produce any replicable results in psionics or to verify
any miraculous power or phenomenon, we have no basis to believe that theory either. Indeed, the
million-dollar James Randi prize remained unclaimed for half a century, despite the fact that just
a single successful demonstration of a miracle would have earned it (see Wikipedia’s entry on the
“One Million Dollar Paranormal Challenge”).
This last is the final clue. It is not only that we lack the evidence we should expect to have
found by now for any of the kinds of miracles commonly claimed (whereas we have all the evi-
dence we need that the other causes of miracle claims not only exist but are frequent). It is also
that we lack the evidence we should expect to have found by now for miracles that are actually
impressive, as in indicative of an overwhelmingly superhuman power, not merely a superhuman
power that is conveniently trivial enough to go unnoticed by all but a few. It’s more than merely
peculiar that a supposedly resurrected Jesus “miraculously appears” to a small number of fanatics
and only one persecutor, and not to all the intellectuals and leaders of the world, from Rome to
China; yet only doing the latter would exhibit the goals and powers indicative of an actual god.
Moses, apparently couldn’t do a single feat earning mention in Egyptian chronicles and inscrip-
tions, despite the authors of Exodus fabricating quite astounding ones. And Muhammad suppos-
edly broke the moon in two and reassembled it on command, yet suspiciously no other culture on
Earth noticed this.
But here the point is not just the poor attestation miracles always have, but that they are always
underwhelming on the scale of what we should expect from a god. We hear of “miracle claims” like
a lone resurrection in some rural community distant from media and medical science, but we do not
hear of stopped tidal waves, vanished pathogens, artillery turning into flowers, cities teleported out
of the path of a collapsing dam. In other words, the miracles that get fabricated are conspicuously
unimpressive compared to what the alleged deity is supposed to be capable of. The coincidence of
this (underwhelming miracles, used to prove the existence of overwhelming powers) is likewise evi-
dence against theism, as atheism can perfectly well explain why this should be, but no theism can,
without elaborate gerrymanders that, for want of evidence for them, reduce theism’s probability
anyway. Atheism requires none.
So the total absence of evidence here is very strong evidence against the authenticity of the
miraculous (Carrier 2012). The inability of science to verify any supernatural cause even after hun-
dreds of years of searching; the peculiar coincidence of miracle claims almost always being made pre-
cisely when they cannot be effectively tested and being scientifically refuted when they can be effec-
tively tested; and the absence of all the evidence that should exist if supernatural causes did;
combined with the fact that miracle traditions all contradict each other—all of that forces no other
conclusion than that atheism, and its concomitant account of natural, fully verified causes for mira-
cle claims, is a better explanation of miracle claims than any theology.
And that entails that the evidence of miracle claims, as we actually have it (with all its suspi-
cious peculiarities and contradictions), actually increases the probability of atheism (Carrier 2005).
Had any theism been true, the evidence of the miraculous would look conspicuously different. To
escape this obvious conclusion, theologians will resort to elaborate gerrymandering to rescue their
preferred theology by explaining away this inconvenient evidence. But they have no evidence that
any of their proposed gerrymanders are true, rather than just what they have to make up to get
the evidence to fit their theory. And in the logic of probability, all such gerrymanders actually
reduce a theory’s probability; they do not rescue it (Carrier 2012, index, “gerrymandering”).
WHY APOLOGETICS IS INEFFECTIVE

It does no good to respond to this, as Craig Keener does (2011), by merely documenting more
miracle claims. Documenting that miracle claims exist, and continue to be made, does not dem-
onstrate any of them are true. This represents a complete lack of any scientific epistemology.
Yet the whole point of the scientific revolution, begun under Thales and completed under

Miracles: Atheism
Isaac Newton, was the realization that merely documenting a story does nothing to verify it. Con-
trast the gullible, premodern, unscientific methodology of Keener with the critical, modern, scien-
tific approach of Joe Nickell to exactly the same kinds of claims (e.g., Nickell 2013) or with the
Indian Rationalist Association (e.g., Burns 1995). That is the difference between modernity and
the Middle Ages. Keener needs to catch up with the times. He seems the sort of fellow who would
have been mocked by William Gilbert in 1600 for merely believing anything told him. Keener’s
two volumes of fabulous tales no more prove any of them true than the six volumes of tales of sor-
cery collected under the direction of Bengt Ankarloo and Stuart Clark (1999–2005) prove that
magic or witchcraft are real. All the problems just surveyed in this section remain. And they con-
tinue to weigh in favor of atheism, not theism.
It is also no response to dismiss this evidence as “bias against the supernatural,” because it is
not. It is evidence against the supernatural. Not bias. The improbability of the supernatural relative
to all the known causes of false claims of it is a documented fact. It is not an assumption. Nor can it
be idly claimed that the universe changed at some point, that the miraculous used to happen at
some point in the past and then God stopped meddling for some reason. That’s a gerrymander. It
is an ad hoc excuse for which there is no evidence. It’s merely something you have to make up to
explain away all the evidence against your belief. And being that there is no evidence for it, presup-
posing it actually makes any theism dependent on it improbable.
Some might instead try to argue that miracles are divinely engineered natural phenomena. Like
saying God arranged the universe at the big bang so that a mighty but wholly natural wind would
blow a pathway through the Red Sea precisely when Moses needed to cross it or so that a nurse
would just happen to be at the same restaurant where your daughter chokes on a bone. But we know
how to test that (Diaconis and Mosteller 1989). And no one has ever shown that any supposed actual
coincidence was beyond the expectations of normal random chance. Either the claimed coincidence
did not exist or was not all that coincidental or was not coincidental enough to require any inter-
vention to have caused it. Similarly, some will claim that “their scriptures” miraculously predicted
modern science. But that has never happened. If any scripture had actually predicted a remarkable
scientific fact, it would not have required discovering that fact a thousand years later before anyone
noticed said fact was “in scripture.” Actually, the evidence of the sciences counts against every known
theism (see this volume’s chapters on “Science: Atheism” and “Doxastic Foundations: Atheism”).
Atheism entails the observation that random good luck and bad luck will be observed in the
world. It therefore predicts that any event that we can confirm that seems miraculous will be physi-
cally explicable (because it is not really miraculous) and rare (because it is actually the outcome of a
random congruence of events). Once again, without an array of excuses for whose truth there is no
evidence, theism predicts instead that miracles will be common and physically inexplicable. We
should expect to see things like Christian or Muslim “healing wings” in hospitals where amputees
have their limbs restored by prayer, or something akin, even if not that. But nothing of the kind
is observed. Observations thus match what atheism predicts, not theism (without those unevidenced
excuses that reduce theism’s probability relative to atheism). Atheism also predicts the only miracle
claims that will “survive scrutiny” are claims that are never reliably investigated and that every time a
miracle claim does get proper scrutiny, it dissolves. Whereas theism of itself does not predict that
should be the case. Yet that is precisely what is observed. Adding excuses to explain that only
reduces the probability of theism. So even the evidence presented of “miracles” increases the proba-
bility of atheism over theism.
This is as true of contemporary miracle claims as of claims in sacred texts, which work even
more to the same conclusion, given that they are as frequent for all theologies (which should be
impossible, again without unevidenced excuses), less credible in their construction (being often
more fabulous than anything of the like that is reliably confirmed in reality), and far more removed
from adequate tests (such as we would ever need to believe them). Living gods don’t need ancient
poorly attested miracles as evidence of their creeds. Living gods can work living miracles. The reli-
ance, therefore, on long dead tales to support the existence of living gods, is a fallacy of the first
order. It would only be necessary in a world without gods. Which is why we can know such is
the world we live in.
Miracles: Atheism
Bibliography
Ahmed, Arif. “W. V. Quine.” In The Oxford Handbook of American Philosophy, Earman, John. Hume’s Abject Failure. Oxford: Oxford University Press,
edited by Cheryl Misak. Oxford: Oxford University Press, 2008. 2000.
Ahmed, Arif. “Religious Belief: A Natural Phenomenon with Natural Edelstein, Emma, and Ludwig Edelstein. Asclepius: A Collection and
Causes.” https://blog.oup.com/2015/09/religious-belief-natural-causes/. Interpretations of the Testimonies. 2 vols. Baltimore: Johns Hopkins
2015a. University Press, 1945.
Ahmed, Arif. “Hume and the Independent Witnesses.” Mind 124 (2015b): Evans-Pritchard, Edward, with Eva Gillies. Witchcraft, Oracles, and Magic
1013–1044. among the Azande. New ed. Oxford: Clarendon, 1991.
Ahmed, Arif. “Do Miracles Provide Evidence for Religious Belief?” In Fogelin, Robert. A Defense of Hume on Miracles. Princeton, NJ: Princeton
Contemporary Debates in Philosophy of Religion, 2nd ed., edited by Michael University Press, 2003.
L. Peterson and Raymond J. VanArragon. Oxford: Blackwell, forthcoming
Frazer, J. G. The Golden Bough. Vol. 2. 3rd ed. London: Macmillan, 1966.
2019.
Goldish, Matt, ed. Spirit Possession in Judaism: Cases and Contexts from the
Ankarloo, Bengt, and Stuart Clark, eds. The Athlone History of Witchcraft and
Middle Ages to the Present. Detroit: Wayne State University Press, 2003.
Magic in Europe. 6 vols. London: Athlone Press, 1999–2005.
Hambourger, Robert. “Belief in Miracles and Hume’s Essay.” Noûs 14, no. 4
Augustine, Saint. De Civitate Dei (The City of God). Translated by H.
(1980): 587–604.
Betteson. London: Penguin Classics, 1984.
Holder, Rodney D. “Hume on Miracles: Bayesian Interpretation, Multiple
Augustine, Saint. De Utilitate Credendi (On the Profit of Believing).
Testimony, and the Existence of God.” British Journal for the Philosophy
Translated by C. Cornish. In Nicene and Post-Nicene Fathers, 1st series,
of Science 49, no. 1 (1998): 49–65.
vol. 3, edited by Philip Schaff. Buffalo, NY: Christian Literature, 1887.
Hume, David. Treatise of Human Nature. Edited by Ernest C. Mossner.
Babbage, Charles. Ninth Bridgewater Treatise. 2nd ed. London: John
London: Penguin, 1969.
Murray, 1838.
Bondeson, Jan. Buried Alive: The Terrifying History of Our Most Primal Fear. Hume, David. Enquiry concerning Human Understanding. In Enquiries
New York: Norton, 2002. concerning Human Understanding and concerning the Principles of Morals.
(1777.) Edited by P. H. Nidditch. Oxford: Clarendon Press, 1975.
Burns, John. “India’s ‘Guru Busters’ Debunk All That’s Mystical.” New York
Times, 10 October, 1995. John, Jeffrey. The Meaning in the Miracles. Grand Rapids, MI: Eerdmans,
2004.
Caciola, Nancy. Discerning Spirits: Divine and Demonic Possession in the
Middle Ages. Ithaca, NY: Cornell University Press, 2003. Keener, Craig. Miracles: The Credibility of the New Testament Accounts. 2
vols. Grand Rapids, MI: Baker Academic, 2011.
Carrier, Richard. Sense and Goodness without God. Bloomington, IN:
AuthorHouse, 2005. Kripke, Saul A. Naming and Necessity. Oxford: Blackwell, 1980.
Carrier, Richard. Not the Impossible Faith: Why Christianity Didn’t Need a Lewis, David. “Causation.” Journal of Philosophy 70, no. 17 (1973): 556–
Miracle to Succeed. Morrisville, NC: Lulu, 2009. 567.
Carrier, Richard. “On Defining Naturalism as a Worldview.” Free Inquiry Lewis, David. “A Subjectivist’s Guide to Objectivist Chance.” In Studies in
30, no. 3 (2010a): 50–51. Inductive Logic and Probability Vol. II, edited by R. C. Jeffrey, 263–294.
Berkeley: University of California Press, 1980. Reprinted in his
Carrier, Richard. “Why the Resurrection Is Unbelievable.” In The Christian Philosophical Papers, vol. 2, 83–132. Oxford: Oxford University Press,
Delusion, edited by John Loftus, 291–315. Amherst, NY: Prometheus 1986a.
Books, 2010b.
Lewis, David. “Counterfactual Dependence and Time’s Arrow.” Noûs 13,
Carrier, Richard. Proving History. Amherst, NY: Prometheus Books, 2012. no. 4 (1979): 455–476. Reprinted in his Philosophical Papers, vol. 2, 32–
Carrier, Richard. On the Historicity of Jesus. Sheffield, UK: Sheffield-Phoenix 66. Oxford: Oxford University Press 1986b.
Press, 2014.
Loxton, Daniel, and Donald Prothero. Abominable Science! Origins of the
Cotter, Wendy. Miracles in Greco-Roman Antiquity. New York: Routledge, Yeti, Nessie, and Other Famous Cryptids. New York: Columbia University
1999. Press, 2012.
Crossan, John Dominic. The Power of Parable: How Fiction by Jesus Became Mackie, J. L. The Miracle of Theism. Oxford: Clarendon Press, 1982.
Fiction about Jesus. New York: HarperOne, 2012.
Mavrodes, George I. “David Hume and the Probability of Miracles.”
Dennis, Geoffrey. The Encyclopedia of Jewish Myth, Magic & Mysticism. 2nd International Journal for Philosophy of Religion 43, no. 3 (1998): 167–
ed. Woodbury, MN: Llewellyn, 2015. 182.
Diaconis, Persi, and Frederick Mosteller. “Methods for Studying Coin- McCormick, Matthew. Atheism and the Case against Christ. Amherst, NY:
cidences.” Journal of the American Statistical Association 84, no. 408 Prometheus Books, 2012.
(1989): 853–861.
McNamara, Jo Ann, John Halborg, and E. Gordon Whatley, eds.
Dodds, E. R. The Greeks and the Irrational. Berkeley: University of California Sainted Women of the Dark Ages. Durham, NC: Duke University
Press, 1951. Press, 1992.

Miracles: Atheism
Mettinger, Tryggve N. D. The Riddle of Resurrection: “Dying and Rising Somé, Malidoma Patrice. The Healing Wisdom of Africa: Finding Life Purpose
Gods” in the Ancient Near East. Stockholm: Almqvist & Wiksell through Nature, Ritual, and Community. New York: Penguin, 1999.
International, 2001. Spinoza, Benedict de. Tractatus Theologico-Politicus. Translated by R. H. M.
Nickell, Joe. The Science of Miracles: Investigating the Incredible. Amherst, Elwes. New York: Dover, 1951.
NY: Prometheus Books, 2013. Strauss, David Friedrich. The Life of Jesus Critically Examined. 4th ed.
O’Sullivan, Suzanne. Is It All in Your Head? True Stories of Imaginary Illness. Translated by George Eliot. London: Swan Sonnenschein, 1902.
New York: Other Press, 2015. Thomas Aquinas, Saint. On the Truth of the Catholic Faith. Translated by
Pflock, Karl. Roswell: Inconvenient Facts and the Will to Believe. Amherst, NY: Vernon J. Bourke. New York: Doubleday, 1956.
Prometheus Books, 2001. Voltaire. Dictionnaire Philosophique (Philosophical Dictionary). Edited and
Quine, W. V. “Two Dogmas of Empiricism.” In his From a Logical Point translated by Theodore Besterman. London: Penguin, 2004. First
published 1764.
of View, 2nd ed., 20–46. Cambridge, MA: Harvard University Press,
1980. Wittgenstein, Ludwig. “A Lecture on Ethics.” (1930.) In Ludwig
Wittgenstein: Philosophical Occasions, edited by James C. Klagge and
Remus, Harold. Pagan-Christian Conflict over Miracle in the Second Century.
Alfred Nordmann, 37–44. Indianapolis, IN: Hackett, 1993a.
Cambridge, MA: Philadelphia Patristic Foundation, 1983.
Wittgenstein, Ludwig. “Remarks on Frazer’s Golden Bough.” In Ludwig
Shoemaker, Sydney. “Causal and Metaphysical Necessity.” Pacific Philosoph- Wittgenstein: Philosophical Occasions, edited by James C. Klagge and
ical Quarterly 79, no. 1 (1998): 59–77. Alfred Nordmann, 118–154. Indianapolis, IN: Hackett, 1993b.
Shorter, Edward. From Paralysis to Fatigue: A History of Psychosomatic Illness Woodward, Kenneth. The Book of Miracles: The Meaning of the Miracle
in the Modern Era. New York: Free Press, 1993. Stories in Christianity, Judaism, Buddhism, Hinduism, Islam. New York:
Smith, Mark S. The Origins of Biblical Monotheism: Israel’s Polytheistic Simon and Schuster, 2000.
Background and the Ugaritic Texts. Oxford: Oxford University Press, Yazicioglu, Isra. Understanding the Qurʾanic Miracle Stories in the Modern
2001. Age. University Park: Pennsylvania State University Press, 2013.
TOPIC 8
Religious Diversity: Theism

Daniel Rynhold
Professor of Jewish Philosophy
Bernard Revel Graduate School of Jewish Studies, Yeshiva University, NY
This chapter considers religious diversity and its philosophical implications. Topics discussed include the
purported challenge to theism posed by the links between religious diversity and geographical and
biographical contingencies; theistic approaches to diversity including exclusivism, inclusivism, pluralism,
and relativism; and the extent to which religious diversity supports theism.
RELIGIOUS DIVERSITY: THE ISSUES
It is a matter of empirical fact that there exists a plurality of religions in the contemporary world.
This has been the case for much of the history of civilization, but in an increasingly interconnected
world, we now know much more about many more of these religious alternatives than at any point
in the past. The question is, what issues does this religious diversity—and our knowledge of it—
raise that require the use of the philosopher’s conceptual toolkit?
There would be no issue were it not for the fact that the beliefs of some of these religions are in
contradiction with those of others, and/or that they sometimes contain incompatible practices. One
might try to cobble together a formula that would describe all or at least most religions worthy of
the name—for example, that they are worldviews that engage their adherents in the service of a
transcendent reality that is thought to be a guarantor of meaning and directs their adherents toward
some ultimate salvific good. But the problem, as Joseph Runzo states it, is that we nonetheless find
among them “conflicts of belief, conflicts over specific claims about how meaning and value are to
be achieved and about what is the desired telos for humankind” (1993, 195). So, for example, theo-
logically speaking, there are those for whom the ultimate divine being is a personal God as depicted
in the Qurʾan, and those for whom it is Brahman, the impersonal reality at the foundation of Hin-
duism. At the doctrinal level, either Jesus was God incarnate, as Christians believe, or God cannot
possibly be thus incarnated, as Jews maintain. And the issues go beyond propositional beliefs. At
the practical level, whether one treats the Qurʾan or the Torah as the revealed word of God makes
all the difference in the world as to how one acts on a daily basis. Even at the intrareligious level we
find diversity among denominations, though they may in principle require entirely separate treat-
ment and as such are not explicitly discussed in this chapter.
Widespread as such disagreements may be, this diversity would be less cause for concern were
it not for the fact that fundamental religious divergences often occur between epistemic peers.
The Muslim and the Christian, teaching side by side in the department of religion at any given
university, are likely to be as qualified, as insightful, and as critically reflective as each other.
And yet they would contradict each other on all manner of religious propositions and would refer
to wildly different symbols, rituals, and narratives, without the semblance of a neutral rational
method that could resolve these disputes. Thus there are issues at various levels that warrant phil-
osophical attention. For the commitment of intellectually sophisticated individuals to mutually
incompatible positions about fundamental claims—and often about what they consider to be
227
saving truths and practices—provokes questions both meta-

KEY CONCEPTS physical and epistemological. We find believers asking how
religious diversity might affect their own commitments, ask-
ing what justifies their commitment to one set of religious
Basicality
claims over those of their epistemic peers. Even in the absence
Exclusivism of epistemic worries concerning one’s own beliefs, many will
Inclusivism wonder how they are to construe the claims of various other
Noumena and Phenomena religions. On any given issue that comes up for discussion,
there appear to be three options: (1) only one religion is cor-
Pluralism rect, (2) more than one religion is correct, or (3) no religion
Problem of Contingency is correct. Thus religious diversity clearly has significant philo-
Soteriological Exclusivism (SE) sophical import, and one that extends over a broad land-
scape—we have already had cause to mention differing beliefs,
Soteriological Inclusivism (SI) practices, symbols, and narratives.
Given that this chapter is written from a theistic perspec-
tive, our interest will be almost exclusively in how believers
address the issues raised by religious diversity, both in relation to the truth claims of other religions,
and the (logically distinct) issue of whether other religions can be understood as routes to salvation.
We also consider the extent to which this diversity can be seen either to support or undermine the-
ism. While on the whole we ignore what we have listed as option (3), we begin by considering a
problem raised by religious diversity that may indeed present a fundamental challenge to theistic
belief altogether.
THE PROBLEM OF CONTINGENCY
One of the more notable factors relating to religious diversity is the major role that geography and
biography apparently play in determining one’s beliefs. The religious options are variously dispersed
throughout the world. Hindus and Buddhists abound in the Asia-Pacific region but barely make a
demographic dent in lands where Judaism and Christianity prevail. Thus the European Christian
cannot help but reflect that, had he been born in India, he would more likely be looking to the
Veda than to the New Testament as the source of religious inspiration. Of course, one need not
go so far, geographically speaking, to generate the problem. Born in a different house around the
corner, one might well have been an observant Jew rather than a believing Christian. This (very
informally) is the so-called problem of contingency, according to which the contingency just illus-
trated is thought to destroy knowledge of one’s own (contingent) religious truth claims. It is enough
to generate what Alvin Plantinga calls a “sense of intellectual vertigo” (2002, 538), and it has the
potential to undermine one’s confidence in one’s own commitments. Indeed, this issue could be
used by the skeptic to undermine confidence in any specific religious commitment below the thresh-
old required for rational belief, making agnosticism the most rational response.
The precise argument operative here is actually more elusive than the impressionistic version
just sketched might suggest. Consider the truth of the following counterfactual statement:
(A) Had Daniel been born and raised in a Hindu family, he would have had different religious
beliefs.
Is this counterfactual enough to undermine Daniel’s conviction in—let us go as far as to say knowl-
edge of—his religious beliefs? Plantinga has convincingly argued that it is not by reasoning along the
following lines:
I believe (truly) that I was born in London.
Now take the following counterfactual:
(B) Had Daniel been born in New York, Daniel would not believe that he had been born in
London.
The truth of counterfactual (B) hardly serves to undermine my knowledge that I was indeed born in
London. Why, therefore, by analogy, should the mere truth of (A) imply that I cannot claim knowl-
edge of my religious beliefs?
The most effective response to Plantinga’s objection appeals to an important distinction
between methods of belief acquisition. Religious beliefs are often instilled uncritically in early child-
hood. Tomas Bogardus (2013) explains that, while it is the case that, for beliefs formed unreflec-
tively, placing oneself in a counterfactual situation would lead one to believe things that one cur-
rently takes to be false, this is not the case for beliefs formed through rational reflection. So it
might be the case that, had I been born and raised a Hindu, I would come to believe things that,
given my current beliefs, I now believe to be false. Here, the counterfactual expressed in (A) that
pertains to a mere change in geography would lead to my holding false beliefs (by the lights of my
current beliefs). So the truth value assigned to religious beliefs that are acquired by habituation
and socialization could well change from true to false (according to one’s current epistemic stan-
dards), given the counterfactual biography contained in (A); and, so the objection goes, if a mere
change of location could switch the truth value assigned to one’s beliefs, something must have gone
wrong in forming these beliefs. In contrast, had I been born in New York, I would, even by my cur-
rent epistemic standards (whereby I know I was born in London), truly believe that I had been born
in New York. Thus, a mere change in geography, while changing the content, does not change the
truth value assigned to my belief concerning my birthplace. Such beliefs do not, therefore, fall foul
of the contingency argument in the manner in which religious beliefs would, since the new (coun-
terfactual) input would not change their truth value. This seems to give us the result that the critic
of theism wants. The response to Plantinga allows religious beliefs to be undermined by contin-
gency in a way that rationally formed beliefs cannot. Hence the phenomenon of contingency as it
relates to religious diversity remains effective against our epistemic confidence in our religious beliefs
but does not take all of our other beliefs with it.
Yet, on further consideration, Bogardus finds that this way of proceeding actually allows the
rational religious believer also to escape the contingency argument. For the believer can argue that,
as long as one’s religious beliefs have been formed rationally, there is no reason why they should be
undermined by contingency any more than one’s belief about where one was born.
Presumably the skeptical counter to this claim would be that religious beliefs are rarely, if ever,
acquired rationally. But, other than begging the question, a modification to the response will take
account of this skeptical countermove. For how the belief is initially formed is surely beside the
point. Take the following thought experiment. A baby boy, born in London a century ago, is taken
by boat to New York. Through a set of tragic circumstances, the baby’s parents die during the cross-
ing, and the baby is left for dead somewhere on the docks in New York. Happily, the baby is found
and adopted by parents who don’t know his origins, and as a result the baby grows up believing he
was born in New York. Later, as a young man, he comes to believe in all manner of irrational super-
stitions. He is told by a palm reader that he was born in London and, given his confidence in such
irrational methods, believes that this is indeed true. Eventually the young man comes to his senses
and rejects such superstitions. Yet, upon subsequently discovering that those whom he regarded as
his parents were not his birth parents but people who found him on the docks, he wishes to address
the nagging doubt regarding his birthplace that is the last vestige of his former superstition.
Researching available birth records, he discovers that he was indeed born in London. Thus he
now holds a belief that was initially acquired through irrational means but was subsequently sub-
jected to critical scrutiny and is maintained rationally. Such a belief will, presumably, be safe from
counterfactual compromise. This example shows that beliefs, however acquired, can, if subsequently
subjected to rational reflection, be safeguarded from counterfactual compromise.
In sum, even if one wished to argue that religious beliefs are never, or at the very least only very
rarely, rationally acquired, and thus can always be thought to be subject to counterfactual compro-
mise, it is surely possible, and is often the case, that intelligent and educated believers subsequently
subject their beliefs to rational scrutiny in a manner that renders them immune to such compro-
mise. If that is conceded (bearing in mind that to assume that such beliefs can never be held ratio-
nally rather begs the question), then it is as yet unclear whether the problem of contingency that

religious diversity allegedly raises can be formulated as a sound argument that presents a genuine
challenge to religious belief. (It should be noted that Bogardus rehearses further variations of the
contingency argument and finds fault with all of them in analogous ways.)
EPISTEMIC ISSUES
Questions of logical validity aside, the sense of intellectual vertigo with which one reacts to religious
diversity appears difficult to shake. But why might this be? One suggestion is that it is grounded in
doubts over whether religious beliefs can be subjected to the requisite type of rational reflection,
either at the acquisition stage or subsequently. The point here is not a fideistic one that would
intentionally place religion beyond the realm of the rational in the first place, and thus beyond ratio-
nal critique. Nor need it be obviously question-begging, as claimed above. For one might generate
the question by arguing that there exists no universal rational method or standard against which
the various religious claims can be measured.
In the absence of such a standard, one might doubt whether any religious believers can have a
rational warrant for their views. For it is one thing to be entitled to one’s views; it is another to have
warrant for them. Presumably, more than one religion could claim that it was entitled to its views
on the grounds of “having broken no epistemic obligations in the forming and maintaining of one’s
beliefs” (Byrne 2011, 30). But it is another matter altogether to claim that one’s views are true (the
question of warrant) in the face of contradictory claims with equal entitlement. In the absence of
universal rational methods and standards, it seems as if warrant remains an issue, and this might
be the nagging sense of having not got the epistemic monkey suggested by the problem of contin-
gency entirely off our backs, at least to the point of being able simply to dismiss the phenomenon
of religious diversity.
What, then, are the possible responses to this epistemic issue? On the one hand, it is of course
possible to wield the breadth and depth of religious diversity against theism altogether much as the
argument from moral disagreement is used in metaethics to undermine moral realism. Written as this
chapter is from a theistic perspective, it is sufficient for our purposes to note that there is clearly no
entailment from diversity to atheism (or to what amounts to the same thing, though with greater
attention to the religious details, namely, metaphysically naturalistic accounts of religion), just as there
is no entailment from moral disagreement to antirealism concerning moral values. No serious propo-
nent of that argument maintains that the fact that two (or more) groups disagree on the moral permis-
sibility of abortion, say, entails that there are no values that could determine the correct answer.
The more powerful versions of the argument in metaethics instead stem from the claim that
moral antirealism is the best explanation of moral disagreement. Yet in the religious sphere it is
not at all clear that atheism is the best explanation for religious disagreement. Indeed, one might
even argue that it is a very poor explanation given that all of the religions we are discussing converge
on the existence of a transcendent reality, despite the very real disagreements that emerge subse-
quently. Moreover, as we will see when we return to this issue in our discussion of pluralism, there
are further areas of convergence among diverse religions for which theism might be argued to be the
best explanation, such that some would claim that religious diversity is best explained by appeal to a
divine ontological reality, with the differences being matters of the manner in which this reality
manifests itself to those from different cultural backgrounds.
As for agnosticism, as an epistemic rather than an ontological view, its rationality or otherwise
will depend on how convincing one finds the epistemic positions discussed in what follows.
THEISTIC RESPONSES TO RELIGIOUS DIVERSITY
Traditionally, philosophers have used the following categories in discussions of religious diversity.
The options in their simplest nontechnical form are:
(1) Exclusivism: the home religion alone is “true.”
(2) Inclusivism: the home religion is superior, though other religions can also have a share of
“truth.”
(3) Pluralism: there are multiple religions that are all on a par with respect to “truth.”
One needs, however, to state matters more precisely in order to see what is at stake here. First,
rather than speak of the “truth” of religion in general, one should distinguish the beliefs or doctrines
of a religion that are truth-apt from the content, such as practices and rituals, that is not truth-apt.
While one can speak loosely of the “truth” of a religion, only its beliefs can be straightforwardly true
or false. Nonetheless, one could presumably claim that exclusivists believe that their religion gets
most things “right,” including beliefs, practices, history, and so on (while presumably all other reli-
gions get most things wrong). The loose claim regarding the “truth” of the religion can therefore be
seen as a convenient way of expressing the view whereby among the true beliefs of certain exclusi-
vists is the belief that their practices are superior in some way to those of other religions, and their
accounts of history are correct. We will continue, then, to use the “loose” manner of expression, but
the reader can render it technically more precise as needed.
This way of proceeding immediately leads, though, to a more important distinction: that
between taking views (1), (2), or (3) with respect to truth and maintaining them in soteriological
form. Taking exclusivism as our example, the soteriological version might maintain the position
extra ecclesiam nulla salus (that is, there is no salvation outside the church). But this view is logically
distinct from exclusivism regarding truth (hereafter ET), for one could take the latter position while
allowing that there are routes to salvation for members of other religions (which sounds prima facie
like a form of soteriological inclusivism).
There are, however, debates surrounding the efficacy of the traditional tripartite taxonomy. If
we begin by asking for a definition of ET, a first attempt might be that the home religion contains
all truth to the exclusion of all other religions. Surely, though, such a view is implausible (which is
not to say that no one maintains it). An exclusivist Christian, say, would surely admit that other
religions have some true beliefs (those that are monotheistic, for example), unless the exclusivist
were willing to take the view that the meaning of a particular doctrinal assertion p was so holistic
that to remove any other belief from the body of doctrine to which it belongs would be to change
the meaning of p to the point that it could not be shared by another religion. This view would seem
to be plausible only—if at all—on an antirealist construal of meaning, which would likely present a
problem for most such exclusivists.
A more plausible form of ET could maintain that, while the home religion contains all “truth”
(or at the very least the vast majority of it), other religions might share some of these truths, but to
the extent that they contain beliefs that are incompatible with the truths of the home religion, these
other religions are false. In discussing various possible definitions of exclusivism, Robert McKim
speaks of the exclusivist as believing that the home religion “outperforms other traditions in terms
of truth” (2012, 31). If that is the case, though, it is unclear how we would distinguish ET from
inclusivism, according to which it is also the case that one religion is superior to all others with
respect to truth but does not have exclusive claims to truth.
It appears, then, that in the realm of truth, at least conceptually speaking we end up with a sim-
pler twofold distinction between:
(1) those who believe that the home religion outperforms all others with regard to truth
(whether by containing all of the core truths or more of them than any other), though some
of its truths might also be found in other religions—a view we can call monism; and
(2) those who believe that there are plural (even incompatible) systems of truth that are all on a
par with each other—a view we can call pluralism.
Certainly it might be possible to further subdivide monism. So, for example, one might speak of:
(1a) those who think that the home religion has all truth (though others can have a lesser share
of it); and

(1b) those who think that the home religion outperforms all others with respect to truth but
accepts that others may contain a number of truths that the home religion lacks and that it
can learn from them.
There might be room for dispute here as to whether the home religion can merely be incomplete,
and thus supplemented by the truths of these other religions, or can actually be wrong about certain
matters such that other religions can act as a corrective. The latter view, however, could complicate
the claims to superiority by the home religion if one were to assign different weightings to different
truths.
Now, there are those who are happy to brand (1b) inclusivism (with a sliding scale regarding
how much the home religion can learn from others), and (1a) exclusivism, and one can certainly
distinguish between those who do and do not believe that one can learn truths from other reli-
gions. Conceptually speaking, however, they agree that there is a single religious truth, and merely
disagree over who has it (or at least most of it). So, it remains unclear (at least to this writer)
whether the exclusivist/inclusivist distinction is conceptually (rather than just factually) signifi-
cant, and thus whether or not setting them up as two distinct metacategories alongside pluralism
is the best way to proceed, rather than having the two categories of monism and pluralism, and
subdividing them.
As for the soteriological options, it seems that the most important distinction is between those
who answer “yes” and those who answer “no” to the question: “Can members of other religions be
saved?” Here, however, there is greater warrant for maintaining the traditional tripartite division
between those who believe that:
(3) only members of the home religion can be saved (soteriological exclusivism);
(4) both members of the home religion and members of other religions can be saved, despite the
latter’s not believing in the true religion (soteriological inclusivism); or
(5) members of all (legitimate) religions can be saved, and there is parity between their
irreducibly different routes to salvation.
In this case, the distinction between inclusivists and exclusivists thus appears clear. Moreover, while
inclusivists allow for the salvation of others despite “false” elements to those religions, for pluralists
different salvific religions can each be on a par with each other (though, as we will see, even a plu-
ralist can allow that some religions are “better” than others). For the pluralist, many different reli-
gions afford their adherents equal access to salvation, often via irreducibly independent routes. For
inclusivists, there is one and only one religion that guarantees a direct path to salvation. Members
of other religions may also be saved, but only inasmuch as they somehow “participate” in the con-
ditions of salvation that are part of the inclusivist’s own religion. As such, members of other reli-
gions might well be at a significant disadvantage.
MONISM
What is one to make of monism? There is a sense in which it is trivially true with respect to reli-
gious belief. Jews do not believe that Jesus was God incarnate; Christians do. These beliefs are
clearly mutually exclusive, and so anyone holding that one is true will indeed believe that the other
is false. What, then, are the alleged problems with monism?
Essentially, through medieval and early modern times, the prevailing view appeared to be that
there were neutral rational methods for assessing religious truth. Thus, with appropriate education,
reflection, and patience, all rational thinkers ought to arrive at the exclusive monistic truth. Moses
Mendelssohn, the eighteenth-century philosopher dubbed the “Jewish Socrates,” was challenged
by Johann Kaspar Lavater either to disprove the arguments for Christianity (as presented in Charles
Bonnet’s La Palingénésie philosophique of 1769) or to convert on account of them. This, however, is
where the existence of religious diversity would at least give one pause for thought. For on the rea-
sonable assumption—given peer conflict—of the absence of neutral rational standards for critical
assessment on these matters, there are those who would ask whether it is rational to continue to
maintain the monistic view.
MONISM AND THE EPISTEMIC CHALLENGE
Must religious diversity inevitably erode one’s epistemic confidence to the point of undermining
monism? Those who deem themselves exclusivists argue not. For Plantinga, religious beliefs can
be thought of as properly basic. They are no more in need of inferential proof than, as G. E. Moore
saw it, was our belief that we have hands. Moreover, Plantinga maintains that the Christian believer
can consistently maintain that he is at a special epistemic advantage as a result of his sense of the
divine and of the work of the Holy Spirit in keeping the Christian free from error (though believers
need not be aware of these mechanisms in order to be in this epistemically privileged position).
From this starting point, the mere existence of competing alternatives need not cast any doubt on
the veracity of one’s own beliefs. Given that, ex hypothesi, there are no independent neutral reasons
for any religious beliefs, why should the existence of alternatives call my beliefs into question when
said alternatives are based on different grounds that by my lights—the only ones that are compelling
to me—are not rationally compelling? Note that even in the absence of proper basicality—imagine
that one does have inferential reasons for one’s beliefs, albeit reasons that are internal to one’s tradi-
tion—the default position would still remain that one’s own beliefs are on a higher epistemic foot-
ing than those of others, and there is thus no reason to lose confidence in one’s views or to engage
in any epistemic soul-searching. While Plantinga concedes that as a matter of empirical fact some
may lose epistemic confidence, he maintains that diversity does not provide us with reason to do
so in the absence of actual warrant for believing the alternatives. For the alternatives rationally to
warrant such attention, they would need to show that they are on an equal footing. Presumably they
would be unable to do this given the absence of neutral rational standards. Thus monism—or for
Plantinga, using the traditional categories, exclusivism—remains an epistemically acceptable position
that need not be undermined by diversity.
Philip Quinn (1995), in contrast, contends that Plantinga is not taking religious diversity seri-
ously enough. The key issue here is where precisely the burden of proof lies. Is the mere presence of
epistemic peers holding contrary views enough to dislodge confidence in one’s own? For Quinn, the
equal levels of (internal) epistemic entitlement enjoyed by the alternatives place the burden of proof
on Plantinga to show why he should not take them seriously ab initio rather than awaiting their epi-
stemic bona fides. For Quinn, reflection on the undeniable empirical fact of diversity must at least
weaken an individual’s justification for claiming any epistemic superiority complex, and this condi-
tion should at the very least bring a measure of epistemic humility to the monist. Thus, while
Quinn believes that it can be rational to “sit tight” (to use William Alston’s [1988] phrase) with
one’s current religious beliefs, it can also be rational to seek what could be seen as a more inclusivist
or even pluralist approach of “thinning” one’s doxastic practices. So, without compromising on the
fruits of one’s own religion, it would be rational to “thin” its core beliefs so as to reconcile divergent
perspectives that appear to be on an epistemic par. This humility will not necessarily rid us of gen-
uine epistemic conflicts, but it can lead to an approach whereby the bases of conflict are reduced.
For example, where traditionally religions based on the Bible understood it to require belief in a
six-day creation, there are those who would now read the Bible in accordance with contemporary
scientific cosmogony (and still others who do not believe that it is trying to teach “science” at all).
Maintaining a core belief in the distinction between God and creation without requiring belief in
a specific cosmogony would be to move to a thinner theology that could remove an area of conflict
without the need for compromise on any religious practices, even if the reduction in the doctrinal
content of religion might be of concern to some.
Where does this leave the monist? There does seem to be something questionable about asking
other religions to provide a justification that is deemed sufficient by our epistemic standards before
we take them seriously, especially when one has more or less ruled out that possibility ex hypothesi.
Given the at-best internal justifications—or straightforward epistemic confidence in the absence of
justification—to which each religion appeals, even in the absence of specific reasons that would

undermine one’s doctrines from an internal perspective, the monist seems to be insulating his doc-
trines from increased rational reflection in a way that ignores the general epistemic landscape. As
such, whether straightforward monism really takes religious diversity seriously enough can at least
be questioned.
MONISM AND THE MORAL CHALLENGE
Epistemic issues aside, there are various moral challenges laid at the door of monism. Again, Plan-
tinga notes the accusation of being “intellectually arrogant, or egotistical, or imperialistic, or oppres-
sive” (Plantinga 2002, 534). It is, however, difficult to pinpoint the precise argument here. A
monist who believes that Φ, and possibly even that this belief has salvific consequences, will obvi-
ously reject the belief that not-Φ. It seems strange, if not entirely arbitrary, to fault someone morally
for this type of “exclusivism.” If one were to be held morally accountable for holding such views, the
entire discipline of academic philosophy would be a moral void (in principle, rather than in practice,
as might well seem to be the case at times). Indeed, it seems very difficult to draw any relationship
of entailment between monism and morally questionable behavior or qualities. Belief in monism is
one thing; how one acts on that belief is an entirely separate matter. But that is the point—it is a
separate matter. Indeed, it may well be that one’s religious beliefs, albeit held exclusivistically, lead
one to act with love toward everyone. My thinking that I am right and others wrong need not lead
to my tyrannizing the other, especially if one believes in objective moral requirements such as to be
tolerant or to “love one’s neighbor,” indeed to “love the sinner,” that stem from the very religious
system that is being questioned.
While judging others to be sinners might be thought to be the moral fault here (at least in all
but obvious cases), once again it is entirely possible that either (a) monism need not lead to this
belief if one is a soteriological inclusivist; or (b) the monist also believes that ultimately such judg-
ments, or at the very least acting on such judgments, are to be left in the hands of God. In addition,
one might note the possibility of questioning whether the religious judgment necessarily implies a
moral judgment, but further discussion of this would take us too far afield.
At this point then, the worst that can be said is that monism can lead to being “judgmental,”
though whether this constitutes a moral fault can certainly be questioned. If it does, it is unclear
whether the pluralist can escape the very same accusation for the “fault” of “judging” the monism
of the monist (though the pluralist may contend that he makes no such judgment, and is merely
giving us an explanatory framework—indeed in his view “the best explanation”—of the phenome-
non of religious diversity).
A stronger moral challenge can be leveled against the morality of a God who would only
vouchsafe the truth to one of the myriad religions that populate his creation. As Mendelssohn
noted, if humanity requires divine revelation for human felicity, “why has the greater part of man-
kind lived without true revelation from time immemorial?” This would clearly “detract from the
omnipotence or the goodness of God” (Mendelssohn 1983 [1783], 94). This highlights a moral
aspect of the problem of contingency, whereby it is implausible that a good God would allow
an accident of birth to give you access to truth, not to mention salvation, while condemning
others to false consciousness at best, eternal torment at worst. Of course, the believer may argue
that there is no such contingency, for everything is the result of divine providence. At this point,
however, the operative assumption is that one somehow “deserved” to be in possession of the
truth while others did not; this would indeed smack of moral smugness, particularly when such
“desert” can be empirically questioned. Moreover, it is difficult to deny the existence of great
moral heroes or heroines in religions other than one’s own, at least as worthy of such providential
protection as oneself, not to mention morally unsavory coreligionists who surely did not merit
their good fortune.
Clearly the truth matters. One would rightly be disturbed to discover that one had based one’s
life on a set of falsehoods, and in the religious sphere, those falsehoods could potentially have signif-
icant effects on how one lives one’s daily life. The problem, however, becomes all the more powerful
if belief in these truths is bound up with salvation. The idea of soteriological exclusivism, whereby
only members of the home religion can be saved, seems to exacerbate the moral problems here.
So, at this point, we move on to consider soteriological exclusivism (SE) together with soteriological
inclusivism (SI).
EXCLUSIVISM, INCLUSIVISM, AND SALVATION
Exclusivism with respect to salvation (in this section, when we use “exclusivist” and “inclusivist”
without qualification, they are intended in the salvific sense) is presumably closely bound up with
monism. The simplest reason for the fact that only members of the home religion (hereafter HR)
can be saved would be that its beliefs—and presumably those that are unique to it, thus excluding
members of other religions—are necessary conditions for salvation. Since we have noted that it is
implausible to think that no other religion shares at least some truths with HR, the exclusivist
here must believe that, while these shared truths are necessary conditions for salvation, only in
combination with its own unique truths that others do not share will we find the sufficient condi-
tion for salvation.
In contrast, an inclusivist could believe that certain of HR’s core beliefs (or practices) that other
religions might share could be both necessary and sufficient for salvation, allowing for members of
other religions to be saved (though as we will see, this is not the only way to be an inclusivist).
So, for example, a Christian exclusivist may believe that monotheism—obviously in combination
with all manner of other factors—in the absence of the trinitarian conception of God, is insufficient
for salvation; an inclusivist could maintain that the monotheistic element (again in combination
with the other factors) would be sufficient even in the absence of—or in a stronger version even
with the denial of—the trinitarian view.
Obviously, the moral challenge to exclusivism here becomes particularly pertinent. Could it
really be that a just God would reveal salvific truths only to one tiny segment of his creation? Or,
to bring the problem of contingency back into the picture, would such a God make something as
important as salvation contingent on an accident of birth? Here inclusivism, with others indeed able
to gain salvation, seems to have the upper hand, though for most inclusivists HR must be superior
in its salvific efficacy, and so the moral challenge is mitigated rather than entirely repelled.
The extent to which the challenge is mitigated depends on which version of inclusivism one
accepts, for inclusivists can differ quite radically regarding the limits and qualifications for “inclusion.”
To take a Jewish example, Moses Maimonides codifies the view that “righteous gentiles have a share in
the world to come” (in his legal code Mishneh Torah, at “Laws of Kings,” 8:11). Here, while obser-
vance of the Jewish commandments is the best route to salvation, salvation is nonetheless open to
non-Jews through adherence to the seven “Noahide laws” that are effectively a very small subset of
the commandments that Jews themselves observe. Note first that it is only because of an overlap with
HR that members of other religions are saved. This type of inclusivism places the necessary means for
salvation in HR, but it allows members of another religion to fulfill them. Second, whether merely ful-
filling these conditions for salvation is sufficient, or only doing so as a result of recognition that they
are divine commands, is a subject of debate in rabbinic literature. Thus we can distinguish inclusivists
for whom the conditions of salvation are ontological—in this case, behavioral—from those for whom
there are, whether alone or in addition, epistemological requirements.
A particularly interesting (and famous) version of inclusivism that allows for the fulfillment of
ontological conditions for salvation in the absence of knowledge of these conditions is that of Karl
Rahner, whose category of “anonymous Christians” who live with integrity by their own lights
can even include “atheists—who are justified by God’s grace and possess the Holy Spirit” (1974,
291). Such people can thus be saved by the atoning death of Christ even in the absence of belief
in that religious doctrine (though some quibble that this is closer to exclusivism than inclusivism,
since those who are saved are actually members of HR, albeit unwittingly).
Whatever brand of inclusivism we may accept (and one could delineate various versions), all
inclusivists maintain that they are in a superior position relative to other religions, presumably since

they either describe the true conditions of salvation (Rahner), contain the best route to salvation
(Maimonides), or can achieve a higher level of salvation (if such things can be thus quantified).
Some such superiority complex is essential if inclusivism is not to collapse into pluralism. It is also
the primary way an inclusivist could be rationally reconciled to a religion that deems it a duty to pros-
elytize in the service of a universal mission—a duty that Rahner’s view was said by some to “demolish”
(Conway 1993, 25), and this despite Rahner’s maintaining the superiority of Christianity.
What makes one route to salvation better than another will again be a subject of debate, but if
we accept this superiority, it is unclear whether inclusivism can entirely dissolve the moral challenge.
In its most extreme form, Rahner’s view helps by allowing for a universal condition of salvation that
can be achieved even absent an awareness of the condition. But, at the risk of sounding overly prag-
matic, to reconcile this position with his view regarding the superiority of Christianity is no simple
task. Certainly it is thought to be the one religion to have correctly described the mechanism for sal-
vation, and presumably very few people would prefer to believe falsehoods over truth. But the
absence of any need to acknowledge this truth means that the benefits that accrue to being a mem-
ber of the correct religion might be at the level of easing the path to salvation at best, but not at the
level of being necessary conditions for salvation.
There is, of course, an important assumption running through all of this—that God is just and
would not therefore unfairly condemn the vast majority of his creatures to a suboptimal existence.
While belief in a just God is hardly a stretch for the theist, believers do sometimes invoke a moral
exclusion clause to solve difficult theological problems, the main examples being various theodicies
that sometimes appeal to our inability to judge God by our moral standards. It seems, however, that
this route should only be an appeal of last resort to prevent it from being abused, whether epistemi-
cally as an all-purpose “get-out” clause for difficult questions, or morally as a justification for all
manner of moral atrocities. Ultimately then, on the assumption of a morally good God, there is a
moral challenge to exclusivism, and to a lesser degree inclusivism, that remains in play.
PLURALISM
Pluralism regarding truth can be clearly distinguished from monism in that it maintains that all
legitimate religions (the conditions of legitimacy are discussed below) contain equally true, though
incompatible, beliefs and doctrines. John Hick (2004), whose version of pluralism is often the main
focus of scholarly discussion, explains this seeming paradox of plural but incompatible truths by
analogy to the Kantian distinction between noumena and phenomena. Hick argues that God is
the ultimate reality, or the real in itself (an sich). This ultimate reality can, however, only be partially
represented by the human mind through various religious schemes that are all merely phenomenal
representations of this noumenal reality. Thus Hick superimposes a culturally plural framework
on the basic Kantian idea that any and all human experience is conditioned by the concepts that
shape it, so as to argue that incompatible religious views simply reflect our differing, culturally con-
ditioned attempts to understand the ineffable divine reality. Each view emerges out of our experi-
ence of the divine, but the divine reality itself is beyond the reach of human conceptual categories
(and it is worth noting that nonpluralists all the way back at least to medieval times—and thus also
non-Kantians—could agree to the idea of a God who is absolutely transcendent and indescribable).
Note that, for Hick, the limits on our religious knowledge are ontological, and are not merely the
result of an epistemological inadequacy. Even if our faculties were working perfectly, and we had
reached the pinnacle of human rationality, we could not advance beyond these partial representa-
tions of the divine.
Hick argues that pluralism gives us the best religious (as opposed to naturalistic) explanatory
account of religious diversity. In allowing that there is one ineffable ultimate reality accessed in vary-
ing culturally conditioned ways, Hick can explain why it is that we have so many religions that all
nonetheless coalesce around a shared belief in an ultimate, transcendent, and meaning-giving reality.
It also helps explain the apparent practical efficacy of so many of these religions. For Hick, religions
(in the plural) successfully achieve a “human transformation from natural self-centeredness to a new
orientation centered in the Transcendent … expressed in an inner peace and joy and in compassion-
ate love for others” (Hick 1997, 162). This transformation (ultimately, a moral transformation) to
“reality-centeredness” is, for Hick, a this-worldly beginning of salvation. The pluralist acceptance
of the truth of all of these religions explains this shared moral-salvific efficacy—for Hick, pluralism
with regard to truth and with regard to salvation go hand in hand.
In general, pluralists seem to elevate the religious significance of morality in a number of ways.
Thus, for Hick, even pluralists can distinguish between superior and inferior religions (though in
their details rather than as totalities) on the basis of ethical criteria that also altogether rule out cer-
tain so-called “religions.” We can, for example, rule out Satanism or cults such as the Waco, Texas,
Branch Davidians by their moral and spiritual fruits—“if the fruits … were hatred, misery, aggres-
sion, unkindness, impatience, violence and lack of self-control this would lead us to deny the
authenticity of the experience” (Hick 1997, 164).
It might be suggested that this fact of moral convergence can also be used as the basis for an
argument that religious diversity supports theism. The argument would begin with the following
question: Given the diverse geographical and biographical backgrounds of religious people and the
very different religions that emerge from them, how are we to account for their moral convergence?
An appeal to theism—in particular to a good God—might be thought to be the explanation for this
convergence. What we have here is a shared starting point—a transcendent reality—and a shared
endpoint—moral convergence. Yet the stuff in the middle is radically different, even at times
opposed. And the argument would be that the shared theistic starting point is the best explanation
of this “convergent diversity,” that is, theism is the best explanation of how such diverse religions
can all yield Hick’s moral fruits.
This is, of course, only the bare outline of an argument, which would need to be worked out in
far greater detail to ultimately succeed. Moreover, atheists will counter that the shared theistic belief
need not be veridical to cause this moral convergence, though theists might counter that without
some ontological grounding, the likelihood that wildly varying theistic beliefs would yield such con-
vergence is vanishingly small. The atheist, in turn, will no doubt appeal to genetics and natural
selection for this ontological grounding. Ultimately, then, the question seems to devolve to the hotly
debated territory of the extent to which morality is bound up with theism or can be accounted for
naturalistically, though it is worth noting that the former view is shared in some sense by both radi-
cal moral thinkers such as Emmanuel Levinas and radical critics of traditional morality such as Frie-
drich Nietzsche, who was also, of course, an atheist.
Placing morality at the center of the pluralist picture acts, however, as a double-edged sword for
the argument just sketched, where God is argued to be the best explanation of the moral conver-
gence among diverse religions. For this can only be the case if we assume a moral God. On what,
though, does the pluralist base this characterization of God? Given that we have no direct access
to the “noumenal” God, it seems that we can only base our moral characterization of God on the
“fruits” of theism, that is, on the moral convergence itself. Thus pluralists would characterize God
as moral on the basis of moral convergence, while explaining moral convergence by appeal to a
moral God. It is not clear, therefore, whether the pluralist’s theistic picture, consistent and self-sup-
porting though it may be, can act as a noncircular foundational explanation of the moral conver-
gence of diverse religions that would convince a neutral observer.
Within the theistic realm, however, pluralists can claim to have the upper hand over their rivals
in relation to the moral challenge, since if all ethically acceptable religions are considered “true” in
some sense, it is no longer the case that God is arbitrary in withholding the doctrinal truth from
the majority of believers, as appears to be the case for rival theories. Indeed, rather than following
Hick’s more metaphysical and epistemological route, Kenneth Einar Himma (2002) develops an
argument for at least soteriological pluralism on purely moral grounds by arguing that a good
God would not punish those who are not morally culpable for their religious commitments. The
role of contingency in the formation of our religious beliefs through the influence of parents and
societies, coupled with the absence of direct volitional control over belief, seems to remove moral
culpability for our religious beliefs. Thus surely a good God would not make adherence to one reli-
gion necessary for salvation (as long as the religions in question are basically moral). On this basis,

combining soteriological pluralism with monism could be perfectly consistent, if rather unusual. A
purely moral argument for the former could allow for belief in the latter. The monist could under-
stand that it is morally problematic to give their doctrinal truth a role in determining who gains sal-
vation. In such a case, any advantages that the monist apportions to his own religion would have to
be limited solely to the realm of theory, at which point the difference between the monist (for
whom other religions are false, though their members can have equal access to salvation nonetheless)
and the pluralist (for whom other religions are equally true, yielding equal access to salvation) seems
to become quite literally academic.
CRITIQUES OF PLURALISM
Hick’s pluralism has been subject to detailed critique. To begin with, if religious beliefs are really
just “phenomenal” representations of the “noumenal” God, then the “personal” God of the main-
stream Western monotheistic religions is only an image of the ultimate reality, which is thus not
actually personal. Hick affirms that this personal image is no truer of the ultimate reality than the
impersonal God of Hindus and Buddhists, and he goes as far as to say that “it cannot be said to
be one or many, person or thing, substance or process.… We cannot even speak of this as a thing
or entity” (Hick 2004, 246). This position, however, would clearly distress many theists, who
believe that God is, in some significant sense, personal, or at the very least is more accurately
described thus than as impersonal.
Hick here is caught between a rock and a hard place, for we either each correctly represent the
ultimate reality and end up with a contradiction—“the ultimate reality is personal and the ultimate
reality is not personal”—or our representations do not correctly describe it, putting us at too far a
remove from God—and truth—to satisfy theists. Indeed, if one is worshipping a personal God,
and it turns out that God is not a personal being, has one really worshipped God at all? Has one’s
worship not been directed toward an imaginary personal being rather than the ultimate reality?
Others might question why we should speak of plural conceptions of one God rather than of
polytheism. Hick writes that polytheism, “although theoretically possible, creates more problems
than it solves.” He asks whether in the neighborhoods of multicultural Birmingham, England,
“the Holy Trinity answer prayers from Edgbaston, but Allah in Small Heath, … and Vishnu in
other parts of Handsworth?” (Hick, 1997, 165). The pluralist maintains that all of these descrip-
tions are aiming at the same referent, and given that the limits on our descriptions are not just epi-
stemic but ontological, we cannot but miss the target in terms of truth. What matters, however, is
that they are all attempts to hit the same target, and once again in pluralism’s favor, it might claim
that this better explains the shared core religious belief and the common transformational effects.
Himma’s moral route to pluralism might avoid some of the aforementioned difficulties caused
by Hick’s Kantian apparatus. But Himma’s view, together with the general pluralist shift to the pri-
macy of morality, leads to the accusation that pluralism at best downplays the significance of the
doctrinal content of religions, and at worst denudes religious doctrine of all meaning. Either Christ
is or is not the son of God. Yet, for the pluralist, either “Christ is the son of God” and “Christ is not
the son of God” are both literally false (Hick); or neither of “Christ is the son of God” and “Christ
is not the son of God” can be objectively known (Himma). Regardless, the belief in soteriological
pluralism seems to render the doctrinal content of religion irrelevant. Furthermore, it does seem dif-
ficult to understand how a pluralist within any given religion can maintain that p is the case and
simultaneously accept the truth of not-p. Yet, as a pluralist, it seems that he must accept the truth
of both.
Runzo attempts to meet some of these issues by suggesting a different metapicture whereby in
the religious realm “first-order truth-claims about reality are relative to the world view of a particular
society” (1993, 545). Runzo understands each system as describing real and irreconcilable facets of
the actual world, much as the elephant is described by the blind men who each feel one real part of
it but have no conception of the whole. In contrast to Hick’s epistemological pluralism, whereby a
single reality is variously misrepresented by different religions, Runzo present a metaphysical
pluralism whereby there are many irreducibly different truths—and really distinct and irreducibly
different paths to salvation—that reflect distinct, irreducible, and incompatible religious “realities,”
each of which is a true, if partial, representation of a rich and multifaceted reality.
For Runzo, however, preventing this from devolving into polytheism becomes particularly
difficult. While Hick’s noumenal divine reality cannot accommodate the truth of either “Jesus was
the son of God” or “Jesus was not the son of God,” Runzo appears to want both to be true, which
cannot be the case if both propositions are supposed to be referring to the same God, and it seems
difficult to solve this problem by reference to “partial” views of the same thing. While we can
understand such an idea in principle by analogy to encountering someone as, say, a judge in a court-
room, but as a loving mother to her children at home, this does not seem to help in the religious
realm. We have little trouble reconciling these differing perspectives in one person, even though
we do not encounter them simultaneously. How, on the other hand, we deal with both the affirma-
tion and denial of Jesus as the son of God or the existence of the human soul is another matter
altogether, whether for Hick or for Runzo.
This is where Hick’s emphasis on the transformation to reality-centeredness becomes increas-
ingly significant, for if religions are simply a means to that end, it may indeed not matter which
doctrines one believes as long as one’s religion as a whole yields this type of moral self-transforma-
tion. Pluralists with respect to truth can accept that there are genuine disagreements between
religions, but soteriological pluralists must deem them nonessential for salvation. They argue for
“tolerant agnosticism” in the doctrinal sphere, for to deny that one religion can indeed be as good
as another with regard to truth will devolve back into a form of monism. Interestingly though, as
mentioned earlier, this emphasis on the centrality of the moral fruits of religion and the noumenal
reality that unifies them might be thought to yield at least one positive cognitive claim about Hick’s
divine being—that it is somehow moral in character.
Where the pluralist differs from the monist, then, is in downplaying the truth of religious doc-
trines. While all are agreed that there is a single ontological being that is the focus of all religion,
monists believe that there is a single truth that their religion has and that can be known, while plur-
alists argue for plural “truths” across various religions, none of which describes the ontological reality
any better than the other, though they all can be understood as “veridical” inasmuch as they effect
the appropriate moral transformation (and may incidentally, for Hick, be subject to eschatological
verification). The most that pluralists can do doctrinally is argue that one can posit of God, or
the ultimate reality, that it has a basically moral character that results in a corresponding require-
ment for human beings to morally transform themselves.
A DUAL PERSPECTIVE?
The problem for the pluralist is that humans effect their moral transformations out of particular reli-
gious systems, within which, from a first-personal perspective, one will generally believe that certain
claims are true, and others false, even if one’s grounds for saying so are indeed “prejudiced” by the
particularistic rationality of the religion in question. At the same time though, pluralists wish, from
an abstract third-personal perspective, to say that their truths cannot be argued to entail the falsity
of the incompatible claims of others. The question for pluralists is whether one can rationally main-
tain this epistemic split-personality.
One might argue that what matters is that from a first-personal perspective we advance our
claims with authenticity, in the genuine belief that they are right—and indeed others wrong—as
does the monist. Taking a leaf from many an exclusivist’s playbook, one could argue that in the
absence of access to neutral rational standards in the religious sphere, we each find ourselves con-
vinced by the religion to which we adhere—Alston speaks here of “self-support”—such that others
are neither live options, nor can they be shown to be superior. Moreover, as Jerome Gellman notes
in arguing for exclusivism, if a religion allows a believer to experience the presence of God, then “if
trying to adjudicate the differences between religions might hurt one’s relationship to God, … our
believer has a good epistemic reason not to engage in such an enterprise” (2000, 405–406).

Nothing could be less rational than leaving behind a religion in which one has epistemic confidence
and entering what from a first-personal perspective is an epistemic void.
Yet from a third-personal perspective, the pluralist, unlike the monist, understands that reality
is set up in such a way that others have equal warrant for their claims, to which, at least from
this neutral perspective, truth cannot be denied, even though it makes no sense to one from a
first-personal perspective. Pluralists can therefore from a first-personal perspective indeed be monis-
tic about their beliefs and reject those of others with all due authenticity, while understanding that
this is something that, given the way the world is epistemically structured, cannot necessarily be jus-
tified from a neutral third-personal perspective. But that should not surprise us, given that most of
the thinkers we have discussed accept that there is no such neutral objective standard by which to
judge religions. Moreover, given this epistemic reality, and a just God who “knows” that matters
are this way, salvation surely cannot be dependent on “the” true set of beliefs (even in the
unlikely—or even impossible—event that it turns out that one religion has been lucky enough to
hit on it). The unique third-personal truth of specific doctrines is, therefore, indeed less important
to the pluralist than their moral and spiritual fruits—and thoroughgoing pluralists may even allow
that this morality can become manifest in varying ways in accordance with different models of prac-
tical rationality. Note, however, the stress on authenticity here from a first-personal perspective;
there seems to be something disingenuous about a religious gourmand, and even something ratio-
nally assailable, if one were to simultaneously adopt inconsistent worldviews.
Here, we should note a distinction between religion, on one hand, and more empirical fields of
study such as science, on the other, where the first-personal and third-personal perspectives will gen-
erally coincide. In the scientific realm, on the whole, pluralism can be denied from a third-personal
perspective, and this seems to be down to the shared epistemic standards in what is after all, in con-
tradistinction to religion, a field of study that deals with decidedly empirical matters. (More inter-
pretive empirical realms of study such as history may split the difference here, being at once more
pluralistic than science yet less so than religion.) Moreover, the correct theory will yield the desired
results and the incorrect theory will not. If that is not the case, then as in the religious case, presum-
ably one could indeed take a pluralistic stance. But there are without question innumerably fewer
scientific than religious options that will yield the same “fruits.” While there is much that could
be written on this point, we simply note that there is a sense in which religious believers, given that
most would willingly assert that their object of study is in the realm of the transcendent, ought to
embrace these differences.
While the knee-jerk reaction would be to criticize this type of split thinking even in the reli-
gious sphere, it is unclear why there is anything particularly problematic about the picture we have
just described, other than the difficulty that certain religious thinkers may have with the idea that
they cannot unequivocally assert that other religions have it all wrong. Soteriological pluralism just
seems morally required by the epistemic structure of God’s world and appears to go hand in hand
with an inescapable first-personal epistemic monism. Yet pluralists find that they cannot deny plu-
ralism from a third-personal perspective.
In sum, it seems that exclusivism and inclusivism in their salvific forms are up against moral
challenges that pluralism can avoid and are less able to give the moral and spiritual fruits of other
religions their due in the manner of the pluralist. But taking the pluralist soteriological position
comes at the cost of not taking the cognitive content of religion as seriously as many believers
may wish, which takes us from soteriological pluralism to pluralism regarding truth. Concerning
the latter, while the monist is able to give literal significance to doctrinal content, the pluralist seems
unable to do so, instead sliding toward a more reductive moral account of religion so as to avoid
having to make sense of a God who, for example, both can and cannot become incarnate. Ulti-
mately, the pluralist will need to see incompatible religious doctrines as equally true and viable
routes to the divine, even if he can only authentically accept the truths of one of those routes. When
it comes to religious diversity, then, it seems that we find ourselves unable to reconcile a version of
the classic Kantian divide between the theoretical and practical. How one prioritizes the two seems
to be the deciding factor in the choice of responses to religious diversity.
Bibliography
Alston, William. “Religious Diversity and the Perceptual Knowledge of Mendelssohn, Moses. Jerusalem; or, On Religious Power and Judaism. (1783.)
God.” Faith and Philosophy 5 (1988): 433–448. Translated by Allan Arkush. Hanover, NH: University Press of New
England, 1983.
Bogardus, Tomas. “The Problem of Contingency for Religious Belief.” Faith
and Philosophy 30, no. 4 (2013): 371–392. Moore, G. E. “Proof of An External World.” Proceedings of the British
Academy 25 (1939): 273–300.
Byrne, Peter. “A Philosophical Approach to Questions about Religious
Diversity.” In The Oxford Handbook of Religious Diversity, edited by Nietzsche, Friedrich. On the Genealogy of Morality. Translated by
Chad Meister. Oxford: Oxford University Press, 2011. Maudemarie Clark and Alan J. Swensen. Indianapolis, IN: Hackett,
Conway, Eamonn. The Anonymous Christian—A Relativised Christianity?: An
Evaluation of Hans Urs von Balthasar’s Criticisms of Karl Rahner’s Theory of the Plantinga, Alvin. “A Defense of Religious Exclusivism.” In Philosophy and
Anonymous Christian. Frankfurt am Main, Germany: P. Lang, 1993. Faith: A Philosophy of Religion Reader, edited by David Shatz. Boston:
McGraw-Hill, 2002.
Gellman, Jerome. “In Defense of a Contented Religious Exclusivism.”
Religious Studies 36 (2000): 401–417. Quinn, Philip L. “Towards Thinner Theologies: Hick and Alston on
Religious Diversity.” International Journal for Philosophy of Religion 38,
Hick, John. “The Possibility of Religious Pluralism: A Reply to Gavin no. 1/3 (1995): 145–164.
d’Costa.” Religious Studies 33, no. 2 (1997): 161–166.
Rahner, Karl. “Observations on the Problem of the ‘Anonymous Christian.’”
Hick, John. An Interpretation of Religion, 2nd ed. New Haven, CT: Yale In Theological Investigations. Translated by David Bourke. New York:
University Press, 2004. Seabury Press, 1974.
Himma, Kenneth Einar. “Finding a High Road: The Moral Case for Salvific Runzo, Joseph. World Views and Perceiving God. New York: St. Martin’s
Pluralism.” International Journal for Philosophy of Religion 52, no. 1 Press, 1993.
(2002): 1–33.
Sacks, Jonathan. One People?: Tradition, Modernity, and Jewish Unity.
Levinas, Emmanuel. Totality and Infinity: An Essay on Exteriority. Translated London: Littman Library of Jewish Civilization, 1993.
by Alphonso Lingis. Pittsburgh: Duquesne University Press, 1969.
Twiss, Sumner B. “The Philosophy of Religious Pluralism: A Critical
McKim, Robert. On Religious Diversity. Oxford: Oxford University Press, Appraisal of Hick and His Critics.” Journal of Religion 70, no. 4 (1990):
2012. 533–568.

TOPIC 8
Religious Diversity: Atheism

Tiddy Smith
University of Otago, Dunedin, New Zealand
This chapter explores the nature of religious diversity, the various theological responses to it, and how the
existence of such diversity serves to undermine theism.
THE PROBLEM
The great diversity of religious opinion demonstrates that human beings are not very good at discov-
ering religious truths. It seems that we are persistently led into error. There is no convergence
around a set of religious beliefs that are probably true, and there is no convergence around a single
set of religious belief-forming practices that are probably reliable. The overwhelming majority of
religious believers must be in a state of ignorance. At most, a tiny minority has got things right.
Religious diversity, then, poses the following problem for the defender of religious belief: On what
grounds can it be reasonable to think that one belongs to the tiny minority of religious experts and
not to the overwhelming majority of religious ignoramuses? One’s own sense of confidence in the
truth of one’s own religious beliefs does not appear to count for much, because this confidence
seems to be equally well shared on all sides and by believers of all stripes.
The problem of religious diversity is hardly a new one. A famous fragment from the pre-
Socratic Greek philosopher Xenophanes marvels at the fact that while the Thracians say their gods
have blue eyes and red hair, the Ethiopians say their gods are snub-nosed and black (Kirk and Raven
1962). Of course, the point of Xenophanes’s wry observation is not that Thracian gods must truly
be red-headed whereas Ethiopian gods must truly be black but rather that religious beliefs typically
have dubious genealogies. Indeed, he adds that if horses had hands, they would draw their own gods
in the image of horses. Our conceptions of the gods appear to say more about our own, very human
natures than about anything supernatural. The diversity of gods, then, may turn out to be nothing
more than a reflection of the diversity of humankind.
Another ancient writer, the Roman Cicero, wrote of religion in the first century BCE:
There is no topic on which not merely the unlearned but even educated people disagree so
much, and since their beliefs range so widely and are so much at odds with each other, two
possibilities exist: it may be that none of them is true, or at any rate no more than one of them
can be. (1997, 4−5)
Cicero’s observation is one that we ourselves are also able to make today. Indeed, we are better
suited to make this observation than Cicero was, because the true extent of religious diversity was
poorly understood by ancient writers like Xenophanes and Cicero. After the voyages of discovery,
a far greater number of divergent and puzzling religious traditions were unveiled. The problem of
religious diversity was exacerbated with each new description of a foreign religious culture. The
probability that one’s own religious opinion was correct continued to shrink with every addition
of a new conflicting view. Today, the list of world religions is not only long but also daily growing.
243
It includes such oddities as UFO religions, psychedelic reli-

KEY CONCEPTS gions, New Age mystical movements, and the famous cargo
cults of Melanesia and Papua. Mainstream religions commonly
laugh these stranger candidates out of contention as serious
Christian Mystical Practice
candidates for religious truth. But why? On what grounds?
Disagreement And let’s take a moment of silence to remember the reli-
Exclusivism gions that are no longer with us. As often as we witness the
Mysticism birth of a new religion, we witness the extinction of another.
The polytheism of the medieval Norsemen left an indelible
Noumenon mark on the culture of northern Europe, and yet in modern
Perennialism times the religion persists only as an assortment of cultural
Pluralism fragments, a word here or there, a superhero movie. The mag-
nificent civilization of ancient Egypt was molded by a reli-
Overrider System
gious system that dominated every aspect of Egyptian life,
yet the artifacts by which we are acquainted with that religion
are now little more than curiosities for the cabinets of Egyp-
tologists. The great pyramids, for all their forgotten sanctity, are glorified tourist attractions.
How many other religions have perished without leaving any monumental headstones such as
these? We may assume that the number cannot be counted. How many more have perished
through deliberate acts of oppression, genocide, and war? Surely, the number is more than cur-
rently exist.
The innumerable variety of religions both extinct and extant, with their contradictory doctrines
and conflicting rituals, makes the task of finding the one true religion, if there is one at all, Hercu-
lean. The probability that one has without much difficulty happened upon the right religion is van-
ishingly small. Yet so often religious ideas are not only believed but, indeed, beloved and held indu-
bitably. This goes as much for the Mormon as it does for the Sikh. Against all odds, religious
persons very often believe that they are among the tiny minority of religious experts.
On what grounds can religious belief be maintained when the chances that one has happened
upon the one true religion are so very low and when it seems that all believers have an equally
strong sense that they are justified in their own beliefs? In answer to the problem, three popular
apologetic strategies have often been presented, and in their simplest forms they run as follows:
1. All religions are basically right.
2. All religions are partly right.
3. My religion is right, and the others are wrong.
The first apologetic strategy belongs to a school of thought called pluralism. According to the plural-
ist position, most famously propounded by John Hick, all the world’s major religious traditions,
while superficially inconsistent, are actually differing conceptualizations of an ineffable, noumenal
ultimate reality. On this account, religious diversity is the result of our diverse religious perceptions
and conceptions, but it is wrongheaded to think of any of these different conceptions as competing
with each other, to imagine that only one may turn out right.
The second apologetic strategy belongs to a school of thought called perennialism. According to
the perennialist position, despite the wide-ranging differences between the world’s religions, there
exists a common core of mystical wisdom that all religions express at the foundations of their creeds.
On this account, religious diversity is largely just accidental; the central, mystical doctrines held in
common by all religious traditions stand in agreement. In the perennialist account, all religions have
things partly right. If we so wished, we could cleave off the disagreement between religions and get
to the agreement at the core.
The third apologetic strategy belongs to a school of thought called exclusivism. Although there
are exclusivists of all stripes, I will examine only William Alston’s exclusivist solution to the problem
of religious diversity. Alston agrees with Cicero that at most one religion can have things right, but
Alston argues that it may nevertheless be rational to maintain religious belief in the face of religious
diversity, because religious diversity does not give religious persons any reason to think that there
exist any superior alternatives to their own religious belief-forming practices.
As I argue, none of these approaches can cure the problem of religious diversity, which remains
the most serious objection to the claim that there exist reliable religious belief-forming practices.
PLURALISM
John Hick is the most famous proponent of religious pluralism. According to Hick, the great world
religions (Christianity, Buddhism, Hinduism, Islam, etc.), while superficially contradicting each
other, are all responses to our experience with a fundamental noumenal reality that Hick terms
“the Real” (Hick 1989, 278). Hick argues that there need not be any overlapping propositional con-
tent between the contents of religious doctrines in order for different religions to be correct. Make
no mistake. Hick is no antirealist; religions make knowledge claims that may be true or false. Yet
the diversity of the world’s religious teachings need not lead us to skepticism concerning religious
belief, because religious belief is grounded in different reactions to a common object, the Real,
and in a desire to transform one’s life from self-centeredness to Reality-centeredness (Hick 1985,
86). Thus, it is mistaken to think that there is just one true religion—one religion that both accu-
rately describes religious truths and efficaciously transforms human life from self-centered to Reality-
centered. Each religion provides a context in which individual human beings encounter and respond
to the Real on different terms.
At this point, a critic may already be feeling impatient. Are we not human beings with shared
cognitive faculties which largely generate intersubjective agreement about the objects of experience?
When a duck is present, everyone agrees a duck is present. When there is a loud noise, everyone
gets a fright. If there is some sort of fundamental religious reality, shouldn’t we expect to find a
common conception of it, given our common cognitive faculties? Hick does not think so. His argu-
ment for religious pluralism begins with the observation that the universe is religiously ambiguous.
The world we inhabit admits of various religious and nonreligious interpretations. None of these
descriptions can claim to be better supported by the evidence, yet they all strive to express some-
thing about how the world appears to the believer and how one ought to live within it. The differ-
ing descriptions of the universe offered by different religions are therefore descriptions of different
phenomena. As Hick writes, the different interpretations are to be taken as “alternative soteriological
‘spaces’ within which, or ‘ways’ along which, men and women can find salvation/liberation/ultimate
fulfilment” (Hick 1989, 240).
By the phrase religiously ambiguous, Hick means something along these lines: Cosmological
arguments, design arguments, arguments from evil, arguments from religious experience, arguments
from divine hiddenness, and so on are not conclusive, and many fail to be convincing at all. The
premises of these arguments are open to either religious or naturalistic interpretations, and we inev-
itably understand them in the light of our own differing interpretative frameworks. Moreover, even
if the arguments could lend support to some religious position, the premises themselves have no
objectively quantifiable probabilistic value. Thus, the strength of any particular argument ultimately
is subjectively determined from within one’s own worldview anyway. Given that this is the situa-
tion—given that there is no reasonable way to settle the disagreement in principle—we are all
within our rights to believe what our experience tells us is the case. That is to say that it is reason-
able to follow some kind of principle of credulity, according to which we are within our epistemic
rights to believe that the universe is the way it appears to us, until we have good reason to doubt
that our experience is veridical. All of our various worldviews are equally reasonable to hold. It just
so happens that the world appears religiously to some people.
Hick’s ambiguity thesis depends on the Kantian distinction between noumena and phenomena.
In adopting a Kantian metaphysics, Hick believes he can account for the diversity of the world’s
religious teachings without impugning their validity. The Real, the fundamental noumenal reality
must be experienced in some way if it is to be experienced at all. We do not have direct epistemic
access to the noumenal Real, but we do have access to the phenomena. Thus, our different

perceptions and conceptions of the Real explain why there is so much religious diversity, despite all
religions having a common object of veneration. Hick says:
The differences between the root concepts and experiences of the different religions, their
different and often conflicting historical and trans-historical beliefs, their incommensurable
mythologies, and the diverse and ramifying belief-systems into which all these are built, are
compatible with the pluralistic hypothesis that the great world religions constitute different
conceptions of and perceptions of, and responses to, the Real from within the different
cultural ways of being human. (Hick 1989, 375–376)
Hick’s pluralism has often been compared to the fable of the blind men and the elephant. One
blind man feels the trunk and declares that it is a tree. Another feels the tail and declares that it is a
rope. Yet another feels the tusk and declares that it is something made of porcelain. Of course, none
of the men know what it is like to actually see the elephant, yet it is clear that they are all, in fact,
experiencing the elephant by way of their existing concepts and expectations. John Godfrey Saxe’s
poem “The Blind Men and the Elephant” concludes:
So, oft in theologic wars,

the disputants, I ween,
tread on in utter ignorance,
of what each other mean,
and prate about the elephant
not one of them has seen!
(SAXE 2007, 29)
For Hick, nobody has seen the elephant as it really is, and it is not even possible to accurately
describe the elephant’s real properties. The fundamental noumenon underlying religious experience
is ineffable, lacking all positive characteristics. Yet because our beliefs about the Real are grounded
in our phenomenal experience, we are right to believe what we do (including that the Real has cer-
tain positive characteristics). We can accept only what seems to us to be the case. What other policy
could we maintain? Our phenomenal experience is our only legitimate guide. In this way, Hick
argues, religious diversity is perfectly compatible with the theory that all religion is an attempt to
articulate and relate to a common noumenon.
What can we say about such an argument? It seems that Hick is right to say that religious
diversity is logically compatible with the hypothesis that religious belief stems from our differing
reactions to a common noumenal substrate. However, bare logical compatibility is hardly good evi-
dence that a hypothesis is true. Indeed, if the world’s religions agreed on a few more important doc-
trines (say, for starters, that religion is grounded in a reaction to an experience with an ineffable
noumenal substrate), this interesting fact not only would be compatible with Hick’s thesis but also
would constitute evidence for it. Unfortunately for Hick, the very fact that the phenomena of reli-
gious experience differ so radically is reason to suspect that the objects of the experiences are not
held in common. After all, even blind men come to agree about what elephants feel like eventually.
“What do you mean it’s a tree?,” says the one to the other. “Come over here and feel this great big
porcelain thing.” Convergence about the basic facts of our experience is the norm, not the excep-
tion, and when we find this convergence breaking down, we attempt to figure out who has gone
wrong and where. We do not just throw up our hands and proclaim that the universe is ambiguous.
Hick’s rebuttal to this point is predictable. He has already argued that the Real is ineffable.
There is no such thing as an accurate description of its positive attributes, and so we should not
expect any convergence around any particular description. As Philip Quinn puts it: “We must
assume that every guise in which [the Real] can appear is a disguise” (Quinn 1995, 149). Thus, it
is unlikely that we would find agreement concerning that which no one can veridically experience
or faithfully describe. Even if agreement is the norm for matters empirical, this is not so for matters
of “capital R” Reality. The Real must be experienced in some way, but this is not its way.
It is true that there is nothing in Hick’s argument that implies that we should expect there to
be any widespread religious agreement. All that can be said is that it is strange that when agreement
is so easy to find for matters concrete and empirical, disagreement should be the norm for matters
religious. In any case, the objection that there is a lack of religious agreement does not show that
Hick’s pluralism is internally inconsistent. However, the charge of inconsistency can be successfully
made once we learn what Hick has to say about soteriology and, specifically, once we learn how
Hick thinks we are able to meaningfully talk about salvation or liberation.
Hick identifies the common motivating factor of “postaxial” religions (a set comprising the
great world religions of the Near East and Far East) as the shift from self-centeredness to Reality-
centeredness (Hick 1989, 32). Salvation, or liberation, is the culmination of this shift. If the Real
has no positive characteristics, how then may we come to shift our focus away from the self and
toward the Real? If salvation depends on relating oneself in the proper way to the Real, then in
which direction ought I to turn? How will I know when I am approaching the Real, and how will
I know when I am distancing myself from it?
To solve this problem, Hick outlines a pragmatic account of truth, which he terms mythological
truth. This is a “practical truthfulness,” which “rightly relates us to a reality about which we cannot
speak in non-mythological [literal] terms” (Hick 1989, 248). If some description of the Real is myth-
ologically true, then it serves the pragmatic purpose of positioning us in the right relation to the Real
with regard to salvation. If a statement is mythologically true, then it is not to be taken literally.
Indeed, if a statement is mythologically true, then it is not to be taken as a metaphor either. Mytho-
logical truth is not any kind of relation between statements and the world. On the contrary, mytho-
logical truths are true only in the sense that they position us correctly with respect to the Real—they
evoke the right kind of attitude. Because Hick claims that mythological truths are not assertoric, their
existence apparently does nothing to undermine his pluralism. Victoria Harrison clarifies this point:
“At the literal level different religions describe different phenomena and hence do not contradict one
another, and at the mythological level there is no contradiction because, not being literally true or
false, myths are just not the sorts of things that can be in contradiction” (Harrison 2015, 263).
Even if mythological truth is not assertoric, even if it only serves the purpose of placing us in
the “right relation” to the Real, we may now reasonably ask what the right relation is and whether
our mythology guides us toward that relation. We may ask whether the mythology is effective.
Whatever the right relation is, it is fair to assume that there must be an opposite wrong relation that
one might adopt. Therefore, if it is the case that mythological truth serves to place us in the right
religion to the Real, it seems to be the case that different religions, through their different mytholog-
ical exhortations, place believers in very different relationships with their divine object(s). Indeed,
different mythologies from different religious cultures often elicit opposite courses of action in the
quest for salvation or liberation. It follows that some may be getting it right and some may be get-
ting it wrong. So which is it? Jihadism or pacifism? Asceticism or sybaritism? Wisdom or crazy wis-
dom? These are very different, even contradictory, relationships with the Real that have been
endorsed by different religious traditions through the course of history.
Different mythologies underwrite these relationships, and there do not seem to be any indepen-
dent grounds for taking up one mythology over any other. How am I to tell whether my religion is
salvific if I have no independent way to tell? It seems that given the extent of religious diversity with
respect to mythologies, it would still be unreasonable to suspect that I am positioning myself cor-
rectly with respect to the Real, whereas others, accepting their opposed mythologies, are not. It
would be unreasonable for me to suspect that I am among the tiny minority of religious experts.
Hick anticipates this problem and suggests that it can be solved in the following way: “The
basic criterion [for grading religions] is the extent to which they promote or hinder the great reli-
gious aim of salvation/liberation. And by salvation or liberation I suggest that we should mean the
realisation of that limitlessly better quality of human existence which comes about in the transition
from self-centredness to Reality-centredness” (Hick 1985, 86). Such a suggestion is patently ques-
tion begging. The standards by which religious believers judge whether a religion promotes salvation
are always internal to some preexistent religious framework. What is required is an independent cri-
terion by which we can identify which religions are successful at achieving the aim of salvation and
which are not. Hick gives this problem very glib treatment. He says that “the ways to salvation/lib-
eration are many and varied” and then continues that “we should respect ways other than our own,

whether or not we can truly appreciate them.” He concludes: “So far as we can tell, they are equally
productive of that transition from self to Reality which we see in the saints of all traditions” (Hick
1985, 86–87).
Are all religions equally salvific so far as we can tell? Frankly, so far as I can tell, I just can’t tell.
Hick fails to provide any criterion independent of one’s own religious tradition by which to judge
whether any religion is producing salvation. In lieu of any such independent criterion, the claim that
every religion is equally salvific is improbable, given their very diverse and incompatible mytholo-
gies. Without any independent corroboration of the salvific effects of, say, violent jihad compared
with vows of silence, it is an odd and improbable assumption that both of these methods will yield
the very same result. In lieu of any further information, we must assume that different processes
produce different results. Thus, the more probable hypothesis—if we assume that salvation is some
kind of state that can be achieved by some kind of method or methods—is that very different meth-
ods are likely to produce very different results.
Perhaps there is an independent measure of salvation to which we can appeal. Hick says that we
see salvation at work in the saints of all traditions. So, then, who are these saints that Hick is refer-
ring to? By what independent criterion can we recognize them as saintly? Alas, the behavior of those
men typically labeled “saintly” often takes us by surprise. Sathya Sai Baba is regarded as a saintly, if
not godly, figure by millions of Indians, and yet it is alleged that he regularly preyed on young boys.
Are we then to assume that he is not a saint or that he is a saint but that pederasty was a weakness
attributable to his mere humanity and therefore not to be considered saintly behavior? Or are we to
assume that pederasty is actually one of the marks of the saintly life? I have no way to tell. The first
guru of Sikhism, Nanak, preached a pacifist doctrine: “No-one is my enemy, and no-one is a
stranger” (Kapoor and Kapoor 2002, 170). The tenth guru, Gobind Singh, preached a doctrine of
just war: “When all strategies brought to bear are exhausted, it is then lawful to take steel in hand”
(Fenech 2013, 55). Muhammad, regarded as the greatest of all men by Muslims around the world,
famously kept sex slaves, had upward of ten wives (in violation of or as a convenient exception to
Islamic law), consummated his marriage with Aisha when she was approximately nine years old,
and executed over six hundred surrendered Jewish men of the Banu Qurayza tribe. Is this what it
means to be Reality-centered? Are these saintly lives? I do not know who Hick’s saints are, and I
cannot pretend that it is just obvious who they are. The lack of any independent measure makes
the identification of true saints a hopeless task.
The fundamental problem is that the salvation that is so essential to the state of sainthood is
always defined within the religious tradition to which one belongs. This may make it obvious to
the believer, from the believer’s point of view, who the true saints are, but it is not clear to any out-
sider. At one point, Hick says that “the fruits of openness to the divine Reality are gloriously evi-
dent” in the lives of religious men and women on the path to Reality-centeredness (Hick 1985,
91). Again, this is just table thumping. Such an answer is of no help to me, because I simply do
not see these “gloriously evident” fruits; rather, what is labeled a fruit by some religions seems often
to me to be a vegetable.
In a discussion on the marks of saintliness, William James states the problem most clearly:
If, for instance, you were to condemn a religion of human or animal sacrifices by virtue of your
subjective sentiments, and if all the while a deity were really there demanding such sacrifices,
you would be making a theoretical mistake by tacitly assuming that the deity must be non-
existent; you would be setting up a theology of your own as much as if you were a scholastic
philosopher. (James 1999 [1902], 360)
Yet faced with this situation, James’s advice is to “test saintliness by common sense, to use human
standards to help us decide how far the religious life commends itself” (James 1999 [1902], 363).
That is to say, the salvific fruits of the religious life can be measured in the same way that we might
measure other, more mundane moral fruits. Perhaps this advice is quite sensible. If Sathya Sai Baba
indeed preyed on young boys, then given that this is morally wrong, we might safely conclude that
this cannot also be saintly behavior. However, the very idea that human moral standards approxi-
mate divine standards for salvation needs justification. Indeed, the very reason that costly religious
practices, such as violent jihad or human sacrifice, place such extreme demands on their practitioners
is that such practices are not measured by human or secular standards but by the standards of super-
natural beings, whose demands from us are rather different from what we might demand from our-
selves. A life of self-torture may recommend itself to no sane or moral person, but such a life may
be precisely what the gods demand from us in exchange for a heavenly reward. The basic problem
for James’s argument is that different religions may be equally effective at producing moral and virtu-
ous people but not equally effective at saving or liberating people, and we have no reason to think that
the one is a reliable indicator of the other (apart from, dare I say, wishful thinking).
To return to Hick’s notion of mythological truth, a final point can be made. If it is not true
that violent jihad effects personal salvation, then the statement “violent jihad effects personal salva-
tion” is assertoric, truth-apt, and false. If selfless service to others produces personal salvation, then
this can be truthfully stated. Indeed, the world’s great religious traditions are often explicit about
the courses of action required for salvation. In speaking directly about the salvific virtues of religious
violence, self-mutilation, silent contemplation, or charitable giving, religions step outside the con-
fines of mythological truth and engage directly with literal statements of fact. If mythological truth
were the only tool available by which we could gesticulate about salvation, then Hick could not
truthfully say, as he does, that “the ways to salvation/liberation are many and varied.” Yet he takes
this latter claim to be quite accurate.
While Hick argues that all the great world religions are basically right, he accepts that some reli-
gious practices are more effective than others. However, Hick accepts that given our limited per-
spective on the matter, we must accept that the great religious traditions are equally salvific so far
as we can tell. His argument depends on the epistemological claim that it is beyond the ken of
the human mind to decide which religions are more likely to be on the money and which ones
are failing dismally. Indeed, on this point I think he is right. It is probably beyond our ken to
decide which religions are succeeding and which are failing, but that’s no defense of religious belief
against the problem of religious diversity; it is, quite the contrary, a recapitulation of the problem.
In Hick’s own words:
For each of these long traditions is so internally diverse, containing so many different kinds of
both good and evil, that it is impossible for human judgement to weigh up and compare their
merits as systems of salvation. It may be that one facilitates human liberation/salvation more
than the others; but if so this is not evident to human vision. (Hick 1985, 86)
Quite so. Likewise, if it is not evident to our various modes of investigation which religions are suc-
ceeding and which are failing, then it is possible (nay, likely) that one’s own system is among the set
of failures, and so one requires a special reason to believe that one is among the privileged group of
religious experts and not among the group of religious ignoramuses. The problem of religious diver-
sity stands just as it did before.
Hick’s final defense is his principle of credulity: we are within our epistemic rights, he says, to
believe that the world is the way it appears to us until such time as disconfirming evidence arrives
on the scene. Perhaps so, but religious diversity is just that: disconfirming evidence arriving on the
scene. Religious belief-forming practices generate widespread disagreement, and this is evidence that
those practices are unreliable. Hick’s pluralism does not escape the problem of religious diversity.
PERENNIALISM
Perennialism is the thesis that all religions agree on a minimal set of claims, typically known via the
practice of mystical experience. This overlapping propositional content, this agreement, is the total-
ity of true supernatural knowledge. It is crucial to the perennialist’s claim that there actually exists
some amount of propositional agreement between the world’s religions. The perennialist does not
argue, in the style of Hick, that every religion engages with the same reality in a different way.
No. The perennialist argues that religions deal with the same fundamental reality in the same way
and typically arrive at the same beliefs about it. Therefore, before discussing any of the perennialist’s
supernatural hypotheses, I will briefly outline what kind of agreement is commonly found between
different and isolated religions.

First, no one disputes that there are examples of religious agreement among remote commu-
nities that derive from a common historico-cultural source. Samoa and Poland, for example, are
predominantly Christian countries. The distance between these two countries is immense, and
there had been, before the arrival of Europeans, no contact between these two cultures for eons.
However, it is no surprise that both communities are now predominantly Christian. The histori-
cal dissemination of Christianity via missionaries during and after the age of discovery is well
documented. Let us imagine, however, that European explorers arrived in Samoa to discover the
isolated native people already wearing crosses and already believing in a Trinitarian god. This dis-
covery, that Samoans had arrived at Christian doctrines independent of cultural diffusion, would
have counted as extremely good prima facie evidence for the existence of reliable religious belief-
forming practices.
Is there any surprising agreement like the unlikely scenario described here? Well, almost. Iso-
lated religions occasionally share unexpected and detailed overlap, particularly with respect to their
origin myths, the existence of human souls, spirits that live without bodies, and the existence and
nature of an afterlife. In some cases, the resemblance is jaw-dropping. For example, in ancient Greek
mythology, a primordial earth mother and sky father bore a progeny of lesser gods, who rule over
certain natural domains, such as the oceans. Likewise, New Zealand Māori mythologies commonly
describe a primordial earth mother and sky father, whose resulting progeny make up the system of
lesser gods who rule over special domains. Another well-distributed mythological idea is that of the
world egg—a primordial egg out of which the universe was born. Such a myth can be found
throughout Eurasia, North Africa, and Oceania. There is simply no reasonable diffusionist hypoth-
esis according to which all such resemblances are the result of a common cultural source, no serious
hypothesis of prehistoric Greco-Polynesian contact. These are surprising examples of religious agree-
ment that cannot be accounted for by cultural diffusion. To borrow the words of the ethnologist
Daniel Garrison Brinton, history cannot unriddle cases such as these.
To explain the nonrandom distribution of ideas about the supernatural, one might hypothesize
that some religious belief-forming methods are reliable. These methods, by way of a mechanism as
yet unknown to us, place practitioners into contact with a supernatural realm. So, one might argue,
religious agreement is evidence that there exist religious methods that give access to religious facts.
Such an argument, call it the Argument from Religious Agreement, might run as follows:
1. Otherwise isolated religions share surprising overlap in their doctrinal contents that cannot be
accounted for as the result of a shared cultural history.
2. If this overlap in content cannot be accounted for as the result of a shared cultural history, it
must result from the independent exercise of reliable religious belief-forming methods.
Therefore, the overlap in content among the world’s religions is accounted for by the independent
exercise of reliable religious belief-forming methods.
This argument, or some version of it, is the cornerstone of the perennialist position. The per-
ennialist argues that there is a common core of religious truth, some deep wisdom or supernatural
experience that all the world’s religions express as a part of their total doctrine. However, the peren-
nialist is usually not particularly interested in the examples of agreement alluded to earlier (world
eggs, sky fathers, ancestor spirits, etc.). Instead, the perennialist takes mystical experience to put
the practitioner into contact with a divine or ultimate reality and believes that this experience, along
with its associated fundamental moral and soteriological insights, is accurately described by many
religious traditions.
Aldous Huxley’s Perennial Philosophy of 1945 is the most famous work drawing together the
supposedly common teachings of the great world religions, claiming that such teachings are com-
mon expressions of deep religious truths. Huxley, like Hick, focuses on the wisdom found within
the “great world religions” in their shared claim to unitive knowledge of the divine. Huxley never-
theless acknowledges that this wisdom is not restricted to scripture-based religions, as he says that
even the “traditional lore of primitive peoples” contains the rudiments of this knowledge (1945,
vii). Huxley is, then, a rather optimistic perennialist. He takes it that a relatively large subset of all
religious knowledge claims should be regarded as accurate.
In contrast, James is somewhat more pessimistic. He maintains that there are nonnegligible dif-
ferences between the doctrines of different religions, that ultimately “feeling is the deeper source of
religion,” and that popular religious doctrines have almost invariably been garbled by philosophical
and theological considerations that corrupt the original content much like “translations of a text into
another tongue” (James 1999 [1902], 470). So far as James can see, the task of uncovering perennial
mystical wisdom is more difficult than Huxley might imagine. This task, he says, belongs to philos-
ophy. Rather than generating a priori conceptions, definitions, and deductive arguments for some
divinity, the role of philosophy is to appraise existing religious doctrines, to reject those that conflict
with scientific or historical facts, and to point us in the direction of those perennial doctrines that
are common and essential to different religions. Lady Philosophy can lead us toward a set of verifi-
able religious claims: “From dogma and from worship she can remove historic incrustations” (James
1999 [1902], 496). While James and Huxley disagree over exactly how much doctrinal agreement
we should expect to find among religions, they nevertheless agree that there is a core of common
religious understanding that can be uncovered by mystical experience.
Note that one may accept the claim that mystical experience tends to generate independent
agreement between practitioners without accepting the claim that mystical experience is thereby
shown to be a reliable source of knowledge. Consider the fact that we all agree that sticks look bent
when stuck into water but do not actually bend at all. Indeed, where mystical experience is con-
cerned, there are at least three reasons why we might be especially doubtful that the beliefs gener-
ated are accurate.
The first reason is that when it is actively sought, mystical experience is often elicited by rather
extreme measures, such as sensory deprivation or hyperstimulation. The role of psychoactive drugs
in bringing about mystical experience is also common. That such drastic, destructive, or painful
measures are necessary to elicit mystical experience might lead one to think that mystical experience
is pathological—the result of a disordered, distressed, or dysfunctional mind. Indeed, Bertrand Rus-
sell was of this opinion, writing: “From a scientific point of view, we can make no distinction
between a man who eats little and sees heaven and the man who drinks much and sees snakes. Each
is in an abnormal physical condition, and therefore has abnormal perceptions” (Russell 1935, 188).
Against Russell, C. D. Broad replied that perhaps a person simply has to be a bit “cracked” to get
“peep-holes” to the supernatural (Broad 2014, 198). That our cognitive faculties are used in an
abnormal way is no evidence that their deliverances will be unreliable when the putative objects
of the experience are themselves abnormal. Broad may have a point. Indeed, anyone familiar with
the Magic Eye illusion series knows that the only way to see the hidden image is to look at the page
with a diverged focus. Russell’s argument nevertheless remains popular, and it seems reasonable to
conclude that, at the very least, we have no positive, supporting evidence showing the mystic’s expe-
rience to be any more reliable than the drunkard’s.
The second reason we may remain doubtful about the mystic’s claim to knowledge is that mys-
tics agree overwhelmingly that mystical knowledge is utterly unlike the mundane knowledge of our
everyday experience, so much so that it is commonly taken to be either inherently self-contradictory
or ineffable. If we assume there are no true contradictions, then given that the knowledge is self-
contradictory, it is no knowledge at all. If, on the other hand, the knowledge is ineffable, then we
have no way to appraise its truth value. Of course, this latter position leaves open the possibility that
mystical experience may be veridical—it may be an experience that places the mystic in contact with
a deeper, spiritual reality—but if this reality is not independently verifiable or even capable of being
described, then mysticism cannot be shown to be reliable. Likewise, if the mystic’s claims are simply
not amenable to independent confirmation, then so much the worse for the mystic’s claims.
The third reason that we may be unconvinced that the mystic is arriving at bona fide knowl-
edge is that the agreement delivered by mystical experience, when it is not self-contradictory or inef-
fable, is not particularly surprising. There is a certain obviousness or banality to most of the unitive
metaphysical and moral claims of the mystic, making the existence of independent agreement unre-
markable. For example, the most common mystical claim is that all things are one. Reality is an
indivisible whole. Good and evil are one and the same and so on. Russell emphasizes the doctrine
of cosmic unity as one shared by all mystics. James agrees, stating that “in mystic states we both

become one with the Absolute and we become aware of our oneness. This is the everlasting and tri-
umphant mystical tradition” (James 1999 [1902], 457). Accordingly, Walter Stace also writes that
the experience of unity is “the one basic, essential, nuclear characteristic” of mysticism, from which
all the other characteristics inevitably follow (Stace 1960, 110).
It is reasonable for us to ask, then, just how unlikely it is that independent practitioners of mys-
ticism would come to the very same conclusion that everything is one. Such agreement does not
seem terribly unlikely on some basic background assumptions. Monism appears to be one of a lim-
ited number of rather obvious metaphysical alternatives. Either there is (1) nothing at all, (2) just
one thing, or (3) two or more things. There are other ways we might carve up the options, of
course, but this triad is roughly in line with our intuitions about how the options lie. Perhaps
monism is correct, and the mystic comes to learn this fact through mystical union with the divine;
alternatively, perhaps monism is false but is just the most attractive option to the human mind
when it is in a state of trance, hypnosis, or ecstasy. In either case, mystics converging on monism
never had a terribly low prior probability to begin with. Consider how strange it would be, in con-
trast, if mystics from different eras and cultures independently agreed that there were precisely sev-
enteen things. This strange agreement would be a much better clue that mystical experience was a
reliable source of knowledge, because the prior probability of an agreement about such a proposition
would be so low. Indeed, if subsequent scientific investigation began to support the claim there were
indeed just seventeen things, this agreement would be further evidence that mystical experience is
reliable.
Indeed, while the question as to which of (1), (2), or (3) is correct remains a matter of philo-
sophical speculation, we have no strong, independent support for the truth of (2). This does not
show that the mystic is wrong by any means. However, it does show that intersubjective agreement
converging on one proposition from among a very limited number of relatively likely alternatives is
not particularly strong evidence for the reliability of the method used and that this is especially so
when the question is an open one regarding the ultimate number of components of the universe.
We have little independent support to suggest that the mystic has indeed hit the nail on the head,
and it is doubtful whether any such support will be forthcoming.
The perennialist is right to accept the problem of religious diversity as genuinely problematic,
and it seems to be a sensible procedure to continue to search for shared doctrines between different
religions and to search for those common doctrines that have been delivered by the same method(s).
By such a procedure, it seems we may discover that some religious belief-forming practices are reli-
able. The trouble is, however, that the doctrines shared by mystics appear to be of very little con-
tent, and whatever content there is seems self-contradictory or ineffable or receives no independent
corroboration from elsewhere. The prospects for the reliability of mystical experience, then, fail to
look very promising.
EXCLUSIVISM
Alston thinks it is quite reasonable to believe that Christianity is the one true religion. That being
said, he also argues that Sikhs may be entitled to think the very same thing about Sikhism and that
Siberian shamans may be entitled to think the same thing about their shamanism so long as certain
conditions hold. Indeed, all religious believers could, in principle, have good reasons to accept their
own traditions as veridical. Alston’s is nevertheless an exclusivist position: at most one of the world’s
religious traditions has got things right, and Alston puts his money on Christianity. To be more spe-
cific, Alston takes it that he is within his epistemic rights to continue to use his own preferred reli-
gious epistemic method, which he terms Christian Mystical Practice (CMP).
Alston has two central arguments that together challenge the claim that religious diversity inva-
lidates religious epistemic methods. First, he argues that different religions use quite different episte-
mic methods to develop and appraise beliefs. Therefore, he contends, religious diversity is not evi-
dence of intra-practice unreliability and establishes only that the results of different methods used
by different religions do not converge. Second, he argues, given that religious diversity results from
the application of diverse religious methods, religious diversity is not evidence indicating that any
particular method is the unreliable one. One is entitled, therefore, to “sit tight” (Alston 1991,
274) with the methods that one is fluent in using, because religious diversity is not evidence that
one’s own method is unreliable.
Alston’s first argument runs along the following lines. The belief-forming practices of different
religions are not identical. There are important differences in the “overrider systems” of the practices
of different religions (Alston 1991, 167). These differences are extensive enough to justify our clas-
sifying them as different kinds of practices altogether. Whenever a religious person applies a partic-
ular belief-forming practice, the belief outputs are typically internally consistent. It seems, then, that
none of these distinct belief-forming practices habitually generates contradictory beliefs. Therefore,
religious diversity is not evidence of internal inconsistency. Religious diversity is just evidence that
different belief-forming practices generate different results, and that hardly seems all that surprising.
What does Alston mean when he speaks of an overrider system? An overrider system is a set of
procedures, accepted by the epistemic community, for evaluating the likelihood that some output of
an epistemic method is true. Its role is especially important with regard to those beliefs that we have
an independent reason to doubt. Alston says that the overrider system of each belief-forming prac-
tice “determines how we go from prima facie to unqualified justification” (1991, 189). The overri-
ders of our beliefs divide into two kinds: rebutters and underminers.
A rebutter is a background belief that contradicts the belief in question, whereas an underminer
is a reason for supposing that the method justifying the belief may not be performing reliably in the
present context. Religious belief-forming practices with very different overrider systems cannot be
counted as variants of one and the same practice. Instead, they are different practices altogether.
Typically, the background beliefs of different religions differ, and so it follows that different rebut-
ters may be brought to bear on any religion’s mystical practices. Moreover, the conditions under
which certain practices are taken to be reliable differ from tradition to tradition.
Note that the overriders for our everyday perceptual belief-forming practices command an
almost universal assent. We are all familiar with them. Imagine that one day I see a red flash dart
past my window. I come to form an unreflective belief that a macaw just flew past. After applying
my overrider system, however, I come to be skeptical about that belief. I have the following rebutter
at hand: there are no wild macaws, as far as I’m aware, in the far south of New Zealand. I also have
the following underminer: the red flash was moving particularly quickly, which would make identi-
fication difficult. There are other overriders, of course: The window was small, and so the object was
in my frame of vision only briefly. I heard no parrot calls, either before or after the event. I was
watching a documentary on macaws at the time, and so I might have had macaws on the brain.
On balance, I reject the belief that a macaw flew past the window.
What about the overrider systems for religious methods? What are their associated rebutters and
underminers? Alston argues that each religious tradition sanctions a different overrider system for its
distinctive belief-forming practices. Moreover, some beliefs of some people who claim to have had
mystical perceptions may be overridden within some traditions. As an example, Alston considers
the case of Jim Jones of the People’s Temple, who reported that God revealed to him that it was
his will to have all those at Jonestown commit suicide. Alston notes that “it seems very unlikely,
given the account of the nature, purposes, and pattern of activity of God in the Christian tradition,
that God would command any such thing” (1991, 190). Thus, concludes Alston, Jones’s revelation
is simply discordant with the Christian tradition. The Christian community, then, is right to think
that Jones was mistaken in some way. The idea is that there was a rebutter of the alleged revelation
that Jones failed to apply.
Alston thinks the least controversial candidate for an overrider is internal inconsistency. If a
belief-forming practice habitually generates a large enough number of internally inconsistent belief
outputs, this would be evidence that the practice is unreliable. Thus, Alston contends that because
religious belief-forming practices, as practiced within each religious tradition, do not generate a large
number of internally inconsistent belief outputs, religious diversity is not evidence that a particular
practice is unreliable. Of course, one mystic may arrive at a belief that contradicts a belief held by a
mystic of another stripe, but this is no problem, because neither mystic is contradicting himself.

The funny thing is that Alston is just plain wrong about this. Mysticism, as practiced within
most religious traditions, does generate a large number of internally inconsistent belief outputs. This
point has been made already in the discussion regarding perennialism. I need only point to the vast
literature on mysticism, according to which avowedly paradoxical, inconsistent, and contradictory
beliefs are characteristic outputs of mystical experience. Eminent theorists of mysticism, such as
James ([1902] 1999), Stace (1960), Matthew Bagger (2007), and Ralph Hood (1975) agree that
paradoxicality is characteristic of the belief outputs of mysticism. There is no shortage of direct
quotes from mystics that make clear that many of the belief outputs are paradoxical. Furthermore,
the paradoxical nature of mystical experience is supposed to be a kind of hallmark of the depth
and awe-inspiring mystery of that kind of knowledge. Either Alston is unaware of the pervasiveness
of paradoxical deliverances of mystical experience, or he simply does not count these mystics as giv-
ing literal descriptions of their experience. If the latter is the case, why should we say that reports of
their experiences are reliable?
I admit, that the latter objection depends on understanding the nature of mysticism somewhat
narrowly. In Alston’s sense of the term, mystical perception applies to a variety of religious practices
from which mystical experience is often difficult to untangle—visions, dreams, the internal testi-
mony of the Holy Spirit, and so on. While there is good evidence that mystical experience, narrowly
defined, routinely generates internally inconsistent outputs and so is unreliable, this is not evidence
that indicts any of these other associated practices. That being said, the objection cited at least
shows that Alston would be forced to reject an important subset of relatively highly esteemed reli-
gious epistemic methods as unreliable on the ground of internal inconsistency.
What can we say about the associated religious practices with which Alston is concerned, such
as dreams and visions? Are these internally inconsistent too? To find out, we would have to identify
a mystical practice within a religion and observe what kinds of outputs resulted. However, it is
unclear exactly how to draw a nontrivial line between different religions according to their different
mystical practices, especially given what Alston had to say about differences in overrider systems. As
Evan Fales has noted: “The mystical practices of snake-handling Pentecostals more closely approach
loa possession in Haitian Voodoo than they do the leadings of the Inner Light experienced by a
sedate Philadelphia Quaker” (Fales 1996, 30). Yet any appeal to different overrider systems just does
not seem to help us draw a line between religions, because this leads us to triviality.
This triviality problem can be seen better when we consider again Alston’s example of Jones’s
suicidal revelation. Alston argues that Jones’s alleged deliverance from the Christian god is overrid-
den by a rebutter that Jones failed to apply. That rebutter is that the Christian god’s purposes
and behavior are inconsistent with Jones’s suicidal revelation. It seems, however, that if Jones’s belief
is overridden by this rebutter, which is a part of CMP, and if Jones failed to take this particular
rebutter seriously, then Jones was using a distinct, non-CMP. Indeed, simply by virtue of accepting
the apocalyptic revelation, Jones accepted a different set of overriders and so does not count as hav-
ing applied CMP. Yet by adopting this kind of procedure, Alston can trivially define his way to
internal consistency. This is a point also noted by Fales: “Rescuing the internal consistency of a
practice by excising from it all conflicting elements is to trivialize the claim of consistency” (Fales
1996, 29).
However, if we broaden the scope of the term mystical practice such that different religious
traditions tend to share many of the same practices, then the problem of internal inconsistency is
doubly intensified. Such practices generate inconsistency with regard to the results of mystical
perception, and they also generate disagreement about what the rebutters and underminers of mys-
tical perception are. There is no convergence with respect to beliefs about the supernatural and
no convergence with respect to beliefs about the bounds of reliability of mystical practices. The
problem of internal inconsistency is much greater than Alston presumes, because it affects both
the belief outputs of the belief-forming practices and related beliefs about their associated rebutters
and underminers.
All this goes to show that there is a stark contrast between the overrider systems of religion and
perception. Regular perception generates agreement both about its results and about its undermi-
ners. Indeed, regular perception has many underminers that garner universal assent. All members
of the epistemic community are familiar with these underminers, even if they cannot verbalize them.
No developmentally normal human being, from any community, accepts that visual identification is
more reliable when applied to small, distant, and fast-moving objects than to large, close, and slow-
moving ones. We reliably discover the underminers of our perceptual practices by a process of cross-
checking with other practices. If I am doubtful that my eyes are telling me the truth, I can reach out
and check with my fingers. Alston himself acknowledges that the underminers of perception are dis-
covered by a process of applying a variety of tests that rely on a variety of different faculties. Yet we
have no ability to cross-check the deliverances of mystical perception, because, as Alston notes, reli-
gious perception is an autonomous epistemological enterprise. “We are simply unable,” he says, “to
go about testing a particular report of perception of God in the ways we can test reports of sense
perception” (Alston 1992, 72).
This all sounds a bit like special pleading. Indeed, if religious perception is autonomous in the
way that Alston says, then how could religious believers from distinct religious communities ever
converge around true beliefs about what the underminers of mystical practice are? In principle, there
could be no cross-checking by which we could establish that a patch of epistemic territory was more
solid than any other. If the underminers are themselves constitutive of mystical practices—that is, if
the underminers determine what kind of mystical practice we’re engaged in—then how could we
ever converge around the truth with respect to which religion is likely true? There could be only
deadlock. We are given no rational procedure for choosing between competing mystical practices,
and the internal inconsistency associated with religious belief-forming practices is, therefore, not
only widespread but in principle unavoidable. In this way, Alston’s first exclusivist argument fails.
Religious diversity is evidence of (unavoidable) intra-practice unreliability.
I turn to Alston’s second argument. This argument runs that the mere existence of religious
diversity is a fact that is silent about which, if any, of the competing religious practices is the reliable
one. He says: “In the absence of any external reason for supposing that one of the competing prac-
tices is more accurate than my own, the only rational course for me is to sit tight with the practice
of which I am a master” (Alston 1991, 274). In other words, the existence of disagreement between
competing mystical practices is no threat to any particular mystical practice, because no particular
mystical practice is more firmly established than any other. By “firmly established,” Alston means
that there is no particular mystical practice that is better socially established or that delivers a more
tangible degree of “self-support” (1991, 171−174). For the Christian, this self-support comes in
terms of an increase in “serenity, peace, joy, fortitude, love and other ‘gifts of the spirit’” (1991,
276). For believers of other religions, self-support may take up a different form. Yes, there is dis-
agreement, but this disagreement does not, on its own, indicate which of the available religious
alternatives is best. Believers are therefore entitled to “sit tight” with whatever practice they are
familiar with.
This is an expert sleight of hand by which Alston stacks the deck. Alston omits to mention that
the problem of religious diversity forces us to decide not only which mystical practice is reliable but
also whether any are. Alston says that he has no external reasons to think that a competing mystical
practice is more reliable. This may be true, but Alston ignores the question of whether there are any
external reasons to think that a nonreligious practice might be. Perhaps there are nonreligious meth-
ods that deliver more accurate information about allegedly supernatural things than the deliverances
of mystical perception. Even if we accept Alston’s claim that the social establishment and self-sup-
port of a mystical practice lend warrant to that practice, reason and sense perception are more
widely socially established and more convincingly self-supported (not to mention independently
supported) than mysticism anyway. Therefore, if the deliverances of mystical perception clash with
the deliverances of reason and sense perception, believers would not be rational to “sit tight” with
their particular mystical practice.
Having said all this, there is another problem to note. Alston’s exclusivism faces the same prob-
lem that the perennialists failed to take seriously. The diversity of the world’s religions is only one
small part of a larger problem. There is also surprising religious agreement that has to be accounted
for. Religions worldwide share puzzling overlap, especially with regard to their basic metaphysical
claims, their mythologies and cosmogonies. Critics of religious belief-forming practices such as I

have good reason to think that there are convincing naturalistic explanations for this overlap. Work
in the cognitive science of religion has seen large strides in explaining the human propensity for
belief in disembodied minds. See, for example, the work of Pascal Boyer (1994) and Justin Barrett
(2004). Naturalistic explanations are also available for the similarities in the mythologies and cos-
mogonies of different religious cultures. See, for example, Edward B. Tylor (1929) and Claude
Lévi-Strauss (1955). These naturalistic explanations of religious belief can account for both the com-
mon ground held between religions and the differences that separate them.
However, for the defender of (especially Abrahamic) religious belief-forming practices, the puz-
zling overlap between religions has to be taken rather more seriously. After all, if there are reliable
religious belief-forming practices, they should generate independent agreement. We would be wise,
then, to take up the perennialist’s project—first identifying any surprising agreements held between
isolated religious communities and then working our way backward to find which belief-forming
practices are generating the agreement in question. We could then single out these practices as the
(most probably) reliable ones. Yet if we search for this kind of surprising independent agreement,
it appears that the most likely candidate for religious truth will be a sort of “primitive” animism.
It is an ironic situation, then, to find that the problem posed by religious agreement is arguably
a severe handicap for a defender of Christian belief-forming practices such as Alston. Moreover, it is
a handicap not necessarily faced by the Christian perennialist or the Christian pluralist. Alston’s own
evangelical Christian practice generates relatively little agreement among separate investigators inde-
pendently applying their various religious practices. Christianity has few doctrines that are indepen-
dently corroborated by a range of alternative religious belief-forming practices. It is very improbable,
therefore, that CMP is reliable. Alston’s personal brand of Christianity is largely an oddity in the
history of religious thought, despite being, for quite accidental, historical reasons, the dominant reli-
gion in the world today.
In contrast, compare the tenets of Christianity with the tenets of the following mock religion
that I dub the Church of the Surest Bet. The Church of the Surest Bet is a syncretist religion that
takes up only those beliefs that exhibit remarkable or surprising recurrence in isolated cultures. The
church’s doctrines are decided by a meta-study of the belief outputs of religious methods. The doc-
trines of the Church of the Surest Bet would probably consist of something like the following: An
earth mother and sky father produced several divine children who embody certain biospheres, such
as the oceans, forests, rivers, and animals. From the waters all things were formed, including dry
land, plants, animals, and human beings. Human spirits survive death and remain in the earthly
realm for an undefined period before either entering the bodies of newborns or departing for the
underworld, depending on the quality of their behavior in this life. These spirits may also be manip-
ulated into entering physical objects, such as statues and amulets, in order to protect their relatives
from spiritual harm. Condemned spirits travel far to the west, where they descend into the under-
world. Even-toed ungulates are sacred.
The Church of the Surest Bet is far better justified by independent agreement than Alston’s
own brand of Christianity. While Alston may argue, quite correctly, that the Church of the Surest
Bet is not a socially established religion, the degree of independent agreement converging around its
doctrines is a better indicator of truth than social establishment anyway. Indeed, it deserves to be
said that the ways in which religious beliefs have traditionally become socially established (through
wars, intermarriage, tax breaks, etc.) are hardly the kinds of processes that lend any support to the
veracity of the beliefs held. If Alston takes independent agreement to be at least some indicator of
reliability (and he does), then it seems he will struggle to explain why Christianity is rationally pref-
erable to the Church of the Surest Bet. Alston’s exclusivist defense of CMP fails.
And so neither the pluralist, nor the perennialist, nor the exclusivist can successfully defend the
reasonableness of religious belief against the evidence of religious diversity, which remains the stron-
gest evidence against the claim that any religious belief-forming practices are reliable. As noted at the
beginning of this chapter, the ongoing emergence of new religious movements and the extinction of
others compound the problem for the defender of reliable religious methods. With the addition of
each new and contradictory religious doctrine the argument that there are reliable religious belief-
forming practices gets weaker and weaker.
Although I have attempted to expose some of the shortcomings peculiar to pluralism, perenni-
alism, and exclusivism, all three apologetic strategies share a common failing. None of the three
strategies recognizes the crucial importance of the independent testing of beliefs by a variety of dis-
tinct methods. It is unfortunate that, to date, no distinctively religious belief-forming practices have
been found to generate beliefs that are independently corroborated by any other methods. The per-
ennialist comes closest, seeking to accept only those religious beliefs that garner widespread intersub-
jective agreement among mystics. This is at least a good place to start. Going beyond this initial
agreement, the claims of the mystic (when they are not paradoxical or vacuous) fail to be corrobo-
rated by any other methods.
Once the disagreement between religions is excised, there appears to be virtually nothing of
substance left over. If we carve off, bit by bit, all the disagreement from the world’s religions, what
remains seems fundamentally impoverished and lacking in any external evidential support. One
might be reminded of Antony Flew’s remark that God has died a death of a thousand qualifications.
The more we refine our religious beliefs to stand up to the rigor of our shared reality, the less of
them we get to keep. Moreover, the more seriously we take the religious views of those with whom
we disagree, the less seriously we must come to take our own. Without any independent reason for
thinking any competing religion to be on the money, we are forced to take the problem of religious
diversity as evidence that no religious belief-forming practices are reliable. If God has died, it was
not a thousand qualifications but a thousand other gods who did the finishing off.
Bibliography
Alston, William P. Perceiving God. Ithaca, NY: Cornell University Press, Hick, John. Problems of Religious Pluralism. New York: St. Martin’s Press,
1991. 1985.
Alston, William P. “The Autonomy of Religious Experience.” International Hick, John. An Interpretation of Religion. New Haven, CT: Yale University
Journal for Philosophy of Religion 31, no. 2/3 (1992): 67−87. Press, 1989.
Bagger, Matthew. The Uses of Paradox: Religion, Self-Transformation and the Hood, Ralph W., Jr. “The Construction and Preliminary Validation of a
Absurd. New York: Columbia University Press, 2007. Measure of Reported Mystical Experience.” Journal for the Scientific
Barrett, Justin L. Why Would Anyone Believe in God? Walnut Creek, CA: Study of Religion 14, no. 1 (1975): 29–41.
AltaMira Press, 2004. Huxley, Aldous. The Perennial Philosophy. London: Harper and Brothers,
Boyer, Pascal. The Naturalness of Religious Ideas: A Cognitive Theory of 1945.
Religion. Berkeley: University of California Press, 1994. James, William. The Varieties of Religious Experience: A Study in Human
Brinton, Daniel Garrison. The Religious Sentiment, Its Source and Aim: A Nature. New York. Random House, 1999. First published 1902.
Contribution to the Science and Philosophy of Religion. Philadelphia: Kapoor, Sukhbir Singh, and Mohinder Kaur Kapoor. Guru Granth Sahib:
Henry Holt, 1876. An Advanced Study. Vol. 1. New Delhi: Hemkunt Press, 2002.
Broad, C. D. Religion, Philosophy and Psychical Research. London: Routledge, Kirk, G. S., and J. E. Raven. The Presocratic Philosophers. Cambridge, UK:
2014. Cambridge University Press, 1962.
Cicero, Marcus Tullius. On the Nature of the Gods. Translated by P. G. Lévi-Strauss, Claude. “The Structural Study of Myth.” Journal of American
Walsh. New York: Oxford University Press, 1997. Folklore 68, no. 270 (1955): 428–444.
Fales, Evan. “Mystical Experience as Evidence.” International Journal for Quinn, Philip L. “Towards Thinner Theologies.” International Journal for
Philosophy of Religion 40, no. 1 (1996): 19–46. Philosophy of Religion 38, no. 1/3 (1995): 145–164.
Fenech, Louis E. The Sikh Ẓafar-nāmah of Guru Gobind Singh: A Discursive Russell, Bertrand. Religion and Science. London. Oxford University Press,
Blade in the Heart of the Mughal Empire. New York: Oxford University 1935.
Press, 2013.
Saxe, John Godfrey. “The Blind Men and the Elephant.” CCPA Monitor 14,
Flew, Antony. “Theology and Falsification.” In Reason and Responsibility:
no. 1 (2007): 29.
Readings in Some Basic Problems in Philosophy, edited by Joel Feinberg,
48–49. Belmont, CA: Dickenson, 1965. Stace, Walter T. Mysticism and Philosophy. London: Macmillan, 1960.
Harrison, Victoria. “Religious Pluralism.” In Routledge Handbook of Tylor, Edward B. Primitive Culture: Researches into the Development of
Contemporary Philosophy of Religion, edited by Graham Oppy, 257– Mythology, Philosophy, Religion, Language, Art and Custom. 5th ed., 2
269. New York: Routledge, 2015. vols. London: John Murray, 1929.

TOPIC 9
Causation and Sufficient Reason: Theism

Victor M. Salas
Sacred Heart Major Seminary, Detroit, MI
This chapter discusses the principles of causality and sufficient reason and their relation to theism.
A HISTORICAL RECONSTRUCTION
The notions of causality and the principle of sufficient reason have been crucial tools in theistic phi-
losophers’ efforts to justify their claims about the existence of a first cause or being (i.e., God).
Not surprisingly, philosophers—both past and present—have inevitably developed their particular
understanding of causality within the parameters of their own metaphysical (or anti-metaphysical)
paradigms; as a consequence, among philosophers the notion of causality itself enjoys an irreducible
diversity. Historically, the first concerted effort to arrive at a theological object vis-à-vis causal rea-
soning can be found in the Aristotelian corpus, in particular the Physics and Metaphysics. With the
“third entry” of Aristotle into the Latin West during the High Middle Ages (Congar 1968, 59–
61), the thought of the Philosopher, as he was commonly called, formed the backbone of many
theological explanations regarding God’s existence as well as the nature of the divine attributes.
Here, the works of Thomas Aquinas, John Duns Scotus, and William of Ockham—to name only
a few—made serious efforts to frame theological science, or sacra doctrina, in terms of Aristotelian
epistēmē and are of unparalleled historical significance insofar as they reveal the growing influence
of Aristotelian philosophy on the development of Christian doctrine.
Aristotle’s influence on religious thinkers is not exclusive to Christianity, of course, since he had
already had a profound impact on the development of Islamic philosophers’ thought, especially that
of Avicenna, al-Ghazālī, and Averroes (or Ibn Rushd). What is more, the interpretive conflicts that
transpired in the wake of Avicenna and Averroes regarding Aristotelian philosophy—especially
the controversy surrounding the proper subject matter of metaphysics—eventually spilled over
into Christian thinkers’ discussions of the same subject and helped frame their own answers. The
Aristotelian-based arguments for the existence of God that many medieval theologians crafted
are further developed in the second scholasticism that flourished on the Iberian Peninsula, in the
writings of thinkers such as, among many others, Francisco Suárez and John of St. Thomas.
The repudiation of the Aristotelian philosophy of nature and its associated metaphysical super-
structure marked a turning point in the history of philosophy and coincided with the birth of the
Enlightenment. René Descartes, himself a student of Jesuit scholasticism, rejected scholasticism’s
realist epistemological orientation and instead grounded his philosophical project—carried out in
terms of methodical doubt—on a form of representationalism wherein one’s immediate epistemic
object is an idea as opposed to an extramental real being. The question for Descartes, then, was
whether the belief in an objective reality (i.e., an outside world) corresponding to his ideas could
be justified, and if so, how? Whether yes or no, the only access to such a reality—if indeed it
exists—would be through what his ideas could “objectively” represent. Here, the relationship
between formal and objective realities (Descartes 1951, 39) served an important role in Descartes’s
259
effort to demonstrate God’s existence and thereby overcome the

KEY CONCEPTS challenge of solipsism. Given that various ideas are distinct in
terms of their content (i.e., their objective reality), the question
whether the objects represented in those ideas have any intrin-
Accidentally Ordered Causal Series
sic connection among themselves became an acute problem
Causality for subsequent representationalist thinkers, whether rationalist
Cosmological Argument for God’s Existence (e.g., Gottfried Wilhelm Leibniz and Nicolas Malebranche) or
Essentially Ordered Causal Series empiricist (e.g., David Hume). Hume’s skeptical empiricism
is particularly noteworthy both in itself and for its shared
Formal Reality insights with the much earlier, aforementioned Arabic thinker
Infinite Causal Regress al-Ghazālī. Likewise, Hume’s logical outgrowth from Male-
Neo-Aristotelianism branche’s occasionalism and the unparalleled influence that
the Scottish empiricist had on Immanuel Kant are of particular
Neoplatonism significance. With Kant, causation, which Hume had framed
Objective Reality skeptically in terms of the probability associated with the suc-
Ontological Argument for God’s Existence cession of events, could only be preserved if it no longer per-
tained to things in themselves (i.e., the noumena). Rather, cau-
Per Accidens Cause sality was reduced to an a priori category of the understanding
Per Se Cause that is imposed on phenomena to render them intelligible and
Phenomenology thereby generate scientific experience.
Physico-theological Argument for God’s The result of Kant’s so-called Copernican revolution was
Existence that any effort to use causal reasoning to infer the existence
of a transcendent creator-God ultimately fails, succumbing to
Principle of Noncontradiction a dialectical illusion. Apart from a moral postulate, God is
Principle of Sufficient Reason merely a transcendental a priori ideal of speculative reason
that pushes the understanding forward to attain a greater syn-
optic vision of the totality of experience. Yet, as the totality of
all experience is not itself a possible object of experience, any
attempt to hypostasize the transcendental ideal of God simply amounts, once again, to a dialecti-
cal illusion. Accordingly, as Kant argues in his Critique of Pure Reason, all theoretical efforts to
offer a demonstration for God’s existence fail. The upshot of the Kantian project is that, with
respect to God, philosophers are essentially condemned to agnosticism.
This ambivalent attitude toward God remains in force up to the twentieth century, as seen
in the comments of thinkers such as Ludwig Wittgenstein, who concluded his Tractatus Logico-
philosophicus with the claim that, “What we cannot speak about we must pass over in silence” (Witt-
genstein 1971, 151). Similarly, postmodern philosophers, such as Gianni Vattimo, operating in the
wake of the “death of metaphysics” announced by Martin Heidegger, find any metaphysical reasons
for disbelief in God’s existence unconvincing for the reason that metaphysics itself is no longer cred-
ible (Vattimo 2002, 17). In fact, Vattimo tells us that “today the silence of philosophy with respect
to God has no basis in any philosophically relevant principle. For the most part, philosophers pro-
fess by force of habit to be atheists or nonreligious, almost out of inertia” (2002, 86). In essence, if
postmodernity is silent about the issue of causal reasoning, it is because causality functions as a
metaphysical category that, together with metaphysics itself, is no longer credible to postmodern
sensibilities. Such a repudiation of metaphysics has not, however, amounted to a repudiation of
faith or lack of theological interest, as is especially evident in the theological turn that has preoccu-
pied many phenomenologists such as Jean-Luc Marion, Michel Henry, Paul Ricoeur, and even
those not usually associated with the phenomenological movement, such as Slavoj Žižek.
Be that as it may, in what follows, I offer a presentation of what is at stake in the various forms of
causal reasoning just sketched, especially as they pertain to theistic concerns. I offer a historical recon-
struction of the topic and discuss how, despite attempts to put an end to causal reasoning as a means
of arriving at God, certain philosophical efforts—namely, those emerging out of the Aristotelian tra-
dition very broadly construed—aimed at offering a coherent commitment to causal reasoning with
respect to justifying belief in God’s existence remain viable. I show that, though philosophers and
theologians have consistently referred to causality in their theorizing, the very notion of causality itself
has always been couched within the particular parameters of their own divergent philosophical para-
digms and the exigencies that pertain to them. The consequence is that—as in most if not all things
in philosophy—“causality” is said in many ways. As such, the growing skepticism toward the explan-
atory power of causal reasoning that one finds in modernity is not necessarily to the point when con-
sidering premodern accounts. If terms and ideas within the history of philosophy—chief among
which is causality—are said in many ways, then the history of philosophy may rightly be regarded
as the history of converging equivocations in which philosophers, despite their earnestness, often
speak past one another.
This history of philosophy has itself borne out this reality. For instance, speaking of causality as
the ground of our experiences of “matters of fact,” Hume tells us, “This part of philosophy, it is
observable, has been little cultivated, either by the ancients or moderns” (Hume 1993, 16). One is
flabbergasted by such a claim if one considers that less than two hundred years earlier, Suárez had
devoted sixteen (12–27) of his fifty-four metaphysical disputations (well over 500 double-columned
pages) to the topic of causality. And Suárez is just one example of such a discussion within the his-
tory of philosophy. It may well be the case, then, that philosophers have always had a tendency to
speak past one another. That tendency notwithstanding, every conversation, much like causality
itself, has its starting point, and the beginning of the conversation regarding the nature of causality
can—without too much violence done to the history of philosophy—be said to begin with Aristotle.
ARISTOTELIAN PHYSICS AND METAPHYSICS
Aristotle’s discussion of the nature of causality emerges first in his Physics, where, together with the
three principles of change (matter, form, and privation), he identifies what is necessary and suffi-
cient for motion. Whereas the three principles of change are necessary antecedent conditions for
the possibility of change or motion, they are not sufficient. In addition, some sort of causal principle
or set of principles must be at play. Here, he gives his famous account of the four-cause doctrine.
Beyond causality’s relationship to physics, causality functions at the very heart of the Aristotelian
notion of human understanding. All human beings by nature desire to know, he tells us at the very
beginning of the Metaphysics. Scientific inquiry, especially metaphysics, simply aims to satisfy the
proper and basic human ambition of understanding. What is more, science (epistēmē), for Aristotle,
is none other than grasping something in terms of its ultimate cause or causes (Aristotle 1941, 218).
In science’s syllogistic structure, the middle term of a demonstration not only links the extremes of
the major and minor premises, it serves as the very explanation for their connection and functions as
the cause of one’s knowledge (Aristotle 1941, 123). The realization of one’s nature as a knower
would be frustrated, on Aristotle’s view, without access to causal reasoning.
As already mentioned, Aristotle teaches that causality is fourfold and enumerated in terms of
material, formal, efficient, and final causes. Each of these causes is distinct and operates within its
own sphere, but they coalesce to bring about the very entity of a thing. In natural substances, as
opposed to artifacts, formal, efficient, and final causes coincide. That is, the formal cause not only
determines the nature or kind of thing that something is through the actualization of it substanding
matter, it also determines the scope of powers with which a substance can act. A human being, for
example, precisely because of his or her nature as a human determined by the formal cause, is able
to exercise acts of rationality in addition to the other actions shared in common with sensitive and
vegetal life. Nevertheless, rational activity is uniquely proper to human nature (Aristotle 1941, 943).
But, as every action aims at some end (Aristotle 1941, 935), it must be the case that one’s properly
human actions aim at a rational end that is coordinate with human nature. For Aristotle, that ratio-
nal end is simply contemplation (theorein), and it is in contemplating that one fully realizes or actua-
lizes one’s own human nature. Indeed, the end (final cause) of any substance is simply the full actu-
alization or unfolding (efficient cause) of its nature (formal cause).
Beyond the consideration of particular substances, nature as a whole manifests a certain order
and telos that gives rise to the idea of an organizing principle or ultimate end. In short, the natural

world, it seems, points to some first being. Aristotle approaches this topic in two principal texts:
Physics, books 7 and 8, and Metaphysics, book 12. He begins his argument with the claim that
“Everything that is in motion must be moved by something” (Aristotle 1941, 340). The question
is: is that “something” (a) distinct from or (b) identical with the object that is moved? A clear exam-
ple of (b) would be the sort of “self-movement” that is found among most living beings. Aristotle
holds that even self-movement is not really an instance of something’s moving itself. As he under-
stands it, for something truly and properly to move itself, that thing would have to move itself in
virtue of the totality of what it is or per se.
One may consider, for example, the movement of a cat in pursuit of a mouse. The activity of
the mouse triggers the various sensory apparatus of the cat, which communicate nerve impulses to
the cat’s brain. The cat’s predatory instincts then direct the cat’s movements. Nerve impulses are
sent from the brain down the spinal column to the various muscle groups in the cat’s limbs; the
nerve impulses (through the mechanism of a sodium-potassium pump within the cat’s musculature)
contract the muscles around the cat’s skeletal structure; that skeletal structure moves and supports
the cat’s body as it thereby pursues its target, the mouse. In essence, what has been described here
is a causal sequence or chain of multiple movements and different motors. At no time is the cat
moving in virtue of the totality of what it itself is, namely, a cat. Rather, the cat is moved acciden-
tally, which is to say, only in virtue of its parts. Aristotle offers an example of what he means by
considering a two-part thing composed of X and Y. If the whole XY were to move itself primarily
and precisely as a whole, then both X and Y would have to be in motion. If, however, Y moves
X, then XY as a whole is not in motion primarily or in virtue of itself. The whole is in motion,
yes, but only accidentally in virtue of one of its parts (Aristotle 1941, 341).
What is more, it is crucial for Aristotle to maintain a difference between a whole moving itself
per se in virtue of itself, on one hand, and, on the other, the whole’s being moved only accidentally
in virtue of one of its parts. Aristotle understands motion fundamentally as the actualization of some
kind of potentiality. Thus, if something is to move, potency, as noted above, must be a precondi-
tion since nothing can become what it already (i.e., actually) is. If it were the case that something
has a potency to move in virtue of itself, it would be simultaneously in act (being) and potency
(nonbeing) at the same time and the same respect, which amounts to a contradiction. Claiming that
everything that is moved is moved by another avoids such a contradiction. If something in motion,
M, is moved by something, P, and if P is also in motion, then there must be yet some other thing,
Q, responsible for moving P. If Q, however, is also moved, there must be something else, R, in vir-
tue of which Q is moved. This sequence of moved movers cannot go on infinitely because then
there would be no cause and thus no explanation of the original motion of M.
We see, then, that causality is meant to provide an explanation of some phenomenon (here,
motion) without which that phenomenon would simply remain unintelligible. But if (as Aristotle
holds) human beings by their nature desire to know (Aristotle 1941, 689), then the lack of a causal
explanation would ultimately frustrate that desire and submerse human existence into absurdity.
While this may have been an acceptable concession for the existentialist Jean-Paul Sartre or the
absurdist Albert Camus, the tension between the embrace of meaninglessness and the phenomenon
of actual intelligibility—whether it be scientific, linguistic, social, or moral, to say nothing of meta-
physics—suggests that absurdity is not necessarily a foregone conclusion. For Aristotle, at least, phil-
osophical inquiry escapes absurdity when, through causal reasoning in the face of an experience that
cannot account for itself, one infers the existence of a first mover (Aristotle 1941, 340–341). This
causal reasoning is more than mere wish fulfillment, for it is predicated on reason’s probing of expe-
rience in a way that prompts, through insight, the disclosure of intelligible patterns latent within
experience (Aristotle 1941, 184–186). It is precisely a reality that is experienced as moving or
changing that both challenges the understanding of meaning and leads to the encounter of that
which is unmoving and cannot be moved. Reason, at last, simultaneously satisfies its quest for
understanding and also realizes its nature as inquiring. Aristotle rejects the idea that there could
be an infinite series of moved movers for the reason that, once again, no explanation of the original
phenomenon (the actual movement of some object M) is rendered. He formulates his rejection of
an infinite causal regress on the basis of two ideas.
First, Aristotle understands motion to be not merely a temporal or sequential affair but a matter
of the simultaneity of the mover’s action with the moved thing’s reaction. (As we shall see, this
becomes a key feature of an essentially ordered causal series, as Duns Scotus will later argue.) Second,
Aristotle thinks that any moved thing is divisible (Aristotle 1941, 340). This follows from the fact that
nothing can move itself essentially as a whole. For something to be in motion it must be so in virtue
of its parts. That is, if some object AB were in motion, it would be either because A moves B or B
moves A, which is to say, AB is divisible into constituent parts (Aristotle 1941, 340). What is more,
if an infinite number of movers (bodies) be posited—contrary to Aristotle’s thesis—then, absurdly, to
move a single body in a finite time, an infinite number of bodies would have to move in a finite time
(Aristotle 1941, 340). Aristotle points out that, while it is not absurd to think that “in a finite time
there may be an infinite motion … of many” (Aristotle 1941, 341), nevertheless insofar as mover
and moved must be in contact in their simultaneous motion, they form one infinite continuous
mobile thing. In their moving, “an infinite motion is passed through in a finite time: and whether
the magnitude in question is finite or infinite this is in either case impossible” (Aristotle 1941, 342).
With these arguments in place, Aristotle goes on to conclude that the first mover must be
unmoved since it can have no admixture of potentiality, which is itself the condition for the possi-
bility of change or motion (Aristotle 1941, 373). Also, since the motion of the universe—for which
the unmoved mover is responsible—is, like the universe, eternal, so must the unmoved mover be
eternal (Aristotle 1941, 374).
In essence, Aristotle’s demonstration for an unmoved mover’s existence as portrayed in the
Physics trades on what he has described as efficient causation. In his Metaphysics he offers yet another
demonstration for a first cause, but this time he argues in terms of final causality. Aristotle begins
his argument in the twelfth book of the Metaphysics with a division of substances into three kinds:
(1) eternal sensible substances (i.e., celestial bodies), (2) perishable sensible substances (i.e., the nat-
ural beings of our terrestrial experience), and (3) immovable substances (Aristotle 1941, 872). The
existence of immovable separate substances (the gods) must be conceded, he thinks, in order to
explain the coming-to-be that properly pertains to the first two kinds of substance. Indeed, we are
told, “substances are the first of existing things, and if they are all destructible, all things are destruc-
tible” (Aristotle 1941, 877). To say that sensible things are destructible is logically equivalent to say-
ing that they are susceptible of change (Aristotle 1941, 872).
If they are changeable, however, then, given Aristotle’s results from his inquiry into the princi-
ples of change, such substances are changeable because of their underlying matter. That matter can
be understood in two ways, each corresponding to our experience of material reality. There is, first,
our experience of celestial bodies and their movements, which are imperishable and susceptible to
change only in space. Second, there is our common experience of terrestrial bodies that are perish-
able and endure generation and corruption according to the cycle of nature’s seasons (Aristotle
1941, 873). Though Aristotle understands the material world in terms of its hylomorphic composi-
tion (i.e., a composition of matter and form), while the material substances in the word come into
and out of being, neither matter nor form taken singularly is subject to generation and corruption
(Aristotle 1941, 873). Rather, it is the whole—through the agency of motion—that comes into
being, for “everything that changes is something and is changed by something and into something”
(Aristotle 1941, 873). If it were the case that motion was not eternal, then nothing would exist, for
the being of sensible reality consists precisely in its coming-to-be. But sensible substances clearly do
exist, and thus it is necessary to infer both that the motion whereby they come to be is eternal and,
not insignificantly, that a prime mover exists.
Aristotle argues for the eternity of motion on the basis of time, which “is either the same thing
as movement or an attribute of movement” (Aristotle 1941, 877). In other words, where there is
time there is always movement—for time is just a measure of motion—and where there is no
motion there is no measure, or time. Thus, were one to deny the eternity of motion, the implication
would be that time has either a beginning or an end. To admit a beginning of motion or an end
would, however, permit one to inquire into a before-time-began and an after-time-concluded. Yet,
“before” and “after” are temporal indices, which is simply to speak of time and, along with it,
motion (Aristotle 1941, 877).

Given that motion is eternal and cannot account for itself, it must follow, concludes Aristotle,
that there is some mover, itself equally eternal, capable of accounting for that motion. What is
more, such a mover cannot in principle be subject to motion since then it would simply form a
member of the series of movers. Were the mover responsible for eternal motion, only a mover in
potentia to motion without actually moving things, then its motion would not be necessary. Because
that motion would not be necessary (Aristotle 1941, 877), it would not be eternal “since that which
is potentially may possibly not be” (Aristotle 1941, 878). Transcending the entire series of moved
movers, the unmoved mover is unique in that it is constituted as a pure act through its own nature.
What is more, since that act simply is without beginning, duration, and/or end, the unmoved
mover is simply eternal (Aristotle 1941, 878).
Unlike the account that Aristotle offers in his Physics, his portrayal of the motion that the
unmoved mover initiates in the Metaphysics is cast in terms of final causality. Given that the
unmoved mover moves without itself being moved, the unmoved mover moves precisely as an
object of desire (Aristotle 1941, 879). Aristotle understands the relationship between the object of
desire and that which it moves in terms of the relationship between act and potency. To say that
an object of desire, X, “moves” Y, is to say that y is rendered actual from prior potentiality for the
sake of attaining X. The quintessential instance of motion begotten by desire is celestial motion,
which, according to Aristotle, is the most perfect kind of spatial motion (viz., circular) since it has
neither beginning nor end (Aristotle 1941, 880). In fact, such motion more perfectly approximates
eternity than rectilinear straight-line motion on account of the fact that the latter has a beginning
and an end. Be that as it may, while celestial motion approximates eternity through its circular
motion, one cannot neglect the fact that there is a variety of distinct celestial motions, which, in
turn, demands a plurality of movers. Aristotle explains that “there must be substances which are
of the same number as the movements of the stars, and in their nature eternal, and in themselves
unmovable” (Aristotle 1941, 882). Given the number of motions that constitute one’s astronomical
experience, Aristotle surmises that there are between forty-seven and fifty-five celestial spheres,
meaning there are as many distinct movers responsible for each (Aristotle 1941, 883).
If the unmoved movers are pure act with no admixture of potentiality, one may inquire what
specific kind of act constitutes the unmoved movers or, what is the same, the gods. Aristotle tells
us in both his Metaphysics and Nicomachean Ethics that the noblest activity that humans carry out,
and also that is most godlike, is thought (Aristotle 1941, 1108). Furthermore, it should be noted
that thought is intentional or object-directed, which raises yet a further question: what do the gods
think about? It is incommensurate with the gods’ dignity that they should think of “nothing,” for
then the gods would be no different from one sleeping (Aristotle 1941, 1099). What is more, Aris-
totle thinks it impossible for the object of the gods’ thoughts to be something other than them-
selves, for they would then depend on that thing for their knowledge and stand in a state of poten-
tiality that is incongruous with their natures as characterized by pure act. The only remaining
option is that the gods think about themselves, which is to say, they are thought thinking thought.
“Since, then, thought and the object of thought are not different in the case of things that have not
matter, the divine thought and its object will be the same, i.e., the thinking will be one with the
object of its thought” (Aristotle 1941, 1099).
THE MEDIEVAL TRADITION
As is well known, the full extent of the Aristotelian corpus was not available to Latin thinkers in the
Christian West until the thirteenth century. What did survive, at least in the early Middle
Ages, were Aristotle’s Organon, or logical works, and some portions of his naturalia. While the
Neoplatonic philosophy that dominated the era can just as properly be identified as a form of
Neo-Aristotelianism, the hierarchical and emanation scheme that functioned at the heart of many
Neoplatonic authors such as Plotinus, Proclus, Pseudo-Dionysius, and John Scotus Eriugena, to
name only a few, tilted the scales in the direction of a metaphysical Neoplatonism. The Liber de
causis (The Book of Causes) fit such a metaphysical perspective well and exercised a great deal of
influence during the High Middle Ages. Coupled with the fact that the Liber de causis was
spuriously attributed to Aristotle, the Neoplatonic perspective was reinforced within the various
metaphysical schemas found throughout the early Middle Ages. The exact author of the work
remains unknown, but it is traced back to ninth-century Baghdad, where it was originally composed
in Arabic (Anonymous 1984, 5). Only after its twelfth-century translation into Latin did the Liber
de causis become accessible to Latin thinkers, who regarded it as Aristotle’s completion of his meta-
physical project. While the causal relations within Aristotle’s Physics and Metaphysics seem horizontal
in the one-to-one relation of causes and their effects, the view of causality presented in the Liber de
causis can be regarded as vertical, as it were, wherein a cause enjoys a greater degree of reality than its
effect to which it emanates or flows. For Latin thinkers such as Albertus Magnus, the difference
between these two schemes generated a basic twofold understanding of causality, namely, causere,
which Albertus regarded as obtaining univocally among beings (Albertus 1964, 95), and fluere,
which obtains among beings within diverse hierarchical orders (Albertus 1993, 42). While Aristo-
tle’s Metaphysics gives a proper account of causere (i.e., causation), Aristotle’s alleged Liber de causis
properly portrays God’s agency in terms of fluere (i.e., flux) and creation as a kind of emanation.
THOMAS AQUINAS
What the Liber de causis had essentially portrayed was a Neoplatonic account of causation cast in
terms of emanation or flux. While one can argue that there was a corresponding causal structure
in the Aristotelian scheme cast in terms of formal causality, the Liber de causis nevertheless repre-
sents a metaphysical vision that is alien to the “Philosopher.” Not surprisingly, Thomas Aquinas,
who, like Albertus, had authored his own commentary on the Liber de causis and who was also thor-
oughly acquainted with the work of Aristotle, was the first to doubt its authenticity. There can be
no doubt that Thomas integrated a great deal of Neoplatonic thought into his own metaphysical
vision (such as the notion of participation and a hierarchical metaphysical vision), but the architec-
tonic structure governing his theological project is fundamentally Aristotelian in method. The Poste-
rior Analytics forms the methodological structure in terms of which sacra doctrina was to be fash-
ioned for Thomas into a scientific theology. As already noted, for Aristotle, a science arrives at
knowledge (epistēmē) of its object through causal reasoning, but such reasoning can occur in two
ways: by reasoning from the fact (hoti) or by a reasoned fact (dihoti). Thomas follows Aristotle
and transposes the discussion into the scholastic Latin terminology of a science that involves either
a propter quid (dihoti) demonstration or a quia (hoti) demonstration. A propter quid demonstration
constitutes science in its proper sense and begins with knowledge of a cause from which one can
determine necessarily what effects follow from that cause and why. Thomas thinks that, on this side
of paradise, homo viator is unable to grasp the divine essence as it is in itself and so one must be con-
tent with a quia demonstration, which reasons from God’s effects (viz., creatures) to an oblique
knowledge of the existence and attributes of God, the creator-cause.
Beyond establishing the structural framework of Thomas’s theological vision, Aristotle’s own
causal approach to a first cause informs much of Thomas’s project. The exact role that Aristotle’s argu-
ment plays in Thomas’s thought remains a point of scholarly controversy. One could argue that
Thomas does little more than reproduce Aristotle’s argument for God’s existence. Some, however,
argue that while the structure of the argument is Aristotelian in tenor, the superstructures of each argu-
ment are radically reframed according to Thomas’s overarching and unique metaphysics of esse (Owens
1980). Whether conducted merely in synchrony with Aristotle’s methodology or utilizing Aristotelian
concepts to express a foreign paradigm, what cannot be doubted is that Thomas utilizes causal reason-
ing in the same manner as Aristotle to arrive at his own conclusions. The Dominican develops a ver-
sion of the proof from motion within his Summa contra gentiles (bk. 1, chap. 13), but, by far, Tho-
mas’s most famous expression of a causal explanation of the divine origin of the universe is found in
his quinquae viae, or five ways, located in the second question of the first part of the Summa theologiae.
For Thomas, much as had been the case with Aristotle, the causal demonstration for God’s
existence capitalizes on two phenomena: (1) the actual existence of the world and (2) the recogni-
tion of the insufficiency of the world with respect to its very existence. That the world actually exists
is not in doubt for Thomas or any other medieval (realist) metaphysician. Accordingly, the principal
challenge for the theistic metaphysician is to manifest the nature of the world’s insufficiency. For

Thomas, it is the task of each of the five ways to expose those metaphysical deficiencies. Among the
five ways, he attributes a certain efficacy to the first, which he describes as the “manifestior via”
(more manifest way). The prima via addresses the phenomenon of motion and reveals how motion,
precisely as such, is incapable of accounting for itself. As Thomas understands it, motion is simply
the reduction of some potentiality into act. He thus understands motion much more broadly than
locomotion or the communication of force that would be characteristic of Newtonian mechanics
or the physical accounts of the same that follow thereon. A log’s “movement” from being potentially
hot (while being actually cold) through the agency of some mover (viz., something actually hot)
serves as Thomas’s example, but other changes (not ordinarily identified as instances of motion)
can also be adduced. For instance, any “alteration” (e.g., the change in color that leaves undergo
in the fall or a person’s attainment of some new knowledge) can satisfy Thomas’s definition of
motion (cf. Wippel 2000, 445). In pointing to motion in terms of the dynamics between potency
and act, Thomas is able to identify the precise insufficiency at the heart of motion. Something Y
that is in potency with respect to being X cannot bring itself actually to be X, for to suppose that
Y can bring itself actually to be X would involve a contradiction. That is, Y would be (act) and
not be (potency) X at the same time and in the same respect. Accordingly, Thomas concludes that
“nothing can move itself” or, cast positively, “everything in motion is put in motion by another.”
The claim that nothing can move itself might seem counterintuitive, given one’s ordinary experi-
ence of self-initiated movement in the natural world, and so it is hardly surprising that this phase of
Thomas’s argument has been challenged by both his immediate successors and more recent thinkers.
Suárez, for example, is critical of Thomas’s claim that “omne quod movetur ab alio movetur” (everything
moved is moved by another) since the Jesuit thinks that it is not sufficiently demonstrated of every
category of movement—for example, living things’ acting through their appetites, water’s self-cooling
from a heated disposition, the downward motion of a stone in free fall, and so on (Suárez 1861,
29.1.7). For this reason, in his own effort to demonstrate God’s existence, Suárez substitutes Thomas’s
claim that “omne quod movetur ab alio movetur” with the claim that “omne quod fit ab alio fit” (every-
thing that is made is made by another) (Suárez 1861, 29.1.20). Even Thomas’s faithful Dominican
commentator, Cajetan, thought that the Angelic Doctor’s proof reached only the existence of an
“intellective soul” and therefore required further completion in order to arrive at the existence of
God Himself (Owens 1980, 160). Beyond Thomas’s medieval and scholastic successors, the emerging
physics of the Enlightenment seems to cast doubt on the credibility of Thomas’s claim that nothing
moves itself. For instance, the first Newtonian law stating that an object in motion will stay in motion
unless acted on by a contrary force suggests that motion does not require continued impetus.
Thomas does not give any consideration to the above-mentioned objections within his Summa
theologiae, but he does address the issue of self-motion in his much lengthier proof for God’s exis-
tence in Summa contra gentiles (bk. 1, chap. 13). Contemporary Thomists, such as Jacques Maritain,
however, do give consideration to the kinds of objections mentioned above and offer responses to a
wide variety of objections stemming from recent developments in physics and quantum mechanics.
Regarding self-motion, in his Summa contra gentiles Thomas, like Aristotle before him, appeals to
the notion that, for something to move itself, it would have to move itself precisely as a whole
and not accidentally or in virtue only of some part. For this reason, even animal motion would still,
on Thomas’s view, cohere with this account since an animal is moved by its appetites through its
perception of some apparent good. As we saw with Aristotle’s proof for a prime mover in Metaphys-
ics, the efficient activities of some agent X initiating motion are a consequence of its efforts to satisfy
its appetites in the attainment of an object. That object exerts final causality on X. Within the con-
text of human operations, especially acts of volition, diverse orders of causality are also at play.
While it is the case that the will efficiently moves other powers of the soul, it too is moved by
the intellect (as well as the senses). Here, Thomas maintains the same point he had argued in his
prima via, namely, that motion involves the reduction of potentiality into act by something already
in act (ST 1a2ae, q. 9, a. 1). In the will’s acting, Thomas thinks it remains the case that the will is
moved by an agent, namely, the desirable object presented by the intellect. An object is desirable
insofar as it has the character of being good, but the good moves according to final causality. John
Wippel observes that, for Thomas, “it would still be true that neither the intellect nor the will
would move and be moved under the same aspect and in the same respect” (2000, 449).
Regarding impetus, two things need to be said. First, even Newton himself conceded that,
though no agent is needed to account for an object’s maintaining motion, some mover is required
for the initiation of motion. “The Vis inertiae is a passive principle by which bodies persist in their
motion or rest, receive motion in proportion to the force impressing it, and resist as much as they
are resisted. By this principle alone there never could have been any motion in the world” (Newton
1721, 372–373). Yet there is motion in the world, which means, as Newton concedes, that inertia
is not a sufficient explanation for that reality. Second, Thomas construes motion as the communi-
cation of existential perfection, which is alien to the nature of the moved object. This metaphysical
dimension is absent from Newton’s considerations, which are cast purely in the mechanical terms of
the communication of force.
It is this underlying metaphysical situation that Maritain thinks forms the proper response to
the inertia objection. Allowing that inertia is established scientifically beyond hypothetical assump-
tions, Maritain argues that Thomas’s claim “everything which moves is moved by another” is logi-
cally equivalent to the claim “Every body which undergoes a change in regard to its state of rest or
of motion changes under the action of another thing” (Maritain 1954, 32). So understood, it is
not motion per se that is considered but its state. That state itself cannot change—that is to say,
come to exist in a different manner—without the causal influence of some agent. Maritain rephrases
Thomas’s initial claim so as to read: “Every change is produced by the agency of something other
than the thing which changes, insofar as it changes” (Maritain 1954, 33). What is more, Maritain
also points out that, in Albert Einstein’s dynamic theory of accelerating motion wherein motion
itself is nonstatic, it still holds that a cause would be required for any acceleration (Maritain 1954,
33, n. 5). Such being the case, Thomas’s claim that “everything that moves is moved by another”
can be maintained consistently in light of developments in modern physics. If that claim still holds
true, it is because Thomas understands motion—broadly construed—as metaphysically insufficient
to account for itself. Thus, though appealing to a phenomenon that is thoroughly treated in the nat-
ural sciences, specifically, physics, Thomas does not draw a conclusion that is itself proper to the
natural sciences. Rather, he infers a metaphysical conclusion: potentiality is not a sufficient explana-
tion for act. The metaphysical aspect of the first way emerges again in the second phase of Thomas’s
argument, as we shall presently see.
The second phase of Thomas’s argument unfolds as follows. If nothing moves itself, given the
existence of motion, there must be an agent or mover responsible for the motion that one encounters
in experience. Thus, if X is in motion, it must be placed in motion by another, Y. If it is the case that
Y is also moved, then it too must be placed in motion by another, Z. Like Aristotle, Thomas denies
that an infinite regress of moved-movers is possible since there would then be nothing to account for
the initial movement of X. It is at this point, as some argue (te Velde 2006, 54–55), that Thomas’s
argument transitions from a physical argument—arguing on the basis of motion—to a metaphysical
argument. Rudi te Velde explains, “If all members of the class of things-in-motion must be explained
by reference to something else, and if outside this class there is nothing, then the existence of motion
would be unintelligible” (Velde 2006, 50). If being-in-motion cannot account for itself, then one
must either (a) advert to that which is just Being itself (God) in order to account for the existence
of being-in-motion or (b) embrace the absurdity of motion’s existence. Whereas Sartre’s character
Antoine Roquentin reacts with existential nausea to being’s superfluity, describing it as “de trop”
(too much), Thomas perceives being as an undeserved gift that signals the beneficence of a creator-
God. He phrases this insight according to scholastic categories: “Therefore it is necessary to arrive
at some first mover [primum movens], which is moved by nothing, and this all understand [to be]
God” (ST 1a, q. 2, a. 3). Thomas makes similar comments at the end of the other five ways. The
second way, arguing from efficient causality, concludes: “Therefore it is necessary to posit some first,
efficient cause, which all call God.” The third way arrives at a necessary being, “which all call God”;
the fourth way concludes to a most perfect being responsible for all creaturely perfections, and this
“we call God”; and, finally, the fifth way argues from the governance of the universe and concludes
to the existence of an intelligence that directs the world, and this “we call God.”
Needless to say, it seems a stretch to identify an “unmoved mover,” “uncaused cause,” and so
on with anything associated with the Christian tradition’s understanding of God to which Thomas

was committed. Within modernity, the God in question here would become known as the “god of
metaphysics,” which “god” Friedrich Nietzsche announces is dead in The Gay Science. Nevertheless,
it is not without reason that Thomas can so unproblematically assert that the God attained at the
conclusion of each of the five ways is the Christian or biblical God. The setting of these five demon-
strations, we recall, is theological in nature. What is more, given that an Aristotelian science always
unfolds by asking two questions—(1) is it? and (2) what is it?—the operative paradigm for Thomas
is a theological narrative, the intelligibility of which is at issue. Further establishing the theological
setting is what Thomas presents in his sed contra, which refers to Exodus 3:14. In that particular
biblical passage, God announces to Moses “I am who am.” Already well established within the
medieval tradition stretching back to Augustine of Hippo was the understanding of God as Being
itself. In the question immediately following the five ways (q. 3, a. 4), Thomas is explicit that
God is “ipsum esse subsistens” (subsisting being itself). Moreover, that which is attained at the conclu-
sion of each of the five ways is simply pure act, which is to say, Being itself. Thomas can confidently
assert, then, that although scrutinized from a metaphysical or philosophical perspective, it is the bib-
lical God that is at issue, for such philosophical speculation occurs on a theological horizon.
DUNS SCOTUS
Another point about the kind of causal reasoning at issue must be made here. In denying an infinite
causal regress, what Thomas and later thinkers have in mind is what is referred to as an essentially
ordered causal series as opposed to an accidentally ordered causal series. This distinction is implicit in
what we find in the Summa theologiae and is only vaguely alluded to in the demonstration presented
in the Summa contra gentiles (bk. 1, chap. 13). In later medieval philosophy, it would become an
explicit feature of the effort to demonstrate God’s existence. The Franciscan theologian Duns Sco-
tus, for instance, clarifies that while an infinite regress with respect to an accidentally ordered causal
series is possible, such a regress is impossible with respect to an essentially ordered causal series.
Duns Scotus’s own demonstration for God’s existence—at least in some versions of his opera—like-
wise trades upon the causal relationship between creation and creator, and so Duns Scotus also
deploys a quia demonstration. His own argument is as follows. Some being, X, is an effect since it
is produced. Now, it must be the case that X (1) produces itself, (2) is produced by nothing, or
(3) is produced by another, Y. Neither (1) nor (2) can be the case, Duns Scotus thinks, since noth-
ing can cause itself nor can nothing serve as a source of being. Thus, X must be produced by
another, Y. If it is the case that Y is itself produced, then it too must be produced by something else,
Z. But, as this cannot go onto infinity, it is necessary to arrive at some first being that is not pro-
duced and is able to produce other things in virtue of its own power (Frank and Wolter 1995, 43).
The parallel here between Duns Scotus’s argument and that of Thomas Aquinas is clear, for
they both utilize causal relations and deny an infinite regress. Whereas Thomas rejects an infinite
causal regress outright, Duns Scotus carefully investigates the philosophical concerns at stake. He
meets an objection to his denial of an infinite causal regress: it is not impossible that productions
enjoying the same character (ratio) should be able to proceed to infinity. For example, it is not
impossible that the generation of human offspring, wherein the child (effect) and parent (cause)
enjoy the same nature (ratio), should proceed to infinity. One might also consider the ignition of
one flame from another prior flame that itself is ignited by yet another prior flame, and so on.
The objection, consequently, seeks to undercut one’s ability to infer the existence of a first produc-
tive cause since an infinite series of causes would seem to be both possible and intelligible.
It is in responding to this objection that Duns Scotus makes explicit the difference between essen-
tially ordered and accidentally ordered causal series. While the latter can admit of an infinite regress,
the former cannot. But, when arguing for the existence of a first being (God), Duns Scotus claims
to make recourse to an essentially ordered causal series. Needless to say, the difference and the rela-
tionship between the two causal series must be established. Before commenting on the nature of the
two kinds of causal series, Duns Scotus clarifies that what is at issue is not the difference between a
per se cause and per accidens cause. As he says, both per se and per accidens causes have a one-to-one
(unius ad unum) relation to their effects such that the per se cause operates on its effect in virtue of
its proper nature. For example, medicine cures a patient in virtue of its chemical content and dosage,
not in virtue of its color, the shape of a pill, its bitter taste, and so on. The latter attributes are all acci-
dental to the proper character of the medicine (Frank and Wolter 1995, 45).
With respect to essentially ordered and accidentally ordered causal series, however, the one-to-one
relationship is displaced by the cooperation of two or more causes operating in concert to bring about
one and the same effect (Frank and Wolter 1995, 45). In such a scenario, an essentially ordered causal
series differs from an accidentally ordered series in three ways: dependency, perfection, and simultaneity.
With respect to dependence, one may consider two causes, M and N, in relation to their effect, O.
Cause N is said to be essentially ordered to M if in order to exercise its proper causal nature and thereby
produce its effect (O), N depends on M precisely as a cause. This is different with respect to accidentally
ordered causes, as Duns Scotus points out using the father-son example. While a son may depend on his
father for the very begetting of his own existence, the son does not depend on his father in exercising his
(the son’s) own causal (reproductive) agencies (Frank and Wolter 1995, 45).
With respect to perfection, again given two causes, M and N, if it is the case that one cause, N,
can exercise its own proper causality only in virtue of M, then Duns Scotus thinks that it can be said
that M is more perfect than N. The reason for this is that, whereas N stands in need of M in order
to be a cause, M does not enjoy such a symmetrical dependence on N to exercise its own causality.
Such relations of dependence and the resulting degrees of perfection do not obtain with accidentally
ordered causes. Again, Duns Scotus gives the father-son example. A son can procreate in the same
manner as his father and does not depend on the latter in order to exercise procreative causality.
What is more, with respect to their nature as causes, one cannot be said to be more perfect than
the other (Frank and Wolter 1995, 45).
With respect to the final difference, Duns Scotus tells us that essentially ordered causes exercise
their causality simultaneously or synchronously (Frank and Wolter 1995, 47). One may consider,
here, the reflection of sunlight via a mirror on some object, Q. Taken on its own, the mirror is inad-
equate or insufficient to account for the effect, namely, the illumination of Q. The luminous agency
of the sun must operate simultaneously with the transmitting of light by the mirror in order to pro-
duce the illumination of Q. Such simultaneity, however, is not required for an accidentally ordered
causal series, wherein causes exercise their causality successively. As an accidentally ordered causal
series one may consider, for example, a row of dominos in which one falls on another successively
without end. It is possible for such an accidentally ordered causal series to stretch infinitely.
The three differences between essentially ordered and accidentally ordered causes, Duns Scotus
thinks, mark how it is impossible for an infinite causal regress to obtain within an essentially ordered
causal series. Can it be the case that each cause within an essentially ordered series is such that it
depends on another? On such a view an infinite number of caused-causes would have to operate
simultaneously to produce some effect. Yet such a conclusion, for Duns Scotus, is absurd since the
infinite sequence of caused-causes is something that is itself dependent, and it will not do to populate
that series with additional causes that are themselves caused. Rather, the sequence must be transcended
to arrive at a first efficient cause. William Frank and Allan Wolter describe Duns Scotus’s account by
marking a difference between “horizontal” and “vertical” lines of causality. A horizontal sequence of
caused causes can increase indefinitely, yet “the fact is that increasing the membership in the sequence
does not overcome the systematic dependency of the set and of each member in it” (Frank and Wolter
1995, 84). To overcome that dependency, the causal sequence is understood as depending on an
entirely different kind of cause, one that stands on a different philosophical plane than within the
indefinite horizontal sequence. To use Frank and Wolter’s example, a hanging chain cannot be
explained by adding ever increasing links higher and higher. Rather, one must arrive at something
beyond the putatively infinite series of links to what is not included within the sequence, for example,
a ceiling beam or a hook (Frank and Wolter 1995, 84).
ENLIGHTENMENT PERSPECTIVES
As can be gathered from the foregoing historico-theoretical considerations, causality for the ancients
and scholastics was fundamentally construed in terms of the communication of being or new

existential perfection. The very being of the things constitutive of one’s ordinary experience presents
itself as a phenomenon that is unable to render an account of itself. For this reason, philosophers
and theologians inferred the existence of what could account for the being of things, which is just
Being itself (viz., God).
DESCARTES AND HIS AFTERMATH

In the transition to the early modern period of philosophy, beginning especially with Descartes, the
Aristotelian causal doctrine underwent gradual but steady erosion. Having rejected the basic Aristo-
telian philosophy of nature—a rejection that had already been fomenting with Galileo and Nicolaus
Copernicus—the notion of causality correlative to the Aristotelian doctrine underwent radical revi-
sion and an eventual overthrow. Descartes does not dispense with all causal reasoning, of course:
“Now it is obvious, according to the light of nature, that there must be at least as much reality in
the total efficient cause as in its effect, for whence can the effect derive its reality, if not from its
cause?” (Descartes 1951, 39). Yet the reality in question for Descartes is not that of beings populat-
ing an extramental world, as had been the case for ancient and medieval philosophers, but instead
that of his ideas. Descartes radically alters the notion of causality when he deploys it to account
for the relation between and among ideas with respect to their objective reality. “In order that an
idea should contain one particular objective reality rather than another, it should no doubt obtain
it from some cause in which there is at least as much formal reality as the idea contains objective
reality” (Descartes 1951, 40). Descartes’s understanding of objective and formal realities has its
remote origin within the scholastic distinction between objective and formal concepts. Transposed
into a Cartesian key, however, the meaning of these scholastic terms is altered in such a way that
objective reality corresponds to the degree of reality that an idea (objectively) expresses. For Des-
cartes, this means that the idea of a substance, for example, has a greater degree of (objective) reality
than the idea of a mode. Nevertheless, from the perspective of their being ideas, both possess the
same degree of formal reality.
In essence, as mentioned above, Descartes moves philosophical investigation away from its real-
ist moorings—wherein one had direct epistemological contact with being—toward a sort of repre-
sentationalism in which one is in contact, instead, with the ideas that represent things. The concern
for Descartes, anxious as he was to avoid skepticism, was to prevent such a representationalism from
lapsing into complete idealism wherein one’s ideas would be entirely severed from reality and leave
one trapped within the immanence of thought itself. That an infinite and good God should exist
was a balm for Descartes’s solipsistic worries, for if such a God existed, then surely God created
Descartes with reliable cognitive faculties. Yet, if the existence of God could be established, it could
only be on the basis of what the idea of God itself represents; and therein lies the difficulty. For
Descartes, it was crucial that he be able to attain such a clarity and distinction among his ideas that
he would be able to reason truthfully to the existence of God.
To secure clear and distinct perception as a criterion of the truth, however, Descartes had to
overcome the hyperbolic doubt that an “evil genius” might exist. He could overcome that doubt
by demonstrating that a good God is in fact the source of his being and, because God is good,
therefore Descartes’s cognitive efficacy is ensured. Famously, Descartes offers two proofs for God’s
existence in his Meditations on First Philosophy. The first—found in the third meditation—attempts
to argue on the basis of the relationship between formal and objective reality; the second—located
in the fifth meditation—offers a version of the ontological argument. In the third meditation, Des-
cartes explains that his idea of God, which is the idea of an infinite substance, contains (infinitely)
more objective reality than Descartes possesses (finite) formal reality. It must be the case, then, that
something enjoying at least as much (infinite) formal reality as the idea of God possesses objective
reality must exist. That would mean, then, that an infinite being (God) must actually exist in order
to account for the existence of the idea of God that Descartes in fact has.
Though this argument has been subjected to severe criticism for its apparent circularity (i.e.,
Descartes appeals to clear and distinct perception in order to ensure that the idea of God is not
materially false, but aims to prove God’s existence in the first place so that he might establish that
clear and distinct perception is a guarantor of truth), what is significant to note is that, unlike his
Aristotelian-minded scholastic predecessors, Descartes does not advert to the existential reality of
contingent finite beings in order to advance to knowledge of a necessary and infinite being’s exis-
tence. For him, the point of departure (and return) is to the mere idea of God and the content of
its objective representation. This is no less true of the argument presented in the fifth meditation.
Just as one cannot turn to the idea of a triangle, for example, and reject the concomitant idea of
three-sidedness without contradiction, one cannot consider the idea of God and reject the idea of
existence without contradiction. The reason for this claim, Descartes thinks, is because just as the
ideas of three-sidedness and three-angled are necessarily contained within the idea of a triangle, so
is the idea of necessary existence contained in the idea of a perfect and infinite being. Were it the
case that necessary existence was lacking in the idea of a perfect being, then what would be con-
ceived would not in fact be a perfect being. Since necessary existence is contained within the very
idea of a perfect being, one cannot think of God without simultaneously thinking of God’s actual
existence. In this case, the very idea of God represents God’s actual existence.
As we shall see, Kant subjects this type of (ontological) argument to severe criticism in his Cri-
tique of Pure Reason. But prior to Kant, another rationalist philosopher, Leibniz, thinks that, in
addition to God’s necessary existence, God’s moral perfection can also be inferred from the mere
idea of God as “an absolutely perfect being” (Leibniz 1993, 3). Leibniz is of course well known
for his espousal of the principle of sufficient reason, which he takes to mean that an effect follows
as a kind of rational necessity from the sheer existence of a cause (Clarke 2001, 181). In his Mon-
adology, Leibniz identifies the principle of sufficient reason, together with the principle of noncon-
tradiction, as the “two greatest principles” on which all reasoning is based. Whereas reason deter-
mines what is false on the basis of contradiction and what is true on the basis of its opposition to
what is false, the principle of sufficient reason is such that “we believe that no fact can be real or
existing and no statement true unless it has a sufficient reason why it should be thus and not other-
wise” (Leibniz 1993, 258).
For his part, Norris Clarke marks the difference between the two principles in terms of the dif-
ference between what is static and dynamic. Whereas the principle of noncontradiction is “static”
and based simply on identity, the principle of sufficient reason, he says, is “dynamic” (Clarke
2001, 20). That is, as Clarke explains, the principle of sufficient reason “enables the mind to pass
from one being to another in the search to make sense out of it, to preserve it from falling into
unintelligibility” (2001, 20). Whether Leibniz’s own take on the principle of sufficient reason yields
a conclusion that is amenable to Clarke’s own speculations is, as the latter notes, not necessarily the
case. For Leibniz seems to argue on the basis of an intrinsic necessary connection between cause,
which, once given, cannot fail but to produce its entailed effect. Thus, given God’s existence as
an absolutely perfect being, Leibniz thinks that God must also be morally perfect and necessarily
produces the best of all possible worlds. For this reason, Étienne Gilson protested against Leibniz’s
theory since it would seem to afford no room for a free creation (Clarke 2001, 21).
Be that as it may, Leibniz is clear: if God is a morally perfect being and a morally perfect being
always acts perfectly, for “one acts imperfectly if he acts with less perfection than he is capable of”
(Leibniz 1993, 5), then, to Leibniz’s mind, one must conclude that the current state of affairs consti-
tutes the best of all possible worlds. While this metaphysical optimism might have been lampooned
by Voltaire’s Candide, it was not without its own host of metaphysical consequences, among which
was that God establishes the world in a completely rational manner. Accordingly, what seems extraor-
dinary to human perspective is, from the perspective of the divine author, thoroughly orderly, for
“God does nothing out of order” (Leibniz 1993, 10). This would seem to collapse the distinction
between the miraculous (i.e., the extra-ordinary) and the natural (i.e., the ordinary): “For it can be
said that this nature is only a custom of God’s which he can change on the occasion of a stronger
reason than that which moved him to use these regulations” (Leibniz 1993, 11).
It had been precisely the concern to open a space for God’s miraculous intervention in the
world that—centuries earlier—had led al-Ghazālī to reject Avicenna’s Aristotelian commitment to
nature as form of necessity. If, as Aristotle suggests, substances act necessarily according to their nat-
ures (e.g., fire burns, light illuminates) and substances are thus limited by those natures (e.g.,
humans cannot fly and fish cannot walk), then divine intervention would seem to require

suspending what is necessary, that is to say, carrying out what is contradictory. To open a space for
miraculous intervention in the world, al-Ghazālī holds that creatures do not exercise any causality of
their own. Nature is not constituted by necessary relations but only by customary connections. “But
[with] any two things,” he writes, “where ‘this’ is not ‘that’ and ‘that’ is not ‘this’ and where neither
the affirmation of the one entails negation of the other, it is not a necessity of the existence of the
one that the other should exist, and it is not a necessity of the nonexistence of the one that the other
should not exist” (al-Ghazālī 2000, 166). The “connection” between the two, then, is not really a
connection at all, for “the connection between what is habitually believed to be a cause and what
is habitually believed to be an effect is not necessary” (al-Ghazālī 2000, 170). Indeed, as the Islamic
thinker sees it, there is no intrinsic causal interaction among things in the world, for “connection is
due to the prior decree of God, who creates them side by side, not to its being necessary in itself,
incapable of separation” (al-Ghazālī 2000, 166). al-Ghazālī, it would seem, offers a version of occa-
sionalism six hundred years before Malebranche.
For his part, Leibniz, while not so much concerned with opening a theoretical space for the
miraculous, agreed with al-Ghazālī and also with Malebranche that God is the sole actor in the
world. Nevertheless, unlike Malebranche, who heaped opprobrium on Aristotle and the schoolmen,
Leibniz views his scholastic predecessors with a certain forbearance: “our moderns do not do suffi-
cient justice to Saint Thomas and to the other great men of that period [for] there is in the theories
of the scholastic philosophers and theologians far more solidity than is imagined” (Leibniz 1993,
18). Leibniz even concedes the existence of substantial forms—an idea Malebranche vehemently
rejects—but thinks the scholastics are entirely wrong to attribute an explanatory role within the
realm of physics to those forms, for “these forms effect no change in the phenomena” (Leibniz
1993, 18). If God is the sole actor in the universe, says Leibniz, it is for the reason that no substance
ever interacts with another. To explain his account, he tells us that the concept of each substance
contains all of the predicates that will ever pertain to it. Imagine a billiard ball described by a variety
of predicates: sphericity, extension, color, weight, and so on. In addition to these predicates, the bil-
liard ball also has others, namely, relations (e.g., next to, behind) and interrelations (e.g., striking
object X, or accelerating toward object Y, or perhaps even shattering a window). As Leibniz under-
stands it, then, each substance contains the reason (e.g., it was because of his predicates that Julius
Caesar crossed the Rubicon [Leibniz 1993, 21]) for its interaction and correspondence with other
substances, but it never causally effects any of them. Leibniz concludes:
It is God alone … who is the cause of this correspondence in their phenomena and who brings
it about that that which is particular to one, is also common to all, otherwise there would be
no relation. In a way, then, we might properly say, although it seems strange, that a particular
substance never acts upon another particular substance nor is it acted upon by it. (1993, 25)
What seems “strange” to Leibniz strikes Malebranche as thoroughly pious and true. Less patient
than Leibniz had been with the “ancients” regarding substantial forms, Malebranche considered
their position a “most dangerous error” (Malebranche 1997, 446). Perhaps bringing Descartes’s
thinking to its ultimate conclusion, Malebranche holds that “no body, large or small, has the
power to move itself” (Malebranche 1997, 448). Indeed, once construed simply in terms of three-
dimensional extension—Descartes’s view of res extensa—how could the idea of motion or commu-
nication of force be derived therefrom? The mind-body problem remained intractable to
the Cartesian system but only because it remained a “problem” in need of solution. For Leibniz’s
preestablished harmony and Malebranche’s occasionalism, there was no problem, for they short-
circuited the issue entirely. No substance or body interacts with any other body whatsoever. All at
once, the Aristotelian doctrine of causality dissolved along with the philosophy of nature it had been
developed in order to explain. If nature, as Aristotle construed it, is an intrinsic principle of opera-
tion or agency, then to do away with nature is to do away with agency itself. This had been precisely
Malebranche’s tactic: “all natural causes are not true causes but only occasional causes” (Malebranche
1997, 448). What is more, if only God is a true cause, then all other putative causes (e.g., the mind
on itself, the mind on the body, or a body on another body) are not truly causes.
If Malebranche had succeeded in driving the final nail in the coffin of the Aristotelian philoso-
phy of nature, then Hume would be the one trying to ensure it stayed buried. Indeed, Hume’s
brand of skeptical empiricism seems to follow entirely as a consequence of occasionalist thinking, for
his critique of causality may be seen as emerging from the “impossibility of proving any transitive
action of one substance upon another” (Gilson 1999, 15) that had been articulated by Leibniz
and Malebranche. Hume was committed to a thoroughgoing empiricism and held that all of our
ideas or concepts are pale copies of our more lively sense impressions (Hume 1993, 11). There is
a difference, for example, between one’s actual experience of a sunset and one’s later recollection
of that experience. The latter is less vivid and paler than the original. All philosophical reasoning
for Hume, then, involves determining the relationship between and among the various ideas one
has attained through sense impressions or experience.
Here, one may well note an apparent point of continuity between Hume, who thinks that all of
one’s ideas derive from experience, and Aristotle, who holds that there is nothing in the intellect
that is not first in the senses. Nevertheless, there is a crucial difference between the two, for Aristotle
thinks that the concept resulting through one’s experience of reality functions as a medium whereby
the thing itself is known. Aristotle is famous for making the claim that the knower becomes the
object known, for the reason that the substantial form determining the object known exists cogni-
tively within the knower and similarly determines his or her intellect. For Hume, in a fashion simi-
lar to another British empiricist, John Locke, in grasping the idea of something one does not
thereby grasp that very thing itself. Hume’s account, like that of Descartes and other thinkers,
may be described as a form of representationalism—as opposed to a direct realism of the Aristotelian
tradition, wherein the object is directly known through the medium of the concept.
For Hume, just as the Cartesian idea of a body contains no true concept of motion, no idea
representing a matter of fact ever implies another. To be sure, Hume thinks one is accustomed to
forming the idea of heat together with the idea of fire, or the idea of communicated motion as a bil-
liard ball strikes another ball, but these conjunctions of ideas are a result of mere experience. Rea-
son, he thinks, cannot discern any intrinsic connection between and among various events. Instead,
it is the similar and repeated experience of a succession of events that produce a psychological habit
or custom in which one anticipates that the future will resemble the past. For example, in every
experience of action one has likewise experienced reaction. It is not the case that reaction can be dis-
cerned a priori; rather, it has always, as a mere matter of fact, followed on action. There would be
no contradiction, however, in denying that reaction will follow on action. One might not have
the belief that reaction will fail to follow on action, but this has nothing to do with what reason
can discern or the relation among one’s ideas. For Hume, then, causality is based entirely on expe-
rience and has nothing to do with reason’s a priori perceptions. If causality were based on experience
and the psychological custom or habit produced thereby, it could not transcend experience. Causal-
ity would then become a matter of the probable, which is simply the perception of a conjunction of
events as opposed to a “connection” of events.
KANT
As Kant himself makes clear, Hume’s critique of the nature of causality provoked a massive shift in
his own philosophical perspective and awoke him from his “dogmatic slumber” (Kant 1950, 8).
Hume’s revision of causality restricted it, as we saw, to the domain of experience and probability
from which one derives the belief that the future will resemble the past. While this preserves Hume
from a complete skepticism, the Scottish philosopher is content to embrace a mitigated form of
skepticism that rejects absolute knowledge for only a probable form of reasoning. Thus, to take
an example from physics, the Newtonian laws, though framed in terms of necessity, can never be
more than a statistical matter of probability given the fact that they are derived from experience.
To the Prussian philosopher’s lights, this approach to physics simply undermined the universal
necessity with which science operates. What is more, Hume’s critique of intrinsic connections
between and among events was even more far-reaching than he himself recognized. For Hume,
mathematical ideas were relations of ideas, whose necessary connection was determined according
to the principle of noncontradiction. Kant was revolutionary, however, in making the argument that
mathematics is composed of synthetic a priori propositions and not analytic propositions. Mathe-
matical propositions are a priori, to be sure, which means the subject and predicate terms of the

propositions stand in a necessary relation to one another. Yet, insofar as the proposition is synthetic,
contrary to Hume’s thinking, that proposition is not determined to be true on the basis of noncon-
tradiction. This situation provoked the question: how is mathematics possible? That it is possible is
not in doubt, Kant thinks, for mathematics is actual. What is more, insofar as physics is likewise
constituted by synthetic a priori propositions, the broader question for Kant was: how are synthetic
a priori propositions in general possible? The answer to that question could serve also as an answer
to the vexed question whether metaphysics itself is possible.
To answer these questions, Kant inaugurates what he describes as a “Copernican revolution” in
philosophy, wherein the relationship between the knower and the object known is inverted such
that the object’s accommodation of cognition’s structures is paramount. Mathematics is possible
because, in addition to empirical intuition, which roughly corresponds to Hume’s notion of a sense
impression, there is a more fundamental a priori intuition, which expresses the a priori forms of sen-
sibility, namely, space and time. With respect to physics, however, Kant turns to yet another a priori
structure of cognition, the understanding. It is the senses’ business to intuit, Kant holds, but it is the
understanding’s role to judge. To judge is simply to bring various intuitions together in a proposi-
tional synthesis. If the synthesis occurs only in one’s own subjective consciousness, then that judg-
ment (of perception) has only subjective validity. If it is the case, however, that the synthesis occurs
according to an a priori rule of the understanding under which one’s intuitions are subsumed, then
the resulting judgment (of experience) has objective and universal validity. The reason for this latter
claim is that the rule for judgment is an a priori structure of the understanding that is universal to
all rational agents. These rules for synthesis, Kant thinks, simply emerge from the logical table of
judgments and can therefore be strictly determined, unlike Aristotle’s own categories, which were
haphazardly enumerated.
Kant advanced the revolutionary claim that causality was precisely one such a priori concept of
the understanding. Corresponding to hypothetical judgments (e.g., if P, then Q) is the a priori con-
cept of cause. To use Kant’s example, to say “when the sun shines on the stone, the stone grows
warm,” is simply a judgment of perception since there is no necessary connection between the ante-
cedent and the consequent. When, however, one says “the sun warms the stone,” because the syn-
thesis of the perceptions “sun” and “warm stone” occurs according to a concept of the understand-
ing—causality—the synthetic judgment enjoys universally necessary and objective validity (Kant
1950, 301–302, n. 3). The novelty of Kant’s account of causality is that it regards causality not
as a feature of things as they are in themselves but as a regulative principle that determines how
those things must appear to the knower.
Causality so understood has the benefit of preserving scientific success, which was called into
question through Hume’s skepticism, and it is able to retain necessity and objectivity. Unlike the
Aristotelian philosophy of nature, however, the Kantian view of causality holds that necessity and
objectivity emerge not on the part of the metaphysical structures of the thing itself, which always
remain unknown, but owing to the fact that experience can occur only in a certain manner given
the a priori transcendental conditions of one’s cognitional structures. The laws of nature, such as
Newton describes them, are not abstracted from experience—as Hume thought, thereby according
them only a probabilistic character—but are instead imposed on the phenomena through under-
standing. Sense intuition and understanding work hand in hand to yield scientific experience such
as what is found in the Newtonian mechanics. What is more, Kant is clear that both faculties are
required, for “thoughts without content are empty, intuitions without concepts are blind” (Kant
1929, A51/B75).
Kant’s understanding of the relationship between intuition and understanding is not without its
implications for the possibility of arriving at natural knowledge of God’s existence. There is the
temptation, Kant thinks, for the understanding to apply its a priori concepts beyond the realm of
any sense experience, that is, beyond space and time. The categories of the understanding only have
applicability within the realm of space and time through which things in themselves present them-
selves phenomenally. Since God, understood as an ens realissimum, can never be a possible object of
experience, one can make no pronouncement on God’s existence. Although Kant will make appeal
to God in his moral philosophy as a postulate of pure practical reason, he remains convinced that
there is no theoretical justification for the claim that God exists. In fact, in his Critique of Pure Rea-
son, Kant argues at length that any effort to demonstrate God’s existence has always involved a dia-
lectical illusion in which reason goes beyond itself.
Kant holds that there are three—and only three—arguments for the existence of an ens realissi-
mum, namely, the ontological, cosmological, and physico-theological arguments. These arguments,
to Kant’s mind, build on each other cumulatively, with the foundational argument being the onto-
logical one. The ontological argument has a long and variegated history and stretches back to at least
Anselm of Canterbury. The version of the argument with which Kant was familiar, however, was
the more proximate one articulated by Leibniz, which itself shares certain traits with Descartes’s
argument from the fifth meditation. The ontological argument attempts to argue from a reductio
ad absurdum and maintains that anyone who rejects the claim “God exists” embraces a contradic-
tion. The reason for this is that the concept of being or existence is contained within the concept
of an ens realissimum such that were one to reject the existence of an ens realissimum, a contradiction
would ensue. For his part, Kant makes the famous claim that being is “not a real predicate” (Kant
1929, A598/B626). To illustrate his point, he turns to the example of the difference between one
hundred actual thalers and one hundred possible thalers. The difference between actuality and
potentiality is not captured conceptually, for if it were, there would be no identity but rather two
irreducibly distinct concepts A (one hundred thalers + existence) and B (one hundred thalers). Being
or existence, for Kant, is just a positing in objective reality that is expressed in a synthetic proposi-
tion validated by one’s experience. Insofar as the ontological argument intends to argue prior to any
experience, it has no ground to justify its assertion of God’s existence without simply begging the
question.
What is more, insofar as the ontological argument functions at the heart of both the cosmolog-
ical and physico-theological arguments, at least according to Kant, they too are predicated on a
problematic foundation and equally succumb to dialectical illusion. The cosmological argument,
“in the endeavor to pass as a second witness[,] merely changes its dress and voice” (Kant 1929,
A606/B634). Be that as it may, the prima facie difference between the ontological and cosmological
arguments is that the former occurs without adverting to any experience, whereas the latter makes a
direct appeal to the contingency of the world. If the world is a contingent reality, it must be the case
that there is a reality that is itself necessary and sufficient to account for contingency. Kant sum-
marizes the argument as follows: “If anything exists, an absolutely necessary being must also exist.
Now I, at least, exist. Therefore an absolutely necessary being exists” (Kant 1929, A604/B632).
This “absolutely necessary being,” of course, is simply God. But, continues Kant, in order for the
understanding to determine the nature of (i.e., assign attributes to) a necessary being, one turns to
the concept of an ens realissimum. To “determine” the concept of a necessary being is, Kant thinks,
to assign it “one out of each possible pair of opposed predicates” (Kant 1929, A605/B633). As he
sees it, the only concept that can “determine” the nature of a necessary being is the concept of an
ens realissimum, which, as we already saw, contains within itself the concept of its own existence.
In other words, at the last stage of the argument, the cosmological argument simply lapses back into
the ontological argument. With respect to the unique character of the cosmological argument as an
effort to argue from contingency to necessity via causal relations, Kant argues that, given that caus-
ality’s scope is limited to the realm of experience, using causality to transcend experience so as to
arrive at an ens realissimum is simply reason’s succumbing to a dialectical temptation. Outside of
the sensible world framed in terms of space and time, causality “has no meaning whatsoever” (Kant
1929, A609/B637).
The physico-theological argument fares no better, according to Kant. It too is dependent on the
ontological argument via the cosmological argument. Nevertheless, Kant appears to have a subdued
fondness for the physico-theological argument: “This proof always deserves to be mentioned with
respect. It is the oldest, the clearest, and the most accordant with the common reason of mankind”
(Kant 1929, A623/B651). Unlike the cosmological proof, which considers the reality of the world
taken as a whole, the physico-theological argument concerns only particular, organized structures
within the world. This particular organization, moreover, is “alien to the things in the world and
only pertains to them contingently” (Kant 1929, A625/B653). Accordingly, there must exist an

intelligent cause (God) capable of organizing the diverse strands of order within the world. Never-
theless, the purposiveness encountered in the world cannot, Kant thinks, ever prove anything more
than an “architect” of the world, and not a “creator” (Kant 1929, A627/B655). Moreover, insofar as
the argument argues from the contingency of the particular structures of the world to a necessary
designer, it has lapsed back into the cosmological argument and then, once again, into the ontolog-
ical argument. Having dismantled the three kinds of argument for God’s existence, Kant declares
that “all attempts to employ reason in theology in any merely speculative manner are altogether
fruitless and by their very nature null and void, and … the principles of its employment in the study
of nature do not lead to any theology whatsoever” (Kant 1929, A636/B664).
RECONSTRUCTION OF CAUSAL PERCEPTION
By the post-Kantian era we have traveled far from an Aristotelian understanding of causality—which
is not to suggest that in the Aristotelian tradition there is a unified view of causality or even a
“view.” Kant’s pessimism regarding the possibility of “pure reason’s” ability to ascend to a first being
via causal reasoning is, as we can see, a direct consequence of Hume’s critique of causality, which is
itself an outgrowth of early modern Cartesianism. As we have also seen, Hume’s critique unfolds
according to the representationalist conception of being that has been operative throughout much
of early modernity, wherein a truncated grasp of being’s otherwise intrinsic dynamism not surpris-
ingly yields a concept—or more properly an idea—that is itself equally anemic and incapable of
expressing the dynamic communication of (existential) act. Accordingly, causation becomes either
merely a matter of the succession of events in which no intrinsic connection can be observed or
an imposition of a priori cognitive structures on appearances. Gilson’s observation sums up this
philosophical situation well: “Hume was perfectly right in refusing to consider causal relations as
deducible from the essences of causes, and Kant was simply dodging the difficulty by transferring
to a category of the understanding the synthetic nature of a relation grounded in things themselves”
(1952, 185). But if causation does not emerge from the contents contained within the ideas of vari-
ous “essences,” does this mean that causality cannot be known at all? An answer to this question, I
suggest, follows on what one means by “to know.”
As Parmenides tells us famously in his poem, “The same thing is for thinking [νοεῖν] and for
being”—an observation whose implications have far-reaching consequences. One such implication
is that misunderstanding cognition not only misconstrues the structures of cognition itself, it also
misunderstands the nature of being. In Insight, his work devoted to cognition, Bernard Lonergan
describes cognition as a “formally dynamic whole” (2005, 205–221). By this, Lonergan understands
that cognition is constituted by parts, namely, experience, understanding, and judgment. As he sees
it, then, any one of these cognitive “parts” taken singularly in isolation from the rest does not con-
stitute knowledge, for, after all, no whole can be reduced merely to one of its parts and retain its
integrity. Rather, each part has meaning and functionality only in relation to one another within
the whole. In fact, the (cognitive) parts build on one another incrementally. What is more, in
describing them as “formally dynamic,” Lonergan means that these “parts” are activities that are
self-assembling and summon one another forth. That is to say, a particular experience first confronts
the knower and generates a question.
This experience is mediated by the knower’s sensation, memory, imagination, previous
insights—in short, what Michael Polanyi describes as “tacit knowledge” (2005, 90)—but still
remains opaque to the knower in terms of its intelligibility. The knower, then, through adverting
to understanding, attempts to answer the question “what is it?” In answering this question, the
knower confronts the sensory data of the experience and thus far traverses the same path as any
empiricist or neo-empiricist, yet, at a certain point, if successful, the knower is, through an insight,
able to discern the intelligible patterns latent within that (sensory) experience itself. While latent
within the sense experience, the intelligibility contained within it transcends the presentation of
sense data. For example, one may hear a speaker uttering a string of vocalizations—“two,” “four,”
“eight,” “sixteen,” and so on—and discern an intelligible pattern within the sense presentation—
namely, a proportional relation double. Here, the intelligible relation of “double” is located within
the concrete and material sensations constituting the utterance, but that intelligibility is certainly
not reducible merely to one’s sense experience. Rather, the intelligibility immanent within one’s
sense presentation transcends that presentation and is attained, not through sensation, but by
insight. Insight moves beyond the mere sensuous presentation of one’s experience; it is that eureka
moment in which the intelligibility enclosed within the sense experience is at last disenclosed.
But even this moment of insight remains tentatively hypothetical. What love-stricken school-
boy, for example, who regards the attention of the girl he fancies—her flirtatious demeanor, her
shy smile, her lingering glances—as precisely the reason for his thinking that perhaps she is smitten
by him too? There is, on one hand, what sensation presents (light, color, sound, etc.) and, on the
other hand, an intelligibility that is conveyed through that sensation and attained through under-
standing. Yet, to continue with our example, what love-stricken schoolboy who thinks the girl he
fancies also fancies him, has also not been paralyzed by the nagging question “is it so”? Perhaps
he has misread—embarrassingly so—the “signs.” Beyond understanding and its achievement of
insight, then, there is judgment, which attempts to answer the question: “is it so?” To answer this
question, judgment must turn to the virtually unconditioned. Any sense experience is a conditioned
reality and can be understood only once those conditions are fulfilled. But those conditions can
themselves be conditional and so their explanatory power is virtual. Whatever its designation, judg-
ment seeks to validate its understanding through evidence. For the empirical researcher, judgment is
the validation of a hypothesis by means of experimentation, which is the space where theory con-
fronts existence. For Juliet, it is awakening from her drug-induced sleep only to find Romeo dead
by her side for grief of her (he did love her); for Galileo, it is seeing the moons of Jupiter transit
the gas giant (“E pur si muove”). The fact of the matter is that cognition has made contact with
being, which is intelligible. But being’s intelligibility is not such that it is disclosed all at once and
exhaustively. Rather, being must be interrogated by cognition, which requires knowing which ques-
tions to ask to attain the answers it seeks.
Yet, if the history of philosophy has called into question cognition’s efficacy, it has also thereby
called into question being’s intelligibility. Again, as it is with knowing, so is it with being. As the
history of philosophy has shown, there has been the frequent effort to reduce knowledge only to
one of its dynamic constituents. “Intellectual operations are related to sensitive operations, not by
similarity,” argues Lonergan, “but by functional complementarity; and intellectual operations have
their objectivity, not because they resemble ocular vision, but because they are what ocular vision
never is, namely, intelligent and rational” (2005, 217–218). In vain, then, would someone such
as Hume turn to mere sense experience and hope to discover therein intrinsic connection rather
than mere constant conjunction. And when the idealist holds that one only ever knows appearances
and laments the ability to reach what is real, does not one detect a subtle legerdemain? “Sense does
not know appearances because sense alone is not human knowing, and because sense alone does not
possess the full objectivity of human knowing. By our senses we are given, not appearance, not real-
ity, but data” (Lonergan 2005, 218). If not through sensation, one might ask: by what act are we
given reality? The answer is: not by any one in particular but only by the total realization of the
complete structure of them all, which is just cognition.
Thomas Aquinas, for example, knew well the value of what Lonergan would later maintain
regarding the perception of being. For Thomas, the structure of cognition mirrors the structure of
being. As being (ens) is twofold—namely, constituted by the two metaphysical principles of exis-
tence (esse) and essence—so too is the cognition of being (ens) twofold. The first operation of the
intellect, known as simple apprehension, is that act by which one knows the quiddities of things.
That is to say, simple apprehension abstracts the intelligible patterns latent within one’s sense expe-
rience and yields a concept expressing the perceived thing’s formal nature, whether substantial or
accidental. The second operation of the intellect, in contrast, is directed to the very existence of
the thing (esse rei). Thomas says: “But the second operation concerns the existence itself [ipsum esse]
of the thing, which indeed results from the coming together of the principles of the composite
thing, or, as is the case in simple substances, is concomitant to the simple nature itself” (Super
Boetium de Trinitate, q. 5, a. 3). Here, Thomas’s point is that simple apprehension alone is not
knowledge, nor for that matter is the second operation of the intellect. The reason for this is that

neither essence nor existence taken singly is a being. One need not be committed to Thomas’s the-
ory of the real “otherness” between existence (esse) and essence to appreciate the notion as old at
least as Parmenides that where there is no being there can be no knowledge.
Accordingly, if, as the Aristotelian tradition has understood it, causality is fundamentally the
communication of existential perfection and if existence is reached finally in the second operation
of the intellect, why should one hope to perceive that causal communication in a bare representa-
tional idea, let alone in mere sense data? And yet this is precisely the tactic that early modern
representationalism has employed. Having failed to discern causal communication either within
sense data alone or within the relation of concepts alone, one either had to despair of perceiving
causal connection at all and reduce it to custom and experience, on one hand, or, on the other,
impose causal relations on experience. Where, however, there is no experience there can be no
custom to fall back on or phenomena to organize. Since God does not present himself within
experience—at least not outside of some mystical or extraordinary experience—one cannot hope
to have any (natural) cognition of the divine being at all. What could not be perceived in sensa-
tion alone or conceptualization alone can, however, readily be known through cognition as a
whole, which also involves judgment. The actual existence of intellectual success within the
domains of mathematics, science, morality, social sciences, linguistics, and so on, and the count-
less moments of cognitive success among individual persons (e.g., problem-solving, innovations,
social communications), give witness to the existence and efficacy of judgment. As Clarke puts
it, “It would be impossible to live rationally or even to sustain human life for long in the often
puzzling and dangerous world we inhabit, where we have to understand it or be destroyed by
it” (2001, 21–22). And yet there is human success, for we have not (yet) vanished from the planet
and have even managed to master it. Thus, to raise anew the question that has guided this chap-
ter: can God be known through causal reasoning? The answer depends on what is meant by
“cause” and how causation is known. It is far from the case, I think, that early modernity, which
witnessed a crisis in metaphysics and the possibility of a natural theology, has the final (or even
the best) word on the matter.
Bibliography
Albertus Magnus. Metaphysica. Edited by Bernhard Geyer. Cologne, Gilson, Étienne. The Unity of Philosophical Experience. San Francisco:
Germany: Aschendorff, 1964. Ignatius Press, 1999.
Albertus Magnus. De causis et processu universitatis a prima causa. Edited by Hume, David. An Enquiry concerning Human Understanding. Edited by Eric
Winfridus Fauser. Cologne, Germany: Aschendorff, 1993. Steinberg. Indianapolis, IN: Hackett, 1993.
Anonymous. The Book of Causes [Liber de causis]. Translated by Dennis J. Kant, Immanuel. Critique of Pure Reason. Translated by Norman Kemp
Brand. Milwaukee, WI: Marquette University Press, 1984. Smith. London: Macmillan, 1929.
Aristotle. The Basic Works of Aristotle. Translated Richard McKeon. New Kant, Immanuel. Prolegomena to Any Future Metaphysics. Translated by
York: Random House, 1941. Lewis White Beck. New York: Macmillan, 1950.
Clarke, W. Norris. The One and the Many: A Contemporary Thomistic Leibniz, Gottfried Wilhelm. Discourse on Metaphysics; Correspondence with
Metaphysics. Notre Dame, IN: University of Notre Dame, 2001. Arnauld; and Monadology. Translated by George Montgomery. La Salle,
IL: Open Court, 1993.
Congar, Yves. A History of Theology. Edited and translated by Hunter
Guthrie. Garden City, NJ: Doubleday, 1968. Lonergan, Bernard. Insight: A Study of Human Understanding. New York:
Philosophical Library, 1970.
Cross, Richard. Duns Scotus on God. Aldershot, UK: Ashgate, 2005.
Lonergan, Bernard. Collected Works. Edited by Frederick E. Crowe and
Descartes, René. Meditations on First Philosophy. Translated by Laurence Robert M. Doran. Toronto: University of Toronto Press, 2005.
Lafleur. Upper Saddle River, NJ: Prentice Hall, 1951.
Malebranche, Nicolas. The Search after Truth. Edited and translated by
Frank, William A., and Allan B. Wolter. Duns Scotus, Metaphysician. West Thomas Lennon and Paul J. Olscamp. Cambridge, UK: Cambridge
Lafayette, IN: Purdue University Press, 1995. University Press, 1997.
Ghazālī, al-. The Incoherence of the Philosophers. Translated by Michael Maritain, Jacques. Approaches to God. Translated by Peter O’Reilly. New
Marmura. Provo, UT: Brigham Young University Press, 2000. York: Harper and Brothers, 1954.
Gilson, Étienne. Being and Some Philosophers. Toronto: Pontifical Institute of Newton, Isaac. Opticks; or, A Treatise of the Reflections, Refractions, Inflections
Mediaeval Studies, 1952. and Colours of Light. 3rd ed. London: William and John Innys, 1721.
Owens, Joseph. St. Thomas Aquinas on the Existence of God: The Collected Vattimo, Gianni. After Christianity. Translated by Luca D’Isanto. New York:
Papers of Joseph Owens. Edited by John R. Catan. Albany: State Columbia University Press, 2002.
University of New York Press, 1980.
Velde, Rudi te. Aquinas on God: The “Divine Science” of the Summa
Polanyi, Michael. Personal Knowledge: Towards a Post-critical Philosophy. theologiae. Burlington, VT: Ashgate, 2006.
London: Routledge, 2005.
Wippel, John F. The Metaphysical Thought of Thomas Aquinas: From Finite to
Suárez, Francisco. Disputationes metaphysicae. Paris: Luis Vivès, 1861. Uncreated Being. Washington, DC: Catholic University of America
Thomas Aquinas. Summa theologia. In Opera omnia. Rome: Ex Typographia Press, 2000.
Polyglotta, 1888–. Wittgenstein, Ludwig. Tractatus Logico-philosophicus. Translated by D. F.
Thomas Aquinas. Super Boetium de Trinitate. In Opera omnia. Paris: Les Pears and B. F. McGuinness. London: Routledge and Kegan Paul,
Éditions du Cerf, 1993. 1971.

TOPIC 9
Causation and Sufficient Reason: Atheism

Felipe Leon
El Camino College, Torrance, CA
This chapter discusses the nature of causation and its fundamental role in the debate between theism and
atheism. The discussion covers cosmological arguments that deploy causal or explanatory principles to prove
God’s existence.
CAUSES, REASONS, AND GOD
Theists argue that God and his activity provide the cause or sufficient reason for the existence and
nature of the universe. Atheists, by contrast, argue that God isn’t needed to cause or explain the
universe’s existence or nature. The notions of causation and sufficient reason thus play a key role
in arguments concerning theism and atheism. The present chapter provides an overview of these
notions and their bearing on the epistemic merits of the two views. The plan of the chapter is as
follows. First, we discuss the core issues and options with respect to the nature of causation and
its bearing on theism and atheism. Second, we explore and critically evaluate the most widely dis-
cussed theistic argument that makes crucial use of a causal premise, namely, the kalām cosmological
argument. Finally, we explore and critically evaluate a representative sampling of the most widely
discussed theistic arguments that make crucial use of an explanatory premise, namely, cosmological
arguments from contingency. In each of these sections, we conclude that considerations respecting
causation and sufficient reason do not favor theism over atheism.
CAUSATION
Before considering causation’s bearing on the epistemic merits of theism and atheism, a brief over-
view of some of the core issues and options with respect to the nature and categories of causation is
in order.
THE NATURE OF CAUSATION: ISSUES AND OPTIONS

There is considerable disagreement about virtually every aspect of causation. (For an overview of the
issues, see Paul and Hall 2013; Beebee, Hitchcock, and Menzies 2009; and Schaffer 2016.) How-
ever, according to the standard view, causation is a relation, where this relation is irreflexive (a thing
can’t cause itself), asymmetric (if A is the cause of B, then B isn’t the cause of A), and transitive (if A
causes B and B causes C, then A causes C).
A number of accounts have been proposed for the causal relation itself, as well as its relata. Pro-
posals for the latter include events, facts, states of affairs, universals, exemplifications of universals,
tropes, and objects. With respect to the causal relation itself, standard views include regularity
accounts (A causes B just in case all As are followed by Bs); counterfactual dependence accounts
(A causes B just in case: if A had not happened, B wouldn’t have happened); probability-raising
281
accounts (A causes B just in case A raises the probability of B);

KEY CONCEPTS conserved quantities accounts (A causes B just in case A trans-
fers a conserved quantity [e.g., mass-energy, charge, momen-
tum, etc.] to B); and agency accounts (A causes B just in case
Big Bang Cosmology
bringing about A would be an effective means for a free agent
Causal Principle to bring about B).
Conjunction of All Contingent Entities There is also considerable disagreement about the meta-
(CCE) physical status of the causal relation. Two main issues are rele-
Contingent Concrete Reality vant for our purposes. The first concerns whether the causal
relation is an objective feature of the world. While realists of
Cosmological Arguments course affirm that it is, eliminativists deny this on a number
Cosmological Arguments from Contingency of grounds. For example, some think the causal relation need
Eliminativism not be posited in our best physical theories. Others think the
problems besetting extant accounts of causation are so severe
Epicurean Cosmological Argument as to warrant abandoning the notion. The second issue, which
Explanatory Principle arises within the realist camp, is whether the causal relation
Explicability Argument can be analyzed in terms of noncausal facts (reductionism) or
not (antireductionism). Reductionistic views are represented in
Formal, Final, Material, and Efficient the proposed analyses of the causal relation mentioned above.
Causes (Aristotle’s table of causes) By contrast, antireductionist views assert that the causal relation
Immortal Counterargument is a basic or primitive feature of reality.
Kalām Cosmological Argument Finally, there is disagreement about how many fundamen-
tal types of causes there are. We can distinguish the types that
Mereological
have been proposed by means of Aristotle’s table of causes: for-
Physical Reality mal, final, material, and efficient. To illustrate these, consider a
Principle of Sufficient Reason (PSR) car that just rolled off the assembly line. The car originated as
an idea in the minds of its designers (its formal cause), which
Reductionism, Antireductionism
they made for a purpose: transportation (the final cause). Fur-
S5 Modal Logic thermore, the car was ultimately made from steel, plastic, glass,
Subtraction Argument and so on (the material causes). Finally, by means of persons
and machines (the efficient cause), these materials were trans-
formed into a shiny new car.
THEISM, ATHEISM, AND FINAL CAUSES
Many think final causes are incompatible with atheism. The issue is usually traced back to Thomas
Aquinas, who is widely thought to have argued that final causes require an agent acting for the sake
of some end. However, it is controversial that this was Aquinas’s view of the matter (see, e.g., Hoff-
man 2009). More important, it’s not at all clear that atheism is incompatible with final causes that
are more fundamental than those produced by human persons.
To see this, let us distinguish between four notions of a final cause that seem to be suggested
in the writings of Aquinas: (a) the tendency to produce one type of effect rather than another, (b)
the tendency toward some endpoint or terminus, (c) the tendency toward some endpoint or ter-
minus that is good in some important sense, and (d) acting for the sake of some end (Hoffman
2009). Now if there are ground-floor-level final causes in sense (d), then this is clearly incompati-
ble with atheism. However, it’s not at all clear that any of the other three senses of a final cause
are incompatible with or otherwise problematic for atheism. For (a) only requires regularity in
nature, which is prima facie compatible with atheism. On this point, it’s worth noting that this
seems to be the only sense of a final cause that Aquinas thought was presupposed by efficient
causes (Hoffman 2009).
Furthermore, (b) and (c) seem no more problematic for atheism than for theism. For consider
the God of theism. Prima facie, his personhood, omnipotence, and omniscience entail that God
has at least an intellect and a will, and these work together in such a way as to have a tendency
toward a number of ends (e.g., to design and create things). Furthermore, these ends appear to
be good in some important sense—at a minimum, they are good for God’s well-being. Prima
facie, then, final causes in senses (b) and (c) are built into God’s nature without a prior cause.
But if that’s right, then classical theism entails the existence of final causes at the metaphysical
ground floor that God cannot create. And if that’s right, then theism entails that nonconscious
final causes in senses (b) and (c) are more fundamental features of reality than final causes pro-
duced by intelligence. And finally, if that’s right, then it’s not clear why atheism would be any
worse off than theism in this regard if it should turn out that there are such final causes at the
metaphysical ground floor that have no prior cause (Hume 1993 [1779]; Leon and Rasmussen
forthcoming).
COSMOLOGICAL ARGUMENTS: PRELIMINARIES
We turn now from consideration of the causal relation itself to arguments for theism that exploit both
it and its near cousin, the principle of sufficient reason. (For further discussion of the latter, see the
second main section of the chapter.) Perhaps the central arguments for theism that exploit the notions
of causation and sufficient reason to infer God’s existence are cosmological arguments. In rough terms,
cosmological arguments aim to show that the sheer existence of the universe, or some features therein,
cannot be adequately accounted for without appeal to God as their causal or explanatory ground.
Standard cosmological arguments can be divided into two main types: those that deploy a causal
principle as one of their core premises, and those that deploy an explanatory principle, or principle of
sufficient reason, as one of their core premises. Each type of cosmological argument is discussed in turn.
THE KALĀM COSMOLOGICAL ARGUMENT

The kalām cosmological argument aims to show that there is a first, uncaused cause of the universe,
and that this cause is God. William Lane Craig is the leading proponent of the argument, and so,
with a few exceptions, this section focuses on his defense of it.
Craig (1979) expresses the core of his argument as follows:
(1) Whatever begins to exist has a cause of its existence.
(2) The universe began to exist.
(3) Therefore, the universe has a cause of its existence.
Craig offers both a priori and a posteriori considerations for the causal principle expressed in prem-
ise (1). According to the first, the causal principle is supported by an a priori intuition that some-
thing cannot come from nothing. According to the second, the principle enjoys strong inductive
empirical support.
For premise (2), Craig offers five main arguments, two a priori and three a posteriori. The
first is that an infinite past is impossible because actual infinites are impossible in general. The sec-
ond is that, even if concrete actual infinites are possible, they can’t be traversed by successive addi-
tion. The third is that the evidence for the big bang indicates an absolute beginning to the uni-
verse. The fourth is that the second law of thermodynamics indicates that the universe is
running down, in which case it must have been “wound up” with an initial input of matter-
energy at the beginning of the universe. Finally, according to the fifth, the Borde-Guth-Vilenkin
theorem (discussed below) entails that any universe that is (on average) expanding has an absolute
beginning.
After defending his core argument, Craig goes on to argue that an analysis of the cause of the
universe reveals it to be the spontaneous, free act of a person of some kind. But since it is the cause
of spatiotemporal, physical reality, it must be a timeless, immaterial person of immense power. And
this, as Aquinas would say, we all call “God.”

EVALUATING PREMISE (2): DID THE UNIVERSE BEGIN TO EXIST?
As mentioned above, Craig’s first argument for premise (2) is that the past can’t be actually infinite
because actual infinites are impossible in general. Craig has offered a large number of sub-arguments
for this claim.
A PRIORI ARGUMENTS AGAINST THE POSSIBILITY OF ACTUAL INFINITES

The present section discusses a representative sampling of Craig’s arguments. (For a fuller exposition
and discussion, see Morriston 1999, 2002a, 2003, 2010, 2013; Craig and Smith 1993; Draper
2008; Sobel 2004; and Oppy 2006a, 2006b.)
Infinite Library, Part I. Craig offers a number of thought experiments to support the claim that
actual infinites are impossible. To begin, consider the following thought experiment involving an
infinite library:
Suppose … that each book in the library has a number printed on its spine so as to create a
one-to-one correspondence with the natural numbers. Because the collection is actually
infinite, this means that every possible natural number is printed on some book. Therefore, it
would be impossible to add another book to this library. For what would be the number of the
new book?… Every possible number already has a counterpart in reality, for corresponding to
every natural number is an already existent book. Therefore, there would be no number for
the new book. But this is absurd, since entities that exist in reality can be numbered. (Craig
1979, 83)
A standard criticism of the present thought experiment is that it is possible to assign a unique nat-
ural number to the new book. This is because one can reassign the natural numbers to the books
in the library so as to free up a natural number for the new book. For example, suppose we assign
“2” to the first book, “3” to the second book, and so on all the way through the rest of the books
in the library. Then we can free up “1” to be assigned to the new book (Morriston 2002a; Oppy
2006a).
Another criticism is that Craig’s argument seems to make a fallacious inference from (a) Any
concrete entity can be uniquely numbered, to (b) concrete entity can be uniquely numbered via a
natural number. However, (b) looks to be false. For consider a library that contains a set of books
that can be put into a one-to-one correspondence with the set of real numbers. Such a set of books
would be non-denumerably infinite; that is, it would be actually infinite, but it couldn’t be put into a
one-to-one correspondence with the natural numbers. Therefore, while all such books in the library
can be uniquely numbered, they can’t all be uniquely numbered via the natural numbers.
In short, either we hold fixed the assignment of natural numbers to books in Craig’s infinite
library or we don’t. If we don’t, then it’s possible to reassign the natural numbers so as to free up
a unique natural number for the new book. On the other hand, if we do hold fixed the original
assignment of numbers to books, then it is impossible to assign a unique natural number to the
new book. But of course there are more numbers than the naturals, and the new book can be num-
bered with one of these. If Craig yet demands that the new book be numbered with a natural num-
ber, even after all the natural numbers have been assigned to other books, then the problem lies not
with the possibility of his infinite library, but rather with the coherence of the task demanded for it.
Either way, Craig’s argument is unsuccessful.
Infinite Library, Part II. Consider next a variant of the library thought experiment (Craig and
Smith 1993). According to this version, the problem is that such a library would have just as many
even-numbered books as it would have odd- and even-numbered books combined. This is because
(i) the set of even numbers (2, 4, 6, …) can be put into a one-to-one correspondence with the set of
natural numbers (1, 2, 3, …), and (ii) any two sets that can be put into a one-to-one correspon-
dence with one another are the same “size”—each set is just as large as the other. But this is absurd;
the set of even-numbered books is a proper subset, or proper part, of the set of all the books in the
library, and no proper part or subset of a set could be just as large as the set of which it is a part.
Therefore, a library containing an actually infinite number of books is impossible. But this problem
isn’t particular to just infinite libraries, but applies to any actually infinite collection of things what-
soever; therefore, concrete actual infinites are metaphysically impossible.
We can express the argument above as follows:
(1) If concrete actual infinites are possible, then a library with an actually infinite number of
books (each labeled with a unique natural number) is possible.
(2) If a library with an actually infinite number of books (each labeled with a unique natural
number) is possible, then it would have just as many even-numbered books as odd- and
even-numbered books combined.
(3) There couldn’t be just as many even-numbered books as odd- and even-numbered books
combined.
(4) Therefore, concrete actual infinites are impossible.
A standard criticism of the argument is that there is an ambiguity in the notion of “just as
many” in premises (2) and (3) (Morriston 2002a, 2002b, 2003; Oppy 2006a; Draper 2008). On
one hand, one might mean that there are just as many even-numbered books as odd- and even-
numbered books together in the sense that the two sets of books can be put into a one-to-one cor-
respondence with one another. Then while (2) is true, it’s not clear that (3) is true. For an actual
infinite is standardly defined as a set that can be put into a one-to-one correspondence with one
of its proper subsets. But, presumably, Craig’s argument is supposed to persuade those who already
know what an actual infinite is supposed to be, and yet remain at least agnostic as to whether actu-
ally infinite sets of things are metaphysically possible. It seems, then, that one wouldn’t accept (3)
unless one has already rejected the possibility of actual infinites.
On the other hand, one might mean that there are just as many even-numbered books as odd-
and even-numbered books together in the sense that neither set of books has other members besides
the ones they share in common. Then while (3) is true on that reading, (2) is not. For of course the
set of odd- and even-numbered books has other books besides the set of books they have in com-
mon, namely, the even-numbered books. The upshot is that the argument gets its intuitive appeal
by equivocating on the notion of “just as many.” Thus, once the different readings of the expression
are disambiguated, the argument loses its force.
Hilbert’s Hotel, Part I. Craig’s most popular arguments against actual infinites appeal to varia-
tions on the Hilbert’s Hotel thought experiment (Hilbert 2013): Imagine a hotel with an actually
infinite number of rooms. If such a hotel were possible, then it would have some remarkable fea-
tures. For example, suppose every one of its rooms were full. Then one could create vacancies in
some of the rooms even if none of the guests left. So, for example, the hotel manager could have
the guests in room 1 move to room 2, the guests in room 2 move to room 3, and so on for all
the rest of the rooms. In this way, room 1 would become vacant for a new guest, even though
not a single guest left the hotel.
In fact, one could create infinitely many vacant rooms without removing a single guest. For one
could move the guest in room 1 to room 2, the guest in room 2 to room 4, the guest in room 3 to
room 6, and in general, for each room n, move the guest in n to 2n. The result would be that an
infinite number of rooms would become vacant for more guests to occupy.
Furthermore, one could also remove any vacancies—that is, fill all the rooms—without adding
a single new guest (again, assuming one has an infinite number of guests to move around). So, for
example, one could just reverse the moving of guests stated in the previous examples, and thereby
go from an infinite number of vacancies to a fully occupied hotel.
These features of an infinite hotel illustrate that no such hotel could exist—not even an omnip-
otent God could create one. For it is metaphysically impossible for a spatial region to transition
from being occupied to being vacant merely by moving its occupants around; without removing
at least some of what occupies a region, that region will remain occupied by something or other.
Similarly, it is metaphysically impossible for a spatial region to transition from being vacant to being

occupied merely by moving its occupants around; without adding at least some new occupants to a
given region, that region cannot become occupied.
We can express the argument in the illustration above as follows:
(1) If concrete actual infinites are possible, then a hotel with an infinite number of rooms and
guests is possible.
(2) If a hotel with an infinite number of rooms and guests is possible, then it can transition from
having no vacancies to having vacancies without removing any of its occupants, and it can
transition from having vacancies to having no vacancies without adding new occupants (i.e.,
merely moving occupants around is sufficient).
(3) No hotel can transition from having no vacancies to having vacancies without removing any
of its occupants, and no hotel can transition from having vacancies to having no vacancies
without adding new occupants (i.e., merely moving occupants around is not sufficient).
A number of worries can be raised against this argument. First, one might worry about (3):
Granted, no hotel occupying a finite region of space and with a finite number of rooms could tran-
sition from having no vacancies to having some (and vice versa) by just moving some of its occu-
pants around. But these things aren’t obviously true for a hotel occupying an infinite region of space
and having infinitely many rooms. For one might think that what makes such transitions impossible
for finite hotels is that having spatial boundaries leaves some occupants with nowhere to go in such
transitions (e.g., the person in the room at the end of the hotel). But since an infinite hotel of the
relevant sort lacks a spatial boundary in at least one direction, there is always somewhere for each
tenant to go (since there is no last room at the end of the hotel).
Second, Morriston (2002a) and Oppy (2006b) have raised worries for the justification for (1),
on the grounds that the supposed absurdities of Hilbert’s Hotel don’t obviously generalize to all
actual infinites. More to the point, the supposedly problematic features of Hilbert’s Hotel aren’t a
part of sets of past events—whether infinite or not. For the supposed problems arise only when
one combines infinites with copresent regions, and only when the occupants of such regions can
be rearranged. But, of course, temporal regions are not copresent (at least not according to Craig’s
preferred A-theory of time). Nor can the events that occupy the temporal regions be rearranged or
moved around. Therefore, even if the combination of features in Hilbert’s Hotel and Craig’s infinite
library give rise to absurdities, such a combination of features is not present in an infinite past. But
if not, then Craig’s justification for premise (1) is undercut.
Hilbert’s Hotel, Part II. One of Craig’s variations on Hilbert’s Hotel raises a more serious charge.
To see this, suppose all the guests in the even-numbered rooms check out. Then the hotel would
still have infinitely many occupied rooms—the odd-numbered rooms. On the other hand, suppose
all the guests in rooms 4, 5, 6, … check out of the hotel. Then the hotel will no longer have infi-
nitely many occupied rooms—it will have three. But this is impossible, since just as many guests
checked out in the first scenario as in the second scenario. This is because (i) the infinite set of
even-numbered rooms (2, 4, 6, …) can be put into a one-to-one correspondence with the set of
rooms numbered 4 and higher (4, 5, 6, …), and (ii) any two sets that can be put into a one-to-
one correspondence with one another are the same “size”—each set has just as many members as
the other. But this is an outright contradiction: subtracting the same amount yields different
amounts. Therefore, Hilbert’s Hotel is metaphysically impossible. But, argues Craig, the absurdity
isn’t particular to just infinite hotels, but generalize to every infinite set of concrete entities whatso-
ever. Therefore, concrete actual infinites are metaphysically impossible.
We can express the argument illustrated above as follows:
(1) If concrete actual infinites are possible, then a hotel with infinitely many rooms and
infinitely many guests is possible.
(2) If a hotel with infinitely many rooms and infinitely many guests is possible, then it’s possible
to remove just as many guests from the hotel and get different amounts of remaining guests.
(3) It’s impossible to remove just as many guests from the hotel and get different amounts of
remaining guests.
What to make of this argument? By now the problem with Craig’s argument is a familiar one: there
is an ambiguity in the expression “just as many” in (2) and (3). On the one hand, it might mean
that “just as many” guests were removed in the two scenarios in the sense that the two sets of rooms
can be put into a one-to-one correspondence with one another. Then while (2) is true, it’s not obvi-
ous that (3) is true. For while the two sets of rooms can be put into a one-to-one correspondence
with one another, one set has all the members of the other set and others besides—the odd-num-
bered rooms. Referring to the coarse-grained similarity of the two sets (sameness of cardinal num-
ber) conceals this crucial difference. But once one shifts one’s focus from the sameness of cardinality
to the finer-grained difference in the particular members of the two sets, the intuitiveness of premise
(3) disappears. For then one would expect different results when the sets removed in the two scenar-
ios are different in this crucial way.
A PRIORI ARGUMENTS AGAINST THE TRAVERSABILITY OF ACTUAL INFINITES

As mentioned at the beginning of our discussion of the kalām argument, Craig’s second a priori
argument for a finite past is that, whether or not concrete actual infinites can exist, they cannot
be crossed or traversed one at a time, by successive addition. But if not, then if the past is actually
infinite, the past couldn’t have been traversed to reach the present. But since this is absurd (after
all, here we are), the number of past moments must be finite, in which case the universe had a
beginning.
Immortal Counterargument. One of Craig’s main arguments against the possibility of traversing
the infinite is compatible with the objection to the previous argument. Craig states the argument as
follows:
Suppose we meet a man who claims to have been counting down from infinity and who is
now finishing: …, -3, -2, -1, 0. We could ask, why didn’t he finish counting yesterday or the
day before or the year before? By then an infinite time had already elapsed, so that he should
already have finished. Thus, at no point in the infinite past could we ever find the man
finishing his countdown, for by that point he should already be done! In fact, no matter how
far back into the past we go, we can never find the man counting at all, for at any point we
reach he will already have finished. But if at no point in the past do we find him counting, this
contradicts the hypothesis that he has been counting from eternity. This shows again that the
formation of an actual infinite by never beginning but reaching an end is as impossible as
beginning at a point and trying to reach infinity. (2009, 121–122)
Call this the immortal counterargument. The immortal counterargument can be expressed as a reduc-
tio, with (1) below as the premise set up for reduction:
(1) The past is beginningless.

(2) If the past is beginningless, then there could have been an immortal counter who counts
down from such a past at the rate of one negative integer per day.
(3) The immortal counter will finish counting if and only if he has an infinite number of days in
which to count them.
(4) If the past is beginningless, then there are an infinite number of days before every day.
(5) Therefore, the immortal counter will have finished counting before every day.
(6) If the immortal counter will have finished counting before every day, then he has never
counted.

(7) Therefore, the immortal counter has both never counted and has been counting down from
a beginningless past (contradiction).
(8) Therefore, the past is not beginningless (from 1 to 7, reductio).
A standard criticism of the immortal counterargument takes issue with premise (3) (Morriston
1999, 2002b, 2013; Leon and Rasmussen forthcoming). The core of the criticism is that Craig
conflates counting an infinite number of negative integers with counting them all, and that com-
pleting the former task does not entail completing the latter. But the problem is that it’s epistemi-
cally possible (i.e., possible for all we know) that he’s counted down an infinite number of nega-
tive integers from a beginningless past, and yet has not counted them all. So, for example, he
could now be counting “-3,” so that he has just finished counting an infinite number of negative
integers—{… -5, -4, -3}—and yet he has not counted down all the negative integers. Given this
epistemic possibility, any reason for believing his (3) is undercut.
Craig’s (1985) reply to this sort of objection is that the believer in the (epistemic) possibility of
a beginningless past is committed to the claim that the counter will finish his count just in case the
days he’s counted can be put into a one-to-one correspondence with the set of natural numbers.
This is because otherwise the objector can’t account for the possibility of an immortal counter
who finishes the task on a particular day, as opposed to any other day.
However, Craig’s response seems to presuppose a substantive version of the principle of suffi-
cient reason, according to which every event has a sufficient reason for why it occurs (Morriston
2003; Oppy 2006a). But this is problematic, as the principle of sufficient reason—especially as
strong as the one assumed here in Craig’s reply—is implausible. (For more on this, see the section
below on cosmological arguments from contingency.) Indeed, Craig (2008) himself has explicitly
rejected strong versions of the sort at issue.
Furthermore, in order for Craig’s reply to work, it must be possible for the series of events of
the universe’s history to occupy different temporal intervals than they in fact occupy. Craig’s reply
therefore seems to presuppose a controversial view of the nature of time, according to which the
temporal segments of the universe’s history are distinct from the events that occupy them (Morris-
ton 2003). If, on the other hand, the temporal intervals of the universe’s history are identical to its
events, or the temporal metric applied to events is a matter of convention, Craig’s reply fails (Mor-
riston 2003.).
A POSTERIORI ARGUMENTS FOR A BEGINNING

In addition to a priori arguments, Craig offers several scientific arguments for a beginning of the
universe. These are discussed in the present section.
Big Bang Cosmology. According to the first scientific argument, big bang cosmology supports the
premise that the universe began to exist (Craig 2008; Craig and Sinclair 2009). The core idea is that
current scientific evidence indicates that our physical universe began less than 14 billion years ago,
when our spatiotemporal manifold and all of the energy within it existed in an extremely condensed
state. From that hot and volatile initial state, our universe expanded and evolved into its present
state. Since big bang models are well confirmed, they imply that our universe is of finite age, and
things with a finite age have a beginning, big bang cosmology confirms premise (2) of the kalām
cosmological argument.
A number of criticisms have been raised against this line of reasoning. First, our best scientific
evidence does not support that the universe had a beginning, since it can’t tell us about the earliest
known stages of the visible universe, before Planck time (i.e., 10–43 seconds). This is because the
general theory of relativity breaks down at that point, and quantum effects dominate. But since
we currently lack a quantum theory of gravity, our best scientific theories can’t tell us what hap-
pened at times prior to Planck time, including even whether there was a beginning to our universe
at all (Morriston 2002b, 2013; Oppy 2006b; Monton 2011).
Second, even if it were true that the evidence for a big bang provided sufficient evidence that
our universe began to exist, it wouldn’t provide sufficient evidence that all physical reality (let alone
all contingent concrete reality) began to exist (Morriston 2002b, 2013; Oppy 2006b). Indeed, on a
wide variety of quantum cosmological theories, our universe has temporal, physical antecedents
(Craig 2008; Craig and Sinclair 2009).
Third, even if it were true that the evidence for the big bang provided sufficient evidence that
all physical reality began to exist, including physical time, it doesn’t follow that there was no time
prior to the big bang (Morriston 2002b, 2002c, 2013). For as Craig (1992) himself has argued,
there is a kind of time—metaphysical time—that is more fundamental than physical time. Thus,
we can imagine or conceive of God’s existing prior to the big bang and counting off, “1, 2, 3, …,
Fiat lux!” (Craig 1992). According to this scenario, the elapsing of God’s mental states consisting
of his countdown to creation constitute events in metaphysical time, and these are temporally prior
to the physical time of our universe. Unless one can rule out this epistemic possibility, one cannot
conclude an absolute temporal beginning from the truth of the big bang theory.
Fourth, even if the universe had a beginning in the sense that there is nothing temporally prior
to the big bang singularity, it doesn’t follow that there is nothing causally or ontologically prior to it
(Morriston 2002b, 2002c, 2013). It’s therefore compatible with the possibility of the universe cre-
ated from a timeless stuff. Indeed, Craig is already committed to saying that, causally and ontologi-
cally prior to the singularity, God exists timelessly. It’s therefore not clear what principled grounds
there might be to rule out the epistemic possibility that, causally and ontologically prior to the sin-
gularity, other things besides God exist timelessly. And if that’s right, it’s not clear how Craig can rule
out the epistemic possibility that God created the universe out of a timeless stuff. Therefore, even if
there was no time prior to the singularity, it doesn’t follow that the universe was created ex nihilo.
In anticipation of the previous objection, Craig (1979, 1991) has argued that he can rule out
the epistemic possibility of a timeless stuff because he thinks (a) the only possible stuff from which
God could make the universe is matter/energy, (b) timeless stuff is quiescent, and (c) matter/energy
is never quiescent. However, it’s not clear that (a) is true, as it’s epistemically possible that the
universe was created out of some timeless stuff that’s distinct from the matter/energy we observe
(Morriston 2002b, 2002c). Indeed, the intuitive and empirical evidence for the need of a material
cause is at least as strong as that for an efficient cause, in which case the need of an efficient cause
of the universe and of a material cause of the universe stand or fall together: rationality requires that
we posit the one just in case it requires that we posit the other (Morriston 2002b, 2002c).
Finally, as Craig acknowledges, the argument presupposes the A-theory of time and real tempo-
ral becoming (Craig and Sinclair 2009). But many scientists and philosophers deny real temporal
becoming, accepting instead the B-theory of time. According to the latter view of time, all past,
present, and future events exist tenselessly, on a par with one another. But if that’s right, then it’s
not at all clear that the universe began to exist in a sense that cries out for a cause. Rather, it exists
timelessly, enjoying a mode of being similar to God’s. And if that’s right, then Craig’s premise (2) is
false.
(Craig has another argument, from the second law of thermodynamics, that is subject to very
similar objections to those that can be raised against his argument from big bang cosmology. For
reasons of space, I pass over consideration of this argument here.)
The Borde-Guth-Vilenkin Theorem. According to Craig’s third and final a posteriori argument,
the Borde-Guth-Vilenkin theorem (named for mathematician Arvind Borde, cosmologist Alexander
Vilenkin, and physicist Alan Guth, who proved it; henceforth, BGV) entails that any universe or
multiverse that is, on average, expanding has an absolute temporal beginning. But if so, then
BGV provides powerful scientific evidence that all physical reality had such a beginning (Craig
2008; Craig and Sinclair 2009).
A number of objections can be raised against the present argument. However, as with Craig’s
argument from the second law of thermodynamics, most of these overlap with those raised against
the previous argument and are thus only briefly summarized here. First, there is currently no scien-
tific consensus about whether BGV has the implication that all physical reality had an absolute

temporal beginning, in which case it is epistemically prudent to suspend judgment about the matter
(Morriston 2013).
Second, even if it could be shown that BGV has the implication that all physical reality had an
absolute beginning in the sense that there was an absolute beginning of physical time, that wouldn’t
entail that there was an absolute beginning to metaphysical time.
Third, even if it could be shown that BGV implies a beginning of metaphysical time, so that
nothing is temporally prior to our universe, it wouldn’t show that there is no material cause of
the universe that is causally or ontologically prior to it.
Finally, even if it could be shown that BGV implies that the universe had an absolute begin-
ning and that it has no causally or ontologically prior material causes, the universe would still lack
the relevant sort of beginning that requires a cause if it turns out that the B-theory of time is cor-
rect.
EVALUATING PREMISE (1): DOES WHATEVER BEGINS TO EXIST HAVE A CAUSE?
Recall that Craig offers both a priori and a posteriori grounds for the causal premise that whatever
begins to exist has a cause of its existence: a priori intuition and enumerative induction from univer-
sal experience. However, a number of criticisms have been raised against both the causal premise of
Craig’s kalām argument and the evidence he offers for it. Three are considered here.
First, the intuitive and empirical support for the causal premise is dependent on what is meant
by “begins to exist” (Draper 2008; Oppy 2006b). According to one gloss, it means “begins to exist
within time.” On this gloss, the empirical and intuitive evidence seems adequate to support premise
(1)—the causal premise. Unfortunately, it does so at the expense of rendering premise (2) false. For
Craig intends the latter premise to mean that the universe began to exist with the first finite interval
of time. By contrast, according to the other gloss, “begins to exist” means the sort of beginning just
mentioned—“begins to exist with time.” But on this gloss, while premise (2) seems true, it’s not at
all clear that the intuitive and empirical evidence supports premise (1). For it appears that the
empirical evidence for the causal premise only supports the “within time” gloss: we have no experi-
ence of things beginning with the origination of time itself. Furthermore, while our intuitions might
be strong that things that begin within time have a cause, it’s not at all clear what our intuitions are
about the need for a cause for things that begin to exist with time (Draper 2008; Oppy 2006b;
Morriston 2002c, 2002d).
Second, the causal premise is called into question by certain apparent counterexamples (Oppy
2006b). In one such counterexample, if, as Craig believes, God and humans have libertarian free
will, then free acts begin to exist and yet lack efficient causes. Furthermore, certain quantum entities
seem to lack efficient causes. To be sure, such events and individuals seem to have material causes,
but that is of no help to Craig’s case for a divine creator of the universe out of nothing (Oppy
2006b). It’s therefore not at all clear that the causal premise is true.
The previous point naturally leads to the third criticism, since the grounds for a requirement of
an efficient cause of the universe are on an epistemic par with the case for a requirement of a mate-
rial cause (Morriston 2002b, 2002c; Fales 2010; Leon and Rasmussen forthcoming), the latter of
which is at odds with the kalām argument’s aim to show that the universe was created by an imma-
terial person who created the universe out of nothing. Both causal principles are intuitive, and both
enjoy strong empirical confirmation. Indeed, the case for material causes is stronger, given the
apparent counterexamples to the existence of efficient causes mentioned above.
Furthermore, the theoretical costs of both are the same. We’ve never observed timeless stuff,
but we’ve never observed a timeless person, either (Morriston 2002b, 2002c; Fales 2010; Leon
and Rasmussen forthcoming). And while it’s odd to think that the material cause of the universe
was timeless sans creation, and then entered time with its creation, Craig thinks the same is true
of the efficient cause of the universe: God is timeless sans creation, but he entered time at the
moment of creation (Morriston 2002b, 2002c; Fales 2010; Leon and Rasmussen forthcoming).
Therefore, given epistemic parity, we have a dilemma: Either our commonsense intuitions about
causation can be justly applied to the beginning of the universe or they can’t. If they can, then cre-
ation of the universe out of timeless stuff is more plausible than creation ex nihilo. But if they can’t,
then we lose our grounds for inferring a cause of a universe (Morriston 2002b, 2002c, 2013).
In a similar vein, Graham Oppy (2016) argues that our empirical and intuitive evidence sup-
ports claims about causation that conflict with Craig’s claim that the universe arose from a timeless,
changeless cause, such as that all causing involves a change in the cause.
A PERSONAL CAUSE?
Recall from the first section above that Craig goes on from defending his core argument that the
universe has a caused beginning to argue that the cause is a timeless, immaterial person of immense
power. This section evaluates Craig’s core argument for a personal cause.
According to Craig (1979, 2008), there are only two main types of efficient cause: personal and
nonpersonal. But the cause of the beginning of the universe can’t be a nonpersonal cause, for any
such cause must be in a state of either quiescence or activity. But neither possibility applies to the
cause in question. For if the cause were in a state of absolute quiescence, then since no events occur
in that state, then it would remain quiescent eternally, in which case the universe would never have
arisen. On the other hand, if the cause of the beginning of the universe were in a state of activity,
then the universe would be eternal. For the effect of a nonpersonal cause occurs as soon as such a
cause is present. And if that’s right, then if the cause is eternal, then the effect is eternal. But assuming
the success of the first stage of the kalām argument, the effect is finite. Therefore, the effect—the uni-
verse—did not arise from a nonpersonal cause, whether quiescent or active (Craig 1979, 2008).
By contrast, a personal cause has the features suitable to produce an effect in the type of situa-
tion at hand. For it can (in principle, at least) exist in a state of eventless quiescence and spring into
action with a spontaneous, creative act of libertarian free will. Therefore, the universe had a begin-
ning, and it was caused by the spontaneous, free act of a person of some kind. But since it is the
efficient cause of all spatiotemporal, physical reality, it must be a timeless, immaterial being of
immense power (Craig 1979, 2008).
A number of objections have been raised against the second stage of Craig’s kalām argument.
First, the same sorts of objections Craig raises against a nonpersonal cause seem to apply with equal
force to a personal cause as well. For if an omnipotent God’s willing to create the universe is suffi-
cient for bringing about its existence (which, on orthodox views of the theistic god, it is), then, just
as with nonpersonal causes, the temporal duration of the universe should stretch back to “when”
God willed it into existence. But if we suppose with Craig that God is timeless sans creation, then
there is no change in God in this state. But if not, then it appears that his willing of the universe
must be eternal, in which case the universe should also be eternal. But this contradicts the conclu-
sion of the first stage of the kalām argument, namely, that the universe had an absolute temporal
beginning (Morriston 2000, 2002b, 2002d).
Craig anticipates the present objection and aims to get around it with the help of the distinc-
tion between God’s intending to create a universe and his undertaking it—that is, his exercising his
will to bring about that intention (2008). Given this distinction, Craig argues that it is possible
for God to eternally intend to bring about the universe, and then to freely and spontaneously exer-
cise his will to create it a finite amount of time ago.
An important rejoinder is that while Craig’s reply may have a good deal of plausibility with
respect to human action, it doesn’t seem plausible in the case of divine action. For intending and
undertaking come apart in three main types of case: (i) when you don’t yet know what you’ll decide
to do; (ii) when the time for carrying out your decision has not yet arrived; and (iii) when you have
weakness of will that (at least temporarily) prevents you from carrying out your decision. But the
problem is that none of these conditions applies to God: God is supposed to be omniscient and
thus, presumably, always knew what he would do; God is in a timeless state “when” he both intends
and undertakes to create, and thus there is no room for a temporal delay in creating; and God is
omnipotent and morally perfect, and thus can’t succumb to weakness of will. But if that’s right,
then appeal to the distinction between deciding and carrying out one’s decision won’t block the pre-
vious objection (Morriston 2000, 2002b).

Perhaps some theists will reply that God is a necessary, but not a sufficient, cause of the origi-
nation of the universe, but then it’s not clear what sense can be made of his omnipotence. It’s also
not clear why a similar account can’t apply to a nonpersonal cause of the universe’s origin (Morriston
2000, 2002b). Relatedly, perhaps some theists will reply that God is a probabilistic cause. But then
again, it’s not clear why similar accounts can’t apply to nonpersonal causes of the universe’s origin
(Oppy 2009, 2010, 2013a, 2013b). Now, to the previous point, one might reply that a purely nat-
uralistic quiescent universe can be ruled out on the grounds that it would be in a state of absolute
rest, from which no event could arise (barring supernatural intervention). However, a similar worry
arises for the hypothesis of the creation of the universe by a God who is quiescent “prior” to the cre-
ation of the universe.
The theist might instead reply that there is a way out if God eternally and timelessly wills the
universe to come into existence at a specified (metaphysical) time. However, as Evan Fales (2010)
has argued, such a reply raises further serious problems. Alternatively, the theist might reply that
there is a way out if God eternally and timelessly wills a universe with a beginning. However, as
Wes Morriston (2000) has argued, such a move generates a contradiction when combined with
other propositions Craig accepts and is committed to:
(1) Alpha (i.e., our universe) has a beginning.

(2) God’s willing-to-create-alpha is eternal.
(3) God’s willing-to-create-alpha is causally sufficient for the existence of alpha.
(4) If a cause is eternal and sufficient for the existence of something, then that thing is also
eternal.
(5) If a thing is eternal then that thing doesn’t have a beginning.
(6) Therefore, alpha both does and doesn’t have a beginning.
Something has to give, but what? Craig needs (1) for the core the kalām argument; (2) seems to fol-
low from God’s timeless eternality and omniscience; those with standard views about God’s omnip-
otence will resist denying (3); and (5) seems analytic (at least assuming an A-theory of time, which,
Craig has argued, is presupposed by his version of the argument). One might think (5) can be
resisted by saying that, while God’s eternally willing alpha makes the statement “there is a world
with a beginning” eternally true, it doesn’t make the world eternal. But this isn’t a move that would
help Craig here. For there seems no principled basis for denying this distinction for a nonpersonal
eternal sufficient condition for the beginning of the universe. And if that’s right, then such a reply
would undercut the second stage of his kalām argument (Morriston 2000).
RECAP
The kalām cosmological argument aims to show that the universe began to exist, and that it was cre-
ated out of nothing by a timeless, immaterial person of immense power. However, we’ve seen that
the argument appears to succumb to at least seven problems: (i) the scientific and philosophical
arguments for a beginning of the universe are unsuccessful; (ii) even if they were, they wouldn’t
entail the origination of the universe ex nihilo; (iii) there are serious worries about how a universe
with a finite past could arise from a timeless personal cause; (iv) even if the arguments for a finite
past and timeless cause succeed, they seem to entail something on the order of an eternal four-
dimensional block universe; (v) even if these issues can be resolved, similar solutions are available
for naturalistic universes with a finite past arising from a timeless nonpersonal cause; and (vi) the
intuitive and empirical evidence for a requirement of a material cause are at least as strong as those
for the requirement of an efficient cause, in which case the grounds for both stand or fall together.
But whether they both stand or both fall, the implications for theism are equally devastating. It
therefore seems that, at the very least, the kalām cosmological argument fails to provide evidence
that favors theism over atheism.
COSMOLOGICAL ARGUMENTS FROM CONTINGENCY
Another form of the cosmological argument makes crucial appeal to some version or other of the
principle of sufficient reason (PSR). Principles of sufficient reason assert that some range of entities
in a given domain have an explanation for why they exist or occur. Versions of the principle can
vary along a number of dimensions, such as modal force (e.g., whether they are supposed to be nec-
essarily, contingently, or possibly true, and whether the modality at issue is de dicto or de re), the
range of entities that are supposed to have an explanation (e.g., all entities whatsoever vs. some spe-
cific subset, such as contingently existing states of affairs or concrete individuals), and type of expla-
nation (e.g., entailment or necessitation vs. some weaker explanatory relation). (For further discus-
sion of the varieties of PSR, see Oppy 2006a, 2006b; Rasmussen 2010a.)
A GENERIC VERSION OF THE ARGUMENT

Once a version of the principle is selected, it can then be pressed into service as a premise in a cos-
mological argument. A huge number of versions of the cosmological argument from contingency
have been proposed, and justice cannot be done to each here. Instead, this section discusses a small
but representative sampling of such arguments. (For a fuller discussion, see Gale and Pruss 1999;
Goldschmidt 2013; Mackie 1982; O’Connor 2008; Oppy 2006a, 2006b, 2009; Pruss 1998,
2006; Rasmussen 2010a; Rowe 1998; Sobel 2004; Swinburne 2004.)
For the purposes of this section, it will prove instructive to begin by considering a fairly stan-
dard version of PSR and how it functions within a generic formulation of the argument:
(1) (a) For every object that exists, there is an explanation for why it exists, and (b) for every state
of affairs that obtains, there is an explanation for why it obtains. (Standard version of PSR)
(2) Some contingent objects exist, or some contingent states of affairs obtain (or both).
(3) Therefore, there is an explanation for why such entities exist or obtain (or both). (From 1
and 2)
(4) The explanation for why such entities exist or obtain is in terms of either (a) contingently
existing or obtaining entities, or (b) at least one necessarily existent or obtaining entity.
(5) There cannot be an explanation for why such entities exist or obtain solely in terms of
contingently existent or obtaining entities.
(6) Therefore, the existence or obtaining of contingent entities is explained in terms of at least
one necessarily existent or obtaining entity. (From 3 through 5)
The argument is valid: (3) follows from (1) and (2), and (6) follows from (3), (4), and (5). That
leaves (1), (2), (4), and (5). What can be said in support of them?
Start with (1)—in other words, PSR. Several lines of evidence have been offered in support of
the various versions of PSR (Rowe 2007; Pruss 2009; Rasmussen 2010a). First, some argue that
PSR is self-evident in the standard sense that once one understands the constituent concepts of
the principle, one can thereby “see” that the principle is true. Second, some argue that PSR enjoys
strong empirical support, where this is cashed out in terms of enumerative induction or an inference
to the best explanation. Third, some argue that PSR is a presupposition of rational thought.
Finally, it has been argued that PSR is supported by the near universal acceptance of the episte-
mic force of explicability arguments (Della Rocca 2010). In an explicability argument, one reasons
that some state of affairs does not obtain, on the grounds that it would be an inexplicable brute fact
if it did obtain. For example, Derek Parfit (1984) famously argued on those grounds that if a person
has his brain cut in half, and each half is placed in a new, different body, neither of the two resultant
persons would be the original person. This is because the grounds for saying that either one of them is
the same person as the original are the same as those for the other. But if so, then it would be an inex-
plicable brute fact if one is the original person and the other is not, and this is implausible (Parfit
1984). Explicability arguments are ubiquitous. But prima facie, all such arguments are legitimate if
any are. But to say that all explicability arguments are legitimate is just to say that PSR is true.

Two main types of evidence are standardly offered in favor of (2): empirical evidence and
modal evidence. According to the first, we observe objects coming to be and passing away, and
we see states of affairs obtaining and then failing to obtain. But necessary beings can’t come to be
or pass away. By contrast, contingent beings can come to be and pass away.
According to the second, we can imagine or conceive of a different universe existing instead of
this one. We can also imagine or conceive of each thing in the universe, and even the universe as
a whole, failing to exist. But whatever we can imagine or conceive is prima facie possible. Therefore,
it’s prima facie possible for the universe to have been different, and to fail to exist (or to fail to have
existed). An important line of modal evidence for the latter claim is the subtraction argument (see,
e.g., Baldwin 1996; Rodriguez-Pereyra 1997, 2013). According to one version of the subtraction
argument, one starts by considering a scenario in which a tiny particle is annihilated from the uni-
verse and asking whether this scenario is metaphysically possible. But if one allows that it is, then
there seems to be no principled grounds for denying the possibility that two, three, or any number
of particles be annihilated, one by one, down to the very last one. But if that’s right, then it follows
that it’s possible for each thing in the universe, and indeed the universe itself, to fail to exist. But no
necessary being can fail to exist, whether in the actual world or in any other possible world. There-
fore, our modal evidence indicates that the universe is not a necessary being but rather a contingent
being.
Premise (3) above can be supported by an argument from elimination. The broadest and most
permissible categorization of entities comprises impossible beings, merely possible beings, contin-
gent beings, and necessary beings. But prima facie, only actual beings can explain actual beings.
But that rules out impossible beings and merely possible beings for explaining the existence of con-
tingent beings. Therefore, the only candidates for explaining some range of contingent beings are
contingent beings and necessary beings.
Finally, in support of (5), theists standardly argue that if everything were a contingent entity,
then at least some such entities would lack an explanation. Consider the totality of contingent con-
crete entities. This totality might be the mereological fusion of all contingent concrete objects, or it
might be the conjunction of all contingent facts, events, truths, or states of affairs (cf. Oppy 2009).
But whichever it is, it will here be called the conjunction of all contingent entities (CCE). Now sup-
pose that nothing exists or obtains but CCE. Then there is no explanation for why CCE exists or
obtains at all, instead of just nothing. There is also no explanation why CCE exists or obtains
instead of some other possible contingent conjunctive entity. Therefore, if everything were a contin-
gent being, then at least some things would lack an explanation for why they exist or obtain. But
that is just to say that (5) is true.
A standard criticism of the argument is that the version of PSR in play absurdly entails that
everything exists and obtains of necessity (Ross 1969; van Inwagen 1983, 2002; Rowe 1998).
The reasoning behind this criticism can be stated as follows. Consider again CCE. By PSR, there
is a sufficient reason for CCE. Now the sufficient reason for CCE is itself either contingent or
necessary. But it can’t be contingent, because then it would itself be a part of CCE. But contingent
facts don’t contain within themselves the sufficient reason for why they obtain, let alone the suffi-
cient reason for why CCE obtains. Thus the sufficient reason for CCE must be necessary. But
whatever is entailed by a necessary truth is itself necessary, in which case all truths would be neces-
sary truths, and the referents they represent would obtain of necessity. But this is absurd. Therefore,
PSR is false.
Another criticism of PSR is that it admits of counterexamples (Oppy 2006a, 2006b; Rasmussen
2010a). In one such counterexample, on standard interpretations of quantum mechanics, certain
quantum events lack sufficient reasons or explanations for their occurrence. Another apparent coun-
terexample is freedom of the will. If humans have libertarian freedom of the will, then our free
actions lack sufficient reasons or explanations that entail their occurrence. Finally, it has been argued
that a counterexample to PSR seems to follow from the problem of the many (Kleinschmidt 2013).
To wit, consider the set of atoms of which you are constituted. There are any number of proper
subsets of this set that are equally good candidates for the set that constitutes you. One is the subset
of all of the original set minus one atom at the top of your left ear, and another is the subset of the
original set minus one atom at the top of your right ear. But there is only one “you” there. There-
fore, there seems to be no sufficient reason for why you are one of these sets of atoms rather than
another.
Another criticism of the argument takes aim at our empirical and modal evidence for contin-
gent entities. Start with the empirical evidence. It’s true that we observe things made out of matter
coming to be and passing away. It’s also true that this is evidence of their contingency. However,
that’s only evidence that things made out of matter are contingent entities; it does nothing to show
that the existence of matter is contingent (Leon and Rasmussen forthcoming). For all that such evi-
dence shows, the latter could turn out to be a necessary being. And while, strictly speaking, such an
epistemic possibility doesn’t call into question the argument’s soundness, it is of course not the kind
of necessary being the theist wants to infer from it.
The modal evidence is even more problematic for the argument (Hume 1993 [1779]; Oppy
2006b; Leon and Rasmussen forthcoming). For either imaginability or conceivability is prima facie
evidence of possibility or it isn’t. If it is, then we can imagine or conceive a different conjunction of
objects than CCE. We can also imagine or conceive anything and everything—including God—
being different than the way it is, and of anything and everything failing to exist. Thus one might
continue the subtraction argument discussed previously and subtract God from the universe after
one subtracts the last physical particle, thereby erasing in thought all concrete objects from the
world. But if so, then one is prima facie justified in believing that it’s possible that anything and
everything could have been different, and that everything—including God—could have failed to
exist. But if that’s right, then one is prima facie justified in believing that there are no necessary
beings. On the other hand, if imaginability or conceivability isn’t prima facie evidence of possibility,
then our ability to imagine or conceive of a different universe, or of no universe at all, isn’t prima
facie evidence that such things are possible. But if not, then we lose our modal evidence of the
CCE’s contingency.
RASMUSSEN’S VERSION
Joshua Rasmussen (2011) offers an original version of the contingency argument. Stripped down to
its essentials, the argument can be stated as follows:
(1) Normally, things that can begin to exist can have a cause of the beginning of their existence.
(2) Contingent concrete reality can begin to exist.
(3) Therefore, there can be a cause of the beginning of contingent concrete reality’s existence.
(4) If there can be a cause of the beginning of contingent concrete reality’s existence, then a
necessary being exists.
(5) Therefore, a necessary being exists.
Rasmussen’s contingency argument is a significant improvement over previous versions. Perhaps the
most obvious improvement is its version of PSR. The principle is considerably more modest in what
it asserts, and in ways that make it easier to support. For example, the “normally” operator in its
PSR indicates that the principle is a defeasible rule of thumb that holds in ordinary circumstances
but which may admit of exceptions. As such, it can’t be refuted by simple counterexamples. To
avoid the demands of a well-supported defeasible principle, one must give principled grounds for
thinking that it admits of an exception in the particular case at stake (Koons 2017). Furthermore,
the principle restricts the requirement for a cause to things that begin to exist. Finally, the principle
only asserts that such things can have a cause, not that they do have a cause.
What can be said on behalf of the premises? Premise (1) can be supported in the standard ways
mentioned earlier. As for premise (2), it asserts that it is metaphysically possible that there is a
beginning to the existence of contingent concrete reality. There are two main ways in which one
might attempt to support (2). First, one might argue that it’s actually true that all contingent con-
crete reality had a beginning to its existence, and since whatever is actually the case is possibly the
case, it follows that it’s possible that all contingent concrete particulars had a beginning to their

existence. So, for example, one might appeal to the big bang theory, or some other line(s) of scien-
tific evidence, to support the claim that the physical universe (actually and thus possibly) began to
exist.
Another way in which one might support (2) is via modal evidence, such as an inference from
imaginability or conceivability to possibility. According to this line of reasoning, our ability to imag-
ine such a scenario—that is, a scenario involving a beginning to the existence of all contingent con-
crete reality—constitutes sufficient prima facie evidence of its metaphysical possibility. This means
of supporting (2) seems suggested by Rasmussen’s remark that one can “imagine a beginning to
the existence of contingent bits of matter as they explode out of an initial singularity” (2011, 4;
emphasis added).
At least three criticisms can be raised against (2). First, the empirical evidence in support of (2)
is inadequate. For as was seen in the section on the kalām argument, the scientific evidence doesn’t
support the claim that the universe had an absolute beginning.
Second, the modal evidence for (2) is likewise inadequate. For there are two different sorts of
states of affairs one might be asked to imagine here: a beginning to the existence of the concrete par-
ticulars of the actual world, or a beginning to the existence of those of some other possible world.
Take the former first. The problem is that we immediately fall afoul of the problem of conceivability-
possibility inferences in contexts involving a posteriori necessities (Kripke 1980). To see this, sup-
pose we give our universe a Kripkean baptism: We say (pointing to the universe), “Let that be
called ‘Uni.’” “Uni” is now a rigid designator, and thus refers only to our universe in all the possi-
ble worlds in which it exists. Holding our universe fixed via the term Uni, we can start considering
modal claims about it. There are two relevant possibilities for us to consider in this regard: (i) Uni
has its origin in the causal activity of a necessary being, and (ii) Uni has no such origin—that is, it
is either eternal or it is a member of a beginningless series of contingent concrete particulars. Now
if (i) is true, then by origin essentialism, this is an essential property of Uni, in which case there is
no possible world in which Uni lacks such an origin. On the other hand, if (ii) is true, then Uni
lacks an origin in the causal activity of a necessary being, and so this fact about Uni is essential
to it, in which case there is no possible world in which it has an origin in the causal activity of a
necessary being.
The moral, then, is that if one accepts origin essentialism, then one will think that facts about
whether our universe has an explanation in terms of a necessary being don’t vary from possible
world to possible world. But if not, then we can’t know whether our universe could have a begin-
ning in the causal activity of a necessary being unless we know beforehand that it in fact had such
a beginning. But of course, if one knew that, then the argument for a necessary being would be
superfluous.
But suppose one takes instead the second sort of candidate referent of one’s act of imagining:
the beginning to the existence of some other possible collection of concrete particulars distinct from
those that actually exist. Would the imaginability of such a scenario adequately support premise (2)?
Not obviously. For even if it is granted that imaginability can provide sufficient justification
for many possibility claims, there is a growing trend in modal epistemology that the justification-
conferring ability of such imaginings does not extend to states of affairs as remote from ordinary
experience as the beginning of all concrete particulars, any more than it does to the states of affairs
denoted by the modal premises in (for example) the modal ontological argument (possibly, an
Anselmian being exists) and conceivability arguments for dualism (possibly, I can exist apart from
my body) (cf. Seddon 1972; van Inwagen 1998; Williamson 2007; Fischer and Leon 2016, 2017;
Leon 2017. For further discussion of this point, see Leon, “A Priori: Atheism,” present volume).
Finally, some argue that all concrete objects that have an originating or sustaining efficient
cause have an originating or sustaining material cause (Leon and Rasmussen forthcoming). But if
so, then there cannot be a beginning to the existence of contingent concrete reality in the sense
required for a theistic conclusion, namely, one where the first contingent concrete particulars were
created ex nihilo.
What about premise (4)? The basic line of reasoning is that (setting aside impossible beings and
merely possible beings, for the reasons given in the earlier discussion of a generic contingency
argument) there are only two types of beings that could explain the beginning of the existence of all
contingent beings: contingent beings and necessary beings. But since nothing can be causally prior
to itself, contingent beings can’t cause the beginning of the existence of contingent beings. There-
fore, a necessary being must be the cause of the beginning of contingent beings if such a beginning
is possible. But by (2), there is at least one such possible world. Therefore a necessary being exists in
at least one possible world. But by axiom S5 of S5 modal logic, whatever is necessary in one possible
world is necessary in every possible world, including the actual world. Therefore, if a beginning of
the existence of contingent concrete reality is possible, then a necessary being exists. But that is just
to say that premise (4) is true.
One criticism of (4) pertains to the twofold categorization of objects assumed by the argument
(Leon and Rasmussen forthcoming). According to this categorization, all contingent beings—beings
that exist in the actual world, but not in others—are dependent on other concrete objects for their
existence. By contrast, necessary beings—beings that exist not only in the actual world, but in all
possible worlds—do not depend on other beings for their existence but are explained by their
own inner nature.
However, contingent independent beings seem epistemically possible as well (Hick 1961;
Rowe 1998; Swinburne 1994). So, for example, Richard Swinburne (1994) argues that the first
person of the trinity of Christian theology is just such a being. According to his account, there
are possible worlds in which God the father doesn’t exist. However, he is an existentially indepen-
dent, freestanding being who is uncaused, uncreated, eternal, and indestructible at all the worlds
in which he does exist. Typically, philosophers of religion who accept such a view of God also take
all the other existing concrete beings to be contingent beings that depend on him for their exis-
tence. It is therefore common for such philosophers to speak of God’s existence as “necessary”
in the relative sense of its being necessary for the existence of other beings—contingent dependent
beings. Let us follow such philosophers in referring to contingent independent beings as factually
necessary beings.
Given the epistemic possibility of this broader categorization of possible types of beings, one
cannot automatically infer “dependent being” from “contingent being.” For then it is epistemically
possible that all contingent dependent beings are ultimately composed of contingent independent
beings, that is, factually necessary beings. So, for example, perhaps matter-energy (or whatever
matter-energy is ultimately composed of) is a factually necessary being. According to such a sce-
nario, the contingent dependent beings (e.g., rocks, trees, planets, you and I) come into being when
two or more contingent independent beings (i.e., factually necessary beings) are combined, and the
contingent dependent beings cease to exist when they decompose into their elements. However, the
fundamental elements of which contingent dependent beings are composed (i.e., the contingent
independent beings/factually necessary beings) cannot pass away, for they are at least de facto
indestructible—in other words, nothing in the actual world has what it takes to knock them out
of existence. Nor can they be created, for they are eternal, existentially independent, and (assuming
origin essentialism and their being uncaused at the actual world) essentially uncaused.
In this scenario, then, we have an explanation for all contingent dependent beings in terms of
contingent independent beings. Furthermore, we have an explanation of contingent independent
beings, partly in terms of the factual necessity of their own nature (i.e., in terms of their being
uncreated, eternal, and existentially independent), and partly in terms of the character of the world
at which they exist. (They are indestructible at least partly in virtue of there being nothing around
in the world that can knock them out of existence.) But since this scenario is epistemically possible,
(4) is undercut.
THEISM, ATHEISM, AND THE CONTINUED EXISTENCE OF

THE UNIVERSE
One can find, through the writings of Lucretius, a powerful yet simple Epicurean argument for mat-
ter’s (factual or metaphysical) necessity. In simplest terms, the argument is that, since matter exists,

and since nothing can come from nothing (in the sense that everything with an originating or sus-
taining efficient cause needs an originating or sustaining material cause, respectively), matter is
eternal and uncreated. The argument can be strengthened in light of the scientific evidence for
the conservation laws, according to which it is at least physically impossible that matter-energy
is created or destroyed. And if there are no supernatural beings that can annihilate matter-
energy, the latter is at least de facto indestructible. Therefore, given the uncreated, eternal,
and de facto indestructibility of matter-energy, it follows that matter-energy (or if matter-energy
is not fundamental, whatever matter-energy is ultimately made of) is at least a factually neces-
sary being.
A stronger version of Epicurus’s core argument can be developed by adding an appeal to some-
thing in the neighborhood of origin essentialism. The basic line of reasoning is that, if being uncre-
ated is a property of matter-energy in the actual world, then it is an essential property of matter-
energy, in which case matter-energy in the actual world is essentially uncreated.
Yet stronger versions of the argument can go on from what is said above by appealing to a
strong version of the PSR to argue that whatever plays the role of being eternal, essentially uncre-
ated, and indestructible does not vary from possible world to possible world. But if not, then matter
is a metaphysically necessary being. On any version of the argument, however, we seem to get the
conclusion that the universe requires no external sustaining cause, in which case, a fortiori, God is
not required to play such a role.
The broadly Epicurean line of reasoning above can be seen as a cosmological argument of sorts,
but one that concludes that matter-energy (or its ultimate constituents), and not an immaterial cre-
ator, is the uncaused cause of contingent, dependent, concrete reality. Let us therefore call any argu-
ment that deploys a material-cause version of the principle ex nihilo nihil fit to infer the factual or
metaphysical necessity of matter (or matter’s ultimate constituents) an Epicurean cosmological argu-
ment.
If successful, Epicurean cosmological arguments can be used to provide evidence in support of
atheism over theism. For such arguments provide prima facie evidence that matter-energy (or its
ultimate constituents) are factually or metaphysically necessary. But if so, then since it is constitutive
of classical theism that God is the creator of any material universe that happens to exist, then since
an essentially uncreated universe exists in the actual world, and since essentially uncreated universes
cannot, by definition, be created, it follows that the God of classical theism does not exist. Indeed,
if, as many classical theists assert, God exists necessarily if he exists at all, then given that he does not
exist in the actual world, God exists in no possible world. In other words, God’s existence is meta-
physically impossible.
FINAL RECKONING
This chapter explores cosmological arguments that deploy causal or explanatory principles to prove
God’s existence. Both the kalām cosmological argument and contingency arguments face a number
of powerful criticisms. Furthermore, it is not at all clear that such criticisms can be successfully
answered. Finally, there are reasons to think that the universe or its ultimate constituents are factu-
ally or metaphysically necessary beings. Therefore, considerations respecting causation and sufficient
reason appear to favor atheism over theism.
Bibliography
Alter, Torin, and Yujin Nagasawa. “What Is Russellian Monism?” In Beebee, Helen, Christopher Hitchcock, and Peter Menzies. The Oxford
Consciousness in the Physical World: Perspectives on Russellian Monism, Handbook of Causation. Oxford: Oxford University Press, 2009.
edited by Alter and Nagasawa, 422–451. Oxford: Oxford University
Craig, William Lane. The Kalām Cosmological Argument. London: Macmillan,
Press, 2015.
1979.
Baldwin, Thomas. “There Might Be Nothing.” Analysis 56, no. 4 (1996): Craig, William Lane. Review of Time, Creation and the Continuum, by Richard
231–238. Sorabji. International Philosophical Quarterly 25, no. 3 (1985): 319–326.
Craig, William Lane. “The Kalām Cosmological Argument and the Hoffman, Paul. “Does Efficient Causation Presuppose Final Causation?
Hypothesis of a Quiescent Universe.” Faith and Philosophy 8, no. 1 Aquinas vs. Early Modern Mechanism.” In Metaphysics and the Good:
(1991): 104–108. Themes from the Philosophy of Robert Merrihew Adams, edited by Samuel
Newlands and Larry M. Jorgensen, 295–312. Oxford: Oxford University
Craig, William Lane. “The Origin and Creation of the Universe: A Reply to
Adolf Grünbaum.” British Journal for the Philosophy of Science 43, no. 2 Press, 2009.
(1992): 233–240. Hume, David. Dialogues concerning Natural Religion. (1779.) In Dialogues
Craig, William Lane. “The Cosmological Argument.” In The Rationality of and Natural History of Religion. Edited by J. A. C. Gaskin. Oxford:
Theism, edited by Paul Copan and Paul K. Moser, 112–131. London: Oxford University Press, 1993.
Routledge, 2003. Kleinschmidt, Shieva. “Reasoning without the Principle of Sufficient
Craig, William Lane. Reasonable Faith: Christian Truth and Apologetics. 3rd Reason.” In The Puzzle of Existence: Why Is There Something Rather
ed. Wheaton, IL: Crossway Books, 2008. Than Nothing?, edited by Tyron Goldschmidt, 64–79. New York:
Routledge, 2013.
Craig, William Lane, and Quentin Smith. Theism, Atheism, and Big Bang
Cosmology. New York: Oxford University Press, 1993. Koons, Robert. “Defeasible Reasoning.” In Stanford Encyclopedia of
Philosophy, edited by Edward N. Zalta. Winter 2017. https://plato
Craig, William Lane, and James Sinclair. “The Kalām Cosmological .stanford.edu/archives/win2017/entries/reasoning-defeasible/.
Argument.” In The Blackwell Companion to Natural Theology, edited
by William Lane Craig and J. P. Moreland, 101–201. Malden, MA: Kripke, Saul. Naming and Necessity. Cambridge, MA: Harvard University
Wiley-Blackwell, 2009. Press, 1980.
Davis, Stephen T. “The Cosmological Argument and the Epistemic Status of Leon, Felipe. “Moreland on the Impossibility of Traversing the Infinite: A
Belief in God.” Philosophia Christi 1, no. 1 (1999): 5–16. Critique.” Philo 14, no. 1 (2011): 32–42.
Della Rocca, Michael. “PSR.” Philosophers’ Imprint 10, no. 7 (2010): 1– Leon, Felipe. “From Modal Skepticism to Modal Empiricism.” In Modal
13. Epistemology after Rationalism, edited by Bob Fischer and Felipe Leon,
247–262. Cham, Switzerland: Springer International, 2017.
Draper, Paul. “A Critique of the Kalām Cosmological Argument.” In
Philosophy of Religion: An Anthology, 5th ed., edited by Louis P. Pojman Leon, Felipe, and Joshua Rasmussen. Is God the Best Explanation? London:
and Michael Rea. Belmont, CA: Thomson/Wadsworth, 2008. Palgrave Macmillan, forthcoming.
Everitt, Nicholas. The Non-existence of God. London: Routledge, 2004. Mackie, J. L. The Miracle of Theism: Arguments for and against the Existence of
God. Oxford: Clarendon Press, 1982.
Fales, Evan. Divine Intervention: Metaphysical and Epistemological Puzzles.
London: Routledge, 2010. Maitzen, Stephen. “Stop Asking Why There’s Anything.” Erkenntnis 77, no.
1 (2012): 51–63.
Fischer, Bob. Modal Justification via Theories. Cham, Switzerland: Springer
International, 2017. Maitzen, Stephen. “Questioning the Question.” In The Puzzle of Existence:
Why Is There Something Rather Than Nothing?, edited by Tyron
Fischer, Bob, and Felipe Leon. “The Modal-Knowno Problem.” Southwest Goldschmidt, 252–271. New York: Routledge, 2013.
Philosophy Review 32, no. 1 (2016): 225–232.
Markosian, Ned. “Restricted Composition.” In Contemporary Debates in
Fischer, Bob, and Felipe Leon, eds. Modal Epistemology after Rationalism. Metaphysics, edited by Theodore Sider, John Hawthorne, and Dean
Cham, Switzerland: Springer International, 2017. Zimmerman, 341–360. Malden, MA: Blackwell, 2008.
Gale, Richard, and Alexander Pruss. “A New Cosmological Argument.” Monton, Bradley. “Prolegomena to Any Future Physics-Based Metaphysics.”
Religious Studies 35 (1999): 461–476. In Oxford Studies in Philosophy of Religion. Vol. 3, edited by Jonathan
Goldschmidt, Tyron, ed. The Puzzle of Existence: Why Is There Something Kvanvig. Oxford: Oxford University Press, 2011.
Rather Than Nothing? New York: Routledge, 2013.
Moreland, J. P. “The Kalām Cosmological Argument.” In Philosophy of
Grünbaum, Adolf. “The Pseudo-problem of Creation in Physical Cosmol- Religion: Selected Readings, 2nd ed., edited by Michael Peterson, William
ogy.” Philosophy of Science 56, no. 3 (1989): 373–394. Hasker, Bruce Reichenbach, and David Basinger, 196–208. New York:
Grünbaum, Adolf. “Creation as a Pseudo-Explanation in Current Physical
Cosmology.” Erkenntnis 35 (1991): 233–254. Morriston, Wes. “Must the Past Have a Beginning?” Philo 2, no. 1 (1999):
5–19.
Hanrahan, Rebecca. “Imagination, Possibility, and Plovers.” In Modal
Epistemology after Rationalism, edited by Bob Fischer and Felipe Leon, Morriston, Wes. “Must the Beginning of the Universe Have a Personal
197–220. Cham, Switzerland: Springer International, 2017. Cause?” Faith and Philosophy 17, no. 2 (2000): 149–169.
Hawke, Peter. “Van Inwagen’s Modal Skepticism.” Philosophical Studies 153, Morriston, Wes. “Craig on the Actual Infinite.” Religious Studies 38, no. 2
no. 3 (2011): 351–364. (2002a): 147–166.
Hick, John. “Necessary Being.” Scottish Journal of Theology 14 (1961): 353– Morriston, Wes. “A Critical Examination of the Kalām Cosmological
369. Argument.” In God Matters: Readings in the Philosophy of Religion, edited
by Raymond Martin and Christopher Bernard, 95–108. New York:
Hilbert, David. “Über das Unendliche.” In David Hilbert’s Lectures on the
Longman, 2002b.
Foundations of Arithmetic and Logic, 1917–1933. Edited by William
Ewald, Wilfried Sieg, Ulrich Majer, and Dirk Schlimm, 668–756. Morriston, Wes. “Creation Ex Nihilo and the Big Bang.” Philo 5, no. 1
Berlin: Springer, 2013. (2002c): 23–33.

Morriston, Wes. “Causes and Beginnings in the Kalām Argument: Reply to Roca-Royes, Sonia. “Similarity and Possibility: An Epistemology of de re
Craig.” Faith and Philosophy 19, no. 2 (2002d): 233–244. Possibility for Concrete Entities.” In Modal Epistemology after Rational-
ism, edited by Bob Fischer and Felipe Leon, 221–246. Cham,
Morriston, Wes. “Must Metaphysical Time Have a Beginning?” Faith and
Switzerland: Springer International, 2017.
Philosophy 20, no. 3 (2003): 288–306.
Rodriguez-Pereyra, Gonzalo. “There Might Be Nothing: The Subtraction
Morriston, Wes. “Beginningless Past, Endless Future, and the Actual
Argument Improved.” Analysis 57, no. 3 (1997): 159–166.
Infinite.” Faith and Philosophy 27, no. 4 (2010): 439–450.
Rodriguez-Pereyra, Gonzalo. “The Subtraction Arguments for Metaphysical
Morriston, Wes. “Doubts about the Kalām Argument.” In Debating
Nihilism: Compared and Defended.” In The Puzzle of Existence: Why Is
Christian Theism, edited by J. P. Moreland, Chad V. Meister, and
There Something Rather Than Nothing?, edited by Tyron Goldschmidt,
Khaldoun A. Sweis., 20–34. Oxford: Oxford University Press, 2013.
197–214. New York: Routledge, 2013.
O’Connor, Timothy. Theism and Ultimate Explanation: The Necessary Shape
Ross, James F. Philosophical Theology. Indianapolis, IN: Bobbs-Merrill, 1969.
of Contingency. Malden, MA: Blackwell, 2008.
Rowe, William L. The Cosmological Argument. New York: Fordham
Oppy, Graham. Philosophical Perspectives on Infinity. Cambridge, UK:
Cambridge University Press, 2006a.
Rowe, William L. Philosophy of Religion: An Introduction. 4th ed. Belmont,
Oppy, Graham. Arguing about Gods. New York: Cambridge University
CA: Wadsworth/Thomson, 2007.
Press, 2006b.
Rundle, Bede. Why There Is Something Rather Than Nothing. Oxford: Oxford
Oppy, Graham. “Cosmological Arguments.” Noûs 43, no. 1 (2009): 31–48.
Oppy, Graham. “Uncaused Beginnings.” Faith and Philosophy 27 (2010):
Schaffer, Jonathan. “The Metaphysics of Causation.” In Stanford Encyclope-
61–71.
dia of Philosophy, edited by Edward N. Zalta. Fall 2016. https://plato
Oppy, Graham. The Best Argument against God. New York: Palgrave .stanford.edu/archives/fall2016/entries/causation-metaphysics/.
Macmillan, 2013a.
Seddon, George. “Logical Possibility.” Mind 81, no. 324 (1972): 481–494.
Oppy, Graham. “Ultimate Naturalistic Causal Explanations.” In The Puzzle
Sobel, Jordan Howard. Logic and Theism: Arguments for and against Belief in
of Existence: Why Is There Something Rather Than Nothing?, edited by
God. Cambridge, UK: Cambridge University Press, 2004.
Tyron Goldschmidt, 46–63. New York: Routledge, 2013b.
Swinburne, Richard. The Christian God. Oxford: Clarendon; New York:
Oppy, Graham. “Divine Causation.” Topoi 36, no. 4 (2016): 641–650.
Parfit, Derek. Reasons and Persons. Oxford: Clarendon, 1984.
Swinburne, Richard. The Existence of God. 2nd ed. Oxford: Clarendon; New
Paul, L. A., and Ned Hall. Causation: A User’s Guide. Oxford: Oxford York: Oxford University Press, 2004.
Taliaferro, Charles. “Sensibility and Possibilia: A Defense of Thought
Pruss, Alexander. “The Hume-Edwards Principle and the Cosmological Experiments.” Philosophia Christi 3, no. 2 (2001): 403–420.
Argument.” International Journal for the Philosophy of Religion 43 (1998):
Taylor, Richard. Metaphysics. Englewood Cliffs, NJ: Prentice-Hall, 1992.
149–165.
Van Inwagen, Peter. An Essay on Free Will. Oxford: Clarendon; New York:
Pruss, Alexander. The Principle of Sufficient Reason: A Reassessment.
Van Inwagen, Peter. Material Beings. Ithaca, NY: Cornell University Press,
Pruss, Alexander. “The Leibnizian Cosmological Argument.” In The
1995.
Blackwell Companion to Natural Theology, edited by William Lane
Craig and J. P. Moreland, 24–100. Malden, MA: Wiley-Blackwell, Van Inwagen, Peter. “Why Is There Anything at All?” Proceedings of the
2009. Aristotelian Society 70 (1996): 95–120.
Rasmussen, Joshua. “Cosmological Arguments from Contingency.” Philoso- Van Inwagen, Peter. “Modal Epistemology.” Philosophical Studies 92, no. 1
phy Compass 5, no. 9 (2010a): 806–819. (1998): 67–84.
Rasmussen, Joshua. “From States of Affairs to a Necessary Being.” Van Inwagen, Peter. Metaphysics. 2nd ed. Boulder, CO: Westview Press,
Philosophical Studies 148, no. 2 (2010b): 183–200. 2002.
Rasmussen, Joshua. “A New Argument for a Necessary Being.” Australasian Williamson, Timothy. The Philosophy of Philosophy. Malden, MA: Blackwell,
Journal of Philosophy 89, no. 2 (2011): 351–356. 2007.
TOPIC 10
A Priori: Theism
Aaron Segal
Lecturer, Department of Philosophy
Hebrew University of Jerusalem, Israel
This chapter discusses whether the existence of a priori knowledge bears positively on theism, and argues
tentatively that it does.
WHAT IS THE A PRIORI? WHAT IS THEISM?
You know as well as I do—and without needing to trust anyone else on the matter—that 2 + 2 = 4,
and that it’s wrong to bludgeon innocent people to death for no reason, and that nothing can be both
entirely red and entirely green. But how do you know these things? It’s not as though you can see
or hear or smell or feel or taste your way to knowing them. It’s true that very often when we put
two things together with two other things we end up with four things. But not always: things go
missing. Haven’t you ever lost a sock? And it’s true that we can see the gory physical consequences
of a bludgeoning, but it doesn’t seem like you can likewise see the wrongness of the bludgeoning.
And it’s true that we’ve never encountered anything that is entirely red and entirely green; and
maybe we can know on that basis that we never will. But how can you know on that basis that
you couldn’t encounter such a thing—in other words, that it’s not possible to encounter such a
thing? It’s not as though you can peer into all the nonactual scenarios and check.
Faced with cases such as these—cases of mathematical, moral, and (so-called) modal knowledge
(knowledge of what is necessary and what is merely possible)—philosophers have historically reacted
in one of three ways. Some philosophers argue that, contrary to initial appearances, you don’t really
know what I took you to know: either because these claims are not of the right sort to be known
(maybe they are not claims at all but expressions of emotion or the laying down of rules) or because
you (like the rest of us) are not properly equipped to know them (see, e.g., Beebe 2011). Other phi-
losophers maintain that you do know these claims but argue that, contrary to appearances, you can
come to know them (ultimately) on the basis of your seeing, hearing, smelling, feeling, and tasting
things (see, e.g., Mill 1884; Quine 1951; Devitt 2005). However, given the difficulties facing both
of these views, some philosophers accept these cases at face value and conclude that you have knowl-
edge that doesn’t depend on sensory experience (see, e.g., BonJour 1998; Bealer 1996, 1998).
Knowledge of that sort is called a priori knowledge.
Going forward, I will simply assume that at least some humans have at least some a priori
knowledge. Indeed, I will assume more specifically that at least some humans have at least some a
priori knowledge of a certain kind of truth. Say that a sentence is analytic if and only if it is synony-
mous with some logical truth, and say that it is synthetic otherwise. I will assume that at least some
humans have at least some a priori knowledge of some synthetic truth. Let us abbreviate that claim
by saying “there is a priori knowledge.” There are arguments for this assumption—including an
argument based on the premise that neither of the other two reactions to the cases I’ve mentioned
is tenable, in conjunction with the plausible (although admittedly controversial) premise that those
truths are synthetic—and to my mind some of the arguments are compelling. But I can hardly do
301
A Priori: Theism
justice to those arguments in the space I have here. (For much

KEY CONCEPTS more discussion, see BonJour 1998; Bealer 1996, 1998.) My
focus will instead be on the question: Granted that there is a
priori knowledge, what follows from that fact, if anything, for
A Priori Knowledge
the theism/atheism debate?
Abstract Objects Before addressing that question, let me say a little more
Bayes’s Theorem about what I mean by “a priori knowledge” and “theism.”
Conditional Likelihood Providing a problem-free and fully illuminating definition
of a priori knowledge is an extremely difficult task (see Casullo
Divine Conceptualism
2009 for a valiant attempt). Even characterizing a priori knowl-
Evolutionary Argument against edge any more precisely than I have already done is nontrivial.
Naturalism But several clarifications are nevertheless in order. First, the
Posterior Probability sense in which a priori knowledge is supposed to be indepen-
dent of experience is rather circumscribed. It is not meant to
Prior Probability
rule out a role for experience in coming to understand the rele-
vant proposition; it is not meant to rule out the possibility of a
future experience undercutting or rebutting or otherwise defeat-
ing whatever (nonexperiential) reason one has for believing the proposition; and it is not meant to
rule out the need for a subject’s environment to “cooperate” in order for the subject to have a priori
knowledge (see Hawthorne 2007). Second, I said that a priori knowledge doesn’t depend on “sen-
sory experience,” but in truth it’s supposed to be independent of experience more broadly. What
this broad notion of experience includes is far from clear: it’s usually taken to include at least sen-
sory experience, kinesthetic experience, memorial experience, introspective experience, and testi-
monial experience. Third, the characterization is silent as to whether a priori knowledge still
depends on something, and if so, on what. This is deliberate. Some have argued that a priori
knowledge (at least when it’s not inferred from other a priori knowledge) is based on a so-called
rational intuition, by which is meant either an “intellectual seeming” of the truth of the proposi-
tion (Bealer 1998), or, more strongly, an “intellectual seeing” that the proposition is true, and
maybe even why it’s true (BonJour 1998). But as far as I’m concerned that’s a substantive claim
about a priori knowledge, not a constituent of the very concept itself. Fourth, I have given an ini-
tial characterization of a priori knowledge. Others would make a priori justification the target of
their analysis and investigation. Since justification is widely thought to be a constituent of knowl-
edge or at least a necessary condition for knowledge, the assumption that there is a priori knowl-
edge is stronger than the assumption that there is a priori justification. Thus it should be
acknowledged that even if the existence of a priori knowledge supports theism, the existence of
a priori justification might not. Fifth, and finally, we can speak not only of a priori knowledge
(and justification), but of a proposition itself being a priori. A proposition is a priori when it is
possible for some (human) subject to know the proposition a priori. (There is another, weaker
sense of a priori proposition: a proposition is a priori in the weaker sense when either it or its
negation is a priori in my stronger sense. In that weaker sense a false elementary arithmetical
proposition is presumably a priori, if anything is a priori at all. But as a priori knowledge is a tech-
nical term, I feel comfortable using it in the sense most convenient for my purposes.) As we shall
see, certain arguments for the bearing of a priori knowledge on theism are best understood as
arguments for the bearing of a priori propositions on theism.
As to theism: It’s important not to pack too much into our characterization of theism. That
would make it unduly hard to argue for it deductively and unduly easy to confirm it probabilisti-
cally. By the same token it’s important not to pack too little into our characterization of theism.
That would make it unduly easy to argue for it deductively and unduly hard to confirm it proba-
bilistically. (Probabilistic confirmation is discussed further in the section “Probabilistic Bearing.”)
More simply even, it’s important not to pack too much or too little into any thesis; one should pack
in no more and no less than what the thesis says.
My best attempt to capture what is standardly meant by “theism” uses the following view: there
is exactly one all-powerful, all-knowing, and wholly good being—a Supreme Being, for short—and
A Priori: Theism
that Supreme Being (at least ultimately) brought into existence all this. When I say “all this,” I am
pointing to, well, everything I can point to; or, more exactly, everything with which I could interact;
or, more exactly still, everything other than the Supreme Being that has any causal powers at all: from
my computer and my children to the sun, the moon, and distant galaxies. Note well that theism, as
I’ve characterized it at least, says nothing of any special concern that the Supreme Being has for
humanity. (Perhaps it follows from being wholly good that the Supreme Being is concerned with
humanity, but not especially concerned.) And theism, again as I’ve characterized it, says nothing of
who or what produced things that lack any causal powers, if such there be.
EPISTEMOLOGICAL BEARING
One way in which the existence of a priori knowledge—or, more exactly, a priori propositions—
might bear on the theism/atheism debate is for it to bear on our knowledge of theism, whether or
not it bears on the truth of theism. Suppose theism is in fact true. Suppose, in fact, that there is a
valid argument for theism, such that (a) if there are a priori propositions, then all of the argument’s
premises can be known, and (b) if all of the argument’s premises can be known, then there are a
priori propositions. For example, at least one version of the so-called ontological argument for the-
ism has the following premises: (1) it is possible for there to be a being that is perfect in every way;
(2) necessarily, any being that is perfect in every way is a Supreme Being, and necessarily so; (3) if it
is possible for there to be a being that is necessarily a Supreme Being, then there is a being that is
necessarily a Supreme Being. It’s pretty clear that each of the premises is such that, if it can be
known at all, then it’s possible for someone to know it a priori. (Even if it can be known on the
basis of testimony, it’s plausible to think that this would require that at least someone know it not
on the basis of testimony, but a priori [but see Lackey 2008].) And, arguably, each of these premises
“commends itself to reason” (at least upon sustained reflection), so that if there are a priori proposi-
tions—if there are things we can know independently of experience—then we can know the pre-
mises. Taken together, that would imply that the ontological argument is such that all of its pre-
mises can be known if and only if there are a priori propositions. And if this is not true of the
ontological argument, it might well be true of at least some argument for theism. If it is, then the
existence of a priori propositions bears on the epistemic status of theistic belief in at least the follow-
ing way: it affords us an otherwise unavailable route to knowledge of theism. That’s not yet to say,
however, that there’d be no other route to knowledge of theism, or even no other “argumentative
route” to knowledge of theism.
But the existence of a priori propositions might bear on the epistemic status of theistic belief in
an even stronger way. Suppose theism is in fact true. Suppose, in fact, that there is a valid argument
for theism, such that (a) if there are a priori propositions then all of the argument’s premises can be
known, and (b) if all of the argument’s premises can be known, then there are a priori propositions.
And suppose finally that every valid argument for theism is such that if all of its premises can be
known, then there are a priori propositions. This is not an implausible cluster of suppositions. It
goes beyond our previous suppositions only insofar as it supposes that either there is no valid argu-
ment for theism all of whose premises can be known, or there are a priori propositions. And that is
quite plausibly true. For an even logically stronger claim is plausibly true, namely, that every valid
argument for theism has at least one premise such that if it can be known at all, either it is itself a
priori or any (human) knowledge of it would ultimately be based on some other a priori knowledge.
Every even slightly compelling argument for theism of which I am aware has at least one premise
with that feature.
A promising version of the cosmological argument—perhaps the most promising version—has
as a premise the principle of sufficient reason (the claim that every truth has a sufficient explanation).
If this can be known at all, it can be known a priori; for as David Hume taught us, it is extremely
problematic to rely on analogy or induction when our conclusion is partly about something that is
so vastly different, or at a vastly different scale, from the things about which we have ordinary evi-
dence. A promising version of the teleological argument—perhaps the most promising version—has
as premises a number of claims about the prior probabilities of various hypotheses. But if these prior

A Priori: Theism
probabilities can be known at all, then they can be known a priori; for these are supposed to be prior
to any empirical knowledge we might have about the world. (Prior probability is discussed further in
the next section.) And Peter van Inwagen (2004) has argued that it is no accident that the argu-
ments for theism of which we are aware have this feature: theism, by its very nature, is just not
the sort of thing for which one could successfully argue using premises all of which can be known
on the basis of experience alone. (The main point for van Inwagen is to argue that there is no good
wholly scientific argument for atheism; but he acknowledges that the same holds, mutatis mutandis,
for theism. And, I take it, if there is no wholly scientific argument for theism, then a fortiori there is
no good argument for theism all of whose premises can be known on the basis of experience alone:
the set of scientific premises includes the set of observation reports—and truths about what we
experience—as a proper subset.) Clearly enough, if every valid argument for theism has at least
one premise such that if it can be known at all, either it is itself a priori or any (human) knowledge
of it would ultimately be based on some other a priori knowledge, then either there is no valid argu-
ment for theism all of whose premises can be known, or there are a priori propositions.
So the final supposition in the cluster is plausible. If it, along with the previous suppositions, is
true, then the existence of a priori propositions bears on the epistemic status of theistic belief in a
very strong way: it affords us an argumentative route to knowledge of theism, where no argumenta-
tive route to knowledge of theism would otherwise be available. That’s still not to say that there
would be no other route to knowledge of theism if there were no a priori propositions. Perhaps
we could have knowledge of theism without any argument at all, and without such knowledge being
a priori. Some philosophers, such as William Alston (1991), have suggested that we can perceive
God. Others, such as Alvin Plantinga (2000), have suggested that as long as certain constraints
are satisfied—having to do with the proper functioning of our cognitive systems and the coopera-
tion of the environment in which we find ourselves—we might come to know that theism is true
by, for example, looking up at the starry heavens above and on the basis of that experience simply
and immediately forming the belief that theism is true. If either of these views is true, then even
if everything we’ve said until now is right, knowledge of theism doesn’t require that there be a priori
propositions; but it’s still noteworthy that argument-based knowledge of theism quite plausibly
hinges on there being a priori propositions.
Now, an astute reader might point out that everything I’ve said about the bearing of a priori
propositions on our knowledge of theism can be said, mutatis mutandis, about the bearing of a priori
propositions on our knowledge of atheism. The existence of a priori propositions admittedly bears
positively on the epistemic status of theism (if theism is true) and bears positively on the epistemic
status of atheism (if atheism is true). There is no tension between these points since neither of them
says that the existence of a priori propositions bears positively on the truth of theism or the truth of
atheism. It is in this regard that theists and atheists can agree as to the significance of there being a
priori propositions. Without them neither side could have any knowledge-producing argument in
favor of her view. Just how significant that is depends on what else we wouldn’t know if there were
no a priori propositions. Some philosophers, such as Laurence BonJour (1998), have argued that we
would know almost nothing at all if we couldn’t know anything a priori: nothing beyond our past
and present experiences (based on memory and introspection, respectively) and our past and present
immediate environment (based on memory and sense perception, respectively). According to such
views, it should come as no surprise that knowledge about a question as recondite and difficult as
whether theism is true is hostage to the fate of the a priori. But such an austere view might itself
give rise to a more direct relationship between the a priori and the question of whether theism is
true. It is to that more direct relationship that I now turn.
PROBABILISTIC BEARING
The existence of a priori knowledge might bear on the theism/atheism debate in a more direct and
less egalitarian way than I discussed in the previous section. In particular, it might bear on the truth
of theism (and thus on that of atheism). But this sort of bearing still comes in two varieties. I start
with the weaker variety.
A Priori: Theism
It is hard, perhaps, to see how the existence of a priori knowledge could absolutely require, or
entail, the truth of theism (and presumably even harder to see how its existence could absolutely
require, or entail, the truth of atheism). It is easier, I think, to see how the existence of a priori
knowledge could make theism more likely.
Here’s a natural thought that would suggest this conclusion. If there is a Supreme Being who
brought “all this” into existence, then in particular there is a Supreme Being who (at least ulti-
mately) brought human beings into existence. But any Supreme Being is wholly good—and so is
likely disposed to ensure that the things that this Supreme Being brought into existence have a
priori knowledge, alongside whatever other knowledge they have. And any Supreme Being is both
all-knowing and all-powerful, and so knows what sorts of cognitive faculties we would need in order
to have a priori knowledge and is capable of endowing us with such faculties. So it is very likely,
given theism, that there would be a priori knowledge.
But if theism is false, then either there is no Supreme Being, or there is, but the Supreme
Being didn’t bring “all this” into existence. If one or the other of these alternatives obtains, then
it’s very unlikely that we would have cognitive faculties that permit us to know anything that we
can’t learn from experience. So it is very unlikely, given atheism, that there would be a priori
knowledge.
The upshot is that it is much more likely given theism than given atheism that there would be
a priori knowledge. But if that’s so, then the existence of a priori knowledge is significant probabi-
listic evidence for theism, in the sense that it significantly raises the probability of theism above its
antecedent, or prior, probability. (That is, its probability before taking into account the evidence
provided by the fact that there is a priori knowledge.)
To help you understand this argument in greater detail—both its formal structure and its sub-
stantive assumptions—I now introduce a bit of probabilistic machinery.
Let us use L (α) to represent the likelihood function. This function assigns a number between 0
and 1 to a proposition (α is to be replaced by propositional variables, like P, Q, and R), in accor-
dance with the degree of belief (the so-called credence) that it is rational to have in the proposition:
it assigns 0 to propositions for which it is rational to be certain that they are false, 1 to propositions
for which it is rational to be certain that they are true, and so on. And let us define the likelihood of
α conditional upon β, L (α | β), as follows:
LðαβÞ
Definition of conditional likelihood: L (α | β) =
LðβÞ
As the Reverend Thomas Bayes pointed out, it follows from this definition (together with the
assumption that L (α & β) = L (β & α)) that:
Bayes’s theorem: L (α | β) * L (β) = L (β | α) * L (α)
From which it of course follows that:
LðαÞ
Corollary of Bayes’s theorem: L (α | β) = L(β|α) *
LðβÞ
Where H is some hypothesis and E is some piece of evidence, then, we have
LðHÞ
Posterior probability: L (H | E) = L(E|H) *
LðEÞ
L (H | E) is known as the posterior probability of hypothesis H (posterior to the bearing of evidence
E); L (H) and L (E) are known as the prior probabilities of hypothesis H and evidence E, respec-
tively; and L (E | H) is known as the likelihood of evidence E given hypothesis H.
As should be evident, the posterior probability of hypothesis H is greater than the prior proba-
bility of H if and only if the likelihood of evidence E given H is greater than the prior probability of
E. Perhaps less evident is that this is so if and only if the likelihood of E given H is greater than the
likelihood of E given the negation of H. That is, we have this schema:
Confirmation: L (H | E) . L(H) iff L (E | H) . L (E) iff L (E | H) . L (E | :H)
In the event that this condition obtains, we say that evidence E probabilistically confirms H.

A Priori: Theism
This provides a rigorous framework in which to state and evaluate the argument. Let T be the-
ism. And let APK be the claim that there is a priori knowledge (where, recall, by that is meant just
that at least some humans have at least some a priori knowledge). Then the argument I sketched
has as its lone substantive premise that:
Likelihood of a priori: L (APK | T) . L (APK | :T)
And from that, together with the relevant instance of confirmation, it follows that:
A priori confirms theism: L (T | APK) . L (T)
What shall we say about this formalized argument, and in particular about its lone substantive
premise? It seems to me that, as it stands, it’s not very persuasive. For as it stands, nothing has been
said to make it plausible that a Supreme Being would have any interest in anything having specifi-
cally a priori knowledge. That is, nothing has been said to make it plausible that a Supreme Being
would have any interest in anything having knowledge that is independent of experience. Why
would a Supreme Being want that? Perhaps some story can be told about the goodness of being
at least somewhat independent and self-reliant, including in epistemic ways. And perhaps that story
makes it plausible that any wholly good being would want some of the beings for whose existence
the Supreme Being is responsible to have knowledge that is a priori (as it would then be indepen-
dent of experience). I am dubious of such claims, and in particular the second, since knowledge that
is independent of experience in the sense relevant to whether it is a priori is still very much depen-
dent on experience, and the contingent features of one’s environment, in senses irrelevant to
whether it is a priori (as discussed at the start of this chapter).
But, in any case, I think that this reply misconstrues the animating idea behind the argument. It’s
not that a Supreme Being is supposed to want creatures to have a priori knowledge per se. It’s that the
Supreme Being is supposed to want creatures to have certain kinds of knowledge—whether mathemati-
cal, moral, modal, or otherwise—and it so happens that we simply cannot have such knowledge unless
we have a priori knowledge. A Supreme Being, working within the constraints there are, is thus disposed
to ensure that the things that this Supreme Being brought into existence have a priori knowledge.
To put the argument properly, we have to add one more element to the formal apparatus: ordi-
nary claims of likelihood and confirmation, including the ones that play a role in my argument,
need to be relativized to some specified background knowledge. John’s fingerprints being on the
gun raises the probability, and hence probabilistically confirms, the hypothesis that John was the
murderer, only given our background knowledge that one can leave a fingerprint by grasping an
object, and that fingerprints don’t usually just appear by chance, and that not everyone shares a fin-
gerprint, and so on. So, the formally correct statement of the relevant part of confirmation is this
(where K is our background knowledge):
Confirmation: L (H | E & K) . L(H | K) iff L (E | H & K) . L (E | :H & K)
And likewise for the lone substantive assumption of our argument:
Likelihood of a priori: L (APK | T & K) . L (APK | :T & K)
And likewise, finally, for our conclusion:
A priori confirms theism: L (T | APK & K) . L (T | K)
And the critical substantive point is that K, our background knowledge, includes at least some claim
of the form:
A priori required: Necessarily, if some human knows X, then some human knows something
a priori (where it can be the case, but need not be the case, that what needs to be known a
priori if X is to be known is X itself).
This is all schematic, of course. What is substituted for X determines the precise contours of
our lone substantive assumption and our conclusion; and what should be substituted for X is
A Priori: Theism
determined, of course, by what it is that we in fact know about what sort(s) of knowledge require a
priori knowledge. We might substitute for X the name of some particular mathematical proposition,
such as “that 2 + 2 = 4” or “Fermat’s last theorem.” Correspondingly, our background knowledge
would include the claim that necessarily, if some human knows that 2 + 2 = 4, then some human
knows something a priori, or the claim that necessarily, if some human knows Fermat’s last theorem, then
some human knows something a priori. Or we might substitute the name of some particular moral
proposition, such as “that it’s wrong to bludgeon innocent people to death for no reason,” or the
name of some particular modal proposition, such as “that nothing can be entirely red and entirely
green.” If BonJour (1998) is right, we might fill in for X the name of some particular, perfectly ordi-
nary proposition that goes beyond our past/present experiences and immediate environment, such as
“that the sun will rise tomorrow.” More generally, we might substitute for X some indefinite
description, such as “an elementary arithmetical proposition” or “an advanced number-theoretic
proposition.” Correspondingly, our background knowledge would include the claim that necessarily,
if some human knows an elementary arithmetical proposition, then some human knows something a
priori, or the claim that necessarily, if some human knows an advanced number-theoretic proposition,
then some human knows something a priori. Of course, these suggestions are not mutually exclusive:
our background knowledge might well include more than one of them.
This version of the argument is in better shape than its predecessor. But it still suffers from sev-
eral shortcomings. The first is that the argument makes a questionable assumption about the
Supreme Being’s will. For the argument seems to assume that the Supreme Being has a special inter-
est in human beings. After all, there could have been—and maybe there actually are—rational beings
other than human beings (and other than the Supreme Being) who have the sort of a priori knowl-
edge that the Supreme Being allegedly wants humans to have. And there could have been—and
maybe there actually are—rational beings other than human beings who lack the sort of a priori
knowledge that the Supreme Being allegedly wants humans to have. There certainly are actual
beings, period, who lack such knowledge! So, why think that the Supreme Being would seek to
ensure that humans, specifically, have such knowledge, unless one thinks that the Supreme Being
has a special interest in human beings? But remember that theism says nothing about any special
concern that the Supreme Being has for humanity.
A second shortcoming is that there seem to be some substitution instances of a priori
required—some specifications of which (human) knowledge absolutely requires a priori (human)
knowledge—that are (a) plausibly part of our background knowledge if any substitution instance
of a priori required is part of our background knowledge, and (b) plausibly such that their con-
junction even with the denial of theism—along with the rest of our background knowledge, includ-
ing our knowledge of evolutionary theory—makes it highly likely that there would be a priori
knowledge. (At least so long as atheism is consistent with there being a priori knowledge. I shall
assume that for now, and revisit the assumption in the next section.) For example, if any substitu-
tion instance of a priori required is part of our background knowledge, then this is plausibly one
such instance: necessarily, if some human knows that 2 + 2 = 4, then some human knows something
a priori. But the conjunction of that and evolutionary theory plausibly makes it highly likely that
some human knows something a priori, even supposing theism is false. For evolutionary theory, even
in conjunction with the denial of theism, makes it highly likely that we would know that 2 + 2 = 4.
Failing to know that 2 + 2 = 4, after all, would seem to be quite a hindrance to survival and repro-
duction. Imagine trying to keep track of predators, counting two to your left, two to your right, and
concluding that there are just three predators to contend with. Not very promising for survival!
(I should note that I am setting aside the central considerations in Beilby’s [2002] evolutionary
argument against naturalism. These considerations would suggest that the likelihood that our cogni-
tive faculties are reliable is quite low, given the conjunction of evolutionary theory and naturalism.
So they would suggest that it’s unlikely that we know much of anything at all, given the conjunc-
tion of evolutionary theory and naturalism. I set these considerations aside not because I don’t find
them compelling but because this chapter is supposed to be about how a priori knowledge in partic-
ular bears on theism. If Plantinga is right, then the likelihood that we would have a priori knowl-
edge, given the conjunction of evolutionary theory and the denial of theism, is approximately equal
to the likelihood that we would have knowledge, period, given the conjunction of evolutionary

A Priori: Theism
theory and the denial of theism. But then the existence of a priori knowledge no more confirms the-
ism than our having knowledge in the first place.) Since it’s part of K that knowledge of 2 + 2 = 4
absolutely requires a priori knowledge, it follows that it’s highly likely, given K and the denial of
theism, that we would have a priori knowledge. It doesn’t much matter that given the denial of the-
ism, nothing deliberately made us in such a way as to have a faculty for producing a priori knowl-
edge. We turned out to have such a faculty because we needed it to survive.
In response to this, one could further refine the argument. Perhaps the evidence in the vicinity
that indeed probabilistically confirms theism is the fact that, for example, (some) humans have knowl-
edge of nonelementary mathematics. It doesn’t seem plausible, after all, that failing to know Fermat’s
last theorem, or even square roots, would be a hindrance to survival and reproduction. Assessing this
more refined argument would take us too far afield, and in any case one has to contend with the other
objections to the argument. On top of all that, it should be noted that this more refined argument is
no longer an argument from a priori knowledge to theism. The a priori status of the knowledge in
question plays no role whatsoever in the refined argument.
A third and final shortcoming of the argument is easy to miss but nonetheless significant. Argu-
ments for probabilistic confirmation of a certain hypothesis always have a certain weakness: even if suc-
cessful they show only that the evidence makes the hypothesis more probable than it would otherwise
be. But that is consistent with the posterior probability of the hypothesis being low, or even extremely
low, since the posterior probability of the hypothesis is partly a function of its prior probability.
Now perhaps we can address that general point as applied to our case by supplementing the argu-
ment that we’ve already given with considerations that support a not-too-low “intrinsic” probability for
theism. Others have suggested such considerations (see Swinburne 2004). But notice that the prior
probability of theism is its likelihood given K. And we’ve suggested that K includes claims to the effect
that certain propositions cannot be (humanly) known except a priori. But such claims don’t comport
well with theism. For if there is a Supreme Being, then that Supreme Being has (or had) a whole range
of options in creating human beings and endowing them with knowledge. (Indeed, such claims don’t
comport well even with the possibility of theism. For if there could have been a Supreme Being, then
there could have been a being who had a whole range of options in creating human beings and endow-
ing them with knowledge.) I would think that the Supreme Being could have attested to any truth that
would otherwise be a candidate for being an a priori proposition, or could have inscribed on the hearts
of humanity—so as to make available for introspection—any truth that would otherwise be a candidate
for being an a priori proposition, or still other ways besides. So the prior probability of theism, that is,
its likelihood given our background knowledge, is quite low indeed. In the best case scenario—where L
(T & K) . 0—our probabilistic argument succeeds, but its conclusion is not very significant. In the
worst-case scenario—where L (T & K) = 0—then as we’ve defined conditional probability, the left-
hand side of likelihood of a priori is undefined, and so our lone assumption is false.
DEDUCTIVE BEARING
This brings us to the second and stronger way in which a priori knowledge might bear on the truth
of theism.
I said above that it is hard to see how the existence of a priori knowledge could entail the
truth of theism. Perhaps it is hard. But there is indeed a case to be made for such an entail-
ment—which derives from a well-known argument due to Paul Benacerraf (1973)—and that case
is not subject to all the difficulties that we raised for the probabilistic argument. (For other pro-
theism arguments with similarities to the arguments in this section, see Adams 1983; Rogers
2008; Thurow 2013.)
The basic thought is this: A priori knowledge—at least a priori knowledge of synthetic truths—
is knowledge about a domain of special objects, like numbers. But then it is quite mysterious how we
could know anything about them at all. Indeed, it seems pretty clear that we couldn’t know anything
about them at all. For numbers and the like (pure sets, features, etc.) are abstract, in the sense of
being causally inert. But then our beliefs about them, even if true, would not amount to knowledge.
A Priori: Theism
More exactly: numbers and the like are abstract if atheism is true. For then there is nothing con-
crete for them to be. So, if atheism is true, then there is no (synthetic) a priori knowledge. But there
is (synthetic) a priori knowledge. So, atheism is false.
Let’s explicitly lay out the argument’s premises:
1. There is (synthetic) a priori knowledge.
2. If there is (synthetic) a priori knowledge, then there is (synthetic) a priori knowledge about a
domain of special objects, such as numbers.
3. If atheism is true, then special objects, such as numbers, are causally inert.
4. There is no a priori knowledge about objects that are causally inert.
Therefore,
Atheism is false.
Suppose that a premise analogous to (3), with “theism” replacing “atheism,” is true if (3) is. Then we
would have a paradox on our hands, not an argument for theism. But the analogue of (3) is not nearly
as plausible as the original (3). If theism is true, then there are concrete things for numbers and the
like to be: ideas in the mind of the Supreme Being (“divine ideas”). In deference to tradition, let us
call the claim that numbers and the like are ideas in the mind of a Supreme Being divine conceptualism
(see Leftow 2012). These ideas are, or at least could be, causally efficacious. There is no great mystery,
then, about how we could come to know about those causally efficacious divine ideas. If it is possible
for us to interact with the Supreme Being—and nothing about the Supreme Being would give us rea-
son to think it’s impossible—then there is no great mystery in our being able to “see” these divine
ideas and what they are like. (Or at least no greater mystery than there is in our interacting with
the Supreme Being in the first place.) Something like this seems to be what Augustine had in mind
with his idea of an inner light (or, divine illumination) (Augustine 2010, bk. 2). Of course, nothing
precludes the Supreme Being from making these things known in other ways, such as attesting to
them in scripture or inscribing them on the hearts of men—unlike the probabilistic argument, we
don’t need to assume any such constraint on the Supreme Being—but there’s also nothing precluding
the Supreme Being in making these things known in this way.
Note well: I do not claim that divine conceptualism straightforwardly follows from theism. The
conjunction of theism and the denial of divine conceptualism is a tenable—indeed, widely
endorsed—position. (At least it’s tenable setting aside considerations having to do with a priori
knowledge!) Remember: theism, as I’ve characterized it, says nothing of who or what produced
things that lack causal powers. But atheism does rule out divine conceptualism. (Nearly enough, at
least. I think we can safely set aside a view according to which there is a Supreme Being, and num-
bers and the like are ideas in the mind of that being, but the Supreme Being didn’t bring into exis-
tence “all this.”) And that’s all that’s needed for the argument.
Let us briefly turn to each of the argument’s premises. (Alas, my brief treatment of each of the
premises can hardly do justice to the rich discussions each has provoked. But I hope it will go some
way toward making them plausible.) As to the first premise: I have already said that I will assume its
truth for the sake of exploring its consequences. As to the second premise: the alternative would be
that all of our a priori knowledge is knowledge about ordinary things, presumably about constraints
on how they are individually and how they stand in relation to one another. But there is both
phenomenological evidence and theoretical evidence for the idea that at least some of our a (synthetic)
priori knowledge—if we have such knowledge—really is about special objects, like numbers. The
phenomenological evidence is that it seems to (at least some of) us when we come to know certain
mathematical truths, for example, that we are in touch with objects different from the ordinary
objects of sense experience. Thus, the great twentieth-century logician Kurt Gödel:
I even think this comes pretty close to the true state of affairs, except that this additional sense
(i.e., reason) is not counted as a sense, because its objects are quite different from those of all
other senses. For while through sense perception we know particular objects and their
properties and relations, with mathematical reason we perceive the most general (namely the

A Priori: Theism
“formal”) concepts and their relations, which are separated from space. (quoted in Parsons
1995, 63)
The theoretical evidence is that many of the claims that we know a priori, if we know anything a
priori, are expressible by sentences that seem to involve reference to and quantification over things
like numbers and features. For example, it seems that we know a priori that there is exactly one
prime number between 6 and 10. But then it seems we know something about numbers.
As to the third premise: Nominalists of a certain stripe—those who deny that there are things
that are causally inefficacious but who accept that there are numbers—have tried to find some spa-
tial, causally efficacious objects with which to identify numbers. (The possibility that such objects
are nonspatial but causally efficacious is not one that they have seriously considered, but it is one
to which we will shortly return.) But any such identification seems arbitrary (Field 2016, 32);
and there isn’t any guarantee that there are infinitely many such objects (Hilbert 1983), let alone
infinities-of-greater-cardinality-many of them.
As to the fourth and final premise: a natural defense of this premise relies on a causal condition
on knowledge. That is, it relies on the idea that a necessary condition for S to know that p is that
there is a causal connection of the appropriate sort between the fact that p and S’s belief that p.
(Spelling out “the appropriate sort” is nontrivial, but it’s supposed to be liberal enough to allow
for knowledge of the future.) Otherwise, it seems that one’s belief that p is just a lucky guess, and
lucky guesses are not knowledge. (One can accept such a condition without committing to a so-
called causal theory of knowledge, where that is the view that the causal condition is both necessary
and sufficient. For the classic presentation of such a theory, see Goldman 1967.)
This condition is admittedly somewhat controversial. David Lewis suggests that in the case of
necessary truths—which are the subject of much, and perhaps all, of our a priori knowledge—the
gap between true belief and knowledge vanishes, and so the causal condition is not necessary for
knowledge of such truths (Lewis 1986, 113). After all, if one’s belief is true, then one can’t really
have been lucky in believing the truth, in that if one held as fixed what one believes, the world
had to cooperate. But John O’Leary-Hawthorne (1996) effectively rebuts this suggestion.
Hawthorne instead considers the more general suggestion that reliabilism—according to which,
roughly, S knows that p if and only if S’s belief that p is the product of a reliable belief-forming pro-
cess—is inconsistent with requiring a causal condition. After all, couldn’t one be using a reliable
belief-forming process about some entities without being in causal contact with those entities? But
as Hawthorne notes, in cases in which “the reliability is not due to any sort of responsiveness to
mathematical reality, direct or indirect, we balk at knowledge attribution” (1996, 193); and even
the reliabilist therefore concedes the need to add some bells and whistles to accommodate the causal
requirement. (The same, it seems to me, goes for the safety accounts that have grown more popu-
lar.) Hawthorne then suggests a contextualist and a conventionalist solution, but as he notes, each
faces serious difficulties. This leaves us with little alternative but to accept the causal condition
(and, with it, what he calls the “Transcendentalist” solution).
It seems to me that perhaps the best reply on behalf of atheism is to challenge the third prem-
ise, not by identifying numbers with some spatial objects that we already took to be causally effica-
cious, but by begrudgingly granting that there are nonspatial things—including numbers—that are
in fact causally efficacious. Since this is what Plato seems to have thought, let us refer to this as the
view that Plato’s heaven exists. The existence of Plato’s heaven is not obviously inconsistent with
atheism. (Note well that, as I’ve stated it, it’s also consistent with divine conceptualism.)
But now we can supplement (a modified version of) this deductive argument with considera-
tions of intrinsic probability. Let us take (a modified version of) the deductive argument to establish
that Plato’s heaven exists. (In premise 3 replace “atheism is true” with “Plato’s heaven does not
exist.”) Now, which of the following is more likely: Plato’s heaven exists and atheism is true, or Pla-
to’s heaven exists and theism is true? It seems to me that the second option is much more likely: set
aside whatever likelihood there is that theism is true, Plato’s heaven exists, and divine conceptualism
is false, and compare just the conjunction of atheism and Plato’s heaven’s existing to the conjunc-
tion of theism, Plato’s heaven’s existing, and divine conceptualism. The latter conjunction is more
parsimonious, more unified, and more powerful than the former.
A Priori: Theism
There is a related argumentative route to a related conclusion—one that derives from Hartry
Field’s (1989) own attempt to improve on Benacerraf’s argument—that does not rely on a causal
condition for knowledge. The argument merely assumes that no viable view according to which
(there are numbers but) Plato’s heaven doesn’t exist permits an explanation of the massive and strik-
ing correlations between how things stand regarding numbers and the like and our beliefs about
those things—correlations that there had better be if we are to have synthetic a priori knowledge.
For the only such viable view is one according to which there are numbers and they are causally
inefficacious, and so it permits no explanation at all. (Let me just forestall confusion: Field takes
his argument to present a problem for what he calls platonism, but that’s because his platonism is
platonist only insofar as it posits numbers that are mind-independent and not in space [or time],
but it is anti-platonist insofar as it says of those things that they are causally inefficacious.) And so
we ought to believe that Plato’s heaven exists.
This conclusion is not exactly that there is such a thing as Plato’s heaven, just that we ought to
believe that there is. But it’s close enough for us to continue on as before. Since we ought to believe
that Plato’s heaven exists, we ought to believe whichever of theism and atheism is more likely con-
ditional on the existence of Plato’s heaven. And as I have already suggested, theism wins that com-
petition.
Again, I cannot do justice to the substantive assumptions of this argument, and the extensive
literature that Field’s (and Benacerraf’s original) argument has generated (See Clarke-Doane 2017
for a helpful discussion). But I will just note that the argument that I’ve given is not meant to favor
the conjunction of theism and the denial of Plato’s heaven over the conjunction of atheism and the
denial of Plato’s heaven. (Thus, while Dan Baras’s [2017] point is well taken, it doesn’t affect my
argument.) Neither of those conjunctions does well in accommodating a priori knowledge or
explaining the correlations such knowledge requires, simply by dint of denying the existence of Pla-
to’s heaven (assuming, as we are, that a priori knowledge is indeed knowledge about a certain class
of objects, and that those objects are not to be found in space: the further denial of Plato’s heaven
leaves us no option but to embrace numbers and the like that are neither spatial nor causally effica-
cious). It is meant to favor theism over atheism, by way of affirming the existence of Plato’s heaven.
THEISM SUPPORTED
Does the existence of a priori knowledge bear favorably on theism? Yes. For if there is a priori
knowledge, then there is a promising case to be made that theism is true, indeed a case that at least
some of us might even know is successful.
Acknowledgments. Thanks to Tyron Goldschmidt, David Shatz, and a reviewer for this volume for
very helpful feedback on earlier drafts.
Bibliography
Adams, Robert Merrihew. “Divine Necessity.” Journal of Philosophy 80, no. Bealer, George. “Intuition and the Autonomy of Philosophy.” In Rethinking
11 (1983): 741–752. Intuition: The Psychology of Intuition and Its Role in Philosophical Inquiry,
edited by Michael DePaul and William Ramsey, 201–240. Lanham,
Alston, William P. Perceiving God: The Epistemology of Religious Experience.
MD: Rowman and Littlefield, 1998.
Ithaca, NY: Cornell University Press, 1991.
Augustine. On the Free Choice of the Will, On Grace and Free Choice, and Beebe, James R. “A Priori Skepticism.” Philosophy and Phenomenological
Other Writings. Edited and translated by Peter King. Cambridge: Research 83, no. 3 (2011): 583–602.
Cambridge University Press, 2010. Beilby, James K., ed. Naturalism Defeated?: Essays on Plantinga’s Evolutionary
Baras, Dan. “A Reliability Challenge to Theistic Platonism.” Analysis 77, no. Argument against Naturalism. Ithaca, NY: Cornell University Press,
3 (2017): 479–487. 2002.
Bealer, George. “A Priori Knowledge and the Scope of Philosophy.” Benacerraf, Paul. “Mathematical Truth.” Journal of Philosophy 70, no. 19
Philosophical Studies 81, no. 2/3 (1996): 121–142. (1973): 661–679.

A Priori: Theism
BonJour, Laurence. In Defense of Pure Reason: A Rationalist Account of a Lackey, Jennifer. Learning from Words: Testimony as a Source of Knowledge.
Priori Justification. Cambridge, UK: Cambridge University Press, Oxford: Oxford University Press, 2008.
1998.
Leftow, Brian. God and Necessity. Oxford: Oxford University Press, 2012.
Casullo, Albert. “Analyzing A Priori Knowledge.” Philosophical Studies 142,
Lewis, David. On the Plurality of Worlds. Oxford: Blackwell, 1986.
no. 1 (2009): 77–90.
Mill, John Stuart. A System of Logic, Ratiocinative and Inductive: Being a
Clarke-Doane, Justin. “What Is the Benacerraf Problem?” In Truth, Objects,
Connected View of the Principles of Evidence and the Methods of Scientific
Infinity: New Perspectives on the Philosophy of Paul Benacerraf: Truth,
Investigation. Vol. 1. London: Longmans, Green, 1884.
Objects, Infinity, edited by Fabrice Pataut, 17–44. Cham, Switzerland:
Springer, 2017. O’Leary-Hawthorne, John. “The Epistemology of Possible Worlds: A
Guided Tour.” Philosophical Studies 84, no. 2/3 (1996): 183–202.
Devitt, Michael. “There Is No A Priori.” In Contemporary Debates in
Epistemology, edited by Matthias Steup and Ernest Sosa, 105–115. Parsons, Charles. “Platonism and Mathematical Intuition in Kurt Gödel’s
Malden, MA: Blackwell, 2005. Thought.” Bulletin of Symbolic Logic 1, no. 1 (1995): 44–74.
Field, Hartry. Realism, Mathematics and Modality. Oxford: Blackwell, Plantinga, Alvin. Warranted Christian Belief. New York: Oxford University
1989. Press, 2000.
Field, Hartry. Science without Numbers: A Defense of Nominalism. 2nd ed. Quine, W. V. “Two Dogmas of Empiricism.” Philosophical Review 60, no. 1
Oxford: Oxford University Press, 2016. (1951): 20–43.
Goldman, Alvin I. “A Causal Theory of Knowing.” Journal of Philosophy 64, Rogers, Katherin A. “Evidence for God from Certainty.” Faith and
no. 12 (1967): 357–372. Philosophy 25, no. 1 (2008): 31–46.
Hawthorne, John. “A Priority and Externalism.” In Internalism and Swinburne, Richard. The Existence of God. Oxford: Clarendon; New York:
Externalism in Semantics and Epistemology, edited by Sanford Goldberg, Oxford University Press, 2004.
201–218. Oxford: Oxford University Press, 2007. Thurow, Joshua C. “The Defeater Version of Benacerraf’s Problem for
Hilbert, David. “On the Infinite.” In Philosophy of Mathematics: Selected A Priori Knowledge.” Synthese 190, no. 9 (2013): 1587–1603.
Readings, 2nd ed., edited by Paul Benacerraf and Hilary Putnam, 183– Van Inwagen, Peter. “Can Science Disprove the Existence of God?”
201. Cambridge, UK: Cambridge University Press, 1983. Philosophic Exchange 34, no. 1 (2004).
TOPIC 10
A Priori: Atheism
Felipe Leon
El Camino College, Torrance, CA
The primary aim of this chapter is to explore whether considerations about a priori domains and abstract
objects favor atheism over theism.
THEISM, ATHEISM, AND A PRIORI KNOWLEDGE
A number of philosophers have thought that, in addition to the familiar world of contingent con-
crete objects known a posteriori, via the senses, there is a realm of necessarily existent abstract objects
that are knowable a priori, that is, prior to or independent of experience. However, it is at least ini-
tially puzzling how such knowledge is possible: how can truths about the world outside the mind be
known merely by means of reflection on items inside the mind?
Some theists have argued that such a correlation between thought and world smacks of pre-estab-
lished harmony, requiring forethought and arrangement by an omniscient, omnipotent deity. Some
theists have also argued that the existence of abstract objects does not fit well within the naturalistic
ontology commonly accepted by atheists. By contrast, they argue, such necessarily existent, immaterial
entities can be handily explained in terms of a necessarily existent, immaterial God.
Atheists disagree. Some atheists deny the existence of a priori knowledge; some deny the exis-
tence of abstract objects; and some accept one or the other (or both) but deny that either one is best
explained in terms of theism.
The present chapter provides an overview of the core issues, arguments, and stances with
respect to a priori knowledge, abstract objects, and what bearing they might have on the epistemic
credentials of theism and atheism. The chapter is divided into two main sections. The first considers
a priori knowledge and its bearing on the epistemic credentials of theism and atheism. The second
considers abstract objects and their bearing on theism and atheism. It will be concluded that neither
one renders theism more likely than atheism.
The present section explores whether a priori knowledge exists, and what evidential bearing
such knowledge might have on the hypotheses of theism and atheism.
PRELIMINARIES
Many accounts of a priori knowledge have been proposed, and to date, there is no consensus about
any given account. For the purposes of this chapter, however, it is enough to provide a rough and
impressionistic account of the notion. Very roughly, then, to say that something is known a priori
is to say that it is known prior to or independent of experience. By “independent of experience,” it
shall be meant that experience plays at most an enabling role, and not a justificatory role, in such
knowledge. That is, when it comes to a priori knowledge of a proposition, experience (at most) either
(a) provides the occasion for one to know the proposition a priori, or (b) provides the relevant concepts
required to grasp the proposition’s truth value. Finally, “experience” is defined broadly so as to denote
not only perception, memory, and testimony but also introspection and proprioception.
313
A Priori: Atheism
KEY CONCEPTS
Analytic Truth Modified Theistic Activism
Aseity-Sovereignty Doctrine One-Over-Many Argument
Bootstrapping Problem Perfect-Being Theology
Cartesian Circle Platonism
Class Nominalism Predicate Nominalism
Concept Nominalism Resemblance Nominalism
Counteractual Conceivability Synthetic Truth
Counteractual Possibility Theistic Activism
Counterfactual Conceivability Theistic Conceptualism
Counterfactual Possibility Theistic Platonism
Divine Simplicity Trope Nominalism
Intuitionism
Truths known a priori are typically taken to be necessarily true. Furthermore, these truths are
commonly taken to be known either directly, by intuition (where these are cashed out in terms of
internally accessible intellectual seemings or externalistically construed, nonexperientially based,
truth-tracking beliefs), or indirectly by deductive, inductive, or abductive reasoning.
Examples of truths knowable a priori include those of mathematics (e.g., 1 + 1 = 2), logic (e.g., If
P is true and P implies Q, then Q is true), modality (e.g., It is necessary that an entity is identical to
itself; it is possible that I will get a flat tire tomorrow), and metaphysics (e.g., If an object was in fact
made from a particular hunk of stuff, then it is impossible for it to have been originally made from a
radically different hunk of stuff).
Truths that are knowable a priori are standardly divided into two types: analytic and synthetic.
As with the distinction between a priori and a posteriori knowledge, there is no agreed upon
account of the distinction between analytic and synthetic a priori truths, and many deny that there
is one. However, it is enough for our purposes to draw a rough and impressionistic distinction
between the two. Thus, say that an a priori truth is analytic if its truth depends on the meaning
of its constituent concepts alone (e.g., “All triangles have three angles”), and an a priori truth is syn-
thetic if its truth does not depend on the meaning of its constituent concepts alone (e.g., according
to some, “nothing can be red all over and green all over at the same time” is such an example, but
this is controversial, as is the category of synthetic a priori truths).
Many philosophers of course deny that there is a distinction between analytic and synthetic
truths, or even that there is such a thing as a priori knowledge. Willard Van Orman Quine, for
example, famously argued that the meanings of terms are determined by the theory of the world in
which they are embedded. Therefore, individual sentences within a theory cannot be tested in isola-
tion but rather must “face the tribunal of experience … as a corporate body” (1951, 41). As such,
even the truths of logic and mathematics are revisable, and earn their keep within one’s theory in vir-
tue of being embedded within our total system or “web” of beliefs, which in turn accrue confirma-
tion just to the extent that they explain and predict the world we experience.
Suppose, though, that at least some knowledge is a priori. What bearing might such knowledge
have on the epistemic credentials of theism and atheism? Some have argued that while theism provides
the materials to explain the possibility and reliability of a priori knowledge, such knowledge is impos-
sible or improbable on the hypothesis that there is no god, and the natural world is all there is. The
discussion that follows explores the issue of whether a priori knowledge makes theism or atheism more
likely. In particular, we will consider whether a priori knowledge or justification across a range of
A Priori: Atheism
domains—mathematics, logic, modality, and philosophy—requires or is best explained in terms of

theism.
To begin, how might a priori knowledge be explained on the hypothesis of theism? Any ade-
quate theistic account of a priori knowledge requires at least two elements: an account of how
God ensures that humans can have a priori knowledge of the truths in the relevant domain, and
an account of how God himself has knowledge in the relevant domain.
Accounts of how God ensures that humans have knowledge of a given domain of truths likewise
fall into two main categories. According to accounts of the first sort, God creates us in such a way that
some range of a priori truths are innate (cf. René Descartes’s doctrine of innate ideas). According to
accounts of the second, God designs our cognitive faculties so that they reliably grasp a priori truths.
Accounts of how God himself can have knowledge in a given domain generally fall into two
main categories: those that take the ground of the relevant domain of truths to be internal to God’s
being, and those that take it to be external to God’s being. Among those of the former type, God
has such knowledge either because he has decreed or otherwise made them true, or because, although
he has not decreed them or otherwise made them true, they are elements of God’s being, and to
which he has internal access. Among accounts of the latter type, God’s omniscience and omnipo-
tence, and/or his metaphysical affinity with the objects of a priori knowledge, somehow enable
him to access, and have reliable knowledge of, the truths in the relevant domain.
Accounts of both of the main sorts are considered in this chapter, with some unavoidable overlap.
Accounts of how God ensures that we have a priori knowledge in a given domain are covered in the
first section. Since accounts of how God himself can have knowledge of truths in a given domain stan-
dardly rely on accounts of God’s relation to abstract objects, they are deferred to the second section.
GENERAL ARGUMENTS FROM INTUITION

In “Two Dozen (or So) Theistic Arguments” (2007), Alvin Plantinga sketches several arguments
from a priori knowledge of logical, philosophical, and mathematical truths to theism. The two most
plausible among them are considered here.
One takes its cue from the Benacerraf problem (1973): a good deal of a priori knowledge is
putatively about abstract objects, such as numbers, sets, and functions. But knowledge requires stand-
ing in causal relations with the thing known. This is impossible with respect to abstract objects, since
they are by definition non-spatiotemporal and acausal; therefore, if the entities in some domain are
abstract, then we cannot have knowledge of them. However, if abstract objects are instead elements
of God’s mind—for example, if properties are divine concepts, propositions are divine thoughts, sets
are divine collections, and so on—then God stands in causal relations with his own thoughts. Fur-
thermore, we can come to stand in causal relations to God’s thoughts by virtue of our causal relation
to God. Therefore, we can have such a priori knowledge only if theism is true.
Atheists can respond to the argument in a number of ways. First, some deny that there is such a
thing as intuitive or a priori knowledge of synthetic truths (e.g., Quine 1951). Second, some have argued
that there are substantive reasons for thinking abstract objects cannot be plausibly construed as divine
thoughts, divine collections, and so on (see the discussion on God and abstract objects in the second sec-
tion). Third, some have argued that since we do have a priori knowledge of synthetic truths, then if such
knowledge requires knowledge of abstract objects, then such knowledge does not require that we stand
in causal relations with them (Plantinga 1993b; Huemer 2016). Alternatively, some have argued that
since we do have such a priori knowledge, if abstract objects are acausal entities, then a priori knowledge
must not involve abstract objects (see the second section for discussion of alternatives to the view that
mathematical and other a priori knowledge involves knowledge of abstract objects).
The other argument begins with the observation that humans seem to have many kinds of intui-
tions: logical, mathematical, moral, and philosophical. It also seems that such intuitions are instances
of knowledge. But it is much easier to see how we could have intuitive knowledge on a theistic
account of the nature of human beings than to see how we could have it on a nontheistic account.
Therefore, theism is the best explanation of intuitive knowledge, in which case theism is probably
true—or at any rate, intuitive knowledge increases the probability of theism (Plantinga 2007).

A Priori: Atheism
An atheist might agree that we have intuitive knowledge of the sorts referred to above, but what
is Plantinga’s reason to think such knowledge is easier to account for on theism than on atheism?
Unfortunately, Plantinga does not say in the paper, leaving the argument for others to fill out.
Perhaps the most natural way to fill it out is to say that if theism is true, then God is omni-
scient, and thereby has direct, unmediated knowledge—that is, intuitive knowledge—of the truths
of logic, mathematics, ethics, philosophy, and so on. He is also omnipotent and perfectly good.
Given these three attributes, it follows that God is willing and able to design any intelligent beings
he might happen to create in such a way that they likewise have such intuitive knowledge. By con-
trast, if atheism is true, then our cognitive faculties are the product of unguided evolution. But evo-
lutionary accounts explain the development of traits in terms of whether they are useful for survival
and reproduction, and it is not at all clear how intuitive knowledge of the necessary truths of math-
ematics, logic, philosophy, and so on—at least those that have no clear bearing on the practical con-
cerns of human beings—are useful for survival and reproduction. Therefore, theism explains intui-
tive knowledge of such truths better than atheism, in which case such knowledge provides at least
some confirmation of the former vis-à-vis the latter.
An atheist might reply in a number of ways. First, some have appealed to evolutionary accounts
of the origin and function of intuitive knowledge, which explain how such knowledge is conducive
to survival and reproduction (De Cruz 2016). Second, Amie Thomasson (2014) and others have
offered conventionalist accounts of truths known via intuition, according to which the objects of
intuitive knowledge are not beyond our cognitive reach. Finally, some have argued that theism is
no better off than atheism with respect to explaining intuitive knowledge, since the theistic explana-
tion just pushes the problem back a step without solving it, leaving God’s knowledge of the relevant
truths unaccounted for. (More on this below and in the second section.)
A more sophisticated version of the argument can be gleaned from Plantinga’s mature (1993b)
account of warranted belief. According to that account, for tokens of any type of belief to have war-
rant—perceptual, mnemonic, introspective, testimonial, intuitive, whatever—that belief must be
produced by a cognitive faculty in accordance with a design plan that is successfully aimed at truth,
and in an epistemic environment of a sort relevantly similar to the ones in which it was designed to
function. (To account for the apparently gradable nature of warrant, Plantinga adds that the degree
of warrant a belief enjoys is a function of the degree of firmness or conviction with which it is held.)
But according to Plantinga, the notion of proper function and cognate notions—purpose, plan,
design, malfunction, and so on—are essentially connected to the notion of intelligent design. Fur-
thermore, Plantinga argues that attempts to give naturalistic accounts of function in terms of natural
selection and evolutionary history have failed miserably, further shoring up this claim. But, argues
Plantinga, the best candidate for the intelligent designer of our cognitive faculties is the god of the-
ism. Therefore, the very existence of warranted beliefs—including warranted intuitive or a priori
beliefs—provides strong evidence for theism (Plantinga 1993b).
A number of criticisms have been raised against both Plantinga’s theistic proper functionalist
account of warranted belief and his argument from proper function to theism. Against the former,
some have argued that the account is subject to counterexamples, and therefore fails as an adequate
account of warranted belief. For example, it has been pointed out that certain sorts of brain lesions
can enhance memory, in which case proper function is not necessary for warrant (Greco 2003).
Against sufficiency: Suppose God designed us in such a way that we come to believe that God exists
with maximal firmness whenever presented with a seemingly impeccable but subtly fallacious argu-
ment for atheism. Plantinga’s account entails that belief in God formed in this way has maximal
warrant—on a par with that enjoyed by one’s belief that one exists. But this is prima facie absurd
(Senor 2002; for other counterexamples, see Feldman 1993; Lehrer 1996).
Against Plantinga’s argument from proper function to theism, some have argued that proper
function is not essentially tied to intelligent design. Toward that end, they have also argued that
adequate naturalistic evolutionary accounts of function can in fact be given that avoid Plantinga’s
criticisms (Bardon 2006; Wunder 2008; Graham 2011).
Finally, one might argue that Plantinga’s account of warrant is self-referentially incoherent. For
we have seen that Plantinga analyzes warrant in terms of beliefs formed by properly functioning,
A Priori: Atheism
(successfully) truth-aimed cognitive faculties in congenial epistemic environments. However, we have

also seen that he thinks appeal to intentional design is required for any adequate account of function.
Yet he also thinks God is a person with cognitive faculties, and that his faculties were not designed.
Therefore, they would seem to lack functions, in which case, a fortiori, they cannot function properly.
But if not, then it seems that God’s beliefs lack warrant. But if God’s beliefs lack warrant, then it is hard
to make intelligible the notion of God as a competent designer of our cognitive faculties. Therefore,
Plantinga’s accounts of warrant and of proper function seem to jointly entail that none of our
beliefs—including, again, our intuitive or a priori beliefs—have warrant. It is therefore not at all clear
that general theistic accounts of intuition render theism more likely than atheism.
MODAL KNOWLEDGE
Some (e.g., O’Connor 2008) have argued that our knowledge of what is possible and what is neces-
sary is puzzling on naturalism but not on theism. The core of the worry is that if naturalism is true,
then it is plausible to think that there is no special faculty of modal intuition that reliably tracks the
truth. As mentioned earlier in relation to knowledge of abstract objects generally, such an idea
smacks of forethought by an intelligent agent that ensured a preestablished harmony between our
modal intuitions and the modal properties of the world. Therefore, if naturalism is true and we have
some modal knowledge, then it is most plausibly construed as empirical knowledge. But empirical
knowledge tells us only about what is actually the case, not what is possibly the case, or necessarily
the case. Therefore, if naturalism is true, then it is prima facie surprising that we have knowledge
or justified beliefs about modal facts.
By contrast, such knowledge is not surprising on theism, since God is omniscient, in which case
he knows all truths, and therefore all modal truths. And since, by hypothesis, he is also omnipotent,
then he can create us with a faculty of modal intuition that reliably tracks the truth about modal facts.
There are three main worries with this line of argument. First, an analogue of the Cartesian cir-
cle threatens such accounts, at least to the extent to which the case for theism relies on arguments
with at least one modal premise. For then the arguments for God’s existence succeed only if one
has independent reason to think our modal intuitions or conceivability-possibility inferences are reli-
able. But one has independent reason to think our modal intuitions or conceivability-possibility
inferences are reliable only if the arguments for God succeed.
Second, such an account merely pushes the problem of modal knowledge back a step, as it
leaves unexplained how God can have knowledge of modal truths. The problem at issue here
reduces to God’s relation to abstract objects and is therefore deferred to the second section.
Finally, it is not at all clear that modal knowledge is surprising on naturalism. For there have
been a number of independently motivated, naturalistically acceptable accounts of justified conceiv-
ability-possibility inference. Standard accounts start with something like a phenomenal conservati-
vist principle, according to which the way things appear or seem provide at least some prima facie
evidence for how things are (Huemer 2007). From there, such accounts can infer that if some state
of affairs seems or appears possible to one, then that is prima facie evidence that it is possible.
Some such accounts construe such appearances as purely rational or intellectual seemings, and
thereby take rational intuition as a guide to possibility (Bealer 2002). Yet others take something like
the phenomenal imagery of imagination to ground modal seemings. Stephen Yablo’s (1993)
account is representative here. The core idea of his account is that possibility claims are prima facie
justified in virtue of imagining possible worlds at which their demodalized counterparts are true. So,
for example, the claim that, possibly, a ball is stuck on a roof is justified if I can imagine a world at
which “a ball is stuck on a roof” is true.
A more sophisticated account is that of David Chalmers (1996, 2002). Chalmers’s account is
similar to Yablo’s in that he views imagination as an independent source of justification for modal
beliefs, and that he accepts his two-step method of verifying a possibility claim. Chalmers’s primary
innovation lies in his wedding of Yablo’s account with the two-dimensionalist distinction between
considering a world as actual and considering a world as counterfactual. To give a rough and brief
illustration: Suppose I imagine an XYZ-world—that is, a world at which XYZ is the stuff that plays
the role of water. If I consider the imagined world as actual, I consider it as a hypothesis about how

A Priori: Atheism
the actual world could turn out to be for all I know a priori—I suppose that XYZ, and not H20,
turns out to be the actual occupant of the watery role. Call this counteractual conceivability. On
Chalmers’s view, the counteractual conceivability of XYZ watery stuff is prima facie evidence for
the metaphysical possibility of XYZ watery stuff. Call this counteractual possibility.
By contrast, when I consider the imagined XYZ world as counterfactual, I hold the actual, H20
world fixed, and imagine (or attempt to imagine) that XYZ is water. Call this counterfactual conceiv-
ability. Of course, XYZ water seems to be counterfactually impossible, but when something is coun-
terfactually conceivable, that is evidence that it is a counterfactual possibility.
Chalmers’s account builds on Yablo’s account and supplements it with key elements of two-
dimensional semantics. Chalmers’s main motivation for the two-dimensionalist twist on Yablo-style
modal epistemology is to handle the problem of a posteriori necessities raised by Saul Kripke and
Hilary Putnam. Thus, Chalmers’s diagnosis of the problem of a posteriori necessities is that one is
liable to modal error when one illicitly uses counteractual conceivability as a guide to counterfactual
possibility, as when one imagines a world with XYZ as the occupant of the watery role and mistak-
enly takes it as depicting a world with XYZ water. And Chalmers’s solution to the problem of a pos-
teriori necessities is to use counteractual conceivability only as a guide to counteractual possibility,
and to use counterfactual conceivability only as a guide to counterfactual possibility.
There has been a growing trend, however, toward empiricism in the epistemology of modality,
according to which our modal knowledge traces back to our knowledge of the actual world via
empirical sources (Fischer and Leon 2016). To start with the most obvious sort of case, a good deal
of modal knowledge can be got via deduction from the actual to the possible (P is actually the case;
whatever is actual is possible; therefore, p is possible). Beyond this, a number of naturalistically
friendly accounts of modal knowledge have been offered. For example, it has been argued that evo-
lutionary pressures have led to a reliable capacity for counterfactual reasoning, as it enables us to
evaluate risks and opportunities, which in turn enhances our ability to survive and reproduce
(Nichols 2006; Williamson 2007). It has also been argued (Williamson 2007; Leon 2016) that
much modal knowledge can be generated from our folk theories of how the world operates (e.g.,
ours is a world where objects can be moved around, painted, held, smashed), which in turn accrue
confirmation by virtue of their ability to explain and predict the world around us.
In addition, there are inductive and track-record accounts (Leon 2016): Observed tokens of a1-
an of F are possible, since actual; so, probably an+1 is possible as well (Leon 2016). There are also
similarity-based accounts, according to which we can justifiably move from the actual to the possible
by means of relevant similarity to the actual world: P is possible (since actual); Q is relevantly similar
to P; therefore, Q is probably possible as well (Hawke 2011; Leon 2016; Roca-Royes 2016).
According to yet other accounts, possibilities are justified by virtue of an inference to the best expla-
nation (Fischer 2016, 2017). According to one important version, our best theories are justified via
embodying the theoretical virtues (simplicity, scope, fit, predictive power, etc.). Such theories entail
that some things are possible. Therefore, such possibilities are justified in this way via abduction.
It might be objected that empiricist accounts cannot explain our knowledge of possibilities that
play a central role in paradigm cases of successful thought experiments. It may also be objected that,
while such accounts provide a naturalistic explanation of our knowledge of “nearby” possibilities, they
cannot account for our knowledge of “far out” possibilities that are remote from ordinary experience.
In response to the first objection, it has been argued that such accounts do, in fact, explain our
knowledge of the paradigm cases of successful thought experiments (Edmund Gettier cases in the
epistemology literature; the trolley problem case in the ethics literature; John Locke’s locked room
case and Harry Frankfurt’s “Black and Jones” counterexamples in the free will literature; the fake
barn case in the epistemology literature; the ship of Theseus case in the personal identity literature;
Peter Singer’s Shallow Pond example and Judith Thomson’s violinist example in the practical
ethics literature; Kripke’s Gödel/Schmidt case, John Perry’s grocery store case, and Tyler Burge’s
“tharthritis” case in the philosophy of language literature, etc. [Leon 2016]). In response to the
second objection, it has been argued that it is a virtue of empiricist accounts that they do not explain
our knowledge of “far out” possibilities remote from ordinary experience. This is because there are
strong reasons to doubt that we have such knowledge. For one thing, knowledge of “far out”
A Priori: Atheism
possibilities cannot be justified on any account of modal epistemology, whether rationalist or

empiricist (Fischer and Leon 2016). This is further evidenced by the widespread and perennially
entrenched disagreement about them (such as those about the possibility of Anselmian beings,
personal fission, disembodied souls, philosophical zombies).
Furthermore, such disagreement is predicted and explained by modal empiricist theories. For if
all of our knowledge of possibility traces back to our sources of knowledge of the actual world via
deductive, inductive, and abductive inference, then we would expect such claims to remain perenni-
ally contentious: On the face of it, such sources cannot justify such claims. By contrast, this is sur-
prising on theism. For prima facie, it is mysterious as to why our modal insight should fizzle out at
all on the theistic hypothesis. It is also prima facie mysterious, on the theistic hypotheses, why the
doxastic and epistemic force of possibility claims would fizzle out at just the point where inferences
from actuality to possibility fizzle out. Again, not so on the naturalism-friendly versions of modal
empiricism discussed presently (Leon 2016). Prima facie, then, it seems that the nature and scope
of human modal knowledge is better explained on atheism than on theism.
In conclusion, we have looked at theistic and atheistic explanations of a priori knowledge (or
the lack thereof) in a representative sampling of domains: modal, mathematical, logical, and philo-
sophical. In most cases, we have seen that it is deeply problematic to account for such knowledge
on theism. We have also seen that, with few exceptions, atheism seems to have an easier time of
explaining such knowledge. It is therefore not at all clear that the existence of a priori knowledge
is more probable on theism than on atheism.
THEISM, ATHEISM, AND ABSTRACT OBJECTS
The present section explores whether abstract objects exist, and what evidential bearing their
existence might have on the hypotheses of theism and atheism.
WHAT ARE ABSTRACT OBJECTS?

There is considerable disagreement about how to characterize abstract objects, and it is not clear
that a single account can clearly individuate them from entities in other ontological categories
(Balaguer 2016). However, in the contemporary discussion surrounding God and abstract objects
in philosophy of religion, the focus is on platonism with respect to abstract objects as that notion
is characterized in the philosophy of mathematics, namely the view that there are timeless, nonspa-
tial, acausal, necessarily existent entities that exist a se—that is, they are self-existent (Craig 2016).
Candidate examples include properties (redness, roundness, etc.), relations (causes, has more mass
than, etc.), propositions (the information content encoded in declarative sentences), possible
worlds (ways a world can be or could have been), numbers, sets, and functions.
Abstract objects stand in contrast with concrete objects, which are spatiotemporal entities that
can stand in causal relations to other entities. Examples include particular substances or individuals
(water molecules, rocks, trees, persons, quantum fields, etc.), events (e.g., sliding into third base),
and tropes (property-instances). An important exception for our purposes is of course God, who is
taken to be a non-spatiotemporal concrete particular object that can stand in causal relations to other
things, including spatiotemporal entities.
KNOWLEDGE OF ABSTRACT OBJECTS

Given the account of abstract objects sketched above, it is natural to suppose that proponents of pla-
tonism might have a hard time explaining our knowledge of abstract objects. For given the peculiar
transcendent, acausal, immaterial nature of abstract objects, it is at least initially puzzling how one
might have knowledge of them.
Two main sorts of accounts of knowledge of abstract objects have standardly been given.
According to the first, which is the more traditional account, abstract objects are knowable immedi-
ately and directly by the mind via acts of rational intuition, whereby the abstract objects themselves
are directly “grasped,” or apprehended, by the intellect. This sort of view is called intuitionism.

A Priori: Atheism
One might find intuitionism mysterious: how can one be in direct contact with an object that
is by its very nature non-spatiotemporal and acausal? This concern is at the root of perhaps the most
important objection to platonism about abstract objects, argued forcefully by Paul Benacerraf
(1973) and discussed in the first section. Recall from that section that the argument runs as follows:
Knowledge requires standing in causal relations with the thing known. But this is impossible with
respect to abstract objects, which are by definition non-spatiotemporal and acausal; therefore, if
the entities in some domain are abstract, we cannot have knowledge of them.
Intuitionists respond to Benacerraf’s objection in a number of ways. The most direct sort of
response requires giving an account of how intuition of abstract objects is possible. One account
of this sort (Chudnoff 2014; Bengson 2015a) offers a noncausal account of rational intuition.
According to this sort of account, there is a certain type of noncausal relation—constitution—that
is epistemically relevant to our cognition of abstract objects. In particular, abstract objects can at
least partly constitute our thoughts, and knowledge of abstract objects can thereby be explained in
terms of such a relation.
Others offer an indirect response to Benacerraf’s objection. We saw some standard replies in the
first section. However, many Platonists are suspicious or skeptical of intuitionism. This leads us to
the second, and perhaps most popular, account of our knowledge of abstract objects. According to
this sort of account, knowledge of abstract objects is a kind of theoretical knowledge (Plantinga
1993b). In particular, abstract objects are theoretical posits that earn their keep in our theories by
virtue of best explaining a range of data. So, for example, perhaps the positing of properties as uni-
versals (i.e., entities that can exist in more than one place at the same time) best explains the phe-
nomena of resemblance, predication, and abstract reference (Moreland 2001; Loux and Crisp
2017). Or perhaps the positing of abstract numbers best explains the metaphysical necessity of
mathematical truths and the infinity of numbers. Relatedly, yet others argue that certain sorts of
abstract objects, such as numbers or sets, are indispensable for our well-confirmed scientific theories,
and are posited on that basis (Quine 1948). (More on this in the next part of this section.)
Finally, some argue that while we cannot stand in causal relations with abstract objects, we can
stand in causal relations with their concrete tokens or property-instances, and that this provides the
epistemic anchor for knowledge of abstract objects (BonJour 1998; Moreland 2001; Loux and Crisp
2017).
ARGUMENTS FOR ABSTRACT OBJECTS

Historically, the central argument for the existence of abstract objects has been the one-over-many
argument (Moreland 2001; Balaguer 2016; Craig 2016; Loux and Crisp 2017). This sort of argu-
ment is typically used to justify belief with respect to a particular kind of abstract object, namely,
universals (i.e., entities that are repeatable or can have more than one instance at the same time).
The argument can be stated as follows: It is a datum of ordinary experience that two or more objects
seem to be the same in some respect. For example, two cars fresh off the assembly line are the same
shade of red. What explains this? The Platonist about universals answers that the best explanation of
this phenomenon is that there is a single entity—namely, the property redness—that is instantiated
in or exemplified by the two cars at the same time.
Although historically the one-over-many argument has been at the center of the debate about
abstract objects, perhaps the most popular contemporary argument for abstract objects is Quine’s
(1948) indispensability argument (Balaguer 2016; Craig 2016). Given its centrality in contemporary
discussions of abstract objects, it warrants more elaboration than the one-over-many argument.
According to this sort of argument, the statements we accept commit us to the existence of the
objects to which they refer or over which they quantify (Quine 1948). Typically, first-order logic is
used as a device to lay bare the underlying structure of the statements we assert, and to thereby
reveal our ontological commitments. For example, consider the statement, “Savannah is a philoso-
pher.” In first-order predicate logic, we may express this statement as follows:
Ps
A Priori: Atheism
where P symbolizes the predicate “is a philosopher,” and s is a singular term that denotes the indi-
vidual, Savannah. If the sentence is to be taken in a literal sense, then if it is true, its components
must genuinely refer to the things they aim to pick out. But if so, then two things must be the case:
(i) Savannah must really exist, and (ii) the predicate “is a philosopher” must correctly apply to her.
So unless one has good reason to think that either (a) the sentence is false, (b) it is more plausibly
given a nonliteral, figurative gloss, or (c) one can replace the sentence with a paraphrase that has dif-
ferent referents, one is ontologically committed to the existence of Savannah, and of her being a phi-
losopher.
Similarly, this holds for statements involving existential quantifiers (i.e., expressions used to
assert that some entity or entities exist, such as “some,” “there is,” and “there exists”). For example,
consider the statement, “A red ball exists.” We can expose the logical structure of the sentence in
terms of first-order logic, as follows:
9x(x is a ball & x is red)
In English, the formalized statement says that there exists at least one entity x, such that x is a ball
and x is red. If the sentence is to be taken in a literal sense, then if the sentence is true, its compo-
nents must all genuinely refer. But if so, then at least two things must be the case: (i) there must be
an object in the domain over which the statement quantifies—in this case, a ball, and (ii) the predi-
cate “is red” must genuinely apply to it. But if so, then the object in the domain must really exist. So
unless one has good reason to think that the sentence is more plausibly given a nonliteral, figurative
gloss, or unless one can replace the sentence with a paraphrase that has different referents, one is
thereby ontologically committed to the existence of the ball.
Given this sketch of how the statements we accept carry ontological commitments, consider
some statements that are relevant to our topic:
(1) Two is an even number.

(2) There is a number between 8 and 10.
(3) Orange resembles red more than it resembles blue.
We can express statements (1)–(3) in first-order logic to clarify what is required for their literal
truth:
(1’) Et
(2’) 9x(x is a number & x is between 8 and 10)
(3’) Rorb
(1’) and (3’) are statements involving singular terms. As such, their literal truth requires that they
have genuine referents that really exist. But the referent of t in (1’) is the number 2 (not the
numeral 2, but the entity to which that numeral refers), and the referents of o, r, and b in (3’)
are apparently universals—namely, redness, orangeness, and blueness. By contrast, (2’) is an existen-
tially quantified statement. Therefore, the literal truth of (2’) requires that there really exists an
object in the domain at issue—that is, a number—such that it is between 8 and 10, namely, the
number 9. But where is the number 9? It certainly does not seem to be a concrete physical object.
Rather, it seems to be an abstract object. Furthermore, it is hard to deny that (1’)–(3’) are true state-
ments. Therefore, unless we can offer adequate paraphrases of (1’)–(3’) that do not make reference
to these entities, it seems that they likewise commit us to saying that abstract objects exist (viz., the
numbers and the color universals they refer to).
Mark Balaguer (2016) helpfully standardizes the indispensability argument as follows:
1. If a simple sentence (i.e., a sentence of the form “a is F,” or “a is R-related to b,” or …) is literally
true, then the objects that its singular terms denote exist. (Likewise, if an existential sentence is
literally true, then there exist objects of the relevant kinds; e.g., if “There is an F ” is true, then
there exist some Fs.)

A Priori: Atheism
2. There are literally true simple sentences containing singular terms that refer to things that could
only be abstract objects. (Likewise, there are literally true existential statements whose existential
quantifiers range over things that could only be abstract objects.)
3. Therefore, Abstract objects exist.
RESPONSES TO ARGUMENTS FOR ABSTRACT OBJECTS

Given the arguments for abstract objects sketched above, how might one respond? Of course, one
response is to accept the arguments’ premises, and thereby to accept the existence of abstract
objects. As we have seen, this view is known as platonism. On this view, one accepts the existence
of abstract objects. However, many philosophers are reluctant to accept such entities. Such philoso-
phers therefore tend to accept one of the views sketched below instead.
Another response is to embrace some version of nominalism about abstract objects. Accord-
ing to views of this sort, there are no abstract objects. Rather, there are only concrete objects.
Nominalists account for the data appealed to in arguments for platonism in a variety of ways.
(To simplify, let us focus on nominalism about universals.) Predicate nominalists try to account
for the data of one-over-many arguments by saying that (e.g.) there is no universal—redness—
that many particular objects have. Rather, there are just particular red things, and we apply the
predicate “red” to all of them. Similarly, concept nominalists (or conceptualists) say that there are
only particular red things, and that to say that each among a plurality of things is red is just to
say that each one falls under our concept of redness. Others—class nominalists—say that redness
is to be identified with the class of all and only red things. Yet others—resemblance nominalists—
say that something is red just in case it resembles the red things. Finally, trope nominalists take
tropes (i.e., concrete, particular property-instances, such as a particular red patch on the surface
of a billiard ball) to be fundamental constituents of concrete reality, in addition to ordinary phys-
ical objects. According to trope nominalists, redness is to be identified with the plurality or class
of red tropes.
With a version of nominalism in hand, one might aim to rationally resist platonism by using it
as the basis for a response to one-over-many and indispensability arguments. In response to the for-
mer, one might argue that a particular nominalist theory of (say) universals explains the data of
(e.g.) resemblance at least as well as platonism. And in response to the latter, one might resist prem-
ise 2 of the indispensability argument by offering nominalistic paraphrases of statements that seem
to commit one to (say) platonism about universals.
A third sort of response is to embrace some version of fictionalism about abstract objects.
According to views of this sort, there are no abstract objects. As such, all statements that seem to
be referring to abstract objects are literally false. However, we pretend or speak as if such entities
exist because of their ability to enhance the representational power of our language (e.g., to help
us express rich and complex truths about the world simply and efficiently).
PROBLEMS ABSTRACT OBJECTS POSE FOR THEISM

Abstract objects raise at least five problems for classical theism. The first four are traditional, and are
associated with the so-called aseity-sovereignty doctrine (AD):
(AD)
i. God does not depend on anything distinct from himself for his existence, and
ii. everything distinct from God depends on God’s creative activity for its existence.
(PLANTINGA 1980; GOULD 2014; CRAIG 2016)
By contrast, the fifth has only recently received focus in the literature (but see Plantinga 2007). It
will therefore be discussed last.
The first problem abstract objects pose for classical theism stems from their aseity (Plantinga
1980; Morris and Menzel 1986; Davidson 1999, 2015; Gould 2014; Craig 2016). According to
classical theism, part of God’s greatness consists in his uniquely enjoying a superior mode of being:
A Priori: Atheism
metaphysically necessary, ontologically independent existence. All other entities are created beings,
or creatures, and thereby have an inferior mode of being: metaphysically contingent, ontologically
dependent existence. Furthermore, prima facie, the chain of ontological dependence cannot be infi-
nite. Therefore, God is the Supreme Being on which all other entities depend for their existence.
But if abstract objects exist, then God is not uniquely supreme in these respects. For as we have
seen, abstract objects likewise enjoy his mode of being. But if so, then God is just one metaphysi-
cally necessary self-existent being among an infinite sea of such beings.
Things are of course worse if God is a contingent being—that is, a being that exists in some pos-
sible worlds, including the actual world, but not others, as (for example) Richard Swinburne (2004,
2016) has argued. For then the mode of God’s being is inferior to that of abstract objects: while
abstract objects exist in all possible worlds, God does not. In either case, God’s greatness is dimin-
ished by the existence of abstract objects.
The second problem pertains to the threat abstract objects pose for God’s sovereignty (Plan-
tinga 1980; Morris and Menzel 1986; Davidson 1999, 2015; Gould 2014; Craig 2016). For if
classical theism is true, then God created and sustains all things apart from himself. But at least
very many abstract objects are entities that are distinct from God that God cannot (and there-
fore did not) create. This is because abstract objects are self-existent, necessarily existent,
uncaused beings. As such, God cannot be their creator or sustainer of their existence. Nor can
he alter their natures, as they have their properties essentially. But these facts are incompatible
with God’s sovereignty. Their existence thus seems to be incompatible with a core thesis of clas-
sical theism.
The third problem is perhaps more worrisome than the last. The idea is that, if the god of clas-
sical theism exists, then he is, at a minimum, all-knowing, all-powerful, and morally perfect. But
these appear to be his properties—indeed, they appear to be some of the core properties that consti-
tute God’s nature or essence. But if properties are necessarily existent abstract objects that cannot be
created, then God’s existence and nature are ontologically posterior to, and dependent on, the exis-
tence of abstract objects (Morris and Menzel 1986; Craig 2016).
This point can be clarified further by means of the larger ontological frameworks standardly
associated with platonism. On platonism, properties are among the basic elements of being, out of
which other things are “built” or constituted. Platonic accounts of properties are usually tied to
either relational ontologies or constituent ontologies of concrete objects. According to Platonic rela-
tional ontologies, a concrete object is a bare particular or a substratum that stands in a primitive
relation of exemplification to its properties. By contrast, on Platonic constituent ontologies, concrete
objects are bundles of properties standing in relations of constitution or fusion. On either sort of
account, it is natural to suppose that God is, in a worrisome sense, a dependent being: he is onto-
logically dependent on more fundamental constituents out of which he is composed. But this seems
to conflict with God’s sovereignty and aseity (Craig 2016).
The fourth problem relates to the third and pertains to the doctrine of divine simplicity—that
is, the doctrine that God is not a composite being in terms of parts and metaphysical structure. As
we saw above, God appears to have “metaphysical parts.” But if God is a simple entity, then he has
no parts. Therefore, since the doctrine of divine simplicity has traditionally been taken to be a cen-
tral doctrine of classical theism, abstract objects seem to pose a formidable problem for the view
(Plantinga 1980; Davidson 2015; Craig 2016).
The fifth problem abstract objects pose for classical theism does not have any obvious connec-
tion to the aseity-sovereignty doctrine. It is the problem of explaining how God can have knowledge
of abstract objects. To see the problem, recall the Benacerraf argument that abstract objects pose for
human knowledge of abstract objects: Prima facie, knowledge requires causal contact with the thing
known. But causal contact is impossible with respect to abstract objects. Therefore, if abstract
objects exist, then we cannot have knowledge of them. The same problem seems to arise for God’s
knowledge of abstract objects. For if abstract objects are mind-independent and acausal, then God
cannot interact with them in the way required for knowledge any more than we can. platonism
about abstract objects therefore poses a serious problem for God’s omniscience (Plantinga 2007;
Baras 2017).

A Priori: Atheism
THEISTIC RESPONSES
The five problems sketched above appear to pose formidable problems for theism. In response, the-
ists have proposed a number of solutions to the problem of God and abstract objects. Of these, four
views have gained a significant number of adherents: platonism, theistic activism, theistic conceptu-
alism, and theistic nominalism.
The first sort of theistic response is theistic platonism (Plantinga 1980; Yandell 2014). According
to this response, abstract objects exist, and they do so independently of God and his causal activity.
Theists who accept this sort of view typically respond to the problems discussed earlier in one or
more of the following ways. First, they argue that such a stance is unproblematic from the stand-
point of revealed theology, on the grounds that sacred scripture or tradition does not clearly teach
that abstract objects (if they exist) were created by God (Yandell 2014). Second, they argue that the-
istic platonism is unproblematic from the standpoint of philosophical theology. For abstract objects
exist and have their natures of necessity, and it is impossible to create or alter an item that exists and
has its nature of necessity. But since omnipotence only ranges over possible states of affairs, it is no
strike against God’s omnipotence or greatness that he cannot create abstract objects.
Despite theistic platonism’s attractions for the theist who accepts abstract objects, a number of
problems have been raised for the view. First, as we have seen, it requires denying AD, which
requires rejecting perfect-being theology (Craig 2016). Many theistic Platonists are willing to accept
this requirement. Because of this, they are also often willing to abandon the doctrine of divine sim-
plicity, and willing to abandon God’s modal supremacy.
However, a second problem is more worrisome for theists who accept the authority of sacred
scripture. For it is widely acknowledged among biblical scholars and theologians that the biblical
writers and ancient church authorities embraced Middle Platonism, according to which putative
abstract objects are ideas and thoughts in the mind of God (Craig 2016). Given this, it is not clear
that theistic platonism is consistent with revealed theology.
Third, while the theistic Platonist who rejects perfect-being theology may find several of the
problems abstract objects pose for theism easy to shrug off, the problem of God’s ontological depen-
dence on abstract objects is not shrugged off so easily. For then God is ontologically posterior to and
dependent on the existence and ontological structure of modal space. But if that is right, then again,
God is, in a worrisome sense, a dependent being.
Another standard view is theistic activism (Morris and Menzel 1986). According to this, abstract
objects are real. However, they are causally or ontologically dependent on either God’s nature or
God’s will.
Most theists reject straight theistic activism for at least two reasons. First, some argue that it is
prima facie incompatible with the classical view of creation (Craig 2016). The root of the problem
lies with the eternality and necessary existence of abstract objects: in order for God to explain their
eternality and necessity, one must posit that God causes them of necessity, whether timelessly or
everlastingly. But then this conflicts with the standard theistic account of divine creative acts. For
according to that account, (i) created entities have a temporal beginning, coinciding with the
moment of their creation by God, and (ii) God’s creative acts are supposed to be acts of free will,
in the libertarian sense that free will requires the ability to do otherwise (Craig 2016). But if so,
then we have a dilemma: either one sticks with the classical view of creation or one revises it so as
to allow creation to be compatible with a denial of (i) and (ii).
Consider each option in turn. If one sticks with the classical view of creation, then God does
not create the vast majority of objects distinct from God; comparatively speaking, God is respon-
sible for creating a tiny portion of the things that exist. Rather, most things—the infinite platonic
horde of abstract objects—are merely ontologically dependent on God, and not created by him
(Craig 2016).
On the other hand, suppose one revises the classical view of creation so that eternal and nonfree
divine acts can count as acts of divine creation. Then most of what God creates is involuntary, in
that it is not “up to” God to create them. But this amounts to an abandonment of divine creation
of abstract objects for a kind of theistic emanationism (Craig 2016).
A Priori: Atheism
The most serious objection to theistic activism is the bootstrapping problem (Davidson 1999,
2015; Bergmann and Brower 2006; Craig 2016). The core of the problem is that at least some
abstract objects must exist and be exemplified by God before God can create them, which is meta-
physically impossible. To illustrate the worry, suppose it is possible for God to create properties, a
kind of abstract object. Now suppose that God is setting out to create them. In order to carry out
his plan, God must first have the power to create properties. But then in order for God to create
properties, God must already have at least one property, namely, having the power to create properties.
Therefore, God must already have some properties before he can create them, which is impossible.
Therefore, since the assumption that God can create all properties logically entails an impossibility,
that assumption must be false—in other words, it is impossible for God to create all properties.
Primarily because of the bootstrapping problem, most theists reject theistic activism. However,
some theists hold on to the view and modify it in ways so that it avoids the bootstrapping problem.
According to this sort of view—modified theistic activism (Gould and Davis 2014)—God does not
cause or create the properties within his nature. Rather, he only causes or creates abstract objects,
so to speak, outside the borders of his being. But what about the status of things inside the borders?
There are two obvious options for the modified theistic activist at this point: (i) adopt a Platonic
construal of God’s nature, so that he has a real nature or essence composed of real yet uncaused
properties, or (ii) adopt a nominalist or fictionalist construal of God’s nature, so that he lacks such
properties.
Option (i) has been subject to at least five criticisms. First, adopting the view comes at the
expense of rejecting the AD, which imperils classical theism (Craig 2016). Second, modifying the-
istic activism in this way comes at the cost of making the view seem unmotivated: if some prop-
erties exist uncreated and independent of God’s causal activity, why not the rest (Davidson 2015;
Craig 2016)? Third, the view entails that God is ontologically dependent on his independently
existing divine nature, in which case God is, in a significant sense, a dependent being, in which
case he does not exist a se (Craig 2016). Fourth, it seems that on such a view, any being who
should exemplify the divine nature would thereby be God, which diminishes God’s uniqueness
and supremacy (Leftow 2012). Finally, the view looks to be incoherent. For one constituent of
the independently existing divine nature—that is, deity—is the property aseity—that is, self-exis-
tence. But then in order to exemplify deity, God must exemplify aseity. But then God’s aseity
depends on his exemplifying the property aseity. And if his aseity depends on exemplifying a prop-
erty, then he does not exist a se. This is incoherent, as it says that God’s aseity ontologically
depends on his exemplifying aseity, and no entity that is dependent on another can be one that
has aseity (Craig 2016).
Some theists who adopt option (i) aim to avoid the problems by conjoining it with the doctrine
of divine simplicity. According to this doctrine, God is absolutely simple: he is identical to his prop-
erties, which in turn are identical to each other. He is therefore not dependent on his ontologically
prior constituents. There are a number of versions of the doctrine. However, the standard criticism
of virtually all of them is that the doctrine entails an incoherent or absurd view of God. For exam-
ple, if God is identical with his properties, and properties are abstract objects, then God is an
abstract object, which is absurd (Plantinga 1980; Davidson 2015; Craig 2016).
Given the problems besetting option (i), a theist with activist leanings might thus look to
option (ii). This sort of view has the obvious benefit of dissolving the bootstrapping problem. It also
seems to avoid most of the problems of option (i). However, the benefits come at the high cost of
making theistic activism look unmotivated. For if one being (i.e., God) can exist without exempli-
fying Platonic properties, it is no longer clear why any being cannot exist without them (Davidson
2015; Craig 2016).
Historically, the most widely accepted response is theistic conceptualism. According to this view,
abstract objects are identified with items in the mind of God. On standard accounts, properties are
divine concepts, propositions are divine thoughts, possible worlds are maximal consistent sets of
divine thoughts, and sets are divine collections (Plantinga 1980, 1993b, 2007; Welty 2014). In this
way, theistic conceptualists aim to account for God as the creator or cause or ground of abstract
objects.

A Priori: Atheism
Despite its historical and contemporary popularity among theists, several criticisms have been
leveled against it. First, if God causes his thoughts, then the bootstrapping problem we saw in our
discussion of theistic activism arises all over again for theistic conceptualism. For on the latter view,
properties are divine thoughts. But if so, then for God to have the property being able to create, he
must first produce the corresponding concept. But before he can produce the corresponding con-
cept, he must first have the property being able to create. Therefore, in order for God to create prop-
erties, he must already have some (Davidson 1999; Craig 2016).
As with theistic activists, it is open to conceptualists to respond to the bootstrapping problem
with a modified version of their view, according to which God’s nature is given a Platonistic or
nominalistic construal. But then the same sorts of worries that arose for that view would seem to
arise with the present view (Craig 2016).
Second, theistic conceptualism seems to unduly restrict God’s freedom. For in order to explain
the objectivity, necessity, and infinity of abstract objects, theistic conceptualism entails that God
must think all of his thoughts (and just those thoughts) and must think them constantly (Oppy
2014; Craig 2016).
Third, God’s thoughts are mental states, and, as such, are concrete particular objects, not uni-
versals. If so, then they cannot be repeatable or multiply exemplifiable entities. But many abstract
objects, such as properties and propositions, are repeatable entities. If so, then prima facie, God’s
mental states are unsuited to play the role assigned to them by theistic conceptualism (Gould and
Davis 2014; Craig 2016).
Fourth, theistic conceptualism seems to imply that God has ungodly thoughts. For if all pro-
positions are divine thoughts, and there are propositions about things that are prima facie beneath
the greatness and moral perfection of God—silly things, evil things, trivial things, naughty things,
and so on—it follows that God has thoughts about things that are beneath him. Indeed, he must
think about them, constantly (Oppy 2014). This seems problematic, given God’s greatness and
moral perfection.
Finally, there are worries about the content and granularity of God’s thoughts. For if
God’s thoughts are inherently intentional or representational, as the theistic conceptualist
insists, it is not clear what his thoughts could be about with respect to merely possible entities.
Some theistic conceptualists (Welty 2006; Leftow 2012) say that his thoughts are about his
power to create. However, as has been pointed out (Adams 2013; Oppy 2014), it is not clear
that thoughts about God’s power alone are sufficiently fine-grained to individuate qualitatively
identical entities.
The fourth and final view theists take in response to the problem of abstract objects is theistic
nominalism. According to this sort of view, there are no abstract objects. Rather, there are only par-
ticular, nonrepeatable, concrete entities. According to this view, then, God is a concrete object, and
he is a person who is all-knowing, all-powerful, and perfectly good. However, God is not this way in
virtue of exemplifying ontologically prior abstract objects (such as universals or individual essences)
on which God ontologically depends. For according to theistic nominalism, there are no such enti-
ties (Craig 2016).
Unlike those who accept one of the other responses, the theistic nominalist denies that there are
abstract objects. As such, they alone have the burden of criticizing the one-over-many argument and
the indispensability argument. Against the one-over-many argument, nominalists naturally take
some version of nominalism about universals (e.g., resemblance nominalism, trope nominalism) to
be at least as good an explanation of resemblance as platonism.
Against the indispensability argument, theistic nominalists have leveled criticisms against both
premises. Since William Lane Craig is the theistic nominalist who has provided the most sustained
and systematic critique of the argument to date, his key points feature heavily in what follows.
Against the first premise, Craig argues that there are a number of ways to resist the claim that
our language commits us to the existence of its referents. For example, neutral logicians propose a
quantifier P (= “for some”) that is not ontologically committing, and which can be used in the place
of the existentially committing “9” (Craig 2016). Jody Azzouni argues that not even the ordinary
A Priori: Atheism
existential quantifier is ontologically committing (Craig 2016). Furthermore, in the place of the
seemingly ontologically committing objectual semantics for quantification, which are presupposed
in premise 1 of the indispensability argument (where the truth of a statement requires objects in a
domain to which its terms refer or over which they quantify), one can instead appeal to substitu-
tional semantics (where the truth of a statement does not require a domain of objects). On the latter
approach, the truth of an existentially quantified statement requires only that at least one term can
be substituted for the variable that yields a true statement.
There are also logical systems that do not require ontological commitments for singular terms. So,
for example, according to free logic, all singular terms in one’s statements have a referent, even if the ref-
erent is not a real object. To get this result, free logics appeal to a dual-domain semantics. In the first
domain are ordinary objects to which singular terms typically refer. In the second domain are fictional
objects of pretense. Given this account, statements can be true even if they fail to refer to real objects.
Theistic nominalists likewise have a range of options for rejecting premise 2 of the indispens-
ability argument (Craig 2016). However, all such options involve an appeal to fictionalism (broadly
construed) about abstract objects. For example, they might appeal to a standard fictionalist account
according to which statements referring to abstract objects are useful fictions that are, strictly speak-
ing, false. By contrast, they might appeal to a figuralist account, according to which such statements
are true when taken as figures of speech (e.g., “He was born with a silver spoon in his mouth”), but
not true when taken literally. Finally, they might appeal to pretense theory, according to which talk
of abstract objects is a form of pretending or make-believe, and therefore does not obviously require
ontological commitments to their referents.
The core criticism of the nominalist response is that it cannot explain the data of abstract objects
better than the hypothesis of platonism. With respect to the one-over-many argument, many contend
that nominalist accounts are implausible or otherwise inadequate. For example, some argue that
“ostrich” nominalism, which asserts that there are many particular red things but denies a univer-
sal—redness—that they all share, implausibly leaves unaccounted for a fact that cries out for explana-
tion (Armstrong 1980). Others argue that resemblance nominalism, which is perhaps the most popu-
lar and sophisticated version of nominalism to date, lacks the explanatory scope of Platonist theories.
For example, they have trouble accounting for distinctions between necessarily coextensive properties,
such as triangularity versus trilaterality (Allen 2016, 2018).
More sophisticated versions of nominalism have been developed and defended that aim to
account for such worries, but such accounts often come at the expense of massively bloating one’s
ontology and adding doctrines that seem no less counterintuitive than platonism’s doctrine of multiple
instantiation. For example, Gonzalo Rodriguez-Pereyra’s (2002) sophisticated version of resemblance
nominalism aims to handle the problem of necessarily coextensive properties. However, the solution
requires accepting modal realism, according to which there is an infinite number of in-principle unob-
servable concrete possible worlds that are just as real as the actual world. It also requires accepting
counterpart theory, according to which no individual can exist in more than one possible world.
ABSTRACT OBJECTS AS A PROBLEM FOR ATHEISM?

Atheism is the view that there is no person such as God. As such, there is no explicit entailment
from atheism to a denial of abstract objects. On the face of it, then, it is not clear how abstract
objects even could pose a problem for atheism.
Sometimes atheism is conflated with materialism—the view that only material entities exist—
and that is of course incompatible with the view that there are abstract objects as we have character-
ized them. However, atheism does not entail materialism. For example, one can consistently accept
both atheism and Russellian monism (the view that the fundamental stuff of concrete reality has
both physical and proto-phenomenal properties). Such a person would be an atheist who is not a
materialist. Similarly, one can consistently accept both atheism and platonism, and thereby be an
atheist who is not a materialist.
Perhaps, though, one will argue that the most plausible forms of atheism entail materialism,
which in turn entails a denial of abstract objects and proto-phenomenal properties at the foundations
of reality. However, it is not clear how such reasoning might go. Perhaps one might argue that if

A Priori: Atheism
Russellian monism is true, atheism cannot account for how proto-phenomenal properties arose from
the physical; nor can they explain why abstract objects exist at all, if abstract objects exist.
But these concerns are confused. As to the first: On Russellian monism, the proto-phenomenal
properties are not derivative entities that arose from physical stuff and its properties. Rather, proto-
phenomenal properties are as fundamental as physical properties, and so did not arise or emerge
from the latter. And as to the second: As we have seen, according to standard Platonic accounts
of abstract objects, they are uncaused, eternal, necessary beings that exist a se. As such, they need
no explanation for their existence any more than God needs such an explanation (if he should turn
out to exist). But it is epistemically possible that a form of atheism that entails both platonism and
Russellian monism turns out to be the best explanation of all the relevant data across the board. It is
therefore not clear that the most plausible versions of atheism entail materialism.
In conclusion, we looked at some of the main lines of evidence for the existence of abstract
objects. We also looked at some of the main ways in which both theism and atheism can account
for such evidence. From our discussion, we saw that, at the very least, theism does not provide a
better explanation of such data than atheism.
Bibliography
Adams, Robert Merrihew. “Necessary by Nature? Possibility and the Mind Oxford Studies in Metaphysics. Vol. 2, edited by Dean W. Zimmerman,
of God.” Review of Brian Leftow’s God and Necessity. Marginalia Review 357–386. Oxford: Oxford University Press, 2006.
of Books, 20 December, 2013 https://marginalia.lareviewofbooks.org
BonJour, Laurence. In Defense of Pure Reason: A Rationalist Account of A
/necessary-by-nature-possibility-and-the-mind-of-god/.
Priori Justification. Cambridge, UK: Cambridge University Press, 1998.
Allen, Sophie R. A Critical Introduction to Properties. London: Bloomsbury
Academic, 2016. Chalmers, David J. The Conscious Mind: In Search of a Fundamental Theory.
New York: Oxford University Press, 1996.
Allen, Sophie R. “Properties.” In Internet Encyclopedia of Philosophy. 2018.
https://www.iep.utm.edu/properties/. Chalmers, David J. “Does Conceivability Entail Possibility?” In Conceivability
and Possibility, edited by Tamar Szabó Gendler, and John Hawthorne,
Alston, William P. “Some Suggestions for Divine Command Theorists.” In 145–200. Oxford: Clarendon; New York: Oxford University Press, 2002.
Divine Nature and Human Language: Essays in Philosophical Theology.
Ithaca, NY: Cornell University Press, 1989. Chudnoff, Elijah. Intuition. Oxford: Oxford University Press, 2014.
Armstrong, D. M. “Against Ostrich Nominalism: A Reply to Devitt.” Pacific Craig, William Lane. God over All: Divine Aseity and the Challenge of
Philosophical Quarterly 61 (1980): 440–449. Platonism. Oxford: Oxford University Press, 2016.
Augustine. On Free Choice of the Will. Translated by Thomas Williams. Davidson, Matthew. “A Demonstration against Theistic Activism.” Religious
Indianapolis, IN: Hackett, 1993. Studies 35, no. 3 (1999): 277–290.
Balaguer, Mark. “Platonism in Metaphysics.” In Stanford Encyclopedia of Davidson, Matthew. “God and Other Necessary Beings.” In Stanford
Philosophy, edited by Edward N. Zalta. Spring 2016. https://plato.stanford Encyclopedia of Philosophy, edited by Edward N. Zalta. Spring 2015.
.edu/entries/platonism/. https://plato.stanford.edu/entries/god-necessary-being/.
Baras, Dan. “A Reliability Challenge to Theistic Platonism.” Analysis 77, De Cruz, Helen. “Numerical Cognition and Mathematical Realism.”
no. 3 (2017): 479–487. Philosopher’s Imprint 16, no. 16 (2016): 1–13.
Bardon, Adrian. “Reliabilism, Proper Function, and Serendipitous Malfunc- Feldman, Richard. “Proper Functionalism.” Noûs 27, no. 1 (1993): 34–50.
tion.” Philosophical Investigations 30, no. 1 (2006): 47–64.
Fischer, Bob. “A Theory-Based Epistemology of Modality.” Canadian
Bealer, George. “Modal epistemology and the rationalist renaissance.” Journal of Philosophy 46, no. 2 (2016): 228–247.
In Conceivability and possibility. Edited by T. Gendler & J. Hawthorne,
71–125. Oxford: Oxford University Press, 2002. Fischer, Bob. Modal Justification via Theories. Cham, Switzerland: Springer,
2017.
Benacerraf, Paul. “Mathematical Truth.” Journal of Philosophy 70, no. 19
(1973): 661–679. Fischer, Bob, and Felipe Leon. “The Modal-Knowno Problem.” Southwest
Philosophy Review 32, no. 1 (2016): 225–232.
Bengson, John. “Grasping the Third Realm.” In Oxford Studies in Epistemology.
Vol. 5, edited by Tamar Szabó and John Hawthorne, 1–38. Oxford: Oxford Fischer, Bob, and Felipe Leon, eds. Modal Epistemology after Rationalism.
University Press, 2015a. Cham, Switzerland: Springer International, 2017.
Bengson, John. “The Intellectual Given.” Mind 124, no. 495 (2015b): Gould, Paul. “Intentionality and God: A Review Essay of R. Scott Smith’s
707–760. Naturalism and Our Knowledge of Reality.” Philo 15, no. 1 (2012): 97–105.
Bergmann, Michael, and Jeffrey Brower. “A Theistic Argument against Gould, Paul, ed. Beyond the Control of God?: Six Views on the Problem of God
Platonism (and in Support of Truthmakers and Divine Simplicity).” In and Abstract Objects. London: Bloomsbury, 2014.
A Priori: Atheism
Gould, Paul, and Richard Brian Davis. “Modified Theistic Activism.” In Plantinga, Alvin. Warranted Christian Belief. New York: Oxford University
Beyond the Control of God?: Six Views on the Problem of God and Abstract Press, 2000.
Objects, edited by Paul Gould, 51–79. London: Bloomsbury, 2014.
Plantinga, Alvin. “Two Dozen (or So) Theistic Arguments.” In Alvin
Graham, Peter J. “Intelligent Design and Selective History: Two Sources of Plantinga, edited by Deane-Peter Baker, 203–228. Cambridge, UK:
Purpose and Plan.” In Oxford Studies in Philosophy of Religion 3, edited Cambridge University Press, 2007.
by Jonathan Kvanvig, 67–88. Oxford: Oxford University Press, 2011.
Quine, W. V. “On What There Is.” Review of Metaphysics 2, no. 1 (1948):
Graham, Peter J. “The Function of Perception.” In Virtue Epistemology 1–19.
Naturalized: Bridges between Virtue Epistemology and Philosophy of Science,
Quine, W. V. “Two Dogmas of Empiricism.” Philosophical Review 60, no. 1
edited by Abrol Fairweather, 13–31. Cham, Switzerland: Springer (1951): 20–43.
International, 2014.
Roca-Royes, Sonia. “Similarity and Possibility: An Epistemology of de re
Greco, John. “Virtue and Luck, Epistemic and Otherwise.” Metaphilosophy Possibility for Concrete Entities.” In Modal Epistemology after Rational-
34, no. 3 (2003): 353–366. ism, edited by Bob Fischer and Felipe Leon, 221–246. Cham,
Hawke, Peter. “Van Inwagen’s Modal Skepticism.” Philosophical Studies 153, Switzerland: Springer International, 2017.
no. 3 (2011): 351–364. Rodriguez-Pereyra, Gonzalo. Resemblance Nominalism: A Solution to the
Huemer, Michael. “Compassionate Phenomenal Conservatism.” Philosophy Problem of Universals. Oxford: Clarendon; New York: Oxford University
and Phenomenological Research 74, no. 1 (2007): 30–55. Press, 2002.
Leftow, Brian. God and Necessity. Oxford: Oxford University Press, 2012. Ruloff, Colin. “Divine Thoughts and Fregean Propositional Realism.”
International Journal for Philosophy of Religion 76, no. 1 (2014): 41–51.
Lehrer, Keith. “Proper Function versus Systematic Coherence.” In Warrant
in Contemporary Epistemology: Essays in Honor of Plantinga’s Theory of Ruloff, Colin. “On Propositional Platonism, Representation, and Divine
Knowledge, edited by Jonathan Kvanvig, 25–46. Lanham, MD: Rowman Conceptualism.” European Journal for Philosophy of Religion 8, no. 4
and Littlefield, 1996. (2016): 195–212.
Leon, Felipe. “From Modal Skepticism to Modal Empiricism.” In Modal Senor, Thomas D. “A Critical Review of Alvin Plantinga’s Warranted
Epistemology after Rationalism, edited by Bob Fischer and Felipe Leon, Christian Belief.” International Philosophical Quarterly 42, no. 3 (2002):
247–262. Cham, Switzerland: Springer International, 2017. 395–396.
Loux, Michael, and Thomas Crisp. Metaphysics: A Contemporary Introduc- Swinburne, Richard. The Existence of God. 2nd ed. Oxford: Clarendon; New
tion. 4th ed. New York: Routledge, 2017. York: Oxford University Press, 2004.
MacLeod, Mary, and Eric M. Rubenstein. “Universals.” In Internet Swinburne, Richard. The Coherence of Theism. 2nd ed. Oxford: Oxford
Encyclopedia of Philosophy. https://www.iep.utm.edu/universa/. University Press, 2016.
Moreland, J. P. Universals. Montreal: McGill-Queen’s University Press, 2001. Thomasson, Amie L. Ontology Made Easy. New York: Oxford University
Press, 2014.
Morris, Thomas V., and Christopher Menzel. “Absolute Creation.”
American Philosophical Quarterly 23, no. 4 (1986): 353–362. Van Inwagen, Peter. “God and Other Uncreated Things.” In Metaphysics
and God: Essays in Honor of Eleonore Stump, edited by Kevin Timpe, 3–
Morriston, Wes. “Must There Be a Standard of Moral Goodness Apart from 20. London: Routledge, 2009.
God?” Philosophia Christi (series 2) 3, no. 1 (2001): 127–138.
Welty, Greg. Theistic Conceptual Realism: The Case for Interpreting Abstract
Nichols, Shaun. “Imaginative Blocks and Impossibility: An Essay in Modal Objects as Divine Ideas. PhD diss., University of Oxford, 2006.
Psychology.” In The Architecture of the Imagination: New Essays on
Welty, Greg. “Theistic Conceptual Realism.” In Beyond the Control of God?:
Pretence, Possibility, and Fiction, edited by Shaun Nichols, 237–256.
Six Views on the Problem of God and Abstract Objects, edited by Paul
Oxford: Clarendon; New York: Oxford University Press, 2006. Gould, 81–96. London: Bloomsbury, 2014.
Nolan, Daniel, Greg Restall, and Caroline West. “Moral Fictionalism versus Wielenberg, Erik J. Value and Virtue in a Godless Universe. Cambridge:
the Rest.” Australasian Journal of Philosophy 83, no. 3 (2004): 307–330. Cambridge University Press, 2005.
O’Connor, Timothy. Theism and Ultimate Explanation: The Necessary Shape Wielenberg, Erik J. Robust Ethics: The Metaphysics and Epistemology of Godless
of Contingency. Malden, MA: Blackwell, 2008. Normative Realism. Oxford: Oxford University Press, 2014.
Oppy, Graham. “Abstract Objects? Who Cares!” In Beyond the Control of Williamson, Timothy. The Philosophy of Philosophy. Malden, MA: Blackwell,
God?: Six Views on the Problem of God and Abstract Objects, edited by 2007.
Paul Gould, 169–181. London: Bloomsbury, 2014.
Wolterstorff, Nicholas. On Universals: An Essay in Ontology. Chicago:
Plantinga, Alvin. Does God Have a Nature? Milwaukee, WI: Marquette University of Chicago Press, 1970.
Wunder, Tyler. “Anti-naturalism and Proper Function.” Religious Studies 44,
Plantinga, Alvin. “How to Be an Anti-realist.” Proceedings and Addresses of the no. 2 (2008): 209–224.
American Philosophical Association 56, no. 1 (1982): 147–170.
Yablo, Stephen. “Is Conceivability a Guide to Possibility?” Philosophy and
Plantinga, Alvin. Warrant: The Current Debate. New York: Oxford Phenomenological Research 53, no. 1 (1993): 1–42.
University Press, 1993a.
Yandell, Keith. “God and Propositions.” In Beyond the Control of God?: Six
Plantinga, Alvin. Warrant and Proper Function. New York: Oxford Views on the Problem of God and Abstract Objects, edited by Paul Gould,
University Press, 1993b. 21–35. London: Bloomsbury, 2014.

TOPIC 11
Our Universe: Theism

Robert J. Spitzer, S.J.
President
Magis Center of Reason and Faith, Garden Grove, CA
James Sinclair
Senior Physicist
United States Navy, Washington, DC
This chapter will discuss the arguments that have been made in favor of theism on the basis of the fine-
tuning of our universe for life. It will discuss the objective basis of fine-tuning, six instances of it in our
universe, and consider six candidates as an ultimate explanation for it—a theory of everything, inflationary
cosmology (and its variants), cyclic cosmologies (including Penrose’s conformal cyclic model), a multiverse,
Tegmark’s Level IV multiverse, and transcendent intelligence. Using a general abductive argument, it
concludes that the most likely ultimate explanation is transcendent intelligence.
FINE-TUNING AND INDICATIONS OF

TRANSCENDENT INTELLIGENCE
As we begin our investigation into fine-tuning, we encounter an unexpected mystery—an orderly uni-
verse. Albert Einstein intuitively recognized this when he noted, “The most incomprehensible thing
about the universe is that it is comprehensible” (1954, 292). In his insightful lecture at New York
University in 1960, the Nobel Prize–winning physicist Eugene Wigner refined Einstein’s intuition
by noting the inexplicable capacity of mathematics to describe whole classes of physical phenomena).
As this chapter explains, this mystery lies at the heart of the fine-tuning question. In order to
explore this mystery, we define fine-tuning and indicate when it occurs (in the first section), and
in the next section address six specific instances of it; in the four sections that follow, we review a
variety of naturalistic explanations—from a theory of everything to inflationary cosmologies, cyclic
cosmologies, multiverse hypotheses, and Max Tegmark’s Level IV multiverse; and we conclude with
an assessment of the need for transcendental intelligence as an ultimate explanation.
WHAT IS FINE-TUNING?
In the literature of physics and cosmology, fine-tuning has a specific meaning that can be best
understood by first defining three terms—free parameters, constants, and initial conditions.
A parameter is a specific measurable quantity that defines the operation or the conditions of a
physical process. Some parameters, termed fixed parameters, are predicted or constrained by a partic-
ular physical-mathematical model or theory, but others are not. These unfixed parameters, termed
free parameters, are not predicted by a physical-mathematical theory, and therefore must be mea-
sured independently of the theory and then inserted into the equations expressing that theory. Thus
the values of free parameters are neither necessitated by nor derivable from a particular physical
331
KEY CONCEPTS
Abductive Argument Inflationary Cosmologies
Big Bang Theory Initial Conditions
Boltzmann Brain Initial Conditions of Our Universe
Brief Brain Mass-Energy
Canon of Parsimony (Ockham’s Razor) Multiverse Hypotheses
Causal Dynamical Triangulations Nonintelligence Proposal/Intelligence
Causal Set Theory Proposal
Conformal Cyclic Cosmology (CCC) Order/Disorder
Constants Past Hypothesis
Cyclic Cosmologies Q Nonuniformity
De Sitter Space Schrödinger Paradox
Elementary Particle Theory Second Law of Thermodynamics
Entropy, Low Entropy, High Entropy Space-Time
Fine-Tuning String Theory Landscape
Four Universal Forces Sum-over-Histories (SOH) Approach to
Quantum Gravity
Free Parameters
Tegmark’s Level IV Multiverse
General Theory of Relativity
Theory of Everything
Hamiltonians
Inflation, Eternal Inflation Transcendental Intelligence
theory or the equations that express them. Such free parameters might be explained by more funda-
mental theories (describing more fundamental laws and equations) or they might be quite ran-
dom—occurring at the big bang or in the very early universe by chance—or by a transphysical
cause. As we shall see, deciding among these three options (more fundamental laws, chance occur-
rence, or supernatural design) is not certain science—or even certain logic, and so it does not offer
the same level of certitude as a metaphysical proof of God (see Spitzer 2010, 110–176). Neverthe-
less, we can list the pros and cons of each option under certain assumptions, and formulate an
abductive argument leading to a reasonable, most likely candidate for an ultimate cause of fine-tun-
ing on the basis of currently understood physics and cosmology.
Universal constants are one kind of free parameter (i.e., a particular measurable quantity that is
not necessitated or predicted by any of the known theories explaining our universe and its laws).
These measurable quantities are thought to be invariant throughout the universe and to remain
“constant” throughout past, present, and future time. Thus these measurable quantities particularize
the equations of physics (expressing the laws of interaction and causation) within our universe. Does
our universe (or any other possible universe) have to have these particular measurable quantities—
these constants? No. Inasmuch as they represent free parameters of the equations describing our
universe or other possible universes, they are not necessary. Yet, as we shall see, some values of
the constants will make a universe life-permitting while others far more frequently will prohibit life.
We discuss some of these constants in the next section, after looking at free parameter space in the
initial conditions of our universe.
“Initial conditions of the universe” refers to “givens” at the big bang that gave rise to the elemen-
tary constituents through which the above-mentioned forces and constants operate. It includes the ini-
tial mass-energy density of our universe, its low entropy, the dimensionality of our universe (three
spatial dimensions and one time dimension), the Q nonuniformity of mass-energy immediately after
the big bang, the specific duration of the initial inflationary period, and the imbalance of matter over
antimatter. As with all free parameters, the measurable quantities associated with these initial condi-
tions are neither necessitated nor predicted by any physical or cosmological theory (or the laws associ-
ated with those physical theories). Thus, as far as we currently know, they seem to have arisen in our
universe independently of the laws (and equations) that they determine and constrain.
We are now in a position to define fine-tuning. This expression refers specifically to fine-tuning
for the emergence of life within our universe—or within any universe with physical laws similar to
those of our universe. An instance of fine-tuning for life occurs if a free parameter has a value
needed for the emergence of life-forms, and if such a value is exceedingly improbable within the
total available parameter space of our universe. So, for example, if the total free parameter space
for the values of a given free parameter (say, the electromagnetic charge or the Q nonuniformity
of mass-energy density) could be mapped on a grid that is three miles by three miles, but the param-
eter space permitting life is only one millimeter by one millimeter on that grid (exceedingly improb-
able), and the value of the free parameter in our universe just happens to fall within that relatively
small area, it appears to be an instance of fine-tuning for life. The fact that this free parameter just
happens to have an exceedingly improbable value needed for life and that it is not necessitated or
predicted by any physical theory (or the laws underlying that theory) requires a causal explanation
beyond a one-off pure chance (which is virtually unrealizable).
As we shall see, there is a multiplicity of these exceedingly improbable coincidences, all of
which are necessary for life to emerge (six of them are explained in the next section). Furthermore,
these coincidences seem to occur on the most fundamental levels of physics and cosmology. The
deeper we go in the probing of the laws of nature to explain the higher levels of fine-tuning, the
more we find additional instances of fine-tuning. This is what has led many physicists and philoso-
phers to appeal to the above natural and transcendent causal options.
The six instances of fine-tuning given below do not require a precise definition of “life” or a
detailed analysis of possible life-forms in other hypothetical universes, because they are concerned with
the most fundamental requirements for all possible life-forms—that is, consistent, lawlike actions and
interactions within a universe that can give rise to complex structures and processes needed for metab-
olism and reproduction. In our universe, this entails physical laws and constants that will not result in
a dispersion of noninteracting (or rarely interacting) matter, a preponderance of black holes, initial
high entropy, and the impossibility (or improbability) of chemistry (atomic and molecular bonding).
It should be mentioned that Victor Stenger (2011) has charged that fine-tuning is merely a
matter of the subjective point of view of physicists. Luke Barnes has definitively responded to Sten-
ger’s charge by showing the fallacy of equivocation in his central argument (2012, 6–9). He goes on
to illustrate how this equivocation produces false conclusions in several areas of physics (e.g., Gali-
lean relativity, Lagrangian dynamics, Newtonian mechanics, general relativity, and the standard
model of particle physics and gauge invariance—Barnes 2012, 8–17). After showing several mistakes
in Stenger’s physics, Barnes concludes to the objective basis of fine-tuning for life.
SIX INSTANCES OF FINE-TUNING
The following instances of fine-tuning are taken from three major areas of physics and cosmol-
ogy: (1) initial conditions of our universe, (2) the four universal forces, and (3) elementary
particle theory.
APPARENT FINE-TUNING IN THE INITIAL CONDITIONS OF OUR UNIVERSE

We consider three instances of fine-tuning in the universe’s initial conditions: (1) the low entropy of
the universe, (2) the density of mass-energy shortly after the big bang, and (3) Q nonuniformity of
mass-energy density immediately after the big bang.
Low Entropy at the Big Bang. Since the publication of Roger Penrose’s classic The Emperor’s New
Mind (1989), the cosmological world has faced a haunting conundrum: despite the fact that the low

entropy necessary for life to emerge and develop is exceedingly improbable, it lavishly occurred at
the big bang. Since the random occurrence of low entropy is not predicted by any physical or cos-
mological theory and, indeed, is opposed by the second law of thermodynamics, we are confronted
by the need to explain how this could possibly happen.
Let us begin with a brief explanation of the second law of thermodynamics and the improbability
of low entropy’s occurrence at the big bang. Entropy is a measurement of disorder within a physical
system. Order refers to the potential of a physical system to produce extractable energy for physical
activity. (For some examples, see Lewis and Barnes 2016, 98–101.) Here is the problem presented
by the second law of thermodynamics: Whenever there is a flow of energy, the order of the system will
never increase; it almost always decreases. This means that every time a physical system produces a
flow of energy, it will lose a little of its order—its potential to produce extractable energy for physical
activities. This has the consequence that no active physical system can last forever—there are no per-
petual motion machines. Every active physical system will eventually run down until it dies. This
includes our universe. (Some hypothesized exceptions, such as eternal inflation, are discussed below.)
Here is where fine-tuning comes into play. Low entropy (high order) is much more improbable
than high entropy (the disorder of a system that has little potential to produce extractable energy).
Note that low entropy is necessary for the emergence, maintenance, and evolution of life, for with-
out sufficient extractable energy, there would be virtually no opportunity for the complexification of
atomic and molecular constituents needed for the emergence of amino acids and proteins, let alone
DNA and cellular structures. This gives rise to a cosmological enigma—if low entropy is highly
improbable, and it is necessary for life to emerge and develop in any universe, then how did our
universe achieve this highly improbable but necessary state at its inception?
This enigma was considerably amplified when Penrose, the mathematician-physicist of the Uni-
versity of Oxford, calculated123the odds against the low entropy of our universe occurring by pure
chance at its inception: 1010 to one! This number is so large that it is tantamount to a monkey’s
typing the whole of William Shakespeare’s Macbeth by random tapping of the keys in a single try—
that is, virtually impossible. Penrose phrases his conclusion, with a bit of tongue in cheek, as fol-
lows: “This now123tells us how precise the Creator’s aim must have been: namely to an accuracy of
one part in 1010 ” (1989, 343–344).
Some hypotheses have been offered to explain this exceedingly improbable condition in our uni-
verse by natural causation—such as eternal inflation, the multiverse, and new versions of the cyclic
universe (including one from Penrose himself). We respond to these below, but for the moment we
can conclude, along with Penrose and the majority of the physics community, that low entropy at
the big bang is a very probative instance of fine-tuning for life in our universe. Failure to find a plau-
sible explanation for it would give a significant epistemic indication of a transcendent intelligence.
The Initial Mass-Energy Density of the Universe. There are two interrelated initial conditions
of the universe that determine its geometry and uniformity, which in turn determine its capacity to
allow for the emergence and development of life—its initial mass-energy density and the Q nonuni-
formity of mass-energy. These initial conditions are free parameters (not explained by any physical-
cosmological theories or the laws underlying them). As we shall see, the values of these free para-
meters of the universe’s initial conditions required for our universe to allow for life are exceedingly
tiny by comparison with their potential available values immediately after the big bang, leading to
an inference of universal fine-tuning for life. We discuss the universe’s critical mass-density in this
section and the Q nonuniformity in the next section.
Einstein’s theory of general relativity shows a causal correlation between the density of mass-
energy and the curvature of space-time. The greater the density of mass energy, the greater the pos-
itivity of the curvature of the universe’s space-time (leading to higher gravitational attraction)—and
vice versa (Einstein 2014, 125–141). In view of this, the density of mass-energy immediately after
the big bang will determine whether the universe will be able to allow and sustain life. There are
three options, only one of which will allow life:
• Too great a density of mass-energy, causing too much positive curvature of the universe (which
would lead to an almost immediate collapse of the universe into a black hole, preventing the
development of any life-form)
• Too small a density of mass-energy, causing insufficient positive curvature of the universe
(leading to a highly accelerated dispersion of mass-energy, preventing the formation of structures,
such as galaxies, stars, and planets, thereby preventing the formation of life)
• The critical (precisely right) density of mass-energy leading to a flat or very nearly flat universe,
which avoids the above two catastrophes, allowing for the emergence and development of life
Geraint Lewis and Luke Barnes calculate that in order to have the nearly flat universe we have
today, the critical density one nanosecond after the big bang would have to have been very close
to 1024 kilograms per cubic meter. If the mass-energy had been but one kilogram per cubic meter
more, the universe would have collapsed in on itself by now (interfering with the formation of life),
and if it had been but one kilogram less (out of 1024 kg per cubic meter), the universe would have
expanded so rapidly that it would never have formed stars or galaxies necessary for life (Lewis and
Barnes 2016, 167; see also Rees 2000, 97–101).
When we consider that the mass of the universe (relative to its gravitational constant) could
have been much higher (having virtually no intrinsic limit explicable by the universe itself), or con-
siderably lower (almost any amount less than the total visible matter and dark matter in the uni-
verse), and that the universe’s mass-energy density is fine-tuned for life to within one kilogram
per cubic meter out of 1024 kilogram per cubic meter—one part in a trillion trillion—one nanosec-
ond after the big bang, we have to wonder how this could have occurred by pure chance. We are
virtually constrained to admit that there must be some causal explanation beyond sheer randomness,
and we are again led to five potential candidates: a theory of everything, inflationary cosmologies,
cyclic cosmologies, a multiverse, or a transcendent intelligence (discussed below).
Q Nonuniformity. The cosmic microwave background radiation reveals that the universe was
exceedingly uniform immediately after the big bang. Patches of the cosmic microwave background
radiation causally unconnected to one another throughout most of universal history are almost identi-
cal in temperature, with only a few very small variations. This could not have happened unless the
universe had been almost perfectly uniform immediately after the big bang. Indeed, the actual unifor-
mity deviated from average uniformity by only one part in 100,000—the Q nonuniformity (Lewis
and Barnes 2016, 167–168).
This precise Q nonuniformity within almost perfect uniformity is required for life to develop in
the universe. Lewis and Barnes (2016) note that relatively small deviations from Q nonuniformity
would be catastrophic for the occurrence of life. If Q had been decreased by only a factor of 10
(to one part in a million), the universe would have been too smooth. There would not have been
enough nonuniformity for stars and galaxies to form during the universe’s deceleration shortly after
the big bang. The absence of this slight nonuniformity would not have enabled gravity to, as it
were, grip a bump or ripple in mass-energy distribution that would later lead to stars and galaxies.
Alternatively, too much nonuniformity—too much “bumpiness”—would have been equally disas-
trous, because gravity would have too much bumpiness to grip, leading to an accelerated formation
of stars and galaxies. If Q had been greater by only a factor of 10 (to one part in 10,000), the resul-
tant acceleration of star and galaxy formation in the early universe would have caused closely com-
pacted stars to interfere with planetary formation and their stable orbits, leading to an absence of
stable planets required for life. Furthermore, if Q had been greater by a factor of 1,000 (to one part
in 100), the entire universe would have become huge black holes in short order (Lewis and Barnes
2016, 169–170; see also Rees 2000, 126–129, and Uzan 2011).
Some physicists have proposed cosmic inflation as a natural explanation for universal smooth-
ness within the very tiny Q nonuniformity, but as we shall see, inflation does not explain this
remarkable initial condition, because cosmic inflation requires initial smoothness before inflation
begins, and what is even more perplexing, cosmic inflation itself must be incredibly fine-tuned in
its duration in order to stop the inflation after about 10–100 of a second after it starts. Without this
precise, incredibly short duration, the universe would have been a virtual void of disbursed energy
without the capacity for interaction needed for life (Ijjas, Steinhardt, and Loeb 2013, 2014; Ijjas
and Steinhardt 2015; Penrose 2016, 300; Lewis and Barnes 2016, 173). Neil Turok (2002) has
called this “the famous fine-tuning problem of inflation.” (See the extended explanation below.)

In conclusion, the above three instances of fine-tuning of initial conditions require a causal expla-
nation, which is examined below. There are several other instances of fine-tuning of initial conditions,
such as the dimensionality of our universe and the asymmetry of matter and antimatter. (For an expla-
nation of these and other such cosmic coincidences, see Lewis and Barnes 2016, 205–209, 221–233).
FINE-TUNING OF THE FUNDAMENTAL FORCES

All four fundamental forces—gravity, the strong nuclear force, the electromagnetic force, and the
weak force—are fine-tuned for life. Here, we discuss only the ways in which gravity and the strong
force are fine-tuned. (Examples of fine-tuning in the other forces may be found in Lewis and Barnes
2016, Barnes 2012, and Rees 2000.)
The Fine-Tuning of Gravity. Gravity is one of the four universal forces, controlling virtually every
reaction from the early universe to the production of stars and the initiation of life over the course
of 13.8 billion years. The strength of gravity dictates the curvature of space-time relative to the den-
sity of mass-energy. The stronger the gravitational force constant (G), the stronger the force of
attraction (the so-called pull of gravity on masses), and vice versa. Gravity is so fundamental that,
as Robin Collins notes, every universe within a hypothetical multiverse would have to have it to
be life-permitting: “Without a universally attractive force between all masses, such as gravity, matter
would not be able to form sufficiently large material bodies (such as planets) for life to develop or
for long-lived stable energy sources such as stars to exist” (2009, 264).
Furthermore, the gravitational force constant would have to be fine-tuned relative to the laws
and forces of each of those universes for life to emerge in them. Such fine-tuning would have to
be similar to that described below for our universe (though the values of the constants could be dif-
ferent in those other universes).
In our universe, the value of G (6.67410−11 N·kg–2·m2) must thread the parameter bound-
aries for life in several areas—the initial expansion of the universe, its initial cooling, and the forma-
tion of stable stars, among others. If gravity were either too weak or too strong, life in our universe
would be impossible. Let us examine each of the above parameter boundaries.
With respect to the initial expansion of the universe, we already noted above (concerning the
critical mass-energy density of the universe) that if the universe had expanded too rapidly, no galax-
ies, stars, or planets would have formed, leaving a weakly interactive dispersion of energy incapable
of producing the stable complexes needed for life. Conversely, if the expansion had been too slow,
the universe would have quickly collapsed into a universal black hole or several black holes (incapa-
ble of producing life). We can see why gravity (the universal force of attraction among all masses
and energy) must play an important role in this regard, for if the value of G had been slightly lower
(and gravity slightly weaker), the universe would have expanded too rapidly to form galaxies, stars,
and planets. Conversely, if the value of G had been slightly higher, we would be in a black hole uni-
verse.
Gravity also plays an important role in the cooling process of the universe in the first fifteen
minutes or so after the big bang. Slight variations in the value of G would have affected the cooling
time of the universe. A stronger gravitational force would lead to the earlier end of neutron produc-
tion, “with roughly equal numbers of neutrons and protons. Almost all the protons will be locked
up inside helium” (Lewis and Barnes 2016, 78). The lack of free protons (and hydrogen) would
have negatively affected the formation of productive stable stars and the emergence of life.
The strength of gravity relative to the strength of electromagnetism also appears to be fine-
tuned for life. If gravity had been substantially weaker (relative to the electromagnetic force), no
stars would have formed, and if it had been only slightly weaker, stars would have been colder
(including our sun) and would not have exploded in supernovae, giving rise to heavier elements.
This would have negatively affected the emergence and development of life (Friederich 2018,
1.1.1; Carr and Rees 1979). Alternatively, if gravity had been slightly stronger (relative to the elec-
tromagnetic force), stars would have been smaller, hotter, and more short-lived—emitting intense
life-threatening gamma rays and dying quickly in supernovae explosions, and negatively affecting life
(Friederich 2018, 1.1.1; Adams 2008).
Gravity is also sensitive to the strong nuclear force. Now, this may seem counterintuitive,
because the strong nuclear force is 1040 times greater than gravity. Fred Adams estimates that if
gravity were only 1035 times weaker than the strong nuclear force (instead of being 1040 times
weaker), there would be no stable stars in our universe, thereby eliminating the possibility of life
(Adams 2008; Lewis and Barnes 2016, 109).
As may now be evident, the precise strength of gravity relative to universal expansion, universal
cooling, and stellar formation and stability seems to be remarkably fine-tuned for life. As noted
above, none of the theories of physics (or the laws underlying them) requires that gravity have this
precise value. As will be seen, superstring theory (our best candidate for quantum gravity and a the-
ory of everything) does not require gravity’s value relative to the other constants and initial condi-
tions, and so we must make recourse to another explanation beyond a one-off, purely random
occurrence.
Fine-Tuning of the Strong Nuclear Force. The value of the strong nuclear force coupling con-
stant is essential for life. If it did not exist or its value was negligible, there would be no atomic
nuclei, and therefore no periodic table and no chemistry. We have already seen how the value of
the strong force constant is so sensitive to the value of the gravitational constant for stable stellar for-
mation and the emergence of life (see Adams 2008).
We can also see a similar sensitivity of the value of this constant relative to electromagnetism
for the emergence of life. If, for example, we were to eliminate or substantially reduce the energy
differential between the strong nuclear force and electromagnetism, the universe would become cha-
otic, because relatively small amounts of heat would transform one kind of nucleus into another.
The periodic table would be in complete flux, making stable chemical structures virtually impossible
(Lewis and Barnes 2016, 74).
The strong force (relative to electromagnetism) is even more precisely fine-tuned. If the strong
force had been about 50 percent stronger, all hydrogen would have been burned up in the early uni-
verse, preventing the formation of stable stars and the emergence of life. Alternatively, if it had been
50 percent weaker, stellar nucleosynthesis would have been much less efficient and few, if any, ele-
ments beyond hydrogen would have formed. Evidently, this would have prevented the emergence
and development of life (Friederich 2018, 1.1.1).
There is more precise fine-tuning of the strong force constant relative to quantum resonance
levels required for the emergence of living organisms. Carbon production is essential for life as we
know it because very few other elements will allow for the development of complex molecules
needed for living organisms. Silicon is the closest element to carbon and it is not nearly as capable
of producing complex molecules. Likewise, oxygen is an important component for life as we know
it because it is ingredient to water and air conducive to respiration.
In 1953 Fred Hoyle discovered that the strong force constant had to be very highly fine-tuned
relative to the resonance levels (quantum resonances that form the ground state and other states of
the atom) to produce the amounts of carbon and oxygen in stars needed for life (Hoyle, Dunbar,
Wenzel, and Whaling 1953). He showed that if carbon had a resonance level equivalent to that
of beryllium and helium and some extra energy, an abundance of carbon could be formed in stars.
Without this resonance, carbon would be quite rare. Additionally, the resonance level of oxygen
(immediately below that of carbon) is also necessary for an abundance of carbon.
The value of the strong force constant is important in the combined production of carbon and
oxygen in stars. If the strong force constant had been greater by only 0.4 percent, there would be a
greater abundance of carbon, but virtually no oxygen. Alternatively, if the strong force constant had
been less by a factor of only 0.4 percent, there would have been an abundance of oxygen, but virtu-
ally no carbon (Lewis and Barnes 2016, 118). Without this precise fine-tuning of the strong force
constant, the universe would have essentially lacked either the capacity to form complex organic
molecules needed for life (i.e., virtually no carbon) or the capacity to produce water and breathable
gas that can be directly metabolically active (i.e., virtually no oxygen). It is said that this fine-tuning
coincidence prompted Hoyle to move from atheism to belief in a supernatural designer (Gingerich
2000, 524–525).

Review. The examples given above show the multiple ways in which the value of the gravitational
constant and the strong force constant must be fine-tuned for the emergence of life. The reader can
infer from the above the degree to which the electromagnetic force must also be fine-tuned (relative
to gravity and the strong force). It will come as no surprise that the value of the weak force constant
is also quite fine-tuned, but this is beyond the scope of this essay. (For an accessible presentation of
this topic, see Lewis and Barnes 2016, 78, 92–93.)
THE FINE-TUNING OF THE MASSES OF FUNDAMENTAL PARTICLES

There are three particles whose masses are particularly important for stable particle physics, chemis-
try, and the emergence of life—the up-quark and the down-quark (the two lightest quarks among
six) and the electron. (For an explanation of quarks, quark masses, and their bearing on protons
and neutrons, see Lewis and Barnes 2016, 42–53.) If the difference in mass between the up- and
down-quark were altered, then the universe would be substantially different, prohibiting life (Hogan
2007). If the mass of the down quark were increased three times (relative to the mass of the up-
quark), there would be a hydrogen-only universe, preventing complex molecules and, therefore, life.
Alternatively, if the mass of the up-quark were increased six times (relative to the down-quark), all
protons would decay, leaving only neutrons incapable of molecular bonding, thereby preventing life
(Hogan 2007, 224; Lewis and Barnes 2016, 50–51).
Furthermore, atomic and molecular bonding would also be eliminated by altering the mass of
the electron relative to the mass differential of the up- and down-quarks. Increasing or decreasing
the mass of the electron relative to the mass differential of the quarks would have similar effects
to that of changing the masses of the quarks themselves—the universe would be devoid of complex
nuclei and molecular bonding, preventing the formation of living organisms (Friederich 2018;
Hogan 2000).
Craig Hogan (2007, 226) and Lewis and Barnes (2016, 52) provide a graph of the differential
of the up- and down-quarks relative to one another and the electron mass, which illustrates the very
tiny zone of parameter space allowing life to develop in relation to the available parameter space for
values of those masses. It indicates just how fine-tuned those three masses must be for life to arise in
our universe. The graph shows a region of parameter space allowing life to develop at about 1 square
centimeter. In contrast to this, note Lewis and Barnes, the region of available parameter space is
about 32 square kilometers (about 12.355 square miles)—an exceedingly tiny zone indicating very
precise fine-tuning (Lewis and Barnes 2016, 52).
S. M. Barr and Almas Khan (2007) also show the fine-tuning of the up- and down-quark masses
relative to the weak force. Hogan’s (2007) illustration of the restricted parameter space for life (in
relation to the huge available parameter space) and Barr and Khan’s (2007) analysis of the quark
masses in relation to the value of the weak force assess multiple parameters in a single analysis. These
kinds of multiple-dimensional-parameter analyses obviate Stenger’s criticism that fine-tuning analysis
fails to vary more than one parameter at a time. Evidently this is not the case. Barnes (2012) gives
other instances of multidimensional parameter analysis (as in Tegmark, Aguirre, Rees, and Wilczek
2006).
IMPLICATIONS
There are several other well-accepted instances of fine-tuning of the constants of our universe
going beyond the scope of this essay, such as the fine-tuning of the mass of the Higgs boson,
the Planck constant, the electromagnetic force, and the weak force. (Readers interested in these
constants may want to consult Lewis and Barnes 2016; Rees 2000; Collins 2009; and Barnes
2012.) The above instances of fine-tuning of initial conditions and universal constants cannot
be explained by a pure chance occurrence or our laws of nature, and so require an explanation
leading us into five areas at the border of physics and metaphysics: a theory of everything, new
inflationary and cyclic cosmologies, a multiverse, Tegmark’s Level IV multiverse, and a transcen-
dent intelligence.
CAN A “THEORY OF EVERYTHING” BE AN ULTIMATE EXPLANATION

OF FINE-TUNING?
What might we conclude from the above six instances of fine-tuning of initial conditions and the
constrained parameters of the four universal forces and the masses of particles?
First, the objectively real parameter space allowing for life is exceedingly tiny by comparison
with the objectively real available parameter space for the laws of nature. This is not a merely subjec-
tive conclusion but an objective one based on the ratio of real life-permitting parameter space amid
the real available life-prohibiting parameter space of our real physical laws. This exceedingly small
probability provokes a legitimate question of why the highest dimensions of physical evolution (e.
g., living organisms, complex organisms, and intelligent life) occurred when the real available param-
eter space of our physical laws dictate that it should not. The answer could be anything from ran-
dom occurrence to an intrinsic cause from within our universe to an extrinsic trans-physical cause
like a designer—or some combination of them.
As noted above, a one-off random occurrence of this fine-tuning is virtually impossible and
should be ruled out as a potential explanation (Penrose 2016, 317). Therefore, a much more sophis-
ticated mechanism that would multiply random natural occurrences, such as a multiverse, will have
to be proposed to make the random natural occurrence argument workable (see below).
For the moment, let’s turn to the second option—a cause intrinsic to our universe. So what
might an intrinsic cause of all the life-permitting initial conditions and constants look like? It would
have to be a unified theory of everything that would be mathematically determinative for all the initial
conditions and constants underlying all the equations expressing all the laws of nature. Does such a
theory exist?
There was some hope several years ago that superstring theory would provide a theory of every-
thing (unifying all four universal forces—the electromagnetic, strong, and weak forces with the grav-
itational force). Though superstring theory provides a complex, mathematically feasible way of
doing this, it does not require the values of the constants and initial conditions necessary for the
occurrence of life. String theory transforms the constants of our equations into initial condition
solutions, but those numbers still remain free parameters—unconstrained by the requirements of
the theory (see Lewis and Barnes 2016, 296; Penrose 2016, 322). An alternative theory of quantum
gravity—loop quantum gravity—does not offer much greater hope of constraining free parameters
toward life, so at present we don’t have any realistic options for an intrinsic explanation, such as a
theory of everything. Though one might speculate that such a theory exists (but lies beyond the
bounds of current knowledge), it will not really resolve the problem of fine-tuning.
Even if a theory of everything were to be discovered, it would only show why the constants in
our laws of nature are mathematically determined toward life. It would not answer the further ques-
tion of why our laws of nature are mathematically determinable toward life in the first place. Why is
it that our universe has physical laws that are determinable toward life when they are not logically
necessary—that is, when there are other logically consistent sets of natural laws that are not life-
permitting and cannot become life-permitting? This problem was set out by Bernard Carr and
Martin Rees as follows: “Even if all apparently anthropic coincidences could be explained in this
way [by a complete physical theory], it would still be remarkable that the relationships dictated by
physical theory happened also to be those propitious for life” (1979, 612).
In 2009 Collins articulated the same problem and applied it not only to a theory of everything
but also to why multiverse theories (which are premised on the laws of physics or consistent adapta-
tions of them) will not be able to provide an ultimate answer to the fine-tuning problem. In so
doing, he set out some of the conditions that prospective new physical laws would have to have
in order to be life-permitting: for example, a multiverse generator that would give rise to laws of
nature allowing a universe to be filled with mass-energy of some kind and that would also require
some kind of universal force of attraction (acting like Einstein’s space-time gravity rather than New-
ton’s gravity)—something akin to the quantization of energy, the Pauli exclusion principle, and
other laws allowing matter to develop into diverse complexes needed for life-forms (Collins 2009,
264). As we shall see, a combination of physical laws to meet Collins’s requirements is exceedingly

improbable, implying not only fine-tuning of a theory of everything in our universe but also a mul-
tiverse that could give rise to our universe. As such, a theory of everything will never be an ultimate
explanation of fine-tuning in our universe. Yet given the proliferation of new multiverse theories
allowing for a virtually infinite number of universes, wouldn’t such theories be able to provide such
an ultimate explanation for fine-tuning? Couldn’t some of the new inflationary and cyclic cosmolo-
gies do the same? We begin with a consideration of the latter.
CAN INFLATION (AND ITS VARIANTS) AND CONFORMAL CYCLIC COSMOLOGY

BE AN ULTIMATE EXPLANATION OF FINE-TUNING?
Recall from above that Penrose calculated the odds against our low entropy occurring at the big
123
bang at about 1010 to one—roughly the chance that a monkey could type Macbeth perfectly by
random tapping of the keys in a single try. This highly improbable objectively real state of affairs,
which indicates an exceedingly high degree of fine-tuning at the big bang, has stubbornly resisted
naturalistic attempts to explain it. As we shall see in this section, our universe’s low entropy is
unlikely to be explained by inflationary cosmology (and its variants) or conformal cyclic cosmology.
It is also the undoing of the multiverse hypothesis (discussed below). In this section, we examine the
implications of our universe’s low entropy in greater depth and then examine inflationary cosmology
and Penrose’s conformal cyclic cosmology in light of it.
Let’s return to the question left unanswered above: If low entropy is so highly improbable, and
it is necessary for life to emerge and develop in any universe, then how did our universe achieve this
highly improbable but necessary state at its inception? Of interest to this study is a theoretical for-
mulation called the past hypothesis. When the problem is posed in terms of our best physics tool
set (statistical thermodynamics), we encounter a bizarre and interesting paradox in regard to
entropy. What is entropy? One generally takes it to be a measure of the number of possible states
a system can be in with a certain description (a “coarse-grained state”) compared to the total num-
ber of configurations. Statistical thermodynamics will say entropy should rise whether looking to the
future or past. To make sense of records of the past (in this case our memories), we must have a past
hypothesis, whereby we postulate that entropy was always lower in the past. That is the second law
of thermodynamics. (See the example of glass with ice from Carroll 2014.)
Penrose’s major contribution has been to show that “something new” is necessary to account
123
for the extreme low-entropy state at the big bang (which is exceedingly improbable at 1010 to
one against). One of the earlier hypothetical explanations of this initial low entropy was thought
to be inflationary theory. As implied above, inflation is a very brief period (lasting from 10–36 sec-
onds to somewhere between 10–33 and 10–32 seconds after the big bang) in which the universe
underwent exponential expansion of space. It is thought to have occurred through a hypothetical
inflation field or cosmological constant. As we shall see below, it allows for the possibility of an
inflationary multiverse in which our universe is but one bubble amid a potentially infinite number
of other bubble universes.
Now let us explore the question: Can inflation adequately explain the low entropy of our uni-
verse? This has led to a serious cosmological debate going back to 2002, although one could say it
stretches back to the nineteenth century and the Austrian physicist Ludwig Boltzmann. Faced with
the problem of “heat death,” which implied a finite past in an age when the universe was thought to
be static and past eternal, Boltzmann capitalized on the concept of a Poincaré recurrence. The
French mathematician Henri Poincaré had suggested that a system with a finite number of possible
configurations would, given enough time, visit/revisit all its possible configurations. Thus, although
extremely unlikely, a move from high to low entropy, even in a closed system, is possible.
How does this apply to cosmology? Boltzmann’s results lead one to believe that if infinite time
has already passed, the universe should be in a lethal state of thermal equilibrium (heat death),
which clearly it is not. If the whole universe undergoes Poincaré recurrences, that could account
for our observations. Boltzmann erred, however, in thinking that a whole universe recurrence is
what we should observe. Rather, the minimum entropy reduction recurrence that accounts for our
existence would be overwhelmingly more likely. What is the minimum (instantaneous) life-
supporting system necessary? It is probably something like a solar system (surrounded by
chaos). We don’t see this, so Boltzmann was/is wrong. This counterargument has been offered
by many physicists going back to Arthur Eddington, Richard Feynman, and John Barrow and
Frank Tipler (Carroll 2008).
But could eternal inflation supply a justification to the past hypothesis? Suppose, for example,
that inflation was eternal into the past as well as the future. For eternal inflation our “big bang” is
not a transcendent event. Rather, it is an inflationary bubble that came into being as a decay process
from an outside realm.
Inflation is said to solve the “flatness” problem (i.e., low entropy) within our own universe, by
smoothing out curvature (gravitational degrees of freedom expressing entropy in terms of high cur-
vature) in the same way a basketball inflated to the size of Earth could appear to be perfectly flat.
This would have to fit within the physics of the recurrence as inflationary degrees of freedom are
part of the larger system. Lisa Dyson, Matthew Kleban, and Leonard Susskind (2002) put this to
the test in their paper “Disturbing Implications of a Cosmological Constant,” which evaluated the
impact of Poincaré recurrences in light of inflationary cosmology. The authors operate within some
fairly complex assumptions beyond the scope of this chapter. Suffice it to say that their rules assume
an observer within a “causal patch,” which implies a finite entropy upper bound, which implies
recurrences. (Banks 2001 gives an extended argument in favor of this view.)
To cue the argument, the idea is that inflation could have started off as a tiny, random, but
highly ordered part of an infinitely sized “generic” manifold. Of interest to us is the fact that the
entropy within the causal patch is exponentially large. An idealized model of the patch corresponds
to one of the first models developed in general relativity by Dutch cosmologist Willem de Sitter. A
de Sitter space is empty except for the presence of a cosmological constant. In the limit as our uni-
verse approaches the far future, its description asymptotically approaches a de Sitter universe. The
entropy of a de Sitter universe is about the same as that represented by a generic singularity. Whether
inflation started like this or came from a nontranscendent event like a local decay from some outer
reality into an inflationary bubble, the beginning point of “our” bubble must account for these sta-
tistical degrees of freedom. Dyson, Kleban, and Susskind indicate:
Inflationary starting points are very rare in time.… The danger is that there are too many
possibilities which are anthropically acceptable, but not like our universe. Let us consider the
entropy in observable matter in today’s universe. It is of order 10100. This means that the
number of microstates that are macroscopically from our world is exp (10100). But only exp
(1010) of these states could have evolved from the low entropy initial state characterizing the
usual inflationary starting point. The overwhelming majority of states which would have
evolved into a world very similar to ours did not start in the usual low entropy ensemble.
(2002, 14)
How are these results to be interpreted? We note for completeness that a multisided conversation in
cosmology has picked up and continued this thread (see, e.g., Albrecht 2002, 2009, 2014; Carroll
2014; Bousso 2012; Bousso and Freivogel 2007; and Penrose 2016).
Sean Carroll (2014) indicates that initial low entropy cannot be accounted for by inflation itself
nor an anthropic principle exercising inflation:
One can show that, in order for inflation to begin, we must not only have very particular
initial conditions in which potential energy dominates over kinetic and gradient energies, but
that the size of the patch over which these conditions obtain must be strictly larger than the
Hubble radius. In other words, even in an inflationary scenario, it is necessary to invoke smooth
initial conditions over super-horizon-sized distances. As compelling as inflation may be, it still
requires some understanding of pre-inflationary conditions to qualify as a successful model.
(Carroll 2014, 12; emphasis added)
And then later:
Reformulating the issue raised by the horizon problem as a matter of the measure on
cosmological trajectories brings the problem with inflation into sharp focus: the fact that most

trajectories that are homogeneous at late times are highly non-homogeneous at early times is
completely independent of physical processes in the early universe. It depends only on the
measure evaluated at relatively late times. Inflation, therefore, cannot solve this problem all
by itself. Indeed, the measure reinforces the argument made by Penrose, that the initial
conditions necessary for getting inflation to start are extremely fine-tuned, more so than
those of the conventional Big Bang model it was meant to help fix. (Carroll 2014, 23;
emphasis added)
As Andreas Albrecht has stated: “Basically, low entropy means ‘in an atypical part of phase space,’ so
how can one expect to argue that such a state is typical?” (2014, 1).
You can have the past hypothesis, or you can have a generic boundary condition, but you can’t
have both. It is Albrecht, however, who has pressed on. His arc of thought has changed from debating
arguments like Carroll’s to seeking out the conditions under which a generic state might give rise to an
apparent low-entropy start. He has grabbed the dilemma by both horns, and suggests, bucking against
most inflationary cosmologists, that one can adequately explain our cosmos by fluctuating from an ini-
tial state (Albrecht 2009, 2014) and without encountering the problems suggested by Dyson, Kleman,
and Susskind, or the related phenomena of “Boltzmann brains” (discussed below).
Albrecht (2009) concedes that the argument by Dyson, Kleman, and Susskind holds for
“normal” inflation, that is, Coleman-De Luccia bubbles (Coleman and De Luccia 1980). But
he argues that another type of inflation, known as FGG inflation (Farhi, Guth, and Guven
1990), may be possible on the grounds that such a fluctuation would constitute an entropy recur-
sion much more probable than a Boltzmann brain—thus typicality. FGG inflation occurs through a
wormhole forming between this reality and an emergent other one. A left-behind observer sees a
black hole.
A wormhole constitutes a topology change for the universe. This is thinkable but has much
going against it. For example, Frank Tipler notes that “all topology changes yield closed timelike
curves (Geroch theorem) and hence singularities (Tipler’s theorem), and hence a violation of unitar-
ity” (Tipler, personal communication with the authors, 3 March 2008; see also Borde 1994). Uni-
tarity (conservation of probability, or information) is something of a holy grail for contemporary
cosmologists, and is not abandoned without much weeping and gnashing of teeth.
Ben Freivogel and his colleagues (2002) claim to uphold this conclusion in string quantum
gravity and list other objectors such as Tom Banks and Susskind who have their own arguments.
Stephen Hawking (2005) weighed in against wormhole formation as well on the basis of unitarity.
A new take on the sum-over-histories (SOH) approach to quantum gravity is that of the causal
dynamical triangulations of Jan Ambjørn, Jerzy Loll, and Renate Jurkiewicz (2010). They suggest,
in their version of non-perturbative quantum gravity, that the success of their prediction comes
from banning causality-violating processes such as wormhole-related topology change.
Another quantum gravity approach that makes use of the SOH is that of the causal set theory
of Rafael Sorkin, David Rideout, Fay Dowker, Arvind Borde, and others (Dowker 2002). They
claim that the FGG process is heavily suppressed in causal set theory quantum gravity for the con-
struction that most likely matches wormhole tunneling with a black hole left behind. “This
implies that in the path integral over spacetime metrics for Einstein gravity in three or more
spacetime dimensions, topology change via a [wormhole] singularity is exponentially suppressed,
whereas appearance or disappearance of a universe [creation ex-nihilo] is exponentially enhanced”
(Sorkin and Louko 1997, 2).
What this means is that causal set quantum gravity predicts that creation of a universe from
“nothing” does happen, but baby universes likely don’t branch off from a mother via a wormhole.
Hence there is a strong theoretical suggestion that an FGG process is prohibited. (Note that this
reasoning would apply to the black hole natural selection theory of Lee Smolin as well.) This implies
that the Albrecht suggestion would not succeed and thereby constitutes a reasonable objection to
the past hypothesis.
What is left? Regarding the past hypothesis, string landscape theorists also posit that one could
start in a high-entropy condition and yet produce a strong thermodynamic arrow. Figure (b)
(Bousso 2012) indicates the concept using a toy model. Imagine four different types of vacua. A
worldline starts off in vacuum C, then undergoes, in effect, a

recurrence by jumping up and settling in vacuum A. A tunnel-
ing event occurs, resulting in a descent and reheating event
(creating stars, galaxies, ordinary observers, etc.), and finishing A
an inflation event by settling in vacuum B. Finally, a tunneling OO
event puts the worldline into the terminal vacuum T with neg-
ative cosmological constant. A negative CC implies an eventual IC
big crunch for the universe.
B
In essence, this is similar to the situation posited by
Albrecht. The difference is that only B is a life-amenable vac- C
uum state. Since the starting point, vacuum C, is not life- (b)
amenable, one does not have the same discussion regarding T
recurrences of different cost. Only the jump to vacuum A
can possibly lead to observers (ordinary, or Boltzmann brains). Vacuum Structure and the Arrow of Time. SOURCE:
So this is unmistakably a case where the initial entropy is REPRINTED FIGURE WITH PERMISSION FROM RAPHAEL
higher than any condition in vacuum B, and yet produces a BOUSSO, PHYS. REV. D, 86,123509, 7 DECEMBER 2012.
standard big bang setting—at least in a toy model. Thus a COPYRIGHT 2012 BY THE AMERICAN PHYSICAL SOCIETY.
fine-tuned low initial entropy for the big bang (a local event
in this scenario) seems unnecessary.
Conceptually, this works. A criticism, however, is that there doesn’t seem to be sufficient
rigor behind the description. Is this (raised to, perhaps 10500 vacua) what the string landscape
looks like? For example, Banks (2012), as well as Ulf Danielsson and Thomas Van Riet (2018),
argue that the only vacua that string theory has successfully described in any type of rigor is a
supersymmetric vacuum with zero cosmological constant (and anti–de Sitter spaces with question-
able connection to de Sitter vacua). The landscape would need to consist mostly of positive cos-
mological constant vacua—de Sitter spaces. Banks’s suggestion is that theories developed so far
for these vacua are independent; the vacua don’t connect in a landscape on the grounds that they
have independent Hamiltonians. A Hamiltonians is the quantum operator related to total energy
of a system. Having a different Hamiltonian implies one has moved from describing multiple
vacua within one theory to having a set of incompatible theories. Furthermore, there are scores
of competing, equally valid constructions in addition to the landscape. Danielsson and Van Riet
seem to suggest that de Sitter spaces and string theory go together like oil and water. Their sug-
gestion, that string theory should be divorced from describing de Sitter spaces and hence falsify
the string landscape, is so new that it is uncertain where it will go.
A connection between string theory and the hypothetical string landscape has not been estab-
lished. That this is the holy grail and resists implementation twenty years after discovery of the cos-
mological constant in our universe seems like a telling fact. An acceptable critique of the adequacy
of string theory and the landscape is, however, well beyond the scope of this chapter. What can
be said is that (unlike Albrecht) Raphael Bousso’s (2012) example can’t be evaluated precisely
because we lack the rigor to test it.
Thus we are brought to Penrose’s own suggestion for accounting for the apparent validity of
the past hypothesis. In order to explain this incredibly low entropy at the big bang, Penrose
hypothesizes a conformal cyclic cosmology (CCC) that evades the above-mentioned inevitability
of cumulatively increasing entropy (leading to a dead universe).
What is the Penrose proposal? This is the idea that something fundamental to the theory (zero-
ing out of the general theory of relativity’s Weyl tensor) suppresses gravitational degrees of freedom
“at” the big bang. Penrose then suggests that an analytic continuation (GTR mathematical tech-
nique) can push time back before the big bang. Now imagine a previous cycle; there is a similar
boundary at future infinity for a de Sitter space. Penrose proposes that one analytically continue
the “end of time” forward such that it connects with a new big bang beyond infinity. Realizing that
going past infinity makes little sense, he suggests that the existence of time just is the existence of
clocks. Clocks are any type of particle except zero mass bosons (like photons). If these particles dis-
appear (in finite time), so does time, and one doesn’t really have to wait till future infinity to get to
a new big bang.

Penrose concedes that this would require completely novel and otherwise unnecessary additions
to particle physics. Even if, for example, there were a grand unified theory that allowed decays of
particles that otherwise seem stable (like the proton), there is still a ground state: the electron.
How does the electron decay? Why should it be suggested to decay except to advance CCC? This
is probably the most serious criticism of CCC.
Now, the “disappearance of time,” given the disappearance of clocks, seems wrong for the same
reason that the disappearance of thermometers would not thereby imply the nonexistence of heat.
Another criticism that is often unappreciated in cosmology involves the problem of the continuity
of time across a boundary. If time “disappears,” then before/after relations disappear with it. The
only justification for saying that an eon exists before the big bang is that one can smoothly follow
a timeline back “before the big bang.” Yet “on the boundary” that must be a theoretically possible
state in CCC, there are no before/after relations with the so-called future or the so-called past. This
implies a contradiction. The boundary is in no sense “before” our eon or “after” the so-called “pre-
vious” eon. Rather, what one has is multiple disconnected realities.
Critical to this is our existence as a monoverse with an asymptotically de Sitter endgame (dS).
But commonly understood de Sitter physics gets in the way of this proposal. For example, an
asymptotic dS state is a maximum entropy condition, with mathematically the same entropy as
the “end state” maximum Weyl curvature singularities that would obtain in a hypothetical future
big crunch (Penrose 2004, 729). The entropy has to match the analytically continued state on the
other side of the boundary (as Penrose would say, from one eon to the next). Now, this brings into
play what happens to black holes (holders of most entropy in the universe) in the far future. They
are thought to decay into radiation and finally disappear! This takes quite a long time—10100 years.
Penrose (2016, 381–386) suggests that the entropy within black holes disappears with them. Given
that this is a unitarity-violating process, he notes that this is a very unpopular proposal among cos-
mologists.
Finally, there is the problem of recurrences themselves. Penrose needs all massive particles
everywhere to disappear. Yet we know that for an asymptotic dS space, there is a nonzero tem-
perature everywhere owing to the cosmological constant. An asymptotic dS space always sits
in a thermal bath. This implies recurrences whereby all one needs to spoil the theory is a single
electron. Penrose would have to argue that the typical time for a recurrence was less than the
time for time to disappear (in his as yet undescribed new physics). Yet that is probably not good
enough. Since horizons are observer-dependent (implying that the whole universe, not just the
visible horizon, must have all electrons decay to make time go away), and our universe may be
spatially infinite besides, there should always be Poincaré recurrent electrons present some-
where. It just doesn’t seem possible to plausibly theorize a universe that becomes empty (in
the most literal sense) given the daunting realities that must be overcome. Therefore, CCC
seems to fail.
It should be mentioned before closing this discussion that other (nonconformal) cyclic cosmol-
ogies are highly unlikely to provide an answer to the fine-tuned low entropy of our universe at the
big bang. Carroll has shown that virtually no such cosmologies can resolve, and instead exacerbate,
the fine-tuning of the universe to a virtually infinite extent:
If you try to invent a cosmology in which you straightforwardly replace the singular Big
Bang by a smooth Big Bounce continuation into a previous space-time, you have one of
two choices: either the entropy continues to decrease as we travel backwards in time
through the Bang, or it changes direction and begins to increase. Sadly, neither makes any
sense.
If you are imagining that the arrow of time is continuous as you travel back through the
Bounce, then you are positing a very strange universe indeed on the other side. It’s one in
which the infinite past has an extremely tiny entropy, which increases only very slightly as
the universe collapses, so that it can come out the other side in our observed low-entropy
state. That requires the state at t= minus infinity state of the universe to be infinitely finely
tuned, for no apparent reason. (The same holds true for the Steinhardt-Turok cyclic
universe.)
On the other hand, if you imagine that the arrow of time reverses direction at the Bounce,
you’ve moved your extremely-finely-tuned-for-no-good-reason condition to the Bounce itself.
(Carroll 2007)
Thus, cyclic cosmologies are powerless to explain the low entropy fine-tuning of our universe.
Implications. Everything we see in observational physics, from galaxies around us today, to the
cosmic background radiation in the far past, tells us that the past hypothesis is correct. Theoretical
calculations, such as the low entropy needed for inflation to start, tell us likewise. There is the sec-
ond law of thermodynamics in operation around us. Yet this implies an unexplained highly fine-
tuned boundary condition imposed sometime to the past. Any initial conditions require explanation.
Very special initial conditions scream for explanation.
Physical attempts to explain it (Albrecht recurrences, Bousso landscape, Penrose CCC, and
other cyclic cosmologies) seem to have powerful theoretical objections to overcome. The landscape
represents the dominant paradigm and will likely continue to be for the foreseeable future. Problems
will be seen as exciting opportunities for new physics. That puts the fine-tuning argument evaluator
in a troublesome position.
Suppose that God really did prepare a low-entropy big bang. How could such a thing be dis-
covered? What would the proposals of the academy (where methodological naturalism is the domi-
nant metaphysic), look like were that true? There won’t be any physics papers telling us they’ve
found God. The problem will persist. The attempt to equate “special” with “generic” will go on.
Proposals will come and go. It seems reasonable to believe the situation would look much as the real
one does. Nevertheless, the significant problems intrinsic to the current formulations of inflationary
cosmologies—Albrecht recurrences, Bousso landscape, and conformal and other cyclic cosmolo-
gies—imply unlikeliness problems for ultimately explaining the highly fine-tuned low entropy of
our universe at the big bang.
CAN THE MULTIVERSE BE AN ULTIMATE EXPLANATION

OF FINE-TUNING?
Multiverse is an unverified, and possibly unverifiable, hypothetical construct of some physicists used
to explain the Schrödinger paradox in quantum mechanics, some problems in the standard big bang
model (such as the absence of magnetic monopoles and the initial smoothness of the universe), and
even fine-tuning for life in our universe. It proposes a metauniverse containing a multiplicity of dis-
tinct alternate or “bubble” universes, one of which is our own. According to this hypothesis, then,
our universe is but one bubble universe amid millions—or in some models an infinity—of other
bubble universes.
Most of the various multiverse models allow for a variation of initial conditions and the
values of constants at the inception of their constituent bubble universes. This potential for vari-
ance in those conditions and constants allows for an explanation of our exceedingly improbable
fine-tuned universe without appealing to a transcendent intelligence. As its proponents suggest,
if there is an infinity or even a virtual infinity of bubble universes, then even the most highly
improbable fine-tuning in any universe can be explained without appealing to a transcendent cre-
ative cause.
We examine the likeliness of a multiverse and its potential for explaining fine-tuning in two
sections: technical problems in the two major theories of the multiverse, and the more general
and vexing problem of Boltzmann brains and brief brains.
TECHNICAL PROBLEMS WITH THE TWO MAJOR MULTIVERSE MODELS

There are literally dozens of multiverse models, but the vast majority fit into the quantum mechan-
ical many worlds model or the eternal inflationary model, both of which frequently include the

string landscape, particularly the latter. We will examine the technical problems with each model as
well as their potential to fully explain fine-tuning for life in our universe.
The Quantum Many Worlds Model. The earliest proposal of a multiverse, the quantum
mechanical Everett many-worlds model, is quite different from the cosmological models discussed
below. In brief, the quantum wave function allows a variety of possible outcomes to arise from
the disposition of any quantum system. The potential for these outcomes in a system is called the
“superposition” of the quantum system. The Copenhagen interpretation asserts that the wave func-
tion really collapses to one real-world outcome. Nevertheless, a thought experiment proposed by
Erwin Schrödinger (“Schrödinger’s cat”) suggests the possibility of the existence of all possible out-
comes of the wave function. In order to avoid an ontological contradiction—such as the cat’s being
alive and dead at the same time—Hugh Everett (1957) proposed that each possibility exists in its
own distinct universe, and so there is a multiplicity of universes (reaching toward a virtual infinity)
accommodating every possible outcome of every wave function that has ever existed.
On the face of it, there are two problems with this proposal that make it difficult for many phy-
sicists to accept—it lacks an empirical ground (and is fundamentally untestable) and it fails a canon
of empirical science called the canon of parsimony, commonly called Ockham’s (or Occam’s) razor.
This principle states that the explanation that is the least complex and has the least number of
assumptions is probably correct. Paul Davies characterized the problem as follows:
To invoke an infinity of other universes just to explain one is surely carrying excess baggage to
cosmic extremes, not to mention the fact that all but a minute proportion of these other
universes go unobserved (except by God perhaps).… It is hard to see how such a purely
theoretical construct can ever be used as an explanation, in the scientific sense, of a feature of
nature. (Davies 1983, 173–174)
Beyond these two problems, Penrose points to an empirically testable potential mechanism to
explain the collapse of the wave function to a single real outcome that, if further validated, would
falsify the Everett many-worlds interpretation. He summarizes it as follows:
I have already made a case … that these [experimental circumstances that predict a deviation
from standard quantum theory] must be those in which gravity starts to be significantly
involved in quantum superpositions. The idea is that when this level is reached, some sort of
nonlinear instability sets in, limiting the duration of such super-positions, and one or
another of the alternatives taking part will have become resolved out, after a certain
estimated time period. Moreover, I am making the claim that all quantum state reductions
(R) arise in this “OR” (objective reduction) way. (Penrose 2016, 354; for background, see
207–210)
Further validation of Penrose’s “claim” would falsify the many-worlds interpretation.
The vast majority of physicists open to the multiverse hypothesis place their bets on cosmolog-
ical models (rather than the quantum mechanical one), preferring one of two models—eternal infla-
tion or a combination of eternal inflation with superstring theory.
The Eternal Inflation Model. As noted above, Andrei Linde and Alan Guth proposed the infla-
tionary model universe to resolve the smoothness problem, and the magnetic monopole problem
in the standard big bang model. They theorized that the problems could be solved by an inflation-
ary period of very short duration (around 10–100 of a second) after the big bang. Some newer mod-
els theorize an inflationary period prior to the big bang (Penrose 2016, 305).
The inflation model requires an extraordinarily large cosmological constant that is far too great
to be explained by the dark energy in the universe, and so a hypothetical “inflation field” was pro-
posed to explain it. Guth’s original formulation of the theory had no way to bring inflation to an
end that would lead to a hot, nearly uniform universe indicated by observation. So Linde, Albrecht,
and Paul Steinhardt proposed a “slow roll” to move from false vacuum to true vacuum without a
concomitant creation of bubbles. However, such a slow roll implies considerable fine-tuning in
the multiverse (Alabidi and Lyth 2006).
Though eternal inflation in the above model implies eternity into the future, Guth (2007) orig-
inally thought that it did not imply eternity into the past. Nevertheless, current proposals suggest
that inflation can be past eternal. As we shall see below, such proposals lead to significant unlikeli-
ness problems, such as Boltzmann brains and brief brains. Before considering these general vexing
problems we consider the unlikeliness of current models of inflationary cosmology to explain the
fine-tuning of our universe. Though this problem was noticed by Turok in 2002, it came to a head
in 2013 when Steinhardt (one of the architects of the model) and two colleagues analyzed data from
the WMAP and Planck satellites.
In 2013 and 2014 data from the Planck satellite convinced Anna Ijjas, Abraham Loeb, and
Steinhardt that the theory as posited eliminated the possibility of most inflationary models and
required the remaining models to have plateau-like inflation potentials that introduced exponentially
more parameters and more fine-tuning, problematic initial conditions, multiverse-unpredictability
issues, and a new unlikeliness problem (Ijjas and Steinhardt 2015, 1). The three authors reiterated
their claim in Scientific American (Ijjas, Steinhardt, and Loeb 2017a). Though Guth, Alexander
Vilenkin, Linde, Hawking, Susskind, and twenty-eight other physicists responded to the trio’s latest
article in a letter to Scientific American (Guth et al. 2017), the trio did not believe the response by
Guth and the others redressed their objections to the contemporary theory and its inability to make
predictions:
Our point is that we should be talking about the contemporary version of inflation, warts
and all, not some defunct relic. Logically, if the outcome of inflation is highly sensitive to
initial conditions that are not yet understood, as the respondents concede, the outcome
cannot be determined. And if inflation produces a multiverse in which, to quote a previous
statement from one of the responding authors (Guth), “anything that can happen will
happen”—it makes no sense whatsoever to talk about predictions. (Ijjas, Steinhardt, and
Loeb 2017b)
Susskind (2005) has offered a refinement of the eternal inflation model termed the string the-
ory landscape. Building on the possible variations within the compactification of extra dimensions
in the Calabi-Yau configuration, string theorists propose that there were about 10500 possible ways
in which eternal inflation could populate a bubble universe within a multiverse. The string land-
scape represents only a “list of all the possible designs of hypothetical universes [within an infla-
tionary multiverse]” (Susskind 2005, 381), while eternal inflation produces and populates bubble
universes in those ways. The number of bubble universes is not limited to 10500, because this
number represents only the possible ways bubbles can be populated. There can be trillions upon
trillions of bubbles for each way.
Banks has written a highly critical assessment of the reality of a string theory landscape multi-
verse, pointing to three ways in which the proposal conflicts with experimental evidence. He sums
up the matter by noting:
The String Landscape is a fantasy.… Theories of Eternal Inflation (EI) have to rely too heavily
on the anthropic principle to be consistent with experiment. Given the vast array of effective
low energy field theories that could be produced by the conventional picture of the string
landscape, one is forced to conclude that the most numerous anthropically allowed theories
will disagree with experiment violently. (Banks 2012)
Penrose agrees with Banks about the unfortunate heavy reliance of the string landscape on the
anthropic principle, and characterizes it as “a sorry place for such a grand theory to have finally
stranded us” (Penrose 2016, 120).
The above technical problems indicate that eternal inflation has significant physical challenges
to overcome in both the eternal inflation conception (of Guth et al.) and the string landscape con-
ception. Moreover, if Ijjas, Steinhardt, and Loeb (2013, 2014, 2015, 2017a, 2017b), Turok (2002),
Lewis and Barnes (2016), and Penrose (2016) are correct, eternal inflation significantly increases the
need for fine-tuning at the initiation and in the parameters of inflation, and presents unlikeliness
problems for the multiverse it implies.

THE UNDOING OF THE MULTIVERSE: BOLTZMANN BRAINS AND BRIEF BRAINS

If we imagine three explanations for fine-tuning—God, chance, and necessity—chance’s best chance
lies in the multiverse. In Bayesian terms, what is P(F/C)—the probability that the fine-tuning evi-
dence is compatible with the chance hypothesis? There is a strongly intuitive mechanism for gener-
ating random values of standard model force strengths and particle masses. This would be the infla-
tionary multiverse.
In this model universe (characterized by Tegmark [2007] as Level II—see below), there are dif-
ferent “false vacuum” states, with different cosmological constants, that thereby inflate at different
rates. New bubbles, usually of lower cosmological constants, nucleate by quantum fluctuation and
energy is released. Inside the bubble, the drama of the big bang is seen whereby the “universe” cools
and undergoes symmetry breaking events that determine the local physics environment by populat-
ing the standard model free parameters. As implied above, the landscape model is one that purports
to show how all the theoretical states are actually populated in an ontologically real multiverse.
Inflation is generically eternal to the future and theorists generally think of the future as already
existing. Thus one encounters infinities with a vengeance, and the infinities are actual, not potential.
In fact each inflationary bubble, though viewed by an outside observer as finite in size, is said to
contain an infinitely sized universe within. Generically the measure problem in cosmology is that
populations of observers and observations from different parts of the multiverse must be compared,
and the result is ambiguous when the populations are all infinite.
Now, a plausible condition for accepting a model is that it should predict that we are likely to
exist. In cosmology, this is called the typicality conjecture. If we are attempting to account for the exis-
tence of human beings, and our theories predict overwhelmingly that sentient observers should be
planet-sized pink elephants with blue ties, then we have a problem. Surprisingly, this very issue exists
within cosmology. For it turns out that the relevant comparison for the fine-tuning argument ceases to
be universes unfit for life versus universes that are fit. Rather, our best theories predict that inflationary
bubbles (universes) may be dominated by differing types of freak observers. It is the comparison of
populations of ordinary folk to freak observers that becomes the relevant metric for determining the
viability of inflationary theories, whether one is limited to the universe or the full multiverse.
We cosmically evolved sentients are called ordinary observers (OOs). So who are the freak obser-
vers? Is there an easier way to (naturalistically) obtain sentient observers? There is, if the atheist hypoth-
esis is correct regarding the relationship of mind to brain. We, the theists, will accommodate this
assumption for argument’s sake. There are two types of freak observers that are possible. The first
can be described by the modern legacy of Boltzmann. It is a variant of the now familiar Poincaré recur-
rence. In theory, if the total number of configurations is finite, given enough time, the system will
explore all of its states. Given infinite time, it will explore them all an infinite number of times.
Now, for the maximally probable and relevant recurrence, we don’t need to propose that a life-
supporting system (e.g., a universe or a solar system that could support OOs like us) must be fluc-
tuated from a thermal bath along with the observer. Producing the observer alone is a much more
probable proposition in entropy terms. The observer could appear, think a few thoughts, and then
die. So, if we have a theory of a universe (or multiverse) that is in a thermal bath, we have the prob-
lem of the prediction of freak Boltzmann brain observers (hereafter BBs). These observers, without
significant probability excursion, can be produced complete with a false set of memories of being
scientists and seemingly confirming themselves as OOs. We thus are vulnerable to a skeptical prob-
lem of serious magnitude. What we care about is whether OOs are typical compared to BB popula-
tions throughout the universe/multiverse. If so, we can continue our comfortable beliefs that we are
OOs and that science is possible in the first place. For if BBs are real, and we are BBs, then we may
never know it. Carroll (2017) argues that theories that predict BB domination are “cognitively
unstable” and should be rejected on philosophical (not physical!) grounds. Most eternal inflationary
theories (naively) predict Boltzmann brain domination. And, in general, there are no halfway cases.
Either the probability is nearly 1.0 that you are a Boltzmann brain, or the probability is nearly zero.
Why do eternal inflation theories (naively) predict Boltzmann brain domination? Imagine a sin-
gle universe of finite size at any one time but an infinite lifetime. In this simple theory there is a
cosmological constant that acts for all time. In such a universe, a finite number of OOs will be
created after a life-supporting system (e.g., a universe or solar system) evolves (within a visible hori-
zon). They will cease to exist when the life-supporting system disappears (due to heat death). Dur-
ing the entire life of the universe, however, the universe sits in a Boltzmann-like heat bath because
the existence of a cosmological constant brings a nonzero temperature to the vacuum along with it.
So, for the entire (possibly) infinite lifetime of the universe, the universe can percolate Boltzmann
brains. Thus the probability that you are an OO in such a universe is a finite number divided by
a number that asymptotically approaches infinity. That probability is a hard zero. You are a Boltz-
mann brain. In such a reality, science is flat out impossible.
How does this apply to the multiverse? Well, what is the multiverse but an infinite collection of
different heat baths of differing temperature that precipitate infinite numbers of new heat baths that
themselves produce infinite numbers of heat baths? It is “turtles all the way down,” as the saying puts
it. There is also the issue that thermal fluctuations in the heat bath can themselves be infant inflationary
universes. This makes the problem of comparing OO and BB populations much more complicated.
We are stuck smack in the measure problem. Note, for clarification, that only a small fraction of these
bubbles allow for any kind of life. BBs infest environments only where some type of OOs can thrive.
What is the measure problem? A major obstacle in assessing the viability of an eternally inflat-
ing multiverse is this: “In an eternally inflating universe the total volume occupied by all, even abso-
lutely rare types of ‘universes,’ is indefinitely large” (Linde and Noorbala 2010, 1). For any scientist,
the problem is that “we are faced with comparing infinities” (Barnes 2012, 60). This means that we
cannot apply the standard measure of testability to any eternally inflating multiverse model, which
makes it fundamentally untestable, and therefore unverifiable. Why so?
Testability and falsifiability of any scientific theorem depends on predictability, and predictabil-
ity in turn depends on the law of statistical sampling: given any population of comparable items, a
random selection of any item will correspond very closely to the presence of those kinds of items
within the total population. But this important statistical law of prediction cannot be applied to
an infinite multiverse. When we are comparing infinities, every item, no matter how rare, occupies
the same nonquantifiable part of the population—an infinite one.
We know that our universe must be exceedingly rare in a population of universes within a mul-
56
tiverse, because ours is so large and so cool that there should be 1010 other universes for every uni-
verse like ours. However, this prediction is untestable because, according to the eternal inflation
model, there are an infinite number of universes like ours as well as an infinite number of universes
not like ours within the multiverse. Though we know that our universe should be rare, the presence
of an infinity of them makes the hypothesis fundamentally untestable. As Steinhardt (2011) vehe-
mently insists, this fails a key requirement of scientific theories.
Some physicists have tried to extricate eternal inflation from its nonscientific status by propos-
ing the possibility of having intelligible ratios among infinities, but there is a problem with this con-
tention given the assumption of an infinite number of bubble universes: all ratios must be infinity
over infinity, which undermines their intelligibility—their potential to describe quantitative differ-
entials. When there is an infinite number of everything, ratios become unintelligible.
The measure problem seems intractable. Consider, for example, cosmologist Don Page’s (2014)
list of 245 references of attempted solutions. Many of these overlap, but there are scores of mutually
exclusive proposals to solve the measure problem in physics today. Page has been one of the most
involved and erudite researchers in discovering, articulating, and attempting to solve the problem
from the start of the twenty-first century all the way to the present day.
The problem is not that there exist no (apparent) solutions. The problem is that there are too
many, and virtually no agreement outside of one’s immediate circle of researchers. In his paper Page
himself suggests three new and differing solutions: MAD = maximal average density, BAD = biased
average density, and UGLY = utility giving light youngness. These give rise to the following joke:
“If GOOD were interpreted as simply meaning Great Ordinary Observer Dominance, then since
all three of my proposed measures suppress Boltzmann brains relative to ordinary observers, one
could say that the MAD, the BAD, and the UGLY are all GOOD” (Page 2014, 14). Funny—
and also suggestive of the helplessness of the community against the mathematical strangeness native
to actual infinities conceived as concrete reality. Could one of his solutions be correct? The

community treats them the same as the other scores of proposals—with polite respect, as all retain
their own mutually exclusive approaches. Another joke comes to mind, that a government-con-
ceived solution to the problem of sixteen redundant and ineffective bureaucracies is to create a sev-
enteenth.
It may be useful to look in another direction, as Ken Olum (2012) has. Olum tries to define a
list of axioms any measure must satisfy. He conducts some thought experiments first in a toy model,
and then extends them to the relevant differing conditions in an eternal inflation model. The experi-
ments he conceives involve a common transaction involving the comparison of individual members
of infinite sets. The result seems to be that no measure is consistent with obeying one or more of
the axioms. Thus, perhaps, no measure is possible.
Olum goes on to show that considering more realistic assumptions about populations in eter-
nally inflating universes doesn’t change the conclusion. He sums things up as follows:
What can we do to make sense of the situation? One might try arguing that anthropic ideas are
wrong, and we should not consider ourselves typical among all observers in a multiverse. But
then it does not seem possible to recover ordinary ideas of probability. In an infinite universe,
everything which can happen will happen an infinite number of times, so what does it mean to
say that one thing is more likely than another? (Olum 2012, 9–10)
Olum mentions the possibility of horizon complementarity (holography), a proposal that suggests
that one shouldn’t take the whole multiverse to simultaneously exist. That does evade the problem.
But complementarity also has a serious Boltzmann brain problem locally because its Hilbert space
(the space of all configurations) is finite; thus there are recurrences. Olum concludes: “Finally, per-
haps we just don’t know how to proceed when the universe is infinite (and not homogeneous). I
don’t think one should conclude from this that the universe must be finite, but rather that new
ideas are needed to make sense of probabilities in an infinite universe” (2012, 10).
We suggest a stronger conclusion may be in order. To answer Olum’s question, perhaps there is
no fact-of-the-matter as to relative population size (or ordered observation count) given a comparison
of infinite populations or observations. Cosmologists can’t find it because, to quote a famous line
from another context, “there is no there there.”
What would the consequences be if—literally—no answer exists? It would mean that OOs
can’t even in principle be shown to be typical in such models. That would produce the skeptical con-
clusion that we can’t know whether we are really Boltzmann brains and thus science is impossible if
we live in such a multiverse.
Better is the approach suggested in two papers by Kimberly Boddy, Carroll, and Jason Pollack
(2015, 2016). The authors’ approach is to avoid the problem by suggesting a physical means
whereby Boltzmann brains are suppressed. Boltzmann brains may be suppressed if the Hilbert space
within which a universal wave function resides is of infinite extent. Poincaré recurrences occur for
finite Hilbert spaces. The authors suggest that the asymptotic end condition of a universe undergo-
ing expansion via cosmological constant (a de Sitter space) is a literal stationary state, that is, noth-
ing dynamical can happen. Such a thing may be a consequence of a particular choice of quantum
gravity interpretation (decoherent histories) and a choice of a consistent set of histories that leads
to this type of stationary state.
Another way this can happen is in landscape theories where the dark energy is metastable and
decays to lower and lower levels until a Minkowski vacuum (or, similarly, a Bunch-Davies vacuum)
obtains. There is no dark energy in such a space, hence a zero temperature, hence no thermal bath.
No thermal bath, no Boltzmann brains. The trajectory of a hypothetical observer would reach the
Bunch-Davies vacuum in a finite time, visiting metastable vacua on the way that decay at a rate fas-
ter than Boltzmann brains percolation.
This theory is opposed by those that feature horizon complementarity (finite Hilbert space;
Banks 2001), by the one put forward by Bousso and Freivogel (2007), and by those that don’t
use the specific quantum interpretation or interpret the decoherent (consistent) histories differently
(Lloyd 2016). Nonetheless, there does seem to be a telling objection against Boddy, Carroll, and
Pollack on the grounds that there is another type of freak observer that is not suppressed on their
proposed model. Matthew Davenport and Olum (2010) explicate this point. These are sometimes
called brief brains, or bbs in Page’s writings. They are fluctuations directly of the quantum vacuum,
and are independent of the phenomena of Boltzmann brains, which are thermal fluctuations. Dav-
enport and Olum point out that these freak observers are much more numerous than Boltzmann
brains, and will occur in every type of vacuum including Minkowski vacua (note, however, that
the vacuum must support minimum conditions for sentient life). They don’t depend on recurrence
phenomena, and thus the size of a quantum wavefunction’s Hilbert space is irrelevant.
Boddy, Carroll, and Pollack are aware of objections such as that of Davenport and Olum, and
cite them in their relevant literature. They don’t refute Davenport and Olum’s conclusion; they
indicate the overall argument is “not without controversy.” Carroll sums the matter up in this man-
ner:
Fortunately, the criterion that random fluctuations dominate the fraction of observers in a
given cosmological model might not be as difficult to evade as might be naively expected, if
Hilbert space is infinite-dimensional and local de Sitter phases settle into a truly stationary
vacuum state, free of dynamical Boltzmann fluctuations. That conclusion depends sensitively
on how one interprets what happens inside the quantum state, an issue that is unfortunately
murky in the current state of the art. If any were needed, this gives further impetus to the
difficult task of reconciling the foundations of quantum mechanics and cosmology. (Carroll
2017, 23)
We believe that the best interpretation of Boddy, Carroll, and Pollack’s work would be that it
is worth consideration and relevant to the Boltzmann brain problem but not the brief brain prob-
lem. The authors may be suggesting a possible philosophical approach to the brief brain problem.
Perhaps the thermodynamic arrow of time just is time itself and therefore nothing can happen
once equilibrium is reached. Time just stops. What could it mean to have a “brief” quantum fluc-
tuation in such a situation? Yet Page, a pure atemporalist in his ontology, does not believe even
that would be a barrier to the presence of brief brains (personal communication with the authors,
13 July 2018). Even within the topic of thermal fluctuation Boltzmann brains, Boddy, Carroll,
and Pollack’s proposal is controversial, but, unfortunately, there isn’t enough discussion of
its promise. This may be due to the fact that most theorists are still tied to the idea of a GOOD
cosmological measure.
Conclusion. The problem of freak observers in landscape eternal inflation multiverse models
points out the larger problem of empirical (in)coherence in today’s cosmological models. The
problem of the multiverse with respect to the fine-tuning argument is not the comparison of life-
less universes to life-amenable ones. Rather, it is discovering whether OOs are typical in the mul-
tiverse compared to freak observers. This is based on the not unreasonable requirement that a
model predict that we, the ordinary evolved observer, exist. As per the best knowledge we have
as of this writing, the eternal inflation multiverse does not do this, and therefore it is likely empiri-
cally incoherent.
It should be noted here that a single universe inflationary theory could still avoid the domi-
nance of freak observers if the universe ends prior to Boltzmann brains/brief brains domination.
This could happen, for example, if the dark energy is of the form of “quintessence” rather than a
cosmological constant. It could decay, and if the structure of the universe is closed, the universe will
end in a finite time in a final singularity (on a general theory of relativity rendering). This will hap-
pen long before Boltzmann brains/brief brains domination could get off the ground, given the
extreme time frames necessary for freak observer domination.
Alternatively, it is thought that the Higgs field is metastable and will decay on a time scale long
compared to current universe age but far shorter than the time for appreciable Boltzmann brains
creation. If so, hydrogen will be the only element in the new periodic table, and hence Boltzmann
brains/brief brains can’t form regardless of the age or size of the universe. It may also be possible to
argue that a theory of time in which the future does not yet exist allows OOs to dominate as obser-
vers (and thus allow rational warrant for belief) in a single universe inflationary model for the entire
lifetime of the OOs. This reasoning is disputed, however, by those who defend the doomsday

hypothesis (Leslie 1996, 249), which depends more on the level of determinism in the model than
the philosophy of time.
One could also reject the inflationary paradigm for a competitor such as the ekpyrotic model
proposed by Steinhardt and Turok (2002). Or one could follow the proposal of Carroll to accept
models such as Boddy, Carroll, and Pollack’s on the grounds that competitors are cognitively unsta-
ble. On these grounds, a possibly low prior probability of Boddy, Carroll, and Pollack’s model
(prior to resolution of its issues) is overcome by ruling out competitors on philosophical grounds.
This seems unsatisfying, though, given more viable one-world alternatives.
Thus, for now, the probability of the fine-tuning evidence in light of the chance (multiverse)
hypothesis P(F/C), should be judged as small, maybe negligible, with a bit of the smokescreen of
obscurity. Continued advocacy of a multiverse under an atheist definition of mind (where the brain
just is the mind), will lead to increasing skepticism—from Boltzmann brains to brief brains, and
then, it seems necessary, to “Boltzmann robots” (machines that can pass a Turing test; see Banks
2012). Ultimately, cosmologists will saw off the branch on which they are sitting and produce the
purest form of skepticism, and, perhaps, rule out cosmology itself. In view of this, it seems unlikely
that the multiverse will provide an ultimate explanation for fine-tuning—even into the future.
TEGMARK’S LEVEL IV MULTIVERSE
Tegmark proposes a four-level multiverse scheme to assess multiverse theories, with higher levels offer-
ing the possibility of greater diversity to explain fine-tuning. His proposal is instructive, because it
brings “multiverse reasoning” to its ultimate conclusion, which ironically reveals that a multiverse is
unlikely to be the ultimate explanation for fine-tuning. He states his position as follows:
Level I: A generic prediction of inflation is an infinite ergodic universe, which contains Hubble
29
volumes realizing all initial conditions—including an identical copy of you about 1010 m
away.
Level II: In chaotic inflation, other thermalized regions may have different physical constants,
dimensionality and particle content.
Level III: In unitary quantum mechanics, other branches of the wave function add nothing
qualitatively new, which is ironic given that this level has historically been the most
controversial.
Level IV: Other mathematical structures give different fundamental equations of physics.
(Tegmark 2007, 99)
We have already discussed Levels I through III with respect to technical problems, the measure
problem, and the Boltzmann brain/brief brain problems. It now remains to discuss Tegmark’s
fourth level, which seems to offer a higher-order diversity of universes and multiverses to explain
fine-tuning—a diversity of universes and multiverses not just with different constants and initial
conditions but different laws of physics.
Recall Collins’s contention above that, even if there were a theory of everything or a real multi-
verse with diverse constants and initial conditions (or both), these would not be sufficient to ulti-
mately explain universal fine-tuning for life, because there is the further question of why our hypo-
thetical multiverse has physical laws that will permit life instead of prohibiting it. He indicates that
the laws of a life-permitting multiverse would have to give rise to bubble universes that could be
filled with mass-energy of some kind and that would also require some kind of universal force of
attraction (acting like Einstein’s space-time gravity rather than Newton’s gravity), something akin
to the quantization of energy, the Pauli exclusion principle, and other laws allowing matter to
develop into diverse complexes needed for life-forms (Collins 2009, 264). Since there are far more
logically possible sets of laws that will not produce multiverses capable of generating life-permitting
bubble universes than sets of laws that will permit life, physical laws too would need to be fine-
tuned to give rise to multiverses capable of generating life-permitting bubble universes (see the ratio-
nale for this below).
In 1996 Tegmark anticipated the need to naturally explain this higher level of fine-tuning for
life. His approach was to postulate a multiverse of all possible physical laws (a Level IV multiverse),
which would effectively obviate the need for fine-tuning of physical laws, because all life-permitting
physical laws (that would allow multiverses generating life-permitting bubble universes) would be
represented alongside the much larger group of non–life-permitting physical laws. In this way, mul-
tiverses allowing life-permitting universes would no longer be anthropically special, but would be a
natural part of a complete array of physical laws giving rise to multiverses.
In order to generate a complete array of all possible physical laws (to govern all possible multi-
verses arising out of these physical laws), Tegmark had to show that the complete array of coherent
mathematical systems capable of underlying physical laws must have real ontological status—not
simply conceptual status in the mind of a mathematician. This requires moving mathematics from
an abstract intramental status to a physical (seemingly concrete) extramental status.
Taking inspiration from Wigner (1960), Plato, and the ancient Pythagoreans, Tegmark in an
interview boldly proclaims:
The reason why mathematics is so effective at describing reality is that it is reality. That is the
mathematical universe hypothesis: Mathematical things actually exist, and they are actually
physical reality.… Stephen Hawking once asked it this way: “What is it that breathes fire into
the equations and makes a universe for them to describe?” If I am right and the cosmos is just
mathematics, then no fire-breathing is required. A mathematical structure doesn’t describe a
universe, it is a universe. (Frank 2008)
Tegmark explains this concept in considerable detail by means of varying mathematical structures in
his 2007 and 2008 papers, but this lies beyond the scope of this chapter.
Though Tegmark may have been inspired by Wigner, Hawking, Plato, and other mathematical
and physical theoreticians, he deviates significantly from them in one major respect. Those theore-
ticians recognized that physical and mathematical equations are abstract (universals)—that is, they
prescind from individuation, and space-time particularity, and, as such, can use unifying concepts
and groups that have no direct physical and spatiotemporal correlates—for example, an equal sign,
or any other symbol representing an abstract operator or a group. Abstractions, such as groups or
mathematical and logical operators, do not exist in any physical world—because they are not and
cannot be individuated. Rather, they exist in ideas originating in conscious intelligence that can
abstract from individuation. Conscious intelligence abstracts intelligible content from the individu-
ation of the physical world to generate what are called “universal concepts” such as, groups, relation-
ships, and “relationships among relationships.” (There is a vast literature on this process summarized
within the context of mathematics and physics in Lonergan 1992, chaps. 5, 10, 12, 13, and 19.)
Philosophers as diverse as Aristotle, Immanuel Kant, and Bernard Lonergan have long recog-
nized that if an idea were to be individuated, it could no longer be abstract and represent groups
and relationships. Individuating an idea eliminates its power to “universalize” (to represent nonindi-
vidual abstractions, such as groups), while universalizing an individuated (material) reality eliminates
its “materiality”—its individual, space-time characteristics, which enable it to interact with other
material (physical) processes and realities. Even quantum systems cannot become universal ideas
without losing their power to interact with other physical systems.
At this point the problem with Tegmark’s proposal comes into focus. Mathematics is a science
built on a system of quantifiable universal ideas (see Lonergan 1992, 334–339). True enough—
mathematics is abstracted from the individuated material world and it can be applied to that world,
but when it is abstracted by a human being, it leaves behind its individual material ground and takes
on a mentative or conceptual ground in conscious intelligence. If it did not leave its individuation
space-time particularity and materiality behind, it would be powerless to represent groups, relation-
ships, and mathematical operators—making it decidedly nonmathematical.
In view of the above, Tegmark’s contention that mathematics is physical reality demonstrates a
nonrecognition of the fundamental distinction between universal ideas and individuated (physical)
reality. In order for Tegmark to make his Level IV multiverse work, he must either infuse mathe-
matical ideas with the individuating, spatiotemporal, particularizing characteristics of material

realities (which forces them to be nonuniversal, and therefore nonmathematical) or he has to elevate
material realities to the level of mathematical ideas, which makes them nonphysical (nonindividu-
ated, not subject to space and time, and nonmaterial). Tegmark can’t eat his cake and have it too.
Trying to combine two contradictories is a category error that invalidates any theory on which such
a category error is based.
In order for Tegmark to rectify his category error and problematic ontology, he will have to let
ideas be ideas (in mathematics) without forcing them to be individuated physical realities and let
individuated physical realities be individuated physical realities without elevating them to the
domain of universal ideas in conscious intelligence. Respecting this distinction entails putting Teg-
mark’s Level IV multiverse in its proper ontological domain—the domain of mathematics (groups,
sets, relationships, and mathematical operations), which is the domain of universal ideas in con-
scious intelligence—but not the domain of individuated physical reality.
One final point: if we concede to the reality of a Level IV multiverse existing through conscious
intelligence, we will have to acknowledge that this conscious intelligence (that conceives it) would
have to transcend not only our material universe, but also the first three levels of material multi-
verses postulated by Tegmark. It would also have to transcend the Level IV multiverse itself, since
it would be the conscious intellectual activity through which the idea of the Level IV multiverse
comes into being. Thus a Level IV multiverse is unlikely to be the ultimate explanation for fine-
tuning, because the existence of such a multiverse (which is fundamentally mathematical) requires
a trans-physical, trans-universal, and trans-multiversal conscious intelligence through which the
interrelationship of all its formal (universal) mathematical systems can exist.
IS TRANSCENDENT INTELLIGENCE A MORE LIKELY ULTIMATE EXPLANATION

OF FINE-TUNING?
As noted above, we have been developing an abductive argument for a transcendent, trans-universal,
and trans-multiversal conscious intelligence as the most likely ultimate explanation for fine-tuning.
An abductive argument begins with a set of observations and seeks, through logical inference, the
simplest and most likely explanation for those observations. Though these arguments have consider-
able probative force, they are less probative than deductive arguments, such as metaphysical argu-
ments for the existence of God (see Spitzer 2010, chaps. 3, 4, and 5). Nevertheless, they have an
objective basis requiring an explanation. This objective ground guides our inquiry through several
possible candidates to arrive at the most adequate and likely ultimate explanation.
We began our investigation by presenting six instances of fine-tuning grounded in objective
reality. Each instance showed the highly improbable occurrence of the value of a free parameter nec-
essary for life amid an exponentially larger available parameter space prohibiting the possibility of
life. Inasmuch as the life-permitting value of a free parameter is not required by any physical theory
(or the physical laws underlying that theory), and a one-off pure chance occurrence is virtually
impossible, we are faced with the legitimate question of how this incredibly improbable anthropic
free parameter value occurred.
We considered several possible natural explanations, including a theory of everything, inflation-
ary cosmologies, conformal and other cyclic cosmologies, and multiverses (on Tegmark’s first three
levels). We found that all three hypotheses have significant unlikeliness problems and are unable to
provide an ultimate explanation for fine-tuning. This led us to the only natural explanation capable
of ultimately explaining the fine-tuning of physical laws—namely, Tegmark’s Level IV multiverse.
We discovered that this hypothesis falls prey to a category error that invalidates it. The only way
of extricating it from this category error is to ground the Level IV multiverse in a trans-universal,
trans-multiversal, transcendent conscious intelligence required for the existence of its universal ideas.
Is a trans-universal, trans-multiversal, transcendent conscious intelligence the most likely ulti-
mate explanation for fine-tuning (beyond simply rescuing Tegmark’s Level IV multiverse)? Let’s
review some salient points in the above investigation that reveal why transcendent intelligence is
probably the only explanation that can assume this ultimate status.
We already discussed Collins’s argument for the necessity of fine-tuning of physical laws. As
noted above, if we suppose there is a multiverse, then the development of life within it will require
physical laws intrinsic to it that permit the development of life. These physical laws would have to
allow the emergence of bubble universes that could be filled with mass-energy of some kind, and
provide for a universal force of attraction, something akin to the quantization of energy, the Pauli exclu-
sion principle, and other conditions allowing matter to develop into diverse complexes needed for life-
forms. These characteristics of bubble universes are necessary to give rise to unifiable, aggregatable, com-
plexifiable, individuated physical constituents of potential life-forms. These ordered properties of physical
entities (and the physical laws that allow for their emergence) are far more improbable than disordered
(non-unifiable, non-aggregatable, non-complexifiable) individuated properties, because statistically, order
(unifiability, aggregatability, and complexifiability) is far more improbable than disorder.
The above statistical principle applies not only to universes but also to bubble universes within
multiverses, multiverses themselves, and even multiverses of multiverses. It applies to every complex
of material realities and, as we saw above, to the physical-mathematical laws governing these mate-
rial complexes. Hence we may infer that physical laws intrinsic to multiverses and universes giving
rise to unifiable, aggregatable, complexifiable individuated properties of physical entities are far more
improbable than physical laws that do not do this. Inasmuch as a one-off pure chance occurrence of
these anthropic physical laws (amid the exponentially greater possibilities of nonanthropic physical
laws) is so exceedingly improbable, we are compelled to seek a higher-level explanation.
What are the possible candidates for a reality that is ontologically higher than physical laws? We
find a clue in the highly precise mathematical determinacy intrinsic to all such sets of physical laws
that can give rise to unifiable, aggregatable, and complexifiable individuated properties. This highly
precise mathematical determinacy is expressible through mathematical formalisms such as linear and
nonlinear equations. We may now infer a partial answer to our inquiry. Any candidate for a higher-
level (and ultimate) explanation of fine-tuning must be able to explain the occurrence of highly
improbable mathematical determinacy (expressible through formalisms) in the laws governing mate-
rial/physical reality. If a potential explanation cannot do this, it is not “higher” than the exceedingly
improbable physical laws that do not explain themselves.
This is precisely what led Tegmark (2007, 2008) to his Level IV multiverse proposal and Col-
lins (2013, 2009) to his transcendent creative intelligence proposal. We might generalize Tegmark’s
effort as a nonintelligence proposal and Collins’s effort as an intelligence proposal. A nonintelligence
proposal seeks to explain the occurrence of highly improbable mathematical formalisms within
physical laws governing material realities without making recourse to a state of consciousness that
would have to be ontologically higher than the mathematical formalisms intrinsic to physical laws.
Conversely, an intelligence proposal does make recourse to such an ontologically higher conscious-
ness (intelligence).
By now the problem with all nonintelligence proposals will be manifest. Giving a higher-level
explanation of mathematical determinacy (expressible by formalisms) within the laws of a physical
system (e.g., a universe or multiverse) without making recourse to conscious intelligence requires
arguing (as Tegmark does) that the universal-ideational status of mathematical formalisms be indi-
viduated physical realities. There is no escape—they will have to argue a category error of synthesiz-
ing universal-individuals. Tegmark’s category error cannot be avoided in any nonintelligence pro-
posal. It does not matter whether a proponent argues a vastly different approach than Tegmark in
such a proposal. Anyone who forecloses the prospect of an ontologically higher-level consciousness
(which can be a “higher” explanation for mathematical formalisms within physical realities) will suc-
cumb to this category error.
In order to avoid the intrinsically contradictory category error in all nonintelligence proposals,
we will have to seriously consider making recourse to an ontologically higher-conscious intelligence
that can conceive of mathematical formalisms and infuse them into physical laws governing individ-
uated material complexes (such as universes or multiverses). This conscious intelligence must be
ontologically higher than the mathematical formalisms it conceives and would have to be creative
in order to infuse the mathematical determinacy (expressed by those formalisms) within physical
laws and complexes of individuated physical realities. It is difficult to see how this implication can

be circumvented in discovering the highest (ultimate) explanation of fine-tuning. In view of this, it

is reasonable and responsible to conclude that a transcendent, trans-universal, trans-multiversal cre-
ative conscious intelligence is the most likely ultimate explanation of fine-tuning.
Is there anything more that can be said about this most likely ultimate explanation of fine-tun-
ing for life in our universe? From the vantage point of this abductive argument (based on the above
six instances of fine-tuning for life in our universe), it is unlikely. However, from the vantage point
of a comprehensive logically based metaphysics, we would be able to connect our conclusion (the
likely existence of a transcendent, trans-universal, trans-multiversal creative conscious intelligence)
with a broader one arising out of the necessity for a completely intelligible, unconditioned
(uncaused) reality in the whole of reality. Lonergan gives a proof of this reality and shows it to be
an unrestricted act of understanding that creates the entire restricted world of intelligibility through
its mentative act (see Lonergan 1992, chap. 19). Evidently, linking the above conclusion to Lone-
rgan’s more extensive one is beyond the scope of this chapter. Nevertheless, we decided to conclude
with a part of Lonergan’s proof that might provide readers with the incentive to connect the above
conclusion with his:
Intelligibility either is material or spiritual [immaterial] or abstract: it is material in the
objects of physics, chemistry, biology, and sensitive psychology; it is spiritual [immaterial]
when it is identical with understanding; and it is abstract in concepts of unities, laws.…
But abstract intelligibility necessarily is incomplete, for it arises only in the self-expression
of spiritual intelligibility. Again, spiritual intelligibility is incomplete as long as it can
inquire. Finally, material intelligibility necessarily is incomplete, for it is contingent in its
existence and in its occurrences, in its genera and species.… Moreover, it includes a merely
empirical residue of individuality, noncountable infinities, particular places and times.…
It follows that the only possibility of complete intelligibility lies in a spiritual intelligibility that
cannot inquire because it understands everything about everything. (Lonergan 1992, 696;
emphasis added)
Bibliography
Adams, Fred. “Stars in Other Universes: Stellar Structure with Different Barnes, Luke. “The Fine-Tuning of the Universe for Intelligent Life.” History
Fundamental Constants.” Journal of Cosmology and Astroparticle Physics and Philosophy of Physics (June 2012): 1–77. Cornell University Library
(August 2008). http://iopscience.iop.org/article/10.1088/1475-7516 archive, https://arxiv.org/pdf/1112.4647.pdf.
/2008/08/010/meta.
Barnes, Luke. “Testing the Multiverse: Bayes, Fine-Tuning, and Typicality.”
Alabidi, Laila, and D. Lyth. “Inflation Models and Observation.” Journal of In The Philosophy of Cosmology, edited by Khalil Chamcham, Joseph Silk,
Cosmology and Astroparticle Physics (May 2006). http://iopscience.iop John D. Barrow, and Simon Saunders, 447–466. Cambridge, UK:
.org/article/10.1088/1475-7516/2006/05/016/pdf. Cambridge University Press, 2017.
Albrecht, Andreas. “Cosmic Inflation and the Arrow of Time.” 2002. Barr, S. M., and Almas Khan. “Anthropic Tuning of the Weak Scale and of
Cornell University Library archive, https://arxiv.org/abs/astro-ph mu/md in Two-Higgs-Doublet Models.” Physical Review D, 76, no. 4
/0210527. (2007). Cornell University Library archive, https://arxiv.org/pdf/hep-ph
Albrecht, Andreas. “De Sitter Equilibrium as a Fundamental Framework for /0703219.pdf.
Cosmology.” Journal of Physics Conference Series 174 012006 (2009). Boddy, Kimberly K., Sean M. Carroll, and Jason Pollack. “Why Boltzmann
Cornell University Library archive, https://arxiv.org/abs/0906.1047v2. Brains Don’t Fluctuate into Existence from the De Sitter Vacuum.”
Albrecht, Andreas. “Tuning, Ergodicity, Equilibrium and Cosmology.” 2015. Cornell University Library archive, https://arxiv.org/abs/1505
Physical Review D 91 (2014), 103510. Cornell University Library .02780.
archive, https://arxiv.org/abs/1401.7309v1. Boddy, Kimberly K., Sean M. Carroll, and Jason Pollack. “De Sitter Space
Ambjørn, Jan, Jerzy Jurkiewicz, and Renate Loll. “Causal Dynamical without Dynamical Quantum Fluctuations.” Foundations of Physics 46,
Triangulations and the Quest for Quantum Gravity.” 2010. Cornell no. 6 (2016): 702–735. https://doi.org/10.1007/s10701-016-9996-8.
University Library archive, https://arxiv.org/abs/1004.0352. Cornell University Library archive, https://arxiv.org/abs/1405.0298.
Banks, Tom. “Cosmological Breaking of Supersymmetry?” International Borde, Arvind. “Topology Change in Classical General Relativity.” 1994.
Journal of Modern Physics A 16, no. 5 (2001): 910–921. Cornell Cornell University Library archive, https://arxiv.org/abs/gr-qc/9406053.
University Library archive. http://xxx.lanl.gov/abs/hep-th/0007146.
Bousso, Raphael. “Vacuum Structure and the Arrow of Time.” Physical
Banks, Tom. “The Top 10500 Reasons Not to Believe in the Landscape.” Review D, 86 (2012), 123509. Cornell University Library archive,
2012. Cornell University Library archive, https://arxiv.org/abs/1208.5715. https://arxiv.org/abs/1112.3341.
Bousso, Raphael, and Ben Freivogel. “A Paradox in the Global Description Dyson, Lisa, Matthew Kleban, and Leonard Susskind. “Disturbing
of the Multiverse.” Journal of High Energy Physics 6 (2007). Cornell Implications of a Cosmological Constant.” Journal of High Energy
University Library archive, https://arxiv.org/abs/hep-th/0610132v2. Physics 10 (2002). http://iopscience.iop.org/article/10.1088/1126-6708
/2002/10/011/pdf.
Carr, Bernard J., ed. Universe or Multiverse? Cambridge, UK: Cambridge
University Press, 2007. Einstein, Albert. “Physics and Reality.” In Ideas and Opinions. Translated by
Carr, Bernard J., and M. J. Rees. “The Anthropic Principle and the Structure Sonja Bargmann. New York: Crown, 1954.
of the Physical World.” Nature 278 (1979): 605–612. http://adsabs Einstein, Albert. Relativity: The Special and the General Theory. Translated by
.harvard.edu/abs/1979Natur.278..605C. Robert W. Lawson. London: Routledge, 2014.
Carroll, Sean. “Against Bounces.” Discover, 2 July 2007. Ellis, George. “Does the Multiverse Really Exist?” Scientific American 305,
Carroll, Sean. “Richard Feynman on Boltzmann Brains.” Preposterous no. 2 (2011): 38–43.
Universe Blog, 29 December 2008. http://www.preposterousuniverse Everett, Hugh. “‘Relative State’ Formulation of Quantum Mechanics.”
.com/blog/2008/12/29/richard-feynman-on-boltzmann-brains/. Review of Modern Physics 29 (1957): 454–462.
Carroll, Sean. “Penrose Cyclic Cosmology.” Preposterous Universe Blog, 7 Farhi, Edward, A. H. Guth, and Jemal Guven. “Is It Possible to Create a
December 2010. http://www.preposterousuniverse.com/blog/2010/12 Universe in the Laboratory by Quantum Tunneling?” Nuclear Physics B
/07/penroses-cyclic-cosmology/. 339, no. 2 (1990): 417–490.
Carroll, Sean M. “In What Sense Is the Early Universe Fine-Tuned?” 2014. Frank, Adam. “Is the Universe Actually Made of Math?” Discover, July 2008.
Cornell University archive, https://arxiv.org/abs/1406.3057v1. http://discovermagazine.com/2008/jul/16-is-the-universe-actually-made
Carroll, Sean M. “Why Boltzmann Brains Are Bad.” 2017. Cornell -of-math.
University Library archive, https://arxiv.org/abs/1702.00850. Freivogel, Ben, Veronika E. Hubeny, Alexander Maloney, et al. “Inflation in
Carter, Brandon. “Large Number Coincidences and the Anthropic Principle AdS/CFT.” 2002. Cornell University Library archive, https://arxiv.org
in Cosmology.” In Confrontation of Cosmological Theory with Observa- /abs/hep-th/0510046.
tional Data, edited by M. S. Longair, 291–298. Dordrecht, Netherlands: Friederich, Simon. “Fine-Tuning.” In Stanford Encyclopedia of Philosophy,
Reidel, 1974. edited by Edward N. Zalta. Spring 2018. https://plato.stanford.edu
Chamcham, Khalil, Joseph Silk, John D. Barrow, and Simon Saunders, eds. /entries/fine-tuning/.
The Philosophy of Cosmology. Cambridge, UK: Cambridge University Gingerich, Owen. “Do the Heavens Declare?” In The Book of the Cosmos, edited
Press, 2017.
by Dennis Richard Danielson, 522–528. Cambridge, MA: Perseus, 2000.
Coleman, Sidney, and Frank De Luccia. “Gravitational Effects on and of
Gurzadyan, V. G., and Roger Penrose. “Concentric Circles in WMAP Data
Vacuum Decay.” Physical Review D 21 (1980) 3305.
May Provide Evidence of Violent Pre-Big-Bang Activity.” 2010. Cornell
Collins, Robin. “The Teleological Argument: An Exploration of the Fine- University Library archive, https://arxiv.org/abs/1011.3706.
Tuning of the Cosmos.” In The Blackwell Companion to Natural
Guth, Alan H. “Eternal Inflation and Its Implications.” Journal of Physics A,
Theology, edited by William Lane Craig and J. P. Moreland, 202–281.
40 (2007): 6811–6826. Cornell University Library archive, https://arxiv
Chichester, UK: Wiley-Blackwell, 2009.
.org/abs/hep-th/0702178.
Collins, Robin. “The Fine-Tuning Evidence Is Convincing.” In Debating
Guth, Alan H., David I. Kaiser, Andrei D. Linde, et al. Letter responding
Christian Theism. Edited by J.P. Moreland, Khaldoun A. Sweis, and
to “Cosmic Inflation Theory Faces Challenges (Pop Goes the
Chad V. Meister, 35–45. Oxford: Oxford University Press, 2013.
Universe),” by Anna Ijjas, Paul Steinhardt, and Abraham Loeb. Scientific
Collins, Robin. “Stenger’s Fallacies.” 2017. http://home.messiah.edu American blog, April 2017. https://blogs.scientificamerican.com
/∼rcollins/Fine-tuning/Stenger-fallacy.pdf. /observations/a-cosmic-controversy/?previewid=7AB91FA0-1546-4DDF
Danielsson, Ulf H., and Thomas Van Riet. “What If String Theory Has No -A4F4C607B797409D.
de Sitter Vacua?” 2018. Cornell University Library archive, https://arxiv Hawking, Stephen W. “Information Loss in Black Holes.” Physical Review D
.org/abs/1804.01120. 72 (2005). http://arxiv.org/abs/hep-th/0507171v2.
Davenport, Matthew, and Ken D. Olum. “Are There Boltzmann Brains in Hogan, Craig J. “Why the Universe Is Just So.” 2000. Cornell University
the Vacuum?” 2010. Cornell University Library archive, https://arxiv Library archive, https://arxiv.org/abs/astro-ph/9909295.
.org/abs/1008.0808.
Hogan, Craig J. “Quarks, Electrons, and Atoms in Closely Related
Davies, Paul. God and the New Physics. New York: Simon and Schuster, 1983. Universes.” In Universe or Multiverse?, edited by Bernard Carr,
Davies, Paul. “A Brief History of the Multiverse.” New York Times, 12 April 221–230. Cambridge, UK: Cambridge University Press, 2007.
2003. https://www.nytimes.com/2003/04/12/opinion/a-brief-history-of Hoyle, Fred, D. N. F. Dunbar, W. A. Wenzel, and W. Whaling. “A State in
-the-multiverse.html. C12 Predicted from Astrophysical Evidence.” Physical Review 92 (1953):
Davies, Paul. The Goldilocks Enigma: Why Is the Universe Just Right for Life? 1095.
Boston: Houghton Mifflin, 2007. Ijjas, Anna, and Paul J. Steinhardt. “Implications of Planck2015 for
Donoghue, John F. 2007 “The Fine-Tuning Problems of Particle Physics and Inflationary, Ekpyrotic and Anamorphic Bouncing Cosmologies.” 2015.
Anthropic Mechanisms.” In Universe or Multiverse?, edited by Bernard Cornell University Library archive, https://arxiv.org/abs/1512.09010.
Carr, 231–246. Cambridge, UK: Cambridge University Press, 2007.
Ijjas, Anna, Paul J. Steinhardt, and Abraham Loeb. “Inflationary Paradigm in
Dowker, Fay. “Topology Change in Quantum Gravity.” 2002. Cornell Trouble after Planck2013.” Physics Letters B 723, nos. 4–5 (2013):
University Library archive, https://arxiv.org/abs/gr-qc/0206020v1. 261–266.

Ijjas, Anna, Paul J. Steinhardt, and Abraham Loeb. “Inflationary Schism.” Polkinghorne, John. Quarks, Chaos and Christianity. London: SPCK Press,
Physics Letters B 7 (2014): 142–146. 1994.
Ijjas, Anna, Paul J. Steinhardt, and Abraham Loeb. “Cosmic Inflation Rees, Martin. Just Six Numbers: The Deep Forces That Shape the Universe.
Theory Faces Challenges (Pop Goes the Universe).” Scientific American, New York: Perseus, 2000.
February 2017a. https://www.scientificamerican.com/article/cosmic-inflation
Sorkin, Rafael D., and Jorma Louko. “Complex Actions in Two-
-theory-faces-challenges/.
Dimensional Topology Change.” 1997. Cornell University Library
Ijjas, Anna, Paul J. Steinhardt, Abraham Loeb. Reply to Guth, Kaiser, Lind archive. http://arxiv.org/abs/gr-qc/9511023v3.
et al. Scientific American, July 2017b. https://blogs.scientificamerican
Spitzer, Robert J. New Proofs for the Existence of God: Contributions of
.com/observations/a-cosmic-controversy/#reply.
Contemporary Physics and Philosophy. Grand Rapids, MI: Eerdmans, 2010.
Kotzen, Matthew. “Selection Biases in Likelihood Arguments.” British
Spitzer, Robert J. “A Contemporary Thomistic Metaphysical Proof of God.”
Journal for the Philosophy of Science 63, no. 4 (2012): 825–839. https://
In Credible Catholic. Vol. 1: The Existence of God. 2016. https://www
academic.oup.com/bjps/article-abstract/63/4/825/1558362?redirected
.crediblecatholic.com/pdf/7E-P2/7E-BB1.pdf.
From=fulltext.
Steinhardt, Paul J. “Inflation Debate: Is the Theory at the Heart of Modern
Leslie, John. “Anthropic Explanations in Cosmology.” PSA: Proceedings of the
Cosmology Deeply Flawed?” Scientific American 304, no. 4 (2011): 36–43.
Biennial Meeting of the Philosophy of Science Association (1986): 87–95.
Steinhardt, Paul J., and Neil Turok. “Cosmic Evolution in a Cyclic
Leslie, John. Universes. London: Routledge, 1989.
Universe.” Physical Review D 65 (2002), 126003. Cornell University
Leslie, John. The End of the World: The Science and Ethics of Human Library archive, https://arxiv.org/abs/hep-th/0111098.
Extinction. London: Routledge, 1996.
Stenger, Victor J. The Fallacy of Fine-Tuning: Why the Universe Is Not
Lewis, Geraint F., and Luke A. Barnes. A Fortunate Universe: Life in a Finely- Designed for Us. Amherst, NY: Prometheus, 2011.
Tuned Cosmos. Cambridge, UK: Cambridge University Press, 2016.
Susskind, Leonard. The Cosmic Landscape: String Theory and the Illusion of
Li, Ming, and Paul Vitányi. An Introduction to Kolmogorov Complexity and Its Intelligent Design. New York: Little, Brown, 2005.
Applications. New York: Springer-Verlag, 2008.
Swinburne, Richard. “The Argument to God from Fine-Tuning Reassessed.”
Linde, Andrei. “Eternal Chaotic Inflation.” Modern Physics Letters A 1 In God and Design: Teleological Argument and Modern Science, edited by
(1986): 81–85. Neil Manson, 105–123. London: Routledge, 2003.
Linde, Andrei, and Mahdiyar Noorbala. “Measure Problem for Eternal and Swinburne, Richard. The Existence of God. 2nd ed. Oxford: Clarendon,
Non-ternal Inflation.” Journal of Cosmology and Astroparticle Physics 2004.
(JCAP 1009:008). 2010. Cornell University Library archive, https://
Tegmark, Max. “The Multiverse Hierarchy.” In Universe or Multiverse?,
arxiv.org/abs/1006.2170.
edited by Bernard Carr, 99–126. Cambridge, UK: Cambridge University
Lloyd, Seth. “Decoherent Histories Approach to the Cosmological Measure Press, 2007. Also available at Cornell University Library archive, https://
Problem.” 2016. Cornell University Library archive, https://arxiv.org arxiv.org/abs/0905.1283.
/abs/1608.05672v1.
Tegmark, Max. “The Mathematical Universe.” Foundations of Physics 38, no.
Lonergan, Bernard. Insight: A Study of Human Understanding. Vol. 3 of 2 (2008): 101–150. Cornell University Library archive, https://arxiv.org
Collected Works of Bernard Lonergan, edited by Frederick E. Crowe and /abs/0704.0646.
Robert M. Doran. Toronto: University of Toronto Press, 1992.
Tegmark, Max. Our Mathematical Universe: My Quest for the Ultimate
Manson, Neil A., ed. God and Design: The Teleological Argument and Modern Nature of Reality. New York: Knopf, 2014.
Science. London: Routledge, 2003.
Tegmark, Max, Anthony Aguirre, Martin J. Rees, and Frank Wilczek.
Manson, Neil A. “The Fine-Tuning Argument.” Philosophy Compass 4, no. 1 “Dimensionless Constants, Cosmology, and Other Dark Matters.”
(2009): 271–286. Physical Review D, 73, no. 2 (2006). http://www.nat.vu.nl/∼wimu
Olum, Ken D. “Is There Any Coherent Measure for Eternal Inflation?” /Varying-Constants-Papers/Tegmark-PRD-DimensionlessConstants.pdf.
Physical Review D 86 (2012), 063509. Cornell University Library Tegmark, Max, and Martin J. Rees. “Why Is the Cosmic Microwave
archive, https://arxiv.org/abs/1202.3376v2. Background Fluctuation Level 10−510−5.” Astrophysical Journal 499,
Page, Don N. “Spacetime Average Density (SAD) Cosmological Measures.” no. 2 (1998): 526–532.
Journal of Cosmology and Astroparticle Physics 2014 (JCAP11(2014)038). Turok, Neil. “A Critical Review of Inflation.” Classical and Quantum Gravity
Cornell University Library archives. https://arxiv.org/abs/1406.0504v2. 19, no. 13 (2002). http://iopscience.iop.org/article/10.1088/0264-9381
Penrose, Roger. The Emperor’s New Mind. Oxford: Oxford University Press, /19/13/305/meta.
1989. Uzan, Jeanne-Philippe. “Varying Constants, Gravitation and Cosmology.”
Penrose, Roger. The Road to Reality: A Complete Guide to the Laws of the Living Reviews in Relativity 14, no. 2 (2011). https://link.springer.com
Universe. New York: Random House, 2004. /article/10.12942/lrr-2011-2.
Penrose, Roger. “Before the Big Bang: An Outrageous New Perspective and Vilenkin, Alex. Many Worlds in One: The Search for Other Universes. New
Its Implications for Particle Physics.” Proceedings of EPAC 2006, York: Hill and Wang, 2007.
Edinburgh, Scotland. https://accelconf.web.cern.ch/accelconf/e06/PAPERS Wigner, Eugene. “The Unreasonable Effectiveness of Mathematics in the
/THESPA01.PDF. Natural Sciences: Richard Courant Lecture in Mathematical Sciences
Penrose, Roger. Fashion, Faith, and Fantasy in the New Physics of the Delivered at New York University, May 11, 1959.” Communications on
Universe. Princeton, NJ: Princeton University Press, 2016. Pure and Applied Mathematics 13, no. 1 (1960): 1–14.
TOPIC 11
Our Universe: Atheism

SCIENTIFIC EVIDENCE FOR GOD? 359
Neil A. Manson
University of Mississippi
THE ARGUMENT FROM DESIGN: BIOLOGICAL COMPLEXITY AND 365

INTERDEPENDENCE
Sahotra Sarkar
Professor, Department of Integrative Biology and Philosophy
DISTRIBUTION OF LIFE ARGUMENTS 371

Cory Juhl
This chapter discusses putative scientific evidence for the existence of God from biology and from physics and
cosmology. After presenting some of those items of evidence and articulating the arguments based on them,
the authors explain why that evidence and those arguments either disconfirm or do not support theism over
atheism.
SCIENTIFIC EVIDENCE FOR GOD?
Neil A. Manson, University of Mississippi
Many theists have thought that consideration of broad-scale features of our universe supports theism
over atheism. This thought is given expression in what is called the design argument (also known as
the teleological argument). Design arguments involve reasoning from seemingly purposeful features
of the natural world to the existence of God. Insofar as they require observational premises, they are
a posteriori arguments. Unlike cosmological arguments, however, the observational premises of
design arguments are not general and incorrigible propositions such as “there exists a contingent
reality.” Rather, the versions of the design argument popular in any given era typically refer to very
specific discoveries of the best available science of that era.
The most famous statement of the design argument is in William Paley’s Natural Theology
(2006 [1802]), in which he compared the exquisitely adapted structures of living creatures to arti-
facts like the watch and concluded that there quite probably exists a supernatural designer of living
things. The force of Paley’s argument diminished greatly when Charles Darwin provided a fully nat-
uralistic explanation of biological adaptation in The Origin of Species (1991 [1859]). From 1859
onward, the design argument lay dormant for more than a century. New scientific discoveries in
the latter half of the twentieth century, however, led to its revival. Physics and cosmology seemed
359
to show that the universe was fine-tuned for life, in the sense
KEY CONCEPTS that life of any sort could not have arisen anywhere in the uni-
verse if the basic parameters of the universe had differed even
slightly. Some theists urge that cosmic fine-tuning shows that
Adaptationism the universe did not arise by chance but rather was designed.
Big Bang Furthermore, big bang cosmology, coupled with knowl-
Cosmic Fine-Tuning edge of the workings of life at the cellular level, opened the
Design Argument door for a new design argument from biology. The temporal
and spatial finitude of the universe implied by the big bang
General Theory of Relativity meant that (contrary to the prior consensus) unlimited oppor-
God of the Gaps tunities for life to originate by chance did not exist. In Darwin’s
Intelligent Design Black Box: The Biochemical Challenge to Evolution (1996) bio-
chemist Michael Behe argues that the allegedly irreducible com-
Microadaptation plexity of life at the cellular level could be explained only as
Microevolution arising from design.
Multiverse Atheists offer several responses to these arguments.
Natural Selection Regarding the design argument from fine-tuning, they say
that it ignores the possibility that our universe is but one in
Neocreationism a much larger multiverse—a plethora of universes all differing
randomly from one another in their basic parameters. Our
universe’s apparently fine-tuned features would then be
explained in terms of the anthropic principle—the fact that
observers should expect to find themselves in a universe within the multiverse that meets whatever
conditions are necessary for the existence of observers. Because in the multiverse it is highly likely
that some universe or other is well suited for life, human observers should not be surprised to find
themselves in just that sort of universe. They also point to the sparseness of intelligent life in the
universe when measured on appropriate temporal and spatial scales—which is not to be expected
if God designed the world as an abode for intelligent life.
Atheists (and others) respond to the design argument from irreducible complexity by claiming
that Darwinian mechanisms are perfectly adequate for explaining even the most complex biological
phenomena. They also claim that design arguments from biology rest on outdated beliefs regarding
adaptationism and ecology. In response to all versions of the design argument, philosophers typically
raise points made by David Hume in Dialogues concerning Natural Religion (1970 [1779]). Even if
there is evidence of design, they say, it cannot support the conclusion that the Christian God exists
but only, at best, that some limited designer (or more than one) exists. Furthermore, the evils and
imperfections of the world block any inference to God, who is supposed to be omnipotent, omni-
scient, and morally perfect. Last, the design argument demands that we do what is beyond us: peer
into the mind of God and say what he is likely to do.
As the discussions that follow indicate, there is no part of the design argument that lacks a
rejoinder from the atheist. On balance, considerations about seemingly purposeful broad-scale fea-
tures of our universe do not clearly favor theism over atheism. Indeed, in some cases such considera-
tions may favor atheism over theism.
THE FINE-TUNING ARGUMENT

The notion that the universe is fine-tuned for life grew out of developments in physics in the twen-
tieth century. In light of the big bang, cosmologists came to recognize that the universe is highly
structured. It has precisely determined parameters such as age, mass, curvature, temperature, den-
sity, and rate of expansion. Take the age of the universe as an example. It is currently estimated
to be 13.772 billion years old, plus or minus 59 million years. Modern particle physics also has
revealed that specific kinds of particles compose the universe and that specific kinds of forces govern
these particles. For example, the mass of the proton is 938.28 mega-electronvolts per square of the
speed of light (Mev/c2), and the mass of the neutron is 939.57 Mev/c2. As we can see, these physi-
cal quantities have been determined to a very high degree of accuracy.
These parameters determine the overall structure and history of the cosmos. In particular, they
enable life to arise somewhere in the universe at some time. For example, the difference between the
mass of the proton and the mass of the neutron is crucial for the existence of stars. Without stars,
there would be no heavy elements, because all elements more massive than hydrogen and helium
were produced via nucleosynthesis inside of stars. Likewise, the rate of expansion of the universe
is crucial to the possibility of life’s arising. Too rapid an expansion would have made matter too dif-
fuse for stars and galaxies to form. Too slow an expansion would have risked the universe’s recollap-
sing as the result of gravitation.
In light of the very precise numerical values of these cosmic parameters, some physicists have
asked what the universe would have been like if those numerical values had been slightly different.
What they found was that the universe would not have been the sort of place in which life could
emerge. The possibility of life is extraordinarily sensitive to the values taken by the cosmic para-
meters. In case after case, the cosmic parameters proved to be like the just-right settings on an
old-style radio dial: if the knob is turned just a bit, the clear signal turns to static. As a result, some
physicists started describing the values of the parameters as fine-tuned for life. For ease of further
discussion, let “cosmic fine-tuning” name the fact that our universe is fit for life when it seems it
so easily might not have been.
To give just one of many possible examples of fine-tuning, the cosmological constant is a cru-
cial term in Albert Einstein’s equations for the general theory of relativity. When it is positive, it acts
as a repulsive force, causing space to expand. When it is negative, it acts as an attractive force, caus-
ing space to contract. If the cosmological constant were not precisely what it is, either space would
expand at such an enormous rate that all matter in the universe would fly apart, or the universe
would collapse back in on itself immediately after the big bang. Either way, life could not possibly
emerge anywhere in the universe. According to some calculations, for life to be possible in the uni-
verse, the cosmological constant could not have differed by more than one part in 1053. That is one
part in a hundred thousand trillion trillion trillion trillion.
Cases like the cosmological constant seem to show that the universe is exquisitely fine-tuned for
the existence of life. Some theists argue that cosmic fine-tuning strongly supports theism over athe-
ism. After all, if we believe in God, we will have an explanation of cosmic fine-tuning, whereas if we
say the cosmos is life-permitting just by chance, we must believe something incredibly improbable
happened. This fine-tuning argument is perhaps the most popular argument for the existence of
God. Despite its popularity, however, there are several problems with it. Here are three of the big-
gest ones.
First Problem: It Is Unclear Whether Cosmic Fine-Tuning Is Genuinely Improbable. Let us

stipulate that, if life is to exist anywhere in the universe, the cosmological constant could not have
differed by more than one part in 1053. Does that fact, by itself, show that it is exceptionally
improbable that the cosmological constant has a life-permitting value? Not at all. Logically speaking,
statements about what would have happened if certain conditions had been slightly different are not
the same as statements about probability. Statements about probability—assuming we are talking
about probabilities in the sense that mathematicians and scientists have in mind—cannot so easily
be derived from counterfactual conditional statements.
Consider that the basketball player Michael Jordan is very nearly 2 meters tall (6 feet, 6 inches).
Had his height not fallen right in the middle of a 1-meter-wide window in the range of possible
human heights, he likely would not have been the greatest basketball player ever. At 1.5 meters in
height, he would have been far too short. Consider that Tyrone “Muggsy” Bogues, the shortest
player in the history of the United States’ National Basketball Association, is 1.6 meters tall (5 feet,
3 inches). Meanwhile, every human over 2.5 meters in height suffers from a medical condition:
gigantism (also called acromegaly). Were Jordan over 2.5 meters tall, he would suffer from the debilitat-
ing effects of this terrible disease. So Jordan’s height is right in the center of a meter-wide window in the
range of possible human heights.
Now consider that a light-year is approximately 1016 meters. That is, the distance a ray of light
travels in one year is roughly 10 million billion meters. Therefore if Jordan’s height had fallen

outside a range the width of which is one part in 1016 of a light-year, Jordan would not have been
the greatest basketball player ever. Does this show that it is extraordinarily improbable that Jordan
had a height that likely enabled him to be the greatest basketball player? Does this show that the
probability that Jordan had the height necessary for him to be the greatest basketball player was
one chance in ten million billion? Of course not.
As basic as this mistake seems, it is sometimes committed in the fine-tuning literature. It used to be
argued that the mass of neutrinos counts as fine-tuned for life, for this reason: neutrinos are 5 10-35
kilograms, but if they were 5 10-34 kilograms, life would not be possible in the universe. This appear-
ance of great precision is simply an artifact of measuring particle masses in kilograms. Doing that is like
measuring Jordan’s height in light-years. This example suggests that the appearance of cosmic fine-tun-
ing might be the result of the choices we make for units of measure. These units are typically geared
toward the scale at which daily life is conducted. Most everyday objects of commerce have masses
within a few orders of magnitude of a kilogram. Indeed, that is why we measure masses in kilograms.
If we take off our anthropocentric lenses and look again at the case of the neutrino, however, we will
see it in a new light. For the sake of argument, let us take the actual value of the mass of the neutrino
as our standard for the possible values of the mass of the neutrino. In that case, we see that, so long
as the mass of the neutrino were not ten times greater than it actually is, life would be possible in the
universe. Put that way, the neutrino’s having a life-permitting mass sounds far less surprising.
To avoid this problem of basing the claim of cosmic fine-tuning on the choice of arbitrary units
of measure, proponents of the fine-tuning argument must restrict themselves to dimensionless para-
meters—parameters that do not involve any units of measure. For example, while the particle
masses of the neutron and the proton are given in kilograms, the ratio of the two masses is a pure
number (1.00138). This mass ratio is the same number regardless of what units are used to measure
mass. Thus, if it turns out that that ratio needs to be very nearly what it is in order for life to arise
(and it seems that it does), then an argument from fine-tuning to the existence of God could pro-
ceed without running into the problem that plagued the neutrino case mentioned earlier.
Even if they restrict themselves to dimensionless parameters, however, proponents of the fine-
tuning argument are left with a fundamental problem. The data regarding cosmic fine-tuning still
do not support the claim that the probability is low that the universe permits life. To defend that
claim, proponents of the fine-tuning argument would need to provide us with extra information.
Specifically, for each of the allegedly fine-tuned cosmic parameters, they would need to tell us (1)
the range of values it could have taken and (2) whether and to what extent some possible values
in that range are more likely to be instantiated than others.
Go back to the case of Jordan. We can talk meaningfully about the probability that a ran-
domly selected human male is 2 meters tall because we know that the range of possible human
heights is approximately 2 meters (0.5 to 2.5). The range is finite. Furthermore, we know that
the heights of human males cluster around 1.75 meters. It is only with respect to that background
information that we can say that it is not vastly improbable that a randomly selected human male
is 2 meters tall.
With regard to the truly fundamental cosmic parameters—the ones the values of which are not
derived from the values of any other cosmic parameters—we do not know the range of possible
values for them, nor do we know that some values are more likely to be instantiated than others.
Because they are fundamental, the values of the truly fundamental cosmic parameters are not
embedded in any deeper, more fundamental background. It looks as if there is no way to define a
probability in the case of the fundamental cosmic parameters. In physics, this is called the measure
problem. In philosophy, it is called the normalizability problem. Either way, it is an enormous prob-
lem for the fine-tuning argument.
Whenever the normalizability problem is brought to the attention of proponents of the fine-
tuning argument, they resort to philosophical, rather than scientific, arguments to restrict the range
of possibilities or to render some possibilities more probable than others. Quite an extensive aca-
demic literature has arisen in response to the normalizability problem. Despite all of the maneuvers,
however, there is still no consensus view regarding just what are the ranges of possible values for the
cosmic parameters from which the actual values are drawn.
Without that information, there are no constraints on what could qualify a cosmic parameter as
finely tuned. Consider, again, the ratio between the mass of the neutron and the mass of the proton.
It is 1.00138. Suppose the physicists who had investigated the effects of altering that ratio had dis-
covered that it was, so to speak, very forgiving when it came to life. Anything between [103, 10-3]
would permit life. Had the ratio been a thousand to one, life would have been possible; had it been
one to a thousand, life would have been possible. In that case, it seems the neutron/proton mass
ratio would not count as fine-tuned for life. If there is no upper limit to its possible values, then
it seems that falling within the life-permitting range [103, 10-3] does, indeed, count as fine-tuning.
After all, it could have been a billion, a quadrillion, a billionth, a quadrillionth, or some other
absurdly large or small number. Even what we would regard as only coarse-tuning would count as
fine-tuning if those are the rules.
Theoretical physicist Paul Davies articulated the basic problem beautifully in The Mind of God:
The problem is that there is no natural way to quantify the intrinsic improbability of the
known “coincidences.” From what range might the value of, say, the strength of the nuclear
force … be selected? If the range is infinite, then any finite range of values might be considered
to have zero probability of being selected. But then we should be equally surprised however
weakly the requirements for life constrain those values. This is surely a reductio ad absurdum of
the whole argument. What is needed is a sort of metatheory—a theory of theories—that
supplies a well-defined probability for any given range of parameter values. No such
metatheory is available, or has to my knowledge even been proposed. (1992, 204–205)
It has been over twenty-five years since Davies made that statement, yet proponents of the fine-
tuning argument still have not put forth such a metatheory. Until they provide one, the central
claim of the fine-tuning argument—that cosmic fine-tuning is improbable—is untenable.
Second Problem: A Fine-Tuned Universe Might Be Likely within a Multiverse. Many phy-
sicists and cosmologists today endorse, or at least have sympathies for, a multiverse theory. While
the details of such theories differ, the basic idea is that our universe is but one in a vast ensemble
of universes. Each universe within the ensemble shares certain basic lawful structures; for example,
each one follows quantum-mechanical laws. Their fundamental parameters are all the same. In all
of them there is a mass for the proton, a mass for the neutron, a speed at which light travels, a cos-
mological constant, and so on.
However, these parameters randomly take different values in the different universes. For exam-
ple, whereas in our universe the mass of a proton is 938.28 Mev/c2, in some other universe it is
something else—say, 627.59 Mev/c2. If there is a multiverse, it is unsurprising that at least one uni-
verse in the ensemble is fit for the production of life. While the probability that any particular uni-
verse is fit for life is very small, the probability that some universe or other is fit for life is now very
high, just because there are so many chances for there to be a fit-for-life universe.
If true, it seems the existence of a multiverse would help explain why some universe is just right
for life. There would be sufficient chances for the parameters to take the right values in at least one
universe. How does the existence of a multiverse explain why our universe is such that life could
arise in it? According to the anthropic principle, observers should expect their environment to meet
whatever conditions are necessary for the existence of observers.
For example, we can never find ourselves observing things from the surface of the sun, just
because life cannot exist on the surface of the sun. The fact that we are observing anything at all tells
us that our environment falls within a certain temperature range—a range compatible with the exis-
tence of observers. Likewise, says Carter, the very fact that we are observing guarantees that we are
not doing so from universes that forbid the existence of living, embodied creatures. If there is a mul-
tiverse, then it is highly probable that some universe is such that conscious observers can exist. If the
thinking behind the anthropic principle is right, such a universe is the only kind of universe we
could observe. That would explain the appearance of fine-tuning without bringing God into the
picture.
Proponents of the fine-tuning argument often respond that the multiverse hypothesis is unsci-
entific and ad hoc. There is just no way to test it. Furthermore, they accuse multiverse proponents

of concocting it in order to avoid the obvious religious implications of fine-tuning. They allege that
the multiverse is the last resort of the desperate atheist. These criticisms of the multiverse hypothesis
are wrong on two counts.
First, even if some physicists promote the multiverse hypothesis because they are atheists, it
would be an ad hominem argument to reject the multiverse hypothesis on that basis. The relevant
question is whether there is independent motivation within current physical theory for the multi-
verse hypothesis. If there is, then proponents of the multiverse hypothesis cannot be accused of fab-
ricating a theory just to block the fine-tuning argument to the existence of God. If we survey the
physicists and cosmologists themselves—as this author has done—they will say that multiverse
hypotheses are motivated by a variety of theoretical considerations, not just by a desire to explain
away the appearance of fine-tuning.
Second, if the multiverse hypothesis is testable scientifically, then it does not matter why theo-
retical physicists came up with the hypothesis. It is still a scientific hypothesis. Here again, we ought
to look to what the physicists say. There are published papers spelling out observational tests of var-
ious multiverse hypotheses—for example, by detecting “ripples” or “wakes” in the cosmic micro-
wave background radiation. Other papers report trying (and failing) to detect such ripples. This
kind of research is ongoing. It is this work by physicists, not objections by philosophers in their
armchairs, that ought to carry the most weight when deciding whether or not the multiverse
hypothesis counts as scientific.
Third Problem: Theism Does a Poor Job of Explaining Cosmic Fine-Tuning. There is a
basic supposition of the fine-tuning argument that is rarely brought to the forefront by its propo-
nents. To most theists, it seems obvious. It is that it is somewhat probable that God would make
a physical universe in which life could arise. Why, exactly, should anyone believe this? Those who
do not already believe in the theist’s account of Creation (as given in Genesis, for example) will need
another reason. They will need an account of how creating a life-friendly physical universe is some-
what likely if a being meeting God’s description exists.
Suppose there exists an all-powerful, all-knowing, morally perfect, eternal, necessarily existing
being but that we otherwise know nothing about it. What would we expect it to do? How would
we expect it to behave? Specifically, would we have any reason to expect it to create a life-permitting
universe? If the theist has no good answers to these questions, then the fine-tuning argument is not
cogent.
It should be noted that, in response to the problem of evil, so-called skeptical theists say we
have little insight into the possible goods available to God or into what God must do in order to
bring those possible goods about. If such theists are to be consistent, they ought also to agree that
we have little insight into God’s motives for creation or into what sort of world (if any) God is likely
to create. Skeptical theists ought not to agree that God is likely to create a universe like ours.
Instead, they ought to take the same attitude toward creation that they take toward evil: God and
his ways are mysterious. If that is one’s view, the fine-tuning argument will have little force.
Traditional theists will be hard-pressed to say they are confident that God is fairly likely to cre-
ate a universe like ours. Each reason given for why God would probably create a universe like ours
seems to be in tension with the theological view that God is the greatest conceivable being and is
absolutely perfect. Being absolutely perfect in every way makes it harder, not easier, to see why
God would create a world like ours or, indeed, any world at all.
Traditional theists cannot say that God created the world because God needs something. As the
perfect being, God is metaphysically independent and utterly self-sufficient. There is nothing God
lacks that God can get only by creating. Furthermore, God is supposed to be omnipotent. He can
make exist anything that can possibly exist (other than God himself). God never needs to create
one thing as a means to some other end. Any result God wants to bring about, God can bring about
directly.
It is worth noting here just how different God is from us in this regard. When we are out in
the wilderness and see a wooden lean-to, we quickly infer there is (or was) another human
around. We infer that for a simple reason: humans need shelter. Humans cannot just keep the
rain and cold away by wishing it so. We must use what is available in our environment to bring
about the ends we desire. More generally, products of design are usually easy to identify because
they fulfill needs humans have and come from whatever materials are available to humans. If a
being has no needs and no obstacles—if it never lacks anything and is never constrained in bring-
ing about what it wants—then it becomes very hard to see what does and does not count as a sign
of its activity.
Theists cannot say it was likely that God created a world like ours because it was necessary to
create it. They would be denying God’s absolute freedom not to create. Creation is an act of God’s
grace; it is in no way compelled. Indeed, according to some theologians, this is part of what it means
to say that God created the world ex nihilo.
If not these reasons, then what reasons can traditional theists give for thinking God would
likely create a world like ours? Proponents of the fine-tuning argument tend to give the same
answer. Typically they say that a world with life in it—more specifically, intelligent life—is a very
good sort of world. Such a world is said to be good for a variety of reasons: because life itself is
good, because intelligent life is good, because intelligent life with the capacity to enter into a mean-
ingful relationship with God is good, and so on. God, being perfectly good, is likely to create a
world that is good. Because a world like ours is good, God is likely to create it.
There are many problems with this line of response. Here are two. First, if the goods alluded to
require the creation of beings distinct from God and if God by his very nature brings about as much
good as he can, then we are back to our earlier problem: God’s creation is necessary, not free. God
will have to create a world like ours on pain of no longer being perfectly good. That is not going to
sit well with many traditional theists. Second, if God created this world because it is a good sort of
world, it stands to reason that God would create not just a good world but the very best possible
world. It seems, however, that, our world is not the very best possible world—given the great evil
in the world. Saying its goodness is the reason God would probably create a world like ours raises
the problem of evil all over again.
TAKING STOCK OF THE FINE-TUNING ARGUMENT

Cosmic fine-tuning is certainly a fascinating phenomenon. Atheists can agree with theists that there
is something strange and awesome about it. Nonetheless, there are grave problems in arguing from
cosmic fine-tuning to the existence of God. First, it is unclear whether cosmic fine-tuning is
improbable. Second, the multiverse hypothesis explains cosmic fine-tuning without bringing God
into the picture. Third, traditional theists cannot say with confidence that God is likely to create
a life-permitting physical universe such as ours. For these reasons and others, it seems the phenom-
enon of cosmic fine-tuning gives theism no advantage over atheism.
THE ARGUMENT FROM DESIGN: BIOLOGICAL COMPLEXITY

AND INTERDEPENDENCE
Sahotra Sarkar, University of Texas at Austin
FUNCTIONAL INTEGRATION IN THE LIVING WORLD

The argument from design is one of the oldest and most venerable attempts to use evidence from
the observable universe to infer the existence of a deity. Rhetorically, the most potent version of this
argument relies on the living world, especially the apparent fit or adaptation between organism and
environment. This fit was taken to be evidence of a deity’s work in creating the universe. Charles
Darwin’s and Alfred Russel Wallace’s mechanism of natural selection provided an alternative expla-
nation of adaptation, one that naturalized adaptation without recourse to extranatural causes such as
inferred deities. This rejection of the argument from design was central to evolutionary theory’s
nineteenth-century challenge to Judeo-Christian-Islamic theology and continues to be debated by
theists and their opponents in the twenty-first century.

That adaptation is present in the living world has been recognized in the Western intellectual
tradition at least since the age of Aristotle. By the nineteenth century Western naturalists exploring
the world tended to find adaptation everywhere. For example, neotropical sloths have often been
regarded as the most ungainly of animals. They were first systematically studied in the early nine-
teenth century by the English naturalist and traveler Charles Waterton. In his 1825 travelogue,
Wanderings in South America, Waterton observed:
On comparing [a sloth] to other animals, you would say that you could perceive deficiency,
deformity and super-abundance in his composition. He has no cutting teeth, and though four
stomachs, he still wants the long intestines of ruminating animals. He has only one inferior
aperture, as in birds. He has no soles to his feet, nor has he the power of moving his toes separately.
His hair is flat, and puts you in mind of grass withered by the wintry blast. His legs are too short;
they appear deformed by the manner in which they are joined to the body; and when he is on the
ground, they seem as if only calculated to be of use in climbing trees. He has forty-six ribs, while the
elephant has only forty; and his claws are disproportionably long. Were you to mark down, upon a
graduated scale, the different claims to superiority amongst the four-footed animals, this poor ill-
formed creature’s claim would be the last upon the lowest degree. (1973 [1825], 5–6)
But, as Waterton perceptively observes later in the book, however, these initial judgments are mis-
leading: “This singular animal is destined by nature to be produced, to live and die in the trees;
and to do justice to him, naturalists must examine him in his upper element” (1973 [1825], 93).
From an arboreal perspective, the sloth’s apparent malformations can be recognized as adaptations
for a life largely spent hanging from branches. Waterton proceeded to give the first reasonably accu-
rate description of its natural history and took the Comte de Buffon to task for assuming that the
sloth must live its life in misery because of the poverty of its design. Modern research largely vindi-
cates Waterton’s judgment (Sarkar 2007).
THE DESIGN ARGUMENT

What explains the sloth’s adaptations, all the more impressive because they initially appear to be a
bundle of malformations? The theological answer is the existence of a deity qua designer of adapta-
tions. While versions of the design argument date back to Plato (Ruse 2003), it was most famously
elaborated by the natural theologian William Paley in his 1802 book, Natural Theology:
In crossing a heath, suppose I pitched my foot against a stone, and were asked how the stone
came to be there; I might possibly answer, that, for any thing I knew to the contrary, it had
lain there for ever: nor would it perhaps be very easy to show the absurdity of this answer. But
suppose I had found a watch upon the ground, and it should be inquired how the watch
happened to be in that place; I should hardly think of the answer which I had before given,
that, for any thing I knew, the watch might have always been there. Yet why should not this
answer serve for the watch as well as for the stone? why is it not as admissible in the second
case, as in the first? For this reason, and for no other, viz. that, when we come to inspect the
watch, we perceive (what we could not discover in the stone) that its several parts are framed
and put together for a purpose, e. g. that they are so formed and adjusted as to produce
motion, and that motion so regulated as to point out the hour of the day; that, if the different
parts had been differently shaped from what they are, of a different size from what they are, or
placed after any other manner, or in any other order, than that in which they are placed, either
no motion at all would have been carried on in the machine, or none which would have
answered the use that is now served by it. (2006 [1802], 1–2)
If the inference about watches and their human designers is tenable, Paley argues, the same infer-
ence about adaptations in living systems and their designers leads to a deity.
Paley’s most compelling example—and one that Darwin felt forced to address explicitly—was
that of the eye, which he compared to the telescope. Paley explored the eye’s complexity in detail,
for instance, how light rays must bend more when they pass from water into the eye than when they
pass from air into the eye. This requires that the lens be more convex in the former case. “Accord-
ingly,” Paley observed, “we find that the eye of a fish, in that part of it called the crystalline lens, is
much rounder than the eye of terrestrial animals” (2006 [1802], 19).
Paley’s design argument is typically treated as a single argument, but it consists of two distinct
ones run together (Sarkar 2007). The first is an argument for functionality: that there is something
special about genuine artefacts such as watches that is also found in living systems due to their com-
posite functionality. By this Paley refers to their purposiveness, their goal-directedness which
requires integrated actions from their components. The second is an argument to a designer—that
functionality implies the existence of a conscious designer and, in most versions, an intelligent
designer.
CRITIQUE
By the time Paley was writing, the argument for functionality had already been subjected to a scathing
critique by David Hume in his 1779 Dialogues concerning Natural Religion. While Paley focused on
individual organisms and their adaptations, Hume focused on the question of whether the world as
a whole could be used as evidence of the handiwork of a deity. Hume presented arguments to suggest
that the world is more like a vegetable or animal than a machine, in which case there is no require-
ment for a conscious designer. (Hume also had other reasons for rejecting the argument from design,
including whether the putative designer had to be intelligent rather than some entity capable of only
rote repetition of an action.) Insofar as Paley was offering a credibly reconstructed design argument
that would survive Hume’s skepticism, that argument depended on two claims: first, that the relevant
entity is an individual organism along with its biological adaptations (and not the world as a whole)
and, second, the argument to a designer as inferred from observed functionality.
This is where Darwin and Wallace enter the story. Independently, they proposed that the survival
of the most fit would lead to the establishment of adaptation between organism and environment.
The plausibility of the mechanism was undeniable after Darwin published the Origin of Species in
1859. Darwin furnished example after example of how a fit between organism and environment
(e.g., the feeding apparatus of insects and the structure of flowers) would contribute to viability or fer-
tility, the two components of fitness. What added force to Darwin’s argument is his careful attention
to potential problems for the theory of natural selection, including the existence of organs of extreme
perfection such as Paley’s most important example, the eye. Darwin accepted that the evolution of
such features through natural selection would be a slow process requiring millions of years. He
believed that the earth was old enough for such evolution to have been possible. Both Darwin and
Wallace also proposed that natural selection could lead to indefinite divergence between lineages of
organisms: speciation was a ubiquitous and almost trivial consequence of this worldview. Beyond
the design argument, the prospect of speciation through the mechanical action of blind variation
and natural selection posed difficult problems from all religious perspectives that demanded the special
creation of each species by a deity. (However, this issue is beyond the scope of this section.)
In the first edition of the Origin (1859), Darwin relied almost entirely on blind variation and nat-
ural selection as the mechanisms of evolutionary change. But, he began to admit and emphasize other
mechanisms of evolutionary change in subsequent editions, partly in response to now-discredited
arguments from physicists that not enough time had been available for the slow process of natural
selection acting on blind variation to have effected the observed evolutionary changes in the history
of the earth. By the end of the nineteenth century, evolution by natural selection alone was often
not deemed to be sufficient to explain the complete history of life on the earth.
However, evolution by natural selection became the standard view of evolution by the end of
the 1920s, when work in theoretical population genetics by J. B. S. Haldane, Sewall Wright, and
Ronald Fisher established that ample time had been available for natural selection to have effected
the observed modifications during the evolutionary history of life on the earth. Empirical data, such
as for the emergence of adaptive melanism in the peppered moth (Biston betularia) in Britain, sup-
ported this theoretical work. By the mid-twentieth century, the argument from biological design
had come to have little relevance for those who took science seriously.
FIRST RESURRECTION: DEMBSKI AND THE EXPLANATORY FILTER

When scientific creationism emerged to challenge evolutionary biology in the United States in the
1960s, one of its major goals was the insertion of creation science into science curricula (Numbers

1992). Most scientific creationists supported literal readings of the Bible and sought direct evidence
for these claims. Arguments such as that from design were peripheral to that strategy. A 1987 US
Supreme Court decision (Edwards v. Aguillard) ended these attempts to corrupt science curricula
in the United States. While the core of the scientific creationism movement went into decline sub-
sequently, in the late 1990s part of that movement morphed into the neo-creationist intelligent
design (ID) movement in which design arguments play a central role. There is a critical way in
which the context had changed from Paley’s time. It was no longer credible to deny that any natural
process could lead to design in the sense of adaptation: natural selection took care of that. In gen-
eral, ID theorists accepted natural selection as a sufficient explanation for many aspects of adaptive
microevolution (that is, relatively minor modifications within species). However, they continued to
deny that it was sufficient for complex and highly integrated microadaptations and especially for
macroevolution (at taxonomic levels higher than that of species).
ID relies on two strategies that map neatly into the two components of the traditional design
argument that were distinguished earlier. The first is usefully dubbed the argument for complexity,
which is the neocreationist reconstruction of Paley’s argument for functionality. The main propo-
nent is ID theorist William Dembski, who has proposed an “explanatory filter” to identify events
that are too complex and, therefore, too improbable to have arisen by chance and that must reflect
some lawlike feature in operation. Suppose that these events are also specified, that is, in some sense,
they fit a pattern. For critics and neutral commentators, this sense has always remained illegitimately
vague. Dembski (2002) later tried to explicate this concept of specification through an invocation of
information theory that almost all commentators have dismissed as vacuous—space constraints will
prevent any further discussion of this issue here, but see Sahotra Sarkar (2007).
Dembski’s explanatory filter aimed to show that if this lawlike feature is not a natural regular-
ity, it must be due to design. Sarkar (2007, 51), following Branden Fitelson, Christopher Stephens,
and Elliott Sober (1999), reconstructs Dembski’s (1998) procedure. Given an event E to be
explained:
i. Assume that three explanations of E, regularity (R), chance (C), and design (D), are mutually
exclusive and jointly exhaustive.
ii. Assume that there is a method for determining if E is specified.
iii. Assume that R is preferred over C and C over D (which leaves D as the last option standing if no
other explanation works).
iv. Because of this ranking, R, C, and D should be examined and eliminated as needed in that order.
v. If E has high probability, R should be accepted; otherwise, it should be rejected in favor of C or D.
vi. If C does not assign a low probability on E or if E is not specified, C should be accepted;
otherwise it should be rejected.
vii. Only D remains and should be accepted.
The filter relies on small probabilities to rule out a nondesign explanation of events, such as com-
plex adaptations (though its first definitive exposition in Dembski’s 1998 book avoided all biological
examples).
Philosophers of science across the board have criticized Dembski’s explanatory filter for making
a mockery out of scientific inference (e.g., Fitelson, Stephens, and Sober 1999; Wilkins and Elsberry
2001; Sarkar 2007):
• Perhaps the most bizarre aspect of the filter is that at no stage does the filter ask what probability
design D confers on event E. This leaves open the possibility that, after passing through the filter,
E is attributed to D even if the conditional probability that D assigns to E, Pr{E|D} is low,
perhaps close to 0.
• Because Dembski never explains what design means (and makes no reference to a designer,
presumably to avoid accusations of illegitimately importing religious assumptions), there is no
method for calculating the probability that D assigns to E, Pr{E|D}.
• The filter starts with the evaluation of the probability of E and not the probability that R confers
on E, that is, Pr{E|R}. Here it is in conflict with both Bayesian and classical modes of inference,
which are the standard modes of statistical inference in contemporary science. (Problems with
assigning Pr{E} with no consideration of what R is are noted below.)
• The filter gives no way of determining what the threshold probability is below which R should be
rejected or what C must assign to E (that is Pr{E|C}) before C must be rejected. Such a threshold
is completely arbitrary.
• Perhaps most oddly, R is considered to be all laws of nature in general rather than some particular
law. It is difficult to imagine what this hypothesis could mean in practice.
• Asking what Pr{E} is with no specification of context is meaningless. Probabilities depend on
reference classes, and these can be constructed only using all the relevant information one has
about the type of event that is being considered—that is, there must be a conditional probability
even if the conditions are left implicit.
Indeed, Dembski’s explanatory filter consists of logical trickeries, as Fitelson and colleagues
(1999) were the first to note. The first piece of trickery consists of considering regularity, chance,
and design in that order when any other would lead to different inferences. The second is that reg-
ularity, chance, and design have to be considered singly; there is no option to consider any two (or
more) of them together, thus excluding by fiat the evolutionary mechanism of natural selection (a
regularity) acting on blind variation (chance). Sarkar (2007) provides many counterexamples to
the explanatory filter.
The critiques have been compelling, and so far Dembski has not offered responses in the peer-
reviewed literature in which the critiques first appeared. It is, therefore, perhaps not surprising
that the explanatory filter has rarely been invoked in the ID corpus since the start of the twenty-first
century, though it has never been officially repudiated.
SECOND RESURRECTION: BEHE AND IRREDUCIBLE COMPLEXITY

In the late 1990s, in parallel with Dembski’s advocacy of the explanatory filter, Michael Behe
(1996) presented a set of biochemical objections to evolution by natural selection. Behe began by
identifying biochemical phenomena for which there was as yet no plausible evolutionary history
and argued that these phenomena present problems for evolutionary theory. Such phenomena are
easy to find: in any science that continues to be a living field of research, there are bound to be open
problems. For theism, the trouble is that such a strategy constitutes deploying a “God of the gaps”
argument, saving for a deity what science cannot currently explain (Drummond 1894). As science
progresses, the role of the deity continually diminishes. Long derided by theologians for thus pro-
gressively demoting the role of God in the universe, to the extent that Behe has relied only on this
strategy, he provides little new reason to embrace a deity.
However, Behe does add some respectability to his objections to evolution by natural selection
by supplementing this God of the gaps strategy with an argument that the unexplained biochemical
features of systems have remained unexplained not because insufficient effort has been dedicated to
their explanation, but because these systems are “irreducibly complex” and therefore, in principle,
immune to evolution through natural selection. Behe’s definition of irreducible complexity seems
innocuous: “By irreducibly complex I mean a single system which is composed of well-matched,
interacting parts that contribute to the basic function, and where the removal of any one of these
parts causes the system to effectively cease functioning” (1996, 39). While Behe seems to suggest
that such biological systems are special, they are in fact ubiquitous (Sarkar 2007).
Behe’s crucial claim is that irreducible complexity cannot evolve through natural selection. His
examples include the bacterial flagellum and the human blood-clotting system. These were well cho-
sen, because in 1996 evolutionary reconstructions of their history were sparse (unlike now—see Sar-
kar 2007); however, the recognition of such cases goes no further that a God of the gaps argument.
Is there any biologically based argument that shows that irreducible complexity cannot evolve
through natural selection?

Given that irreducible complexity is ubiquitous, it would be surprising—and perhaps telling—

if biologists have paid no attention to the pathways by which they may emerge through standard
biological mechanisms. Indeed, they have. There are two common pathways: lost functional redun-
dancy and structure co-option.
H. Allen Orr succinctly explained the redundancy pathway in a review of Behe’s 1996 book:
An irreducibly complex system can be built gradually by adding parts that, while initially just
advantageous [later] … become essential. The logic is very simple. Some part (A) initially does
some job (and not very well, perhaps). Another part (B) later gets added because it helps A.
The new part isn’t essential, it merely improves things. But later on A … may change in such a
way that B now becomes indispensable. (1996–1997, 29)
The system is irreducibly complex in Behe’s sense. Examples include the mitochondria in cells,
which were initially free-living organisms but are no longer capable of independent existence (Sarkar
2007). They have lost many genes required for independent existence, because the functions of
those genes were taken over by genes in the nuclear genome of the cell of which the mitochondria
are a part.
The logic of the co-option pathway is only slightly more complicated: Suppose two features A
and B are present because of a function F for which they were selected. A also contributes to a func-
tion G. If G becomes more important, A may be modified to serve G even better and could lose
function F. B will have become indispensable, and the system would become irreducibly complex
in Behe’s sense. In one example, bird feathers are supposed to have evolved because they initially
contributed to thermoregulation but later became modified for flight (Sarkar 2007).
In later work Behe (2008) does not engage his critics, and there is no evidence that any ade-
quate response exists. Irreducible complexity remains a poor argument for design. Worse, is it even
a sign of good design? Though there is ample evidence of evolved irreducible complexity in living
systems, they may be no more than indicative of how such systems may persist if selection against
them is not strong. Irreducible complexity may well be very poor design (Miller 1999). It results
in potentially devastating fragility: the system cannot persist if a single component fails. In contrast,
functional redundancy, the opposite of irreducible complexity, may be very good design, allowing a
system to survive the failure of some parts. Functional redundancy has evolved in the living world
many times more often than irreducible complexity. From the perspective of evolution by natural
selection, this is no surprise.
The 2005 decision by a Pennsylvania federal court ruling that ID was religion rather than sci-
ence (Kitzmiller v. Dover) defused the neocreationist movement in the United States, at least tempo-
rarily. Since then, the new design arguments of Dembski, Behe, and their ilk have receded from
public view. Do biological design arguments have a future? Probably not, because design, even in
the naturalized form of adaptation, may not be as common in nature as once thought.
NATURE AND DESIGN IN THE LIGHT OF RECENT FINDINGS IN BIOLOGY

The design argument relies on the ubiquity of well-constructed biological systems, ones in which
the interdependent parts interact harmoniously and thereby contribute to the betterment of the
whole. At the level of the individual organism, such integrated operation would be manifested as
adaptation of each part in which it contributes to individual fitness. At the level of ecological com-
munities, consisting of individuals of multiple species, being well constructed would suggest exhibit-
ing a “balance of nature.” Twentieth-century biology inherited both these assumptions from earlier
eras. While the role of natural theology in promoting the first of these two assumptions is well
known (and mentioned earlier, in the discussion of Paley), the balance of nature assumption also
has theological roots (see, e.g., Egerton 1973), though they have not so far played a prominent role
in creationist arguments. Both have been challenged and largely abandoned in evolutionary biology
and ecology since the 1950s.
In evolutionary biology, the claim that there is pervasive adaptation is known as adaptationism.
It has been highly controversial since a critique by Stephen J. Gould and Richard C. Lewontin
(1979) called it the Panglossian paradigm and showed how it was typically based on “just so” stories
rather than empirical data. Genomic data have since challenged adaptationism even more strongly
(Sarkar 2015). It is hard even to find “just so” stories for why rice has almost three times as many
genes as humans or why most of the DNA in eukaryotic genomes appear not to have any function
at all. Adaptationism requires the decisive action of natural selection, but since the 1960s it has also
been recognized that evolution at the molecular level may be driven by neutral mutations and their
spread through drift (statistical fluctuation effects during reproduction) at least as much as by selec-
tion (Dietrich 2007). It may be the case that living organisms are far from Paley’s well-designed
watches. Most living features may be as much a result of contingency and chance as natural selec-
tion. If this is correct, in general, there is no putative design to explain at all. Rather, theology
should address the problem of why so many features of organisms may not be functional and may
even be dysfunctional, though not dysfunctional enough to be eliminated by natural selection.
In ecology, the idea of a balance of nature also has not survived the twentieth century. Even in
the early part of the twentieth century disagreement existed between those who viewed ecological
assemblages of species as integrated communities forming a superorganism and those who regarded
them as a result of the fortuitous mixing of species placed in the same spatial locale (Cooper 2007).
While this controversy remains unresolved, the wealth of evidence collected since the 1950s favors
the fortuitous assembly of species (Wu and Loucks 1995). The balance of nature assumption also
underlies a once-common belief in ecology that, by and large, natural communities were at equilib-
rium or reached equilibrium soon after disturbance (Egerton 1973). Empirical data from the last
half-century have discredited this belief. Equilibrium systems are nonexistent among natural com-
munities; what has emerged as being of most interest is the structure of disequilibrial communities
(Wu and Loucks 1995). As in the case of evolutionary features of organisms, contingency and his-
tory are central to the structure of ecological communities. To summarize: Just as contemporary
evolutionary biology denies the preponderance of well-designed features within living organisms,
contemporary ecology denies a balance of nature and the preponderance of fully functionally inte-
grated ecological communities. There just may be insufficient design in nature to support an infer-
ence to a deity, even if both components of the biological design argument survive the objections
leveled against them.
Both components of the biological design argument—that design similar to that of human arte-
facts occurs in the living world and calls out for special explanation and that the best such explana-
tion is the postulation of a conscious designer—are no longer plausible in the light of modern biol-
ogy. Whether there is design in nature similar to anthropogenic design has always been a matter of
controversy; developments in evolutionary biology and ecology in the late twentieth century show
that any assumption of such design is scientifically unwarranted. Meanwhile, Darwin’s and Wal-
lace’s elaboration of the mechanism of natural selection indicates that any design in nature has a
straightforward natural explanation. The design argument adds very little, if any, warrant for the
claim that a deity governs the universe.
DISTRIBUTION OF LIFE ARGUMENTS
Cory Juhl, University of Texas at Austin
In his book The Non-existence of God (2004), Nicholas Everitt presents what he dubs “arguments
from scale” (see Chapter 11) in support of atheism. He thinks that such arguments, while not con-
clusive, provide significant inductive support for atheism. In this section we examine both argu-
ments from scale and the broader question of whether spatiotemporal distribution patterns pertain-
ing to life carry any significant evidential weight concerning some forms of theism or atheism that
are of broad interest.
Everitt provides us with a fictional example as a way of making clear how he conceives the struc-
ture of his arguments from scale. Suppose that Robinson Crusoe is on an island and wonders whether
there are any other shipwreck survivors or any other human beings on the island. He should consider
what other humans in similar situations might reasonably do and then check the island for such signs

or patterns. One potential limitation of Everitt’s analogy is that we are to imagine that other humans
are inclined to attempt to show that they are on the island. If we build this into the argument from
scale, then that argument is an instance of other arguments from nonbelief (see Drange 1998).
According to this argument, if God existed, he would do more than he seems to have done to make
himself known. The actual arguments that Everitt makes do not hinge on this additional assumption.
Following Everitt, in this section we ignore questions pertaining to whether God would reveal his
presence more clearly via facts about life and its distribution throughout the universe.
The general form of Everitt’s argument is this:
• If there is a world/universe creator G with nature N, intentions I, and beliefs B, then it will
(likely) produce a world/universe with feature C.
• The world does not display C.
• Therefore, there is likely no such G (with N, I, and B).
Everitt thinks that, given a widespread NIB profile attributed to God, features of the distribu-
tion of life in the universe provide evidence against these forms of theism. Everitt further thinks that
a standard NIB profile makes it probable that God’s intentions in creating the universe place
humankind’s existence at the forefront of his purposes. This, Everitt believes, would make some-
thing like a Genesis account, with humans present throughout most of the history of the universe,
highly likely. This motivates Everitt’s argument from temporal scale.
ARGUMENTS FROM TEMPORAL SCALE

Our best estimates of the age of the universe and of the period during which creatures plausibly
identified as humans have existed yield a very surprising and different result. Humans have existed
only for a tiny fraction of the age of the universe, roughly 1/100,000. If one of the main purposes
of creating a universe is to create humans, God would not form a system that includes humans in
only a relatively minuscule fraction of the overall age of the universe, according to Everitt.
One might object that it is inappropriate to take the entirety of human existence to be limited
to our existence up to the present. Humans may be around for some additional, indefinitely long
time. If so, the human existence time ratio might look better for theism. This point connects to a
general issue that arises for arguments concerning the relevance to theism of facts about the distribu-
tion of life in the universe. What might we evaluate when we assess whether known facts concerning
the distribution of life in the universe are evidence for or against theism? We might want to consider
just the bearing of distribution of life and leave aside religious experiences or miracle reports or fine-
tuning arguments. This could turn out to be misleading, to the extent that the evidential weight of a
given collection of facts can vary depending upon what other facts are taken for granted.
We might nevertheless think that the distribution-of-life and, say, fine-tuning arguments are inde-
pendent. If this is correct, given what participants in the debate are liable to accept as known, then we
might want to consider the bearing of, in some interesting sense, just the distribution of life. To the
extent that this is feasible (and it would be too much of a tangent to go into the many complications),
we still may not want to consider only what is known, as opposed to what is highly likely, given what
we know. For example, we might argue that given the many ways that humans might cease to exist—
including epidemics, asteroid impact, a nearby neutron star gamma-ray event, nuclear or biological or
artificial intelligence or nanobot warfare, ultrapersuasive anti-childbearing memes, the destruction of
food sources via environmental pollution, and so on—it is unbelievable that humans will exist for bil-
lions of years. Survival for the next century is highly uncertain. Given these other considerations, while
we cannot predict how long humankind will survive, this uncertainty does not undermine the appro-
priateness of Everitt’s presumption that humans will ultimately exist for only a tiny fraction of the
time during which the universe exists from the time of the big bang.
ARGUMENTS FROM SPATIAL SCALE

The earth-sun system is minuscule relative to the known universe. Apparently such considerations
were thought to count against Copernicanism (see, e.g., Kuhn 1957). Astronomers understood that
if Renaissance astronomer Nicolaus Copernicus was correct in his assertion that the earth orbits the
sun, given that no parallax was observable (at the time) for stars, this fact entailed an inconceivably
large chasm of space separating the solar system from the celestial sphere. This inconceivably
immense wasted void was taken to be highly implausible, given the universally assumed background
theism. Fellow Renaissance astronomer Galileo felt obliged to argue against these objections in his
Dialogue Concerning Two Chief World Systems (1967; originally published in 1632). That Galileo
felt it necessary to deal with the issue provides data to support that many educated people, when ini-
tially considering the question and without already knowing the answer, thought that the likelihood
was low, given theism, that the universe would be much larger than the dimensions of the earth-sun
system.
Whatever bearing historical intuitions may have, Everitt argues that the scale of the known uni-
verse is mind-bogglingly larger than the scale of human existence, and he thinks that this fact pro-
vides evidence against the existence of a God who created the universe with purposes that suppose
a great importance for humans in the divine scheme of things. There is no known plausible reason
why God, paying special attention to humans, would produce—seemingly without pattern—so
many other distributed planetary systems, all so inconceivably distant, in addition to one planet
containing humans. Everitt notes that the scale of the human-oblivious universe is so large (to the
best of our knowledge) compared with the region occupied by humans that no foreseeable discover-
ies are likely to undermine the force of the argument.
One sort of objection to his arguments that Everitt considers is that just because God as repre-
sented in biblical texts appears to be interested in human affairs, this is not a doctrine adhered to
within of all forms of theism. Everitt argues that phenomena pertaining to spatial and temporal scale
raise doubts about the existence of any God whose existence we permit empirical evidence to count
for or against. While Everitt may be correct that various details of the biblical conception are dis-
pensable while preserving the force of the arguments, it is doubtful that Everitt’s arguments from
scale and their generalizations to patterns of distribution can have much force against God con-
ceived in a highly deistic manner. Nevertheless, it is of interest to focus attention on the question
of whether arguments concerning the distribution of life in the universe provide evidence against
the existence of God as conceived in the most widespread Judeo-Christian traditions. This focus
enables the argument to retain force against the existence of (such a) God. Excessive ambitions of
scope arguably sap the strength of the arguments, whereas the more focused versions against the
existence of God as widely understood can be assured of some success.
ARGUMENTS FROM POSSIBLE OR ACTUAL PATTERNS OF DISTRIBUTION OF LIFE

Although Everitt emphasizes the greater scales of cosmological space and time compared with
human-involved spaces and times, it may be of interest to introduce spatial or temporal patterns
as well. Objectors can point out that there is no logical entailment between sheer sizes of spatial
or temporal expanses, on one hand, and importance of the items existing at those times or places,
on the other. One can accept the force of this objection but then appeal to the seeming absence
of anything special about the location, either temporal or spatial, of the realm of human activities.
Earth is in one of the outer bands of a fairly unremarkable galaxy, which is in an unremarkable
supercluster of galaxies. The past 100,000 years, give or take an order of magnitude, does not appear
to be a special phase in the history of the cosmos, aside from the history of one branch of primates
wandering about a tiny spherical speck. There is no sign that there is anything special about this
time or place in the known cosmos, aside from the existence of humans. We might call such con-
siderations the “nothing special” distribution problem.
By way of contrast, consider the Aristotelian cosmological scheme. The earth, the domain of
humans, is at the center of the entire scheme. Such a scheme fits naturally with a creator who cre-
ated the cosmos with an eye toward human affairs, but the cosmos that we observe appears dramat-
ically different and at best indifferent. Admittedly, it is notoriously difficult to adjudicate in general
whether some complex situation is special, because practically any complex structure, such as the
dust pattern on my desk, is special in innumerable ways. This observation makes it difficult to see
how to convincingly support or undermine Everitt’s argument via such considerations.

An additional consideration concerns the fact that the universe appears not just indifferent to
humans but, indeed, unremittingly hostile, except for one body, the earth. Although astronomers
have discovered an enormous plethora of planets, there are reasons for thinking that the conditions
required to support life, especially multicellular life, are extremely narrow. Although there remains
much uncertainty, it is plausible that the earth is either unique or among an infinitesimal minority
in the universe with the right Goldilocks conditions, we might say (see Ward and Brownlee 2000).
This crushing hostility to human life is not what one would expect to observe if there is a creator
who produced the universe in significant part to enable humans to flourish.
Hostility to human expansion and flourishing is a matter of degree, of course. Maybe within a
century or two there will be human outposts on an extraterrestrial body or two, such as Mars or one
of the moons of Jupiter or Saturn. Even so, given that humans are told to be fruitful and multiply,
there appear to be limits to such fruitfulness, considering the very limited parts of our universe that
are at all compatible with human life. Perhaps a novel technology will be developed to readily ter-
restrialize planets, as occurs in various science fiction scenarios. Even if so and even if the population
can grow by an order of magnitude or two by inhabiting the oceans and underground cities, the
prospects for indefinite fruitfulness and multiplication are dim. Yet suppose we had been presented
with purported divine revelations telling humans to be fruitful and multiply, suggesting that there is
a creator who has human fruitfulness as an important goal or desideratum. If it then turned out
upon investigation that the entire earth was utterly uninhabitable besides a few acres near Eden, this
discovery would naturally generate perplexity as to why this creator would produce a world so
unfriendly to multiplication. Analogously, the discovery of the hostility of the vast majority of the
universe to human existence counts against the existence of a God who values human lives.
A possible rejoinder might be that God meant that we should be fruitful and multiply to some
degree but to stop around 10 billion, give or take an order of magnitude. But any de facto numeri-
cal bound on human population growth would seem unmotivated, to the extent that additional
meaningful lives are of value to God. This argument, as Everitt repeatedly emphasizes for his argu-
ments from scale, is as defeasible as any that has to conjecture about possible intentions or further
developments, either revelatory or technological. On the face of it, the discovery that the universe
is almost completely and unremittingly hostile to human life counts somewhat against the existence
of a God who supports the sort of human expansive flourishing for Abraham’s descendants that is
suggested in Judeo-Christian accounts.
Because of their dependence on assumptions concerning the importance of humankind to
God’s affairs, the significance of facts concerning the distribution of life vary depending upon what
intentions are attributed to God. Arguments from astronomical patterns pertaining to life or distri-
bution of life arguments, of which Everitt’s arguments from scale are examples, provide independent
reasons to doubt the existence of God as commonly conceived within orthodox Judaic and Chris-
tian doctrine.
Bibliography
Behe, Michael J. Darwin’s Black Box: The Biochemical Challenge to Evolution. Dembski, William A. No Free Lunch: Why Specified Complexity Cannot Be
New York: Simon and Schuster, 1996. Purchased without Intelligence. Lanham, MD: Rowman and Littlefield,
2002.
Behe, Michael J. The Edge of Evolution: The Search for the Limits of
Darwinism. New York: Simon and Schuster, 2008. Dietrich, Michael R. “Molecular Evolution.” In A Companion to the
Philosophy of Biology, edited by Sahotra Sarkar and Anya Plutynski, 157–
Cooper, Gregory J. Science of the Struggle for Existence: On the Foundations of
168. Malden, MA: Blackwell, 2007.
Ecology. New York: Cambridge University Press, 2007.
Drange, Theodore M. Nonbelief and Evil: Two Arguments for the
Darwin, Charles. The Origin of Species. Amherst, NY: Prometheus Books,
Nonexistence of God. Amherst, NY: Prometheus Books, 1998.
Drummond, Henry. Ascent of Man. New York: James Pott, 1894.
Davies, Paul. The Mind of God: The Scientific Basis for a Rational World. New
York: Simon and Schuster, 1992. Egerton, Frank N. “Changing Concepts of the Balance of Nature.” Quarterly
Review of Biology 48, no. 2 (1973): 322–350.
Dembski, William A. The Design Inference: Eliminating Chance through Small
Probabilities. New York: Cambridge University Press, 1998. Everitt, Nicholas. The Non-existence of God. London: Routledge, 2004.
Fitelson, Branden, Christopher Stephens, and Elliott Sober. “How Numbers, Ronald L. The Creationists: The Evolution of Scientific Creationism.
Not to Detect Design.” Philosophy of Science 66, no. 3 (1999): Berkeley: University of California Press, 1992.
472–488.
Orr, H. Allen. “Darwin v. Intelligent Design (Again): The Latest Attack on
Galilei, Galileo. Dialogue concerning the Two Chief World Systems: Ptolemaic Evolution Is Cleverly Argued, Biologically Informed—and Wrong.”
and Copernican. (1632.) Translated by Stillman Drake. Berkeley: Boston Review (December 1996/January 1996–1997): 28–31.
University of California Press, 1967.
Paley, William. Natural Theology; or, Evidences of the Existence and Attributes of
Gould, Stephen J., and Richard C. Lewontin. “The Spandrels of San Marco the Deity. (1802.) Oxford: Oxford University Press, 2006.
and the Panglossian Paradigm: A Critique of the Adaptationist
Ruse, Michael. Darwin and Design: Does Evolution Have a Purpose?
Programme.” Proceedings of the Royal Society (London) Series B, Biological
Cambridge, MA: Harvard University Press, 2003.
Sciences 205, no. 1161 (1979): 581–598.
Sarkar, Sahotra. Doubting Darwin? Creationist Designs on Evolution. Oxford:
Holder, Rodney D. God, the Multiverse, and Everything: Modern Cosmology
Blackwell, 2007.
and the Argument from Design. Burlington, VT: Ashgate, 2004.
Sarkar, Sahotra. “The Genomic Challenge to Adaptationism.” British Journal
Hume, David. Dialogues concerning Natural Religion. Indianapolis, IN:
for the Philosophy of Science 66, no. 3 (2015): 505–536.
Bobbs-Merrill Educational, 1970. First published 1779.
Ward, Peter D., and Donald Brownlee. Rare Earth: Why Complex Life Is
Kuhn, Thomas. The Copernican Revolution: Planetary Astronomy in the
Uncommon in the Universe. New York: Copernicus Books, 2000.
Development of Western Thought. Cambridge, MA: Harvard University
Press, 1957. Waterton, Charles. Wanderings in South America, the North-west of the
United States and the Antilles, in the Years 1812, 1816, 1820 and 1824.
Leslie, John. Universes. New York: Routledge, 1989.
Oxford: Oxford University Press, 1973. First published 1825.
Lewis, Geraint F., and Luke A. Barnes. A Fortunate Universe: Life in a Finely
Wilkins, John S., and Wesley R. Elsberry. “The Advantages of Theft over
Tuned Cosmos. Cambridge, UK: Cambridge University Press, 2016.
Toil: The Design Argument and the Argument from Ignorance.” Biology
Manson, Neil A., ed. God and Design: The Teleological Argument and Modern and Philosophy 16, no. 5 (2001): 711–724.
Science. New York: Routledge, 2003.
Wu, Jianguo, and Orie L. Loucks. “From Balance of Nature to Hierarchical
Miller, Kenneth R. Finding Darwin’s God: A Scientist’s Search for Common Patch Dynamics: A Paradigm Shift in Ecology.” Quarterly Review of
Ground between God and Evolution. New York: Cliff Street Books, 1999. Biology 70, no. 4 (1995): 439–466.

TOPIC 12
Human History: Theism

Brendan Sweetman
Rockhurst University, Kansas City, MO
This essay discusses the arguments in favor of theism and atheism through a consideration of human
evolutionary history, including the relationship between evolution and scriptural revelation, the role of
chance in evolution and science, as well as the problem of evil and suffering in the universe. The question of
design in nature, along with the development of human civilizations and the demographics of theistic belief,
is also a focus.
THE THEORY OF EVOLUTION
One of the most interesting and indeed challenging topics to emerge in recent times in the dispute
between theists and atheists is the significance that should be accorded to the theory of evolution.
Issues of disagreement concern current interpretations of the theory, the implications of evolution
for issues outside of science, as well as the appropriate relationship between religion and science,
more generally. Our aim here must be modest: to bring some clarity to the topic so as to better
appreciate the issues raised for the theism/atheism debate, to argue that the claim that evolution
provides a good case for atheism is greatly exaggerated, and that, with certain qualifications, the
design evident in the process of evolution is, in fact, a further argument for the existence of a
designer.
The theory of evolution is one of the most influential and thought-provoking scientific theories
of all time. No one who is interested in the ultimate questions of existence—who is pondering the
place of human beings in the overall scheme of things—can afford to be ignorant of the theory. To
bring clarity and order to the issues, it is helpful to organize our thinking around three main ques-
tions: (1) What factual claims does the theory make (e.g., about the origin and development of spe-
cies, and especially about the human species)? (2) What evidence is offered to support the factual
(scientific) claims? (3) What are the implications of the theory for other important issues in religion
and philosophy (e.g., what is its significance for our understanding of the nature of human beings,
and our place in the universe)? We will assume for the sake of argument that the evidence for evo-
lution is pretty good (question 2), and so our concern must primarily be about the implications of
the theory (question 3). It is important to have a sense of how much one can extrapolate from the
evidence, and to note that, because evolution is employed to support atheistic conclusions on a vari-
ety of topics, it is often the case that the evidence is exaggerated and the theory co-opted to support
preordained conclusions.
SCIENCE AND NATURALISM
We must first distinguish carefully between science and naturalism. Naturalism in this discussion is
usually defined as the view that all that exists is physical in nature, consisting of some configuration
377
of matter and energy. It is a philosophical thesis about reality.

KEY CONCEPTS Since science is the method for studying the realm of the phys-
ical, then science, on this worldview, becomes the principal way
of trying to understand and explain all of reality. Some well-
Chance
known naturalists of the twentieth and twenty-first centuries
Design are Francis Crick (1995), Carl Sagan (2002), Daniel Dennett
Evolution (1995), Jerry Coyne (2010), and Richard Dawkins (1986).
Moral Evil These thinkers have realized that it is not enough (logically)
simply to criticize a worldview that one does not hold (e.g., reli-
Natural Evil gion) as a way of defending one’s own worldview (naturalism).
Naturalism So, they have turned to scientific theories in an attempt to offer
Problem of Evil positive support for their atheistic conclusions, and especially to
the theory of evolution, thereby giving it an atheistic spin. But
Randomness this move only succeeds in creating confusion in the minds of
Science the general public between science and naturalism, with many
Teleology convinced that modern science is really a form of atheism and
often (unfortunately) adopting a hostile attitude toward it.
Unpredictability
It is essential to remember that naturalism must not be
confused with science. Naturalists will usually appeal to sci-
ence to defend their views, and therefore they place a great
deal of faith in science, but a scientist need not necessarily be a philosophical naturalist, and
indeed most scientists are not naturalists. Science is the study of the physical realm, using scien-
tific method (gathering data, performing experiments, forming and testing hypotheses, etc.), and
makes no claims about the whole of reality. Moreover, most scientists do not believe that all that
exists is physical and that science can explain all of reality (including the origin of the universe,
the scientific laws that govern the universe, along with human phenomena such as consciousness,
reason, the freedom of the will, the moral order). For as soon as one goes beyond the scientific evi-
dence and makes a claim about the ultimate origin of the universe or about the nature of human
beings or about the correct moral theory, one is crossing the line from science proper and moving
into philosophy/religion, and the general area of worldviews. Put another way, one would have
moved from methodological naturalism (considering only physical, testable explanations when
studying the physical universe) to metaphysical naturalism (the position that the only possible
explanations for any aspect of reality must be scientific ones). When naturalists appeal to science,
there is a real danger of confusing the lines of demarcation that must exist between science and
philosophy, and so of pressing science into the service of atheism.
When we keep this important distinction in mind and consider evolution primarily as a scien-
tific theory, it allows for a better understanding of what evolution entails and for a more objective
analysis of its significance in the debate concerning theism and atheism. It is often claimed that
the theory of evolution undermines the religious view of reality in at least three ways. First, it is said
to undermine revealed texts such as the Bible because the scientific evidence that supports the the-
ory of evolution seems to contradict the biblical creation story in the book of Genesis, which cannot
therefore be true. Second, there is the argument that the process of evolution appears to be signifi-
cantly governed by chance, which would appear to undermine the presence of teleology, or purpose
in nature. If it is true that it is largely a matter of chance whether the human species, for instance,
arose at all from the process of evolution (not to mention the contingency of its structure, including
its defining qualities such as consciousness, reason, and moral agency), then it is unlikely that there
is a divine mind guiding the process. Third, it is often contended that evolution contributes signifi-
cantly to the strength of one of the traditional arguments against the existence of God, the problem
of evil, because it shows that this problem is even worse than originally thought. The brutal struggle
for survival that goes on in evolution reveals that there is a parasitical and predatory battle at the
very heart of nature, in which species live off each other and the survival of some is at the expense
of others. It is hard to argue, critics contend, that there can be any underlying purpose for what
appears to be large-scale gratuitous evil (Kitcher 2007, 123ff).
EVOLUTION AND CHRISTIAN REVELATION
These arguments are interesting and must be taken seriously by the theist. However, they are not
nearly as convincing as some seem to think. If one approaches the topic with an open mind, while
these concerns might give one pause with regard to some matters, they are far from establishing a
strong case for atheism. Let us turn first to the argument that evolution undermines biblical revela-
tion (for the purposes of this discussion we will confine ourselves to the Christian religion). Long
before the theory of evolution came along, many theologians and philosophers held the position
that some parts of the Bible might be read as metaphors intended to convey deeper theological
truths and that other parts should be understood more literally, intended to convey both factual
and theological truths. Developing a consistent theory of biblical interpretation involved, among
other things, taking a stand on these matters.
The most developed view on this question came from Saint Augustine, in his “On Genesis: A
Refutation of the Manichees” (2004). Influenced by the argument that God is outside of time,
Augustine proposed that the biblical story of creation may be read as a series of metaphors intended
to convey deeper points that are the real focus of the narrative. These foundational points include
the claims that God created the universe, that God created all living things, and that human beings
differ in kind and not just in degree from other species. These truths, Augustine held, are not revis-
able, but the manner in which God brought all of this about is a question that has not been dis-
closed by divine revelation and is a legitimate field for study, and so is revisable.
Augustine developed this view by appeal to the Stoic idea of the rationes seminales (seedlike
principles) that were envisaged as present from the beginning of creation. Each of them, he said
in “The Literal Meaning of Genesis” (2004), contained the potential for the later development of
a specific living kind. This view has been further developed by Ernan McMullin (2004), who argues
that these seedlike principles might have contained in them the potentialities for all the living kinds
that would later appear, and so there would be no need for any kind of special intervention in
nature by God (as argued for by the intelligent design theorists). McMullin further suggests that
such a view could be elaborated by appeal to the theory of evolution, so as to respect both the find-
ings of the natural sciences and the deepest insights of the Christian theology of creation. But the
general point is that evolution would undermine perhaps only a strict literal interpretation of Gene-
sis that reads the text as if it were making scientific claims, whereas well-established theological read-
ings of Genesis would be unaffected by the theory.
EVOLUTION AND CHANCE
There is considerable confusion surrounding the notion of chance in science and particularly in evo-
lution. Indeed, we often see chance and randomness confused with each other, and randomness
confused with unpredictability. My own view is that, from the point of view of scientific causation
(and excluding such factors as free actions by human beings and any intervention by God in the
universe), we live in a completely deterministic universe (Sweetman 2015). However, whether or
not one accepts this deterministic thesis is not crucial for an effective response to the argument that,
because evolution operates with a significant element of chance, it would undermine the presence of
design in the universe. In order to elaborate and understand these concepts and arguments a bit
more fully we need to say more about the difficulties that chance and evolution are supposed to
generate.
In their explanatory accounts of how evolution works, evolutionary biologists frequently appeal
to notions of chance and randomness. For many thinkers, especially those under the sway of natu-
ralism, the notion of chance is a vital aspect of the evolutionary process because it seems to under-
mine the view that there is purpose in evolution, and in nature more generally. One thinker whose
work did much to give credence to the claim that evolution operates by chance, and to make it part
of the way the theory is currently explained and taught, is Stephen J. Gould, who introduced what
became a well-known metaphor: the notion of “replaying the tape of life.” Gould proposes that we

imagine the whole of evolutionary history as playing a tape and then suggests that, if we were to
rewind the tape and play it a second time, making sure first to erase everything that actually hap-
pened, evolution would take a different path and that almost certainly there would be no species
of Homo sapiens. He is suggesting that history, and causal happenings, could have gone differently
than they did in fact go, and so there is a large element of chance involved in natural events. Gould
calls this random aspect of evolutionary development “contingency,” and argues that the evolution-
ary process exhibits contingency, not necessity (Gould 1989, 51, 309ff; see also Ayala 2004, 55–80;
Miller 2007, 232–239; Peacocke 1993, 64–65).
Dawkins has also contributed to this discussion by focusing mainly on the notion of randomness,
which he believes affects the mutations that occur in the development of DNA in various life-forms
during the evolutionary process. The term “mutation” describes changes in the DNA structure of
organisms that are then passed on to offspring in the process of reproduction. Mutations can be
advantageous, detrimental, or neutral with regard to the success or “fitness” of an organism. Although
many in biology describe these mutations as occurring randomly, Dawkins acknowledges that this ran-
domness cannot mean that they have no causes, that they “just happen.” He believes that the best way
of understanding this claim about randomness is to say that any changes in DNA do not occur in
anticipation of what would make life better for an organism or animal. The process of natural selec-
tion simply works with whatever mutations occur, and so it is a matter of luck or chance that some
features of an organism end up conferring an advantage on it. There is no designer involved, no initial
causal forces guiding nature toward one particular end result rather than another (Dawkins 1986,
305). On this point, Dawkins is in fundamental agreement with Gould. Elliott Sober makes the point
in a slightly different way. He notes that, when various random changes in mutations, which then lead
to changes in varieties and in species, occur, there is no known physical mechanism that detects which
mutations would be beneficial to an organism and that then causes these mutations to occur (Sober
2011, 191). This definition is an attempt to capture Sober’s belief that the effect of a change in
DNA on an organism is a matter of chance. (He does acknowledge, however, that we cannot rule
out that there could be a nonphysical cause for mutations.)
So in this discussion, chance refers to the fact that the causes of events could have been differ-
ent, and so the events might not have occurred (and so history might have gone differently); and
randomness refers to the fact that mutations that occur in organisms do so without any regard to
an organism’s fitness to survive in its environment. They are closely related notions: it is because
chance plays a role in causation that events can be said to occur in a random way, according to this
understanding of how nature behaves; that is, it is because causal events could have gone differently
that changes in an organism’s development (through, for example, mutation) can be said to have no
intended beneficial effect on its ability to survive, or adapt to its environment.
These are interesting arguments, but they do not establish the conclusion that there is no
design in evolution or, more generally, in the universe. The first point to make in response is that
we can make a good case that from the point of view of scientific explanation we live in a determin-
istic universe. This means that there is no chance operating anywhere in nature. As we have seen,
Dawkins and Gould understand chance to mean not that evolutionary change has no cause, but
that the causes could have gone differently than they did in fact go. As Gould notes, some other event
could easily have occurred along the way that would have altered the subsequent path of the chain
of evolutionary causes, and then we would have gotten a different result (Gould 1989, 51ff). Or in
Dawkins’s approach, a different causal chain could have occurred that would then have caused a dif-
ferent mutation, which would then have led to a different outcome in an organism. However, the
problem with this position is that it is not true that the cause (or causes) of any event in the universe
could have been otherwise that they in fact were. And if a cause could not have been otherwise, this
means that the end result of a causal chain could not have been otherwise either. In short, we are
looking at a completely deterministic universe in the realm of the laws of physics and science (leav-
ing out human freedom of will and any action by God in the world).
The deterministic nature of the universe is suggested by several considerations. An important
one is the remarkable fact that our universe operates according to the laws of science (sometimes
referred to as laws of physics or of nature). Our work in the various sciences shows us that there
are “regularities” that occur in nature—empirically detectable patterns in how nature behaves. Such
regularities have been discovered and established as the laws of science. Examples include Newton’s
first law of motion (which states that for every action on an object there is an equal and opposite
reaction, a law utilized in such events as the operation of thrusters on NASA’s space shuttle
engines); Kepler’s laws of planetary motion; the laws of chemistry based on the elements in the peri-
odic table; and many others (on which all scientific predictions are based). So, given these laws we
need to ask if there is anything that happens in nature, or indeed in the physical universe as a whole,
that is not governed by the laws of physics. The answer in modern physics at the practical level of
doing science is generally no, especially at the atomic level (I have argued elsewhere that this is also
true at the subatomic level [Sweetman 2015, 141–145]). What this means is that whenever any-
thing happens in the universe, there is a cause for it, and the cause operates according to, or obeys,
or can be explained in terms of, the laws of science. But it follows that there is, then, an underlying
determinism in how nature behaves.
Determinism may be understood as the view that every present state of the universe is caused
directly (or is completely determined) by prior states together with the laws of physics. This means,
for example, that the formation of the Grand Canyon has been completely brought about by its
chain of prior causes, and so the Grand Canyon could not have been otherwise, and had to occur.
An implication of determinism is that there are no chance events in the universe. This means that
nothing occurs without a cause, of course, but it also means that the causes that bring about events
could not have been otherwise, and so the events had to happen. Determinism changes our understand-
ing of causal happenings in nature because, if the causes could not have been otherwise, then the
end results must come about, and (as we will see below) this has great implications for the question
of teleology in nature. It is also essential to point out that this is also true of the interaction of causal
chains such as one would find in the formation of the Grand Canyon—for example, the coming
together of various gases in a specific type of atmosphere that would later contribute to the forma-
tion of glaciers on the earth’s surface. We must also be careful to maintain a clear distinction
between (1) how the universe obeys laws and acts in itself and (2) our attempts to predict how it will
act. Our inability to predict with perfect accuracy how things will turn out—that is, to say which
causes will lead to which effects—must not be confused with the position that nature is unpredict-
able in itself (indeterminate in itself). This confusion is very common in the general discussion
about chance in the universe. But, given the initial ingredients of the universe and the laws that gov-
ern their interactions, all of the events in the universe had to happen, and this then inevitably
prompts the question: who set up the initial ingredients and why?
One need not be a determinist, however, to question the significance of chance in evolution.
Even if one were to hold that biology, and even physics and chemistry, operate with some element
of chance or indeterminism, it still would not follow that there is no purpose in the universe. For, as
theologians John Polkinghorne and Arthur Peacocke, among others, have argued, it could be the
case that God has allowed some element of chance into creation and yet still directs the final out-
comes. Polkinghorne holds that there could be quite a bit of chance involved: “The fruitful history
of an evolving cosmos has involved an interplay between [chance and necessity], not only in biolog-
ical evolution on Earth but also in the physical development of the universe itself” (1998, 39). Nev-
ertheless, he thinks that God is directing a plan in an overall sense because the ingredients of life are
built into the universe from the beginning, and he believes that some kind of conscious life must
inevitably emerge, though he does not think it would necessarily have to be Homo sapiens. Develop-
ing a similar argument, Peacocke (1993, 49ff) holds that the behavior of reality itself at the sub-
atomic level involves some elements of chance, and not just unpredictability (a view I reject above).
Yet, despite this, he suggests that many effects emerge from within systems which consist of their
own structures, environments and ingredients, together with the laws of science, all interacting with
each other. Given a particular system, there is a range of possibilities that can come about in creation.
But the range is limited, according to Peacocke, citing the work of such figures as John Maynard
Smith and Karl Popper. So, even if there were an element of chance involved and we could not pre-
dict the outcomes of any system with complete accuracy, it would still be the case that, because the
system constrains what can emerge, there would be reason to say it is moving toward a goal, as it

were. So the universe contains a combination of chance and law, and the outcomes may still be
directed by an intelligence.
Although the views of Polkinghorne and Peacocke are a bit vague and not without problems,
many accept their general point that, although there could be an element of chance in creation,
there is clearly an overall purpose because of the obvious complexity and progress in nature gener-
ally, and in evolution in particular. In fact many naturalistic interpretations of evolution struggle
to explain why the process leads to increasing complexity and progress in species like that of Homo
sapiens if their emergence is significantly governed by chance. This is why so many look at the
nature of human life and doubt the current evolutionary account of it, especially the claims about
randomness, the fact that events happen in nature without regard to the eventual outcomes (or,
in evolution, to the “fitness” of the organism). If nature is moving toward unusual and suggestive
complexity, this is a good reason to think that it was set up this way, which suggests a designer.
SUFFERING AND EVIL IN NATURE
Let us turn to consider how the absence of purpose in evolution is also supposed to strengthen the
“problem of evil argument” against the existence of God (this will also give us an opportunity to
develop further the argument just made). The general argument from evil (treated elsewhere in this
volume) is based on the reasoning that the pain and suffering evident in nature make it unlikely that
the theistic position is true, namely, that God exists—that is, a God who is all-powerful, all-good,
all-knowing, and so on. The theory of evolution, some suggest, strengthens this argument because
the process of evolution reveals that there is a great deal of waste in nature, and that the existence
of evil is on a much larger scale than we had realized. It reveals also that much of this evil occurs
by chance; it just happens randomly with no rhyme or reason, and could just as easily not have
occurred, and so there seems to be no overall purpose in nature. The evil in nature revealed by evo-
lution would include the large number of extinctions that have occurred since the process began, as
well as the general predatory nature of life both at the plant and animal levels. These types of evil
are often referred to as “natural evil” by philosophers because they occur naturally and can be distin-
guished from “moral evil,” which describes evil acts carried out by free human beings. Evolution
shows that there is a struggle for survival going on in nature, often in situations of limited resources,
and that many species and individuals live off of other species and individuals. Charles Darwin, for
instance, noted the way in which the ichneumon wasp parasitically feeds off the bodies of caterpil-
lars in its attempt to survive, a case that made him question the idea that nature has a purpose and is
created by a benevolent deity (see Gould 1983, 32–45; also Kitcher 2007, 123ff). The predatory,
competitive battle constantly going on among living things leads to hardship, suffering, death, and
extinction, and all of this is said to be an amount of natural evil that is inconsistent with there being
a benevolent, omnipotent, and omniscient God.
Another consideration here is the question of imperfect designs in nature. It is often claimed
that, because the process of natural selection is random, haphazard, and affected by so many vari-
ables, we end up with many adaptations in nature that are less than optimum for the survival of spe-
cies or individuals. Examples of “bad designs” include the blind spot in the eye (which is caused by
the optic nerve’s running through a hole in the eye, according to Guttman 2005, 10–11). Some
argue that in our breathing apparatus the trachea are on the wrong side of the esophagus (Brockman
2006). If God had designed nature, the argument goes, we would not have imperfections of this
type—imperfections that cause suffering and hardship, problems with survival, and so forth.
One reply that the theist may make to the point that there is no underlying rationale for the
natural evil is the deterministic argument I suggested earlier, namely, that the absence of a rationale
only seems to follow if it is true that evolution operates by chance. It is not true if all of the events
that happen in nature are determined by the existence and the specific type of initial conditions and
ingredients (together with the laws of science). This is because the determinism that we have been
describing suggests an underlying purpose in the universe after all—a type of order or design to
which the complex evolutionary process is moving. Therefore what happens in nature, including
natural evil in the sense of the harm, suffering, and destruction of some species along the way, may
be an intended part of this purpose. We may not be able to discern the purpose completely, but
that is simply the result of our status in time as inquirers who are striving to understand the process;
the discussion brings us back to the question of who started it all off, and why. And so it is a very
interesting point that evolution gave rise to complex, conscious, intelligent observers who have some
control over the process. If there is no chance involved in evolution, Gould would be wrong to say
that, because the existence of every species is due to chance, no particular species had to come into
existence, and Dawkins would be wrong that the final structure an organism ends up with is due to
chance (not even to mention the exceptional qualities of Homo sapiens—reason, logic, freedom of
will, and moral agency—that create very difficult problems for evolutionary explanations), and
Sober would be wrong that there is no mechanism driving the process of evolution. This latter is
because deterministic causal chains may themselves be regarded as physical mechanisms that must lead
to their outcomes.
The naturalist may press the point that we are then saying that all the evil in nature is part of
God’s purpose. But this objection misses the mark, for this is the theistic position before considering
the theory of evolution. It is why we are able to generate the problem of evil in the first place. It
simply leaves us back where we started—with regard to the problem of evil, but this time after evo-
lution, and so we can say that evolution makes no substantial difference to the nature of the prob-
lem. The problem concerning what God’s purpose may be in allowing evil is difficult to solve, and
it is not my aim to solve it here; the argument that I am trying to refute is not that there is no prob-
lem of evil but that neither determinism about nature nor the claim that evolution operates by
chance makes the problem of evil worse. Our discussion of the work of Polkinghorne and Peacocke
showed that even some element of chance, as long as it is a part of larger law-guided systems, can
produce complexity and order. This suggests that there is purpose in the universe, and so gives us
some reason to think that evil has some place within this larger purpose, even if we may find it dif-
ficult to discern what that purpose is. (For some responses to the general problem of evil, see Augus-
tine 2003; Lewis 1962; Hick 1966; Murray 2011.) We must also remember that many of the spe-
cific claims about how changes in nature came about through the evolutionary process are based on
a certain kind of guesswork, such as the number of species that became extinct. There are so many
“just so” stories used to explain the path of evolution in particular cases that we should be wary of
accepting these speculations as the basis for taking them to generate some new version of the prob-
lem of evil. To claim that scientific evidence alone allows us to conclude that the existence, the
nature, or the life span of any species (or member of a species) is a “waste” in any clear sense of that
term is highly controversial. The very fact that an organism exists means that it should not be
regarded as a waste, and this would be especially true from God’s point of view.
Similar points can be made in response to the argument considered above concerning certain
imperfections in nature, such as the blind spot of the eye. For the sake of argument, let’s assume
that the blind spot originated as a by-product of inefficient, random evolutionary processes (i.e.,
processes that are not aimed at any particular goal or end result). This fact about the blind spot
seems similar to saying that the eye is not perfect because our vision fades as we age, or that hip
joints are not perfect because they wear out, or, more generally, that the body is not perfect because
it is subject to disease. These are all problematic features of nature that force us to consider the fact
of evil, but the evidence from evolution does not seem to make this problem any worse than before
the theory came along. We already knew that nature was blemished with flaws, limitations, ineffi-
ciencies, and other problems that contribute to the problem of natural evil. Evolution may reveal
a few more of these flaws, but this seems to be akin to the discovery of new diseases rather than fun-
damentally changing the nature of the problem with which we are concerned. Imperfections in
nature do not seem to alter our understanding of the problem of evil in a fundamental way.
EVOLUTION, DESIGN, AND COMPLEXITY
In this context, it is often argued that the theory of evolution is an effective response to the argu-
ment from design advanced by the eighteenth-century thinker William Paley (2010), which

appealed mainly to organized patterns in nature, patterns for which it was later thought Darwin’s
theory provided a natural explanation. But Paley was also making a more general argument. In his
famous watch example, he noted that the conclusion that an object so intricate and so complex
must be designed is not lessened if one also recognizes imperfections (or design flaws) in the watch;
indeed, the evidence that the watch is designed is still indisputable. I have no doubt that this would
be Paley’s response to the theory of evolution. Paley’s general point was that what needs to be
explained is the emergence in nature of complex species, culminating in the arrival of conscious
observers, who have consciousness, rationality, and moral agency. Such beings have features that
cannot be explained without recourse to the idea that they were designed for a particular purpose.
It is this complexity that makes us doubt that the process can be governed entirely by chance—
because the evidence from nature (and today from evolution itself) is against it. This is why it is cru-
cial to appreciate that if one wishes to claim that the complexity can be explained by natural selec-
tion operating without any guidance, one must show two things together: how the process operates
by chance, and how it can produce complex, machinelike organs and species. Promissory notes with
regard to these crucial questions are only a way of avoiding the difficulty.
Sometimes secularists will appeal to evolution to explain all traits and characteristics of human
beings, including not just obvious remarkable qualities such as consciousness and reason but also
phenomena such as morality and religion (as indeed they must if all such features originated by
means of a chance process). The moral nature of human beings, as well as the fact that the vast
majority of human beings both now and in the past have been religious believers, suggest both that
human beings are naturally religious and also that there is an objective moral order. These two
points are further evidence of an overall purpose, and therefore of a designer. The enormous contri-
bution of religion to the development of human civilization is also something we must take into
account when appraising these issues. Secularism or evolutionary naturalism is basically an untested
view of reality and of the meaning of life, and if we are to be honest I think we must acknowledge
that the early signs are not promising. The history of secularism in the twentieth century, in the
regimes of Adolf Hitler, Joseph Stalin, and Pol Pot, is not reassuring, as many thinkers have noted.
Moreover, secularism seems to contain destructive forces that threaten the happiness of the individ-
ual, and that often lead to alienation, despair, and loss of meaning. It is simply a fact that human
beings have found it very difficult to live meaningful and fulfilled lives in the absence of a religious
worldview. A purely secular evolutionary explanation has great difficulties with these facts. This by
itself would not justify the religious worldview (perhaps), but it is yet another argument in its favor.
One may make this same point in a more positive way, inspired by Saint Thomas Aquinas
(1998), among other great thinkers. Many philosophers, including Aquinas, held that it is perfectly
reasonable to argue that human life has an overall intelligibility and meaning, and therefore, by
extension, purpose, and that this purpose cannot be accounted for by any atheistic theory or
approach. Such considerations also played a leading role in Immanuel Kant’s version of the moral
argument for God’s existence. We should not overlook the fact that evolution is a “survival of the
fittest” theory; a species that has a selective advantage over others (relative to the environment) is
more likely to survive, and to thrive. So if, for instance, one were to attempt to explain morality
along evolutionary lines, one may have to argue that morally bad behaviors must have had a selec-
tive advantage for some of our ancestors in their particular context. Selfish behavior on the part of
an individual would be more likely to lead to survival and success than altruistic behavior. From a
purely evolutionary naturalistic view of evolution, one can argue that our moral values are just
another tool, like our limbs or our intelligence, in our set of survival skills, and indeed secularists
Michael Ruse and Edward O. Wilson argue as much. Ruse has argued that Darwinian theory shows
that, in fact, morality is a function of (subjective) feelings; but it shows also that we have (and must
have) “an illusion of objectivity.” He suggests that morality is “a collective illusion foisted upon us
by our genes,” and can perhaps be “explained” as an evolutionary adaptation (Ruse 1998, 253;
see also Wilson 1975). Dawkins agrees that morality has no absolute basis from an evolutionary
point of view (and it is a form of enlightenment to realize this) (1995, 131–133).
Ruse has attempted to respond to the difficulties in justifying objective morality and purpose in
life by noting that we could no more get rid of objective morality than we would our eyes. Yet it
does not follow from this that we could not abandon objective moral values (i.e., basically act like
moral relativists), when it would seem to advance our own development and survival; nor does it
even seem to follow that we could not get rid of objective morality altogether. Indeed, thinkers like
Dennett (2006) and Dawkins (2006) have stridently tried to explain religious beliefs this way, argu-
ing that such beliefs evolved and survived because they must have given humans some adaptive
advantage in the struggle for survival. Why not offer the same explanation for moral beliefs and
draw the same conclusion—that they are false, and we should get rid of them? These evolutionary
accounts and responses, of course, are speculative, based on little or no direct evidence, and are used
to prop up a troubled secularist worldview. However, the considerable difficulties that afflict a secu-
larist interpretation of evolution are among the reasons that religious believers hold that religious
belief and evolution are not only compatible but that evolution considered as a scientific theory is
no threat to the religious view of reality.
Bibliography
Augustine, Saint. Concerning the City of God against the Pagans. Translated Guttman, Burton S. Evolution: A Beginner’s Guide. Oxford: Oneworld,
by Henry Bettenson. London: Penguin, 2003. 2005.
Augustine, Saint. On Genesis: A Refutation of the Manichees, Unfinished Haught, John F. God after Darwin: A Theology of Evolution. Boulder, CO:
Literal Commentary on Genesis, the Literal Meaning of Genesis. Edited by Westview Press, 2000.
Boniface Ramsey. New York: New City Press, 2004.
Hick, John. Evil and the God of Love. New York: Harper and Row, 1966.
Ayala, Francisco. “Design without Designer: Darwin’s Greatest Discovery.” Kitcher, Philip. Living with Darwin: Evolution, Design, and the Future of
In Debating Design: From Darwin to DNA, edited by William A. Faith. Oxford: Oxford University Press, 2007.
Dembski and Michael Ruse, 55–80. New York: Cambridge University
Press, 2004. Lewis, C. S. The Problem of Pain. New York: Macmillan, 1962.
Bartholomew, David J. God, Chance, and Purpose: Can God Have It Both McGrath, Alister. Dawkins’ God: Genes, Memes, and the Meaning of Life.
Ways? Cambridge, UK: Cambridge University Press, 2008. Oxford: Blackwell, 2005.
Brockman, John, ed. Intelligent Thought: Science versus the Intelligent Design McMullin, Ernan, ed. Evolution and Creation. Notre Dame, IN: University
Movement. New York: Vintage, 2006. of Notre Dame, 1985.
McMullin, Ernan. “Evolution as a Christian Theme.” Reynolds Lecture, Baylor
Coyne, Jerry A. Why Evolution Is True. New York: Penguin, 2010.
University, 2004. https://www.baylor.edu/content/services/document
Crick, Francis. The Astonishing Hypothesis: The Scientific Search for the Soul. .php/36443.pdf.
New York: Touchstone, 1995.
Miller, Kenneth R. Finding Darwin’s God: A Scientist’s Search for Common
Darwin, Charles. The Origin of Species. London: Penguin, 1985. First Ground between God and Evolution. New York: Harper, 2007.
published 1859.
Morris, Simon Conway. Life’s Solution. New York: Cambridge University
Dawkins, Richard. The Blind Watchmaker: Why the Evidence of Press, 2003.
Evolution Reveals a Universe without Design. New York: Norton,
Murphy, Nancey. Bodies and Souls, or Spirited Bodies? Cambridge, UK:
1986.
Dawkins, Richard. River out of Eden: A Darwinian View of Life. New York:
Murray, Michael. Nature Red in Tooth and Claw: Theism and the Problem of
Basic Books, 1995.
Animal Suffering. Oxford: Oxford University Press, 2011.
Dawkins, Richard. The God Delusion. Boston: Houghton Mifflin, 2006. Paley, William. Natural Theology. New York: Oxford University Press, 2010.
Dembski, William A., and Michael Ruse, eds. Debating Design: From Peacocke, Arthur. Theology for a Scientific Age: Being and Becoming—
Darwin to DNA. New York: Cambridge University Press, 2004. Natural, Divine, and Human. Minneapolis, MN: Fortress, 1993.
Dennett, Daniel. Darwin’s Dangerous Idea. New York: Simon and Schuster, Petrik, James. Evil beyond Belief. Armonk, NY: M. E. Sharpe, 2000.
1995.
Plantinga, Alvin. Where the Conflict Really Lies: Science, Religion, and
Dennett, Daniel. Breaking the Spell: Religion as a Natural Phenomenon. New Naturalism. New York: Oxford University Press, 2011.
York: Viking, 2006.
Polkinghorne, John. Science and Theology: An Introduction. Minneapolis,
Goetz, Stewart, and Charles Taliaferro. Naturalism. Grand Rapids, MI: MN: Fortress, 1998.
Eerdmans, 2008.
Ruse, Michael. Taking Darwin Seriously: A Naturalistic Approach to
Gould, Stephen J. Hen’s Teeth and Horse’s Toes. New York: Norton, 1983. Philosophy. Amherst, NY: Prometheus, 1998.
Gould, Stephen J. Wonderful Life: The Burgess Shale and the Nature of Sagan, Carl. Dragons of Eden: Speculations on the Evolution of Human
History. New York: Norton, 1989. Intelligence. New York: Ballantine, 1986.

Sagan, Carl. Cosmos. New York: Random House, 2002. Sweetman, Brendan. Evolution, Chance, and God: Understanding the Relation-
ship between Evolution and Religion. New York: Bloomsbury, 2015.
Schönborn, Christoph. Chance or Purpose?: Creation, Evolution, and a
Rational Faith. Edited by Hubert Philip Weber. Translated by Henry Thomas Aquinas, Saint. Selected Writings. Edited and translated by Ralph
Taylor. San Francisco: Ignatius, 2007. McInerny. New York: Penguin, 1998.
Ward, Keith. God, Chance and Necessity. Oxford: Oneworld, 1996.
Sober, Elliott. “Evolution without Naturalism.” In Oxford Studies in
Philosophy of Religion. Vol. 3, edited by Jonathan Kvanvig, 187–221. Wilson, Edward O. Sociobiology: The New Synthesis. Cambridge, MA:
Oxford: Oxford University Press, 2011. Harvard University Press, 1975.
TOPIC 12
Human History: Atheism

DARWINISM AND HUMAN ORIGINS 387
Michael Ruse
Lucyle T. Werkmeister Professor and Director of HPS Program,
Department of Philosophy
Florida State University
THE EVOLUTIONARY STUDY OF RELIGION 391

Susana Nuccetelli
St. Cloud State University, MN
IS HISTORY GUIDED BY THE HAND OF PROVIDENCE? 394

Keith Parsons
Professor of Philosophy, College of Human Sciences and Humanities
University of Houston–Clear Lake, TX
THE EVOLUTIONARY DOWNSIDE OF MASS THEISM AND 397

THE RISE OF POPULAR ATHEISM
Gregory Paul
Independent Scholar
Baltimore, MD
ACCOUNTING FOR MONOTHEISM’S HISTORIC DISPLACEMENT 400

OF NON-MONOTHEISM WITHOUT A DIVINE HAND
Matthew Wade Ferguson
Doctoral Candidate in Classics
University of California, Irvine
Theism, the belief that the God of Abrahamic religions objectively exists, faces a challenge from evolutionary
accounts of belief in supernatural agencies. It does put the burden of argument on theists, who must argue
persuasively for either the epistemic justification of their fundamental belief or for the inaccuracy of the
evolutionary hypothesis.
DARWINISM AND HUMAN ORIGINS
Michael Ruse, Florida State University
The Darwinian evolutionist’s position with respect to human origins is straightforward (Ruse 2012).
The basic picture is known, as are many of the causes, the chief of which is natural selection.
387
KEY CONCEPTS
Abrahamic Religions Hinduism
Agnosticism Hunter-Gatherers
Axial Age Islam
Buddhism Meaning of Life
Christianity Monotheism
Daoism Moral Realism
Deism Natural Selection
Democracy Origin of Civilization
Divine or Providential Guidance Original Sin
Epistemic Warrant Prophecy
Evil Scientific Revolution
Evolution Skeptics
Existential Anxiety Supernatural Agencies
Folk Religions Truth
Although there are disagreements, no paleoanthropologist doubts that the processes were entirely
natural, in the sense of being governed by laws and no supernatural interventions. This is not a field
in crisis looking for a new paradigm—inside or outside science.
Humankind, Homo sapiens, is a primate species, that started to evolve independently about five or
so million years ago. The animals who are our closest relatives are the chimpanzees, and we are more
closely related to them (98 percent shared DNA) than either species is to other higher primates like
the gorillas. Exactly why we started to go it alone is still a matter of some discussion. It is generally
thought that it was related to climate and the consequent move from a shrinking forest out onto
the plain. (Ruse 2012 discusses some of the huge literature.) New adaptations were needed, and these
included the ability to walk upright—bipedalism—and, increasingly, more brainpower. Our brains
increased from around 400cc to the current 1200cc. It is known from the fossil record that bipedalism
came before the large brain, and there are some intermediate fossils—the most famous being “Lucy,”
Australopithecus afarensis—that walked on two feet and had a brain about the size of a chimp (Johan-
son and Edey 1981). A highly important aspect of human evolution was the move toward sociality.
Overall, a powerful human adaptation is the ability to plan and cooperate.
How far does any of this mesh with Christianity? In On the Origin of Species, Charles Darwin
stressed that his theory did not refute divinity per se and even had things to commend it in that
respect. “To my mind it accords better with what we know of the laws impressed on matter by
the Creator, that the production and extinction of the past and present inhabitants of the world
should have been due to secondary causes, like those determining the birth and death of the individ-
ual” (Darwin 1859, 488). However, this is the god of deism, not the Christian God of theism, who
created and cares for His Creation. The God who made humans in His image, clearly regards us as
something special, and the possibility of redemption goes to the heart of the Christian story. Unlike
God, we are contingent beings; nevertheless, our existence and our nature are not matters of chance.
On the Christian view, we had to occur in some sort of sense (Ruse 2001).
Here there does seem to be a clash with Darwinism, for the essence of the Darwinian process is
nondirectionality, randomness in the sense of not occurring to need. As the paleontologist John J.
Sepkoski noted colorfully, there is nothing special about human intelligence. “I see intelligence as
just one of a variety of adaptations among tetrapods for survival. Running fast in a herd while being
as dumb as shit, I think, is a very good adaptation for survival” (Ruse 1996, 486).
Some obvious theological solutions to this problem—that our existence seems to be contin-
gent rather than necessary—involve God getting into the process directly. Every now and then
God gives things a shove in one direction rather than another. It is open to God to gather up
in His hand the molecules of an existing organism and rearrange them to make a new organ-
ism—fox into dog in one step. Many suspect, however, that God works through directed varia-
tions—“mutations”—that give us dogs in a few generations. Obviously, these options—going
from direct Creationism at one end through various forms of so-called intelligent design theory
and onto the other end positing unseen processes of theist evolution—are unacceptable to the
Darwinian (Dembski and Ruse 2004). They are not natural in the sense of appealing to and only
to unbroken law, and hence they are not real science. Nor does it really help if one suggests that
the direction comes at the quantum level and is thus undetectable (Russell 2008). One is still
going against the spirit of modern science. Not to mention the theological difficulties. If God is
prepared to intervene to create humans, why did He not intervene to make better-quality
humans, for instance without Tay-Sachs disease?
Are there scientific solutions to this problem, solutions that are entirely natural? Three at least
have been proposed. One is through a form of arms race, where organisms compete against each
other and improvement ensues (Huxley 1912). It is often thought—Darwin thought—that intelli-
gence is bound to succeed in such circumstances.
If we look at the differentiation and specialisation of the several organs of each being when
adult (and this will include the advancement of the brain for intellectual purposes) as the best
standard of highness of organisation, natural selection clearly leads towards highness; for all
physiologists admit that the specialisation of organs, inasmuch as they perform in this state
their functions better, is an advantage to each being; and hence the accumulation of variations
tending towards specialisation is within the scope of natural selection. (Darwin 1861, 134)
In recent years, Richard Dawkins (1986) has been a strong proponent of this approach.
A second option focuses on ecological niches; the argument is that they exist independently of
their occupiers, that a culture niche exists, that we have found it, but that if we had not found it,
some other species of organism would almost certainly have succeeded. The paleontologist Simon
Conway Morris writes, “If brains can get big independently and provide a neural machine capable
of handling a highly complex environment, then perhaps there are other parallels, other conver-
gences that drive some groups towards complexity” (Conway Morris 2003, 196). He continues:
“We may be unique, but paradoxically those properties that define our uniqueness can still be inher-
ent in the evolutionary process. In other words, if we humans had not evolved then something
more-or-less identical would have emerged sooner or later.”
The third option, not very Darwinian, lets the processes of physics and chemistry do the heavy
lifting. In the eyes of biologist Daniel McShea and philosopher Robert Brandon, “biology’s first
law”—the “zero-force evolutionary law” (ZFEL)—can do it all: “In any evolutionary system in
which there is variation and heredity, there is a tendency for diversity and complexity to increase,
one that is always present but that may be opposed or augmented by natural selection, other forces,
or constraints acting on diversity or complexity” (McShea and Brandon 2010, 57).
As one might expect, all of these suggestions have their (scientific) critics. With respect to arms
races, we have already seen Sepkoski’s sardonic comment that winning does not necessarily require
or even want human excellence. A simpler solution may often be better. With respect to niches,
there is serious question about whether they exist independently of organisms. Would human body
lice exist and thrive if there were not human hair? The culture niche may never have been occupied
or might have taken a very different form had things gone otherwise. Here is Darwin, again, sug-
gesting that had evolution gone differently, we might have ended with a very different moral code,
one hardly recognizably human:
If, for instance, to take an extreme case, men were reared under precisely the same conditions
as hive-bees, there can hardly be a doubt that our unmarried females would, like the worker-
bees, think it a sacred duty to kill their brothers, and mothers would strive to kill their fertile
daughters; and no one would think of interfering. (Darwin 1871, 73)

As for ZFEL, that really is for those who believe in the tooth fairy—at least, with respect to the
furnishing of any evidence, empirical or theoretical—although there is an interesting idea lurking
here. Presumably, since we humans have evolved, we could in principle do so again. So given time
and space—no big problem for God, since (at least if we believe He exists outside space and time)
He is outside both—if, there were an infinite number of possibilities—perhaps thanks to multi-
verses—then would there not be the chance, the actuality, of something humanlike evolving else-
where? We might perhaps have green skin, but God could surely work with that. What God might
not be able to work with are the side effects. You don’t just have one set of humans; you have many
sets (Ruse 2017). The philosopher J. J. C. (Jack) Smart, a British-born fanatic about cricket,
hypothesized that all over the universe there is an infinite number of teams capable of beating the
Australians! Smart felt even more strongly on the subject of cricket played in colors other than the
traditional white. Presumably, all over the universe there are teams clad in shocking-pink hot pants,
capable of beating the English by an inning and several wickets. Is this a price an All-Perfect God
would think worth paying to get humans in the first place?
What about the problem of evil? Darwin worried that this was exacerbated by his theory.
With respect to the theological view of the question; this is always painful to me.—I am
bewildered.—I had no intention to write atheistically. But I own that I cannot see, as plainly
as others do, & as I shd. wish to do, evidence of design & beneficence on all sides of us. There
seems to me too much misery in the world. I cannot persuade myself that a beneficent &
omnipotent God would have designedly created the Ichneumonidae with the express intention
of their feeding within the living bodies of caterpillars, or that a cat should play with mice. Not
believing this, I see no necessity in the belief that the eye was expressly designed. (Darwin
1985–, 8, 224)
Actually, things are not this bad (Ruse 2001). In some respects, Darwinism is a friend of Chris-
tianity. When one makes the traditional distinction between natural evil and moral evil—the evil of
the Lisbon earthquake versus the evil of Auschwitz—one can argue that natural evil is simply the
consequence of nature’s laws. More than that, if Dawkins (1983) is right that the only natural
way you get design-like features is through natural selection brought on by struggle, one might
argue that natural evil is necessary to get functioning organisms. At least if you believe that God cre-
ated by means of natural causes (rather than by a series of miraculous actions), then he cannot be
blamed for necessity. As far as moral evil is concerned, the Darwinian is not condoning it, but does
agree with St. Paul that we are conflicted on these matters. “For that which I do I allow not: for
what I would, that do I not; but what I hate, that do I” (Romans 7:15). We have evolved as social
beings, but, beneath it all, our selfish genes are at play. We are bound to be torn in different ways
and sometimes do that which we judge wrong.
There is a major problem: original sin. Christians in the Augustinian tradition think we sin
because of the disobedience of Adam. We are tainted and can be saved only through the blood of
the Lamb (Ruse 2018). But if Darwin is right, Adam and Eve did not exist, and at any time in
our prehistory, not only were our ancestors as conflicted as we, but so were their parents and grand-
parents. Darwinism says that if you reject Adam and Eve (as you must), then, in order to go on
being a Christian, you have to give up atonement theory and argue for something like an Incarna-
tional theology, where Jesus is an example and where we see God’s creation developmentally with
humans as part of this process. We are not created complete and perfect.
Finally, what of the soul? Darwinism accepts that consciousness evolved, although whether a
follower of Darwin will have any more to say than anyone else on the relationship between body
and mind is another question. Some evolutionists think that the theory pushes one toward some
form of panpsychism, where in a kind of Spinozistic fashion mind is all-pervasive. In the nineteenth
century, mathematician and philosopher W. K. Clifford wrote:
We cannot suppose that so enormous a jump from one creature to another should have
occurred at any point in the process of evolution as the introduction of a fact entirely different
and absolutely separate from the physical fact. It is impossible for anybody to point out the
particular place in the line of descent where that event can be supposed to have taken place.
The only thing that we can come to, if we accept the doctrine of evolution at all, is that even in
the very lowest organism, even in the Amoeba which swims about in our own blood, there is
something or other, inconceivably simple to us, which is of the same nature with our own
consciousness, although not of the same complexity. (Clifford 1879, 53)
No one is now saying that plants think but rather that mind is not a thing apart. All of creation
is filled with potential. None of this is anti-Christian, especially given that God is creator of all and
that He is immanent always throughout His creation. The heresy would be to identify mind with
God, perhaps in the way of the Gaia hypothesis identifying our planet with a living organism that,
one presumes, has some cognitive abilities (Ruse 2013).
The extent to which the natural phenomenon of mind is to be identified with or connected to
the theological notion of the soul is obviously another matter. They cannot be identical because
newborn babies, hardly thinking beings, have souls. Pope John Paul II took the whole question of
souls out of science and left at that. “If the origin of the human body comes through living matter
which existed previously, the spiritual soul is created directly by God” (John Paul II 1997, 5).
Someone like Dawkins will not find this satisfactory, but the question to be asked is whether this
dissatisfaction comes from Darwin’s theory or from more general theological and philosophical wor-
ries about souls. One might, for example, reject the kind of immortal souls the pope is speaking of
because one cannot see how they can transcend death and become eternal. But that is not a conclu-
sion derived from Darwinism. Perhaps it is best to say that if Christians want to believe in souls,
that is their business. It doesn’t have to be the business of the rest of us.
Does Darwinian thinking on human evolution refute Christianity? No, but it is more of a chal-
lenge than many Christians realize.
THE EVOLUTIONARY STUDY OF RELIGION
Susana Nuccetelli, St. Cloud State University
The science of religion aims at producing an empirically sound account of the cause of the develop-
ment of religion among humans. Evolutionary religious studies, a part of this thriving interdisciplin-
ary field, is devoted entirely to providing an account of the evolutionary origin of religious beliefs
and practices. The question for us here is what implications, if any, may the evolutionary genealogy
of these have fo theism? Theism is a form of realism about God, a supernatural being with the qual-
ities standardly ascribed to Him in the Abrahamic religions. Given theism, the belief that God
objectively exists is true in a manner completely independent of any human responses to it.
Evolutionary religious studies’ findings about the etiology of religious belief may or may not
undermine theism, depending on some considerations about epistemic warrant. These are relevant
because faulty processes in the origin of a belief cannot by themselves undermine its warrant. Let’s
illustrate with someone’s belief that the likelihood of developing Gaucher’s disease for Ashkenazic Jews
is greater than average among ethnic groups. Suppose the believer is persuaded about this on the basis
of a bias against Ashkenazic Jews. In this case, the faulty way in which this belief was formed fails to
undermine its epistemic justification, for there is plenty of comparative genetic data warranting that
belief. Examples of this sort call for caution about rejecting theism solely on the basis of the findings
of evolutionary religious studies. But, as discussed below, in the absence of independent reasons for
theism, given those findings, the warrant of religious belief would be undermined. After all, the fun-
damental premises of evolutionary religious studies have overwhelming support within the various sci-
entific fields that contribute to the science of religion, which range from biology, neurobiology, and
game theory, to scientific psychology and anthropology. According to those premises, evolutionary
pressures are responsible for the evolved tendency in humans to entertain and spread belief in super-
natural agency. Whether the cognitive and emotional processes that trigger that tendency are adapta-
tions that have evolved to fulfill certain functions or are simply by-products of mechanisms that have
evolved earlier to be functional need not concern us here. However this disagreement gets resolved,
the consensus in evolutionary religious studies supports the view that religious belief plays a role in
promoting reproductive success and survival (Bulbulia et al. 2008; Nola 2013).

This hypothesis squares well with Charles Darwin’s views on the origin of religion, especially as
presented in the section “Belief in God—Religion” of his Descent of Man. There Darwin speculates
that humans appear to have a universal tendency to develop belief in supernatural agency, which with
time will “easily pass into the belief in the existence of one or more gods” (Darwin 1981, 67). As to
why humans might have felt the need to postulate supernatural agencies, Darwin hypothesizes that
our ancestors “would naturally have craved to understand what was passing around him, and have
vaguely speculated on his own existence.” According to Darwin, not only humans but any beings with
some imagination, wonder, curiosity, and power of reasoning can have this tendency to project super-
natural agencies onto the world. His dog, for example, seems to have had it since Darwin has observed
it to bark at the movements of a parasol with the wind, as if a strange living agent was causing that
movement. Cognitive psychologists, however, might doubt that a dog or even a chimp could go on
to develop belief in supernatural agencies, for these require having among other cognitive abilities what
is known as “a theory of the mind” that allows ascription of beliefs, desires, and other intentional
states to others. Nonhuman animals appear to lack that ability (Bering 2002). But if we limit Dar-
win’s hypothesis to humans, it might be on track. Current evolutionary accounts of theism associate
the origin of belief in supernatural agencies with a number of evolutionary functions.
RELIGIOUS BELIEF’S ETIOLOGY

Evolutionary religious studies attempts to account for the human tendency to form and maintain
supernaturalist beliefs by associating these with a number of Darwinian functions. For instance,
evolutionary-minded philosopher Philip Kitcher agrees with Darwin: supernaturalist religions
respond to a human need to answer fundamental questions concerning the meaning and value of
life. Thus he notes:
There is truth in Marx’s dictum that religion, more precisely supernaturalist and providentialist
religion, is the opium of the people, but the consumption should be seen as medical rather than
recreational. The most ardent apostles of science and reason recommend immediate withdrawal
of the drug—but they do not acknowledge the pain that would be left unpalliated, pain too
intense for their stark atheism to be a viable solution. (Kitcher 2007, 165)
In a parallel account of the function of religion, cognitive anthropologist and psychologist Scott Atran
(2002) associates humans’ evolved tendency to have and spread belief in the existence of God with the
promotion of positive feelings. These can help humans cope with existential anxieties created by such
factors as the inevitability of death and the constant threat of deception by others. Religious belief
might then have some health benefits, thus promoting reproduction and survival. Other candidates
within evolutionary accounts of the origin of religion focus on its contribution to cooperation, social
cohesion, and group identity. Proponents of these accounts regard religion as a response to problems
of cooperation that originally afflicted our hominid ancestors. But evolutionary religious studies need
not focus exclusively on the infancy of religions. It may also devote itself to investigating the spread
of religions as a process that coevolves with increases in social complexity (Purzycki et al. 2018).
EVOLUTIONARY DEBUNKING ARGUMENTS

Either way, evolutionary religious studies takes theism and other beliefs in supernatural agencies to
be products of evolution, a hypothesis that by itself is not enough to support skepticism about
beliefs of this sort. For quite likely all of our beliefs have been shaped by evolution, including math-
ematical beliefs and empirical beliefs such as those central to evolutionary theory. Therefore, evolu-
tionary debunking arguments against theism must avoid premises that entail a global type of skep-
ticism.
But these arguments can easily do that by drawing a disanalogy between the evolved mechan-
isms responsible for mathematical and perceptual beliefs, and those responsible for religious beliefs.
After all, it is only in the case of the former beliefs that success in reproduction and survival requires
that they track the truth. Our hunter-gatherer ancestors might have gone extinct had they systemat-
ically failed to count the number of predators before them or discriminate between poisonous and
edible mushrooms growing nearby.
By contrast, beliefs in the existence of supernatural agents might contribute to survival and
reproductive success in the ways suggested above even if they were all false. The evolved mechan-
isms responsible for these beliefs need not track the truth. Consider theism, or the belief in the
mind-independent existence of the God of the major monotheistic traditions. If some hypothesis
widely accepted in the science of religion is correct, theism results from a psychological mechanism
that has evolved to fulfill the crucial cognitive function of detecting predators.
Part of the reason people believe in gods, ghosts, and goblins also comes from the way in
which our minds, particularly our agency detection device (ADD) functions. Our ADD
suffers from some hyperactivity, making it prone to find agents around us, including
supernatural ones, given fairly modest evidence of their presence. This tendency encourages
the generation and spread of god concepts. (J. Barrett 2004, 31)
On this account, something like the ADD provides the cause of religion. Although this mecha-
nism has the basic cognitive function of detecting real threats, because of its hyperactivity it some-
times yields beliefs that fail to track the truth. But it nonetheless promotes reproduction and sur-
vival because evolution has tuned it in a way that the false positives thus produced outweigh the
false negatives. Whereas a failure to detect a predator (false negative) can undermine survival
and reproduction, seeming to detect a predator when there is none (false positive) may have no
such negative result. Thus, a faulty cognitive mechanism can lead some natives to have the false
belief that all mushrooms nearby are poisonous. Yet since some mushrooms nearby are indeed
poisonous, having the false belief that all mushrooms are poisonous carries an evolutionary benefit
for these people.
Now if evolutionary pressures have shaped the human capacity to form and spread religious
beliefs in a way insensitive to their being true, that amounts to saying that the mechanism responsi-
ble for them fails to track the truth. In the absence of independent reasons for their epistemic war-
rant, it follows that these beliefs’ evolutionary genealogy undermines their epistemic warrant. From
an evolutionary debunking argument along these lines some even draw the conclusion that theism is
false (Pidgen 2013). However, skepticism about theism requires only that there be no good reason
available to think that belief in the existence of God has been produced in a way that tracks the
truth. If the belief turned out to be true, that would be a miracle or cosmic coincidence. Since
countenancing beliefs whose truth would depend on a miracle is unacceptable, it follows that the
evolutionary etiology of religious belief provides a skeptical argument against theism.
AN ANALOGY WITH MORAL REALISM

Theism is not alone in facing the challenge of evolutionary debunking: nonnaturalistic moral real-
ism, or the belief that there are mind-independent right and wrong actions in moral matters, also
faces it. Note, however, that not all evolutionary challenges to moral belief aim at debunking it.
The first wave of evolutionary challenges to morality associated “more evolved,” “the fittest to sur-
vive,” and other Darwinian notions with “right,” “best,” and other moral notions. But by appeal
to ordinary semantic intuitions about what asserted by using notions of either kind, moral realists
soon put this challenge to rest.
Since the 1970s, however, a better understanding of the Darwinian benefit of cooperation has
paved the way for evolutionary debunkers aiming at moral realism, a view that is the counterpart of
the ethics of theism. Evolutionary debunkers now argue that humans have an evolved propensity to
entertain certain moral beliefs because of their tendency to promote reproductive success and sur-
vival. Within that category are beliefs about the value of helping others, caring about our own chil-
dren, punishing cheaters, and so on. Debunkers of moral realism note that, since those beliefs are
shaped by a faculty that fails to track their truth, without independent good reasons for their episte-
mic warrant, moral skepticism or even nihilism follows. Michael Ruse argues that “the evolutionist’s
case is that ethics is a collective illusion of the human race, fashioned and maintained by natural
selection in order to promote individual reproduction.… What is really important to the evolution-
ist’s case is the claim that ethics is illusory inasmuch as it persuades us that it has an objective refer-
ence” (Ruse [1986] 1995, 235).

THE UPSHOT
Theism and moral realism seem vulnerable to an objection from the evolutionary etiology of their
most fundamental beliefs. In the case of theism, that is the belief that God, as standardly conceived
in the Abrahamic religions, objectively exists. Plausible hypotheses from evolutionary religious stud-
ies threaten the epistemic warrant of theism by indicating that it results from evolved mechanisms
that are insensitive to truth. Although this provides no apodictic reason for either agnosticism or
atheism, it does suggest that theists now face a difficult dilemma. They must produce a compelling
argument for the justification of their fundamental belief. Or they must show what’s wrong with an
evolutionary hypothesis that is widely accepted within the science of religion.
IS HISTORY GUIDED BY THE HAND OF PROVIDENCE?
Keith Parsons, University of Houston–Clear Lake
How would we expect the course of human history to have been different had it been providentially
directed toward certain divinely ordained ends as opposed to having had no such direction or guid-
ance? To ask this is not to imply that the choice is between history as providentially directed and
history as having no trends or directions. It is not “God in control” versus “one damn thing after
another.” A secular historian may conclude that history obeys intrinsic and deterministic laws, as
Georg Wilhelm Friedrich Hegel and Karl Marx held, or merely that history displays particular
trends contingently conditioned by influences such as religion or technology. The question is, then,
whether history displays distinctive patterns—or remarkable particular events—that are likely given
the providential guidance of history and less likely given no such guidance.
Where would we look for evidence of such purported providential guidance? Certainly, reli-
gious people have often believed that they discerned the hand of God in certain remarkable events.
Some of these events, such as the story of the Exodus, when the people of Israel were delivered from
bondage in Egypt by the mighty works of God, are so shrouded in myth and legend that it is hard
for the objective historian to ascertain exactly what happened. Other events are more reliably histor-
ical, such as the conquest of Babylon by Cyrus the Great of Persia and his decision to return the
Jews to Zion. Naturally, the Jews of that time and later saw this event as a clear case of God’s
actions through the human agency of Cyrus.
In general, Judaism and Christianity are historical religions. That is, it is essential to their doc-
trines that certain things happened at certain times and places in history, such as God’s covenant
with Abraham and Jesus’s resurrection. If these events did not happen, then Judaism and Christian-
ity rest on false claims, as Paul candidly admits with respect to the resurrection (1 Corinthians
15:14). The problem with citing such purported events as evidence of divine operation in history
is that they inevitably get bogged down in the preliminary debate over their actual occurrence. Skep-
tical secular historians, of course, do not accept these events and will cite arguments derived from
David Hume’s “Of Miracles.” So, you have interminable debates between apologists and skeptics
over the very occurrence of epiphanies or miracles.
Rather than stagnate in intractable debates about controversial claims, here it will be better to
look at events of unquestionable historicity and ask whether they are better understood as manifesting
providential guidance or intent, or as owing to mundane causes that evince no such telos. Here the
burden of proof must be on the one making the claim of providential guidance or direction. It would,
of course, be vacuous to claim that history displays evidence of a divine plan but concede that the
course of history looks no different than it would if there were no divine plan. That history evinces
no providential guidance will therefore be our null hypothesis. We will look at three crucial turning
points in history to see if they are most reasonably construed as evincing divine guidance or intent.
The three crucial episodes I choose are these:
1. The origin of civilization.
2. The origin of the great world religions in the Axial Age.
3. The scientific revolution.
Why these three? I am betting that anyone interested in finding evidence of providential guidance in
history would expect to find it in these loci if anywhere. By modus tollens, then, failure to discern
the hand of providence here means that our null hypothesis is not rejected.
But could not other events or episodes have been chosen? Sure, but the issue is big, and space
here is small, so, as in any test of a null hypothesis, we have to make our best guess about where the
supposed effect, if real, would be found. I am betting that civilization, the world’s great religious tra-
ditions, and modern science will be seen as very good things, and therefore plausibly of great interest
to any provident deity.
THE ORIGIN OF CIVILIZATION

The origin of civilization is perhaps best understood as the emergence of significantly greater com-
plexity in human social, political, and economic organization. Prior to the rise of civilization, human
populations were small, dispersed, and organized relatively simply. Complex social and political
organization probably began when environmental challenges required the coordination of the labor
of large numbers of people (Feder 2004, 388). The development of agricultural techniques and
technology, notably irrigation, provided the food surplus that was a necessary condition for civiliza-
tion:
A food surplus made possible by the agricultural revolution set the stage for the differential
access to wealth, and with the concentration of wealth came the concentration of social and
political power. “Civilization,” characterized by the existence of a formal government, social
stratification, large and dense settlements, monumental edifices, elaborate burials, large armies,
full-time artisans, and a system of record-keeping developed in some parts of the Old World as
fewer people were needed for the subsistence quest and its rulers attempted to legitimize,
reinforce, and magnify their position of power and level of wealth. (Feder 2004, 424)
The earliest civilizations so characterized began in Mesopotamia and the Nile Valley in the fourth
millennium BCE and somewhat later in the Americas.
It is a perennial temptation when a development thought to be favorable occurs to see it in ret-
rospect as having been intended or at least “in the cards” all along. This has certainly been the case
with the existence of the human species. Many have found it difficult to accept that human evolu-
tion was as chancy and contingent as the evolution of any other species, yet this is precisely what
Darwinian theory implies. That is, the process of organic evolution did not aim at our existence
any more than, say, the existence of the zebra mussel or the oak tree.
Likewise, in looking back at the development of civilization, it is tempting to read providential
intent in the process, since, after all, most of us value the possibilities and possessions that civilization
provides. However, human history seems to be as radically contingent as evolutionary history.
Anatomically, modern humans had existed for tens of thousands of years before the emergence of civ-
ilization, and, when civilization did emerge, the causes appear to have been dependent on local factors,
such as the enormous food surplus that irrigation can produce in places like Mesopotamia and the
Nile Valley, and the need for the coordination of many workers to achieve and maintain that surplus.
Yet the food surplus was a necessary but not sufficient condition for the development of civili-
zation, as prehistorian Kenneth L. Feder observes:
The shift to social, economic, and political complexity was not inevitable or universal; the
archaeological record of Neolithic food producers does not reflect a pattern by which all
peoples marched in lockstep towards such complexity. Instead, the ability to produce and
control a surplus of food, with few exceptions made possible exclusively by agriculture, made
such complexity not certain but possible. (Feder 2004, 396)
In fact, it seems quite conceivable that human beings might have been hunter-gatherers until this
day. Hunter-gatherer societies persist (or until quite recently did persist) in the world today.
However, if there are deterministic historical factors that make it inevitable that civilization
does eventually develop wherever it can, none of the identifiable causes of civilization appears to
require divine planning or guidance. Such planning or guidance cannot be ruled out, but, as noted

earlier, if history allegedly shaped by providence looks no different from what we would expect with
no providence, then to claim such providence is vacuous.
THE AXIAL AGE

Historians of religion refer to the period from roughly 800 to 200 BCE as the Axial Age. As Karen
Armstrong puts it:
In all the main regions of the civilized world, people created new ideologies that have
continued to be crucial and informative.… Taoism [Daoism] and Confucianism in China,
Hinduism and Buddhism in India and philosophical rationalism in Europe. The Middle East
did not produce a uniform solution, but in Iran and Israel, Zoroaster and the Hebrew
prophets respectively evolved different versions of monotheism. (Armstrong 1993, 27)
As Armstrong sees it, each of these developments was a response to similar problems caused by sig-
nificant social and economic changes in these diverse societies.
According to John Hick, religion as developed in the Axial Age had a distinctly different character
from “archaic” or pre-Axial religion (Hick 2004, 22–29). Archaic religions were, for instance, the poly-
theistic “pagan” cults of the Norse/Germanic peoples who worshipped Odin, Thor, and associated dei-
ties, and of the early Greeks who worshipped the Olympians. Characteristic of archaic religion, accord-
ing to Hick, was a sense of the natural world as pervaded by spiritual forces or occupied by supernatural
beings. Indeed, there was no clear distinction between “natural” and “supernatural” or the secular and
the religious. Forests, hills, rivers, and the sky were home to unseen beings—gods, demigods, ancestors,
totem animals—that were to be worshipped, appeased, or bargained with, and there were accepted
rites, spells, and magical practices for doing so. The purpose of archaic religion was to tame the chaotic
unpredictability of life and to provide, via the vehicle of myth, a sense of the unity of the individual
with the community, and of humans with all of organic life and with the earth itself.
By contrast, post-Axial religion, in diverse ways, placed emphasis on salvation, either of the
individual or the community. It is this soteriological emphasis that, for Hick, is the definitive dis-
tinction between pre- and post-Axial religion (Hick 2004, 32–33). All of the post-Axial religions
emphasize the radical deficiency or defectiveness of the world and human beings. Christianity, for
instance, sees the human condition as “fallen,” that is, as one of radical alienation from God and
from each other. Only redemption and justification through Christ can cure this condition and
restore the primordial unity with God that is everlasting life. Analogous conceptions are found even
in the very different religious traditions of India. For Hindus, human life is lived in a world of illu-
sion in which karma propels us into the painful cycle of recurring birth and death. Release from
samsara, the wheel of birth and death, is moksha. In Judaism, instead of individual salvation, the
emphasis is on the coming of the Kingdom of God.
So, should we see in the diverse but parallel developments of religions of salvation during the Axial
Age evidence of providential intent? Surely, devotees of these religions will interpret these developments
as having been divinely guided or inspired. What if, however, one holds that the pagan, pre-Axial reli-
giosity offered in many ways a more genuine and authentic approach to the sacred than the religions of
salvation? Such a stance is eminently defensible. The post-Axial religious denigration of “the world”
may be regarded as problematic, even pathological, and the proffered “salvation” as illusory. One could
argue at a deeper level that post-Axial religions did a great disservice to humanity by shattering the
organic unity of humans with the divine, and by, in effect, making the sacred remote and difficult to
access—except, allegedly, via the particular doctrines, disciplines, and practices of specific religions.
Of course, these are deep issues that need to be argued out at much greater length. My only
purpose here is to adumbrate a possible response to the claim that the movement from the pre-Axial
to the post-Axial religious landscape is a kind of spiritual progress that is indicative of providential
intent. Such a claim rests on a notion of the superiority of post-Axial religion, an assumption that
is eminently open to challenge.
THE SCIENTIFIC REVOLUTION

Historian Steven Shapin begins his book on the scientific revolution by denying that there was a sci-
entific revolution (Shapin 1996, 1). In fact, among historians of science in general there has for
some time been a tendency to downplay the notion that anything of particularly great significance
happened in the early seventeenth century, certainly not a Copernican revolution. Howard Margolis
amusingly comments on this peculiarity:
But if you look closely at what was going on in science around the year 1600, you will have no
trouble seeing the appropriateness of a story of the Old West about a cowboy wandering over
the plateau of northern Arizona. Innocently, he rides right up to the rim of the Grand
Canyon. The cowboy sits for a long time, contemplating the vast gorge. Eventually he
mutters, “Something happened here.” (Margolis 2002, 3)
Assuming that the period marked by the publication of Nicolaus Copernicus’s Revolutions in 1543
and Isaac Newton’s Principia in 1687 encompasses a time of discovery and innovation in the theo-
ries and practice of science sufficient to be declared “revolutionary,” did these events evince provi-
dential guidance or design?
In response to attacks on Christianity as antiscientific, some apologists have claimed that, on the
contrary, it is not a coincidence that modern science developed in Christian Europe. That is, they
argue that Christianity, with its emphasis on a creation by a rational God, provided a more fertile field
for the growth of science than did pagan antiquity or non-Christian societies of the day (see, e.g., Jaki
2000). Such a claim is highly debatable, but even if we accept that Christianity was, for various rea-
sons, the most propitious matrix for the development of modern science, this is irrelevant. That Chris-
tianity, as an intellectual milieu, was favorable to the development of science, if true, would be true
whether or not these developments revealed the guiding hand of the Christian God. In other words,
even if false, Christianity could have provided a fruitful set of heuristic assumptions.
Where, then, in the events of the scientific revolution do we see evidence of providential plan-
ning or guidance? A salient question to ask here is why scientific revolutions (plural) occur in the
ostensibly sporadic fashion that they do. Despite the multiple deficiencies of his account of science,
Thomas Kuhn did effectively demonstrate that science does not develop in a linear or monotonic
manner, but in fits and starts. Greek science reached a peak in the Hellenistic period (323–31
BCE), and Islamic science in the ninth and tenth centuries. The Darwinian revolution was in the
nineteenth century, and quantum and relativity theory in the twentieth. The reasons that these per-
iods of rapid advance occurred when and where they did seem to depend on a complex of contin-
gent factors and do not evince an overall pattern of fated development. (The best recent histories
of science, e.g., Lindberg 2007, emphasize the complexity and contingency of the development of
science.) Again, apparent lack of pattern is evidence for actual lack of pattern. Prima facie, one
would expect that had Providence guided the course of science, it would be one of more continuous
and less sporadic development.
In summary, we have seen that three crucial events in human history—occurrences where we
would expect to see providential guidance if anywhere—are well understood as contingently and
locally caused and, prima facie, do not display the inevitability or patterns indicative of guidance
or planning. Again, nothing prohibits those so inclined from retrospectively reading providential
intent into those events. However, there seems to be no objective basis for a providential historiog-
raphy or philosophy of history.
THE EVOLUTIONARY DOWNSIDE OF MASS THEISM AND THE RISE

OF POPULAR ATHEISM
Gregory Paul, Independent Scholar
Presumably unique to Homo, most especially H. sapiens, supernaturalistic theism has been widely
popular in human societies probably for many tens of thousands of years. Enough so that a number
of researchers have characterized religiosity to be as integral to being human as is language (Bloom
2007). Many have also presumed that high levels of belief in deities have been vital to societal cohe-
sion in one way or another, contributing to high levels of prosocial behavior (Stark 2006; Putnam

and Campbell 2010). In this scenario theism is a selectively advantageous evolutionary adaptation
that has done much to improve the human condition. Yet until recently little effort was made to
test these important hypotheses. The results of the pertinent research indicate that theism is neither
as ubiquitous as typically thought nor as benign and beneficial as commonly believed. Modern athe-
ism is associated with the best societal conditions yet seen in human history.
Early theism probably appeared in species of Homo in the middle or late Pleistocene. Appar-
ently deliberate burials of H. heidelbergensis and, more especially, by H. neanderthalensis and archaic
H. sapiens with grave goods, suggest a belief in an afterlife and perhaps some degree of religiosity.
However, there are reasons to believe that hunter-gatherers are not always religious. A survey of liv-
ing hunter-gatherer tribes found that while all possess some level of animism, a quarter lack either
belief in an afterlife or shamanism, with a fifth lacking at least one of those features (Peoples, Duda,
and Marlowe 2016). Some of these peoples do not have a tradition of burial of the dead; other
hunter-gatherer groups worship ancestors and believe in gods. These tribal peoples have demo-
graphic roots extending back into the Pleistocene, so it is probable that levels of religiosity were sim-
ilarly variable since the origins of H. sapiens, and it is possible that theism was even less common
among prehistoric hunter-gatherers (Paul 2012).
The onset of agriculture and the civilizations it allowed saw a major increase in the sophistica-
tion, organization, and perhaps popularity of theism, including the development of priesthoods and
state-connected religions, as well as monotheism. Although widely distributed, the intensity of these
religions has varied considerably. The religion of China, for example, is not generally theistic, in the
sense of the theism of the Abrahamic faiths.
The onset of modernity via advanced science and technology with a correspondingly greater
degree of material comfort and security has brought with it a major decline in the popularity of the-
ism (D. Barrett, Kurian, and Johnson 2001; Bruce 2011; Norris and Inglehart 2004; Harris Poll
2013; Gallup 2016b, 2017a, 2017b; Zuckerman 2009; Paul 2009, 2010, 2012; Pew 2014,
2015a, 2015b; RedC 2012). This has been most true in the Western democracies. The United
States has tended to lag behind in its loss of religion, but despite massive efforts by organized the-
ism, America is rapidly catching up with the rest of the West, with the nonreligious expanding from
about 30 percent to about 40 percent of the population in a decade, about two-thirds or more of
them qualifying as atheists (Harris Poll 2013; Gervais and Najle 2018). Bible literalism has dropped
from about 40 percent to about 30 percent since the Reagan era, and creationist beliefs are edging
down while support for god-free Darwinian evolution rises. Globally the nonreligious have risen
from about 30 percent to about 40 percent since the start of the twenty-first century, although (per-
haps by way of reaction) there has been a rise of hard-line, conservative religion in some places.
That the level of theism has proven so variable in prehistoric, historic, and modern cultures,
and has been so easily shed in democratic societies, is extremely interesting. The disparity proves
that “theoreligion” (a religion involving belief in gods) is not as universal as are other human attri-
butes. Not consistently popular even in archaic cultures, supernaturalistic theism has proven to be
highly vulnerable to the materialistic and intellectual forces of modernity.
These findings make it possible to test the commonly touted hypothesis that supernaturalistic
religion is an overall positive factor for people and for societies. Hunter-gatherer cultures are prone
to being afflicted by high levels of societal dysfunction, sometimes lethal. There is currently no evi-
dence that high levels of supernaturalism produce lower rates of social illness in tribal peoples.
Instead, the core problem for noncivilized societies is the absence of a criminal justice system, which
leaves individuals and groups free to rampage as long as they feel able to fend off counterattacks
(Pinker 2018).
The agriculture that made civilization possible may have been invented mainly to satisfy the
desire for mass consumption of alcohol (Hayden, Manuel, and Shane 2012), but there is no reason
to think that supernaturalism played a critical role in this development. The onset of organized the-
ism may have played a part in the beginnings of large-scale settlements for purposes of mass wor-
ship. Organized civilization and the law enforcement systems that accompany them have been criti-
cal in suppressing the level of violence while improving overall living (Pinker 2018). The degree to
which religiosity was critical to this success is uncertain. It is possible that had early civilizations
been atheistic they would have been at least as successful. Nor is there evidence that highly theistic
India enjoyed superior socioeconomic conditions compared to less theistic China.
It is notable that religion had little to do with the first experiments in democracy in the classical
world. An opportunity to develop the industrial complex that suppresses slavery was missed by the
Greeks and Romans despite, for example, their knowledge of steam power. Instead the Roman
Empire was taken over by a form of monotheistic theism that severely suppressed contrary opinions
and invented the theological anti-Semitism (Carroll 2001) that was carried into Lutheranism. Chris-
tianity can be credited with a strong decline in infanticide, saving up to a tenth of those born (Paul
2012). However, war-torn medieval Europe was afflicted by rates of homicide so extreme that Leo-
nard Beeghley (2003) characterized the populace as a “wild bunch.” Organized theism was behind
the liquidation of the Cathars and other heretics as well as the war on women accused of sorcery.
Contact between the hemispheres was catastrophic for the western half, with the collapse of New
World populations by Old World diseases and the aggressive imperialism of Christian Europeans.
Islam also spread largely by military conquest, the mainly peaceful conversion of the East Indies
being a notable exception.
Europe was the center of the radical modernization that led to the developed democracies,
although whether this was due to its Christian population or geopolitics is a matter of dispute (Stark
2006 versus Diamond 2005; Morris 2014). It is true that Christianity was at least compatible with
the factors that led to the appearance of Western modernity. A rejection of divine autocratic rule for
largely secular reasons was initiated in the United States and began to spread to a number of Euro-
pean countries in the 1800s. The Jewish and Christian portions of the Bible endorse chattel slavery,
and Euro-American Christians developed a new, race-based version of the institution that helped lay
the foundations for a self-styled “scientific” racism (Richards 2013). As predicted by Adam Smith
(1776) the secular economics of corporate-consumer capitalism was crucial to the end of mass slav-
ery, which ceased peacefully in the relatively secular British Empire. The strongly religious antebel-
lum American South saw a brutal form of slavery (Baptist 2015; Cashin 2017), which came to an
end because the war to reunite the nation accidentally led to the end of slavery. In the reunited
United States the theo-conservative South developed a terroristic Jim Crow apartheid lynching
(Dray 2003; Cashin 2017). Although eugenics theory was putatively Darwinian/scientific, this did
not prevent eugenics laws from being enthusiastically passed in a majority of American states, many
theo-conservative and some with laws against the teaching of evolution.
Most wars have involved a religious component, as in the case of the Islamic conquests, the
Christian counterattack via the Crusades, which employed terroristic cannibalism (Rubenstein
2008), the long Catholic-Protestant wars of Europe, and the Aztec capture of sacrificial victims.
Many of the French revolutionaries were militant deists, and the later mass slaughter of the Paris
Commune was driven by Catholics (Merriman 2014). The rise of fascism often occurred in alliance
with religious powers (Carroll 2001; Paul 2003, 2003/2004; Kertzer 2014). Far from being a direct
expression of atheistic Darwinian materialistic science (Richards 2013), Nazism was a mystic theol-
ogy mixed with pseudoscience. Adolf Hitler became dictator only because of the direct aid of the
Catholic Church he was nominally a member of (Paul 2003, 2004). The Euro-Christian alliance
with fascism has in part been driven by opposition to atheistic communism, which itself led to enor-
mous atrocities. Since the end of the Cold War, however, those who have been killed by violence
induced by atheistic movements number in the very low six figures, while those who have died at
the hands of believers number several million.
Violence has generally declined with modernity (Paul 2012; Pinker 2018), in parallel with the
waning of theism. Hence homicide rates dropped over recent centuries in secularizing Europe and
have remained that way. American theo-conservatives often blamed the (perhaps temporary) rise
in crime, sexually transmitted diseases, teen pregnancy, and abortion in the late 1900s on the rise
of secular, Darwinian and counterculture values, including the judicial limitations on school reli-
gious services. But these events also followed the federal endorsement of religion, including placing
a reference to God in the Pledge of Allegiance.
A number of studies have demonstrated a pattern in which the more secular and atheistic
regional and national societies are, the better they tend to be in terms of crime and socially

dysfunctional behavior (Norris and Inglehart 2004; Zuckerman 2009; Paul 2009, 2010, 2012; Rees
2010; Barber 2011). This pattern is most obviously expressed in the first-world developed democra-
cies, in which the exceptionally theistic United States is the most exceptionally dysfunctional, partic-
ularly in the red (Republican) states, which tend to be more religious (correlating Gallup 2016a
with Gallup and Healthways 2017). The same tends to be true globally. For example, atheistic
China is enjoying markedly better economic conditions than is the more pious India, despite the
latter being a fairly stable democracy. In general, the most liberal-progressive democracies featuring
hybrid capitalist economies with universal health care and a welfare state have been the most suc-
cessful and most atheistic in human history.
To date, proponents of the hypothesis that mass religiosity is a true societal positive have failed
to document their case. Although analyses often indicate that religious Americans are doing better
than nontheists in a number of regards (Norris and Inglehart 2004), this may reflect the difficulties
of being a skeptic in a majority theistic society. Luke Galen and James D. Kloet’s (2011) statistical
reanalysis indicates that similarly strong atheists and theists share similar levels of quality of life—
theistic fence-sitters are the ones who appear to be doing less well—whereas other studies cast doubt
on the hypothesis that the less religious are less well-off (Putnam and Campbell 2010; Paul 2012).
This evidence strongly indicates that socioeconomic factors play a leading role in the decline in
popularity of supernaturalistic theism (Bruce 2011; Norris and Inglehart 2004; Pew 2017). There
are no examples of highly successful societies that are highly religious, and the ongoing worldwide
rise in nonreligiosity closely tracks with the rise of the global middle class. It therefore appears that
religion is most popular when the life circumstances are dysfunctional and insecure for the majority
of the population (Paul 2009, 2010, 2014). The combined forces of modernity are so powerful, and
the human propensity to be religious is so weak, that it is not surprising that mass theism cannot
survive the socioeconomic onslaught of modernity. But although it is not practical for theism to
thrive in successful societies, this does not mean that it consistently thrives in dysfunctional cultures,
as is shown by the low religiosity levels of hunter-gatherers, some ancient civilizations, and the
dechurching of much of the American white working class.
The evidence given above suggests that religiosity declines with social success. But could it also
be the case that religiosity is a cause of societal ills? There are a number of reasons to conclude that
both factors are operative. In many forms of theism, the gods are not benign. Even when the gods
are thought of as benign, it is arguable that their teachings are morally harmful, promoting misog-
yny, severe punishments, genocidal conquest, and slavery. Some studies have indicated a reduced
level of altruism in more religious children (Decety et al. 2015), while higher levels of conservative
religious practice are associated with elevated levels of racial and ethnic prejudice (Scheepers, Gijs-
berts, and Hello 2002; Hall, Matz, and Wood 2010).
Contrary to the claims of both religious apologists and some recent evolutionary psychologists
and sociologists, it seems that atheism in a democratic context produces the most functional kind
of society.
ACCOUNTING FOR MONOTHEISM’S HISTORIC DISPLACEMENT OF

NON-MONOTHEISM WITHOUT A DIVINE HAND
Matthew Wade Ferguson, University of California, Irvine
Today over half the world’s population adheres to two monotheistic religions: Christianity and
Islam. This proportion of monotheistic belief has historically not always been the case, and it invites
the question of whether the historic rise and distribution of monotheism around the globe is better
explained by a theistic account of divinely ordained ends, or if a nontheistic account of human his-
tory can equally explain the same phenomenon. To answer this is no easy matter primarily because
we only have one data set to work with: human history on Earth. Could history have turned out
differently, where monotheism did not become predominant? We may never know. Likewise, even
if it is true that monotheism has a proclivity to replace non-monotheism among intelligent species,
this would still not necessarily entail that divine guidance is the best explanation. There may be per-
fectly natural reasons for why monotheism has a tendency to mimetically evolve alongside other
developments in civilization.
To form a robust theistic theory for the spread and distribution of monotheism, therefore, the
burden of proof is to show that this would have less probably happened, without supernatural
causes. It makes little sense to say that God has intervened in history, when history would look
the same without God’s involvement. Otherwise, the null hypothesis is that our religious landscape
is expected to look the same, due to nothing but natural causes.
First, the operative terminology must be established for what is being investigated. For the sake
of defining traditional monotheism, this chapter will use the following definition of H. S. Versnel
(2011, 241): “The conviction that only one god exists (involving the cultic corollary of exclusive
worship), while other gods do not.” Out of the current world population, a Pew survey (2012) esti-
mates that 31.5 percent are Christian, 23.2 percent are Muslim, and 0.2 percent are Jewish. These
religions are classified as monotheistic. Fifteen percent are Hindu and 7.1 percent are Buddhist.
These religions are classified as non-monotheistic. “Folk Religionists” constitute 5.9 percent, and
0.8 percent are “Other Religions.” These are considered ambiguous. The religiously unaffiliated
constitute 16.3 percent. In terms of the global religious landscape, approximately 54.9 percent is
monotheist, therefore, a demographic majority.
Precise statistics are difficult to assess for religious trends in many periods, but a general historical
survey should suggest that the world was probably less monotheistic than it is today. The earliest known
monotheism dates to the fourteenth century BCE, when Pharaoh Amenhotep IV declared the sun god
Aten as the sole God and sought to eradicate Egyptian polytheism; but his religious reforms ended
upon his death (Gregory 2009, 13n.20). The Mediterranean was less monotheistic during the early
Roman Empire, which was dominated primarily by a plurality of polytheistic religions. Judaism repre-
sented approximately 10 percent of the population during the first century CE (Potter 2006, 565),
whereas Christians by the end of the century had a population below 10,000 (Stark 1996, 6).
A significant chapter in monotheism must have existed during the Achaemenid and Sassanid Per-
sian Empires, which respectively controlled 44 percent and 37 percent of the world’s population (Sha-
jari 2015, 26). Since both empires established Zoroastrianism as their official religion—which held a
theology of cosmogonic dualism but eschatological monotheism (Boyd and Crosby 1979)—there
must have been millions of adherents who practiced the faith within their borders. The Achaemenid
Empire showed tolerance for foreign religion (Curtis and Simpson 2010, 86), however, and there
was considerable religious diversity in the Sassanid Empire (Daryaee 2009, 96). As such, it cannot
be assumed that the whole population of these empires practiced the religion.
There is also the question of whether monotheism was prevalent in the religious beliefs of human-
kind’s anthropological history. An older view once advocated by Austrian scholar Wilhelm Schmidt
(1931) was that monotheism was the earliest religious belief, and that animism and polytheism were
later developments. This evolutionary understanding was justified on the grounds that monotheism is
allegedly the simplest and therefore the oldest. As anthropologist Robert Winzeler (2012, 54) points
out, however, “Such arguments do not seem to have held up well.” Winzeler notes that while the reli-
gions of “small-scale” and “technologically simple peoples” can have “high god beliefs,” they tend to dif-
fer from the “thoroughgoing monotheism of Judaism, Christianity, and Islam.”
The survey of historical monotheism above does not provide any concrete evidence that mono-
theistic religion was ever as widespread as it is today, in reaching over half the world’s population.
Since my academic background specializes in early Christian history, which is also the world’s largest
monotheistic faith, the remainder of this chapter primarily examines its spread and distribution. As
previously noted, to overcome the null hypothesis that the current religious landscape can be equally
explained by natural causes, a theistic account must demonstrate that supernatural causes are a more
probable explanation. Four theories of supernatural causes are considered in this chapter:
1. Christianity spreading to uncontacted parts of the world without missionaries, suggesting divine
revelation.
2. Christianity spreading at a speed so rapid that it suggests divine intervention.

3. Christianity having an inexplicable proclivity to replace non-monotheistic religion, suggesting

divine intervention.
4. Christianity spreading in a manner predicted by a fulfilled prophecy, which can meet evidential
criteria.
The first of these theories does not have much evidential support. The Mormon faith, for example,
claims that Christian religion was revealed to inhabitants of the Americas not by missionaries but
through a revelation of Jesus. But as Warren Steinkraus and Michael Mitias (1998, 48) point out,
“No corroboration, outside evidence, or archaeological support exists, and the idea that Jesus came
to North America before [Christopher] Columbus and any known history is totally without founda-
tion.” For the first of these theories, it is generally safe to assume that there is no bona fide example
of a discrete monotheistic system of belief appearing on one part of the globe apart from another
without human contact, which has been subjected to historical and archaeological scrutiny.
The second theory offers the most promise for Christianity during its earliest stages. Christian-
ity first emerged as a minority religion in the Roman Empire, when it could be less expected to
eventually become the empire’s dominant religion or to reach its global influence today. But Chris-
tianity’s growth is less extraordinary when factoring in time and the rate of change. Sociologist Rod-
ney Stark (1996, 5–7) estimates that Christianity spread, on average, at a rate of roughly 40 percent
per decade. Notably, Stark’s calculation may be too high, since Bart Ehrman (2018, 170) has
revised his rate to an average of 30 percent, but 40 percent shall be granted a fortiori. Starting with
an estimate of 1,000 Christians in 40 CE, Christianity would have represented approximately 10
percent of the empire’s population in 300 CE, enough to easily gain Constantine’s attention by nat-
ural means (despite legendary claims about his conversion to the contrary).
What is remarkable about Stark’s findings is that he notes a parallel between early Christian
growth rates and that of the Mormon Church over the past century, which maintained an average
growth rate of 43 percent per decade. As Stark (1996, 7) remarks, “Thus we know that the numeri-
cal goals Christianity needed to achieve are entirely in keeping with modern experience, and we
are not forced to seek exceptional explanations.” After Christianity had gained ascendance in late
antiquity, post-Roman breakaway states and the rising Catholic Church subsequently spread the
religion throughout the rest of Europe, and eventually European colonial powers exported Chris-
tianity around the globe. All of these developments are explicable enough through political and eco-
nomic factors for supernatural causes to be no more probable than natural ones.
Worth noting is that the rate of Islamic growth was no more remarkable, following the Muslim
expansions of the seventh and eighth centuries CE. As Hugh Kennedy (2007, 6) explains, “While
conquest and settlement took place comparatively quickly … the conversion of the subject people
to Islam was a slow and long-drawn-out process, and it was not until the tenth and early eleventh
centuries that the majority of the population was converted to Islam.”
On the third theory, the first observation that should be made about ancient Christianity,
which is perhaps counterintuitive to how religious affiliation works today, is that it was a translocal,
transethnic religion. As Larry Hurtado (2016, 79) points out, most pagan deities were tied to city
and ethnicity (e.g., Roman gods), but Christian faith could be practiced regardless of one’s demo-
graphic background. This made the religion transferable and proliferant in ways that most contem-
porary polytheistic religion was not. From the time of the apostle Paul (Romans 1:5–7), Christianity
likewise placed a heavy emphasis on proselytizing, helping to bolster its numbers. Another atypical
characteristic of Christianity pertains to the nature of its exclusive monotheism. By teaching a the-
ology of one God and not others, it inherently excludes the worship of other deities. Whereas the
pluralistic pagan religions of the Roman Empire could be multiply practiced by a single individual,
Christians were prohibited from practicing any religion but their own (1 Corinthians 8). As Chris-
tianity grew in numbers, therefore, other religions shrank; but the introduction of a new local deity
in a sea of polytheistic pluralism had no such consequence. Mimetically, this had the effect of
crowding out other religions as Christianity spread, giving exclusive monotheism a tendency to sup-
plant more tolerant non-monotheistic religions. When Christianity became dominant in the late
Roman Empire, its exclusivism turned into outright suppression of polytheistic religions, with
emperors like Theodosius I issuing decrees that imposed anti-polytheist prohibitions. Once Chris-
tian exclusivism had political backing in subsequent European and colonial history, its natural
spread at the expense of other religions was no great surprise. Although the observations above are
made about Christianity, it should be noted that most apply to Islam as well.
For the fourth theory, evidential criteria must first be established regarding what constitutes a
genuine fulfillment of prophecy. A prophecy must be (1) made prior to the time of the fulfilling
event, (2) unlikely given ordinary background knowledge, (3) unambiguous enough not to consti-
tute a vague guess or to be applied to random events, (4) not self-fulfilling, and (5) not one out
of a larger tradition of failed prophecies, so as to only have succeeded by chance. Some verses in
the New Testament can be read as prophesying the spread of Christianity throughout the world.
As part of the Great Commission, for example, Jesus tells his followers to make disciples of all
nations (Matthew 28:19).
To begin with, the Greek word for “nations” in Matthew 28:19 is éthnē, which Daniel
Harrington (1990, 53) explains means “Gentiles” in its historical context. Interestingly, éthnē is also
the etymological origin of the English word “ethnicity.” One could argue that this passage predicts,
therefore, that Christianity would spread to all the ethnicities of the world. That would include the
indigenous peoples of the Americas, if taken literally. In context, it is doubtful that Matthew can be
read with any such meaning. In Matthew 4:8, Satan takes Jesus onto a mountain to show him (hor-
izontally) all the kingdoms of the world. Even Christian scholar Kyle Greenwood (2015, 136) in
Scripture and Cosmology argues that this passage “precludes the possibility of people living on the
opposite side of a sphere.” Perhaps granting a metaphorical reading, however, it is still likely Mat-
thew thought that the planet was much smaller than is known today.
Another issue with interpreting this verse to mean that Christianity would reach the peoples of
the Americas pertains to Matthew’s expectation about the chronological length of the Christian
ministry. In Matthew 24:1–35, several signs are given of the end-times, and Jesus says that this
geneá (generation) will not pass until all these things are brought to fulfillment. A number of issues
abound regarding whether Matthew intends an imminent parousia (Second Coming of Christ) near
the end of the first century CE or one that is deferred at some point in the future, but they are dis-
cussed at length by David Sim (1996, 155–174) in Apocalyptic Eschatology in the Gospel of Matthew.
Sim (1996, 174) concludes that interpretations of deferred parousia “have no real basis,” but even
more critically he points out, “The one major unfulfilled prediction is that the gospel must be
preached throughout the world but … Matthew probably considered that this was close to fulfill-
ment.” This final point is crucial to the present section.
If Matthew maintained that Christianity could be adequately preached to the Gentiles before
the parousia, then the gospel certainly does not prophesy a chronologically distant spread of Chris-
tianity to the Americas. This implication compels a more narrow reading of the prediction to mean
only that Christianity would be preached to the immediate world surrounding Palestine, for about
half a century or so after Jesus’s death. This is not really that remarkable of a prediction, however,
given the land and sea networks established by the Roman Empire, and Palestine’s proximity to
the Near East. If one interprets a literal prophecy of Christianity’s spread to the entire world, how-
ever, then Matthew also made failed apocalyptic predictions elsewhere. The third theory is thus
caught between the horns. On the narrow view, the theory does not make a prediction that is
unlikely given general background knowledge, failing the second evidential criterion. On the literal
view, the prophecy was made among other failed prophecies, and so doesn’t meet the fifth criterion.
Either way, on both interpretations, since Christian missionaries fulfilled the prediction (rather than
an act of divine intervention), the prophecy was self-fulfilled, failing criterion four, and the details of
how Christianity would spread are rather ambiguous to begin with, failing criterion three.
Out of the four theories considered in this chapter, I would argue that the third has the most
force. There is very little evidence that a monotheistic religion has spread without human contact,
the growth rates of Christianity and Islam are not so remarkable as to suggest supernatural interfer-
ence, and biblical prophecies about the spread of Abrahamic religion are vague if not counterba-
lanced by failed prophecies. It is a peculiar fact of world history, however, that monotheism has gen-
erally displaced non-monotheistic religion, at least for roughly half the world’s population, in

comparatively recent anthropological history. One could certainly imagine an alternate timeline in
which events turned out differently and non-monotheistic religions were still predominant. If I
had to put my finger on the root cause for this, I would argue that it rests primarily with the exclu-
sivism of the world’s two major monotheistic faiths. Both Christianity and Islam have historically
had more of a drive to gain converts and to stamp out other religions, and as a result they have done
just that.
Bibliography
Armstrong, Karen. A History of God: The 4000-Year Quest of Judaism, Darwin, Charles. The Descent of Man, and Selection in Relation to Sex.
Christianity, and Islam. New York: Gramercy Books, 1993. London: John Murray, 1871.
Atran, Scott. In Gods We Trust: The Evolutionary Landscape of Religion. Darwin, Charles. The Descent of Man, and Selection in Relation to Sex.
Oxford: Oxford University Press, 2002. (1891.) Princeton, NJ: Princeton University Press, 1981.
Baptist, Edward E. The Half Has Never Been Told: Slavery and the Making of Darwin, Charles. The Correspondence of Charles Darwin. Cambridge, UK:
American Capitalism. New York: Basic Books, 2015. Cambridge University Press, 1985–.
Barber, Nigel. “A Cross-National Test of the Uncertainty Hypothesis of Daryaee, Touraj. Sasanian Persia: The Rise and Fall of an Empire. London: I.
Religious Belief.” Cross-Cultural Research 45, no. 3 (2011): 318–333. B. Tauris, 2009.
Barrett, David, George Thomas Kurian, and Todd M. Johnson, eds. World
Dawkins, Richard. “Universal Darwinism.” In Evolution from Molecules to
Christian Encyclopedia. Oxford: Oxford University Press, 2001.
Men, edited by D. S. Bendall, 403–425. Cambridge, UK: Cambridge
Barrett, Justin L. Why Would Anyone Believe in God? Walnut Creek, CA: University Press, 1983.
Altamira Press, 2004.
Dawkins, Richard. The Blind Watchmaker. New York: Norton, 1986.
Beeghley, Leonard. Homicide: A Sociological Explanation. Lanham, MD:
Rowman and Littlefield, 2003. Decety, Jean, et al. “The Negative Association between Religiousness and
Children’s Altruism across the World.” Current Biology 25, no. 22
Bering, Jesse M. “The Existential Theory of the Mind.” Review of General (2015): 2951–2955.
Psychology 6, no. 1 (2002): 3–24.
Dembski, William A., and Michael Ruse, eds. Debating Design: Darwin to
Bloom, Paul. “Religion Is Natural.” Developmental Science 10, no. 1 (2007): DNA. Cambridge, UK: Cambridge University Press, 2004.
147–151.
Diamond, Jared M. Guns, Germs, and Steel: The Fates of Human Societies.
Boyd, James, and Donald A. Crosby. “Is Zoroastrianism Dualistic or
London: Vintage, 2005.
Monotheistic?” Journal of the American Academy of Religion 47, no. 4
(1979): 557–588. Dray, Philip. At the Hands of Persons Unknown: The Lynching of Black
America. New York: Random House, 2003.
Brooks, Arthur C. Who Really Cares? New York: Basic Books, 2006.
Bruce, Steve. Secularization: In Defence of an Unfashionable Theory. Oxford: Ehrman, Bart. The Triumph of Christianity: How a Forbidden Religion Swept
Oxford University Press, 2011. the World. New York: Simon and Schuster, 2018.
Bulbulia, Joseph, Richard Sosis, Erica Harris, et al., eds. The Evolution of Feder, Kenneth L. The Past in Perspective: An Introduction to Human
Religion: Studies, Theories, and Critiques. Santa Margarita, CA: Collins Prehistory. 3rd ed. New York: McGraw-Hill, 2004.
Foundation Press, 2008. Galen, Luke, and James D. Kloet. “Mental Well-Being in the Religious and
Carroll, James. Constantine’s Sword: The Church and the Jews: A History. Nonreligious: Evidence for a Curvilinear Relationship.” Mental Health,
Boston: Houghton Mifflin, 2001. Religion and Culture 14, no. 7 (2011): 673–689.
Cashin, Sheryll. Loving: Interracial Intimacy and the Threat to White Gallup. “New Hampshire Now Least Religious State in US.” 2016a. http://
Supremacy. Boston: Beacon, 2017. news.gallup.com/poll/189038/new-hampshire-least-religious-state.aspx.
Clifford, William Kingdon. Lectures and Essays. Edited by Leslie Stephen and Gallup. “Five Key Findings on Religion in the US.” 2016b. http://news
Frederick Pollock. London: Macmillan, 1879. .gallup.com/poll/200186/five-key-findings-religion.aspx.
Conway Morris, Simon. Life’s Solution: Inevitable Humans in a Lonely Gallup. “In US, Belief in Creationist View of Humans at New Low.” 2017a.
Universe. Cambridge, UK: Cambridge University Press, 2003. http://news.gallup.com/poll/210956/belief-creationist-view-humans-new
Curtis, John, and St. John Simpson. The World of Achaemenid Persia: -low.aspx.
History, Art, and Society in Iran and the Ancient Near East. London: I. B. Gallup. “Record Few Americans Believe Bible Is Literal Word of God.”
Tauris, 2010. 2017b. http://news.gallup.com/poll/210704/record-few-americans-believe
Darwin, Charles. On the Origin of Species by Means of Natural Selection; or, -bible-literal-word-god.aspx.
The Preservation of Favoured Races in the Struggle for Life. London: John
Gallup and Healthways. “State of American Well-Being.” 2017. http://info
Murray, 1859.
.healthways.com/hubfs/Gallup-Healthways%20State%20of%20American
Darwin, Charles. Origin of Species. 3rd ed. London: John Murray, 1861. %20Well-Being_2016%20State%20Rankings%20vFINAL.pdf.
Gellately, Robert. Backing Hitler: Consent and Coercion in Nazi Germany. Morris, Ian. War! What Is It Good For?: Conflict and the Progress of
Oxford: Oxford University Press, 2001. Civilization from Primates to Robots. New York: Farrar, Straus and
Giroux, 2014.
Gervais, Will M., and Maxine Najle. “How Many Atheists Are There?”
Social Psychological and Personality Science 9, no. 1 (2018): 3–10. Nola, Robert. “Do Naturalistic Explanations of Religious Belief Debunk
Religion?” In A New Science of Religion, edited by Gregory W. Dawes
Greenwood, Kyle. Scripture and Cosmology: Reading the Bible between the
and James Maclaurin, 162–188. New York: Routledge, 2013.
Ancient World and Modern Science. Downers Grove, IL: IVP Academic,
2015. Norenzayan, Ara, and Azim F. Shariff. “The Origin and Evolution of
Religious Prosociality.” Science 322, no. 5898 (2008): 58–62.
Gregory, Tobias. From Many Gods to One. Chicago: University of Chicago
Press, 2009. Norris, Pippa, and Ronald Inglehart. Sacred and Secular. Cambridge, UK:
Hall, Deborah L., David C. Matz, and Wendy Wood. “Why Don’t We
Practice What We Preach? A Meta-analytic Review of Religious Racism.” Paul, Gregory S. “The Great Scandal: Christianity’s Role in the Rise of the
Personality and Social Psychology Review 14, no. 1 (2010): Nazis.” Part 1. Free Inquiry 23, no. 4 (October/November 2003): 20–
126–139. 29.
Harrington, Daniel. Interpreting the New Testament: A Practical Guide. Paul, Gregory S. “The Great Scandal: Christianity’s Role in the Rise of the
Collegeville, MN: Liturgical Press, 1990. Nazis.” Part 2. Free Inquiry 24, no. 1 (December 2003/January 2004):
28–33.
Harris Poll. “Americans’ Belief in God, Miracles and Heaven Declines: Belief
in Darwin’s Theory Rises.” 2013. http://www.harrisinteractive.com Paul, Gregory S. “The Chronic Dependence of Popular Religiosity upon
/NewsRoom/HarrisPolls/tabid/447/ctl/ReadCustom.percent20Default Dysfunctional Psychosociological Conditions.” Evolutionary Psychology 7,
/mid/1508/ArticleId/1353/Default.aspx. no. 3 (2009): 398–441.
Hayden, Brian, Neil Manuel, and Jennifer Shane. “What Was Brewing in Paul, Gregory S. “The Evolution of Popular Religiosity and Secularism: How
the Natufian? An Archaeological Assessment of Brewing Technology in 1st World Statistics Reveal Why Religion Exists, Why It Has Been
the Epipaleolithic.” Journal of Archaeological Methodology and Theory 20 Popular, and Why the Most Successful Democracies Are the Most
(2012): 102–150. Secular.” In Atheism and Secularity, edited by Phil Zuckerman, 49–209.
Santa Barbara, CA: Praeger, 2010.
Hick, John. An Interpretation of Religion: Human Responses to the
Transcendent. 2nd ed. New Haven, CT: Yale University Press, 2004. Paul, Gregory S. “Why Religion Is Unable to Minimize Lethal and
Nonlethal Societal Dysfunction within and between Nations.” In The
Hurtado, Larry W. Destroyer of the Gods: Early Christian Distinctiveness in the
Oxford Handbook on Evolutionary Perspectives on Violence, Homicide, and
Roman World. Waco, TX: Baylor University Press, 2016.
War, edited by Todd Kennedy Shackelford and Viviana A. Weekes-
Huxley, Julian S. The Individual in the Animal Kingdom. Cambridge, UK: Shackelford, 435–470. Oxford: Oxford University Press, 2012.
Peoples, Hervey C., Pavel Duda, and Frank W. Marlowe. “Hunter-
Jaki, Stanley L. The Savior of Science. Grand Rapids, MI: Eerdmans, Gatherers and the Origins of Religion.” Human Nature 27, no. 3 (2016):
2000. 261–282.
Johanson, Donald C., and Maitland A. Edey. Lucy: The Beginnings of Pew Research Center. “The Global Religious Landscape.” 2012. http://
Humankind. New York: Simon and Schuster, 1981. www.pewforum.org/2012/12/18/global-religious-landscape-exec/.
John Paul II. “The Pope’s Message on Evolution.” Quarterly Review of Pew Research Center. “Millennials in Adulthood.” 2014. http://www
Biology 72, no. 4 (1997): 5, 377–383. .pewsocialtrends.org/2014/03/07/millennials-in-adulthood.
Kennedy, Hugh. The Great Arab Conquests: How the Spread of Islam Changed Pew Research Center. “America’s Changing Religious Landscapes.” 2015a.
the World We Live In. Philadelphia: Da Capo, 2007. http://www.pewforum.org/2015/05/12/americas-changing-religious
-landscape.
Kertzer, David I. The Pope and Mussolini. New York: Random House,
2014. Pew Research Center. “Evolution and Perceptions of Scientific Consensus.”
2015b. http://www.pewinternet.org/2015/07/01/chapter-4-evolution-and
Kitcher, Philip. Living with Darwin: Evolution, Design, and the Future of
-perceptions-of-scientific-consensus.
Faith. Oxford: Oxford University Press, 2007.
Pew Research Center. “In America, Does More Education Equal Less
Lindberg, David C. The Beginnings of Western Science: The European
Religion?” 2017. http://assets.pewresearch.org/wp-content/uploads/sites
Scientific Tradition in Philosophical, Religious, and Institutional Context,
/11/2017/04/25115302/Religion-and-education-FULL-REPORT.pdf.
Prehistory to A.D. 1450. 2nd ed. Chicago: University of Chicago Press,
2007. Pidgen, Charles. “Subversive Explanations.” In A New Science of Religion,
edited by Gregory W. Dawes and James Maclaurin, 147–161. New York:
Margolis, Howard. It Started with Copernicus: How Turning the World
Routledge, 2013.
Inside Out Led to the Scientific Revolution. New York: McGraw-Hill,
2002. Pinker, Steven. Enlightenment Now: The Case for Reason, Science, Humanism,
and Progress. New York: Viking, 2018.
McShea, Daniel W., and Robert N. Brandon. Biology’s First Law: The
Tendency for Diversity and Complexity to Increase in Evolutionary Systems. Potter, David. A Companion to the Roman Empire. Malden, MA: Blackwell,
Chicago: University of Chicago Press, 2010. 2006.
Merriman, John M. Massacre: The Life and Death of the Paris Commune. Purzycki, Benjamin, Joseph Henrich, Coren Apicella, et al. “The Evolution
New York: Basic Books, 2014. of Religion and Morality: A Synthesis of Ethnographic and Experimental

Evidence from Eight Societies.” Religion, Brain and Behavior 8, no. 2 Russell, Robert J. Cosmology: From Alpha to Omega: The Creative Mutual
(2018): 101–132. doi: 10.1080/2153599X.2016.1267027. Interaction of Theology and Science. Minneapolis, MN: Fortress Press,
2008.
Putnam, Robert D., and David E. Campbell. American Grace. New York:
Simon and Schuster, 2010. Scheepers, Peer, Mèrove Gijsberts, and Evelyn Hello. “Religiosity and
Prejudice against Ethnic Minorities in Europe: Cross-National Tests on
RedC Research. “Global Index of Religion and Atheism.” 2012. https://
a Controversial Relationship.” Review of Religious Research 43, no. 3
sidmennt.is/wp-content/uploads/Gallup-International-um-tr%C3%BA-og
(2002): 242–265.
-tr%C3%BAleysi-2012.pdf.
Schmidt, Wilhelm, and H. J. Rose. The Origin and Growth of Religion: Facts
Rees, Tomas James. “Is Personal Insecurity a Cause of Cross-National
and Theories. London: Methuen, 1931.
Differences in the Intensity of Religious Belief?” Journal of Religion and
Society 11 (2010): 1–24. Shajari, Sadegh. Chain Reaction and Chaos: Toward Modern Persia. Lanham,
MD: University Press of America, 2015.
Richards, Robert John. Was Hitler a Darwinian?: Disputed Questions in the
History of Evolutionary Theory. Chicago: University of Chicago Press, 2013. Shapin, Steven. The Scientific Revolution. Chicago: University of Chicago
Press, 1996.
Rubenstein, Jay. “Cannibals and Crusaders.” French Historical Studies 31,
no. 4 (2008): 525–552. Sim, David. Apocalyptic Eschatology in the Gospel of Matthew. Cambridge,
UK: Cambridge University Press, 1996.
Ruse, Michael. “Evolutionary Ethics: A Phoenix Arisen.” Zygon 21, no. 1
(1986): 95–112. Reprinted in Issues in Evolutionary Ethics, edited by Paul Smith, Adam. An Inquiry into the Nature and Cause of the Wealth of Nations.
Thompson, 225–247. Albany: State University of New York Press, 1995. London: W. Strahan and T. Cadell, 1776.
Ruse, Michael. Monad to Man: The Concept of Progress in Evolutionary Stark, Rodney. The Rise of Christianity: A Sociologist Reconsiders History.
Biology. Cambridge, MA: Harvard University Press, 1996. Princeton, NJ: Princeton University Press, 1996.
Ruse, Michael. Can a Darwinian Be a Christian? The Relationship between Stark, Rodney. The Victory of Reason: How Christianity Led to Freedom,
Science and Religion. Cambridge, UK: Cambridge University Press, Capitalism, and Western Success. New York: Random House, 2006.
2001.
Steinkraus, Warren, and Michael Mitias. Taking Religious Claims Seriously: A
Ruse, Michael. The Philosophy of Human Evolution. Cambridge, UK: Philosophy of Religion. Amsterdam: Rodopi, 1998.
Versnel, H. S. Coping with the Gods: Wayward Readings in Greek Theology.
Ruse, Michael. The Gaia Hypothesis: Science on a Pagan Planet. Chicago: Leiden, Netherlands: Brill, 2011.
University of Chicago Press, 2013.
Winzeler, Robert L. Anthropology and Religion: What We Know, Think, and
Ruse, Michael. On Purpose. Princeton, NJ: Princeton University Press, Question. 2nd ed. Lanham, MD: AltaMira Press, 2012.
2017.
Zuckerman, Phil. “Atheism, Secularity, and Well-Being: How the Findings
Ruse, Michael. The Problem of War: Darwinism, Christianity, and Their Battle of Social Science Counter Negative Stereotypes and Assumptions.”
to Understand Human Conflict. Oxford: Oxford University Press, 2018. Sociology Compass 3, no. 6 (2009): 949–971.
TOPIC 13
Human Beings: Theism

Brendan Sweetman
Rockhurst University, Kansas City, MO
This chapter defends the view that theism is a better explanation than naturalism of the remarkable
phenomenon of the human mind and its activities. It discusses arguments concerning dualism and
materialism, consciousness and intentionality, personal identity, and free will and moral agency. Related
issues such as determinism, God and freedom, true beliefs and knowledge, and the significance of “God of
the gaps” objections are also considered.
HUMAN UNIQUENESS
In the course of our day-to-day lives, we are inclined to take for granted the extraordinary nature of
human beings, including our remarkable abilities, capacities, and achievements. Given our appar-
ently humble origins through the evolutionary process, our success in making sense of and manag-
ing our lives and the world is simply astonishing, never mind what the future may hold as the
human race continues to advance. Human beings can write symphonies that seem to express the
essence of the human spirit, construct engines, understand the past, perform remarkable feats in
medicine, develop world communication systems, explore space; we can practice philosophy, sci-
ence, mathematics, and art; we can reach toward God and infinity, and achieve inspiring acts of rea-
soning ability, understanding, love, sacrifice, and moral integrity. Given our early savage nature,
these developments are not only surprising but almost incredible. As Alfred Russel Wallace (a con-
temporary of Charles Darwin) noted, “Natural selection could only have endowed savage man with
a brain a little superior to an ape, whereas he actually possesses one little inferior to that of a philos-
opher” (Wallace 1870, 356).
According to the theist, the nature and achievements of humans suggest two things: that there
is an overall design to the universe in which human beings are most likely the intended outcome,
and that human beings have a higher purpose or telos within God’s creation. This chapter defends
the theistic view, especially as it concerns more specifically the human mind and its activities. For
a defining feature of human beings is that we have consciousness and self-awareness, are capable
of language, logic, abstract thinking, and rational understanding, along with free will and moral
agency, all of which make our intellectual, moral, and spiritual achievements possible.
Thinkers across many disciplines have long been fascinated by the emergence and structure, or
makeup, of human life on Earth—but perhaps none more so than philosophers, for whom these
subjects raise some of the ultimate questions about human existence. Pondering our origin and
nature also prompts philosophers to develop a unified approach to these questions. This means that
it is possible to bring together under the umbrella of the topic of the origin and nature of human
beings many key philosophical issues that are often discussed separately. Questions relating to the
special qualities of humans, such as consciousness, personal identity, reason, and free will, along
with questions of the purpose of human life more generally, prompt philosophers to develop a more
comprehensive view in which these features would fit as part of a coherent theory of reality. In fact,
407
one might say that any complete theory of reality, any world-
KEY CONCEPTS view that one subscribes to, must provide some account of
these key features of human life, features that define so much
of our understanding of ourselves and that enable us to live suc-
Compatibilism
cessfully in the world.
Consciousness The theist argues that these human features, both individ-
Designer ually and, even more so, when taken together in a cumulative
Determinism argument, support a theistic view of reality, and that such a
view is also more reasonable than atheistic alternatives. The tra-
Dualism
ditional understanding of special human qualities such as
Eliminative Materialism higher-level consciousness (including self-awareness and self-
Epiphenomenalism identity), the unique tool of reason that allows us to make sense
Free Will of an intelligible world, along with our notable capacity to be
moral agents, is that such qualities not only make human
Intentionality beings special among all other species but that they point to
Interactionism an overall meaning and purpose to human life that requires an
Materialism explanation. The question of the explanation can be
approached from two directions. The first is a top-down point
Metaphysical Naturalism of view, meaning that the arguments for the existence of God
Methodological Naturalism together with other considerations, such as the inability of the
Mind-Body Problem scientific method to explain the origin of the universe, or the
inability of evolution to explain the design in the cosmos that
Moral Agency
is revealed in the laws of science, support a theistic explanation
Naturalism for the origin and nature of the universe, including human
Nonreductive Materialism beings. But one may look at the issue in a bottom-up way as
Personal Identity well, as this chapter does.
The bottom-up approach begins from a philosophical anal-
Property Dualism ysis of those unique features of human beings already men-
Reductive (or extreme) Materialism tioned and argues that those features cannot be explained, even
Self in principle, on an atheistic approach, and that when taken
individually, and then together, they point toward the conclu-
Soul
sion that a theistic explanation is the most reasonable one.
Substance Dualism The theistic explanation is then further supported by the vari-
True Beliefs ous parts of the cumulative case for the existence of God that
come from other sources, such as the arguments from natural
theology for the existence of a Supreme Being, the moral argu-
ment, the argument from religious experience, the argument
from miracles, and so forth (all discussed elsewhere in this volume). The aim of this chapter is to
show that the human mind and its activities are best explained by the existence of God, and that
this explanation therefore supports a theistic account of reality.
DUALISM AND MATERIALISM
The phenomenon of human consciousness and its activities, along with their relationship to the rest
of the human body, especially the brain, has long fascinated philosophers. The mind-body problem,
as it is often called, raises two central questions: first, what is the relationship between the mind and
the body; second, which other important issues hinge on the answer to this question? The mind
refers to human consciousness (which includes self-awareness and subjectivity), thoughts, beliefs,
experiences, reasoning and logic, imagination, memories, and emotions. The body (or brain) refers
to the physical organ of the brain and its regions and parts (its hemispheres, lobes, cerebellum, hip-
pocampus), as well as its molecular and chemical composition, which includes cells, neurons, synap-
ses, dendrites, blood and other chemicals. The crucial (and challenging) question is: what is the
relationship between the latter and the former? Philosophers who have discussed this question
include, from the history of philosophy, Plato, Augustine, Thomas Aquinas, René Descartes, John
Locke, David Hume, Thomas Reid, and Immanuel Kant, as well as many contemporary philoso-
phers. It is no coincidence that the topic has attracted great interest in this century and indeed is
now one of the hottest topics in philosophy. This is because contemporary philosophers are well
aware of its significance for the larger debate involving theism versus atheism; indeed, a desire to
defend atheism is one reason why the mind-body problem is such a popular area of philosophical
work. The mind and its activities are recognized as posing a special problem for naturalistically
motivated views of reality.
Naturalism is the view that everything that exists is physical in nature, consisting of some con-
figuration of matter and energy, and can, at least in principle, be explained by science. This view has
become increasingly dominant in contemporary philosophy and science. It is a positive thesis in the
sense that it is a claim about what is really the case, what is true in reality; but if correct, it would
entail that God does not exist, that there is no soul, no nonphysical mind, and so forth. In this
way it differs from atheism, which mainly involves negative claims about what does not exist, what
is not true, what is not the case. Naturalism is the modern face of atheism, and has become an influ-
ential, if not dominant, view in certain intellectual circles, especially in Western universities (Sweet-
man 2010). The theist regards naturalism as a predisposition toward a certain view of reality rather
than as a plausible thesis that can be supported with evidence. Some philosophers, such as Thomas
Nagel, have admitted that it is the way they would like the world to be; they then seek explanations
for parts of reality that appear not to be this way, especially the phenomenon of mind and its activ-
ities (Nagel 1997). In a way, it is not an exaggeration to say that naturalism is to contemporary phi-
losophy, at least in the English-speaking world, what theology was to medieval philosophy. It is a
mostly unargued-for backdrop to philosophical work, rather than an established view that then pro-
vides the foundation for future thinking and research.
Although we should be careful to distinguish naturalism from science, it remains closely allied
with science. The reason for this is that it is a positive thesis about reality and so needs to be sup-
ported with positive arguments (rather than simply with criticizing religious views that one rejects).
Supporters of naturalism often appeal to science for their positive evidence, particularly to current
theories in evolution, genetics, cosmology, and neurology.
Naturalism, however, makes philosophical claims and so it requires a philosophical examina-
tion. In fact, those who are inclined to propose a naturalistic account of reality run into a series
of very difficult problems with regard to consciousness and its activities. This is because the human
mind seems to be one area of life that is not accessible to scientific explanation or even investigation
in the way that naturalism suggests. If one cannot explain the mind by means of science, then one’s
thesis that all of reality can be explained naturalistically is seriously undermined. Reflection on the
nature of the human mind suggests that human beings are not simply biological beings. We do have
obvious biological features such as breathing, digestion, and perhaps some of our emotional behav-
ior (though not all of it), but we are also rational, thinking beings who pursue knowledge and truth
and who are capable of moral agency.
There is obviously a relationship between the biological realm and the mental realm, but what
kind of relationship? How is it possible to explain the activities of the mind, especially conscious
experience, subjectivity, and free will to make choices, in terms of the physical? This is sometimes
referred to as the “hard problem” of consciousness (Chalmers 1995). These unusual features also
raise the question of whether we differ in degree or in kind from other species. The theist argues
for the latter. The continued failure of naturalism to explain consciousness represents a serious road-
block to an atheistic view of reality. Naturalism becomes a mere aspiration, a promissory note,
rather than a plausible position, and its theistic opposite becomes increasingly credible, given that
its main arguments are founded on the general claim that the activities of the mind transcend the
physical realm and so the mind as such is not amenable to a scientific explanation in principle.
The various positions that philosophers have taken on the relationship between mind and body
can be broadly divided into two main categories: dualism and materialism. The first category, dual-
ism, was famously defended by Descartes, whose arguments we will mostly concentrate on here

(Descartes 1999). Descartes provided one of the best-known accounts of the relationship of mind to
body in the history of ideas, and some version of this position (though there are important differ-
ences between versions) was held by most major thinkers in the history of philosophy, including
Plato, Aristotle, Augustine, Aquinas, Locke, Reid, and Kant.
Descartes argued in his Meditations that the mind and body are best understood as two distinct
entities, two substances. The mind is a mental, perhaps even a spiritual, substance; the brain is a phys-
ical substance. Moreover, the mind should be understood as a nonphysical substance, and this fact,
according to Descartes, would place it outside of the physical, and therefore the scientific, domain.
He further argued that the mind is independent of the brain, is an entity in its own right; it is not
produced by the brain, and does not necessarily depend on the brain for its existence. The strict dual-
ist holds that the mind and body have a relationship similar to that in a car engine between the alter-
nator and the battery. They are distinct, but related, entities; they work together in the engine, but
one does not produce the other, and if either one were removed from the engine, the other would still
exist. Although not a perfect analogy, this comparison does give a good sense of the dualist position.
Descartes and other dualists make a crucial observation about the mind and its operations that
further emphasizes its independence from the brain. He notes that the mind has causal power over
the brain, a clear indication of its metaphysical status as an entity in its own right. Dualists generally
hold that we know this from attending carefully to what occurs in the process of human decision
making. This is best illustrated by considering a few simple examples. First, let us consider the phys-
ical and mental processes that are involved when one feels hungry. What happens in this case? Well,
we know from our work on the human nervous system that the stomach sends signals to the brain,
and these signals cause a person to start thinking about food. I may begin thinking that it is time for
lunch, and about what I will have for lunch today. Will I have soup and a sandwich or a hamburger
and fries in the college cafeteria? The dualist argument is that the brain prompts the mind to begin
thought processes and reasoning about food, but as one thinks through the options and makes deci-
sions with regard to the task of eating lunch, the mind then causes the brain to send signals to the
body to cause one to move to the cafeteria. But equally well the mind could direct that one not
move toward eating, whether because one is on a diet or a deadline. The ability of the mind to
choose against a strong bodily inclination is a strong indicator that there is real independence here.
An accurate description of the process thus seems to show that the mind and the brain interact,
and that both have causal power over each other. The dualist argues that if one wishes to deny this
point, then one must show that decision making does not involve mental causation of this sort, and
also, crucially, how one’s free decision about what to have for lunch or whether to have lunch could
have a scientific explanation at all (we will return to the issue of free will later). Descartes in fact
held that it would not be possible to figure out the nature of the causal interaction between brain
and mind because the mind is nonphysical and we have no way of studying the nature of the non-
physical. So, although the mind has causal power, it is not ordinary scientific causation; this allows
us to infer that there is a kind of psychophysical causation that is not subject to the usual scientific
laws because one side of it, the mental side, is outside of the physical realm. Descartes himself
regarded the body as a type of machine; the mind is a completely different kind of substance, which
is nonphysical or mental.
The causal power of the mind can also be illustrated, second, by considering more abstract,
higher-level forms of thinking that are not triggered by bodily impulses, such as might be involved
in thinking about the proper relationship between religion and politics in a democratic society, or
how to develop a banking system that would be immune to fraud, or how to test for gas pressure
in a machine, and so forth. These are all interesting puzzles that require abstract thinking, which
will usually have the consequence of initiating some further causal activity on the part of the
agent—for example, the chemist will design and perform an experiment based on an abstract analy-
sis of the task, and so forth. These puzzles also involve causal relationships among mental states
themselves, in addition to the relationships between mental and physical states. Everyday, common-
sense examples such as these, the dualist holds, show clearly that the mind is more than the body,
that it is not reducible to the brain, and that it has causal power over the brain. (We will revisit
the topic of mental causation in a later section.)
The second main category of positions on the mind-body relationship is materialism (some-
times also called physicalism). It too has a long history, and can be found as far back as the work
of the pre-Socratic philosophical school known as the atomists, which included Democritus and
Leucippus. It became more popular during the Enlightenment in the work of such thinkers as Pierre
Gassendi, Thomas Hobbes, and Julien Offray de La Mettrie. Materialism in general is the view that
everything is made of matter, or depends on something that is made of matter. Although material-
ism is not an atheistic position in itself (because this position can be considered apart from its role
in the larger debate), it is usually presented as part of the case for atheism. As noted earlier, one’s
motivation for being a materialist is often to further support the atheistic view.
In the philosophy of mind, materialism has two main versions. The first, which is often called
nonreductive (or moderate) materialism, holds that it is possible that the mind consists of some type
of nonphysical properties, which are not fully reducible to the brain. Nevertheless, these are pro-
duced by, and are therefore dependent on, the brain for their existence, and would cease to exist
if the brain did not exist. Mental operations are therefore a kind of by-product of the brain; they
are sometimes described as supervening on the brain; moreover, the mind does not exist in its
own right, and has no causal power over the brain. This position is also called epiphenomenalism.
The second version is called reductive (or extreme) materialism (or sometimes mind-brain
identity theory, depending on the version under discussion). This view holds that the mind and
the brain are the same thing, that even though we talk of two different things, there is really only
one entity there, the brain, and that all the operations we classify under the heading of the mind,
such as consciousness, self-awareness, beliefs, and so forth are really just sophisticated brain opera-
tions. They are not properties that supervene on brain states, or properties that need brain states
in order to be expressed; they are simply parts of the brain. Extreme materialists acknowledge that
we do not yet know how these operations work or even much about them, but they believe that
we will figure it all out in the future through further scientific research.
It is important to note that there are quite a view variations of all of these positions within phi-
losophy of mind that we do not have space to go into here. These include property dualism (which
holds that mental states are properties of the brain but do not belong to or constitute a substance in
their own right, and are usually understood as having no causal power); and eliminative materialism
(which holds that much of what we describe as belonging to the realm of the mental is really just
brain processes, and therefore we need to realize that our current way of talking about the mental
is likely to be radically mistaken, something we will see more clearly as neuroscience advances)
(Matthews 2005; Robinson 2008; Taliaferro 1994; Murphy 2006).
CONSCIOUSNESS, INTENTIONALITY, AND QUALIA
The debate between dualists and materialists involves a theoretical analysis of mental properties and
activities, and the development of philosophical arguments to show that these properties and activi-
ties cannot be explained by physics and science, in the case of the dualist, or that it is scientifically
plausible that they could be, in the case of the materialist. In defense of dualism, a thinker like Des-
cartes appeals to a number of different arguments; modern dualists and nonmaterialists have
expanded his various arguments to develop what the theist regards as a very strong case for some
version of dualism. Descartes, like many philosophers, was initially very impressed by the straight-
forward logical point that if two things have different properties they must be different things. So
he argued that among the properties of the brain are that it is divisible, is extended in space (which
means it is in principle visible), is measurable, and can be examined and studied like any physical
object. The mind, on the other hand, is indivisible, is not in space, is not measurable, and cannot
be studied directly by means of the scientific method. This establishes a plausible initial case that
we are dealing with two different entities.
In addition, he argued that the essence of mind is thinking, and that it must be therefore a dif-
ferent kind of entity than the body. He was led to this view by his method of doubt, whereby he
placed all his beliefs, including those of common sense, under doubt as part of his attempt to

establish the certainty of knowledge (including the new and growing domain of scientific knowl-
edge). He then realized that, though he could doubt whether he had a body, he could not doubt
that he had a mind: he could not doubt that he had consciousness. As he famously argued, even if
I subject all my beliefs to doubt I cannot doubt the fact that I (the one who is doing the doubting)
exist; I cannot doubt my own being. This led him to his famous cogito argument (“I think, therefore
I am”), which he believed was another indication that the mind belongs to a different order than
that of the body.
Descartes’s approach became influential, and his argument from properties was later expanded
to include other significant and quite peculiar properties of the mental. One of these properties is
the privileged access that we have to our own minds, the claim that our experiences of our own men-
tal states are incorrigible, that we cannot be wrong about our awareness of what we are experiencing,
even when we are wrong about the nature of the objects that we are thinking about. Not only do we
have this certainty about our mental states, but we can have it while knowing absolutely nothing
about the nature or workings of the brain.
Another unusual mental phenomenon is that of intentionality, a feature of the mental that was
given serious attention in the attempt to understand the nature of human knowledge by a number
of important philosophers, including Franz Brentano, Edmund Husserl, Jean-Paul Sartre, and sev-
eral twentieth-century analytical philosophers, including Hilary Putnam, John Searle, Ruth Milli-
kan, and Jerry Fodor (Putnam 1975; Moran 2000; Searle 1983; Millikan 1984; Fodor 1987).
The phenomenon of intentionality is one that we often take for granted in our ordinary, everyday
experiences of consciousness and other mental activities. Intentionality refers to what we can
describe as the “ofness” or “aboutness” of our mental life, of our ideas, beliefs, reasoning processes,
along with attendant notions of truth and knowledge. (The word comes from the Latin tendere,
meaning “to direct” or “point toward” something, and should not be confused with the ethical
meaning of “intention” in English.)
A thought experiment will help to illustrate the odd mental property of intentionality. Think
for a moment about your home, perhaps forming a picture of it in your mind. Or think about what
you will have for breakfast tomorrow morning. If someone asks you, “what are you thinking about?”
you will answer, “I am thinking about my home,” or “I am thinking about having a full Irish break-
fast tomorrow morning!” You are not thinking about your car or your watch. It is a remarkable fea-
ture of the mental that it has “intentional content”—meaning that consciousness is consciousness of
something, that it is, as Brentano noted, directed toward an object: one may think of a car, hope
for a new job, or dream about winning the lottery. Intentionality has causal power: I can intend to
travel to New York City next week, make plans, and then travel there. Now, the dualist or imma-
terialist about the mind argues that intentionality cannot be explained in physical terms, that it is
an odd feature of the mental that is not open to a scientific explanation, even in principle. This is
because it does not seem possible to explain in scientific terms how the atomic or molecular struc-
ture of a physical object (e.g., brain cells) could be about or of or directed toward another object dis-
tinct from it. It does not even seem to make sense to say this of any physical phenomenon. Inten-
tionality (also sometimes referred to as “semantic content”) is a very unusual property and one that
is not found in the rest of nature.
Intentionality is a difficult problem for a materialist view because it draws attention to a feature
of mind that we do not see in the world of physical atoms, chemicals, and scientific laws. It would
seem to require the breaking of new bounds in science to explain how a physical object could be
“about” or “of” or “for” another physical object. Nevertheless, materialists must hold, if they are
to remain materialists, that this aboutness has a physical explanation. Much work since the late
twentieth century has focused on attempts to state necessary and sufficient conditions for how
(and which) mental phenomena exhibit intentionality, but even a successful project of this sort
would not show that intentionality can be naturalized, that is, that it can be reduced to physical
matter and causal laws.
One of the main difficulties facing attempts to naturalize intentionality is the meaning that
seems to be an essential feature of mental states, sometimes referred to as their mental content or
their referent (though these terms are not always interchangeable). One suggestion is that objects
that are independent of the mind must in some sense cause the ideas we have of them; these objects
are then the referent of these ideas, and so some philosophers suggest that it might be possible to
explain intentionality in causal terms. However, certain puzzles arise that are difficult to resolve
on the view that intentionality can be explained in causal terms (Fodor 1987; Millikan 1993).
One of these difficulties can be illustrated by invoking the example of the planet Venus, which is
known as the “morning star” and the “evening star” in different parts of the world. Suppose a per-
son believes that his idea of the evening star refers to the planet Venus; this might sound like he also
should believe that his idea of the morning star refers to the planet Venus. Yet he does not believe
that the morning star refers to the planet Venus because he does not know that the morning star
and the evening star are the same star. If this person really wishes to see the evening star but is indif-
ferent toward the morning star, and is then told that the morning star is now visible in the sky, he
will not rush outdoors to see it. It seems that these interesting variations of meaning are an essential
part of mental life, and that they involve a particular point of view. The problem arises because most
attempts to explain intentionality in naturalistic terms appeal to a causal account, which claims that
a mind-independent object causes an idea in the mind, and determines what that idea is an idea of.
Yet, in the case just mentioned, the person can look at the morning star and not believe he is seeing
the planet Venus, and still he must be seeing it! The opposite can also occur: I can look at an object
under certain conditions that might mistakenly cause a false belief about another object (for example,
I may see a rabbit in the dark and believe it to be a cat). So the content of certain beliefs and mental
states is difficult to explain on causal theories of intentionality.
Yet further difficulties for naturalistic theories of intentionality are caused by related cases, espe-
cially those where one thinks that one’s idea has reference but it does not; for instance, in cases
where one is hallucinating, or cases where nothing is present but one mistakenly thinks one is seeing
a real object. There must be a difference between cases where the object really exists and where it
does not (and so between our two ideas and their “aboutnesses,” the idea of the real object, and
the idea of the object that is not actually there), but developing a theory of intentionality to cover
both kinds of case has proved elusive. Moreover, reflecting about our thoughts themselves can raise
similar problems, if objects in the world are supposed to be causally responsible for the content of
thought.
Analytic philosophers in particular have attempted to develop various ingenious ways around
these problems, but without much success (Dretske 1995; Millikan 1984). One attempt involves
arguing that it may be possible somehow to explain intentionality separately from consciousness,
so that even if we were to agree that consciousness is nonphysical, intentionality could still be phys-
ical (though such attempts would still need to account for the “aboutness” of mental states). It is
also counterintuitive, to say the least, to suggest that we could make sense of conscious states with-
out introducing the experience of intentionally. Other philosophers have focused on the notions of
reference and truth, in an attempt to lay out necessary and sufficient conditions that account for
these notions while minimizing the appeal to mental content or meaning as much as possible. This
research project is intriguing and still ongoing, but it is fair to say that it has run into considerable
obstacles (Searle 1983; Siewert 1998).
Sometimes materialists will suggest that nonmental phenomena can exhibit intentionality, such
as pictures, physical symbols, and perhaps words or sentences, but these are not really intentional
because their aboutness in fact comes from their interaction with the human mind, and so it is
the mind that is intentional and not the physical object. It has also been suggested that intentional-
ity might be some kind of illusion produced as an aftereffect of the mental, but this aftereffect
would still require a scientific explanation; this suggestion seems merely to shift the problem rather
than solve it.
Perhaps, a nonreductive materialist might claim, it is possible that a physical object, one as
sophisticated as the brain, could produce a nonphysical effect that would have as one of its features
the phenomenon of intentionality. What is required is an explanation of intentionality in terms of a
scientific description involving an account of the workings of the human brain, an explanation in
terms of physical properties and scientific causal laws. The nonmaterialist argues that this cannot
be done, even in principle. Robert Adams has developed this nonmaterialist argument by appealing

to the uniqueness of what are sometimes called qualia (Adams 1992). Qualia is a technical term in
the philosophy of mind to describe various human feelings and experiences; for example, what it is
like to experience pain, say from a toothache or a broken bone, what the experience of sorrow is
like, what the experience of seeing the color red is like, and so on. When one has the experience
of the color red (say, when looking at a red rose), does the experience involve nonphysical, mental
properties, or physical properties, or perhaps both? Adams argues that philosophers sometimes miss
the point of his question, or deliberately downplay its significance. Some tend to confuse this ques-
tion with a different question about whether the color experience is correlated with a particular
brain state, or pattern of brain activity, and also with the question of whether the brain state or pat-
tern then causes the experience. These are obviously important questions but miss the most impor-
tant one, according to Adams.
Suppose for the sake of argument, Adams contends, that we grant to the materialist two hotly
debated and controversial points: that the experience is correlated with a corresponding brain event
(every time you see red you are in the same brain state), and furthermore, let us also agree that the
brain state causes the experience of red, a standard materialist claim. Nevertheless, Adams argues,
the brain state is not the same thing as the experience of red. If materialism were true, we would
require a scientific causal explanation for the experience of redness. How would we go about getting
such an explanation? Adams says that, from the scientific point of view, it seems any progress that
we could make would be in learning more about the brain state (its internal structure, composition
of its parts, relation to other parts of the brain, etc.) that causes the experience of redness. Suppose
we were to discover the complete physical makeup of the brain state—would this then also be an
explanation for what it feels like to experience redness? It seems not. We would almost have to
undergo the experience of seeing red as a result of studying the scientific account of what causes
the experience. The account would almost need to include the experience of redness in some way,
and this is not plausible.
Nagel (himself a materialist) has suggested that it does not seem possible logically to provide a
third-person (objective, scientific) explanation of what is essentially a first-person point of view, such
as the experience of seeing red, or of feeling pain, of what Nagel describes as “the subjective charac-
ter of experience” (Nagel 2004). Nagel and Adams conclude that these subjective experiences can-
not in principle be explained in scientific terms. It is not simply that we do not yet have the expla-
nation for recalcitrant phenomena, but that there seems no possible way to obtain an explanation.
These philosophers present this problem as an intractable one for materialist views, and as a strong
argument in favor of the view that the human mind, and so the human person, cannot be reduced
to physical properties and causal laws. Adams draws the further conclusion that the best explanation
is the one given by Descartes and especially Locke, who held that matter could not possibly produce
thought because of its physical nature. Locke added that the best explanation for why the mind
works the way it does is that it has been designed to work that way by God (Locke [1689] 1979;
Adams 1992; Swinburne 1991).
REASON, KNOWLEDGE, AND TRUTH
We have been considering some of the remarkable properties of human beings and whether their
nature can be the basis of a reasonable argument for a supreme being, as both designer and cause
of these properties. Although the nature of consciousness and of intentionality are notable
features of life, in another sense they are obvious qualities of human nature. Consciousness and
self-awareness seem to be part of the essence of what it means to be human, as Descartes pointed
out. However, sometimes we overlook the fact that human consciousness exhibits other significant
properties that make it possible for us to survive and indeed thrive in the world, especially over time.
These include our capacity for reason, knowledge, and objective truth about ourselves and the
universe. The theist argues that these phenomena are so remarkable that they require an explanation
and that a theistic explanation is best; in short that these human phenomena were created by an
intelligence to enable human beings to pursue meaning, truth, and purpose in life on the way to
the larger goal of salvation.
One of the things that distinguishes man from other species is our capacity to reason about the
universe. Reason involves the ability to discover knowledge about reality merely by thinking about
it. Knowledge and truth are often used interchangeably in philosophy. Knowledge may be under-
stood as the accurate representation of reality, on an appropriate basis of thought and experience;
truth refers to the way things really are independently of our beliefs, opinions, or cultural perspec-
tives about them. Knowledge involves truth, but we sometimes use the word truth to describe the
nature of reality even if knowers had never existed. Be that as it may, the main point is that the
human mind can discover what really obtains in the world outside the mind. Knowledge also
involves beliefs; a belief is a proposition about reality, which may or may not be true. Indeed, a
key feature of beliefs is that they can be true or false, rational or irrational, depending on whether
or how much they match up with reality outside the mind.
It is not clear how a brain state, or any physical process, could be true or false, rational or
irrational—surely it just is? And this seems to be yet another example of properties that belong
to the mental but not the physical, thereby showing that they are different. Our cognitive pro-
cesses are so unusual that many philosophers have proposed the view that they suggest a harmony
between mind and world that did not come about because of accident but is likely the product of
design. It is a remarkable fact that the mind can form true beliefs, beliefs that tell us the way the
world really is, and that we can use these beliefs to survive, to improve ourselves, as well as to
think about higher-level issues such as the composition of the atom, the origin of the species,
the nature of mathematics, and the origin of the universe. Philosophers who have drawn attention
to these phenomena include Thomas Aquinas, Richard Taylor, C. S. Lewis, John Henry New-
man, and, in recent work, Nagel and Alvin Plantinga. If the human species originated by chance,
as the naturalistic view claims, it seems to be just a fortunate accident that we ended up with cog-
nitive processes that give us knowledge and truth about reality, a conclusion that many thinkers
regard as not credible.
Plantinga has taken the argument a step further to propose that, if evolution together with the
naturalistic worldview of which it is often a part were true, then the probability that our cognitive
faculties are reliable would be low (Plantinga 2011; Nagel 2012). But then the belief in the truth
of evolution together with naturalism would itself be undermined, and would be, according to Plan-
tinga, self-referentially incoherent! By our cognitive faculties, Plantinga is referring to our faculties
for gaining true beliefs about the world and distinguishing them from false beliefs, faculties such
as perception, intuition, sympathy, introspection, testimony, and induction.
Plantinga’s argument is straightforward. He claims that, on an assumption of evolution and
naturalism, the probability of our beliefs being reliable or true coming from these sources is low.
The assumption of naturalism is important, because it means that our cognitive faculties came about
by chance, and that a chance process is not concerned with and is unlikely to lead to reliable cogni-
tive faculties (a conclusion, as Plantinga notes, that is acknowledged by many naturalistic philoso-
phers, including Patricia Churchland and Barry Stroud) (Plantinga 2011). So, those who believe
both evolution and naturalism inadvertently have within their own position a “defeater” for the reli-
ability of their cognitive beliefs, that is, their cognitive beliefs are undermined. But this defeater
undermines the reliability of all one’s cognitive beliefs, including the belief in evolutionary natural-
ism itself. Plantinga’s basic point is that, if our beliefs came about because of evolutionary natural-
ism, we could not therefore trust any belief that comes from this basis, including the belief that evo-
lutionary naturalism itself is true. He thinks that Darwin himself considered this possibility, as
suggested by remarks such as the following: “With me the horrid doubt always arises whether the
convictions of man’s mind, which has been developed from the mind of the lower animals, are of
any value or at all trustworthy. Would any one trust in the convictions of a monkey’s mind, if there
are any convictions in such a mind?” (Plantinga 2011, 316).
A common reply to Plantinga’s argument is that it is just obvious that our beliefs are reliable,
and that they contribute to adaptive behavior and therefore to our survival and development as a
species. Doesn’t this count as good evidence that our cognitive faculties are reliable, and so can’t
we then develop an argument for how they came about, which some believe is as a result of evolu-
tionary naturalism? Even if such a view has not been proved, his critics argue, it is certainly

reasonable to believe. Plantinga rejects this obvious rejoinder as missing the main point of his argu-
ment. He believes that we need to be concerned not with whether we have reliable cognitive pro-
cesses or with whether these processes are adaptive (he agrees they are), but with whether we can
regard our beliefs as reliable if we also believe that they came about according to evolutionary natu-
ralism. Central to Plantinga’s argument is his claim that Homo sapiens or indeed any other species
could have false beliefs and yet still have adaptive behavior. This is the heart of his argument,
because one obvious objection to his view seems to be that if we had false beliefs we would not sur-
vive for long. Plantinga thinks that this is not true and points out correctly that many lower species
that have no beliefs at all have adaptive behavior, and others, if they have beliefs, may have unreli-
able, inaccurate, or false beliefs.
He develops this point by accepting the materialist’s understanding of belief, at least for the sake
of argument. On this view, beliefs are neural structures in the brain with electrochemical
or neurophysiological properties, but they also have mental content; for example, the belief that
“Marcel Proust is more subtle than Louis L’Amour” has two features, according to both reductive
materialists and nonreductive materialists, (1) a particular brain state with a certain neurological struc-
ture, and also (2) content (that “Marcel Proust is more subtle than Louis L’Amour”). The content of
the belief on the nonreductive view is said to supervene on the physical properties and is caused by
the physical properties (let us say); on the reductive view, it somehow is the same thing as the physical
properties, meaning that it is (somehow) a part of (and is made up of) the neurological properties.
Plantinga notes that according to evolutionary naturalism it is the neurological properties that
cause adaptive behavior, and not the beliefs associated with them. He further suggests that on an
assumption of evolution and naturalism we have no reason to believe that the associated beliefs
are true. He makes a crucial distinction between “indicators” and beliefs, arguing that species have
indicators that are a kind of instinct that is attached to the neural properties. These indicators are
adaptive but do not require that true beliefs be associated with them. The belief associated with a
particular indicator could just as easily be false. On nonreductive materialism, the neurological
properties are also selected, according to evolutionary theory, because they lead to adaptive behavior
and not because they lead to true beliefs. The same story is true for reductive materialism because
beliefs that are a physical part of the neurological properties do not need to be true in order for these
properties to be adaptive.
Plantinga illustrates this point with an example of a gazelle who escapes a predatory lion. The
neural properties in the brain of the gazelle are presumably selected for adaptive behavior and there-
fore help the gazelle survive, but the gazelle does not have true beliefs. One might be tempted to
reply that it does not work the same way with human beings. Plantinga, however, reminds us that
this is the way natural selection is supposed to work in nature and in millions of species. (Indeed the
objection sounds like an ad hoc move and is not consistent with current claims for evolution.) Mil-
lions of lower-level species lack beliefs altogether, and yet they all exhibit adaptive behavior, accord-
ing to the theory. So, isn’t it likely to be the same with human beings, especially when we take into
account that it is probably only lately in our evolutionary development that we formed the capacity
for true beliefs, if the full evolutionary story (together with naturalism) is true? In our early stages we
either had no beliefs, inaccurate reasoning processes, or even false beliefs, and yet our behavior was
still adaptive.
We also need to be careful not to fall into the temptation of a circular argument that has always
been a concern for evolutionary theory: according to the theory, if a species survives, it is because its
behavior is adaptive, and if its behavior is adaptive, this then confirms the truth of the theory. This
raises a general point about whether behavior and neural pathways are connected in such a way that
they lead to the survival of a species (as the theory claims), something that is far from established.
Or to put this point another way: the fact of adaptive behavior is one thing, its explanation is
another, and the explanation that it was designed this way is, as Plantinga notes, more plausible
than any atheistic explanation that assigns to chance a large role in the process. The arrival in the
universe of conscious observers like us, who not only understand the process that gave rise to them
but who have ever-increasing control over this process, is an enormously significant development,
and one that suggests intelligence and purpose (Sweetman 2015). If we regard our cognitive
faculties as giving us true knowledge about reality, then we must take them as being designed, and
design, of course, requires a designer.
PERSONAL IDENTITY AND LIFE AFTER DEATH
Many philosophers have given consideration to the question of personal identity, especially to how
to make sense of the idea of human identity over time, a particularly perplexing problem. The exis-
tence of a substantial personal self also raises the question of whether this self may be able to survive
bodily death (as the vast majority of people believe). These questions have an obvious bearing on
the issue of theistic versus atheistic explanations of human existence. The conventional view is that
human beings have direct experience of a unified, substantial self that consists of conscious aware-
ness, thoughts, ideas, beliefs, memories, and so forth. There is disagreement about whether one’s
body is a necessary part of one’s personal identity; on one hand, self-understanding and a strong
sense of self seem to be possible without reference to the body, and yet on the other hand some exis-
tentialist philosophers have held that our body is an important part of who we are, of how we expe-
rience the world, and of our identity for others. It has also seemed to many that consciousness may
be able to survive the death of the body.
Locke argued that we could distinguish the identity of the “person” from the identity of the
“man,” where the latter would include the body but the former would refer only to the nonphysical
mind. Locke thought that mental life was sufficient for personal identity; he defined a person as a
“thinking intelligent being, that has reason and reflection, and can consider itself as itself, the same
thinking thing, in different times and places; which it does only by that consciousness which is
inseparable from thinking, and as it seems to be, essential to it” (Locke [1689] 1979, 312). He fur-
ther held that memories were a necessary condition for personal identity, for our deep-seated sense
of a psychologically continuous self over time. Refinement of Locke’s view has suggested that mem-
ories need not be always accessible for personal identity to be maintained (Martin and Barresi
2003). Locke captures the commonsense view that our personal identity consists of a substantial
unified self, that our present and past experiences belong to one and the same person, even though
our conscious states, perceptions, beliefs, even memories are constantly changing.
There are several questions raised by this discussion: first, is this a correct description of per-
sonal identity, and do we have such an identity?; second, could the self survive after bodily death?;
third, how do these questions bear on the theism/atheism debate? We do not need to get bogged
down debating the nature of identity as a metaphysical principle. Needless to say, philosophers dis-
agree and have often generated puzzles for any definition of identity that is proposed in general
metaphysics, so it will suffice to note that it is very difficult to define identity exhaustively because
there always seem to be counterexamples. One obvious definition is “sameness over time,” yet many
things that we take to be the same over time sometimes undergo radical changes; for example, all of
the wood in a ship may be changed over a long period, and even in the human body many cells are
renewed in a relatively short period, which would suggest that identity involves something more
than maintaining continuous properties over time. It seems best to conclude, as some metaphysi-
cians do, that identity is real but indefinable, and it is wise to reflect on our experience of identity
and not get distracted with logical conundrums.
Some philosophers deny that human beings have a personal identity over time or a substantial
self. This view was held by Hume, Sartre, and, in more recent times, Derek Parfit (Hume [1739]
2007; Sartre 1991; Parfit 1984). Influenced by empiricism, Hume held that every idea must come
from an impression on which it is based, but we have no impression of a substantial self. Impres-
sions connected with the self are fleeting and changing, not constant and invariable, and so, though
we believe we have a substantial self to which individual impressions belong, Hume was skeptical of
its existence because there is no corresponding impression of it in an empirical sense. Sartre argued
that our intentional consciousness is exhausted by our present intentional states; each present con-
scious state carries a memory of the former state and projects to the next state, but there is no sub-
stantial self to which each individual state belongs. Consciousness is always at a particular time, and

at that time it is exhausted by its object; if the object is removed there is nothing left. For Sartre,
consciousness is nothingness. Parfit’s approach is based on an appeal to neurology. He notes the fact
that it is possible to split the cerebral cortex into two and have more than one center of conscious-
ness, and this, together with other neurological evidence, prompts him to conclude that there is
likely no substantial self.
Sartre’s view, and indeed Hume’s too, is based on a kind of phenomenological description of
the self, an attempt to describe carefully what we experience. The counterargument to Sartre and
Hume is that they have misdescribed our experience of consciousness. When we attend closely to
the nature of our conscious life, we are aware of a unified, substantial self over time through intro-
spection. Our conscious experience includes memories, thoughts, and beliefs, free decisions (dis-
cussed in the next section), and a strong sense of a unified self that exists through time and change.
It is an important part of the general case for nonmaterialist views that we have a strong experience
of a substantial unified self and that this experience can count as evidence in the absence of scientific
alternatives. We cannot dismiss it under the supposition that materialism is true and adopt the atti-
tude that there must be something badly mistaken about our experience. Materialists such as Paul
Churchland glibly dismiss our psychological experiences under the disparaging name of “folk psy-
chology,” but in the absence of solving the very difficult conceptual problems that accompany a
materialist view of consciousness, or in the absence of actual scientific evidence, we must regard
such an attitude as wishful thinking (Churchland 1984).
The materialist attempt to deny that there is a substantial self is a counterintuitive view and
another indication that materialism is false. It would also leave us with problems concerning how
to explain memories (whose memories of the past am I recalling), and also why we should be con-
cerned about the future (say future goals or plans) if it is not one and the same self that set the goals
and will experience their fulfillment. Moreover, punishment and responsibility seem to require that
personhood be understood as a substantial self existing over time, otherwise why punish a person for
crimes committed in the past? A nonphysical explanation for personal identity is what all the evi-
dence points to. Moreover, only someone antecedently committed to a materialist view could take
seriously the notion that there is no self, the theist argues.
Can the existence of such a substantial self support an argument for immortality, and thereby
add to the general case for theism? Many philosophers have thought so, including Plato, Descartes,
and Locke. Theistic philosophers and theologians believe that dualist arguments show that the mind
has a certain independence from the brain, and it follows that it is possible that it may be able to
survive bodily death. Although there is clearly a radical difference between the body and the mind,
and arguments like Parfit’s appeal to the fact that the mind and the brain certainly work together in
this life, there is no logical objection to saying that at bodily death the mind could survive. Because
the mind also includes consciousness, subjectivity, and memories, this would also mean that it
would be me that survives—personal identity would be maintained in the afterlife.
Materialists sometimes invoke what can be called a “brain injuries” argument to cast doubt on the
survival of consciousness after death. This is the argument that when the brain is injured (say by
trauma, stroke, or illnesses such as Alzheimer’s disease), there is often a corresponding loss of mental
function, and so it is reasonable to conclude that the mind is dependent on the brain for its existence.
This is an important argument, but what does it actually show? There are a number of logical possibil-
ities. It could mean, as the materialist claims, that the brain causes the mind (or a group of mental
properties, perhaps) to exist, and so the mind needs the brain to exist. Or it might mean that the mind
needs the brain to function, but not to exist (just as the car battery needs the alternator). And just as a
car battery could operate if it was placed in an alternative electrical configuration that did not involve
an alternator, logically it could be the same with the mind. All we can conclude with certainty there-
fore from the brain injuries argument is that the brain is a necessary condition for mental life in this
bodily existence, but it does not follow that it is a sufficient condition for mental life, nor that mental
life could not occur outside the realm of bodily existence. It could still be the case that the mind can
function on its own, without needing brain events, and that this can happen in an afterlife.
Descartes thought that the nature of consciousness suggests that it does survive the death of the
body. It is also important to remember that other arguments can be offered for immortality
independently of the mind-body discussion, such as the moral argument and the argument from
design. Because we have several independent arguments for immortality, the theist holds, they rein-
force each other with cumulative effect, and strengthen the overall case for immortality.
This discussion prompts us to consider the nature of the soul, and its relation to the body and
the mind. Many philosophers have believed that the soul is a nonphysical entity or principle, and
some simply identified it with the mind; others (such as Locke) thought the mind was part of,
but not identical with, the soul (Green and Palmer 2005). The soul is what animates—in the lan-
guage of Plato, Aristotle, Thomas Aquinas, and others—the physical matter in the body. Indeed,
the Latin word for “soul,” anima, gives us animates, and the Greek word, psyche, gives us psychology.
Aquinas has suggested that the “the soul is the principle of life” (Thomas Aquinas 1964). Perhaps
also, according to some thinkers, including Aristotle (in De Anima), it is not just human beings that
have souls, but lower animals and even plants may have a type of soul in the sense that they are
alive, so the soul may explain their animation. Descartes later denied this view and held that only
human beings have souls, and that animals (and indeed the human body) are best thought of as
machines (among other reasons, he held this view because he thought that the developing scientific
method showed great promise in approaching the study of the body as a machine). The relationship
between the soul and the mind is more complicated, and many traditional religious thinkers did not
fully work out their views on this matter. Aristotle thought the soul could not exist independently,
but he did propose that thought could exist independently of the body. Plato, Augustine, Aquinas,
and Descartes believed the soul could exist independently of the body, but Aquinas argued that this
would not be its natural state, and eventually it would be reunited with the resurrected body by
means of divine action (Davies 2004; Koterski 2009; Feser 2005).
FREE WILL, DETERMINISM, AND MORAL AGENCY
For many philosophers (and of course nonphilosophers), free will is an indispensable feature of the
human person (Kane 2005). Indeed, human life as we experience it is unthinkable without reference
to free choice. Free will is not only an important concept in the religious view of the world but a
central feature of any view of the world, including the atheistic one, even though we mostly take
it for granted in the course of ordinary living. It takes only a moment’s reflection to realize what
an extraordinary and complex property of human beings free will is, and its existence is regarded
by many as a decisive argument against any theory of materialism about the human mind.
Free will may be defined as the ability of human beings to make real choices between genuine
alternatives, choices that are not determined by the operation of atomic or molecular particles or
combinations of particles that constitute the physical structure of the brain, along with the relevant
scientific laws. Although there appear to be some cause and effect processes involved in choice, there
also has to be an essential noncausal aspect to it, otherwise it would not be free. Free will is not the
same thing as chance, because human will and intention are involved and so one’s choices are
directed toward a particular end; whereas chance means that events could have gone otherwise than
they did in fact go because their causes occur randomly. Nor is it the same thing as instinct (which
we find in animals), because it can involve a process of rational deliberation. It must also not be
confused with freedom of action, because action can be restricted by external constraint; the ability
to carry out an action is to be distinguished from the ability to will an action.
We can illustrate with a few concrete examples of the use of free will in ordinary life. Let us
return to our earlier example of feeling hungry and examine it from the point of view of the phe-
nomenon of freedom. As in our earlier example, one feels hungry and, speaking very generally, this
biological event sends signals through the nervous system to the brain, and the brain then causes
one to start thinking about food. Working in my office at 11:00 a.m., I may start thinking that it
is time for a cup of tea and a snack. I decide to go downstairs to the coffee shop, where I will order
tea and a doughnut. I won’t order a scone, because I had one yesterday. I then get up from my
desk, leave my office, and take the elevator downstairs. It is important to attend closely to what
has happened here. My body has initiated the process of my thinking about food. I have undergone

a physical process first (though whether the process is totally physical even at this initial stage is not
certain), followed by complex mental operations involving thoughts, beliefs, images, and reasoning
concerning my tea break. I have made several rational decisions leading to my exercising causal
power over my brain and body in order to physically move my body to the coffee shop.
The decisions I have made are arrived at freely in some mysterious way as part of the mental
process of thinking about my 11:00 a.m. tea break. But I do have reasons for these decisions; for
example, the reason I am having a doughnut today and not a scone is that I had a scone yesterday,
and also because this morning I feel like a sweet pastry. Yet these reasons do not causally compel me
in the scientific sense to choose the doughnut; I could just as easily order the scone. Nor, obviously,
do I choose to buy a doughnut by chance. Unlike a machine that has been programmed to select a
predetermined option, I am really free to make a decision, whereas the machine’s “choice” would be
simply the last step of a fully causal process. Proponents of immaterialist views and those committed
to the doctrine of free will argue that it is not possible to give a scientific account of the process of
free decision making because this is a contradiction in terms. It would be asking for a scientific,
causal account of a phenomenon that is not subject to a causal explanation. What we can therefore
say about free choice is that it is a real phenomenon but outside of physics and scientific law, and
therefore beyond the scientific method. This is not a criticism of the scientific method but simply
a recognition that it is limited to the physical realm, and that there are some things beyond the
physical realm.
This is a simple case taken from ordinary human experience, but free choice is present in so
many of the decisions that we make as human beings. I can choose to think about higher-level,
more abstract problems, say, the problem of universals in metaphysics or which theory of literary
criticism is the most insightful, or I can decide to go to the game tonight, or to take a course in
photography, or to pursue countless other endeavors; all of these decisions are based on my belief
in and experience of free will. In fact, such freedom goes even deeper than this in human experi-
ence. It is the foundation of love and friendship, and rewards and punishments. Freedom is the
basis of morality, because the whole enterprise depends on human beings having a real choice
between morally good and morally bad alternatives; as a result, civil law, responsibility, and punish-
ment also depend on the prior belief that human beings are free, because it is unfair and unjust to
hold somebody responsible for an action if that person was compelled to do it (this claim of unfair-
ness itself presupposes free will). The dualist argues, then, that because human beings have the
power to make free choices, materialism is false. This is a very difficult problem indeed for materi-
alists, who appear to be faced with saying that, because all our actions are rooted in the brain and
central nervous system, then all our “choices” are really explicable in terms of scientific causal laws
and the structure of the brain on which these laws operate. We would be like sophisticated robots,
whose very operations are determined by causal sequences operating according to scientific laws.
There appears to be no place for free will in a naturalistic universe.
Materialists are reluctant to give any causal status to those phenomena that make up the human
mind. This is not a problem for extreme or reductive materialists, given that they deny altogether
the existence of the nonphysical (although they still have to explain free will itself); but it is a prob-
lem for moderate or nonreductive materialists because they usually deny any causal power to the
nonphysical aspects of consciousness, such as mental states, thinking, intentionality, perhaps even
reasoning. A similar problem confronts property dualists, who usually deny any causal power to
the nonphysical properties. This seems to make our thinking and reasoning redundant, a view that
is impossible to believe. Sometimes materialists will defend such a view by saying that it is difficult
to conceive how a nonphysical property could have a causal relationship with a physical part of the
brain. Indeed, this criticism is often raised about substance dualism as well, and is one that was
advanced against Descartes’s position in particular. Often referred to as the problem of interaction-
ism, it was an issue that Descartes himself discussed.
Some philosophers think there is some kind of conceptual problem involved in the claim that
the mind could cause the brain to act. However, it is hard to see where the problem lies. For if
the objection involves a demand for a scientific account of the causal interaction, this is a contradic-
tion because the phenomenon of free will is precisely not subject to a causal, or a scientific,
explanation. This mistake is made by Taylor, who suggests that “to say that a diffusion of sodium
ions [in the neurons in the brain] was caused by an ‘idea’ is not to give any causal explanation at
all” (Taylor 1992, 23). This is true if one means a “scientific causal explanation,” but this is pre-
cisely what the immaterialist denies can be given (Taliaferro 2005). Although human reasons are
causes in some sense, they are not causes in the scientific sense. If one means that one wishes to
know from the side of the brain, as it were, how the causal process works, one may discover more
about this as neuroscience advances. Descartes suggested that, from the physical side, it may be
the pineal gland that is the conduit, as it were, to the physical from the mental. He may have been
incorrect, but this is a minor and irrelevant objection to his main point—that the mind causes the
brain to act in some way. Are there perhaps psychophysical laws governing at least some aspects of
mind-brain interaction? There may well be, but they would not be the whole story in that free
choice does not involve fully deterministic causes. The human person is a rational agent, and a
rational agent makes decisions based on reasons that are not causally compelling in the scientific
sense (that is, they are not sufficient conditions for the action to occur) but that nevertheless can
result in the action (O’Connor 2000). The problem from the materialists’ side is that they appear
to fall into the view that there is no free will at all.
This radical view, known as determinism, is one that some materialists embrace, but it is also
one from which others shrink because they are honest enough to see that it leads to great difficulties.
Determinism may be defined as the view that every current event in the universe and state of the
universe is determined by previous events and states of the universe. Applied to the physical uni-
verse, this would mean that the current state of everything on Earth was caused by earlier states
in history, and that these states in turn were caused by preceding states, all the way back to the
earth’s formation. On this view, no event occurs in the universe that is not brought about through
causal interaction with previous events, right back to the beginning of time. Of course, we may not
know the cause of a specific event, according to the determinist, but it always has a cause. What
does this deterministic approach to the understanding of causal happenings in the universe mean
when it is applied to the workings of the human mind?
Both reductive and nonreductive materialists hold that all causal events take place in the physi-
cal brain; moreover, the brain consists of physical matter that behaves according to causal laws. In
this way, the mind is like any other organ in the body, only much more sophisticated in its physical,
chemical composition and its operation. But as with the human heart, for instance, the current state
of one’s brain (and mind) is determined by its past states, present structure, and specific environ-
ment, and by scientific laws. This explanation would also apply to our so-called free decisions.
The determinist argues that human actions are not really free because there is a causal process going
on that results necessarily in the action that we take, and our belief and our feeling that our actions
are free are therefore illusory and mistaken. For example, when we consider whether to take a coffee
break, the determinist argues that our “choice” is in fact caused, and so we are compelled to do what
comes at the end of the causal process, just as a machine must operate in the way it has been
mechanically designed to (assuming it is in proper working order). This is true of the machine no
matter how sophisticated its operations (as in, say, an advanced robot); so it is with human action,
claims the determinist. We may believe, and act as if, we are free, but really we are not. The deter-
minist does not claim to know how the process of decision making in human beings works in terms
of specifics. They see it as a research program for the future, to be developed as the field of neuro-
science advances, but they are committed to the general position that this is how it must work. It is
not so much perhaps that determinists set out to deny free will; it is more that they cannot see any
way to fit it into a worldview that holds that everything is physical in nature and governed by causal
laws. So, despite our strong experience of free choice, this, according to the determinists, is an illu-
sion.
The theist argues that the evidence for human free will is very strong, and that human existence
is almost unthinkable without it. So it is not surprising that determinists have been disturbed not
only by the counterintuitiveness of their position but by the consequences the doctrine of determin-
ism has for human life in general. Because of this, they feel they have to put forward some other
explanatory account of human action and decision making that offers something more than (hard)

determinism. Some thinkers have developed what are called compatibility theories of freedom as a
way to hold us responsible for our decisions, because it seems that responsibility (as we have noted)
must be abandoned if free choice is abandoned, a position many regard as a reductio ad absurdum
argument against the denial of freedom.
Compatibilism is based on the view that, though determinism is true, it might still be possible
to defend some version of freedom, especially one that allows for moral responsibility (this general
view is sometimes called soft determinism). The view goes back at least to Hobbes; one of the most
sustained contemporary versions of it is to be found in the work of John Martin Fischer and Mark
Ravizza (Fischer and Ravizza 1998). These authors claim that the notion of “alternative possibilities”
appears to be central to the notion of moral responsibility, that to be responsible a person must have
a genuine choice between various possibilities in his experience. For example, we hold a student
responsible for cheating on an exam because we believe he had a choice between cheating and not
cheating. If the student was hypnotized and then somehow caused to cheat, we would not hold
him responsible.
Fischer and Ravizza wish to argue that we do not need “alternative possibilities” to be morally
responsible. They defend a “compatibilist” view of moral action, one in which it is compatible to
hold that all events in the universe, including human actions, have causal explanations (because
determinism is true) and yet that human actions may still be regarded as free. Determinism under-
mines moral responsibility because it would follow from it that we have no choice when faced with
alternatives but rather are compelled through a combination of scientific law and the nature of phys-
ical matter to follow one course over the other. The Fischer and Ravizza view revolves around trying
to draw a meaningful distinction (by appeal to contrived, implausible examples, first developed by
Harry Frankfurt) between cases in which we do something “freely”—in ordinary instances of choos-
ing something where we seem to have some control over what we do, such as choosing what to have
for lunch—and cases in which we might do something because we have been hypnotized, brain-
washed, or coerced.
Yet this view prompts the question: is there a meaningful difference on a determinist thesis
between “deciding” to have a particular sandwich for lunch, or being hypnotized and therefore com-
pelled by the hypnotist to choose a particular sandwich? It is hard to see how there can be if deter-
minism is true. If determinism holds, then it seems that I am not acting freely on either scenario
when I “choose” a particular sandwich for lunch. Compatibilists like Fischer and Ravizza downplay
the full implications of causal determinism. If causal determinism were true, then all current events
would be determined by past events and the laws of nature, including my brain states, reasoning
processes, logical deductions, beliefs, and decisions. Therefore, assuming causal determinism, I would
have no choice about which sandwich to order in either case, the case where I appear to choose it
freely, or the case where I am operating under hypnosis. The only difference between them is the
actual causal process involved, yet crucially there is a complete causal story to be told in each case,
with the same action as the end result. I do not have real freedom of will to make a genuine choice.
Compatibilist theories are interesting, but it is hard to see how they could be true since there is an
apparent contradiction at the heart of them: such theories seem to (subtly) obfuscate the meaning of
causal determinism, yet one must resort to something like these theories in order to deal with this
problem for materialism (Shatz 2016; Vihvelin 2013).
Theists, and supporters of free will, have also argued that there is a logical contradiction at the
heart of all deterministic views about human freedom. We can appeal to the work of one of the
most influential psychologists of the twentieth century, B. F. Skinner, to illustrate this point. In
Beyond Freedom and Dignity, Skinner defended a type of determinism that he thought we could
use to mold individuals, particularly as infants, into better citizens (Skinner 2002). Because human
behavior is determined by a combination of our physical, biological structure, which includes our
genetic makeup, together with the environment in which we were raised, it might be possible to dis-
cover as science advances which “inputs” (e.g., childhood upbringing, educational system, external
rewards and punishments) in a person’s experience would produce the correct “outputs” (that is,
most desirable actions from a moral and social point of view). Human beings would be treated as
if they were machines: if one operates the machine in the appropriate environmental conditions,
and implements the right “inputs” (oil, replacement parts, etc.), one will get a more optimal perfor-
mance. Just as it is with machines, so it is with human beings.
However, there is a fatal flaw in this far-fetched thesis. It assumes that it is possible to choose
the “right” inputs freely! Otherwise, the experiment could not succeed because, since there is no
freedom of will, any “inputs” chosen to contribute to the formation of infants could not be chosen
freely. The “choice” of inputs would, in fact, be causally determined by the physical makeup of the
person responsible for the inputs. So the thesis is based on the contradictory claims that determin-
ism is true and that it is necessary to have free control over the inputs required to lead to a prede-
termined outcome. We can apply this criticism to determinist theories more generally. Proponents
of determinism wish to convince people that determinism is true; they want those who believe in
free will to abandon this belief, and to accept that determinism is the correct thesis about the pro-
cess of human decision making. However, if determinism were true, this would mean that the pro-
ponent of determinism did not freely arrive at this view but was caused to hold it by a scientific pro-
cess of cause and effect. It also means that if someone comes to be convinced by the determinist’s
arguments, she too would have been caused to do so, and not because she concluded freely that
the arguments for determinism were convincing.
Such scenarios are wildly implausible, and are part of the reason why free will remains a central
feature of human beings’ understanding of themselves—and a feature that plays an essential role in
religious views of human life and the universe. For free will, of course, makes the moral order pos-
sible. This moral order is not only an essential feature of human nature, but of the universe. It also
needs an explanation. This is why the moral argument for God’s existence is thought by many thin-
kers to have considerable force, and also why it is an important part of the cumulative case for the
truth of the religious view of reality.
GOD AND FREEDOM
Many theists hold that the universe operates according to a set of scientific laws that do not require
God’s constant working in the world (but perhaps only his occasional working). The one exception
is free will, which, as we have seen, is outside scientific law. Free will can be understood as a limita-
tion that God has placed on his own power, because it is not possible for God to control human
decisions and yet for those decisions to be truly free. Theistic philosophers argue that this is because
God cannot do the logically impossible; he cannot create free beings who at the same time are
guaranteed, in the sense of being causally compelled, to make the right choices. We have other
resources to guide us toward correct morality: reason, revelation, conscience, prayer, and grace.
The existence of free will also gives the theist a way to explain the evil that is done by human beings
(often called moral evil). This evil is possible because human beings have free will, that is, moral evil
is the price of creating human creatures who have been given God’s highest gift, which is freedom.
No matter how bad this world is, as several philosophers have argued, it is better than the alterna-
tives: a world with perfect moral beings who lack free will, or no world at all (Hick 1966). Of these
possibilities, God decided to create our world. This raises the question of whether God is free to
choose among various possible worlds, and if so, is our world the best world that he could have cre-
ated, and if not, why did he not choose the best? Some atheists argue that the combination of God’s
freedom and the evil in our world can generate a problem with regard to God’s nature, perhaps even
a contradiction.
God is usually conceived of in traditional philosophy and theology as a being of infinite power,
wisdom, and goodness, who is morally perfect and also a free being. These perfections are also
thought to belong to God necessarily, and not just contingently, though this conclusion is disputed
by some thinkers at least as regards the attribute of freedom. But some atheistic philosophers, such
as William Rowe, have argued that these perfections can generate a problem, one that was recog-
nized by several philosophers of the past (Rowe 2006). Given God’s perfections, Rowe believes
that God would have no choice but to create the best possible world. If he does not, as Gottfried
Wilhelm Leibniz argued, he would seem to be lacking in either wisdom or goodness. But, according

to Rowe, “if God could not do otherwise than create the best world, he created the world of neces-
sity, not freely” (Rowe 2006, 2).
And there would be another difficulty. We have to believe that our world, with its innumerable
evils, is the best possible world that God could have created. Rowe thinks that this puzzle may force
us to rethink our understanding of God’s freedom, because if it is true that God had to create this
world out of necessity—because he has to create the best possible world—then perhaps we should
ascribe a compatibility conception of freedom to God, which would mean that God acted freely
but could not have done otherwise than he did (in line with the view of Fischer and Ravizza dis-
cussed earlier). This is a solution that will hardly appeal to most theists, who are likely, as we have
seen, to reject compatibility views of freedom as contradictory (Beilby and Eddy 2001).
The puzzle is generated because, as Leibniz argued, God must create the best possible world
because he is necessarily good. If there is a better world than ours that God could create, he would
necessarily have to create it because of his perfect goodness. So Leibniz’s view seems to run into the
problem that God was not free to create this world (freedom understood here as the ability to make
a different choice from the one actually made). One might solve this problem by saying that God’s
goodness is analogous to human goodness in one key respect: that while God could have made a less
perfect world, he would not because of his perfect goodness. Rowe rejects this common answer
because he thinks that God cannot become less than absolutely perfect, because this is what it means
to say that God is necessarily good, rather than (like us) just contingently good. Samuel Clarke,
quoted by Rowe, suggested that one could dissipate the problem by arguing that “God always dis-
cerns and approves what is just and good, necessarily, and cannot do otherwise: But he always acts
or does what is just and good freely; that is, having at the same time a full natural or physical power
of acting differently” (Rowe 2006, 24). This means that God necessarily sees what is right (and
always does it, in part because he has perfect rationality and no undesirable impulses) but is free
to do what is wrong. Although this raises the question of whether the necessity in God is logical
necessity or some kind of metaphysical necessity (which means that although God can do things like
sin, he never does), Rowe also rejects Clarke’s argument because he thinks it is inconsistent with
what logical necessity means: the existence of a (necessarily) perfectly good God is not compatible
with the claim that God can do otherwise.
In replying to a puzzle like Rowe’s, the theist usually insists that God’s moral goodness
requires that he have true freedom; in this sense God’s freedom is similar to human freedom. In
the case of God, an all-good being who cannot help being good is not as perfect as an all-good
being who can help being good (that is, a being who always does what is right because he cannot
do otherwise is not as good as a being who, though he can commit evil, never does!). So God is not
necessarily good in the logical sense because a necessarily good being is not as perfect as a contin-
gently good being because of the nature (and value) of freedom. The success of this line of reason-
ing all depends on whether or not “necessarily good” precludes genuine freedom. Clarke thinks it
does not. The theist will insist that there is a sense in which we can say that God is necessarily
good as long as we mean not logical necessity (God must choose such and such) but absolute or
metaphysical necessity, meaning that God is God—a being in the highest possible, indeed in a per-
fect, moral state—because he always chooses what is right even though he could do otherwise. Many
have thought that this kind of answer will work fine, and indeed it appears to have been Leibniz’s
own answer.
Another important, and to many, convincing line of reply to an argument like Rowe’s is that of
Thomas Aquinas, who argued that there is no best possible world. So, on Aquinas’s view, it would
not be possible to generate Rowe’s problem (because God could have created any number of differ-
ent worlds, and so retains his freedom). Aquinas believed that the type of world God can create is
limited by the parts he chooses to create it. Because there is an apparently infinite choice of parts,
there is no end to the number and type of worlds that God can create. Adams has also suggested
a reason for why God might create an inferior world—because it would give him a greater opportu-
nity to manifest his grace. Rowe replies that this would only show that God is not doing anything
morally wrong in creating such a world, but this is not the same “as showing that God’s perfect
goodness does not render it necessary that he create the best world he can” (Rowe 2006, 82). This
problem does raise interesting questions that force us to think further through our conceptions of
the nature of God; but it is clear, the theist will argue, that it is far from supporting any kind of
compatibility account of free will, or generating a more general problem concerning the notion of
freedom.
GOD OF THE GAPS?
Sometimes materialists invoke a “God of the gaps” accusation as a rejoinder to all nonmaterialist
positions about the status of the mind. A God of the gaps argument is thought to be logically prob-
lematic. The accusation is that the theist points to many features of the mind that cannot be
explained in scientific terms, and then invokes the intelligence and purposeful activity of God to
explain them. But, the materialist rejoinder continues, eventually we will discover a full scientific
explanation for mental phenomena, such as consciousness, intentionality, subjectivity, and free will,
and so there will be no need to appeal to the causal activity of God as an explanation. Just because
we do not know the explanation now, the argument goes, it does not follow that we will not dis-
cover it in the future. The objection is also often accompanied by the suggestion that appeal to
God is really no explanation at all. It is true that it is not a scientific explanation, but it would be
the simplest and the most complete explanation in a general sense. This is why, the theist argues,
we only need to establish its reasonability and plausibility based on a cumulative case approach.
With regard to the problem of consciousness, we must assess the materialist rejoinder by making
a judgment on whether the arguments that claim that mental life cannot be explained in principle
by science are good arguments. We are required to make a decision on this issue because one cannot
suspend judgment indefinitely in the hope that one day we will find a scientific explanation.
Theists often make a similar “in principle” argument with regard to other issues that we cannot
explain through science, such as the origin of life, the origin of the universe, and the laws of physics.
The cosmological argument for God’s existence claims, for instance, that science (in principle) will
never be able to explain the ultimate origin of the universe because this would require showing
how something came out of nothing, and it is irrational to think that one day we will discover
how this is possible. This raises the question, how do we know when we have reached the limits
of scientific explanation? Obviously this question needs to be handled very carefully, and perhaps
on a case-by-case basis, for if we conclude too quickly that there is no possibility of a scientific
explanation for a phenomenon, we run the risk of later being proved wrong. Yet if we refuse to ever
consider the possibility that there might not be a scientific explanation, we run the risk of adopting
a naturalistic approach as a matter of policy, that is, of ruling out explanations in terms of a super-
natural personal agent by definition.
Many thinkers have suggested that the disciplines of science and philosophy today are atheistic
or naturalistic in practice and outlook (Johnson 1993). This naturalistic attitude is particularly prev-
alent in the philosophy of mind where nonphysical explanations appear to be ruled out by defini-
tion and a whole host of increasingly implausible arguments are offered as “thought experiments”
or as theoretical underpinnings to what must in the end be scientific explanations of mental phe-
nomena. It is no wonder that materialists are frustrated, because the best way to defend their view
would be to provide actual scientific explanations, yet so far none are available. Indeed, some thin-
kers such as Colin McGinn are skeptical about ever finding a scientific explanation. Although com-
mitted to materialism, McGinn observes that
we just don’t have the faculties of comprehension that would enable us to remove the sense of
mystery.… People sometimes ask me if … the growth of neuroscience has given me pause;
they fail to grasp the depth of mystery I sense in the problem. The more we know of the brain,
the less it looks like a device for creating consciousness: it’s just a big collection of biological
cells and a blur of electrical activity—all machine and no ghost. (McGinn 2012)
Lynne Rudder Baker has expressed the point very well: “The first-person perspective … can neither
be explained by science nor explained away; it can neither be eliminated nor be reduced to anything
non-perspectival or non-first personal.… Since the scientific picture of the world is of a totally

objective, totally impersonal reality, the scientific picture cannot accommodate the first-person per-
spective … [which] is a serious challenge to naturalism” (Baker 2013, xxi).
Contemporary work in science officially adopts a working assumption toward the physical realm
that has been called methodological naturalism, which means that only physical, testable explanations
will be pursued. But it is a logical mistake to conclude from this that the only possible explanations
for any aspect of reality are physical ones (this would be metaphysical naturalism). When one is think-
ing about whether or not immaterialist accounts of the mind are correct, or perhaps rely too much on
an appeal to a God of the gaps argument, one must be extremely careful not to allow an assumption
of metaphysical naturalism to influence one’s thinking, just as one should not rule out too quickly the
possibility of a scientific explanation. Perhaps the best way to walk this line is to recognize that science
is confined to physical explanations but that this does not exhaust the explanatory possibilities in rela-
tion to the universe and the human person. We need to appreciate that the discipline of philosophy,
or the realm of the rational, is broader than that of science. Philosophers must consider whether some
areas pointed to by reason might be outside of science (as science is normally understood) but never-
theless are important to our understanding of the nature of the human person and the universe as a
whole. These areas include the origin of the universe and of life, questions concerning the laws of
physics, and, crucially for our purposes here, consciousness and its properties and activities, including
subjectivity, the experience of a unified self, in addition to reason, free will, and moral agency.
Materialists often reply more generally to the accusation of metaphysical naturalism by appeal-
ing to what might be called a kind of “scientific faith” argument in the philosophy of mind. This
involves beginning essentially from a premise or from a supposition of naturalism, and then arguing
that the mind must be physical. One then confronts difficult problems, like that of consciousness,
subjectivity, and intentionality, by arguing that, though these challenges are both logically and prac-
tically very difficult, they will be eventually solved when we have better knowledge of how the brain
works. This argument is based on an appeal to inductive evidence (to take an example from U. T.
Place) that, since we have approached many phenomena in the past that we thought we could not
explain, but for which we eventually did find a scientific explanation (for example, lightning), it will
be the same with the recalcitrant features of the mind (Place 2002).
Obviously, this type of scientific faith argument is a good approach in general in science, a
point the immaterialist about the mind does not deny. Those who accept some version of dualism
agree that it is a good argument when applied to a physical phenomenon like lightning but deny
that it is a good argument when applied to complex features of the mental, such as consciousness,
intentionality, qualia, and free will. This is because human consciousness and its properties are quite
obviously not just additional physical objects in need of an explanation; rather, they seem to have
no basis in physical matter, such as atoms, and causal laws. The implausibility of the scientific faith
argument may be likened to a supporter of cryonics who claims that one day in the future we will
be able to bring people back from death and so it is reasonable to preserve a corpse today in liquid
nitrogen at a cryonics facility. Is this a plausible argument? Many believe (including most natural-
ists) that it is very implausible. But, the immaterialist argues, the claim that the mind will one
day be explained by science is even more implausible because we are dealing with differences in kind,
whereas at least in the cryonics case (if we confine ourselves to talking only about the physical body)
it is a difference in degree, and there seems to be no objection in principle (especially on an assump-
tion of naturalism) to the possibility of resuscitating a corpse.
This argument shows why any analogy between the human mind and other biological or phys-
ical phenomena always breaks down. Some materialists believe, for instance, that if we were to fig-
ure out precisely how each part of the brain works, down to basic units of neurons and synapses, if
there are such things, and we were then able to reproduce such processes artificially (say, in a
machine), the machine would likely have some kind of consciousness, or mental activity. This, they
continue, would be similar to constructing an artificial heart that would be capable of performing
the functions of a real heart, such as pumping blood and maintaining blood pressure. However, this
analogy does not work because the heart and the brain are not sufficiently analogous. The immate-
rialist argument is that, in addition to the brain and its structure, we also have consciousness and its
activities, and it does not seem possible in principle that these could be just neurons and synapses.
That is the key point in the dispute that we must never overlook or paper over. Even if it were pos-
sible to reproduce brain structure in an artificially exact way, it is still hard to see how consciousness
and its activities would arise from this arrangement. Therefore, to treat consciousness as if it were
just another physical object is far-fetched. Finally, but crucially, it is also not legitimate logically
to assume a starting point of naturalism in one’s investigation into the nature of reality and what
exists if the question of the status of the human mind is yet to be decided.
Bibliography
Adams, Robert. “Flavors, Colors and God.” In Contemporary Perspectives on McGinn, Colin. “All Machine and No Ghost?” New Statesman, 20 February
Religious Epistemology, edited by R. D. Geivett and Brendan Sweetman, 2012. https://www.newstatesman.com/ideas/2012/02/consciousness-mind
225–240. New York: Oxford University Press, 1992. -brain.
Baker, Lynne Rudder. Naturalism and the First-Person Perspective. Oxford: Millikan, Ruth Garrett. Language, Thought, and Other Biological Objects.
Oxford University Press, 2013. Cambridge, MA: MIT Press, 1984.
Beilby, James K., and Paul R. Eddy, eds. Divine Foreknowledge: Four Views. Millikan, Ruth Garrett. White Queen Psychology and Other Essays for Alice.
Downers Grove, IL: InterVarsity Press, 2001. Cambridge, MA: MIT Press, 1993.
Chalmers, David J. “Facing Up to the Problem of Consciousness.” Journal of Moran, Dermot. Introduction to Phenomenology. London: Routledge,
Consciousness Studies 2, no. 3 (1995): 200–219. 2000.
Churchland, Paul M. Matter and Consciousness. Cambridge, MA: MIT Press, Murphy, Nancey. Bodies and Souls, or Spirited Bodies? Cambridge, UK:
Davies, Brian. Aquinas. New York: Bloomsbury, 2004. Nagel, Thomas. The Last Word. New York: Oxford University Press,
1997.
Descartes, René. Meditations and Other Metaphysical Writings. Translated by
Desmond M. Clarke. London: Penguin, 1999. Nagel, Thomas. “What Is It Like to Be a Bat?” In Philosophy of Mind: A
Guide and Anthology, edited by John Heil, 528–538. Oxford: Oxford
Dretske, Fred. Naturalizing the Mind. Cambridge, MA: MIT Press, 1995. University Press, 2004.
Feser, Edward. Philosophy of Mind: A Short Introduction. Oxford: Oneworld, Nagel, Thomas. Mind and Cosmos: Why the Materialist Neo-Darwinian
2005. Conception of Nature Is Almost Certainly False. New York: Oxford
Fischer, John Martin, and Mark Ravizza. Responsibility and Control: A Theory University Press, 2012.
of Moral Responsibility. Cambridge, UK: Cambridge University Press, O’Connor, Timothy. Persons and Causes: The Metaphysics of Free Will. New
Fodor, Jerry. Psychosemantics: The Problem of Meaning in the Philosophy of Parfit, Derek. Reasons and Persons. Oxford: Clarendon Press, 1984.
Mind. Cambridge, MA: MIT Press, 1987.
Place, U. T. “Is Consciousness a Brain Process?” In Philosophy of Mind:
Green, Joel B., and Stuart L. Palmer, eds. In Search of the Soul: Four Views of Classical and Contemporary Readings, edited by David Chalmers, 55–60.
the Mind-Body Problem. Downers Grove, IL: InterVarsity Press, 2005. New York: Oxford University Press, 2002.
Hick, John. Evil and the God of Love. New York: Harper and Row, Plantinga, Alvin. Where the Conflict Really Lies: Science, Religion, and
1966. Naturalism. New York: Oxford University Press, 2011.
Hume, David. A Treatise of Human Nature: A Critical Edition. (1739.) Putnam, Hilary. Mind, Language and Reality: Philosophical Papers. Vol. 2.
Oxford: Clarendon; New York: Oxford University Press, 2007. Cambridge, UK: Cambridge University Press, 1975.
Johnson, Phillip E. Darwin on Trial. 2nd ed. Downers Grove, IL: Robinson, Daniel N. Consciousness and Mental Life. New York: Columbia
InterVarsity Press, 1993. University Press, 2008.
Kane, Robert. A Contemporary Introduction to Free Will. Oxford: Oxford Rowe, William L. Can God Be Free? Oxford: Clarendon Press, 2006.
Sartre, Jean-Paul. The Transcendence of the Ego: An Existentialist Theory of
Koterski, Joseph. An Introduction to Medieval Philosophy: Basic Concepts. Consciousness. Translated by Forrest Williams and Robert Kirkpatrick.
Oxford: Wiley-Blackwell, 2009. New York: Hill and Wang, 1991.
Locke, John. An Essay concerning Human Understanding. Edited by Peter H. Searle, John R. Intentionality: An Essay in the Philosophy of Mind. New York:
Nidditch. Oxford: Clarendon; New York: Oxford University Press, Cambridge University Press, 1983.
Shatz, David. “Irresistible Goodness and Alternative Possibilities.” In David
Martin, Raymond, and John Barresi, eds. Personal Identity. Oxford: Shatz: Torah, Philosophy, and Culture, edited by Hava Tirosh-Samuelson
Blackwell, 2003. and Aaron W. Hughes, 35–68. Leiden, Netherlands: Brill, 2016.
Matthews, Eric. Mind: Key Concepts in Philosophy. London: Continuum, Siewert, Charles P. The Significance of Consciousness. Princeton, NJ:
2005. Princeton University Press, 1998.

Skinner, B. F. Beyond Freedom and Dignity. Indianapolis, IN: Hackett, 2002. Taliaferro, Charles. “The Give and Take of Biological Naturalism: John Searle
and the Case for Dualism.” Philosophia Christi 7, no. 2 (2005): 447–462.
Sweetman, Brendan. Religion and Science: An Introduction. New York:
Continuum, 2010. Taylor, Richard. Metaphysics. Englewood Cliffs, NJ: Prentice Hall, 1992.
Sweetman, Brendan. Evolution, Chance, and God: Understanding the Thomas Aquinas. Summa theologiae. Translated by Timothy McDermott.
Relationship between Evolution and Religion. New York: Bloomsbury, London: Hafner, 1964.
2015.
Vihvelin, Kadri. Causes, Laws, and Free Will: Why Determinism Doesn’t
Swinburne, Richard. The Existence of God. Rev. ed. Oxford: Clarendon; New Matter. New York: Oxford University Press, 2013.
York: Oxford University Press, 1991.
Wallace, Alfred Russel. “The Limits of Natural Selection as Applied to
Taliaferro, Charles. Consciousness and the Mind of God. Cambridge, UK: Man.” In Contributions to the Theory of Natural Selection, 332–371.
Cambridge University Press, 1994. London: Macmillan, 1870.
TOPIC 13
Human Beings: Atheism

CONSCIOUSNESS AND INTENTIONALITY 429
Kenneth Williford
Associate Professor and Chair, Department of Philosophy & Humanities
University of Texas at Arlington
REASON 434
The Head of the Society and Cognition Unit
PERSONAL IDENTITY AND LIFE AFTER DEATH 436

Diane Proudfoot
University of Canterbury, New Zealand
HUMAN FREEDOM 441

Mariam Thalos
University of Tennessee, Knoxville
This chapter observes the way theists have supposed that metaphysical considerations about human beings have
supported theism over atheism. The chapter further discusses questions about consciousness and intentionality,
reason, personal identity, and freedom. It is argued that the incompleteness of current neuroscientific accounts
of these phenomena does not lend any significant support to a theistic account of them.
CONSCIOUSNESS AND INTENTIONALITY
Kenneth Williford, University of Texas at Arlington
Though one can find precursor arguments in the early modern period (and probably before), since
at least the late 1980s some theistic philosophers have been arguing that the (supposed) naturalistic
or neuroscientific inexplicability of phenomenal consciousness (Adams 1987; Moreland 2009) or of
our rational capacities, which, of course, presuppose representational capacities (Reppert 2009),
strongly suggests a supernatural, theistic explanation. Everyone can grant that at present we do
not have a complete neuroscientific understanding of consciousness, intentionality, and reasoning,
but does theism offer a superior explanatory framework in this connection?
THE HARD PROBLEM OF CONSCIOUSNESS

In 1995 David Chalmers introduced the hard/easy problem distinction into philosophical and scien-
tific discussions of consciousness (Chalmers 1995). Though formulations vary, the basic idea is that
429
the problems of cognition and naturalistically defined represen-

KEY CONCEPTS tation are relatively easy and can be fruitfully approached using
the concepts and tools of the cognitive sciences, while certain
phenomena essential to consciousness are impervious to such
Agent Causation
treatment. For example, neurocomputationally and biofunc-
Compatibilism tionally plausible models of how representational contents can
Consequence Argument become available for the guidance of conscious action, reason-
ing, and report may well tell us much about how the conscious
Duplication Problem
mind interfaces with the brain understood as a physico-
God of the Gaps functional system. However, such models, because they deal
Guidance Control only with structure and dynamics, cannot tell us anything illu-
Iff minating about what it is like (Nagel 1974) to experience those
contents or explain why there should be any such thing as
Life-Form phenomenal experience or manifestation in the first place.
Narrative Freedom In the terms that Chalmers made famous, solutions to the
Near-Death Experiences (NDEs) easy problems would apply to philosophical zombies—beings
that are behaviorally, physically, and functionally indistinguish-
Nonnaturalistic able from us but lack phenomenal consciousness. The hard
Panpsychism problem boils down to this question: In terms of what can we
Patternism explain the existence and range of variability of phenomenal
consciousness? Ideally, we would like it if consciousness could
Philosophe
be explained in terms native to contemporary science (physics,
Principle of Alternative Possibilities (PAP) chemistry, biology, computer science) and if such an explana-
Quantum soul tion could tell us, specifically, why we experience colors, flavors,
Techno-supernaturalists moods, and so on the way we do or tell us what it would be
like to experience the world the way a bat or octopus does.
Famously, Chalmers (1996) argues that consciousness
cannot be explained in terms acceptable to contemporary
science and that we must therefore accept some form of dualism, panpsychism, or pan-proto-
psychism (cf. Shear 1997; Brüntrup and Jaskolla 2016). All of these views entail that consciousness
is, in one way or another, metaphysically irreducible or fundamental (without, of course, entailing
theism).
The hard problem and the closely related notion of the explanatory gap (Levine 1983) are just
new ways of formulating old and seductive intuitions, but we can say that a new level of dialectical
clarity has been achieved since consciousness became a topic of intense philosophical and scientific
concentration in the 1990s. Conceivability (zombie) and related arguments for the claim that con-
sciousness is metaphysically irreducible to physical processes have been explored thoroughly. The
emerging consensus about these arguments seems to be that they rather too naively assume that the
sort of conceivability we are capable of with respect to consciousness entails metaphysical possibility
(Williford 2007; Elpidorou 2016). As Antoine Arnauld pointed out to René Descartes (see Arnauld’s
Fourth Set of Objections to Descartes’s Meditations [Descartes (1641) 1984, 140ff.]) in a similar con-
text, simply being able to conceive of a property as not being essential to some phenomenon should
not in general be taken to entail that it is not essential (its being possibly inessential entailing actual
inessentiality). It seems easy to conceive of consciousness without a brain and brains just like ours
without consciousness. It seems easy to conceive of consciousness in the absence of computational,
neurobiological, and physical substrates, as we now understand them, and vice versa.
It is, however, relatively simple to explain why it should be easy for us to conceive of these
putative possibilities: the metaphysics of consciousness is phenomenologically undecidable. We get
our concept of consciousness from reflection on first-person conscious experience, but at most we
get only rather general properties from this reflection. Consciousness seems to involve, notably,
qualitative character (“qualia” neutrally understood or “what-it-is-like-ness”), temporality, subjective
character (“for-me-ness”), intentionality, synchronic multimodal integration, global availability, and
so on (Williford, Rudrauf, and Landini 2012). One cannot tell just by phenomenological reflection,
however, whether consciousness is a substance, a process, a bundle of tropes, a brain process, a

computational process, a divine spark, and so on. All of these seem to be epistemic possibilities
just because we have essentially no information about them from the first-person point of view.
Moreover, we seem to be prone to “headless woman” (Armstrong 1968) or “what you see is all
there is” (Kahneman 2011) illusions rather generally—we take an absence of information with
respect to P to positively indicate the absence of P; thus, it is natural for us to assume that con-
sciousness has complete introspective access to all its essential properties. If it did, of course, then
conceivability would be a good guide to metaphysical possibility with respect to consciousness.
But there is no reason to think that consciousness has complete access to its essential properties,
and there are many good reasons for thinking that it does not (Williford 2007). Moreover, we
fool ourselves if we think that we can meaningfully imagine brains like ours in their relevant com-
plexity and thus that philosophical zombies are ideally conceivable.
The fact that our phenomenologically derived concept of consciousness contains such a pau-
city of information about the metaphysics of consciousness ought to move us in the direction of
neurophenomenology in terms of methodology (Varela 1996; Rudrauf et al. 2003, Petitot et al.
1999) and a posteriori identity theory in terms of metaphysics (Elpidorou 2016). According to
the neurophenomenological paradigm, progress can be made on the problem of consciousness
by (1) codifying an appropriately abstract phenomenological description of the structural and
dynamic invariants of consciousness; (2) identifying the likely cognitive, affective, and biological
functions of consciousness; (3) developing a neurobiologically realizable, generative mathematical
model of consciousness that captures these structural, dynamical, and functional components; (4)
experimentally identifying the neural correlates of consciousness and demonstrating their relation
to the model; and (5) supporting the model with simulations and applications (in artificial intel-
ligence, etc.).
Evidently this method is in its infancy, but already strides have been made (see, in particular,
Rudrauf et al. 2017; Oizumi, Albantakis, and Tononi 2014). What is particularly appealing about
the method is that it allows us to make an intelligible connection between the phenomenology of
first-person conscious experience and the brain processes that realize consciousness via the interme-
diary of mathematico-computational generative models that preserve structural and dynamical prop-
erties between these apparently disparate domains. If the method were successful, we would be able
to identify what processes in the brain realize consciousness and their general neurocomputational
architecture. However, it is unknown just how much leverage this would give us when it comes
to explaining the precise phenomenal character of various types of experiences. It may give us a gen-
eral, high-level understanding of what quality spaces are, but it may or may not be able to say much
that is illuminating about what it was like to see the sunset on a particular day in a particular mood.
It may or may not tell us in an illuminating way what it is like to be a bat. Still, it has been a fault
of much thinking on this problem to take such a limit in our explanatory frameworks to be indica-
tive of a special problem in relation to consciousness.
It has generally been forgotten in many discussions of this problem that even in areas of phys-
ical science that no one would regard as particularly mysterious, there are serious limits to the
applicability of our mathematical models. The theory of the onset of turbulent fluid flows is a case
in point (see, e.g., Ruelle 1993; Wilson 2008). No one would think that because our mathemati-
cal models of fluid flows do not allow us to deal with the onset of turbulence in precise and ade-
quately predictive ways we must resort to divine intervention or make the onset of turbulence into
some nonphysical property. In the case of consciousness, which is, no doubt, one of the most
complex processes in the physical world, we should accept a certain explanatory modesty. As the
old adage has it, if human brains were so simple that we could understand them, then we would
be so simple that we could not. Given this, if we can capture the general structural, dynamical,
and functional features of consciousness in a phenomenologically motivated generative mathemat-
ical model and intelligibly relate them to specific brain processes, or (neuro)computational pro-
cesses, we will have done a great deal—perhaps all we can reasonably expect to do. If we cannot
predict exactly what pistachio-chipotle ice cream would taste like to an agitated manatee on
LSD, so what?

In the absence of compelling arguments for dualism (or panpsychism, idealism, etc.), the posi-
tion on the metaphysics of consciousness that seems most plausible is some form of a posteriori iden-
tity theory. Given that we cannot decide the metaphysics of consciousness by phenomenological
means or by conceivability arguments or by a priori thought experiments, the most epistemically
responsible thing we can do is simply accept that, most probably, consciousness is some sort of
brain process (or perhaps computational process realizable in brain hardware and perhaps other sub-
strates), even though that is hard for us to understand. The limits of phenomenology and conceiv-
ability in this connection entail that whatever consciousness is, metaphysically speaking, the state-
ment expressing this identity (e.g., to be conscious is to realize computational process C under
suitable temporal and informational conditions S) will, at some level of analysis, look irremediably
contingent and brute to us, though perhaps somewhat less so once the theory is developed and
we get used to it. This appearance of contingency at the heart of the theory of consciousness should
not bother us. It is not, at the end of the day, importantly different from other famous theoretical
identities (e.g., water = H2O, pace Kripke 1980 and others). Moreover, even in pure mathematics
there is the necessary appearance of contingency (see, e.g., Chaitin 2003). Given these limits on
our powers of conceivability and explanatory insight, the lack of any compelling arguments for dual-
ism (or related positions), and what neuroscience seems to tell us is in our skulls, Ockham’s razor
would suggest accepting the claim that consciousness must be a physical process realized in our
brains that we could perhaps identify a posteriori. We must not get confused: in a certain sense,
identities are to be discovered and belief in them epistemically justified, but they are not to be
explained in terms of something more basic (cf. Papineau 2002). If this is right, the hard problem
is, at least in the abstract, solved (or dissolved).
However, even if some form of dualism or panpsychism turns out to be true, nothing would be
gained here by positing the creator God of traditional theism. It may be that certain versions of
panpsychism, for example, entail, in a way reminiscent of Benedict de Spinoza, that the whole uni-
verse is basically a single mind (Goff 2018), but this, again like Spinoza’s Deus sive Natura, would
clearly not be the God of traditional theism. Even if the neo-dualist and neo-panpsychist arguments
were compelling, they are far from lending support to traditional theism specifically—again, a lesson
that should have been learned from Spinoza. Invoking the God of traditional theism to explain how
it is that consciousness gets correlated with human brain function has all the virtues of theft over
honest toil, to steal a phrase from Bertrand Russell. It would be a fairly paradigmatic “God of the
gaps” sort of move.
We cannot explain consciousness to our own satisfaction in terms of science we can easily
understand, the story would go. So we simply postulate that there is a God, himself possessed of
consciousness, who has the power to make human brains conscious (or correlate consciousness with
brain function, etc.). God’s consciousness is taken to be irreducible, fundamental, and not in need
of further explanation, on pain of a vicious explanatory regress reminiscent of the regress that looms
for the design argument. No one can deny the trivial conditional claim that if there were an all-pow-
erful deity who wanted to make human brains conscious—brains, let us assume, otherwise incapa-
ble of hosting consciousness—then he could and would do so. If the antecedent were true, we
would indeed expect to find consciousness associated with human brain activity and not be able
to understand how or why this was so in theoretically satisfying terms. Thinking of it in Bayesian
terms, we could say that, indeed, the fact that human brains are associated with consciousness in
some theoretically impenetrable way (modulo contemporary science) may be evidence that there is
a conscious God with the power to forge this association, because the relevant observations are
indeed very probable on the hypothesis.
However, carrying the Bayesian analysis further, in a way analogous to Bayesian analyses of
the design argument (see, e.g., Sober 2008), we have to ask about the prior probability of there
being such a God. This is, of course, a prior probability we have no clear way of establishing. It
could be zero for all we can tell a priori. Because we cannot do much by way of updating in this
case, the usual Bayesian epistemological point about priors not mattering (as long they are not 1
or 0) does not seem to apply in any straightforward way. If we have some other arguments for
theism that establish a high prior for the existence of such a God, well and good, but then this
argument from consciousness is rendered dependent on those other arguments and thus is argu-
ably obviated.
The competing naturalistic hypothesis is that consciousness in humans and animals is a phys-
ically realized brain process that we simply do not have a good theory of and may never have
because of our inherent conceptual and epistemic limitations (cf. McGinn 1991, 1993). This
hypothesis, too, would predict that consciousness would be found in association with properly
functioning human (and animal) brains and also that we would not now understand exactly
how or why this is so. What is the prior probability of this hypothesis? The prior probability that
consciousness is a physically realized process that we simply do not understand yet is high. We
can get a sense of this by considering the progress of science, which generally shows that it is fre-
quent that we begin with phenomena that we do not understand and, by a combination of empir-
ical work and conceptual and often mathematical developments, can come to a certain (often lim-
ited) theoretical understanding of them. The central question here is this: Is it more probable that
consciousness is a physical phenomenon that we do not (and maybe cannot) fully understand or
that there exists a supernatural almighty consciousness that, for some rather mysterious reasons
and by incomprehensible means, wanted to make little consciousnesses and associate them with
complex, organized biofunctional arrangements of cells? The latter hypothesis would, again, pre-
dict what we observe. But surely the former is a more reasonable hypothesis. It has the same pre-
dictive power and certainly involves fewer otherwise unmotivated independent parameters. In
terms of both hypotheses we must accept epistemic and conceptual limitations on humans, but
with the first hypothesis that is essentially all we must accept; in the latter case we have not only
a version of the same epistemic limitation but also a suddenly bloated ontology with no compen-
sating explanatory gains.
INTENTIONALITY
A similar dialectic unfolds with respect to the problem of intentionality. (Indeed, on some views, the
problem of intentionality is not separable from the problem of consciousness.) There are many com-
peting naturalistic theories of intentionality, mostly worked out in the 1980s (see, e.g., Dretske
1981; Millikan 1984; Fodor 1987; Dennett 1989) and with little significant progress since; none
of them are completely satisfactory (Kriegel 2011). If it should turn out that real or “underived”
or “intrinsic” intentionality (Searle 1983, 1992) is an irreducible feature of human (or animal)
minds, what then? We would have roughly the same set of explanatory options: panpsychism
(etc.), a posteriori identity theory cum phenomenological undecidability, and traditional theism.
Again, two out of three of these options are of no help to the traditional theist. Moreover, the sec-
ond of these options seems to be the most well motivated both epistemologically, because it is famil-
iar that there are natural phenomena we do not understand very well, and metaphysically, because,
absent powerful independent arguments, nothing is gained by positing a supernatural being that has
the target property (intentionality, phenomenal consciousness) and the inexplicable power to make
complicated brains to have the property as well—and for equally mysterious reasons.
To sum up, traditional theists can derive little to nothing of use for their case from considera-
tions about our epistemic and theoretical limitations vis-à-vis phenomenal consciousness and
intentionality. The hard problem of consciousness and the problem of intentionality may have
well-motivated solutions or dissolutions in neurophenomenology and a posteriori identity theory
(modulo general considerations about the limits of our theoretical understanding and the irremedia-
ble appearance of contingency, even in pure mathematics). Even if the conceivability and related
arguments for anti-physicalist views of consciousness (dualism, panpsychism, etc.) work, they do
not generally support traditional theism. We cannot say how much evidence these phenomena of
mind provide the theistic hypothesis unless we have an independent way of determining the likeli-
hood (in the Bayesian technical sense) of the hypothesis and its prior probability—which means, at
best, that the argument is not an independent line of evidence for theism. We can say, however,
that Ockham’s razor suggests that consciousness and intentionality are physical processes or features
realized in human brains in some way that we do not (and may not ever) fully understand. There is
no help for theism here and no specific problem for atheism.

REASON
Konrad Talmont-Kaminski, University of Bialystok
The Enlightenment is known as the age of reason, intoxicated as it was with fresh realization of the
possibilities that the application of reason offered. When the light of reason was applied to look at
our own mental capabilities, however, what it revealed was all too often tawdry, with superstition,
irrationality, and willful ignorance being the norm. The contrast between the ideal and the reality
was striking—reason appearing like something very much out of this world. The philosophes could
not for the most part imagine how reason could have a source in the physical world.
In various forms, much the same argument is employed by today’s theists to claim that reality
cannot be just physical, that the existence of reason shows that the fundamental ontology of the uni-
verse must include the rational in some form. Alvin Plantinga (1993) presents one line of this
thought in his evolutionary argument against naturalism, where he argues that cognitive abilities
produced by unguided evolution would not give rise to reasoning that sought out the truth. (The
argument is considered in detail in the context of evolutionary-debunking arguments in the chapter
on theories of religion.) On the basis of arguments put forward by C. S. Lewis, Victor Reppert has
been developing his version of the argument from reason in a series of publications (Reppert 1999,
2003, 2009).
Reppert claims that reason could not have emerged from purely physical processes. Putting the
argument in a somewhat different form, he also argues that any explanation of reason must involve
nonphysical entities, which entails an ontological commitment to the existence of those entities.
Reppert (2009) sees the argument as ultimately referring to six different aspects of reason: the inten-
tionality of our thoughts, our knowledge of truth, the way the content of our beliefs causes us to
come to have other beliefs, the psychological relevance of logical laws, the unity of our conscious-
ness, and, finally, the reliability of our rational faculties. Breaking up the argument, it is possible
to see how it is closely connected to the arguments considered in other chapters. In particular, it
can be seen that at least in part it runs over much the same ground as Plantinga’s argument—
Reppert, similarly to Plantinga, thinks that evolution provides no basis for a naturalist account of
the development of reason.
More generally, however, Reppert’s argument is, in effect, a version of the argument from
design. The difference is that instead of claiming that the existence of complex life-forms could
not be explained without recall to a designer, Reppert claims that naturalism lacks the means to
explain how it is that some of those life-forms came to be rational. This means that many of the
standard problems with the argument from design will also affect the argument from (the design
of) reason. In Daniel Dennett’s (1995) terminology, by claiming that reason cannot be reduced to
or emerge from physical processes, Reppert is proposing a skyhook. The particular significance of
Reppert’s argument is what he calls its transcendental aspect—if he is right and reason requires a
nonnaturalist explanation, then arguing for naturalism is self-defeating because arguing assumes
the existence of reason.
Explanations that involve agents, including rational agents, come naturally to humans and are
readily understood by us. If we are told that someone stole a loaf of bread because their children
were hungry, we do not feel the need for any further explication but that we know what happened.
This is because we have an excellent intuitive grasp of folk psychology—the main driver of the
development of primate intellect having most probably been social interaction (Dunbar 2004).
While this is a boon in our everyday dealings with other humans, it has held back scientific devel-
opment on many occasions by leading us to accept explanations that are far from adequate.
This is typically the case with arguments from design. We find it easy to accept that complex-
ity, life, or reason must be explained by the actions of a designer. If we do, inquiry stops there. In
the case of reason, what does it mean to say that in some way reason must be a fundamental part
of reality? Normally, we consider reason to be a trait of a person, so we seem to be talking about
some reasoning being. How did that being spread the capacity for reasoning to others, including
us? Reppert does not seem to be particularly interested in providing a detailed explanation, only
in arguing that naturalists cannot. Why would he think that, however? The answer in part would
seem to have much to do with the traditional view of what reason is. The conception of reason as
something perfect and godlike, rather than arising out of the daily struggles of all life-forms, is
one that has been the norm rather than the exception since the times of the Enlightenment, making
the argument from reason intuitively attractive not just for Reppert. However, that is not the pic-
ture of reason that empirical research is revealing. What we are seeing is extensive evidence of bot-
tom-up, cranelike (to use Dennett’s terminology) development of reasoning abilities.
Cognitive science is the multidisciplinary scientific field of research that has been investigating,
among other things, the nature of reasoning. The field has been developing rapidly for more than
fifty years. Explaining how reason can come from purely physical beginnings is precisely what it
does. Yet Reppert essentially ignores it, claiming that what thousands of scientists are doing every
day is impossible, a position that is particularly incomprehensible given that Reppert by title and
at the start of the book C.S. Lewis’s Dangerous Idea (2003) refers to Dennett’s book Darwin’s Dan-
gerous Idea (1995), in which Dennett presents much of that work. Thus, what ultimately shows the
argument from reason to be a nonstarter is not so much a philosophical counterargument as a whole
field of scientific research. What might have looked like a white gap on the map of our knowledge
during Lewis’s times has been well and truly explored by now.
What are the kinds of answers that cognitive science has for Reppert? There are the comparative
studies that show how human reasoning capabilities are a development on the capabilities of other
living organisms (Penn and Povinelli 2007). Indeed, there exists a journal focused upon, as the
name suggests, Animal Cognition. At the same time, work within artificial intelligence has sought
to understand how random variation and selection—the very mechanism of evolution—can lead
to effective problem solving (Sivanandam and Deepa 2008). An altogether different line of research
is looking at the way that making explicit the norms, including the epistemic norms, that function
in a society helps that society. Together, these and thousands of other strands of research are helping
us put together a picture of how it is that reason can come to be in a reasonless universe. This pic-
ture is incomplete, as is always the case with scientific knowledge. However, it is being rapidly filled
in with new results being published daily. Of course, Reppert does not have to agree with the con-
clusions of this research, though such a stance is more than a little quixotic. Neither he nor anyone
else who would like to put forward the argument from reason has any excuse, however, if they sim-
ply choose to ignore a whole field of scientific research.
For theists, the difficulty raised by research in cognitive science goes further than to just under-
mine the main contention of an argument from reason. The problem is that the picture of reason
that cognitive science has been revealing renders implausible key aspects of the traditional view of
God as a perfectly rational being. Empirical research shows that reasons are developed bottom-up,
tied to the function of surviving in one’s immediate environment and based upon a variety of pro-
cesses. A prime example is Herbert Simon’s (1996) conception of bounded rationality, according to
which human reasoning abilities consist of a selection of heuristics, each of which is capable of deal-
ing with a limited range of problems and provides solutions that are merely adequate in the given
circumstances. On such views the traditional idea of a perfect reasoner is at best an idealization
and not even a particularly useful one (Wimsatt 2007; Talmont-Kaminski 2013).
It could be claimed, of course, that the empirical studies are revealing only what human reason
is like, with no implications for what God’s reasons could be like. The problem with that claim,
however, is that the cognitive abilities of humans and other evolved life-forms are the only ones that
we know. When no reference to a perfect rationality is necessary to understand them, the shortcom-
ings of that concept of rationality come into sharp relief. Something of the problems can be seen if
we simply consider that God is deemed to be all-knowing. This creates fundamental difficulties for
considering God to be capable of reasoning. After all, reasoning is a process by which we change our
beliefs upon the basis of other beliefs or other evidence. However, a change in God’s beliefs would
entail that before or after the change he either had some beliefs that were false or failed to have some
that were true—neither of which claims appear to be consistent with the claim that God is all-
knowing. An all-knowing God would be no more capable of reasoning than an omnipresent God
could move from place to place. How much of a difficulty is that for theists? Could they not allow

that God cannot reason? The problem is that our understanding of what it means to be a person is
profoundly tied up with being able to reason. Without the capacity to reason it is impossible to
make decisions, and without the capacity to make decisions the capacity for intentional action
breaks down. If God is not capable of reasoning, it becomes hard to see in what sense he is a person.
Of course, this conceptual difficulty has not meant that theists who claim that God is all-
knowing thereby have ceased to think of him as a person. The reason for this, however, is much
more psychological than philosophical. Justin Barrett and Frank Keil (1996) have shown that when
they are remembering stories that involve God, people will tend automatically to think of him as
having many of the traits of normal humans, such as being able to act on things only in one loca-
tion, even if they are theists who have just a few minutes earlier stated that they believe God to
be omnipresent. The theologically correct (Slone 2004) picture of what God is like may be possible
to learn but it is very hard to keep in mind when thinking about him. The question has to be to
what degree this difficulty has affected theological positions regarding God—in particular, to what
degree theist thinkers have felt able to claim that God has personhood, is all-knowing, and is
perfectly rational because their thinking about him is drawn back to the human model.
PERSONAL IDENTITY AND LIFE AFTER DEATH
Diane Proudfoot, University of Canterbury
“I used to like traveling, but now I am a homebody.” “Have you seen her recently? She has totally
changed.” These mundane statements point to a deep question: How could I, or any human being,
be both the same and different? This question becomes pressing in the context of the promise by
many religions of a life after death. The Hebrew Bible says, “Many of those that sleep in the
dust of the earth will awake, some to eternal life, others to reproaches, to everlasting abhorrence”
(Daniel 12:2). What makes the premortem human being identical to the reanimated corpse?
Philosophers distinguish between numerical and qualitative identity: a human being A at time1
(an earlier time) and a human being B at time2 (a later time) may be numerically identical—that
is, one and the very same human being—even though qualitatively very different. This allows iden-
tity despite change. Introducing the numerical/qualitative distinction, however, merely gives labels
to our question. For under what conditions, if any, are A (at time1) and B (at time2) numerically
identical—one human being?
THE DUPLICATION PROBLEM

This question is harder to answer than one might at first think. Here is a sample answer: because a
human being is a biological organism (i.e., a member of the species Homo sapiens), A and B are one
human being if and only if (iff) they are “biologically” identical. We might speculate further that A
and B are biologically identical iff they have exactly similar DNA.
This answer leads to a puzzle. Let us assume that in the actual world no two humans (even
monozygotic twins) have exactly similar DNA. It is still possible that two humans have exactly sim-
ilar DNA. In that case A (at time1) could be identical with two human beings (at time2), B and C,
both of whom have exactly similar DNA to A’s (and, therefore, to each other’s). If A is identical to
B and A is also identical to C, then (by the transitivity rule) B is identical to C. B and C surely must
be different human beings, however, because they are in different places, having different experi-
ences. Philosophers call this puzzle the duplication problem. A satisfactory account of personal iden-
tity must avoid it.
PSYCHOLOGICAL CONTINUITY
If we want to leave open the logical (or metaphysical) possibility of an afterlife, perhaps our question
about identity despite change should not be put in terms of human beings. We can imagine that a
person might survive even if the biological organism that (in “this” life) the person was did not
survive. Let us then reframe our question as this: Under what conditions are A (at time1) and B (at
time2) one person?
Once our question concerns persons rather than human beings, it may seem obvious that the
answer concerns psychology. “Person” and “personality” seem to fit together. Perhaps what makes
A and B one person is that their core personality traits, psychological characteristics, and perspective
on the world are exactly similar. This theory succumbs immediately to the duplication problem,
however—and has the counterintuitive result that anyone who altered his or her core traits would
cease to exist.
Someone very like me in personality, though, would not have my history. Perhaps, then, what
matters is that B remembers (“from the inside”) doing what A did or, more plausibly, remembers
doing what B-5-minutes-earlier did and B-5-minutes-earlier remembers doing what B-10-minutes-
earlier did … and so on, all the way back to A. Still, the duplication problem remains: both B
and C could sincerely report memories exactly similar to A’s and also have the corresponding
“movie” playing in their heads. (We can use suggestion to create false memories in other people.)
So which one is A?
To avoid this difficulty, we might say that B rather than C is identical to A iff the person B (at
time2) remembers doing what A did (at time1) and the human being B—that is, the physical
entity—actually did do this. Yet keeping the requirement that people remember their earlier lives
implies that, if B has total retrograde amnesia, A and B are not one person. In turn, this means that
B has not forgotten doing what A did, because B is a different person. Moreover, in this theory, A,
B, and C all cease to exist every time they are asleep (on the assumption that unconscious “remem-
bering” is not really remembering). These are counterintuitive results.
SPATIOTEMPORAL CONTINUITY
Perhaps looking for a psychological criterion of persistence of persons is looking in the wrong place.
It may be, instead, that a person A (at time1) and a person B (at time2) are one person iff there is
spatiotemporal continuity between A’s body and B’s body. Alternatively, science—and the claims
of cryonics firms—might suggest that only the brain is necessary: in this view, A and B are one per-
son iff there is spatiotemporal continuity between A’s brain and B’s brain. Assuming a tight connec-
tion between physical and psychological states, both approaches have the advantage of incorporating
the psychological (and so might even enable us to salvage the requirement that B remembers doing
what A did).
Philosophers, however, have invented hypothetical counterexamples to defeat this spatiotempo-
ral continuity approach. What if A split into two fully formed human beings (in the way that tape-
worms and jellyfish reproduce)? Each of B and C (at time2) would be spatiotemporally continuous
with A (at time1), so which is A? Taking an example slightly closer to the actual world, what if the
connection between the two hemispheres of A’s brain is severed and one hemisphere is placed in a
donor body? Which of two human beings (at time1) is A? Is it B, who has A’s original body and left
hemisphere, or C, who has A’s right hemisphere? (B and C might even have the same personalities
and memories.) They cannot both be A, and if neither is A, A does not survive the transplant. Per-
haps, though, we should reject imaginary counterexamples such as these. Human beings do not
reproduce by binary fission or undergo brain transplants, and we do not have reliable intuitions
about what we would say if they did.
There are other problems with this approach. First, it leads to another tough question: What is
the relation between the human being and the person? Second, it rules out persistence after death in
the many cases where bodies are destroyed (e.g., by cremation).
Perhaps in the afterlife case B is a re-creation of A. In the Qur ʾan, unbelievers say, “When we
are turned to bones and dust, shall we really be raised up in a new act of creation?” (17:49). Yet if A
is to be re-created, it seems that A (at time1) and B (at time2) will be one person iff they are exactly
similar human beings (or exactly similar persons). What if, then, two humans (or persons) exactly
similar to A are created after A’s death? The re-creation hypothesis leads again to the duplication
problem.

THEISM AND SOULS

The theist might reply that God would not create duplicates. However, can we know God’s mind
(without mistakenly assuming that God is just like us)? Even if the theist could be confident, this
reply is a merely practical solution to the duplication problem, which is a logical problem—it is like
saying that the solution to a mathematical problem is to give up math.
Nevertheless, it might seem that theists—and proponents of impersonal religions such as non-
theistic Buddhism or Hinduism—have an advantage with respect to the duplication problem: they
believe that human beings have souls. According to the Christian Bible, a human being “is sown a
physical body [and] is raised a spiritual body” (1 Corinthians 15:44). The Bhagavad Gita says, “As a
man leaves an old garment and puts on one that is new, the spirit leaves his mortal body and then
puts on one that is new” (2:22). Using the soul (or spirit) criterion of persistence, A (at time1) and
B (at time2) are one person iff they possess—or simply are—one soul. The person persists iff the
soul persists, even if the physical entity that is the human being is destroyed. Here there is no scope
for the duplication problem, unless the soul can reproduce by binary fission.
The concept of the soul has a complex history, but let us consider two views: the substance
theory, in which the soul is a concrete immaterial entity or substance, and the form theory, in
which the soul is an abstract principle or form organizing matter. Both views generate problems.
The substance theory leads to such questions as these: What differentiates one soul from another?
How does an immaterial soul interact causally with a human body? If the soul is the location of
consciousness, can this be reconciled with scientific evidence of the link between consciousness
and the brain?
Theists have various responses to these questions. First, they may, of course, simply reject sci-
entific discoveries that appear inconsistent with the substance theory; so much the worse for modern
science, they may say. Alternatively, theists may say that the soul is a mystery; for example, the Cat-
echism of the Catholic Church says that the doctrine of the resurrection of the dead “exceeds our
imagination and understanding: it is accessible only to faith” (1994, 1000). This sort of theist is like
the philosophical “mysterian” who claims that, although the mind emerges from the brain, humans
(with biologically limited brains) are unable to “resolve the mystery” of how or why this happens
(McGinn 1989, 349).
Yet again, some nonrealist theists accept that modern science and philosophy demonstrate that
we must “give up the belief in substances, including the belief that we are or have ‘immortal souls’”
(Cupitt 2000, 64). Afterlife doctrines and ideas of the supernatural are not literally true, nonrealists
say; instead, they are “poetry to live by,” guiding the religious person to a fulfilling life (Cupitt
2002, 48).
There may be other answers to these questions, but these replies, at least, seem unsatisfactory.
Rejecting science is a major loss; embracing mystery is too similar to adopting a merely practical
solution—and is there any real difference between the nonrealist theist and the atheist?
THE SOUL AND MODERN SCIENCE

Several theists (and members of nontheistic religions) aim, by contrast, to use science to answer
questions about the soul. Theology is a branch of physics, they say. For example, some proponents
of the hypothesis that the brain is a quantum system talk of a quantum soul and claim that such an
entity accounts for near-death experiences. This hypothesis has the advantage that, because physi-
cists must in any case explain interaction between quantum and classical systems, the question of
how the soul can interact with the body is not a special problem.
Turning now to the form theory of the soul, some thinkers use the software/hardware distinc-
tion to explain the relation between soul and body; from this perspective, the soul is a program run-
ning on the brain. According to these techno-supernaturalists (as I call them), the soul could run
just as well on a mainframe computer and in this way enable a person to survive death (see Proud-
foot 2013). Persons are re-created, techno-supernaturalists claim, insofar as they are simulated in
some virtual world, which may be heavenly or hellish. A simulated person is akin to a program-
controlled avatar in a video game, but this simulation is sufficiently rich, techno-supernaturalists
say, for a virtual person to have subjective experience. This approach can be developed into a full-
scale “digital theology” (Steinhart 2014, 107).
Underlying this approach is the following theory of personal identity: where, for example, A is
an organic human being and B a simulation, A (at time1) and B (at time2) are one person iff A and
B instantiate the same pattern. This is patternism. However, patternism leads to puzzles (Proudfoot
2013, 2016). For one, would A not want backups—simulations C, D, E, and so on—in case of
computer failure? Even the mere possibility of backups runs straightway into the duplication prob-
lem, because, in principle, we can simulate “one” person in many different virtual worlds.
One last question can be posed about souls: If human beings persist iff their souls persist, what
about God? According to the Christian Bible, “God is spirit” (John 4:24), so we might theorize that
God, too, persists iff God’s soul persists. Difficulties for the soul criterion of persistence in the case
of humans, however, would then also apply to this theory of the identity of God over time.
SOLVING THE DUPLICATION PROBLEM

Can the theist or atheist find a solution to the duplication problem that is less problematic than pos-
iting souls? Potential fixes all have heavy costs.
First, we might say that A (at time1) and B (at time2) are identical only if there is no duplicate
that also satisfies our criterion of identity (whatever it is). This seems to imply, however, that
whether or not A is B is merely a matter of convention—like my ability to exchange a $5 bill for
a cappuccino—rather than a fact about the universe. Moreover, using this fix, affirmative judgments
of persistence are always tentative; my claim that A is B will be defeated if a duplicate appears. Both
results are counterintuitive.
Second, we might argue that in the case of human brains duplication is physically impossible,
because brains are quantum systems, which cannot be (perfectly) duplicated (Aaronson 2013). Still,
whether the brain is a quantum system is a controversial question. This argument, too, is another
practical solution to a conceptual problem.
Third, we might say that a person is a type; just as this book is one copy (or instance) of Theism
and Atheism, so the person writing these paragraphs is merely one instance of the Diane Proudfoot
type—and more instances of that type could in principle be created, just as there are many copies of
the book. In this approach, B and C are merely two instances of the A type, and two instances can
be in different places at once without generating a puzzle. This solution, too, seems counterintui-
tive. If a person were to lose this book, that person can replace it with another, equally good copy;
one cannot, it seems, simply replace me with a copy. A person is a particular, not a type.
A fourth option is to use fuzzy logic in order to say that, while it is not completely true that A
(at time1) is identical to either B or C (at time2), it is true enough. At the same time, it is false that
B is identical to C (just because, as before, these are distinct human beings in different places)
(Proudfoot 2013). Fuzzy identity is not transitive, and so no puzzle arises. However, even if we
accept fuzzy identities (this position faces several objections), is its being “true enough” that I am
one person over time really what we want from a theory of identity—or from an afterlife?
NEAR-DEATH EXPERIENCES
Should we require that a theory of personal identity leave room for life after death? Is there any rea-
son, after all, to think that there is an afterlife?
In reply, several theorists point to reports of “near-death experiences” (NDEs). Those who
report an NDE (the NDEr) typically say that while they were in cardiac arrest and their brains were
deprived of oxygen, they had an “out-of-body” experience; they saw a beckoning white light but
reluctantly “returned” to their bodies. NDEs, theorists claim, demonstrate that the soul (or mind)
can survive death or, strictly speaking, absence of brain activity.
These theorists offer three arguments. First, NDEs have positive consequences, and this implies
a distinctive cause. Surviving cardiac arrest also has positive consequences, however. Second, pro-
spective studies (where data are collected on subjects before the NDE) find no physiological or psy-
chological difference between NDErs and non-NDErs; thus, according to theorists, NDEs cannot

be explained in natural terms (van Lommel et al. 2001). Perhaps the experimenters simply did not
investigate all relevant factors. Third, NDEs are accurate descriptions of events that occurred during
cardiac arrest, and the only way that the NDEr could learn of these events, theorists argue, is by
supernatural means. Studies designed precisely to test the latter claim (e.g., Parnia et al. 2014) have
not, however, produced any support for it. NDErs may later learn of events that occurred while
they were unconscious and unwittingly construct false memories.
NATURAL VERSUS SUPERNATURAL EXPLANATION

Some theorists who claim that naturalistic explanations explain away NDE reports also go beyond
the evidence. Suggested biological causes of NDEs include low levels of oxygen in the brain, release
of neurotransmitters, medication, and seizures. Psychological causes proposed for NDEs include the
hypothesis that NDErs tend more to dissociation (as a way of managing fear of death) than non-
NDErs. Theorists also offer naturalistic explanations for other empirical phenomena that might be
considered proof of an afterlife. For example, numerous people report, apparently without prior sug-
gestion and in the same locations, unusual experiences that they attribute to the presence of
ghosts—that is, dead people—but scientists have attributed these experiences to environmental fac-
tors and also to suggestibility (Wiseman, Greening, and Smith 2003). Similarly, some people take
accurate readings by psychics as evidence of spirits, but studies that control for various mundane
ways in which a medium can acquire information about a sitter have failed to find any evidence
of genuine psychic ability (O’Keeffe and Wiseman 2005).
Nevertheless, finding a correlation between NDEs and abnormal brain chemistry—or between
ghost sightings and light levels—does not exclude a supernatural cause of these experiences (assum-
ing the latter is possible). Similarly, if (on testing) a psychic is unconvincing, it may be that the psy-
chic is not genuine or simply that spirits refuse to participate in experiments.
NO SELF OR NO CRITERIA
Even if the empirical evidence were stronger, we would still have to construct a theory of personal
identity on which an afterlife is possible. Two further approaches are available to both theist and
atheist, but each has a high price. What if a person is not one person over time? What if it is simply
false—for a human being, person, soul, or any other entity—that A (at time1) and B (at time2) are
identical? This approach is usually associated with impersonal religions. In Buddhism, for example,
it is said that “in the absolute sense there is no ego”; the word I is a “figment” (Radhakrishnan and
Moore 1957, 284, 285). This view has the advantage that, because identity is a fiction, the duplica-
tion problem does not arise.
Philosophers taking this approach may say that A survives as B (and C and so on), but this is
not survival in the strict sense, because A is not identical to any later entity. We could not look,
then, to an afterlife for enlightenment or redemption as, for example, Job does. In the Hebrew
Bible, Job stresses that he, Job, will receive postmortem compensation for his suffering; after his
death, he says, “I would behold … I myself, not another, would behold [God] … with my own
eyes” (Job 19:26–27). On the “survival as” account, poor Job is entirely mistaken.
The second approach has similarities with mysterianism. According to the anti-criterialist, there
is nothing that is necessary and sufficient for A (at time1) to be identical to B (at time2)—nothing,
that is, over and above A’s identity with B (Merricks 1998). That is to say, there is no nontrivial or
informative answer to our original question. This looks like good news, for both theist and atheist.
In this approach, destruction of the body is no barrier to life after death, because neither spatiotem-
poral continuity nor anything else is a criterion of personal identity over time (Merricks 2001). Sim-
ilarly, the duplication problem cannot arise, because there are no criteria for a duplicate to satisfy. If
the anti-criterialist is correct, however, how do we know or even reasonably believe that A persists,
in this world or the next? There seems no answer, which is pretty bad news.
In sum, the theist and the atheist have an equally difficult task in formulating a satisfactory the-
ory of personal identity over time. Neither has a promising account of postmortem persistence
(which in any case is not supported by the extant empirical evidence).
HUMAN FREEDOM
Mariam Thalos, University of Tennessee
When it comes to freedom, the philosophical problems confronting its attribution to humans are
substantially different from those confronting its attribution to God. With regard to God, the core
questions circle around the topic of whether attributing freedom to God is compatible with also
attributing the qualities of omnipotence, omniscience, eternality, and perfect goodness; a God
who has all these qualities could not also be free to accept limitations upon them—to take any steps
that contribute to becoming ignorant or weak, for instance. Furthermore, a God who is not per-
fectly good might be barred from creating just any possible world; only the best possible, it has been
argued, is compatible with perfect goodness. (The debate has been at the center of celebrated con-
troversies for centuries [cf. Rowe 1993, 2004]). Thus the philosophical questions around God’s
freedom are to do with how to square its possession with such features as certain theological tradi-
tions are bound to attribute to a sovereign creator of the universe.
By contrast, the traditional philosophical questions around human freedom are to do with how
to square freedom for human organisms with increasingly scientific understandings of the universe
itself. At the beginning of Western philosophical consciousness, Plato, unlike later philosophers eli-
gible of the label rationalist, maintained that there are obstacles to free and rational agency, owing in
no small measure to pressures exerted by the human psyche from what later were referred to as bio-
logical drives and drives for social status. In subsequent eras, these obstacles came under the heading
of “weakness of will”—a conception shared widely in popular and traditional culture. Today the
realities of addiction, social stresses, forces of habit, culture, social position, and group identity,
together with psychological findings about cognitive biases, have come to dominate the landscape
of empirically-based obstacles to agency. Can the image of the human being under the microscope
of contemporary science, including cognitive and social science, be reconciled with the “manifest
image” of human beings as sources of their individual behaviors (and held responsible accordingly),
particularly if the scientific image is best construed in terms of governance by laws of nature? (The
phrases manifest image and scientific image were coined by the twentieth-century American philoso-
pher Wilfrid Sellars, see Sellars 1962.) How is it possible to square the idea of laws of nature with
the manifest image in which humans exercise free agency as they play their various parts in the
drama that is life? Does the manifest image thus square better with a theistic image of the world,
as both seventeenth-century French philosopher René Descartes and eighteenth-century German
philosopher Immanuel Kant seem to have thought, and by contrast the scientific image square bet-
ter with atheism?
A deterministic universe, such as was proposed during the Enlightenment, was early on thought
to pose a threat to the idea of human freedom. (It does not pose the same threat to the idea of a
sovereign creator, however, because the notion of a creator of the universe already puts the entity
who qualifies for the office outside the regime of those laws, ahead of them both in time and in
ontological status.) One traditional answer to the threat has been the notion of compatibilism—
which simply asserts that freedom and determinism are not incompatible. The devil, of course,
has been in the details of explaining how this can be so, and many schools of thought on the matter
exist.
It is worthwhile noting that there are enormously compelling reasons, from the quantum
regime so well described by quantum theory, to reject determinism: the laws governing the universe
are simply not fully deterministic. This, of course, vitiates the need to confront the threat as origi-
nally posed. However, the trouble, if one follows along this line of thought, is to make sense of
the nature of freedom in a world governed partly by deterministic laws and partly by nondetermin-
istic ones that admit of an element of chance. Is such a world itself any more compatible with the
idea of human freedom?
In this section we grapple with the question of human freedom, what it could mean, and
whether it is best accepted in the context of a form of theism. We find that theism has no more
resources for addressing these difficulties than atheism does.

LAWS OF NATURE
If laws of nature exist, they pose a certain challenge to human freedom, because human freedom
seems to be contrary to the ineluctable connections spoken of in laws of nature. The most defensi-
ble response here is to maintain that laws, understood in causal terms, refer to networks of events,
not individual ones. In such structures, there is great variety, as attested to by the various sciences,
but the only true contrast to a pattern of causation in such a network is a pattern of complete
and unmitigated randomness. Randomness, as such, is not incompatible with causal order; it can
be overlaid, more or less stringently, by some of the multitude of patterns of causation that the
sciences speak of (lessons that flow from Pearl 2009 [2000]; Spirtes Glymour, and Scheines 2001;
Pearl Glymour, and Jewell 2016; cf. Thalos 2013). Thus the contrary of causal law is not freedom
but chance. And if the scientists are right, they can and do coexist: randomness is not ruled out by
causal laws; randomness is their foil, there to be overlaid by and merged with any number of causal
patterns. An important question remains: What could freedom amount to in this picture of causal
patterns merging with randomness, this new modern scientific picture? If “free” is not to be con-
trasted with “caused,” what is it to be contrasted with?
THE CONSEQUENCE ARGUMENT

The true source of unfreedom—if unfreedom is the true condition of human life—must be that,
when the specific event or reality X appears in my life, I have no options but to respond to it with
Y—if, in other words, there is a freedom-subverting connection between the X and the Y in my sit-
uation. Whatever might be true in other contexts where X occurs, I am unfree when I cannot do
otherwise than Y in my context, when I simply have no option but the one I execute. When this
is so, it can legitimately be said that I have no power over the sequence of events in the relevant
context. This is one way of understanding the conclusion of the so-called consequence argument
(called so by Carl Ginet [1966]).
The consequence argument draws inspiration also from the vision of that network of all events
in the universe connected by consequence relations owing to laws of nature, the vision spelled out
so succinctly by French natural scientist Pierre-Simon Laplace ([1902] 1951) early in the nineteenth
century. The consequence argument advances another principal idea: if A has B as a consequence,
then if I have no power over A, because nothing I can do can alter it, then I also have no power over
B. On the basis of this principle, with roots in Laplace’s conception of the universe as a series of
necessitated events, with later events consequent upon earlier ones, we infer that the events that
occur today and into the future are the natural-law consequences of events that occurred long before
humanity even existed. It is patently true that human beings have no power over events occurring
long before they even existed, in exactly the same way that they have no power over laws of nature
themselves. It thus follows that human beings have no power over events occurring today and into
the future. So human beings cannot be free in a straightforward way. Human freedom is thus
incompatible with this conception of the connections between natural events—a conception pur-
portedly deriving from a scientific view of the universe. The result is a picture of the world without
free agents in an ordinary sense that contrasts directly with the determinism of the Laplacean vision.
If human agency, whatever else it might be, is not free in that it is fully caused by preexisting
condition, then we are obliged to produce an account of freedom—if we still hold to it—not as a
contrast to event determination but as a complement to it. We are obliged to produce an account
of freedom as compatible and in lockstep with the determinism manifest in the network of conse-
quence relations envisioned by Laplace. This is a familiar project—known as compatibilism—in
our philosophical era.
FORMS OF COMPATIBILISM
Compatibilisms are perhaps the most subscribed views on the topic of freedom in philosophy today.
These are perhaps best characterized as employing the strategy of identifying patterns of events
within the very broad class that can be described as causal or natural-law-abiding but involving
agents in such a manner that we would accept them also as expressing or manifesting freedom.
Examples might be cases in which we would describe the relevant agents as acting according to laws
of nature (even human nature), but also where we are comfortable speaking of those agents as
exercising control over the options or prospects open to them or responding to reasons and concerns
that they themselves have or care about.
For what it is worth, I think that both the following are true: (1) Science is committed neither
to the network of events being explainable entirely causally, nor to determinism in the Laplacean
sense; the laws of science can be stated fully without any reference to causes and furthermore with-
out a commitment to a unique and ineluctable sequence of events (Thalos 2013); and furthermore
(2) Agency is compatible with both causality and determinism anyway. It also seems obvious that
(2)—which is another way of saying “compatibilism”—does not settle the question of freedom.
For the most part, compatibilists have assumed the following (false) proposition: if only we could
show that there is genuine agency, then we could show that freedom exists. (Ismael [2016] falls prey
to this assumption.) This is simply to accept a certain form of compatibilism without argument and
against an illustrious philosophical tradition resisting it. Agency, as such, is not freedom (but merely
what Fischer [1994] refers to as guidance control).
To make the point clearer, let’s say that an agent is an entity that can bend the world to its will.
Let’s assume that agents exist—hence there are entities that can bend the world to their will. Now,
not to put too fine a point on it, if those wills to which agents bend the world are not themselves
free, then we might have agency but still no freedom. Like the nineteenth-century German philos-
opher Arthur Schopenhauer, I think this argument is decisive and compatible with the existence of
agency. Freedom, in other words, is more about the origins of one’s will than about how one exer-
cises that will. Nineteenth-century British philosopher John Stuart Mill, author of the much-
beloved On Liberty (1869), penned words evocative of this sentiment:
Among the works of man, which human life is rightly employed in perfecting and beautifying,
the first in importance surely is man himself.… Human nature is not a machine to be built
after a model, and set to do exactly the work prescribed for it, but a tree, which requires to
grow and develop itself on all sides, according to the tendency of the inward forces which
make it a living thing. (4)
My point is this: supposing that freedom depends on the workability of some account of agency in
the context of science is the wrong tactic, because being free is not a matter of being able to bend
the world as agents. Dogs are agents when they bend the world to their wills at the behest of a
human master. Human slaves are even better at it, but they are not free for all that. (What I say
about the will here is in no way uncontroversial: cf., for instance, Baggini [2015].)
What is wanted is an account of how the will itself—whatsoever this shall be deemed to be—
itself is free. Might it help at this point to appeal to the theistic image of the world? Of course, nei-
ther Schopenhauer nor Mill could have made such an appeal. But one could appeal to the idea that
human beings are simply created to will freely, and in that way to resemble their creator—an idea
that free will is simply a brute fact of creation. The theist would then incur the problem of having
to explain how some humans end up being unfree—end up enslaved in one way or another—
because an account of free will that renders all human agents free, always and everywhere, is no bet-
ter a theory of freedom than one that renders no one free anywhere or at any time. Freedom is
alienable, and obviously so. Why else would it be a rallying cry for people everywhere whose lives
are crushed under the wheels of tyranny?
REINVENTING COMPATIBILISM
Rather than offer a positive account of freedom, compatibilists have traditionally proceeded by tak-
ing it as a contrary to a notion of consequence such as that of Laplace, but the strategy is flawed. If
freedom and natural consequence are even prima facie contraries, this renders permanently vulnera-
ble the project of identifying a subclass of law-governed events that are also manifestations of free-
dom.
What is required, before the notion of compatibilism can even make sense, is a philosophical
dissociation of freedom and determinism as concepts. This can happen straightforwardly if one gives
a positive characterization of freedom that renders the two at least conceptually separate. In other

words, there is no better way to defend compatibilism than by offering an account of freedom, as
such, that does not use the concept of natural consequence by way of explicating it. For our pur-
poses here, we can consider a conception of freedom that a theist might wish to defend. What, in
the economy of theistic thought and faith, is freedom good for? One list is supplied by Meghan
Griffith:
It is plausible to assume that theism requires that many, if not all, sins involve full-blooded
agency.… Some are instances of weakness of will, in which the agent succumbs to temptation
against her best judgment of what she ought to do. Some sins [involve] unprecedented and
pivotal immoral decision[s]. Other sins may be instances of perversity, in which the agent does
something he believes or knows not to be good. Some may even be instances of defiance or
rebellion … simply because he believes it is sinful. (2016, 181)
Weakness of will, pivotal decisions, perversity, and rebellion enjoy a commonality: they fit neatly
into storytelling, into the narration of a life; in addition to describing, they are also ways of explain-
ing action. I will refer to the kind of freedom that the theist requires as narrative freedom. (It is an
interesting and separate question whether it is also the sort of freedom the theist requires in connec-
tion with the deity; that topic is for another time.)
Is narrative freedom compatible with a world governed by laws of nature, as revealed by sci-
ence? Assuming also that these laws must be causal in nature, operating within a network of events
under the sovereignty of causal laws, events involving the forming of intentions will presumably be
correlated with events of bodily movements in the patterns we would normally recognize as actions.
Such patterns may be consistent with freedom, but they also may fail to be. What could be the dif-
ference between being free and being unfree?
SUGAR AND PAP

A straight route to freedom that steers clear of drawing attention to the idea of determinism is one
that invokes the idea of an open future: it is sometimes referred to as the garden of forking paths.
This is the idea that free agents are ones who could have done otherwise than they did and in their
doings realize one of multiple alternatives—formulated in the so-called Principle of Alternative Possi-
bilities (PAP). Unfortunately, this simple idea has been proven flawed via a strategic coordination of
counterexamples, so much so that the counterexample type now has a name: Frankfurt-style coun-
terexample (Frankfurt 1969). The consensus seems to be that PAP cannot be a true measure of
(narrative) freedom, for the reasons that it runs up against the realities of addiction, social stresses,
forces of habit, and acculturation—basically, it runs up against everything that contemporary sci-
ence says keeps us from being able to do otherwise than we do. In spite of these we persist in repuls-
ing the label of unfreedom for the actions in question.
Thus, less-direct strategies of articulating the notion of freedom are favored at present, via
identifying concepts that are subsequently argued to track or be tracked by freedom (cf. Thalos
[2016] for a direct strategy that maintains that freedom rests in resisting social forces of a partic-
ular kind). The most frequently associated feature of freedom is the quality of being subject to
evaluation, especially moral evaluation. One cannot be held to answer for some action if one can-
not be reckoned free in its execution. Indeed, the performance could not even be called an action
otherwise. What, then, are the defining features of actions, such that we can be held to account
for them?
John Perry sought to unsettle a certain dogma—the iconic desire-belief model of action, which
views action simply as an event consequent upon a pair of other events, understood as the occasion-
ing of two mental attitudes: one, a desire, and the other a companion belief about how that desire
can be satisfied in the circumstances. Perry wrote:
I once followed a trail of sugar on a supermarket floor, pushing my cart down the aisle on one
side of a tall counter and back the aisle on the other, seeking the shopper with the torn sack to
tell him he was making a mess. With each trip around the counter, the trail became thicker.
But I seemed unable to catch up. Finally it dawned on me. I was the shopper I was trying to
catch.
I believed at the outset that the shopper with a torn sack was making a mess. And I was right.
But I didn’t believe that I was making a mess. That seems to be something I came to believe.
And when I came to believe that, I stopped following the trail around the counter, and
rearranged the torn sack in my cart. (1979, 3)
Perry writes that before he believed it was he himself who was making the mess, the only appropri-
ate response open to him was to reach out to the mess maker and intercede on behalf of the uni-
verse. When he learned that he himself was the offender, the options suddenly changed. The moral,
to my mind (though this is not the moral Perry himself drew), is that for something to qualify as
action, the subject in question at the very least must enjoy a specific command over the details of
a personal point of view in relation to the target environment of action. Action is thus demanding;
it is something that only persons described in a sufficiently rich way can undertake. It demands an
appreciation of one’s point of view in relation to a scene or environment in view. The conclusion
this line of argument seems to point to is that actions can be ascribed only to persons, not to events
as happening in a network and certainly not to mental states divorced from other features of the
person (cf. Frankfurt 1971).
There is an idea in the philosophical literature called “agent causation” (with recent defenses of
it to be found in Franklin 2016; cf. O’Connor 2000, Pereboom 2004) that is taken up in the con-
text of a theistic view of freedom (Griffith 2016). This is essentially the view I just described and
pressed as an alternative to the belief-desire model of action. Critics of it have tended to claim that
this notion of agent causation is both obscure and mystical (Bratman 2000; Enç 2003).
Even if the obscurity criticism of the doctrine of agent causation can be overcome, is the
account proffered really enough for freedom—could not a slave, too, experience a point a view while
at the same time proceeding only at the bidding of a master? The response here must be simply to
reiterate what Mill said—that being a free agent also must involve the making of oneself as a free
person—becoming in a sense self-made (an idea fleshed out at some length in Thalos 2016).
If this is the right answer, then the question for us becomes whether an account of being self-
made is better supported within a theistic worldview or an atheistic one. This question is wide open.
It is in no way obvious that theists are better placed to offer an account of self-made persons than
atheists, and perhaps they are disadvantaged. After all, theists routinely urge upon mortals the sub-
ordination of one’s will to that of the deity; they urge voluntary denial of the self, a rejection of the
ability to guide oneself in preference to taking guidance in life from the deity construed as a figure
of authority in matters of moral and personal significance. As a consequence, when theists stand up
for something, this is not easily read as self-chosen or reflecting directly on them as people; if theists
come to possess certain qualities, this is not easily read as something the rest of us are entitled to put
down to them. So how is it possible for a theist to affirm themselves as self-made. On its face, there-
fore, theism does not easily support the notion of freedom as a matter of being self-made.
CHAPTER REVIEW
In summary, then, certain features of human beings have posed beguiling philosophical challenges
from very antiquity: consciousness and intentionality (first-personal experience as well as its
“aboutness”), the fact that human beings can reason, that they think of themselves as the same
people over a lifetime of undeniable change and development, and, finally, that they experience
freedom or suppose that they do. Contentions that a theistic philosophical framework can explain
these features of human beings, at least better than atheistic ones, are wildly exaggerated. Even
more exaggerated is the contention that the reality of these features can itself provide support
for the doctrine of theism.
With regard to formulating an account of the sameness of a person through a lifetime of
change, the theist and the atheist have an equally difficult task, and neither has a promising account
of postmortem existence that is supported by evidence of a scientific caliber. With respect to con-
sciousness, reason, and freedom, certain considerations urge that the theist is at an argumentative
disadvantage: naturalistic forms of inquiry are increasingly better able to make sense of all these mat-
ters, and naturalistic accounts put human features on a continuum with those of other animals in
the natural world gives a certain explanatory advantage to the atheist—namely, that the naturalist

approach is favored by the axiom that the simplest explanation is the best. Finally, as we explore in
greater and more scientific detail the human versions of reason and freedom, we come to appreciate
that the capacities exhibited by humans are distinct from those going under the same names as per-
fections traditionally attributed to the deity, so they cannot really benefit from association within a
theistic framework. Consequently a focus on the most convincing accounts of human reason and
freedom, born of social and behavioral science, is just as liable, if not more so, to work against tra-
ditional theistic doctrines as to support them.
Bibliography
Aaronson, Scott. “The Ghost in the Quantum Turing Machine.” 7 June Elpidorou, Andreas. “A Posteriori Physicalism and Introspection.” Pacific
2013. https://arxiv.org/pdf/1306.0159v2. Philosophical Quarterly 97, no. 4 (2016): 474–500.
Adams, Robert Merrihew. “Flavors, Colors, and God.” In The Virtue of Faith Enç, Berent. How We Act: Causes, Reasons, and Intentions. New York: Oxford
and Other Essays in Philosophical Theology, 243–262. Oxford: Oxford University Press, 2003.
Fischer, John Martin. The Metaphysics of Free Will. Oxford: Blackwell,
Armstrong, David M. “The Headless Woman Illusion and the Defense of 1994.
Materialism.” Analysis 29, no. 2 (1968): 48–49.
Fodor, Jerry A. Psychosemantics: The Problem of Meaning in the Philosophy of
Baggini, Julian. Freedom Regained: The possibility of Free Will. University of Mind. Cambridge, MA: MIT Press, 1987.
Chicago, 2015.
Frankfurt, Harry G. “Alternate Possibilities and Moral Responsibility.”
Barrett, Justin L., and Frank C. Keil. “Conceptualizing a Nonnatural Entity: Journal of Philosophy 66, no. 23 (1969): 829–839.
Anthropomorphism in God Concepts.” Cognitive Psychology 31, no. 3
(1996): 219–247. https://doi.org/10.1006/cogp.1996.0017. Frankfurt, Harry G. “Freedom of the Will and the Concept of a Person.”
Journal of Philosophy 68, no. 1 (1971): 5–20.
The Bhagavad Gita. Translated by Juan Mascaro. London: Penguin, 2003.
Franklin, Christopher Evan. “If Anyone Should Be an Agent-Causalist, Then
Bratman, Michael. “Reflection, Planning, and Temporally Extended Everyone Should Be an Agent-Causalist.” Mind 125, no. 500 (2016):
Agency.” Philosophical Review 109, no. 1 (2000): 35–61. 1101–1131.
Brüntrup, Godehard, and Ludwig Jaskolla, eds. Panpsychism: Contemporary
Ginet, Carl. “Might We Have No Choice?” In Freedom and Determinism,
Perspectives. Oxford: Oxford University Press, 2016.
edited by Keith Lehrer, 87–104. New York: Random House, 1966.
Catechism of the Catholic Church, with Modifications from the Editio Typica.
Goff, Philip. “Is the Universe a Conscious Mind?” Aeon (February 2018).
New York: Doubleday, 1994.
https://aeon.co/essays/cosmopsychism-explains-why-the-universe-is-fine
Chaitin, Gregory J. The Limits of Mathematics: A Course on Information -tuned-for-life.
Theory and the Limits of Formal Reasoning. Berlin: Springer, 2003.
Griffith, Meghan. “Agent Causation and Theism.” In Free Will and Theism:
Chalmers, David J. “Facing Up to the Problem of Consciousness.” Journal of Connections, Contingencies, and Concerns, edited by Kevin Timpe and
Consciousness Studies 2, no. 3 (1995): 200–219. Daniel Speak, 172–194. Oxford: Oxford University Press, 2016.
Chalmers, David J. The Conscious Mind: In Search of a Fundamental Theory. The Holy Bible, New Revised Standard Version with Apocrypha. New York:
Oxford: Oxford University Press, 1996. Oxford University Press, 1992.
Cupitt, Don. “The Radical Christian Worldview.” CrossCurrents 50, nos. 1/2 Ismael, Jenann. How Physics Makes Us Free. Oxford: Oxford University
(2000): 56–67. https://www.thefreelibrary.com/The+Radical+Christian Press, 2016.
+Worldview.-a063300893.
Kahneman, Daniel. Thinking, Fast and Slow. New York: Macmillan, 2011.
Cupitt, Don. “Free Christianity.” In Is Nothing Sacred? The Non-realist
Philosophy of Religion: Selected Essays, 46–60. New York: Fordham Kriegel, Uriah. The Sources of Intentionality. Oxford: Oxford University
University Press, 2002. Press, 2011.
Dennett, Daniel C. The Intentional Stance. Cambridge, MA: MIT Press, 1989. Kripke, Saul. Naming and Necessity. Cambridge, MA: Harvard University
Press, 1980.
Dennett, Daniel C. Darwin’s Dangerous Idea: Evolution and the Meanings of
Life. London: Allen Lane/Penguin Press, 1995. Laplace, Pierre-Simon. A Philosophical Essay on Probabilities. Translated by
Frederick Wilson Truscott and Frederick Lincoln Emory. New York:
Descartes, René. The Philosophical Writings of Descartes. Vol. 2. Edited by
Dover, 1951. First published 1902.
John Cottingham, Robert Stoothoff, and Dugald Murdoch. Cambridge,
UK: Cambridge University Press, 1984. Levine, Joseph. “Materialism and Qualia: The Explanatory Gap.” Pacific
Philosophical Quarterly 64, no. 4 (1983): 354–361.
Dretske, Fred I. Knowledge and the Flow of Information. Cambridge, MA:
MIT Press, 1981. McGinn, Colin. “Can We Solve the Mind-Body Problem?” Mind 98, no.
391 (1989): 349–366.
Dunbar, R. I. M. “The Social Brain: Mind, Language, and Society in
Evolutionary Perspective.” Annual Review of Anthropology 32, no. 1 (2004): McGinn, Colin. The Problem of Consciousness: Essays towards a Resolution.
163–181. https://doi.org/10.1146/annurev.anthro.32.061002.093158. Oxford: Wiley-Blackwell, 1991.
McGinn, Colin. Problems in Philosophy: The Limits of Inquiry. Oxford: The Qur ʾan. Translated by M. A. S. Haleem. Oxford: Oxford University
Wiley-Blackwell, 1993. Press, 2004.
Merricks, Trenton. “There Are No Criteria of Identity over Time.” Noûs 32, Radhakrishnan, Sarvepalli, and Charles A. Moore, eds. A Source Book in
no. 1 (1998): 106–124. Indian Philosophy. Princeton, NJ: Princeton University Press, 1957.
Merricks, Trenton. “How to Live Forever without Saving Your Soul: Reppert, Victor. “The Argument from Reason.” Philo 2, no. 1 (1999):
Physicalism and Immortality.” In Soul, Body, and Survival, edited by Kevin 33–45. https://doi.org/10.5840/philo1999214.
J. Corcoran, 183–201. Ithaca, NY: Cornell University Press, 2001.
Reppert, Victor. C. S. Lewis’s Dangerous Idea: A Philosophical Defense of Lewis’s
Mill, John Stuart. On Liberty. 4th ed. London: Longman, Roberts & Green, Argument from Reason. Downers Grove, IL: InterVarsity Press, 2003.
1869.
Reppert, Victor. “The Argument from Reason.” In The Blackwell Companion
Millikan, Ruth Garrett. Language, Thought, and Other Biological Categories:
to Natural Theology, edited by William Lane Craig and J. P. Moreland,
New Foundations for Realism. Cambridge, MA: MIT Press, 1984.
344–390. Oxford: Wiley-Blackwell, 2009.
Moreland, J. P. “The Argument from Consciousness.” In The Blackwell
Rowe, W. L. “The Problem of Divine Perfection and Freedom.” In Reasoned
Companion to Natural Theology, edited by William Lane Craig and J. P.
Faith: Essays in Philosophical Theology in Honor of Norman Kretzmann, edited
Moreland, 282–343. Oxford: Wiley-Blackwell, 2009.
by Eleonore Stump, 223–233. Ithaca, NY: Cornell University Press, 1993.
Nagel, Thomas. “What Is It Like to Be a Bat?” Philosophical Review 83, no. 4
Rowe, W. L. Can God Be Free? Oxford: Clarendon Press, 2004.
(1974): 435–450.
O’Connor, Timothy. Persons and Causes: The Metaphysics of Free Will. New Rudrauf, David, Daniel Bennequin, Isabela Granic, et al. “A Mathematical
York: Oxford University Press, 2000. Model of Embodied Consciousness.” Journal of Theoretical Biology 428
(2017): 106–131.
Oizumi, Masafumi, Larissa Albantakis, and Giulio Tononi. “From the
Phenomenology to the Mechanisms of Consciousness: Integrated Informa- Rudrauf, David, Antoine Lutz, Diego Cosmelli, et al. “From Autopoiesis to
tion Theory 3.0.” PLoS Computational Biology 10, no. 5 (2014): e1003588. Neurophenomenology: Francisco Varela’s Exploration of the Biophysics
of Being.” Biological Research 36, no. 1 (2003): 27–65.
O’Keeffe, Ciarán, and Richard Wiseman. “Testing Alleged Mediumship:
Methods and Results.” British Journal of Psychology 96, no. 2 (2005): Ruelle, David. Chance and Chaos. Princeton, NJ: Princeton University Press,
165–179. 1993.
Papineau, David. Thinking about Consciousness. Oxford: Oxford University Searle, John R. Intentionality: An Essay in the Philosophy of Mind. Cambridge,
Press, 2002. UK: Cambridge University Press, 1983.
Parfit, Derek. Reasons and Persons. Oxford: Oxford University Press, 1987. Searle, John R. The Rediscovery of the Mind. Cambridge, MA: MIT Press, 1992.
Parnia, Sam, Ken Spearpoint, Gabriele de Vos, et al. “AWARE— Sellars, Wilfrid. “Philosophy and the Scientific Image of Man.” In Frontiers
AWAreness during REsuscitation—A Prospective Study.” Resuscitation of Science and Philosophy. Edited by Robert Colodny, 35–78. Pittsburgh,
85, no. 12 (2014): 1799–1805. PA: University of Pittsburgh Press, 1962. Reprinted in Science,
Pearl, Judea. Causality: Models, Reasoning, and Inference. 2nd ed. Cambridge, Perception and Reality, 1–40. Routledge & Kegan Paul Ltd; London,
UK: Cambridge University Press, 2009. First published 2000. and The Humanities Press: New York, 1963. Reissued by Ridgeview
Publishing Co., Atascadero, CA, 1991.
Pearl, Judea, Madelyn Glymour, and Nicholas P. Jewell. Causal Inference in
Statistics: A Primer. Chichester, UK: Wiley, 2016. Shear, Jonathan, ed. Explaining Consciousness: The Hard Problem. Cam-
bridge, MA: MIT Press, 1997.
Penn, Derek C., and Daniel J. Povinelli. “Causal Cognition in Human and
Nonhuman Animals: A Comparative, Critical Review.” Annual Review of Simon, Herbert A. The Sciences of the Artificial. 3rd ed. Cambridge, MA:
Psychology 58, no. 1 (2007): 97–118. https://doi.org/10.1146/annurev MIT Press, 1996.
.psych.58.110405.085555.
Sivanandam, S. N., and S. N. Deepa. Introduction to Genetic Algorithms.
Pereboom, Derk. “Is Our Conception of Agent-Causation Coherent?” Berlin: Springer, 2008.
Philosophical Topics 32 nos. 1–2 (2004): 275–286.
Slone, D. Jason. Theological Incorrectness: Why Religious People Believe What
Perry, John. “The Problem of the Essential Indexical.” Noûs 13, no. 1 They Shouldn’t. New York: Oxford University Press, 2004. https://doi
(1979): 3−21. .org/10.1093/0195169263.001.0001.
Petitot, Jean, Francisco J. Varela, Bernard Pachoud, and Jean-Michel Roy, Sober, Elliott. Evidence and Evolution: The Logic behind the Science.
eds. Naturalizing Phenomenology: Issues in Contemporary Phenomenology Cambridge, UK: Cambridge University Press, 2008.
and Cognitive Science. Stanford, CA: Stanford University Press, 1999.
Spirtes, Peter, Clark Glymour, and Richard Scheines. Causation, Prediction,
Plantinga, Alvin. Warrant and Proper Function. Oxford: Oxford University and Search, 2nd ed. Cambridge, MA: MIT Press, 2001.
Press, 1993.
Steinhart, Eric Charles. Your Digital Afterlives: Computational Theories of Life
Proudfoot, Diane. “Software Immortals—Science or Faith?” In The after Death. London: Palgrave Macmillan, 2014.
Singularity Hypothesis—A Scientific and Philosophical Assessment, edited
by Amnon H. Eden, Johnny Hartz Søraker, James Moore, and Eric Talmont-Kaminski, Konrad. Religion as Magical Ideology: How the Supernatu-
Steinhart, 367–389. Berlin: Springer, 2013. ral Reflects Rationality. New York: Routledge, 2013.
Proudfoot, Diane. “Heavenly Computation: Digital Metaphysics and the Tanakh: A New Translation of the Holy Scriptures according to the Traditional
New Theology.” Metaphilosophy 47, no. 1 (2016): 147–153. Hebrew Text. Philadelphia: Jewish Publication Society, 1985.

Thalos, Mariam. Without Hierarchy: The Scale Freedom of the Universe. Williford, Kenneth, David Rudrauf, and Gregory Landini. “The Paradoxes
Oxford: Oxford University Press, 2013. of Subjectivity and the Projective Structure of Consciousness.” In
Thalos, Mariam. A Social Theory of Freedom. New York: Routledge, 2016. Consciousness and Subjectivity, edited by Sofia Miguens and Gerhard
Preyer, 3217–353. Frankfurt, Germany: Ontos Verlag, 2012.
Van Lommel, Pirn, Rudd van Wees, Vincent Meyers, and Ingrid
Elfferich. “Near-Death Experience in Survivors of Cardiac Arrest: A Wilson, Mark. Wandering Significance: An Essay on Conceptual Behaviour.
Prospective Study in the Netherlands.” Lancet 358, no. 9298 (2001): Oxford: Oxford University Press, 2008.
2039–2045. Wimsatt, William C. Re-engineering Philosophy for Limited Beings: Piecewise
Varela, Francisco J. “Neurophenomenology: A Methodological Remedy for the Approximations to Reality. Cambridge, MA: Harvard University Press, 2007.
Hard Problem.” Journal of Consciousness Studies 3, no. 4 (1996): 330–349. Wiseman, Richard, Emma Greening, and Matthew Smith. “Belief in the
Williford, Kenneth. “The Logic of Phenomenal Transparency.” Soochow Paranormal and Suggestion in the Séance Room.” British Journal of
Journal of Philosophical Studies 16 (2007): 181–195. Psychology 94, no. 3 (2003): 285–297.
TOPIC 14
Ethics: Theism
Michael J. Harris
Faculty of Divinity
This chapter first discusses whether the existence of moral norms constitutes proof of the existence of
God. It then evaluates various forms of the divine command theory of ethics. Finally, the chapter
considers certain concepts such as human rights, conscience, and virtue as supporting theism, atheism, or
neither.
MORALITY AND THE EXISTENCE OF GOD
Morality has important connections with theism and atheism. An influential argument in favor of
atheism, for example, is that there is no morally sufficient reason for an omnipotent, omniscient,
and omnibenevolent God to allow evil or suffering in the world. An influential argument in favor
of theism, to take another example, is that the very existence of moral norms is dependent on
God in some significant way (see the discussion of divine command theory later in this chapter).
In this section, the focus is on the idea that the existence of moral norms is itself proof of theism,
that morality itself demonstrates the existence of God.
Three of the most important arguments from morality to the existence of God are Immanuel
Kant’s argument that belief in God is a postulate of moral reason, John Henry Newman’s argument
from conscience, and the argument from the objectivity of value.
KANT’S ARGUMENT
While Kant critiques the traditional ontological or cosmological arguments for the existence of
God, he argues that morality provides rational grounds for belief in God—grounds that are prag-
matic. Belief in God, according to Kant in his Critique of Practical Reason, is a “postulate of prac-
tical reason.” Kant basically argues that practical reason contradicts itself unless it holds that the
world is under the ultimate moral direction of God, who in the long run dispenses happiness in
proportion to virtue and desert. As David Baggett and Jerry Walls put it, “Morality, for Kant…
makes full rational sense only if acting morally is ultimately in our self-interest” (2011, 13; see
also Green 1978, chap. 3).
There are different interpretations of Kant’s argument, but one useful way of presenting it in
outline (adapted from Wainwright 2016, 18) is as follows:
1. Any rational and moral being must necessarily will the “highest good,” a state of affairs in which
every person is happy proportionately to that person’s moral worthiness.
2. Happiness must be brought about by virtue or moral action.
3. It is not rational to pursue ends the realization of which one believes to be impossible. Therefore
agents must believe that the causal structure of nature is such that happiness can be attained by
practicing morality.
449
Ethics: Theism
4. There is no evidence in the phenomenal world of such a

KEY CONCEPTS connection. The distribution of happiness and unhappiness
among people seems to have no systematic connection to
their moral worthiness or unworthiness.
Constructivism
5. Moral reason must therefore necessarily postulate some-
Divine Command Morality thing beyond nature and the phenomenal world that does
Divine Command Theory systematically link moral worth and happiness. The
Ethical Nonnaturalism existence of God, a moral being who determines the
ultimate character of the natural world and its causal
Ethical Nonobjectivism
structure, is therefore a necessary postulate of practical
Ethical Objectivism reason.
Euthyphro Dilemma
An obvious question about Kant’s argument is why the guar-
Moral Realism antor of the link between virtue and happiness has to be
Retributivist Punishment God. Kant’s answer is that a being able to provide this con-
Rightness (in deontology) nection must have God’s attributes: for example, omni-
science, in order to know people’s true moral worthiness;
Theological Voluntarism omnipotence, in order to be able to provide the happiness
Utilitarianism proportionate to people’s worthiness; and moral perfection,
since it is the “highest good” that we are looking to this
being to provide.
One obvious line of objection to Kant’s argument is that at
least some of the moral claims on which it rests are controversial. That one ought to aim at a state of
affairs in which every person is happy in proportion to his or her moral worthiness, for example,
would be rejected by utilitarianism, which considers the degree of happiness of the morally unwor-
thy as well as that of the worthy. Just as utilitarian views of punishment are not retributivist and are
not primarily focused on the desert of the offender, so happiness need not be indexed to moral wor-
thiness. Overall utility might well be increased by allowing happiness for the morally unworthy in
addition to the worthy, not only in terms of more people enjoying happiness but perhaps also in
overall benefits for society as a whole, such as increased economic productivity.
In fact, it seems that Kant’s argument works only for those who accept his moral claims and
that he himself is quite aware of this. He is concerned about proving God’s existence effectively
for all, because we would then obey the moral law simply in order to receive divine reward and
avoid punishment, and our actions would thus lack true moral worth. The proof from morality,
however, protects the purity of the intentions from which moral actions are performed precisely
because it is persuasive only for those who are already committed to pursuing what Kant believes
is the “highest good,” the situation in which every person is happy in proportion to his or her moral
worthiness.
A further objection to Kant’s moral argument for the existence of God is that the fact (if it is a
fact) that practical reason needs to postulate the existence of God does not prove that God, the
object of this need, is indeed a reality. Kant’s response is that the need of practical moral reason
for the existence of God is not a need based on inclination but a need of reason itself. His argument
is that one cannot rationally participate in the moral enterprise without it. However, this response
fails to establish that postulates of reason are any more than postulates, and so the objection that
Kant’s argument from morality fails to establish the existence of God seems valid.
Of course, that Kant’s particular argument from morality to theism seems open to strong objec-
tions does not entail that no such argument can be successfully mounted. Two further important
arguments are considered next.
NEWMAN’S ARGUMENT
In An Essay in Aid of a Grammar of Assent (1979 [1870]), Newman argues that “the most authori-
tative” means of acquiring knowledge of God is “our own minds,” that is, conscience. In “Proof of
Theism” (1961 [1859]) Newman claims that the argument from conscience is available to all from
Ethics: Theism
earliest childhood. His argument from conscience runs essentially as follows (adapted from Wain-
wright 2016, 28, 31). Conscience, like reason, is a natural faculty. It is reasonable for us to trust
our natural faculties in the absence of compelling reasons not to do so. Conscience indicates the
existence of a just, powerful, and omniscient Supreme Governor and Judge. Therefore it is reason-
able to believe that there is a Judge.
Let us expand a little on how, for Newman, conscience indicates the existence of God.
Conscience, according to Newman, makes one vaguely aware of a higher sanction than oneself for
one’s decisions. Conscience is uniquely forceful: its “voice” is “imperative and constraining, like
no other dictate in the whole of our experience” (Newman [1870] 1979, 99). The stirrings of con-
science are accompanied by feelings that intimate “a tribunal in the future” (Newman [1859] 1961,
118–119). Conscience also has what he terms a “magisterial” or “judicial” dimension that affects
our feelings and in particular stimulates fear in us. That we feel responsibility, guilt, shame, and fright
upon transgressing the dictates of conscience suggests, for Newman, the existence of an intelligent
being before whom we have these feelings. Since the cause of these emotions cannot be found in this
world, conscience must be caused by a supernatural Supreme Governor and Judge, namely, God.
Several lines of objection to Newman’s argument are possible. First, John Mackie suggests that
conscience does not necessarily point to anything beyond itself: “what conscience, taken at its face
value, tells us is that this is how one should feel about a wrong action simply in itself” (1982,
104–105). Although, for example, feelings such as guilt, shame, and regret usually assume the exis-
tence of another person, conscience may be a special case in which such emotions are appropriate
even without any other person being involved. Conscience does not make one feel guilty or
ashamed because one is in the presence of another person but simply guilty or ashamed about the
wrong action per se.
The theist might respond to Mackie by returning to the notion of trust that constitutes a key
premise of Newman’s argument. We generally trust our moral beliefs and judgments. While it is
not difficult to provide naturalistic explanations of such beliefs and judgments, how are these to be
justified ? Thomas Nagel, himself an atheist, argues (1997, chap. 7) that an evolutionary explanation
of human reason ought to provide a possible naturalistic account of the existence of reason but by
itself does not give us grounds for confidence in our reason. If our rational capacity is the product
of natural selection, there would be no reason to trust its results in, for example, mathematics and
science (Nagel 1997, 135). Moreover, we cannot justify our trust in our moral beliefs (see also Plan-
tinga 1993). It is therefore plausible—and this is one way of casting the argument from con-
science—that we need belief in a good being such as God in order to have trust in and justify,
rather than merely explain, our moral beliefs and judgments.
A second line of objection to Newman’s argument is suggested by S. A. Grave, who raises the
possibility (1989, 184–185) that conscience points to God only if theistic belief is presupposed. Per-
haps one would not sense through conscience any special constraint or fear of future judgment
unless one already believed in God’s existence. It might be countered that many people who do
not believe in God experience these kinds of feelings as part of conscience and that such emotions
do not necessarily therefore already assume theistic commitment. The inference from these emo-
tions to God’s existence as the cause of conscience, whether or not a given individual in fact makes
this inference, is therefore justified.
Third, and perhaps most significantly, one might object that the phenomena described in New-
man’s account of conscience can be explained in a perfectly adequate way without reference to God.
William Wainwright suggests, for example, that our sense of conscience indicating something
beyond itself may amount to no more than the fact that the moral demands of conscience are
imposed by our reason independently of our desires and are thus experienced by us as coming from
“outside” (2016, 35).
A related, fourth objection to Newman’s argument is based on the plausibility of entirely natu-
ralistic explanations of conscience and the phenomena such as guilt and fear of punishment identi-
fied by him. Mackie argues that, if we investigate the origins and operations of conscience, “we do
indeed find persons in the background, but human persons, not a divine one” (1982, 105). Mackie
suggests that conscience is most plausibly understood as the “introjection into each individual of

Ethics: Theism
demands that come from other people,” most directly from “parents and immediate associates,” but
ultimately from the society in which the individual was raised (1982, 105). One is strongly
reminded here of Friedrich Nietzsche’s dictum that “the belief in authorities is the source of the
conscience: it is therefore not the voice of God in the heart of man but the voice of some men in
man” ([1879] 1986, “The Wanderer and His Shadow,” sec. 52). Sigmund Freud famously identifies
as one major region of the mind the superego, which is formed by the child’s internalization of
parental demands to restrain its instinctive drives, arguing that “the sense of guilt, the severity of
the super-ego, is … the same thing as the rigor of conscience” (1958, 82). Nietzsche himself, in
the second essay of On the Genealogy of Morality ([1887] 1998), offers an account of conscience
the precise contours of which are debated in the secondary literature but that is certainly naturalis-
tic. Or one might plausibly offer an evolutionary account of the development of conscience.
The theist might respond, however, that the kinds of naturalistic accounts proposed by thinkers
such as Freud, Nietzsche, and Mackie, or evolutionary accounts, even if correct, do not succeed in
demonstrating the incompatibility of theism with the natural processes that the accounts highlight.
Perhaps it was God who set the natural processes in motion.
THE ARGUMENT FROM THE OBJECTIVITY OF MORAL VALUE

Many people believe in the objectivity of moral value or the existence of “moral facts”—moral pro-
positions, for example, that are not just matters of taste, opinion, or perspective but that reflect the
facts of the matter in the way that (on a yet more widespread view) the proposition “The earth
orbits the sun rather than the sun orbiting the earth” faithfully reflects a fact, a way in which things
actually are in the world. An example of such a proposition would be “It is morally wrong to kill a
person for no reason.” That it is morally wrong to kill a person for no reason is, in the view of many
people, a fact.
Some theists, as will be illustrated shortly, may wish to argue from the existence of such moral
facts (if indeed we agree that such facts exist) to the existence of God. Some might go a step further
and argue that what makes it a fact that, for example, it is morally wrong to commit murder is that
this moral prohibition has been ordained by God. This is a version of the divine command theory
of ethics discussed in detail below. However, although many theists find some version of divine
command ethics attractive, theists are not committed to advocating such a theory and may assert
that there are moral facts on other grounds—or indeed they may, consistent with their theism, deny
that there are moral facts altogether. Atheists, of course, by definition reject divine command theo-
ries of ethics. But they are not committed to denying the existence of moral facts. Like theists who
reject divine command theories, they may ground the existence of moral facts in some other way—
or again, like some theists, they may deny the existence of moral facts altogether. It is clear that mat-
ters are much more complex than a straightforward opposition between theists who accept the exis-
tence of moral facts and ground these in the existence and command of God, and atheists who deny
the existence of moral facts since to concede their existence would supposedly entail theism.
Theists and atheists alike have several options from among major objectivist theories of morality
that do not entail theism, including ethical naturalism (a naturalistic form of moral realism
that holds that there are objective and natural moral facts and properties) and ethical nonnaturalism
(a form of moral realism most famously exemplified by G. E. Moore’s account of goodness in
Principia Ethica ([1903] 1962). A further objectivist option is constructivism, which maintains that
ethics can be constructed by practical reason. Moral facts, according to the constructivist, are
produced by rational deliberation. Constructivism argues for an objective morality without the
metaphysical commitments involved in moral realism, but precisely where constructivism should
be situated on the moral realism/antirealism spectrum is a matter of controversy.
The difficulty of arguing from moral facts to the existence of God (given the numerous alterna-
tives of sound philosophical pedigree) is compounded by the fact that whether moral values are
indeed objective is, of course, a fundamental controversy in metaethics. Robert Merrihew Adams
provides a salutary corrective to the instinct of many contemporary philosophers to dismiss the idea
of objective moral values when he points out that “we believe quite firmly that certain things are
morally right and others are morally wrong (for example, that it is wrong to torture another person
Ethics: Theism
to death just for fun)” (R. Adams 1987, 144). Nevertheless, it cannot be denied that arguments for
theism from the objectivity of moral value do face the disadvantage of beginning from a highly con-
tentious premise even before they reach the task of showing that God’s existence is the most plausi-
ble explanation of this objectivity, if it obtains.
Nevertheless, some have attempted to meet this dual challenge, or at least to assume the exis-
tence of moral facts as a premise. W. R. Sorley argues that, since moral values are objective, they
must exist somewhere (1918, 351–356). Because they clearly do not exist in material objects or,
in their entirety, in the mind of any individual person, they must exist in a “Supreme Mind,” the
Mind of God. A similar argument is presented in some detail by A. C. Ewing, who argues that,
in order to be aware of the entire moral law, the superhuman mind must be supremely intelligent
(1973, 191–197). Moreover, Ewing contends, since it is implausible to suppose that a being that
possessed perfect moral knowledge would not act on it, and since a being having such a unique
position in relation to morality cannot reasonably be supposed to be just one finite being among
others, the being in question must be God and the existence of objective moral values therefore con-
stitutes an independent argument for God’s existence.
One way of attacking this argument is to question the premise that the objectivity of moral
values must consist in actual existence somewhere. As Ewing himself formulates this objection:
“We cannot reasonably hold that there is an exact correlation between each true judgement and a fact
or element in reality to which it corresponds” (1973, 193–194). Ewing offers the example of hypo-
thetical judgments. A hypothetical judgment can be true even when its antecedent is never realized
in actual fact, but we are not compelled to accept that, therefore, there exists in reality a hypothetical
fact corresponding to every counterfactual conditional. Ewing responds to this objection that if our
moral attitudes and judgments are true they must conform to some law or standard, and this law
or standard must have some place in reality. To this, however, it may be replied that even if we
accept the premise that if moral values are objective they must exist somewhere, they may exist as,
for example, Platonic forms rather than in the mind of God, and so theism cannot necessarily be
inferred even if the idea that moral values must exist somewhere in metaphysical reality is conceded.
Some have argued from the premise that there exist objective moral facts or truths to a theistic
conclusion on the grounds that such truths are metaphysically odd. They might be considered odder
than logical or mathematical facts, logical and mathematical facts being necessary and analytic
truths. Even if moral truths are taken to be necessary, they are not analytic and hence require more
explanation than the truths of logic and mathematics (Wainwright 2016, 66). Theistic metaphysics
can perhaps address the issue of the oddness of moral facts more satisfactorily than a naturalistic
view of the world since, as Robert Gay puts it, “the theist has a suitably ‘queer’ world-view into
which to fit these ‘queer’ objects” (Gay 1987, 123). On a naturalistic conception according to
which what ultimately exists is space, time, and atoms, it is difficult to see how moral values can
be part of the fundamental structure of reality. The existence of objective values arguably seems
much less anomalous in a world in which the Good is, as a key attribute of the divine creator of
the world, basic to reality. And such a world is the world as conceived by religious metaphysical out-
looks such as theism (Wainwright 2016, 67). This argument appears to be a promising version of
the argument for theism from the objectivity of moral value.
C. S. Lewis (1960) argues for objective morality and from it infers the existence of God. He
argues that we know the difference between right and wrong naturally and do not need to be taught
it. Both sides in disagreements where ethical matters are relevant tend to appeal to the same laws or
principles of proper human conduct even if they disagree about how to apply them in a given prac-
tical instance. For Lewis, the existence of the objective moral law can be accounted for either natu-
ralistically or in religious terms. He argues, in somewhat similar fashion to Newman’s argument
from conscience, that the moral law within us, pushing us toward certain ways of behaving, is an
intimation of something beyond the natural world, something transcendent.
An obvious naturalistic objection to this approach is that one can offer an evolutionary account
of our belief in or feelings of moral obligation and indeed of belief in the objectivity of moral
norms. Baggett and Walls respond that such an account would fail to explain obligation itself—
collections of atoms could not generate genuinely binding moral requirements (2011, 11). Like

Ethics: Theism
others, they also draw attention to the peculiarly exacting nature of moral obligations: they demand
to be obeyed even if they conflict with our interests, praise and blame are strongly attached to them,
and they require pure motivation (2011, 14–17). Such features, they argue, are very difficult to
account for on a purely naturalistic conception of the universe (2011, 18). Nagel (1997) argues that
naturalistic accounts of ethical principles are inadequate because they overlook the normative force
of ethical reasoning. Naturalistic accounts provide information about what is happening when one
reasons ethically but miss the “inside” perspective that compels the ethical reasoner to action. A fur-
ther objection to Lewis’s line of argument comes from Henry Sidgwick (1962). Like Lewis, Sidg-
wick believes that the fundamental principles of morality are self-evident. But rather than arguing
from self-evident morality to the existence of God in order to account for objective morality, Sidg-
wick maintains that basic moral principles do not require any explanation precisely because of their
self-evident nature. The theist may respond, however, that the existence of God is needed to explain
a priori knowledge. (See Aaron Segal’s chapter in this book, which argues that a priori knowledge is
more compatible with theism than with atheism.)
It thus appears that objections to important forms of the argument for theism from the objec-
tivity of moral value can be successfully countered, and the argument from the metaphysical odd-
ness of objective moral truths may be especially promising.
THE DIVINE COMMAND THEORY OF ETHICS
As indicated above, a divine command theory of ethics, at least in a straightforward form, goes signif-
icantly further than arguing from morality to the existence of God. In such a form, it argues that
moral requirements are dependent on God’s commands. Fundamental moral concepts such as right
and wrong are defined in terms of what God prescribes or forbids.
Discussions of divine command theory often begin with a celebrated passage in Plato’s Euthyphro
(c. 380 BCE) in which Socrates poses the following question: “Is what is holy holy because the gods
approve it, or do they approve it because it is holy?” The contemporary philosophical literature tends
to recast Socrates’s query, which it terms the Euthyphro dilemma, in monotheistic and more modern
terminology. The horns of the dilemma are then refashioned in the following way: Is it the case that
(1) an act is right because God commanded (or wanted or willed or approved) it, or
alternatively, is it the case that
(2) God commanded (or wanted or willed or approved) this act because it is right?
According to horn (1), God—or more precisely his command or will—determines morality; moral-
ity depends on God. This is divine command theory (henceforth DCT), also sometimes referred to
as divine command morality or, because of the central role played by God’s will in the theory, theo-
logical voluntarism. On horn (2), by contrast, morality is independent of God; God commands or
wills certain things because of moral considerations that are not contingent upon him. (2) is often
referred to by the term “autonomy,” since on (2) ethics is autonomous in the sense of being inde-
pendent of God’s commands or will.
Formulation (1) as it stands expresses a rough and rather strong form of DCT; there are, how-
ever, as discussed below, more precise interpretations, and there are also weaker interpretations.
Similarly, (2) expresses an autonomy thesis only in broad and rather strong terms; there exist both
more exact readings and weaker readings.
Parenthetically, it is important to note that (1) and (2) use the deontological concept of right-
ness. There are indeed versions of DCT on which the theory applies only to deontological and not
to axiological concepts such as goodness (see, e.g., R. Adams 1981, 93−97; Quinn 1979, 309–310).
Nevertheless, here and throughout this chapter, the category of the rightness of actions is intended
to stand for all moral ascriptions to any kind of thing to which they can properly be ascribed.
The various forms of DCT to be considered below share with (1) and with each other the core
idea that morality is in some sense dependent on God. An important point to emphasize at the
Ethics: Theism
outset is that, as various philosophers have pointed out, even the idea that, for example, God’s com-
mand is both a necessary and a sufficient condition of the moral rightness of an act fails adequately
to convey the dependence of the act’s moral rightness on God’s command. If God’s command is a
necessary and sufficient condition of the moral rightness of an act, then the following equivalence
holds: (x) (RxCx) where R = “right” and C = “commanded by God.” On this equivalence, the
actions that God has commanded are coextensive with whatever actions are morally right. But this
still does not capture the idea that an action is right because God has commanded it; the equivalence
does not express what these philosophers term the asymmetry of DCT (see, e.g., Burch 1980, 281–
284; Wierenga 1983, 389; Sagi and Statman 1995, 20). What might be termed a “full” version of
DCT involves both conveying this dependence and explaining its nature.
The most straightforward full version of DCT is an ontic version on which it is claimed that
the moral rightness or wrongness of an act depends for its very existence on God’s command or pro-
hibition. Put more formally, God’s command is a necessary and sufficient condition of the moral
rightness or wrongness of an act, and in addition the moral rightness or wrongness of an act
depends for its existence on God’s command or prohibition.
Ontic DCT seems to be well illustrated in the writings of the scholastic philosophers John
Duns Scotus and William of Ockham, with whom attempts to trace the origins of DCT in the his-
tory of philosophy often begin. Though several medievalists have challenged the claim that Duns
Scotus espouses DCT, Duns Scotus does sometimes appear to espouse a radical kind of ontic
DCT. For instance: “it can become legitimate to kill … a man, namely, if God should revoke this
precept, Do not kill … and not only legitimate, but meritorious, namely, if God should give a com-
mand to kill, as He gave a command to Abraham concerning Isaac” (Duns Scotus, bk. 3, chap. 38,
q. 1; translation from Idziak 1979, 52). William of Ockham writes:
The hatred of God, theft, adultery, and actions similar to these according to the common law
… can even be performed meritoriously by an earthly pilgrim if they should come under a
divine precept, just as now the opposite of these in fact fall under a divine command. And,
with a divine command in effect for the contrary of these, it is not possible that anyone
meritoriously or rightly engage in such acts, because these are not performed meritoriously
unless they come under a divine command. (bk. 2, q. 19, O; translation from Idziak 1979,
56–57)
The import of this passage is that the moral status of an action depends solely on God’s com-
mand. Theft, adultery, and so on are morally wrong at present only because “the opposite of these
in fact fall[s] under a divine command”—that is, because God forbids these actions. If these actions
were commanded by God, they would be “performed meritoriously.” Passages such as this have led
many to interpret William of Ockham as an advocate of ontic DCT, although some scholars have
argued that his position is more complex (see, e.g., M. Adams 1986).
DCT IN THE HISTORY OF PHILOSOPHY AND THEOLOGY

Before discussing forms of DCT other than ontic DCT, it is appropriate to indicate briefly the broad
contours of the debate concerning DCT and autonomy in the history of philosophy and in some
major religious traditions. Martin Luther and John Calvin endorse DCT, and this Christian tradition
of support for DCT continued into the twentieth century with writers including Emil Brunner, Karl
Barth, and Carl Henry. There are, however, conflicting traditions within Christianity (see Rooney
1996, chaps. 4–5). The DCT versus autonomy issue was also keenly debated in Islamic theology, with
the Ashʿarites endorsing DCT and the Muʿtazilites supporting autonomy (in the sense defined earlier).
The Jewish tradition has embraced a variety of positions, with support for DCT becoming more evi-
dent in relatively more recent texts (Harris 2003). More recent advocates of DCT in the history of
philosophy include René Descartes, John Locke, and Ludwig Wittgenstein.
The Western philosophical tradition has also witnessed much opposition to DCT. In response
to the dilemma that Euthyphro puts to him, Socrates argues that the good is not good because the
gods approve it; rather the gods approve it because it is good. Similarly, the Cambridge Platonists,
such as Ralph Cudworth (1731), emphasized the existence of an absolute standard of right and
wrong, grounded in reason and independent of appeals to divine authority. Richard Price is another

Ethics: Theism
figure in the history of philosophy who has supported autonomy. Kant was the author of a seminal
critique of DCT, famously arguing that obeying commands only because God issued those com-
mands cannot count as truly moral behavior. If behavior is to have moral worth, Kant insisted, it
must be not heteronomous but rather autonomous, that is, self-legislated or self-imposed. Thus
divinely revealed morality cannot justifiably be considered moral. Revealed morality is, on the Kan-
tian view, a contradiction in terms.
DCT AS AN EPISTEMIC THESIS

A further form of DCT understands morality to be dependent on divine command not ontically
but epistemically. That is to say, action A may be morally right independently of God’s command,
but it is only through God’s command that we know that it is morally right. This epistemic version
of DCT can itself take different forms. One strong form is:
(3) We would have no moral knowledge without divine commands.

(3) clearly makes no ontic claim: it does not deny the existence of morality or moral obligation in
the absence of God’s commands. It claims merely that, without those commands, we would not
know anything about morality. (3) is nevertheless plainly a very radical version of DCT as an episte-
mic thesis. It claims that moral knowledge is totally unachievable by human beings without God’s
assistance.
Arguably, the most cogent version of the radical (3) assumes not only that human beings had no
moral knowledge at all prior to receiving God’s commands but also that they were unaware of this lack
of knowledge (Sagi and Statman 1995, 85). On this version of (3), human beings were completely
unaware not only of any moral obligations but even of moral concepts. Through divine commands
they became aware not only of their actual moral obligations but also of moral language itself.
(3) runs into serious empirical difficulties (see, e.g., Sagi and Statman 1995, 85–86). Its
assumption that the world lacked all moral knowledge prior to God’s revelation is implausible in
the light of our awareness of significant human cultures that claim that they attained moral knowl-
edge without divine revelation. This objection might be countered by the assertion that the moral
knowledge of atheist societies ultimately does rely on divine revelation, and that this fact has simply
been forgotten by those societies in the course of time. It is very difficult, however, to see how this
counterclaim could ever be substantiated.
Although the term “divine command theory” tends to be used in the literature as a broad
description of any theory that asserts the dependence of morality on God, at this point it will be
helpful to introduce a distinction made by some philosophers between forms of DCT that focus
on God’s command and forms that concentrate on his will (see, e.g., Murphy 1998; Harris
2003). Some (e.g., Murphy 1998) in fact prefer to interpret DCT in terms of God’s will, while
others favor a command interpretation (e.g., R. Adams 1979, 76–77).
One way to make the distinction between God’s commands and his will in the context of DCT
is to understand God’s will to mean how God wills human beings to conduct themselves without
having communicated this in any act of explicit revelation, and to take God’s command to mean
an explicit divinely revealed command or commands (such as, to cite a famous biblical example,
the Ten Commandments). With this distinction in mind, and returning to the issue of epistemic
DCT, we can see that (3) can be taken in two ways. First:
(3C) We would have no moral knowledge without God’s revealed commands.

Alternatively:
(3W) We would have no moral knowledge without our access to God’s will.
(3W) is intended to convey the thought that we would have enjoyed at least some moral knowledge
without God’s revealed commands, but that we would not have any moral knowledge without access
to his will through our cognitive equipment.
Ethics: Theism
(3W) enjoys at least significant initial plausibility from a religious perspective. Presumably,
from a religious point of view, we would want to explain any moral knowledge not received through
explicit revelation in terms of cognitive apprehension of God’s will (though, interestingly, some have
argued that morality, in the form of the principle of utility, gives us access to what God’s will is
rather than vice versa (Paley [1786] 2002, bk. 2, chap. 4). Moreover, the empirically based objec-
tions that can be leveled at (3) seem to have more force when applied to (3C) than when directed
at (3W). (3W) is able to account for the moral knowledge claimed by atheist societies without
resorting to the claim that such knowledge ultimately depends on a distant revelation of which
the atheist society has no recollection; moral knowledge is available through cognitive apprehension
of the divine will. Of course, the atheist denies that moral knowledge is attained in this way, but
what the atheist denies is at least closer to the issue of God’s existence that is by definition at stake
between the atheist and the theist. That is to say: the atheist may well be prepared to concede that if
there were a God, he might well make some moral knowledge easily available to human beings
through their cognitive faculties. But the atheist would perhaps be less prepared to admit that
(3C) was anything more than utterly fanciful.
As an alternative to adopting (3W), advocates of epistemic DCT might attempt to modify (3C)
in such a way as to give it greater plausibility. One way of doing this would be to claim, more mod-
erately than (3C), that what God’s commands provide us with knowledge of is whether certain
types of acts count as instances of a morally forbidden (or obligatory) broader category of act. Thus,
for example, a moderate alternative to (3C) might concede that we know, independently of divine
revelation, that all acts of murder are wrong, but that what God’s revealed commands supply us
with is the ability to identify precisely which kinds of acts count as murder. Without this revelation
we may not have known, for example, whether killing an assailant in self-defense, or killing an
enemy in battle, or capital punishment counts as murder.
This kind of moderate form of (3C) seems to be advocated by the great medieval Jewish philos-
opher Saadia Gaon:
Although reason considers stealing objectionable, there is nothing in it to inform us how a
person comes to acquire property so that it becomes his possession. It does not state, for
instance, whether this comes about as a result of labor, or is effected by means of barter, or by
way of inheritance.… Besides these, there are many other uncertainties pertaining to this
subject which would take too long and would be too difficult to enumerate. The prophets,
therefore, came along with a clear-cut decision for each instance. (1955, 146)
There are, however, additional difficulties that even moderate versions of (3C) need to address, and
further refinement may be required in order to yield a defensible version of epistemic DCT (for fur-
ther discussion see Harris 2003, 10–15). Other versions of DCT as an epistemic thesis are also pos-
sible (see, e.g., Sagi and Statman 1995, 88–90).
It is important to emphasize the independence in general of ontic and epistemic forms of DCT.
For example, even on quite a strong version of the epistemic independence of morality from God, it
is still a logical possibility that morality is ontically dependent on God, in other words, that some
ontic version of DCT is true. By the same token, morality might be ontically independent of
God, but we might at the same time not be capable of knowing anything about morality without
divine assistance, so that DCT would obtain at the epistemic level. It turns out, therefore, that some
versions of DCT and some versions of autonomy are logically compatible (but see the discussion of
“The Objection from Epistemic Asymmetries” in Quinn 1978, 43–45, and Sagi and Statman 1995,
93–96).
DCT AS A SEMANTIC THESIS

Another possible form of DCT is as a semantic thesis. DCT might be taken as claiming that:
(4) “x is morally right” means “x is commanded by God.”

One way in which (4) can be understood is as a factual statement about the way in which moral
concepts are used. In other words, when people use moral expressions like “good,” “right,” and

Ethics: Theism
“wrong,” what they truly intend is statements about God’s commands. This position is the focus of
an important discussion by Robert Adams that itself serves as a good illustration of important issues
concerning semantic DCT. Let us term Adams’s precise initial formulation (4a):
(4a) “the word ‘wrong’ in ethical contexts means ‘contrary to God’s commands.’” (R. Adams
1981, 83; emphasis in original).
(4a) faces an obvious objection that is noted by Adams himself: even if, in some cases, what people
mean by “x is wrong” is “x is contrary to God’s commands,” there are clearly very many instances in
which this is not the case (1981, 84). For example, atheists who articulate moral judgments obvi-
ously do not intend to make statements about God’s commands.
A further objection to (4a) relies on G. E. Moore’s celebrated “open question argument” (one
version of the argument appears in Moore 1962, 16). Moore’s argument runs as follows: if the
meaning of “good” is identified with natural property x, then it should make no sense to ask “But
is something with property x indeed good?” That would be like asking “Is a single man one who
is unmarried?”—a question that makes no sense, since the word “single” means “unmarried.” If it
does always make sense to ask “Is something with property x indeed good?” then it follows that
the meaning of “good” cannot be identified with property x. Similarly, the claim of (4a) that the
word “wrong” in ethical contexts means “contrary to God’s commands” can be challenged, since
the question whether an act that God has commanded us not to do is indeed wrong is always a
meaningful question. However, the open question argument as deployed in the context of DCT
has not convinced many (see, e.g., Rooney 1996, 18).
Adams’s awareness of the first objection to (4a), that people often manifestly do not mean “x is
contrary to God’s commands” when they say “x is wrong,” prompts him to suggest a modified ver-
sion of (4a): “The theory cannot reasonably be offered except as a theory about what the word
‘wrong’ means as used by some but not all people in ethical contexts. Let us say that the theory
offers an analysis of the meaning of ‘wrong’ in Judeo-Christian religious ethical discourse”
(R. Adams 1981, 84).
Let us express Adams’s modified position as follows:
(4b) The word “wrong” in Judeo-Christian ethical discourse means “contrary to God’s
commands.”
As Adams is aware, (4b) must still explain how it is that theists and atheists engage in moral dia-
logue if they are speaking, as it were, totally different languages. For theists use “wrong” in the sense
of “contrary to God’s commands,” whereas atheists make no reference to God whatsoever. Adams
attempts to show that sufficient common ground still exists between these two languages for dia-
logue to be possible.
Whether or not Adams is successful in that task, however, serious difficulties with (4b) remain.
One might first ask a very straightforward question concerning the truth of (4b). When Jewish and
Christian believers use the word “wrong” in ethical discourse, do they really always mean “contrary
to God’s commands”? At one point, Adams writes: “‘It is wrong to do X’ will be assented to by the
sincere Jewish or Christian believer if and only if he assents to ‘It is contrary to God’s commands to
do X.’ This is a fact sufficiently well known that the known believer who says the one commits him-
self publicly to the other” (Adams 1981, 89).
This seems to be just false. For example, a Jewish or Christian believer might very well hold
that the content of ethics is not exhausted by divine commands, that there are many things that
are morally wrong about which God has issued no commands. There may, for instance, be matters
“beyond the letter of the law,” to cite a well-known concept in rabbinic literature, that God’s laws
permit but that the believer considers to be ethically prohibited. Or there may simply be lacunae
where there are things that the believer thinks are wrong but concerning which God has issued
no commands. Or consider the sphere of ritual law, such as the prohibitions in Judaism against eat-
ing nonkosher food, eating leaven on Passover, and working on the Sabbath. The traditional Jewish
believer takes these actions to be contrary to God’s commands, but he or she probably does not
Ethics: Theism
deem them ethically wrong. Yet another possibility is that a Judeo-Christian believer might even
hold (see the discussion of the “conflict thesis” below) that there are things that God has com-
manded that are themselves morally wrong, but that divine commands override morality. For exam-
ple, Søren Kierkegaard ([1843] 1954) argues that God’s command to Abraham in Genesis 22 to
sacrifice Isaac was morally wrong but religiously right, appealing to his celebrated notion of the “tel-
eological suspension of the ethical.”
A further important point, slightly adapted here from an observation made by Avi Sagi and
Daniel Statman (1995, 30–31), concerns semantic versions of DCT in general, and therefore
applies not only to (4b) but equally to (4a) and (4). This point has to do with the status of seman-
tic forms of DCT as interpretations of DCT. Propositions like (4b), (4a), and (4) claim that there
is a logical equivalence between the term “wrong” and a claim about God, and assert that this
equivalence originates in a linguistic usage. But, as mentioned above, logical equivalence is too
weak to capture the notion of morality’s dependence on God: DCT (or certainly what is termed
above “full DCT”) is “asymmetrical.” A “full” semantic interpretation of DCT needs to claim both
that (i) x’s being wrong is synonymous with x’s being contrary to God’s command, and that (ii) x’s
being contrary to God’s command is explanatorily prior to x’s being wrong. But (4b), like (4a) and
(4), claims only (i) and not (ii); it is symmetrical rather than asymmetrical. Adams tries to resolve
this difficulty by saying that, on his theory, x’s being contrary to God’s command “conceptually
precedes” its being wrong. But Adams does not explicate the meaning of this conceptual prece-
dence. The idea of conceptual precedence thus seems to smuggle in kinds of dependence other
than semantic dependence, since without them we cannot understand why and in what sense
God’s command conceptually precedes moral judgments. If this is correct, then (4b), like (4a)
and (4), is either not a full version of DCT, or it is full but does not claim that the dependence
of morality on God is semantic.
MODIFIED DCT
An important recent development in discussion of DCT is the novel approach of those Christian
philosophers who have advocated a “modified” version of DCT. Modified DCT was pioneered by
Robert Adams (see, e.g., R. Adams 1981). The crucial notion for modified DCT is that of a God
who loves his human creatures and is moved by concern for them. Adams holds that if God com-
manded cruelty for its own sake it would not be wrong to disobey God. Indeed, were God to com-
mand cruelty for its own sake, the religious believer’s concept of ethical wrongness would “break
down” and he or she could no longer call any action ethically wrong. This is because the religious
believer’s statement that x is ethically wrong, for Adams, not only carries implications concerning
God’s commands but also presupposes that certain conditions for the applicability of the believer’s
concepts of right and wrong are met. One important condition of this sort is that God loves us,
or, more specifically, that he does not command cruelty for its own sake.
What is gained by modifying DCT in this way, of course—and Adams’s explicit motivation
in developing modified DCT—is that it apparently facilitates both commitment to DCT, with
all DCT’s religious attractions, and the avoidance of an apparently morally unpalatable ramifica-
tion of DCT, namely, that it would be wrong not to practice cruelty for its own sake if God
commanded it.
The modified DCT clearly conceives of believers as valuing kindness and disvaluing cruelty
independently of their relation to God’s commands. The central philosophical difficulty with
Adams’s modified DCT, as he is himself aware, is that these independent valuations that modified
DCT allows appear fatal to the core DCT notion that morality depends on divine commands.
For these independent valuations, focused (positively) on kindness and (negatively) on cruelty, are
clearly moral valuations. And if they are, then morality is not after all dependent on divine com-
mands. Adams seems unable to surmount this crucial difficulty.
A further possible objection to modified DCT is that it is theologically unnecessary (see Levine
1994). Some theists argue that it is logically impossible for God to command things like cruelty:
that God has a loving character is, it is therefore suggested, an assumption that the unmodified
divine command theorist already makes. Since it is a central claim of many forms of theism that

Ethics: Theism
God is perfectly good, the unmodified divine command theorist already assumes that God possesses
a loving character with which commanding cruelty is incompatible. Part of many theists’ conception
of God is that God cannot command cruelty. This objection provides a reason for considering mod-
ified DCT theologically unnecessary from the perspective of many theists in many important reli-
gious traditions.
WEAK DEPENDENCE
If major religions such as Judaism and Christianity endorse DCT, it is easy to identify and account
for the role of God in the moral arena in these traditions. But if these traditions contain strands—as
they do—that endorse autonomy rather than DCT, how do these strands account for this divine
role? Why is divine revelation commanding moral laws necessary at all if DCT is rejected? There
must be some significant role that, according to these strands, God plays in the moral sphere. For
their view to be plausible, those who claim that religious traditions in which both God and morality
are central substantially advocate autonomy rather than DCT need to clarify the role occupied by
God in relation to morality. Here the concept termed by Sagi and Statman “weak dependence”
becomes important. Weak dependence of morality on divine commands or religion is some kind
of dependence that stops short of DCT (Sagi and Statman 1995, 81–83, though the authors extend
the term to cover epistemic DCT as well). Possible roles for God and religion in the moral arena
that come under the rubric of weak dependence include the idea that religion is necessary for moral
activity (Sagi and Statman 1995, 97). It might be argued that human beings, even when conscious
of their moral obligations, are incapable of acting in accordance with their awareness because of
their nature, and that only religion can help them overcome the conflict between their nature and
moral requirements and enable them to act morally. Less radically, it might be urged that, though
capable of moral action, human beings find it very difficult to behave ethically without religion.
More moderately still, it might be claimed that religion heightens the likelihood that individuals
or societies will behave morally, because of the capacity of religion to foster moral activity. Needless
to say, all of these claims are empirical and highly controversial, and counterclaims that religion
makes immoral behavior more likely are legion. A salient example comes from the American physi-
cist Steven Weinberg: “With or without religion, good people can behave well and bad people can
do evil; but for good people to do evil—that takes religion” (1999).
THEOLOGICAL MOTIVATIONS FOR DCT

There are powerful theological motivations for asserting the dependence of morality on God. A
strong and widespread religious intuition wishes to assert God’s absolute sovereignty, independence,
and omnipotence. According to this intuition, God is sovereign over everything and does not
depend on anything; on the contrary, everything depends on him. If God’s command is subject
to standards of good and evil or right and wrong that are independent of him, then his sovereignty,
freedom, and supreme power are apparently compromised.
A brief consideration of the situation regarding the truths of logic and mathematics may, how-
ever, be illuminating in this context. For it might be thought, for instance, that God’s sovereignty is
compromised by the mere existence of truths that are independent of him, or that his omnipotence
is vitiated if he cannot do the logically impossible. And we might therefore expect that religious tra-
ditions such as Christianity and Judaism would unequivocally champion the dependence of logical
and mathematical truths on God. Indeed, Descartes, for example, asserted clearly on several occa-
sions the existence of such dependence, for example in the following passage: “The mathematical
truths … have been laid down by God and depend on Him entirely no less than the rest of his crea-
tures. Indeed to say that these truths are independent of God is to talk of Him as if He were Jupiter
or Saturn and to subject Him to the Styx and the Fates” (letter to Mersenne, 15 April 1630; trans-
lation from Kenny 1970, 11).
Nevertheless, this kind of view has equally distinguished opponents. Thomas Aquinas explicitly
maintained that God’s sovereignty is not undermined by being restricted to the realm of what is log-
ically possible (Summa theologiae Ia.25.3). Similarly, in the Jewish tradition there are conflicting
views. Saadia Gaon and Joseph Albo state explicitly that God cannot do the logically impossible
Ethics: Theism
(Saadia Gaon 1955, 134; Albo 1929, vol. 1, 179), and Saadia adds in the same passage that this
implies no restriction on God’s omnipotence: God, says Saadia, is able to do everything, and the
absurd is nothing. Some Hasidic thinkers take issue with the medieval view. In light of the different
positions adopted regarding the truths of logic and mathematics in both Christian and Jewish the-
ology, we cannot rule out in advance the possibility of finding support for autonomy as opposed
to DCT in those traditions simply on the assumption that such endorsement would conflict with
qualities such as omnipotence, sovereignty, and freedom that these traditions ascribe to God. And
indeed, some philosophers and theologians such as Cudworth have endorsed autonomy, arguing
that moral truths, like the truths of logic and mathematics, are necessarily true and that, since
omnipotence is the power to do what is possible rather than impossible, God’s inability to make
moral truths false does not compromise his omnipotence.
THEOLOGICAL AND MORAL OBJECTIONS TO DCT

At the same time, however, DCT has clear theological and moral disadvantages that must be
addressed if DCT is to be successfully defended and that affect it particularly in its ontic iterations.
As Cudworth points out and many have highlighted since, DCT raises the problem of arbitrariness:
“nothing can be imagined so grossly wicked … but if it were supposed to be commanded by this
omnipotent deity, must needs … forthwith become holy, just and righteous” (1731, 10). If God
were to command the torture of innocent children, that would become morally right. There is a fur-
ther problem concerning arbitrariness at a more general level: God’s total freedom in determining
morality seems to undermine the rationality and supreme moral goodness, indeed perfection, that
many theists would want to ascribe to him, since according to DCT God prohibited, for example,
murder but might just as easily have permitted or commanded it. Adams’s modified DCT discussed
above, according to which morally obligatory acts are those commanded by a loving God, is an
attempt to block the problem of arbitrariness.
A further standard objection to ontic DCT focuses on one of its clear implications: as Fyodor
Dostoevsky formulates it in his novel The Brothers Karamazov (1879−1880), without God, every-
thing is allowed. Since, according to ontic DCT, God’s command is (minimally) a necessary condi-
tion of morality, without divine command there is no morality, and therefore in possible worlds
without God nothing is morally wrong. But this runs up against the plausible assumption that mor-
ally wrong acts such as killing for no reason are wrong in all possible worlds, that is, are necessarily
and not merely contingently wrong. Despite extensive attempts to resolve this problem (see, e.g.,
Burch 1980; R. Adams 1987), ultimately it seems to show that either moral obligations are neces-
sary and independent of God, or dependent on God as ontic DCT claims but contingent (Sagi
and Statman 1995, 55–61).
Yet a further standard objection to DCT, closely related to Kant’s celebrated critique men-
tioned earlier, is that it undermines the autonomy of moral agents. DCT suggests that agents should
not exercise their ability to make moral judgments. They should ignore their own consciences and
instead simply accept as right or wrong whatever is commanded or forbidden by God. James
Rachels (1981) goes so far as to argue not merely against DCT but against theism on the basis of
moral autonomy. For Rachels, there cannot exist any being who is God because the worship of
God would entail total subservience and would therefore be inconsistent with the autonomous
moral decision making that characterizes moral agency. (For a critique of Rachels’s assumption that
worship and moral autonomy must conflict, see Sagi and Statman 1995, 139–143.) It might be
countered that DCT still allows for the exercise of autonomous moral reasoning by agents in inter-
preting or inferring practical conclusions from God’s commands. However, this might not be suffi-
cient for those who value a robust moral autonomy unconstrained by the parameters of divine com-
mands.
Finally, some point out that a divine command cannot create a moral obligation unless there is
a moral reason independent of God’s commands for obeying them. The additional premise that one
ought to obey God’s commands, it is argued, is the only thing that can allow God’s commanding x
to be sufficient moral reason to do x. There are thus several major difficulties that need to be
resolved if DCT, particularly in its ontic form, can be confidently espoused. If DCT were the best

Ethics: Theism
theory of what is right and wrong, this would lend support to theism, but whether indeed it is the
best such theory is far from clear. Moreover, as this section has demonstrated, even theists are
divided between those who advocate some version or other of DCT and those who endorse auton-
omy.
DCT AND THE CONFLICT THESIS

A final important clarification regarding DCT is in order. Kierkegaard in Fear and Trembling
famously argues that the narrative of the binding of Isaac, the Akedah, in Genesis 22 involves “the
teleological suspension of the ethical” ([1843] 1954, 47). The standard interpretation of Kierke-
gaard here is that he considers the biblical story to show that religious faith can sometimes demand
that morality be put aside in order to fulfill the higher, religious obligation to obey God. Abraham,
who is willing to disregard a clear moral requirement because of God’s command, is, as Kierkegaard
([1843] 1954) calls him, a “knight of faith.” Kierkegaard thus supports what Sagi and Statman label
the “conflict thesis”: God’s command can sometimes conflict with the requirements of morality, and
when it does, it overrides these requirements (Sagi and Staman 1995, chaps. 6–8). In the medieval
era Martin Luther already drew attention to a very important corollary of DCT that is often over-
looked: if God’s will (or command) determines morality, it follows logically that there can be no
conflict between his will (or command) and morality. Although, from a logical perspective, DCT
and the conflict thesis are therefore mutually exclusive, Sagi and Statman correctly point out that
the conflict thesis expresses religious intuitions similar to those of DCT, such as those that empha-
size God’s sovereignty and freedom (1995, 117). God is sovereign according to the conflict thesis
because he alone determines how human beings ought to behave; he is free because no law, not
even the moral law, limits his actions or commands.
FURTHER CONSIDERATIONS
It remains in this chapter to discuss briefly some further ethical concepts and considerations and
their possible impact on the theism/atheism controversy. It would appear that none of these factors
decisively militates in favor of either theism or atheism, and each factor is compatible with both
positions.
ETHICAL DIVERSITY
Ethical diversity across historical periods and across different societies and cultures, a much-
discussed phenomenon in several disciplines, is also of philosophical import. Some ethical nonobjec-
tivists might argue that ethical diversity supports the view that morality is not an objective matter
and is simply relative to particular cultures or places. It is sometimes argued that disagreement about
objective, factual matters is far more limited than disagreement and diversity on moral issues and
therefore that morality cannot be objective. There will be little disagreement that it is raining out-
side my house today if in fact it is, whereas disagreements on capital punishment or economic jus-
tice abound.
It can be effectively responded, however, that one must compare like with like, and one could
also contrast the consensus that a child about to undergo a medical operation should be anesthetized
with many cases of widely divergent views among scientists regarding the interpretation of factual
data (Bambrough 1979, 18–19). More important, the fact that there is disagreement, however long-
standing and deep, does not exclude the possibility that some parties to the dispute are right and
others wrong (Bambrough 1979, 18–19; Byrne 1999, 11). The existence of widely divergent moral
codes across human historical, geographical, and cultural axes likewise does not demonstrate the
falsehood of ethical objectivism. Ethical objectivists can also explain ethical diversity on the basis
that moral conceptions are embedded in broader worldviews that are diverse, for example, views
about human nature or about history (Byrne 1999, 11). Moreover, the contextual nature of moral
thought must be acknowledged. What the moral truth consists in can vary with circumstances.
Peter Byrne suggests the analogy of engineers building bridges. The engineers may build according
Ethics: Theism
to true principles of civil engineering and correct calculations of the strengths of the particular struc-
tures they erect, but how any particular bridge is built varies according to the specific geological cir-
cumstances, the materials from which that bridge is constructed, and so on. Civil engineering
applies correct general notions in the light of specific facts to provide a unique plan for how any par-
ticular bridge is to be built. Similarly, from a moral objectivist’s perspective, what is right and good
is a matter of objective truth, but this does not entail that the specifics of what is right and good will
remain the same regardless of circumstances (Byrne 1999, 11). It may be objectively morally right
to defend a work colleague from malicious accusations that one knows to be false, but the specifics
of how to do this will depend on one’s professional role in the workplace, the specific allegations,
and so on.
A further way of accounting for ethical diversity while maintaining the position that morality is
an objective matter is to point out that ethical disputes are often over facts rather than values. For
example, there may be agreement that military action by a state is justified in self-defense in the face
of real and immediate threat but disagreement over the assessment of the threat. For all these rea-
sons, it would thus seem that the undisputed fact of ethical diversity is compatible with objectivism.
However, as discussed earlier in this chapter, even once objectivism is established or assumed, there
is still a major task involved in showing theism to be the best explanation of it.
NATURAL LAW
The natural law tradition maintains that there is a natural moral law that is knowable by and acces-
sible to all human beings. The best known natural law theory of ethics is that of Aquinas in Summa
theologiae (1485). On Aquinas’s natural law theory, natural law is given by God and is an aspect of
divine providence. Aquinas’s natural law theory therefore clearly assumes theism and is incompatible
with atheism, so that if one were convinced of the soundness of Aquinas’s natural law theory this
would entail a commitment to theism. “Natural law ethics evaporates if nature is purposeless” (Pope
2005, 159). Some philosophers have written within the natural law tradition affirming the authority
and epistemic accessibility of the natural law to all people but denying that the natural law is God-
given (see, e.g., M. Moore 1996; Foot 2001). On this secular version of natural law theory, some
moral principles can be established by reason.
Whether the idea of natural law supports theism cannot be assessed in the absence of an agreed
definition of natural law. But there is no such agreed definition, since while classical natural law the-
ory depends on theistic assumptions (on traditional natural law theory, see Hittinger 2003), atheists
can argue that the concept of natural law can be coherently secularized.
VIRTUE
In the eighteenth century, Jonathan Edwards argued that true virtue must be intimately connected
with God: it mainly consists in “a supreme love of God” (Edwards [1765] 1989, 551). This is
because virtue, for Edwards, encompasses not only benevolence toward being in general but also
what he calls “complacence,” that is, delight in the intrinsic excellence of benevolence. Since God
is infinitely the most excellent being, virtue essentially involves the love of God. Thus benevolence,
however extensive, does not count as genuine virtue if it is not intimately connected to the love of
God. Yet there seems to be no compelling reason for an atheist to accept either that virtue must
encompass love for all being or for God. The fact that the virtuous actions of nonreligious people
are usually indistinguishable from those of theistic believers in terms of their content and beneficial
consequences seems to place the onus of proof firmly on those who, like Edwards, insist on an
intrinsic connection between theism and virtue.
HUMAN RIGHTS
In the contemporary world it is widely accepted that there are human rights, but it does not seem
that this acceptance supports theism rather than atheism or vice versa. This is because there are
plausible groundings for human rights in both theistic and atheistic terms, though both kinds of
grounding also face significant challenges. One could argue that human rights are innate in human
beings because they are God-given. Nicholas Wolterstorff suggests a more indirect grounding of

Ethics: Theism
human rights in theism in which such rights are grounded in the equal worth of every human being,
that worth existing by virtue of God’s love (Wolterstorff 2008). This approach has the advantage of
rooting human rights in something putatively eternal and metaphysical, but its obvious disadvantage
is that it will persuade only those who already accept or can be convinced to accept the relevant
assumptions about God, unless the argument is so strong that it shows that human rights can only
be grounded in a theistic outlook, in which case the atheist will still have the option of denying the
existence of human rights rather than embracing theism. Some ground the existence of human
rights totally independently of God in national or international law. This is clearly less controversial
than a theistic basis for human rights but has the drawback of making human rights subject to
changes in human laws. One could bolster this grounding of human rights in law by appealing to
moral justification for human rights, but, again, this moral justification could reasonably rest on
either a theistic or an atheistic metaethics.
This chapter initially considered three of the most important arguments from morality to the
existence of God and concluded that the argument for theism from the objectivity of moral value
may constitute the strongest such argument. The chapter then focused in detail on DCT, showing
how DCT takes many different forms and is controversial even among theists. It concluded that it is
far from clear that DCT can lend real support to theism. Finally, the chapter considered a range of
ethical concepts and considerations, including ethical diversity, natural law, and human rights, argu-
ing that none of these factors decisively militates in favor of either theism or atheism.
Bibliography
Adams, Marilyn McCord. “The Structure of Ockham’s Moral Theory.” Edwards, Jonathan. The Nature of True Virtue In Ethical Writings, edited by
Franciscan Studies 46 (1986): 1–35. Paul Ramsey (Vol. 8 of The Works of Jonathan Edwards). New Haven,
CT: Yale University Press, 1989. First published 1765.
Adams, Robert Merrihew. “Divine Command Metaethics Modified Again.”
Journal of Religious Ethics 7, no. 1 (1979): 66–79. Ewing, A. C. Value and Reality: The Philosophical Case for Theism. London:
Allen and Unwin, 1973.
Adams, Robert Merrihew. “A Modified Divine Command Theory of Ethical
Foot, Philippa. Natural Goodness. Oxford: Clarendon; New York: Oxford
Wrongness.” In Divine Commands and Morality, edited by Paul Helm,
83–108. Oxford: Oxford University Press, 1981.
Freud, Sigmund. Civilization and Its Discontents. Garden City, NY:
Adams, Robert Merrihew. “Moral Arguments for Theistic Belief.” In The Doubleday, 1958.
Virtue of Faith and Other Essays in Philosophical Theology, 144–163. New
York: Oxford University Press, 1987. Gay, Robert. “Moral Arguments for the Existence of God.” Modern Theology
3, no. 2 (1987): 117–136.
Adams, Robert Merrihew. Finite and Infinite Goods: A Framework for Ethics.
Grave, S. A. Conscience in Newman’s Thought. Oxford: Clarendon; New
Albo, Joseph. Sefer Ha-’Ikkarim (Book of Principles). Translated by Isaac Green, Ronald M. Religious Reason: The Rational and Moral Basis of Religious
Husik. Philadelphia: Jewish Publication Society, 1929. Belief. New York: Oxford University Press, 1978.
Baggett, David, and Jerry L. Walls. Good God: The Theistic Foundations of Harris, Michael J. Divine Command Ethics: Jewish and Christian Perspectives.
Morality. New York: Oxford University Press, 2011. London: RoutledgeCurzon, 2003.
Hittinger, Russell. The First Grace: Rediscovering the Natural Law in a Post-
Bambrough, Renford. Moral Scepticism and Moral Knowledge. London:
Christian World. Wilmington, DE: ISI Books, 2003.
Routledge and Kegan Paul, 1979.
Idziak, Janine Marie, ed. Divine Command Morality: Historical and
Burch, Robert. “Objective Values and the Divine Command Theory of Contemporary Readings. New York: Edwin Mellen Press, 1979.
Morality.” New Scholasticism 54, no. 3 (1980): 279–304.
Kant, Immanuel. Critique of Practical Reason. Translated by Lewis White
Byrne, Peter. The Philosophical and Theological Foundations of Ethics: An Beck. New York: Liberal Arts Press, 1956. First published 1788.
Introduction to Moral Theory and Its Relation to Religious Belief. 2nd ed.
Kenny, Anthony, ed. and trans. Descartes: Philosophical Letters. Oxford:
Basingstoke, UK: Macmillan; New York: St. Martin’s Press, 1999.
Clarendon, 1970.
Cudworth, Ralph. A Treatise concerning Eternal and Immutable Morality. Kierkegaard, Søren. Fear and Trembling. Translated by Walter Lowrie.
London: J. and J. Knapton, 1731. Princeton, NJ: Princeton University Press, 1954. First published 1843.
Duns Scotus, John. The Oxford Commentary on the Four Books of the Levine, Michael. “Adams’s Modified Divine Command Theory of Ethics.”
Sentences. In Opera Omnia, 2nd ed. Paris: Vives, 1891–1895. Sophia 33, no. 2 (1994): 63–77.
Ethics: Theism
Lewis, C. S. Mere Christianity. New York: Macmillan, 1960. First published Quinn, Philip L. Divine Commands and Moral Requirements. Oxford:
1942−1944. Clarendon, 1978.
Mackie, John L. The Miracle of Theism: Arguments for and against the Quinn, Philip L. “Divine Command Ethics: A Causal Theory.” In
Existence of God. Oxford: Clarendon, 1982. Divine Command Morality: Historical and Contemporary Readings,
Moore, G. E. Principia Ethica. Cambridge, UK: Cambridge University Press, edited by Janine Marie Idziak, 305–325. New York: Edwin Mellen
1962. First published 1903. Press, 1979.
Moore, Michael. “Good without God.” In Natural Law, Liberalism, and Rachels, James. “God and Human Attitudes.” In Divine Commands and
Morality: Contemporary Essays, edited by Robert P. George. Oxford: Morality, edited by Paul Helm, 34–48. Oxford: Oxford University Press,
Clarendon; New York: Oxford University Press, 1996. 1981.
Murphy, Mark C. “Divine Command, Divine Will, and Moral Obligation.” Rooney, Paul. Divine Command Morality. Aldershot, UK, and Brookfield,
Faith and Philosophy 15, no. 1 (1998): 3–27. VT: Avebury, 1996.
Nagel, Thomas. The Last Word. New York: Oxford University Press, 1997. Saadia Gaon. The Book of Beliefs and Opinions. Translated by Samuel
Rosenblatt. New Haven, CT: Yale University Press, 1955.
Newman, John Henry. “Proof of Theism” (1859.) In The Argument from
Conscience to the Existence of God according to J. H. Newman, edited by Sagi, Avi, and Daniel Statman. Religion and Morality. Amsterdam: Rodopi,
Adrian I. Boekraad and Henry Tristram. Louvain, Belgium: Editions 1995.
Nauwelaerts, 1961.
Sidgwick, Henry. The Methods of Ethics. 7th ed. Chicago: University of
Newman, John Henry. An Essay in Aid of a Grammar of Assent. Notre Dame, Chicago Press, 1962.
IN: University of Notre Dame Press, 1979. First published 1870.
Sorley, W. R. Moral Values and the Idea of God. Cambridge, UK: Cambridge
Nietzsche, Friedrich. “The Wanderer and His Shadow” (1879.) In University Press, 1918.
Human, All Too Human: A Book for Free Spirits. Translated by R. J.
Wainwright, William. Religion and Morality. London: Routledge, 2016.
Hollingdale. Cambridge, UK: Cambridge University Press, 1986.
Weinberg, Steven. “A Designer Universe.” Based on a talk given at the
Nietzsche, Friedrich. On the Genealogy of Morality. Translated by
Maudemarie Clark and Alan J. Swensen. Indianapolis, IN: Hackett, Conference on Cosmic Design of the American Association for the
1998. First published 1887. Advancement of Science, Washington, DC, 1999. http://www.physlink
.com/Education/essay_weinberg.cfm.
Paley, William. The Principles of Moral and Political Philosophy. Indianapolis,
IN: Liberty Fund, 2002. First published 1786. Wierenga, Edward. “A Defensible Divine Command Theory.” Noûs 17, no.
3 (1983): 387–407.
Plantinga, Alvin. Warrant and Proper Function. Oxford: Oxford University
Press, 1993. William of Ockham. On the Four Books of the Sentences. In Opera Plurima.
Lyon, 1495.
Pope, Stephen. “Reason and Natural Law.” In The Oxford Handbook of
Theological Ethics, edited by Gilbert Meilaender and William Werpehowski, Wolterstorff, Nicholas. Justice: Rights and Wrongs. Princeton, NJ: Princeton
148–167. Oxford: Oxford University Press, 2005. University Press, 2008.

TOPIC 14
Ethics: Atheism
WHY MORAL REALISM DOES NOT FAVOR THEISM 467
Jason Thibodeau
Assistant Professor of Philosophy
Cypress College, Cypress, CA
ACCOUNTING FOR SIMILARITIES AND DIFFERENCES IN 472

MORAL BELIEF
Thaddeus Metz
University of Johannesburg, South Africa
THE NATURE OF MORALITY AND OUR CAPACITY FOR 477

MORAL KNOWLEDGE
Bruce Russell
Wayne State University, MI
CONCEPTIONS OF HUMAN RIGHTS 482

David Neil
Lecturer in Philosophy
University of Wollongong, Australia
The aim of this chapter is not to consider particular ethical questions; it is to ask whether there are general
facts about morality and our ability to make moral judgments that count in favor of either theism or
atheism.
WHY MORAL REALISM DOES NOT FAVOR THEISM
Jason Thibodeau, Cypress College
MORAL REASONS WITHOUT GOD

A young child is suffering from malnutrition caused by a severe famine, itself caused by political
conflict in her home country. Her suffering is extremely severe, as is that of her parents, who, in
addition to their own physical suffering, are also suffering emotionally because of their reasonable
fears that their daughter will die. A group of kind people who work for a humanitarian organization
locate the family and bring them the water, food, and medical supplies needed to nurse them back
to health. The group also provides them with transportation to a refugee camp in a neighboring
country and help the family emigrate out of the conflict zone.
467
Ethics: Atheism
We can make the following moral claims about this sce-

KEY CONCEPTS nario: The child’s suffering and that of her parents is bad. It
is good that the child’s suffering has been relieved. It is good
that her parents’ anguish has been relieved. The aid workers
Arbitrariness Problem
who helped the family acted with virtue. They had strong rea-
Contingency Problem sons to provide such assistance. Those responsible for the
Divine Command Theory of Moral conflict ought to do everything they can to end the conflict.
Obligation People who are in a position to help ought to provide other
famine victims with lifesaving assistance.
Euthyphro Problem
Consider now the claim that moral facts depend for their
Metaethical Divine Command Theory existence on the existence of God. Is this claim true?
(MDCT) The first paragraph of this section lists only natural facts
Normative Divine Command Theory about the child, her family, and their predicament. The sec-
(NDCT) ond paragraph lists moral facts about them. It is possible
that God does not exist but that all the natural facts in the
Problem of Counterintuitive Possibilities
first paragraph are true. If moral facts depend on the exis-
Supervenience tence of God, then, if God does not exist, the natural
facts listed in the first paragraph are true but the moral facts
in the second paragraph are false. An atheist moral realist
will find this claim to be implausible. For such an atheist,
the natural facts in the first paragraph are sufficient to make the moral facts in the second para-
graph true.
All moral realists agree that the moral facts are dependent on the natural facts in the follow-
ing sense: if there are changes in the moral facts, there must be corresponding changes in
the natural facts. If the child’s situation is morally better, then her suffering must have
been relieved. If there is no longer an obligation to render aid, then the natural facts must have
changed—for example, the family is no longer suffering from malnutrition. Furthermore, the
right kind of changes in the natural facts will result in changes to the moral facts. If the
child is not suffering from malnutrition, then her situation is better. Philosophers call this
kind of dependence relation supervenience. An atheist moral realist will say that this superveni-
ence can be explained on the premise that the natural facts are sufficient to make the moral
facts true.
It is natural to ask such an atheist moral realist such questions as, “Why does the suffering of
the child make her situation bad? In virtue of what is the child’s suffering bad?” The best response
to such a question is, I believe, to say that suffering is a kind of mental state whose nature gives us
reasons to avoid it (see Parfit 2011). That is what it is for something to be morally bad. We might
also want to ask, “In virtue of what do governments have a moral obligation to work to end armed
conflicts?” An atheist could respond by saying that the fact that armed conflict causes great suffering
is a moral reason to work to avoid them.
Many people are unsatisfied by the kind of answers I just provided. To such people it is
implausible that a natural fact could be sufficient to make it the case that any moral claim is true.
To some of them, only the existence of God could account for moral facts. Let’s consider this. Sup-
pose that we think that God does something in virtue of which the moral facts hold. We can ask,
“Why is it that what God does makes it the case that, for example, the child’s suffering is bad?” I
do not think that there is a response to this question that is any more satisfying than the response
that I gave above to the question, “Why does the suffering of the child make her situation bad?”
These questions are exactly parallel, and, for someone who is inclined to be skeptical about moral
facts, there is probably no answer that will satisfy. The fact that the atheist cannot answer such ques-
tions in a manner that satisfies the moral skeptic is no more reason to doubt the truth of atheistic
moral realism than the fact that the theist cannot satisfy the skeptic is a reason to doubt the truth
of theistic moral realism.
The atheist moral realist will believe the following two claims: First, that certain natural facts,
such as those stated in the first paragraph of this section, are sufficient to give us moral reasons.
Ethics: Atheism
Second, that there is nothing God could do, which would add to the natural facts, in virtue of
which and only in virtue of which we would have moral reasons. Some philosophers have argued
that moral properties are metaphysically and epistemically peculiar (see Mackie 1977, and Parfit
2011, esp. 464–510, for a response to such arguments); others have argued that, because of such
peculiarity, the existence of moral properties is evidence that favors theism over atheism (see Craig
2007). To show that moral realism does not give us reason to accept theism, we do not have to
establish the first of the above two claims; we need only show that the second is true. In this section,
I show that the attempt to ground moral facts in some activity, attitude, or judgment of God inevi-
tably fails because no activity of God could account for the existence of moral reasons. We can see
this with the help of a proper understanding of the Euthyphro problem.
THE EUTHYPHRO PROBLEM

Does God command morally right actions because they are right, or are they right because God
commands them? Divine command theories of moral obligation are committed to the second
option. According to metaethical divine command theory (MDCT), a morally right action is right
in virtue of the fact that God commands that we do it, and a morally wrong action is wrong in vir-
tue of the fact that God commands that we not do it. But this view seems to have the following
problematic consequences:
1. Morality is arbitrary: Since moral properties depend only on God’s will, God’s commands are
logically prior to moral properties. Thus God can have no reasons for his commands. It follows
that morality itself is arbitrary, that is, there is no particular reason why a given action is morally
wrong, morally right, or morally neutral.
2. Morality is contingent: Since God is all powerful, it is possible for God to issue any command
whatsoever. In the actual world, we’ll assume, God has issued one set of commands. However,
there are possible worlds where God has issued alternative sets of commands. Therefore, for any
moral claim that is true in the actual world, there is some possible world in which the claim is
not true.
3. There are counterintuitive possibilities: It is possible for God to issue commands with any
content whatsoever. We can distinguish four distinct possibilities:
i. God could command something cruel, for example, the torture of infants.
ii. God could command that we not engage in kind acts, for example, helping the poor.
iii. God could command that we engage in acts that are, intuitively, morally neutral, for
example, flossing our teeth every day.
iv. God could command that we not engage in acts that are, intuitively, morally neutral, for
example, taking walks in the evening.
What is sometimes referred to as the Euthyphro problem (so called because it was first raised
in Plato’s Euthyphro [c. 380 BCE], a dialogue on the nature of piety) is actually three problems:
(1) the arbitrariness problem, (2) the contingency problem, and (3) the problem of counterintui-
tive possibilities. Since the late 1970s and early 1980s there has developed a standard response to
the Euthyphro problem (see, e.g., Wierenga 1983; Adams 1979; Milliken 2009), which I will call
the appeal to love (ATL). ATL reminds us that God is essentially omnibenevolent and that his
commands are constrained by his nature. It is not consistent with the character of an omnibenevo-
lent being to issue horrible commands. Furthermore, his character is the same in every possible
world. Since his commands are grounded in and constrained by his character, he will issue the
same (or at least very nearly the same) set of commands in every possible world. Most important,
there is no possible world in which God commands something horrible or commands that we not
do something kind. It seems, therefore, that ATL eliminates the most problematic of the above
consequences. God’s commands are grounded in his essential nature; thus, he will not issue horri-
ble commands, nor are there possible worlds in which the content of morality differs dramatically
from the actual world.

Ethics: Atheism
Whether this response adequately addresses all three of the above problems is not my concern
here (see Thibodeau 2017 for a discussion of this question). Since we are interested in normativity
and reasons, I am going to focus exclusively on the arbitrariness problem and show that ATL does
not resolve the problem.
ARBITRARINESS AND WHY IT IS A PROBLEM

The reason that arbitrariness is a problem is that it shows that MDCT undermines the connection
between moral facts and reasons. If God has no reason to command that we A, then there is no rea-
son why A is obligatory rather than wrong. A just happens to be morally right because God just
happened to command it; nothing more can be said about why A is right. So, if I ask, “Why should
I A?” the only answer possible is, “Because God commanded that you A.” But if God had no reason
to command that I A, then how can I have a reason to A? If God’s commands are not grounded in
normative reasons, then, in a fundamental sense, there is no normative reason for me to do what he
commands.
In this context, it is important to bear in mind the difference between metaethical divine com-
mand theory—or, as I have been calling it, MDCT—and normative divine command theory (see
Murphy 2002, 6–7, for a thorough discussion of this distinction). MDCT claims that all moral
obligations are grounded in God’s commands. The normative divine command theory (NDCT)
claims that if God commands that we A, we are obligated to A, but only because of the existence
of a supreme ethical principle according to which God’s commands are to be obeyed. NDCT does
not account for the source of this principle, but for the theory to work, it must be external to and
independent of God and his commands. Since MDCT is meant to be an account of all moral obli-
gations, it is not consistent with NDCT. In other words, if NDCT is true, then MDCT is false
because there is at least one moral obligation that is independent of God’s commands, namely the
obligation to obey God.
Let’s connect this to reasons. NDCT’s supreme principle amounts to the claim that we always
have reason to obey God. Thus, on NDCT, the reason-giving force of a divine command is rooted
in that supreme principle. Since the existence of such an independent principle is not consistent
with MDCT, MDCT cannot appeal to it to account for the reason-giving force of divine com-
mands. That force, on MDCT, must be internal to the commands themselves. But arbitrary com-
mands have no reason-giving force.
Does MDCT, modified by ATL, imply that God has reasons for his commands? The answer
depends on what we mean by reason. In one sense, offering a person’s reasons is offering an account
of the motives on which the person acted. In this sense a reason is part of an explanation of an
intentional action: offering a reason means appealing to the factor(s) on the basis of which the agent
acted. In a different sense, however, when we appeal to reasons we are not trying to provide an
explanation of the person’s actions but rather account for how, whether, and/or to what extent
the action is justified. To avoid this ambiguity, philosophers typically speak of motivational reasons
(or simply motives) when talking about factors on the basis of which the agent acts, and normative
reasons (or simply reasons) when talking about factors that tend to justify actions.
The arbitrariness problem should not be understood as the concern that, on MDCT, God’s
commands would be unmotivated. Rather, the problem is that they would not be grounded in nor-
mative reasons. To see this, consider the possibility that there exists a deity that is very much like
the God of the monotheistic faiths, except that, whereas God is perfectly benevolent, this other
deity, whom we will call Asura, is perfectly malevolent. Asura is motivated by hatred and his com-
mands flow from his hateful nature; he commands horrible things, including that we engage in the
torture of children. Of course, Asura’s commands are not unmotivated: he commands that we tor-
ture because he is filled with hatred. But his commands are nonetheless arbitrary in the sense that
they stem merely from his own personal motives; they are not grounded in normative reasons.
The affective character of Asura’s motivational profile, by itself, does not provide reasons for his
commands. The same is true, by parity of reasoning, of God. Of course, it is true that the motives
that a loving being acts on will, unlike Asura’s motives, frequently also be normative reasons. But
this is not the same as the claim that having a loving nature gives one normative reasons. This
Ethics: Atheism
argument shows that for God to have normative reasons for his commands, those reasons must be
grounded in factors that are external to God’s motivational profile.
A normative reason is a factor that counts in favor of an action (see Dancy 2006). (It is worth
noting that, by counting in favor of an action, a consideration tends to justify the action. However,
it would be wrong to think that the mere presence of a factor that favors some action justifies that
action. In many cases there will be factors that favor engaging in some action and other factors that
favor not engaging in that action or favor some alternative action.) This is not the place to offer a
complete account of what reasons are and what they do. One point, though, is important for the
present discussion. For a factor to favor one person’s action, it must be the case that any other per-
son, who is in relevantly similar circumstances, would be equally justified to act on its basis. Let’s
call this the universal character of reasons. This universal character has an important implication
for any discussion of God’s reasons. If there are factors that count in favor of God’s commanding
that humans engage in some action, the universal character of reasons implies that these same fac-
tors count in favor of our engaging in the action God has commanded that we engage in. If so, they
would count in favor of our so acting even if God had not commanded the action.
So, if there are normative reasons for God’s commands, there must be factors that count in
favor of God’s commands, and these factors must be external to God’s motivational profile. It seems
fair to suggest that a factor that counts in favor of a divine command can only be some feature of
the action that is the object of the command. In other words, if God has a normative reason to
command, for example, that we help the poor, there must be some fact (or facts) about helping
the poor that counts in favor of God’s commanding that we do it. Similarly, there must be some
fact about, for example, killing that counts in favor of God’s commanding that we not kill.
This is perhaps what defenders of MDCT who appeal to God’s love have in mind. Given that
God is loving, he will command that we do good things and not do bad things. It is the goodness or
badness of an action that provides God with normative reasons (see Milliken 2009 for a defense of
precisely this view). But if that is the correct analysis, then it is equivalent to the view that the good-
ness of, for example, the act of telling the truth gives God reasons to command that we tell the
truth. And the badness of killing provides God with reasons to command that we not kill. The
problem for MDCT is that, even in the absence of God’s commands, these actions still have these
features. And if these features provide God with reasons to command, then, given the universal
character of reasons, they provide us with reasons to do what God commands. Since these factors
are independent of God’s commands, they would be reasons even if God issued no commands. In
other words, these factors, which count in favor of God’s commands, would count in favor of our
engaging in the actions God commands that we engage in (and refrain from engaging in the actions
that God commands that we not engage in) even if God does not exist.
If God has reasons for his commands, then it is these reasons, rather than God’s commands,
that are moral reasons for us. If God does not have reasons for his commands, then his commands
are arbitrary. MDCT claims that an action is morally wrong just in virtue of the fact that God com-
mands that we not do it. Given the nature of reasons and the universality of reasons, this can only
mean that, on MDCT, God does not have reasons for his commands. But, given the connection
between morality and reasons, arbitrary commands cannot ground moral obligations.
THE IRRELEVANCE OF GOD

Given all of this, we can now say, with some confidence, that the existence of God is almost cer-
tainly irrelevant to the existence of moral facts and properties. If God exists, then God has beliefs
and desires about, attitudes toward, and judgments concerning the moral value of human actions.
Since morality is normative, true moral judgments and beliefs, as well as appropriate moral attitudes
and desires, are not arbitrary. Thus God’s judgments, attitudes, and so on must be grounded in rea-
sons. Whatever reasons God has that ground his moral judgments, attitudes, and so on will, because
of the universal character of reasons, also be reasons for us. If God does not exist, those reasons still
exist.
Consider an act of deliberately torturing and killing a small child. Now abstract away any and
all facts that have to do with God, including God’s attitudes toward and judgments about those

Ethics: Atheism
facts. The natural properties that will not be abstracted away include the fact that it is an act of
deliberate cruelty, that the child suffers severe physical and emotional pain, and that the people
who love this child also suffer intensely. An atheist moral realist will think that these natural facts
are sufficient to give us overriding reasons to not engage in such actions. True, if God exists and
is loving, then these factors will motivate him to command that we not torture children, and these
same factors will count in favor of (i.e., be reasons for) his command. But, even if God does not
exist, these same factors still exist. If they are reasons for God to command that we not torture,
then, given the universal character of reasons, they are reasons for us to not torture. The reason-giv-
ing force of these factors exists completely independently of any divine command.
On any view according to which the existence of God is necessary for the existence of moral
properties, the natural properties of an act of torturing a child are not sufficient to make it the case
that there are moral reasons to refrain from such actions. This claim is highly implausible. If the fact
that an act deliberately causes the undue, unnecessary suffering and death of an innocent child does
not give us moral reasons, then it is not clear what the existence of God could do to change this.
The challenge for any defender of the claim that God is necessary for the existence of objective moral
properties is to explain (a) why natural properties, such as the fact that an act involves the deliberate
causing of intense physical and emotional suffering, are not sufficient to ground moral reasons, and
(b) what God could do in virtue of which the action, which in the absence of such divine activity
lacks all moral properties, would be morally wrong. In the absence of good explanations, it is more
plausible to suppose that the natural features of actions are sufficient to ground their moral properties.
ACCOUNTING FOR SIMILARITIES AND DIFFERENCES IN MORAL BELIEF
Thaddeus Metz, University of Johannesburg
This section addresses one specific facet of the debate between theists and atheists with respect to
morality. Both philosophers who believe in God and those who do not sometimes argue in support
of their views by contending that they do a better job than their rivals of accounting for certain fea-
tures of our moral lives. For instance, the previous section supposes, for the sake of argument, that
there are moral facts and considers whether theism or atheism best explains them. The present sec-
tion considers whether theism or atheism best explains relatively uncontroversial features of people’s
moral beliefs.
Think for a moment about the differences between moral facts and moral beliefs. Beliefs are
features of human psychology. They are states of mind, specifically, ones that are capable of being
true or false. While beliefs can be appraised as either true or false, other mental states, such as feel-
ings, desires, and moods, cannot. It does not make any sense to ascribe truth or falsity to, say, the
feeling of pleasure, the desire for an apple, or an irritable mood (although one can sometimes
appraise them as appropriate or not). In contrast, beliefs are what can represent the world accurately
or inaccurately; they are judgments.
Whereas beliefs are features of the mind, facts are features of the world. A fact is a state of
affairs at a certain time; it is a way that some part of the world is or has been (or, perhaps, will
be). Uncontested examples of facts include: nothing can be green all over and red all over at the
same time; it has not rained much in Cape Town this year; it is proper to drive on the right side
of the road in the United States; I like chocolate. Some facts are necessary, that is, could not be oth-
erwise, as in the first example, while others are contingent, as in the latter three. Some facts are
absolute, in the sense of not depending on humans for their reality, as per the first two examples,
while others are relative, that is, constituted by human choices, preferences, or perspectives, as in
the latter two.
For many philosophers, beliefs are about facts, such that when beliefs represent the facts accu-
rately the beliefs are true, and when they fail to do so they are false. If you believe the earth is flat,
your belief does not correspond to the facts and is thereby false. In contrast, if you believe the earth
is round(ish), your belief does correspond to the facts and is thereby true.
Ethics: Atheism
The previous section of this chapter was about moral facts, the idea that, say, injustice is a
property of the world about which our beliefs can be either true or false. In contrast, this section
is mainly about moral beliefs, the ways that human beings judge right and wrong actions, just
and unjust institutions, good and bad character. Considering the similarities and differences in peo-
ple’s moral beliefs, which best explains them, theism or atheism? If God existed, what would our
moral beliefs be like, and are our beliefs like that? Or are our beliefs what one would expect them
to be if there were no God?
This section proceeds by considering respects in which the content of what human persons
tend to believe about morality is similar, after which it considers respects in which it is different.
Then, the section addresses the extent to which theism can explain the similarities and differences
that exist among people’s moral beliefs and argues that there are serious problems with its ability
to do so, insofar as the existence of God would mean a uniformity in judgment that does not seem
to exist. Finally, this section sketches the ways in which atheism could explain the content of peo-
ple’s moral beliefs, and argues that, by comparison with theism, it does a reasonable job.
SIMILARITIES AND DIFFERENCES IN PEOPLE’S MORAL BELIEFS

In suggesting that people hold many similar judgments about right and wrong and related catego-
ries, the claim is not that literally all people in the universe, or even all human persons, hold these
same judgments. Consider an analogy. Not everyone can hear, and few who are deaf will judge a
certain piece of music to be beautiful. Those who can hear might not have heard enough music,
or enough of a certain kind of music, to be in a position to judge. Furthermore, some might not
have been paying close enough attention to a piece of music, and so failed to detect certain features
of it. Despite these kinds of exceptions, it could be fair to say that “people generally judge” a certain
piece of music to be beautiful. Similar remarks apply to moral beliefs. Those capable of understand-
ing right and wrong and who are paying close attention (on which see Audi 2014) tend to converge
on some of the same beliefs.
Specifically, here are some apparent similarities among people when it comes to moral beliefs at
the level of general principle. First, consider that virtually all of the world’s long-standing philoso-
phies and religions include some version of the Golden Rule (Küng and Kuschel 1993). Confucian-
ism, Christianity, Islam, Hinduism, and many other worldviews are explicit that one is to treat
others as one would want to be treated. Second, many worldviews also accept the idea that human
beings have a dignity that elevates them above the rest of the animal, vegetable, and mineral king-
doms (Christians 2010). The monotheist religions tend to say that human beings were created in
the image of God or are sacred, while other worldviews without a deity, such as Confucianism,
maintain that human beings have a higher moral standing and merit greater consideration than
other beings.
There is also substantial agreement at the level of particular cases. If you were driving a bus and
had to choose between striking a cat and a human person (who is not threatening severe harm to
others), just about everyone would agree that you should aim for the cat. If there were a runaway
train that would mow down ten (innocent) people tied to the track unless you pulled a switch, in
which case it would kill only one person on a different track, a very large majority would agree that
it would be permissible to pull the switch. If there had been a shipwreck in which your father and a
stranger were stranded in the water, and you could reach only one in your life raft, you should save
your father (supposing he is not a wicked man). If two men were trying to hack an elderly woman
to death simply because she has a different ethnicity, and the only way to prevent the men from
doing so were to shoot them with your machine gun, you would not do wrong to shoot them.
Notice that these are all cases about life-and-death matters at the core of morality, that it is
extremely difficult to disagree with them sincerely, and that they could be multiplied.
And yet it is implausible to maintain that there are no differences in moral beliefs, and substan-
tial ones at that, even when it comes to central life-and-death matters. Is abortion in the first trimes-
ter, before the fetus is aware and can feel any pain, justified? Is that sort of abortion permissible just
because one would rather have a girl than a boy? Supposing that meat is not needed on grounds of
nutrition, is it wrong to eat meat merely because one likes the taste? Should people who have,

Ethics: Atheism
beyond a reasonable doubt, tortured and killed other people be put to death by the state? Is it
immoral to commit suicide? Would doctors be right to deliver a lethal dose of painkiller to a termi-
nally ill patient who is mentally competent and wants it? Controversy abounds.
One might suspect that one could resolve these particular disputes by appealing to putatively
shared general principles. However, the Golden Rule does not help when it comes to suicide, or
any situations in which the choice is between saving two different lives—regardless of whose life
is threatened, you yourself would want to be saved. In addition, the principle of treating human
beings as having dignity seems too vague to prevent disagreement. Does capital punishment respect
or disrespect human dignity? What about euthanasia?
In addition, there are other general principles that are frequently, but not universally, held that
conflict with the Golden Rule and the principle of respect. Utilitarians hold that one should maximize
happiness and minimize unhappiness, wherever and however one can. Many people in indigenous
African cultures think that one should do whatever would enhance life force, an imperceptible energy
that permeates everything (e.g., Dzobo 1992; Magesa 1997). Confucians maintain that you ought to
strive to foster harmony, particularly harmonious relationships in which less-qualified inferiors obey
more-qualified superiors who prescribe what they judge to be best for all (Chan 2014; Li 2014).
Some philosophers suggest that many moral differences are merely superficial and are really a
function of different metaphysical views or varying contexts. For example, some Hindu people deem
it a grave wrong to kill a cow, which seems to differ from what most Christians and Muslims
believe; but these Hindus think their relatives were reborn in a cow’s body, and so they share the
underlying principle of not killing relatives. For another example, some Inuit people practice infan-
ticide, and one might conclude from this that they do not value human life; but it turns out they
practice infanticide only when there is a serious scarcity of food, and so, again, they can be viewed
to share the principle of valuing human life (for both examples, see Rachels 2003, 23–26).
Yet it should be noted that the examples of disagreement about abortion and the death penalty
do not seem similar to these sorts of cases, concerning Hindu and Inuit cultures. There can still be
robust differences about abortion, the death penalty, and other topics when all the interlocutors
share the same metaphysical views and are focused on the same context.
DOES GOD’S PRESENCE EXPLAIN THE SIMILARITIES AND DIFFERENCES?

Which account of the nature of reality best explains the similarities and differences in moral beliefs
discussed in the preceding paragraphs? What must the world be like, when it comes to God’s exis-
tence, in order to find the sorts of moral judgments that we encounter across the world? The discus-
sion that follows argues that, while theism can easily account for the similarities, it has a difficult
time explaining the existence of widespread and deep differences.
Theism holds that there exists an all-powerful, all-knowing, and all-good being who created the
universe with certain goals in mind. God has intended human beings to help realize at least some of
those goals, which include moral purposes. With respect to these purposes, God has issued com-
mands to human beings to perform certain actions and not to perform certain others. On the most
influential versions of theism, God has issued the same commands to all human beings, which
might include to glorify God (the Qur ʾan) or to love one’s neighbor (the Bible).
God is usually thought to have conveyed these commands in one of two ways, and perhaps
both. On one hand, God has communicated directly with prophets who have, in turn, written
God’s word down in the form of long books (scripture), while, on the other, God has enabled
human beings to use their reason to figure out the difference between right and wrong on their
own (natural law). The latter approach can make sense of the intuition that even atheists who avoid
religious texts could learn about morality from God (on which see Austin 2017, sec. 7d).
Those who have done what God has instructed us to do are thought by many to be given
entrance into heaven, an eternal afterlife in His presence, whether because that is a deserved reward
(Islam) or a generous gift (Christianity). In contrast, those who have disobeyed God’s commands are
thought to be destined for hell, which, at the very least, means not living forever with God, and
might involve substantial suffering for an eternity.
Ethics: Atheism
If this account of how God interacts with us were true, one would expect a very high degree of
similarity in the content of our moral beliefs. God has the same moral purposes for us, and, being
omnipotent, is obviously able to inform us of them. Since it would clearly be unfair of God to hold
us accountable for failing to do what He wants if He did not make us all aware of His commands,
theism suggests that we should all share the same understanding of morality. However, as has been
shown in the various examples above, there are substantial differences in the content of our moral
beliefs.
How might the theist try to account for these differences? One strategy is to point out that the
key holy text was set down in one language, and that some of God’s commands have become “lost
in translation” as they have been conveyed in languages other than the original Hebrew, Greek, or
Arabic (depending on one’s religion).
Upon reflection, though, this explanation is a poor one. Knowing everything, God would have
foreseen that meanings would not get adequately conveyed in different languages and so would have
laid down His commandments in all human languages. Being omnipotent, if God created human
beings, He surely could compose books in all languages Himself in order to get His point across.
God would be foolish to depend on unreliable human beings to do the translating. Not doing
everything God could do to inform us of His commands would be both imprudent on His part,
insofar as He would be unlikely to advance His goals, and also unfair to us, insofar as we could
be damned for failure to do what He wants. However, God would be neither imprudent nor unfair,
and so He would have ensured that everyone had access to His word in their respective languages.
At this point, the theist is likely to invoke the other way that God might convey His will,
namely, by giving us the innate ability to apprehend it in the absence of a text. It would be fair
of God to threaten us all with hell for not having achieved His moral ends, if we all carried inside
us the ability to ascertain them. In addition, imparting such a capacity to us in the womb would
be a prudent way for God to teach us, too. Just as we all can use our reason to ascertain that the
shortest distance between two points on a flat surface is a straight line and that 2 + 2 = 4, so we
can reflect carefully on moral matters and come to the same judgments of what God wants us to
do and not to do. By this account, it is part of human nature to be able to figure out God’s com-
mands.
The problem with this natural law rationale is that it continues to predict too much uniformity
in people’s moral beliefs. If ascertaining the content of God’s will were akin to doing basic geometry
and arithmetic, then we would all have more or less the same moral views. However, we do not.
Of course, some human beings cannot judge that 2 + 2 = 4. By the same token, one might sug-
gest that not all human beings should be expected to arrive at the same answers when it comes to
morality. Some individuals are “defective,” that is, cannot reason well, and God would not hold
them comparably accountable. However, it is only adults who are severely mentally incapacitated
who do not know that 2 + 2 = 4, whereas plenty of mentally capacitated adults disagree about what
the right course of action is when it comes to, say, abortion and the death penalty. A wide array of
intelligent, mentally healthy people can and do judge what is moral, yet they are not certain of their
judgments, and their judgments are hardly uniform.
Furthermore, their inability to know for sure, or to come to a consensus, is not because they are
prideful or uninterested. Some philosophers humbly devote their forty-year careers to trying to
determine what is right and what is wrong, but even they have not come to the same conclusions.
DOES GOD’S ABSENCE EXPLAIN THE SIMILARITIES AND DIFFERENCES?

Consider how we might have come to judge morality on an atheist worldview. Atheism holds that
there is no God who created us to achieve particular purposes, and we arose as a chance product
of natural selection. One atheist view that is consistent with this picture is that our deepest value
judgments developed because making them helped our species to survive and to reproduce in a par-
ticular environment.
More specifically, the capacity to make moral judgments, which at the core are about how to
treat other people evenhandedly, probably helped us to flourish. Those who lacked a disposition

Ethics: Atheism
to cooperate, and instead were inclined to steal, break promises, and cheat, would have been isolated
by others, making it hard to find a mate. In contrast, those who were less selfish would have had
greater ease in finding a mate and thereby passing on their genes. According to this picture, it would
have been useful for human beings to develop a language by which to name desirable and undesir-
able actions and attitudes. Humans would have started calling ways of behaving “moral” and
“immoral.”
Even though natural selection would have prompted human beings to call certain behaviors
moral and immoral, it does not follow that they would have a clear awareness of what precisely they
were labeling with these terms (Brink 1989; Miller 1992). Consider an analogy that elucidates this
point: It would have been useful for human beings to call certain conditions of the human body “ill-
ness” or “disease.” Picking out that feature of our bodies would have been particularly important
when it comes to survival, reproduction, and flourishing. It does not follow, though, that human
beings have known much about what disease or injury essentially is. Indeed, it is only recently that
we seem to have learned much at all about how the human body works and what interferes with
its proper functioning. We have known the “surface” properties of disease, such as seeing skin dis-
coloration, or of injury, such as broken bones, for a long while, but we have not similarly known
“deep” properties, such as how that discoloration is related to a malfunctioning liver or how bones
really work, and what is going on when there is a malfunction in either (see Putnam 1975 for the
surface/deep distinction).
Similar things probably go for what the word “immoral” picks out on an atheist worldview. We
readily recognize surface properties, as in the instance of torturing others for fun or using innocent
people for target practice, but we have found it hard to obtain a firm grasp of the deep properties of
immorality that account for these instances and an array of others. Instead, on this picture of human
development, we have to learn over time, with argumentation, what immoral behavior objectively
consists of, analogous to learning experimentally over time about the nature of disease or illness.
We cannot apprehend the moral facts, or at least not many of them, merely by reflecting on them
as individuals and with some degree of certainty (a priori), in the way we do facts of geometry and
arithmetic. Instead, we must learn about moral facts by drawing inferences in critical discussion with
lots of other inquirers and on the basis of probabilities (a posteriori), similar to the way that science
is done.
What does this atheist account of moral knowledge entail for the content of moral beliefs? It sug-
gests that there will be both similarities and differences. On one hand, one would expect some simi-
larities when it comes to the surface properties of wrongness and vice. Everyone should agree that cer-
tain actions are characteristically picked out with the word “immoral.” Torturing others for fun is
obviously wrong in the way that a broken arm is obviously an injury. And that is what we find—peo-
ple do tend to share some core ideas about what immorality is, as per the previous section.
Furthermore, at some point over time, one would expect more of a consensus to emerge, at
least among those who have studied morality carefully. This learning process would be similar to
the way that physicians have lately in the history of humanity begun to find some degree of consen-
sus about the nature of illness, and laypeople have come to share their views. Arguably, human
beings have recently come to learn that all persons have an equal moral status, that slavery for the
sake of profit is immoral, that it is arbitrary to assign educational opportunities on the basis of race
or gender, and that it is unjust to colonize other peoples (see Gilbert 1990).
On the other hand, the atheist picture of moral knowledge entails that there would be some
substantial disagreement about what is moral and why. In the history of humanity there have been
widespread differences about what is an illness and what makes it one, and, furthermore, not every-
one currently believes that the scientific method has indeed revealed something substantial about
those matters. By analogy, one would expect the history of humanity to reveal that there have been
widespread differences in belief about what is wrong and what makes it so, and, furthermore, not
everyone currently believes that systematic moral inquiry (philosophy) has indeed revealed some-
thing substantial about those matters.
On the atheist worldview, knowledge of morality would not be evenly distributed over genera-
tions of human beings or even among those of the same generation. Instead, some societies would
Ethics: Atheism
be more mistaken than others, and their ability to learn would depend on contingent matters such
as whether they have the resources to allow people to devote themselves to study and whether they
have let go of unjustified beliefs about God, spirits, and souls in a spiritual realm (cf. Parfit 1984,
sec. 154).
THE NATURE OF MORALITY AND OUR CAPACITY FOR

MORAL KNOWLEDGE
Bruce Russell, Wayne State University
How we can know, or at least be justified in believing, what morality requires, permits, and prohibits
depends on what the nature of morality is. I first consider what atheists might say if morality is based
on convention or is in some way subjective, depending on human sentiments or approval when
appropriately situated. I then consider what atheists might say if morality is objective and more like
mathematics than etiquette. I conclude that in many cases theists can agree with atheists, who are
moral objectivists, regarding how we can know what is morally required, permitted, and prohibited.
THE NATURE OF MORALITY

If there were no mind-independent moral facts, it would be relatively easy for atheists to explain
how humans could make, and be justified in making, moral judgments. For instance, if morality
were like etiquette in being based on accepted conventions, then a person could learn to make,
and be justified in making, moral judgments by learning the relevant conventions, say, to keep a
promise unless great harm will result if you do. But morality seems different from etiquette. Break-
ing a promise, cheating, and stealing, not to mention rape and murder, are different from eating
peas with a spoon. They don’t seem to be merely violations of certain conventions that are adopted
to coordinate behavior (as happens with conventions about which side of the road to drive on) or to
relieve anxiety about how to behave in social settings (as is the case with etiquette).
Perhaps morality is not based on actual conventions but on the conventions people would
adopt to govern their interactions were they to engage in rational bargaining with one another
knowing their actual social positions, abilities and weaknesses, race and gender, and so on. Gilbert
Harman has argued that it is in the interest of both the rich and powerful and the poor and power-
less not to be harmed (1977, 110–112). However, the poor and the powerless will want a strong
principle of bringing aid since they will need aid more often than the rich and powerful, who will
want a weaker principle since “they would end up doing most of the helping and would receive little
in return” (1977, 111). The poor and the powerless might refuse to adhere to a principle of nonin-
jury and noninterference unless the rich and powerful agree to some principle of mutual aid. A ratio-
nal bargain might then be struck where all agree to a strong principle of noninjury and noninterfer-
ence and a weaker principle of mutual aid. Harman thinks that our actual morality reflects just such
a distinction: people believe that duties not to harm are stricter than duties to help. If morality were
based on this sort of rational bargaining (Harman calls it implicit or tacit bargaining), we could fig-
ure out what morality requires by reasoning about what a rational bargain between people who
will interact with each other would be, given that they all have full knowledge of the situation they
are in.
But this account of morality also seems to allow for immoral rules or conventions. A strong reli-
gious or racist majority could bargain to restrict the days on which a religious minority could prac-
tice its religion, or bargain to ban interracial marriage in exchange, say, for civil unions for interracial
couples. A way around this objection would be to require that the bargainers reason from behind a
veil of ignorance, which prevents them from knowing any particular facts about themselves, say,
whether they are rich and powerful or poor and powerless, whether they are men or women, gay
or straight, a member of some majority or of a minority, of this or that ethnic or racial origin,
whether they have skills that are much in demand or lack them, and so on. Reasoning from behind
a veil of ignorance would ensure that the rules or conventions agreed to are impartial, which seems a

Ethics: Atheism
requirement of moral rules. Of course, this is the sort of approach to questions of social justice
adopted by John Rawls in his famous work A Theory of Justice (1971). Perhaps it could be expanded
to become the basis for a complete theory of morality.
Other approaches to morality that atheists might accept do not involve bargaining, whether
actual or hypothetical, or choice from behind a veil of ignorance. According to these views, what
is morally required is determined by what people would want or approve of when ideally situated.
So, on this account, what morality requires is determined by the rules they would want, or approve
of, to govern their interactions were they fully informed and “procedurally rational,” meaning that
they do not make mistakes in reasoning by accepting invalid arguments or thinking that something
counts as evidence for something else when it really does not. Some might add, as David Hume did,
that what these people want or approve of must rest on sentiments “common to all mankind,”
which for him meant sympathy. To avoid the problem with actual bargaining views of morality,
the sentiments and desires of the ideal agents should, in some sense, be impartial, not self-interested.
Otherwise, we may either end up with what are intuitively immoral rules or no rules at all because
none are in everyone’s self-interest.
The worry about this view, which we might call ideal subjectivism, is that it still allows what are
intuitively immoral rules to be approved or chosen. If only humans are ideal agents, they might
approve of destroying an alien civilization, if there are any, or allowing animals to be harmed and
to suffer needlessly. If ideal agents consist of all rational agents, they still might approve of rules that
permit current generations to use up resources that leave future generations very badly off and may
not solve the problem of animal suffering. If we required that future generations also be represented
as ideal subjects, we face the opposite problem: future generations might only approve of rules that
require the current generation to make huge sacrifices for their sakes.
Many atheists reject convention-, sentiment-, and desire-based views of morality, at least
partly because of the objections to them that have been raised. They think that morality is mind-
independent, that it is like science, mathematics, or logic. Both Peter Railton and Derek Parfit see
morality this way. They reject conceptual analysis as the right approach to determine what the fun-
damental moral principles are, which some think is the right approach to determine what the
fundamental natures of knowledge, persons, and personal identity are. Conceptual analysis aims to
give necessary and sufficient conditions of a concept that are meant to capture its meaning. Though
not of much philosophical interest, standard examples of such analyses include: (1) X is a bachelor if
and only if X is an unmarried male of marriageable status; (2) X is a square if and only if X is a four-
sided plane figure all of whose sides and angles are equal. Philosophers tried to analyze the concept
of knowledge and ran into the Gettier problem. Edmund Gettier (1963) showed that knowledge is
not justified true belief because there are examples of justified true belief that are not cases of knowl-
edge, in that a person is in some sense “lucky” to believe what is true. Attempts were made to give
an account of “lucky,” and none gained wide acceptance because a host of counterexamples to every
proposed solution were produced. Perhaps because of these sorts of failure, or perhaps because of
G. E. Moore’s open question argument (for more information, see Moore in his Principia Ethica
[1903]) and similar arguments to show that it is impossible to analyze normative concepts in terms
that are purely descriptive, most moral philosophers have given up trying to offer an analysis of
“right” or “wrong” (cf. Adams 1999, 15). Even if they think that it is necessarily true that: (3) An
action is morally obligatory if and only if it produces more good, on balance, than any other alter-
native open to the agent (or less bad, on balance, if none produces good on balance), they do not
think that (3) gives the meaning of “morally obligatory.”
Railton has been influenced by Saul Kripke, whose work allows us to distinguish between con-
cepts, the analysis of concepts, and properties. The concept ,water. is a different concept from
the concept ,H2O. even though those concepts refer to the same property, namely, being com-
posed of two hydrogen atoms and one oxygen atom. And it is necessarily true that water is H2O
even though this can only be known empirically, not a priori. Similar things are true with respect
to the concept ,heat. and the concept ,average molecular kinetic energy.: necessarily, heat is
average molecular kinetic energy, but this could only be discovered empirically, not a priori, and
this necessary truth does not give the meaning of “heat.”
Ethics: Atheism
Railton’s approach is first to figure out what the job description of, say, ,wrong. is, which
some call the conceptual role, reference fixing description, or complex role of ,wrong.. He says
the following about the job description for ,heat.:
Such a “job description” need not be a conceptual analysis properly so called, since it involves a
great deal of substantive empirical content and since any of its particular elements—detailing
conceptual interrelations, causal or constitutive connections, paradigm cases, etc.—might in
turn be revised or abandoned without leading us to abandon the phenomenological concept
,heat. altogether. (Railton 2017a, 49; Railton’s italics)
I assume that Railton would say something similar about the job description for ,morally wrong..
It would “not be a conceptual analysis properly so called,” and would include empirical content and
even what seem to be necessary moral truths such as: it is wrong to torture children to death for the
fun of it, or cruelty for its own sake is wrong. It can contain statements about the relation of obli-
gation to motivation and what it means to say “ought implies can.” The job description might also
contain what might be called statements of the pragmatics of ,wrong., that is, claims about what
people typically do when they say, or think, that an action is wrong. Of course, if the point of for-
mulating the job description is to help us find what the essence of wrong is, as it is in the case of
water and heat, the worry is that by including empirical and pragmatic elements in the job descrip-
tion we will be led astray.
The second step in the method Railton proposes is to find what property uniquely, or best, fits
the job description. Railton says that he accepts what he calls “methodological naturalism,” which is
not a metaphysical position but a view about how to acquire justified beliefs. Railton says that
methodological naturalism is “the view that, in inquiry, we should try to proceed by following the
ways of observing, hypothesis-forming, testing, formal modeling, axiomatizing, and explaining that
have been so successful in the natural, biological, social, and mathematical sciences” (2017a, 45).
In this passage, Railton seems to be endorsing an empirical approach to discovering necessary truths
about what makes actions right and wrong, though perhaps not entirely empirical given his inclu-
sion of the “mathematical sciences” at the end of his characterization of the method.
Parfit is a nonnaturalist about normative (including moral) properties because he thinks that
they are not reducible to natural properties. Natural properties are properties that, in principle,
can be discovered by empirical means. He believes that there are normative truths; these truths
“are not natural, empirically discoverable facts, since we could not have empirical evidence either
for or against our beliefs in these truths” (2017, Vol. 3, 3; cf. Vol. 3, 58; see also Adams 1999,
81, 365). Even if utilitarianism were true, it seems impossible to know this through empirical evi-
dence. We might learn through empirical investigation that most people are utilitarians, or that they
become utilitarians once they have reflected on what morality requires, but that empirical evidence
would not be evidence that utilitarianism is true, only that people accept it.
T. M. Scanlon offers an a priori approach to discovering what morality requires and prohibits.
He defends what he calls a domain-specific account of standards that should be used to discover
truths in various domains. Science has its standards, which include induction and inference to the
best explanation, to discover truths about the world. Mathematics has its standards, which include
deduction from axioms to discover mathematical truths. Scanlon thinks that the method for discov-
ering normative, including moral, truths is reflective equilibrium.
This method is founded on what, following Rawls, Scanlon calls considered moral judgments,
by which he means judgments that are founded on what seems obviously true to someone when that
person is “in the right conditions” (2014, 82–83, 85–86). He says that a judgment is a considered
judgment only if the proposition that is its object seems clearly true after the person has been unable
to discover any “implausible implications or presuppositions” of that proposition (2014, 84). Those
judgments may be about cases (say, that it is wrong to take the vital organs from one healthy person
without his consent to transplant into five people to save their lives) or about principles, say, that
you ought to do what produces the most good on balance. Reflective equilibrium is the process
by which you reconcile any inconsistencies between considered moral judgments at any level of gen-
erality. This can be done by either giving up considered judgments about cases or by giving up or
modifying principles that are the objects of considered judgments. In the end, you look for a

Ethics: Atheism
coherent system of judgments where the principles explain the considered judgments on cases that
survive the search for consistency. A utilitarian principle could explain why in triage situations
you should do what saves the most lives, and a principle prohibiting using people as mere means
could explain why we should not take the organs from one to save five. So a principle consistent
with considered judgments in transplant, triage, and other examples might be: Do what produces
the most good on balance unless that requires using someone as a mere means; in that case, do
not use the person as a mere means unless that is the only way to prevent harm or death to a large
number of people.
Robert Adams accepts a divine command theory of moral obligation. He thinks that wrongness is
the property of being contrary to the revealed commands of a loving God (1979, 76). Adams thinks
that it is necessarily true that we are obligated to do what a loving God commands and that it is
wrong to do what is contrary to his commands (1979, 67; 1999, chap. 11). He accepts this view
on the basis of a semantic view like Railton’s. Adams even notes his agreement with Railton at
the semantic level (1999, 27). He says that wrongness “will be the property of actions (if there is
one) that best fills the role assigned to wrongness by the concept” (1979, 74; 1999, 15–18). By
“the concept” Adams means what Railton means by the “job description.” However, Adams differs
from Railton at the metaphysical level in that he is a supernaturalist while Railton is a naturalist (cf.
Adams 1999, 27). They differ on what best fits the relevant job description or conceptual role of
wrong.
MORAL EPISTEMOLOGY
To settle the metaphysical question about the nature of morality we would have to determine what
the appropriate job description is and then look carefully to see if Adams’s divine command theory,
another nonnaturalistic theory like Parfit’s, or some naturalistic theory like Railton’s, best fits that
job description. But those semantic and metaphysical issues are not the concern of this discussion.
What is partly of concern is whether a theistic ethics must adopt a different epistemology from an
atheistic ethics. Adams argues that it need not. According to him, what is right and wrong is deter-
mined by God’s commands. But what justifies us in thinking that God commands this rather than
that? Adams finds it unimaginable that God would command Abraham to kill his son Isaac, so not
even what is written in holy books can be uncritically accepted (1999, 290). Adams adds that a
command not to commit murder requires interpretation (1999, 273). Is turning off a respirator
murder (1999, 273)? How about abortion? Suppose we know that God commands us to love
him with all our heart, and to love our neighbors as ourselves. Still, what behavior is required by this
command? How do we know what that behavior is?
Even if we adopt a conceptual role semantics and a theistic metaphysics of morals, for a com-
plete metaethics we need some epistemology of value (as Adams calls it) that tells us what justifies
us in believing, or provides knowledge, that God commands this and not that. Following William
Alston, Adams adopts a doxastic practice epistemology (1999, 357). By a “doxastic practice,” Adams
means “a suitably integrated set of such skills and habits [highly developed, socially acquired, and
which we have compelling reason to trust], a system of forming and holding beliefs, learned in a
social context” (1999, 357).
Evaluative doxastic practices involve inputs that include feelings, emotions, inclinations, desires,
intuitions, and also evaluative beliefs, and they have evaluative beliefs or judgments as outputs.
What the outputs are is influenced by background beliefs and reasons that can affect the weighting
of the various inputs. Adams thinks that the most fundamental doxastic practices were taught to us
in childhood by our parents and others (1999, 354, 360). And he thinks people who have grown up
in different doxastic practices can have different beliefs as a result. (See Adams 1999, 356–366, for
his general discussion of doxastic practices.)
Atheists, as well as theists, can engage in evaluative doxastic practices to reach evaluative conclu-
sions. Adams says that his epistemology “does not require us to know anything about God, as such,
before we can have knowledge, or adequately grounded belief, in ethics” (1999, 355). He thinks
what he calls “general revelation” from God can be accessed through evaluative doxastic practices
and says, “I cannot regard the recognition of moral truths as an exclusive possession of theists, let
Ethics: Atheism
alone a particular religious tradition.… The generality of general revelation means that it can be
received by adherents of any religious tradition or none at all” (1999, 364).
Adams compares moral knowledge of what actions are right and wrong possessed by people who
are ignorant of the metaphysical nature of right and wrong to knowledge of the ordinary properties of
water possessed by people who are ignorant of the metaphysical nature of water, namely its being
identical to H2O (1999, 355). And he points out that a person could know a lot of mathematical
truths even if he was ignorant of the metaphysical status of numbers (1999, 355–356). Adams goes
even further in saying beliefs about what is right and wrong precede knowledge of the metaphysical
nature of right and wrong because these beliefs enter into the determination of the conceptual role
(the job description) of right and wrong (1999, 356). That conceptual role fixes the reference of
these moral terms and the metaphysical nature of right and wrong is what best fits that role.
People can disagree about what the correct normative and moral epistemology is: for Railton it
is empirical, for Scanlon it is largely a priori and involves reflective equilibrium, and for Adams it
involves doxastic practices. While the metaphysical foundations of morality will be different for the-
ists and atheists, both can hold that the correct epistemology of value allows atheists and theists to
have “knowledge, or adequately grounded belief, in ethics” (Adams 1999, 355). A different meta-
physics of morals does not necessarily imply a different epistemology of morals, and there are several
different moral epistemologies that permit both theists and atheists to have moral knowledge of
what actions are right and wrong, and even of what a loving person, whether human or God, would
will.
MORAL CAPACITY
Even if atheists and theists agree that we can know what is required, permitted, and prohibited by
morality, either empirically or through reason, a puzzle remains about our having a capacity to
know the moral status of actions and whether people are virtuous or vicious. We can literally see
some boys pour gasoline on a cat and set it on fire (Harman 1977, 4), but we cannot see with
our eyes the wrongness of their action. We can see Mother Teresa relieve the suffering of starving,
injured, or ill people, but we cannot literally see her goodness. So if there are objective moral facts,
theists might seem to have the best answer: God implanted a conscience in everyone, which is a sort
of moral sense, and conscience allows all of us, whether theists or atheists, to grasp objective moral
principles that we can then apply to specific cases. Atheists can say the capacity to reason has an evo-
lutionary advantage for creatures who possess it and that reason can develop in a way that allows us
to grasp objective truths that are not always advantageous from the standpoint of reproduction and
survival. No doubt the ability to do simple arithmetic was advantageous to our ancestors. As Parfit
has said, if one of our ancestors saw three lions go into a cave and only two come out, it would be
advantageous for him if he could do subtraction and thereby conclude that one still remains in the
cave. Starting from these humble practical beginnings, humans have been able through their reason
to develop sophisticated mathematical theories, including number theories (Parfit 2011, vol. 2, 496).
Similarly, it is sometimes advantageous to cooperate with others rather than going it alone. It
seems plausible to think that people with a disposition to cooperate with others in mutually advan-
tageous ways would be favored by evolution. And such a disposition to cooperate would seem to
require a disposition to keep promises, to tell the truth, not to cheat, and so on. Morality developed
from these humble beginnings, when people applied reason to draw further moral conclusions. Why
should we treat members of a rival tribe differently from ourselves when it would be mutually ben-
eficial to cooperate with them instead of always being at war? And why treat them differently even if
cooperation was not beneficial, at least if it was not harmful to us? Why should women be treated
differently than men? They have the same feelings and intellectual capacities that men have. Why
should we cause animals to suffer needlessly? They experience pain just as we do, and we could live
just as well without harming them. Reason, which initially provided an evolutionary advantage by
helping us to determine the best means for satisfying our ends of survival and reproduction, came
with a power of its own to address such questions. Having reason with powers beyond what is use-
ful for survival and reproduction is like having eyes that not only allow us to see predators, distant
forest fires, and approaching storms but also what is happening on the moon.

Ethics: Atheism
Theists might reply that it would be a weakness in evolutionary theory if it could not explain
why people have a capacity to see with their eyes what is happening on the moon, and why reason
has a capacity to answer moral and mathematical questions that have no evolutionary advantage.
But as long as having that capacity is not disadvantageous, evolutionary theory need not be able
to explain it. Evolutionary theory should be able to explain why gazelles can run faster than lions,
but it would be no mark against it if it could not explain why they are much faster than lions. No
doubt there would be some as yet unknown detailed genetic explanation of their ability to run
much faster that goes beyond any evolutionary explanation. It seems plausible to think that there
is some as yet unknown detailed genetic explanation of the brain’s capacity to grasp objective math-
ematical and moral truths. Even if Scanlon is wrong in thinking that moral truths just are the moral
propositions arrived at when reflective equilibrium is correctly applied, it does seem necessarily true
that we are justified in believing that something is a moral truth when reflective equilibrium is cor-
rectly applied.
Not only can atheists explain how we know, or are justified in believing, objective moral truths,
if it is not reasonable to believe that God exists, then it will not be reasonable to believe that God
created us with a conscience, or even a kind of reason, that allows us to grasp moral truths. If it
is not unreasonable to believe that God exists, it will be hard to explain why there is so much moral
disagreement if God created everyone with the same sort of moral conscience or why he did not cre-
ate everyone with that sort of conscience if he created different people with different moral con-
sciences. Perhaps theists will say that people are born with the same moral conscience, but they have
corrupted it through sin. But such an auxiliary hypothesis needs support if it is to save the view that
God created everyone with the same sort of moral conscience.
Atheists also have to account for moral disagreement. It can partly be explained by partiality:
people often favor themselves and those for whom they care, and this can skew their moral judg-
ments. At the end of Reasons and Persons, Parfit is optimistic about the prospects of moral agree-
ment, saying that moral reasoning, freed from the influence of religion, “is a very recent event”
(1984, 454). He sees religion itself as an obstacle that must be overcome for greater moral agree-
ment to emerge.
Suppose, for the sake of argument, it were a toss-up whether the atheist or theist has the best
explanation of moral disagreement, leaving questions about the existence of God aside. Still, I
believe that arguments against the existence of God, and in particular the argument from evil, would
break the tie. So, all things considered, we should accept the atheistic explanation of our capacity to
know objective moral truths. It is a plausible account that parallels an atheistic account of how we
have a capacity to know objective mathematical truths. Furthermore, it does not require positing
the existence of an immaterial being who somehow implants a moral sense or capacity in people.
CONCEPTIONS OF HUMAN RIGHTS
David Neil, University of Wollongong
So far this chapter has assessed theistic and nontheistic conceptions of ethics with regard to moral
realism, the diversity of moral belief, and the nature of moral judgment. Another approach to com-
parative evaluation is to focus on substantive normative claims that are widely accepted, and to ask
whether theistic or nontheistic approaches can provide the best explanation of those norms. For the
purpose of this discussion, the focus will be on human rights. Is the global ascent of human rights a
secular phenomenon, or does the very idea of human rights require a religious underpinning?
Charles Beitz makes an important distinction between “orthodox” and “practical” conceptions
of human rights. The orthodox conception “is the idea that human rights have an existence in the
moral order that is independent of their expression in international doctrine” (Beitz 2004, 196).
On this view, persons possess human rights simply in virtue of being human, and the instruments
of human rights law, such as international covenants, ultimately draw their authority from this
moral foundation. The practical conception, by contrast, “takes the doctrine and discourse of
Ethics: Atheism
human rights as we find them in international political practice as basic” (Beitz 2004, 197). Human
rights have functional roles in international discourse, and those functions determine both what
human rights are and what human rights there are. The practical conception is not committed to
denying any fundamental moral foundation for international human rights. Rather the practical
approach treats questions about the nature of human rights as separable from questions about the
moral justification of human rights. On this view, human rights advocates could conceivably agree
about the form and content of human rights while disagreeing as to how those rights are grounded
(Beitz 2004, 197).
The practical conception appears to align with the strategy adopted in drafting the 1948 Uni-
versal Declaration of Human Rights (UDHR), such that forty-eight of the then fifty-eight members
of the United Nations (UN) were able to ratify it (UN General Assembly 1948). The UDHR con-
tained no enforcement mechanism, and it was not legally binding. The decades since the UN Gen-
eral Assembly’s adoption of the UDHR have seen no shortage of crimes against humanity, and the
world still lacks effective institutions and procedures that can protect the vulnerable against state-
sponsored violence. Yet the UDHR is undoubtedly one of the most consequential documents of
the twentieth century. The prominent human rights lawyer Geoffrey Robertson describes the
UDHR as an “imperishable statement that has inspired more than 200 international treaties, con-
ventions and declarations, and the bills of rights found in almost every national constitution adopted
since the war” (Robertson 2013, 42).
Along with the Geneva Conventions, the UDHR was drafted in the climate of moral urgency
that followed the Nuremberg trials (1945–1946), in which former Nazi leaders were brought before
the International Military Tribunal on charges of war crimes. Its preamble asserts that “the barbarous
acts which have outraged the conscience of mankind” were the consequence of “disregard and con-
tempt for human rights” and declares “freedom of speech and belief and freedom from fear and want”
as the “highest aspiration of the common people.” Viewed in historical context, the UDHR was not
an assertion of the universal validity of Western values. Rather, in the words of Michael Ignatieff, it
was “a war-weary generation’s reflection on European nihilism and its consequences” (Ignatieff
2001, 4). Ignatieff describes the UDHR as representing a return to the natural law heritage of Euro-
pean law, intended to promote the agency of citizens to stand up to the orders of an unjust state. The
Nuremberg trials were a pivotal moment in international law because they brought the first indict-
ments of political leaders for crimes against humanity. By asserting that a state’s treatment of its citi-
zens must manifest a basic respect for their human dignity, the trials directly challenged the Westpha-
lian principle of state sovereignty and the international law principle of sovereign immunity. It is
perhaps because the postwar understanding of human rights stresses their function of imposing moral
limits on sovereign power that human rights discourse proved so effective for a broad range of politi-
cal movements in the latter half of the twentieth century. From the 1950s the so-called human rights
revolution (that is, the global deployment of human rights discourse in anticolonial revolutions, labor
struggles, civil rights struggles, and other forms of political activism) transformed human rights into
the primary tool of argumentation for social justice movements (Iriye, Goedde, and Hitchcock 2012).
However, the success of human rights in international politics and law has been achieved by
defining human rights only at the level of practice and abstaining from theoretical commitments
as to their basis and justification. The UDHR was able to be ratified by communist and capitalist
states, by countries with state religions and countries with constitutional separation of church and
state, because the text maintained a strict neutrality with respect to both religion and ideology.
The Thomist philosopher Jacques Maritain, who was involved in the drafting of the UDHR, relates
an anecdote: at a meeting of the UNESCO National Commission where human rights were being
discussed, “someone expressed astonishment that certain champions of violently opposed ideologies
had agreed on a list of those rights. ‘Yes,’ they said, ‘we agree about the rights but on condition that
no one asks us why”’ (UNESCO 1949, 1). In his introduction to the UNESCO symposium on
human rights, Maritain further states:
Where it is a question of rational interpretation and justifications of speculation or theory, the
problem of Human Rights involves the whole structure of moral and metaphysical (or anti-
metaphysical) convictions held by each of us. So long as minds are not united in faith or

Ethics: Atheism
philosophy, there will be mutual conflicts between interpretations and justifications. In the
field of practical conclusions, on the other hand, agreement on a joint declaration is possible,
given an approach pragmatic rather than theoretical and cooperation in the comparison,
recasting and fixing of formulae, to make them acceptable to both parties as points of
convergence in practice, however opposed the theoretic viewpoints. (UNESCO 1949, 3)
This conviction about the possibility and value of “convergence in practice” lies at the core of John
Rawls’s approach to human rights in The Law of Peoples. In pluralist societies, neither religious, philo-
sophical, nor moral unity is possible. Therefore, social stability “must be rooted in a reasonable political
conception of right and justice affirmed by an overlapping consensus of comprehensive doctrines”
(Rawls 1999, 16). For Rawls, human rights belong to a “reasonable political conception of right” pre-
cisely insofar as they do not presuppose any particular theistic, metaphysical, or metaethical premises.
These rights do not depend on any particular comprehensive religious doctrine or
philosophical doctrine of human nature. The Law of Peoples does not say, for example,
that human beings are moral persons and have equal worth in the eyes of God; or that they
have certain moral and intellectual powers that entitle them to these rights.… Still, the Law of
Peoples does not deny these doctrines. (Rawls 1999, 68)
One important reason why human rights discourse has been able to acquire the political and legal
authority that it has is that the human rights project has been conceived as a project of formalizing
what Rawls called an “overlapping consensus.” At the level of practice, there is broad agreement
between mainstream theistic and atheistic ethical systems on the importance of human rights as
goals and standards for moral progress in the world.
Yet, despite the pragmatism with which the foundational instruments of human rights protec-
tion were constructed, the UDHR and the key treaties that followed it cannot actually be classed as
“practical” conceptions of human rights in Beitz’s sense. This is because UN-authored human rights
documents allude to an objective moral foundation for human rights, in human dignity. In the
UDHR, a connection between human rights and dignity is implied but unspecified. Subsequently,
in both of the 1966 International Covenants on Civil and Political Rights we find the explicit dec-
laration that the human rights they enumerate are derived from human dignity (UN General Assem-
bly 1966). This claim that human rights derive from human dignity has since been reiterated in
other UN declarations (e.g., the 1993 Vienna Declaration), but all official references to human dig-
nity carefully avoid imputing any religious or metaphysical content to that term.
For any orthodox theory of human rights, human rights declarations, treaties, charters, and laws
do not create human rights. Human rights exist prior to and independently of the texts that aim to
define, enact, and protect them. The concept of a human right is a right that one has simply in vir-
tue of being human. Although the UN nowhere defines the meaning of “dignity,” the reference to
human dignity as the source of human rights can only be read as the claim that humanness carries
some innate value or moral worth, from which substantive, deontic norms can be derived.
The most basic theoretical problem for any orthodox theory is explaining why being a member
of the species Homo sapiens is both a necessary and sufficient condition for possessing human rights.
Here we find a crucial difference between theistic and nontheistic accounts. A number of theolo-
gians and religious philosophers have argued that human rights require a religious foundation and
that secular theories of human rights are untenable (see Perry 2006; Küng 1991; Stackhouse
2005; and the Organization of the Islamic Conference 1990). Any secular theory that endorses
the uniquely human character of human rights needs to explain the moral significance of human-
ness without appealing to notions of human sacredness or the sanctity of human life. Because a sur-
vey of the key figures on this terrain is well beyond the scope of this section, for the purpose of illus-
tration I briefly contrast just two approaches to this question, one theistic and one nontheistic.
A THEISTIC APPROACH
The Christian philosopher Nicholas Wolterstorff asks what the project of trying to ground human
rights in human dignity requires, arguing that no secular ground for human rights is possible (Wol-
terstorff 2008, 319, 323–341). That project demands an explanation of why all and only human
Ethics: Atheism
beings possess a dignity that makes them morally inviolate in just the ways specified by human
rights declarations. Given that any secular grounding of human rights cannot appeal to sacredness
to explain the inviolability of every human, such approaches must identify some capacity humans
possess, which is posited as the source and ground of human dignity.
Immanuel Kant provides perhaps the most influential capacity-based account of human dig-
nity. For Kant, personhood arises from the capacity for rationality, through which humans are able
to grasp the moral law and to make the moral law the governing motive of the will. Against Kant,
Wolterstorff contends that rationality cannot be the ground of human dignity because it is not pos-
sessed by all humans (it is lacking in babies and the cognitively impaired, for example). It is a capac-
ity that a human person can lose without thereby losing his or her moral worth. It is also likely that
some nonhuman animals possess reasoning abilities superior to some humans. Most important,
however, rationality is a spectrum capacity. Some humans have significantly greater rational capaci-
ties than others. Any analysis that reduces dignity to the capacity for reason has the consequence
that humans with greater rational powers also possess greater dignity, and hence greater moral
worth, than others. Wolterstorff claims that an objection of the same form can be extended to
any capacity-based account of human moral worth. For any capacity that might plausibly by identi-
fied with human dignity, not all humans will possess that capacity, even potentially. And given the
unequal distribution of every capacity across the human race, capacity approaches cannot explain
why dignity attaches to humanness and only to humanness in such a way as to confer dignity and
moral value equally on every human being.
In Wolterstorff’s view, a commitment to human rights requires that every member of the spe-
cies Homo sapiens is accorded a sacredness that other species do not possess. Human sacredness can
only be grounded theistically, and that ground is found in the biblical doctrine that humans bear
the image of God. The argument proceeds via a lengthy discussion about how the imago dei should
be interpreted, which cannot be summarized here. The crucial claim, however, is that (correctly
interpreted) the imago dei is an essential aspect of human nature, and that God loves “equally and
permanently” (Wolterstorff 2008, 352) every creature that bears the imago dei, thus bestowing all
humans with a special moral worth.
A NONTHEISTIC APPROACH
James Griffin, in his highly influential book On Human Rights (2008), takes on the task that Wolter-
storff declares impossible for secular ethics. Griffin traces the concept of human rights to medieval
theories of natural law, a law that was believed to be manifest in the form of innate, action-guiding,
moral dispositions implanted in humans by God. Natural rights, understood as entitlements a person
has, first appeared as a corollary of natural law. During the Enlightenment the theological content of
natural rights was stripped away, and natural rights came to be understood as accessible through rea-
son alone. The French Revolution’s 1789 Declaration of the Rights of Man formalizes both the sec-
ularization of natural rights and their rebranding as human rights. The language of human rights
largely fell out of favor in the nineteenth century, and was emphatically rehabilitated after World
War II. The contemporary understanding of a human right is continuous with the Enlightenment
notion of “a right that we have simply in virtue of being human” (Griffin 2008, 2). However, when
the theological content of natural right was abandoned, “nothing was put in its place” (Griffin 2008,
2): the concept of human rights we have inherited from the Enlightenment is indeterminate. A con-
sequence of this indeterminacy of sense is that arguments about what human rights there are, how
those rights should be understood, and how conflicts of rights should be adjudicated, are undecid-
able. Lacking clear criteria for the application of the concept of a human right, “we often have only
a tenuous, and sometimes a plainly inadequate, grasp on what is at issue” (Griffin 2008, 2).
Griffin’s project is to remedy the indeterminacy of human rights, articulating what they are,
how they are derived, and the role they have in ethical theory. Rather than attempt here to give a
précis of a book-length argument, I will focus on Griffin’s interpretation of dignity in the phrase
“the dignity of the human person” (Griffin 2008, 6). For Griffin, the notion of dignity that is rele-
vant for human rights is profoundly connected with human agency: “Human life is different from
the life of other animals. We human beings have a conception of ourselves and of our past and

Ethics: Atheism
future. We reflect and assess. We form pictures of what a good life would be.… And we try to real-
ize these pictures. This is what we mean by a distinctively human existence” (Griffin 2008, 32).
Griffin variously refers to this essentially human power as “personhood,” “autonomy,” and
“normative agency.” The claim is not that human rights are derived from normative agency. Rather,
the proposal is that defining the basic function of human rights as the protection of normative
agency makes the concept determinate in a way that best preserves its role in ethics. We can see
how the most basic and uncontroversial human rights—the rights to life, liberty, physical security,
freedom of speech, conscience, religion, and association—are all essential for the enjoyment of
meaningful normative agency.
Griffin insists that normative agency is not merely instrumentally valuable. It should be under-
stood as an end in itself. The effort of self-direction makes our lives more meaningful, indepen-
dently of whether our plans and our efforts are successful. Our dignity as human persons, then, con-
sists in our status as autonomous beings, a status that can be destroyed or compromised when our
human rights are violated.
Atheists and theists can agree, to a large extent, on what rights belong on the list of human
rights. However, religious and secular theorists follow very different paths in explaining the basis
of human rights, and each path has distinct advantages and disadvantages. Those who believe in a
loving God have a more straightforward explanation of how it could be that all human beings pos-
sess an innate moral worth to an equal degree. That special form of value is bestowed by God on all
humans and only on humans. Without recourse to the notion of sacredness, secular ethics faces the
challenge of identifying the human capacity or attributes that warrant the protection of human
rights, and that are sufficiently universal to explain the universality of human rights. Griffin’s
account of normative agency is one way. Of course, there are several other theories. Griffin is right,
though, that human rights discourse is vague. This indeterminacy of human rights makes it impos-
sible to give principled answers to practical problems about the scope and boundaries of specific
rights, about their application in hard cases, and about how to resolve conflicts between rights.
Some human rights aim to protect people from violence and oppression. Others aim to guarantee
some minimum level of opportunity to lead a worthwhile life. A theoretical specification of the
rights protections a person needs in order to construct a life that is properly her own must be
informed by a theory of personhood, of what persons need to flourish, and of the ways in which
personhood can be damaged or lost. If we accept that modern human rights doctrine is an outline
plan from which a lot of essential detail is missing, then we should also accept that the resources
of secular ethics are needed to supply the details.
An account of why humans have an innate moral value, which makes each human irreplace-
able, is not the same as an account of what makes a human life valuable to the person whose life
it is. These are distinct kinds of value. Crucially, the second kind of value can be gained or lost,
but not the first kind. For example, a person experiencing unbearable suffering in the final stages
of a terminal illness may rationally judge that life has lost all value and may wish to die. Yet that
person still possesses the first kind of value and rightly commands the respect that follows from
the fact that he or she is still a human being. Indeed, we extend that form of respect to the dead.
Our conception of human rights is answerable to both of these types of moral value. One way to
characterize the difference between theistic and nontheistic theories of human rights is that theistic
accounts tend to treat the first kind of value—the innate value of human life—as both independent
of, and morally more important than, the second kind of value—the value one places on one’s own
life as it is subjectively experienced. Secular accounts largely focus on the second kind of value—on
how life is valued from the inside—and tend to treat the first kind of value as, in some way, derived
from the second. As a generalization, that is, of course, also an oversimplification. This contrast is by
no means offered as an exceptionless taxonomic rule. However, I hope it is a useful way to bring into
focus one fundamental difference between religious and secular theories of human rights.
CHAPTER REVIEW
This chapter examines whether general facts about morality, and the human capacity to make moral
judgments, count in favor of theism or atheism.
Ethics: Atheism
The first section assumes moral realism and argues that God’s will cannot be the source of
moral reasons. Divine command theories face the Euthyphro problem: the dilemma of whether
God commands right actions because they are right, or whether right actions are right because
God commands them. In order to hold that moral reasons are born from God’s will, the theist must
believe that moral obligations do not arise from the natural facts alone. Thibodeau shows that a
divine command account of morality cannot meet this challenge. The price of insisting on a divine
command morality, in which God’s commands are not determined by objective moral properties, is
that such a view renders the connection between normative reasons and moral obligation unintelli-
gible.
The second section considers whether the similarities and differences in the moral codes of dif-
ferent peoples are more readily explained from a theistic or atheistic standpoint. Metz contends that,
if an omnipotent God communicated a single set of moral commands to all human beings, and
made entry to heaven conditional on observance of those commands, it is hard to explain why we
do not find far greater convergence in moral belief than there actually is. Atheism is consistent with
the view that morality developed because it enabled forms of social cooperation that contributed to
human survival and flourishing. If systems of morality evolved separately and in parallel, we would
expect some fundamental similarities, arising from our common biological needs and vulnerabil-
ities—and we would also expect many differences arising from the divergent developmental paths
of different peoples and their cultures. An account of morality as something that emerges and devel-
ops within the history of human development provides a more plausible explanation of the plurality
of moral systems.
The third section assesses whether atheists can hold that moral judgments are objective and
truth-assessable. A common charge against atheistic ethics is that the practice of moral judgment
appears to presuppose moral objectivism, and moral objectivism, the theist claims, requires a reli-
gious foundation. Atheistic ethics, here, refers to any metaethical theory in which God does not play
an explanatory role. Bruce Russell shows that even views that find ethics in convention, social con-
tract, or subjective moral dispositions can still make sense of practices of moral reasoning, and can
hold that there are right answers to moral questions. However, there are atheistic ethical theories
that comprehend morality as objective and mind-independent. These views assert that there are
moral truths than can be known, just as there are knowable scientific truths. So it is not true that
atheists must deny either rationality of moral judgment or the objectivity of moral facts. Russell also
points out that moral epistemology is substantially independent of the metaphysics of morals.
Despite their metaphysical disagreements, theists and atheists can share an epistemology that ade-
quately grounds their ethical beliefs, irrespective of the truth or falsity of their moral metaphysics.
The final section looks at the rapid embrace of human rights, after World War II, as moving
toward universal acceptance of the moral value of every human individual, and considers whether
the idea of human rights has religious or secular foundations. The political project of achieving inter-
national acceptance of human rights, and of creating enforcement mechanisms in international law,
has been successful because declarations and treaties have largely avoided any metaethical claims about
the foundation of human rights. Here also we find that theists and atheists, despite their metaphysical
differences, can find substantial agreement on the content of human rights. However, the idea of
human rights does appear to presuppose that membership in the human species carries an inherent
dignity that attaches to all human beings and only to human beings. Theists claim that humans can
possess an inherent dignity only if it is bestowed by a loving God. There are, however, compelling
atheistic accounts of human dignity that focus on what constitutes a distinctively human mode of exis-
tence. David Neil suggests that the dispute between theistic and atheistic accounts of human rights
derives from fundamental differences concerning the source of human values and the value of humans.
A commitment to human rights does not present a risk of self-contradiction for either theists or athe-
ists, but theists and atheists must construct their theories of human rights from different resources.
This chapter considers moral realism, our capacity for moral judgment, the diversity of systems
of moral belief, and the normative force of human rights. It argues that the justification of moral
practice, as we ordinarily understand it, does not require belief in God. Indeed, in some areas of eth-
ical theory, atheism has explanatory advantages over theism.

Ethics: Atheism
Bibliography
Adams, Robert Merrihew. “Divine Command Metaethics Modified Again.” Ignatieff, Michael. Human Rights as Politics and Idolatry. Edited by Amy
Journal of Religious Ethics 7, no. 1 (1979): 66–79. Gutmann. Princeton, NJ: Princeton University Press, 2001.
Adams, Robert Merrihew. Finite and Infinite Goods: A Framework for Ethics. Iriye, Akira, Petra Goedde, and William I. Hitchcock, eds. The Human
New York: Oxford University Press, 1999. Rights Revolution: An International History. New York: Oxford University
Press, 2012.
Audi, Robert. “Normative Disagreement as a Challenge to Moral Philosophy
and Philosophical Theology.” In Challenges to Moral and Religious Belief: Küng, Hans. Global Responsibility: In Search of a New World Ethic.
Disagreement and Evolution, edited by Michael Bergmann and Patrick Translated by John Bowden. New York: Crossroad, 1991.
Kain, 61–79. Oxford: Oxford University Press, 2014.
Küng, Hans, and Karl-Josef Kuschel, eds. A Global Ethic: The Declaration of
Austin, Michael W. “Divine Command Theory.” In Internet Encyclopedia of the Parliament of the World’s Religions. London: SCM Press; New York:
Philosophy. 2017. https://www.iep.utm.edu/divine-c/#SH7d. Continuum, 1993.
Beitz, Charles. “Human Rights and the Law of Peoples.” In The Ethics of Li, Chenyang. The Confucian Philosophy of Harmony. London: Routledge/
Assistance: Morality and the Distant Needy, edited by Deen K. Taylor and Francis, 2014.
Chatterjee, 193–214. Cambridge, UK: Cambridge University Press,
Mackie, J. L. Ethics: Inventing Right and Wrong. New York: Penguin, 1977.
2004.
Magesa, Laurenti. African Religion: The Moral Traditions of Abundant Life.
Brink, David O. Moral Realism and the Foundations of Ethics. Cambridge,
Maryknoll, NY: Orbis, 1997.
UK: Cambridge University Press, 1989.
Miller, Richard W. Moral Differences: Truth, Justice, and Conscience in a
Chan, Joseph. Confucian Perfectionism: A Political Philosophy for Modern
World of Conflict. Princeton, NJ: Princeton University Press, 1992.
Times. Princeton, NJ: Princeton University Press. 2014.
Milliken, John. “Euthyphro, the Good, and the Right.” Philosophia Christi
Christians, Clifford G. “The Ethics of Universal Being.” In Media Ethics
11, no. 1 (2009): 149–159.
beyond Borders: A Global Perspective, edited by Stephen J. A. Ward and
Herman Wasserman, 6–23. New York: Routledge, 2010. Moore, G. E. Principia Ethica. Cambridge, UK: Cambridge University Press,
1903.
Craig, William Lane. “Theistic Critiques of Atheism.” In The Cambridge
Companion to Atheism, edited by Michael Martin, 69–85. Cambridge, Murphy, Mark C. An Essay on Divine Authority. Ithaca, NY: Cornell
UK: Cambridge University Press, 2007. University Press, 2002.
Dancy, Jonathan. “What Do Reasons Do?” In Metaethics after Moore, edited Organization of the Islamic Conference. Cairo Declaration on Human
by Terry Horgan and Mark Timmons, 39–59. Oxford: Clarendon; New Rights in Islam. 5 August 1990. http://www.fmreview.org/sites/fmr/files
York: Oxford University Press, 2006. /FMRdownloads/en/FMRpdfs/Human-Rights/cairo.pdf.
Daniels, Norman. “Wide Reflective Equilibrium and Theory Acceptance in Parfit, Derek. Reasons and Persons. Oxford: Clarendon, 1984.
Ethics.” Journal of Philosophy 76, no. 5 (1979): 256−282.
Parfit, Derek. On What Matters. Vols. 1 and 2. Oxford: Oxford University
Daniels, Norman. “Reflective Equilibrium.” In Stanford Encyclopedia of Press, 2011. Referred to as OWM I and OWM II, respectively.
Philosophy, edited by Edward N. Zalta. Spring 2017. https://stanford.
Parfit, Derek. On What Matters. Vol. 3. Oxford: Oxford University Press,
plato.edu/archives/spr2018/entries/reflective-equilibrium/.
2017.
De Lazari-Radek, Katarzyna, and Peter Singer. “Parfit on Objectivity and
Perry, Michael J. Toward a Theory of Human Rights: Religion, Law, Courts.
‘The Profoundest Problem of Ethics.’” In Does Anything Really Matter?
Essays on Parfit on Objectivity, edited by Peter Singer, 278−296. Oxford:
Oxford University Press, 2017. Putnam, Hilary. “The Meaning of ‘Meaning.’” In Mind, Language, and
Reality. Cambridge, UK: Cambridge University Press, 1975.
Dzobo, N. K. 1992. “Values in a Changing Society: Man, Ancestors, and
God.” In Person and Community: Ghanaian Philosophical Studies. Vol. 1, Rachels, James. The Elements of Moral Philosophy. 4th ed. Boston: McGraw-
edited by Kwasi Wiredu and Kwame Gyekye, 223–240. Washington, Hill, 2003.
D.C.: Council for Research in Values and Philosophy.
Railton, Peter. “Two Sides of the Meta-ethical Mountain?” In Does Anything
Evans, C. Stephen. God and Moral Obligation. Oxford: Oxford University Really Matter? Essays on Parfit on Objectivity, edited by Peter Singer,
Press, 2013. 35–59. Oxford: Oxford University Press 2017a.
Gettier, Edmund L. “Is Justified True Belief Knowledge?” Analysis 23, no. 6 Railton, Peter. “Railton’s Commentary.” In On What Matters. Vol. 3, by
(1963): 121−123. Derek Parfit, 113–127. Oxford: Oxford University Press, 2017b.
Gilbert, Alan. Democratic Individuality. Cambridge, UK: Cambridge Rawls, John. A Theory of Justice. Cambridge, MA: Belknap Press of Harvard
University Press, 1990. University Press, 1971.
Griffin, James. On Human Rights. Oxford: Oxford University Press, Rawls, John. The Law of Peoples. Cambridge, MA: Harvard University Press,
2008. 1999.
Harman, Gilbert. The Nature of Morality. Oxford: Oxford University Press, Robertson, Geoffrey. Crimes against Humanity: The Struggle for Global
1977. Justice. 4th ed. New York: New Press, 2013.
Ethics: Atheism
Russell, Bruce. Review of Being Realistic about Reasons, T. M. Scanlon. Notre UNESCO. Human Rights: Comments and Interpretations. New York:
Dame Philosophical Reviews: An Electronic Journal, 07/36. 2014. https:// Columbia University Press, 1949. http://unesdoc.unesco.org/images
ndpr.nd.edu/news/being-realistic-about-reasons/. /0015/001550/155042eb.pdf.
Russell, Bruce. “A Defense of Moral Intuitionism.” In Does Anything Really United Nations General Assembly. Universal Declaration of Human Rights.
Matter? Essays on Parfit on Objectivity, edited by Peter Singer, 230–258. 1948. http://www.un.org/en/universal-declaration-human-rights/pdf.
Oxford: Oxford University Press, 2017. United Nations General Assembly. International Covenant on Civil and
Political Rights. 16 December 1966. Treaty Series, vol. 999. https://
Scanlon, T. M. Being Realistic about Reasons. Oxford: Oxford University treaties.un.org/doc/publication/unts/volume%20999/volume-999-i-14668
Press, 2014. -english.pdf.
Shafer-Landau, Russ. Moral Realism: A Defence. Oxford: Oxford University United Nations General Assembly. International Covenant on Economic,
Press, 2003. Social and Cultural Rights. 16 December 1966. Treaty Series, vol. 993.
https://treaties.un.org/doc/Publication/UNTS/Volume%20993/volume
Singer, Peter, ed. Does Anything Really Matter? Essays on Parfit on Objectivity. -993-I-14531-English.
Wierenga, Edward. “A Defensible Divine Command Theory.” Noûs 17, no. 3
Stackhouse, Max. “Why Human Rights Needs God: A Christian (1983): 387–407.
Perspective.” In Does Human Rights Need God?, edited by Elizabeth Wolterstorff, Nicholas. Justice: Rights and Wrongs. Princeton, NJ: Princeton
M. Bucar and Barbra Barnett, 25–40. Grand Rapids, MI: Eerdmans, University Press, 2008.
2005.
World Conference on Human Rights. Vienna Declaration and Program of
Thibodeau, Jason. “God’s Love Is Irrelevant to the Euthyphro Problem.” Action. 25 June 1993. http://www.ohchr.org/Documents/Professional
Sophia (2017). https://doi.org/10.1007/s11841-017-0615-8. Interest/vienna.pdf.

TOPIC 15
Meaning: Theism
Mirela Oliva
Associate Professor, Center for Thomistic Studies
University of St. Thomas, Houston, TX
This chapter discusses the ways in which theism provides answers to the question of the meaning of life.
Theists read this modern question as a quest for the ultimate meaning of our life that entails the
understanding of our existence in the great scheme of things created by God. Theistic positions fall under
two main categories, namely pluralism and singularism. This latter is in turn specified in four classes:
metaphysical, experiential, narrative, and subjective.
THE MEANING OF LIFE
“The meaning of life” is a modern expression that entails a wide spectrum of questions about the
significance of our existence. Why are we born? Why do we die? What can we do to make our life
meaningful? Is there a story of our life? The wide spectrum of such questions could make the
expression sound bombastic and even impossible to answer. The novelty of the expression “the
meaning of life,” however, is precisely its ability to gather up distinct aspects of life that are often
treated separately. The scholarship on the meaning of life has grown exponentially since the late
twentieth century, revealing the philosophical potential of this expression, not only in analytic phi-
losophy but also in continental thought.
Within this scholarship, many naturalistic answers try to bypass the cosmic question regarding
the why of our existence and our place in the universe. Some philosophers especially focus on pro-
viding a rationale for human action (Wolf 2010; Frankfurt 1998; Metz 2013). Others tackle the
narrative question to show how our life can be understood as a story (Velleman 2003). Theists give
priority to the cosmic question, whose answer informs the other questions. From a theistic perspec-
tive, the meaning of life is union with God. The theistic contribution to this field lies, prima facie,
in the formal completeness of its treatment, which covers all the distinct aspects embedded in the
use of “the meaning of life” in everyday language. Theism offers a substantive vision of the meaning
of life that is larger and deeper than the naturalistic answers.
Many theists admit that nontheistic theories can provide some criteria for life’s meaning that
intuitively correspond to the human quest for personal fulfillment, existential intelligibility, and
happiness. However, they point to the incompleteness of accounts that do not take into consider-
ation the full extent of the question. Naturalists most often replace the preposition “of” in the
expression “the meaning of life” with the preposition “in,” and eliminate the article “the” on the
ground that it raises unrealistic expectations of a grand, all-inclusive answer to the question about
the significance of human life (Metz 2013, 3). “Meaning in life” seems a better option for natural-
ism because it allows one to examine the meaningfulness of an individual life without assuming the
burden of discussing the meaning of the whole universe. By contrast, theism strives to honor the
original meaning and format of the question.
Theistic reflections on the meaning of life take a broad array of positions. This chapter is a sur-
vey of the main ones. It classifies them in two categories: pluralism and singularism (each of which
491
Meaning: Theism
is discussed in detail). Theistic pluralism (Mawson 2016) sees

KEY CONCEPTS the quest for the meaning of life as a collection of questions,
each of them yielding a different answer. Thus life has not
one but several meanings. This account makes some conces-
Surrogate Theory sions to naturalism, insofar as it accepts that some meanings
Theistic Pluralism (such as personal autonomy) are independent of God’s purpose.
Theistic Singularism At the same time, it maintains that a life committed to honor-
ing God’s purpose offers an ultimate and deep significance that
other visions of meaning for our lives do not reach.
Theistic singularism, embraced by the majority of theists,
formulates the question of the meaning of life as a singular request that gathers all aspects of life’s
meaningfulness into a unique line of inquiry. It offers a unified account that grounds every aspect
of our life in God’s creation and purpose. There are various subsets of singularism, which we will
classify according to the path that they propose for finding the meaning of life: (1) metaphysical,
(2) narrative, (3) experiential, (4) subjective.
(1) Theistic metaphysical singularism answers the question of the meaning of life by unfolding
a general conception of reality grounded in God as meaning-giving creator (Nozick 1981, 1989;
Cottingham 2003, 2016; McGrath 2004, 2011, 2014, 2015; Nancy 1997, 2008, 2013; Waghorn
2014). It employs meaning as a category for understanding reality, which is conceived in various
modes (patterning, integration, enlightenment, relation). (2) Theistic narrative singularism empha-
sizes how the meaning of life is narrated by a story or cluster of stories across all fundamental aspects
of human existence (Seachris 2009, 2013, 2016). (3) Theistic experiential singularism points to the
discovery of life’s meaning through experience. It is based either on the perception of God’s action
in our life (Sauter 1995) or on the example of a faithful community (Tolstoy 1987). (4) Theistic
subjective singularism starts from the human conscience and sees the meaning of life as perfect hap-
piness made possible by the creator of a purposeful universe (Goetz 2012).
THE CHALLENGE OF THE SURROGATE THEORY
From the outset, theism must justify the philosophical validity of the concept “meaning of life.”
Some scholars deny its theoretical relevance by referring to its psychological origin and historical
development (Hochschild 2017; Freud 1960; Flew 1976). For them, the meaning of life is not a
concept in its own right but a surrogate for something else, and a ghostly concept that reflects sub-
jective unease.
If we look at the historical development of the concept of the meaning of life, we can note
that it substitutes the classic concept of happiness by adding to it the modern importance of sub-
jective orientation (Marquard 1991, 38). The substitution occurred after Immanuel Kant’s deon-
tology supplanted the Aristotelian and Scholastic eudaimonism. Indeed, the birth of the expres-
sion “the meaning of life” coincides with the philosophical development of the notion of
“value” starting in the eighteenth century. The expression appears in philosophers such as Frie-
drich Nietzsche and Wilhelm Dilthey, for whom it means value, purpose, or significance of life.
Since then, the expression has crossed the border from philosophy into everyday language. We
all have an implicit understanding of what the question is about, although we might not know
the answer. However, connecting the question with certain naturalistic assumptions that are typi-
cal of modernity undermines any possible theistic development of the idea. A metatheoretical con-
sideration of what this question entails in modernity shows that these assumptions eliminate alto-
gether any possible theistic account.
This is the challenge of what we can call the surrogate theory, which claims that the question of
the meaning of life is ill-conceived and parasitic on classic problems. It urges us to drop the ques-
tion. Following this scenario, some scholars believe that the concept “meaning of life” does not
entail a positive and constructive theory. It is, on the contrary, a cul-de-sac out of which no theoret-
ical paths lead, for two main reasons.
Meaning: Theism
First, the very notion of “the meaning of life” emerges from a modern worldview dominated by
individualism, skepticism, and nihilism. It is entirely different from a framework that might allow
for the genuine discovery of the human purpose (Hochschild 2017, 6). In Joshua Hochschild’s
view, the modern conditions under which theorists coined this expression preclude in advance any
meaningful vision of life beyond the subjective horizon of one’s personal feelings. The meaning of
life is an existential question disconnected from any moral or metaphysical framework and incom-
patible with any theistic view of human life.
Second, the everyday context for questioning the meaning of life appears most often to be a
context of crisis. The simple fact of asking “what is the meaning of life?” indicates a symptom of
sickness. As soon as we ask about the meaning of life, we can be assured that we are sick (Freud
1960, 436). The question thus has no philosophical value. It only expresses unhealthy emotions
(Flew 1976, 167). Coincidentally, the birth of the idea was soon followed by a widespread cultural
discourse around “the crisis of meaning” or “the loss of meaning.” From this perspective, the ques-
tion is a pathological inquiry parasitic on genuine problems regarding human life.
It seems thus that the modern expression “the meaning of life” has created nothing but confu-
sion and frustration that undermine a rational treatment of the question. Yet the surrogate theory
can be debunked through an integrative approach that emphasizes the expression’s innovative force.
Just like the very concept of meaning in general, “the meaning of life” offers a new opportunity for
philosophy as a concept whose philosophical potential must be discovered and expanded (Figal
2006, 84). The wide spectrum of questions covered by “the meaning of life” suggests a way to inte-
grate various philosophical problems that were usually treated separately. In this view, the meaning
of life is understood not just as a surrogate for happiness. On the contrary, it includes happiness
among its questions but also poses other questions that are equally important.
This innovative version of the term “meaning of life” has been wholeheartedly adopted by the-
ists. John Paul II, for example, has taken it up in his encyclical Fides et ratio, which deals with the
role of philosophy in the contemporary world. The modern quest for the “meaning of life” is a new
moment in the development of humanity that can associate a deeper knowledge of reality with a
higher level of personal self-consciousness (John Paul II, 1998, 9). Its integrative character lies not
only in its invitation to see these questions in relation to one another, but also in its stress on find-
ing a fundamental balance between objective and subjective dimensions. The very possibility of dis-
covering the real meaning of life seems to be undermined by modern skepticism, undifferentiated
pluralism, and a lack of confidence in the possibility of ever obtaining truth. The question of the
meaning of life entails, nonetheless, a search for ultimate truth that can never be eradicated from
the human mind and that has always animated all religious and philosophical texts and practices.
The pursuit of the meaning of life represents a radicalization of questions about “the meaning and
ultimate foundation of human, personal and social existence” (John Paul II, 1998, 12–15). In John
Paul II’s perspective, such a radical search can reach its end only by reaching the absolute (1998,
45). Theism seems thus the best candidate to tackle this question because theism is both radical
in its focus on God at the summit of the scale of being and integrative in its topical outreach.
THEISTIC PLURALISM
The theistic approach on the meaning of life has taken two paths: pluralism and singularism. The
latter is favored by most theists.
Theistic pluralism claims that there are several meanings of life but that, among these, the rela-
tion with God provides the deepest significance (Mawson 2016). In the pluralist view, the term
“meaning of life” is nothing but a placeholder for a set of questions: Why are we here? What is
the purpose of life? What is the significance of events in my life? What can I do to make my life
meaningful? The challenge is to understand the relation between all the relevant questions and also
to identify which levels of intelligibility are perhaps higher (or deeper) than others. The amalgam
theory advanced by T. J. Mawson holds that there are several ways in which to understand the
terms “meaning” and “life.” Mawson believes that the general question of the meaning of life

Meaning: Theism
encompasses several questions, and that each question has a specific answer. One could even say that
there are many meanings of life (Mawson 2016, 15). One example of the range of meanings typical
in this approach is the pairing of divine purpose and self-creative autonomy. Assuming that God
exists, God’s plan gives meaning to our life by providing us with a purpose projected on the canvas
of the whole universe (meaning simpliciter, which is the ultimate meaning); on the other hand, there
is a sense in which our life is meaningful if our self-creative autonomy is realized independently of
any purpose that we receive from God (meaning for us). These two meanings can conflict: by pro-
viding us with a purpose, God delimits in some way our self-creative autonomy. A meaningful life
requires therefore both meanings: purposiveness and autonomy. None of these meanings, Mawson
contends, can be the meaning of life (not even God’s purpose).
Mawson finds that there is some truth in Jean-Paul Sartre’s idea that God, if He were to exist
and if He had assigned us a purpose, would make our life meaningless. Sartre uses the metaphor of a
paper knife to highlight the importance of human self-determination. The paper knife is created by
an artisan to serve a certain purpose—to slit the uncut pages of a book. The human being in a the-
istic view, claims Sartre, is similarly created by an artisan God to serve a certain purpose. Thus a
human being is like a paper knife in the sense that the latter has no initiative and no responsibility
of its own (Sartre 2007, 21). The theistic view is, according to this account, counterintuitive, for it
annihilates the very dignity of the human person.
Mawson holds that Sartre is right to assert the role of autonomy in determining the meaning of
life. For instance, somebody who works in a car factory as a junior widget-affixer might find his job
meaningless in a Sartrean sense because all details of his job are prescribed by management. Mawson
notes, on the other hand, that Sartre’s position is equally counterintuitive, for it fails to recognize
human beings’ importance in the larger scheme of things. In this case, the junior widget-affixer is
a significant part of the overall structure of the car factory, without which the car factory could
not function properly. Although the junior widget-affixer is not able to make decisions about all
the details of his work, he still plays a role in the factory. Similarly, a divinely conceived purpose
enhances the meaningfulness of our lives because it confers on it a high-scale significance that
Sartre’s self-autonomy does not possess. Indeed, should the junior widget-affixer become a manager,
he might find his new position meaningless in a new way. He would then have maximal freedom
with regard to running the factory, he could decide to produce something else or nothing at all,
and he could even award himself the employee-of-the-month badge. However, in this case, too,
his life would be meaningless in a certain respect. Although he would have maximal self-autonomy,
he would have lost a clear role in the larger scheme of things. What is the point, indeed, of getting
the honor badge if it is not received in comparison to other workers’ job performance? It would be
like playing chess against oneself.
This collection of meanings that can be at odds with each other need not be reduced to a grab-
bag polyvalence in which these various meanings just happen to be together. The trade-off between
these meanings is rendered possible by the relative depth of certain sorts of meaningfulness. On this
account, the polyvalence of meanings could be ranked by noticing the meanings with higher value
in amplitude and depth. There can certainly be a godless meaningful life that exhibits especially
self-creative autonomy. But the prospect of union with God offers a deeper sort of meaning that
would enhance the overall meaning, even though this would come at the cost of diminishment of
individual self-autonomy. The theistic formula that Mawson proposes for this amalgam polyvalence
theory is a line from the book of Job: “The Lord giveth, and the Lord taketh away” (Job 1:21). The
formula is not to be interpreted as a zero-sum game but rather as an abundance of giving that over-
flows the loss: “God ends up bringing more to the party (or, if you will, afterparty) than He takes
away” (Mawson 2016, 17). The afterlife is for Mawson the crucial point in this perspective of abun-
dance. The afterlife supports both types of meaning: the ultimate significance meaning (meaning sim-
pliciter), because our actions will have everlasting significance in the ultimate context in which they
are placed, and the meaning for us, insofar as we will exist to appropriate this everlasting signifi-
cance. One might imagine a scenario in which our life remains everlasting only in God’s memory,
but in which we do not survive our death. In this case, our life will still be meaningful in the larger
scheme of things but less meaningful for us because we will not exist anymore to appropriate it. For
Meaning: Theism
our own life to remain meaningful, there is need for a guarantee of our immortality, which would
secure an ultimate significance meaning that atheism is not able to provide.
THEISTIC SINGULARISM
The second option within the theistic approach on the meaning of life is singularism, which almost
all theists promote. In this view, the question of life’s meaning is posed as a singular request. Even
though there are various questions included in the big one, like individual bricks they all converge
into a single structure, supported by a common foundation. When we inquire into the meaning
of life we are searching for the ultimate meaning of our life, which surpasses regional and partial
meanings unveiled by science or by action theories (John Paul II 1998, 45). The ultimate meaning
is all-inclusive, covering a multitude of questions and sustaining all directions of human life.
We will classify the various singularist positions in four main types: (1) metaphysical singular-
ism, (2) narrative singularism, (3) experiential singularism, (4) subjective singularism.
THEISTIC METAPHYSICAL SINGULARISM

Theistic metaphysical singularism looks for the ultimate meaning of life against the backdrop of a
general conception of reality (Nozick 1981, 1989; Cottingham 2003, 2016; McGrath 2004,
2011, 2014, 2015; Nancy 1997, 2008, 2013; Waghorn 2014). It employs meaning as a category
able to integrate our lives into the bigger picture of the universe. Thus the understanding of mean-
ing involves a relational account that unifies all the aspects of life on a foundational basis. All theistic
accounts that can be labeled “metaphysical” are indeed operating with this categorical positing.
They also emphasize the shortcomings of naturalism, which fails to provide an adequate conception
of meaning.
Robert Nozick’s theistic conception of life’s meaning is based on the idea of meaning as rela-
tional. Meaning constitutes a dimension of reality defined by integration, just as value does. Both
meaning and value are dimensions of reality that represent an integration of diversity into unity:
value entails an inherent integration, while meaning entails a relational integration. Something has
intrinsic value insofar as it is organically unified, that is, has a high degree of unity in diversity
(Nozick 1989, 164). For instance, the aesthetic value of a painting resides in the capacity of the
painting to integrate a diversity of elements (colors, shapes, etc.) into a tight unity. The greater
the diversity, the higher the organic unity. A monochromatic painting would have a high degree
of unity but not a high degree of organic unity. On the other hand, something has meaning insofar
as it connects to something else, outside itself: “Value involves something being integrated within its
own boundaries, while meaning involves its having some connection beyond these boundaries”
(Nozick 1989, 166). While value must realize unity, meaning must surpass boundaries. For
instance, we question the meaning of our life because we have to grapple with various limits: the
limit of information, the limit of space and time, the limit of resources, the limit of personal skills,
the limit of social constraints, and so on, until, finally, we have to grapple with the limit that death
poses to our existence. When we ask about the meaning of an event or an action, it is because we do
not have an immediate indication and have to look for something beyond that event or action: we
look for a cause, a purpose, a link that renders intelligible the connection between various elements
(Nozick 1981, 601). A limited thing thus receives its meaning from something outside itself.
One’s life can be oriented toward high values, such as family harmony, fulfilling professional
ambitions, and so on, and still be limited by the threat of one’s coming death. Thus the highest
degree of meaningfulness arises from transcending all limits and reaching something without limits,
something unlimited. This something does not, itself, have a meaning. If it had one, it would need
to have a relation to its meaning as something outside of itself. The concatenation of relations of
meaning must thus stop at something that is unlimited and without relation to any further mean-
ing, so that there is no vantage point from which one can assess its limits. Here Nozick refers to
the kabbalistic term Ein Sof, the Unlimited, which surpasses all inquiries of meaning precisely
because the question “what is the meaning of the Unlimited?” is by itself absurd. There is no

Meaning: Theism
external thing to which the Ein Sof could connect through a meaning giving relation. The concepts
“meaningful” and “meaningless” cannot be applied to the Unlimited since the Unlimited transcends
their distinction. The Unlimited is its own meaning; it stands as an absolute value irreducible to
anything else.
This idea of the Unlimited is more powerful than the idea that God’s purpose for human
beings gives meaning to life. Centering the meaning of life on God’s purpose does not fit the bill,
for Nozick, because this option does not provide us with sufficient material for the significant role
that we ought to play according to the divine plan. Simply being part of God’s plan is not enough.
If we make an analogy with painting, we wouldn’t be happy, for instance, if we would play the role
of the rag used to wipe off brushes. We would need a more significant role to obtain meaningful-
ness. Similarly, even if an extragalactic civilization created us with a purpose in mind, this perhaps
would not suffice to provide fulfilling meaning: maybe we were created just to serve as food for
intergalactic travelers. Therefore it is not enough to say that God has some purpose for us. His pur-
pose for us must be meaningful (Nozick 1981, 590). What really grounds our life’s meaning is not
just the purpose of God but the fact that God is, in Himself, unlimited.
A life can thus obtain meaning only from something that is unlimited and all-inclusive. But this
statement does not entail a claim about the actual existence of that thing. We can know that our life
would be fully meaningful in relation with an Unlimited, but how do we know that such an Unlim-
ited really exists? For Nozick, the only way to come close to the question about the existence of this
Unlimited would be a type of understanding that would swipe away the reductionism rampant in
science, humanities, and even philosophy. Reductionism, he notes, operates in various ways. It
might reduce the more valuable to the less valuable, the more meaningful to the less meaningful
(e.g., love to chemical reaction, religion to expression of fear, political principles to tools of powerful
elites). It might reduce what has a certain value to what does not have that value (for instance, by
explaining creativity in terms of parts that do not exhibit the value of creativity; love in terms of
processes that are not “in love”). It might consider the whole as sum of the parts, without integra-
tion at the level of the whole. It might diminish the role of the human being in the order of things
and envision human beings as the instruments of forces that they do not control. A nonreductionist
understanding must first and foremost approach valuable human traits as having an integrity of their
own (being a self, freely choosing). Then, it must follow our common sense by placing actions in a
network of patterns. Finally, it must reason by analogy, drawing similarities and moving from one
analogical focus to another. What we obtain through this type of understanding is not a deductive
proof in the sense of the classical arguments for God’s existence, but a rational understanding and a
response to value and meaning. Understanding meaning thus entails a work of patterning and inte-
gration, which is realized with a creative philosophical spirit that always remains under the con-
straint of truth. This constraint constantly underpins meanings, ideas, and values that the philoso-
pher experiences as given, not self-imposed. One comes close to the Unlimited through this
experience of understanding, in which value and meaning are each time attained anew (Nozick
1981, 647).
The reductionism criticized by Nozick especially affects naturalism, which undermines the orig-
inal interest that sparks the question about the meaning of life. There are indeed several problems
that a naturalistic account leaves unsolved and that only a theistic account can handle. John Cot-
tingham’s theistic view on meaning, for instance, focuses on two issues: the challenge of modern sci-
ence and the human vulnerability. He contends that naturalism is not sufficiently supported by the
results of science, and it does not give a satisfactory account of human suffering. First, our relation-
ship with the universe is a core issue in the question of the meaning of life. While objective achieve-
ments and a subjective sense of fulfillment, promoted by naturalistic accounts, are indispensable for
a meaningful life, they still do not exhaust questions such as why we are here in the universe. This
question is surely rendered more difficult by modern cosmology and evolutionism, which depict the
universe as a vast machine that functions according to laws indifferent to human concerns. The evo-
lution of our species appears, from this perspective, as an impersonal and accidental chain of blind
natural forces. There seems to be in this picture no room for a principle of goodness or human pur-
posiveness. However, this skeptical reading of the scientific forma mentis is not the only one
Meaning: Theism
possible. As a matter of fact, Cottingham believes that neither the mechanical view of the universe
nor biological Darwinism exclude de jure a metaphysical teleology. The idea of an unbounded cos-
mos containing innumerable worlds is not opposed to God’s infinity and omnipotence. Besides, the
inevitable chain of causes in the physical realm does not exclude the operation of providential pur-
poses in the metaphysical realm (Cottingham 2003, 37). Similarly, in the evolutionary process there
is no logical obstacle that would prevent a mechanical system of efficient causality from coexisting
with a purposive system. For instance, our brain is a blind mechanical system, but at the same time
its outputs constitute the purposive plannings and doings of conscious agents (Cottingham 2003,
48). The ability of the universe to produce life and intelligence does not exclude its own teleology.
It is thus rationally possible to hold that our lives belong to a creative order that is inherently good.
Cottingham admits that the idea of the intrinsic goodness of the universe needs to be elaborated fur-
ther if it is to resist doubt. In his assessment, however, the naturalistic accounts of the meaning of
life that suspend the issue of our place in the universe do not have sufficient support in the modern
conception of the universe. There is, in his eyes, not enough evidence to support the conviction that
the universe is meaningless and our lives cannot find any meaning in relation to it.
A second problem left unsolved by naturalists regards the futility and fragility of human life.
The human flourishing at the core of many naturalistic accounts is rightly grounded in what is
noblest and best within us and can become a normative ideal, an objective value. However, this
approach neglects the negative experiences of human suffering, failure, vulnerability and, lastly,
death. The idea of the rationally autonomous moral agent that makes self-sufficient decisions about
her good life is a fantasy, because it neglects the origin and destination of our life. We were born in
a context of vulnerability, for we did not make ourselves. We carry since birth this creaturely depen-
dency on what (or who) made us be here. Furthermore, the final destination of our earthly life is
death, which limits all future projects. A theistic view incorporates these negative experiences into
a global conception. It sees our human nature as more than a list of characteristics that our species
happens intermittently to possess. In this view, all worthy human endeavors, including those
marked by fragility, contribute to
the establishment of a moral order that the cosmos was created to realize. To act in light of
such an attitude is to act in the faith that our struggles mean something beyond the local
expression of a contingently evolving genetic lottery; that despite the cruelty and misery in the
world, the struggle for goodness will always enjoy a certain kind of buoyancy. (Cottingham,
2003, 72)
The gap between our aspirations and our weakness and mortality is covered by an inner conversion
that connects one with the core of life. This change makes us see life as a gift bestowed by an origin
that generates everything good, true, and beautiful.
At this point, Cottingham switches from the theoretical to the practical discourse, so as to show
that the essence of religion is more than adherence to doctrine. It is a commitment to a certain way
of life. Like Nozick, Cottingham ultimately advocates for a responsiveness to meaning rather than a
demonstration of meaning. His theistic view on the meaning of life takes the form of an epistemol-
ogy of receptivity that associates evidence with certain transformations in the subject (Cottingham
2016, 47–58). The rational orientation toward good is not enough. Our fragility requires faith in
the moral order of the universe. Such faith is not self-delusional because it is rooted in an aesthetic
wonder that responds to the objective beauty of the world. As a matter of fact, our modern idea of
the meaninglessness of the universe is the artifact of “our own confusions and bitterness as we will-
fully turn away from the light, as we steadily advance with our bulldozers until we cover the whole
planet in concrete and then complain that the cosmos we live in is no more than meaningless rub-
ble” (Cottingham 2003, 102). We thus need an inner conversion to overcome the modern alien-
ation that has blurred our vision of the universe and our role in it. The meaning of life is ultimately
disclosed by this inner change, which sees human flourishing on the large canvas of an ordered uni-
verse.
While Cottingham conceives the relation of science and the meaning of life as a relation of
nonexclusion, Alister McGrath proposes a stronger relation. He believes that science supports the
theistic idea of a purposeful life integrated into a purposeful universe created by God. In his view,

Meaning: Theism
the problem of meaning is a problem of connecting our scientific understanding of the world with
matters of value and purpose. Science teaches us how things work but not what they mean. Mean-
ing has to do with the “why?” question: Why am I here? What is the point of life? What is the point
of the universe? McGrath distinguishes between understanding and meaning by assigning under-
standing to the internal functioning of things and meaning to their value and purpose. It is quite
possible that someone understands something but does not capture its meaning. McGrath confesses
that as scientist he had the experience of the beauty and wonder of nature but missed its meaning
(McGrath 2015, 8). The two should, however, be combined into a big picture that links our
thoughts about the world with our thoughts about value and purpose. Religion offers such a big pic-
ture, sorely needed by science. McGrath claims that the conflict between science and religion is a
historical contingency, a social construction that does not necessarily arise from the nature of science
and religion. There is plenty of evidence to support coherence rather than conflict between the two.
For instance, recent scientific explorations have shown how the initial conditions of the universe
have supported the emergence of life. The universe has been fine-tuned for life, and even a slight
variation from the initial conditions would have made life impossible. Some fundamental cosmolog-
ical constants are at the right scale to allow for human existence. If the strong coupling constant
would have been smaller, hydrogen would be the only element in the universe, thus excluding car-
bon on which life depends. Life could not emerge without some hydrogen being converted into car-
bon by fusion. If the same constant would have been larger, the hydrogen would have been con-
verted to helium, impeding the formation of long-lived stars, crucial for the emergence of life
(McGrath 2015, 89).
Scientific descriptions do not obstruct theism. In fact, a mental map is needed to make sense of
them. “Making sense” is the main role of a theistic big picture. It provides a framework of intelligi-
bility for all aspects of reality. This is a sort of intellectual evidence that emerges not as a result of
deductive reasoning but rather as an enlightenment. Following C. S. Lewis, McGrath describes
the Christian faith as a lens that brings things into focus. The faith enables an intellectual seeing
that was not available in its absence: “It was like turning on a light and seeing things properly for
the first time” (McGrath 2014, 15).
Such discovery of meaning is vital for those who have to cope with trauma and suffering. As
recent psychological studies demonstrate, those who can see an order and meaning underneath the
surface of random and painful events are more likely to cope with their suffering than those who
cannot (McGrath 2014, 6). In the psychological literature, human well-being is related to the grasp
of the meaning of life, which enhances the ability to face life challenges. The meaning-focused cop-
ing is grounded in beliefs, values, and existential goals that help a person in a difficult time
(McGrath 2015, 173). These statistics, McGrath shows, are confirmed also by personal testimonies
like that of Viktor Frankl, who was a Holocaust survivor. Frankl wrote that survival in Nazi concen-
tration camps depended on the way in which the prisoners were able to discern meaning and pur-
pose in their atrocious situations. Man’s search for meaning represents the primary motivation for
his life. This motivation, Frankl insists, is not just consolation but a way to bear suffering (Frankl
2006, 77). As the experience of Frankl attests, there is a peculiar relation between meaning and
action, insofar as grasping the bigger picture calls for an existential transformation. Meaning is, from
this viewpoint, more than simple knowledge because it loads significance into information
(McGrath 2011, 112).
Theistic metaphysical singularism thus incorporates our human life into the larger picture of
the universe. The meaning of life arises from the global intelligibility of the world, whose founda-
tion is God. The method through which we attain this meaning is not the usual deductive reason-
ing, but an understanding that provides intellectual evidence. At the same time, this meaningfulness
does not just require vision, it also implies experience. A peculiar version of this metaphysical singu-
larism comes from phenomenology, which employs meaning as a core category expressing the
intrinsic intelligibility of the world (Oliva 2018). Jean-Luc Nancy (1997, 2008, 2013) and Nicholas
Waghorn (2014) propose a phenomenology of the meaning of life under the influence of Martin
Heidegger. This latter warns that being as such differs from entities. The search for a deeper onto-
logical level leads Nancy to move from signification (entity) to the sense itself (being). In the same
way, Waghorn points to nothingness, where the infinite regress of the chain of entities stops.
Meaning: Theism
Nancy understands meaning as relation and as an opening of relations. The meaning of life, in
his perspective, arises from a practice of sense that unveils our embedding in a world constituted by
relations and grounded in God, who is self-relational. Nancy radicalizes the distinction between
sense/meaning (in French, sens) and signification (French, signification) by placing the concept of
sense/meaning at a foundational level, prior to all significations. Sense for Nancy is the very opening
of all possible significations. It must be understood verbally as a process of signifying from which all
significations emerge (Nancy 1997, 10). Sense is a relation, a referring or sending toward some-
where else, toward others (Nancy 2013, 9). To be sure, although sense does not end in signification,
neither does it end in the unsignifiable, a nihilistic danger that Nancy seems eager to avoid (Nancy
2008, 127). The referral of sense is infinite but, by being infinite, it discloses the world: “the infin-
ity of a sense that no signification can fill, and that, let it be said, envelops together with mankind
the totality of the world with all its existents” (Nancy 2013, 4).
Corresponding to the distinction between sense and signification is the distinction between
“being sense” and “having sense.” When we talk about the sense of the world or the meaning of life
we usually intend “having sense.” We are looking for a signification, a completion that embraces, as
it were, the whole world or the whole life: the world has a sense, the life has a sense. It is, however,
necessary to understand the sense of the world also as being sense. This way is disclosed by the
Christian experience. The Christian world has its sense beyond itself because it is a creation of an
omnipotent and loving God. At the same time, the incarnation and resurrection of Jesus Christ
reveal the world as being sense: “they come from the region of history where human culture grasps
itself as a ‘world’ detached from any cosmogony and where the ordering of a kosmos no longer could
or should be looked for except beginning with ‘this world here’” (Nancy 2013, 52). The death of
Jesus on the cross depicts the image of a life exposed to a suffering that lacks an immediate signifi-
cation. In this sense, resurrection does not offer a signification, something like a second life, but
rather reorients this life toward a vertical dimension (Nancy 2013, 52).
This account differs from classical theism as well as from atheism. Classical theism assigns life a
signification conceived as purpose of God or grounded in God who is His own meaning. Atheism,
on the other hand, conceives the signification of life as a value or set of values defined objectively or
subjectively. They both miss the mark, from Nancy’s perspective, because they both remain at the
level of signification without going deeper into the movement of sense that makes possible significa-
tions. “Being sense” must be grasped through a practice of sense. One example is music. Sound has
a peculiar character. First, sound has no hidden surface, thus no hidden significations: there is noth-
ing behind a sound. A word or an image can disguise a hidden signification, but sound does not
conceal anything, it is the “in-significant opening of sense” (Nancy 1997, 85). Second, sound is
the line of contact between interior and exterior. For example, when we listen to music, our body
vibrates in the rhythm of the music. At the same time, the sound diffuses itself in its environment,
so that even after the sound stops, its vibrations still remain in the air for a few seconds. Music is, in
this sense, pure relation because it does not preexist its representation and it fully engages the
human body.
The ability to grasp our existence within the web of relations is, for Nancy, not a purely cogni-
tive matter. What is at stake here is not a signification that one can possess. It is rather an opening
of relations that cannot be mastered in the way we master a signification and thus requires faith.
Faith is trust in this opening, which constitutes the world and, at the same time, is not from this
world. Although his downgrading of significations might bring Nancy to the limit of agnosticism,
he acknowledges that one cannot grasp the meaning of our existence in this world without orienting
oneself toward a divine figure that is relational in itself and thus makes possible all relations in this
world: “The mystery of the Trinity strikes this spark: sense is relation itself, the outside of the world
is therefore within the world without being of the world” (Nancy 2013, 52). Faith seems, from this
perspective, unavoidable: renouncing faith would mean to shut down the opening of relations.
Another phenomenological position in the debate on the meaning of life proposes nothingness
as the key to the meaning of life. Whereas Nancy looks to the relational functioning of meaning,
Waghorn questions the very possibility of a positive determination of the ultimate meaning of our
life. Although he does not frame his position as a theistic one sensu stricto, Waghorn shows how

Meaning: Theism
the concept of nothing might help settle the problem of infinite regression that theism tries to solve.
When we talk about meaning, we are accustomed to associate meaning with something. Thus,
when we talk about the meaning of life, we expect to point to an entity of whatever nature (ideal,
spiritual, physical) that confers meaning on life. However, all positive determinations of the mean-
ing of life are subject to an infinite regress. On this account, the naturalistic determinations of life’s
meaning (pleasure, altruism, aesthetic experience, love, objective achievements) are the first to fail
the test of regress. In all cases, one can always imagine a further request of justification: “as far as
love is something cognitive, that is, as far as it involves an intentional object, we are given resources
to perform a reflexive iteration on that object and ask ‘But what is the point of loving it?’” (Wag-
horn 2014, 183).
The established theistic positions fail this reflexive iteration as well. For instance, the thesis that
the meaning of our life is the purpose that God has for us is plagued with various deficiencies. What
if God’s purpose is one that we do not like, such as to serve as food for intergalactic travelers? Even
a greater purpose is still not sealing the deal, because it raises the question: “What is the point of
this purpose?” (Waghorn 2014, 199). Similarly, the concept of an afterlife offers a good trade-off
for problems in the earthly life, but it seems to have some limits. For instance, hell does not make
the afterlife very suitable for those assigned there. The question still remains: “What makes the after-
life meaningful?”
What can stop this infinite regress? Waghorn thinks that “nothing” is the perfect candidate
because “nothing has no limits which a reflexive iteration can cross, as it is not characterized in
any way” (Waghorn 2014, 188). From this apophatic perspective, the ultimate meaning of life is
reached when one goes to the bottom of everything and in so doing overcomes every single object
(be it material or spiritual). The result of this rational regress through the chain of entities is noth-
ingness, which by default cannot be positively defined and thus cannot make the object of a further
regress. The reflexive iteration differs from the deductive method insofar as its result is not an entity
that can be positively determined. One uses reason to reach, as it were, the very limits of reason. In
this sense, naturalism, although it claims to be rationally sound, does not go far enough in the exer-
cise of reason.
To sum up, theistic metaphysical singularism focuses on the integration of human life into the
big picture of reality—an integration that naturalism is not able to provide. This integration occurs
by means of the concept of meaning, employed in several ways: surpassing the limits of our exis-
tence (Nozick), finding value and purpose in the connection between the universe and human exis-
tence (Cottingham, McGrath), grasping the web of relations that makes up our world (Nancy), and
performing a reflexive iteration that stops at nothingness (Waghorn). On all accounts, metaphysical
theism, as compared to naturalism, claims to provide a deeper meaning: naturalism is reductive and
does not grasp meaning in its surpassing of limits (Nozick); it does not capture the meaning behind
scientific discoveries and does not see how human fragility connects to the purposeful universe (Cot-
tingham, McGrath); it remains at the level of significations without seeing the relational sense that
makes them possible (Nancy); it falls short in the exercise of reason (Waghorn).
THEISTIC NARRATIVE SINGULARISM

Narrative singularism regards the meaning of life in terms of a narrative that covers all aspects of our
life, from its place within the universe to the normative conditions of action. According to this
account, explanatory approaches to the meaning of life lack completeness. Joshua Seachris, for
instance, argues that only a story or a cluster of stories can fully address the fundamental issues of
our life. He offers several reasons why the narrative account might be the best suited to address
the question of the meaning of life (Seachris 2009, 5–23).
First, a theistic narrative account best fits the application of the concept of meaning to nonlin-
guistic phenomena such as human life. Much confusion in the discussion of the meaning of life
arises precisely because the nonlinguistic use of the concept of meaning is not properly captured.
If we consider an individual concrete life situation, we realize that in asking for its meaning we
are asking for a story that makes sense of it. Seachris takes the example of a father who finds his chil-
dren fighting in the playroom. When he asks “what is the meaning of this?” he is looking for an
Meaning: Theism
explanation that includes information about how the scuffling started, who did what, and so on. He
needs the accurate story of his children’s scuffle, which provides a more substantial understanding
than a mere description of the kicking and screaming. The meaning of the situation is thus con-
veyed by the story of the fight.
Moreover, the narrative account not only clarifies why we use “meaning” when talking about
human life but also why we use “the” in talking about “the meaning of life.” When the father is ask-
ing for the meaning of his children’s conflict, he requests a story, the accurate story, that explains
what happened. Similarly, when we ask for the meaning of life, we search for one narrative, not
for several alternative narratives. The narrative account preserves, in this way, the original nature
of the question, which many naturalists abolish by breaking it down into various questions and
eliminating “the” and replacing “of” in “the meaning of life” to obtain a more modest “meaning
in life.” The narrative account offers a single, unifying interpretation of life.
Second, the theistic narrative account is not only singular but also all-inclusive because it
addresses all the issues raised by the meaning of life question. While the naturalistic accounts reduce
the question of life’s meaning to problems of value, worth, and purpose associated with human
action, the narrative account has a wider range. It also includes the cosmic questions about the ori-
gin of our life and our relation with the universe. For instance, many religious narratives such as the
biblical narrative or the Qurʾan tell stories about our role in the universe.
Third, the theistic narrative account can make room for the issue of death, evil, and futility,
which merely explanatory accounts have a hard time addressing. The ability of the narrative inter-
pretation to address such issues comes from its own structure, which includes an ending. The nar-
rative ending is fundamental in every story. The way a story ends sheds significant light on the
whole plot. Events narrated during the story acquire new significance in light of the final word. This
explains why, when we look for a narrative that provides the meaning of our life, we are so con-
cerned with the end of our life and even with the end of the universe.
What does a story of human life look like? According to Seachris, such a narrative must address
the fundamental dimensions of the human condition and provide the ultimate metanarrative frame-
work that can clarify those dimensions and offer practical guidance. This narrative must thus satisfy
two criteria: formal and epistemic. Formally, the narrative must narrate across all existentially
charged aspects of our life and our world (Seachris 2016, 19):
(1) Why do we, or why does anything, exist at all?

(2) Does life have any purpose (or purposes), and if so, what is its nature and source?
(3) What, if anything, grounds the value, significance, and worth of these pursuits and projects?
(4) Why is there pain, suffering, and evil?
(5) How does it all end? Is death final?
The category of narrative suits theism better than naturalism because theism satisfies all five of these
formal criteria in a substantial way. There are surely also naturalistic narratives (for instance, Albert
Camus’s interpretation of the story of Sisyphus), but many of them offer no robust narrative ending
required by the fifth criterion above. Christian theism, for instance, offers a narrative ending in the
form of the afterlife, which is an ending that itself never ends (Seachris 2013, 469). Although the
Christian narrative cannot offer an ultimate stance that regards the whole of the beatific afterlife,
it can still bring a closure to the earthly life, which solves the problems of significance and futility
that naturalistic accounts struggle to settle.
Epistemically, the narrative must satisfy the truth condition. As with the story of the father of
children in conflict, we are looking not for several stories but one, namely, the accurate one that
corresponds to reality. In this sense, “the narrative interpretation presupposes that there can be only
one meaning of life though this does not preclude a rich variety in the ways meaningful life can be
realized” (Seachris 2016, 20).
Seachris believes that the Bible is a good candidate for a meaning of life narrative because it
embraces both the cosmic-intelligibility and the individualist-normative dimensions of our life. The

Meaning: Theism
redemptive-historical narrative of the Bible addresses ultimate origin, purpose, value, significance, pain
and suffering, futility, death, and ultimate ending (Seachris 2016, 23). The story of Genesis, the stor-
ies about the making of a covenant between men and God, the stories about the life, death, and Res-
urrection of Jesus Christ, and the eschatological narrative of the end of the world are all stories that
satisfy the formal conditions of a meaning of life narrative. (1) The origin of life is narrated in Genesis;
(2) and (3) the purpose of life is narrated across several sequences that talk about love as the telos of
humanity and as grounded in God’s infinite love; (4) the problem of evil and suffering is addressed
in the story of Job, Ecclesiastes, the story of the Jewish exile, and the story of the Crucifixion of Jesus;
(5) the issue of death and afterlife comes up in the story of Jesus’s resurrection.
The truth condition of the biblical narrative is satisfied, Seachris seems to suggest, by a co-natural
relation between the narrative, our own human capabilities, and the reality narrated. For instance, the
human desire for a positive end-of-life narrative, including immortality, is not mere wishful thinking;
like our desire for water, it is a “truly natural desire that points to a referent capable of fulfilling it”
(Seachris 2009, 20). In this sense, a narrative of human life as an instrument of an intergalactic super-
ior population does not square well with the ingrained capabilities of creativity and intelligence dis-
played by humankind.
THEISTIC EXPERIENTIAL SINGULARISM

Theistic experiential singularism (Sauter 1995; Tolstoy 1987) claims that the meaning of life can
only be grasped through experience. Experiential singularists warn against the danger of the singular
request of the meaning of life to succumb to the human need to control the world. Gerhard Sauter
believes that understanding the meaning of life as a totality of meaning can sometimes be vitiated by
an intellectual attempt to achieve security and mastery over life. In such case, the meaning of life is
only an idol that humans create (Sauter 1995, 106). One can notice that the modern quest for the
meaning of life has moved from the given meaning to the subjective giving of meaning. This shift is
visible especially in the restriction of the concept of meaning/sense (in German, Sinn) to an intellec-
tual content, away from the sensorial signification. The German word Sinn means both intellectual
signification and perceptive sense, an ambivalence already present in the Latin sensus. The modern
quest for the meaning of life has abandoned the perceptive nature of Sinn. In this reduced intellec-
tual field, meaning is known only as the opposite of meaninglessness and contradiction. The mean-
ing of life is thus conveyed through intellectual constructions that appeal to God as a guarantor of
their coherence. Even the concept of meaning as relation can be misused by artificial constructions
that eliminate a concrete relation with God: “The question of God, when translated into the ques-
tion of meaning, leaves no place for talk of the revelation and silence of God. Looking for meaning
as an eloquent relation, it adopts a position beyond any specific divine communication” (Sauter
1995, 81).
A theistic approach should thus go back to the perception of the given meaning (what Sauter
calls “appointed meaning”), which always takes place in the experience of faith. “To perceive our
existence in such a way as to see it as God’s work is to inquire into meaning” (Sauter 1995, 152).
Instead of deploying an intellectual construction of a totality of meaning, the inquirer into the
meaning of life needs to understand her life experientially. She has to try to gather her life’s meaning
from the experiences that she lives and not from a ready-made conception.
For instance, one of the experiences that call for a strengthening of our perception is suffering.
Suffering triggers a receptivity in which we let life happen, we accept and appropriate what is given
to us. In Christianity, the fellowship with Jesus Christ is not an intellectual consolation that helps us
restructure our existence but a personal surrender to the presence of Christ in our suffering (Sauter
1995, 138). The Pauline epistles, but also Ecclesiastes, deploy this kind of faithful passivity that is
incorporated in the perceptive dimension of meaning. When the author of Ecclesiastes stops asking
questions, he does not succumb to skepticism. Rather, he turns away from an intellectual quest for a
context that he can control. He thus focuses on responding to God’s call.
Leo Tolstoy (1987) proposes another form of experiential meaning of life. For him the meaning
of life arises from the example of a faithful community. As soon as one realizes that individual suf-
fering makes sense within the scope of faith, one’s entire life begins to have meaning. In this sense,
Meaning: Theism
traditions reveal how faith sustains life. The recipe for happiness comes from happy people. Tolstoy
suggests that one adopt the beliefs of happy people in order to achieve the same results. The discov-
ery of the meaning of life is therefore a blend of inner transformation and social influence.
Tolstoy’s personal story became paradigmatic for the scholarship on the meaning of life. In the
middle of his life, he suffered from a crisis of meaning. Despite all his personal achievements,
wealth, and social prestige, the writer fell into despair for not being able to find a reason for living.
“Why do I live? What will become of all this after I die?” These questions pushed all his achieve-
ments into the background. Neither the sciences nor metaphysics were able to provide him with
an answer. The sciences are equipped with detailed and concrete information, but they do not take
the question of the meaning of life seriously. Metaphysics poses the question, but Tolstoy felt that
such an approach cannot give a concrete and detailed answer. The failure of science and metaphysics
to provide the meaning of life motivated Tolstoy to look within life itself by observing other people.
His struggle made him aware of the gap between the privileged elites and the majority of people
who carry on with their lives laboriously and faithfully. The privileged elites of his time were not
able to grasp the meaning of life because they were looking for an escape from what is dreadful
about it, always feeling threatened by pain, suffering, and death. Their escape took various forms
such as ignorance, epicureanism, suicide, and self-deception.
To this escape Tolstoy opposes the life example of the millions of people who endured, worked,
and created despite the sorrow and destruction inherent in life: “Indeed, since those long ago days
when life began and of which I know nothing, people who knew the argument about the vanity
of life (which seems to me to prove its senselessness), have nevertheless lived and brought to life a
meaning of their own” (Tolstoy 1987, 48). This understanding, which has sustained countless peo-
ple in the history of humankind, is grounded in faith. Faith represents the force of life and provides
a meaning that is not destroyed by suffering, deprivation, or death because it accounts for our life
comprehensively and positively.
Following the example of the peasants around him, Tolstoy understood that the meaning of life
can only be grasped while being embedded in the movement of life and letting oneself be carried by it.
The difference between the intellectual quest and the experiential quest can be captured through two
metaphors, the watch and the pump. The intellectual quest is similar to the dismantling of a watch by
a child: once the child pulls all the pieces apart, the watch stops working. Similarly, when we disman-
tle the pieces of life, life stops living, so to speak. The experiential quest, on the other hand, is similar
to the situation of a naked and hungry beggar who is brought to a splendid establishment where he is
given food and drink and asked to move a handle up and down. Before deciding why he was brought
there, whether the establishment is good enough, and why he has to move the handle, the beggar
must move the handle. Once he does so, he realizes that the handle operates a pump that draws water
to flow into a garden. From there, he will be moved to the garden to gather fruits. In this way, Tol-
stoy says, the beggar “will enter the joy of his Lord. As he progresses from lower to higher tasks he will
continue to understand more and more about the structure of the establishment and participate in it,
and he will never stop to ask why he is here, and he will never come to reproach his master” (Tolstoy
1987, 61). Simple working people behave like the beggar, following the will of the master without
criticizing him. The intellectual elites, on the other hand, eat the food given by the Lord without fol-
lowing His will but debating whether the master is stupid, or whether he does not exist.
Consequently, faith in God is not difficult to achieve: one has only to submit to the force of life
that is at work in us from the dawn of humanity and that keeps us alive and provides us with the
meaning of our lives. This submission might be rendered difficult by personal frustrations, weak-
ness, or vanity, but it can be facilitated by looking at the example of all the people who carried
on with their lives despite hardships and privations.
THEISTIC SUBJECTIVE SINGULARISM

Most of the theistic accounts see life’s meaning as a result of a comprehensive and objective world-
view grounded in God. This worldview is constituted either through knowledge (metaphysical or
narrative) or through experience. The subjective theism of Stewart Goetz (2012) proposes a slightly
different perspective: it starts from the consciousness of a person.

Meaning: Theism
Goetz introduces an epicurean theme into a theistic framework. His thesis is that the pur-
pose of life is perfect and unending happiness, which he identifies with pleasure. All our desires
are wired toward perfect pleasure—nobody wants pain for its own sake. The reason for this
identification is mainly that pleasure is an intrinsic good: it is not pursued for the sake of some-
thing else but for its own sake. The value of pleasure is not assigned by our desire. On the con-
trary, our desire is oriented toward pleasure. Pleasure is not good because we want it, but
because it is good in and of itself. As such, pleasure is not a moral good, because it is not a fea-
ture of our action, but of experiences with respect to which we are passive: “the experience of
pleasure that constitutes happiness is not an action that we perform but a passion that we
undergo” (Goetz 2012, 34). While the actions through which we obtain pleasure are morally
charged, pleasure is good in a nonmoral way. Of course, pleasure must not be disconnected
from action. It would be indeed counterintuitive to defend pleasures obtained through immoral
actions such as stealing, rape, and so on. The link between pleasure and action is justice. The
pursuit of pleasure must meet the requirement of justice, so that only virtuous people deserve
to experience pleasure.
In Goetz’s eyes, pleasure itself calls for theism. A perfect happiness defined as pleasure makes
sense only if our life does not end but continues into an afterlife, in which the experience of pain
is completely absent. A subjective naturalistic account can focus on the fulfillment of desires and
conceive it also as the ground of meaning (Frankfurt 1998). However, denying the afterlife under-
mines the long-term desirability of pleasure. What is the point of pleasure if it is fleeting? The
two conditions of the meaning of life, namely (1) a singular purpose of the whole life and (2) the
survival after death, open Goetz’s subjective account to the larger theistic worldview advocated by
other theists. First, the purposiveness of our life can only fit into a purposeful universe that is cre-
ated by an omnipotent and loving creator. It can surely not fit into a Russellian hostile universe that
is purposeless and headed toward extinction, nor into a naturalistic universe caught in causal con-
nections that exclude every form of purposefulness.
Second, the requirement of the afterlife is associated with the existence of a soul, which is the
locus of pleasure. Goetz underlines indeed that although there are various things good for the
human body, such as food or water, they are only instrumentally good. A human body has no good
of its own, in the sense in which the soul has pleasure as its own good (Goetz 2012, 91). One can
wonder whether beatitude (the heavenly bliss) is something different from pleasure, but since we do
not have full information about the outlook of the afterlife, we can at least take pleasure as an
approximation of the happiness we will experience after death.
The everlasting happiness in the afterlife also provides a good response to the problem of
evil. God has created us with the purpose of experiencing perfect pleasure, which must be
deserved by each person according to her moral actions. Thus God must grant each person
the freedom of choice that implicitly allows her to commit evil. If God would grant perfect hap-
piness to everybody independently of personal choice, evil would defeat good, for there would
be no point in being moral in that case: “it is the intrinsic good of justice that requires the pro-
vision of free will. The experience of perfect happiness is a great good and, because it is, it can-
not simply be bestowed on created persons independent of considerations of justice” (Goetz
2012, 164). Conceiving the purpose of life on the basis of the psychological state of pleasure
also indicates that the problem of evil must be addressed from an individual perspective, in
terms of the final good of created persons, rather than in terms of possible worlds and their
goods (Goetz 2012, 163).
REVIEW
The theistic answers to the question of “the meaning of life” enjoy a clear advantage over their nat-
uralistic opponents. Only theism provides a full account of the human existence. It addresses the
cosmic question of our purpose in the great scheme of things, which naturalism usually leaves unan-
swered. Most theists agree that one can find some sort of meaning in life also without any relation
Meaning: Theism
with God. Personal satisfaction, achievements, and love might well fulfill the task. However, their
meaning is not the ultimate meaning that the question of the meaning of life seeks when it uses
the preposition “the.” Death, evil, and human vulnerability challenge us to find a radical under-
standing of the way in which human limits can be confronted.
Theists offer several solutions. Above we outlined the most significant among them, especially
focusing on pluralist and singularist theism. Before that, we addressed the challenge of the surrogate
theory, according to which there is no such thing as “the meaning of life.” It is rather a ghostly ques-
tion that conveys human unease (Sigmund Freud) or modern nihilism and subjectivism (Hochschild).
Theists who write on the meaning of life see the novelty of this expression in a positive light. Theistic
pluralism recognizes that “the meaning of life” can collect independent meanings that go under that
expression. However, it also shows that God provides the deepest kind of meaning. Singular theism
reinforces this last idea by arguing for a single meaning of life in relation to God. It contains four
kinds of theories: (1) metaphysical singularism (Nozick, Cottingham, McGrath, Nancy, Waghorn)
(2) narrative singularism (Seachris), (3) experiential singularism (Sauter, Tolstoy), and (4) subjective
singularism (Goetz).
Yet a few issues remain. First, the relation between meaning and truth needs further clarifica-
tion. Theism consistently shows how a relation with the creator of the universe grants deep and ulti-
mate meaning to our life. This relation is conceived within a specific epistemological framework.
The evidence that shows the truth of our knowledge of life’s meaning has various sources, but none
of them pertains to deductive reasoning. On the contrary, among them are the analogical knowledge
(Nozick), the receptivity that requires a personal transformation (Cottingham), the practice of sense
(Nancy), and the example of a faithful community (Tolstoy). The theistic scholarship on the mean-
ing of life is not deductive and thus seems inadequate to prove the existence of God on which it
largely relies. However, this scholarship might not need any deduction. Perhaps tools such as anal-
ogy, receptivity, and example could well replace any classic proof of God’s existence. In this sense,
the very question “what is the meaning of life?” already offers a new way to speak about God’s exis-
tence. Yet this line of inquiry has not been fully developed.
The second issue pertains to the narrative aspect of life’s meaning. Narrative singularism under-
stands human life in general in conformity with the great narrative accounts provided by religions.
How do we make the shift from the general narrative of human life to the individual narrative of
specific lives? What tools can enable us to interpret our own life narrative? What role does divine
providence play in this individual narrative? These questions remain unanswered.
Bibliography
Cottingham, John. On the Meaning of Life. London: Routledge, 2003. Goetz, Stewart. The Purpose of Life. A Theistic Perspective. London:
Continuum, 2012.
Cottingham, John. “Theism and Meaning in Life.” European Journal for the
Philosophy of Religion 8, no. 2 (2016): 47–58. Hochschild, Joshua. “John Paul II’s Gamble with ‘the Meaning of Life.’”
Unpublished manuscript. 2017.
Figal, Günter. “Sinn. Zur Bedeutung eines philosophischen Schlüsselsbe-
griffs.” In Anfang und Grenzen des Sinns, edited by Brigitte Hilmer, John Paul II. Fides et ratio. Encyclical Letter on the Relationship Between
Georg Lohmann, and Tilo Wesche, 83–92. Weilerswist, Germany: Faith and Reason. Boston: Pauline, 1998.
Velbrück Wissenschaft, 2006.
Marquard, Odo. In Defense of the Accidental: Philosophical Studies. Translated
Flew, Antony. “What Does It Mean to Ask: ‘What Is the Meaning of Life?’” by Robert M. Wallace. New York: Oxford University Press, 1991.
In The Presumption of Atheism and Other Philosophical Essays on God,
Freedom and Immortality, 155–167. London: Elek Books, 1976. Mawson, T. J. God and the Meanings of Life: What God Could and Couldn’t
Do to Make Our Lives More Meaningful. London: Bloomsbury, 2016.
Frankfurt, Harry. “The Importance of What We Care About.” In
The Importance of What We Care About, by Harry Frankfurt, 80–94. McGrath, Alister. Dawkins’ God: Genes, Memes, and the Meaning of Life.
New York: Cambridge University Press, 1998. Oxford: Blackwell, 2004.
Frankl, Viktor E. Man’s Search for Meaning. Translated by Ilse Lasch. McGrath, Alister. Surprised by Meaning: Science, Faith, and How We Make
Boston: Beacon Press, 2006. Sense of Things. Lousiville, KY: Westminster John Knox Press, 2011.
Freud, Sigmund. Letters of Sigmund Freud. Edited by Ernst L. Freud. McGrath, Alister. If I Had Lunch with C. S. Lewis: Exploring the Ideas of C. S.
Translated by Tania and James Stern. New York: Basic Books, 1960. Lewis on the Meaning of Life. Carol Stream, IL: Tyndale House, 2014.

Meaning: Theism
McGrath, Alister. The Big Question: Why We Can’t Stop Talking about Sauter, Gerhard. The Question of Meaning: A Theological and Philosophical
Science, Faith, and God. New York: St. Martin’s Press, 2015. Orientation. Edited and translated by Geoffrey W. Bromiley. Grand
Rapids, MI: Eerdmans, 1995.
Metz, Thaddeus. Meaning in Life: An Analytic Study. Oxford: Oxford
University Press, 2013. Seachris, Joshua W. “The Meaning of Life as Narrative: A New Proposal for
Interpreting Philosophy’s ‘Primary’ Question.” Philo 12, no. 1 (2009): 5–23.
Nancy, Jean-Luc. The Sense of the World. Translated by Jeffrey S. Librett.
Minneapolis: University of Minnesota Press, 1997. Seachris, Joshua W. “Death, Futility, and the Proleptic Power of Narrative
Ending.” In Exploring the Meaning of Life. An Anthology and Guide,
Nancy, Jean-Luc. Dis-Enclosure: The Deconstruction of Christianity. Trans-
edited by Joshua W. Seachris, 461–480. Oxford: Wiley-Blackwell, 2013.
lated by Bettina Bergo, Gabriel Malenfant, and Michael B. Smith. New
York: Fordham University Press, 2008. Seachris, Joshua W. “The Meaning of Life and Scripture’s Redemptive-
Historical Narrative: Illuminating Convergences.” In God and Meaning:
Nancy, Jean-Luc. Adoration: The Deconstruction of Christianity II. Translated New Essays, edited by Joshua W. Seachris and Stewart Goetz, 13–36.
by John McKeane. New York: Fordham University Press, 2013. New York: Bloomsbury, 2016.
Nozick, Robert. Philosophical Explanations. Cambridge, MA: Belknap Press Tolstoy, Leo. A Confession. In A Confession and Other Religious Writings.
of Harvard University Press, 1981. Translated by Jane Kentish. London: Penguin, 1987.
Nozick, Robert. The Examined Life: Philosophical Meditations. New York: Velleman, J. David. “Narrative Explanation.” Philosophical Review 112, no. 1
Simon and Schuster, 1989. (2003): 1–25.
Oliva, Mirela. “Hermeneutics and the Meaning of Life.” Epoché: A Journal Waghorn, Nicholas. Nothingness and the Meaning of Life: Philosophical
for the History of Philosophy 22, no. 2 (2018): 523–539. Approaches to Ultimate Meaning through Nothing and Reflexivity. London:
Sartre, Jean-Paul. Existentialism Is a Humanism. Edited by John Kulka. Bloomsbury, 2014.
Translated by Carol Macomber. New Haven, CT: Yale University Press, Wolf, Susan. Meaning in Life and Why It Matters. Princeton, NJ: Princeton
TOPIC 15
Meaning: Atheism
Thaddeus Metz
University of Johannesburg, South Africa
This chapter explores what it means to live a meaningful life. It also evaluates the idea that belief in God is
required for such a life.
DEBATES ABOUT MEANING IN LIFE
What is it for a life to be meaningful? If one wants a meaningful life, how should one live in order
to make it so? In particular, must there be a God to whom one relates in a particular way, say, by
having faith that He exists? Why might one think that meaning is impossible without God, or, con-
versely, that meaning could come in a world in which neither God nor anything else spiritual exists?
This chapter is devoted to answering such questions.
The section after this one begins by pointing out that English-speaking philosophers did not
address questions of life’s meaning for a long time because they thought that the questions did
not make sense. After providing some reason to reject that perspective and instead to think that
these questions are intelligible, the section aims to articulate what they mean. It contends that when
speaking about meaning in life, we are not necessarily discussing whether it is happy for the one
whose life it is. Meaning often brings well-being in its wake, but that is not always the case. Simi-
larly, although a moral life is probably meaningful to some degree, there are intuitively many ways
to live meaningfully that do not involve acting rightly. Instead, the terms meaningfulness, happiness,
and rightness connote different goods that a life can exhibit to varying degrees, where presumably the
best life, or at least a really good one, would have enough of all three.
The third section of this chapter considers which facets of our lives are capable of exhibiting
meaning, considered as a value distinct from well-being and morality. Is a meaningful life simply
one with lots of meaningful parts, or, in contrast, can life as a whole be meaningful in a way that
is not merely the sum of its parts? This chapter defends an affirmative answer to this latter question,
contending that meaning can be found either in the parts of a life such as particular actions or in the
entire life as a narrative or “life story.”
Supposing, then, that meaning can be present in both the parts of one’s life and one’s life as a
patterned whole, how should one live so as to have enough of both sorts? Would God be the sole
possible source of meaning, the only person who could ground an important purpose in our lives?
If God did not exist, could meaning come from anywhere else? For example, could human beings
simply create their own meaning by achieving whichever purposes they happen to have? This chap-
ter’s fourth section seeks to explain why most English-speaking philosophers since at least the 1990s
believe that, although fulfilling a purpose from God is not necessary for a meaningful life, fulfilling
just any of our own purposes would fail to be sufficient. For most who have put their minds to the
matter, a meaningful life probably comes from “objective” factors in our world as known by science,
such as helping others, making artworks, or acquiring knowledge, and ideally in a way that improves
over time.
507
Meaning: Atheism
The fifth and final section of this chapter considers not

KEY CONCEPTS whether God must exist for our lives to be meaningful, but
rather whether belief that He does exist would help make our
lives meaningful. Even if the existence of God were not neces-
Faith
sary for our lives to be meaningful, and even if there were no
God’s Purpose conclusive evidence that God exists, it might be that faith in
Meaningful Lives Him would be useful. Those who have faith that there is an
all-powerful, benevolent force who will grant us eternal paradise
Meaningful Lives without God
are likely more resilient and able to face challenges in their lives.
Naturalism The chapter concludes by spelling out how an atheist, one who
Objectivism does not believe in God, could achieve some reasonable degree
Part-Life Meaning of meaning, suggesting a respect in which faith in God would
in fact reduce meaning in one’s life.
Reason
Subjectivism
Supernaturalism MEANING, WELL-BEING, AND MORALITY
Whole-Life Meaning For much of the twentieth century, English-speaking philosophers
did not put much thought into what would make a life meaning-
ful, because they thought that the question does not make sense.
After canvassing the major reasons for denying that it is sensible to inquire into life’s meaning and
explaining why these reasons are no longer commonly accepted, this section sketches some of the core
elements of what English-speaking philosophers, at least, tend to mean by talk of “life’s meaning” and
how it relates to goods such as well-being and morality.
“LIFE’S MEANING” AS MEANINGLESS?

Some twentieth-century philosophers inspired by logical positivism (for example, Ayer [1947] 2000)
argued that the only questions that make sense are ones that can be evaluated with evidence from
our five senses. It can make sense to ask how long it takes for light to reach us from the sun, since
we can (with some effort) measure that with our eyes. In contrast, since we cannot see, smell, hear,
taste, or touch whether meaning exists in our lives and what constitutes it, these questions lack gen-
uine answers. Other thinkers at that time thought that talk about “meaning” is essentially linguistic,
that is, that only things such as words can be meaningful or not. As the novelist and onetime
philosophy professor William H. Gass remarked, “There is no meaning to life. It is not a sign”
(quoted in Moorhead 1988, 70).
These abstract considerations about the nature of language kept many philosophers in the
English-speaking world from making any claims about meaning in life. Toward the end of the
twentieth century, however, most philosophers revised their views of language and hence of whether
it is worth spending time thinking about what, if anything, makes life meaningful.
First off, many questions that intuitively make sense and are worth seeking to answer cannot be
evaluated with sense-data. For example, we cannot literally see one thing causing another; we can
see only one thing appearing in time after another, which is not the same thing. Yet it surely makes
sense to ask whether one thing causes another. For another example, we cannot see democracy, but
it is reasonable to ask whether democracy exists and what tends to promote it. Similarly, even if we
cannot see or otherwise immediately sense meaning in life, it could be intelligible to inquire into it.
That can be true even if life is not a word or essentially a symbol of any sort. It appears to many
now that the word meaning is a homonym, that is, people use this word with the same spelling and
same pronunciation to mean different things. There is the meaning of a word, on the one hand, and
the meaning of a life, on the other (and even meaning in the sense of “smoke means fire”), where
meaning in these contexts means different things; a life’s meaning does not have to be the same
thing as a word’s meaning. Philosophers also came to recognize that the question of life’s meaning
could be posed without the word meaning at all! The same question appears expressed by asking
whether an existence is significant or important, for example.
Meaning: Atheism
UNCONTESTED ELEMENTS OF LIFE’S MEANING

One project that philosophers have been undertaking since the late twentieth century is working to
clarify what exactly we are asking when posing the question of how life could be meaningful (or sig-
nificant or important) (see the contributions to Seachris 2012, part 1). Although there remains
debate about that, this section sketches some of the less contested claims.
First, most of us are inquiring into some facet of our lives as individuals and not so much about
why the human species exists or, even broader, what might be the point ofthe universe. No doubt
we sometimes think about the latter, more cosmic questions (see Seachris 2009). When we do,
however, we are often interested ultimately in how they bear, if at all, on answering the first ques-
tion, about whether our own lives are important or how to make them more so. Sometimes philo-
sophers distinguish between the meaning “of” life, which is about the point of the human race as a
whole, and the meaning “in” a life, about whether a given one of us has some importance. Most of
the contemporary debate has focused on the latter issue.
Second, most of us are interested in what an individual’s life could exhibit that is valuable. We
want to know how to live in ways that would impart more of something that is good for its own
sake and not merely valuable as a means to something else, let alone not valuable in any respect.
Meaningfulness is intuitively something that is supposed to provide a good reason for living one
way rather than another, which would be well explained by meaning having some noninstrumental
value.
Third, most of us believe that meaning in life comes in degrees. A given one of us can have
more or less of it in our lives, depending on the choices we make. Similarly, one person can have
a more meaningful life than someone else. Notice that when contending that one person’s life is
more meaningful, or even more important, than another person’s, one is not necessarily suggesting
anything from a moral point of view. That is, one should not be understood to be saying that the
person with a less important life matters less when it comes to the just distribution of resources,
for example, that he or she should be sacrificed for the sake of others. Probably for most philoso-
phers, people can have differential degrees of meaning in their lives while having the same value
from the standpoint of morality and justice. In fact, it is perfectly coherent to suggest that, because
of people’s inalienable equal dignity, those who have less meaning in their lives are entitled a greater
share of resources than those who already have more meaning (Metz 2013, 4).
Fourth, there are clear examples of when this variable final value is present in an individual life
and, conversely, when it is absent. Two famous thought experiments are meant to illustrate an
absence of meaning in life. First, there is the case of Sisyphus (discussed by Camus [1942] 1955),
imagined to have been doomed by the gods to roll a large rock up a hill forever. Each time the rock
reaches the top, it rolls back down, with Sisyphus having to roll it back up, and so on, into eternity.
Second, there is the case of the experience machine (discussed by Nozick 1974, 42–45), a sophisti-
cated virtual reality device that enables a person to have any kind and number of pleasant feelings
that he or she would like. Although one might be excited, blissful, or content when spending time
in the experience machine, virtually no philosophers believe that one would be living meaningfully
when inside it. Less fantastic and more realistic examples of a lack of meaning include: killing one’s
healthy, innocent spouse for the insurance money; being subjected to racial or gender discrimina-
tion; chewing gum; and scratching an itch. The latter two might be worth doing, but not because
they would make your existence significant.
Additional quintessential examples, this time of meaning being present, often involve “big
names.” Many think of the likes of Albert Einstein, Martin Luther King Jr., Mother Teresa, Toni
Morrison, and Vincent van Gogh, at least in our limited familiarity with them. These are often
deemed to be people who have made major accomplishments. They are taken to have made reveal-
ing discoveries about our world or helped lots of people or created powerful artworks. However, few
philosophers believe that one has to be quite so “great” in order to have some real meaning in one’s
life, which could intuitively come from getting a university education, rearing children with love, or
tending a garden.
The point of these cases is to suggest that if someone thought that Sisyphus’s life would be
meaningful, or that Einstein’s life was not meaningful, then that person would not be using the

Meaning: Atheism
word meaningful in the same way as most of us, or at least a large majority of twenty-first century
English-speaking philosophers. Debate about meaning in life presumes that, by definition, it is
about a final value that can be realized to varying degrees in a human person’s life that is absent
in the case of Sisyphus but present in the case of Einstein.
Beyond this common ground, philosophers disagree about what else talk of “life’s meaning”
might essentially mean. Additional suggestions include that it connotes the ideas of achieving purposes
that are particularly worth striving for, connecting with goods beyond oneself or at least one’s animal
nature, making a positive difference to the world, doing what merits great esteem or admiration, or
living in a way that would make for a good biography to read. These suggestions are much more con-
troversial, however. For example, life in a world where the wicked receive benefits and the upright
undergo burdens is probably less meaningful than one where the opposite is true, but, as it is not
within your power to make the wicked undergo burdens, it is not a purpose you can achieve or some-
thing in which you could sensibly take pride in having achieved. The safest suggestion, at the current
stage of reflection involving lots of disagreement among philosophers, is that these other properties are
typical of meaning in life, even if they have not been shown to be invariably true of it by definition.
LIFE’S MEANING CONTRASTED WITH RELATED GOODS

King and Mother Teresa performed moral actions that went beyond the call of duty, whereas nei-
ther the fate of Sisyphus nor a life spent in the experience machine involves having done anything
morally notable. One might suspect, therefore, that talk of “meaning in life” is just another way
of speaking about how we ought to treat others in a morally ideal way.
Almost no philosophers, however, believe that the meaningful and the moral are equivalent.
Most in the field accept that there are intuitively cases of actions that confer meaning on people’s
lives but not in virtue of morality. Some of these are cases in which there is nothing morally posi-
tive present and yet there appears to be some meaning. Consider, for instance, the cases of Mor-
rison and van Gogh, where it appears that it was their creativity that made their lives particularly
meaningful.
In reply, one might suggest that their creativity ultimately had a moral dimension, in that it was
good for many people. It improved many people’s awareness of race in the case of Morrison’s books
and pleased many people in the case of van Gogh’s paintings.
Although it is plausible to think the positive influence on others greatly contributed to the
meaning of these two lives, few philosophers would say that this means that meaning is reducible
to morality. For one, part of the meaningful influence on others is not having made them better
off or better people but just having received recognition for a significant accomplishment. For
another, imagine that no one had ever read Morrison’s books or seen van Gogh’s paintings, that
they were locked up never to see the light of day after they had been composed. Although there
would have been less meaning in Morrison’s and van Gogh’s lives in such a case, arguably some
meaning would still have been present, simply in virtue of their having created art objects that
deserve to have benefited many others and to have been recognized by them.
Other cases, perhaps stronger ones, that lead philosophers to doubt that what is meaningful is
always what is moral are those in which it appears at least possible for meaning to come in the wake
of immorality. Immorality, in this context, means a violation of moral prohibitions against harming
or degrading others, not merely a failure to live up to moral prescriptions to go out of one’s way to
help others as much as one can. To illustrate how such immorality might be meaning-conferring, or
at least facilitate meaning, consider the case of another painter, Paul Gauguin, that is often discussed
by philosophers. Gauguin is reputed to have abandoned his wife and children so that he could move
to Polynesian islands and concentrate strictly on his painting. It was wrong both to break his vow to
his wife and not to live up to his responsibility to his children, and yet it seems plausible to think
that Gauguin’s life acquired some substantial meaning by virtue of his artworks, which presumably
would not have been created without the distance and isolation.
Just as for most philosophers it is at least thinkable that nonmoral or even some immoral
behavior could be meaningful, so it is possible that nonhappy or even unhappy states could be,
too. Although the best sort of meaningful conditions might be those in which we are fully absorbed
Meaning: Atheism
by what we are doing and enjoy doing it, it is unlikely that such mental states are essential to mean-
ing in life, at least not by definition of “meaning in life.” When you think of Mother Teresa’s life
being meaningful, at least in our limited understanding of it, do you think of her being happy,
understood principally in terms of having pleasant experiences? Chances are that you instead attend
to the idea that she made great sacrifices for the sake of other people’s well-being, not that she did
what made herself well off.
Furthermore, if you now explicitly imagine that Mother Teresa was downright unhappy chang-
ing bandages, healing wounds, cleaning up human waste, and trying to comfort those with fatal ill-
nesses, it will likely still appear to you that her life could have been meaningful to some degree. You
might even wonder whether Mother Teresa evincing a cheery disposition while telling others they
will soon die would have reduced the meaningfulness of her engagements with them. Sometimes
negative emotions are appropriate, and confer some meaning on a life, so you might think.
Summing up the last two sections, it appears true that meaning in life essentially involves some-
thing good for its own sake that a human person’s life can exhibit to a variable degree and that was
clearly present to a substantial extent in the actual lives of Einstein, King, and van Gogh and clearly
absent in the hypothetical lives of Sisyphus and of one spent in an experience machine. When it
comes to additional conditions, such as achieving purposes, doing what is morally right, and being
well off, these are probably often involved with meaning in life, but there is no consensus at this
stage that they are necessarily so, at least not by definition.
WHICH DIMENSIONS OF LIFE HAVE MEANING?
Up to now, this chapter has been vague in speaking about “life” having meaning in it. This way of
speaking could mean one of three different things that should be distinguished.
First, one might be thinking of a particular part of a life, perhaps a stage such as adolescence, or
even just a particular action, say, of helping an old lady cross the street.
Second, one may have in mind the life as a biographical whole. One might be inquiring into
whether the all-encompassing pattern of the entire life, from birth to death and even to posthumous
influences on others, was significant.
Third, one could be asking whether a life has enough of the first two sorts of meaning to count
as a meaningful one on balance, supposing that meaning could come from both part-life and whole-
life dimensions. To speak of a “meaningless” life in this context would mean that there was either
literally no episodic or patterned meaning in the life, or just not enough to warrant the title of
“meaningful.” It would be like saying a life was “unhappy”; there were presumably some happy per-
iods in it, but there were not enough to deserve the “happy” label overall, on weighing them all up
relative to the unhappy ones.
MEANING IN LIFE VIA PART AND WHOLE

There remains disagreement among philosophers about whether meaning can in fact come from
both part-life and whole-life dimensions. There are some pure part-lifers who maintain that, if
you want to know whether someone’s life has been meaningful on balance, all there is to do is to
add up the particular meaningful actions, decisions, and other episodic conditions. There is nothing
about the pattern of a life story, from this perspective, that can in principle exhibit meaning.
In contrast, there are also pure whole-lifers who contend the opposite. For them, if you want to
know whether someone’s life has been meaningful on balance, all there is to do is to consider a per-
son’s actions, decisions, and other conditions as they have formed a narrative that covers the entire
course of his or her life. There is nothing about the parts of a life in themselves, from this perspec-
tive, that can in principle exhibit meaning.
This section does not address arguments for either of these extreme views (see Metz 2013, 37–
51). Instead, it provides some reason to think that a middle view is probably correct, that is, that
meaning can inhere in both the parts of a life and the life as a patterned whole.

Meaning: Atheism
The main evidence that the parts of a life can be meaningful is that we routinely judge episodes
in this way. Imagine that, because of your education and determination, you were the first to meet a
species of intelligent aliens from another planet. That would be a meaningful encounter for you,
and we are inclined to judge it meaningful without knowing about earlier or later features of your
life. More strongly, we would continue to judge it to have been meaningful even if we learned that
you afterward became a drunkard who did nothing else with your life.
The second thing to note about this latter case, however, is that another sort of meaning is
intuitively missing from it, namely, a whole-life sort. A life is somewhat more meaningful for pro-
gressing over time, moving from one accomplishment to other, even greater ones. A life story is
somewhat less meaningful in virtue of undergoing a decline, of having done something great when
young and then wasting away into trivialities.
Additional sorts of whole-life considerations that undercut meaning are randomness and repeti-
tion. A life without any pattern at all appears to be missing a kind of meaning, while a life with
parts that are too similar to each other too often, for example, a life in prison or on an assembly line,
also appears to be missing some meaningfulness (for these and other considerations, see Kauppinen
2012). These are poor life stories and hence lack some meaning, it appears.
It is tempting for a pure part-lifer to suggest that what seems to be a meaningful pattern is
merely a sum of meaningful parts, but that is implausible, particularly on considering redemption.
Redemption is a matter of making something good come from the bad. Think, for instance, of
someone’s having been an abuser (of drugs, of people), changing his or her character, and then
teaching others how to stop being abusers. Here, the pattern includes a clear lack of meaning at
the start, and it appears that some whole-life meaning is enhanced upon having made some part-life
meaning emerge from that bad initial state.
A MEANINGFUL LIFE ON BALANCE

Supposing, then, that there are two ways in which a life can exhibit meaning, either episodically in
its parts or in the life as a biographical whole, if one wants to know whether a person’s life has been
meaningful on balance, one must find a way of weighing up both sorts (perhaps while subtracting
harmful or degrading behaviors that may have reduced meaning). It is natural to want to know
how much meaning is enough for a life to count as meaningful on balance, but that is hard to spec-
ify, even in principle.
One philosopher has suggested that a meaningful life on balance is one that has more meaning
in it than average: “Ordinarily, the standard we employ is the average of the kind. We call a man
and a tree tall if they are well above the average of their kind.… The same principles must apply
to judging lives” (Baier [1957] 2000, 127).
One serious problem with this principle, however, is that it entails that no one’s life would be
meaningful if everyone’s lives had the exact same, extremely high amount of meaning in them; since
no one would be above average, no one’s life would count as meaningful on balance by the present
standard. In the real world, it is of course unlikely that everyone would in fact have the same
amount of meaning. The point is that, in a possible world in which everyone did have the exact
same, extremely high amount of meaning, the standard of above average would counterintuitively
entail that no one’s life is meaningful on balance, meaning that it is probably the incorrect standard
even for the actual world.
Another problem with the above-average standard for whether a life is meaningful on balance is
that we often appraise lives not by appeal to the average but to something more absolute, in the
sense of not relative to the ways people have actually lived. As another philosopher has remarked
of a parallel issue, “Normality seems to have nothing to do with it, for the fact that we will all inev-
itably die in a few score years cannot by itself imply that it would not be good to live longer. Sup-
pose that we were all inevitably going to die in agony—physical agony lasting six months. Would
inevitability make that prospect any less unpleasant?” (Nagel [1970] 1979, 10). By analogy, the
amount of meaning in the average human being’s life might have been very low up to now, making
it a poor standard by which to evaluate whether someone’s life counts as meaningful.
Meaning: Atheism
As with many other considerations about life’s meaning, English-speaking philosophers are con-
tinuing to debate the present topic about how much part-life and whole-life meaning (supposing
both exist) a life must have in order to count as meaningful on balance. Although few accept the
above-average standard, and some alternatives have been proposed (canvassed in Metz 2013, 146–
158), there is no consensus about how to judge how much meaning is enough. Notice that this
issue abstracts from contested considerations of precisely what might constitute meaning, for exam-
ple, whether meaning essentially comes from God or not, to which this chapter now turns.
THE SOURCE OF MEANING I: WHAT DOES GOD HAVE TO DO

WITH IT?
So far, little of this chapter has directly addressed the question of what (if anything) makes a per-
son’s life meaningful. Instead, it has focused on more abstract questions such as what it means to
talk about “life’s meaning,” how meaning as a value differs from other goods, and what in a life,
its parts or its entire pattern, can exhibit meaning. Now, in this section and the next one, the chap-
ter will take up the question that most people have when they think about meaning: What should I
do in order to make my life more meaningful?
This section critically discusses three major answers: fulfill a purpose God has assigned to you;
achieve whichever goals you think are most worth achieving; and live in a way that includes many
objective goods such as wisdom, love, and beauty. Although pros and cons of all three positions
are considered, the chapter tentatively concludes in favor of the last, objective view.
A GOD-BASED APPROACH TO MEANING

In the Western tradition of philosophy, the biggest split between philosophers of meaning in life is
between supernaturalists and naturalists. The former believe that one’s life would have no meaning,
or at least not be meaningful on balance, if there were no spiritual realm that one related to in the
right way. A spiritual realm, for most of Western philosophy, is understood to be one in which
there are persons that live beyond (our) space and time and that are not composed of subatomic
particles. These persons are typically thought to include God, a perfect being who is the source of
the physical world, and souls, those who once lived in the physical world but have survived the
deaths of their bodies to live on in a Heaven or Hell. According to supernaturalism, if there is no
God and there are no souls, or if these exist but one fails to orient one’s life in the right way toward
them, then one’s life is not meaningful. Naturalism is the view that supernaturalism is incorrect and
that, even if there is no spiritual realm, and only a physical one exists, a meaningful life on balance
is still possible.
Supernaturalism and Theism. Before considering precisely how a spiritual realm might be the
sole source of meaning in our lives, notice the difference between supernaturalism and theism.
Supernaturalism is a claim about what would make our lives meaningful: namely, that only relating
to a spiritual realm in a certain way could do so. In contrast, theism is a claim about what exists:
that there is enough evidence to know that a spiritual realm is real.
Although in practice many supernaturalists are theists (and vice versa), there is no incoherence,
or at least no logical contradiction, in a supernaturalist being an atheist. Such a combination of
views would mean that a person believes, for instance, that God is necessary for our lives to be
meaningful, but that God does not exist and hence that our lives are meaningless. Albert Camus
appears to have held this position in his famous book The Myth of Sisyphus, as did Leo Tolstoy, at
least for a time, as he reports in My Confession.
Conversely, one could also be a naturalist who is a theist. That is, one might, without inconsis-
tency, hold that although the evidence indicates that God does exist, relating to Him is not neces-
sary to live meaningfully. Some naturalists believe that God would in fact reduce the meaning in
our lives, while others think He could enhance the meaning (without being essential for enough
of it), and still others believe He would make no real difference either way.

Meaning: Atheism
This chapter does not address any metaphysical questions about the nature of reality. It focuses
strictly on questions of value, of what is involved in a life’s being meaningful or not. For now, the
issue is what it might be about God that makes Him (if you are a theist) or would make Him (if
you are an atheist) essential for our lives to be meaningful.
God’s Purpose. The most influential form of supernaturalism about meaning in life has been that
one’s life is more meaningful, the more one fulfills a purpose that God has assigned. If there were no
God, then one’s life would have zero meaning, or at least not enough meaning to count as having a
meaningful life on balance. Or if there were a God but He did not give one a purpose, or one failed
to fulfill it, then, again, one’s life would be meaningless.
The standard version of this view is not that having been created by God for a reason is suffi-
cient for our lives to be meaningful, for then everyone’s life would be equally meaningful. Instead,
the usual idea is that people’s lives vary in the degree to which they are meaningful, in virtue of
the extent to which they have in fact freely done what God intended them to do.
Notice that it is in principle possible for an atheist to live meaningfully by the supernaturalist
account, as stated so far. Even if a person did not believe that God exists, he or she could do what
God wants him or her to do with his or her life, depending on what that might be. For instance, if
God wanted one to love one’s neighbor, a person could do that without believing in God. If God’s
purpose for one, however, included believing in God, then, only in that particular case, an atheist in
principle could not live meaningfully.
This issue brings out the point that there are different accounts of what God’s purpose for us
might be. Some philosophers think that it would be to enter into a relationship with God, while
others maintain that it would be to do good works with respect to humans and perhaps animals.
Some believe that God would assign the same purpose to all human beings, while others think that
God would tailor particular purposes to us as individuals. Finally, there is disagreement about
whether fulfilling the purpose, or receiving some share of happiness for having done so, would come
in an afterlife, or whether meaning would come simply by doing God’s bidding while on earth, with
no expectation of surviving the inevitable death of one’s body.
This chapter cannot address all the variations of the view that meaning comes solely or princi-
pally from fulfilling God’s purpose and instead focuses on the general expression of it. The core idea
that a life is more meaningful the more it does what God intends for it is powerful. Who has the
authority to determine what you should aim for in life? A perfect person, and a perfect person alone.
Where does the higher value of meaning come from, a value that is beyond the animal world of
desires and pleasures? From a holy being in a spiritual realm. What explains why some people have
more meaning in their lives than others? Some have done more of what God wants them to do.
What accounts for the whole-life dimension of meaning? Realizing God’s purpose in increasingly
devoted, sophisticated, and impactful ways over time.
Problems with God’s Purpose. The initial objection to thinking that fulfilling God’s purpose is
necessary and sufficient for one’s life to acquire meaning concerns the content of the purpose.
Above we imagined that the gods had assigned to Sisyphus the purpose of rolling a rock up a hill
forever. What if God were to assign us that purpose or something similar, such as serving as food
for intergalactic travelers (see Nozick [1981] 2004, 74–76)? It is hard to accept that these kinds
of purposes would be enough to confer meaning on our lives upon realizing them. The natural reply
is to say that meaning would come from “beneficial,” “moral,” or otherwise intuitively “positive”
purposes that God has assigned. It is not just any purpose that is meaningful, but one from God
that is good.
However, the deep worry about this reply, a version of the “Euthyphro Problem,” is that it
appears that it is the desirable content of the purpose that explains the meaning, not the fact that
it has come from God. Why not hold that meaning would come from “beneficial,” “moral,” or oth-
erwise intuitively “positive” purposes, period? What would God’s having assigned these good pur-
poses add to anything? If God were to assign the purposes because they are good, then it is their
goodness that would be doing the work, not the fact that God assigned them.
Meaning: Atheism
At this point, the debate turns on whether there could be anything good without God. One
might be tempted to suggest that since God is the source of the entire universe, He is obviously also
the source of the values within it. However, this claim is not really relevant. Pointing out that God
exists and is the source of everything fundamental to the human condition is not the same as show-
ing that what is fundamental to the human condition, at least as concerns values, could not have
existed without God. It is the latter claim that needs to be motivated, at this stage of debate with
the naturalist.
One influential version of the latter claim (found in, e.g., Cottingham 2005, 49–54; Craig
[1994] 2012, 161–162, 165–167) is that only the realization of a moral purpose would confer
meaning on our lives and that only God could be the source of morality. Supposing that morality
is a system of rules that applies universally, that is, to all human persons, and is not relative to what
particular groups of people happen to believe about how to act, where else could morality come
from but God? It seems that some agent or other must have created the moral rules, and if human
beings are not the source of them, and could not be since they lack the authority, it must be God,
standing apart and above us all, in whom alone they could have originated.
As this chapter addresses meaning in the first instance, and not morality, readers should consult
other parts of this book for further reflection on whether this response is plausible. The key question
at this point is whether morality could exist without God. If so, then it appears that doing good
deeds in a world without God would be sufficient for meaning. If not, then good deeds would be
possible only in a world with God, making God necessary for meaning.
There is a second influential objection to the idea that fulfilling God’s purpose is essential for
meaning in life. Whereas the first one was about the content of the purpose, the current one is
instead about the fact that it would be assigned to us by someone else. Suppose you go up to a
stranger and say to him or her, “I am assigning you the following purpose” and then indicate what
it is you are expecting him or her to do (cf. Baier [1957] 2000, 104). Even if it were to do good
deeds, there would intuitively be something morally troublesome about having treated someone in
this way. A natural thing to say is that you would be treating this stranger merely as a means to your
ends. Similar things might go for God. Even if God assigned us a purpose with a moral content
(one that He alone could have created), it might seem to be degrading for Him to assign it to us.
This objection was initially voiced, it appears, in the 1940s by French existentialist Jean-Paul
Sartre, who, in a famous lecture titled “Existentialism Is a Humanism,” argued that for God to have
created us for a purpose would be for Him to treat us like a knife. Just as a human makes a knife for
the purpose of cutting things, so God would make us for some purpose or other. We, too, would be
treated as mere tools to God’s ends. Sartre said of his naturalist view of meaning in life, namely, that
meaning is merely a function of whichever choices a person has made for himself or herself (dis-
cussed in the next section), that it “alone is compatible with the dignity of man, it is the only
one which does not make man into an object” (Sartre [1946] 1956, 302). The key point is probably
that God would be immoral to create us for a purpose, which means that He would not in fact
assign us a purpose since He is by definition morally perfect. Or the point could be that even if it
would not be immoral for God to create us for a purpose, doing so would still degrade us and
thereby undercut our ability to obtain meaning from fulfilling the purpose.
There are a number of ways that supernaturalists and others have responded to this objection.
Some suggest that, even if being assigned a purpose were somewhat degrading of a person’s autonomy,
meaning could still come from realizing it (Walker 2002). Others contend that one of God’s purposes
for us would be for us to pursue our own, self-chosen purposes (Mawson 2016, 112). Still others con-
tend that, even if God’s purpose were not for each of us to “do our own thing” and instead were to
obey some specific rules, there still need not be anything degrading about it, depending on precisely
how God were to assign the purpose to us (Metz 2013, 99–104). If God made us with a certain
end in mind but did not prompt us to realize it by threatening, deceiving, tricking, exploiting, or oth-
erwise manipulating our wills, then He would not treat us merely as a means. Imagine that instead of
making a command backed up with the threat of Hell for disobedience, God were to make a request,
something like, “Would you please love one another, the reason for which I made you?” Philosophers
continue to debate whether these and other responses resolve the problem.

Meaning: Atheism
There is a third objection that, if successful, applies not just to the idea that meaning must
come from God, but to any supernaturalist theory. The objection is that meaning intuitively seems
possible despite atheism. Suppose, for the sake of argument, that there is no spiritual realm, and all
that exists is instead ultimately a material world best known by the scientific method. Despite our
presuming that there is neither a God nor a soul, it still seems as though some lives have meaning
relative to others. Surely, so this argument goes, a firefighter who routinely risks life and limb in
order to rescue people has meaning in his or her life, more than “The Blob” (i.e., couch potato),
someone who spends as much time as he can drinking beer while watching television sitcoms at
home alone (Wolf [2010] 2012, 306–307). If fulfilling God’s purpose, or some other spiritual con-
dition, were necessary for meaning in life, then we should have the intuition that the firefighter and
The Blob would have equally meaningless lives in a purely physical world, but we do not.
Although there are some supernaturalists who reject this argument and do have the intuition
that such a firefighter’s life would be meaningless (perhaps Craig [1994] 2012, and probably Cot-
tingham 2005), what most have done is to modify the thesis of supernaturalism. Since the late
twentieth century, those who believe that God is central to meaning have tended to admit that a
shallow or limited kind of meaning in life would be possible without Him, but to contend that only
He could ground a deeper or more substantial kind.
Sometimes the idea is that God would make possible an eternal afterlife that would be much
more meaningful than what is available to us in an eighty-or-so-year life span on Earth (Mawson
2016, 134–158). Other times, the thought is that only God could satisfy the strongest desires that
are typical of human nature (Seachris 2011). Still other times, the proposal is that participation in
God’s plan would give our lives “a cosmic significance … instead of a significance very limited in
time and space … (in that) the whole universe and all its inhabitants including ourselves were cre-
ated and are sustained by God who loves each one of us” (Swinburne 2016, 154).
As the debate stands, naturalists have yet to address these rationales thoroughly. One response
that merits consideration, however, is to suggest that if the prospective gains to meaning with God
would be much greater, then so would the prospective losses (Metz 2018). For example, if God’s
being pleased about one’s good deeds would produce a much greater significance for one’s existence
than would be possible in a world without Him, then presumably His displeasure at one’s bad deeds
would reduce the significance of one’s life to a correspondingly much greater degree. If Heaven would
produce much more meaning than is possible in an earthly life, then presumably Hell would reduce
much more meaning. It is not obvious which world, theist or atheist, would be reasonable to wish
for, given an interest in living a maximally meaningful life. The philosophical debate continues.
A SUBJECTIVE APPROACH TO MEANING

If we suppose, for the rest of this chapter, that a life can be meaningful to some real degree in the
absence of a spiritual realm, the next question is how to make it so. How should one live so as to
obtain meaning in life, if there is no God?
An initially tempting suggestion is that you should achieve your own goals. If fulfilling
God’s purpose is not necessary for your life to be meaningful, perhaps what would be sufficient
is fulfilling your own purposes (e.g., R. Taylor [1970] 2004; Frankfurt [1982] 1988). Presum-
ably the relevant purposes would not be ones forcibly thrust upon you. For instance, if you were
brainwashed into supporting the New York Yankees, that would not count. The relevant aims
must come from deep inside you, or at least be confirmed by something within you, not imposed
from the outside.
This approach to meaning is often called “subjective,” since it more or less says that what is
meaningful for people varies, depending on what each wants out of life. Some people want to
become zookeepers, and if they are able to do so, then their lives are meaningful. No meaning is
lost, however, to those who do not want to become zookeepers and are not.
Instead, meaningless lives, by the present account, are those in which people fail to get what-
ever they most want out of life. It does not matter what the reasons are for having failed to fulfill
one’s purposes. Perhaps it is one’s own fault for having made poor decisions, or perhaps it is others’
Meaning: Atheism
fault for having taken property that did not belong to them, leaving one financially poor and
thereby unable to get what one wants.
Sartre advocated a version of subjectivism, and recall that one argument for it that he advanced
is that this account of meaning does not reduce a person to the status of an artifact, something cre-
ated to serve a function. If meaning is something that an agent creates, then it must be created by
and for oneself, not by another (not even God), to avoid being treated like a mere tool.
An additional motivation for subjectivism is authenticity. Many have the sense that a meaning-
ful life is an authentic one, in the sense of a person being true to herself or himself in the choices she
or he makes. Many would say that choosing a field of study at university simply because your par-
ents want you to study it, when it has no resonance with your particular inclinations, is a poor can-
didate for conferring meaning on your life.
A third rationale for subjectivism is the intuition that many different meaningful lives are
possible. The most common versions of supernaturalism maintain that God would assign the
same purpose to everyone, say, to live morally or to glorify Him, which seems overly narrow as
a source of meaning. Supposing that a variety of meaningful lives are on offer, subjectivism is a
good explanation of why that is the case—meaning depends on what people want, and what
people want differs greatly.
Although these are strong arguments for subjectivism, few philosophers in the twenty-first century
think they are as powerful as the most prominent argument against it, which is that it entails absurd-
ities. Imagine that a person’s strongest desire were to maintain 3,732 hairs on his or her head (C. Tay-
lor 1992, 36). Or suppose that a person’s foremost goal were to torture and kill as many innocent
people as he or she could before getting caught. Then, by subjectivism, these lives would be particu-
larly meaningful insofar as they were able to achieve these aims. However, that is counterintuitive. The
natural thing to say that these are aims that one should get rid of, if one wants a meaningful life. It is
not just any desires or goals that can confer meaning, but instead ones with a certain kind of content.
Some subjectivists reply by maintaining that no people in fact have such aims. No one really
wants to keep 3,732 hairs on his or her head, they might say. However, it appears that some people
have really wanted to be serial killers. Furthermore, the deeper point is that whether anyone actually
has wanted these things is not really what is at stake. The objection can be formulated hypotheti-
cally: if someone most wanted to be a serial killer, then subjectivism entails that becoming one
would be meaningful, but it would not.
Other subjectivists are tempted to “bite the bullet,” to point out that maintaining 3,732 hairs
on one’s head would be “meaningful for” that person, or would be something he or she “finds
meaningful.” It is, however, trivially true that if a person really wants to keep 3,732 hairs on his
or her head and commits his or her life to that, then it will probably be “meaningful to” him or
her. The real question is whether a project’s being meaningful “to” or “for” a person is enough to
make it really meaningful, to be the sort of thing that is good for its own sake and worth pursuing.
Few contemporary philosophers believe that it is.
Yet the arguments for subjectivism still look good, especially the points about authenticity and
variation. A life does seem meaningful insofar as one is true to oneself in one’s choices, while a wide
array of different lives could be meaningful. Many recent theorists about meaning have therefore
been led to hold that subjective conditions are part but not all of the story about how to live mean-
ingfully in a purely physical world.
AN OBJECTIVE APPROACH TO MEANING

The dominant view among naturalists these days includes an “objective” component, which means
that meaning has a mind-independent dimension to it. Something is meaningful not merely because
you like it or think it is meaningful, but instead, or at least also, because of what it is “in itself.”
Some projects are by their nature meaningful and are such as to merit being desired or chosen if
they have not yet been.
One influential version of this view has been pithily summed up with the phrase “Meaning
arises when subjective attraction meets objective attractiveness” (Wolf 1997, 211). On the one

Meaning: Atheism
hand, for one’s life to be meaningful, one must engage in activities that are objectively attractive, that
is, that are worth doing because of the kinds of activities they are. Different theorists have different
views of what counts as objectively attractive, but a recurrent theme among contemporary Western
philosophers is the idea of using one’s intelligence in creative, moral, or theoretical ways. Objectively
attractive projects might include using one’s reason to make a work of art, to learn how to understand
a poem, to love a spouse or children, to go beyond the call of duty, to learn fundamental truths about
human nature or the world, or to come to understand oneself (cf. Nozick [1981] 2004, 81–82;
Wielenberg [2005] 2012). In addition, a life would arguably be more objectively attractive in virtue
of certain patterns over time: for example, if it progressed, making ever better artworks or forming ever
deeper relationships, rather than if the life were repetitive or substantially declined.
On the other hand, objectively attractive activities are not enough for meaning, by the present
view. In addition, a person must be subjectively attracted to them. Again, different theorists have
different views of what counts as subjective attraction. However, the cluster of attitudes that are
salient include liking what one is doing, freely choosing to do it, and thinking that it is worth doing.
Contra subjectivism, these attitudes are not sufficient for meaning, which instead arises when they
are directed toward activities that merit them. With subjectivism, though, if these attitudes are
not present, then no meaning is either.
This form of objectivism avoids the problems facing subjectivism, since presumably neither
maintaining a certain number of hairs on one’s head nor becoming a serial killer is objectively
attractive. Although these involve the use of reason in a way, the use of reason is not exercised posi-
tively or toward the right states of affairs, typified by morality, inquiry, and creativity. In addition,
this view can capture much of the motivation for subjectivism, insofar as there are a variety of objec-
tively attractive activities that one could perform, and being subjectively attracted to some of them
would be a matter of being true to oneself. One can see why this view has been so popular an
account of how to obtain meaning in the sort of world that an atheist thinks exists.
To criticize this objective view, one might wonder whether it is objective enough. Even if sub-
jective attraction would make a life more meaningful when oriented toward an objectively attractive
project, perhaps it is not necessary for meaning (Metz 2013, 183–184). Return to earlier reflection
on Mother Teresa, in particular the case in which we imagined that she was unhappy while helping
others. In that scenario, it appeared that she was forsaking some of her own wants and likes so that
others could have more of theirs. More generally, then, perhaps one source of meaning is precisely
not being subjectively attracted to certain objectively attractive activities, when that is expected to
enable other people to be subjectively attracted to them.
A further sort of objection to consider, one that applies to all forms of objectivism, is to doubt
that there are in fact any projects that are objectively attractive in the absence of God. As we have
seen, many supernaturalists believe that only God could be the source of moral rules that apply to
all human beings, and they would readily generalize the point, that is, that only God could be the
source of anything that a given human being has a good reason to do regardless of whether he or
she wants to do so (objective attractiveness), making God essential for meaning. Even some subjec-
tivists would make a similar point. For example, Sartre appeared to think that only God could create
universal standards for human beings, by making us for the same purpose. However, since Sartre
believes that God does not exist, he also believes that there are no such standards, and that only sub-
jectivism can account for the presence of meaning in our lives. Resolving the debate among super-
naturalists, subjectivists, and objectivists crucially depends on resolving meta-ethical debates about
the origin and status of morality and objective value in general.
THE SOURCE OF MEANING II: WHAT DOES BELIEF IN GOD HAVE TO DO WITH IT?
The preceding section considered whether God’s existence is necessary for our lives to be meaningful.
This section differs in that it addresses the bearing of belief in God’s existence on meaning.
Obviously, if a standard Christian account of meaning in life were true, then belief in God
would be necessary and sufficient for meaning. By this account, a meaningful life is one that fulfills
Meaning: Atheism
God’s purpose, where God’s purpose for us is to have faith in Him, and, more specifically, to accept
Jesus as our savior. Muslims, too, make belief in God central, with faith being one of the five pillars
of Islam, and the Qur ʾan is explicit that unbelievers will be doomed to Hell.
Instead of considering whether belief in God is essential for meeting Him in Heaven, this sec-
tion addresses the question of whether belief in God would be useful for living meaningfully on
earth. The question is not quite whether one should believe in God, all things considered. It is
instead whether considerations of meaning in life support belief in God. It could be that those rea-
sons are outweighed by other sorts, relating to, say, happiness or epistemic justification.
There is some reason to think that one should believe in God if one wants a meaningful life in
this world, a reason that does not depend on God actually existing. There is some evidence that the
most successful people on average harbor more illusions about what they can achieve than those
who are less successful. It can help to have an inflated sense of what one can do—for this misappre-
hension about oneself or the world one lives in tends to prompt one to get more done than one
would have otherwise. In the context of life’s meaning, supposing that one needs to undertake some
laborious projects in order to obtain it, having unjustifiably or inaccurately optimistic beliefs could
be useful, even essential in some cases, for getting them done.
Now, precisely which beliefs would be useful in this respect? Consider that if the world were as
atheists portray it, then the odds of achieving much meaning in life might seem poor. One influen-
tial supernaturalist speaks of the “futility and fragility” of acting in the universe (Cottingham 2003);
the apparent facts are that, for all we know of the physical world, anything we accomplish will not
leave much of a trace into the next generation, and that most of us are unlikely to accomplish much
with regard to morality, inquiry, or creativity in the first place. Given that some success is required
for meaning in life and that success is unlikely in a purely material world, it appears that it would
help motivate many of us, and thereby enable us to succeed, if we believed that there is a God
who is on our side and will help us achieve our meaningful ends.
Now if the ultimate nature of reality contains no bias towards the good as opposed to the
vicious, if there is nothing to support the hope that the good will ultimately triumph, if
essentially we are on our own … then at the very least it is hard to see how we can achieve the
necessary confidence and resolution to follow the path of goodness. (Cottingham 2003, 72)
In contrast, one is more likely to strive for a significant existence, if, despite the lack of evidence of
God’s existence, one believes that God is there to back one up. Perhaps the most motivating cluster
of beliefs would be that God created the world for certain purposes, that one is integral to the reali-
zation of His purposes, that God will assist one in achieving them, and that one will live happily
forever if one achieves them.
Note that if this argument is sound, it applies not merely to supernaturalists but to naturalists
as well. Recall that there are crucial differences between supernaturalism and theism and between
naturalism and atheism, such that there is no inconsistency in being a naturalist theist. Even those
who believe that God is not essential for or central to making one’s life meaningful could be more
likely to realize their goals on earth if they came to believe that there is a God who is helping them
in their quest. Otherwise, typical naturalists would lack the psychological strength to continue to
confront the obstacles in their endeavor to achieve objectively attractive goals. If naturalists were
atheists, they might end up choosing to live like The Blob.
One way to question this argument is to suggest that “an illusion sooner or later leads to an
inappropriate response and to the failure to accomplish whatever we set out to do. This makes it
unreasonable to allow illusions to guide our actions, especially when something important is at
stake” (Kekes 2010, 211). If there is a lack of evidence for God’s existence, such that belief in
Him is an illusion, then we are likely to rely on Him at the wrong time or in the wrong way and
thereby fail to achieve our goals.
However, what if the evidence for God is equivocal, does not point toward either theism or
atheism definitively? Can there not be cases when an illusory belief is on the whole pragmatically
useful—although we might rely on Him in unproductive ways sometimes, might some people’s
lives be on balance more productive with faith in Him?

Meaning: Atheism
A second way to question the argument for faith is to grant that belief in God would produce
meaning in one respect but to maintain that it would reduce meaning in another one. Faith would
sometimes enhance meaning in life when it comes to motivation, but it would always detract from
meaning in life when it comes to deliberation.
Recall that many philosophers contend that objectively attractive projects, ones that are
meaning-conferring, are those that involve the exercise of our intelligence in certain ways. If a
large part of what constitutes meaning in life is the development and exercise of our rational
nature, then there can be meaning in the processes of deliberative choosing—for example, in select-
ing our aims based on practical reasons and in deciding to act in light of the epistemic reasons
that exist for judging aims to be realizable. Even if such processes were to lead to fewer meaning-
ful goals being achieved down the road, the processes themselves would arguably enhance meaning
in that agents are free of illusion. Conversely, those who acted on faith would have somewhat less
meaning in their lives by virtue of having failed to exercise their reason and perhaps for having
degraded it. If belief in God is not supported by the evidence, then believing in God would be
irrational and hence reduce meaning in life, particularly since the belief would be about the fun-
damental nature of our environment. “The pessimist is less likely to win the race, complete the
task, achieve the goal; but he takes home the consolation prize: his judgment was sound, his
assessment was true” (Waller 2003, 196).
Of course, one might deny that faith in God is degrading of our rational nature or otherwise
detracts from meaning in life. However, another consideration at this point is whether faith, suppos-
ing that it did reduce meaning in terms of deliberation, would be on balance warranted when fac-
toring in motivation. On the one hand, there is, let us suppose, the certain outcome of obtaining
some meaning, or at least not reducing meaning in one’s life, by believing in accordance with evi-
dence about the basic way the world operates, which, we are also supposing, means not believing
in God. On the other hand, there is the likelihood of obtaining some meaning by achieving more
goals due to one’s faith that there is a God who is “my rock, my fortress, and my deliverer” (Psalm
18:2).
The tension dissolves if there is conclusive evidence that God exists, but what if there is not? In
that case, another way to dissolve the tension would be to avoid belief in God but to develop the
confidence in oneself and obtain the requisite knowledge of one’s society that is needed to achieve
one’s aims. At least among people living in modern societies, unwarranted confidence in God’s exis-
tence does not appear essential for many (Hooker 2008, 196–199). Plenty of people are able to
undertake charity work, scientific experiments, and creative activities, not to mention get married
and rear children, without thinking they are doing anything that is part of God’s plan. The final
question is: Are you one of them?
Bibliography
Ayer, A. J. “The Claims of Philosophy.” (1947.) In The Meaning of Life, 2nd Frankfurt, Harry G. “The Importance of What We Care About.” (1982.) In
ed., edited by E. D. Klemke, 219–233. New York: Oxford University The Importance of What We Care About, by Harry G. Frankfurt, 80–94.
Press, 2000. New York: Cambridge University Press, 1988.
Baier, Kurt. “The Meaning of Life.” (1957.) In The Meaning of Life, 2nd ed., Hooker, Brad. “The Meaning of Life: Subjectivism, Objectivism, and Divine
edited by E. D. Klemke, 101–132. New York: Oxford University Press, Support.” In The Moral Life: Essays in Honour of John Cottingham, edited
2000. by Nafsika Athanassoulis and Samantha Vice, 184–200. New York:
Palgrave Macmillan, 2008.
Camus, Albert. The Myth of Sisyphus. Translated by Justin O’Brien. London:
Kauppinen, Antti. “Meaningfulness and Time.” Philosophy and Phenomeno-
H. Hamilton, 1955. First published 1942. logical Research 84, no. 2 (2012): 345–377.
Cottingham, John. On the Meaning of Life. London: Routledge, 2003. Kekes, John. The Human Condition. New York: Oxford University Press, 2010.
Cottingham, John. The Spiritual Dimension. Cambridge, UK: Cambridge Mawson, T. J. God and the Meanings of Life. London: Bloomsbury, 2016.
University Press, 2005. Metz, Thaddeus. Meaning in Life: An Analytic Study. Oxford: Oxford
Craig, William Lane. “The Absurdity of Life without God.” (1994.) In University Press, 2013.
Exploring the Meaning of Life: An Anthology and Guide, edited by Joshua Metz, Thaddeus. “Meaning.” In Blackwell Companion to Atheism and
W. Seachris, 153–172. Oxford: Wiley-Blackwell, 2012. Philosophy, edited by Graham Oppy, chap. 24. Oxford: Blackwell, 2018.
Meaning: Atheism
Moorhead, Hugh, ed. The Meaning of Life. Chicago: Chicago Review Press, Taylor, Charles. The Ethics of Authenticity. Cambridge, MA: Harvard
Nagel, Thomas. “Death.” (1970.) In Mortal Questions, by Thomas Nagel, Taylor, Richard. “The Meaning of Life.” (1970.) In Life, Death, and
1–10. New York: Cambridge University Press, 1979. Meaning, edited by David Benatar, 19–28. Lanham, MD: Rowman and
Littlefield, 2004.
Nozick, Robert. Anarchy, State, and Utopia. New York: Basic Books, 1974.
Tolstoy, Leo. “My Confession.” (1884.) Translated by Leo Wiener. In The
Nozick, Robert. “Philosophy and the Meaning of Life.” (1981.) In Life,
Meaning of Life, 2nd ed., edited by E. D. Klemke, 11–20. New York:
Death, and Meaning, edited by David Benatar, 63–90. Lanham, MD:
Rowman and Littlefield, 2004.
Walker, Lois Hope. “Religion Gives Meaning to Life.” In Philosophy: The
Sartre, Jean-Paul. “Existentialism Is a Humanism.” 1946. Translated by Quest for Truth, 5th ed., edited by Louis P. Pojman, 519−522. Oxford:
Walter Kaufmann. In Existentialism from Dostoyevsky to Sartre, edited by Oxford University Press, 2002.
Walter Kaufmann, 287–311. New York: World, 1956.
Waller, Bruce N. “The Sad Truth: Optimism, Pessimism, and Pragmatism.”
Seachris, Joshua W. “The Meaning of Life as Narrative: A New Proposal for Ratio 16, no. 2 (2003): 189–197.
Interpreting Philosophy’s ‘Primary’ Question.” Philo 12, no. 1 (2009): 5–23.
Wielenberg, Erik J. “God and the Meaning of Life.” (2005.) In
Seachris, Joshua W. “Death, Futility, and the Proleptic Power of Narrative Exploring the Meaning of Life: An Anthology and Guide, edited by
Ending.” Religious Studies 47, no. 2 (2011): 141–163. Joshua W. Seachris, 335–352. Oxford: Wiley-Blackwell, 2012.
Seachris, Joshua W., ed. Exploring the Meaning of Life: An Anthology and Wolf, Susan. “Happiness and Meaning.” Social Philosophy and Policy 14, no.
Guide. Oxford: Wiley-Blackwell, 2012. 1 (1997): 207–225.
Swinburne, Richard. “How God Makes Life a Lot More Meaningful.” In Wolf, Susan. “The Meanings of Lives.” (2010.) In Exploring the Meaning of
God and Meaning, edited by Joshua W. Seachris and Stewart Goetz, Life: An Anthology and Guide, edited by Joshua W. Seachris, 304–318.
149–166. London: Bloomsbury Academic, 2016. Oxford: Wiley-Blackwell, 2012.

TOPIC 16
Suffering: Theism
Siobhan Nash-Marshall
Mary T. Clark Chair of Christian Philosophy
Manhattanville College, Purchase, NY
This chapter discusses what evils exist in the world with a specific focus on suffering. Natural evil and
moral evil are explained in this chapter as well as the problem of evil.
THE PROBLEM OF EVIL
Much noise was made in twentieth-century philosophical discussions about the incompatibility of
evil and a good, omnipotent God, with evil meaning “suffering,” good meaning “moral,” and incom-
patibility of meaning the “impossibility of the coexistence of.” Such was the din caused by these dis-
cussions that it is not uncommon in this century to hear nonacademics, even religious Christians,
complain about it. How can a good—loving is the term most often used—God allow innocent peo-
ple to suffer? Is there not something contradictory about this?
There is nothing odd about a person of faith asking how and why God allows the innocent to
suffer. It is a natural and legitimate question. Even the long-suffering Job, who everyone would
agree was a level-headed person of faith, asked it. But, to take an example from my own tradition,
there is something decidedly odd about devout Christians asking if there is something contradic-
tory about God allowing the innocent to suffer. Christians believe that innocent suffering played
a crucial role in the unfolding of human salvation. Jesus of Nazareth was tortured to death. He
died on a cross after having been betrayed, scourged, mocked, and declared innocent. Surely if
Christians believe that Jesus of Nazareth suffered, was innocent, and that he is the Son of God,
they must also believe that the claims that “God is good”—that “God is moral”—and that
“God allows the innocent to suffer” are necessarily compatible. Christians, if they are true to their
faith, that is, must believe that God the Son innocently suffered. There is something truly amiss,
then, when Christians ask each other if there is something contradictory about the coexistence of
God and innocent suffering.
Yet today many still ask if there is not something contradictory about a loving God allowing the
innocent to suffer. Such is the lure of the question that it has become quite common in religious
settings. It has been the question that, as M. B. Ahern indicates, intellectuals have been asking in
academic settings for some time: “one speculative problem about evil has been considered so impor-
tant that it has been called ‘the problem of evil.’ This is the problem whether the existence of an
omnipotent and wholly good God is compatible with the existence of any evil” (1971, 1).
The question concerning the compatibility of “the existence of an omnipotent and wholly good
God … with the existence of any evil,” with an omnipotent being clearly also meaning an omni-
scient one, is not a particularly interesting one. It leaves the crucial and thorny issue of the very exis-
tence of evil unaddressed. It is also not a valid question, unless one supplements it with a few plau-
sible premises that would logically connect evil in one being with lack of goodness or power in God.
It is an illuminating question only for accidental reasons: it lays bare some of the problematic
assumptions of the modernist philosophical approach to reality. Its occurrence and reoccurrence in
523
Suffering: Theism
polite society today obliges thinkers to shed some clear light on

KEY CONCEPTS it. In what follows I attempt to do so, and I defend two claims
about evil and the “problem of evil,” namely, that:
Akrasia 1. The crucial problem raised by the existence of evil is its
Aporia reality: evil is an ἀπορία (aporia)—literally “that which has
Evidential Argument from Evil no path”—that is fundamental to the actually existent
world.
Inherent Anticipability
2. If one properly and rationally approaches and defines the
Intrinsic Intelligibility
reality of evil itself, one will see that its compatibility with
Logical Argument from Evil the existence of a good and omnipotent God is not a true
Probabilistic Argument from Evil problem at all. (This is not to suggest that one should not
reflect on the reasons why God allows evil to exist and
persist. Aporetic realities, especially those that are
fundamental to the actually existent world, should be
contemplated. They are, as Aristotle claims, the source of that wonder that is the spring of all
wisdom.)
This article is divided into five primary sections. Each responds to a question. The questions are
these:
i. How and why is evil in se problematic?
ii. What do those who raise the problem of evil take to be the reasons why evil is problematic?
iii. What presuppositions allow those who so raise the problem of evil to interpret what makes evil
problematic?
iv. Are the presuppositions legitimate?
v. What should one make of the problem of evil?
I begin with the question concerning evil itself—rather than the problem of evil—for several rea-
sons. The first is that it is always a good idea to know what one is talking about when one is going
to discuss it at any length. Tackling the problem of evil without first discussing the facts of the
case—evil itself—lends itself to getting lost in a forest of words and mental constructions. The
second reason is that the nature of evil is of crucial importance to understanding whether or not
existing evils are compatible with the existence of a good and omnipotent God. It is precisely
the nature of evil—what evil is—that demonstrates that questions concerning the possibility of
the coexistence of evil and a good God—the problem of evil—are not well formed. If evil is an
aporia, then any attempt to deduce something therefrom is a priori a nonstarter. The problem
of evil is an attempt to deduce something from evil: the nonexistence of God. That problem can-
not be well formed.
THE REALITY OF EVIL: HOW AND WHY IT IS AN INHERENTLY PROBLEMATIC FACT
Evil is an inherently problematic fact. It is destructive, if not of physical integrity, then of mental
soundness, and all too often of both things. Evil causes living things to feel fear and pain, both of
which are genuinely trying and profoundly disorienting for those who experience them. Both make
living things incapable of properly functioning. Evil is above all intrinsically unintelligible. Evil is a
surd, an anti-miracle of sorts: a break in anticipated and intrinsically anticipable patterns of action
and behavior that contradicts the very laws that inform it. I am borrowing the word surd here from
mathematics, where it denotes an irrational number (a real number that cannot be expressed as a ratio
between two integers) and using it metaphorically. It is because it is a surd that it is destructive, that
it causes fear and pain, that it is disorienting. Evil shocks reality and the human mind. Experiencing
it, in Marilyn Adams’s apt phrasing, “is part of what it is to be insane” (1999, 143).
Suffering: Theism
TRIVIAL MORAL EVILS AND THE INANITY OF EVIL ITSELF

Among the most shocking and illuminating evils that we can experience are those of our own direct
making. It is in our own destructiveness that the intrinsic unintelligibility and the inherent unanti-
cipability, the absurdity and incongruity, the fundamental contradictoriness of evil are on full dis-
play for us plainly to see. Years and years after the fact, Augustine could not, despite conspicuous
efforts, truly understand why he had stolen and thrown out pears that he did not even want or
need. The paltriness of the object of his theft and destruction made the matter only more mysteri-
ous to him. Had the pears been overwhelmingly desirable, he claims, his theft—though not his dis-
posal of the pears—might have made some sense to him. So too would the theft have had a plausi-
ble cause had he been hungry and destitute, although that cause too would have made his throwing
out the pears inexplicable. As it was, the pears were not particularly beautiful, they did not taste very
good, Augustine did not need them, and he could later find no rational and impelling cause for
either one of his deeds. “By what passion, then, was I animated? It was undoubtedly depraved
and a great misfortune for me to feel it. But still, what was it? Who can understand his errors?”
(Confessions, II.x). Both the theft and destruction were wanton: undertaken independently of the
intrinsic desirability of those desires of Augustine’s that “animated” him as he performed them.
Augustine describes something with which we are all well acquainted. Anyone who is willing to
honestly examine his own conscience and stare at his own pears will find that, like Augustine, he will
be unable to find a rational impelling cause for his own petty lies, thefts, and deceptions, for his little
abuses of power, vanity, smugness, arrogance, hasty judgments, and indifference. What rational reason
could we have for lying about inconsequential matters, pilfering trifles, casting judgments on people
whom we have never really bothered to get to know, taking offense when people do not acknowledge
our importance, or doing any one of those many things for which we were reprimanded when we
were children, and for which we could find no truly impelling rational cause then?
Brief reflection should show us that our misbehavior has no cogent purpose at all. Mr. Smith’s
gravely nodding at the claim that we have climbed Mount Vesuvius does not alter the fact that we
haven’t. Nor, in the best of circumstances, can his apparent increased respect for us ultimately elicit
in us anything but contempt for both him—and other suckers—and the truth, thereby subtly
undermining our own capacity to relate to others and to the very reality in which we live, as Harry
Frankfurt indicates so eloquently (2006, 78–79). The momentary relief that comes from vilifying
Jill does not alter the fact that we have not accomplished much in the last few years, magically make
our disgust with ourselves disappear, nor suddenly make us willing to roll up our sleeves and get our
work done.
Trivial moral evils do not just lack a cogent purpose. They run counter to both our better judg-
ment and to the inherent desire that underlies all of our actions: the desire to preserve and protect
ourselves, and to be happy. Fibbing, swiping, and drinking excessively are all inherently risky endea-
vors, as are puffing ourselves up, casting stones, and so on. The nodding Mr. Smith, who appeared
so taken with the account of our climb of Mount Vesuvius, might just stun us with a series of
pointed questions about the geography of the volcano, and thereby turn our contempt for him
against ourselves. Sally, with whom we enjoyed a bout of lambasting Jill, might just turn out to
be a friend of Jill’s, and subject us to a public blow-by-blow account of every nasty thing we said
about her. We all know this before and while we misbehave.
Our awareness (and experience) of the purposelessness, the dangers, and downright incoherence
inherent in engaging in trivial moral evils should suffice to steer us clear of them. Who in his right
mind would set himself up for ridicule—if not a worse fate—in order to attain nothing worth his
efforts and disgust himself in the process? Yet, experience and better judgment do not always keep
us from doing stupid and damaging things. We give in to the destructiveness, to the inanity, as it
were, unexpectedly, as Augustine points out, and more often than we care to admit. We jump head-
first into doing what we know we should not do. And when we honestly reflect on our wayward-
ness—especially after it has boomeranged on us—and try to pinpoint the reasons why we did what
we did, why we gave in to whatever passion that animated us, we cannot but say with Paul: “I do
not understand my own behavior; I do not act as I mean to, but I do things that I hate. While I
am acting as I do not want to, I still acknowledge the Law as good” (Romans 7:15–16).

Suffering: Theism
It is in trivial moral evils that the sheer senselessness, the intrinsic unintelligibility, the inherent
unanticipability, the absurdity and incoherence of evil are most distinct. Trivial moral evils not only
obviously lack impelling rational causes (by definition they concern insignificant things, things, that
is, for which we cannot formulate rationally impelling reasons). We engage in them knowing that they
run counter to what we should do. They stun us while—or after—we engage in them, if they have
not yet so bent us that we do not recognize their inanity. They reveal our own ἀκρασία (akrasia), lack
of control over our actions. They show us that some of our actions can not only not logically follow
from what we actually are, they can contradict what we are: rational beings who want naturally to
be happy, preserve our existence, and protect ourselves from danger. They show that our actions
can be intrinsically unintelligible and inherently unanticipable even to ourselves. They show us that
we can by our own hand and choice turn ourselves against ourselves.
MORAL EVIL
The inherent unintelligibility and intrinsic unanticipability, the absurdity and contradictoriness that
are immediately discernible in trivial moral evils stretch across all moral evils. One look at Adolf
Hitler’s obsessive destructiveness should more than amply illustrate the point. What rational cause
could have impelled Hitler in 1942 to assign 330,000 troops to slaughter Eastern European Jews,
while he launched his preemptive strike on Russia? The key to defeating Russia on land, as Napo-
leon’s disastrous Russian adventure had amply demonstrated, was a quick and decisive blow: an
overwhelming Blitzkrieg that would break the Russian will to defend the motherland. Decisive land
blows require the allocation of as many troops as one can muster. Napoleon’s 675,000 troops did
not suffice to strike that decisive blow in 1812. Borodino, with its vast bloodshed, was nothing
but a Pyrrhic victory, as Napoleon himself later lamented. Given these facts, which Hitler knew,
the decision to divert more than 300,000 troops from the Blitzkrieg so that they could slaughter a
nonmilitary target was downright irrational. It not only lacked clear military justification, it contra-
dicted military laws. It gainsaid the very purpose of the Russia invasion.
It is precisely because Hitler’s actions were irrational, because they contradicted his own explicit
purpose—not to mention the intrinsic need of every person to preserve and protect himself, to flour-
ish—that they were intrinsically unintelligible. It is because they were intrinsically unintelligible that
they were inherently unanticipable. It took years for the Allies to declare war on Hitler. Hindsight
might make us question Neville Chamberlain’s appeasement tactics: their lucidity, intentions, and
even morality. But hindsight operates with clear knowledge of objects, goals, plans, and outcomes.
Foresight and midsight do not. They must work with a part of the picture, as it were—events whose
purpose is not yet fully known, on plans as they are unfolding—in order to grasp the object of the
actions, the true nature of the events, and thereby predict future actions and formulate adequate
responses to them. Grasping and predicting make use of logic and operate with the expectation that
both the acts themselves and their purpose are intrinsically intelligible and inherently anticipable: that
they do not contradict their agent’s purposes or common human needs and desires. Hitler’s actions
and purpose did contradict his purposes and natural needs and desires. He did not operate within
the bounds of rationality. Nor do serial killers, school shooters, and other perpetrators of grave moral
evils. This is why, in so many cases, we cannot find them before it is too late.
DISEASE AND OTHER PHYSICAL EVILS

The inherent unintelligibility and intrinsic unanticipability of moral evil, its incoherence and run-
ning counter to the very laws that inform it, is replicated in physical evils. It is not by chance that
modern medicine places great emphasis on preventive and prophylactic measures: checkups, screen-
ing, cleaning, vaccinating, quarantining, and so forth. It is also not by chance that on the whole it
describes the development and progression of diseases in statistical, rather than causal, terms. It is,
finally, not by chance that medical doctors often have such vastly different prognoses for and
approaches to diseases, and that we know to get second, third, and even fourth opinions before
making important decisions regarding our health.
Like moral evils, physical evils do not have what ancient and medieval philosophers called per
se causes. Like moral evils, physical evils are not, in other words, the direct intended effects of the
acts of properly functioning entities or organisms or of other natural events. Like moral evils, as
Suffering: Theism
such, physical evils cannot be a priori deduced from any single cause. Like moral evils, they cannot
be approached with the expectation that they are intrinsically anticipable, or inherently intelligible.
Medical doctors know this. It is why they use primarily a posteriori methods to deal with dis-
eases, concentrating on discovering them in their early stages. It is why there is a plethora of differ-
ent approaches to diseases, why radiologists, oncologists, and gastroenterologists can give contradic-
tory opinions with respect to what should be done with a pancreatic mass that causes no pain. It is
above all why the object of modern medicine is to contain the diffusion of diseases.
Were physical evils to have per se causes, were they to be inherently intelligible, medical doctors
would certainly not use primarily a posteriori methods to deal with them. Nor would they give con-
tradictory opinions, prognoses, and cures. Nor again would the object of medicine be to contain the
diffusion of diseases. It would be to address their causes.
Autoimmune diseases, for instance, are inherently mysterious. Why does a living organism sud-
denly begin to attack itself—its immune system attack its own bowels, its own joints, its own
nerves, and so forth? It cannot be the basic rules, the encoding of the immune system per se, that
directly causes it to do so. That encoding is intended to preserve a body’s integrity and to defend
it from foreign invaders, as it were. What, then, makes that system turn against the very body whose
integrity it is intended to preserve? What makes it “mistake” the very organism that it is intended to
preserve for a “foreign invader”? How can it do so? What, in other words, is the cause of autoim-
mune diseases? For that matter, what is the cause of cancer, and of aging? Why do organisms sud-
denly begin to act in a way that is contrary to their own encoding? Why do we die?
Like moral evils, autoimmune diseases, cancer, and even aging are not just destructive. They
lack per se causes. They are intrinsically unintelligible. They are surds, anti-miracles of sorts: breaks
in anticipated and intrinsically anticipable patterns of action and behavior that contradict the very
laws that inform them and are thereby destructive. These breaks are explainable, it would seem,
only as Aristotle thought chance is, as “the unintended, undesired result of a clash of causalities.
… When a cancer develops in the body [it] … cannot be explained by per se causes: it happens
unexpectedly, as it were at random as an adventitious event” (Steel 1998, 97–98).
THINKING ABOUT EVIL

As should be clear, the very fact of evil ought to provoke any number of fundamental and interre-
lated philosophical questions: What is evil? How can we understand the inherently unintelligible?
Why is the inherently unintelligible that contradicts the very laws that inform it, why, in other
words, are surds present in our world? Why do things act in a way that contradicts their own encod-
ing, as it were? How can they do it? How can there be breaks in the intelligible order of the moral
and material universes? How should one confront evil and deal with its aftermath? What is the
proper response to the fact itself of evil?
A person reading about “The Problem of Evil” could justifiably assume that it deals with these
and other fundamental questions that are elicited by the problematic fact itself of evil. The presence
of evil in the world is not just a threat to us. It is not just painful and frightening. Evil challenges
our assumptions about reality itself.
Things’ actions and behavior must follow rules in order for us both to understand them and to be
able to coexist with them. We assume that people will act rationally: that they will not intentionally
act in a way that they know will put them in harm’s way. We assume, in other words, that people
are morally good. We could not live in society or interact with people if we did not so assume. We also
assume that material things must act in accordance with the laws that govern material existence—bio-
logical laws, laws of chemistry, laws of physics, and so forth. An apple that shot up into space rather
than falling to the ground would be truly shocking to us, as would a lioness in the wilderness giving
birth to a goat. We assume that things are naturally good: that they act in an orderly and intelligible
way. We could, again, not live in the world or interact with it if we did not so assume.
Our assumptions are not groundless. They are rooted in two facts: (1) that the actions of things
must follow laws, or rules, in order to be intelligible, and (2) that the actions of things must follow
laws, or rules, in order for things to persist: to maintain their integrity.

Suffering: Theism
Given our assumptions, given the facts, it should be impossible for evil to exist. It should be
impossible for the actions of things encoded for integrity and persistence to contradict their very
encoding: the laws, the rules that ensure their integrity and persistence. It should be impossible
for rational beings intentionally to act irrationally. And yet it is not impossible. Evil is very real.
It is the absurdity of the combination of these two facts—that inherent intelligibility and
intrinsic lawfulness are necessary conditions of existence itself and yet that there exists that which
is inherently unintelligible and intrinsically unlawful—that makes evil intellectually shocking to
us, that makes experiencing evil “part of what it is to be insane.” Evil is not an intellectual problem
to be solved, or a problem that human beings can completely grasp. Evil is, as has been noted, an
aporia, a quandary.
APORIA
Aporiai, in the sense intended here, are Gordian knots about the basic elements of reality and
human thought that resist human attempts to think them through. Aristotle’s classical example of
an aporia, in Book B of the Metaphysics, concerns universals. The gist of that aporia is this: univer-
sals, he claims, must exist in themselves, else human beings would not be able intellectually or ratio-
nally to know reality, and real change would be impossible; but if universals actually did exist in
themselves they could not be universal, since everything that is, insofar as it is, must be a particular
thing (Aristotle, B.4 999a, 24–999b 23). Universals must, Aristotle concluded as such, both be and
not be, exist and not exist in themselves.
When it comes to aporiai, one ought neither to minimize them—to ignore or deny one of the
factual elements that constitutes them—nor to exaggerate their impassibility by refusing to approach
them altogether. Neither of these responses allows one properly to address the reality of the prob-
lematic issues to be faced in the aporiai themselves. Rather, one should see in the aporiai, as Aristo-
tle intended and as Søren Kierkegaard claims, “the passion of thought.” It is the aporiai that cause
the wonder and puzzlement from which wisdom is engendered. It is again such aporiai that keep
the human mind anchored to reality. Like matter, aporiai are stubborn facts. They do not bend
to our feelings, minds, or wills. They are constant reminders that our minds are, exist in, and are
subject to a reality that we have not created, and over whose rules and domain we have no real
power: that the object of knowing is the adequatio intellectus ad rem—conforming our minds to
reality—and not the adequatio rei ad intellectum—conforming reality to our minds. This is why
Kierkegaard continues his point about the aporiai by claiming that “a thinker without” them “is like
a lover without passion: a mediocre fellow” (1985, 32). The human response to quandaries, to apor-
iai, is and should be to recognize their irreducibility, to acknowledge their centrality in our attempts
to come to know reality, and to reflect with other thinkers on the ways in which one can and should
approach them: how and where one can find a path between the factual elements that constitute
them without denying the reality of either of them.
EVIL AND APORIAI

There are many reasons why evil is aporetic, or rather, why like the ontological status of universals,
evil resists the human attempt to understand it. The reason on which this chapter has been focusing
concerns the fact that evil is an inherently problematic reality. It is intrinsically unintelligible. It is a
surd, an anti-miracle of sorts: a break in anticipated and intrinsically anticipable patterns of action
and behavior that contradicts the very laws that inform it, and is thereby destructive. Plainly put,
surds should not exist. But they do. They are what we call evil. Hence, the aporia: the inherently
unintelligible and unlawful—evil—both cannot and does exist.
That moral evils are surds should not be a surprising claim. It is a fact that is a posteriori dem-
onstrated by countless human acts. It should also be evident a priori.
One of the necessary conditions of morality itself is that the moral good be intrinsically intelli-
gible: formulatable, justifiable, and teachable. There could be no practical syllogisms, natural moral
laws, categorical imperatives, utilitarian calculi, or any other understandable, learnable, and applica-
ble moral principle or standard were the moral good not intrinsically intelligible.
Suffering: Theism
The point has many consequences. It clearly entails that actions whose impelling causes are
moral are, to the extent to which they are moral, necessarily inherently intelligible, anticipable,
and lawful. It also entails that moral evil—that which by definition lacks the moral good—can-
not but be inherently unintelligible, unanticipable, and unlawful: that their grounds and reasons
cannot be rationally articulated. It is this point that led Plato in the Protagoras (358c), for
instance, to claim that moral evil is ignorance, Boethius in the Consolation of Philosophy (IV.2)
to claim, with Plato, that only the wise can do what is good, and Hannah Arendt to claim that
evil is banal.
There is another dimension to the problem. The Pauline claim that the evildoer does “not act
as” he means to, but does “things that” he hates, that he still acknowledges “the Law as good” while
he performs evil actions, is as true of great moral evils as it is of trivial ones. It is what grounds the
Kantian claim that perpetrators of moral evils actually hold contradictory beliefs: that the “Law is
good” and “binds” everyone, and that the “Law is good” and does not “bind” them (Kant 1981,
32). Moral evil is inherently contradictory. But contradictions are neither inherently intelligible,
nor do they have intrinsically foreseeable consequences. If I am both dead and alive at the same time
and in the same way, I can equally do both anything and nothing. This is why evil is inherently
unanticipable.
That physical evils are surds should also not be a surprising claim. We have all of the a poster-
iori proof of the fact that anyone should want. It should also be a priori evident. The entire scien-
tific adventure, the human adventure is predicated on four primary presuppositions: (a) that mate-
rial events occur with regularity; (b) that they are necessarily inherently informed by universal
regulae—laws—or per se causes that make them so occur; (c) that the regulae are intelligible; and
(d) that the regulae are that through which we know material things.
The point has many consequences. It clearly entails that those acts, events, and things that are
informed by the regulae, in other words the normal (that which follows the νόμος [law]), are neces-
sarily inherently intelligible. It also entails that acts, events, and things that are not informed by the
regulae insofar as they are not are necessarily not inherently intelligible. Crazed immune systems are
not inherently intelligible insofar as they are crazed. As crazed, the actions of the immune system are
not caused by that which would make the immune system intelligible: its regulae.
The point also entails that those destructive events that are caused by “clashes of causalities” are
not inherently unintelligible: they cannot be explained by the regulae informing either cause, and
since the clash is a singular event, it cannot itself have a regula. To be clear, that an old, frayed Inca
rope bridge collapses is a normal and inherently intelligible event. That Doña María crosses that
bridge on her way to a pilgrimage is also an inherently intelligible event. That Doña María crosses
the bridge when the bridge collapses is not informed either by the regula that makes the frayed
bridge collapse—gravity—or by the regula that led Doña María to cross it—piety. Doña María’s
death is consequently an event that is per se unintelligible and per se unanticipable. It lacks a per
se cause.
Breaks in anticipated and intrinsically anticipable patterns of action and behavior that contra-
dict the very laws that inform them, should not be. Such breaks fly in the face of the necessary
assumptions of all epistemic activities. They fly in the face of the normal functioning of things in
the universe. But such breaks do exist. They are instances of what we call evil. We have, thus, that
evil both cannot and does exist.
EVIL AND THE “PROBLEM OF EVIL”
What thinkers commonly call the problem of evil today does not at all directly address the aporia
inherent in the very existence of evil, or even allow for the fact that evil is an aporetic reality. What
modern—especially late modern—and contemporary thinkers call the problem of evil would more
aptly be called the problem of God, or better yet, the problem of religious belief. That problem is primar-
ily a logical one that concerns the compatibility of four terms: God, omnipotent, good, and evil. It asks
if the very possibility of existing “evil” logically demonstrates the impossibility of the existence of an

Suffering: Theism
“omnipotent” and “good” “God” and the irrationality of faith in him. The arguments through which
thinkers would demonstrate the incompatibility of these four terms are often called atheological argu-
ments. Their intent is to demonstrate either that God does not exist and that belief in God is, as J. L.
Mackie claimed, “positively irrational,” by which he meant that it can be “maintained” only by an
“extreme rejection of reason” (1955, 200), or, the more moderate claim that it is improbable that
God exists and that belief in God is implausible.
THE STRUCTURE AND AIM OF ALL FORMULATIONS OF THE PROBLEM OF EVIL

As Mackie indicates in his article “Evil and Omnipotence,” the problem of evil can be plainly stated:
“In its simplest form … [it] is this: God is omnipotent; God is wholly good; and yet evil exists”
(1955, 200). All three of these statements, Mackie claims (1955, 200), are:
(1) “essential parts of most theological positions: the theologian … must adhere … to all three.”
Yet, he adds, having necessarily to adhere to all three statements is inherently problematic:
(2) “There seems to be some contradiction between these three propositions, so that if any two
of them were true the third would be false.”
It is the conjunction of (1) and (2) that constitutes the problem of evil. (1) requires that the theist
hold that:
(3) (a) “God is omnipotent,” (b) “God is wholly good,” and (c) “evil exists” are all true
statements.
(2) would entail that:
(4) it is impossible for (a) “God is omnipotent,” (b) “God is wholly good,” and (c) “evil exists”
all to be true statements.
Clearly (3) and (4) are contradictory propositions: they respectively state and deny that all of the
members of one and the same set of statements are true. But (4) is just a qualified restatement of
(2), and (3) simply makes explicit the content of (1). This means, of course, that if (3) and (4)
are contradictory propositions, so too must (1) and (2) be. Thus, if it is true that (1)—the theist
must believe that (a) “God is omnipotent,” (b) “God is wholly good,” and (c) “evil exists” are all true
statements—and that (2)—there is a “contradiction between these three propositions, so that if any
two of them were true the third would be false”—there is a problem. That problem is what is today
referred to as the problem of evil.
The problem of evil holds only for the theist. One of the conditions of its arising for a person,
Mackie acknowledges, is that he “adhere to” (1): “If you are prepared to say that God is not wholly
good, or not quite omnipotent, or that evil does not exist … then the problem of evil will not arise
for you” (1955, 201). But one cannot “adhere to” (1) unless he also “adheres to” (a) “God is
omnipotent” and (b) “God is wholly good.” One of the necessary requisites of so “adhering” is
clearly believing that “God exists” is a true statement. Only the theist can hold (1). Thus Mackie’s
concession that: “The problem of evil … is a problem only for someone who believes that there is a
God who is both omnipotent and wholly good” (1955, 200).
It is not by chance that the modern problem of evil holds only for the theist. Mackie, like
Pierre Bayle and David Hume before him and H. J. McCloskey and Antony Flew with him, only
raises the issue of evil and its ostensible incompatibility with a good and omnipotent God in order
to demonstrate that “religious beliefs … are positively irrational” (Mackie 1955, 200).
It is not the problem of evil that leads Mackie to consider theism rationally bankrupt. In his
brief presentation of the reasons why he raises the problem of evil, he makes clear that, like the later
philosopher Thomas Nagel (1997, 130), he is not keen on the very idea of God. Theism, Mackie
brusquely asserts, lacks a proper rational foundation: “The traditional arguments for the existence
of God have been thoroughly criticized by philosophers” (1955, 200). He infers from this purported
lack of philosophically adequate proofs (and knowledge) of God’s existence not just that “no
Suffering: Theism
rational proof of God’s existence is possible” but that there can be no such thing as rational knowl-
edge of God’s existence: that belief in God is inherently nonrational. Theism, he concludes, requires
a “rejection of reason” (1955, 200).
The problem of evil is significant for Mackie because it offers a “more telling criticism” of the-
ism than the “thorough critiques” of “traditional arguments for the existence of God.” The latter
critiques, he thinks, afford the theist an escape route. The believer can “admit that no rational proof
of God’s existence is possible. And he can still retain all that is essential to his position, by holding
that God’s existence is known in some other, non-rational way” (1955, 200). Evil, Mackie thinks,
affords no such path. It is, he is convinced, theism’s Armageddon, its Trojan horse: it provides proof
positive that belief in the existence of a good and omnipotent God requires an “extreme rejection of
reason” (1955, 200), that God does not exist:
Here it can be shown … that the theologian can maintain his position as a whole only by a
much more extreme rejection of reason than in the former case [i.e., by claiming that
knowledge of God can be nonrational]. He must now be prepared to believe, not merely
what cannot be proved, but what can be disproved from other beliefs that he also holds.
(1955, 200)
SUPPLEMENTARY CLAUSES
As sure as he is that evil is theism’s Trojan horse, Mackie realizes that demonstrating—as opposed
to claiming—that it is so takes more than simply laying our three propositions common to the great
forms of monotheism and claiming that they are contradictory. As David Conway points out,
despite the fact that “on a common sense pre-philosophical level it seems clear enough that the evil
that exists in the world constitutes, if not an insuperable obstacle, at least a real difficulty for the-
ism,” a strong philosophical case can be made that “evil in the world presents no problem of logic
or of evidence for the theist” (1988, 35).
What Conway points to, and Mackie seems to realize, is that, emotions and prejudices aside,
there is not even a prima facie contradiction between Mackie’s three propositions. Certain condi-
tions must hold in order for propositions to be contradictory. Among them are that:
(a) the propositions must concern one and the same thing (or state of affairs) T;
(b) one of the propositions (or a combination thereof) must affirm (or necessarily imply) that T
is (or has the property or quality) x;
(c) one of the propositions (or a combination thereof) must deny (or necessarily imply the
denial) that T is (has the property or quality) x.
Mackie’s three propositions do not even meet the first of these conditions. There is no T common
to all of those propositions. The first two concern God, the third concerns a reality other than God.
Whoever wants to jump to the conclusion that anyone who simultaneously holds the three proposi-
tions that:
i. All living human beings are capable of abstract rational thought
ii. All living human beings are capable of free acts
iii. Sue Venir is not capable of abstract rational thought or free acts
“rejects reason” because he is guilty of holding contradictory beliefs, should first check to see if Sue
Venir is actually a living human being. Sue Venir might be dead. Then again, she might be a ham-
ster or a souvenir shop. In none of these cases is there a contradiction between the three proposi-
tions. They concern different things. The same thing holds for Mackie’s propositions. They simply
do not concern one and the same thing or state of affairs.
What is needed to make an atheological argument out of the three propositions that constitute
the problem of evil are supplementary propositions that would logically link the two propositions
that concern God and the third proposition that concerns evil: that would make all three

Suffering: Theism
propositions concern one and the same subject, and affirm and deny one and the same quality or
property of that subject.
MACKIE AND THE LOGICAL LINK

Mackie, like all thinkers who formulate atheological arguments, presumably takes the link between
the two propositions concerning God and the third proposition stating that evil exists to be self-
evidently provided by the terms “good,” “omnipotent,” and “evil.” Such was, for instance,
McCloskey’s (1960, 97) understanding of the link. What is not evident to “atheologians”—or to
anyone else who thinks carefully about the matter—is how and why the terms of their argument
actually are contradictory. Mackie recognizes this and responds that “to show it we need some addi-
tional premises, or perhaps some quasi-logical rules connecting the terms” (1955, 200–201). These
“quasi-logical rules” are two:
(a) “Good is opposed to evil, in such a way that a good thing always eliminates evil as far as it
can.”
(b) “There are no limits to what an omnipotent thing can do.”
“From these,” he concludes, it “follows that a good omnipotent thing eliminates evil completely.”
This last proposition, he then adds, shows that “the propositions that a good omnipotent thing
exists, and that evil exists, are incompatible” (1955, 201).
Mackie takes the “incompatibility” that his supplementary premises have yielded to show that
there is sufficient reason to hold that there really is a contradiction between the three propositions
of the problem of evil, and therefore that he has successfully demonstrated both that theism requires
an “extreme rejection of reason,” and that God does not exist. The argument that claims that any
instance of evil is proof positive that God does not exist is typically called the logical argument from
evil.
PIKE AND PLANTINGA AND THE LOGICAL ARGUMENT FROM EVIL

Mackie’s attempt to articulate the logical connection between the three propositions of the problem
of evil and to generate what atheologians consider a self-evident contradiction in theistic belief was
quickly shown to be unsuccessful. Both of his supplementary premises are false.
After specifying that by the term “good” atheologians actually mean “morally good,” and that
by the term “evil” they mean “suffering,” Nelson Pike (1958) indicates that supplementary pre-
mises, like Mackie’s first one, actually state that:
(ab) a moral being always eliminates all instances of suffering as far as he can, else he will cease
to be a moral being.
This statement is simply false. Not only is it possible for a moral being to act morally while not
eliminating all instances of suffering that are in his power to eliminate, but there are cases in which
it is morally necessary for a moral being not to eliminate suffering. These cases—such as the case of
a parent who must allow medical doctors to give his three-year-old child painful rabies shots, or who
forces his child to return stolen candy—are those in which morality requires the acceptance and per-
mission of suffering. As moral persons generally can be in a position where they must not eliminate
suffering in order to be moral, so too a perfectly morally good omnipotent person can be in a posi-
tion of having to permit suffering in order to be moral:
If the proposition, “There is a good reason for evil in the theistic universe” (i.e., “there are
motives or other factual conditions which, if known, would render blaming God for evil
inappropriate”) could be true, then the logic of the phrase “perfectly good person” allows that
the propositions “God is a perfectly good person” and “God allows evil in the world even
though he could prevent it” could be true together. This point rests on the fact that a perfectly
good person can allow evil, provided he has a good reason. Since the first of the three
propositions just mentioned is clearly not contradictory and thus could be true, the
conjunction of the latter propositions is also free of contradiction and the contention that a
Suffering: Theism
perfectly good person would of necessity prevent evil if he could is shown, beyond question, to
be in error. (Pike 1958, 119)
Mackie’s second supplementary premise is also false, as Pike—and Alvin Plantinga after him—
point out. It calls on a faulty understanding of the term “omnipotent.” It is simply not true that
“there are no limits to what an omnipotent thing can do.” Like every other being, an omnipotent
being is limited by logical possibility. No being, omnipotent or otherwise, can do what is logically
impossible. No being can, for instance, violate the principle of noncontradiction by creating crea-
tures with free choice of the will and simultaneously creating them so that their actions will not have
nasty consequences.
The significance of this point is manifold. It indicates, as Pike argued, for instance, that existing
evil is necessarily logically compatible with the existence of an omnipotent, good God, or, what is
the same thing, that there can be no necessary contradiction between the claims that “God is good
and omnipotent” and that “evil exists.” Were evil, for instance, a logically necessary feature of reality
(which is a logical possibility), not even an omnipotent being could eliminate it (Pike 1958, 119;
Plantinga 1974, 189). It is, of course, also true, as Plantinga claims in his “free will defense,” that
it is logically possible for God to create free beings whose wills it is logically impossible for him to
determine. In such a case, it would be logically impossible for God to prevent all suffering. If a cre-
ated free being were, for instance, to inflict eternal suffering on himself through the single and irrev-
ocable free act of choosing to rebel against God, let us say, it would be logically impossible for God
to eliminate just that suffering. As this is so, it is not logically impossible for God at once to be
omnipotent and morally good while not eliminating “evil completely.”
CONWAY, TOOLEY, ROWE, AND EVIDENTIAL ARGUMENTS

The work of Pike and Plantinga to expose the flaws inherent in the logical attempt to define the
supplementary premises that would “reveal” the contradiction that is purportedly inherent in the
three propositions that make up the problem of evil, however, did not quiet the atheological call
to arms. Rather, it led atheologians to rearticulate how and why the terms of the problem of
evil—“good,” “omnipotent,” and “evil”—are conflicting terms. The newer attempts have come to
be known as evidential or probabilistic arguments from evil.
What is common to both evidential and logical arguments is their general structure. Both make
use of the same three propositions. Both attempt convincingly to articulate what those thinkers who
formulate them take to be a self-evident contradiction inherent in those propositions. The articula-
tions of this contradiction in both forms of the argument make use as logical links between the
three propositions new variants of the claims that:
(ab) a moral being always eliminates all instances of suffering as far as he can, else he will cease to
be a moral being;
(bb) there are no limits to what an omnipotent being can do.
Where evidential arguments differ from the logical ones is that they do not claim that any instance
at all of evil suffices to generate the contradiction that the problem of evil requires. Nor do those
who would articulate evidential arguments claim that they can conclusively demonstrate that evil
makes belief in God positively irrational. Evidential arguments claim that there is some “fact” about
concrete evils that is such that it makes God’s existence “unlikely” (Tooley 1991, 90).
Evidential arguments diverge on the nature and identity of this specific “fact.” Michael Martin
argues that the “fact” concerns the quantity of evil in this world of ours. It is “evil in great abun-
dance,” he claims, that “would falsify the existence of God unless one assumes either that God
has sufficient reason for allowing the existence of evil in great abundance or that evil in great abun-
dance is logically necessary” (Martin 1978, 430). William Rowe argues that the “fact” concerns the
quality of evil that is to be found in this world. Specifically, he claims, it is cases of “intense human
and animal suffering” (Rowe 1979, 335) that falsifies the existence of God. Quentin Smith also
holds that the “fact” about evil that would show that God does not exist concerns animal suffering.
He argues that it is the specific suffering of animals being attacked by other animals (Smith 1991,

Suffering: Theism
159). Conway argues that the troubling “fact” about evil concerns the “lingering suffering and even-
tual death of the child of cancer” (Conway 1988, 40).
Michael Tooley’s generic structure of evidential arguments from evil is a good summary of the
arguments of this type:
(1) Cases where animals die agonizing deaths in forest fires, or where children undergo lingering
suffering and eventual death due to cancer, are cases of intrinsically undesirable states of
affairs that an omnipotent and omniscient person could have prevented without thereby
either allowing an equal or greater evil, or preventing an equal or greater good.
(2) An omniscient and morally perfect person would prevent the existence of any intrinsically
undesirable state of affairs whose prevention it could achieve without either allowing an
equal or greater evil, or preventing an equal or greater good.
(3) There does not exist an omnipotent, omniscient, and morally perfect person. (Tooley
1991, 95)
What is one to make of evidential arguments from evil? Is there some “fact” about evil that
makes God’s existence “unlikely”? Can concrete evils be evidence of God’s nonexistence? One must
understand both what “evil” means in the context of the evidential problem of evil and how that
problem must understand the nature of “evidence” in order to answer the question.
THE PRESUPPOSITIONS OF THE PROBLEM OF EVIL
As both the logical and the evidential formulations of the problem of evil make clear, atheologians
must make a number of crucial assumptions about both the nature of evil and the relation between
human rationality and reality in order to make their case. The most significant of those concerning
evil are two. The first is that:
(1) Evil is not an inherently problematic fact, an intrinsically unintelligible fact: a break in
anticipated and intrinsically anticipable patterns of actions and behavior that contradicts the
very laws that inform it.
Atheologians presuppose that (1) is true. The strongest confirmation of the fact that they do are
Tooley’s three claims, that the problem of evil concerns logical relations between ethical judgments,
that the logical relations between propositions are unaffected by “whether ethical judgments have
truth values,” and that arguments from evil indeed need not involve “reference to correct moral
principles” (Tooley 1991, 99–102). These claims clearly entail that the reality of evil itself, the fact
of evil, what evil is, is not a concern for Tooley. One assumes that Tooley would be concerned
about evil itself, the “truth values” of judgments like “x = evil,” and “correct moral principles” if
he did consider evil to be an inherently problematic reality.
Further confirmation of the fact that atheologians presuppose that (1) is true comes from the
fact that atheological arguments explicitly treat evil as though it were a problematic fact only in rela-
tion to God’s goodness and omnipotence. They all unequivocally state that evil is a problem only for
those who believe that “God is both omnipotent and wholly good” (Mackie 1955, 201). If this is
so, however, then atheological arguments must presuppose that evil is not an inherently problematic
fact. If E is only problematic in relation to G, then E cannot be inherently problematic.
There are two primary reasons why atheologians believe (1) to be true. The first and most
obvious is that they all assume that the nature of evil—what evil is—both is self-evident and is
self-evidently to be identified with pain and suffering. Smith claims this explicitly when he states
that: “It seemed to me self-evident that the natural law that animals must savagely kill and devour each
other in order to survive was an evil natural law” (Smith 1991, 159; emphasis added). Tooley thinks
it so obvious that one would necessarily recognize “cases where animals die agonizing deaths in forest
fires, or where children undergo lingering suffering and eventual death due to cancer” (1991, 95;
emphasis added) as clear instances of evil that, like Smith, he does not explain why the evils he cites
Suffering: Theism
are evil. This particular assumption makes for the most striking characteristic of atheological argu-
ments. They often describe the most horrendous examples of suffering in order to demonstrate that
existing evils preclude the existence of God. McCloskey is an atheologian who is most guilty of this
(1962, 187). Clearly, if atheologians consider evil to be self-evidently identified with pain and suf-
fering, they cannot take it to be an inherently problematic fact in the sense that it is an inherently
unintelligible fact, what we have been calling a surd.
A second obvious reason is closely related to this first one. Atheologians assume that evil is a
positive property like any other positive property, such as might be the property of having a specific
color—green, red, or blue—or of being a specific weight or height, with the notable difference that
it is, as Tooley claims, “intrinsically undesirable.” Mackie claims that, were evil a privation, there
would be no problem of evil (1955, 201), but then simply dismisses the claim that evil is a privation
without giving reasons why it is not so. In one of his more colorful passages, McCloskey gives voice
to what one assumes are the reasons why atheologians assume that evil is a positive property: “Pain
is positive. It has a real nature, and its evilness flows from that nature, not from its being the
absence of something else” (McCloskey 1962, 187).
The second significant presupposition regularly made by atheologians regarding evil is that:
(2) Evil is necessarily eliminable, removable, or preventable.
Like the first presupposition, this second one is both explicit and crucial to the problem of evil
itself. If, as Mackie explicitly claims in his first supplementary premise, a “good thing is opposed
to evil in such a way that a good thing always eliminates evil as far as it can,” he cannot but hold
that evil is eliminable. Likewise, when Conway and Tooley claim that an omnipotent being
“could”—or indeed “should”—“prevent the existence of any intrinsically undesirable state of
affairs,” they presuppose that evil is always preventable. As both Pike and Plantinga demonstrate,
atheologians could not reasonably raise the problem of evil were they not to presuppose that evil is
eliminable or preventable. One cannot reasonably claim that an omnipotent being is not moral or
omnipotent (or that he does not exist) if he does not eliminate or prevent a necessary feature of
reality.
The third crucial presupposition of all atheological arguments concerns human rationality and
is common to much modern thought. It is commonly formulated as:
(3) Inconceivability necessarily entails metaphysical impossibility.
The linchpin of atheological arguments is the belief that it is inconceivable that a good and omnip-
otent God and evil can both exist. It is a belief that precedes atheologians’ attempts to demonstrate
that God and evil are logically (or evidentially) incompatible terms. It is a belief that drives the
attempts to so demonstrate. As noted above, Conway makes this point explicitly: “On a common
sense pre-philosophical level it seems clear enough that the evil that exists in the world constitutes,
if not an insuperable obstacle, at least a real difficulty for theism” (Conway 1988, 35). It is a belief
that persists despite the many powerful criticisms of the attempts to so demonstrate. It is a belief
that is predicated on the principle used by many of the classical modern philosophers: that incon-
ceivability entails metaphysical impossibility.
ARE THE PRESUPPOSITIONS OF THE PROBLEM OF

EVIL LEGITIMATE?
All of the presuppositions of the problem of evil mentioned above are questionable. The first—that
evil is not an inherently problematic reality, that it is not what we have been calling a surd, but is
problematic only in relation to God—is simply not tenable. Peter van Inwagen (2006, 12–17) to
the contrary, evil itself—and not just “radical evil”—is inherently problematic. It is, as the first sec-
tion of this chapter argues, both a priori and a posteriori evident that evil is an inherently problematic
fact, an intrinsically unintelligible surd: a break in anticipated and intrinsically anticipable patterns of
behavior that contradicts the very laws that inform it. The very existence of evil throws a wrench in

Suffering: Theism
our attempts to understand and survive in the world and with other people. As that section argues,
those attempts presuppose both that (a) material reality is knowable, explainable, and regular, that it
is what the ancients and medievals called naturally good, and (b) that people are morally good, that
their behavior is knowable, explainable, and regular—that their actions, in other words, conform to
the natural desires of every human person to persist, protect himself, and flourish. Evil contradicts
these assumptions. It is inherently unintelligible: unknowable, inexplicable, and irregular. It is because
evil contradicts these assumptions that it is destructive, painful, and shocking. One cannot simply
ignore these facts.
The belief that evil is self-evidently to be identified with pain and suffering, as popular as it has
become, is also eminently questionable. Pain and evil are not, as Aristotle pointed out in response to
Eudoxus, simply self-evidently identical. History is rife with people who believed pleasure to be an
evil thing, which would be impossible were evil self-evidently to be identified with pain (Aristotle,
Nicomachean Ethics, 1171a, 26–35). Nor are pain and evil coextensive realities. There are evils that
are not in themselves painful—like taking heroin, being in a coma, or being senile. There are pains
that are not in themselves evil—like the pain of muscles that have had a good workout, like just
remorse, like healing pains. As this is so, there are no grounds to claim that evil is self-evidently
identical with pain and suffering.
The belief that evil is necessarily eliminable or preventable is also not a new one. John Stuart
Mill and many nineteenth-century writers were quite convinced that “all the grand sources … of
human suffering are in a great degree, many of them almost entirely, conquerable by human care
and effort” (Mill 2001, 15).
And to be sure, it is possible for all evil to be eliminated if evil is nothing but “suffering.” All it
would take in order to eliminate all suffering would be to eliminate all life. There can be no suffer-
ing if there is no life. But surely this is not what Mackie, Conway, Tooley, or, for that matter, Mill,
mean to presuppose when they assume that evil is eliminable, or preventable. What they seem,
rather, minimally to presuppose is that:
It is possible for a thing to remain essentially like what it is and not to include evil.
What is interesting about the above belief is that it flies straight in the face of the atheological presup-
position that evil is not an inherently problematic fact: that it is neither inherently unintelligible, nor
intrinsically unanticipable. If it is indeed possible for a thing to remain essentially like what it is and
not to include evil, and if that is indeed what that thing ought to be (as atheological arguments explic-
itly claim), then the fact that there is evil cannot but be an inherently unintelligible and intrinsically
unanticipable one. If (T + e) = (T – e), then clearly T does not imply e, (∼ (T e)), which is to
say both that: (i) T is necessarily not a sufficient condition of e, and that (ii) whatever makes T be
T is necessarily not a sufficient condition of e. If this is so, then given T, we have both no reason
to expect e, and no way to explain why e exists. In other words, if (T + e) = (T – e), then given T
(iii) e is intrinsically unanticipable and (iv) e is inherently unintelligible. This point is applicable not just
to single things, but to the material universe as a whole.
If it is possible for a thing—or the whole of the material universe itself—to remain essentially
like what it is and not to include evil, what is the cause of evil? That cause clearly cannot be what
constitutes that thing—or the universe itself—per se. It cannot, in other words, be what makes that
thing—or the set of all material things—what it is. Were that cause to be what makes the thing—or
set of all material things—be what it is, it would not be possible for the thing—or that set—to
remain essentially like what it is and not to include evil. Nor can the cause of evil be the thing—
or set of all material things—per se, else again it would not be possible for the thing—or set—to
remain essentially like what it is and not to include evil. What then is the cause of evil? How can
evil exist, if it is possible for a thing (or set of things) to remain essentially what it is like when it
does include evil and when it does not? How can a reality that remains essentially like itself act dif-
ferently from itself? How can a reality that remains essentially like itself with and without evil be in
radically different states? The point is, that if it is possible for a thing—or for the set of all material
things—to remain essentially like what it is and not include evil, then the fact that there is evil is
inexplicable. And if that is the case, the very existence of evil cannot but be aporetic.
Suffering: Theism
WHAT IS ONE TO MAKE OF THE PROBLEM OF EVIL?
Digging beneath the presuppositions of the problem of evil leads us thus back to where we started:
evil is not an intellectual problem to be solved, or a problem that we can completely grasp. Evil is
what the Greeks called an aporia. Recognizing this alone is enough to demonstrate that the problem
of evil is ill formed. No valid deductions can be drawn from aporetic premises, but the problem of
evil attempts to do just that.
That the contemporary problem of evil has come to concern God and religious belief, rather
than evil in se, is an odd fact. It is only in limited circumstances that a being who cannot by defini-
tion have anything to do with a given inherently problematic fact can be taken to be part of what
makes that fact inherently problematic. It would be truly odd, for instance, to claim that Angela
Merkel, the chancellor of Germany, not only has something to do with the opioid epidemic in
the United States but is what makes that epidemic problematic. No doubt, some conspiracy theorist
might find some far-fetched reason to connect Merkel and drug use in the United States. The more
levelheaded of us would, I hope, simply shrug at the claim and respond both that the abuse of
opioids is a problem no matter who Merkel is (and even if or whether or not she exists), and that
given who she actually is, she cannot have had anything to do with the opioid epidemic itself. So
too for God and evil. Like the abuse of opioids, evil is a problematic fact in se. It was inherently
problematic for ancient pagan philosophers. It is inherently problematic for nonbelievers in any
age. It is inherently problematic, period. And God has less to do with evil than Merkel does with
the opioid epidemic in the United States.
It is also a curiously unsurprising fact that many thinkers would attempt to turn the problem
of evil primarily into the problem of God or the problem of religious belief. Aporiai are uncomfort-
able realities. No one really wants to think about how it is possible for parallel lines to intersect
at infinity, any more than about Georg Cantor’s transfinite number theory and its claim that
there are different orders of infinity, or to contemplate the fact that universals both must and can-
not exist in themselves (to mention again Aristotle’s aporia). This is especially true of thinkers
who work under the shadow of René Descartes’s search for absolute clarté and believe that a
thing’s conceivability is a condition of the possibility of its existence. Aporiai are not just uncom-
fortable for Descartes’s heirs, they are systemic threats. If reality does contain aporetic facts, as
Aristotle would insist that it does, conceivability and metaphysical possibility cannot be coexten-
sive: conceivability cannot, in other words, mark the divide between what is possible and what
is not. Aporiai thus challenge the foundations of the modern approach to reality. It is not and
should not, therefore, be surprising that many modern and contemporary thinkers attempt to
avoid addressing the aporia inherent in evil.
Nor should it be surprising that many modern and contemporary thinkers attempt to reduce
the aporia that is evil to a contradiction, which is precisely what turning the problem of evil into
the problem of God and religious belief attempts to do. That attempt is nothing but the attempt
to deny that evil is an aporia.
But aporiai are stubborn facts. Like matter, they resist our attempts to control reality. For every
atheological attempt to reduce the aporetic reality of evil into a contradiction concerning God, there
is a theistic response that would show the inadequacy of that attempt. Thinkers who articulate
atheological arguments complain about this fact (Conway 1988, 38). The history of the modern
problem of evil is in many ways a contest between atheists and theists, whereby atheists attempt val-
idly to formulate the reasons why they believe that the existence of evil contradicts the very possibil-
ity of the existence of a good and omnipotent God, and theists invalidate the attempt.
More important than the back and forth between atheists and theists is the fact itself of evil.
Both atheists and theists realize that evil ought not to be there: that it is a surd. It is on the response
to this point that they differ. Augustine reflected on the reasons why evil is unintelligible and shock-
ing. He reflected on the reasons why it ought not to be there. He begged for forgiveness for being
one of the reasons for its existing. The atheists feel that evil is a reality that ought to be eliminated,
and, rather than try to understand what its very existence implies about reality itself and about
themselves, prefer to claim that it cannot but be someone else’s responsibility.

Suffering: Theism
Bibliography
Adams, Marilyn McCord. Horrendous Evils and the Goodness of God. Ithaca, Nagel, Thomas. The Last Word. Oxford: Oxford University Press, 1997.
NY: Cornell University Press, 1999.
Pike, Nelson “God and Evil: A Reconsideration.” Ethics 68, no. 2 (1958):
Ahern, M. B. The Problem of Evil. London: Routledge and Kegan Paul, 1971. 116–124.
Conway, David A. “The Philosophical Problem of Evil.” International Plantinga, Alvin. The Nature of Necessity. Oxford: Clarendon, 1974.
Journal for Philosophy of Religion 24, no. 1/2 (1988): 35–66.
Rowe, William L. “The Problem of Evil and Some Varieties of Atheism.”
Frankfurt, Harry G. On Truth. New York: Knopf, 2006. American Philosophical Quarterly 16, no. 4 (1979): 335–341.
Kant, Immanuel. Grounding for the Metaphysics of Morals. 2nd ed. Edited Rowe, William L. “Ruminations about Evil.” Philosophical Perspectives 5
and translated by James W. Ellington. Indianapolis, IN: Hackett, 1981. (1991): 69–88.
Kierkegaard, Søren. Philosophical Fragments. Edited and translated by Smith, Quentin. “An Atheological Argument from Evil Natural Laws.”
Howard V. Hong and Edna H. Hong. Princeton, NJ: Princeton International Journal for the Philosophy of Religion 29, no. 3 (1991):
University Press, 1985. 159–174.
Mackie, J. L. “Evil and Omnipotence.” Mind, n.s., 64, no. 254 (1955): 200–212. Steel, Carlos. “Proclus on the Existence of Evil.” In Proceedings of the
Boston Area Colloquium in Ancient Philosophy. Vol. 14, edited by John
Martin, Michael. “Is Evil Evidence against the Existence of God?” Mind 87,
J. Cleary and Gary M. Gurtler, 83–102. Leiden, Netherlands
no. 347 (1978): 429–432.
Cologne: Brill, 1998.
McCloskey, H. J. “God and Evil.” Philosophical Quarterly 10 (1960): 97–114.
Tooley, Michael. “The Argument from Evil.” Philosophical Perspectives 5
McCloskey, H. J. “The Problem of Evil.” Journal of Bible and Religion 30, (1991): 89–134.
no. 3 (1962): 187–197.
Van Inwagen, Peter. The Problem of Evil: The Gifford Lectures Delivered in the
Mill, John Stuart. Utilitarianism. Edited by George Sher. Indianapolis, IN: University of St. Andrews in 2003. Oxford: Clarendon; New York:
Hackett, 2001. Oxford University Press, 2006.
TOPIC 16
Suffering: Atheism
THE PROBLEM OF EVIL 539
Bruce Russell
Wayne State University, MI
THE PROBLEMS OF DIVINE HIDDENNESS AND DIVINE INSCRUTABILITY 549

Daniel Linford
Graduate Student, Department of Philosophy
THE EXPLANATION OF THE DISTRIBUTION OF SUFFERING 554

Daniel Linford
Graduate Student, Department of Philosophy
This chapter discusses the existence of suffering as it pertains, in several dimensions, to traditional theism. A
central component of this chapter concerns the conflict between theism and the premise that excessive
gratuitous evil exists.
THE PROBLEM OF EVIL
Bruce Russell, Wayne State University
The problem of evil is the problem of how an all-knowing, all-powerful, wholly good God could allow
the existence of so many bad things, especially the suffering of innocent people and animals. After dis-
cussing the nature of suffering and what makes it bad, I offer three versions of the argument from evil
against the existence of God, the last of which concludes that God does not exist because there is much
more suffering in the world than he need allow to achieve important goods. I consider defenses against
this version of the argument that hold that we are in no position to judge that the suffering is excessive
and unnecessary, and then consider replies that claim that the defenses lead to unacceptable forms of
moral or epistemic skepticism. In the last section I consider arguments against theism based on Bayes’s
theorem, criticize part of an argument against theism put forth by Paul Draper, and end by arguing
that theism, even when supplemented by certain auxiliary hypotheses, should be rejected.
As several philosophers have said, the problem of evil is really the problem of “bad things”
(van Inwagen 1991, 4; Swinburne 1998, 4). There are lots of bad things in the world: injustice,
exploitation, torture, rape, murder, and suffering, to name a few. How can a God who is all-knowing,
all-powerful, and wholly good allow these bad things to occur? A good parent would not allow her
innocent children to be brutally beaten, raped, and then killed if she knew it was happening and could
prevent it. So how can God, who knows what is happening and has the power to prevent it, allow his
539
Suffering: Atheism
innocent children to be treated in these ways? Why does he allow

KEY CONCEPTS them to suffer horribly, so much and so often?
THE NATURE OF SUFFERING

Argument from Evil
Among the world’s many bad things, the argument from evil
Bayes’s Theorem often focuses on suffering and how a wholly good God can
Core Theism allow it. We need to understand the nature of suffering and
Defense what makes it bad to see why this is a problem.
Evidential Problem of Evil Of pain, Derek Parfit writes: “When we are in pain, what is
bad is not our sensation but our conscious state of having a sensa-
Global Problem of Evil tion we dislike. If we didn’t dislike this sensation, our conscious
Horizon Blocker state would not be bad” (2011, vol. 1, 54). He gives examples
Inference to the Best Explanation (IBE) of sensations some people dislike but others like or are indifferent
to, such as the squeaking of chalk against a blackboard or the
Local Problem of Evil buzzing of houseflies, the sensation of touching velvet, the taste
Logical Problem of Evil of milk chocolate. Parfit distinguishes between, on the one hand,
Naturalism inherently disliking or liking something, and, on the other, not
wanting or wanting it. Masochists, and those who are doing pen-
Skeptical Strategy ance, might want pain that they dislike, and some puritanical peo-
Skeptical Theism ple might not want sexual pleasure that they like. It is inherent
Special Theism disliking that makes a mental state inherently bad for a person
and inherent liking that makes it inherently good.
Suffering
While people suffer when they have pain they inherently
Theodicy dislike, sometimes suffering of that sort does not raise the prob-
Young Earthers lem of why God would allow it. Eleonore Stump (2010, 6)
gives the example of women who would rather have natural
childbirth with the accompanying pain than give birth with
an anesthetic to relieve the pain. I think she is mistaken when
she says, “the pains of childbirth do not count as suffering” (Stump 2010, 11). Parfit could reply
that the women who prefer natural childbirth with pain to childbirth without pain suffer because
they do not like the pain. However, they value natural childbirth because of the feelings of joy
and exhilaration that they experience, and inherently like. These goods outweigh the suffering the
pain represents so, on balance, the women prefer natural childbirth with suffering to childbirth with
an anesthetic and no suffering. There is no puzzle why a good parent or God would allow people
they love to experience what is, on balance, good for them.
Humans, as well as nonhuman animals, can suffer from pain that serves no good end, as when
they are burned or whipped, their flesh is cut, or their bones broken. Why God allows so much of
this suffering to occur is a puzzle since, unlike in the case of natural childbirth, very little of it is
accompanied by outweighing goods.
Still, Stump thinks there are other types of suffering that people endure that are even more seri-
ous than painful suffering. She defines “desires of the heart” as those desires that are at “the core of a
person’s volitional structure” (2010, 7). If a person fails to get, or loses, a desire of her heart, she suf-
fers. She is heartbroken, heartsick, or experiences heartache when her lover leaves or her child dies.
In addition, Stump says that “undermining a person’s flourishing, regardless of a person’s will [her
deepest desires], is also a source of suffering” (2010, 8). She gives examples of enslaved women who
no longer care about consenting to sex and so do not even consider whether they have been raped.
Stump defines a person’s flourishing in terms of being the best sort of person she can be. She illus-
trates the difference between a person’s getting his heart’s desire and his flourishing by citing the case
of the seventeenth-century English poet John Milton, who at one time strongly wanted to be involved
in politics and thus served in government, even though the best life for him, the one in which he
would flourish, was that of a poet (Stump 2010, 10). At one point in his life, Milton’s heart’s desire
conflicted with his flourishing. Different people will be the best they can be by pursuing different
things, say, by being a poet, a plumber, a parent, or a philosopher. Presumably, being the best one
Suffering: Atheism
can be in Stump’s sense is incompatible with being a certain sort of person, say, with being a pimp, a
philanderer, or a flimflam man, even if being that sort of person is someone’s heart’s desire.
Stump summarizes her view about suffering and what makes it bad as follows: “What is bad
about the evil a human being suffers is that it undermines (partly or entirely) her flourishing, or it
deprives her (in part or in whole) of the desires of her heart, or both” (2010, 11). I think Stump
should also include mental states like pain, grief, and depression that a person inherently dislikes.
People in those mental states, and nonhuman animals in pain, also suffer because they inherently
dislike being in those states. And it is this inherent dislike that makes those states bad. Parfit says
that to be in agony is to have sensations that you inherently and intensely dislike (2011, vol. 1,
56). Why is there so much agony in the world if God exists?
THE RELATION BETWEEN GOD AND SUFFERING: THREE VERSIONS

OF THE ARGUMENT FROM EVIL
There are various versions of the argument from evil that conclude that God does not exist. The
simplest version goes like this:
1A. Necessarily, if God exists, there is no suffering.
2A. But there is suffering.
3A. So God does not exist.
The idea behind 1A is that suffering is bad and a good God would want to prevent bad things
from happening if he knew about them and could prevent them. Since God is assumed to be
wholly good, all-knowing, and all-powerful, if he exists, there would be no bad in the world and
so no suffering.
There are two standard criticisms of 1A in this argument. First, it is argued that some goods are
possible only if there are certain bads, and it’s better to have these goods with the bads than neither.
For instance, there can be forgiveness only if there has been harm or wrong, which are bad in them-
selves, and it is better to have forgiveness and those harms or wrongs than neither. Second, it is argued
that having significant freedom requires being able to freely will, and bring about through your will,
both great goods and great evils. So God would want to create a world where such creatures would
eventually exist since having significant freedom is a great good in itself. But if beings with significant
freedom exist, they may will, and bring about through their will, awful suffering. So contra 1A, it’s not
true that, necessarily, if God exists, there is no suffering. His creatures who have significant freedom
can cause awful suffering (Plantinga 1977; van Inwagen 2006; Swinburne 1998).
In light of these criticisms, philosophers have modified the argument from evil to read as
follows:
1B. Necessarily, if God exists, he would not allow any unnecessary suffering (that is, suffering that
he need not allow to prevent something even worse or for there to be some greater good).
2B. But there is such unnecessary suffering.
3B. So God does not exist.
Two criticisms of 1B have been put forward by Peter van Inwagen (1996b, 234–235). First, he
appeals to an analogy involving a good ship’s captain who is faced with saving at least some of the
one thousand people on the island of Atlantis, which is sinking. For every person he puts on board
to sail to safety, the chance that the ship will go under increases by 0.1 percent. So if he takes all
aboard, the ship will definitely sink; if he takes none, then the ship will make it safely to port but
all will die when Atlantis goes under. Van Inwagen says that a good captain would do nothing
wrong if he took at least a handful (say, 10) or all but a handful (say, 990). However, if he took just
a handful, he might leave behind some who could have come aboard without the ship’s sinking. If
those people then drown, there will be unnecessary suffering. But this shows that it could be mor-
ally permissible for a good captain to allow unnecessary suffering. If that’s true of a good captain,
why not of a good God? God, too, can allow unnecessary suffering.

Suffering: Atheism
Of course, this analogy is not apt because the good captain is not omniscient and so does not
know exactly how many he can take on board and make it safely to port. So the captain has an
excuse, via his nonculpable ignorance, that God would lack if he were the captain. The analogy
gives us no reason to think that God could allow unnecessary suffering. He should allow the mini-
mum amount of suffering needed to bring about some relevant good or to prevent some relevant
bad, if there is such a minimum.
Van Inwagen’s second objection to 1B is subtler and makes the case that God could allow unnec-
essary suffering in circumstances where there is no such minimum. Call a unit of painful suffering a
“dolor.” Suppose allowing nine dolors of suffering would not achieve some great good but allowing
only a smidgen more than nine would. Perhaps, then, one would not be justified in allowing ten
dolors of pain to achieve that good. We might fault a person for allowing ten dolors since he could
have allowed fewer, say, 9.5 dolors and got the same result. But, of course, he could have done the
same by allowing 9.25, 9.125, and so on. Still, if he allows an amount of dolors very close to nine,
he does nothing wrong even though he could have allowed fewer and got the same result, that is, even
though he has allowed unnecessary suffering. So a good being could allow unnecessary suffering, con-
trary to what 1B asserts. That will be the case if there is no minimum amount of suffering that must
be allowed to bring about some great good or prevent some great bad, but only an approachable limit
to some amount of suffering that is not itself enough to bring about that good or prevent that bad.
Perhaps there is a way to defend 1B against this subtle objection, say, by arguing that it is true
except in rare cases where no minimum, but the relevant sort of limit, exists. But I think the best
route for the defender of the argument from evil is to offer a third version of that argument that starts
from a different principle. Let’s grant that in the example a good person could allow unnecessary suf-
fering very close to nine dolors. Still, he could not allow, say, a hundred dolors as a means of bringing
about the relevant good. In other words, he could not allow much more suffering than he need allow
to achieve that good. Allowing much more suffering than need be allowed is what I’ll call allowing
excessive, unnecessary suffering. This is how the third version of the argument from evil goes.
1C. Necessarily, if God exists, he would not allow excessive, unnecessary suffering (that is, much
more suffering than he need allow to prevent something even worse or for there to be some
greater good).
2C. But there is such suffering.
3C. So God does not exist.
I do not think there are objections to 1C. The objections now focus on 2C. Arguments from evil
that focus on the first premise of the argument as I’ve presented it raise what is called the logical
problem of evil because they are concerned with whether that premise is logically necessary. Those
forms of the argument that focus on the second premise raise what is called the evidential problem
of evil because they are concerned with the evidence that is offered in support of that premise.
Another distinction that is made is between local and global versions of the argument from evil.
Local versions focus on particular evils, say, the rape, beating, and strangulation of a little girl (Rus-
sell 1989); a fawn that has been burned in a forest fire and lingers on, suffering for days before
dying (Rowe 1979); or even the Holocaust, which is a particular instance of evil even if it involves
a vast amount of suffering. Global versions of the argument from evil focus on the vast amounts
of evil, or horrible evil, in the world, and how it is distributed (say, on the heads of innocent or
good people) (van Inwagen 1991). The third version of the argument from evil that I have offered
is a global version of the argument.
People distinguish between “natural evil” and “moral evil.” Natural evils include suffering that
results from natural events like hurricanes, earthquakes, tsunamis, volcanic eruptions, and diseases
spread by animals. Moral evils include suffering that is caused by moral agents who torture, kill,
rape, and exploit other humans and animals. When I say that there is excessive, unnecessary suffer-
ing I mean to include both “natural” and “moral” suffering.
While we can see how some amount of suffering might be needed to develop certain virtues
like compassion, perseverance, courage, mercy, trust, and forgiveness, we can’t see how all the
Suffering: Atheism
suffering of innocents that we observe in the world is needed to produce those virtues. Certainly,
after reflection, it seems that there is more than enough suffering to allow people to develop the rele-
vant virtues. And after reflection it seems that there would be plenty of room for significant freedom
even if there were a lot less suffering. Sometimes the suffering that results from natural events is
partly due to the choices of free agents, say, when people build their houses near earthquake faults
or in flood- or hurricane-prone areas, but many times it does not, unless one assumes, implausibly,
that Satan or other free agents cause these natural events.
There are two possible explanations of why it seems that there is excessive, unnecessary suffer-
ing: (1) because there really is much more suffering than is needed to develop important virtues and
for the exercise of significant freedom, or (2) because God’s allowing all this suffering (or nearly all
of it) is needed to achieve important goods or prevent important bads, but we are ignorant of the
reasons why it is needed. (1) is a better explanation than (2) because it is simpler than (2). Explain-
ing why something seems to be the case in terms of its being the case is a simple explanation. In
contrast, (2) posits mysterious goods and bads that we cannot access even after much reflection. It
does not help that (2) includes the idea of an immaterial perfect being who has reasons beyond
our ken for allowing all the suffering we see. If we walked into a seemingly empty room and saw
the drapes moving, the best explanation of their moving would not be that it is the work of ghosts
that cannot be seen, touched, smelled, tasted, or heard, even if we looked all around and could not
find the reason why they were moving. For a similar reason, (2) is not the best explanation of what
seems to be excessive, unnecessary suffering. In addition, in the case of suffering, (1) is an alternative
explanation, unlike with the drapes where perhaps we should suspend judgment about what is caus-
ing them to move. So we should accept (1) and reject (2).
Note, first, that this argument is not what has been called a “noseeum” argument (Wykstra
1996). The argument is not that because we do not see any reason why there is so much unneces-
sary suffering it is reasonable to believe there is none. Opponents of that type of argument point out
that it is not always true that failing to see something (say, bacteria on your hand) gives you reason
to believe it’s not there. Rather, the argument I offered is that the best explanation of why it seems
that there is excessive, unnecessary suffering is that there is such suffering. In the case of bacteria on
your hand, the best explanation of why you don’t see them is not that they are not there. The best
explanation is that they are there but too small to see.
Some opponents of the sort of argument I just gave for 2C think that failing to see a reason for
something (here, why there is so much suffering of innocents) gives you reason to believe it is not
there only if you have reason to believe you would see it if it were there. (Stephen Wykstra intro-
duces a much discussed principle that he calls the “Condition of Reasonable Epistemic Access,” or
“CORNEA” [1984, 152]; the principle I give here is supposed to capture the intuitive underpin-
ning of CORNEA.) So failing to see why God would allow so much suffering of the innocent does
not give us reason to believe that there is no reason to allow it, for we do not have reason to believe
that we would see God’s reason if it existed. There is no reason to think that our finite minds could
grasp all of God’s reasons for doing or allowing certain things.
Sometimes the principle that opponents of “noseeum” arguments rely on gives the right answer.
Failing to see bacteria on your hand gives you no reason to believe there are none if there is no rea-
son to think that you would see them if they were there. A novice chess player’s failing to see the
point of a master’s move does not give the novice reason to believe there is none, since he has no
reason to believe that he would grasp the point if there were one. But that principle gives the wrong
answer when it comes to skeptical hypotheses that claim that we are brains in a vat (BIVs), are being
deceived by an evil demon, or are in the Matrix. We can have reason to believe that we are not
BIVs, are not being deceived by an evil demon, and are not in the Matrix even though we do not
have reason to believe that if these things were true we would have reason to believe that they were.
The reason for rejecting these skeptical hypotheses, and accepting the commonsense view that we
are in a real world independent of us that is pretty much like what we think it is, is that the
commonsense hypothesis is the best explanation of what we observe. Sometimes the results of
applying the principle that theists use to defend theism agree with the results of applying inference
to the best explanation (IBE) to the situation, and sometimes they do not. When they do not, the

Suffering: Atheism
results of applying IBE seem the right ones. This suggests that IBE is more fundamental and that
the principle that skeptical theists apply is derivative; that principle is perhaps a heuristic that usu-
ally, but not always, gives the right answer about when failing to see something does not give us rea-
son to believe it is not there.
I have given an argument for 3C and responded to two criticisms of that argument. One criti-
cism mistakenly construes it as a “noseeum” argument. The other involves a counterargument that
relies on a false principle to conclude that failure to see reasons that God might have if he exists
gives us no reason to think that there are no such reasons.
OBJECTIONS TO THE ARGUMENT FROM EVIL

Critics of the second premise in the argument from evil as I have formulated it offer either what are
called theodicies or defenses against that second premise. A theodicy purports to give reasons for
thinking that that premise is false, and a defense purports to give reasons for thinking we are in
no position to judge whether it is true or false (van Inwagen 2006, 7). Van Inwagen says that a the-
odicy and a defense may even have the same content, but they have different purposes: to give
God’s actual reasons for allowing all the suffering that exists versus to give a coherent account of
what, “for all we know,” are God’s reasons for allowing that suffering. An attorney for the defendant
in a trial might argue that there is reason to believe that her client, Bobby, is innocent because there
is reason to believe that his identical twin, Billy, robbed the convenience store; or she might argue
only that, “for all we know,” Billy robbed the store, which means that we are not justified in believ-
ing he did not even if we are also not justified in believing he did. The first sort of argument would
be the analogue of a theodicy; the second, of a defense.
Most philosophers have given up on producing a successful theodicy that will explain all the
apparently pointless suffering in the world. Theodicies may show how God and some evil are consis-
tent, but they have a hard time showing that God and the existence of so much apparently unnec-
essary evil are consistent. Often that is because they do not provide good reason to believe that all
that suffering is really necessary. However, a defense might be more successful, because all it has
to establish is that, “for all we know,” all that suffering is necessary.
It’s impossible here to consider all the defenses that have been offered, but all the ones I know
of have some implausible feature in them that prevents them from satisfying the “for all we know”
requirement. For instance, van Inwagen’s defense includes a story that he claims is true “for all we
know.” According to this story, God miraculously changed our primate ancestors—who had preter-
natural powers that enabled them to protect themselves from wild beasts that they could tame with
a look, to cure disease with a touch, and to foresee natural disasters—into rational creatures with
libertarian freedom of the will. Because of their animal origins, early humans exercised their newly
acquired freedom of the will in a self-centered way, rebelled against God, and thereby became sepa-
rated from him. God’s plan for redemption includes letting even innocent children and adults suffer
to nudge everyone to freely return to him in loving union, which is the highest good for everyone
(see van Inwagen 2006, 85–90, for his defense).
Evolution gives us reason to believe that the first part of van Inwagen’s story is false: why would
creatures suddenly become rational and have libertarian freedom of the will; why would they lose
those advantageous preternatural powers? The second part of the story seems implausible given
our knowledge of morality and human psychology. It does not seem morally permissible to allow
innocent children to suffer in order to encourage others, including evil people, to seek union with
God. Furthermore, it seems psychologically implausible to think that children estranged from their
parents would be encouraged to restore their relationship with them if they thought that their par-
ents were letting them suffer when the parents could easily do something about it. It is not the case
that van Inwagen’s defense is true “for all we know.” There is reason to think that it is false (see
Russell 2017, 95–97, for further discussion of van Inwagen’s defense).
Richard Swinburne thinks that God would not allow any sentient being to have a life that is
not, on balance, worth living for that being. So, returning to William Rowe’s example of the fawn
dying slowly after being burned, Swinburne says that its life is worth living because it can serve as a
warning to other animals who observe its plight; being of use to others in that way makes its life
Suffering: Atheism
worth living. This, of course, ignores fawns that die lingering deaths that are never observed by any
other animal. Furthermore, in defense of allowing the suffering of innocent children, Swinburne
says that God, being more the author of their being than humans are of their children, has a greater
right to allow them to suffer than human parents have to allow their children to suffer. This is
doubtful since it does not seem that Dr. Frankenstein has a greater right to allow his creature to suf-
fer than we have to allow our children to suffer. But even if it were true, it does not seem that God’s
right is so broad as to permit him to allow children to be tortured, abused, and then painfully killed
(see Swinburne 1998 for his defense).
Another attempt to show that we are not justified in believing that there is unnecessary suffer-
ing, let alone excessive, unnecessary suffering, assumes that if we were justified in believing that
there is such suffering, we would have to be justified in believing that there are no goods that
God is aware of, but we are not, that would justify him in allowing what appears to be unnecessary
suffering. It is further argued that we could be justified in believing this only if we were justified in
believing that the goods and bads of which we are aware are a representative sample of all goods and
bads, including those of which God is aware. But then it is argued that we are not justified in
believing that they are a representative sample. It follows that we are not justified in believing that
there is unnecessary suffering (Bergmann 2014).
Defenders of the argument from evil respond that this line of argument will drive skeptical the-
ists to an unacceptable form of moral skepticism where we will often not be justified in believing
that some action is wrong where, intuitively, we are. It is assumed that in many contexts the objec-
tive rightness and wrongness of an action is determined by its total consequences. It would be objec-
tively wrong for you to give a thirsty child a drink of water if, contrary to the evidence, the water
was poisoned. However, if the child were Adolf Hitler, it would be objectively permissible, perhaps
even obligatory, to give young Hitler the poisoned water. So it seems that at least sometimes the
objective rightness and wrongness of an action is a function of the totality of its good and bad con-
sequences. But if there are good and bad consequences of which God is aware but we are not, the
objective rightness and wrongness of actions will partly be a result of those consequences. But
because those good and bad consequences are beyond our ken, in cases where consequences are
all that matter to the moral status of an action, we can never know, or even be justified in believing,
that any particular action is objectively obligatory or wrong. That is just what general moral skepti-
cism says. But surely sometimes we can know that an action is objectively right, say, that giving
unpoisoned water to the thirsty child who is not Hitler is objectively right. So this skeptical theist’s
objection leads to unacceptable results.
Skeptical theists respond by distinguishing between the morally relevant consequences for us
and those for God. Wykstra says that “our delimitation of morally relevant facts will need to be
indexed to the person in question, and to the kind of agent that person is” (2017, 134). His
account of our objective duties is given in terms of our optimal moral horizon, which consists of
all the morally relevant facts that we would be aware of if “horizon blockers” did not get in our
way, that is, if moral insensitivity, lack of empirical information (say, about whether the water is
poisoned), time pressure, and so on, were absent. On Wykstra’s view, what is objectively right
and wrong for us is determined by the morally relevant facts in our optimal moral horizon. That
does not include all the morally relevant facts in God’s optimal moral horizon, which includes all
morally relevant facts. So we are in no position to judge that God has no good reason to allow much
of the suffering we see because what he has good reason to allow is relative to his optimal moral
horizon, not ours. But just because we do not have access to God’s moral horizon, it does not follow
that often we cannot know, or even be justified in believing, what is objectively morally required of
us. That is because our optimal moral horizon is accessible to us when we are not subject to distort-
ing influences, including nonculpable ignorance, and that horizon determines what is objectively
right and wrong for humans.
While I think that more has to be said about the idea of an optimal moral horizon for humans,
and the correlative idea of horizon blockers, to meet the objection from moral skepticism to skepti-
cal theism, the skeptical theists face a bigger problem involving skepticism about beliefs, that is, con-
cerning epistemic rather than moral skepticism. The defense against the objection from moral

Suffering: Atheism
skepticism is that we are in no position to judge that there is unnecessary suffering because we can-
not presume to be aware of all the goods and bads that God is aware of. But young earthers offer a
similar defense of their view based on our ignorance of God’s reasons. The young earthers believe that
the earth was created recently: maybe six thousand years ago, maybe a hundred years ago, maybe just
five minutes ago, as Bertrand Russell once speculated. Their view is that God created the earth
recently with all its signs of age (deep river valleys, fossils, cave paintings, etc.) so that we could learn
the lessons of history and science without all the suffering of innocent animals and people that would
occur if there were a real history. They grant that it is prima facie wrong for God to deceive people
this way but argue that it is not all things considered wrong (see Swinburne 1998 and van Inwagen
1991, who argue God would not deceive, and Wielenberg 2014, who argues otherwise). Sometimes
it is permissible for people to deceive others if that is the best and only way to save a large number
of innocent people, for example, if lying to the Nazis about where a large number of Jews are hiding
is the best and only way to save them. Something similar seems true of God.
Of course, the young earthers will admit that they do not know why God created the earth
when he did, not earlier or later, why innocent people still seem to suffer horribly, and so on. Those
reasons are beyond our ken. But the skeptical theists say the same sort of things about parallel par-
ticulars. While they have general views about why there are rational creatures like us with significant
freedom, about what the ultimate good is (which they believe consists in a loving relationship with
God), about why people are sinful, and about what God’s plan for redemption is, they grant that
God’s reasons for allowing what appears to be unnecessary suffering are beyond our ken, and grant
that we often do not know why God would allow a particular instance of horrific suffering.
So if skeptical theists are right that we are in no position to judge that there is excessive, unnec-
essary suffering, the young earthers are justified by the same reasoning in holding that we are in no
position to judge that the earth was not created recently. But, intuitively, young earthers are not jus-
tified in holding that we are in no position to judge that the earth was created recently. So neither
are skeptical theists justified in believing that we are in no position to judge that there is excessive,
unnecessary suffering.
BAYES’S THEOREM AND COMPETING WORLDVIEWS

Bayes’s theorem helps us compare how likely one hypothesis is on some evidence, e, to another. To
determine how likely H1 is on evidence, e, [P(H1/e)] we should see how likely the evidence is on
H1 [P(e/H1)] and how likely H1 is on its own [P(H1)]. The same holds for a competing hypothe-
sis, H2. The probability of H2 on the evidence, e, [P(H2)/e] depends on how likely e is on H2 [P
(e/H2)] and the probability of H2 itself [P(H2)]. Suppose e is the evidence that the bottom half of a
ship sailing away from us disappears first, and the mast and the upper parts last. Let H1 = the earth
is round; H2 = the earth is flat. Then the probability that the earth is round on e [P(H1/e)] is
greater than the probability of H2 on e [P(H2/e)] if it is more likely that the bottom of a ship will
disappear first on sailing away if the earth is round [P(e/H1)] than if it is flat [P(e/H2)], and the
probability of the hypotheses, H1 and H2, are about the same [P(H1) = P(H2)].
Bayes’s theorem tells us how to compute the probability of a hypothesis on the evidence. Let k
= background evidence that we must take into account besides e, where e usually refers to the rele-
vant observational evidence. Then Bayes’s theorem says: P(H/(e & k)) = [P(H/k) x P(e/(H & k))]/P
(e/k). P(H/(e & k)) is called the posterior probability of H on the evidence and background evidence
(the total evidence); P(e/(H & k)) the likelihood of the evidence on the hypothesis and the back-
ground evidence; P(H/k) the prior probability of the hypothesis on background evidence; and P(e/
k) the priori probability of the evidence on background evidence. Bayes’s theorem allows us to com-
pare the probabilities of two hypotheses, H1 and H2, on the total evidence (e & k) by using Bayes’s
theorem twice, plugging in the relevant probability values on the right side of Bayes’s theorem for
H1 and then for H2.
In principle, Bayes’s theorem allows us to compare the probability of theism with naturalism on
relevant evidence. Suppose we let e = our observations about the amount, kind, and distribution of
goods and bads among sentient beings, and rational beings like us. Let T = core theism, the hypoth-
esis that there is an all-knowing, all-powerful, wholly good being who created and now sustains the
Suffering: Atheism
physical universe. Let N = naturalism, the hypothesis that the mental world either does not exist, or
it does exist but is asymmetrically dependent on the existence of the physical world (Draper and
Dougherty 2013). Naturalism is incompatible with theism since theism holds that there is an intel-
ligent being with a mental life who is independent of the physical world. So which is more probable
on e and relevant background evidence, P(T/(e & k)) or P(N/(e & k))?
To answer this question by employing Bayes’s theorem we need some way of assessing the rel-
evant probabilities on the right side of the equation, in particular, the relevant likelihoods. Paul
Draper proposes a way (Draper and Dougherty 2013). He imagines that two alien beings observe
the course of evolution on our planet. Natty is a naturalist, Theo, a theist. Before any sentient
beings have evolved, Natty and Theo observe the evolution of plant life and discover the ratio of
flourishing to languishing on earth and find it very low. They are then asked to predict the ratio
of flourishing to languishing in sentient beings who can experience pain and pleasure. Draper says
that Natty will predict that this ratio is about the same as the ratio of flourishing to languishing
in plants. Like Theo, she thinks that the goodness of pleasure and the badness of pain are morally
significant, but unlike Theo, she does not think that the course of evolution is being watched over
by a wholly good God. But because Theo believes it is, he will wonder whether God does not alter
the course of evolution to increase the ratio of flourishing to languishing in sentient creatures as
compared to what it is in plants. So Theo will be disposed to predict a higher ratio.
Let’s suppose that it turns out that he is wrong and that Natty is right. Suppose that, having
observed the lives of sentient creatures, Natty and Theo are then asked to predict the ratio of flour-
ishing to languishing when rational creatures come on the scene, and again when rational creatures
that are free and have a moral sense appear. Each time Natty will assume that new morally relevant
considerations appear but that they have no effect on how evolution will go. So she will predict that
the ratio of flourishing to languishing at each stage is similar to what it was at the previous stage,
which ultimately will mean that it is similar to the first stage involving plants. However, Theo will
recognize the appearance of new morally relevant considerations and, because he believes that God
is watching over things, predict at each stage that the ratio of flourishing to languishing will be
higher than at the previous stage. Assume that Natty is right each time and Theo wrong. Wouldn’t
this give us reason to believe that naturalism is true and theism false? Draper thinks it would. P(e/
(N & k)) is greater than P(e/(T & k)) where e includes the predictive successes of naturalism versus
theism at each morally relevant new stage of evolution.
Wykstra is right, I think, in criticizing this argument for comparing naturalism to bare, or core,
theism instead of to some version of theism that combines it with certain auxiliary hypotheses, that
is, to what Wykstra calls special theism. Assume that special theism, because of those auxiliary
hypotheses, can account as well as naturalism for the observations that are made as sentient crea-
tures, rational creatures, and creatures with freedom of the will and a moral sense arise. Won’t spe-
cial theism be as probable on Bayes’s theorem as naturalism, and maybe more probable once the rel-
evant prior probabilities are taken into account?
Here is a relevant analogy that, I believe, will make this point. Suppose Ida holds that the own-
ers of restaurants are indifferent to the cost of replacing broken glassware and plates and that Connie
holds that they are cost conscious about such breakages. Connie and Ida first observe the percentage
of glassware and plates that are broken in inexpensive restaurants and find, say, that it is about 20
percent a year. They then predict what the breakage will be at moderately priced restaurants. Ida
predicts that it will be about the same as in the inexpensive restaurants; Connie predicts that it will
be lower because those restaurants use more expensive glassware and plates: surely owners will
implement policies that will reduce breakage and thereby reduce costs. Assume, however, that Con-
nie and Ida observe that there is about a 20 percent breakage in moderately priced restaurants, too.
Finally, they predict what percentage of breakage there will be at very expensive restaurants. Again,
Ida predicts about 20 percent; Connie predicts that it will be much lower because these high-priced
restaurants use very expensive glassware and plates. But again they observe that the percentage of
breakage is about 20 percent, as Ida predicted. It looks like Ida’s theory, IND, beats Connie’s.
Assume that Connie’s core theory, C, is that restaurant owners want to maximize profits. She
had assumed that the best way for them to maximize profits when they started using more expensive

Suffering: Atheism
glassware and plates was to reduce breakage in order to reduce costs. But observation showed that
the percentage of breakage did not go down as the cost of glassware and plates went up. Connie
could save her core theory about maximizing profits by adopting an auxiliary hypothesis that says
that the least expensive way for restaurant owners to reduce costs was to buy breakage insurance
instead of implementing expensive policies or buying expensive equipment to reduce breakage.
Let Connie’s new special theory be C supplemented by Insurance = owners buy breakage insurance
when they buy more expensive glassware and plates.
Connie’s core theory supplemented by her hypotheses about buying insurance (C & Insurance)
does predict as well as IND. So now the relevant likelihoods when Bayes’s theorem is applied are
the same. However, it seems plausible that the priori probability [P(C & Insurance)/k)] of Connie’s
special theory is greater on background evidence, k, than that of Ida’s [P(IND/k)]. Among back-
ground evidence, k, is the evidence we have that owners of restaurants want to maximize their prof-
its, so of course they will be concerned about the cost of breakage of glassware and plates. That’s
what Connie’s hypothesis says but Ida’s denies. All things considered, it is more reasonable to accept
Connie’s special theory than Ida’s indifference theory. The theories explain equally well the observa-
tion about unchanging percentages of breakage across different types of restaurants, and Connie’s is
more likely on background evidence than Ida’s.
Draper tends to emphasize comparison of the relevant likelihood of the evidence involving the
ratio of flourishing to languishing across different sorts of beings (plants, sentient beings, rational
beings, and rational beings with freedom of will and a moral sense) on naturalism and background
evidence to the likelihood of that evidence on core theism and that same background evidence. But
the glassware and plates analogy shows that we should look at the likelihood of the evidence on spe-
cial theism and background evidence, where special theism consists of core theism plus relevant aux-
iliary hypotheses. That is the relevant competitor to naturalism, not core theism alone.
Of course, that is not the end of the story, as the analogy also shows. We must also consider the
prior probability of special theism, and the priori probability of naturalism, on background evidence
just as we had to consider the prior probabilities of Ida’s theory and Connie’s supplemented theory
on background evidence. But here, I think, is where special theism loses, not on the comparison of
the relevant likelihoods (Dougherty and Draper 2013; Lipton 2004). The prior probabilities of
worldviews like naturalism and special theism have to be assessed, I believe, largely on the basis of
a priori considerations about what makes one hypothesis better than another when they have the
same observational implications as, for example, in the analogous case where the skeptical hypothesis
that we are in the Matrix is compared to the commonsense view that there is an external world
pretty much like what we think it is.
Suppose Swinburne is right that God would not allow there to be some sentient creature whose
life is, on balance, not worth living for that being. In the light of babies born with Tay-Sachs dis-
ease, who often live brief lives full of so much suffering that their earthly lives are not, on balance,
worth living for them, theism must posit an afterlife so that the entire lives of these children are
worth living for them. If we imagine a young fawn that is badly burned in a forest fire and suffers
for more days than it had lived before the fire, and that no human or nonhuman animal observes
the fawn’s suffering, then it seems that theism would have to posit an afterlife for such fawns,
too, to meet the requirement that God not allow lives that are not worth living on balance.
If special theism requires positing an afterlife not only for humans but for fawns, it comes at a
cost. It is hard to understand how there could be an afterlife for people because that requires posit-
ing either the existence of immaterial souls or a physical resurrection world where the molecules that
make up a person are instantly transported by God to that world when that person dies, and that
person is re-created, molecule for molecule, just as he was on Earth. While that “body-snatching”
theory is possible, it seems implausible on our background knowledge of how the physical world
works. And the notion of an immaterial soul is puzzling in itself. First, it is hard to see what the cri-
terion of identity for souls would be. In principle, two different physical people might have exactly
the same mental contents (thoughts, desires, emotions, etc.), but they can be distinguished by hav-
ing two different physical bodies that occupy different places in space. But souls do not occupy
space. So what would make an immaterial soul that exists on one day the same as one with the same
Suffering: Atheism
mental contents that exists on another day, rather than just a duplicate? Furthermore, it is hard to
see how immaterial souls could cause physical events or physical events could have causal effects
on souls. This is a problem for the existence of God, too, since God is an immaterial being. I call
it the mind-body problem writ large. The mind-body problem is the problem of how, if we had
immaterial souls, they could interact with our brains and vice versa.
So while I agree with Wykstra that versions of theism should be compared to naturalism, or
even versions of naturalism, once that is done, I think that the prior probability of these versions
of theism will be low, lower than the prior probability of naturalism, or versions of naturalism. It
is relatively easy to formulate versions of theism whose likelihoods are at least as great as the likeli-
hood of naturalism, or versions of naturalism, understanding “likelihood” in the technical sense used
in discussions of Bayes’s theorem. As Timothy Perrine and Wykstra have said about a hypothetical
example where an auxiliary hypothesis is added to make an unsupplemented hypothesis better fit
the facts, “What we’ve gained with our right hand, we’ve taken away with our left.… There’s no
such thing as a free lunch” (Perrine and Wykstra 2014, 150). Higher likelihoods for special theism
come at the cost of a lower prior probability.
THE PROBLEMS OF DIVINE HIDDENNESS AND

DIVINE INSCRUTABILITY
Daniel Linford, Purdue University
Theism is held captive between two problems, as Odysseus was between Scylla and Charybdis,
without a plausible route through the middle. In this section I discuss two families of arguments:
(i) the problem of divine hiddenness (PDH) and (ii) the problem of divine inscrutability (PDI). The
PDH claims that empirical facts concerning nontheists provide evidence for God’s nonexistence,
whereas the PDI claims that our difficulty in knowing anything about God provides reason to with-
hold belief in God. To explicate the PDH, I begin by summarizing three versions of the argument
(provided by J. L. Schellenberg [1993], Stephen Maitzen [2006], and Jason Marsh [2013]). I then
distinguish the PDH from the problem of evil and consider some objections. Having done so, I cat-
alog the varieties of nonbelief. As I show, a tempting solution to the PDH leads to the PDI.
THE PROBLEM OF DIVINE HIDDENNESS

Divine hiddenness traditionally referred to God’s silence or absence from the lives of many. The
theme is old in theological literature. The Psalmist complains that God has hid His face from us
(Psalms 88:14). In Isaiah 45:15, we read, “Truly, you are a God who hides himself.” Others have
felt that God’s absence poses a threat to their faith or to their sense of existential purpose. A son
of Holocaust survivors says of his mother, “she told me when she’s called before God in final judg-
ment, she will … demand to know why he stood by silently … as her large family was being
destroyed” (“Holocaust Survivors” 2001). Upon visiting Auschwitz, Pope Benedict XVI stated,
“In a place like this, words fail. In the end, there can only be a dread silence—a silence which is
itself a heartfelt cry to God: Why, Lord, did you remain silent?” (Whitlock 2006).
Beginning with Schellenberg’s Divine Hiddenness and Human Reason (1993) and continuing
with his subsequent works (2010, 2015, 2017a, 2017b), the term divine hiddenness has taken
on a different meaning in a family of arguments for atheism. According to the PDH, empirically
observed features of (dis)belief in God’s existence undermine the hypothesis that any such being
exists (see Drange 1993, 1998; Maitzen 2006; Marsh 2013; Howard-Snyder 2006; Howard-Sny-
der and Green 2016; De Cruz 2015). Schellenberg’s most recent version (2015, 2017a, 2017b)
proceeds:
(1) If God exists, there are no nonresistant nonbelievers.
(2) There are nonresistant nonbelievers.
(3) So, God does not exist.

Suffering: Atheism
Why should we think (1) is true? God would be open to and desirous of personal, loving relation-
ships with her creatures. Though some humans may resist relationship, God would never close her-
self off from relationship. The belief that God exists is prerequisite to loving relationship with God.
So, if God exists, everyone who is open to relationship would believe that God exists. Premise (2) is
an empirical observation.
Stephen Maitzen (2006) argues that the geographic distribution of nonbelief is surprising on
theism but not surprising on naturalism. On naturalism, theistic belief is explained as a cultural
product without supernatural intervention, predicting that theistic belief varies with the cultural
forces postulated by the social sciences. Theism, however, needs to explain away nonbelief; typical
theistic explanatory strategies are implausible. Some argue that nonbelievers might not be ready
for a relationship with God. Compare two groups of people, Danes and Texans: one study
reported that 80 percent of Danes are nonbelievers (Zuckerman 2007), while a study of religiosity
in Texas found that only 5 percent are nonbelievers (agnostics and atheists combined; Pew
Research Center 2014). Danes do not seem less prepared than Texans for loving relationship with
God. Other theists concede that nonbelievers would accept God’s existence if God’s existence
were apparent, but would believe for reasons contrary to loving relationship. Again, the right rea-
sons seem equally available to Texans and Danes. Still others propose that the judgment of non-
believers is clouded by sin, disabling nonbelievers from seeing the clear truth of theism. But
Danes are no more sinful than Texans.
Jason Marsh (2013) argues that the evolutionary and historical development of theistic belief
seems incompatible with theism. The cognitive and social sciences have significant success in pro-
viding a naturalistic explanation for the development of theistic belief. Though religion has been
commonplace throughout human history, theistic belief was not. Ancestor worship, animism,
polytheism, and henotheism preceded theism. The cognitive capacities that gave rise to supernat-
ural beliefs are not particularly conducive to theism, as evidenced by the late arrival of theistic
belief in our evolutionary history. Again, if God exists, God is open to and desires loving relation-
ship; the unavailability of the concept of God would undermine God’s desires and purposes. The
earliest populations of humans were probably not more resistant to or less in need of relationship
with God. But if they were, why would God bring about a state of affairs in which her creatures
were resistant, unprepared for relationship, or sinful? God could have created humans in a state
similar to that of medieval Europe, when, apparently, most humans fulfilled the prerequisites
for belief in God.
The PDH and the Problem of Evil. To complete my discussion of the PDH, I turn to distin-
guishing the PDH from the problem of evil (POE). To be sure, a version of the POE could be con-
structed based on some facts appealed to by friends of the PDH. Various goods accrue from rela-
tionship with God. Individuals in loving relationship come to share their moral evaluations;
relating to God would allow one to grow in virtue, come into one’s greatest fulfillment, and under-
stand one’s purpose. One would understand God as a moral exemplar and follow God’s example in
living one’s life. Nonresistant nonbelievers are (apparently) deprived of great goods for no compel-
ling reason, perhaps undermining God’s perfect justness and goodness.
Nonetheless, whether divine hiddenness is an evil is tangential to the PDH. The PDH con-
cerns divine love and not God’s opposition to badness or desire for goodness (Schellenberg
2017a). Those who enter into loving relationships do so because they value relationship for its
own sake and not because the beloved’s well-being improves in virtue of the relationship (though
that may be the case) (Schellenberg 2015, 43). God would seek loving relationship regardless of
whether loving relationship promoted well-being. Van Inwagen (2002) provides a useful thought
experiment. Consider a possible world where all existent evils can be plausibly explained. Some peo-
ple, though not resistant to relationship with God, find themselves unable to believe God exists.
One day, they realize their nonbelief is itself evidence for God’s nonexistence—that is, they have
formulated Schellenberg’s argument without reflecting on evil.
While the PDH is distinct from the POE, the POE bolsters the case for the PDH. Loving par-
ents sometimes inflict justified suffering on their children that the children do not understand, as
Suffering: Atheism
when a child requires a painful lifesaving medical procedure. Even though God’s reasons may not
be epistemically available to us, God might possess sufficient reason for all observed suffering. Still,
loving parents are present with and reassure their children. Parents who neglect suffering children
are not loving parents. Many people who suffer are unable to believe God exists, even though they
desperately desire to believe. Others suffer but live in contexts in which theism is conceptually
unavailable.
Some Possible Objections. Schellenberg’s, Maitzen’s, and Marsh’s versions of the PDH assumed
God would desire and be open to loving relationships. But perhaps God would desire and be open
to loving relationships only with some select group of people and unconcerned for others. Schellen-
berg replies that God is defined as a perfect person—a person such that none more perfect can be
conceived (2015, 96–97). A perfectly loving person is more perfect than a person who either loves
imperfectly or fails to love at all. Therefore, God is perfectly loving. Human beings are sometimes
blocked from relationship because they have finite resources. For example, time spent with one per-
son might conflict with time spent with another. But God has unlimited resources. Nothing blocks
God from relationship with nonresistant persons. Moreover, being perfectly loving involves seeking,
and openness to, relationship for its own sake. Thus, if God exists, God seeks and is open to loving
relationship with all her children.
Others might object that most people have had a general belief in divine beings, so that most
people have had relationships with God. To the contrary, theism was not the inevitable consequence
of our cognitive architecture. The cognitive machinery historically responsible for the development
of supernatural beliefs involved our agent detection capacities, our tendencies toward anthropomor-
phism, and so on—none of these fine-tuned for theism. As Marsh argues, our ancestors came to
theism late in human evolution and largely by happenstance. (Also see the argument presented in
Linford and Megill [forthcoming].)
Perhaps we were wrong to have supposed that loving relationship requires one to consciously
acknowledge the beloved. If loving relationship does not require conscious acknowledgment of the
beloved, then, contrary to appearances, nonresistant nonbelievers might be in loving relationship
with God. For example, if God is numerically identical with the Good, and nonresistant nonbelie-
vers love the Good, perhaps we should understand them to love God. This is implausible. Consider
the following state of affairs. Pam tells us Roy loves her. Prima facie, we have reason to believe Pam.
Nonetheless, we have reason to revise our belief if we learn that Pam is radically mistaken about
Roy’s identity. Roy does not exist; the man with whom Pam has a relationship is Jim. When asked
to describe the man Pam takes to be Roy, Pam does not accurately describe Jim’s attributes and can-
not describe what Jim looks like. Suppose we learn Jim has somehow orchestrated Pam’s confusion.
We would rightfully say this state of affairs is conceptually incompatible with loving relationship.
Nonresistant nonbelievers include ancient polytheists. According to the objection under consider-
ation, ancient polytheists loved God under another name. But, if ancient polytheists did love God
under another name, ancient polytheists were wrong about whom they claimed to love, wrong
about the attributes of whom they loved, and wrongly depicted God in their artwork.
Consider the plausible principle that someone who loves me would not bring about a state of
affairs in which I predictably relate to them inappropriately, especially when the inappropriate
ways of relating are costly. Ancient ways of relating to the divine typically involved practices
now understood as costly or theologically inappropriate, such as the offering of human or animal
sacrifices or the conception of the divine as male or monarchical. The latter is theologically inap-
propriate because the latter assumes a political ideology no longer considered just (monarchism)
together with gender essentialist assumptions (God as male), while simultaneously costly because
conceiving of God as a male monarch potentially legitimates unjust patriarchal political systems
(see, for example, Marcus Borg’s discussion of patriarchal images of God [1997, 64–71, 73]).
(Further examples are offered in Linford and Megill forthcoming.) If God exists, God providen-
tially ensured the conditions antecedent to the historical development of religion, and therefore
ensured that most people would relate inappropriately, and in costly ways, to God. None of this
is compatible with God’s love.

Suffering: Atheism
Cataloging Varieties of Nonbelief. Having explained some initial objections to the PDH, I pro-
ceed to catalog varieties of nonbelief. To enter relationship with God, there must exist some finite
sequence of steps an agent A could perform that, if God were open to and desired relationship,
would culminate with A’s loving relationship with God. Call the state of affairs in which there exists
such a sequence unblocked and in which there is no such sequence blocked.
Blocking factors can be organized into cognitive and sociological categories. A’s relationship
with God may be cognitively blocked in the first sense (CB1) if (i) relationship with God is possible
but (ii) A’s cognitive capacities hinder relationship with God. A’s relationship with God may be cog-
nitively blocked in the second sense (CB2) if no possible finite cognitive capacity would have
enabled relationship with God. In contrast, sociological blocking factors originate with one’s sociolog-
ical context (encompassing the cultural, geographic, and/or temporal context). There are sociological
blocking factors in the first sense (SB1): (i) relationship with God is possible but (ii) A’s sociological
context hinders relationship with God. There are also sociological blocking factors in the second
sense (SB2): no possible sociological context would enable relationship with God. The relationship
between cognitive and sociological blocking factors is complex, depending on, for example, the
intertheoretic relationships between the cognitive and social sciences.
Examples of Cognitive Blocking Factors. CB2 includes instances in which God’s characteristics
disable humans from forming relationship with God. For example, no finite person could literally love
God if God is too incomprehensible. By contrast, instances of CB1 have to do with contingent char-
acteristics of human psychology. On contemporary psychological theory, the capacity for theory of
mind should co-vary with theistic belief; indeed, autistic individuals are observed to endorse theistic
beliefs at a reduced frequency as compared with nonautistic individuals (Caldwell-Harris et al. 2011;
Gervais 2013, 18; Norenzayan, Gervais, and Trzesniewski 2012; De Cruz 2015, 57–58). Or consider
that some Christian theologies have supposed moral conscience to play an important role in religious
conversion. Many Christians suppose that instances of nonbelief are explained by immorality. In light
of the global/temporal distributions of belief, immorality is a bad explanation for all instances of dis-
belief (should we really believe that Danes are more immoral than Texans?). But perhaps immorality
can explain some instances of nonbelief. Moral behavior plausibly requires particular cognitive
mechanisms. As a result of their neuroanatomy, psychopaths have a reduced capacity for moral con-
science and for recognizing their moral failings. Therefore, some individuals may be cognitively disad-
vantaged in their ability to come into relationship with God. Additionally, many theists argue that
moral facts provide evidence for theism. Insofar as psychopaths do not recognize moral facts, they
may be unable to recognize an important argument for theism (Wielenberg 2008, 80–82). So, some
persons may be disadvantaged from coming into relationship with God because they are cognitively
unable to recognize evidence for God (Megill and Linford 2017, 8–9).
Others may be exposed to such tremendous pain, suffering, and misery that their lives no lon-
ger seem worth living; they may be psychologically unable to believe God exists, because they have
an incorrigible conviction that their lives are purposeless and that God would not produce purpose-
less lives. (For an argument that God would not bring about purposeless lives, see Megill and Lin-
ford 2016.) Or they may be psychologically unable to form a relationship with God, because,
though they believe God exists, they cannot come to see God as having more than an instrumental
purpose for their lives. Others may have been horrifically abused by religious officials, or aware of
the history of religious violence, and this may psychologically disable theistic belief. For some abuse
victims, participation in spiritual activities diminishes their well-being, which might foreclose rela-
tionship with God. Traditional theologies have asserted the importance of participation in religious
community for relationship with God. Again, participation in religious communities might be a
nonoption for some abuse victims.
Examples of Sociological Blocking Factors. Instances of SB1 include the cultural unavailability
of theistic belief. Agents may be born before the advent of theistic belief. Agents may, through no
fault of their own, live in societies where theistic belief is forbidden; alternatively, if relationship
with God requires the enactment of particular rituals (e.g., prayer), such relationships would not
be possible if the particular rituals are culturally impermissible. For others, though theistic belief is
culturally available, relationship with God might not be culturally available. Some cultures have
Suffering: Atheism
understood the gods to be morally undesirable or some gods as malevolent. If the gods are morally
undesirable, or if anger toward the gods is permissible, some may be unable to enter loving relation-
ship with God.
Alan Kors (1990) argues that early modern European atheism originated with orthodox theol-
ogy’s endeavor to determine the best argument(s) for God’s existence. Finding that no argument
could withstand scrutiny, orthodox theology’s endeavor to come to know God led into (perhaps cul-
minated with) atheism. The history of science offers a similar lesson. There is no a priori reason for
science’s present secularism. First, according to broad consensus in philosophy of science and reli-
gious studies, the terms science and religion do not carve nature at the joints (Laudan 1983; Cava-
naugh 2009), undermining any a priori argument for their categorical distinctiveness. Second,
according to historiographical consensus, the historical science/religion relationship is complex
(Brooke 1991), not mutually supportive or combative. Past science invoked supernatural hypotheses
to explain natural phenomena and sometimes succeeded—for example, Pierre-Louis Moreau de
Maupertuis developed the principle of least action on theological grounds. Research focused on
the religious or supernatural was pursued in physics (in early modernity), biology (Kitcher 1993,
13–18), geology (Kitcher 2007, 25–42), archaeology (Davis 2004), and psychology (Alcock
1987), but such programs degenerated. Science secularized because religious or supernatural hypoth-
eses bore less empirical fruit than their naturalistic rivals (Papineau 2000, 2016; Boudry 2010).
Those standing on the shoulders of their informed, cultural forebears may be rationally led to
doubt. And those doubts may block relationship with God.
Synthesis. Schellenberg’s focus on nonresistant nonbelievers sells the evidence short. Many
blocking factors produce individuals who are resistant to theistic belief but whose disbelief intui-
tively undermines theism (also see Linford and Megill forthcoming). The examples for CB1/SB1
describe the way people have been contingently organized/situated; God could have done things dif-
ferently. Consider the individuals who incorrigibly disbelieve because they were abused by religious
officials. Abuse victims do not choose to be abused, so their resistance is not the result of their free
choice. Nor is the victim’s disbelief the consequence of the victim’s wrongdoing. The victim suffers
the consequence, while the abuser may continue in the abuser’s relationship with God.
Why would God produce states of affairs of that kind in the first place? Theology offers tradi-
tional answers but was constructed to reconcile theism with the world. As philosophers, we should
instead ask what we would expect if theism were true. God could have created us so that, even if we
were abused, we would not lose our capacity for theistic belief. Similar stories can be told for each
instance of CB1/SB1: for example, God could have ensured that individuals with diminished theory
of mind would not be diminished in their capacity to form relationships with God. Or, by organiz-
ing the world in other ways, God could have ensured that the trajectory of scientific inquiry did not
lead away from her.
THE PROBLEM OF DIVINE INSCRUTABILITY

We’ve been supposing that we understand enough about God to understand consequences of God’s
love. Theists may invoke God’s inscrutability to reject this supposition. God’s love is distinct from
creaturely love and may find expression in unpredictable ways. Perhaps we cannot see how, for
example, the nonbelief of those who are abused or have diminished theory of mind is compatible
with God’s love. That’s to be expected if we cannot generally understand the consequences of God’s
love. Perhaps God’s inscrutability doomed the early modern quest for arguments establishing theism
and similarly doomed theistic research programs. Some cultures may have radically misapprehended
God, but radical misapprehensions of God are to be expected if God is inscrutable. Nonetheless, in
alleviating the PDH, the theist has turned from Scylla to Charybdis. God’s inscrutability generates a
new family of arguments.
According to the problem of divine inscrutability (PDI), God’s transcendent nature undermines
theistic belief. Many traditional theologies understand God to transcend the cognitive capacities,
conceptions, and vocabularies humans ordinarily employ. Three problems follow. First, if God suf-
ficiently exceeds our cognitive abilities, the theist’s God-beliefs were acquired through unreliable
belief-forming mechanisms. We have reason to suspend beliefs we discover we acquired via

Suffering: Atheism
unreliable mechanisms. Hence, if God does sufficiently exceed our cognitive abilities, we have rea-
son to suspend our God-beliefs. Second, if God sufficiently exceeds our conceptions/vocabulary,
we cannot grasp what we mean when we talk about God. If so, we have no reason to think God
exists and we ought to suspend judgment. Third, we turned to inscrutability in part because of
the (apparent) failure of the historical search for arguments establishing theism or of theistic research
programs. If God is inscrutable, both are doomed. No amount of future inquiry would establish
theism. But if so, there can be no rational grounds on which to believe theism.
The theist may seek a moderate position requiring only that facts pertaining to God’s love are
inscrutable. Skeptical theists assert that there are many facts pertinent to loving relationship that are
known only to God. We are not epistemically positioned to judge that what we observe is incom-
patible with God’s love. Consequently, we are not epistemically positioned to know whether the
PDH succeeds (McBrayer and Swenson 2012).
Here are three responses. First, as the skeptical theist agrees, in order for God’s love to comport
with observation, we would have to be sufficiently ignorant of the facts pertaining to God’s love.
But then why do we say God is loving? We couldn’t say that God is loving based on any empirical
observation because that would require knowing what sort of empirical observations are consistent
with God’s love. To be sure, Schellenberg argues that God is loving because God is defined as a per-
fect being. Nonetheless, we can hypothesize a less-than-perfect ultimate person who doesn’t love.
No empirical evidence could decide between the two hypotheses. The only philosophical argument
that could distinguish them—the ontological argument—is largely regarded as unsuccessful.
Second, if God’s love is compatible with most cultures remaining radically mistaken about
God—as the skeptical theist maintains—why should theists think their own religious tradition accu-
rately describes God? On skeptical theism (discussed further below), we couldn’t be epistemically
positioned to judge.
Third, to justify a substantive religious life, theists require a conception of the facts pertinent to
divine love sufficiently robust to explain our God-given purpose and obligations. But such concep-
tions undermine skeptical theism.
On one end of a spectrum, we find anthropomorphic conceptions of God susceptible to the
PDH. In the other direction, God is increasingly incomprehensible. The more incomprehensible
God is, the more incomprehensible God’s love, and so the less susceptible to the PDH. But the
greater the divine’s inscrutability, the more theism faces the PDI.
THE EXPLANATION OF THE DISTRIBUTION OF SUFFERING
Daniel Linford, Purdue University
Suffering in our universe does not pose a special explanatory problem for naturalism. To be sure,
instances of suffering may be implicated in broader anti-naturalist arguments. Naturalists may have
difficulty explaining suffering qualia or the existence of anything, including suffering, at all. None-
theless, naturalistic theories of qualia, consciousness, and cosmology have been offered. Their evalu-
ation is beyond the scope of this chapter.
On naturalism, the explanation of suffering is a job for the cognitive and social sciences, per-
haps in conjunction with political philosophy and philosophy of mind—not a job for philosophy
of religion or for theologians. The distribution of suffering tracks socioeconomic and biological
facts. Those who suffer are not typically morally worse than those who flourish. There is no grand,
cosmic justice. Nonetheless, naturalism does not commit us to nihilism; many naturalists endorse
objective moral facts. For politically or ethically concerned naturalists, suffering provides powerful
reasons for action.
The nature, distribution, and causes of suffering pose a special explanatory problem for theism.
Let’s pump the relevant intuitions. Parasitic flies gestate within the flesh of other animals, painfully
bursting from their hosts. Cats instinctively toy with their wounded prey for extended periods.
People who suffer often do not deserve to suffer. Theists believe that God desires our well-being,
Suffering: Atheism
and knows how to and can bring it about, because God is perfectly loving, perfectly good, all know-
ing, and all-powerful. Nonetheless, suffering exists. Theists have developed two strategies to recon-
cile our world’s spectacle of fear and pain with God’s goodness and lovingness:
1. The story strategy. Some theists offer putatively probable stories (theodicies) in which God allows
some instances of suffering. Other theists offer epistemically possible stories (defenses) where
God and suffering coexist.
2. The skeptical strategy. In this strategy, theists justify rationally maintaining their theism without
explaining suffering.
Various stories have been offered. Perhaps God allows Greg to suffer so that Anne can demonstrate
moral bravery and fortitude in saving him. Perhaps God desires for her creatures to have freedom of
the will, because freedom of the will is a great good, but freedom of the will could not exist if we
did not have the ability to cause each other to suffer. So God allows us to inflict suffering on each
other in order that we may have freedom of the will. Perhaps some suffering is necessary for moral
development. No single story is currently thought to explain all suffering; instead, insofar as theists
can explain suffering, different kinds of stories are expected to explain different kinds of suffering.
Nonetheless, most theists endorse the skeptical strategy because we can easily identify instances
of suffering for which no plausible story has been constructed. Consider a fawn who is trapped
under a log in a forest fire. (This example is from Rowe 1979.) The fawn slowly and agonizingly
burns to death over the course of several hours. No traces of the fawn are left. No one will know
the fawn’s suffering. We cannot see a purpose for the fawn’s suffering. There was no opportunity
to demonstrate moral fortitude in saving the fawn. The fawn was deprived of freedom—if fawns
have freedom—when she became trapped. No one could exercise freedom in rescuing the fawn,
for that would require knowledge of the fawn. The fawn died, so the fawn was not able to grow
in virtue (if fawns do grow in virtue) through the experience. No one else grew in virtue, either.
The distribution of suffering can also seem to lack God-justifying reasons. Consider the follow-
ing eminently plausible moral principle:
Equality: A just society treats person A equivalently to person B, unless A and B differ in some
morally relevant way. (See Linford and Patterson 2015, 191; Aristotle Nichomachean Ethics
1130b–1132b; Gosepath 2011)
Equality explains why racially segregated seating on buses is unjust: race is not a relevant character-
istic for deciding seating. The principle finds support from diverse ethical theories (utilitarianism,
deontology, and virtue ethics; see Linford and Patterson 2015, 191). Political philosophers, across
the spectrum, support equality, though they widely diverge on equality’s implications. The principle
is sometimes understood to follow from necessary principles of rationality (Berlin 1956); if so,
equality will not be overturned in future ethical inquiry.
Equality does not entail that societies afford all persons the same (or equivalent) material pos-
sessions. One way persons can be equivalently treated involves affording them equivalent opportu-
nities. Applying this principle to the theological domain yields:
Divine Equality: Because God is just, God would ensure that persons A and B have equivalent
opportunities to obtain goods or evils, or to avoid evils, unless there are morally relevant
differences between A and B, or unless some factor overrides the demands of equality. (See
Linford and Patterson 2015, 191–192)
The opportunities to obtain goods or evils, or to avoid evils, are not distributed according to morally
relevant characteristics. Amelioration is often not possible for those who suffer the most. People who
have the misfortune to have been born into underprivileged families suffer disproportionately. Moral
progress is made when we recognize that opportunities for obtaining goods or avoiding evils are
unfairly distributed in our world. Those who mistakenly believe opportunities for obtaining goods
or evils or avoiding evils are fairly distributed are guilty of the just world fallacy: they blame vic-
tims—Sue must have done something horrible to deserve what happened to her—and praise evil-
doers—Jake’s actions must have been good, because otherwise he wouldn’t be so rich. The just

Suffering: Atheism
world fallacy motivates morally repugnant theology. If the history of the suffering of people of color
is interpreted as just, then it is difficult to avoid the racist conclusion that “God favored the white
race” (Baker 2011, 168; see also Jones 1998, 19). Likewise, the prosperity gospel is wrong to inter-
pret wealth accumulation as God’s blessing on the virtuous. (Similar considerations for gender are
offered in Overall 2007, 242–243.)
The most plausible story-response involves an appeal to freedom of the will. Consider the case
that, for example, disenfranchised ethnic groups appear to be singled out for suffering. Much of
their suffering is the result of unjust social institutions freely formed by human beings. God recog-
nizes the good of freedom of the will and would not contravene the will of her creatures. This
response is implausible. First, unjust social institutions are formed through the collective actions
of individuals, but no single individual is wholly responsibility for the institutions within which they
participate. (Some might be responsible for, e.g., furthering the agenda of an unjust institution,
however.) Many members of oppressive classes participate in unjust social institutions though they
desire not to. Many lack the power to end the oppression and ameliorate conditions. And many
societies inherit unjust social institutions from their ancestors, so that, for example, racial inequality
persists even if racial prejudice dies out. Therefore present unjust social institutions should not be
understood as an expression of freedom of the will.
Second, the autonomy of oppressive groups often suppresses the autonomy of the oppressed.
Again, the opportunity for amelioration is often not afforded to the world’s most vulnerable people.
If God does recognize the goodness of freedom of the will, why wouldn’t God ensure the freedom
of the will of oppressed classes? Why would God prize the freedom of the will of Nazis and slave
owners above those whom they oppress?
Third, the conditions antecedent to the construction of unjust social conditions often require
widespread misconceptions, or outright fabrications, concerning the oppressed classes, for example,
the anti-Semite’s fantastical claim that Jews eat babies, the horrific white supremacist claim that peo-
ple of color are civilized through slavery, or the dehumanizing partriarchal claim that the place of
women is in the home. When God designed our world, God could have ensured that misconcep-
tions of that sort never arose.
Some entire regions, inhabited by economically vulnerable peoples, or others who lack adequate
resources, are susceptible to plagues, natural disasters, and starvation. Theists invoke freedom of the
will in another way to explain these cases: perhaps a group A living in impoverished conditions
experiences natural evils so that another group B can demonstrate a morally excellent choice in help-
ing A. This response is implausible for two reasons. First, most ethicists recognize supererogatory
actions, that is, those good actions that rise above our moral duties. Some utilitarian views are criti-
cized because they do not seem to allow for supererogation or are otherwise overly demanding (for a
recent discussion, see McElwee 2017). Many regard Peter Singer’s view—that those who are better-
off have a duty to donate as much of their money as they can, without sacrificing something of
equal or greater moral significance, to those worse off—as implausible (1972). If B does not help
A, then God’s purpose in A’s suffering was not met. And if God’s purpose was not met, A’s suffer-
ing would have been for nothing. Therefore, B probably has a duty to help A. If ethicists have been
rightly skeptical of Singer’s view, then we should be skeptical that the world’s most vulnerable peo-
ples suffer so that the better-off may help them. Second, consider the plausible principle that one
ought always to treat others as ends unto themselves and never as mere means. If A is made to suffer
for the sake of B’s ability to demonstrate moral excellence, then A has been used as a mere means
for A’s betterment. God would have failed to recognize A’s humanity. But God is perfectly good,
loving, and knowing; God could not fail to recognize A’s humanity. Accounts according to which
A suffers so that B can grow in, for example, moral or spiritual maturity fail for similar reasons,
for they likewise treat the world’s most vulnerable peoples as instruments for the betterment of
others.
In response, theists should not endorse the story strategy, for known stories lead to the just
world fallacy. Instead, they should concede that opportunities for obtaining goods and evils are
not fairly distributed in our world, while resisting the charge that God favors some groups of people
over others. According to the definition of Divine Equality, if theism is true and opportunities are
Suffering: Atheism
distributed unfairly, the distribution must be due to some reason overriding the moral demands of
equality. If so, the distribution of opportunities seems to lack God-justifying reasons only because
the overriding reasons have yet to be successfully identified.
Most theists have turned to adopting the skeptical strategy. The most popular skeptical strategy is
skeptical theism (Wykstra 1984, 1996; Draper 1996, 2013; Bergmann 2001; Almeida and Oppy
2003; McBrayer n.d.; Dougherty 2008, 2011, 2012, 2016). Roughly, skeptical theism is the conjunc-
tion of two claims: (i) theism and (ii) that we have no good reason to believe our knowledge of the
moral facts, and the entailment relations between the moral facts, are representative of all the moral
facts, and entailment relations between them, that there are. William Rowe (1979) provides a useful
distinction. Inscrutable suffering is suffering for which we cannot see justifying reasons, such as the
fawn’s; similarly, an inscrutable distribution of opportunity is a distribution of opportunities for goods
and evils for which we cannot see justifying reasons. Gratuitous suffering is suffering for which there
are no justifying reasons; again, similarly, a gratuitous maldistribution of opportunity is a distribution
of opportunities for obtaining goods or evils for which there are no justifying reasons. Skeptical theists
argue that, given (ii), we are not epistemically positioned to judge whether an instance of inscrutable
suffering is probably an instance of gratuitous suffering. Likewise, skeptical theists should say we are
not epistemically positioned to judge whether any instance of the inscrutable distribution of opportu-
nity is an instance of the gratuitous maldistribution of opportunity.
Currently, we are concerned with whether successful theistic explanations of the nature, distribu-
tion, and causes of suffering can be offered and not whether suffering provides a good argument
against theism. The two strategies for explaining suffering suggest theists cannot explain suffering.
Most theists have moved to skeptical theism because they have little hope for the story strategy’s global
explanatory success. On the other hand, if skeptical theism is true, theistic explanations for inscrutable
suffering or for the inscrutable distribution of opportunity are beyond our epistemic abilities.
Bibliography
Alcock, James E. “Parapsychology: Science of the Anomalous or Search for Brooke, John Hedley. Science and Religion: Some Historical Perspectives.
the Soul?” Behavioral and Brain Sciences 10, no. 4 (1987): 553–565. Cambridge, UK: Cambridge University Press, 1991.
Almeida, Michael J., and Graham Oppy. “Sceptical Theism and Evidential Caldwell-Harris, Catherine, Caitlin Fox Murphy, Tessa Velazquez, and Patrick
Arguments from Evil.” Australasian Journal of Philosophy 81, no. 4 McNamara. “Religious Belief Systems of Persons with High Functioning
(2003): 496–516. Autism.” Paper presented at the Annual Meeting of the Cognitive Science
Baker, Kelly J. Gospel according to the Klan: The KKK’s Appeal to Protestant Society, Boston, 20–23 July 2011. http://autismodiario.org/wp-content
America, 1915–1930. Lawrence: University Press of Kansas, 2011. /uploads/2011/09/Religious-Belief-Systems-of-Persons-with-High-Functioning
-Autism-.pdf.
Bergmann, Michael. “Skeptical Theism and Rowe’s New Evidential
Argument from Evil.” Noûs 35, no. 2 (2001): 278–296. Cavanaugh, William T. The Myth of Religious Violence: Secular Ideology
and the Roots of Modern Conflict. Oxford: Oxford University Press,
Bergmann, Michael. “Skeptical Theism, Atheism, and Total Evidence
2009.
Skepticism.” In Skeptical Theism: New Essays, edited by Trent Dougherty
and Justin P. McBrayer, 208–220. Oxford: Oxford University Press, 2014. Davis, Thomas W. Shifting Sands: The Rise and Fall of Biblical Archaeology.
Berlin, Isaiah. “Equality.” Proceedings of the Aristotelian Society 56, no 1 Oxford: Oxford University Press, 2004.
(1956): 301–326. De Cruz, Helen. “Divine Hiddenness and the Cognitive Science of
Borg, Marcus J. The God We Never Knew: Beyond Dogmatic Religion to a Religion.” In Hidden Divinity and Religious Belief: New Perspectives,
More Authentic Contemporary Faith. San Francisco: HarperSanFrancisco, edited by Adam Green and Eleonore Stump, 53–68. Cambridge, UK:
Boudry, Maarten, Stefaan Blancke, and Johan Braeckman. “How Not to Dougherty, Trent. “Epistemological Considerations Concerning Skeptical
Attack Intelligent Design Creationism: Philosophical Misconceptions Theism.” Faith and Philosophy 25, no. 2 (2008): 172–176.
about Methodological Naturalism.” Foundations of Science 15, no. 3
(2010): 227–244. Dougherty, Trent. “Further Epistemological Considerations concerning
Skeptical Theism.” Faith and Philosophy 28, no. 3 (2011): 332–340.
Boyce, Kenneth. “Some Considerations concerning CORNEA, Global
Skepticism, and Trust.” In Skeptical Theism: New Essays, edited by Dougherty, Trent. “Reconsidering the Parent Analogy: Further Work for
Trent Dougherty and Justin P. McBrayer, 103–114. Oxford: Oxford Skeptical Theists.” International Journal for Philosophy of Religion 72, no.
University Press, 2014. 1 (2012): 17–25.

Suffering: Atheism
Dougherty, Trent. “Skeptical Theism.” In Stanford Encyclopedia of Linford, Daniel, and Jason L. Megill. “Idolatry, Indifference, and the
Philosophy, edited by Edward N. Zalta. Winter 2016. https://plato. Scientific Study of Religion: Two New Humean Arguments.” Religious
stanford.edu/archives/win2016/entries/skeptical-theism/. Studies (forthcoming).
Dougherty, Trent, and Justin P. McBrayer, eds. Skeptical Theism: New Linford, Daniel, and William R. Patterson. “God, Geography, and Justice.”
Essays. Oxford: Oxford University Press, 2014. Essays in the Philosophy of Humanism 23, no. 2 (2015): 189–216.
Drange, Theodore M. “The Argument from Non-belief.” Religious Studies Lipton, Peter. Inference to the Best Explanation. 2nd ed. London: Routledge,
29, no. 4 (1993): 417–432. 2004.
Drange, Theodore M. Nonbelief and Evil: Two Arguments for the Maitzen, Stephen. “Divine Hiddenness and the Demographics of Theism.”
Nonexistence of God. Amherst, NY: Prometheus Books, 1998. Religious Studies 42, no. 2 (2006): 177–191.
Draper, Paul. “The Skeptical Theist.” In The Evidential Argument from Evil, Marsh, Jason. “Darwin and the Problem of Natural Nonbelief.” Monist 96,
edited by Daniel Howard-Snyder. 175–192. Bloomington: Indiana no. 3 (2013): 349–376.
McBrayer, Justin P. “Skeptical Theism.” In Internet Encyclopedia of
Draper, Paul. “The Limitations of Pure Skeptical Theism.” Res Philosophica Philosophy. N.d. https://www.iep.utm.edu/skept-th/.
90 (2013): 97–111.
McBrayer, Justin P., and Daniel Howard-Snyder, eds. The Blackwell
Draper, Paul, and Trent Dougherty. “Explanation and the Problem of Evil.” Companion to the Problem of Evil. Malden, MA: Wiley Blackwell, 2013.
In The Blackwell Companion to the Problem of Evil, edited by Justin P.
McBrayer and Daniel Howard-Snyder. Draper, “Humean Arguments McBrayer, Justin P., and Philip Swenson. “Scepticism about the Argument
from Evil,” 71–80; Dougherty, “Reflections on Explanation and from Divine Hiddenness.” Religious Studies 48, no. 2 (2012): 129–150.
Draper’s Argument,” 80–87. Malden, MA: Wiley Blackwell, 2013. McElwee, Brian. “Demandingness Objections in Ethics.” Philosophical
Gervais, Will M. “Perceiving Minds and Gods: How Mind Perception Quarterly 67, no. 266 (2017): 84–105.
Enables, Constrains, and Is Triggered by Belief in Gods.” Perspectives on Megill, Jason L., and Daniel Linford. “God, the Meaning of Life, and a New
Psychological Science 8, no. 4 (2013): 18, 380–394. https://doi.org Argument for Atheism.” International Journal for Philosophy of Religion
/10.1177/1745691613489836. 79, no. 1 (2016): 31–47.
Gosepath, Stefan. “Equality.” In Stanford Encyclopedia of Philosophy, edited Megill, Jason L., and Daniel Linford. “On the Unimportance of Theistic
by Edward N. Zalta. Spring 2011. https://plato.stanford.edu/archives Belief.” Essays in the Philosophy of Humanism 25, no. 2 (2017): 8–9,
/spr2011/entries/equality/. 187–207.
“Holocaust Survivors: The Search for Faith.” Religion and Ethics News Weekly. Norenzayan, Ara, Will M. Gervais, and Kali H. Trzesniewski. “Mentalizing
PBS. 2001. http://www.pbs.org/wnet/religionandethics/2001/08/03/august
Deficits Constrain Belief in a Personal God.” PLoS One 7 (2012).
-3-2001-holocaust-survivors-the-search-for-faith/13853/.
https://doi.org/10.1371/journal.pone.0036880.
Howard-Snyder, Daniel, ed. The Evidential Argument from Evil. Blooming-
Overall, Christine. “Feminism and Atheism.” In Cambridge Companion to
ton: Indiana University Press, 1996.
Atheism, edited by Michael Martin, 233–249. Cambridge, UK: Cambridge
Howard-Snyder, Daniel. “Hiddenness of God.” In The Encyclopedia of University Press, 2007.
Philosophy, 2nd ed., edited by Donald M. Borchert. Detroit: Macmillan
Reference USA, 2006. Papineau, David. “The Rise of Physicalism.” In The Proper Ambition of
Science, edited by M. W. F. Stone and Jonathan Wolff, 74–208. 1
Howard-Snyder, Daniel. “Agnosticism, the Moral Skepticism Objection, London: Routledge, 2000.
and Commonsense Morality.” In Skeptical Theism: New Essays, edited by
Trent Dougherty and Justin P. McBrayer, 293–306. Oxford: Oxford Papineau, David. “Naturalism.” In Stanford Encyclopedia of Philosophy,
University Press, 2014. edited by Edward N. Zalta. Winter 2016. https://plato.stanford
.edu/archives/win2016/entries/naturalism/.
Howard-Snyder, Daniel, and Adam Green. “Hiddenness of God.” In Stanford
Encyclopedia of Philosophy, edited by Edward N. Zalta. Winter 2016. Parfit, Derek. On What Matters. Vols. 1 and 2. Oxford: Oxford University
https://plato.stanford.edu/archives/win2016/entries/divine-hiddenness/. Press, 2011.
Jones, William R. Is God a White Racist? A Preamble to Black Theology. Perrine, Timothy, and Stephen J. Wykstra. “Skeptical Theism, Abductive
Boston: Beacon Press, 1998. Atheology, and Theory Versioning.” In Skeptical Theism: New Essays,
edited by Trent Dougherty and Justin P. McBrayer, 142–163. Oxford:
Kitcher, Philip. The Advancement of Science: Science without Legend, Oxford University Press, 2014.
Objectivity without Illusions. New York: Oxford University Press,
1993. Peterson, Michael L., ed. The Problem of Evil: Selected Readings. 2nd ed. Notre
Dame, IN: University of Notre Dame Press, 2017. First published 1992.
Kitcher, Philip. Living with Darwin: Evolution, Design, and the Future of
Faith. Oxford: Oxford University Press, 2007. Pew Research Center, Religion and Public Life. “Adults in Texas: Religious
Composition of Adults in Texas.” 2014 U.S. Religious Landscape Study.
Kors, Alan Charles. Atheism in France, 1650–1729. Vol. 1: The Orthodox
http://www.pewforum.org/religious-landscape-study/state/texas/.
Sources of Disbelief. Princeton, NJ: Princeton University Press, 1990.
Plantinga, Alvin. God, Freedom, and Evil. Grand Rapids, MI: Eerdmans,
Laudan, Larry L. “The Demise of the Demarcation Problem.” In Physics,
1977.
Philosophy and Psychoanalysis: Essays in Honor of Adolf Grünbaum, edited
by Robert S. Cohen and Larry L. Laudan, 111–127. Dordrecht, Rowe, William L. “The Problem of Evil and Some Varieties of Atheism.”
Netherlands: D. Reidel, 1983. American Philosophical Quarterly 16, no. 4 (1979): 335–341. Reprinted
Suffering: Atheism
in The Evidential Argument from Evil, edited by Daniel Howard-Snyder, Van Inwagen, Peter. “The Problem of Evil, the Problem of Air, and the
1–11. Bloomington: Indiana University Press, 1996. Problem of Silence.” In Philosophical Perspectives, 5: Philosophy of Religion,
edited by James E. Tomberlin, 135–165. Atascadero, CA: Ridgeview,
Russell, Bruce. “The Persistent Problem of Evil.” Faith and Philosophy 6,
1991. Reprinted in The Evidential Argument from Evil, edited by Daniel
no. 2 (1989): 121–139.
Howard-Snyder, 151–174. Bloomington: Indiana University Press, 1996a.
Russell, Bruce. “Defenseless.” In The Evidential Argument from Evil, edited
Van Inwagen, Peter. “Reflections on the Chapters by Draper, Russell, and
by Daniel Howard-Snyder, 193–205. Bloomington: Indiana University
Gale.” In The Evidential Argument from Evil, edited by Daniel Howard-
Press, 1996.
Snyder, 219–243. Bloomington: Indiana University Press, 1996b.
Russell, Bruce. Review of Richard Swinburne’s Providence and the Problem of
Van Inwagen, Peter. “What Is the Problem of the Hiddenness of God?” In
Evil. Philosophical Books 41, no. 3 (2000): 222–224.
Divine Hiddenness: New Essays, edited by Daniel Howard-Snyder and
Russell, Bruce. “The Problem of Evil: Why Is There So Much Suffering?” In Paul K. Moser, 24–32. Cambridge, UK: Cambridge University Press,
Introduction to Philosophy: Classical and Contemporary Readings, 3rd ed., 2002.
edited by Louis P. Pojman, 207−213. New York: Oxford University
Van Inwagen, Peter. The Problem of Evil: The Gifford Lectures Delivered in the
Press, 2004.
University of St. Andrews in 2003. Oxford: Clarendon. New York:
Russell, Bruce. “The Problem of Evil: Excessive Unnecessary Suffering.” In Oxford University Press, 2006.
Ethics and the Problem of Evil, edited by James P. Sterba, 90–107.
Whitlock, Craig. “At Auschwitz, Pope Invokes a ‘Heartfelt Cry.’” Washington
Bloomington: Indiana University Press, 2017.
Post Foreign Service, 29 May 2006. http://www.washingtonpost.com/wp
Russell, Bruce, and Stephen Wykstra. “The Inductive Argument from Evil: -dyn/content/article/2006/05/28/AR2006052800453.html.
A Dialogue.” Philosophical Topics 16, no. 2 (1988): 133–160.
Wielenberg, Erik J. God and the Reach of Reason: C. S. Lewis, David
Schellenberg, J. L. Divine Hiddenness and Human Reason. Ithaca, NY: Hume, and Bertrand Russell. Cambridge, UK: Cambridge University
Cornell University Press, 1993. Press, 2008.
Schellenberg, J. L. “Divine Hiddenness.” In A Companion to Philosophy of Wielenberg, Erik J. “Divine Deception.” In Skeptical Theism: New Essays,
Religion, 2nd ed., edited by Charles Taliaferro and Paul Draper, edited by Trent Dougherty and Justin P. McBrayer, 236–249. Oxford:
509–518. Malden, MA: Wiley-Blackwell, 2010. Oxford University Press, 2014.
Schellenberg, J. L. The Hiddenness Argument: Philosophy’s New Challenge to Wykstra, Stephen J. “The Humean Obstacle to Evidential Arguments from
Belief in God. Oxford: Oxford University Press, 2015. Suffering: On Avoiding the Evils of ‘Appearance.’” International Journal
Schellenberg, J. L. “Divine Hiddenness: Part 1—Recent Work on the for Philosophy of Religion 16, no. 2 (1984): 73–93, 152.
Hiddenness Argument.” Philosophy Compass 12, no. 4 (2017a): 1–9. Wykstra, Stephen J. “Rowe’s Noseeum Arguments from Evil.” In The
Schellenberg, J. L. “Divine Hiddenness: Part 2—Recent Enlargements on Evidential Argument from Evil, edited by Daniel Howard-Snyder,
the Discussion.” Philosophy Compass 12, no. 4 (2017b): 1–10. 126–150. Bloomington: Indiana University Press, 1986.
Singer, Peter. “Famine, Affluence, and Morality.” Philosophy and Public Wykstra, Stephen J. “Rowe’s Noseeum Arguments from Evil.” In The
Affairs 1, no. 1 (1972): 229–243. Evidential Argument from Evil, edited by Daniel Howard-Snyder,
126–150. Bloomington: Indiana University Press, 1996.
Stump, Eleonore. Wandering in Darkness: Narrative and the Problem
Wykstra, Stephen J. “Beyond the Impasse: Contemporary Moral Theory and
of Suffering. Oxford: Clarendon; New York: Oxford University Press,
the Crisis of Skeptical Theism.” In Ethics and the Problem of Evil, edited
2010.
by James P. Sterba, 108–140. Bloomington: Indiana University Press,
Swinburne, Richard. Providence and the Problem of Evil. Oxford: Clarendon; 2017.
Zuckerman, Phil. “Atheism: Contemporary Numbers and Patterns.” In
Tomberlin, James E., ed. Philosophical Perspectives, 5: Philosophy of Religion. Cambridge Companion to Atheism, edited by Michael Martin, 47–68.
Atascadero, CA: Ridgeview, 1991. New York: Cambridge University Press, 2007.

TOPIC 17
Science: Theism
Guy Consolmagno
Director of the Vatican Observatory, President of the Vatican
Observatory Foundation
Tucson, AZ
This chapter discusses the relation between religion and science, providing an overview of contemporary
science and the very nature of scientific knowledge. Predominant methods of science are detailed as well as
its proper subject matter and goals, which are essential for addressing questions about whether scientific
knowledge seems to support or to conflict with the sacred texts of theism.
ONE IS NEXT TO NOTHING
In pursuing the theme of the dialogue between believers and nonbelievers, the editors of this book
have specifically charged me to provide “an essay that discusses the relation between religion and sci-
ence, with an overview of contemporary science and the very nature of scientific knowledge,”
including a discussion of whether “its predominant methods, its proper subject matter, and its goals
are essential for addressing questions about whether scientific knowledge seems to support or to
conflict with the sacred texts of theism.”
In particular, they have posed eight questions to me. To begin this essay, I answer those ques-
tions directly:
1. Does science favor theism or atheism? No.
2. Does scientific progress favor theism or atheism? No.
3. Do recent advances in cognitive science favor theism or atheism? No.
4. What should we make of claims by scientists to have made discoveries about religious experience
and religious belief formation? Do they simply reveal some of God’s ways of communicating
with us? No.
5. Do recent neurophysiological studies—for example, of mystical experiences—favor theism or
atheism? No.
6. Do studies of the efficacy of petitionary prayer and near-death experiences (or other paranormal
phenomena) favor theism or atheism? No.
7. Do considerations from astrophysics and physical cosmology favor one view over the other? No.
8. Do considerations from quantum mechanics favor one view over the other? No.
I suspect that these were not originally intended to be questions answered by a simple yes or
no. Even so, I think that it is important to recognize from the beginning that both theism and athe-
ism can find equal support (or lack of support) in science. To put it another way, if the editors were
expecting me to argue in favor of God because of something I find in my science, then I am on the
opposite side of the theism/atheism divide. I do not believe in the kind of God who could be dem-
onstrated by science.
561
Science: Theism
It is not surprising that I do not think the atheists and

KEY TERMS I are so far apart. There are a lot of versions of God out there,
and not only do I think that most of them wrong, but I also
think that they are well worth not believing in. I believe
Bible Belt
in only one God; I reject the others—which is to say, I believe
Big Bang in only one more God than my atheist colleagues. Because we
Church Fathers are not so far apart, it may well be that science is a place
Creation where the worlds of theism and atheism can find common
ground, a theme to which I return in a later part of this
God of the Gaps chapter.
Gospel(s) One reason why I am so curt in dismissing these questions
Incarnation is that they carry hidden assumptions about the nature of sci-
People of the Book ence and the nature of faith—assumptions that I, as a scientist
and believer, do not accept. These are quite common assump-
Religions of the Book tions and assumptions that get in the way of dialogue. Thus,
Religious Right in order to find common ground in science, it is important
Resurrection for both believers and nonbelievers to identify and deal with
these assumptions and to identify assumptions that I believe
Scripture
both sides can accept.
Space-Time From that starting point, I want to examine a new aspect
of the faith-science discussion. I do not propose to examine
faith or science but rather to examine the dialogue, to try to
see what a fruitful discussion between those of faith and those of science might entail. My intent
is not only to examine how believers might engage the nonbelieving scientists but also how we
might engage a notable group who probably would not feel represented by my discussion as a theist:
the evangelical fundamentalists who seem to have rejected science for reasons that both I and the
atheists find false.
THE TWO BOOKS
The idea that the study of nature and the study of sacred texts make up “two books” has a long
and noble history. Indeed, Saint Paul asserts in the first chapter of his Letter to the Romans that
since the beginning of time, God has revealed himself in the things that were created. Olaf Ped-
ersen, in his introduction to The Two Books: Historical Notes on Some Interactions between Natural
Science and Theology (published posthumously in 2007), traces back to the church fathers Origen
and Tertullian the actual imagery of nature as a “book” parallel to scripture that teaches us about
the divine.
This metaphor can be taken in a very misleading way. The request to examine where sci-
entific knowledge seems to support or conflict with sacred texts embodies that misconception.
Those who imagine a war between religion and science do not think of using science and
scripture as books to be read in parallel; rather, they see each as its own Big Book of Facts,
closed and complete. They picture science and faith as two competing sets of truths. What
happens when the facts in one book contradict the facts in the other? Which do you choose
to believe?
This misconception follows from a naive understanding of both faith and science. Neither is
based on rigid certainties that can fit into a book. Indeed, the case is just the opposite. Neither
faith nor science rightly understood can make claims to absolute certainty. To push the meta-
phor further, if faith and science are each books, neither of them is a closed book; they are unfin-
ished books. Science books go out of date notoriously quickly after all, while at the same time,
even after thousands of years, scholars seem to always find new ways to interpret and argue about
scripture.
Science: Theism
FAITH AND CERTAINTY
In Dynamics of Faith the theologian Paul Tillich writes of faith as

an act of a finite being who is grasped by and turned to the infinite … certain in so far as it is
an experience of the holy. But faith is uncertain in so far as the infinite to which it is related is
received by a finite being. This element of uncertainty in faith cannot be removed. (1957, 16)
The writer Anne Lamott, in Plan B: Further Thoughts on Faith, put it more succinctly: “The oppo-
site of faith is not doubt, but certainty” (2004, 256–257). If we had certain knowledge, facts in a
book, then we would not need faith. In the same way, science does not consist of just the formulas
and answers in the back of the book—even if sometimes that is how it is taught in high school.
While both science and faith have doubt as an essential part of their character, Tillich remarks:
The doubt which is implicit in faith is not a doubt about facts or conclusions. It is not the
same doubt which is the lifeblood of scientific research. Even the most orthodox theologian
does not deny the right of methodological doubt in matters of empirical inquiry or logical
deduction. A scientist who would say that scientific theory is beyond doubt would at that
moment cease to be scientific. He may believe that the theory can be trusted for all practical
purposes.… Doubt in this case points to the preliminary character of the underlying theory.
(1957, 19)
He goes on to discuss other forms of doubt, but after dismissing skepticism and cynicism, he recog-
nizes the nature of religious doubt as
the doubt which accompanies every risk. It is not the permanent doubt of the scientist, and it
is not the transitory doubt of the skeptic, but it is the doubt of him who is ultimately
concerned about a concrete content. One could call it the existential doubt.… It is aware of
the element of insecurity in every existential truth. (1957, 20–21)
He concludes, “Serious doubt is confirmation of faith. It indicates the seriousness of the concern, its
unconditional character” (1957, 22).
Scientific uncertainty might involve questions of precision and accuracy. How good are our
data? How complete are our theories? Religious uncertainty in Tillich’s view, however, is akin to
that encountered in attempting to ask questions for which no certain answer can be found: Whom
should we marry? What career should we pursue? Where should we live? All of life consists in mak-
ing crucial decisions on the basis of inadequate information. Yet even though we do not have full
knowledge of the truth, we still have to choose. Just such a choice faces us when we must decide
to admit or reject the presence of God in our lives.
Beyond pointing out the role of doubt in faith, however, Tillich also recognizes that the origin
of faith lies in experience. The time when we feel called upon to make a decision about God is often
recognized as a “religious experience.” Every religious experience begins with, well, an experience.
Every act of faith begins with a new and startling thing that has happened to us: a sudden tragedy
in life or an unexpected and inexplicable moment of holiness. It can sound like the voice of God
on a mountaintop or a still small voice within us that calls us. It can reaffirm us in the choices
we have made or call us to some new and unexpected action. Faith starts with the data point of that
experience in which we are confronted with an event that demands that we make a choice. The idea
that religion depends only on obedience to authority without connection to experience is false.
SCIENCE AND UNCERTAINTY
As Tillich points out, science is marked by uncertainty. He notes that the uncertainty arising from
an imprecision in our data or limitations in our theories is of a nature different from the uncertainty
of making personal decisions. I think he overlooks the fact that science as a human activity is also
subject to the same risk-accompanying doubts as religion. When we identify science only with rea-
son, we miss out some essential points of what makes science what it is. Science is not just a

Science: Theism
recitation of facts (or data points with associated error bars, one sort of doubt) found in a science
textbook. It is not even just the quest for an accurate description, finding the right equation to
describe nature as best it can. Rather, science is the act of a human being looking for a deep under-
standing of nature, and understanding is more than just data.
Merely finding the most accurate equation to match the data is not science. A computer
could do that. Instead, what we seek in science are insights about why nature behaves the way
it does. We might begin by searching for all the different solutions that a computer could give
us to fit our data, but then we have to use human judgment and intuition to decide which solu-
tion seems most likely to lead us to understand the bigger puzzle we are trying to solve. That deci-
sion must be made intuitively, as an act of faith. (This idea is also developed in Stanley Jaki’s The
Relevance of Physics, 1965, 129.)
Consider one of the first examples of rational, mathematical science in action, the description of
the motions of the planets. If all we want to do is to predict eclipses and cast better horoscopes, then
the Ptolemaic system of circular epicycles about a fixed earth can calculate planetary positions just
fine. In fact, if we refine the system by adding more and more epicycles, we can calculate those posi-
tions ever more accurately. It is an example of Fourier’s theorem: an infinite number of epicycles in
a Ptolemaic formulation gives a perfectly accurate description of planetary positions, indeed, one far
more accurate than relying on Johannes Kepler’s ellipses. Kepler’s version of the Copernican system
(elliptical orbits about a fixed sun), however, led to and supported Isaac Newton’s laws of motion
and of gravity. These laws completely changed the way we understand physics. No version of Ptol-
emy, no matter how accurate, was going to do that.
Notice what this means about the nature of science in terms of the “big book” idea. First, it
reminds us that no science is permanent; the book is unfinished. The pre-Copernican system of
planetary positions held for fifteen hundred years, but even that was not fixed and eternal. Our best
science of today may well be seen to be wrong in a thousand years or ten years or tomorrow. There
is no predicting ahead of time when it will come up short. Note that it is not even or only that we
will replace our theories with newer and better theories. If we can learn anything from the history of
science it is that sooner or later our questions will change so much that today’s certainties will be
not just wrong but, indeed, irrelevant.
More essentially, science is not the same thing as “getting the right answer.” It is coming up
with an answer that leads to new insights. It is impossible to calculate ahead of time whether or
not a given explanation will lead to fruitful insights. Indeed, there is no way to know ahead of time
just how useful any given insight is likely to be. Generally, it takes years of work before a clear judg-
ment can be confirmed about the value of a particular piece of work, and even then such judgments
are never final. Yet the scientist must decide, in the absence of this certainty, which explanation is
worth pursuing now—in time for the next grant proposal or the next thesis defense. Such decisions
are just those that Tillich says faith must make without certainty, and they are made on the basis of
something that looks a lot like faith.
As an example, the current state of theories of gravity in twenty-first-century cosmology are
reminiscent of the Aristotelian/Copernican debate of the seventeenth century. Given the evidence
for an accelerating universe, one can either attempt to adapt Albert Einstein’s general theory of rel-
ativity, with perhaps the addition of a cosmological constant, or one can abandon Einstein in favor
of any of a half dozen different exotic theories of gravity. The consensus today favors staying with
Einstein, but then the consensus for most of the seventeenth century favored staying with Aristotle.
REASON AND AXIOMS
There is an even more fundamental way, however, in which reason requires faith. Every system of
reason must begin with axioms. An axiom is generally considered a point of knowledge so obvious
that its truth can be perceived immediately and without proof. The role of axioms is most visible in
mathematics, but, in fact, the existence of axioms, accepted on faith, is fundamental to all systems of
reason.
Science: Theism
One may consider that mathematics is a more pure sort of reasoning system than physics. We
may no longer read Aristotle for his physics, but we still use Euclid’s geometry. In that sense, unlike
science, mathematics never goes out of date. Euclid’s geometry, however, depends on Euclid’s
axioms, and in the two thousand years that followed his work mathematicians have developed all
manner of rich geometries precisely by asking what happens if one or another of those axioms are
relaxed. Do parallel lines never meet? What about on the surface of a sphere?
If science is a system of logic, then it, too, must have its axioms. Three examined here are
the axioms of the existence of reality, the existence of scientific laws, and the innate value of
the scientific enterprise itself. What is more, I argue that these axioms are religious in nature.
As evidence, I point out that these axioms are supported by only a small subset of religions;
one can find religions that do not accept one or another of these axioms. If our choice of
religion affects our faith in these axioms, then only certain religions are going to give the nec-
essary conditions for science to flourish. Let us look more closely at the axioms that we must
accept on faith before we can do any kind of meaningful science—axioms that depend on one’s
religion.
REALISM
To do science, we have to believe in reality. The universe exists; it is not just a dream. Consider the
famous story of the Chinese Daoist Zhuangzi. One day he dreamed that he was a butterfly, and, as
a carefree butterfly, he did not know that he was the philosopher Zhuangzi. When he woke up, he
did not know whether he was Zhuangzi who had dreamed that he was a butterfly or a butterfly
dreaming that he was Zhuangzi.
It is possible that everything is illusion. The idea of the existence of physical reality is something
we have to assume, knowing full well that our ability to perceive that reality is strongly limited by
the limits of our senses. Even quantum physicists have to accept that the physical manifestation of
their experiments in the macroscopic world are real, even as they ponder the nature of the reality
they are revealing in the subatomic universe. Otherwise, there is nothing to study, or at least what
we wind up studying is no longer physics but instead metaphysics. There is more to this axiom,
however, than simply believing that reality is real. There must also be the belief that the reality of
the universe matters. With that comes an axiomatic faith that spending time to become intimate
with this reality is time worth spending.
Ancient India was far ahead of the rest of the world when it came to mathematics; what we call
Arabic numerals were an Indian invention. The Indians, however, never developed the study of the
natural universe the way that the medieval universities did. For their culture, the physical world was
a place of test and trial. For the ancient Chinese, the highest achievement was to be a person of cul-
ture and ethical behavior. For them, as for the Manichaeans in ancient Rome, the physical universe
was the source of temptation, evil, and corruption. We find echoes of that thought even in some
passages of our scripture.
The People of the Book—Jews, Christians, and Muslims—have a belief that this physical
universe was created by our God, who looked at it in each stage of Creation and declared that
it was good. God reveals himself in the things he has created. For Christianity, in particular, we
believe in a God who loved the world—not good people and not nice ideals and ethics, but the
World—so much that he sent his only son not to carry us out of this world but to redeem this
world.
As Saint Athanasius said in his book On the Incarnation around 335 CE, this physical universe
has been “cleansed and quickened” and made sacred by the Resurrection (Section 17). Even the
Christian belief in an afterlife is not a belief in some ethereal nonreal existence. Read carefully what
Jesus says about what happens after death: he promises not a paradise elsewhere but eternal life.
His resurrection, the only example we have of what life after death is supposed to look like, was into
a body more than ours but containing all that ours contains—including its scars, including an
appreciation of a good fish dinner.

Science: Theism
Thus, studying the physical universe for a Christian is an act of worship, not because the uni-
verse is God in a pantheistic way but because it is the expression, the word, of a supernatural God.
Jaki argued (c.f. The Savior of Science) that this was the crucial difference between West and East,
the reason why science was born in Padua and Paris, Cambridge and Berlin, and not in Bihar or
Beijing.
Some religions, including the Religions of the Book, insist on realism. Atheism, on the other
hand, is agnostic about it; it has no inherent reason to insist that the universe is real.
SCIENCE AND ITS LAWS
Another axiom necessary to take on faith in order to do science is that nature follows regular laws.
We have to believe in the existence of these laws before we can look for them. Again, not every reli-
gion thinks that the universe works that way. The nature gods of ancient Greece and Rome did not
behave according to rigid laws external to themselves. Yet by assuming the nature gods, we can, as
with the Ptolemaic system of planet motion, arrive at a complete and accurate description of how
nature behaves.
Consider a scene from an ancient legend made famous in Aeschylus’s play Prometheus Bound.
To quote from one scene, “The earth-born dweller of the Cilician caves … impetuous Typhon
… withstood all the gods, hissing out terror with horrid jaws, while from his eyes lightened a hid-
eous glare, as though he would storm by force the sovereignty of Zeus. But the unsleeping bolt of
Zeus came upon him” (1926, lines 350–360). The nature gods here are Zeus, with the lightning
bolts, and Typhon, the monster who lives in the caves under the mountains of Cilicia. Cilicia is
an ancient kingdom in present-day Turkey, located between the Carpathian Mountains and the
sea. Those mountains are the site of many active volcanoes. Think volcanoes, and listen again to
the description of that monster. He is “hissing out terror with horrid jaws, while from his eyes light-
ened a hideous glare, as though he would storm by force the sovereignty of Zeus”—fire and smoke
rising up into the sky, challenging the realm of Zeus.
Lightning strikes occur all the time in the dust of volcanic eruptions. To us, lightning follows
Maxwell’s equations and Ampere’s law; it is a flow of electricity that is promoted by the presence
of dust in the atmosphere. To the ancients, lightning was a battle between Typhon, the monster
under the mountain, and Zeus, the god in the air who threw lightning bolts. If we think that vol-
canoes erupt because there is a monster under the mountain and that lightning strikes the mountain
because there is a god in the air throwing bolts at the monster, then we already have an explanation
for these spectacular natural events: things happen because the nature gods decide to make them
happen. If we deny the existence of nature gods, then we have to come up with some other expla-
nation for volcanoes and lightning. If we do not reject the presence of nature gods, it would never
occur to us to look for that other explanation. One religion allows for science; the other has no
room for it.
Who was the first person to think there might be laws to describe nature? Who was the first
person who went searching for those laws? Who was the first to find laws that we still think are true
today? Arguably, it was Aristotle. But it is interesting that for a thousand years after him no further
progress was made in physics. The Romans were great practical engineers; I can still see their aque-
ducts out the train window every time I travel from my home in Castel Gandolfo to Rome. Engi-
neering, or building for a purpose, is not the same as science, or studying simply for knowledge.
The Romans made no progress at all in asking questions of how the natural universe works, much
less why it works the way it does. Indeed, the one name in Roman astronomy who is remembered
today, Ptolemy, was actually a Greek living in Egypt, not a Roman. He worked out the motions of
the planets with all their epicycles as a mathematical trick to let him cast horoscopes, which is as
close as astronomy gets to engineering.
Some religions, including the Religions of the Book, deny nature gods and insist on a universe
that is rational. Once again, atheism is agnostic on the question. While it denies gods, it is also free
to deny laws.
Science: Theism
THE SUPPORT OF THE VILLAGE
There is one final axiom that makes science possible. Science happens only when we are convinced
that science itself is worth doing.
Why is this so? First, doing science comes at a cost of many resources, both physical and
human. It happens only when there is a will to allow those resources to be used for science. That
is not an obvious choice. Science is not engineering; it does not exist to put food on the table or
make our lives more comfortable. Even if, in the long run, the knowledge that comes from science
can lead to those results, it is not clear that those who benefit are the ones whose investments make
it possible. Thus, there has to be some other purpose besides pragmatism to pay for the science.
Furthermore, science itself is a conversation about how the universe works. That conversation by
necessity requires more than one participant. Therefore, there must be a significant portion of the
members of a culture interested in pursuing such a conversation.
I noted earlier how in every rational system one must rely on intuition to make crucial deci-
sions—what to study, how to study it, when to persevere, and when to give up. Making these deci-
sions is not all that different from making the other decisions that require choices made with inade-
quate data. How do we make the decisions about what to study, where to live, whom to marry? The
decisions that change our lives are made not only on the basis of inadequate and often incorrect
information; indeed, they are made by ignorant, stubborn, and inexperienced teenagers—us. We
all were teenagers, or at best young adults, back when we made those decisions for ourselves.
In reality, we do not make these decisions in a vacuum. We rely on advice from our friends and
our elders; we listen to the stories of people who have gone before us and made these decisions, try-
ing to learn from their mistakes and their successes. We talk to our parents. We talk to people who
are in a position to know more than we do—authorities—whom we trust because of how well their
advice has guided our community in the past.
Notice how this is similar to the conversations we have concerning religious questions: we rely
on faith in our elders, our advisers, our parents and teachers and priests. We listen to their experi-
ences in making those decisions and, even more, their experiences of those moments when they,
or those they spoke with, experienced the numinous. It is a rare moment that any of us experience
God, so it is important that we pool those experiences, to compare and learn from the similarities
and differences and, just as important, to sort out the real experiences from the spurious ones. That
is why we have religions.
In the same way, scientists who succeed in becoming full participants in the conversation of sci-
ence are those who are willing to spend the time to be taught, by authorities, about the successes
and failures of the authorities in our past. They put their faith in data from the past that describe
unusual events like comets or supernovas. They are willing to submit their work to referees and deal
with the (always unfair!) referee reports. They are willing to sacrifice their most valuable commodity,
their time, to serve as authorities and referees for others. If we do not pay our dues by learning,
teaching, refereeing, and editing, then we are not full members of the scientific conversation.
Instead, we are no better than those folks who send off emails full of Capitalized Words to everyone
we can find, establishing that Einstein was wrong and that the gravitational constant can be proved
to be equal to two-thirds.
In other words, the scientific enterprise is too large for any one person to accomplish. It
requires that many people support and correct one another, many people divide the labor and build
on the labor of those gone before, many people run the universities where the knowledge of the past
is kept alive and the scientists of the future can be brought into the conversation. It takes a village to
do science. We need a community of people willing to waste their time asking these sorts of ques-
tions about rocks and leaves and their origin and their motion. We also need a space where it is safe
to ask those questions, a place where we can be free to admit we do not already have the answers.
Whether it is pagan gods or the physics of Aristotle or some marvelous theorem that came in a
dream, if we think we already know how nature works, we are not going to ask any further ques-
tions. Certainty is the death of science.

Science: Theism
THE RELIGIOUS ORIGIN OF SCIENCE
Where did science get started? The medieval universities for the first time had the magic combina-
tion. They were places where scholars could gather, where the conversation could take place. They
were backed by a culture that was big enough and rich enough to support a small number of smart
monks—Albert the Great being one of the first and Roger Bacon perhaps the most notable—to col-
lect leaves and rocks, perform experiments, ask questions, look for patterns, and describe those pat-
terns in terms of rules. As monks, they had the education and free time to pursue these studies. As
Christians, they had a belief system that had cleared away the easy answers of both the pagans and
the ancient philosophers.
Pierre Duhem, in Le système du monde: Histoire des doctrines cosmologiques de Platon à Copernic,
[1913] 1954) identified a key moment in science in 1277, when Bishop Tempier of Paris, home of
the then-new University of Paris, challenged Aristotelian dominance of science by pointing out that
it could not be the final word, because it appeared to have no room for God’s creative power. For
example, to insist, as Aristotle did, that there could not be other worlds was to deny God’s power
to make whatever he wanted (Duhem 1954, vol. VI, 19ff). Thomas Aquinas eventually reconciled
Aristotle’s philosophy with Christian theology, but (as described in Crombie 1969, Augustine to
Galileo) the impetus to look beyond his physics inspired John Buridan (in 1355) to write about
the motions of bodies in terms of what he called impetus (Crombie 1969, 250); it was the forerun-
ner of Newton’s concept of momentum. Nicholas Oresme (in 1377) speculated about the relative
motions of earth and sky, in anticipation of Nicolaus Copernicus (Crombie 1969, 250). Nicholas
of Cusa (in 1440) even proposed that every star is composed of the four elements (fire, air, water,
earth) making up a world with inhabitants like our own (Crombie 1969, 243–244).
Modern science was ready to be born right then, until the plague killed a third of Europe. The
population of Europe would not recover until around 1600, when at last William Gilbert and
Tycho Brahe and Galileo and Kepler could give birth to what we finally can recognize as science,
with its careful observations of nature described with mathematical laws. Remember, this all hap-
pened in the context of the universities. (Even Galileo did most of his actual science while employed
at universities in Pisa and Padua.) The universities were the places funded by the church to train lea-
ders of the church in philosophy and theology.
Large institutions can do things that individuals cannot. It comes at a cost, of course; everyone
hates bureaucracy. Still, bureaucracy makes sure that a certain minimal amount of work will get done,
regardless of who is in charge or who has called in sick. That is why we put up with big science. That
is why we have universities. There are rare examples like Srinivasa Ramanujan, the remarkable self-
taught mathematician from India who was able to reinvent calculus on his own, but most of us would
rather take advantage of a textbook and a teacher. It saves a lot of time, if nothing else, but it means
placing our faith in that professor, that institution. It also means, at the very least, that somebody has
to feed and house all those nonproductive members of society. We cannot expect a society to support
a culture of science unless it is big enough and rich enough to be able to afford it. None of that would
happen unless the culture decided that it was worth doing.
WHEN THE VILLAGE DISAPPROVES
The presence of a culture that supports science provides the resources for the large project that sci-
ence is and also supports each individual who must choose to go into a life of science. I once gave a
talk at the College of Charleston, a beautiful campus in Charleston, South Carolina, and after the
talk an undergraduate came up full of enthusiasm. “I want to be a geologist!” he told me. “Sounds
great,” I told him. “Yeah,” he said. Then his face fell. “But what do I tell my mom?” He was from
South Carolina, in the Bible Belt. In the culture in which he grew up, our geological ideas of
billion-year-old rock formations directly contradict what he had been taught from the Bible. To
be a geologist, for him, would be like declaring war against his religion, his home, his family. His
mom would be ashamed of him.
Science: Theism
Scientists are people. We have families; we have desires. Like every human being, each of us is a
mixture of reason and heart, with hearts that have—in the famous words of Blaise Pascal—“reasons
that reason does not know” (Pascal 1958, line 277). Like that student, we have to answer not only
those desires inside us but also the desires inside the people whom we love. If our students come
from families who are not proud of them for studying science, they will not study science.
After all, consider this: there are a lot of mountains and snow in India, and more people live in
India than in Switzerland or Norway or Colorado (or, indeed, in all of the United States or Europe).
Yet India has never won any medal in the Winter Olympics. Winter sports at that level just are not
supported by Indian society. There are not enough mothers dreaming that their daughters will
become figure skaters or fathers willing to pay for the equipment or the trips to the mountains to
have their sons train in downhill skiing, and there are not enough competitors or competitions or
trainers to hone the skills of those who might want to do that. If the society a person lives in does
not think that doing science is the sort of thing that will make a mother proud, not very many chil-
dren will choose to study science, much less persevere in their study. Aspiring students also will have
a hard time finding anyone who can teach them what they want to learn, much less anyone who
will want to learn from them.
Science, like faith, is done as a part of a community. If we do not have the support of a com-
munity, it is not going to happen. It cannot happen. Without a community, it cannot possibly be
passed on to the next generation. If the society we live in does not think that doing science is the
sort of thing that will make a mother proud, we will not find very many young people who choose
to study science. We will not find anyone who can teach science. Likewise, we will not find anyone
who will want to learn what scientists have to teach.
Slanders about Galileo aside, the Catholic Church throughout its history has provided just such
support. Likewise, the support of science in classical Muslim cultures is one reason that ancient
knowledge survived to our present time. The scholarship of Judaism has also been supportive of
studying the natural world, as evidenced by the notable number of Jewish scientists throughout
the history of science. If one believes that Creation is the act of one’s God, then studying it is
almost a necessity. In the absence of such a belief, there is no particular imperative in atheism to
study science.
AXIOMS AND BELIEF IN GOD
It is clear from the way I have formulated these axioms that I see in my Christianity precisely the
source of what is needed to support and feed science (while noting that Judaism and Islam should
be comparably supportive of science). Christianity insists that its followers assume that the uni-
verse is real and worth studying. Genesis describes the creation of a universe as real as the God
who made it, and in the Incarnation one finds that this religion recognizes not only that reality
exists but also that it is blessed. Christianity explicitly rejects nature gods and demands that the
only God to be believed in is supernatural not natural and, furthermore, one who is not the cap-
tive of whims but changeless from age to age. In the absence of nature gods, we are left with both
room and necessity to find natural rules as an explanation for how nature works. Moreover, Chris-
tianity provides the physical and intellectual space to allow science to happen. It feeds scholars,
and it also honors them. The great saints in the Christian tradition are not generals or wealthy
donors; even when kings and nobles get the honor of sainthood, it is because they went beyond
their wealth and privilege. The things that saints do to earn sainthood are typically helping the
poor and suffering, feeding the body, or learning and teaching—feeding the soul. Christianity is
a religion that reminds us that we do not live by bread alone. We honor doctors who help the
physically sick; we give equal honor to the “doctors of the church” who respond to our intellectual
and spiritual hungers.
There is a deeper reason why I find in reason a place for my belief in God. Recall that we must
accept axioms on faith before we can reason. Change the axioms, and it is possible to prove or dis-
prove anything. That is why no “proof” of the existence of God can ever be valid, nor can we use

Science: Theism
reason to disprove the existence of God. We do not arrive at God by the end of some chain of rea-
soning. Rather, God—or God’s absence—is the key axiom that determines how our reasoning
develops. It is no surprise that devout people can claim to see evidence of God everywhere they
look; they see it because they assume it is there in the first place. It is wonderfully pious, but it is
not a proof, just a recovering of a key assumption. The same must be said for those who fail to
see God.
It is possible to assume the existence of many different ideas about God and then see if the sys-
tem that results from that assumption is consistent with the evidence of our senses (which is similar
to but hardly the same thing as science). As we have already seen, not every choice of theology
allows for the axioms that science needs. Certainly, many choices do allow and even promote those
axioms. Catholicism is one in particular that does so. By the same token, we can also declare axiom-
atically that we do not believe in God and construct a logically self-consistent system that explains
the universe in terms of, say, accident or rigid necessity. (It is tougher to explain where those laws
of chance or rigid necessity came from, of course.)
How do we choose between assuming the existence of God and assuming God’s nonexistence?
The same way we make any decision based on inadequate a priori data. Study the universe. Listen
to the authorities who have proved trustworthy in the past. At the end of the day, ask which expla-
nation, the one with God or the one without, satisfies both the data of science and the instinct of
the human heart.
A RETURN TO THE QUESTIONS
With this discussion in mind, perhaps we can return to the original questions and provide more
than a snarky one-word answer.
Does science favor theism or atheism? I hope that now readers can see why I say that neither is
favored. The choice to be a theist or atheist is an axiom choice, and it does not affect the results of
the science being done. It does, however, affect the choices an individual makes about what science
is done and how.
Does scientific progress favor theism or atheism? The key difference of this question compared
with the previous one is the word progress. What counts as progress certainly depends on the goals in
mind, and those goals can be tied to the axioms of theism/atheism. Each system has its own goals,
its own idea of success, and it makes those judgments accordingly.
Do recent advances in cognitive science favor theism or atheism? Even without being cognitive
scientists, we can see the fallacy of using its results, like any scientific results, as a way to prove or
disprove God. God cannot be proved or disproved. Therefore, anything proved or disproved by
any science is something different from God. What is more, no science proves or disproves any-
thing; it simply comes up with richer or more complete descriptions. All of those descriptions are
open to further interpretation in the light of future results and in the light of one’s axioms.
What should we make of claims by scientists to have made discoveries about religious experi-
ence and religious belief formation? Do they simply reveal some of God’s ways of communicating
with us? Do recent neurophysiological studies—for example, of mystical experiences—favor theism
or atheism?
Let us say that we do a magnetic resonance imaging scan on the brain of someone undergoing a
religious experience. (Never mind, for the moment, the difficulty of defining such an experience or
calling it up to order for the benefit of our machinery.) If such an experience can be seen to excite a
certain part of the brain, the theist could claim, “See! I told you it was real!” The atheist could
claim, “Aha! What you call a religious experience is nothing more than chemistry in the brain!” If
no such signal was found, the atheist could use that as proof that the experience was not real, while
the theist would use it as proof of the supernatural nature of the experience. In other words, regard-
less of the results, each side would simply recover the assumptions that they brought to the experi-
ment. Such an experiment proves nothing either way.
Science: Theism
Do studies of the efficacy of petitionary prayer and near-death experiences (or other paranormal
phenomena) favor theism or atheism? Not in my understanding of prayer. One of the most subtle
points of spiritual direction is the ability to discern true impulses from God compared with impulses
that have a less lofty cause. One tell for me of divine impulses is that they always come with a cer-
tain plausible deniability. God does not want us to have certainty; he wants us to have faith. Only
then will we invest enough of ourselves in the subsequent decision to make the partnership between
us work.
Do considerations from astrophysics and physical cosmology favor one view over the other?
This question deserves a little more explication, especially because this is closer to my own field of
study.
A few years ago Stephen Hawking claimed that he could answer the Leibniz question—Why is
there something instead of nothing?—by invoking a quantum fluctuation of the primordial space-
time continuum, a bit of gravity so to speak, as the starting point of the big bang. Consider Hawk-
ing’s idea logically. He says that the big bang might be the result of a quantum fluctuation of grav-
ity. He then identifies the origin of the universe at the big bang with the act of God’s Creation. He
concludes that because he can explain that big bang with gravity, there is no such act and thus there
is no such God. By his own reasoning, that is not true. If God is defined as the instigator of the big
bang and the instigator of the big bang is shown to be gravity, then we must conclude logically that
gravity is God. (At this point I am tempted to interject that perhaps he thinks that is why Catholics
celebrate Mass.)
He misunderstands two concepts here. He thinks that Creation is something that occurs only
at the starting point, and he thinks that God is the thing that kicks off everything at the starting
point (and nothing else). A god who was responsible for the big bang, within a universe of preexist-
ing space-time and scientific laws, would be no more than a nature god like Zeus or Typhon,
throwing lightning bolts or causing volcanoes to erupt—one more force alongside gravity and elec-
tromagnetism. Hawking is right not to believe in such a god. No Christian should believe in such a
god either. We rejected the nature gods of the ancient Romans and Greeks (and suffered persecu-
tions as a result), and we must reject this idea of God as well.
The God of Genesis is one who is already present before space and time have been created,
before there are “laws of physics” to be used to explain how things work. Because God is outside
of time, the act of will that makes this universe exist is also an act that is not confined to any one
place or time. Creation is occurring all the time, at every moment. Every second that exists so that
we can be aware of its existence is a miracle; there is no reason within nature itself to explain why
nature exists.
This is a classic point of philosophy and of reason, and it relates to the idea that every act of
reason must start with axioms that are outside the realm of reason to prove or disprove. The mean-
ing of this universe cannot be found within the universe, any more than the reason why there are
chairs can be explained without noting the existence of something that is not a chair, using that
chair to sit down. Incidentally, the Belgian scientist who first proposed the big bang theory, Georges
Lemaître, understood this point very well. He refused to identify the origin point of the big bang
with the act of God’s Creation. Of course, unlike Hawking, Lemaître had also studied theology
and philosophy; along with being a scientist with doctorates in math and astrophysics, Lemaître
was a Catholic priest.
Do considerations from quantum mechanics favor one view over the other? This is the latest
temptation in the naturalist view of religion. Some religious people fear that Newton’s laws lead
to a deterministic universe where no freedom of will could exist, and they find in quantum uncer-
tainties a place to hide free choice of the will after all. This is just another version of the “God of
the gaps,” holding God hostage to future developments in quantum theory, if nothing else. The
theologically more honest answer, both then and now, is to recognize that freedom of will does
exist, whether or not Newton (or quantum theory) has room for it. Nor should one be surprised
that there are aspects to reality that go beyond scientific explanation. Science is human; it is limited.
By design it chooses carefully the phenomena it is capable of describing. It must remain agnostic
about experiences outside of its expertise.

Science: Theism
WHY DOES AN INDIVIDUAL CHOOSE TO BE A SCIENTIST?
Our story does not end here. Making such a choice to accept the existence of a Creator God (and,
what is more, a Personal God who wishes to be in contact with us) affects the other choices that we
make as individuals. In order to understand what science looks like to a believer-scientist, it is per-
haps useful to explore some of the ways that the choice for theism reflects the choice for science.
The effects are not seen in the scientific results themselves. An atheist and I will come up with
the same results to our equations (if we do the math the right way, or at least in the same way).
Where theism makes a difference is at a different level, a meta-level, the level of why we choose
to do science and what we choose to accept as success in that science.
Why does any particular person choose to become a scientist? What benefits does being a sci-
entist impart, benefits that no other career can give? Being a scientist lets us do what? What is it that
we are trying to do when we do science? What constitutes success?
When I was a young man, a few years after my schooling was finished, I entered the US Peace
Corps and trained with a group of about eighty teachers, most of them right out of college and full
of idealism. During the breaks between teacher training and Swahili classes, we used to play a game
called hacky sack. Someone tosses a beanbag into the air, and then everyone takes turns keeping it
up in the air by kicking it or bopping it with their heads. They use anything except their hands
and arms, as in soccer. It is the kind of “cooperative” feel-good activity to be expected of a bunch
of do-gooders like Peace Corps types. During stateside training before we went overseas, the boyfriend
of one of our group came by to visit and watched us playing hacky sack. He was from New York City.
This New Yorker could not figure out what the heck we were doing. “You just kick it up in the air?
No goals? No contact? No trying to stop the other guy from getting it? How do you win this game?”
Science is like hacky sack. It is a cooperative activity. We gain status by how much we give,
what we can add to the field, how much our work helps out the others in the field. We try to throw
lots of ideas up in the air, keep them alive, and pass them on to the next player. There is never a
moment when we say, “I won, and you lost!” What drives an individual to do it? What is one’s per-
sonal goal? How is this game won? What counts as winning?
There are plenty of obvious criteria for what a university might consider to be signs of success,
the sorts of things that lead to tenure. The approval of others is one sign of success; good letters of
recommendation lead to tenure. Tenure itself denotes success. Other markers are grant money or
prizes, honors, and awards, among them, for example, a MacArthur Fellowship or a Nobel Prize.
Fame in general is a mark of success, like being mentioned in the New York Times. Another sign
of success is having successful students. Like parenthood, sometimes we live through our offspring.
The favorite metric of tenure committees is whether a person has written papers that other people
use. The more papers or the more citations, the more a person is seen as successful.
Those may sound a bit crass, and clearly they do not show the whole picture. Another carrot
driving a scientist to excel is the hope of academic freedom. In other words, people are deemed suc-
cessful if they have reached a state where they can set the criteria for what they want to study, if they
are free to choose their own topics and are not just working for someone else. A successful scientist
is not just an employee or a student and is not limited by what the National Aeronautics and Space
Administration (NASA) or the National Institutes of Health (NIH) will pay for but rather a success-
ful scientist is one who is in a position to sit on the committees that decide what NASA or the NIH
will pay for.
Notice what that means: the reward for winning at science is to be able to do more science. It
avoids the question of why anyone finds it rewarding to do science in the first place. Indeed, are any
of these things ends in themselves? Look at the table of contents of any recent issue of Science or
Nature or any other top scientific journal. Having a paper published in Science is a sign of success.
But does anyone wake up in the morning and jump out of bed to rush to the lab because by golly
they are going be one of the “et al.” authors who will tell the world something new and exciting and
groundbreaking about “the effect of collective molecular reorientations on Brownian motion of col-
loids in nematic liquid crystals” (to quote the title of one recent Science paper)? It is important
work—really, it is—and it may be cited twenty times in the next three years.
Science: Theism
Is that what truly motivates the individual scientist from moment to moment? What are the
personal motivations that one might conceivably have for working in science? Curiosity is one.
There is also pleasure to be found in solving problems or merely in finding patterns. That is cer-
tainly what makes doing jigsaw puzzles or crossword puzzles so much fun. But surely science is
about more than that. One obvious central pillar of the doing of science is the search for truth.
At the end of the day, there is more to truth than merely finding the answer that solves the puzzle.
Here, for the first time, we get a glimpse that something akin to a desire for God is at the center of
all science; that is, if we worship the God who is, as the Gospel of John tells us, “the way, the truth,
and the life” (14:6). To put it another way, if we did not experience love in the work, we would not
do it. God is love.
How do these criteria—pleasure and truth and love—compare against the first set? Which can-
didate would a tenure committee choose—the faculty member who loves the material or the one
who publishes more papers? Could you argue with their decision? In real life, it is never such a clean
choice. Would we give up the approval of others to satisfy our own curiosity? Or vice versa? Give up
tenure to do the science we love? Or vice versa? Give up academic freedom, the freedom to pursue
our own research, to satisfy our pleasure in solving problems? Or vice versa? Give up fame to satisfy
the pleasure in finding patterns? Or vice versa? Give up grant money to satisfy truth? Or vice versa?
Where is one’s heart? Where is one’s God? Before we jump to one conclusion or another, remem-
ber that if we are not getting paid to do the work (assuming we are not independently wealthy), then
we cannot do the work; without eating, we cannot do any work at all. If we are not doing work that
the culture respects and supports, then we are not going to have very many people to engage in that
conversation that we call science. At the very least we are going to be facing the scorn (or worse) of the
people whom we rely on to support us emotionally as well as financially. What is the ultimate crite-
rion? Without internal motivations, without some sort of internal payout, we would not do this work
very long; without external motivations, we would not be able to do the work.
It is safe to say that the goal of science is the search for truth. I suspect, however, that what gets
us up in the morning is something else. After all, why do we play hacky sack? We play because it
brings players together in a bonded community with a common goal; it helps nourish friendship
among the players. It is also fun.
SCIENCE AS JOY
I remember once, a few years ago when I had a sabbatical year teaching physics at Fordham Univer-
sity, I had a class of bright students taking the introduction to electricity and magnetism. We had
covered Maxwell’s equations, and I was writing them on the board in front of the class, doing the
mathematical manipulations that James Clerk Maxwell had done back in 1865 on how the equation
for electric fields gave rise to magnetism and how magnetic fields could give rise to electric fields.
Then, if we took a derivative here and put in a substitution there.… As I wrote down the final
equation, the result of all this manipulation—a complicated scrawl of E’s and t’s and del’s and
mu-sub-zeros—and before I had a chance to turn around and explain what it all meant, my bright-
est student in the front row whispered, under his breath but loud enough for everyone to hear him,
“Oh my God. It’s a wave.”
Every bit of science we can extract from our telescopes starts with Maxwell’s equations and the
fact that—Oh My God—it is a wave. The fact that it is a wave gives us radio; electric power trans-
mission of alternating current; and, eventually, special relativity. It takes a couple of semesters of
physics to get there, but when we get there—take my word for it and take my student’s word for
it—it is an Oh My God moment. In my forty years of research, I had had a handful of those
moments. Nothing was as big as Maxwell’s, of course, but a couple were big enough to publish
in Science. It is not the final paper that I remember; it is the gasp of amazement when suddenly I
saw a pattern in nature that I had not anticipated.
Where is God in science? Science is about the universe—everything. Likewise, God is omni-
present. Science is about truth; as Christians, we follow a God who tells us that he is “the way,

Science: Theism
the truth, and the life.” Most immediately, science is a source of satisfaction and joy. Ignatian spir-
ituality tells us that God is the source of “consolation … without previous cause” (Ignatius of
Loyola 1951, 330). It is a joy that surprises.
CARRYING THE CONVERSATION FURTHER
I have no doubt that many atheists who pursue science do so for exactly the same reasons I have
stated here: the desire to know, the joy of knowledge. I know very few scientists of any religious
stripe who would fake their data, knowing that they could get away with it until after their grant
has been spent and they are long dead—or, at least, I know of no scientist who would think that
such an act was a praiseworthy act of science. Lies are not science. In that sense we have the same
motivations, and those motives are directed toward something transcendent, not measurable, like
truth or joy.
The difference between the atheist and me, however, is that I identify this desire for something
transcendent, something more than material gain, as a desire for God. The meta-reasons underlying
science are for me exactly the pointers that direct me toward God. The very thing that makes sci-
ence worth doing and desirable to do are the places where I see God. It is in that common ground
that I find the hope for conversation and connection.
One goal of this book is to foster communication between theists and atheists. As a traditional
believer and a traditional scientist, I find that I have more than one front to fight in this battle. For
this reason, I want to end this chapter with a discussion of what I have learned so far in promoting
such dialogues not only with the atheists but also with the fundamentalists. The parallels, to me, are
striking.
I was once asked to give a lecture on how to talk on frequently answered questions on science
and faith, and the answers that satisfy. As with this chapter, I tried to fulfill the charge of the title,
but I found that I do not always have answers that satisfy. What I found that I do have, however, is
a lot of experience dealing with the public: twenty-five years at the Vatican Observatory, serving as
an example how one can be both a scientist and a person of faith.
As it happens, someone in my position, a very public scientist and a Jesuit (working at the Vat-
ican no less) is likely to find himself under attack from two different sides. There are, of course, the
skeptics who think that a real scientist must reject religion. They tend to be suspicious either of my
scientific bona fides or, even worse, of my religious sincerity. (“He’s figured out a clever way to get
funding for his research,” I overheard one fellow comment.) Even more problematical in my expe-
rience are those people who think they are being devout Christians by rejecting my science.
How, then, do I answer tricky questions, especially questions from folks whom I consider my
scientific colleagues or fellow believers? I do not claim to be a master or exemplar on the matter,
but I know someone who was. What would Jesus say? He spent three years of his life wandering
the Holy Land, engaging both the ultra-devout and those outside religion, in the temple and in
the market square. How did he handle the questions he received?
Of course, unlike Jesus we do not meet in the temple or the market square or even the parish
hall to debate points of theology. That is not to say we do not have such debates. The difference is
that nowadays we find these kinds of conversations online. In fact, the verse-by-verse way that these
stories are told in the New Testament reminds me of the conversations I see on Twitter. I can just
see the scribes tweeting: “Imagine 7 brothers; the 1st married, died childless; his widow marries his
brother; the 2nd did the same, also the 3rd, down to the 7th. Last of all, the woman herself died. In
the resurrection, whose wife of the seven will she be? For all of them had married her. #gotcha!”
Jesus tweets back: “You know neither scripture nor God’s power. In the resurrection they neither
marry nor are given in marriage, but are like angels in heaven. Haven’t you read ‘I am God of Abra-
ham, Isaac, Jacob’? He is God not of the dead, but of the living. #notsofast.”
Notice what Jesus does. First of all, he does not hesitate to engage those who would try to trap
him. There is a time when it is better not to feed the trolls; there are other times, however, when
the trolls must be confronted. When they bring the argument to him, he is not afraid to respond.
Science: Theism
That is his job, and he is good at it. (That does not mean that all of us must do that; we may not be
as good at it as he is.)
Next, just as on Twitter, he is aware that he has two audiences. He is speaking to his oppo-
nents, but he is also speaking to the wider audience that is listening. Sometimes we tweet (or argue)
directly with the person talking to us and forget that each of us has a hundred or a thousand fol-
lowers, with more who might find the retweets, and that audience may not know the backstory
to our argument. Are we friends bantering, familiar with each other’s sense of humor, or are we
strangers speaking rudely to each other? Even though we may be confident that the person to whom
we are talking knows scripture, or science, we can in no way be confident that the other thousand
followers (if we should be so popular) understand the references. We have to spell them out. Even
worse, of course, is playing only to the audience in our tweets and forgetting that the poor person
who first asked the question, who might be sincere, has merely become the butt of our clever cut-
ting humor. We forget that they, too, can bleed.
Finally, notice how Jesus wins this argument. He changes the grounds. He does not attack the
logic of the scribes and Pharisees. These are smart people; they can reason as closely as anyone can.
Instead, he attacks the assumption that they made. Who says the widow has to be anyone’s wife?
Indeed, most arguments are based on assumptions made without even realizing they are being
made. Then he pulls back from the confines of the argument itself to the larger frame. The question
of the Resurrection is not about who gets to be married to whom. It is a much bigger question: it is
a question about the nature of God. Who is God? Who does scripture say that God is? If our con-
clusion runs against something as fundamental as that, then all the cleverness in the world will not
save us. We have made a mistake somewhere.
CONVERSING WITH THE RELIGIOUS
Notice another thing about encounters like these. For the most part, Jesus gets grief not from those
who have lost faith but rather from those who think they have already got faith and do not have to
look any further for it.
This parallels my own experience. Hostility from scientists does still happen, of course. Still,
except for a few aging white males (who tend to be biologists and British) who are found more often
on TV than in the lab, in my experience the scientists who do not follow a faith are more likely to
be quietly amused by my odd behavior than to show any overt hostility. For example, Neil deGrasse
Tyson, the host of Cosmos: A Spacetime Odyssey, is hardly a friend of religion, but he is a friend of
mine, and he is not afraid to share that friendship in public.
Instead, I find that the most troubling place where the science/faith debate occurs is in the pews.
Our problems start with the “pharisees” of our own faiths who want to secure their sanctity by heap-
ing abuse on science. That makes me notice an interesting aspect in the example of Jesus engaging his
Pharisees. That was risky for him to do, and he knew it. He did not have to talk to them. There were
plenty of religious sects of his time who separated themselves from the general run of society and its
authorities and hid out in monasteries in the desert. Likewise, it would be easy to just let the antisci-
ence faithful of our own faiths have their own way, to let them continue with their foolish preconcep-
tions about science, as long as they do not bother us. Jesus did not do that. Neither should we.
Recall my story of the student from College of Charleston, South Carolina. To a Catholic like
me from the urban North of the United States, South Carolina evokes images of those terrible
southern fundamentalists. It is easy and tempting to merely dismiss such people as ignorant. Too
easy. Too tempting.
For most of us, the failures of fundamentalism are so obvious that it is boring to even talk
about them. I remember one time, during the 1990s, when I was traveling through Italy by train.
As an American I had bought a rail pass that let me travel in first class. Of course, the only other
people one would meet in first class were other Americans also traveling on rail passes. As it
happened, in the car with me was a couple from Cambridge, Massachusetts, home of Harvard Uni-
versity and the Massachusetts Institute of Technology (MIT), and all they wanted to talk about was

Science: Theism
American politics. I wanted to look out the window and enjoy being in Italy; they wanted to gripe
about the Religious Right.
I had lived in Cambridge for fifteen years. I had studied at MIT and taught at Harvard. There
was not a thing they were saying that I had not heard a million times. In fact, I already agreed with
most of it. How boring! So how could I shut them up? Because I knew their culture, I knew just
what to say. I put on my most serious face and intoned, “You realize that the Religious Right have
been marginalized by the intelligentsia in America.” Oh! A marginalized minority!
It did shut them up. It also shut me up, because as soon as I had said it, I realized that it is
true. Those of us who are bleeding heart liberals—I am a proud member of that tribe—need to
remember it and treat the fundamentalists with the same care we would treat any other marginalized
population. With respect. With careful listening. With the recognition that no matter how “primi-
tive” their strange rites and rituals may seem to us, maybe we can learn from them some things we
have lost in ourselves.
SCIENCE AND CULTURAL IDENTITY
Notice an important element of that story. I knew the attitudes of the Cambridge couple because I
was one of them. I was also different from them. As a result, I was very hesitant to dare to express,
in that small space on the train, anything that I knew would be a red flag to them—like my own
religion. I did not want to be ostracized. I liked Cambridge. I liked living there. I admired the peo-
ple I met there. I wanted to be identified as one of them. That is why I tried to divert their circu-
larly endless discussion with what I had thought was a joke.
It turns out that that sort of cultural identity is an important element in any conversation we
have with people whose ideas about anything, much less of science and religion, are different from
ours. Yale University’s law school hosts the Cultural Cognition Project. They have an excellent web-
site exploring how cultural values shape public policy arguments and decisions. When they look at
the great polarization between political groups and the ways different groups deal with questions of
science, they come up with surprising results.
First, we have all seen how some people tend to bend scientific evidence to fit their own pre-
conceptions. By “some people,” we generally mean those with whom we disagree. It is absolutely
true; it happens. It is not the result of some sort of conservative penchant for authoritarianism; lib-
erals do it too. It is not a case of some people using too much emotion instead of rationality when
they approach these issues; in fact, one surprising result from the Yale study was that the more ratio-
nal one’s thinking, the more likely one is to take a polarized position. In fact, such bending of the
facts comes about precisely as the result of reading the evidence rationally but entirely in the light of
previous assumptions. We see all evidence as a way of validating that which we already believe or
want to believe, because it is a marker of the social group with which we identify.
That is exactly what we do in science all the time. For example, we scientists know the laws of
thermodynamics; they are now a part of the set of assumptions that make up our scientific cosmol-
ogy. When someone brings us a supposed perpetual motion machine, we know it is fake. When we
devise an experiment of our own, it already has built into it many assumptions about what sort of
result we are likely to see. If we did not have that idea, we would not know what to look for or
how to recognize it when we saw it. All data is theory-laden. More than that, we all use these data
as a way of reinforcing our identity with our colleagues who hold the same worldview that we have.
As Dan M. Kahan, one of the researchers from the Yale group wrote, “Ideologically motivated cog-
nition is a form of information processing that promotes an individual’s interest in forming and
maintaining beliefs that signify their loyalty to important affinity groups” (Kahan 2013, 407).
I truly believe that fundamentalists are wrong: wrong in their science, wrong in their religion.
My belief is so firm that I have to recognize it derives in no small part from the community of scho-
lars, both scientists and educated Catholics, who hold the same beliefs I do. I rarely bother question-
ing the wrongness of fundamentalism any more than as a scientist I bother worrying about testing
the laws of thermodynamics. I have to recognize that the fundamentalists believe they are right,
Science: Theism
for equally strong affinity-group loyalties. It is important for that boy in Charleston to be able to go
home to a mom who will be proud of him, just as it is important for me to be able to go back to
MIT without hearing people whisper behind my back about my religion.
How do we get around this conundrum? How does one communicate across different affinity
groups? The Yale group suggests that the way to do it is to find areas of common viewpoints
between ourselves and those with whom we would argue. Jesus relies on scripture when speaking
to the scribes, but he does not when he speaks to a Roman centurion elsewhere in the Gospel.
Of course, it might seem tricky to use religion or science as our common points, because those
are the very points under dispute. In fact, those might be the very grounds that we can share.
SCIENCE AS COMMON GROUND
Perhaps this desire for science is a common ground where we can meet. A couple of years ago, the
Vatican Observatory hosted a group of scholars from Iran. This followed a visit that our director at
the time, Father José Funes, had made to Iran. Even though our approaches to science and religion
questions are different, the simple fact that these meetings took place points out a fundamental fact:
among people of different religions (and no religion), our common love for science is a place where
we can begin to engage one another, to learn to understand, trust, and eventually even like one
another. As astronomers put it, we all live under the same sky.
People of every religion love science. They recognize the same joys we do at the beauty of a
nebula in a telescope or the intricacy of a living cell, but they want to be able to learn science from
people who do not threaten their core identity. Let us face it, most of the public faces of science on
TV or the other media are also publicly atheists, who often express specific hostility against religion.
Tyson is my friend. Hawking was a friend of the late Father Bill Stoeger, an astronomer at the
Vatican Observatory—they had worked together at Cambridge University—and he was a proud
member of the Pontifical Academy of Sciences. These atheists are not evil people. If they have naive
views of religion and philosophy, well, so do most people. (Professional theologians would say the
same of me.) All too often the way they present science looks like an attack on religion. No wonder
the fundamentalists do not trust them. On such matters, I do not trust them either. Opposed to
such a point of view, I guess I am on the same side as the fundamentalists.
THE FUNDAMENTALIST FALLACY
I admire the faith of the fundamentalists, the commitment that they have to hold fast to their faith
in the face of strong societal pressure. Those are admirable traits. So why do I not join them? Why
am I not a fundamentalist? It is not that fundamentalism is provably wrong. Science is not in the
proof business (a point both fundamentalists and atheist scientists often forget). A thousand years
from now, some new way of understanding the laws of physics may suggest that the world is just
six thousand years old. It is not likely, but I cannot rule it out. Even if that were to happen, it
would not mean that the fundamentalists were right.
The problem with fundamentalism is not that it gets this fact or that fact of science right or
wrong. I get various facts of science wrong all the time. That comes with being a scientist. Rather,
the real fault in fundamentalism is that it insists that its set of “facts” is unassailable, and unlike real
science it leaves no room for growth. We return to the fallacy of the two big books. Science must
always say, “I may know nature pretty well, but I can always know nature better.” Religion must
always say, “I may know God pretty well, but I can always know God better.” Both require the abil-
ity to admit the need to grow. Both require the strength to admit when we are wrong—or, more
commonly, when we were right but we had only part of the story. The problem with fundamental-
ism is not that it is wrong; I am used to being wrong. The problem is that it is certain it is right,
and certainty is the opposite of faith.
Yet fundamentalists have great confidence in their religion because, in their day-to-day lives, it
works. It describes right and wrong, good and evil, how to live and how to die, far better (in their

Science: Theism
opinion and mine too) than our modern penchant for making it up as we go along, taking our cues
for life from pop music and Hollywood movies. They have no reason to doubt it any more than we
have reason to doubt conservation of energy. When they hear things from us that attempt to chal-
lenge those beliefs, we are the ones who have the burden of proving both that what we say about
Creation is true and that it in no way violates what scripture says about Creation. Scripture says,
in many different ways and in different books written at different times that God created the uni-
verse. We need to remember that and remember to say that our science merely tells us (in many dif-
ferent ways and in different theories devised at different times) how he did it.
QUESTION AUTHORITY? SAYS WHO?
A fundamentalist relies on authority, but so does science. We cannot possibly derive every equation,
produce every data point, define every term anew by ourselves. (I would not trust my own work
enough to even try.) We trust the giants upon whose shoulders we, like Newton, must stand. Those
of us in science are familiar with authority, and we respect authority precisely because we are authorities
ourselves. It is an attitude familiar to us from the story of the centurion’s servant, as told in Matthew:
A centurion came up to Jesus, asking for help. “Lord,” he said, “my servant lies at home
paralyzed, suffering terribly.” Jesus said to him, “Shall I come and heal him?” The centurion
replied, “Lord, I do not deserve to have you come under my roof. But just say the word, and
my servant will be healed. For I myself am a man under authority, with soldiers under me. I
tell this one, ‘Go,’ and he goes; and that one, ‘Come,’ and he comes. I say to my servant, ‘Do
this,’ and he does it.”
When Jesus heard this, he was amazed and said to those following him, “Truly I tell you, I
have not found anyone in Israel with such great faith. I say to you that many will come from
the east and the west, and will take their places at the feast with Abraham, Isaac and Jacob in
the kingdom of heaven. But the subjects of the kingdom will be thrown outside, into the
darkness, where there will be weeping and gnashing of teeth.” Then Jesus said to the
centurion, “Go! Let it be done just as you believed it would.” And his servant was healed at
that moment. (8:5–13, NIV):
It is a lovely illustration of how someone who is used to authority can understand and appreciate
authority in others. Jesus teaches with authority, we are told in many places in the Gospels. That
understanding of authority is a common ground between Jesus and the centurion. He speaks to
the centurion in a different way than he speaks to the scribes and Pharisees. He does not cite scrip-
ture or argue points of theology. He also does not reject the centurion just because he is a Roman.
He takes the centurion for who he is; meets him where he lives; and, indeed, even literally offers to
go to where the centurion lives.
Notice also, once again, how Jesus is aware of both the person to whom he is talking and the
audience that is overhearing them both. He uses the opportunity to teach the crowd, making it clear
that outsiders like the centurion are closer to the kingdom than the self-righteous in the audience.
That is something for us believers to think about; it may well be that our closest allies in the long
run are the skeptics and the atheists. (Also notice one last but important thing. In all the teaching
that Jesus does at the moment, he does not ignore the concern at hand—he does get around to cur-
ing the centurion’s servant.)
THE FOUNDATION OF ATHEISM
When it comes to understanding atheism, I rely heavily on the work of Father Michael J. Buckley. At
the Origins of Modern Atheism is his great work, widely read and widely quoted and justly so. One of
Buckley’s key insights is that “a bond of necessity stretches between [theism and atheism]: atheism
depends upon theism for its vocabulary, for its meaning, and for the hypotheses it rejects” (1987,
15). In other words, to be an atheist one must have a clear idea of the God one does not believe in.
Science: Theism
I remember seeing a photograph of one of the world’s most famous atheists, the elderly British
biologist Richard Dawkins, posing with a smirk on his face in front of the advertising he had put on
the side of London buses as a publicity stunt a few years back. The sign on the bus read, “There’s
probably no God. Now stop worrying and enjoy your life.” To show what he meant by enjoying
life, he had an attractive young woman draped around his shoulder; her T-shirt bore the same slo-
gan. Notice the idea of God that he assumes. To Dawkins, God is primarily a source of worry,
someone who would get in the way of our enjoyment of life. Also notice that his idea of enjoying
life is to have a “babe” on his arm; meanwhile, her idea of success is having a famous sugar daddy.
There is more than one bit of advertising going on in this photo.
Why does the popular culture, which Dawkins and his friends embrace and exploit, equate
science with atheism? After all, while Kepler and Newton had pretty odd ideas of theology, nei-
ther one of them would have accepted or appreciated being called an atheist. Both were strong
believers in their own versions of God. Buckley attempts to answer that question historically.
He argues that modern atheism can be traced to the very attempts to use the new certainties of
modern science (the stuff we now recognize is not so certain) as a means to prove that God indeed
does exist.
For example, Newton laws of motion and gravity described the motions of the planets around
the sun in a way no one had ever been able to do before, but Newton could not explain why they
did not disturb each other with their own gravities. Instead, he used the obvious stability of the solar
system as evidence of God’s benevolent interference. When the French mathematician Pierre-Simon
Laplace developed a far more advanced mathematical description of planetary orbits, a hundred
years later, that “proof” of God’s action was made moot.
Likewise, when Aristotle argued that motion in the universe demanded a prime mover, theolo-
gians identified that prime mover as God. Newton happily kept that idea, insisting only that once
the universe was set in motion, its laws inexorably described its future course. Such a picture of
God, the deist view, no longer saw God personally involved in the lives of individual humans but
rather as merely the prime mover and someone responsible for bridging the gaps in our understand-
ing of physics.
This God of the gaps, Buckley argues, is precisely what led to atheism. If our proof of God is
the existence of such gaps, then once the gaps are filled by natural laws, we wind up showing that
there is no need for God after all. Thus, atheism. Notice that the god that was lost was not much
of a god; he had already been reduced to just one force alongside all the other forces in nature.
Rather than being supernatural—outside space and time, existence before the beginning, before
the “before”—this kind of god went back to being a sort of pagan nature deity.
THE FOUNDATION OF THEISM
If most scientific atheists have rejected a deist god (which is far from the Christian God in any
event), we can ask, what is religion to a skeptic? Does one of my skeptic friends have it right when
she says that religion is just a behavior-modification system to persuade people to be good? My
experience with my techie friends suggests that this idea of religion, defining it in terms of its func-
tion, is much more widespread than outright atheism. This idea assumes an axiomatic kind of
agnosticism. Those techie friends assume that they can never find the truth about whether or not
there is a God, and so to them the only honest response is to throw up their hands and move on.
The absence of evidence may not be evidence of an absence, but it surely is no proof of a presence
either.
We who live in contemporary Western culture are all scientists now. Even the fundamentalists
claim to be scientists. We all look for evidence for everything. The fundamentalists fall back on all
the bits of science that they cannot understand (how does life evolve?) as proof of the need of God;
the skeptics rightly reject such arguments as a God of the gaps. I am neither a skeptic nor a creation-
ist. What is the evidence on which I base my faith? It is not the orbits of the planets or the position
of the sun or the moon. It is not the authority of ancient sages, either saints or scientists. Instead, in

Science: Theism
a later book, Denying and Disclosing God, Buckley suggests a different sort of evidence that brings
people to religion. He quotes a twentieth-century philosopher, Raïssa Maritain, who came to
Christianity late in life. She encountered, in her study of history and in her own life, the lives of saints.
The essential data point for her was what she called “the fact of sanctity” (Buckley 2004, 129).
Holiness exists. Any theory of the universe that fails to take into account that existence is
incomplete. Joy is a sign of holiness; joy is why we do science. Holiness goes beyond joy. When I
spoke to the student in South Carolina who wanted to know how to tell his mom that he would
be studying geology, we both came to the conclusion that no sound bite would do the job. Instead,
if he was going to be a geologist, he would also have to be a good, upstanding person—a holy per-
son. That, of course, would be the effort of a lifetime, but only that would convince her.
How did Jesus convince people? I doubt that any of his clever arguments changed any minds
among the scribes and Pharisees. No one has ever been converted by a syllogism. Instead, consider
how Jesus converts the woman at the well, described in the fourth chapter of John’s Gospel. It starts
when Jesus makes the first approach to the Samaritan woman, asking as a supplicant for a cup of
water from her. He soon offers her the “living water” of eternal life. As the conversation proceeds,
more about this woman’s life comes out, and it is clear that Jesus knows the worst about her.
Though she is a Samaritan and someone who has been married five times, she is overcome by the
fact that Jesus knows all about her and still loves her. She is moved by who Jesus is, what he tells
her about herself, his obvious sanctity. That makes her start to question the cosmology that up to
now she has lived inside. It is the first step toward conversion.
This leads me to ask, finally, just what kind of success is it that we hope to accomplish in our
conversations about science and faith? What conversion experience do we hope to trigger? Once
more we return to the Master in this final Gospel story, from Mark:
One of the teachers of the law came and heard them debating. Noticing that Jesus had given
them a good answer, he asked him, “Of all the commandments, which is the most important?”
“The most important one,” answered Jesus, “is this: ‘Hear, O Israel: The Lord our God, the
Lord is one. Love the Lord your God with all your heart and with all your soul and with all
your mind and with all your strength.’ The second is this: ‘Love your neighbor as yourself.’
There is no commandment greater than these.”
“Well said, teacher,” the man replied. “You are right in saying that God is one and there is no
other but him. To love him with all your heart, with all your understanding and with all your
strength, and to love your neighbor as yourself is more important than all burnt offerings and
sacrifices.”
When Jesus saw that he had answered wisely, he said to him, “You are not far from the
kingdom of God.” And from then on no one dared ask him any more questions. (12:28–34,
NIV)
Once again, Jesus succeeds by changing the ground of the discussion—in this case, from law to
love. For once, he gets the scribe to agree with him, but the story does not end there. In fact, the
final success occurs when Jesus agrees with the scribe. We know we have won the argument when
we can acknowledge that the other side is right.
Bibliography
Aeschylus. Prometheus Bound. In Aeschylus, with an English translation by Buckley, Michael. Denying and Disclosing God. New Haven, CT: Yale
Herbert Weir Smyth, Ph. D. in two volumes. 1. Prometheus Bound. University Press, 2004.
Translated by Herbert Weir Smyth. Cambridge, MA: Harvard
Chesterton, G. K. Orthodoxy. San Francisco: Ignatius Press, 1995. First
published 1908.
Athanasius. On the Incarnation. Translated by John Behr. Yonkers, NY: St. Consolmagno, Guy. “Cosmology.” In New Catholic Encyclopedia Supplement
Vladimir’s Seminary Press, 2012. 2009. Detroit: Gale, 2009.
Buckley, Michael. At the Origins of Modern Atheism. New Haven, CT: Yale Crombie, A. C. Augustine to Galileo. 2 vols. London: Harmondsworth,
University Press, 1987. 1969.
Science: Theism
Crowe, Michael J. Theories of the World from Antiquity to the Copernican Lamott, Anne. Plan B: Further Thoughts on Faith. New York: Riverhead Books,
Revolution. New York: Dover, 1990. 2004.
Cultural Cognition Project. http://www.culturalcognition.net. Lewis, C. S. The Discarded Image: An Introduction to Medieval and
Renaissance Literature. Cambridge, UK: Cambridge University Press,
Danielson, Dennis Richard, ed. The Book of the Cosmos: Imagining the
1967.
Universe from Heraclitus to Hawking. Cambridge, MA: Perseus, 2000.
Pascal, Blaise. Pascal’s Pensées. New York: Dutton, 1958.
Duhem, Pierre. Le système du monde: Histoire des doctrines cosmologiques de
Platon à Copernic. 10 vols. Paris: Librairie Scientifique Hermann et Cie, Pedersen, Olaf. The Two Books: Historical Notes on Some Interactions
1954. First published 1913. between Natural Science and Theology. Edited by George V. Coyne and
Tadeusz Sierotowicz. Notre Dame, IN: University of Notre Dame
Ignatius of Loyola. The Spiritual Exercises of St. Ignatius. Translated by Louis
Press, 2007.
J. Puhl. Westminster, MD: Newman Press, 1951.
Stoeger, William R. “The Big Bang, Quantum Cosmology and creatio ex
Jaki, Stanley. The Relevance of Physics. Chicago: University of Chicago Press, 1965.
nihilo.” In Creation and the God of Abraham, edited by David B. Burrell,
Jaki, Stanley. The Savior of Science. Grand Rapids, MI: Eerdmans, 1988. Carlo Cogliati, Janet M. Soskice, and William R. Stoeger, 152–176.
Kahan, Daniel M. “Ideology, Motivated Reasoning, and Cognitive
Reflection.” Judgment and Decision Making 8, no. 4 (2013): 407–424. Tillich, Paul. Dynamics of Faith. New York: Harper and Row, 1957.

TOPIC 17
Science: Atheism
DOES SCIENCE FAVOR THEISM OR ATHEISM? 583
Herman Philipse
Distinguished Professor of Philosophy
Utrecht University, Netherlands
THE EFFECT OF SCIENTIFIC PROGRESS AND THE SCIENCE 588

OF RELIGION ON THE CREDIBILITY OF THEISM
Richard Carrier
Educator (PhD)
RELIGIOUS EXPERIENCE AND NEUROSCIENCE 591

Kenneth Williford
Associate Professor and Chair, Department of Philosophy & Humanities
University of Texas at Arlington
NEAR-DEATH EXPERIENCES ARE NOT EVIDENCE FOR 594

EITHER ATHEISM OR THEISM
Keith Augustine
Executive Director and Editor-in-Chief
Internet Infidels, Carson City, NV
ARGUMENTS INVOLVING COSMOLOGY AND QUANTUM PHYSICS 596

Taner Edis
Professor, Department of Physics
Truman State University, Kirksville, MO
A proper assessment of the bearing of scientific inquiry on theistic religion requires recognition that conflict,
mutual consistency, independence, or concilience are possible but depends on the methods accepted and the
claims made in each domain at particular times and places—which can, and have, varied. This chapter,
therefore, focuses on questions relating to whether the best methods, findings, and theories in contemporary
scientific disciplines support, cohere with, or conflict with commitments made by theistic theologies.
DOES SCIENCE FAVOR THEISM OR ATHEISM?
Herman Philipse, Utrecht University
With regard to nearly all gods that humans ever believed in, theists and atheists fully agree that these
divinities do not exist. Homeric gods such as Zeus or Poseidon and ancient Egyptian deities like
583
Science: Atheism
solar Ra or mother Isis are considered to be fictitious characters.

KEY CONCEPTS Social scientists and historians investigate the mental and com-
munal functions of these cultural constructs, which allegedly
explained striking empirical phenomena, reinforced group
Aurora Borealis cohesion, or legitimized the reign of rulers. According to priests
Big Bang in ancient Egypt, for example, a god called Khnum produced
Entheogen the annual Nile flood, and several creation myths accounted
for the existence of the cosmos.
Exoplanet
Because contemporary theists are atheists with regard to all
Light-Year of these gods, we should distinguish sharply between two varie-
Near-Death Experience ties of atheism. On one hand, universal atheists such as I hold
New Age it to be highly improbable that any god humans ever believed
in exists. On the other hand, (mono-)theists may be called
Out-of-Body Experience “exceptionalist” atheists: they endorse universal atheism with
Quantum Cosmology one exception, which is concerned with the god they call
Singularity God, Yahweh, or Allah. Theists believe that one god exists,
whereas all other gods do not.
Space-Time
Well-educated theists will concede that, typically, the
empirical phenomena people explained by postulating divine
actions in ancient Egypt, Greece, or India can be better
accounted for by empirical science. Whereas in ancient Greece it was believed that the sky god
Zeus caused lightning and thunder, using it as weapons against his opponents, from the nine-
teenth century onward scientists developed physical explanations of thunder as a shock wave in
the air caused by lightning, which is a spectacular electrostatic discharge in the atmosphere.
Hence, with regard to the many gods of traditional polytheisms, informed theists agree with
atheists that science favors atheism, because well-confirmed scientific explanations of empirical
phenomena rule out traditional religious accounts. In this section, I argue that the same holds
with regard to God, that is, concerning the god of theism. My thesis is that science favors univer-
sal atheism.
Contemporary theists have invented various apologetic strategies to avoid incompatibilities
between well-established results of scientific research and the contents of their theistic beliefs, legiti-
mizing the latter. Using such a strategy, they argue that although science may not favor theism, it
does not favor atheism either, at least not with regard to God. As I have argued elsewhere, none
of these apologetic strategies is convincing (Philipse 2012).
Let me defend briefly three claims (discussed in the sections below) in order to support the the-
sis that science favors atheism with regard to God, that is, the thesis that scientific progress makes it
increasingly unlikely that the god of theism exists. In order to provide content to this thesis, we have
to know how theists characterize the being to which they refer with the proper name God. Tradi-
tionally, this god is described as “a person without a body (i.e. a spirit) who … is eternal, perfectly
free, omnipotent, omniscient, perfectly good, and the creator of all things” (Swinburne 2004, 7).
How do science and theism relate to each other if theism is the view that this god exists?
SCIENTIFIC PROGRESS CAUSED GRADUAL THEISTIC RETREAT

Many scientists who endorse a monotheist creed argue that there can be no incompatibility between
science and theism, because they are concerned with different domains—such as nature and the
supernatural, or the world of facts and the domain of values. However, these compatibilist views
have been triggered by a long learning experience, during which biblical and other religious accounts
of many things have been superseded by superior scientific explanations. In other words, since the
times of Nicolaus Copernicus and Galileo, scientific progress has caused a gradual theistic retreat.
This has happened in all domains of science, from cosmology to medicine. Let me mention three
examples to illustrate my first claim.
One of the greatest mathematical physicists ever, Isaac Newton, was a passionately pious mono-
theist. His religious writings are more voluminous than the Principia (1687) and Opticks (1704)
Science: Atheism
taken together. It is often forgotten today that although Newton developed the first well-con-
firmed universal mechanics, he also argued that some observed phenomena could be explained
only with reference to God. For example, he thought that no physical explanation could be given
for (1) the fact that all planets rotate around the sun in the same direction “in Orbs concentrick”
and (2) the fact that the solar system does not collapse because of the gravitational attractions
between the planets. He concluded that these two striking aspects of our solar system “can be
the effect of nothing else than the Wisdom and Skill of a powerful ever-living Agent” (Newton
[1704] 1979, 402–403).
Nearly a century after Newton had put forward this argument, the French mathematician
and astronomer Pierre-Simon Laplace (1749–1827) eliminated God from scientific cosmology.
He explained the unidirectional rotations of the planets by his hypothesis that the sun’s atmo-
sphere had started to rotate when contracting because of gravitation, and he accounted for the
noncollapse of the solar system by a superior calculation of gravitational planetary perturbations,
showing that they remained within a limited range (Gillispie 1997, chap. 21). According to a
famous anecdote, Napoléon Bonaparte would have asked his former teacher Laplace: “Monsieur
Laplace, où est Dieu dans votre système?” (Mr. Laplace, where is God in your system?), and
Laplace would have answered: “Sire, je n’ai pas besoin de cette hypothèse” (Your Majesty, I do
not need this hypothesis).
When Laplace eliminated God from cosmology, many Christian philosopher-scientists still
believed that God had created all biological species separately, as Genesis tells us. For example, Wil-
liam Paley published his celebrated book Natural Theology in 1802, in which he applied the so-
called watchmaker analogy to show that the existence of animals and plants amounts to a “proof
of an intelligent Creator” (Paley [1802] 2006, 277). This is why Charles Darwin argued extensively
in On the Origin of Species by Means of Natural Selection (1859) that his evolutionary hypothesis
refuted the biblical view according to which “each species has been independently created” (Darwin
[1859] 1996, 105–106). The impressive progress of evolutionary biology combined with research in
genetics further confirms Darwin’s hypothesis.
My third example is concerned with the temporal dimension of the universe. In the Christian
tradition, many biblical experts, from Theophilus of Antioch (second century CE) to Bishop James
Ussher (1581–1656) have attempted to calculate the age of the universe on the basis of biblical
chronologies. Ussher’s conclusion that God created the world in 4004 BCE has become famous,
because this date is mentioned in the King James Bible. Very few Christians still believe that biblical
chronologies can be trusted. With regard to the age of the earth, we rely today on radiometric dat-
ing of meteorite material, which shows that our planet is about 4.54 billion years old. Because big
bang cosmology has been empirically confirmed, scientists are able to investigate the chronology
of the universe as well, on the basis of measurements of its expansion rate. The initiating expansion
of our universe probably occurred around 13.8 billion years ago. We might mention many other
instances of scientific progress that caused religious retreat. Let us now wonder why this happened.
THERE ARE NO RELIABLE SOURCES OF RELIGIOUS BELIEF

If theism were true, we might expect that scientific research would disclose ever more traces of
God’s activity and authorship in the universe, because the universe would be God’s work of art.
As indicated, however, the reverse has happened. Scientific progress increasingly caused religious
retreat, in the sense that experts eliminated religious explanations of phenomena in favor of scientific
explanations. Why did this happen, precisely? Let us define science in a very broad sense, including
the humanities and everyday know-how, as the human endeavor to obtain knowledge by using
methods of research and epistemic sources that have been well tested and shown to be reliable.
My second claim is that there are no reliable sources of religious belief. This became increasingly
clear during the ages of scientific progress, which explains why scientific advancement caused reli-
gious retreat. Science favors universal atheism because the belief that God exists, or that there are
gods, cannot be validated by using any reliable epistemic source or method.
Theists may object that our access to God must be different from the ways in which we exam-
ine the natural world. Because God is a supernatural being, sensory perception and well-calibrated

Science: Atheism
methods of scientific research will be irrelevant for getting in touch with him. According to most
religious traditions, gods communicate with us via dreams, states of trance, oracle experiences, mira-
cles, revelations, repeated rituals, and so on. In his classic biography of Muhammad, for example,
the eighth-century Arab historian and hagiographer Ibn Ishaq tells us that the archangel Gabriel
provided a first revelation to Muhammad while the latter was dreaming on a mountain in the cave
of Hira. If God exists, he also may have implanted in humans a special cognitive mechanism via
which he reveals himself to us directly, a so-called sensus divinitatis (Plantinga 2000, 148 and pas-
sim).
According to my second claim, all these special sources of religious belief are clearly unreliable.
One well-known argument to this effect is the argument from the plurality of religions. Since many
religions rely on sources of the same kinds in order to legitimize their claims to truth concerning
God or gods (such as divine revelations), whereas these contentions contradict each other, the
alleged special sources of religious knowledge cannot be relied on. Dreams should not be trusted
as sources of truth, for example.
The argument from the plurality of religions may not apply to extraordinary events that legiti-
mize one religion only and which are claimed to be unique. Let me discuss in some detail the con-
version experience of Saint Paul on the road to Damascus. This extraordinary event has been crucial
for Christianity. If Paul had not been converted, he would not have propagated faith in Jesus
throughout the Roman Empire, and the Jesus sect would not have developed into a world religion.
Why and how did the pitiless Jewish persecutor of Jesus adepts Saul transform into Paul, the most
prominent early propagator of Christianity and author of some of the oldest texts of the New Tes-
tament?
Let us assume that we can rely on the biblical book Acts with regard to the natural facts con-
cerning Saul’s conversion experience. According to Acts 9:1–9, the following things occurred sud-
denly to Saul on the way to Damascus, when he was “still breathing threats and murder against
the disciples of the Lord”: (1) a light from heaven flashed about Saul, (2) he fell to the ground,
(3) he heard a voice, and (4) he was without sight during three days. According to Acts 22:9, those
who were traveling with Saul (5) “saw the light but did not hear the voice,” whereas according to
Acts 26:13, (6) the light from heaven was “brighter than the Sun,” and (7) the travelers “had all
fallen to the ground.” Saul assumed that the resurrected Jesus was speaking to him during this baf-
fling experience. As a consequence, he converted from a cruel persecutor of Jesus’s followers to the
main Jesus propagandist.
There is no doubt that natural facts (1–7) must have been deeply disturbing and perplexing to
Saul and his fellow travelers. How should we explain them? Some Christians still think that the res-
urrected Jesus truly appeared to Saul, but this leaves unexplained quite a few aspects of the event.
For example, why would a supernatural appearance of Jesus cause all travelers to fall to the ground?
The best explanation of facts (1–7) was proposed in 2015 by the astrophysicist William K. Hart-
mann, who is cofounder of the Planetary Science Institute in Tucson, Arizona.
According to Hartmann, the description of what Saul experienced on his way to Damascus
offers “a strikingly good match” to eyewitness accounts of the “explosive entry of an asteroid frag-
ment over Chelyabinsk in 2013” (Hartmann 2015, 368). Let us suppose, then, that something sim-
ilar happened to Saul and his companions on the way to Damascus. The explosion of a meteoroid
in the air will have caused a shock wave that blew the travelers to the ground (2, 7). The light from
heaven brighter than the sun (1, 6) was an intense ultraviolet radiation resulting from the fireball
event. This caused Saul’s temporary blindness, because he looked longer at the light than his fellow
travelers, so that he suffered from photokeratitis afterward (4). Apparently, Saul interpreted the loud
noise resulting from the explosion as a voice (3), which he attributed to the resurrected Jesus,
whereas his companions did not hear a voice (5).
It is no wonder that, living in the first century and being ignorant of astrophysics, Saul and his
companions interpreted their overwhelming experience in religious terms. Saul’s belief that Jesus
appeared to him during this shocking event should be explained psychologically. Saul may have felt
a deep guilt or fear of revenge because of his cruel persecution of Jesus adepts. Perhaps he also real-
ized that converting to Jesus would liberate him from the exigent demands of Jewish law. However
Science: Atheism
this may be, Saul’s conversion experience can be accounted for completely in scientific terms with-
out assuming any supernatural intervention. Hence, what happened to Saul on the way to Damas-
cus cannot be relied on as a trustworthy source of religious belief.
THE TRUTH OF THEISM IS IMPROBABLE, GIVEN MANY SCIENTIFIC RESULTS

My first two claims may be interpreted as favoring agnosticism instead of universal atheism. They
tell us merely that alleged evidence for theism is illusory, without showing that God does not exist.
Let me finish, then, by indicating some of the many scientific results that make it improbable that
theism is true.
Cosmology. Whereas according to the biblical book of Genesis God created the cosmos in order to
house humanity, cosmological investigations show that life is extremely rare in the universe and that
many regions of it are empty, such as the Giant Void, which measures some 1.3 billion light-years
across. It is unlikely that God would have created such an inhospitable universe.
Neuroscience. Ever-more neuroscientific research shows that all mental phenomena depend on
complex brain processes. Hence, it is improbable that God exists, if he is defined as a bodiless mind.
Evolutionary Biology. An infinitely good God would never have created humans by means of
biological evolution, because this would have been an excessively cruel creative procedure. For
example, during the evolution of life on the earth, 99 percent of all species have perished. Further-
more, humans would not have evolved if about 66 million years ago the Cretaceous-Paleogene
extinction event had not happened, when a massive asteroid impact caused the extinction of about
three-quarters of animal and plant species.
Paleoanthropology, Genetics, Archaeology, Sociology. As fossils from early hominins and

progress in DNA sequencing have revealed, hominins must have diverged from chimps more than
7 million years ago. Modern humans speciated at least some 200,000 years ago. When they devel-
oped religious beliefs is less clear, but archaeological and historical research reveals that probably
all known religions were polytheistic until the Egyptian pharaoh Amenhotep IV became a monothe-
ist around 1343 BCE. These data imply that if God exists, he hid himself to humanity during many
millennia. Furthermore, he would still hide himself today from all those who do not believe that he
exists. Because God loves all humans as a father, however, he would never hide himself from any
decent human being. Hence, God does not exist (Schellenberg 2015).
Theism Explained. When we wonder what explains the occurrence of theism in human history,
two hypotheses should be compared. Either God caused its occurrence, or there is a purely secular
explanation. If the latter accounts for the facts of religions on the earth better than the former, this
is a further scientific confirmation of atheism. One of these facts is that monotheism or theism in
the sense defined started to appear quite late in human history. As I said earlier, theism cannot
explain this fact.
Inspired by David Hume, I propose the following secular explanation. As we read in the Old
Testament, the Jewish people appealed to God often in situations of war with other tribes (e.g.,
Deuteronomy 20). My hypothesis is that during warfare, people often thought that they could tri-
umph if their god was cleverer and more powerful than the gods of their opponents. This motivated
a tendency to ascribe ever-greater properties to their god(s) “till at last they arrive at infinity itself,
beyond which there is no farther progress” (Hume [1757] 1976, 52). As soon as people thought that
their god had infinite properties, such as being omnipotent, omniscient, and perfectly good, they real-
ized gradually that their theism excluded the existence of other gods, so that they became monotheists.
In many respects, this secular explanation of theism’s origin is empirically superior to a theistic
explanation. It accounts for the facts that theism arose only locally and late in human history and
that in the Old Testament God is portrayed as a god of one tribe, the Jews, who supports them

Science: Atheism
against other tribes, which are butchered by this god. The same point applies to the histories of
Christianity and of Islam.
THE EFFECT OF SCIENTIFIC PROGRESS AND THE SCIENCE OF RELIGION

ON THE CREDIBILITY OF THEISM
DOES SCIENTIFIC PROGRESS FAVOR THEISM OR ATHEISM?

In the aggregate, the evidence of scientific progress has decreased the probability of theism and
increased the probability of nontheistic naturalism (Carrier 2010b). This started to become apparent
in the ancient history of the West. Since the seventh century BCE, when the Greek philosopher and
astronomer Thales realized that lunar and solar eclipses were not caused by magical or divine forces
but were merely the shadows cast by objects moving in their natural courses, science for many cen-
turies thereafter continued discovering that theistic and supernaturalistic explanations of the phe-
nomena of nature turned out to be false (Lloyd 1979, 2016; Carrier 2017), until scientific progress
ended under Christian dominance in the fourth century CE. A thousand years later a few Christian
intellectuals fought hard to recover that neglected tradition and reinvigorate scientific advancement
(while before that time a brief attempt at such a revival by Muslims failed to take hold). The out-
come was an even more comprehensive and rapid pace of disproving theological assertions about
the world that continues to this day.
Theology has responded by continually “redefining” God or beliefs about God to fit these
newly revealed facts. Nonetheless, the singular fact that theism (in any form, worldwide) never
anticipated any of these developments and always has to be changed to accommodate them demon-
strates that this evidence has lowered the probability that theism holds true. If that evidence hadn’t
been lowering its probability, theism would not have had to continually change to fit what was
being revealed. In the logic of probability, the more a theory’s predictions are disconfirmed and,
consequently, the more that theory has to be revised with more additions after the fact to fit those
disconfirmations, the less likely that theory is to be true (Carrier 2012)—unless evidence is acquired
that independently confirms that those ad hoc additions are true. That, however, has never hap-
pened in the history of theism.
Balance of Probability. In terms of prior probability, if naturalism wins every time two horses—
theism and naturalism—run a race to substantiate whose explanation of some fact of the world is
correct, then, after thousands of races, the conclusion must be that theology is least likely to provide
the true model for reality. That is what has happened. No theistic explanation of any phenomenon
has turned out to be true. Some hypotheses remain to be fully tested against one another (princi-
pally, the grounding theories of cosmological and cognitive science: how the universe began and
acquired the characteristics it has and why consciousness is experienced in the form of veridical qua-
lia). Given past experience, the odds do not favor theism in those races; that is true by a consider-
able margin, owing to a vast accumulated past of scientific progress, which progress only ever con-
tinues in the absence of theistic hypotheses (Carroll 2016; Carrier 2005).
For example, in the early 200s CE, in his monumental treatise On the Uses of the Parts of the
Body, the Greco-Roman biologist Galen produced an extended scientific defense of the theory of
intelligent design in the construction of animal and human bodies; shortly before that, the Greco-
Roman astronomer Ptolemy suggested the same conclusion from the elegantly complex organization
of the geocentric universe in Planetary Hypotheses. English mathematician and astronomer Isaac
Newton and English naturalist Charles Darwin subsequently overthrew both conclusions by advanc-
ing science with the discovery of what actually causes apparent biological design and all the balanced
motions of the planets and stars. In neither case was it an intelligent engineer but instead the inevi-
table outcome of mindless objects and forces acting on one another, without the meddling of any
deity. In the twentieth century, science has progressed to the point of discovering that the absence
Science: Atheism
of divine involvement better explains observations in every domain, including astrophysics and the
origin and evolution of life (Carrier 2011; Carroll 2016). The ancient battle of naturalism versus
theism in the creationism debate was mostly undecidable on the science known before modern
times (Sedley 2008) but has since gone in every way for naturalism. The same has occurred in the
domains of weather, disease, astronomical and geological events, and all of biology, psychology,
chemistry, and physics.
Although theism can be reconstructed in a way compatible with science, it is never a conclusion
actually confirmed by science. That the most widely believed and fiercely defended theisms are the
most hostile to scientific conclusions and values is further evidence that theism is likely untrue (Cro-
mer 1993; Loftus 2016; Carrier 2017). Even moral and social progress has proceeded on nontheistic
scientific conclusions against the opposition of popular theisms (Shermer 2015; Carrier 2005). This
has all been true not only in the context of Christendom but under Islamic (Ofek 2011) and Hindu
(Nanda 2011) regimes as well.
Apologetic Responses. Theistic apologetics aims to deny these conclusions. At least three strate-
gies have been attempted. One approach is to argue that scientific progress may have failed to vin-
dicate theism but nevertheless requires a foundation of assumptions dependent on theism. The
historical facts never support this claim (e.g., Carrier 2010a). Another approach is to argue that
theistic hypotheses are the only ones viable for answering a certain few as yet unanswered ques-
tions, but the analytical facts never support this claim. For example, theism does not even have
a working theory that properly predicts the many peculiar features of human consciousness,
whereas evidence continues to mount that the only working theory will derive from computa-
tional physics (see below). Likewise, theism has never advanced any cosmological theory that
makes any uniquely testable predictions—most theistic theories have been falsified (Carrier
2011)—whereas many nontheistic cosmological theories have been developed that do explain
many unusual observations, including features of the universe that are theistically unnecessary
and even contrary to expectation in terms of any popular theism (Carrier 2011, 2005). Though
no theory has yet been confirmed, the evidence indicates that some nontheistic model will explain
most observations, including the most peculiar of them.
A third approach is to argue that theism has occasionally made successful scientific predictions,
but there is no actual case where this happened. Even the theosophist Rudolf Steiner’s 1923 predic-
tion of “colony collapse” in honey bee populations derived from no theological premise; it was
argued entirely from natural science (Steiner 1975). Theories that require no deity are out of
account, as by that very fact their being true does not support theism over naturalism. No theory
requiring a deity has ever been verified by science. Attempts to claim that evolution by natural selec-
tion requires prohibitively large steps in random information gain have never survived peer review,
and to this day no empirical evidence supports any such claim (Young and Edis 2004). The claim
that Christian theology predicted the big bang theory is false. Christian theology predicted a radi-
cally different origin, in both time and process, and was disproved by the discovery of deep cosmo-
logical time and a particularly mechanical evolution of the cosmos (in accord with nuclear physics),
which in no credible way corresponded to the account of Genesis or the writings of any theologian.
Theology’s only correct guess was that there is an origin point to the current cosmos in time, but
that same guess was made by many nontheistic cosmologies. It still has not been scientifically
proved that time itself had an origin (Veneziano 2004), the only unique prediction Christian theol-
ogy can claim. Even that prediction is now made by several nontheistic cosmological theories.
DO RECENT ADVANCES IN COGNITIVE SCIENCE FAVOR THEISM OR ATHEISM?

Over the course of the twentieth century, and especially the early years of the twenty-first, cognitive
science has comprehensively established the dependence of the human mind on physical structures
and events in the brain (Carrier 2005, 2011, 298–302; Churchland 2013). Even some theologians
are conceding this point (Murphy 2006). There is no actual science establishing that the mind
can survive the death of the brain. A vast scale of scientific evidence has accumulated establishing
countless defects of the mind’s functioning as the result of dependence on its physical structure

Science: Atheism
(Martin and Augustine 2015). The entry for “List of Cognitive Biases” at Wikipedia, for example, is
not even comprehensive yet is already daunting in scale. These facts favor atheism.
Scientists have identified where in a brain different kinds of memories are stored, where emo-
tions and reason operate, where each kind of sensory experience is processed, and so on. They have
observed that if any one of these parts is physically damaged, removed, or deactivated, the memories
or abilities it contains are then lost. It follows that if all the parts are taken away, everything that
human beings are will cease to exist. The same is true if oxygen or nutrients are removed or electri-
cal discharges or psychoactive chemicals are introduced. Atheistic naturalism, without any additional
assumptions, fully predicts observations of this kind, because no other way exists to have conscious-
ness except as the product of a large, delicate, flawed, yet complex physical system, one that sits at
the end of an extremely long, meandering, faulty process of trial and error over billions of years.
Yet in no version of theism are those same observations expected, without adding an extraordinary
array of excuses; but such additions guarantee the low probability of any theism that requires them
(Carrier 2012, 80–81). Quite simply, if theism is true, minds do not require brains. For if God
exists, it is necessarily true that minds can exist without brains. Therefore, if theism is true, it
remains inexplicable why our minds require brains—at all, much less brains so flawed, vulnerable,
and resource-dependent as the human brain is. The evidence therefore falls far more in favor of god-
less natural selection than theism, which greatly limits the probability of theism.
Theistic apologists attempt to answer this issue by noting that science has yet to confirm a
physical-causal explanation of the subjective qualities of conscious experience (the argument from
qualia), which fact alone does not support theism, because the prior probability remains strongly
in favor of an eventual naturalist explanation (cf. Carrier 2011). Some apologists attempt, then, to
argue further that a physical-causal explanation of the subjective qualities of conscious experience
is impossible. There has yet to be any analytically sound demonstration of that claim (Melnyk
2003; Cottrell 1999). If the subjective quality of experience is wholly without remainder caused
by, and thus dependent on, physical structures and events, theism is not required to account for
it. Whether one then chooses to classify the qualia of such experience as a separate ontology from
the physical would be mere semantics; their causation would remain fully explained. Theism has
produced no theory capable of predicting the specific correlations of brain event to qualia abun-
dantly documented. Given past experience in discovering how the brain produces mental experi-
ence, it is unlikely any such theory will succeed. To date, computational physics remains far more
promising (Koch and Tononi 2017; Cottrell 1999).
DO SCIENTIFIC DISCOVERIES ABOUT RELIGIOUS EXPERIENCE AND RELIGIOUS BELIEF

FORMATION SIMPLY REVEAL GOD’S WAYS OF COMMUNICATING WITH US?
Many advances have been made in the cognitive science of religious experience (Newberg 2010),
but none has confirmed the reality of any gods or anything at all beyond the natural. To the con-
trary, the endless diversity and demonstrated cultural dependency of religious experience and belief
mean that there is no true source of the information they convey about the world (Loftus 2010,
2013). The findings of scientific anthropology must also be added to the findings of cognitive sci-
ence, including the discovery that religious experiences serve identifiable social functions and are
culturally constructed (Lewis 2003; Fales 1999). All of this reveals a basic truth: what humans expe-
rience and believe in the domain of religion has other causes, wholly internal to the mind, as influ-
enced externally by its culture and environmental circumstances and without the involvement of
gods or the supernatural (Guthrie 1995; Dennett 2007; Vyse 2013).
That the content of religious experience changes over time, culture, and circumstance and
that all these contradictory intuitions and communications feel equally real to their percipients
(yet are then often confirmed to be false) means that religious experience is not a reliable access
point to any truth about reality. This author was fully convinced of the supernatural truth of
Daoism by religious experiences (Carrier 2005), in the same way as every convert to Christianity,
Islam, Buddhism, or any other tradition. This verifies that religious experience is all a mental con-
struct, which percipients then convince themselves is true. The construct may be useful. That is a
matter of debate. But whether it is factually true requires more objective and consistently
Science: Atheism
verifiable evidence, which is never to be found. By contrast, were gods genuinely interested in and
capable of communicating with human beings, humans should have a data set on the matter that
is consistent, at least to some notable extent, across the entire five thousand or so years of
recorded human history and all human cultures (present and past, which number in the thou-
sands). Explaining away the lack of this data set does not rescue the conclusion that religious
experience has any real source outside the individual mind affected by culture and circumstances.
Such tactics only reduce the probability of the conclusion through the accumulated additions of
unverified assumptions.
That religious experience is never scientifically verified to contain any knowledge not already
possessed by percipients or the cultures they are influenced by and thus routinely conforms to the
ignorance and factually false beliefs of that culture and time further verifies this conclusion. The
very evidence needed to verify that a religious experience came from any superhuman source is
never to be found outside unverifiable legends and popular tales. This is to be expected in atheism
but not in theism, thereby increasing the probable validity of atheism over theism. The authors of
the Bible and Qurʾan, for example, despite claiming a channel to the divine, never learned of the
immorality of slavery or gender inequality or of the truth of heliocentrism or the germ theory of dis-
ease or anything at all remarkable for their time. For all these reasons, the science of religious expe-
rience actually supports atheism over theism.
RELIGIOUS EXPERIENCE AND NEUROSCIENCE
Kenneth Williford, University of Texas at Arlington
The neuroscience of religious and mystical experience and practice is still in its infancy, though it
has made many strides in the last half of the twentieth century and since the beginning of
the twenty-first century (for overviews see Newberg, d’Aquili, and Rause 2001; Horgan 2003;
McNamara 2006, 2009; Schjoedt 2009; Nelson 2010; McNamara and Butler 2013). It is, of
course, virtually impossible to capture spontaneously unfolding religious or mystical experiences
(RMEs) in a laboratory setting. Thus, the neuroscientific study of RMEs has tended to focus, on
one hand, on deliberate contemplative practices, like prayer and meditation (see, e.g., Austin
1999; Newberg et al. 2003; Lutz, Dunne, and Davidson 2007; Ricard, Lutz, and Davidson
2014; Neubauer 2014), that are not as sensitive to circumstance and surroundings as, say, conver-
sion experiences or beatific visions and, on the other hand, on relatively reliably reproducible
RMEs, for example, psychedelic drug-induced mystical experiences (see, e.g., Masters and Hous-
ton [1966] 2000; Siegel [1989] 2005; Smith 2000; Hobson 2001; Shanon 2002; Strassman
2005; Griffiths et al. 2006, 2008, 2011) or related experiences, for instance, induced out-of-body
experiences (see, e.g., Blanke et al. 2002, 2005; Blanke and Metzinger 2009; Lenggenhager et al.
2007). To date, the neuroscientific study of RMEs has not provided any significant, independent
evidence in favor of theism (or any other religious outlook for that matter, e.g., Buddhism).
Naturalism provides a more parsimonious framework for understanding these experiences (cf. Fales
2010). In that sense, atheism (at least with respect to a God that would want to provide unequiv-
ocal evidence for its existence via the induction of RMEs) is more supported by the neuroscientific
study of RMEs than is theism.
There are several sources of confusion that must be guarded against in any discussion of the
bearing of neuroscientific studies on the evidential import of RMEs. First, there need be no com-
mitment to methodological naturalism when it comes to the interpretation of these studies. Con-
trary to a naturalistically tendentious philosophy of science, it is perfectly conceivable that the results
of neuroscientific studies of RMEs could favor a nonnaturalistic or specifically theistic interpreta-
tion. If, for example, contemplative practitioners could regularly enter into, say, beatific visions
through prayer, and if this could reliably be seen to correlate with an influx of energy into their
brains that could not be accounted for by sensory stimulation or by endogenous neuronal activity,
this would count as some evidence that the practitioner’s experience correlates with nonnatural

Science: Atheism
activity making a difference in the physical world. This is somewhat analogous to John C. Eccles’s
ideas about empirically testing interactionist dualism (see, e.g., Eccles 1953). If, moreover, the prac-
titioner could make reliable predictions while in such states or, say, give us demonstrable answers
to unsolved problems of mathematics and science (e.g., dictate a proof, disproof, or proof of the
undecidability of the Goldbach conjecture, tell us how and where to detect evidence of life on
exoplanets, dictate a better theory of the aurora borealis), this would constitute even more evidence
that the experiences put the practitioner into contact with an intelligent, supernatural source. It is
safe to say that, to date, no such experiences have been reliably and replicably documented in labo-
ratory settings. If there is an intelligent, supernatural being that one is in contact with during a
veridical religious experience, it would seem that that being is reluctant to allow those experiences
to be an unequivocal source of evidence accessible by scientific means.
Second, it is also important that one avoid simplistic “debunking” arguments about RMEs. The
fact that RMEs are associated with a range of techniques, practices, and conditions that, in one way or
another, alter consciousness and thus brain processes (e.g., prayer, meditation, flagellation, fasting,
sleep deprivation, ingestion of entheogens, twirling, chanting, tantric sex, dimmed lights, organ music,
snake handling, sensory deprivation, sensory overstimulation, dancing, hand raising to sappy praise-
and-worship songs, the Sun Dance, seizures, dreams) does not, in and of itself, constitute a
knockdown argument against the veridicality of these experiences. Presumably, every sort of cognitive
and affective experience, from everyday visual experience to falling in love to the understanding of a
mathematical proof, has some sort of neural substrate; the substrates of relatively unusual or extraordi-
nary experience will, of course, be as unusual and extraordinary as the experiences themselves. That
said, the fact that RMEs are associated with such a variety of “spiritual technologies” or practices
and that there appears to be no independent way of ruling some in and others out as reliable ought
to give the proponent of the evidential value of RMEs pause. If a plausible general explanation (or
family of related explanations) not couched in terms of any particular religious tradition and, more
specifically, not couched in theistic terms can be found for this wide variety of RMEs (and their meth-
ods of inducement), this would seem, ceteris paribus, to be the more reasonable approach.
Third, the number of methods for inducing RMEs and the variety of traditions and subtraditions,
theistic and nontheistic, in which these methods have emerged and developed cannot be conveniently
ignored in any scientific study of the neural underpinnings of RMEs. The classic “many contenders”
problem plaguing the argument from religious experience reaches into the neuroscientific study of
RMEs. As indicated earlier, there is a considerable body of scientific literature now on meditative prac-
tice and psychedelic (or entheogenic) drug-induced RMEs. Much of so-called contemplative neurosci-
ence is actually motivated by varying degrees of religious (or quasi-religious) commitment to Bud-
dhism. The Dalai Lama himself has actively cooperated with neuroscientists for many years (see, e.
g., Goleman and Thurman 1991; Hayward and Varela 1992; Varela 1997; Houshmand, Livingston,
and Wallace 1999; Dalai Lama 2005). While almost no one involved in this endeavor (including Ten-
zin Gyatso himself) would argue that contemplative neuroscience provides evidence of the signal meta-
physical doctrines of Buddhism (e.g., karma and rebirth), considerable evidence for the generally posi-
tive effects of meditation on long-term practitioners has accumulated, and plausible models of the
underlying neuronal processes supporting these effects have been articulated (see Ricard Lutz, and
Davidson 2014 for an accessible overview, but see Khalsa et al. 2008, 2015, for cautionary studies).
If a neuroscientifically based “ye shall know them by their fruits” argument can be mounted for any
specific religious tradition, it can be mounted for Buddhism (see, e.g., Wallace 2003, 2007; Wallace
and Hodel 2008; Flanagan 2011; Wright 2017; McMahan and Braun 2017; though see Victoria
2003, 2006, and Jerryson and Juergensmeyer 2010 for some caution here about “fruits”—equally
applicable, by the way, to Judaism, Christianity, Islam, and Hinduism).
Most theists would, of course, be quick to point out that a plausible non-Buddhist explanation
of the generally positive effects of meditation (and the neural substrates of those effects) would not
be all that hard to articulate. After all, any sort of long-term practice (e.g., musical training, brain
games, athletic practice) correlates with brain changes mediated by neuronal plasticity. If one
devotes oneself to the control of attention (as in single-pointed meditation), the control of anxio-
genic and “grasping” tendencies (as in mindfulness or open monitoring meditation), and the
Science: Atheism
cultivation of compassion and altruism (as in loving-kindness meditation), it will not be all that sur-
prising to find neuronal changes in the relevant brain regions—the attention control centers (e.g.,
prefrontal cortex), startle-reflex-related areas (e.g., amygdala), and so on (see Ricard Lutz, and
Davidson 2014 for references and details). This eminently plausible response to a neuroscience-
based argument for Buddhism would, of course, cut equally against any analogous argument
mounted for a theistic conclusion, unless some special phenomena (like those suggested earlier) were
reliably shown to attend specifically theistic religious experience.
RMEs induced by psilocybin, DMT, LSD, mescaline, and other psychotropic substances cannot
be dismissed a priori by the theist or by a superficial a posteriori firsthand or secondhand survey (pace
Zaehner 1972; for a more careful classic study see Wainwright 1981 and Richards 2005 for a more
recent discussion). In fact, the history of the use of plants and concoctions containing such substances
for religious purposes seems to be more or less coincident with human history (see, e.g., Harner 1973;
Furst 1976; Grinspoon and Bakalar 1979; Anderson 1996). Still, today there are many religious tradi-
tions, most of them theistic, that recognize the use of such substances as a legitimate spiritual practices
(e.g., Rastafari, Native American Church, Santo Daime, União do Vegetal). While the RME-inducing
potential of such substances was studied scientifically in the heyday of psychedelic enthusiasm in the
1960s (before such study was ended essentially for political reasons by the Nixon administration; see
Baum 2016), since the 1990s and the next decade this study has been renewed for the classical psy-
chedelics and other substances (in addition to the studies by Griffiths et al. 2006, 2011, 2016, see
Carhart-Harris et al., 2012, 2016; Muthukumaraswamy et al. 2013). The studies on psilocybin indi-
cate that many of the RMEs it engenders are regarded as profound and positive by the experimental
subjects, and the palliative potential of psilocybin for terminally ill patients is being actively explored
(Griffiths et al. 2016). The theist, at least one wedded to Christianity, Judaism, Islam, or Hinduism,
is in no good position to argue on independent grounds against the genuineness or veridicality of the
experiences delivered by entheogenic substances. Again, altering consciousness is the name of the game
when it comes to RMEs; it is only the means that differ. There is no good independent argument that
prayer and fasting, say, are the genuine means God has chosen for delivering these experiences while
the ingestion of psilocybin (etc.) is not.
Of course, the theist averse to the idea that God would use tryptamines, phenethylamines, and
other molecules crossing the blood-brain barrier as vehicles of veridical RMEs might hasten to point
out that imaging studies on psilocybin have suggested something about the unusual neuronal processes
that seem to be the substrate of such experiences and that there is no need to appeal to the truth of
any particular psychedelic vision to account for these experiences (see, e.g., Lebedev et al. 2015 on the
mechanisms at play in experiences of “ego dissolution” engendered by psilocybin, e.g., “decreased func-
tional connectivity between the medial temporal lobe and high-level cortical regions” [3137]). This
response would, of course, be spot on. Here again, however, the response would cut against, mutatis
mutandis, any argument for theism drawn from an analogous neuroscientific study of theistic RMEs
(or some preferred subset thereof, e.g., Christian, Islamic, Hindu), were such an analogous study to exist.
In the absence of dramatic, replicable discoveries that relate theistic RMEs to an influx of
energy into the brain that cannot be accounted for by endogenous brain processing or known
sources of sensory stimulation or that relate them to reliable predictions (this latter, of course, not
really requiring any specifically neuroscientific methods), there is no good reason to think that the-
istic RMEs are not to be explained by the same sorts of processes (differential activity in brain
regions induced by a variety of causes, neural plasticity, etc.) that explain nontheistic RMEs or psy-
chedelic RMEs. So far, there is no reason to appeal to special explanatory posits (e.g., the activities
of Jesus, Krishna, or bodhisattvas) to account for the range of these experiences. Though the details
will, of course, vary (e.g., the effects of DMT on the brain are, evidently, very different from the
effects of prayer vigils and fasting or meditation), so far there is nothing that demands we go beyond
the basic explanatory frameworks of the neurosciences, psychology, and the social sciences to
account for the variety of RMEs. Any specific theistic framework would entail that only some such
experiences are veridical or legitimate, but no such framework gives us an independent way to tell
the difference. It would also entail that God, at least as conceived of by exclusivist traditions, is
something of a trickster when it comes to RMEs, allowing for various “spiritual technologies” and

Science: Atheism
related religio-cultural traditions and interpretive frameworks to grow up that all seek to draw evi-
dence from the RMEs the “technologies” and traditions legitimate. This trickster God would also
seem to allow this with some regularity, since RMEs seem to be more or less reliably inducible in
people across traditions and cultures. A naturalistic explanation that combines neuroscience, psy-
chology, sociology, anthropology, and perhaps evolutionary biology and can explain the wide variety
of RMEs seems to be both more parsimonious and more comprehensive (cf., again, Fales 2010). It
is therefore to be preferred, on elementary, well understood, and generally accepted explanatory
grounds, to a theistic explanation.
NEAR-DEATH EXPERIENCES ARE NOT EVIDENCE FOR EITHER ATHEISM

OR THEISM
Keith Augustine, Internet Infidels
Near-death experiences (NDEs) are ostensibly otherworldly experiences reported by a small percent-
age of people who were dying but returned to life (van Lommel et al. 2001) or who at least believed
that they were dying though they were not medically close to death (Gabbard and Twemlow 1991;
Gabbard, Twemlow, and Jones 1981; Stevenson, Cook, and McClean-Rice 1989–1990). The pop-
ular image of an NDE has become ubiquitous: feelings of joy, an out-of-body experience (OBE) in
which a person seems to view the physical world from a perspective different from that of the body,
rushing through a tunnel or darkness toward a brilliant light, encountering deceased loved ones,
undergoing a review of one’s earthly life, and perhaps strolling through a beatific environment or
running into an uncrossable point of no return.
All of these motifs are rarely reported within a single NDE, however, and the most prominent
of them are characteristically absent from reports of NDEs by those with minimal exposure to
Western ideas. Reports of NDEs from India, for example, typically feature being brought in front
of the Hindu god of death before being returned to life because the wrong person was retrieved
(Pasricha and Stevenson 1986). In addition to such stark cross-cultural differences, in the mere
one-quarter of Western NDE reports that include an OBE component (van Lommel et al. 2001),
the NDEs occasionally feature visions of the physical world at odds with what is actually happening
at the time, such as a vision of a nonexistent addressed envelope on a table (Lindley, Bryan, and
Conley 1981). Other features suggesting that at least some NDEs are hallucinatory include reports
of feeling bodily sensations while ostensibly outside the body and visiting a spiritual realm (Augus-
tine 2015), receiving prophetic visions of the future during NDEs that fail to come pass (Ring
1982, 1988), and encountering living persons (Kelly 2001) while supposedly “on the other side.”
Not every NDE report includes such blatantly hallucinatory features, however.
If there is nevertheless good reason to infer that all NDEs are probably hallucinations, then
NDEs are irrelevant to whether either atheism or theism is true, as well as to the more salient ques-
tion of whether spiritual realms exist at all. In principle, NDE reports cannot provide evidence
against the existence of some sort of spiritual reality. While they do have the potential to provide
evidence for the existence of some such reality, in practice, they fail to provide such evidence.
NDES ARE NOT EVIDENCE FOR ATHEISM/NATURALISM

In the best-case scenario for atheism—or, more accurately, naturalism (the view that no spiritual
realms exist)—all NDEs would be demonstrably hallucinatory. But the existence of one kind of hallu-
cination, NDEs, would no more support naturalism than would that of another, such as dreaming. As
a simple point of logic, that people hallucinate spiritual realms during NDEs cannot provide evidence
that spiritual realms do not exist any more than that people hallucinate trips to Paris while dreaming
can provide evidence that Paris does not exist. In other words, spiritual realms might exist even if no
NDE is ever an experience of them. Thus, the only NDEs that could be relevant to whether spiritual
realms exist are those that seem to be experiences of such realms—that is, NDEs that are in all prob-
ability not hallucinatory. Forceful claims have been made for the accuracy of purported visions of the
Science: Atheism
physical world said to have been obtained during some NDEs (Habermas 2018; Holden 2009; Rivas,
Dirvin, and Smit 2016), however dependent on the frailties of human testimony these claims may be.
Their value ultimately depends on how well they can be corroborated by independent testimony (i.e.,
testimony where agreement between witnesses is not simply the result of witnesses having talked
among themselves before an investigator interviewed them, or the result of independent witnesses
gleaning congruent information from the same third party).
Those experiences that are said to have been accurate are often called veridical NDEs, but it is
important to be mindful that those who have not experienced them have access only to the reports
of such experiences, not to the experiences themselves. No one is in a position to say that when the
details of NDE reports are accurate, their accuracy derives from some paranormal source rather than
from information gleaned through the normal senses (such as hearing) during such experiences, or
from information gleaned through normal channels before or after such experiences, but interpo-
lated into reconstructed memories of NDEs.
NDES ARE NOT EVIDENCE FOR THEISM/SUPERNATURALISM

Only one way exists to overcome this inherent weakness of testimonial evidence: controlled experi-
ments. In principle, veridical paranormal perception during NDEs could suggest the existence of
spiritual realms, thus supporting a kind of broad supernaturalism (though perhaps not theism per
se). Is there any good evidence that veridical paranormal perception actually occurs during NDEs?
This discussion summarizes the results of experiments that have aimed to establish the occurrence
of veridical paranormal perception during OBEs and NDEs by implementing safeguards meant to
prevent normal sources of information from influencing the content of reports of such experiences.
From the late 1960s to the early 1980s, parapsychologists carried out controlled experiments
designed to register the effects of OBE adepts’ “astral bodies” on various instruments and sub-
jects—from strain-gauge sensors to animals to human “sensitives”—when adepts “projected” to spe-
cific physical locations at particular times. Bucking the trend, a few of these “detection studies” pro-
duced some promising early results, but they either fell apart under further scrutiny or were never
further investigated. In other experiments, OBE adepts “projected” to particular locations in order
to report back on the characteristics of visual targets there. With two exceptions—one from an
pseudonymous Miss Z and the other from the remote viewer Ingo Swann—these target-identifica-
tion experiments produced no clear-cut hits. In the two cases where there were indeed clear-cut hits,
the adepts were left alone with the visual targets for significant periods, essentially running the
experiments with no controls at all (Augustine and Fishman 2015).
From 1990 to 2014, the results of six target-identification experiments aimed at spontaneously
occurring NDEs that include an OBE component were published. Not one of them produced a sin-
gle hit (Augustine 2007; Holden 2009; Parnia et al. 2014). The AWARE study, for example,
involved multiple NDE target-identification experiments running concurrently at different medical
institutions across the globe.
Although the AWARE study was designed to vindicate the view that NDEs are not hallucinations,
the results ironically have had the opposite effect. The study found that of the fifty-five reported cardiac
arrest experiences, forty-six (84 percent) were clearly dreamlike hallucinations, with the remaining clas-
sic NDEs constituting only 16 percent (nine of fifty-five) of the total (Parnia et al. 2014). These results
alone are sufficient to refute premature arguments that it is simply impossible for the brain to generate
any experiences during cardiac arrest, and thus NDEs cannot be brain-generated hallucinations (see,
e.g., Greyson 2010; van Lommel 2006). They also raise the possibility that classic NDEs are simply
a subset of these dreamlike hallucinations. Perhaps the more coherent of the dreamlike narratives simply
get labeled as reports of NDEs because they happen to have an otherworldly theme (because expecta-
tion of imminent death during cardiac arrest calls up afterlife imagery at the time).
TAKING STOCK
The failure to secure replicable positive results in NDE target-identification experiments does not
establish the nonexistence of any spiritual realms, but it does serve to substantially challenge positive
arguments in favor of the existence of spiritual realms from NDE reports. For if veridical

Science: Atheism
paranormal perception occurs during OBEs or NDEs, why the failure to find it in all of the con-
trolled experiments that have been undertaken to document it thus far? Various explanations can
be put forward (Augustine and Fishman 2015), but in the absence of ad hoc maneuvering, the hal-
lucination hypothesis predicts only one set of possible results: the results actually found. Until the
time that properly controlled NDE target-identification experiments yield replicable positive results,
they will take their place as historical curiosities akin to similarly unsuccessful tests of survival after
death (Augustine and Fishman 2015; Berger 1996; Fox 2007; Gay et al. 1955; Journal of the Amer-
ican Society for Psychical Research 1989; Lodge 1905; Perry and Fontana 2009; Schwartz and Russek
2001; Stevenson, Oram, and Markwick 1989). While some eagerly await the results of the follow-
up AWARE II study (which is recruiting subjects until 2020), at the moment the unsuccessful his-
tory of comparably easier-to-implement research into the paranormality of non-near-death OBEs
does not bode well for those results.
ARGUMENTS INVOLVING COSMOLOGY AND QUANTUM PHYSICS
Taner Edis, Truman State University
PHYSICAL COSMOLOGY
Before modern physics, arguments about the creation or design of the universe depended on every-
day, intuitive ideas of space and time. Creation has often been understood as an event in time
linked to supernatural agency. Hence the notion of divine causation involved in creation has also
been an intuitive, social concept of causality. Theistic traditions have placed creation and the design
of the cosmos within a richly layered story, in which a god interacts with humans and reveals some-
thing about its purposes. In the context of such a story, supernatural agents have genuine explana-
tory roles. Abstract discussions about a god of the philosophers have still depended on background
stories and intuitive notions of time and causality to anchor the metaphysical intuitions in play.
Historically, critical responses to creation claims have preserved everyday intuitions while
attempting to deny divinity any explanatory role. With the doctrine of creation from nothing
becoming dominant in established theologies, critics have often defended an infinitely old universe,
denying that there was a creation event. Ideas of an always existing universe have typically been asso-
ciated with pagan philosophy, heterodoxy, and doubt.
Skeptics have also had to respond to the suggestion that cosmic order was due to intelligent
design. They have typically intensified the perception of order to an extent where the flexibility
implied in personal causation became implausible. The laws of physics, in such a view, are where
natural explanations come to an end, while divine agency could be manifest in miraculous violations
of an otherwise rigidly impersonal order of nature.
Modern Views. Today’s skepticism about theistic cosmology has been shaped by the way modern
physics has undermined intuitive views of time and causality and the collapse in plausibility of the
traditional stories that provided a context for divine agency. Cosmology has become a subfield of
physics devoted to impersonal forms of explanation, so that supernatural agency has no productive
role in advancing cosmological understanding.
Therefore, in an inversion of the historical pattern, in today’s cosmology, supernatural creation
and design have become marginalized claims. Conservative religious thinkers loyal to the traditional
stories assert failures or limitations in physical cosmology to be remedied by divine agency. The big
bang might be best understood as a creation event. Possible fine-tuning of physical constants or the
low entropy of the early universe might indicate a universe intelligently designed to favor the pres-
ence of life and mind. Liberal theologians tend not to seek such direct employment for their gods;
they more often point out that cosmology cannot decisively rule out personal supernatural powers.
If there are reasons to think gods exist, these reasons may be found outside cosmology.
In this environment, atheists often point out either that physical cosmology has good candi-
dates for solutions to the problems theistic apologists bring up or that the prospects for physics to
Science: Atheism
make progress are good. If the gods are unemployed in as important an area as making the universe,
this casts doubt on their existence (Carroll 2005). Occasionally, a more ambitious atheist may argue
that physical cosmology can more rigorously formulate concepts such as “nothing” and supplant tra-
ditionally metaphysical debates about how anything happens to exist (Krauss 2013).
Beginnings. The modern theoretical context for doing cosmology begins with general relativity.
Time is mixed with space when changing reference frames; no such thing as a “now” is common
to all observers, and physicists are compelled to think of space-time as a single geometric object, past
and future included. Relativity and astronomy also indicate that time can be extended backward
only until a point where all space-time and all matter and energy collapse into a singularity.
For physicists, this suggests a breakdown in the applicability of general relativity; for some religious
thinkers, it indicates a creation event.
It is, however, misleading to think of the big bang as a beginning or as an explosion into a pre-
existing empty space. Space and time cannot be extended earlier from the big bang in much the way
that it makes no sense to speak of a point north of the North Pole. Attaching a divine cause to the
big bang is motivated by metaphysical intuitions, not physics. Moreover, general relativity is a
classical theory. It cannot be applied in circumstances close to the big bang, where quantum
gravitational effects dominate.
No adequate theory of quantum gravity is yet available. Nevertheless, cosmologists have
approximate models that, while speculative, promise better understanding of the early universe.
With only weak theoretical and experimental constraints on such models, however, there are a wide
variety of scenarios. Some models take quantum uncertainties to smear out the big bang, retaining a
space-time that cannot be extended infinitely backward while discarding the big bang as a unique
infinite-density point in space-time. Some models rely on inflationary cosmology to generate many,
perhaps an infinity of universes. Models that depend on string theory also can extend time back
infinitely, because the string-length scale provides a limit beyond which the universe cannot become
smaller.
In any case, quantum cosmology inherits the randomness in quantum mechanics. Macroscopic
causality emerges from a quantum substrate where uncaused microscopic events are the rule. There-
fore physical cosmology does not support traditional ideas either of creation or of an infinitely old
universe with time understood in everyday terms (Edis 2002; Halper and Nayeri 2016). Some theo-
logians advocate atemporal gods, avoiding some of the difficulties with the concept of a creation in
time, but such views are also detached from physics and irrelevant to cosmology.
Fine-Tuning. Theists sometimes claim that physical constants have been fine-tuned for a universe
that allows the development of complex structures, which is best explained as a divine design. In
some circumstances, the fine-tuning appears exaggerated; regardless of the details, a theoretical con-
text that would allow us to reliably assign probabilities to fundamental constants does not yet exist.
Fine-tuning is highly model-dependent, and cosmological models are not always well constrained.
Indeed, fine-tuning problems are known from many areas of physics and often indicate a need for
novel physical approaches. Even within cosmology, solving fine-tuning problems has partially moti-
vated important developments such as inflationary cosmology. From a physical point of view, there-
fore, fine-tuning appears as one of the many puzzles to be expected in a cutting-edge, highly unset-
tled area of physics (Stenger 2011).
The low entropy of the early universe may also suggest design, but again this is a physics prob-
lem on which science can claim progress—for example, noting that an expanding universe is driven
away from equilibrium as the maximum possible entropy grows faster than the actual entropy. The
arrow of time is not a completely solved puzzle, but it does not call for supernatural intervention
(Carroll 2010).
Design as an explanation for fine-tuning or low-entropy states has its own weaknesses. A highly
abstract god of the philosophers implies almost nothing about what sort of universe such a god
might design. A more traditional divinity dressed up in stories does not help either, because the tra-
ditional stories are radically misinformed about the sort of universe human beings inhabit. A claim

Science: Atheism
of supernatural design responds to a puzzle with “God did it,” without engaging in the explanatory
work relevant to a physical puzzle.
Cosmology as Physics. The present state of cosmology supports nontheistic views but only weakly
and mainly through cosmology becoming a specialization within physics, where supernatural claims
are not useful. If it is not part of a broader evidence-based argument against supernatural agency,
cosmology has little significance on its own. Therefore, emphasis on physical cosmology tends to
accompany more broadly naturalist and physicalist views. While such views are often associated with
atheism, they are not strictly necessary for doubting the gods.
QUANTUM MECHANICS
Quantum mechanics appears in debates about religion because some believe that quantum physics
supports the possibility of supernatural action or that it shows that consciousness is not reducible
to mindless processes. In the nineteenth century, many thought that classical physics depicted a
causally closed universe. Theists could therefore look for signs of the supernatural in exceptions to
this closure. If miracles that violated the laws of physics took place, this indicated a power beyond
the natural order. Free will meant that human beings are not determined by prior physical condi-
tions, which shows that consciousness is not bound by the laws of physics. Skeptics, however, have
usually thought that the case for miracles is weak. Determinism is not a comfortable position, but
the rigidly impersonal order glimpsed in fundamental physics strongly contrasts with ideas of a uni-
verse subject to the whims of capricious gods.
Quantum Loopholes. The advent of quantum mechanics challenged the hyperrationalistic picture
of rigid natural order. After all, the results of quantum measurements were random, and the equa-
tions of classical physics now referred to expectation values rather than completely determined out-
comes. Moreover, quantum measurements were defined by a classical, macroscopic limit, where a
superposition state randomly “collapsed” onto one of many possible outcomes. To some of the phy-
sicists working to formulate the early versions of quantum theory, state collapse in measurement
suggested a fundamental role for consciousness in physics, because definite results were associated
with the presence of an observer.
Quantum mechanics, therefore, changed the debate about supernatural intervention. In classi-
cal physics, an interaction between the universe and an outside agent would violate fundamental
laws such as the conservation of energy and momentum. Quantum randomness, however, means
that conservation laws apply not to the result of any single measurement but to the expectation
values that are approached over a long run of identical measurements. A god could, for example,
intervene in any particular event or in any small set of events. As long as the magnitude of the inter-
vention is small enough to be lost in the statistical noise, no violation of conservation laws could be
measured. If a god wanted to create humans by way of evolution, that god could make sure just the
right mutations took place over a long timescale, and this intervention would be undetectable by
physical means.
New Age Magic. In popular culture, quantum means “magic.” The supernaturalism that claims
support from quantum mechanics is closer to that of Hindu religions or the more occult or mystical
variants of the Middle Eastern religious traditions; its allegedly scientific basis is parapsychology
rather than physics.
Neither New Age religious movements nor the more intellectual varieties of quantum mysti-
cism currently enjoy mainstream scientific support (Stenger 1995). Parapsychology has not pro-
duced reliable experimental evidence for psychic phenomena and has not established robust theoret-
ical links with the sciences—particularly not with quantum mechanics. Cognitive neuroscience
makes progress in understanding minds without taking seriously speculations about quantum con-
sciousness.
The measurement problem in quantum mechanics is genuinely interesting, because state col-
lapse is not invertible, whereas all time evolution in quantum mechanics is invertible. This is not
Science: Atheism
a question unique to quantum mechanics, however; it is very closely related to the problem in sta-
tistical mechanics of macroscopic irreversibility deriving from microscopically reversible dynamics.
While questions remain outstanding, this is at least a partially solved problem. Quantum measure-
ments involve interactions with extremely complex, noisy environments, which lead to noninvertible
macroscopic approximate descriptions or a time evolution toward states with classical properties.
Schrödinger’s cat will be either dead or alive long before any observer investigates. Conscious obser-
vers are not special to the measurement process—they are just another part of a messy environment
(Edis 2002).
Miracles. The claim that quantum randomness covers up supernatural intervention without violat-
ing conservation laws is associated with more conventional forms of theism. It is also scientifically
sterile (Sansbury 2007).
If any set of quantum measurements is to be claimed as evidence for intervention, it would
mean that the results of the measurements were not random—that there was discernible structure
in the data and that therefore quantum mechanics was violated. Currently there is not even the
smallest experimental hint of violations of quantum mechanics. If, on the other hand, the interven-
tions claimed are very few and lost in the statistical noise, they would not produce a discernible pat-
tern, and therefore the data alone could not support supernatural intervention as an explanation.
There could still be reasons to believe in intelligent design, but these reasons would have to come
entirely from outside the data and theories of physics. With undetectable interventions following
a purpose revealed only to those privy to special knowledge, we end up with a cosmic conspiracy
theory (Edis 2018).
In other words, the current state of affairs is not very different from that with classical physics.
Quantum randomness means humans live in a universe of uncaused events, which undermines
older notions of causal closure. Randomness, however, does not open up the universe to alleged
nonphysical interventions. Physical evidence for supernatural agency still requires robust signals
combined with a successful theory of intelligent design.
Chance-and-Necessity Physicalism. The fundamental randomness manifested in quantum

mechanics also affects how we conceive of physical explanations. In current physics, even the low-
temperature laws of physics are often seen as an outcome of a cascade of spontaneous symmetry-
breaking events, with random results. The most fundamental laws of physics, as in the standard
model of particle physics, are statements about highly symmetric conditions with very low informa-
tion content, while the complexities of our universe arise from symmetry breaking. Highly symmet-
ric fundamental laws, in other words, describe the dice that were rolled to generate our universe
(Edis 2002).
Acknowledging the centrality of randomness in modern physics can lead to arguments that cast
doubt on all supernatural and theistic claims. Physical explanations combine rules and randomness,
both of which are mindless. Therefore, as some intelligent design proponents also recognize, the sig-
nature of an agent not reducible to physical processes would be data that could not be produced by
any combination of rules and randomness. In fact, possible functions exist that require infinite
computational resources, which would be available to gods with traditional omni-attributes. Data
that fit such functions might best be explained by an agent that is not limited by rules and random-
ness and therefore is beyond fundamentally mindless physical processes. Such data would not just
violate quantum mechanics but also defeat any physical theory. However, none of the data available
to science even remotely suggests such a possibility.
Hence quantum mechanics has an important role in formulating chance-and-necessity physical-
ism, according to which everything is physical, a combination of rule-bound and random processes,
regardless of whether the most fundamental physical theory has yet been formulated (Edis and Bou-
dry 2014). Religions usually take a top-down view, starting with an irreducible mind to shape the
material world from above. Physicalism, whatever form it takes, supports a bottom-up understand-
ing of the world, where life and mind are the results of complex interactions of fundamentally

Science: Atheism
mindless components. The current state of science, including quantum mechanics, supports chance-
and-necessity physicalism.
If physicalism appears plausible, this does not imply certainty that there are no gods. Future
data may come to support nonphysical agents. Arguments that make no reference to publicly avail-
able information may yet seem compelling. Today, however, humans live in an environment where
the successful sciences have no use for the supernatural. This state of affairs puts claims for divine
agency on the defensive, and it means that the burden of proof for such claims is very high.
Bibliography
Anderson, Edward F. Peyote: The Divine Cactus. Tucson: University of Carrier, Richard. “Christianity Was Not Responsible for Modern Science.”
Arizona Press, 1996. In The Christian Delusion, edited by John Loftus, 397–419. Amherst,
NY: Prometheus Books, 2010a.
Augustine, Keith. “Does Paranormal Perception Occur in Near-Death
Experiences?” Journal of Near-Death Studies 25, no. 4 (Summer 2007): Carrier, Richard. “On Defining Naturalism as a Worldview.” Free Inquiry
203–236. 30, no. 3 (2010b): 50–51.
Augustine, Keith. “Near-Death Experiences Are Hallucinations.” In The Carrier, Richard. “Neither Life nor the Universe Appear Intelligently
Myth of an Afterlife: The Case against Life after Death, edited by Michael Designed.” In The End of Christianity, edited by John Loftus, 279–304.
Martin and Keith Augustine, 529–569. Lanham, MD: Rowman and Amherst, NY: Prometheus Books, 2011.
Littlefield, 2015. Carrier, Richard. Proving History: Bayes’s Theorem and the Quest for the
Augustine, Keith, and Yonatan I. Fishman. “The Dualist’s Dilemma: The Historical Jesus. Amherst, NY: Prometheus Books, 2012.
High Cost of Reconciling Neuroscience with a Soul.” In The Myth of an Carrier, Richard. The Scientist in the Early Roman Empire. Durham, NC:
Afterlife: The Case against Life after Death, edited by Michael Martin and Pitchstone, 2017.
Keith Augustine, 203–292. Lanham, MD: Rowman and Littlefield,
Carroll, Sean. “Why (Almost All) Cosmologists Are Atheists.” Faith and
2015.
Philosophy 22, no. 5 (2005): 622–635.
Austin, James H. Zen and the Brain: Toward an Understanding of Meditation
Carroll, Sean. From Eternity to Here: The Quest for the Ultimate Theory of
and Consciousness. Cambridge, MA: MIT Press, 1999.
Time. New York: Dutton, 2010.
Baum, Dan. “Legalize It All: How to Win the War on Drugs.” Harper’s Carroll, Sean. The Big Picture: On the Origins of Life, Meaning, and the
Magazine (April 2016). https://harpers.org/archive/2016/04/legalize
Universe Itself. New York: Dutton, 2016.
-it-all/.
Churchland, Patricia. Touching a Nerve: The Self as Brain. New York:
Berger, Arthur S. “Thouless Test for Survival: Failures and Claims.” Journal Norton, 2013.
of the American Society for Psychical Research 90, no. 1 (January 1996):
44–53. Cottrell, Allin. “Sniffing the Camembert: On the Conceivability of
Zombies.” Journal of Consciousness Studies 6, no. 1 (1999): 4–12.
Blanke, Olaf, and Thomas Metzinger. “Full-Body Illusions and Minimal
Phenomenal Selfhood.” Trends in Cognitive Sciences 13, no. 1 (2009): Cromer, Alan. Uncommon Sense: The Heretical Nature of Science. New York:
7–13. Oxford University Press, 1993.
Dalai Lama. The Universe in a Single Atom: The Convergence of Science and
Blanke, Olaf, Christine Mohr, Christoph M. Michel, et al. “Linking Out-of-
Spirituality. New York: Morgan Road Books, 2005.
Body Experience and Self Processing to Mental Own-Body Imagery at
the Temporoparietal Junction.” Journal of Neuroscience 25, no. 3 (2005): Darwin, Charles. On the Origin of Species by Means of Natural Selection.
550–557. Oxford: Oxford University Press, 1996. First published 1859.
Blanke, Olaf, Stéphanie Ortigue, Theodor Landis, and Margitta Seeck. Dennett, Daniel C. Breaking the Spell: Religion as a Natural Phenomenon.
“Neuropsychology: Stimulating Illusory Own-Body Perceptions.” New York: Penguin Books, 2007.
Nature 419, no. 6904 (2002): 269–270. Eccles, John Carew. The Neurophysiological Basis of Mind: The Principles of
Carhart-Harris, Robin L., David Erritzoe, Tim Williams, et al. “Neural Neurophysiology. Oxford: Oxford University Press, 1953.
Correlates of the Psychedelic State as Determined by fMRI Studies with Edis, Taner. The Ghost in the Universe: God in Light of Modern Science.
Psilocybin.” Proceedings of the National Academy of Sciences 109, no. 6 Amherst, NY: Prometheus Books, 2002.
(2012): 2138–2143.
Edis, Taner. “Cosmic Conspiracy Theories: How Theologies Evade
Carhart-Harris, Robin L., Suresh Muthukumaraswamy, Leor Roseman, et al. Science.” In Theology and Science: From Genesis to Astrobiology, edited
“Neural Correlates of the LSD Experience Revealed by Multimodal by Joseph Seckbach and Richard Gordon, 143–163. Hackensack, NJ:
Neuroimaging.” Proceedings of the National Academy of Sciences 113, no. World Scientific, 2018.
17 (2016): 4853–4858.
Edis, Taner, and Maarten Boudry. “Beyond Physics? On the Prospects of
Carrier, Richard. Sense and Goodness without God: A Defense of Metaphysical Finding a Meaningful Oracle.” Foundations of Science 19, no. 4 (2014):
Naturalism. Bloomington, IN: AuthorHouse, 2005. 403–422. doi: 10.1007/s10699-014-9349-z.
Science: Atheism
Fales, Evan. “Can Science Explain Mysticism?” Religious Studies 35, no. 2 Harner, Michael J., ed. Hallucinogens and Shamanism. Oxford: Oxford
(1999): 213–227. University Press, 1973.
Fales, Evan. Divine Intervention: Metaphysical and Epistemological Puzzles. Hartmann, William K. “Chelyabinsk, Zond IV, and a Possible First-Century
London: Routledge, 2010. Fireball of Historical Importance.” Meteoritics and Planetary Science 50,
Flanagan, Owen. The Bodhisattva’s Brain: Buddhism Naturalized. Cam- no. 3 (2015): 368–381.
bridge, MA: MIT Press, 2011. Hayward, Jeremy W., and Francisco J. Varela, eds. Gentle Bridges:
Fox, Margalit. “Ian Stevenson Dies at 88; Studied Claims of Past Lives.” Conversations with the Dalai Lama on the Sciences of Mind. Boston:
New York Times, 18 February 2007. Shambhala, 1992.
Furst, Peter T. Hallucinogens and Culture. Novato, CA: Chandler and Sharp, Hobson, J. Allan. The Dream Drugstore: Chemically Altered States of
1976. Consciousness. Cambridge, MA: MIT Press, 2001.
Gabbard, Glen O., and Stuart W. Twemlow. “Do ‘Near-Death Experiences’ Holden, Janice Miner. “Veridical Perception in Near-Death Experiences.” In
Occur Only Near Death?—Revisited.” Journal of Near-Death Studies 10, The Handbook of Near-Death Experiences: Thirty Years of Investigation,
no. 1 (September 1991): 41–47. edited by Janice Miner Holden, Bruce Greyson, and Debbie James,
185–211. Santa Barbara, CA: Praeger/ABC-CLIO, 2009.
Gabbard, Glen O., Stuart W. Twemlow, and Fowler C. Jones. “Do ‘Near-
Death Experiences’ Occur Only Near Death?” Journal of Nervous and Horgan, John. Rational Mysticism: Dispatches from the Border between Science
Mental Disease 169, no. 6 (June 1981): 374–377. and Spirituality. New York: Houghton Mifflin, 2003.
Gay, Kathleen, William Henry Salter, Robert H. Thouless, et al. “Report on Houshmand, Zara, Robert B. Livingston, and B. Alan Wallace, eds.
the Oliver Lodge Posthumous Test.” Journal of the Society for Psychical Consciousness at the Crossroads: Conversations with the Dalai Lama on
Research 38, no. 685 (September 1955): 121–134. Brain Science and Buddhism. Ithaca, NY: Snow Lion, 1999.
Gillispie, Charles Coulston. Pierre-Simon Laplace 1749–1827: A Life in Hume, David. The Natural History of Religion. Oxford: Clarendon Press,
Exact Science. Princeton, NJ: Princeton University Press, 1997. 1976. First published 1757.
Goleman, Daniel, and Robert A. F. Thurman, eds. MindScience: An East- Jerryson, Michael, and Mark Juergensmeyer, eds. Buddhist Warfare. Oxford:
West Dialogue. Sommerville, MA: Wisdom, 1991. Oxford University Press, 2010.
Greyson, Bruce. “Implications of Near-Death Experiences for a Postmateri- Journal of the American Society for Psychical Research. “Has Arnold Barber
alist Psychology.” Psychology of Religion and Spirituality 2, no. 1 (2010): Survived Death?” 83, no. 3 (July 1989): 286.
37–45.
Kelly, Emily Williams. “Near-Death Experiences with Reports of Meeting
Griffiths, Roland R., Matthew W. Johnson, Michael A. Carducci, et al. Deceased People.” Death Studies 25 (2001): 229–249.
“Psilocybin Produces Substantial and Sustained Decreases in Depression
and Anxiety in Patients with Life-Threatening Cancer: A Randomized Khalsa, Sahib S., David Rudrauf, Antonio R. Damasio, et al. “Interoceptive
Double-Blind Trial.” Journal of Psychopharmacology 30, no. 12 (2016): Awareness in Experienced Meditators.” Psychophysiology 45, no. 4
1181–1197. (2008): 671–677.
Griffiths, Roland R., Matthew W. Johnson, William A. Richards, et al. Khalsa, Sahib S., David Rudrauf, Richard J. Davidson, and Daniel Tranel.
“Psilocybin Occasioned Mystical-Type Experiences: Immediate and “The Effect of Meditation on Regulation of Internal Body States.”
Persisting Dose-Related Effects.” Psychopharmacology 218, no. 4 (2011): Frontiers in Psychology 6 (2015): 924.
649–665.
Koch, Christof, and Giulio Tononi. “Can We Quantify Machine
Griffiths, Roland R., William A. Richards, Matthew W. Johnson, et al. Consciousness?” IEEE Spectrum 54, no. 6 (2017): 64–69.
“Mystical-Type Experiences Occasioned by Psilocybin Mediate the
Krauss, Lawrence M. A Universe from Nothing: Why There Is Something
Attribution of Personal Meaning and Spiritual Significance 14 Months
Rather Than Nothing. New York: Free Press, 2013.
Later.” Journal of Psychopharmacology 22, no. 6 (2008): 621–632.
Lebedev, Alexander V., Martin Lövdén, Gidon Rosenthal, et al. “Finding the
Griffiths, Roland R., William A. Richards, Una McCann, and Robert Jesse.
Self by Losing the Self: Neural Correlates of Ego-Dissolution under
“Psilocybin Can Occasion Mystical-Type Experiences Having Substan-
Psilocybin.” Human Brain Mapping 36, no. 8 (2015): 3137–3153.
tial and Sustained Personal Meaning and Spiritual Significance.”
Psychopharmacology 187, no. 3 (2006): 268–283. Lenggenhager, Bigna, Tej Tadi, Thomas Metzinger, and Olaf Blanke.
Grinspoon, Lester, and James B. Bakalar. Psychedelic Drugs Reconsidered. “Video Ergo Sum: Manipulating Bodily Self-Consciousness.” Science
New York: Basic Books, 1979. 317, no. 5841 (2007): 1096–1099.
Guthrie, Stewart. Faces in the Clouds: A New Theory of Religion. New York: Lewis, I. M. Ecstatic Religion: A Study of Shamanism and Spirit Possession. 3rd
Oxford University Press, 1995. ed. New York: Routledge, 2003.
Habermas, Gary R. “Evidential Near-Death Experiences.” In The Blackwell Lindley, James H., Sethyn Bryan, and Bob Conley. “Near-Death
Companion to Substance Dualism, edited by Jonathan J. Loose, Angus J. L. Experiences in a Pacific Northwest American Population: The Evergreen
Menuge, and J. P. Moreland, 227–246. West Sussex, UK: Wiley-Blackwell, Study.” Anabiosis: The Journal of Near-Death Studies 1, no. 2 (1981):
2018. 104–124.
Halper, Phil, and Ali Nayeri. “Before the Big Bang.” In Christianity in the Lloyd, G. E. R. Magic, Reason and Experience: Studies in the Origin and
Light of Science: Critically Examining the World’s Largest Religion, edited Development of Greek Science. Cambridge, UK: Cambridge University
by John W. Loftus, 119−140. Amherst, NY: Prometheus Books, 2016. Press, 1979.

Science: Atheism
Lloyd, G. E. R. The Ideals of Inquiry: An Ancient History. New York: Oxford Findings and Methodological Issues.” Perceptual and Motor Skills 97, no.
University Press, 2016. 2 (2003): 625–630.
Lodge, Oliver. “Opening of an Envelope Containing a Posthumous Note Newton, Isaac. Opticks. New York: Dover, 1979. First published 1704.
Left by Mr. Myers.” Journal of the Society for Psychical Research 12, no.
Ofek, Hillel. “Why the Arabic World Turned Away from Science.” New
215 (January 1905): 11–13.
Atlantis: A Journal of Technology and Society 30 (Spring 2011): 3–23.
Loftus, John, ed. The Christian Delusion: Why Faith Fails. Amherst, NY:
Paley, William. Natural Theology. Oxford: Oxford University Press, 2006.
First published 1802.
Loftus, John. The Outsider Test for Faith: How to Know Which Religion Is Parnia, Sam, Ken Spearpoint, Gabriele de Vos, et al. “AWARE—Awareness
True. Amherst, NY: Prometheus Books, 2013. during Resuscitation—A Prospective Study.” Resuscitation: Official
Loftus, John, ed. Christianity in the Light of Science: Critically Examining the Journal of the European Resuscitation Council 85, no. 12 (December
World’s Largest Religion. Amherst, NY: Prometheus Books, 2016. 2014): 1799–1805.
Lutz, Antoine, John D. Dunne, and Richard J. Davidson. “Meditation and the Pasricha, Satwant, and Ian Stevenson. “Near-Death Experiences in India: A
Neuroscience of Consciousness.” In The Cambridge Handbook of Preliminary Report.” Journal of Nervous and Mental Disease 174, no. 3
Consciousness, edited by Philip D. Zelazo, Morris Moscovitch, and Evan (March 1986): 165–170.
Thompson, 499–555. Cambridge, UK: Cambridge University Press, 2007. Perry, Michael, and David Fontana. “Charles Fryer and His Tape.”
Martin, Michael, and Keith Augustine. The Myth of an Afterlife: The Case Paranormal Review 52 (October 2009): 12–13.
against Life after Death. Lanham, MD: Rowman and Littlefield, 2015. Philipse, Herman. God in the Age of Science: A Critique of Religious Reason.
Masters, Robert, and Jean Houston. The Varieties of Psychedelic Experience: Oxford: Oxford University Press, 2012.
The Classic Guide to the Effects of LSD on the Human Psyche. Rochester, Plantinga, Alvin. Warranted Christian Belief. New York: Oxford University
VT: Park Street Press, 2000. First published 1966. Press, 2000.
McMahan, David L., and Erik Braun, eds. Meditation, Buddhism, and Ricard, Matthieu, Antoine Lutz, and Richard J. Davidson. “Mind of the
Science. Oxford: Oxford University Press, 2017. Meditator.” Scientific American 311, no. 5 (2014): 38–45.
McNamara, Patrick, ed. Where God and Science Meet: How Brain and Richards, William A. “Entheogens in the Study of Religious Experiences:
Evolutionary Studies Alter Our Understanding of Religion. 3 vols. London: Current Status.” Journal of Religion and Health 44, no. 4 (2005): 377–389.
Praeger, 2006.
Ring, Kenneth. “Precognitive and Prophetic Visions in Near-Death
McNamara, Patrick. The Neuroscience of Religious Experience. Cambridge, Experiences.” Anabiosis: The Journal of Near-Death Studies 2, no. 1
UK: Cambridge University Press, 2009. (1982): 47–74.
McNamara, Patrick, and P. Monroe Butler. “The Neuropsychology of Ring, Kenneth. “Prophetic Visions in 1988: A Critical Reappraisal.” Journal
Religious Experience.” In Handbook of the Psychology of Religion and of Near-Death Studies 7, no. 1 (September 1988): 4–18.
Spirituality, 2nd ed., edited by Raymond F. Paloutzian and Crystal L.
Park, 215–233. New York: Guilford Press, 2013. Rivas, Titus, Anny Dirven, and Rudolf H. Smit. The Self Does Not Die:
Verified Paranormal Phenomena from Near-Death Experiences. Durham,
Melnyk, Andrew. A Physicalist Manifesto: Thoroughly Modern Materialism. NC: International Association for Near-Death Studies, 2016.
Sansbury, Timothy. “The False Promise of Quantum Mechanics.” Zygon 42,
Murphy, Nancy. Bodies and Souls, or Spirited Bodies? Cambridge, UK: no. 1 (2007): 111–122.
Schellenberg, J. L. The Hiddenness Argument: Philosophy’s New Challenge to
Muthukumaraswamy, Suresh D., Robin L. Carhart-Harris, Rosalyn J. Belief in God. Oxford: Oxford University Press, 2015.
Moran, et al. “Broadband Cortical Desynchronization Underlies the
Human Psychedelic State.” Journal of Neuroscience 33, no. 38 (2013): Schjoedt, Uffe. “The Religious Brain: A General Introduction to the
15171–15183. Experimental Neuroscience of Religion.” Method and Theory in the Study
of Religion 21, no. 3 (2009): 310–339.
Nanda, Meera. The God Market: How Globalization Is Making India More
Schwartz, Gary E. R., and Linda G. S. Russek. “Celebrating Susy Smith’s
Hindu. New York: Monthly Review Press, 2011.
Soul: Preliminary Evidence for the Continuance of Smith’s Conscious-
Nelson, Kevin. The Spiritual Doorway in the Brain: A Neurologist’s Search for ness after Her Physical Death.” Journal of Religion and Psychical Research
the God Experience. New York: Penguin, 2010. 24, no. 2 (April 2001): 82–91.
Neubauer, Raymond L. “Prayer as an Interpersonal Relationship: A Sedley, David. Creationism and Its Critics in Antiquity. Berkeley: University
Neuroimaging Study.” Religion, Brain and Behavior 4, no. 2 (2014): of California Press, 2008.
92–103.
Shanon, Benny. The Antipodes of the Mind: Charting the Phenomenology of the
Newberg, Andrew. Principles of Neurotheology. Farnham, UK: Ashgate, 2010. Ayahuasca Experience. Oxford: Oxford University Press, 2002.
Newberg, Andrew, Eugene D. d’Aquili, and Vince Rause. Why God Won’t Shermer, Michael. The Moral Arc: How Science and Reason Lead Humanity
Go Away: Brain Science and the Biology of Belief. New York: Ballantine toward Truth, Justice, and Freedom. New York: Henry Holt, 2015.
Books, 2001.
Siegel, Ronald K. Intoxication: The Universal Drive for Mind-Altering
Newberg, Andrew, Michael Pourdehnad, Abass Alavi, and Eugene D. Substances. Rochester, VT: Park Street Press, 2005. First published
d’Aquili. “Cerebral Blood Flow during Meditative Prayer: Preliminary 1989.
Science: Atheism
Smith, Huston. Cleansing the Doors of Perception: The Religious Significance of Study in the Netherlands.” Lancet 358, no. 9298 (December 15, 2001):
Entheogenic Plants and Chemicals. New York: Jeremy P. Tarcher/ 2039–2045.
Putnam, 2000.
Varela, Francisco, ed. Sleeping, Dreaming and Dying: An Exploration of
Steiner, Rudolf. Nine Lectures on Bees. Spring Valley, NY: St. George, 1975. Consciousness with the Dalai Lama. Boston: Wisdom, 1997.
Stenger, Victor. The Unconscious Quantum: Metaphysics in Modern Physics Veneziano, Gabriele. “The Myth of the Beginning of Time.” Scientific
and Cosmology. Amherst, NY: Prometheus Books, 1995. American 290, no. 5 (2004): 54–65.
Stenger, Victor J. The Fallacy of Fine-Tuning: Why the Universe Is Not Victoria, Brian Daizen. Zen War Stories. London: Routledge, 2003.
Designed for Us. Amherst, NY: Prometheus Books, 2011.
Victoria, Brian Daizen. Zen at War. Lanham, MD: Rowman and Littlefield,
Stevenson, Ian, Emily Williams Cook, and Nicholas McClean-Rice. “Are 2006.
Persons Reporting ‘Near-Death Experiences’ Really Near Death? A
Vyse, Stuart A. Believing in Magic: The Psychology of Superstition. 2nd ed.
Study of Medical Records.” Omega: Journal of Death and Dying 20, no. 1
(1989–1990): 45–54.
Wainwright, William J. Mysticism: A Study of Its Nature, Cognitive Value and
Stevenson, Ian, Arthur T. Oram, and Betty Markwick. “Two Tests of
Moral Implications. Madison: University of Wisconsin Press, 1981.
Survival after Death: Report on Negative Results.” Journal of the Society
for Psychical Research 55, no. 815 (1989): 329–336. Wallace, B. Alan, ed. Buddhism and Science: Breaking New Ground. New
York: Columbia University Press, 2003.
Strassman, Rick. “Hallucinogens.” In Mind-Altering Drugs: The Science of
Subjective Experience, edited by Mitch Earlywine, 49–85. Oxford: Oxford Wallace, B. Alan. Contemplative Science: Where Buddhism and Neuroscience
University Press, 2005, 49–85. Converge. New York: Columbia University Press, 2007.
Swinburne, Richard. The Existence of God. 2nd ed. Oxford: Oxford Wallace, B. Alan, and Brian Hodel. Embracing Mind: The Common Ground
University Press, 2004. of Science and Spirituality. Boston: Shambhala, 2008.
Van Lommel, Pim. “Near-Death Experience, Consciousness, and the Brain: Wright, Robert. Why Buddhism Is True: The Science and Philosophy of
A New Concept about the Continuity of Our Consciousness Based on Meditation and Enlightenment. New York: Simon and Schuster, 2017.
Recent Scientific Research on Near-Death Experience in Survivors of Young, Matt, and Taner Edis, eds. Why Intelligent Design Fails: A Scientific
Cardiac Arrest.” World Futures: The Journal of General Evolution 62, nos. Critique of the New Creationism. New Brunswick, NJ: Rutgers University
1–2 (2006): 134–151. Press, 2004.
Van Lommel, Pim, Ruud van Wees, Vincent Meyers, and Ingrid Elfferich. Zaehner, R. C. Drugs, Mysticism, and Make-Believe. London: Collins,
“Near-Death Experiences in Survivors of Cardiac Arrest: A Prospective 1972.

TOPIC 18
Theories of Religion: Theism

Margaret I. Hughes
Tutor
Thomas Aquinas College, Santa Paula, CA
After giving a brief overview of the history of the development of theories of religion, this chapter delves into
scientism, which is the foundation on which many contemporary theories rest. It considers the way in which
scientism has come to dominate a certain approach to studying religion, and then offers a critique of this
approach. It concludes that faith is necessary in order to come to a fuller understanding of the phenomenon
of religion because, unlike an approach based in scientism, faith allows for an openness to finding what is
true in any religion.
THE AIM OF A THEORY OF RELIGION
A theory of religion is aimed at understanding religion. The word “theory” derives from the Greek
word theoria, which contrasts with the word praxis. Whereas praxis is knowledge of how to act or
how to make—practical knowledge—theoria is knowledge for its own sake—knowledge simply for
the sake of understanding. While, of course, theory must underlie praxis—we cannot know how
to act or to make without some preceding understanding—the final aim of theory is simply under-
standing. Understanding comes, ultimately, from knowing “why.” And so, theories of religion tend
to look for an explanation of religion.
For most theorists of religion, the aim is not to understand a particular religion, although their
method may involve studying a particular religion in depth, but rather to understand religion itself.
Theories of religion, then, are interested in explaining the phenomena of religion. They aim at
explaining why there is religion, why human beings would engage in religious practices, or why they
would assent to religious beliefs.
THE DEVELOPMENT OF THEORIES OF RELIGION
While the study of religion as such came to the fore especially in the nineteenth century, a case
could be made for tracing the questions that theorists of religion investigate even to the ancient
Greeks. Herodotus, for example, is aware of different names used for Greek and Egyptian gods
and engages in a kind of study of comparative religion. Plato famously bans the poets from the
ideal city, in Book X of the Republic, because they present the gods as behaving in a way not fit-
ting with the divine. Later, Marcus Tullius Cicero writes a treatise On Divination, in which he
examines the validity of divination, the practice of reading signs in nature as a communication
from the gods. These are a few instances of ancient authors examining religion and thinking about
its purpose and place in human life.
The Western intellectual tradition as it developed in the Middle Ages was dominated by three
monotheistic religions—Judaism, Islam, and unfractured Christianity. (Although Christianity was
divided in 1054 between East and West by the Great Schism, medieval Christianity had not yet
605
broken into the multifarious denominations that are present

KEY CONCEPTS today.) The viewpoint of each religion allowed for an under-
standing of the others. Put quite basically—and lacking much
nuance and the sort of explanation that space here does not
Fideism
allow—to Christians, Judaism was the precursor of and prepa-
Natural Faith ration for Christianity and Islam a heresy, which means that
Nominalism Islam possessed some of the truth but misunderstood it in some
way. To Muslims, Judaism and Christianity were precursors to
Scientism
the final revelation given to Muhammad. And to Judaism,
Supernatural Faith Christianity and Islam were heresies. It is certainly the case that
thinkers representing each tradition engaged in what is now
called “philosophy of religion,” in which reason is brought to
bear on questions about the nature of God, the relation
between God and believers, and between God and nonbelievers. These thinkers, however, were
operating at a place and time in which monotheism was an essential part of the normal worldview,
along with the worship of God through prescribed rituals. This made the question of how to
account for the existence of these other religions less complicated.
In the fifteenth and sixteenth centuries, however, Europeans began to come into contact with
peoples who engaged in practices that were recognizably religious but who were not monotheists.
The breakdown of the Mongol Empire and the subsequent need to replace overland trade routes
with safer sea routes led to an age of exploration in which Europeans discovered and explored south-
ern Africa, the Americas, and Australia. As trade with India, China, and Japan increased, Europeans
encountered, for the first time since the Christian evangelization of the barbarians, peoples who
believed in gods, not a single God, and whose religious practices differed vastly from those of Chris-
tianity, Judaism, and Islam. That there were other religious beliefs and practices than those of the
three monotheistic religions cast a new light on religion as something that could be studied from
the outside, as opposed to living and thinking about it from within.
The realization that there are distinct religions and religious practices raises the question of
what religion is at all—what constitutes a religious belief and a religious practice. It is Plato’s ques-
tion (Meno 72b) that has reverberated throughout philosophy and is now applied to religion: How
do we account for there being one among many? How can it be that there are many different beliefs
and practices, and yet all belong to the category “religion”? What is it that everything that is religion
has in common and that sets religion apart from that which is not religion?
This chapter considers especially the intellectual foundation of the attempts of modern and
contemporary thinkers to answer these questions. Those who undertake the study of religion itself
often do so from the position of scientism. The discussion that follows considers first what scientism
is and then various ways of understanding the relation between faith and reason in order to show
that attempts to develop a “scientific” theory of religion (that is, a theory of religion that adheres
to the standards of modern science) violates the standard of objectivity that is the attraction of such
a method. All theories of religion, it is asserted here, are affected by whether the researcher himself
has some religious faith or not.
RELIGION AND THE STUDY OF HUMAN BEINGS
The most obvious element of the phenomenon of religion, of course, is that it is something specifi-
cally human. Other animals give no indication of engaging in religious practices or of holding reli-
gious beliefs. And, unlike many other subjects that we study, the life cycle of plants or the move-
ment of the planets, for example, it is clear that, if there were no human beings, there would be
no religion. As a result, many theorists of religion conclude that an understanding of religion can
come only through a study of human beings.
Consequently, theories of religion emerge especially out of the social sciences, especially psy-
chology and sociology. Beginning with the observation that religion is a human phenomenon, most
theorists of religion look for an explanation of that phenomenon that is grounded in the study of
human beings. Generally, these theories look for an explanation of the religiosity of human beings
as somehow fulfilling a human need or as the outgrowth of some human malady.
Sigmund Freud, for example, whose psychological theories increasingly have been called into
question but who has left a lasting mark on the way many people think of religion, explains religion
as being caused by psychological neuroses. People who are religious, he holds—and here, he is
thinking primarily of Judaism and Christianity—are so because of a psychological need left unful-
filled, especially through the relations of individuals with their fathers or mothers. Religion is an
illusion, Freud (1989) asserts, caused by the restraining of impulses in ways that leave needs unful-
filled. Religion is the illusion that allows a person to feel as though those needs are fulfilled. All reli-
gion, then, according to Freud, can be explained as a response to mental disturbances.
Karl Marx, likewise, looks for the cause of religion and finds it in a distortion of human society.
He begins with the presumption that all that is, is material, and with the understanding that human
beings are producers. They work on the material world to change it so as to fulfill needs necessary
for survival. As those needs are fulfilled, however, they create other needs, and still further needs.
The story of history, according to Marx (1994), is the story of the development of forms of society
that arise from each human being making what is necessary for his survival, a development from
tribes that hold everything in common to an economic system in which there are owners and those
who work for the owners. The workers sell their labor in order to earn a wage to survive but then
work longer than is needed for survival. This surplus time becomes the profit of the owner, who
sells the product that the worker has produced. The product, then, does not belong to the worker
or contribute to his survival. In this way, the worker becomes alienated from his product, from
his work, from others, and ultimately from himself. This alienation is the cause of great misery
and strife. If it is not addressed, it will lead to workers rising up against owners. As a result, owners
invent and spread religion so that workers are both comforted and dissuaded from revolution. Reli-
gious belief and practice provide a means of dulling the pain caused by capitalism.
While Freud and Marx understand religion as a symptom of the distortion of human life, other
theorists of religion also look for the cause of religion in human life, even though they do not
understand the cause to be necessarily nefarious. Émile Durkheim (1976), for example, understands
the cause of religion to be the vital importance of the community to human life. Religion is the
worship of the community, as evidenced by the role of totems in religion, which, Durkheim asserts,
are concrete images of the community. Living in a community is essential to human well-being, and
it easily becomes the focus of the community’s life. Religious practices and beliefs are for the sake of
the preservation and strengthening of the community.
Another proposed cause of religion is the human desire to understand and to seek for explanations.
According to Edward B. Tylor (2010) and James George Frazer (1996), who are often called the first
theorists of religion, religion develops out of the human impulse to explain why the world, and espe-
cially nature, acts as it does. Like Durkheim, they focus their studies on so-called primitive religions
rather than on the monotheistic religions of the West. They look for an explanation of the phenomena
of religion by trying to identify its historical starting point. Both take an evolutionary stance toward
religion, by beginning the explanation of religion with musings about prehistoric religion and how it
developed. Religion, they assert, is a human invention that attempts to explain the world and so to sat-
isfy the human desire to know. They conclude that human beings are evolving beyond religion because
science can now do what religion once did. They understand science and religion as having the same
function, which is to explain why the world works as it does, but as offering contradictory explanations.
As a result, they understand science and religion as opposed to each other.
This position is one that has gripped the popular imagination. For example, there have been
many books and movies that present the trial of Galileo as a contest of science against religion.
The same is true in accounts of figures such as Charles Darwin and Giordano Bruno. Prominent
figures in the “new atheism” movement, such as Richard Dawkins (2008) and Christopher Hitch-
ens (2007), have written popular books for a wider audience that begin with the assumption that
there is a tension between science and religion. A 2015 Pew Research study (reported on by Cary
Funk and Becka A. Alper) found that 59 percent of adults in the United States think that religion
and science are in conflict with each other.

The study of religion as such—that is, not the study of theology or of a particular religion, but
the study whose end is religion itself—is pursued today almost exclusively in the academy. Theorists
of religion tend to be university professors and other researchers working within a university setting.
It is important to note the university context of the study of religion because of the ways that con-
text shapes such study.
SCIENTISM
Contemporary university thought and culture tends toward scientism. While it can be used as a
pejorative term, scientism is used here simply to convey a particular epistemology and consequent
worldview, which this chapter critiques. Underlying scientism is the notion that the model for all
areas of study should be the so-called hard sciences. That is, scientism is the view that the methods
used in the investigation of the physical world—physics, chemistry, biology, astronomy, and so
on—are the best, and, in extreme versions, the only, way of coming to knowledge about anything.
Scientism adheres to a standard of knowledge that privileges the notion of certainty. The more cer-
tain a conclusion is—that is, the less it admits of doubt—the more readily it is accepted as knowl-
edge. In a culture shaped by scientism, only what cannot be doubted is taken to be knowledge.
The method of the sciences, based as it is on mathematics, is taken to offer a way to the greatest
possible certainty because of the static and definite nature of numbers and their relationships. With-
out entering into the discussion about whether numbers are real or are ideas abstracted from what is
real, it is possible to see that numbers are fixed and exact. The number two is simply that; it does
not change, it is neither less nor more, but simply is what it is. Mathematical operations, which con-
sider the relations between definite numbers, such that those relations are definite and lead to defi-
nite conclusions, are the source of definite, and so certain, conclusions.
Hence, only studies that are based on numbers appear to lead to knowledge. This need to pro-
vide a mathematical foundation for all inquiry requires developing an empirical account of any sub-
ject under investigation through numerical measurement. Such measurement allows for an assess-
ment of the elements of the subject into units to which numbers are assigned. Those numbers, in
turn, can be used in mathematical operations so as to reach conclusions that the method regards
as certain, and so they are accepted as knowledge.
The proper use of measurement, however, requires comparison between the standard that is the
measure and the thing that is being measured; the one must be held up against the other. As a
result, only things that can be sensed, or that produce sensible effects, can be measured in numbered
units. And, for something to be sensible, it must be physical. Only something that is physical—that
is, having a body—can have length, width, height, weight, and so on. All numerical measurements
are ultimately grounded in observations about the physical world.
Consequently, in scientism, the subject of study is phenomena, what we can observe with our
senses, measure, and so come to undoubtable knowledge. Scientism concentrates on appearances
and on applying methodology that as closely resembles mathematical operations as possible in order
to come to definite conclusions. Only phenomena, what appears to our senses, can be measured,
and so the only possible knowledge is of phenomena. Appearance, therefore, is considered to be
all that is knowable and so is regarded as most real.
As a result, the only fields of study thought to be deserving of attention and capable of leading
to knowledge are those that are restricted to phenomena and able to be measured scientifically. This
move is apparent in the grouping under the title “social science” of the academic areas of economics,
psychology, sociology, politics, and occasionally history. The aim of the social sciences, as the term
suggests, is to study human beings and their interactions by applying the “scientific” method to
them. This means, then, that the social sciences, like the hard sciences, study measurable phenom-
ena in order to come to conclusions that are taken to be certain because they are the result of math-
ematical operations.
This approach to the study of human beings seems appealing because it appears to come to
objective conclusions without being tainted by the partiality or preconceptions of the researcher.
The scientific method, applied to human subjects, aims to discover how things actually are, and not
how the researcher already thinks they are or wants them to be. Just as a researcher might study the
movement of particles or the life cycle of a pine tree objectively, without his own commitments or
prejudices influencing the outcome, by distancing himself from the subject and studying only the
measurable appearance without any personal involvement on his part, social scientists aim to study
human phenomena in the same way. Their aim is to observe and to measure those observations,
which is possible only if they stand at a distance with an “objective” point of view. The researcher
must consider the human phenomena as he does the particles or the pine trees, with an openness
to whatever he might find and with care not to allow his own commitments and preconceptions
to blind him.
Many theories of religion, especially those that come out of the social sciences, attempt to adopt
such a scientific approach to the study of religion. The subject of the study, therefore, is observable
phenomena, and the object of the study is to come to discover the cause of the phenomena through
a method that allows no room for doubt. This goal can only be achieved, however, when the
researcher takes an objective stance, such that his own commitments do not color his observations
and he is able to see religious behavior and belief as it really is.
THE PROBLEM WITH SCIENTISM
For a theist, developing a theory of religion by using the methods of “science” is problematic
because of the presumptions that scientism makes. Scientism begins with the assumption that
human reason can only access appearances, such that human knowledge can only be of what can
be observed with the senses and measured. It also presumes that an objective stance toward religion
is possible.
A theorist of religion who approaches the study of religion as if it were a science, however,
begins with the presumption that whatever claims practitioners of that religion make about the
divine, the supernatural, and what is beyond material appearances, is, at best, beyond the scope of
investigation by reason, and, at worst, is simply false. At any rate, he rejects religious belief as a kind
of knowledge. But this is not an objective stance.
Freud and Marx, most famously, are good examples of this tendency. Both develop their theo-
ries of religion after having already rejected the possibility that any religion or religious claims might
be true. Each sees the phenomena of religion as a perplexing question that requires explanation pre-
cisely because they do not accept the explanation that the claims of any religion might be true. They
aim to explain how it is that believers could come to assent to theistic beliefs and practice a religion,
given that human beings tend to develop beliefs and actions based on concrete reality, but that the
claims of religion are not based on reality. As the penultimate section of this essay will suggest, this
distrust of faith ultimately undermines the very possibility of a valid study of religion because such a
study in principle fails to meet the scientific standards of “objectivity.”
The work of theorists of religion who argue that religion must be studied religiously, that is,
that religion can only be understood if the faith claims of a religion are taken seriously, is derided
and dismissed by those theorists of religion who begin from an atheistic or agnostic position. For
example, Mircea Eliade (1987) suggests that the explanation for a religion may, in fact, be the expla-
nation that the participant in the religion offers. He argues that a theorist of religion who himself is
not religious is at a disadvantage in studying religion, because he will look to other areas of study—
psychology, economics, and so on—to try to understand religion, rather than looking at religion
itself for its explanation. This position has been roundly criticized by other theorists of religion,
who claim that it stems not from a scientific, objective stance but from religious prejudice.
There are other theorists of religion who take a less extreme view than that of Freud and Marx.
Rather than begin with the presumption that religious belief is untrue and simply the result of
human construction, they begin with the presumption that any religious belief is unreasonable—
that is, not grounded within the realm of reason and so unable to be evaluated by human reason.
As a result, they draw a distinction between faith and reason or knowledge. Faith, they hold, cannot

be a source of knowledge, since knowledge is the fruit of reason, but faith is opposed to reason. As a
result, there is no way to study the contents of faith. All that is open to study is the phenomenon of
religion; religion can be studied only from the outside, since study requires human reason, but the
contents of religious belief are outside the realm of human reason. Robert Segal maintains that most
twentieth-century theorists of religion hold this position (2010, 76).
Given that much of the study of religion and development of theories of religion takes place
within the field of social science, which is founded on the notion that “science” is the best or only
route to knowledge, such that all that can be known arises from observable phenomena, much of
the study of religion and many theories of religion have a fundamental distrust of faith, since the
contents of faith are beyond the realm of observable phenomena and so cannot be submitted to
the methods of science.
THE QUESTION OF FAITH AND REASON
Ironically, given that many proponents of scientism understand science as countering and presenting
an alternative to religion, both the understanding of religion as centered around faith and the rejec-
tion of faith as a means of knowledge grew out of religion, and especially Christian, thought. This is
important to realize because many theorists of religion, particularly those who themselves are not
religious, identify all religion as distinguished by the centrality of faith and all religious people as dis-
tinguished by activity that is motivated by faith.
Although religion in general is often dismissed because it is based on faith claims, the associa-
tion of faith with religion—the emphasis on faith as an essential aspect of religion—is primarily
Christian. While Christianity necessarily involves practice, what makes someone a Christian is not
primarily some practice but assent to Christian belief, which admittedly should lead to particular
practices and ways of living. By contrast, Judaism and Islam focus primarily on adherence to the
law (Torah, sharia). While that adherence may entail faith, it is not essential to that practice. What
matters is that the law is followed. There is little emphasis on the importance of faith as the moti-
vation for following the law. Whereas Christianity is concerned first with orthodoxy—true
belief—which entails a certain way of life and religious practice, Judaism and Islam tend toward
an emphasis on orthopraxy—true practice.
Likewise, in the “primitive” religions (so called because they became known to the Western
world later than Judaism and Christianity) and as a result of the application of the theory of evolu-
tion to religion, such that religion is understood to evolve or progress from what is primitive to what
is more sophisticated or complex, we have evidence of practice and explanatory stories. While it is
clear that, for “primitive” people, religion was a practice with specific rituals, it is less clear that those
practices required belief. To see that religious practice does not necessarily entail belief, one need
only consider ancient Rome, in which religious practice was taken very seriously, but few, at least
among the learned, actually took the stories about the Roman gods to be true.
Christianity, on the other hand, is first and foremost a religion based on faith. What deter-
mines whether someone is a Christian is not whether he has been born to a particular group of
people, or some external practice, but whether he chooses to believe the witnesses to the words
and actions of Jesus. Indeed, some of these words include the very emphasis on faith that is at
the heart of Christianity: “Let it be done for you according to your faith” (Matthew 9:28); Jesus’s
admonishment of Peter: “O you of little faith, why did you doubt?” (Matthew 14:31); his com-
forting of the woman suffering from hemorrhaging: “Your faith has healed you” (Mark 5:34);
and his chastisement of Thomas: “Blessed are those who have not seen and yet have believed”
(John 20:29). The Gospels are filled with similar instances of Jesus’s exhorting his followers to
believe in Him and to be faithful.
It was the relation between faith and practice, and the question of which should be primary in
the Christian life, that played a role in bringing to the fore scientism and the rejection of faith as a
legitimate source of knowledge, even in the study of religion. The Protestant Reformation stressed
that only faith is necessary for salvation. While good works may follow on faith, the works
themselves cannot be conducive to salvation because that would suggest that, through works, a
human being can coerce God to save that human being. If, through works, human beings can bring
about their own salvation, then it is human beings, and not God, who choose who will be saved.
But if it is human beings, and not God, who determine who will be saved, then God’s freedom
to choose—even to choose to damn someone who does good works—is limited. But this conclusion
would involve a logical contradiction, for God is an unlimited being and so God’s freedom, too
must be unlimited.
The position that Catholics hold, and have held, on this position is that both faith and
practice—faith and works—are necessary for salvation. Neither one nor the other is sufficient in
itself, and neither is independent of the other. Genuine faith necessarily leads to practice and good
works, and underlying all genuine practice and good works is faith. This position follows from an
understanding of the human being as a whole. A human being, while he has different faculties, is
a single being, a unified whole. As a result, the faith to which he assents must necessarily affect
his external actions, and those external actions will further shape his inner life. Because the human
being is not a set of parts that operate parallel to each other without intersecting but a single, inte-
grated whole, the faith and works of a human being must be seen as coming from individuals, each
of whom is inwardly unified as a whole.
To say that the human being is a single, integrated whole is to say that he is a substance with a
particular nature that causes him to be the kind of thing that he is. “Substance” is, in Aristotelian
metaphysics, one of the most basic concepts for understanding all that is. Aristotle comes to the
notion of substance by thinking through the problem of change and investigating how it is that
something can both change and yet remain what it is. How can it be, for example, that an apple
can change from green to red, and yet still be the same apple? Aristotle’s answer is that there must
be something underlying the change that remains the same throughout the change, such that,
although a quality changes, the thing itself maintains its identity. What remains the same while
undergoing a change in its features and relationships is what he calls a substance.
It is this understanding of relative permanence and the changes that substances can undergo
that gives rise to what we now think of as science. While what the substance is cannot be known
directly by the senses but only by the mind, its qualities are accessible to the senses, and those qual-
ities point to what kind of thing a given substance is. Science investigates those qualities so that,
based on the knowledge accrued through science properly understood, it is possible to come to
metaphysical conclusions about what the substance is. Empirical science, then—the study of observ-
able phenomena—is at the service of the knowledge of the nature of things and ultimately the
knowledge of metaphysics—the study of substances by abstracting from observable phenomena in
order to come to know the substance itself.
It is worth noting here what Aristotle, and those who pursue Aristotelian metaphysics, may claim
that they can come to know. While the study of substance leads to knowledge, and we can come to
know the natures of things, this knowledge is never complete or final. No matter how much we dis-
cover about the nature of something, there will always be more to know about it. As Thomas Aquinas
observes in the prologue to Exposition on the Apostles’ Creed, “Our manner of knowing is so weak that
no philosopher could perfectly investigate the nature of even one little fly” (1939). In making this
observation, Thomas echoes Aristotle, who writes, “For as the eyes of bats are to the blaze of day,
so is the reason in our soul to the things which are by nature most evident of all” (Metaphysics
993b). This position both has confidence in human reason to attain knowledge about the physical
world, and yet also maintains an epistemic humility that recognizes that, however much we know,
there will always be more to discover. It recognizes the importance of empirical knowledge through
measurement, but it does not restrict knowledge to what can be attained through empirical study.
FAITH OVER REASON?
Historically, the Catholic position on the role of faith and works found confirmation in its scrip-
tures and in experience for having confidence in the human ability to know the sensory world

and what is beyond the sensible. The philosophical basis for science and for metaphysics comes
from reflection on the human ability to abstract, tempered with the humility involved in recogniz-
ing that, no matter how much a human being knows, there will always be more to know. But dur-
ing the Protestant Reformation there was an emphasis on faith over works as the result of a suspi-
cion of human reason.
In response to the temporal corruption in the Catholic Church, a number of reformers—chief
among them Martin Luther, John Calvin, and Huldrych Zwingli—broke from the Catholic
Church, declaring that it could not be the legitimate church founded by Christ because of the cor-
ruption of its members. Indeed, one of the positions that the reformers adopted, contra the Catholic
position on the effects of the Fall, is the absolute depravity of human beings. While the Catholic
position holds that original sin wounded human nature, such that people continue to be inclined
toward the proper human end of seeking what is true and good but struggle in achieving that
end, the Protestant position holds that original sin totally deforms human nature, such that it is
so thoroughly shot through with an inclination against God that the human being is utterly incapa-
ble of reaching his end. Instead of maintaining a tempered confidence in the human ability to know
through the power of reason, the Protestant position often tends toward a complete distrust of
human reason and a reliance on revelation and faith that is totally independent of it. Human reason
is so deformed that it cannot be trusted to reach knowledge. Some forms of Protestantism still
maintain fideistic positions of this type.
This epistemological position develops out of the late Scholastic concern about the freedom of
God and the rise of nominalism. Nominalism is the notion that there are no natures, or at least that
human beings are incapable of knowing natures. Instead, human beings recognize similar character-
istics among things, and give those things that have enough characteristics in common the same
name. The names (Latin, nomina), however, do not point to a common underlying nature that
accounts for those characteristics. The result is a metaphysics of extrinsic denomination.
The reason for rejecting the reality of natures and the possibility of our knowing these natures
arise from a concern about the freedom of God. Since God is omnipotent, he must be able to do
whatever he chooses to do. That is, because God is omnipotent, he must be understood to be
completely free, with no limitations placed on his actions. A nature, however, is a principle of limi-
tation. So long as something has any particular nature, that thing is limited to acting in the ways
typical of that kind of thing and no other. Curiously, this recognition of the way in which things
are limited by their nature limits not only those things but also God. If a thing has a particular
nature, then its end (that toward which its internal principle of motion is aimed) is determined. A
kitten, for example, grows into a cat; an acorn grows into an oak tree. But, if this is the case, then
even God cannot choose to cause a kitten to grow into something other than a cat, or an acorn to
grow into something other than an oak tree. The notion of natures, according to this position,
seems to remove power from God and to place it in the hands of “nature.” To do so thus apparently
contradicts the omnipotence and freedom of God.
To hold that there are no natures that human beings can know, therefore, was regarded as con-
sonant both with the notion that human beings are so deformed by original sin that they cannot,
through their own power, know reality, and with the emphasis on the power of God in the face
of the total impotence and depravity of human beings. As a result, many Protestant reformers reject
Aristotelian metaphysics, with its confidence in human reason and its concern for natures and the
investigation of substances.
Instead, in order to preserve the freedom of God and their commitment to the depravity of
human beings, the reformers emphasized faith as the best and only sure way of having knowledge,
especially knowledge of God. Calvin, for example, in the Institutes of the Christian Religion, holds
that, while God implants in every human being a natural knowledge of him, and although the order
of creation points to its creator, this knowledge, as a result of original sin and depraved human
nature, lies “confused in our minds” ([1536] 1989, bk. 1, ch. 6) until scripture clarifies it. Clear
knowledge of God can come only through accepting what God tells man of himself, through the
revelation found in scripture. Even tradition—the handing down of beliefs and practices from one
human being to the next, and from one generation to another—is suspect because of the deformed
nature of those handing it down. Knowledge of God and any practice that is pleasing to God comes
directly from God to man through grace.
The decoupling of faith and reason that characterizes scientism emerged from a position that
originally aimed at the exaltation of faith over reason and that was motivated by a basic distrust of
human reason and its ability to reach any knowledge beyond what human beings sense.
REASON OVER FAITH?
The unfolding of history after the initial break of the reformers from the Catholic Church, however,
led to a distrust of faith and so to an exclusive embrace of human reason, tempered by the Protestant
distrust in any claims of reason to know with certainty anything beyond what can be sensed. In this
way, we arrive at the scientism that marks so much of the methods underlying various theories of reli-
gion. After the subsequent splintering of the reformers, there has never again been unity within West-
ern Christendom. This disunity had political consequences. Henry VIII, for instance, declared himself
the head of the church in England. Elsewhere in Europe, political leaders took advantage of the dis-
unity among the faithful in order to gain political power. This is a complex, intricate history that can-
not be recounted here, even in sweeping overview. It suffices to observe that the wars of religion in
France (1562–1598), the bloody clashes between Catholics and Protestants in England and Ireland,
the eighty years of war between the Dutch and the Spanish (1568–1648), and the thirty years of
war in central Europe (1618–1648) led to the notion that religion and its foundation in faith are sim-
ply causes of strife and excuses to manipulate the unthinking faithful while seizing power.
Since faith led to such disagreement, dissension, and strife, it came to be seen as something
merely subjective and not as a legitimate way of accessing reality. Whereas the reformers divorced
faith from reason as a result of a lack of confidence in reason and as a way of emphasizing the neces-
sity of faith in God, a mere two hundred years after Luther issued his Ninety-Five Theses in
Wittenberg, Benjamin Franklin, the purveyor of much popular culture through his Poor Richard’s
Almanack, wrote: “The way to see by faith is to shut the eye of reason” (1914, 51). Because faith
leads to division, it must be that faith is not about the objective world, which is something that
everyone shares and on which everyone ultimately can come to agree as the result of observations,
testing, and other aspects of the scientific method. Faith, on the other hand, since it leads to dis-
agreement, must not be grounded in the objective world. If it were grounded in the objective world
that we share, it would lead to agreement. Instead, faith is taken to be a subjective position that
individuals come to on the basis of their own feelings, perceptions, and personal experiences.
Consequently, very few theorists of religion investigate whether a religion or religious faith is
true. Indeed, few theorists even accept the notion that someone might accept the propositions of
faith simply because he thinks that those propositions correspond with reality. Instead they look
for an explanation of religion beyond religion itself. They seek other motivations for the people
who practice and have religious faith.
ANOTHER NOTION OF FAITH
There is, however, a way of understanding faith as making claims rooted in objective reality and so
as either true or false. When we remove from the word “belief” the connotations that it acquired in
the modern period, it becomes clear that we can take the statement “I believe” to mean “I accept as
true.” Someone who says “I believe x, but I think x is false” would make no sense. We would not
even be sure what that person means. If someone believes, it is because he accepts the content of
that belief as true.
The difference between reason and faith is not that reason is concerned with objective reality and
truth, while faith is subjective and untrue. Rather, reason and faith are two different means by which
we access reality. Knowledge is often the result of a direct encounter with the content of that knowl-
edge. We can come to know something because we see it directly, or it might be that we see the

effects of something and then reason in order to come to know something about the cause of those
effects, even if we cannot directly see the cause. This is, indeed, the method that scientific investiga-
tion employs. If an experiment is designed so that there can be only one explanation for its results,
the experiment can lead us to a discovery of something true. We might know something because
we see it immediately, or because we see the effects of something, and we may have to undertake a
considerable amount of reasoning in order to understand how and why those effects occur.
The demands of everyday life do not permit us to see directly everything that we must know in
order to live. It is impossible for any one human being to have discovered, through seeing directly or
through experimentation, even those few things that are most basic to natural survival. Much of the
“knowledge” that we act upon in daily life is not our own knowledge but the knowledge of someone
else whom we believe. This is especially the case during a time such as ours of great technological
innovation. There is so much that a human being uses on a daily basis of which he has no direct
knowledge; he himself has not performed the experimentation to see for himself how and why
something is as it is. As a result, much of even basic human survival depends on natural faith.
Faith of any sort is aimed at getting to the truth about things. Belief is distinguished from knowl-
edge not in its content but in its method of reaching that content. While knowing attains the truth
about something by seeing the thing or the reasoning about something for oneself, faith attains the
truth about the content of a faith-claim by assenting to the testimony of someone else who is a credi-
ble witness. Faith is believing not just something but someone. Faith allows us to attain to the truth
about the content without witnessing the reality for ourselves. Instead, we come to knowledge by
trusting someone who is himself a witness of the reality of the content. For many people, eating food
from a grocery store, being a passenger on a plane, having a car repaired or a drain unplugged requires
trusting others who know the reality of the safety of the food or the plane or the reality of the car’s or
drain’s inner workings. These beliefs are not cases in which we know the content directly on our own.
Rather, we place faith in those people who do know. That is, we believe them.
In both cases of natural faith and supernatural faith, faith requires a choice. Because the content
of the belief is not immediately evident, the prospective believer must choose to believe the witness
or not. While the prospective believer can certainly measure the consistency of the content of belief
against what he knows in some other way and can come to some judgment as to the witness’s trust-
worthiness, ultimately, his acceptance of the witness’s testimony is a choice—he chooses to believe
the witness or not.
This is not to say that one may choose to believe whatever one wants. The warrant for an act of
belief ultimately rests not on the will of the believer but on the reliability of the testimony of the
witness. The choice to believe is not a choice of content, but a choice of witness. The believer
chooses to accept a particular witness as trustworthy as a justification for the truth of the content
of his belief.
Knowing seems to be more certain because it does not involve such a choice. We cannot help
but know once direct evidence is before us. Someone who knows that 2 + 2 = 4 and that 3 + 1 = 4
cannot help but grant that 2 + 2 = 3 + 1. There is no choice involved except the choice to accept
(not to resist) the reasoning that two things that are equal to a given quantity (in this case, 4) are
also equal to each other; the content is so evident that the knower simply receives reality directly,
and the only thing that resembles a choice is the willingness of the person considering the question
to submit to evidence and argument.
Believing is less certain because it relies on a witness whom we choose to believe, and we can
choose well or ill. When we choose well—when we choose to believe a trustworthy witness, a wit-
ness who is telling the truth—then the content of a claim that we hold as an act of faith is as true as
the content of a claim for which we have knowledge by our own grasp of evidence and argument.
NATURAL AND SUPERNATURAL FAITH
That we must rely on the witness of others in order to function in the world can hardly be a con-
troversial idea. It is through trusting their parents and other adults, textbooks, and so on that
children become educated. While education, ideally, leads people to see things for themselves, and
so to know, much of education relies on the students’ faith in their teachers, such that the students
accept the testimony of their teachers, as their teachers have accepted the testimony of their own
teachers, and so on. A single human being cannot, on his own, come to know all that he must
know in order to live and thrive. He relies on the knowledge of others that he accepts through faith.
So far, however, this section and the previous one have discussed natural faith. The content
of natural faith is something that a human being can come to know on his own, provided that
he has the time, the aptitude, and the resources, to discover it. Natural faith is, in a way, a short-
cut to attain the content that could be had by a longer, more arduous process of coming to see
for oneself. The content of natural faith is something that can be discovered through human
powers.
Supernatural faith—the faith that is sometimes termed “religious”—differs from natural faith in
that the content of supernatural faith cannot be discovered through human ability alone. The origi-
nal knower whose testimony is accepted as an act of faith cannot be a human being alone. Instead,
as Josef Pieper (1997) points out in his treatise on faith, in religious faith the witness and the con-
tent are identical. Who is believed is also what is believed. That is, the ultimate witness in religious
faith, the one who has direct knowledge of God, is God Himself. Likewise, the content of what is
believed in religious faith is content about God. Religious faith is the choice to believe that God’s
self-revelation is true because of the source, who is God himself.
Admittedly, for someone who does not believe, this appears to be an insurmountable obstacle
to believing. The obstacle comes not from the content being unknowable but from the way in
which we need to access the content of faith. As in the case of natural faith, the content of supernat-
ural faith is knowable. The content itself can be understood and known, even if an individual
human being, or indeed, all human beings together, do not yet know it or understand it fully.
Understanding the content of faith is not what is difficult in this case but rather the warrant
for our assent to the truth of that content, since to assent to the truth of the content requires an
assent to the credibility of the witness of the content, which is one and the same thing. To believe
about God is also to believe God. To do so seems thoroughly irrational, until one observes, as Pie-
per (1997) does, that we do this regularly in human relations, especially in receiving the expres-
sion “I love you.” It may be true in the expression of any aspect of a person’s inner life, but it
is especially so in his reporting of what he loves, that the witness and content of one’s belief are
the same. It must be belief, and not knowledge, because no human being has access to another
human being’s inner life except through what that person chooses to express about himself and
his inner life. To accept that someone loves us is an act of faith that most closely resembles the
act of faith in believing God, since the only way to come to accept the truth of that content is
to choose to trust the testimony of the witness, who is the only one who can know the content
directly.
THE POSSIBILITY OF AN “OBJECTIVE” THEORY OF RELIGION
If God does in fact reveal himself, then to choose to believe him is thoroughly rational and leads
us to extend our intellects to more of reality than we can reach on our own. If God does reveal him-
self, but we choose not to believe him, then we limit ourselves and cut ourselves off from all that
there is.
At the start of the development of any theory of religion, therefore, the thinker who develops
that theory has already made a choice that will color his whole understanding of religion. Either
he has already made the decision that there is no trustworthy witness in supernatural matters, such
that no religion can be true, or he has made the decision to accept the testimony of a witness, such
that he accepts the truth claims of a religion. If the thinker begins with the presumption that a belief
is false or unwarranted, he will look to other human needs, and so on, for an explanation. If the
thinker begins with the presumption that the belief is true, then he will examine the religion differ-
ently. Someone who rejects the possibility of religion being true approaches the study of religion

from as “subjective” a position—or perhaps more so—as that of someone who approaches the study
of religion from a position of faith.
It is, of course, possible that a thinker may claim agnosticism. He may have suspended the
question of the truth or falsity of religious claims, such that he holds off from coming to a conclu-
sion about whether religious claims are true or false, or he may not yet have come to a conclusion,
and so simply does not know, at least for now, whether religious claims are true or false, although he
may leave open the possibility of coming to such a conclusion in the future.
Agnosticism may appear unbiased and objective, but it is not. While in the hard sciences
remaining “agnostic” is an objective position, in studying religion it is not. Since a theory of religion
attempts to account for the phenomenon of religion and to discover the cause of religious belief and
practice, whether the contents of that religion’s beliefs are true or not shapes the question and the
way in which the cause is sought.
The position laid out here about the impossibility of agnosticism is similar to one taken by
Joseph Ratzinger in “What Does It Mean to Believe?” (2006), in which he points out that agnosti-
cism is practically impossible. Either one lives as though God does exist, or one does not. When
choosing the everyday, practical activities that shape one’s life, one chooses either to shape one’s life
in reference to God or to live as though God does not exist; there is no other alternative. Likewise,
in looking for the cause of religion, a thinker investigates either as though God exists or as though
God does not exist. The beginning presumption about the truth or falsity of the content of religious
belief will shape thoroughly the conclusions of the study.
As such, the objective stance of the scientist that social science and many theorists of religion
aim to emulate in their study of religion is impossible in developing a theory of religion. The theo-
rist of religion must begin from an already formed position about the truth or falsity of the claims of
a religion. It is practically impossible to study religion as an agnostic.
A CHRISTIAN THEORY OF RELIGION
When someone looks for an explanation of religion from the perspective of faith, therefore, his
questions must be posed differently from those of theorists of religion who by and large begin from
the position of rejecting supernatural faith. For those who do think that God reveals himself to
human beings and makes clear to them both how they should act and how they should worship,
the number and variety of religions and religious practices across cultures and times is perplexing
in a certain way; those who reject the truth of any religion are perplexed by that variety in a differ-
ent way. Theorists who are perplexed by the phenomenon of religion as a whole look for an expla-
nation for religious belief and practice that is rooted in something other than that belief and prac-
tice. Someone who begins from a position of faith, however, must look for an explanation of
religious beliefs and practices other than his own, given that his faith is a commitment to the truth
of his own beliefs and practices. That is, he must explain why, if religion should be based on what is
true, and his religion is true, there are religions other than his. If his religion is revealed by God,
why are there other religions making other claims and engaging in other practices? What accounts
for the phenomena of false religions?
Beginning from the position outlined above, that there are natures that are knowable, such that
it is possible for reason to discover what something is, even if not yet fully real or fully disclosed to
us, it is possible to give an account of the religious nature of human beings. Human beings, because
they are by nature knowers, are by nature inclined toward religion and supernatural belief.
As noted at the beginning of this chapter, what it is to know, ultimately, is to understand why.
Knowledge involves a grasp of causes. While we may know that something acts in a particular way,
or how something works, we have a better understanding of that thing when we understand why.
Someone who recognizes that there is a human nature that we can know, and who begins from
a position of Christian faith, sees that human beings, because of their very nature, have an inclina-
tion toward understanding why something is as it is, and, even more, to understanding why all of
existence is as it is. It is natural to human beings to ask why there is something and not nothing;
why does anything exist at all? And, because human beings are capable of self-reflection and self-
knowledge, they have a natural inclination toward understanding themselves and how they fit into
the rest of existence.
The human desire for knowledge will only be fully satisfied in knowing the cause of all that is,
which is God. Short of directly gazing on God in the beatific vision, human beings remain always
on the way toward fulfillment, always never yet fully satisfied. The condition of human beings in
this life is one of searching for the fulfillment of their nature in understanding all that is by under-
standing the cause of all that is. (It should be noted that, while this is a position that is consonant
with Christian faith, it is possible to arrive at this conclusion through reason. See, for example,
Thomas Aquinas’s argument in the Summa theologiae, 1a2ae, q. 1.)
Because human beings are temporal, it is part of human nature to discover the truth about
things gradually. Human beings do not come to know everything all at once, in a single instant,
but rather their knowledge and understanding can continue to deepen over a lifetime.
For those human beings to whom the revelation of God is accessible and who choose to believe
it, their inclination to know has a guide and a clear destination. They have found what it is that will
fulfill their nature and will satisfy their desires, and so they spend their lives pursuing the deepening
of the understanding of the truth that they accept by faith.
This leaves the question, however, of how to account for religions that are not a response to the
revelation of God. Why, if there is a true religion, are there any number of other religions?
A possible response to this question is first to point out the historical and cultural fact that,
until the rise of secularism in the contemporary West, all peoples at all places and times manifested
some religious inclinations. This fact supports the claim that human beings have natures oriented
toward understanding all that is and that observations of the physical world on their own have
not satisfied this desire.
Given that human nature is aimed ultimately at understanding all that is, which comes through
union with God, when human beings do not have access to God through supernatural faith in rev-
elation, it makes sense that they would develop religious stories and practices as a way of attempting
to satisfy their natures.
This explanation imitates, in some ways, those of Tylor and Frazer, who see that religion offers the
explanations that human beings desire. It is consonant with the theories of Durkheim and others, who
see that religion serves human needs, such as the strengthening of the community. What it allows for,
however, is that it is possible that there is a religion that is not simply a human invention and that no
genuine religion is a purely human invention. If religions develop out of the human desire to know, out
of human nature, then all religions have some trace of God, who is the creator of human nature.
Because God has given human beings the desire to know Him, the human invention of religion is a
response to how God has created human beings. This is not to say that religions invented by human
beings are true but rather that the human invention of religions points to the God-given nature of
human beings to pursue him, which is fulfilled in the religion that is not humanmade but God-given.
It is possible, then, for all religions to possess some truth. If all religions other than some one
true religion were human inventions, then they would have to have been invented by truth seekers
who have the capacity, however impaired, to discover what is true through the exercise of their rea-
son. What is true in religions points to the human desire to know and to discover the truth. What is
not true in religions points to the inability, or great difficulty, of human reason, on its own, to dis-
cover all that is.
Someone who begins the study of religion from the perspective of faith, then, rather than with
a research bias that limits his view, is open to discovering what truth a religion may possess. Instead
of dismissing the beliefs and practices of religion as mere human invention, someone who
approaches the study of religion recognizing that a religion can be grounded in reality and that
the truth that religion possesses matters is far more likely to see religion as it actually is and to study
it without prejudice. He can do so because he does not limit his view of reality to mere appearances
and the physical but instead recognizes the possibility of knowing beyond the physical and of receiv-
ing knowledge, even from what is divine.

Bibliography
Calvin, John. Institutes of the Christian Religion. (1536.) Translated by Henry Marx, Karl. Selected Writings. Edited by Lawrence H. Simon. Indianapolis,
Beveridge. Grand Rapids, MI: Eerdmans, 1989. Available from: IN: Hackett, 1994.
Christian Classics Ethereal Library. Calvin College. http://www.ccel.
Pals, Daniel L. Nine Theories of Religion. 3rd ed. New York: Oxford
org/ccel/calvin/institutes.html.
Cicero, Marcus Tullius. On Divination. Translated by David Wardle.
Pieper, Josef. “Faith.” In Faith, Hope, Love, translated by Richard and Clara
Oxford: Clarendon; New York: Oxford University Press, 2006.
Winston. San Francisco: Ignatius Press, 1997.
Dawkins, Richard. The God Delusion. Boston: Houghton Mifflin, 2008.
Pieper, Josef. For the Love of Wisdom. Edited by Berthold Wald. Translated
Dougherty, Jude P. The Logic of Religion. Washington, DC: Catholic by Roger Wasserman. San Francisco: Ignatius Press, 2006.
University of America Press, 2003.
Pieper, Josef. The Platonic Myths. Translated by Dan Farrelly. South Bend,
Durkheim, Émile. The Elementary Forms of the Religious Life: A Study in IN: St. Augustine’s Press, 2011.
Religious Sociology. 2nd ed. Translated by Joseph Ward Swain. London:
Plato. The Republic. Translated by C. D. C. Reeve. Indianapolis, IN:
Allen and Unwin, 1976.
Hackett, 2004.
Eliade, Mircea. The Sacred and the Profane: The Nature of Religion.
Preus, J. Samuel. Explaining Religion: Criticism and Theory from Bodin to
Translated by Willard Trask. New York: Harcourt, 1987.
Freud. New Haven, CT: Yale University Press, 1987.
Evans-Pritchard, E. E. Witchcraft, Oracles, and Magic among the Azande.
Oxford: Clarendon, 1965. Ratzinger, Joseph. Introduction to Christianity. Translated by J. R. Foster.
San Francisco: Ignatius Press, 2004.
Franklin, Benjamin. Poor Richard’s Almanack. Waterloo, IA: USC Publishing
Company, 1914. Ratzinger, Joseph. “What Does It Mean to Believe?” In Christianity and the
Crisis of Cultures. Translated by Brian McNeil, 89–91. San Francisco:
Frazer, James George. The Golden Bough: A Study in Magic and Religion. Ignatius Press, 2006.
New York: Penguin, 1996.
Segal, Robert A. “Theories of Religion.” In The Routledge Companion to the
Freud, Sigmund. The Future of an Illusion. Edited and translated by James
Study of Religion, 2nd ed., edited by John Hinnells, 75–92. London:
Strachey. New York: Norton, 1989.
Routledge, 2010.
Funk, Cary, and Becka A. Alper. “Religion and Science.” Pew Research
Strausberg, Michael, and Steven Engler, eds. Oxford Handbook of the Study of
Center. 22 October 2015. http://www.pewinternet.org/2015/10/22
Religion. New York: Oxford University Press, 2016.
/science-and-religion/.
Strenski, Ivan. Understanding Theories of Religion: An Introduction. 2nd ed.
Gilson, Étienne. The Unity of Philosophical Experience. New York: Scribner,
Malden, MA: Wiley Blackwell, 2015.
1937.
Thomas Aquinas. Exposition on the Apostles’ Creed. Translated by Joseph B.
Gregory, Brad S. The Unintended Reformation: How a Religious Revolution
Secularized Society. Cambridge, MA: Belknap Press of Harvard Collins. Edited by Joseph Kenny. New York, 1939. Available from
University Press, 2012. https://dhspriory.org/thomas/creed.htm.
Herodotus. The Landmark Herodotus: The Histories. Edited by Robert B. Tylor, Edward B. Primitive Culture: Researches into the Development of
Strassler. Translated by Andrea L. Purvis. New York: Anchor Books, Mythology, Philosophy, Religion, Art, and Custom. New York: Cambridge
Hitchens, Christopher. God Is Not Great: How Religion Poisons Everything. Wellmuth, John James. The Nature and Origins of Scientism. Milwaukee,
New York: Twelve, 2007. WI: Marquette University Press, 1944.
Larsen, Timothy. The Slain God: Anthropologists and the Christian Faith. Williams, Richard N., and Daniel N. Robinson, eds. Scientism: The New
New York: Oxford University Press, 2014. Orthodoxy. New York: Bloomsbury Academic, 2015.
TOPIC 18
Theories of Religion: Atheism

RELIGION AS A NATURAL PHENOMENON 619
Head of the Society and Cognition Unit
THE RELEVANCE OF THE HUMAN SCIENCES TO THEISM 623

Evan Fales
Emeritus
University of Iowa
EVOLUTIONARY AND COGNITIVE THEORIES OF RELIGION 627

Todd Tremlin
Lecturer, Department of Philosophy and Religion
Central Michigan University, Mount Pleasant
MYTH AND MAGIC 632

Gregory Dawes
Professor, Philosophy and Religion
University of Otago, New Zealand
This chapter discusses whether there are satisfactory natural theories of religion. Furthermore, the authors
consider whether these natural theories favor atheism.
RELIGION AS A NATURAL PHENOMENON
Konrad Talmont-Kaminski, University of Bialystok
Prima facie, naturalist theories of religion present theism with a problem that hits particularly close
to heart. Being able to explain people’s belief in a god without the need to posit the existence of that
god would seem to give grounds to treat that belief with a great degree of skepticism. This is much
the same problem as has been faced by theism for hundreds of years, owing to science’s general ten-
dency to supplant previously used religious explanations, but it has the added tang that this problem
concerns religion itself.
A common strategy to thwart the significance of such explanations, used by Alvin Plantinga
(1991), is to claim that science has hobbled itself by needlessly insisting on methodological natural-
ism, unlike theists, who have access to the full range of explanations. According to this interpreta-
tion, naturalist scientific theories of religion are essentially incomplete. However, as Maarten Bou-
dry, Stefaan Blancke, and Johan Braeckman (2010) have shown, the naturalism of science is not
an assumption but a conclusion, resulting from hundreds of years of failure to find any need for
supernatural explanations. Science is supremely opportunistic and pragmatic about the kinds of
619
explanations it uses. It has just never found a use for theist

KEY CONCEPTS ones—as in Pierre-Simon Laplace’s apocryphal dictum that he
had no need for the theist hypothesis in his scientific system.
The phenomenon of religious belief and practice is just one
Assertives
more domain where this holds true—there is no need to
Batesian Mimicry assume the truth of any theological claims in order to explain
Big Bang religion as a social and psychological phenomenon.
Cognitive Science of Religion The first few years of the twenty-first century have seen the
appearance of complex and sophisticated scientific theories of
Debunking Arguments religion, with the major focus being on the evolutionarily based
Dialectical Materialism cognitive accounts under development by researchers working
Evolutionary Argument against within the cognitive science of religion. Most researchers have
Naturalism focused on issues of why people espouse religious beliefs and
engage in religious behavior, avoiding the philosophical impli-
Non-alethic Function of Belief cations of their work. Yet the question of what these kinds
R· igveda of scientific accounts of religion entail regarding the truth of
Sensus Divinitatis religion needs to be considered.
Social Darwinism EXPLAINING AND DEBUNKING
Theory of Mind Work on the significance of naturalist scientific explanations of
Transitivity Rule religion with respect to whether religious beliefs are justified has
Transmutation, Transmutationism focused on evolutionary debunking arguments. The basic form
of these arguments is to claim that particular beliefs are
explained by a mechanism that does not track the truth of such
beliefs and that therefore those beliefs are not justified (Kahane
2011). An example would be finding out that the phone number a person thought belonged to
Bono was generated randomly, so there is no reason to believe it works. In the case of religious
beliefs, debunking has involved showing that leading cognitive explanations of such beliefs make
no reference to their accuracy (Wilkins and Griffiths 2012). Thus, the cognitive science of religion
explains many aspects of religious belief as by-products of cognitive mechanisms whose function is
not connected to religion (Boyer 2001). These arguments are predated, however, by Plantinga’s
(1993) evolutionary argument against naturalism. Plantinga argued that cognitive abilities shaped
by evolutionary pressures to ensure adaptation cannot be relied on to provide true beliefs; thus, if
unguided evolution were true, there would be no reason to have faith in naturalism. The reason,
according to Plantinga, is that unguided evolution will lead to belief-forming mechanisms that are
fitness-tracking rather than truth-tracking, that is, that form beliefs which ensure survival rather
than accuracy. For many, Plantinga’s reductio argument is striking in its counterintuitiveness,
but it provides a useful counterpoint to understanding arguments that aim to debunk religious
claims. Is there a reason to be skeptical about the religious beliefs generated by our evolved mental
capacities while generally accepting common sense and scientific beliefs?
A basic problem of potential relevance to all debunking arguments is the objection that they are
all making the genetic fallacy. All debunking arguments base their critique of particular beliefs on a
critique of the means by which people have arrived at those beliefs. Yet it is a commonplace that
invalid arguments may lead to true conclusions and that profoundly faulty mechanisms may gener-
ate true beliefs by happenstance. Thus, the philosophically inclined TV series The Good Place
invents Doug Forcett, a stoner whose description of the afterlife given under the influence of a
particular mushroom trip happened to be “92 percent correct”—a means of achieving religious
insights that, while popular, is not generally accepted as truth-tracking. Another way of making
the point is via the traditional distinction between the context of discovery and the context of justi-
fication—the means by which a belief is originally arrived at need not be the same as the means by
which it is justified, especially when it faces unexpected criticism.
What, then, is the point of examining the mechanism that generates beliefs? Clearly, showing
that a belief was formed by a means that does not track the truth does not mean that we can be sure
that belief is false. Nor does it mean that we cannot be presented with further evidence that should
have a positive impact on our evaluation of that belief. However, in very many contexts, the fact
that someone (who is deemed to be an appropriate epistemic authority) holds a belief is treated as
prima facie evidence for that belief—prima facie evidence that is undermined if we are shown that
the reasons why that person holds that belief are not connected to its truth. This seems a remarkably
weak form of criticism and not necessarily relevant to theological claims. Naturalist explanations for
religious beliefs mostly concern what is known as folk religion—the largely unreflective and often
theologically incorrect (Slone 2004) religious beliefs of typical believers—rather than the arguments
presented by their theologically and philosophically trained coreligionists. Yet proper justification for
religious belief is more likely to be sought among the latter. This seems a prime example of the kind
of case where while the basic means by which particular beliefs are arrived at appears not to be con-
nected to their truth, there is further evidence—another means for justifying them—that renders
irrelevant all considerations of the problems with the original means of arriving at belief.
However, the situation for the theist is nowhere near as unproblematic as it might seem. First,
even if philosophically sophisticated arguments can be presented for theist claims, they do not jus-
tify the believers who do not appreciate those arguments—just as getting the lotto numbers right
does not show one knew them beforehand. Second, for the point to hold in reference to the philo-
sophically trained, their theological positions would have to be held independently from folk reli-
gious beliefs. Were this the case, however, the conclusions reached by theologians and philosophers
of religion would not covary with the religious beliefs commonly held by the general populace. Yet
that is obviously not the case. In countries with large Catholic populations it is common for theo-
logians and philosophers to find that it is Catholic religious beliefs that their intellectual reflection
leads them to. The same is the case for Protestant, Muslim, and Jewish communities and scholars.
This suggests that intellectual reflection, which should provide independent evidence for religious
claims, is not really independent of folk belief and should therefore be treated with a similar level
of skepticism—a point that does not settle the question of the soundness of the arguments for those
claims but casts doubt on the capacity of theist thinkers to determine it, in a way analogous to how
Platinga intended to undermine naturalism. From the point of view of the cognitive science of reli-
gion this is far from surprising. All people, including philosophers, share the basic cognitive traits
that render particular kinds of supernatural beliefs attractive. As previously noted, even if we can dis-
count folk belief and philosophical argumentation as suspect evidence, we cannot conclude that reli-
gious claims are necessarily false. Yet concluding that they are so unlikely as to be best ignored
seems just as justified in their case as in the case of the randomly generated number for Bono or
in the case of the stoner’s mushroom-fueled imaginings.
Of course, it might be maintained that an analogous argument could be raised for atheist posi-
tions; after all, philosophical argumentation that leads to atheist conclusions is most common in
atheist communities. To a significant degree this is correct. Atheist thinkers are subject to the same
biases as everyone; it would be hubris to talk about cognitive biases, accept general fallibilism, and
yet think oneself immune. There are several reasons, however, to argue that the situation is not
quite symmetrical—seeing why also helps to see the problems with Plantinga’s debunking of natu-
ralism. As Robert McCauley (2011) has observed, religious beliefs are natural in that they arise
unprompted and become popular in most places and cultures. No special institutional means are
necessary for religious beliefs to be reliably spread in a culture; indeed, it is the naturalness of the
beliefs that acts as a foundation for religious institutions. In the case of science, the very opposite
is the case. For scientific beliefs to be maintained—and not be reduced to their pale pseudoscientific
shadows—it is necessary that a society maintain specialized scientific institutions. Unlike religious
beliefs, the naturalist scientific worldview is far from natural to humans. Thus, the naturalism of sci-
ence is not caused by the acceptance of atheist views in the general public. Indeed, if we are to dis-
trust our naturally evolved reasoning abilities as Plantinga would hold, it is the religious beliefs that
should be the more suspect, given that they do come more naturally to us. The argument can be
pushed further.
As already pointed out, debunking arguments rely on the claim that the belief-forming mecha-
nism is not truth-tracking. John Wilkins and Paul Griffiths argue, however, that Plantinga’s

dichotomy between truth-tracking and fitness-tracking is a false one—belief-forming mechanisms in

general increase fitness by ensuring that we have an accurate idea of our environment. Thanks to the
accuracy of our beliefs, we can act appropriately to avoid threats and seek out resources. Indeed,
as Wilkins and Griffiths note, it really makes no sense to ask whether a trait is an adaptation for
fitness—adaptations by definition are traits that evolved because they increase fitness. Yet truth-
tracking is not the only way that belief-forming systems can increase our fitness—a distinction that
might seem to give solace to Plantinga but, in fact, is his bugbear. Crucially, insofar as religious
claims are deemed by scientific accounts to have a function, it is that of maintaining cooperation
(D. Wilson 2002); however, their functionality is not reliant on their truth but merely on whether
they are believed in. Having such an nonalethic function requires that such beliefs engage in a form
of Batesian mimicry; to maintain stability the beliefs must appear to be justified even while they
avoid potentially destabilizing counterevidence (Talmont-Kaminski 2013a, 2013b). In short, to
ensure functionality, it is not enough that the mechanisms that generate religious beliefs not track
the truth; they must actively ensure that truth does not play a role in determining which religious
beliefs are held. A prime example of this is the way in which religious claims are held to be sacred
and thereby not to be examined or criticized. This puts into sharp relief the differences between
the epistemic effect of scientific and religious institutions. Scientific institutions serve to ensure that
beliefs come to be accepted on the basis of evidence and maintained despite how radically counter-
intuitive they are for most humans. Religious institutions ensure that beliefs avoid potential counter-
evidence, thereby helping to stabilize those beliefs and the institutions themselves. In both cases, the
very different epistemic stances are the result of the relationship between the accuracy and the func-
tionality of the beliefs in question. The difference, however, quite ironically ensures that while sci-
entific naturalism is shielded against Plantinga’s debunking by the general reliability of our reason-
ing and the additional mechanisms developed by scientific institutions to further ensure the
accuracy of scientific beliefs, it is the religious beliefs that are, by design, exposed to the very biases
that Plantinga felt naturalism threatened.
NATURAL AND NONNATURAL EXPLANATIONS

What is the relationship between the kinds of natural explanations for religious belief that have been
presented thus far and nonnatural explanations, such as miraculous divine revelation? In the sim-
plest of analyses, they are competing accounts, and David Hume’s (1748) critique of belief in mira-
cles straightforwardly applies to the nonnaturalist explanations. Further to that critique, however, it
should be considered that claims of personal revelation are the exception rather than the norm
among theists. Most believers come to accept their religious beliefs for unquestionably natural rea-
sons of exactly the kind that the cognitive science of religion seeks to cast light on. It could be
argued that by these natural processes believers are passing on divinely revealed truths, but, as has
been pointed out, religious institutions specifically serve to ensure that true revelation should have
no particular purchase compared with the myriad false ones. Claims of revelation may rise or fall
but not for reasons of accuracy.
Thinking of nonnatural explanations of belief as simply false, however, is hardly the best use we
can make of them. The seeming dichotomy between natural and nonnatural explanations of belief
invites the idea that they should be understood as functioning on the same level. While we know
a nonnatural explanation when we see one, the distinction is a highly problematic one. Science is
often presented as making an a priori methodological or metaphysical resolution to accept only nat-
ural explanations, but natural explanations are typically defined as those that are accepted by sci-
ence—thereby creating a very tight and uninformative circle of definitions. At the same time, the
phlogiston theory of combustion is not accepted by science, but it could hardly be deemed to be
supernatural or even nonnatural. It is just false.
As shown by Boudry, Blancke, and Braeckman (2010), natural explanations as employed by
science are not the result of an a priori dictum against any particular kinds of claims but, simply,
those explanations that have been found to be adequate by our very best epistemic practices—as
institutionalized within science. Talk of supernatural claims as lying beyond the ken of science
may make it easier to maintain civil discourse around the table or funding in societies where
religions still have influence, but it is no more correct to claim that science remains mute about the
existence of God than it is to claim that it has nothing to say about the existence of phlogiston. The
nonnatural explanations that have any acceptance owe their popularity to particular cognitive by-
products, shielded from factuality by religious institutions. Their content thus reflects aspects of
the structure of our cognitive system. This means that both natural and nonnatural explanations,
including explanations of religious belief, are informative, but they are informative in very different
ways. In particular, nonnatural explanations of religious belief tell us about the nature of the cogni-
tive by-products that our cognitive system produces. In effect, they do help us to explain religious
belief but not in the way that theists might intend. To put it another way, nonnatural explanations
are not really competing with natural explanations but rather are among the phenomena to be
explained naturally.
This reveals a particular strength of the modern scientific approach to religion. It shows not
only that believers are not justified in their beliefs but also, very importantly, why it is that they
maintain those beliefs nonetheless.
DIVINE GUIDANCE
A further line of argument is open to the theist, which is to claim that all of the features of our cog-
nitive system that generate religious beliefs have arisen by divine guidance so that what seems a nat-
urally occurring feature of our mind is actually a sensus divinitatis—a sense that provides us with
knowledge of God, first proposed by John Calvin. Plantinga (2000) argues for the existence of a sen-
sus divinitatis, and Justin Barrett (2004) thinks that his research into the cognitive basis of religious
belief has revealed it.
The general idea that this or any other feature of the natural world arose owing to divine guid-
ance is as impossible to completely disprove as it is lacking in empirical evidence. We have been
studying the evolutionary roots of the mechanisms that generate religious belief, and their history
is as lacking in signs of guidance as evolutionary history in general. Once again, a nonnatural expla-
nation is otiose. What of the information about God with which the sensus divinitatis is supposedly
providing us? As already noted, science is both opportunistic and pragmatic. Were there such a
sense providing us with such data, science would have gleefully explored the implications.
What of the detailed claims that Plantinga makes, which do have specific empirical implica-
tions? In the case of Plantinga, the claim is that sin has rendered the sensus divinitatis ineffective
in some people—in simplistic terms, atheists do not feel God’s presence because they are bad (for
a related critique of Plantinga, see Van Eyghen 2016). Were this the case, one would expect that
there would be a correlation between secularization and antisocial behavior. However, the very
opposite is the case—the most secularized societies are also the societies with the lowest levels of
crime (Paul 2009). Indeed, patterns of secularization fit the claim that religion’s prosocial functions
have been usurped by secular institutions (Talmont-Kaminski 2013b). The same problem recurs on
the individual level, with causes for the various types of atheism having nothing to do with criminal-
ity (Norenzayan and Gervais 2013). Given what has been set out regarding the different ways by
which religious as opposed to scientific claims come to be stabilized within our culture, this failure
for the two to fit together should not be at all surprising.
THE RELEVANCE OF THE HUMAN SCIENCES TO THEISM
Evan Fales, University of Iowa
METHODOLOGICAL CONSIDERATIONS
The work of the Scottish philosopher David Hume on the historical and psychological sources of
theistic belief—The Natural History of Religion ([1757] 2009) and Dialogues concerning Natural Reli-
gion ([1779] 1947)—is a watershed moment in the scientific study of religion. I shall not enter into
the details of most of the eighteenth- and nineteenth-century thinkers who followed in his footsteps
in offering naturalistic explanations for religious practice and belief. It is nonetheless worth reflecting

briefly on Hume’s argumentative strategy, because it applies to understanding the implications of

the work of those who have followed Hume for assessing the truth claims of religious doctrines.
In The Natural History of Religion and Dialogues concerning Natural Religion, Hume offers alter-
native explanations for a range of facts (apparent design in the first case and religious worship in the
second). The theist attributes orderly patterns in nature to a designing God and explains human
religiosity by appeal to divine/human interactions of various kinds. Hume’s proposed explanations
are, of course, nontheistic. His argumentative strategy is to confront the theist with multiple alter-
native explanations for certain data that the theist thinks can be explained—or can be best
explained—only by invoking a divine creator. To make his case, Hume does not have to show that
one of his alternative explanations is the best explanation or better than the theistic one; all he needs
to establish is that it is more likely that some one or another of his naturalistic alternatives is true.
Suppose we want to assign betting odds to horses in a race. We need to consider all the horses
that are eligible to race, past track records, odds of mishaps, and so on. The odds that a horse will
win depend on how many competitors it has and how strong they are. In general, the more horses
and the stronger they are, the lower the odds will be for any given horse. Even weak contenders
might win—for example, if the others stumble. Think of competing hypotheses/explanations as
horses crowding each other on the track: the more, the meaner the race.
An important point is that, where a hypothesis is introduced to explain some range of data, its
credentials can be significantly enhanced by its ability independently to predict/explain other empir-
ical data. Failure to generate new lines of investigation and independently testable predictions is a
mark of sterility.
Early Naturalistic Theories of Religion. The articulation of early naturalistic theories of religion
coincided, more or less, with the emergence of the disciplines of Darwinian biology, sociology,
anthropology, and psychology in the nineteenth and early twentieth centuries. It was recognized
that, for all its diversity of forms, religion or something like religion existed in every known society.
Therefore, the hunt was on to find some universal feature(s) of human nature that could account
for this fact. The notion that a single God was in some way accountable for or lay behind all the
diversity seemed just unlikely. Under debate was, inter alia, not just which aspects of human nature
accounted for the religious impulse but also how those traits arose and how the development of full-
blown religious practices emerged.
A number of nineteenth-century theorists searched for the wellsprings of religion in facets of
the human emotional constitution—for example, fear of death and other threats, feelings of solidar-
ity with fellow tribe members, the experience of dreams, and so forth. Others focused on material
conditions—most prominently the German nineteenth-century philosopher Karl Marx, who saw
religion as a tool of the ruling classes to motivate cooperation with and acceptance of their govern-
ing power and the institutions that preserved that power. Still others, baffled by the apparently out-
landish religious beliefs and practices of the various newfound tribal cultures, tried to explain reli-
gion as the product of various cognitive errors—mistaken reasoning from experience to the
existence of a supernatural world. (Perhaps the most radical position—that of Lucien Lévy-Bruhl
early in his career—was that the thought of tribal peoples was intellectually primitive, easily prone
to errors in or absence of logic. Such a view may have seemed plausible, given the prevalent Social
Darwinist view that societies whose material culture is preagricultural represented not only a stage
closest to the evolutionarily earliest stage of human cultural development but also equivalently prim-
itive intellectual endowments. Lévy-Bruhl’s version, outlined in Primitive Mentality ([1922]
1923)—which he later retracted—may be considered a precursor of more recent relativisms, such
as those of Peter Winch ([1958] 1970).
Here let me contrast an important methodological divide between two major turn-of-the-
century figures, Austrian neurologist and psychoanalyst Sigmund Freud and French sociologist
Émile Durkheim. Freud ([1913] 1950; [1939] 1955; [1927] 1989) naturally sought the sources
of religion in the emotional constitution of human beings. Emphasizing the role religion can play
in coping with human insecurities, Freud entered the fray over the historical emergence of religion
with a fable about the transition of our hominid ancestors to humanity, possessed of distinctive
traits of conscience, guilt, and prohibition of incest. This was a fable about the origin of the
Oedipus complex and father hatred transformed into a sublimated form of father worship: the first
religion—or “fable,” because Freud must have known that his ([1913] 1950) offers a just-so story
that cannot possibly be true. (For more information, see Fales [2010, 118, n. 22].)
Durkheim, who fathered the discipline of sociology and argued vigorously for its independence
from other fields, both methodologically and ontologically (rules, forms of religious life), was per-
haps the first major thinker to suggest that the supernatural beings that populate the spiritual world
are psychologically manageable “representations” of complex social realities; they emerged from the
efforts of early humans to explain the origins and give substance to the felt authority of their social
arrangements (see Durkheim [1912] 1965, 245−272). Durkheim, again with a Social Darwinist
focus on hunter-gatherer societies as our best indication of the aboriginal form of religion, recog-
nized that the denizens of their spirit worlds were representations of social forces and institutions;
for example, clan totems were symbolic personifications of the clans themselves, understood in per-
sonal terms. (He thought these identifications were unconsciously made, but that supposition is not
necessary to his position and is not particularly well supported.)
Developments: Anthropology. The debate over the cognitive dimensions of religious belief has
continued but with an increasing recognition (1) that religious beliefs are taken to be true (or false),
that is, to have factual content; (2) except that doctrinal declarations sometimes have a kind of per-
formative force—in that their promulgation makes true what they declare; (3) that those who
espouse them are rational—indeed, rationally on a par with human beings in modern societies;
and (4) that their theological reflections bear a structural resemblance to modern (including scien-
tific) theorizing.
Let me comment on each of these:
(1) Factual Content. During the last century, a widespread view among analytic philosophers
was that religious statements actually have no factual content at all; instead, they serve other com-
municative purposes—for example, expressing moral attitudes or other emotional states. This non-
cognitivism was probably encouraged by a strong thread of theological apophaticism and a convic-
tion that matters of faith transcend empirical investigation; it influenced some anthropological
theories that sought to explain religion, or central doxastic features of it, as a noncognitive enterprise
(Leach 1954; Rappaport 1968, 1999).
Such a position dovetails naturally with functionalist explanations of religion, though it neither
entails nor is entailed by functionalism. A functionalist may allow, for example, that religious doc-
trines have cognitive content but maintain that, like social arrangements generally, they arise in
the service of practices and institutions that are themselves the product of adaptive processes, often
understood in terms modeled on Darwinian natural selection. A central tenet of functionalism,
then, is that social structures, including religious institutions, emerge not for the most part as a
result of consciously planned measures aimed at an intended result but instead as unintended con-
sequences of otherwise directed actions.
As a result, social actors typically do not correctly understand the purposes or functions that
their social arrangements serve and commonly try to close the gap in knowledge by constructing
quite fanciful rationales, or rationalizations, of their practices. Thus, while believers will be certain
that their worship practices are maintained as services properly owed to the gods, a functionalist will
dismiss that story in favor of an explanation, for example, in terms of the adaptive value to a society
of rituals and institutions that bind a people together into a cohesive, resilient social order with sta-
ble lines of authority and a strong sense of group identity. Views of this sort were developed
during the twentieth century, in good measure reflecting the influence of Durkheim, by such figures
as Alfred Radcliffe-Brown ([1922] 1964), Bronislaw Malinowski ([1922] 1961), Edward Evans-
Pritchard ([1937] 1976), and many others.
In this camp one might also place Marx and neo-Marxist anthropologists such as Marvin Harris
(1974). In Marx’s case, however, social ideologies, including prominently religious belief systems,
were specifically devised by members of a ruling class (in stratified societies), with the intended pur-
pose of justifying and legitimating their own dominant position and protecting the means of

production of wealth that, in fact, was the source of their power. Nevertheless, Marx’s dialectical
materialism entrenched functionalist commitments: those in power are themselves driven by social
and material forces that they neither control nor understand, often to policies that ensure their even-
tual destruction. A given religion—the “opium of the people,” as Marx called it (1843)—can main-
tain its authority only so long as it is attuned to the economic processes that maintain the power of
the ruling class, and those remain functional.
(2) Performative Force. One advance over the false choice between factual versus nonfactual
utterances came with the recognition that even declarative sentences can be used, not to state a fact,
but to bring facts—at least social facts—into existence. Such pronouncements are called performatives
(see Austin and Urmson 1962). Familiar examples include such expressions as “I promise” (which cre-
ates the promise) and “I pronounce you husband and wife” (which marries the couple). Legal and rit-
ual contexts offer wide scope for this kind of language-use: witness the law-creating force of the US
Constitution and the legal force of formal decisions of the Supreme Court of the United States.
Recognition of this role of language opens up the prospect that certain “religious” utterances are
not straightforwardly factual claims but rather serve to institute congregational social arrangements
(compare the finality of Supreme Court rulings with the doctrine of papal infallibility).
(3) Rationality in Religious Belief. This suggests another major anthropological approach to
understanding religious thinking, namely intellectualism—the view that underlying religious
thought is the fundamental human urge to comprehend and order the world all the way down to
its roots. Some early attempts might strike us as amateurish or highly fanciful, but anthropologists
have suggested that this lack of sophistication reflects either undisciplined attempts at theoretical
inference that lacked the benefit of long and careful reflection on proper scientific method or that
represented serious intellectual effort on the part of people handicapped by lack of technology, time
for pure research, or general background knowledge.
Among earlier figures, British anthropologist James Frazer’s theory of magical thinking and its
evolution into religious belief in The Golden Bough (1890) represents, together with, for example,
the work of Lévy-Bruhl, the intellectual blunders form of the view, which did not survive scrutiny.
With growing recognition of the epistemic and inferential sophistication of so-called primitive peo-
ples, a variety of more sophisticated forms of theorizing have been proposed as underlying religious
thought. While some progress along these lines was made by Evans-Pritchard, major contributions
were made by French anthropologist Claude Lévi-Strauss (who first understood the capacity for
abstract thought that lay behind Australian aborigine kinship systems and who proposed that myth-
making was a highly conceptual enterprise aimed at ironing out what natives allegedly saw as contra-
dictions in the forces that ruled their lives).
(4) Sacred Texts as Quasi-scientific Reasoning. Structural analysis of myths (including biblical
texts) has gone a considerable distance beyond Lévi-Strauss’s techniques, with important work by,
inter alia, Edmund Leach (1969), Terry Turner (1969), and Mary Douglas (1966). Finally, Robin
Horton (1993) has argued that the mythical thinking of tribal peoples in many ways mirrors sophis-
ticated scientific explanation.
It will naturally be impossible for me to assess here these varying approaches. Instead, I want to
make a final point. Some theists, sufficiently impressed by the progress made in the human sciences
(including psychology and now neurology), have proposed a kind of détente: “We theists,” they will
say, “accept your findings, the power and fruitfulness of your explanations and research programs.
But you are forgetting God. There is no real incompatibility between everything you’ve discovered
and our claim that this is God’s world which he made and governs.” In short, the theist proposes to
add God to do additional explanatory work not accomplished by scientific means.
This has repeatedly proved to be a misguided strategy. It is a species of “God of the gaps”
theology, with a twist: theists now hunt for or postulate extra gaps where divine providence can
play a role. This is the sort of strategy deployed by guided-evolution proponents. They propose
inscrutable mutation events that bespeak design, when the Darwinian has no need of them. That
means more ontology with no improvement in explanatory or predictive power. Things are often
worse than that.
Here are two examples. William Alston (1964) proposed that theists can take Freudian psychol-
ogy onboard, by suggesting roughly that the creation of the Oedipus complex was by design—as
God’s way of bringing human beings into a condition of responsiveness to the divine. Surely God
could have and obviously should have found a better strategy than that; Freudian psychology and
theism are not logically contradictory, but they are probabilistically at odds.
A similar maneuver has attracted theists to the suggestion that, although anthropology and neu-
roscience can jointly go a long way toward explaining mystical experiences, they are incomplete,
because they omit consideration of the fact that God is the cause of the brain processes that underlie
the experiences. There are many reasons why, so far from enhancing the explanatory power of the
scientific explanations, this suggestion wreaks havoc with them. First, theists will have a devil of a
time explaining why the same basic brain engineering delivers so many mystical insights that are
wildly alien to and in conflict with the deliverances of the home religion’s mystical experiences. Sec-
ond, it will be hard to explain why many of these processes are brain malfunctions and quite regu-
larly associated with mental illness. Adding the God hypothesis to the scientific picture of religious
experience does nothing to improve and a good deal to create problems for our understanding of
these phenomena.
EVOLUTIONARY AND COGNITIVE THEORIES OF RELIGION
Todd Tremlin, Central Michigan University
Charles Darwin’s five-year round-the-world journey aboard the HMS Beagle sowed the seed for the
idea of evolution, but it was the decades of down-to-earth research back home at Down House that
brought his revolutionary theory to bloom. Darwin investigated the mechanics of domestication and
selective breeding. He observed the flight of bumblebees and the construction of honeycomb. He
tested whether fish would eat pods and snails cling to ducks’ feet. He explored the structure of orch-
ids, the behavior of carnivorous plants, and the varieties of beetles and barnacles. He cataloged the
genealogies of specimens sent from afar and witnessed the battle for survival taking place in his own
backyard (Costa 2017). These “fools’ experiments,” as Darwin called them, led the would-be clergy-
man qua naturalist, stepwise, from creationism to transmutationism, finally articulated in On the
Origin of Species (1859) as descent with modification by natural selection (Lankester 1896, 4391).
Over that same period a transmutation took place in Darwin as well. Once an ardent fan of
William Paley’s natural theology, Darwin’s homespun experiments moved him further and further
from belief in a divinely designed world to one shaped by “laws acting around us” (C. Darwin
1859, 489). That honeybees build perfect hexagonal cells is not evidence of supernatural inspiration,
nor is the human eye irreducibly complex. To the contrary, Darwin repeatedly voiced, sometimes with
notable melancholy, his sense that living things are not the handiwork of a purposeful and benevolent
creator but the outcome of an indifferent and uncaring nature.
Darwin’s recognition that the biological world requires no divine guidance parallels French sci-
entist Pierre-Simon Laplace’s précis of the physical world half a century earlier. It is said that upon
inspection of Laplace’s work on the solar system, Mécanique Céleste (Celestial Mechanics, 1798–
1825), Napoléon Bonaparte complained that it contained no mention of God. Laplace famously
quipped: “I have no need of that hypothesis.” Similar transmutations are common among contem-
porary dignitaries of science. Richard Dawkins has written that Darwin “made it possible to be an
intellectually fulfilled atheist” (1986, 6). Steven Weinberg affirms that science makes it possible
not to believe in God. Edward O. Wilson believes the empirical evidence is massive enough to
undermine any search for meaning: “We are, it seems, completely alone” (2014, 173). Polling data
show that the adverse effect of science on religious faith is hardly eccentric: atheism among scientists
runs at ten times that of the general public (Masci 2009). Generally speaking, scientific knowledge
is dangerous to spiritual well-being.
What does this mean for the discipline of religion studies, which has taken a decidedly scientific
turn? Today’s two most compelling approaches to explaining religious thinking and behavior draw

from evolutionary theory and cognitive science. In the new millennium, it is intellectually recalci-
trant to discuss religion without direct reference to either adaptation or mental function. In keeping
with the theme of this chapter, it is worth asking whether these naturalistic theories of religion favor
theism or atheism.
NATURALISTIC APPROACHES TO EXPLAINING RELIGION

Though they share common elements, evolutionary and cognitive theories of religion emphasize dif-
ferent mechanisms for the origin and persistence of religious thinking and behavior. Evolutionary
theories usually explain religion as a human adaptation. The cognitive science of religion typically
envisions religion as a by-product of mental tools evolved for different though functionally related
purposes.
Evolutionary Theories of Religion. Religion is being reexamined through the lens of evolution-
ary theory with the aim of explaining why it arose as an evolved trait in the course of human devel-
opment. In a Darwinian world, religious behavior should be subject to selection if it contributes to
survival and reproduction. Religion’s ubiquity, antiquity, species typicality, and nonrandom com-
plexity are persuasive indicators of a genuine adaptation. George C. Williams, however, insisted that
true adaptations pass a more rigorous test: design for a particular function (1966). A growing num-
ber of scholars believe that religion rises to this challenge. Broadly speaking, the adaptive functions
of religion are sought at two different levels: benefits to individuals competing with other individuals
(individual selection) and benefits for groups competing with other groups (group selection).
Possible benefits that religion confers to individuals include problem solving, physical fitness,
and successful reproductive strategies. Wulf Schiefenhövel proposes that religious beliefs and rituals
helped humans’ ancestors make sense of a bewildering, often threatening world (2009). Extrapola-
tions from contemporary data correlating religious activities with personal health and longevity sug-
gest that it produced the same effects in ancestral environments (Koenig 2008). Religion might also
play a role in mate choice. Nature is full of uncanny behaviors and physical features—consider the
peacock’s tail—that arose through sexual selection. Geoffrey Miller argues that religion likewise pro-
vides a medium for displaying talents or character traits desired by potential partners (2000).
The centerpiece of the group-selection argument is the claim that groups with religion are more
unified, stable, and cooperative than groups without. In the course of human evolution, groups
bonded by shared ideology and practice outcompeted less unified groups and left more descendants,
resulting in the emergence of religiosity as an adaptive trait. David Sloan Wilson’s (2002) historical
examples of successful religious communities have been supplemented with studies of the durability
of religious versus secular kibbutzim (Sosis and Ruffle 2003).
Other scholars focus on the question of how religion strengthens groups. Most common is the
use of signaling theory. Costly, time-consuming activities belie evolutionary efficiency, especially
when there is no immediate payoff. However, if such activities signal commitment to others, then
individuals can reap the benefits conferred by group membership. Acceptance of shared beliefs
and participation in communal practices demonstrate an individual’s loyalty, cooperation, and trust-
worthiness (Sosis and Alcorta 2003). Religion therefore began as a vehicle for group bonding and
alliance building, functions crucial to survival in both natural and social environments. Signal theory
also makes sense of initiation rites, rites of passage, and other extravagant, sometimes painful rituals.
These activities not only signal group commitment but also motivate members to effectively work
toward public goals and engage in collective actions, including, when necessary, armed conflict
(Johnson 2008).
Dominic Johnson (2016) and Ara Norenzayan (2013) offer a noteworthy twist on the unity-
promoting role played by religious belief: that assent to the presence of interested, watchful, reactive
gods and spirits helped combat the problems of free-riding, antisocial behavior, and group defection.
Fascinating empirical evidence supports this contention. When people suspect that they are being
watched, they are more honest, generous, and cooperative and more scrupulously conform to group
norms. This is a natural outcome of hypersensitivity to social scrutiny in a milieu of intense gregari-
ousness. Once the conceptualization of gods and other supernatural agents emerged, it, too, served
this function. Jesse Bering sharpens the focus on divine vigilance yet further, suggesting that belief
in gods who actively punish and reward human intentions and behaviors helped humans’ ancestors
address the problem of gossip (2011). Belief in an observant, moralizing, punitive god curbed
immoral acts that might be witnessed and reported by others, thereby ruining reputations and
reproductive interests.
The Cognitive Science of Religion. Employing an interdisciplinary set of tools borrowed from
cognitive science, evolutionary psychology, and religion studies, the cognitive science of religion
explores how innate mental mechanisms originate, transmit, and shape religious thinking and
behavior (Barrett 2004; Boyer 2001; Tremlin 2006). The basic principles of the cognitive science
of religion can be succinctly stated.
1. The human brain is the product of evolution. Natural selection has designed neural circuitry just as
it has designed eyeballs, stomachs, and hands.
2. The human brain is an adapted organ. The brain has its specific architecture and set of mental
processes because these advantaged survival and reproduction.
3. The human brain is hardwired with innate intelligences that guide perception, emotion, thought, and
action. The mind is not a blank slate but possesses at birth the capacities to interface, often
automatically and not consciously, with the natural and social worlds it inhabits.
4. No special mental structures or processes are needed to explain religion. Religious thinking arises
from garden-variety mental mechanisms and is tightly constrained by the functional design of
the brain.
As a complex phenomenon, religion derives from mental, emotional, and behavioral propensi-
ties wired throughout the brain system. It turns out, however, that thinking about supernatural
agents—the conceptual cornerstone of religion everywhere—pivots on just a handful of mental tools
connected with social intelligence. Two mental mechanisms, in particular, accomplish most of the
heavy lifting: the agency detection device (ADD), which identifies agents in the environment, and
theory of mind (ToM), which attributes mental states to agents. These two capacities, which are
present at birth and mature in the first years of life, account for some of the most critical operations
of human intelligence, from the attention needed to survive in a predatory world to the cunning
involved in social exchange and to the reflective capacities involved in imagination.
ADD and ToM are undoubtedly evolved mental structures. The ability to quickly discern and
understand agents in the environment is crucial to survival. A person who fails to recognize a spot-
ted pattern as a leopard is unlikely to fare well for long. A person who cannot interpret the minds of
others is not fit for social life. Working in concert, these two brain processes promote mentalistic
perception of the world. Significantly, the operation of ADD and ToM is not only profound but
also promiscuous. The brain is prone to attribute agency and intent to ambiguous objects, signs,
and events. Humans see faces in clouds, sense presences in the night, and glance at uncertain
sounds. Humans are not unlike Darwin’s dog spooked by a parasol moving in the wind (C. Darwin
1871). A great deal of human reasoning involves the mental representation of agents who are both
present and absent, real and unreal. This incorrigible predisposition is the result of a risk asymmetry
in humans’ ancestral environment. In a dangerous, often inchoate world, oversensitivity to agents is
the best survival strategy (Guthrie 1993).
With this set of mental mechanisms alone humans have the cognitive base upon which religion
is founded and a set of adapted psychological biases that religious thinking and behavior continue to
exploit. The hyperactive nature of ADD and ToM leads us to some rather peculiar though perfectly
natural ways of thinking. People make causal connections where none exist, see signs and design in
natural events, intuit unseen forces and intended purposes behind experiences, and conceive the
existence of supernatural agents, be they the spirits encountered by shamans, ancestors lingering at
the edge of a hunter-gatherer village, or Christianity’s omniscient creator.
Similar cognitive mechanisms shape humans’ thinking about supernatural agents. The brain
automatically classifies things according to natural categories: person, animal, plant, and object (Boyer

1994). Gods and spirits, while they include important counterintuitive properties, are at bottom
person concepts. Person is the only category that the mind possesses for comprehending the type
of entities that gods are said to be. Because the conceptual blueprint for building a god is based
on the mental template for a person, cognitive inference systems activate a set of natural expecta-
tions about what a person is like. Gods are said to think, feel, care, and act. Consequently, people
represent supernatural agents in relational terms and interact with them in personal ways.
Viewed in this light, religious behaviors and ritual practices are no more mysterious than the act
of shaking hands with a friend or buying a wedding gift. The machinery of social cognition that
evolved to help humans negotiate competitive/cooperative interpersonal relationships also guides,
in by-product fashion, their relations with supernatural beings. Humans take oaths, make vows,
and enter into covenants. They offer sacrifices and give tithes. They praise, entreat, and confide
in. They perform acts of dedication, acts of confession, and acts of contrition. At every turn, reli-
gious practice pirates the social adaptations humans use to interact with one another.
As these examples show, the cognitive approach offers an incisive account of the natural foun-
dations of religion. Religion is dependent on cognitive and behavioral scaffolding erected by natural
selection to help humans’ ancestors meet recurrent challenges in the human niche. Chiefly, the
social psychology that orchestrates thinking and behavior in the real world also guides thinking
about imagined ones. Owing to a long history of adaptations designed for interaction with others,
humans possess a powerful set of cognitive endowments that make their minds susceptible to pro-
ducing supernatural concepts and, as a consequence, religion itself.
From a cognitive perspective, then, religion is not an adaptive trait per se but an accidental by-
product of mental adaptations. Religion originated for the same reason that it is recurrent: it capita-
lizes on evolved tendencies basic to everyday human life. Religion can be seen as an enticing error,
the outcome of a brain prone to make perceptual mistakes. Evolution has molded the human mind
with powerful cognitive biases, which, when employed in the natural environment, provide an effec-
tive means of survival. However, these same tightly strung biases also produce false perceptions and
cognitive blunders that are the wellspring of religious thinking.
THEISM OR ATHEISM?
While both evolutionary and cognitive explanations of religion are explicitly naturalistic, it is worth
asking whether they necessarily entail atheism or whether they admit of theistic interpretations.
Methodological naturalism is not the same as metaphysical naturalism. It is evident, as Jerry Coyne
points out, that practicing science erodes religious belief (2015), but it takes a further philosophical
step to assert that empirical facts about how the world works amount, in Darwin’s memorable
phrase, to “confessing a murder” (F. Darwin 1887, 23).
For the atheist, the immediate response is to suggest that the stunning success of naturalism in
other fields is now unraveling the mysteries of religion—a phenomenon long treated as sacred, even
sui generis. From physics to biology to psychology, no other method has proved nearly as effective
at discovering reliable facts about reality. The God hypothesis has simply held no utility. Natural-
ism, as Evan Fales writes, is an ally of atheism (2007). Why should not the same be true for the
study of religion, save for the detail that it treads on hallowed ground?
Theism does not dispute the accomplishments of scientific naturalism, but it does consider it
an unnecessary restriction on the overall search for truth, because scientific naturalism assumes from
the start that there is no supernatural realm or causation. No one can know in advance that only
natural causes are operative in the world. Theism also highlights the oft-used distinction between
the how (naturalism) and the why (theism). There are both necessary and sufficient causes to be
explained. Finally, naturalism has yet to offer complete answers to some large questions, such as
the generation of matter-energy, the arising of life, and the emergence of consciousness—the “three
big bangs,” as Holmes Rolston III calls them (2010). The same can be argued of current theories of
religion. Identifying the mechanisms of religious thought and behavior, be they adaptive benefits or
cognitive tools, does not satisfactorily account for all features of religion, such as the subjective feel-
ing of the sacred or the nature of mystical experiences.
A second theistic strategy aims at accommodation rather than confrontation. As with any field
of inquiry that calls on evolutionary theory to do work, here, too, the straightforward move is sim-
ply to declare that God orchestrated mutations and selection to produce believing primates. The
same approach can be applied to cognition. Whereas Bering calls the idea of God an “adaptive illu-
sion” (2011, 7) and E. Thomas Lawson and Robert N. McCauley end their seminal book by saying
that “Dionysus dances not in heaven but in our heads” (1990, 184), Justin Barrett believes that the
mental apparatus that leads us to religious belief is “God-endowed” (2004, 123), whereas V. S.
Ramachandran at least allows for the possibility that God put an “antenna” in humans’ heads to
facilitate divine encounters (BBC Active 2003).
For the atheist, this grasp at rapprochement is fraught with fabrication. First, to assert that
God masterminded the evolutionary processes is often to misconstrue the subject matter. One
repeatedly finds professional theologians either twisting traditional religious beliefs beyond recogni-
tion or unjustifiably inserting supernatural assumptions into Darwinian evolution. Impersonal ran-
domness and chance are inconsistent with any notion of divine guidance or directedness in the
evolutionary process. As Steve Stewart-Williams points out, “Every attempt to effect this reconcili-
ation does grievous damage either to the religion, or to evolutionary theory, or to both” (2010,
61). Second, theism is used like a philosopher’s stone; naturalistic explanation simply cannot pre-
clude religious interpretation. “God” is a placeholder with infinite utility. Theists do not need
“gaps” in which to insert God; God can be made to fit into—or, rather, behind—any preexisting
naturalistic system. As Keith Thomson rightly surmises, “Experience suggests that even if a scien-
tific proof of the origin of life from nonlife by purely material means were to be demonstrated in
a laboratory experiment, there would still be a huge number of people who would say, ‘Yes, but
that doesn’t mean that God didn’t do it first’” (2009, 12).
Stalemate, however, need not be the final word. In the present context, there are two rather
powerful rejoinders to the position of theism. The first, interestingly and certainly ironically, comes
from the cognitive science of religion itself. If the cognitive account shows anything, it is how eas-
ily—how naturally—theism is conceived and maintained. The suite of cognitive dispositions out-
lined earlier make it extraordinarily difficult to render the world in wholly naturalistic terms. Super-
natural thinking and teleological reasoning begin in early childhood (Barrett 2012; Kelemen 1999),
and these same proclivities continue throughout life. Theism is intuitively irresistible. It feels right.
Compared with theism, atheism is unnatural. Atheism actually cuts against humans’ innate intui-
tions. But human intuitions, of course, are often profoundly mistaken.
Second, it is possible, at least tentatively, to weigh the balance of atheism and theism. As
Michael Ruse points out, “Probably nothing necessitates metaphysical naturalism, but it would
be naïve simply to say that there is nothing in methodological naturalism and its pursuit that
has implications for God and his existence and nature” (2013, 394). Inference to best explana-
tion is a form of abduction embraced by both science and religion as a reasonable guide to truth.
Paley’s natural theology and contemporary intelligent design depend on it, as does Richard Swin-
burne, who argues that God is the best explanation for the complex world humans observe
around them (2004). Yet it is precisely this path to understanding that demonstrates the elegance
and superiority of naturalistic explanations, which are able to explain more and more without
resorting to the supernatural. The more one knows of physics, the less the universe appears to
be the product of divine creation. The more one knows of evolutionary biology, the less con-
trived seems life.
The same is true for with today’s evolutionary and cognitive theories of religion. The more one
knows about the adaptive origins of human behavior, the more sense is made of the ritualized con-
duct ubiquitous to religious practice. The more one knows about the structures and functions of the
human brain, the less special seem the supernatural agency, purpose, and design that characterize
religious thought. Granted, these naturalistic explanations do not and cannot disprove religion as
such. Given the ease with which theism can be made to fit into any naturalistic account, they need
not threaten scientifically minded believers. But if inference to the best explanation is to be the
guide, the continuing success of these approaches at exposing the naturalistic foundations of religion
most strongly suggests that theism is much more likely an idea than an actuality.

MYTH AND MAGIC
Gregory W. Dawes, University of Otago
Before we can assess religious utterances—before we can answer the question “Do we have reason to
regard them as true?”—we need to understand their character. Philosophers discussing religion com-
monly make two assumptions. They assume that when devotees speak of gods, spirits, demons, or
some hidden, nonnatural realm of existence, they are intending to state facts. Their utterances are
(to use the language of speech act theory) assertives, making claims about the way the world is. Phi-
losophers also assume that such assertions are, or at least should be assessed as, explanations of
things we can observe. Gods are, in this view, comparable to the unobservable entities and forces
posited by modern physics, even if we have less reason to believe in the former than the latter.
AGAINST INTELLECTUALISM
We can combine these two assumptions in a single thesis, which I shall call the intellectualist thesis:
religious utterances are assertives that posit unseen entities in an attempt to explain some group of
observable phenomena. This thesis is associated with the founding figure of modern anthropology,
Edward B. Tylor (1832–1917), who held that “spirits are simply personified causes” (Tylor 1913,
108). In this view, religious claims should be assessed by (a) examining the evidence that can be
offered in their support and (b) asking if there are better, nonreligious explanations on offer. There
is a role for assessments of this kind, for many religious utterances do make explanatory claims,
albeit of a non-scientific kind. Critics of the intellectualist (or “Tylorean”) thesis insist, however,
that this way of treating religious utterances misses some of their most important and distinctive
characteristics.
Symbolist Views. The first thinkers to oppose intellectualism were those who offered a symbolist
view of religion. This view held that religious utterances cannot be understood as “scientific propo-
sitions, based on experience and on a belief in the uniformity of nature” (Beattie 1966, 72). Even if
they do make explanatory claims, their primary function is expressive. “They provide people with a
means of representing abstract ideas, often ideas of great practical importance of themselves indi-
rectly, ideas which it would be difficult or even impossible for them to represent to themselves
directly” (Beattie 1964, 70).
The symbolist view can take two forms. From what I call a quasi-symbolist perspective, reli-
gious utterances are assertives—they do assert the existence of gods and other nonnatural entities,
but they do not originate as attempts to provide explanations of observable phenomena. They origi-
nate in order to serve some other purpose, such as the legitimation of a certain kind of social order.
From a pure symbolist view, by way of contrast, religious utterances may appear to be about gods
(and other nonnatural entities), but they should not be thought of as making assertions about such
entities. They are serving a quite different function, such as that of committing the speaker to a cer-
tain way of life.
Modes of Thought. The view of religious utterances I am about to describe is closer to the quasi-
symbolist than the pure symbolist position. But it emerges from a wider debate, which has to do
with differing modes of thought, although these went by other names as well, such as collective
representations, mentalités, and orientations to the world. This debate was prompted by the study
of small-scale tribal societies, whose members appeared to employ categories of thought and modes
of inference very different from those characteristic of the modern sciences. A classic instance is the
claim made by the Nuer people of southern Sudan that “twins are birds” (Beattie 1964, 68). Newer
studies have generalized this observation, suggesting not merely that differing modes of thought are
found in all societies but also that they are often contextual, being employed by the same people at
different times and for different purposes (Lloyd 1990, 28, 93–95).
How can we think of these differing modes of thought? First, the phrase modes of thought may
be misleading. The peoples in question do sometimes use differing forms of inference, but these
commonly reflect differing ontologies and epistemologies: conceptions of what kinds of things exist
and how we obtain knowledge of them (Nisbett et al. 2001). Second, we need to be careful about
how we classify modes of thought. Those studying this issue have sometimes made sharp distinc-
tions, such as that between “logical” and “prelogical” (Lévy-Bruhl 1926, 78) or “mythic” and “the-
oretic” thought (Donald 1991, 269–275). It is more helpful, however, to think of modes of thought
as falling on a spectrum. At one end of this spectrum are forms of speech that attempt to be as lit-
eral as possible and which employ carefully structured arguments. Such thinking distinguishes
between different kinds of knowledge claims—religious, ethical, political, and scientific—and tries
to keep them distinct. At the other end of the spectrum are forms of speech that use narrative rather
than argument and are rich in imagery and metaphor (Bruner 1986, 11–19). Discourse of the latter
kind combines knowledge claims of differing kinds, to create complex texts with layers of meaning.
Astronomical knowledge, for instance, derived from careful observation, may be presented in reli-
gious terms (Best 1922; Breen 2006). Practical knowledge, such as that relating to navigation,
may be shot through with allusions to history and narratives of origin (Biggs 1994; Salesa 2004).
Among the forms of speech that employ narrative, imagery, and metaphor we find what are called
“myths.”
MYTH AND RELIGION

The term myth, like religion, is best regarded as a kind of family resemblance concept. In one version
of the family-resemblance view, myths would be linked by threads made up of features that are
common to two or more instances, although there may be no thread that links them all. My idea
of a family resemblance concept is a little different. It begins with a paradigmatic instance, lists its
features, and incorporates other instances to the extent that they share these features. A paradigmatic
myth, I assume, encompasses the following characteristics (Cohen 1969): (1) it has a narrative form,
(2) describes origins or transformations, (3) speaks of nonnatural agents and activities, (4) has nor-
mative significance, (5) uses stories about individuals to refer to abstractions and relations, (6) is
recited or reenacted in ritual, and (7) is treated as self-evidentally true, requiring no external legiti-
mation.
The relation between the major theistic religions and myth is a complex one. The gods of myth
are intimately involved, or even identified, with aspects of what we call the “natural” world. During
the period of what is sometimes called the Axial Age (c. 800–200 BCE), however, some theists
began making a sharp distinction between God and the world, insisting that God was entirely dis-
tinct from his creation (Jaspers 1953, 3). We find a parallel development occurring among the ear-
liest Greek philosophers, who begin to develop the idea of a natural world (Descola 2013, 63–68), a
world distinct from that of the gods.
One way of stating this is to say that the major theistic faiths have undertaken a partial demy-
thologization of their central claims. The demythologization remains partial, however, because their
foundational texts retain many of the characteristics of myth. The biblical book of Genesis, for
instance, includes two creation myths, one of which is echoed in the Qurʾan’s repeated claim that
God created the world in six days. The R·igveda (10.90) includes a myth (the Purus.asūkta) about
the sacrifice of a giant man, from whose body parts the world was created. The biblical stories of
the Exodus and the death of Jesus function mythically, insofar as they are recounted in ritual con-
texts. Insofar as all these stories are thought of as having an other-than-human origin, they share
the taken-for-granted authority of myths. Demythologization, it seems, can go only so far. Religion
requires myth, and “mythology is finally extruded from religion only when religion itself perishes
and gives place to philosophy” (Collingwood 1924, 152).
INTERPRETING MYTH
Insofar as religions continue to employ mythic texts, these texts need to be interpreted before their
claims can be assessed. The interpretation of religious myth began as part of the process of demy-
thologization. Within the Christian and Islamic traditions, theologians employed the categories of
Platonic and Aristotelian thought to create rational reconstructions of their central beliefs. This pro-
cess was always controversial. Not only were particular interpretations debated, but the legitimacy of

applying philosophical analysis to religious traditions was itself contested. A striking example is the
medieval debate between the mutakallimūn (theologians) and the falāsifa (philosophers) of Islam.
Some of the former, such as Abū Ḥāmid Muḥammad al-Ghazālī (c. 1058–1111), rejected the idea
that one may use philosophy to explicate matters of faith.
This process of interpretation continues to the present day. It is a complex process, for two rea-
sons. The first is that mythic texts are notoriously ambiguous. The biblical story of the creation of
Eve from the rib of Adam, for instance, seems intended to say something about the relation between
men and women, but it is unclear just what its message is. The second reason is that myths can
embody ways of understanding the world that are quite different from those with which we are
familiar. Taking seriously other people’s myths can involve calling into question the assumptions
that underlie our own view of the world.
MYTH AND EXPLANATION

At some point, however, the philosopher will want to assess mythic texts. In doing so, we should
avoid the error that characterized pure symbolist accounts of religion. Symbolist accounts rightly
recognized that religious utterances have functions other than those of assertion and explanation.
The pure symbolist view, however, wrongly concluded that such utterances have no assertive or
explanatory role. Mythic discourse does make assertions. Its claims about the origins of the world,
for instance, really are claims about origins. It is true that myths are not presented as scientific
hypotheses, to be accepted or rejected on the basis of their explanatory power. Nor is their explana-
tory form that of a scientific theory (Beattie 1966, 72; Dawes 2007, 68–74). Myths are, however,
still used to explain observable facts.
It is at this point that atheists can seek to undermine the authority of religious myths. They can do
so by offering a more adequate explanation of the facts in question, one that makes no reference to non-
natural entities. An argument of this kind, however, rests on a further assumption. It is that we should
accept the existence of unobservable entities only when they are required in order to offer an adequate
explanation of some observable state of affairs. This assumption is not shared by many religious thinkers,
who hold that their myth, at least, is revealed by God and requires no further evidential support. This is
one reason why theistic and atheistic thinkers so often seem to be talking past one another.
I have spoken about myth and said nothing about magic. The two are closely related since magi-
cal practices often draw upon myth (Malinowski [1926] 1948, 114–119). Insofar as magic has a prac-
tical rather than a theoretical orientation, however, it lies outside the scope of the present discussion.
Let me make just one suggestion. If myths do have functions other than that of explanation, they
could be useful even if their explanations are false. In a similar way, magic can have an efficacy that
does not depend on the existence of the occult forces it invokes. (For a striking illustration, see Firth
1959, 244–279.) It follows that magic, like myth, should be taken seriously by scholars of religion.
Bibliography
Alston, William. “Psychoanalytic Theory and Theistic Belief.” In Faith and Beattie, John. Other Cultures: Aims, Methods and Achievements in Social
the Philosophers, edited by John Hick, 63−102. New York: St. Martin’s Anthropology. London: Cohen and West, 1964.
Press, 1964.
Beattie, John. “Ritual and Social Change.” Man, n.s., 1, no. 1 (1966): 60–
Austin, John Langshaw, and James Opie Urmson. How to Do Things with 74.
Words: The William James Lectures Delivered at Harvard University in Bering, Jesse. The Belief Instinct: The Psychology of Souls, Destiny, and the
1955. Oxford: Clarendon Press, 1962. Meaning of Life. New York: Norton, 2011.
Barrett, Justin L. Why Would Anyone Believe in God? Lanham, MD: AltaMira Best, Elsdon. The Astronomical Knowledge of the Maori, Genuine and
Press, 2004. Empirical. Dominion Museum Monograph 3. Wellington, New
Zealand: Dominion Museum, 1922.
Barrett, Justin L. Born Believers: The Science of Children’s Religious Belief.
New York: Free Press, 2012. Biggs, Bruce. “Knowledge as Allegory.” In Science of Pacific Island Peoples.
Vol. 4: Education, Language, Patterns and Policy, edited by John
BBC Active. God on the Brain: A Neurological Basis for the Religious Impulse. Morrison, Paul Geraghty, and Linda Crowl, 1–11. Suva, Fiji: Institute of
Directed by Liz Tucker. 2003. Documentary. Pacific Studies, 1994.
Boudry, Maarten, Stefaan Blancke, and Johan Braeckman. “How Not to Freud, Sigmund. Totem and Taboo: Resemblances between the Mental Lives of
Attack Intelligent Design Creationism: Philosophical Misconceptions Savages and Neurotics. Translated by James Strachey. New York: Norton,
about Methodological Naturalism.” Foundations of Science 15, no. 3 1950. First published 1913.
(2010): 227–244. https://doi.org/10.1007/s10699-010-9178-7.
Freud, Sigmund. Moses and Monotheism. Translated by Katherine Jones.
Boyer, Pascal. The Naturalness of Religious Ideas: A Cognitive Theory of New York: Vintage, 1955. First published 1939.
Religion. Berkeley: University of California Press, 1994.
Freud, Sigmund. The Future of an Illusion. Translated by James Strachey.
Boyer, Pascal. Religion Explained: The Evolutionary Origins of Religious New York: Norton, 1989. First published 1927.
Thought. New York: Basic Books, 2001.
Guthrie, Stewart Elliott. Faces in the Clouds: A New Theory of Religion. New
Breen, John. “Inside Tokugawa Religion: Stars, Planets and the Calendar-as- York: Oxford University Press, 1993.
Method.” Culture and Cosmos 10 (2006): 63–78.
Harris, Marvin. Cows, Pigs, Wars, and Witches: The Riddles of Culture. New
Bruner, Jerome. Actual Minds, Possible Worlds. Cambridge, MA: Harvard York: Random House, 1974.
Horton, Robin. Patterns of Thought in Africa and the West: Essays in Magic,
Cohen, Percy S. “Theories of Myth.” Man, n.s., 4, no. 3 (1969): 337–353. Religion and Science. Cambridge, UK: Cambridge University Press,
Collingwood, R. G. Speculum Mentis; or, The Map of Knowledge. Oxford: 1993.
Clarendon Press, 1924. Hume, David. An Enquiry concerning Human Understanding. London:
Costa, James T. Darwin’s Backyard: How Small Experiments Led to a Big A. Millar, 1748.
Theory. New York: Norton, 2017. Hume, David. Dialogues concerning Natural Religion. Edited by Norman
Coyne, Jerry A. Faith vs. Fact: Why Science and Religion Are Incompatible. Kemp Smith. Indianapolis, IN: Bobbs-Merrill, 1947. First published
New York: Viking, 2015. 1779.
Darwin, Charles. On the Origin of Species by Means of Natural Selection. Hume, David. The Natural History of Religion. In Dialogues concerning
London: John Murray, 1859. Natural Religion and The Natural History of Religion, edited by J. C. A.
Gaskin, 134−196. Oxford: Oxford University Press, 2009. First
Darwin, Charles. The Descent of Man, and Selection in Relation to Sex. published 1757.
London: John Murray, 1871.
Jaspers, Karl. The Origin and Goal of History. Translated by Michael Bullock.
Darwin, Francis. The Life and Letters of Charles Darwin. Vol. 2. New York: New Haven, CT: Yale University Press, 1953.
Appleton, 1887.
Johnson, Dominic. “Gods of War: The Adaptive Logic of Religious
Dawes, Gregory W. “Paradigmatic Explanation: Strauss’s Dangerous Idea.” Conflict.” In The Evolution of Religion: Studies, Theories, and Critiques,
Louvain Studies 32, nos. 1–2 (2007): 64–77. edited by Joseph Bulbulia, Richard Sosis, Erica Harris, et al., 111–118.
Dawkins, Richard. The Blind Watchmaker: Why the Evidence of Evolution Santa Margarita, CA: Collins Foundation Press, 2008.
Reveals a Universe without Design. New York: Norton, 1986. Johnson, Dominic. God Is Watching You: How the Fear of God Makes Us
Descola, Philippe. Beyond Nature and Culture. Translated by Janet Lloyd. Human. New York: Oxford University Press, 2016.
Chicago: University of Chicago Press, 2013. Kahane, Guy. “Evolutionary Debunking Arguments.” Nous 45, no. 1
Donald, Merlin. Origins of the Modern Mind: Three Stages in the Evolution of (2011): 103–125. https://doi.org/10.1111/j.1468-0068.2010.00770.x.
Culture and Cognition. Cambridge, MA: Harvard University Press, 1991. Kelemen, Deborah. “Why Are Rocks Pointy? Children’s Preference for
Douglas, Mary. Purity and Danger. New York: Routledge and Kegan Paul, Teleological Explanations of the Natural World.” Developmental
1966. Psychology 35, no. 6 (1999): 1440–1453.
Durkheim, Émile. The Rules of Sociological Method. 8th ed. Edited by George Koenig, Harold G. Medicine, Religion, and Health: Where Science and
E. G. Catlin. Translated by Sarah A. Solovay and John H. Mueller. New Spirituality Meet. West Conshohocken, PA: Templeton Foundation
York: Free Press, 1938. First published 1895. Press, 2008.
Durkheim, Émile. The Elementary Forms of the Religious Life. Translated by Lankester, E. Ray. “Charles Robert Darwin.” In The Library of the World’s
Joseph Ward Swain. New York: Free Press, 1965. First published 1912. Best Literature: An Anthology in Thirty Volumes. Vol. 2, edited by C. D.
Warner, 4391. New York: J. A. Hill, 1896.
Evans-Pritchard, Edward Evan. Witchcraft, Oracles and Magic among the
Azande. Abr. ed. Oxford: Oxford University Press, 1976. First published Lawson, E. Thomas, and Robert N. McCauley. Rethinking Religion:
1937. Connecting Cognition and Culture. Cambridge, UK: Cambridge
University Press 1990.
Fales, Evan. “Naturalism and Physicalism.” In The Cambridge Companion to
Atheism, edited by Michael Martin, 118–134. Cambridge, UK: Leach, Edmund. Political Systems of Highland Burma: A Study of Kachin
Cambridge University Press, 2007. Social Structure. London: Athlone Press, 1954.
Fales, Evan. Divine Intervention: Metaphysical and Epistemological Puzzles. Leach, Edmund. Genesis as Myth and Other Essays. London: Jonathan Cape,
New York: Routledge, 2010. 1969.
Firth, Raymond. Economics of the New Zealand Maori. 2nd ed. Wellington, Lévy-Bruhl, Lucien. Primitive Mentality. Translated by Lilian A. Clare.
New Zealand: R. E. Owen, 1959. London: George Allen and Unwin, 1923. First published 1922.
Frazer, James George. The Golden Bough: A Study in Comparative Religion. Lévy-Bruhl, Lucien. How Natives Think. Translated by Lilian A. Clare. New
New York: Macmillan, 1890. York: Knopf, 1926.

Lloyd, G. E. R. Demystifying Mentalities. Themes in the Social Sciences. Schiefenhövel, Wulf. “Explaining the Inexplicable: Traditional and
Cambridge, UK: Cambridge University Press, 1990. Syncretistic Religiosity in Melanesia.” In The Biological Evolution of
Religious Mind and Behavior, edited by Eckhart Voland and Wulf
Malinowski, Bronisław. “Myth in Primitive Psychology.” In Magic, Science
Schiefenhövel, 143–164. Berlin: Springer, 2009.
and Religion and Other Essays, edited by Robert Redfield, 72–124.
Glencoe, IL: Free Press, 1948. First published 1926. Slone, D. Jason. Theological Incorrectness: Why Religious People Believe What
They Shouldn’t. Oxford: Oxford University Press, 2004.
Malinowski, Bronisław. Argonauts of the Western Pacific. New York: E. P.
Dutton, 1961. First published 1922. Sosis, Richard, and Candace Alcorta. “Signaling, Solidarity, and the Sacred:
The Evolution of Religious Behavior.” Evolutionary Anthropology 12,
Marx, Karl. A Contribution to the Critique of Hegel’s Philosophy of Right? 1843. no. 6 (2003): 264–274.
https://www.marxists.org/archive/marx/works/1843/critique-hpr/intro.htm.
Sosis, Richard, and Bradley J. Ruffle. “Religious Ritual and Cooperation:
Masci, David. “Scientists and Belief.” Pew Research Center Religion and Testing for a Relationship on Israeli Religious and Secular Kibbutzim.”
Public Life Project. 2009. http://www.pewforum.org/2009/11/05 Current Anthropology 44, no. 5 (2003): 713–722.
/scientists-and-belief/.
Stewart-Williams, Steve. Darwin, God, and the Meaning of Life: How
McCauley, Robert N. Why Religion Is Natural and Science Is Not. New York: Evolutionary Theory Undermines Everything You Thought You Knew.
Oxford University Press, 2011. Cambridge, UK: Cambridge University Press, 2010.
Miller, Geoffrey. The Mating Mind: How Sexual Choice Shaped the Evolution Swinburne, Richard. The Existence of God. 2nd ed. Oxford: Oxford
of Human Nature. New York: Doubleday, 2000. University Press, 2004.
Nisbett, Richard E., Kaiping Peng, Incheol Choi, and Ara Norenzayan. Talmont-Kaminski, Konrad. “For God and Country, Not Necessarily for
“Culture and Systems of Thought: Holistic versus Analytic Cognition.” Truth: The Nonalethic Function of Superempirical Beliefs.” Monist 96,
Psychological Review 108, no. 2 (2001): 291–310. no. 3 (2013a): 447–461. https://doi.org/10.5840/monist201396320.
Norenzayan, Ara. Big Gods: How Religion Transformed Cooperation and Talmont-Kaminski, Konrad. Religion as Magical Ideology: How the
Conflict. Princeton, NJ: Princeton University Press, 2013. Supernatural Reflects Rationality. Durham, NC: Acumen, 2013b.
Norenzayan, Ara, and Will M. Gervais. “The Origins of Religious Disbelief.” Thomson, Keith. “Introduction.” In The Religion and Science Debate: Why
Trends in Cognitive Sciences 17, no. 1 (2013): 20–25.
Does It Continue?, edited by Harold W. Attridge, 1–14. New Haven,
Paul, Gregory. “The Chronic Dependence of Popular Religiosity upon CT: Yale University Press, 2009.
Dysfunctional Psychosociological Conditions.” Evolutionary Psychology 7,
Tremlin, Todd. Minds and Gods: The Cognitive Foundations of Religion. New
no. 3 (2009): 398–441.
Plantinga, Alvin. “When Faith and Reason Clash: Evolution and the Bible.”
Turner, Terrence S. “Oedipus: Time and Structure in Narrative Form.” In
Christian Scholar’s Review 21, no. 1 (1991): 8–32.
Forms of Symbolic Action: The Proceedings of the 1969 Annual Spring
Plantinga, Alvin. Warrant and Proper Function. Oxford: Oxford University Meeting of the American Ethnological Society, edited by Robert F. Spencer,
Press, 1993. 26−69. Seattle: University of Washington Press, 1969.
Plantinga, Alvin. Warranted Christian Belief. New York: Oxford University Tylor, Edward B. Primitive Culture: Researches into the Development of
Press, 2000. Mythology, Philosophy, Religion, Language, Art and Custom. Vol. 2, 5th
Radcliffe-Brown, Alfred Reginald. The Andaman Islanders: A Study in Social ed. London: John Murray, 1913.
Anthropology. New York: Simon and Schuster, Free Press, 1964. First Van Eyghen, Hans. “There Is No Sensus Divinitatis.” Journal for the Study of
published 1922. Religions and Ideologies 15, no. 45 (2016): 24–40.
Rappaport, Roy A. Pigs for the Ancestors: Ritual in the Ecology of a New Wilkins, John S., and Paul E. Griffiths. “Evolutionary Debunking
Guinea People. New Haven, CT: Yale University Press, 1968. Arguments in Three Domains: Fact, Value, and Religion.” In A New
Rappaport, Roy A. Ritual and Religion in the Making of Humanity. Science of Religion, edited by Gregory W. Dawes and James Maclaurin,
Cambridge, UK: Cambridge University Press, 1999. 133–146. New York: Routledge, 2012.
Rolston, Holmes, III. Three Big Bangs: Matter-Energy, Life, Mind. New York: Williams, George C. Adaptation and Natural Selection: A Critique of Some
Columbia University Press, 2010. Current Evolutionary Thought. Princeton, NJ: Princeton University Press,
1966.
Ruse, Michael. “Naturalism and the Scientific Method.” In The Oxford
Wilson, David Sloan. Darwin’s Cathedral: Evolution, Religion, and the Nature
Handbook of Atheism, edited by Stephen Bullivant and Michael Ruse,
of Society. Chicago: University of Chicago Press, 2002.
383–397. Oxford: Oxford University Press, 2013.
Wilson, Edward O. The Meaning of Human Existence. New York: Liveright,
Salesa, Damon. “Finding and Forgetting the Way: Navigation and Knowledge
2014.
in Samoa and Polynesia.” Princeton Workshop in the History of Science.
2004. https://www.princeton.edu/∼hos/Workshop%20II%20papers/Salesa Winch, Peter. The Idea of a Social Science and Its Relation to Philosophy.
%20paper.pdf (cited with permission). New York: Routledge and Kegan Paul, 1970. First published 1958.
TOPIC 19
Prudential/Pragmatic Arguments: Theism

Joshua Golding
Bellarmine University, Louisville, KY
This chapter discusses pragmatic arguments for religious commitment, such as the wager-approach, the will-
to-believe, and an argument for pragmatic faith.
PRAGMATIC VERSUS COGNITIVE ARGUMENTS
Throughout the ages, philosophers have debated whether it is rational to believe or have faith in
God. Many philosophers have given arguments in favor of theistic commitment, and many others
have opposed such arguments. In general, it is possible to distinguish two kinds of arguments: cog-
nitive and pragmatic. A cognitive argument is one that seeks to show that it is rational to believe or
accept some claim, on the grounds that it can be demonstrated that the claim is either true or very
likely to be true. Thus a cognitive argument for theistic belief seeks to show that we can know with
certainty or a high degree of probability that God exists. On the other hand, a pragmatic argument
seeks to show that it is rational to believe or accept some claim because of some great potential value
to be gained by believing or accepting that claim. Thus a pragmatic argument for theistic commit-
ment seeks to show that it is rational to believe or have faith that God exists, on the grounds that
there is some great potential value in having such a commitment.
Generally speaking, the classic arguments for belief in God are cognitive. These include the
ontological, the cosmological, and the teleological arguments, as well as the argument from religious
experience. Such arguments aim to show that it is rational to believe in God, on the grounds that
the very idea of God implies that God exists, or on the grounds that there is evidence that God
exists. The French philosopher and religious thinker Blaise Pascal offered what seems to be the first
major instance of a pragmatic argument for religious belief. The argument has come to be known as
Pascal’s wager. Simply put, Pascal argued that it is rational to believe in God because one has more
to gain than to lose by being a religious believer. Various versions of such an argument have been
discussed ever since. Notably, a version of this argument occurs in the work of William James. This
chapter aims to show that, while the arguments of both Pascal and James are flawed, some form of
pragmatic argument in favor of religious commitment is quite plausible. Indeed, it is partly by rec-
ognizing the flaws of these earlier versions that we may construct a more convincing, if also more
modest, pragmatic argument for religious commitment.
Before proceeding, two preliminaries are necessary. First, whenever philosophers claim that it is
rational to believe or accept some claim, there are two senses in which they may be using the term
rational. One might argue that some claim is rationally compelling, or that some claim is rationally
defensible. To argue that a claim is rationally compelling is to claim that all rational persons should
believe or accept that claim. Stated differently, to argue that a claim is rationally compelling is to
argue that anyone who does not accept that claim is irrational. On the other hand, to argue that
a claim is rationally defensible is to argue that a person is within his or her “rational rights” to accept
that claim, even though not all rational persons must accept that claim. Clearly, the claim that some
637
position is rationally defensible is weaker than the claim that

KEY CONCEPTS some position is rationally compelling. This distinction pertains
to all arguments, both cognitive and pragmatic. However, as we
Argument from Religious Experience shall see, there may be an especially apt application of the
notion of rational defensibility to the case of a pragmatic argu-
Cognitive Argument for Faith ment, and especially the case of a pragmatic argument for theis-
Expected Value Calculation tic commitment.
Pascal’s Wager Second, insofar as any pragmatic argument is bound to
Pragmatic Argument for Faith have steps or premises, it is to be expected that there should
be some justification for those premises. For example, if one
Rationally Compelling Claim were to argue that it is pragmatically justifiable to believe
Rationally Defensible Claim that one is well liked by one’s peers on the grounds that hav-
ing this belief is likely to bring it about that one makes
friends more easily, one still requires justification for the
claim that having the stated belief is likely to bring about
the stated result. Such a claim might be justified by citing cognitive evidence that a certain belief
in one’s own popularity makes it more likely that one will make friends. Similarly, the pragmatic
arguments for religious commitment considered here involve premises that themselves may have
some cognitive justification. In that sense, none of these arguments is “purely” pragmatic. Nev-
ertheless, what is pragmatic is the overall strategy of trying to show that it is rational to have reli-
gious commitment because of some great potential value in having such commitment.
SUMMARY OF PASCAL’S WAGER
Roughly, the wager runs as follows. First, Pascal claims that inasmuch as the human mind is finite
and God is conceived as an infinite being, it is impossible to know whether God exists. Some things
in the world seem to point to God’s existence; other things seem to indicate there is no God. For
Pascal, we are in position to regard the proposition that God exists and the proposition that God does
not exist as “equiprobable”—that is, each of these propositions is as likely (or unlikely) as the other.
Pascal seems to infer from this that there is a one-out-of-two chance that God exists, and a one-out-
of-two chance that God does not exist. Yet, Pascal continues, we are “forced” to decide whether to
believe in God or not. One cannot remain on the fence, for doing so is tantamount to not believing
in God. Thus one is forced to choose whether or not to believe in God, in the face of cognitive
doubt about whether God exists. Under such circumstances, the only reasonable way to decide what
to believe is to consider the potential effects of belief (and disbelief) in God on one’s “goodness” or
“happiness.” In modern terms, one must resort to pragmatic considerations in order to make this
decision.
Next, Pascal argues as follows. First, let us consider what might happen if one chooses to
believe in God. If one believes in God and it turns out that indeed God exists, one will gain an
“infinity of infinitely happy lives,” that is, an infinitely valuable good. Evidently, Pascal has in mind
the eternal bliss of heaven, which God (if he exists) will grant as a reward in the afterlife to those
who believe in him. On the other hand, if it turns out that God does not exist, the believer will
have lost only something that is finite. For, by living a religious life, he will presumably forgo many
worldly pleasures that he would have had, had he led the life of an atheist (especially a debauched
one!). All these worldly pleasures are only finite, both in quality and duration. In summary so far,
if one chooses to believe in God, one has (in the best case) some chance of gaining an infinite value,
and (in the worst case) some chance of losing some amount of finite value.
On the other hand, suppose one chooses not to believe in God. In that case, Pascal assumes
that, whether or not God exists, one has no chance of going to heaven, and therefore no chance
of gaining an infinite value. The best life one can lead as an atheist will be a life that contains only
finite value. The atheist may enjoy many fine things in this life that the religious believer gives up,
but all of those things are only finite. It turns out, then, that a person has a chance of getting infi-
nite value only if he chooses to believe in God.
Based on these assumptions, Pascal’s argument continues as follows. In any decision that is
made under uncertainty, if the prize for winning a bet is large enough, it is rational to risk the loss
of some lesser prize in order to secure even a small chance for some very great prize. In modern
terms, such a strategy for decision making is known as an expected value calculation. Roughly, the
expected value of an option is determined by multiplying the probability of gaining a certain result
by the value of that result. To take a simple example, assuming the cost of two bets is the same, a
one-in-ten chance of winning $1,000,000 is a better bet than a one-in-two chance of winning
$100. The expected value of the first bet is 1/10 times $1,000,000, which is $100,000; the
expected value of the second bet is 1/2 times $100, which is only $50. Returning to the case at
hand, the value of attaining eternal bliss is infinite. Moreover, Pascal has already argued that, in this
case, the probability of “winning” eternal bliss is rather high, namely, one-in-two. For, as stated ear-
lier, the existence of God is equiprobable with God’s nonexistence. Clearly, a one-in-two chance of
attaining infinite value is a better bet than even a very high chance of gaining only finite value. The
expected value of belief in God is 1/2 times infinity, which equals infinity; the expected value of dis-
belief is 1/2 times some finite number, which equals some finite number. Hence it is pragmatically
rational to believe in God rather than not to believe in God.
At this point, the major claim of the wager has been reached. However, Pascal considers a
practical objection. Even though someone may accept the entire argument up to this point, what
can a person do if he just so happens not to be a believer? It does not seem that one can simply
decide to believe in God (or any proposition for that matter). In response, Pascal proposes a rem-
edy, namely, that such a person should “take holy water,” that is, go to church, act like a believer,
talk like a believer, and sooner or later he will find that he is a believer in God. In other words,
although one may not be able to decide immediately to believe in God, one can slowly but surely
engender such a belief in oneself by engaging in religious behavior. Hence the practical objection
is overcome.
CRITICISMS OF PASCAL’S WAGER
Pascal’s wager is subject to numerous criticisms. Before stating the criticisms that constitute genuine
flaws, it is worth disposing of one potential objection that is off the mark. One might be inclined to
object that it is somehow morally or religiously inappropriate to believe in God on the basis of pru-
dential calculations. Shouldn’t a belief in God be motivated by something holier or purer than
the desire for personal gain, whether in this world or the hereafter?
There are several responses to this objection. First, as explained above, Pascal has already fore-
stalled it. Given the assumption that one is forced to decide what to believe, and that one cannot
make the decision on cognitive grounds, it is claimed that we have no choice but to make the deci-
sion on pragmatic grounds. (Whether those assumptions are correct is another matter.) Second,
the objection that one should not be a religious believer on prudential grounds is rather sanctimo-
nious. Many decisions in life are based on self-interest, and it is not clear that religious commitment
should be any different. The Bible itself declares it is in our best interest to believe in God. Putting
the point bluntly, the kind of God about which Pascal is concerned is the kind of God that seems
not to mind if people believe in him out of self-interest. Third, nothing in the wager rules out the
possibility that indeed there may be some “higher way” of believing in God that is not motivated by
self-interest alone. Whether or not there is such a “higher way,” the purpose of the wager is to show
that belief in God is justified on pragmatic grounds. Finally, Pascal could easily argue that the ulti-
mate end of achieving the infinite gain in the next world is not a matter of sheer self-interest but
rather of our attaining what is objectively our best possible state. This is because the attainment
of “heaven” is conceived not only as blissfully pleasurable but also as the best possible state for
our soul. Thus the value at stake in being a religious believer is in fact not a matter of sheer self-
interest but rather of attaining what is best.

However, there are several other criticisms of the wager that are not easily dismissed. Here we
shall consider five criticisms. First, Pascal does not sufficiently support his claim that reason is inca-
pable of forming a plausible cognitive judgment about whether God exists. Despite the notion that
human mind is finite and God is infinite, it does not follow that one cannot make some cognitive
estimation of whether there is an all-powerful and omnibenevolent God. What might follow is that
the mind cannot be certain of whether there is a God or not; but Pascal needs more than that to
support his claim that the only way to make the decision about whether to believe is on pragmatic
grounds.
Second, Pascal does not adequately support his claim that the choice to believe is “forced.” If
indeed from a cognitive point of view it were the case that God’s existence is equiprobable with
God’s nonexistence, it seems that the rational thing to do is to suspend judgment or be an agnostic,
that is, neither to believe that God exists nor to believe that God does not exist. It may very well be
true that agnosticism is “tantamount” to atheism in the sense that neither the agnostic nor the athe-
ist will go to heaven (if God exists), but from a cognitive point of view, these are genuinely different
positions. Hence the decision is not forced after all.
Third, Pascal assumes without argument that one will attain the infinite gain only if God exists
and one believes in him. Indeed, Pascal assumes that if there is a God at all, God is a certain sort
of God. Numerous other possibilities seem to have escaped consideration. Perhaps there is a God,
but God is not the kind of God described by traditional Christianity. Perhaps it is not God as
understood by Christians that exists but God as understood by Judaism, or God as understood by
Islam, or perhaps some other deity. If so, how might one decide which sort of God to wager on?
Perhaps God is “all-forgiving,” that is, he will grant all individuals of a certain sort (including agnos-
tics and atheists) infinite bliss in the next world. In this case, there is nothing to be gained by being
a believer. Furthermore, some critics have posed the notion of a “nonstandard” God, such as one
who might grant the infinite reward only to atheists and agnostics. Another possibility is that God
is Satanic, in the sense that he rewards only those who are morally evil. In any of these cases, one
would actually lose the infinite reward by choosing to believe in a traditional God! Still another pos-
sibility is that there might be some way to attain infinite value even if there is no God at all. While
to some persons these may seem like remote possibilities, how can Pascal rule them out?
On behalf of Pascal, one might be inclined to respond that one can make some reasonable esti-
mation of such possibilities and choose the option that is most likely to result in eternal bliss. (Later
in this chapter, we pursue such a strategy.) However, given Pascal’s insistence that cognitive reason
is at a loss when it comes to evaluating whether God exists, such a strategy seems off-limits for Pas-
cal. If God is beyond our comprehension, how can we expect to make rational estimations of how
God would make decisions about granting eternal bliss?
Moreover, suppose Pascal were to claim that somehow one can make reasonable estimations
about what the conditions for attaining eternal bliss are. Still, this would bring up a fourth criticism,
which concerns any expected value calculation in a situation where an infinite gain is deemed to be
at stake. Suppose one could somehow estimate that believing in a traditional God is more likely to
result in an infinite gain than believing in some nonstandard or Satanic God. The problem is that
even a very minimal chance of attaining infinite value seems to have the same expected value as a
relatively large chance of attaining infinite value! For example, suppose that on one option a person
has a one-in-two probability of gaining infinity. The expected value of that option is 1/2 times
infinity, which equals infinity. Now suppose that on a second option a person has a one-in-twenty
probability of gaining infinity. The expected value of this second option is 1/20 times infinity,
which also equals infinity. Indeed, this curious mathematical fact should make us suspicious of
any expected value calculation where an infinite gain is deemed to be at stake. It also renders point-
less the strategy of trying to defend the wager by claiming that there is a rational way of estimating
what the most likely conditions are for attaining the eternal reward. Probabilities will not make a
difference where an infinite value is at stake.
Finally, a fifth objection concerns the process that Pascal recommends for those who find his
argument convincing but still don’t believe in God. Perhaps it is possible for one to engender within
oneself a belief in some proposition by constantly repeating that proposition and acting as if one
believes it. However, a genuine belief in some proposition seems to involve a belief in some proposi-
tion that a person really thinks is true. In the case at hand, the person doesn’t really think it is true
that God exists, yet he makes an effort to cultivate that belief. Arguably, Pascal’s recommendation
involves what might be called self-brainwashing. Thus, even for those who find the argument
sound, it yields a kind of belief that is disingenuous or inauthentic.
WILLIAM JAMES AND THE WILL TO BELIEVE
William James proposed an argument that is similar to Pascal’s wager but also different from it in
important respects. James makes clear that his argument is not intended to persuade everyone to
believe in God but rather to defend the “right” of those who believe in God to do so. In other
words, James argues that the belief in God is rationally defensible, if not rationally compelling. He
proposes a general principle that describes those circumstances under which a person is entitled to
believe some proposition on pragmatic grounds. Basically, the principle runs as follows: A person
not only may but must make a decision to believe on pragmatic grounds in those cases where the
options are live, the choice is forced, the value that one stands to gain or lose is great, and one can-
not make the decision on empirical or cognitive grounds. For James, two or more options are live if
those options are psychologically realistic choices for some individual. Clearly, what is live for one
person may not be live for another person. For example, it may not be a live option for someone
in today’s world to choose to believe in Zoroastrianism. Indeed, for some people it may not even
be a live option to believe in God. James would readily admit that his argument does not apply
to such persons. Yet surely there are many people for whom belief or disbelief in God are live
options.
Like Pascal, James argues that cognitive evidence neither proves nor disproves God’s existence,
and that we are nevertheless forced to choose whether or not to believe in God. Unlike Pascal,
James does not rely on the claim that the only way to achieve infinite bliss is to believe in God.
In fact, he does not appeal to the notion of infinite value at all. For James, what is at stake in having
a belief in God is something very valuable for the believer, in this world. He seems to have in mind
the positive psychological benefits of believing in God. A theist, as James sees it, is someone who
believes that “the best things in life are eternal” and that, in the grand scheme of things, everything
works out for the best. Such psychological benefits are unavailable to the nonbeliever. Although he
does not explicitly say so, he would admit that such benefits accrue to the believer, even if it turns
out that God doesn’t exist. Moreover, James would concede that a person who does not find the
belief in God psychologically uplifting is entitled to not believe in God. James’s claim is only that
those who do find the belief in God salutary are rationally entitled to do so.
Additionally, James suggests that those who believe in God may subsequently have some kind
of experience that would validate their belief. He claims that this is yet a further reason for holding
that a “leap” into religious belief is rationally defensible. Without that initial leap, a person may
never experience that validation. Here, James seems to backtrack on his own claim that cognitive
reasoning cannot support a belief in God. Yet he may respond that this kind of experience is not
sufficient to constitute empirical evidence that God exists. It is a kind of experience that confirms
the belief of the believer but that cannot be used to convince others who do not believe.
James’s claim is more modest than Pascal’s wager. In displacing Pascal’s appeal to “infinite
value” with something much lesser, James’s argument has an advantage and also a disadvantage in
comparison to Pascal’s argument. An advantage is that, since there is no appeal to infinite value,
James’s argument avoids the fourth criticism stated above. If the gain deemed to be at stake in being
a religious believer is finite, it makes sense to take into account what the probabilities of attaining
that gain are, and it makes sense (if one can) to estimate what kind of God one should believe in,
in order to have the best chance of attaining that end. On the other hand, a disadvantage is that
by watering down the value of what is at stake in being a religious believer, it turns out that reli-
gious belief is less pragmatically attractive than it is on Pascal’s wager. If the potential gain at stake
in being a religious believer is not infinite, the option of religious belief will not necessarily trump

competing options. James would concede that, at least for some persons, the value that is at stake in
being a religious believer may simply not be worth the bother.
When faced with the objection that there are many different conceptions of God and many dif-
ferent views about what God (if he exists) demands or expects of human beings, James could simply
respond, “just pick your favorite—the one that gives you the greatest psychological lift.” Neverthe-
less, James’s claim is subject to some of the same criticisms as Pascal’s wager. First, it is not clear
that the choice is forced. James argues that the great potential benefits of religious belief are unavail-
able to the nonbeliever, but that this consideration does not imply that the choice is forced. Agnos-
ticism seems like a genuine alternative to both theism and atheism. Nor is it clear that cognitive rea-
soning cannot form some reasonable estimation about whether there is a providential God. Granted,
there are differences of opinion about this question, but perhaps the most rational thing to do is to
keep an open mind and not commit one way or the other. Although James claims that a person
who believes in God is more likely to experience evidence that there is a God, it is equally arguable
that a person who commits to belief in God without already having such evidence is biasing himself
against potential evidence that counts against God’s existence. Finally, James’s claim that validating
religious experience of God is available only to persons who believe in God is subject to the criti-
cism that such “experience” is merely the result of wishful thinking.
AN ARGUMENT FOR PRAGMATIC FAITH IN THE EXISTENCE OF GOD
In this section we consider a pragmatic argument that aims to circumvent the criticisms discussed
above. This version aims not to justify the belief that God exists but rather the rational defensibility
of pragmatic faith in God’s existence. In addition, this version replaces Pascal’s problematic use of
an infinite value with a finite but transcendent value. These terms require explanation.
To define the notion of pragmatic faith, we must first define the notion of living a religious way
of life. A person lives a religious way of life if she pursues a good relationship with God, that is, she
does the things that she thinks are most likely to promote a good relationship with God (if God
exists). The exact nature of that relationship may be a matter of theological debate or religious con-
troversy; we shall return to that point below. Next, we may say that a person has pragmatic faith if
she makes an action-guiding assumption that God exists, for the purpose of pursuing a good relationship
with God. Essentially, such a person conducts her affairs “as if” there is a God. She is inclined to do
the actions that she thinks that God (if he exists) would endorse, and avoid the actions that she
thinks God (if he exists) would forbid. Needless to say, different religious persons will behave differ-
ently, in accord with different beliefs about what God would endorse or forbid. Let us postpone for
the moment the question of how one arrives at such beliefs.
Next, the notion of a transcendent value is the notion of a certain kind of good that is finite but
qualitatively superior to other kinds of goods. One kind of good is qualitatively superior to another
kind of good if it is so vastly better that any small amount of the first kind is better than a vast amount
of the second kind. For example, one might reasonably regard the value of a loving relationship as
qualitatively superior to the kind of good that one attains in experiencing physically pleasurable
states. No matter how much pleasure one might have, it would not equal the kind of good that
one might attain in a loving relationship. Taking this one step further, a transcendent good is a kind
of good that is qualitatively superior to any other kind of good. Let us postpone for the moment the
question of whether such a good is available for humans to attain. If there were such a transcendent
good, a small chance of attaining it would have a higher expected value than even a large chance of
attaining any lesser good. For, by hypothesis, any fraction of a transcendent good is better than a
large amount of a lesser good. Moreover, since a transcendent value is not infinite, a larger chance
of attaining it will have a higher expected value than a lower chance of attaining such a value.
We may now pose the following question. Assume for the sake of argument that a person is in
doubt about whether God exists. Is it pragmatically rational for such a person to adopt pragmatic
faith in God’s existence? Even if the chance of success is deemed to be relatively low, it will be ratio-
nal to adopt pragmatic faith so long as the value of attaining the good relationship with God is
deemed sufficiently high to trump any nonreligious way of living. More precisely, it is pragmatically
rational for a person to pursue a religious way of life and have pragmatic faith, if and only if it is
rational for a person to believe the following three propositions: (1) there is at least some small prob-
ability that God exists, (2) the potential value of attaining a good relationship with God is transcen-
dent, and (3) a religious way of life is more likely than a nonreligious way of life to result in attain-
ing the good relationship with God. For any person who accepts these propositions, the expected
value of choosing to live a life of faith will be higher than any nonreligious alternative. Let us discuss
each of these three propositions in turn.
First Proposition. Clearly, a person who has pragmatic faith must believe that there is at least
some small probability that God exists, for otherwise it would not make sense to pursue the goal
of attaining a good relationship with God. Moreover, such a belief is rationally plausible, on cogni-
tive grounds. After the manner of Pascal, one can argue that it is impossible for us finite humans to
rule out conclusively the possibility that God exists. Furthermore, there is (at least some minimal)
evidence that there is a God, based on the orderliness and existence of the universe and based on
purported religious experiences. Even many atheists agree that there is some (small) probability that
God exists. They just think that there is a much greater probability that God does not exist. Thus
this first proposition is rationally defensible, at least for many persons.
Second Proposition. It is rational to believe that the potential value of attaining a good relation-
ship with God is transcendent in the sense defined above. Surely, different religions understand
the details of this relationship differently. By the same token, at least among the great world reli-
gions, there is much overlap. Generally, the good relationship with God is conceived as some kind
of union between the religious person and God, such that the person somehow partakes or partici-
pates in the divine nature, whether in this world or in some afterlife. Now, if God is conceived as
the Supreme Being, and the good relationship with God is conceived as one in which a person par-
takes of the divine nature, it is reasonable to think that attaining this relationship would be qualita-
tively superior to any other kind of good. Such a good would be vastly better than any other imag-
inable kind of good.
Third Proposition. Third, at least for many persons, it is rational to believe that a religious way of
life is more likely than a nonreligious way of life to result in attaining a good relationship with God.
A religious way of life involves devotional practices such as prayer and meditation, and in general
doing those things that one thinks that God (if he exists) endorses, and avoiding those things that
God (if he exists) forbids. Exactly how one chooses a particular religious way of life is a good ques-
tion. But the most rational thing for a person to do would be to choose the religious way that seems
to make the most sense; on this issue, people will naturally disagree. Still, it is plausible to think that
at least some religious way of life is more likely to lead to a good relationship with God than a
completely nonreligious way of life.
Furthermore, even if one cannot make a clear-cut determination among competing religious
ways about which one is the most likely to result in a good relationship with God, one may simply
make this choice based on other considerations, such as familiarity or family tradition. In fact, this is
what many (modern) religious people do. They think that some religious way of life is more likely to
lead to a good relationship with God, but they concede that other religious ways might also be just
as plausible, especially for other people. Unless they become convinced that their own tradition is
wrong, they adhere to their own tradition. This is actually quite a reasonable procedure to follow.
In sum, for many persons, it is rationally defensible to believe the three propositions stated
above. For such persons, it is rational to pursue a good relationship with God, and it is rational
to have pragmatic faith. Since the value at stake in attaining this relationship is transcendent, it is
assured that the pragmatic value of choosing faith is higher than nonreligious choices. Also, since
the value of attaining a good relationship with God is conceived as finite but transcendent, it will
be relevant to take into account any estimation of what is the most probable way of attaining this
end. It may be conceded that “nonstandard gods” are a real possibility, but, for many persons, the
most plausible conception of a God that can supply a transcendent good is a God who is a

supremely perfect being. Note that this argument does not claim that a full-blown belief in God’s
existence is rational on pragmatic grounds. Thus it avoids the criticism that it endorses a disingenu-
ous or inauthentic form of belief. Finally, this argument is pluralistic in that it claims that religious
persons have good reason to follow some religious way, while still recognizing that there may be
other viable alternatives toward attaining a good relationship with God.
Bibliography
Bartha, Paul, and Lawrence Pasternack, eds. Pascal’s Wager. New York: James, William. “The Will to Believe.” In The Will to Believe and Other
Cambridge University Press, 2018. Essays in Popular Philosophy, 1–31. New York: Longmans, Green,
Cargile, James. “Pascal’s Wager.” Philosophy 41 (1966): 250–257. 1912.
Clifford, W. K. “The Ethics of Belief.” (1877.) In Readings in the Philosophy Jordan, Jeff, ed. Gambling on God: Essays on Pascal’s Wager. Lanham, MD:
of Religion: An Analytic Approach, edited by Baruch A. Brody. Englewood Rowman and Littlefield, 1994.
Cliffs, NJ: Prentice-Hall, 1974.
Jordan, Jeff. Pascal’s Wager: Pragmatic Arguments and Belief in God. Oxford:
Gale, Richard M. The Divided Self of William James. Cambridge, UK: Clarendon; New York: Oxford University Press, 2006.
Martin, Michael. “On Four Critiques of Pascal’s Wager.” Sophia 14, no. 1
Golding, Joshua L. “Toward a Pragmatic Conception of Religious Faith.”
(1975): 1–11.
Faith and Philosophy 7, no. 4 (1990): 486–503.
Morris, Thomas V. “Pascalian Wagering.” Canadian Journal of Philosophy
Golding, Joshua L. “Pascal’s Wager.” Modern Schoolman 71, no. 2 (1994):
16, no. 3 (1986): 437–453.
115–143.
Golding, Joshua L. Rationality and Religious Theism. Aldershot, UK: Ashgate, Pascal, Blaise. Pensées and Other Writings. Translated by Honor Levi.
2003. Oxford: Oxford University Press, 1995.
Hacking, Ian. The Emergence of Probability: A Philosophical Study of Early Rescher, Nicholas. Pascal’s Wager: A Study of Practical Reasoning in
Ideas about Probability, Induction, and Statistical Inference. 2nd ed. Philosophical Theology. Notre Dame, IN: University of Notre Dame
Cambridge, UK: Cambridge University Press, 2006. Press, 1985.
TOPIC 19
Prudential/Pragmatic
Arguments: Atheism
EVIDENCE, PRUDENCE, AND RELIGIOUS BELIEF 645
Richard Feldman
University of Rochester, NY
PASCAL’S WAGER 649

Malcolm Murray
University of Prince Edward Island, Canada
BELIEVING BECAUSE IT PAYS: PRAGMATIC REASONS FOR 654

RELIGIOUS BELIEF
Charles Pigden
Associate Professor, Department of Philosophy
University of Otago, New Zealand
PASCAL’S WAGER REVISITED 658

Evan Fales
Emeritus
University of Iowa
The aim of this chapter is to decide whether, in the absence of adequate evidence of the (probable) truth or
falsity of theism, we may be justified in making a religious commitment on prudential or pragmatic
grounds.
EVIDENCE, PRUDENCE, AND RELIGIOUS BELIEF
Richard Feldman, University of Rochester
This chapter examines the connection between the prudential defense of religious belief and a
prominent epistemological doctrine about rational belief known as evidentialism. The prudential
defense emphasizes the benefits religious beliefs may confer on those who hold them. The idea
is that such beliefs contribute to a better, more satisfying life. In the discussion that follows, it is
assumed that religious beliefs do tend to be beneficial in this way, not because it is obvious that
they impart these benefits but rather because the goal here is to examine the relationship
between the prudential defense and the evidentialist view about the right way to form beliefs.
Evidentialism traces back at least to Socrates, who said that “wherever the argument, like a
wind, tends, there we must go” (Plato, Republic 394d). A particularly influential formulation
645
Prudential/Pragmatic Arguments: Atheism
of the idea is found in William Clifford’s essay “The Ethics

KEY CONCEPTS of Belief.” Clifford writes: “It is wrong always, everywhere,
and for anyone to believe anything on insufficient evidence”
([1877] 1999, 77).
Belief-Forming Strategy
Our question is thus about the implications of evidential-
Doctrine of Salvation by Grace ism for the prudential defense of religious belief. It seems that
Doctrine of Vicarious Atonement evidentialism rules out or declares as wrong beliefs based on
Evidentialism prudential considerations. After clarifying evidentialism, we
consider whether it really does have this implication.
Gnostics
Many-Contenders Objection CLARIFYING EVIDENTIALISM
Pascal’s Wager Several aspects of Clifford’s statement of evidentialism require
Practical Reasoning explanation. First, by “believe” he means accept a statement as
Prudentialism true. Beliefs can include firm convictions as well as cautious
affirmations. It is important to understand that evidentialism
Theoretical Reasoning
is a doctrine about belief, a mental attitude toward a proposi-
Utile tion. It does not say anything about what actions are wrong
(or right). Suppose, for example, that a young child asks you
whether Santa Claus will come to her house for Christmas.
Assuming that you do not have evidence supporting the propo-
sition that Santa Claus will come to her house, evidentialism implies that it would be wrong for you
to believe that he will come. Evidentialism, however, is silent on the question of what you should
say to the child. Evidentialism, strictly speaking, rules only on the status of the belief and not on
related actions, including speech acts that may disappoint children.
Second, by “evidence” Clifford means information indicative of the truth or falsity of a propo-
sition. This information may include things one has observed oneself, things one has learned from
others, and things one has somehow figured out by thinking through an issue (e.g., a mathematical
topic). Notably, it does not typically include facts about what one would like to be true. For exam-
ple, if someone is planning a picnic for the next day, information about weather patterns might pro-
vide evidence about the likelihood of rain on that day. The fact that the person would like it to be
warm and sunny is not evidence supporting the proposition that it will be warm and sunny. (Infor-
mation about what the person wants may be evidentially related to some propositions, for example,
propositions about the person’s character or dispositions.)
Third, by “insufficient evidence,” Clifford means evidence that is not strong enough to warrant
belief. But what is strong enough? Clifford is not especially clear about this, but we can identify a
lower limit on what might count as sufficient. When considering a proposition, a person can believe
it, disbelieve it, or do neither (suspend judgment). Suppose that a person is considering a proposi-
tion and that the person’s evidence suggests that the proposition is false. For example, suppose
our picnic planner’s evidence makes it look more likely than not that it will rain on the day of
the planned picnic. In that case, a person does not have sufficient evidence to believe that it will
be warm and sunny: the evidence goes against it. To take another example, suppose a person is con-
sidering which way a coin that is about to be tossed will land, heads or tails. The evidence is that it
is a fair coin, so the person has no evidence favoring heads or favoring tails. In this case, the evi-
dence is insufficient for believing that it will land heads but also insufficient for believing that it will
not be heads. These examples suggest an intuitively plausible idea: for evidence to be sufficient for
believing a proposition, it must at least better support the truth of the proposition than the falsity
of the proposition. In other words, evidence is sufficient for believing a proposition only if it sup-
ports the truth of the proposition; it is insufficient if it goes against the proposition (as in the picnic
example) or is neutral (as in the coin example).
One could argue for a stronger standard for sufficiency according to which evidence is sufficient
only if it makes the proposition in question very likely to be true. For present purposes, however, we
can set that stronger condition aside and rely on the weaker standard according to which evidence
that is neutral or supports the falsity of a proposition is insufficient evidence to believe it and evi-
dence that supports the truth of a proposition is sufficient.
It is worth highlighting that the standards just described for having sufficient evidence for a
proposition require more than simply having some evidence supporting it. There can be cases in
which one has a mix of evidence, some indicating the truth of a proposition and some indicating
its falsity. To have sufficient evidence to believe a proposition, it must be that one’s evidence, taken
together, adds up in favor of its truth. It is the total evidence that matters.
Next, note that according to Clifford it is wrong to believe on insufficient evidence. Suppose the
evidence a person has really does add up in favor of a proposition, but the person believes it not as a
result of this supporting evidence but rather because it is comforting to believe it. In that case, the per-
son is not believing on (the basis of) sufficient evidence, even if sufficient evidence exists. Therefore,
evidentialism implies that it is wrong to believe in this way. Finally, although the passage from Clif-
ford does not say this, evidentialists typically endorse a corresponding principle according to which
beliefs held on sufficient evidence are not wrong (i.e., are right, or proper, beliefs).
EVIDENTIALISM AND THE PRUDENTIAL DEFENSE

We now have a clearer understanding of evidentialism. Why might it undermine the prudential
defense of religious belief? It is not that evidentialism implies that religious belief is not true or that
it implies that there is not sufficient evidence for religious belief. Evidentialism itself is silent on
these issues. The way in which evidentialism undermines the prudential defense is via an additional
premise that is highly plausible. This additional premise is that the prudential considerations given
in support of religious belief do not constitute sufficient evidence for the truth of those beliefs.
We can accept for the sake of argument that it is in one’s interest to hold religious beliefs, that such
beliefs make one’s life, or one’s afterlife, better. The fact that the beliefs have these benefits, how-
ever, is not evidence that they are true. In general, the fact that beliefs have beneficial effects is
not evidence for their truth. For example, if a person is waiting for a bus on a cold day, believing
that the bus will arrive soon may make the wait more tolerable. Still, that in no way supports the
truth of the proposition that the bus will arrive soon.
There are also beliefs that are in a certain way self-fulfilling, in that having the belief makes it
more likely to be true. For example, in some cases, people’s believing that they will succeed may
make them more likely to succeed. If one is aware of this, then having the belief can add to one’s
evidence for its truth. Of course, at most this would mean that the fact of having the belief is some
reason to think it is true. It does not show that the belief is on balance supported by the evidence.
In any case, religious belief is not like that. It is not evidence for its own truth.
Thus, if evidentialism is correct, then holding religious beliefs on the grounds cited by the pru-
dential defense is wrong. Given the plausibility of evidentialism, this seems to undermine the pru-
dential defense.
DEFENDING THE PRUDENTIAL DEFENSE

Defenders of the prudential defense might respond in several different ways. This section reviews
three possible responses.
First Response: Reject Evidentialism. Evidentialism is a purist doctrine. It holds that the wrong-
ness of a belief is entirely a matter of evidence. This view has widespread support as a descendant of
the compelling Socratic idea of following the argument wherever it leads. Still, one can resist. What
follows describes one line of resistance.
We often say that certain actions, often those that harm others, are wrong. Sometimes we add a
qualifier, saying that such actions are morally wrong. Where other people are not involved, we some-
times say of people who make choices that are not in their own interest that their behavior is impru-
dent or prudentially wrong. It is this last concept that is particularly significant here. One’s beliefs
can have a major impact on the quality of one’s life. If, for example, the only way to save one’s life
is to jump from the roof of a burning building onto the next roof, it is likely to be in one’s interest

to believe that it is possible to jump far enough. The evidence about one’s jumping ability seems
irrelevant.
Consider, then, whether it is wrong to believe something when (1) sufficient evidence does not
exist for its truth, but (2) it is in one’s interest (and harmless to others) to believe it. Evidentialists
say that the evidence alone determines what is wrong (or right) to believe. Even so, advocates of
the prudential defense might argue that the prudential and evidential considerations must be
weighed against one another: strong prudential considerations in favor of a belief can make it not
wrong to hold the belief even if the evidence does not support it. They can claim that this is exactly
what happens in the case of religious belief.
This view about proper belief formation may reflect an appealing kind of moderateness and
inclusiveness, and some philosophers do endorse it. However, it must be noted that it fails to do
justice to the insight about the centrality of evidence that prompts evidentialism. This can be seen
by considering a simple example. Suppose that two people see a reliable weather report for the next
day saying that rain is quite likely. One of them believes that it will rain the next day. The other,
who is planning a picnic, does not believe this. When they discuss the topic, their difference
emerges. When pressed to explain why the picnic planner believes that it will not rain, in spite of
the forecast, the planner says that there is nothing wrong with so believing, given that it imparts
positive feelings. (Assume that the belief does not have other harmful consequences.) Something
seems clearly amiss in this story. It is understandable that someone might believe contrary to the
evidence in a case like this. Doing so does not make that person bad, but there is something plainly
unreasonable about it. That is because something seems clearly right about the evidentialist insight
into the rationality of following the evidence.
Second Response: Different Senses of Wrong. A second response to the evidentialist objection
to the prudential defense acknowledges the evidentialist insight but also accommodates the ideas
motivating the first response. The starting point for this response is the fact that we can evaluate
things along many dimensions. Suppose a family is looking to buy a house. They have numerous
preferences regarding the number and size of rooms and other features important to them. They
are looking through a magazine containing pictures and descriptions of houses and find one that
has just the features they want. They say that the one depicted is a “good house.” They go on to
wonder whether there are houses similar to that one in the city they will live in, since the one in
the photo is thousands of miles away. When thinking about location, they say that the house
depicted is “not good.” Both of these judgments can be true—the house is good (for them) in terms
of features but not good (for them) in terms of location.
Something similar can be true about beliefs. We can examine beliefs from the perspective of
rationality and evidence. A belief supported by one’s evidence is a good one from this perspective.
A belief not supported by one’s evidence is wrong from this perspective. The word epistemic often
is used to describe this perspective, and thus we can say that evidentially unsupported beliefs are epi-
stemically wrong. At the same time, we can look at beliefs along prudential lines. It can be that hav-
ing a certain belief is beneficial (or not), and thus we can evaluate beliefs as prudentially right (or
wrong). The prudential evaluations and the epistemic evaluations need not coincide. One belief
might be epistemically wrong but prudentially right.
An advocate of the prudential defense can apply this thinking to religious belief, acknowledg-
ing, perhaps, that such beliefs are epistemically wrong (as the evidentialist says) but contending that
they are not prudentially wrong. This response has considerable merit. It is true that beliefs without
evidential support can be beneficial, and it is surely possible to evaluate such beliefs from that per-
spective. However, this response might disappoint some advocates of the prudential defense,
namely, those who want to say that somehow, in this case, the prudential considerations trump
the evidential considerations and make religious beliefs simply not wrong. For these defenders, it
is not enough to say that they are prudentially but not epistemically right.
Third Response: Acceptance Rather Than Belief. A final response concedes that the evidential-
ist objection to the prudential defense of religious belief succeeds but argues that there is another
attitude, called acceptance, that is something like belief. Furthermore, acceptance of religious propo-
sitions is immune to evidentialist objections.
It is sometimes the case that we regard a proposition as true for a certain purpose even if we do
not actually believe it. For example, early in this chapter it said that “it is assumed that religious
beliefs do tend to be beneficial.” Readers were not being asked to believe that this is true. Instead,
they were being asked to accept it as true for the purposes of thinking through the issues under dis-
cussion.
We can accept propositions for a variety of reasons. Accepting something for the sake of a par-
ticular discussion is a familiar instance. We might also accept something because doing so helps sim-
plify and guide action. For example, scientists are sometimes said to accept certain theories because
the use of a theory to shape one’s research can be extremely helpful. Without endorsing the theory
in some way, it might be difficult to motivate oneself to do one’s work or to think through the
issues effectively. One might not actually believe that the theory is, in fact, true. Acceptance is thus
a kind of voluntary endorsement of a proposition and often is associated with acting in ways like
those that one who believed the proposition would act, whereas belief is typically an involuntary
response.
A supporter of the prudential defense can concede that the evidentialist objection shows that
religious belief on prudential grounds is (epistemically) wrong but maintain that accepting religious
propositions on prudential grounds is not wrong. If accepting religious propositions organizes and
sustains one’s life—if acting as if they are true is helpful—then a lack of evidential support need
not count against the merits of accepting them. Since evidentialism does not have any implications
regarding acceptance, it does not undermine this version of the prudential defense.
Whether this way of spelling out the prudential defense is successful remains an open question.
One question is whether it constitutes the kind of robust support for religious attitudes that its
advocates seek. Acceptance can be a kind of practical tool that falls short of any kind of genuine
endorsement. Consider, for example, the request to accept for the purposes of this discussion the
proposition that religious belief is beneficial. Even people who strongly disbelieve this proposition
can nevertheless play along for the sake of discussion. That hardly seems to be the status proponents
of the prudential defense seek for religious belief.
Furthermore, it may be a mistake to think of acceptance as a belief-like attitude at all. Although
we say that people accept things for the sake of argument, what is most clearly true is that they
examine the consequences of certain propositions. When readers were asked to assume for the sake
of discussion the proposition that religious belief is beneficial, it is not clear that they were being
asked to do anything more than examine the consequences of that proposition in light of evidenti-
alism and other propositions concerning religious belief. That also seems to fall short of what pru-
dential defenders seek.
These considerations leave us with a question about whether the appeal to acceptance actually
provides the kind of support that advocates of the prudential defense are seeking. It might be not
wrong to accept religious propositions for the sake of argument or to facilitate thinking through
their consequences or to encourage related behavior. Supporters of the prudential defense may still
hope for something more than this in support of religious propositions.
PASCAL’S WAGER
Malcolm Murray, University of Prince Edward Island
One November night in 1654, a religious experience convinced Blaise Pascal that God alone—not
philosophy, not science—provided certainty, so he abandoned science and philosophy in pursuit of
religion. In his Pensées, he offers the argument we now know as Pascal’s wager ([1670] 1941).
Pascal’s wager shows that it is a better bet to believe in God than not to believe in God. We
would not need such a bet if we could know God through reason or revelation. In De l’esprit géomé-
trique (Of the Geometrical Spirit, [1657/1658] 1980), Pascal shows why reason falls short of

certainty. All arguments rest on some proposition the truth of which we cannot establish without
presupposing a further ungrounded proposition. Absent revelation, then, atheists need another
means to believe: hence Pascal’s wager.
THREE VERSIONS
According to Ian Hacking (1972), Pascal offers us three versions of the wager. In its simplest form,
if we believe God exists and God does exist, then we gain salvation. If we believe God exists and
God does not exist, we lose nothing. If we do not believe in God, yet God exists, we are damned.
If we do not believe in God and God does not exist, we break even. Because salvation beats damna-
tion and the other alternatives are equal, believing in God is a better bet than not believing in God.
The second version allows that a secular life may have more merit than a religious life absent any
eschatology. Not everyone believes this. Perhaps the cost of theism is positive even if God does
not exist. William James (1970), for example, thinks religious belief may provide meaning in one’s
life even without any god. Even still, because eternal reward in heaven far surpasses eternal damna-
tion in hell, any difference between secular and religious life would make no difference.
So far, Pascal’s argument would tell us to buy a lottery ticket. Because the payoff of a lottery
is so high (say $100) and we lose next to nothing in buying a ticket (say $2), we should buy it
(see Table 1).
Table 1. Lottery without Probabilities
Buy Do Not Buy
Win $100 $0
Do Not Win −$2 $0
Sum $98 $0
What has gone wrong, obviously, is the failure to consider probabilities. If the odds of our win-
ning the lottery were one in a thousand, reason would tell us not to take the bet. Table 2 captures
the expected utility of buying that lottery ticket.
Table 2. Lottery with Probabilities
Buy Do Not Buy
Win 98 (0.0001) 0 (0.0001)
Do Not Win −2 (0.9999) 0 (0.9999)
Sum −$1.99 $0
Because $0 is better than -$1.99, we should not buy the lottery ticket. To apply probability to
Pascal’s wager, if we assume a 10 percent chance that God exists, 100 utiles for heaven, and 15
utiles for secular life, then nonbelief would be the better bet (see Table 3).
Table 3. Pascal’s Wager with Probabilities
Believe Do Not Believe
God Exists 100 (0.1) −100 (0.1)
God Does Not Exist −15 (0.9) 15 (0.9)
Sum −3.5 3.5
Both the first and second versions assume that the probability of God’s existence is 50 percent.
Because “God is, or He is not,” Pascal seemed to think that “there is an equal risk of winning and of
losing,” just as a coin will turn up either “heads or tails” ([1670] 1941, 81). This is implausible. I
cannot assume that the odds of my winning a lottery are 50-50 simply because my lottery ticket
either wins or loses. Later, Pascal claims that reason cannot defend God’s existence or nonexistence.
This suggests that the 50 percent odds of God’s existing are not derived merely from the binary
nature of existence but instead on an epistemic standoff. In any event, Pascal rectifies the implausi-
ble 50 percent odds given to God’s existence in the third version. In the third version, the utility of
heaven is given an infinite value, as is the negative utility of hell. If we can assume that utilities can
be given an infinite value—a question to which I shall return—then no matter what probability we
give to God’s existence—assuming only that it is greater than zero—a fraction of infinity is still infi-
nite. Belief in God wins.
We can represent the three versions of Pascal’s wager in tabular form. In the first version, let us
assign successful entrance to heaven +100, and hell -100 (see Table 4).
Table 4. Basic Version
God Exists 100 −100
God Does Not Exist 0 0
Sum 100 −100
For the second version, assume the secular inconvenience of belief equals -10, whereas the
earthly benefit of nonbelief equals +10 (see Table 5).
Table 5. Secular Life Version
God Exists 100 −100
God Does Not Exist −10 10
Sum 90 −90
For the third version, we change a finite utility for heaven and hell to an infinite utility and
assign a probability of 10 percent to God’s existence (see Table 6).
Table 6. Infinity Version
God Exists +∞ (0.1) −∞ (0.9)
God Does Not Exist −10 (0.9) 10 (0.1)
Sum +∞ −∞
No matter the variations, each table shows that belief is more reasonable than disbelief. Even a
self-interested atheist ought to see its logic—assuming that heaven is heavenly for atheists, that
heaven comes only to those who believe in God, that mere belief is sufficient for access to heaven,
that we believe in the right god, that we have the values right, that we can assign an infinite utility
to anything, and that we believe the probability of the existence of God and heaven is greater than
zero.
As an aside, I am doubtful that Pascal conceived the wager as three distinct arguments as
opposed to one argument with two responses to potential objections. He never credits secular life
ahead of religious life, and he never obviously modifies his notion that the probability of God’s exis-
tence may be deemed less than 50 percent. All he says is that disagreement on these points would
not matter.
We could even modify the argument in one further respect. For those who think a good god
would not permit hell (e.g., Hick 1966), we can remove hell without upsetting the argument. As
Richard Swinburne (1981) maintains, the risk of not getting to heaven ought to be motivating
enough.
SELF-INTEREST AND HABITS

Many theists reject Pascal’s wager for offering too secular a reason for believing in God. The impe-
tus for accepting the bet is selfish gain, which is counterproductive to forming a relation with God.
Such an attitude is unfortunate on two grounds. Many think that acting on self-interested desires is
simply what it means to be a reasonable agent, religious or not (e.g., Stark and Finke 2000, 36–41;

Schlesinger 1994). The opposite of acting on self-interest would be to endorse a kind of irrational-
ity, as if, with Søren Kierkegaard (2006), the religious are proudly irrational. Ignoring that, Pascal
clarifies that he offers merely a foot-in-the-door approach for nonbelievers. Appealing to their self-
interest would be the only way to get atheists to start to believe; once started, true belief would
develop. “By acting as if [one] believed, taking the holy water, having masses said, etc. … will nat-
urally make you believe, and deaden your acuteness” (Pascal [1670] 1941, 83). In other words,
proper habit inculcates correct attitude.
Analogously, consider a boy who picks up a guitar to impress girls. In time, with proper prac-
tice, the boy comes to love the guitar for its own sake and not for getting girls. If nothing else had
prompted him to take up the guitar, the ignominious beginning may have been worth it. Aristotle
(1941) advocated this approach to ethics. If we do as the virtuous do, we will become virtuous in
time. Neither Pascal nor Aristotle confused doing with being.
The habit-forming part of Pascalian belief admits the possibility of latching onto wrong habits,
however. As David Hume observes, “Many of the votaries … seek the divine favour, not by virtue
and good morals, which alone can be acceptable to a perfect being, but … by frivolous observances
… and absurd opinions” (Hume 1956, 70). For Swinburne, “You can only get to Heaven if you do
certain actions and it is impossible in practice to do those actions unless you believe” (1981, 93).
Until we settle the odds of stumbling onto the right practice, Pascal’s wager cannot motivate us.
RATIONALITY
Pascal offers a fideist approach to belief in God. “We know truth not only by the reason, but also by the
heart, and it is in this last way that we know first principles; and reason, which has no part in it, tries in
vain to impugn them” (Pascal [1670] 1941, 95). Swinburne is dubious. One needs to apply reasoning
skills to accept Pascal’s wager. If reason tells us to accept the bet, then abandoning reasoning skills will
tell us not to accept the bet (Swinburne 1981, 96). Acting without reason is a kind of insanity.
Swinburne overstates matters. When I commit to do sit-ups every morning, I also commit to
not thinking about whether I really have to do sit-ups every morning. Rationally thinking about
whether I really have to do sit-ups in the mornings when I am still sleepy will likely undermine
my resolution to do sit-ups. My intentionally abandoning my rationality when approaching sit-
ups, however, does not mean my doing sit-ups is as predictable as running naked through the streets
reciting poetry. No insanity gap exists between rationally recognizing the need for commitment and
commitment.
For Swinburne, though, “it is rather unlikely that God has set up a world in which both God
rewards beliefs that he exists highly and the only way to acquire it is on Pascalian-type grounds”
(Swinburne 1981, 96). Pointing out Pascal’s audience ought to soften the arationality critique.
Assume most people come to believe in God in the traditional way. God made the world for these
people. That still leaves out those others who fail to form a belief in God. Because Pascal speaks
only to those unfortunates, Pascal would deny Swinburne’s suggestion that wagering is the only
way to know God.
RELIGIOUS DIVERSITY
In The Natural History of Religion (1956), Hume points out the problem about multiple religions.
Assume that only one true god exists among seventeen false gods. A bettor has a 6 percent chance
of betting on the right god. If we have a 94 percent chance of believing in the wrong god, Pascal
exaggerates the promise of heaven. Antony Flew’s (1966) worry is similar. Perhaps the true God lets
atheists into heaven and burns theists in hell (Flew 1966, 9). Pascal has first to show that his god is
more likely the true God than all other gods combined (Swinburne 1981, 94–95). Because Pascal
thinks reason alone cannot suffice, his only way of persuading aspirant believers to believe in his
god is to get them to make the leap and see what happens. Because every other religion can make
the same appeal, Pascal can persuade no aspirant believer to wager (see Jordan 1994; Mackie
1982; Martin 1983). Matters are worse. The wager ought to apply for each god who promises an
infinitely wonderful heaven. Rational bargainers ought to believe in every god, just to hedge their
bets. Assuming we gain heaven through exclusive belief only, Pascal’s wager prevents access to
heaven. Some people, like Baháis, believe only one god exists but one who manifests himself in
countless ways. If so, the many gods objection causes no trouble.
INFINITY
Pascal refused to give heaven any finite value. As with Anselm (1962), heaven always deserves a
higher utility than mortals can conceive. For Pascal, heaven provides “an infinity of an infinitely
happy life” (Pascal [1670] 1941, 82). If we give heaven an infinite positive number, then no prob-
ability we assign greater than zero will matter. A fraction of infinity is still infinite. This makes the
appeal to infinite utility seem a bit too facile. While Edward McClennen (1994) rejects the concept
of infinite value because it prevents our ability to maximize expected utility, Anthony Duff (1986)
and Alan Hájek (2003) use infinite utility against Pascal. Pascal offers two pure strategies: believe
or disbelieve. Countless mixed strategies are possible, such as believe if a coin lands heads; disbelieve
otherwise. This mixed strategy would give you all that Pascal promised, because infinite happiness
multiplied by 0.5 (the odds of heads) is still infinite. Choosing to believe only if struck by lightning
would also reap the same reward. Pascal’s trickery backfires.
Paul Bartha (2007) defends infinite utility against the Duff/Hájek critique by moving from
direct representation of preferences with a utility value to indirect representation with utility ratios.
Whether Bartha’s move successfully handles infinitesimal probabilities, incorporating infinity into
the wager calculations succumbs to the same complaint Immanuel Kant (1969) raised against the
ontological argument. Agreeing that heaven, if it exists, is infinitely great cannot rule out the possi-
bility that no heaven exists. In the Christian tradition, belief in God and belief in heaven are a pack-
age deal. Pascal’s argument for belief in God requires the presupposition that heaven be infinitely
great. To believe such metaphysics presumes supernatural agency, that is, God. We have gone in
a loop. Pascal’s wager was supposed to lead us to believe in God, not presuppose our belief in
God. The insertion of a probability-defeating infinity into our values simply begs the question.
NONZERO PROBABILITY
As Michael Rota (2016a) maintains, Pascal’s wager works for those who believe God’s existence is as
likely as not. Similarly, lotteries are good bets for those who think they have a 50 percent chance of
winning. Whether anyone ought to think so is another matter. Still, perhaps Pascal is talking only
to agnostics, not atheists—at least agnostics in the sense of believing that the evidence for God’s
existence is 50-50—as opposed to agnostics, who, contra gnostics, believe God is beyond compre-
hension (e.g., Kenny 2004). Concerning God’s existence, Pascal says that “the evidence is such that
it surpasses, or at least equals, the evidence to the contrary” (Pascal [1670] 1941, 185).
That the evidence in favor of God is equal to the evidence against God is either false or mis-
leading. On the misleading side, I might confess that I have no evidence for God’s existence and
no evidence for God’s nonexistence, but only in the way that I have no evidence that a magical
monster either exists or does not exist under my bed. By magical, I mean that kind of detection-
avoiding monster whose presence or absence makes no empirical difference. In such cases, however,
no one assumes the evidence for and against the monster is 50-50. The burden of proof lies with
the claim that a magical thing exists. Failing sound evidence, nonbelief, the default, is justified
(see Murray 2010, 23–28).
On the false side, Pascal supposes that a posteriori arguments for God’s existence are 50 percent
convincing. Whether agnostics believe that, they ought not. Concerning the design argument, evolu-
tion accounts for the appearance of design better than the design theory. A design model cannot well
explain the fact that 99 percent of known species are extinct, whereas that is what natural selection
predicts (see Dawkins 1986). With respect to the fine-tuning argument, that the odds of the universe
coming from chance are astronomically small does not show that the odds of the universe coming
from a supernatural agent are equal, let alone higher (see Sober 2005). We would first need to com-
pute the probability of a god’s existing. We cannot calculate that as 1 minus whatever the probability
of the universe coming about by chance. For example, if the odds of the universe coming about by
chance are 1/1055, the odds of God creating the world are not 1−(1/1055)! That would make it about
100 percent probable. Concerning the cosmological argument, if we start with the belief in

supernatural agents who can create universes, then we can certainly believe that a god’s creating our
universe is more probable than that our universe came about by chance. If, however, we start with
the empirical observation that our universe exists and try to infer a god from that (the proper direction
of a posteriori arguments), things are less rosy for theists. We have prior experience of naturally occur-
ring events. We have no prior experience of gods making universes (see Hume 1985).
Once we assign an infinite utility to heaven, any nonzero probability that the right God exists
will not matter. Apart from rejecting the concept of an infinite value, an atheist can reject the ascrip-
tion of a nonzero probability to God’s existence. For one, all atheists will assign a zero probability to
God’s existence if we mean God is a necessarily existent being. The mere probability of a necessarily
existent being guarantees it exists. For Pascal’s wager, however, the problem concerns the difference
between possibility and probability. Probability is a utility function; possibility is an admission of
fallibilism. When atheists admit that a god is possible, they admit that, despite all evidence available
to them, they may be wrong. To assign a probability to a god, however, requires more than an
admission of fallibility. Assigning probability requires either past experience or logical permutation.
In the first sense, assigning an 80 percent chance of getting lung cancer from smoking cigarettes is
derived from witnessing past cases of smokers and nonsmokers. In the second sense, when we assign
a 17 percent chance of getting a three on a die, we recognize that there is only one three and six
sides. To assign a probability greater than zero to a god, then, we need to have either experienced
gods before, or we can calculate the probability from mathematical possibility. Since neither is possible,
assigning a probability greater than zero to a god is unmerited.
BELIEVING BECAUSE IT PAYS: PRAGMATIC REASONS FOR RELIGIOUS BELIEF
Charles Pigden, University of Otago
In the late tenth century, according to Snorri Sturluson’s Heimskringla, the king of Norway, Olaf
Tryggvason, was cruising off the coast of Scotland with his battle fleet when he chanced upon the
much smaller fleet of Sigurd the Stout, the earl of Orkney:
When the king was informed that the earl was there, he made him be called; and when the earl
came on board to speak with the king, after a few words only had passed between them, the king
says the earl must allow himself to be baptized, and all the people of the country also, or he
should be put to death directly; and he assured the earl he would lay waste the islands with fire
and sword, if the people did not adopt Christianity. In the position the earl found himself, he
preferred becoming Christian, and he and all who were with him were baptized. (1906, 175)
King Olaf had just given Earl Sigurd a prudential reason for adopting a rather elaborate set of beliefs (a
reason that the earl seems to have found convincing), though he gave him neither an argument nor any
evidence for any of the propositions believed. (Of course, there is a lot more to becoming a Christian—
or a Muslim or a Buddhist, etc.—than adopting a set of beliefs, but that is a big part of it. Hence the
widespread emphasis on creeds, shahadas, and professions of faith.) Things would have been different
if one of King Olaf’s attendant priests had worked some spectacular miracle (on the assumption that
only the representatives of a powerful god could do such a thing) or if Olaf had supplied Sigurd with
some brilliant apologia for Christianity by one of the church fathers. But in fact there was neither evi-
dence nor argument: it was a case of “believe it or else!” And Sigurd, it seems, decided to believe.
There are other well-authenticated cases of incentivized belief that are rather less brutal. Why is
it that nowadays most of the inhabitants of what were once the Umayyad and Abbasid caliphates are
believing Muslims? Well, after the Arab conquests had subdued both the Persian Empire and
the greater part of the Empire of Justinian, the caliphs established an Islamic polity. But Islam
was not, in the main, a persecuting religion. The “People of the Book”—Jews, Christians, and
Zoroastrians—were allowed to practice their religions undisturbed and even to serve in the govern-
ment, but they were taxed at a higher rate than Muslims. Thus the reason that most of the modern
inhabitants of North Africa and the Middle East are believing Muslims is that many of their ances-
tors converted to get a tax break.
Now one response is to deny that King Olaf gave Sigurd a reason to believe in Christianity: he
only gave him a reason for pretending to believe or for behaving like a believer. Likewise those Mus-
lim “converts” who were trying to minimize their tax obligations. Some philosophers back this up
with the claim that, since belief is not under the control of the will, it is impossible to give some-
body a prudential reason to believe anything. It is impossible to adopt the belief that Jesus Christ
is your personal savior or that there is no God but God and Muhammad is his prophet because it pays
you to do so, since that would involve choosing to believe, and belief is not under the control of the
will. And you cannot have a prudential reason for doing what it is impossible for you to do. Even if
I am surrounded by ravening wolves who are about to tear me limb from limb, I do not have a pru-
dential reason for transforming myself into a bird and flying away since (not being a shape-shifter)
this is not something that I am able to do. Thus something like the ought-implies-can principle in
its contrapositive form, not-can-implies-not-ought, is supposed to apply to prudential reasons, ruling
our prudential reasons for adopting religious (or any other) beliefs.
Here is a more pedantic reformulation of the argument.
Argument A: The Impossibility of Prudential Belief

(1) Necessarily nobody has a prudential reason to do what it is impossible for them to do.
(Ought-implies-can applies to prudential reasons excluding reasons to do the
impossible.) [Premise.]
(2) In order for it to be possible to adopt a set of genuine beliefs for prudential reasons,
beliefs would have to be (sometimes) under the direct control of the will; it would have
to be possible (sometimes) to choose to believe. [Premise.]
(3) Genuine beliefs are never under the control of the will; it is never possible to choose to
believe. [Premise.]
(4) Therefore it is impossible to adopt a set of genuine beliefs for prudential reasons. [From
2 and 3.]
(5) Therefore it is impossible to have a prudential reason to adopt a set of genuine beliefs.
[From 1 and 4, resting on 1, 2, and 3.]
From this it follows that King Olaf cannot have given Earl Sigurd a prudential reason to convert to
Christianity, nor can the caliphs have given their infidel subjects a rational incentive to convert to
Islam, at least not if conversion implies a genuine belief in the Creed, or the Shahada. In both cases
they can only have given the supposed converts a prudential reason to fake belief.
This argument carries an interesting corollary. Campaigns of persecution on the part of kings and
princes are necessarily ineffective if their object is to save souls by inducing genuine belief. They are
therefore illegitimate. This is the basis of one of John Locke’s key arguments against persecution in his
famous “Letter concerning Toleration.” In an unusually succinct formulation, Locke puts the point like
this:
The care of souls cannot belong to the civil magistrate, because his power consists only in
outward force: but true and saving religion consists in the inward persuasion of the mind,
without which nothing can be acceptable to God. And such is the nature of the
understanding, that it cannot be compelled to the belief of any thing by outward force. (Locke
[1689] 1991, 18)
Following Jeremy Waldron (1991), we can develop this argument as a continuation of Argument A:
Argument B: The Impossibility of Legitimate Coercion

(6) Governments can only have a legitimate reason to do what it is possible for them to do.
[Premise.]
(7) Governments can only induce their subjects to do things via coercion or material
incentives if the coercion or the material incentives supply those subjects with
prudential reasons for doing those things. [Premise.]

(8) Necessarily, salvation is only possible on condition that people adopt genuine
(nonfake) beliefs in more or less the right God or Gods. [Premise.]
(9) Therefore governments can only save the souls of their subjects via coercion or the use
of material incentives if coercion or material incentives can induce them to adopt
genuine (nonfake) beliefs in more or less the right God or Gods. [From 8.]
(10) Governments can only save the souls of their subjects via coercion or the use of
material incentives if persecution or material incentives can give their subjects
prudential reasons to adopt genuine beliefs. [From 7 and 9, resting on 8.]
(11) Governments cannot give their subjects prudential reasons to adopt genuine beliefs,
since it is impossible to have a prudential reason to adopt a set of beliefs. [From 5,
resting on 1, 2, & 3]
(12) Therefore it is impossible to save the souls of their subjects via coercion or the use of
material incentives. [From 10, resting on 7 and 8, and 11, resting on 1, 2, and 3]
(13) Therefore no government has a legitimate reason to save the souls of its subjects via
coercion or the use of material incentives. [From 12, resting on 1, 2, 3, 7, and 8 and
from 6, which is itself a premise.]
Thus we have valid (and nonridiculous) arguments for two theses: (5) that it is impossible to have a
prudential reason to adopt a set of genuine beliefs; and (13) that no government has a legitimate
reason to save the souls of its subjects via coercion or the use of material incentives. However,
although the first argument might survive without the second, the second cannot survive without
the first, which means unfortunately that it cannot survive (and thus that toleration must be
defended by other arguments). For two of the key premises of the first argument are false.
I concede premise (1), but premise (2) is false. It is not the case that for it to be possible to
believe a set of propositions X for prudential reasons, the belief-forming faculty must be under
the direct control of the will, so that it is possible to choose to believe the propositions X. For it is
often possible to (choose to) adopt a belief-forming strategy that will gradually induce the relevant
beliefs. Thus it is possible, for prudential reasons, to acquire a perfectly genuine set of beliefs even
though the belief-forming faculty is not under the direct control of the will and the propositions
X are not such that one can simply choose to believe them. Locke’s contemporary opponent Jonas
Proast makes this point, arguing that the judicious use of mild persecution can induce heretics to
adopt a belief-forming strategy that will make it more likely that they will become convinced of
the correct view by encouraging them to pay close attention to the relevant arguments (Waldron
1991, 119). The point also comes up in connection with Pascal’s wager, an argument that we con-
sider below. Pascal believes himself to have supplied a strong prudential reason for his opponent (an
atheistically inclined French aristocrat with a taste for worldly pleasures) to adopt a belief in Catho-
lic Christianity. The opponent accepts Pascal’s argument, but he can’t quite bring himself to
believe. “I am so made that I cannot believe. What then would you have me do?” (Pascal 1901,
98). Pascal accepts that, in this instance, at any rate, this is not a belief under the direct control
of his opponent’s will and hence not something that he can chose to believe.
But what Pascal’s opponent can choose to do is to adopt a belief-forming strategy that is likely to
bring about the requisite belief. “Follow the way by which [others] began, by making believe that
they believed, taking the holy water, having masses said, etc. Thus you will naturally be brought to
believe, and will lose your acuteness” (Pascal 1901, 99; emphasis added). In other words, if you act
like a believer, the chances are that you will gradually come to believe. This is probably what hap-
pened to Sigurd and his followers. It is unlikely that they all became sincere, believing Christians
at one fell blow (so to speak). But Olaf had forced them to act like believing Christians and had
no doubt supplied priests who forced them to partake in Christian rituals and who filled their heads
every Sunday with Christian stories (which to begin with they probably thought of as just stories),
so that gradually, like Pascal’s opponent, they “lost their acuteness” and the folk of the Orkneys
became sincere Christians. And the same goes, mutatis mutandis, for the tax-incentivized converts
to Islam.
However, it not just Premise (2) of the argument that is mistaken: Premise (3) is false too. It
states that genuine beliefs are never under the control of the will and that it is never possible to
choose to believe. Proponents of this thesis sometimes use a questionable tactic. They select some
arbitrary proposition Y and challenge you to believe it. When you can’t, they declare victory (Mus-
grave 2009, 12). But the negation of the claim that it is never possible to choose to believe is not
that it is always possible to choose to believe but that it is sometimes possible to choose to believe.
And the fact that it may be impossible to choose to believe some arbitrary proposition Y does not
mean that, for many of us, there are not some propositions Z that we can choose to believe. Indeed,
William James (who develops a pragmatic argument that “our passional nature not only lawfully
may, but must, decide an option between propositions, whenever it is a genuine option that cannot
by its nature be decided on intellectual grounds” (James 1897, 11), that is, that when the evidence
is insufficient it is sometimes rational and even moral to choose to believe a religious proposition
because believing it makes life go with a swing) takes it for granted that, sometimes, at least, choos-
ing to believe is perfectly possible:
A living option is one in which both hypotheses are live ones. If I say to you: “Be a theosophist
or be a Mohammedan,” it is probably a dead option, because for you neither hypothesis is
likely to be alive. But if I say: “Be an agnostic or be a Christian,” it is otherwise: trained as you
are, [and James was addressing an audience of New Haven philosophers in 1896] each
hypothesis makes some appeal, however small, to your belief. (James 1897, 3)
In other words, for any given individual there are some things that she can choose to believe and
other things that she can’t, but for most of us there are some hypotheses that are “live,” that is, such
that we can indeed believe them if we choose to do so. (See Coady 2012.) Once we admit this (and
if it were really impossible to choose to believe, wishful thinking would not be so widely condemned
as an intellectual vice) there is no bar whatsoever to prudential or pragmatic reasons for belief, and if
there is no bar to such reasons there is no bar to arguments that conclude with such reasons. (For
robust refutations of the view that we cannot choose to believe, see Coady 2012, 12–17; and Turri,
Rose, and Buckwalter 2018.)
But the reader may be getting restive. Sure, Olaf could provide Sigurd with a prudential reason
to believe, and the caliphs could do likewise with their financially motivated subjects. But were they
good reasons? What I am suggesting is that there is no absolute answer to this question. A consider-
ation Z (a fact, a threat, or an item of information) constitutes a reason for an agent B to do A
(where A could involve adopting a belief) if-and-only-if B became aware of Z, and if B conformed
to some norm of rationality R, then B would have some motivation to do A. Thus whether or not
a consideration constitutes a reason for action is a norm-relative affair: one and the same fact could
constitute a reason for action according to one norm of rationality R but not according to another
norm R*. A consideration constitutes a good reason for an agent B to do A (according to some norm
of rationality R) if-and-only-if B conformed to some norm of rationality R, then B would be strongly
motivated to do A. And there is no norm of rationality that is absolutely correct. Thus whether or
not Olaf’s threats or the caliphs’ incentives constituted good reasons to adopt the requisite beliefs
depends on the norms in question. If we choose a norm that emphasizes truth or the avoidance
of error, then the answer is probably no. If we choose a norm that emphasizes survival or material
self-interest, then the answer is probably yes. In other words there can be good self-interested rea-
sons to believe what is false (or probably false) even though these would be nonreasons (or even rea-
sons to disbelieve) given a norm that makes truth or the avoidance of error the sole considerations.
(What about a norm that mixed them up? That would depend on the mix.)
Thus far I have been discussing prudential or pragmatic reasons to believe that depend on the
contingent threats and promises of terroristic kings and taxing caliphs. Could there be prudential or
pragmatic reasons to believe that hold in the absence of such contingencies? There are at least two
philosophers who have thought so, Blaise Pascal and James. James would take us too far afield, so
I conclude with a brief discussion of Pascal’s wager.
Pascal imagines an opponent addicted to games of chance. The opponent has two alternatives:
to carry on in his worldly and atheistic way (gambling and having love affairs with beautiful mar-
quises and countesses) or to choose to believe (or to bring himself to believe) in Catholicism, which

entails a radical reform of his way of life. If God exists, the first entails a finite gain and an infinite
loss (an eternity of damnation) while the second entails a finite loss (no more sex with beautiful
marquises plus a lot of tedious church services) but an infinite gain (an eternity of heavenly bliss).
If God does not exist, then the first entails a finite gain and the second entails a finite loss (though
Pascal tries to reassure his opponent that living a godly life isn’t that bad). Thus the expected utility
of choosing not to believe is a finite gain minus an infinite loss multiplied by the probability of
God’s existence. Hence, the expected utility of choosing not to believe is infinitely bad no matter
how low the probability of God’s existence, provided that it is less than zero. And the expected utility
of choosing to believe is a finite loss, plus an infinite gain multiplied by the probability of God’s
existence—that is, an infinitely good expected utility no matter how low the probability of God’s exis-
tence. Thus, to a rational gambler, choosing to believe in God is the best bet since it has the highest
expected utility (see Pascal 1901, 97–99).
What is wrong with this argument is that it assumes just two alternatives: the existence of a
God who will reward the good Catholic with an eternity of bliss or no God whatsoever. What about
a God who has got it in for gamblers who are convinced by Pascal’s wager? Or a God who rewards
those with the guts to carry on with the beautiful marquises? Or a Protestant God who condemns
Catholics to Hell as idolaters? Or an Origenist God who saves everybody in the end? I could go
on and on. Once we admit that there are many alternatives, often with infinite payoffs, Pascal’s
expected-utility calculations go out the window. (Pascal seems dimly aware of the problem and
makes some rather feeble efforts elsewhere in the Pensées to dismiss some of the alternatives, but
to no avail.) When it comes to what it is prudent to believe (if we are trying to maximize our
post-death expected utilities) it is impossible to determine the best bet.
Thus Pascal’s argument is a failure. He does not provide us with prudential reason (where the
norm of rationality consists in pursuing expected utility) for choosing to believe in the Catholic
God. However, once we admit that it is sometimes possible to choose to believe (or to adopt
belief-forming strategies that induce belief) the possibility opens up that somebody may succeed
where Pascal failed, providing us with good pragmatic or prudential reasons to adopt a religious
belief even if there is insufficient evidence or argument for the proposition believed. And this might
be true even in the absence of persecuting kings or bribing caliphs.
PASCAL’S WAGER REVISITED
Evan Fales, University of Iowa
It has long been recognized that the capacity to reason confers on human beings two forms of ratio-
nality: the ability to make reasoned judgments and the ability to use reason to guide action—or, to
use Aristotle’s terms for the distinction, the ability to deploy both theoretical and practical reason.
The former guides proper belief, the latter proper action. (It is problematic whether believing—or
coming to believe—is an action, but if so, the distinction assigns it to a special category.) One dis-
tinctive feature of pragmatic arguments for the existence of God is that they blur the distinction,
insofar as they do take coming to believe in God to be a kind of action one can undertake to per-
form, or at least to be a change of state one can undertake to effect by way of actions that lie beyond
the scope of theoretical reasoning. The conflation arises because, after all, believing can have (and
not just in this case) practical consequences.
In this section, however, we focus on a further distinction that can be made, this time within
the arena of practical reason. Aristotle noted that theoretical reasoning and practical reasoning differ
in form. In theoretical reasoning, we undertake to justify belief in a proposition by providing evi-
dence for (and against) the truth of the proposition, in the form of other defensible propositions
and/or the data of empirical experience, to show that, all things considered, that proposition is likely
to be true. But in practical reasoning, the inputs to a deliberation come in the form of propositions
of certain kinds (about one’s circumstances, resources, and abilities) and what might generically be
called desires, or desiderata. The purpose is to select one course of action, among alternatives, as
the right or a good one. Desires provide a menu of potential aims or ends, without which action
would have no point. It is sometimes misleadingly alleged that desires populate the irrational side
of human nature. But desires, at least insofar as they do not lie within our control or spring into
being unbidden, are neither rational nor irrational: they are “given.” Their role, however, lies at
the very heart of rationality: without them, we would lack reason to do anything. (It is true that
desires can hinder the proper operation of reason, when they over-insist on their importance; but
that is just to say that preferring certain desires over other competing ones can be a mistake.)
Now, then, the further distinction that is worth attending to is that between felt desires, like
hunger, thirst, cessation of pain, being loved by one’s beloved, and so forth, and ends intellectually
recognized to be desirable or good, such as the desire to be treated fairly by others, or the desire that
one’s hard work will result in a job well done. Admittedly, no sharp line can be drawn between
these, but it is important to recognize that the goodness of certain ends is to be adjudged, not just
on the strength of visceral forces that impel one to act, but on cognitive judgments about what
bestows felicity on natures such as ours.
What does all this have to do with Pascal’s wager? The wager is proposed under conditions of
cognitive uncertainty about the existence of God. Absent such certainty, and recognizing the prac-
tical consequences that (un)belief may itself entail, it is urged that, all things considered, it is wiser
to err on the side of belief than disbelief. Thus, one might suppose, it is pragmatically rational to
choose theism—supposing one can so choose, and provided that believing satisfies desires suffi-
ciently important to the quality of one’s life—even if it is theoretically irrational to do so. The
qualifier is important. In general, our beliefs do not float free of the evidence we see to be rele-
vant. But credulity is a well-known (even if not well-understood) phenomenon; most of us some-
times manage it.
However, one of the teloi of human nature is knowledge: to believe what’s true, and to do so
properly—that is, on sufficient evidence. We pursue knowledge as a means to other ends, but also
for its own sake—presumably as an intellectual satisfaction, though curiosity, seemingly fundamen-
tal to human nature, appears to be a kind of hybrid, or perhaps a desire situated on the borderline
between the intellectual and the visceral. How important is the satisfaction of that (presumably
intellectual) aim or purpose, especially when it conflicts with the satisfaction of other, perhaps vis-
ceral desires and passions? It is not, in general, easy to say. A parallel case may be instructive.
How important is it not to lie? Clearly, honesty is among the most fundamental of virtues; yet, just
as clearly, there are circumstances in which it can be trumped by other obligations (you have hidden
fleeing Jews in your basement; Nazi soldiers in pursuit knock at your door and ask whether you
have any knowledge of their whereabouts).
An atheist might, for example, allow it to be better that a theist whose life of beneficent self-
sacrifice and care for others is centered on deep religious convictions should not have those convic-
tions shattered (supposing they could be by the evidence), if doing so would result in existential
crisis, demoralization, and paralysis of generous instincts. Still, it is a rather different question
whether an agnostic should be converted to theism, not on the evidence but in some other way,
even if doing so would provide an incentive for virtue. The desideratum of epistemically legitimate
beliefs is, that is to say, not easily trumped by independent considerations.
But we may fairly raise the question: how will the moral terrain appear to the atheist, and with
what implications for the pursuit of the life well-lived? At one pole, we have the view, not uncom-
mon among theists, that to deny the existence of a theistic God is to abandon all hope of an intel-
lectually cogent grounding for moral and axiological principles: if there is no God, then, allegedly,
everything is permissible, whatever moral instincts one might nevertheless obey. That is, roughly,
the opinion of William Lane Craig, for example. At the other pole, there are moral objections to
appeal to prudential arguments such as the wager, for example, the often-made point that such argu-
ments trade on motives of self-interest, which are antithetical to genuine love of God. Righteousness
is commended to us just as insurance against the prospect of divine retribution.
Michael Rota (2016a, 2016b) has offered thoughtful responses to objections of this latter kind.
Although we do not pursue them here, something wants to be said to answer the charge that athe-
ism is (at least intellectually) morally bankrupt. The most obvious response—and one that should

suffice—is to point out that there are numerous theories regarding the foundations and nature of
moral truths that make no appeal to a deity, and a number of these are at least as defensible as the-
istic theories of morality (or more so). Pursuing that response would take us too far afield; instead,
let us attend to another feature of the atheistic moral landscape that is too often ignored.
The thoughtful atheist, theists often argue, is forced ultimately into moral despair. A mate-
rialistic world (to which the atheist is allegedly committed) is a world without any place for gen-
uine purposes, aims, or ultimate goals: it is, we might say, a telos-free world. And so our lives
have no ultimate purpose or rationale, no meaning. The universe is but a congeries of atoms,
blindly obeying physical laws that are indifferent to right and wrong; thus human beings, with
their aspirations, ideals, and desire for ultimate justice, are fundamentally aliens who, at best,
suffer idle delusions regarding the fundamentality and objectivity of right and wrong, good
and evil.
There is some truth in these somber reflections. Even if, as must be recognized, it is not a fore-
gone conclusion that a material world cannot account for consciousness, value, or teloi, and even if
there are defensible nontheistic accounts of an objective moral order, still, an atheist might well feel
a certain sense of loss, of anomie, with the realization that we human beings are inconsiderable
blobs of protein in the vastly greater machinery of nature. Such a sense of loss is historically ampli-
fied, at least in Western cultures, by post-Enlightenment scientific revolutions, critical Bible scholar-
ship, and other intellectual developments that have bolstered the intellectual credentials of atheism,
the price being loss of a secure intellectual framework in which a benevolent God remains
enthroned in Heaven.
With this in mind, Pascal’s wager might seem to offer at least a welcome palliative to despair.
Living with unremitting and irremediable despair is a hard task. Psychologically, it can pose severe
threats to a life well-lived, to the undertaking of moral challenges, to any sense of fulfillment or
accomplishment. For one thing, in renouncing theism one abandons a realistic hope for an afterlife,
and with it, any real hope that the injustices of this life will somehow be ultimately set right. To
push on in the face of such dauntingly sober recognitions requires both a deep commitment to
the pursuit of such justice and goods as this life affords opportunity for, and unusual perseverance.
It requires, furthermore, a special kind of courage, the exercise of which can be just as passionate as
hope for an afterlife or for an ultimate balancing of the scales of justice.
A life lived in pursuit of righteousness in the face of such an understanding of the human con-
dition should be, surely, cause for admiration by atheists and theists alike. Despair is, of course, not
the only response with which an atheist might meet the challenges posed by a Godless universe.
Nihilism and optimism are among the others (see Fales 2008 for further discussion). But those the-
ists who demean the worth and integrity of good lives lived in the face of honest despair, or who
suggest, in their formulation of Pascal’s wager, that we can generously suppose that atheists will feel
free to live a life of carefree self-indulgence do a disservice to the existential significance of the atheist
perspective. Exemplary and poignant expressions of atheistic reflections on the dimensions of such a
life can be found, for example, in Albert Camus (1991a, 1991b), Bertrand Russell ([1903] 1981),
Elie Wiesel (1995, 2006), and Miguel de Unamuno (1996).
CHAPTER REVIEW
Pragmatic arguments for the existence of God raise some basic questions about how sharp the dis-
tinction is between theoretical and practical reason, and force attention to the question whether
(coming to a) propositional belief is itself a kind of action. It is a merit of Pascal’s wager that it
forces us to attend to dimensions of religious faith other than purely doxastic ones. Recognizing
that, and allowing that religious faith may often function in profoundly beneficial ways in the lives
of the convinced, it remains problematic whether such arguments are (a) successful on their own
terms, and (b) show that faith can offer moral or eudaimonistic benefits in this life that cannot be
equaled by any comparable concomitants of an atheistic outlook.
Two of the most significant difficulties that stand in the way of a favorable evaluation of the
wager are closely related: they concern the initial probabilities to be assigned to theism and its
denial. First, there is the many-contenders objection: if there is a God who promises an immensely
valuable afterlife, what conditions must be met to win the prize? Are we dealing with a highly ecu-
menical Supreme Being, or with one who has strong exclusivist proclivities? If the former, perhaps it
doesn’t much matter what one believes, beyond a bare commitment to (mono?)theism. If the latter,
what are the odds that we will plump for the correct theology and praxis? Under pressure from these
difficulties, wager defenders have tended to frame the argument as one in which (some favored ver-
sion of) theism is to be assigned a probability of (at least) ½. Such an assignment might be reason-
able, given a sufficiently strong body of evidence favoring (that version of) theism over all its com-
petitors, including atheism. But that is a tall order.
Others, following the lead of James, deploy a different strategy to render the many-contenders
objection harmless: they invoke the psychological fact that (because of acculturation, etc.), an indi-
vidual might simply not be capable of adopting certain religious traditions; for such a one, the only
live options are the faiths he or she can envision embracing. But (a) this will leave many atheists out
in the cold (James would rightly say that this is not incompatible with his defense of faith, but it
does serve to highlight the unfairness of Christian exclusivism—e.g., John 14:6), (b) it fails if there
are even just two theistic competitors, neither of which reaches a probability of ½, (c) it invites a
form of despair that is, if anything, more severe than that faced by atheists (since one or more of
the competitors may also promise a hell for those who lack a ticket of admission to paradise). For,
once one considers the fact that one’s pool of available faiths has been determined largely by factors
beyond one’s control—one’s culture, family background, and so on—one realizes that the prospects
of a favorable outcome are, objectively speaking, small to vanishing. (For a vigorous but modest
defense of James’s view, see Bishop 2007.)
Two final considerations should be mentioned in closing. One is directed at Christian soteriol-
ogy and praxis, the other notes the relevance of a specimen of the problem of evil. One should per-
haps rather speak of Christian soteriologies, since there are many, often with profound differences.
Here, however, we single out certain doctrines and practices that have been historically important
and theologically central to Christian life and ideas about salvation and moral education. These
include (1) the doctrine of vicarious atonement, (2) the doctrine of salvation by grace, not works,
and (3) the doctrine that sincere repentance, often signaled by confession, suffices for absolution
of many sins. The question one might raise is whether these doctrines (and implied praxis) play,
on the whole, a salutary role in the moral education of young Christians and the development of
true moral maturity.
Of course this must be largely a matter of speculation, but common sense and theological his-
tory suggest the following observations: (1) It is notoriously hard to give a morally cogent concep-
tion of and defense for vicarious atonement, as is attested by the decided and persistent lack of con-
sensus among theologians on how to understand this doctrine. Surely, we should take responsibility
for our own sins? (2) The doctrine of salvation by grace is predicated on the supposed incapacity of
human beings to live righteously or to square accounts with God by their own efforts to make res-
titution for their sins. Is such a teaching of (in some versions, abject) moral helplessness a help or a
hindrance to the development of morally mature adults? (3) Do not the doctrines and practices of
confession, ritual forgiveness of sin, and proper absolution prior to death as a necessary and suffi-
cient condition for salvation tend to promote a perfunctory attitude toward repentance, and a pro-
crastinatory attitude toward sanctification? It is not that these questions have clear or ready answers;
but they should serve a cautionary role for those tempted to think that a religious moral upbringing
and life must be superior to an atheistic moral realism, coupled with a strong commitment to the
imperative of taking responsibility for one’s own derelictions, a view of human nature as capable,
with proper character development, of taking such responsibility, and an understanding that moral
goods are to be pursued for the sake of humanity and of one’s own proper ends in this life. A fair
posing of Pascal’s wager must recognize this.
Finally, the wager effectively concedes the hiddenness of God, a circumstance that has
prompted considerable discussion over whether that hiddenness, insofar as it denies the powers of
reason and experience to deliver to sincere inquirers any rational assurance of God’s existence or
nature, does not itself constitute a form of evil or a dereliction on God’s part that is incompatible
with faith. The wager, particularly insofar as it enjoins unbelievers to encourage the growth of faith

not by cognitive means but by religious exercises that work on their affective makeup, encourages
engagement in practices designed to induce faith, not by means that acquaint the mind with truth
but that promote a kind of self-deception. That is not a bargain serious atheists will find attractive.
They are unwilling to perform a sacrificio del intelecto on the altar of contentment. For them, only
the truth, that stern teacher, is worthy of a passion that endures.
Bibliography
Anselm. Proslogion. In Basic Writings. Translated by S. N. Deane, 1–34. Kierkegaard, Søren. Fear and Trembling. Edited by C. Stephen Evans and
La Salle, IL: Open Court, 1962. Sylvia Walsh. Translated by Sylvia Walsh. Cambridge, UK: Cambridge
Aristotle. Nichomaechean Ethics. In The Basic Works of Aristotle. Edited by
Richard McKeon, 935–1126. New York: Random House, 1941. Locke, John. “A Letter concerning Toleration.” (1689.) Translated by
William Popple. In John Locke: A Letter concerning Toleration in Focus,
Bartha, Paul. “Taking Stock of Infinite Value: Pascal’s Wager and Relative
edited by John Horton and Susan Mendus, 98–124. London:
Utilities.” Synthese 154, no. 1 (2007): 5–52.
Routledge, 1991.
Bishop, John. Believing by Faith: An Essay in the Epistemology and Ethics of
Mackie, J. L. The Miracle of Theism: Arguments for and against the Existence of
Religious Belief. Oxford: Oxford University Press, 2007.
God. Oxford: Oxford University Press, 1982.
Camus, Albert. The Myth of Sisyphus and Other Essays. Translated by Justin
Martin, Michael. “Pascal’s Wager as an Argument for Not Believing in
O’Brien. New York: Vintage, 1991a.
God.” Religious Studies 19, no. 1 (1983): 57–64.
Camus, Albert. The Plague. Translated by Stuart Gilbert. New York:
McClennen, Edward F. “Pascal’s Wager and Finite Decision Theory.”
Vintage, 1991b.
In Gambling on God: Essays on Pascal’s Wager, edited by Jeff Jordan,
Clifford, William. “The Ethics of Belief.” (1877.) In The Ethics of Belief and 113–137. Lanham, MD: Rowman and Littlefield, 1994.
Other Essays, 70–96. Amherst, NY: Prometheus Books, 1999.
Murray, Malcolm. The Atheist’s Primer. Peterborough, ON: Broadview Press,
Coady, David. What to Believe Now. Oxford: Wiley-Blackwell, 2012. 2010.
Dawkins, Richard. The Blind Watchmaker. New York: Norton, 1986. Musgrave, Alan. Secular Sermons: Essays on Science and Philosophy. Dunedin:
Duff, Anthony. “Pascal’s Wager and Infinite Utilities.” Analysis 46, no. 2 University of Otago Press, 2009.
(1986): 107–109. Pascal, Blaise. The Thoughts of Blaise Pascal. Translated by C. Kegan Paul.
Fales, Evan. “Despair, Optimism, and Rebellion.” In The Future of Atheism: London: George Bell and Sons, 1901.
Alister McGrath and Daniel Dennett in Dialogue, edited by Robert B. Pascal, Blaise. Pensées: The Provincial Letters. Translated by W. F. Trotter.
Stewart, 97–112. Minneapolis, MN: Fortress Press, 2008. New York: Random House, 1941. First published 1670.
Flew, Antony. God and Philosophy. London: Hutchinson, 1966. Pascal, Blaise. Of the Geometrical Spirit. Translated by O. W. Wright.
Hacking, Ian. “The Logic of Pascal’s Wager.” American Philosophical University of Adelaide, 2014. First published 1657/1658. https://
Quarterly 9, no. 2 (1972): 186–192. ebooks.adelaide.edu.au/p/pascal/blaise/geom/.
Proast, Jonas. The Argument of the Letter concerning Toleration Briefly
Hájek, Alan. “Waging War on Pascal’s Wager.” Philosophical Review 112,
Considered and Answered. (1690.) In Locke on Toleration, edited by
no. 1 (2003): 27–56.
Richard Vernon. Cambridge, UK: Cambridge University Press, 2010.
Hick, John. Evil and the God of Love. New York: Macmillan, 1966.
Rota, Michael. “A Better Version of Pascal’s Wager.” American Catholic
Hume, David. The Natural History of Religion. Edited by H. E. Root. Philosophical Quarterly 90, no. 3 (2016a): 425–439.
Stanford, CA: Stanford University Press, 1956.
Rota, Michael. Taking Pascal’s Wager: Faith, Evidence, and the Abundant Life.
Hume, David. Dialogues concerning Natural Religion. Edited by Norman Downers Grove, IL: InterVarsity Press, 2016b.
Kemp Smith. Indianapolis, IN: Bobbs-Merrill, 1985.
Russell, Bertrand. “A Free Man’s Worship.” (1903.) In Mysticism and Logic,
James, William. The Will to Believe and Other Essays in Popular Philosophy. and Other Essays. Totowa, NJ: Barnes and Noble, 1981.
London: Longmans Green, 1897.
Schlesinger, George. “A Central Theistic Argument.” In Gambling on God:
James, William. Pragmatism. New York: Longman Green, 1970. Essays on Pascal’s Wager, edited by Jeff Jordan, 83–99. Lanham, MD:
Jordan, Jeff. “The Many-Gods Objection.” In Gambling on God: Essays on Rowman and Littlefield, 1994.
Pascal’s Wager, edited by Jeff Jordan, 101–113. Lanham, MD: Rowman Snorri Sturluson. The Heimskringla; or, Chronicle of the Kings of
and Littlefield, 1994. Norway. Translated by Samuel Laing. London: Norroena Society, 1906.
Kant, Immanuel. “The Impossibility of an Ontological Proof of the Sober, Elliott. “The Design Argument.” In The Blackwell Guide to the
Existence of God.” In Critique of Pure Reason. Translated by Norman Philosophy of Religion, edited by William E. Mann, 117–147. Oxford:
Kemp Smith, 500–507. London: St. Martin’s Press, 1969. Blackwell, 2005.
Kenny, Anthony. The Unknown God: Agnostic Essays. London: Continuum, Stark, Rodney, and Roger Finke. Acts of Faith: Explaining the Human Side of
2004. Religion. Berkeley: University of California Press, 2000.
Swinburne, Richard. Faith and Reason. Oxford: Clarendon Press, 1981. Waldron, Jeremy. “Locke, Toleration and the Rationality of Persecution.” In
John Locke: A Letter concerning Toleration in Focus, edited by John
Turri, John, David Rose, and Wesley Buckwalter. “Choosing and Refusing:
Horton and Susan Mendus, 12–56, 119. London: Routledge, 1991.
Doxastic Voluntarism and Folk Psychology.” Philosophical Studies 175
(2018): 2507–2537. Wiesel, Elie. The Trial of God. Translated by Marion Wiesel. New York:
Schocken, 1995.
Unamuno, Miguel de. “Saint Emmanuel the Good, Martyr.” In Abel
Sanchez and Other Stories, translated by Anthony Kerrigan. Washington, Wiesel, Elie. Night. Translated by Marion Wiesel. New York: Hill and
DC: Regnery, 1996. Wang, 2006.

TOPIC 20
Final Reckonings: Theism

Fordham University, NY
The first half of this chapter examines Augustine of Hippo’s use of philosophical distinctions to clear away
various difficulties that stand in the way of pursuing such questions as the existence of God and the relations
between some of the attributes normally attributed to God (omnipotence, omniscience, omnibenevolence)
and the problems presented by evil and freedom. The second half of the chapter examines natural law
theory to make a case that morality requires the existence of God.
ENCOURAGEMENT FOR THE JOURNEY
Even the vast size of this book cannot exhaust discussion about the merits of the arguments for and
against theism. Rather than review what has been said, however, I would like to use this final chap-
ter as a way to encourage readers to continue the quest for the truth about this subject. In my view,
there are no more important questions than whether God exists, whether God made the cosmos,
and whether God loves creatures like us enough to preserve us in existence, to help us on our jour-
ney, and to keep us free enough to make good choices about how we stand toward him and one
another. These are choices that will determine our everlasting destiny. Should the answer to the first
of these questions be negative, one will be free of nagging worries about the rest of the list. Con-
versely, if the answer to the first question is affirmative, or even could be so, then the supreme
importance of the others is clear. Every human generation has felt the significance of these ques-
tions.
In the first half of this chapter, I would like to consider the way Augustine of Hippo found phi-
losophy useful in the service of faith. Prior to accepting religious faith, he employed a number of
crucial philosophical distinctions to help resolve various difficulties that had prevented him from
seeing clearly how best to address the great questions just mentioned. Following Augustine’s story
will help us to see that religious faith is not irrational and that, quite to the contrary, it allows us
to fulfill our human drive to know the truth about things. In the second half, I consider a part of
the case for the existence of God that has long struck me as profoundly compelling—the need for
a transcendent source of morality. To examine this theme, we will consider the philosophical under-
pinnings of traditional natural law theory.
THE ROLE OF PHILOSOPHY IN CLEARING AWAY OBSTACLES

TO FAITH
As a personal record of the journey that led Augustine to accept religious faith, the Confessions draws
its name from two senses of the term confession. It refers not only to the admission of faults with real
contrition but also to the admission of creaturely dependence on God as the creator of all things,
the redeemer of the human race, and the sanctifier of souls. Augustine understood his profession
665
of faith as a choice that he made in response to a gift of grace

KEY CONCEPTS from God. His decision to believe was not simply the result
of seeing the implications of any philosophical argument; in
the process of coming to his decision, philosophical meditation
Dao
played an indispensable role both in awakening his quest and in
Dharma clearing the path of various obstacles. We shall see, for example,
Manichaeism that philosophy allowed Augustine a way to think of God as
Natural Law spiritual rather than material, to cease holding material things
in suspicion and instead to see them as made by God but capa-
Neoplatonism ble of misuse by us, and to grasp that to admit the reality of
Perdurance human free choice does not imperil the omniscience, omnipo-
Representationalism tence, or omnibenevolence of God. Gaining these philosophical
insights helped him to get beyond what had been major stum-
Supereminent Theology bling blocks for him.
It was while studying rhetoric in Carthage at age sixteen,
he explains, that he fell into a “cauldron of illicit loves”
(Confessions, III.1). In that same period of his life, he read the Hortensius of Marcus Tullius
Cicero and felt a burning desire “to seek and win and hold and embrace not this or that philo-
sophical school but Wisdom itself, whatever it might be” (III.4). Hard as it proved for him to
sort through various intense but shallow friendships, the arousal of his passions at the theater,
the heady joys of professional success as a rhetorician, and the seductive promises of Manichae-
ism, Augustine from this time forward displayed a passion for finding knowledge, truth, and
wisdom and a discomfort with ersatz substitutes: “O truth, truth, how inwardly did the very
marrow of my soul pant for you” (III.6).
By pondering a few unnamed books of Platonic philosophy, Augustine began working his way
through an interlocking set of problems. Scholars speculate that he may have read portions of the
Enneads of Plotinus or perhaps works by his disciple Porphyry, a bitter critic of Christianity. The
problems were admittedly those that had come to bother him in the course of his own history,
but they do not seem terribly different from the kind that anyone thinking about these issues has
to face at some point.
Throughout the Confessions, Augustine shows himself grateful to have received from God the
gift of faith in his early thirties, but his embrace of religious faith came only after a period of intense
struggle over such questions as the origin of evil, the freedom of the will, and the nature of God.
The outcome of his inquiry was not clear to him in advance. Like the chapters in this volume on
Theism and Atheism, the pages of the Confessions (especially book 7) show the need to take the ques-
tions seriously and the obligation in conscience to follow where truth leads.
DISTINGUISHING THINKING FROM SENSING AND IMAGINING
From the Confession’s record of Augustine’s experience it is possible to learn the paradoxical lesson
that good and careful reasoning can help us break free from the constraints of rationalism. By that
term I mean the philosophical position that tries to make reality conform to the categories of one’s
own typical ways of reasoning. I use the term realism here to designate the philosophical stance that
tries instead to conform our thinking to reality. The relevance of this distinction for the case of
Augustine can be seen in his long struggle to resist the tendency to impose on God the categories
of thought appropriate to the spatiotemporal realm (e.g., having sensory experience of specific
shades of color, particular fragrances and tastes, and so on). Because all the objects that Augustine
had ever come to know experientially had such properties, he quite naturally assumed that all possi-
ble objects of knowledge have such properties, and for this reason he invariably imagined any divine
being as having them too.
As Augustine envisioned his situation, there was an interwoven cluster of problems facing him.
If there were such a thing as a god, that god (however powerful) would have to be a being within
the cosmos (where else is there to be?) and thus a finite being (because there are other things in the
cosmos that are not this god, there must be some limit to the being of god, and thus god must be
finite). A finite being (however knowledgeable and well disposed toward us), however, would at best
be locked into an endless struggle with other forces (for Augustine at the time, the forces of darkness
and fate) that also appeared to be at work in the cosmos and in human choices. In addition to such
metaphysical questions, questions of moral psychology also pressed on his mind, such as finding a
satisfying explanation for how one could possibly know what was good and right and yet still some-
times do what was bad or wrong. These problems long seemed deeply mystifying to Augustine.
His reading in platonism did not seem to give him a totally satisfying answer, but it did strike
him as a better way to grapple with the difficulties than other approaches he had considered. Years
earlier, Augustine had read Aristotle’s Categories with understanding and appreciation, but at the
time, he explains, he had misapplied its teachings when considering the case of God. For he had
imagined that anything real should be classifiable within those ten categories (substance, quality,
quantity, relation, action, passion, time, place, habit, and posture). God, he presumed, must be
some corporeal substance in which qualities such as beauty and powers like intelligence adhered
(IV.16). With the candor that his conversion made possible, Augustine later admits that at the time
his mind was still enthralled by disordered desires and that he could not conceive of an incorporeal
substance (IV.15), and so he could not yet see clearly enough to get beyond thinking that God was
some “luminous immeasurable body” and that he himself was “a kind of particle broken from that
body” (V.10).
Real openness to belief came only with a better understanding of what kind of being was at
issue in questions of religious belief. To come to this understanding there was first the need to rec-
ognize a crucial distinction in kind between sensation (the power to perceive things with our senses,
for instance, specific shades of red or green that we experience by seeing an individual apple or a
leafy tree or particular degrees of sweet or sour in tasting a specific drink) and understanding (the
power to grasp mentally what, for instance, color in general or flavor in general is or what an apple
in general or a tree in general is).
As Augustine discovered by reading treatises in rhetoric and in philosophy, the terms by which
we know these things in general are called universal terms, and that by which we hold them before
the mind’s gaze is called an idea or concept. So long as he thought of an idea or concept as merely a
picture or an image, these ideas seemed just as particular and material as the images from which
they were drawn. What made progress possible for Augustine was grasping the difference in kind
between sensory images and mental concepts. The former are particular and material, and the latter
are universal and immaterial. The former are sensory impressions, and the latter are the result of our
powers of judgment about what the things under consideration are and what they are like.
The intrinsic materiality of our sensory knowledge is such that every sensation we have is a sen-
sation of some particular quality within a given range on a continuum appropriate to each sense
organ. When we understand what a color is or what a tree is, however, there is something of general
applicability that we hold before the mind by means of our ideas or concepts. It is not only when
we understand immaterial things such as God or angels but even when we turn the attention of
the mind to understanding material objects, our mental grasp of them has a spiritual (that is, imma-
terial) aspect. When we understand what a sensory quality such as red or blue or what a sensory
object like an apple or a planet is, we do so in an immaterial way, that is, by a general or universal
term. As an orator with a deeply philosophical bent, Augustine, in his love of words and how they
signify, was able to reflect on the difference between sensation and understanding, and this distinc-
tion proved crucial for his consideration of the question of God.
While initially the human mind has nothing on which to focus without receiving images from
outside through the senses, it is able to reflect on what it receives and to go beyond the limits of
sensory experience. In fact, it must do so in order to attain real understanding of anything. Augus-
tine writes: “My mind was in search of such images as the forms that my eye was accustomed to see;
and I did not realize that the mental act by which I formed these images was not itself a bodily
image; yet it could not have formed them unless it were something and something great” (VII.1).
Augustine here records a moment of insight about the spiritual character of the mind—that it is

something real but immaterial. This recognition would be crucial for all of the other steps that he
would come to make on this journey to accepting religious faith, and yet it was not itself directed
by faith, for he did not have such faith at the time of the insight.
In due time Augustine applied to the question of God the recognition that what we attain in
knowing is not merely another sort of picture or image but an idea by which to grasp what some-
thing is and what it is like. This insight about the nature of the faculties of the mind gave him
important tools for seeing that God must be understood by thinking rather than by sensation or
imagination. It had been perplexing to Augustine what the Supreme Being could possibly be. He
employed this distinction to make the sort of progress that he could not make so long as he envi-
sioned knowing as simply another kind of sensing or imagining. So long as he still regarded all
thinking as picture-thinking, his thinking about God had foundered on the latent discrepancies that
emerged from picturing God in one or another finite form (VII.1–2).
CLEARING AWAY PROBLEMS SET BY TOO LIMITED AN IMAGINATION
With these insights about the nature of human cognition and with certain metaphysical arguments
in place for the existence of a first cause, Augustine was equipped to understand how God could be
a being but an incorruptible one. It amounted to seeing that God is real and present everywhere but
is not limited by some finite extension in space. God is present to all moments of time but not with
the finite perdurance for a limited period that is typical of temporal beings.
This realization about the nature of divine being did not force Augustine’s hand or bring him
to the point of a decision to make a profession of faith (for that, see VIII.12), but it freed him from
one of the stumbling blocks of the rationalism that had conscripted his intellectual vision, namely,
the presumption that for God to be real, God would have to be a finite being of the same sensory
sort that daily fills anyone’s experiences and imagination.
Later religious thinkers would take up the same lines of reflection and articulate the insight sys-
tematically. In Anselm, for example, this is a constitutive part of thinking through the idea of God
as “that than which none greater can be conceived” (Proslogion, chap. 3). For Pseudo-Dionysius, it
would be expressed (e.g., The Divine Names, chap. 4) through the combination of the methods
proper to negative theology, the denial of any limitation, and to supereminent theology, the use
of finite qualities as a base from which to point out the direction of absolute perfections that lie
beyond anything that the human mind can encompass.
These refinements are a crucial part of the history of theology, for they express with greater pre-
cision what we can come to know of a being that always exceeds our own categorization. To under-
take these refinements, the first steps have to be of the sort that Augustine describes in the process
of coming to understand more about what God is. Such knowledge comes through thinking in a
way that can get beyond the contradictions that invariably surface so long as we imagine the divine
nature along the lines that we are used to from ordinary experience.
For us today, the situation is no different in principle even though the details of our back-
ground worldview may have changed. So long as we presume that for anything to be real it must
be material, there will be no adequate way to consider the question of a divine being or to answer
questions about the sufficient reason for what we directly experience. We tend to picture the cos-
mos as immeasurably vast but entirely constituted by beings that are in one way or another forms
of matter and energy. It is entirely understandable that we do so, for the advances in each
domain of science have come about by a rigorous application of methodological materialism—
that is, the commitment of scholars in a given discipline to seek answers to the questions proper
to that discipline only by recourse to objects and forces within a given set of physical entities and
energies.
There is a pervasive tendency to mistake this methodological materialism for metaphysical
materialism—the position that the only sort of things that are real are material entities and forces.
As Augustine’s experience shows us, there are various sorts of problem that flow from coalescing
these two forms of materialism. What can truly help us make progress within the sphere of a given
discipline (methodological materialism) can render us blind to the sphere of what is real but imma-
terial and not just invisible to the color range accessible to human vision (metaphysical materialism).
Coming to an awareness of immaterial reality began for Augustine with his recognition that our
human knowledge is not identical with the images and sensations that are the basis for our thinking.
His further progress in understanding other aspects of human experience also required him to grant
that there are immaterial realities. One sees this point in Augustine’s efforts to consider the meta-
physics of freedom. He came to grasp that no human choice could be free unless it was the result
of something truly independent of the fully determined nexus of physical causality. Still further
advances came for him when he learned to envision divine reality as semper maior—always
greater—than any of the limiting forms by which we distinguish one kind of thing or quality from
another.
Our reflection on Augustine’s experiential account can show us, among other things, the need
to get beyond a representationalist epistemology of the type championed by John Locke and popu-
lar in many forms today. There are various strategies for doing so. One might think, for instance, of
the devastating critique of representationalism offered by works like Richard Rorty’s Philosophy and
the Mirror of Nature and of the inexorable logic that Thomas Nagel uses in Mind and Cosmos to
point to the irreducibility of mind to matter. Even though Nagel stops short of accepting theism,
his book does invaluable service in the work of clearing the ground in epistemology.
Happily, the alternatives to representationalism are not limited to theories that envision human
knowledge as ultimately mind-constructed, for the position that all that we can ever know is limited
to our own perceptions provides no framework for ever getting knowledge of the other as other.
Rather, a better alternative is an epistemology in the Aristotelian and Augustinian tradition that
respects the primacy of intellectual receptivity and a rational openness to independent reality.
Among contemporary efforts in this direction we might think, for example, of the one developed
by John McDowell’s Mind and World as part of his account of how the active work of conceptuali-
zation requires ongoing receptivity to the world known by our senses. It is by considering such con-
temporary ventures into philosophical realism that we can become confident about developing the
kind of openness that was crucial to Augustine’s pathway.
DEALING WITH EVIL AS AN OBJECTION TO THEISM
The next major obstacles that Augustine found to block his way were the puzzles that he perceived
in regard to the paired issues of evil and freedom. Granting the differences between the circum-
stances of his situation and ours, both of these issues can still prove to be stumbling blocks to faith
today. Part of Augustine’s early attraction to Manichaeism was its apparent ability to provide an
explanation for evil in the world by tracing it back to a principle somehow envisioned as equal to
God but qualitatively opposite: material rather than spiritual, dark rather than light, and so on
(Confessions, III.6–10, V.3–7). The dualist metaphysical underpinnings required to support such a
notion led him to a dualist notion of the human being as a light-filled spirit trapped in the darkness
and disorder of a corporeal body (III.6). As Patrick Lee and Robert George have convincingly
shown in Body-Self Dualism in Contemporary Ethics and Politics, there are strong proclivities to a
related form of dualism in the intellectual culture of our own day.
Augustine’s two-pronged response to this challenge was to develop certain hints found in the
Platonic books about the privative character of evil and others that he drew from the Stoics and
Saint Paul about weakness of the will. Stymied for a long while by the dilemma of thinking that
the presence of evil in the cosmos would force us to choose between unlimited divine power and
unlimited divine goodness, Augustine discovered an alternative position in Platonic thought on
the question of the nature of evil that helped him to resolve that dilemma. To hold that God was
all-good had seemed to require that there must be some other force responsible for whatever is evil
or destructive. To hold that the omnipotent God was the cause of absolutely everything that exists,
however, seemed to require that the creator was not all-good, or perhaps not all-knowing, or that
(inconceivably) evil, suffering, and pain were not real.

For Augustine, it was possible to make progress on the issue of evil by developing an insight
traceable to Plotinus. Admittedly, Plotinus provides an inconsistent treatment of the topic, some-
times regarding evil as a positive reality in its own right and taking matter as evil and as a source
of evil in the soul (see Enneads I.8) and at other times recognizing that evil is the absence of good
(Enneads III.3–4). The notion that evil may be understood as the absence of something sparked
the realization in Augustine that it was possible to account for the reality of suffering, pain, and
destruction without compromising the omnipotence or the perfect goodness of God.
To do this requires that we come to understand the privative character of evil not merely as the
lack of something but as the lack of some due good in a being that has positive reality in its own
right. The phrase due good here refers to a good that is expected in a being on account of its nature.
That plants do not have vision is not an evil or defect in their being, but for a dog or a cat to be
blind (whether it comes about by a genetic defect or by an injury received in an accident or a fight)
would be deleterious for such animals.
The difficulty of the problem should not be understated, for we are inclined to label things that
are the source of pain and destructive as evil. Such things are not mere absences. They are powerful
precisely because they are real beings. Our attribution of “evil” to them does not pertain to a lack of
being in such objects but instead to the effects they have, including the damage or destruction that
they bring about and perhaps to some disorder (that is, a lack of due order) within them that
explains why they act in such a way as to bring about damage or destruction.
To use the terminology developed later in the history of thought for the articulation of Augus-
tine’s insights in this area, both physical evil and moral evil may (with the appropriate distinctions
in place) be described as privations. Moral evil refers to those free choices that a person deliberately
makes that exhibit the absence of the right order of love. Physical evil refers not to a thing in itself
(VII.16) but to the absence of some feature that belongs to a being’s nature (for instance, the deaf-
ness of an animal that by its nature has the power of hearing). For Augustine, the absence of a due
good of a physical nature connotes no moral wickedness in the one who suffers from a defect or
injury, except in those cases where the privation of some good quality or feature is the result of
deliberate choices made by the one who then suffers as the result of those choices.
At issue for Augustine in this account of physical evil is not merely the absence of some feature
that might be desirable (human beings, for instance, do not by nature have wings for flying or gills
for breathing under water) but, indeed, the absence of some due good (congenital defects, the dam-
age done to an organ or limb by disease or by accident, and so on). Putting the matter in this way
allows Augustine to affirm the reality of pain and suffering and at the same time to explain the ori-
gin of physical evil by recourse to this worldly causation, for example, the damage that unfolds from
a genetic defect or the damage that one object does to another. Considered in this way, a physical
evil is something that God allows rather than something that God directly and positively chooses
to bring about, except for cases of the punishments that God brings about for moral evil.
Like physical evil, moral evil also has a privative character. Moral evil, for Augustine, occurs
when our deliberate choices contradict the objective hierarchy of goods in the universe. This hierar-
chy of goods has God as the Supreme Being, rational beings such as angels and humans below God,
and then animals, plants, and inanimate objects lower still. Augustine holds that moral evil involves
the absence of the proper order of a person’s loves (IV.12), whether a person chooses to love the
right object at the wrong time or in the wrong way, or to love the wrong object, or to love a higher
good too little or a lower good too much.
In giving human beings the power of free choice of the will, God has given human creatures a
great gift. For those creatures truly to be free, God allows them to exercise that power in ways that
can transgress the proper order of loves that ought to govern their use of this gift. As Augustine pon-
ders this matter, it becomes ever clearer to him that there is a goodness in all things and that all
things derive their being from God, either directly or by the mediation of worldly causes (VII.12).
Although Augustine’s account of evil as privative has its roots in various Neoplatonic texts, his
understanding of moral evil moves beyond the Platonic position that moral wrongdoing may be laid
at the feet of ignorance of the good. As Hannah Arendt explains, Augustine follows the lead of Saint
Paul (see Romans 7:19) in creating a genuinely new idea of the will in the course of his reflection on
the times when we fail to do the good we want to do and when we do the evil that we do not want
to do.
In addition to consulting the texts of Augustine himself on this question (e.g., VII.3), one may
find an insightful elaboration of this position in the opening chapters of C. S. Lewis’s The Problem
of Pain. In good Augustinian fashion, Lewis makes the case that God could have created a world
without suffering or pain but that there would be a logical contradiction if we were to suppose that
God could have simultaneously created a world in which there could be no damage or destruction
and one in which rational creatures have free choice of the will. The reason is that genuinely to have
free choice of the will implies that freely chosen actions will have their consequences, and for actions
to have their consequences requires that there be a world with stable natural laws in which the inter-
actions among physical objects have their effects. Such interactions can bring about damage and
even destruction when things collide.
In passing, I should point out that Lewis, like G. K. Chesterton (see “The Ethics of Elfland” in
Orthodoxy), deliberately resists what he takes to be mistaken in the Humean criticism of causality as
if it were a misplaced claim about some logical necessity; rather, he takes causality as Aristotle,
Augustine, and their traditions took it, as the actualization of a potency for change by a real being
with sufficient power to induce that change.
To put Augustine’s point rather succinctly, the possibility of pain and suffering is an inescap-
able aspect of physical reality if there are to be creatures with the power of free choice. He
explains his approach to this problem biographically. During his Manichaean period, he had been
partial to the notion that we could take credit for our good choices while chalking up our moral
failings to the flesh. At the time, he thought human flesh derived from material that was knocked
off the co-principle of darkness and evil in the course of its endless clashes with the co-principle of
light and goodness (III.8). Augustine examined this notion carefully, especially through his con-
versations with the Manichaean bishop Faustus. Faustus, however, proved unable to resolve the
questions that had been troubling him. Augustine reports that he could not abide the contradic-
tion of claiming credit for the good that he made while avoiding blame for his bad choices
(V.6–7) and that, as a result, he came to see human free choice and not some cosmic principle
as the cause of wickedness, pain, and suffering.
A bit later, in connection with the famous garden scene in which he records his conversion
(VIII.10–11), Augustine formally renounces the Manichaean doctrine that our inner conflicts of
will are evidence of a dual nature within us. He insists instead that having a power of genuinely free
choice means that we can be attracted to any number of apparent goods. We manage to make a
choice only by saying no (at least for the time being) to all the options before us except one. While
human beings find themselves necessarily attracted at least to some extent by any apparent good that
comes to our awareness, there is no guarantee that an apparent good will be a genuine good. There
is need to use our minds to examine the attractions and revulsions that we feel and to make judg-
ments and choices.
It is worthy of note that neither Augustine nor any thinker in the Latin-speaking tradition of
the patristic age and Middle Ages ever spoke about free will (libera voluntas). They invariably used
the phrase free choice of the will (liberum arbitrium voluntatis). Their reason, I believe, was that they
took the power of will in human beings to have as a natural property an orientation of being neces-
sarily attracted to anything that appeared to be good. In this respect human beings are like other
creatures that have appetites and desires. What makes human beings free not merely in the sense
of being able to act spontaneously (as any other animal would do) but also in the sense that is dis-
tinctive of our species is the capacity to make a judgment or choice (arbitrium) about whether or
not to take the means necessary to pursue the apparent good to which we are attracted. To make
such choices well, reason needs to consider whether these apparent goods are genuinely good and
thus choice-worthy as well as whether we are willing to take the means necessary to pursue them.
In describing this situation, Augustine resists what might be called the hydraulic model of the
will, according to which we deterministically yield to the strongest pressure (in pursuit of an appar-
ent good or in flight from an apparent evil). Rather, he sees that it is possible for us to act against
even a very strong inclination and to choose even that to which we are strongly averse, for the will

is a spiritual power that transcends the causal nexus of physical determinations (see especially Augus-
tine’s postbaptismal account of the natural powers of memory, understanding, and will and the
effects of divine grace in restoring these powers, X.29–41).
As the story of Augustine’s moment of conversion that is recounted in the eighth book of the
Confessions makes clear, it was the grace of God and not Augustine’s own reasoning that strength-
ened his will and made possible his choice to make a profession of faith (VIII.12). As this chapter
has tried to show, however, philosophical considerations played an enormous role in clearing away
certain difficulties that came from mistaken assumptions, overly limited imagination about possible
explanations, and self-serving rationalizations.
My hope is that this review of the use that Augustine made of philosophical reasoning in search
of wisdom and knowledge may be useful for readers at some stage or other of their own journey to
answer the great questions. Let me now turn from the consideration of a figure such as Augustine
and his concerns with understanding the claims of revealed theology to a different aspect of the the-
istic tradition—one that is primarily rooted in natural (that is, philosophical) theology. This is the
argument that reflection on morality gives its own support for taking the theistic position. Examin-
ing this argument should help us to see some of the differences between the theism to which Augus-
tine came by grace and the theism that one might feel the need to accept by the force of certain
philosophical arguments.
THE DEPENDENCE OF MORALITY ON GOD
In Real Ethics, a book on the foundations of morality, John Rist has argued that the practical result
of much contemporary discussion of ethics has been moral confusion, inside the academy and with-
out. Like Alasdair MacIntyre’s observations in After Virtue about the shrill tones typical of intermi-
nable debates that seem to make no discernible progress, Rist’s diagnosis of professional ethics has
been severe.
What Rist calls the “mindless attachment to an ethics of rules” (whether consequentialist or
deontological) devised by calculative rationality has proved to be forgetful of what makes human life
truly happy and deeply satisfying (2002, 10). The tendency to rationalize even certain brutalities in
practice and the self-deluding readiness to indulge certain personal proclivities while fending off
their natural consequences have made for a dire situation in philosophical ethics. Rist sees two alter-
natives: either a genuine reconstruction of the foundations of moral theory along Platonic and
Augustinian lines or a readiness to embrace an amoral individualism in which people construct their
own moral universes and then use the powers put at their disposal by a highly technological culture
to stave off any unpleasant consequences of their actions, regardless of the cultural implications.
As we might expect, Rist finds the moral nihilism of the second option, often disguised as dem-
ocratic egalitarianism, to be repugnant. That some thinkers take the route of religious fundamental-
ism in the face of the impasse created by skepticism and nihilism he considers somewhat under-
standable but not a properly philosophical response to the problem. At the root of the situation,
in Rist’s view, is the tendency to understand ethics as a kind of free-floating discussion that depends
only on prudential judgments or on arbitrary beliefs about the good.
In Real Ethics he observes that either an ethical theory has a transcendent foundation (whether
explicit or implicit), or it lacks one. In the latter case, he believes that it can be shown to be ulti-
mately incoherent and that whatever strength it has rests on some philosophically inadequate foun-
dation such as power. His book displays in great detail the emptiness of attempts to do moral rea-
soning without recourse to adequate metaphysical foundations and the inevitable sophistry that
comes to the fore once we make the assumption often championed today that human beings are
merely the autonomous products of blind evolution.
What is the alternative? Rist argues that “real ethics” involves discovering values that we do not
invent (2002, 47). His view is that the divine mind alone offers a suitable explanation about the
source of the transcendent reality of values that is needed for an adequate foundation for ethics, that
is, an understanding of a provident God as the final cause whose goodness orders the universe.
One of the most compelling themes of the book is its recurrent attention to a basic mistake in
regard to what ethical reasoning is about. It is not about how to shape some self-justifying vision of
the good that may then be used to rationalize the conduct needed to accommodate our wants and
desires. The point of undertaking an academic study of ethics is not to become a logical enabler.
Rather, Rist argues, ethics is about how to shape our souls so that we will want and desire what is
really satisfying and really perfective.
Moral realism requires that we learn how to form our characters, not how to shape our argu-
ments for self-serving purposes. What most needs to happen in moral formation is that it would
help us to come to want the sort of things that are genuinely good, not that it would teach us
how to manipulate reasons and arguments so as to get what we happen to want, especially given
the ebbs and flows of emotional life and the shifts in what we find desirable and gratifying.
Speaking more broadly, ethics is about genuine human flourishing and so needs to be built on
a solid philosophical anthropology, and that, in turn, needs secure metaphysical foundations. The
alternative would be to create some moral bungalow in the hope of enjoying the pleasures of living
near the shore, regardless of the fact that the way in which we have propped it up will be insuffi-
cient to withstand beach erosion, let alone the gales and storms that regularly occur in life. Building
more securely, in a more reasonable location, is not only more prudent as a calculation but also
more reasonable in a universal sense. It involves cultivating an appreciation for goodness that is
not of our own making through cultivating respect for the structures of reality, including reverence
for the author of all reality. Following in the footsteps of Augustine, Rist thus moves us beyond the
admittedly important contributions to the recovery of moral realism made by such neo-Aristotelians
as Philippa Foot to a sense of why we need recourse to God for a realist account of the good (see
especially Rist, chap. 9).
NATURAL LAW THEORY
In my judgment, the form of moral theory that is best suited for this task is traditional natural law
theory. Despite the reference to law in the name, this approach to ethics is not primarily a system of
rules, obligations, and prohibitions. (For a discussion of the role of virtue in moral realism see
Anscombe 1958.) Rather, it is a program for guiding human behavior (individual and social) toward
the kind of free choices and to an ordered life of the virtues (see Sanford 2015) that will respect the
intentions of the designer of human nature and will bring individual human beings to the happiness
without end that comes from union with God.
One can find proponents of some aspects of natural law reasoning in the ancient strata of various
cultures (e.g., the Dao of the Chinese Confucian tradition and Dharma in the Indian Buddhist tradi-
tion), but its classic articulation comes from Thomas Aquinas. His Summa theologica locates the natu-
ral moral law as our human participation in the eternal law. By the term eternal law, he means God’s
providential plan for each type of being within creation. As I have argued elsewhere, it is precisely in
Aquinas’s treatment of the natural moral law that he has integrated the Augustinian understanding of
the transcendent source of ethical normativity at the core of biblical ethics with the Aristotelian con-
cept of nature and natural virtue (see Koterski 2016). Within his treatise on the natural moral law
(Summa theologica, I–II, qq. 90–97), Aquinas gives a detailed consideration of how the natural law
matches what has been divinely revealed and recorded in the scriptures (see q. 91, aa. 4–5). There
he argues for a considerable overlap between the natural law and the divine law that God explicitly
revealed in the Ten Commandments and that God did this so that people could know with greater
certainty what they could, in principle, discover on their own about the natural law.
Within the Summa, the treatise on natural law is part of a larger theological program that sees
beatitude as the happiness of everlasting union with God and that outlines a program of law and
virtue that is not so much about laying out a rule-based ethics of what we ought to do as it is about
articulating a virtue-based ethics of what we ought to be (see especially q. 94, a.2).
A central claim of this approach to ethics is that the natural moral law is an objective reality
that can be discovered within each human being in every culture and every epoch. Its proponents

admit that the knowledge of this moral law may decline in regressive cultures and increase in cul-
tures that are more alert to what human flourishing requires. Likewise, they grant that the sensitivity
of individuals to its moral demands may diminish when a person’s conscience grows dull or when a
culture’s formation of its members becomes distorted.
To articulate the natural law as a body of knowledge, complete with an adequately formulated
set of justifications, much reflection is needed. It is for just this reason that a work like Aquinas’s
Summa theologica devotes so much attention to providing a metaphysical grounding for all his asser-
tions about moral obligation by offering a detailed account of human nature, of the intrinsic orien-
tation of rational creatures to find the fulfillment of their potentialities in union with God, and of a
description of the virtues that are constitutive of a good human life.
A proper understanding of traditional natural law theory requires preliminary comments, for a
broad range of moral theorists have laid claim to the term. In particular, we need to distinguish
between the laws of nature in the moral sense and the “laws of nature” that theorists in various dis-
ciplines have formulated to articulate the regularities that have been discovered about the way phys-
ical objects and forces work. As Robert Spitzer notes, discoveries about the fine-tuning of various
cosmological constants reinforce the notion of divine design throughout the cosmos and permit us
to make a compelling argument for the existence of God on the basis of the laws of nature operative
in the physical realm. At issue here, however, is a separate but not unrelated concern with the laws
of nature that govern human conduct—which is what has traditionally been meant by the natural
moral law.
The precepts of the natural moral law, in its classical understanding, refer to the self-evident
standards by which human conduct is to be judged (Summa theologica, I–II, q. 94, a.2). The most
fundamental of these standards (e.g., that good is to be sought and done, that evil is to be shunned
and avoided) are so basic that they are as inescapable for practical reasoning as are the first principles
of theoretical reason, for example, the principle of noncontradiction (that one cannot meaningfully
assert and deny the same proposition at the same time) and the principle of sufficient reason (that
for anything whatsoever, there must be an adequate explanation, either in itself or in its causes).
While there is, in principle, no way to derive these first principles from anything yet more basic,
there are informal routes for illustrating their self-evidence, for there is no way to go forward in rea-
soning if one holds that something both is and is not the case at the same time, considered from the
same respect. Likewise, it makes no sense in the sphere of practical reasoning to imagine holding as
a basic principle that good is to be avoided and evil sought. Even villains pursue their villainy under
the formality of goodness, for they desire something that appears good to them and choose the
means by which to pursue it.
The specification of just what is authentically good requires further work in the sphere of prac-
tical reasoning, to be sure. For this reason the mainstream of classical natural law theorists (from
Sophocles and the Stoics, Cicero and Aquinas, Locke and the judges at Nuremberg, through a vari-
ety of present-day proponents) have relied on our knowledge of what human beings are and what
sorts of things genuinely fulfill or frustrate their growth and their lives (individually and as members
of societies of various sorts).
The proponents of the “basic goods” school of thought sometimes label their efforts a “new
natural law” theory that is derived analytically by examining a range of propositions about the
objects of human desire to identify self-evident fundamental goods (see, e.g., Finnis 2011). The
classical theory of the natural moral law, however, provides more compelling arguments than do
such intuitionist accounts. Proponents of the traditional form of natural law theory insist on the
need to understand human nature in order to see why a given practice is virtuous or vicious, why
it deserves to be regarded as obligatory or permissible or forbidden. Even though it goes behind
the scope of this chapter to deal with the brutalities that have unfortunately been committed in
the name of religion or the natural law, there is a suitable response, namely, that the very fact that
religious people have sometimes acted against their religion and against the natural moral law shows
us all the more reason why there must be an objective moral order with a transcendent basis, for it is
on this basis that we can recognize and condemn these brutalities. In fact, this is why any sound
ethics must distinguish between authority and the power with which authority is entrusted. It is
only by properly understanding the difference between authority and power that there is a basis for
distinguishing between what is genuinely authoritative and what is authoritarian. (For further dis-
cussion of this, see Simon 1962.)
Admittedly, anyone who does not recognize the possibility of moral precepts that are always
and everywhere in force will not find this argument for theism compelling, but for most of
human history the statement of these obligations and prohibitions has been taken to express
what we already know about their universally binding character. These are such precepts as that
the innocent may never be deliberately killed, that what belongs to others may not be taken
without their permission, that the spouses of others may not be legitimately pursued, and so
on. The content of these precepts is prior to its formulation in various religious traditions,
whether as the Golden Rule or as the Dao or as the Dharma. Yet the recognition of these basic
points in diverse traditions shows that there are invariant norms for human life that are recog-
nized across cultures.
As Russell Hittinger argues (2003), there are three main assumptions in any genuine form of
natural law theory: one that is anthropological, one of an epistemological sort, and one that is theo-
logical. With regard to the first point, unless there is such a thing as human nature (the thing, not
necessarily the specific terminology of “human nature”), there would be no justification for holding
that there are basic human rights that need to be respected regardless of whether it is convenient
and independently of whether those rights are yet codified in the positive law of any given culture.
The alternative to this position would inevitably require us to hold that ethical norms are imposi-
tions of restraint by force or by choice rather than discoveries by reason. The forcefulness of this
position is readily seen by considering the arguments against Thrasymachus voiced in the first book
of Plato’s Republic and again in the Gorgias.
The second point is epistemological. Unless human reason is capable of understanding human
nature and its intrinsic orientation toward things genuinely good, there will be no foundation for
any claims about moral truth that can legitimately be thought to be objective, universal, and intelligi-
ble. To be sure, establishing this point is a project beyond the scope of the present chapter, which is
concerned only to show the reasonableness of theism on the basis of certain assumptions (in this case,
the reality of such moral norms). There is much in the area of moral psychology that depends on prac-
tical reasoning about how to deliberate about possible courses of action so as to achieve various goals.
There would be no reason, however, to take such reasoning as based on anything but hypothetical
imperatives unless theoretical reasoning is able, in principle, to recognize ends that are given in our
nature prior to our purposive choice of means suitable to the realization of those ends (see “What Is
Natural Law? Human Purposes and Natural Ends” in Sokolowski 2006, 214–233).
The third aspect is theological, and it is, of course, this aspect that is most relevant to our
concerns in this volume. Unless there is a transcendent source for the natural law that human rea-
son can discover within human nature, there will be no sufficient ground for its categorical nor-
mativity. For the mainstream natural law thinkers such as Aquinas, this means a recognition that
God is the author of human nature and that the human nature that God designed includes an
intrinsic directedness (teleology) of human beings to what is for their genuine good. In his
accounts of this telic orientation, he includes detailed arguments about the inclination of human
beings at various levels, including the inclinations that human beings feel to sustain their lives by
seeking healthful food, cleanliness, and exercise; the inclinations to procreate and to live in famil-
ial societies; the inclinations to seek knowledge useful for their lives and for their eternal destiny;
the inclinations to love and be loved in appropriate ways; and the inclinations to create and sus-
tain other forms of society in which they can live and prosper. Aquinas is ever mindful that there
may be wayward inclinations as well as genuine ones, and so he has a detailed treatment of the
prudential discernments that reason must make to sort out the trustworthy inclinations from
those that are untrustworthy. In describing these aspects of moral psychology, he is ever at pains,
however, to locate the metaphysical grounding for all these claims in God’s creation of human
nature. For him, it is precisely by the way in which God included trustworthy telic inclinations
and the power of reason to discriminate trustworthy from unworthy inclinations that God pro-
mulgated the natural moral law for human beings.

The naturalized versions of ethics that flourished in the early modern period (Hugo Grotius,
Samuel von Pufendorf, Locke, and others) had adopted as their starting point a deliberate rejection
of this transcendent source (that is, as if God did not exist—si velut Deus non daretur). By doing so,
however, they also lost the grounding for their claim to categorical normativity and could at best
claim a kind of hypothetical imperative that quickly degenerated into an ethics of rules made on
the basis of enlightened self-interest. From this perspective, Immanuel Kant (whose third formula-
tion of the categorical imperative preserves the very terminology natural law) saw deeply into the
character of the theoretical problem here and tried to replace the missing third leg of this stool
(the theological) by reassigning the role of promulgating the natural law to human reason. In doing
so, however, he needed to stress that it was practical reason (not theoretical reason) that generates
the categorical imperative (see Rist, chap. 7). No longer is ethics rooted in a divine source but rather
in practical reason and the primacy of autonomy. The danger of this approach is that what is
asserted on the basis of will may be denied on the same grounds. To resist the charge of arbitrari-
ness, however, requires that there be a basis in what we do not create, control, or alter—that is,
in what has traditionally been called nature and the designer of nature.
FINAL RECKONING
The purpose of the second portion of this chapter has been to make the case that morality is ulti-
mately dependent on God. In its first part we saw the usefulness of philosophical reason for clearing
away various obstacles that may prevent us from seeing how well theism answers some of the peren-
nial questions of human existence.
It remains my hope that the discussion of these great questions between contemporary theists
and atheists may be an occasion for fruitful dialogue. The ultimate grounding of moral norms seems
to me to be a question on which much turns, as is the question of the ultimate source of the deep
intelligibility of all reality. The story of Augustine’s search for compelling answers may prove to be
an opening to others to conduct a similar quest, and the study of the cross-cultural recognition of
certain moral truths may be conducive to an investigation of the requirements for norms that tran-
scend particular societies and contingent choices.
Bibliography
Anscombe, Elizabeth. “Modern Moral Philosophy.” Philosophy 33, no. 124 Ralph McInerny, edited by Matthew Lamb, 54–73. Washington, DC:
(1958): 1–19. Catholic University of America Press, 2016.
Anselm. Proslogion. In Anselm of Canterbury: The Major Works, edited by Lee, Patrick, and Robert P. George. Body-Self Dualism in Contemporary
Brian Davies and G. R. Evans, 82–104. New York: Oxford University Ethics and Politics. New York: Cambridge University Press, 2008.
Press, 1998.
Lewis, C. S. The Problem of Pain. In The Complete C.S. Lewis Signature
Arendt, Hannah. Love and Saint Augustine. Edited, with an interpretive Classics. New York: HarperOne, 2017.
essay, by Joanna Vecchiarelli Scott and Judith Chelius Stark. Chicago:
University of Chicago Press, 1996. First published 1929. Locke, John. An Essay concerning Human Understanding. Edited by Peter H.
Nidditch. Oxford: Clarendon Press, 1975.
Augustine. Confessions. Translated by F. J. Sheed. Edited by Michael P.
Foley. Indianapolis, IN: Hackett, 1993. MacIntyre, Alasdair. After Virtue. London: Duckworth, 1981.
Chesterton, G. K. Orthodoxy. Wheaton, IL: H. Shaw, 1994. First published McDowell, John. Mind and World. Cambridge, MA: Harvard University
1908. Press, 1996.
Finnis, John. Natural Law and Natural Rights. 2nd ed. New York: Oxford Nagel, Thomas. Mind and Cosmos: Why the Materialist Neo-Darwinian
University Press, 2011. Conception of Nature Is Almost Certainly False. New York: Oxford
Foot, Philippa. Natural Goodness. New York: Oxford University Press, 2001. University Press, 2012.
Hittinger, Russell. The First Grace: Rediscovering the Natural Law in a Post- Plato. Republic and Gorgias. In Plato: Complete Works, edited by John M.
Christian World. Wilmington, DE: ISI Books, 2003. Cooper. Indianapolis, IN: Hackett, 1997.
Koterski, Joseph W. “The Concept of Nature: Philosophical Reflections in Plotinus. Enneads. 7 vols. Translated by A. H. Armstrong. Cambridge, MA:
Service of Theology.” In Theology Needs Philosophy: Essays in Honor of Harvard University Press, 1966–1988.
Pseudo-Dionysius. The Divine Names. In Pseudo-Dionysius: The Complete Simon, Yves R. A General Theory of Authority. Notre Dame, IN: University
Works. Translated by Colm Luibheid and Paul Rorem. Mahwah, NJ: of Notre Dame Press, 1962.
Paulist Press, 1987.
Sokolowski, Robert. Christian Faith and Human Understanding: Studies on
Rist, John M. Real Ethics: Rethinking the Foundations of Morality. New York: the Eucharist, Trinity, and the Human Person. Washington, DC: Catholic
Cambridge University Press, 2002. University of America Press, 2006.
Spitzer, Robert. New Proofs for the Existence of God: Contributions of
Rorty, Richard. Philosophy and the Mirror of Nature. Princeton, NJ:
Contemporary Physics and Philosophy. Grand Rapids, MI: Eerdmans,
Princeton University Press, 2009.
2010.
Sanford, J. J. Before Virtue: Assessing Contemporary Virtue Ethics. Washing- Thomas Aquinas. Summa theologica. 3 vols. Translated by the Dominican
ton, DC: Catholic University of America Press, 2015. Fathers of the English Province. New York: Benziger Bros., 1947–1948.

TOPIC 20
Final Reckonings: Atheism

Graham Oppy
Monash University, Australia
This chapter explains how the sum of the considerations in the previous chapters fit together in a
comprehensive case for preferring atheism to theism.
WORLDVIEWS AND WAYS OF LIFE
There are so many different controversies about theism and atheism that it is hard to obtain a syn-
optic view of all of them. There is disagreement about what theism and atheism are. Some take
them to be claims. Some take them to be beliefs. Some take them to be theories. Some take them
to be worldviews (comprehensive theories). Some take them to be ways of life. I think that we do
best to suppose that they are claims. The claims in question can be believed; they can belong to the-
ories; they can belong to worldviews; their acceptance can be central to ways of life.
Given that theism and atheism are claims, there is disagreement about the nature of these
claims. Some take theism to be the claim that there are gods; others take theism to be the claim that
God exists. Some take atheism to be the claim that there are no gods; others take atheism to be the
claim that God does not exist. I think that we do best to take theism and atheism to be general
claims: that there are gods and that there are no gods, respectively. Note that, following standard
contemporary practice in philosophy, I suppose that there are Fs is true if and only if there is at least
one F. Although some do not like this way of stating matters, I shall suppose, moreover, that God
exists if and only if there is exactly one god. (Some suppose that if God is a god, then it follows that
there could be other gods. But this does not follow. Suppose, as many theists do, that if there is a
god, then there is just one god and it is necessary that that god exists and also necessary that no
other god exists. Suppose that there is a god. Given our assumptions, it follows not only that this
god is God but also that there could not possibly be any other god or God.)
Even given this account of theism and atheism, there is disagreement about what it takes to be
a theist or an atheist. For example, some think that atheists are those who fail to accept that there
are gods. I think that we do best to adopt a fourfold scheme of classification: theists accept that
there are gods; atheists accept that there are no gods; agnostics suspend judgment on the question
of whether there are gods; and innocents have never so much as entertained the thought that there
are gods. Atheists, agnostics, and innocents alike fail to accept that there are gods.
Even given this account of theists and atheists, there is disagreement about exactly how accep-
tance should be understood. Must one believe the central claim (or, perhaps, other claims that entail
the central claim)? Is it enough if one merely hopes that the central claim is true? How should we
classify someone who, while failing to believe that God exists, nonetheless hopes that God exists?
While I propose to work with characterizations that invoke belief, I note that there are other atti-
tudes—including hope and faith—that relate to belief in interesting and important ways.
There is disagreement about whether belief alone suffices to characterize theists and atheists.
Some suppose that the characterization should advert to claims to knowledge. Some suppose that
679
the characterization should advert to claims to proof. Some

KEY CONCEPTS suppose that the characterization should advert to claims to cer-
tainty. Some suppose that the characterization should advert to
claims to resistance to revision of opinion. I think that we
Infinite Regress should reject all of these suggestions. For theists and atheists
Initial Singularity alike, there is a distribution across the entire spectrum from
James’s Wager dogmatic conviction to hesitant acceptance. Some make claims
to proof and knowledge; others do not. Here, as elsewhere, we
Natural Causal Reality/Nonnatural should not mistake the part for the whole.
Causal Reality There is disagreement about what we are primarily inter-
Pascal’s Wager ested in evaluating when we turn our attention to atheism and
Quantum Mechanics theism. Some suppose that we wish to evaluate the truth of the
central claims. Some suppose that we wish to assess the rational-
Relativity ity of the central beliefs. Some suppose that our focus should be
on theories that include or entail the central claims. Some sup-
pose that our focus should be on worldviews that include or
entail the central claims. Some suppose that our focus should
be on ways of life that are appropriately related to the central claims. I think that we do not need
to choose. The questions about truth of claim and rationality of belief are important, but they can
properly be addressed only by thinking about the truth of theory and worldview and the rationality
of way of life.
Given that we wish to assess the truth of worldviews and the rationality of ways of life, we imme-
diately run into significant difficulties. There is enormous diversity in theistic worldview and theistic
way of life, and there is enormous diversity in atheistic worldview and atheistic way of life. Many,
but not all, theists are religious; many, but not all, theists belong to organized religions. Many, but
not all, theists are monotheists; many, but not all, theists believe in an Abrahamic God. Some mono-
theists are politically and doctrinally conservative; some are not. Among Christian theists, some think
that God is literally a person; some do not. Similar diversities of opinion exist among atheists. Many,
but not all, atheists are nonreligious; many, but not all atheists, do not belong to organized religions.
Many, but not all, atheists are naturalists; many, but not all, atheists suppose that there are none but
natural causal entities with none but natural causal powers. Some atheists are politically and socially
liberal; some are not. Some atheists are materialists; some atheists are physicalists; some atheists are
humanists; some atheists are communists; many atheists are none of these things.
Given the diversity of worldviews and ways of life, it is massively implausible to suppose that all
of the worldviews and ways of life on one side are better than all of the worldviews and ways of life
on the other side. Some theistic worldviews and ways of life are better than some atheistic world-
views and ways of life; some atheistic worldviews and ways of life are better than some theistic
worldviews and ways of life. Indeed, some theistic worldviews and ways of life are better than other
theistic worldviews and ways of life; and some atheistic worldviews and ways of life are better than
other atheistic worldviews and ways of life.
To undertake a philosophically interesting project, it seems that we ought to be comparing best
theistic worldviews and ways of life with best atheistic worldviews and ways of life. At least in prin-
ciple, given that we have a best theistic worldview WT and a best atheistic worldview WA, we can
see how to compare them. First, we formulate the worldviews: we make them fully explicit. Second,
we weigh the virtues of the worldviews: we check to see which one is better, working from an agreed
set of criteria. Similarly, at least in principle, given that we have a best theistic way of life LT and a
best atheistic way of life LA, we can see how to compare them. First, we formulate the ways of life:
we make them fully explicit. Second, we weigh the virtues of the ways of life: we check to see which
one is better, working from an agreed set of criteria.
In practice, there are serious obstacles. If best worldviews are theories of everything, then we
cannot make them fully explicit. Even if best worldviews are merely our best current approximations
to theories of everything, we cannot make them fully explicit. No single person or small team of
people has knowledge of more than the tiniest fraction of our collective best current approximations
to theories of everything. The best we can do, it seems, is to put together theories that address
everything that is currently taken to be relevant to the question of whether there are gods. If the
claim that p is relevant to the question of whether there are gods, then exactly one of p and ∼p is
included—perhaps by entailment—in each fully articulated worldview.
Given that we have WT and WA, we can proceed to evaluate them. How do we do that? A first
step is to check WT and WA for logical consistency. If one is logically consistent and the other is
not, then the logically consistent one wins immediately. Supposing that WT and WA are logically
consistent, we move on to the next step: determining which is more virtuous. To make this deter-
mination, we need a canonical list of virtues and a means of weighing them. There is disagreement
about both the members of the canonical list of virtues and the means of weighing them.
When we compare WT and WA, we will find that there are many claims on which WT and WA
agree and many claims on which WT and WA disagree. I think that, in the context of the assessment
of the disagreement between WT and WA, we should take the claims on which they agree to be the
evidence that is relevant to the assessment of the disagreement. Furthermore, I think that we should
suppose that we should prefer whichever of the theories best manages the trade-off between mini-
mizing theoretical commitments and maximizing explanatory breadth and depth with respect to evi-
dence. On this view, there is just one theoretical virtue: optimizing the trade-off between minimiz-
ing theoretical commitments and maximizing breadth and depth of explanations of evidence. There
are two dimensions to this theoretical virtue. On the one hand, holding fixed explanatory breadth
and depth, theories are better insofar as they minimize commitments to numbers and kinds of enti-
ties, primitive ideology, and primitive principles. On the other hand, holding fixed theoretical com-
mitments, theories are better insofar as they give broader and deeper explanations of all of the rele-
vant evidence. There are many other qualities that are commonly said to be theoretical virtues: fit
with established knowledge, beauty, simplicity, unification, and so forth. In the context of our com-
parison between WT and WA, however, these are all plausibly incorporated in our single theoretical
virtue. Beauty, simplicity, unification, and so forth are all properly subsumed under the minimiza-
tion of primitive principles, primitive ideology, and numbers and kinds of entities. Fit with estab-
lished knowledge is guaranteed by the fact that all relevant evidence is taken into account.
Even if we agree that our principal interest lies in determining which of WT and WA best man-
ages the trade-off between minimizing theoretical commitments and maximizing explanatory
breadth and depth, we find significant disagreement about how to make this determination. Many
philosophers suppose that we can make a Bayesian calculation:

Pr WT =E Pr WT Pr E=WT
¼ , where E is the total evidence.
Pr ðWA =EÞ Pr WA Pr E=WA
Often enough, to make this calculation, we need a method for arriving at numerical values for
the prior probabilities—Pr (WT) and Pr (WA)—and the likelihoods—Pr (E/WT) and Pr (E/WA). In
the case of interest, it is highly controversial whether we can get agreed assignments for these
numerical values. There are special cases in which we do not need numerical values. If one theory
does better than a second on both the count of minimization of commitments (prior probability)
and the count of maximizing explanatory breadth and depth (likelihood), then, without any calcu-
lation, we can conclude that the first theory is more virtuous than the second.
If we suppose that, at least in principle, we can determine which of WT and WA is most virtu-
ous, then perhaps we can use this determination to decide which of LT and LA is better. The guid-
ing idea is that goodness of worldview is a trumping consideration. If one worldview is better than a
second, then the best ways of life that embed the first worldview are better than the best ways of life
that embed the second worldview. We shall turn our attention to the dispute about worldviews and
come back to the dispute about ways of life toward the end of this chapter.
The theistic contributors to this volume disagree among themselves about the full list of claims
that belong to WT, and the atheistic contributors to this volume disagree among themselves about
the full list of claims that belong to WA. Nonetheless, we can identify some claims that almost all
contributors to this volume agree belong only to WT: God exists, God is omniscient, God is omnip-
otent, God is perfectly good, God is the sole creator of natural reality, God is the sole designer of

natural reality, God is worthy of worship, and science is our touchstone for identifying merely nat-
ural causal entities, merely natural causal powers, and merely natural causal histories. Moreover, I
think, we can identify some claims that almost all contributors to this volume agree belong only
to WA: there are none but natural causal entities with none but natural causal powers, nothing is
omniscient, nothing is omnipotent, nothing is perfectly good, natural reality has no cause, natural
reality has no designer, nothing is worthy of worship, and science is our touchstone for identifying
all causal entities, all causal powers, and all causal histories.
Supposing—or, perhaps, pretending—that we have a good enough grasp of the content of WT
and WA, what can we say about their comparative theoretical virtue? I start by considering what WT
and WA have to say about natural causal reality. The most important point to note is that, by and
large, WT and WA are in agreement about natural causal reality. By and large, WT and WA agree
about what natural objects there are, what natural events occur, what natural causal powers are pos-
sessed by natural objects, and so on. However, despite this broad agreement about natural causal
reality, there is no similar agreement about nonnatural causal reality. Whereas according to WA
there are only natural objects, natural events, natural causal powers, and so forth, according to
WT there are, in addition, nonnatural objects, nonnatural events, nonnatural causal powers, and
so on. Because there is broad agreement between WT and WA on natural causal reality and because
WA supposes that causal reality is exhausted by natural causal reality and WT supposes that there is
also nonnatural causal reality—involving additional objects, events, powers, principles, and so on—
it is clear that WA scores better than WT on the count of theoretical commitments. The theoretical
commitments of WT outweigh the theoretical commitments of WA.
What I just said is not exactly right. For many theists, there is only broad agreement with athe-
ists about natural causal reality. Some theists suppose that there is special divine action in natural
causal reality: some historical events involve more than merely natural objects with merely natural
causal powers. Some theists suppose that there is also general divine action that enables and con-
serves natural causal reality. However, setting aside disputes about the details of history, these sup-
positions all involve objects, events, powers, and principles to which atheists have no commitment.
Where there is the kind of disagreement about natural causal reality that I have just identified, it is
always because theists take on additional theoretical commitments that atheists do not have. I think
that this point about the weight of theoretical commitment is uncontroversial: theists, but not athe-
ists, invest in “transcendence of natural causal reality.”
What about the details of history? Suppose, for example, that WT states that at a certain point
in history God became incarnate in a particular human being. Of course, WA does not include—or
entail—such a claim. However, WA and WT agree that there are many people who believe that God
became incarnate in a particular human being at a particular point in history, and there are many
old documents that purport to be records of a time when God became incarnate in a particular
human being. When we consider the explanatory breadth and depth of WA and WT, one thing that
we shall take into account is the adequacy of the explanations that they give of the distribution of
belief and the contents of the historical documents. However, so long as we restrict our attention
to agreed history—the historical claims on which WA and WT agree—we see that the theoretical
commitments of WT outweigh the theoretical commitments of WA. Of course, it remains to be
considered which of WT and WA gives the best explanation of the total data; if the explanation
given by WT is sufficiently superior to the explanation given by WA, then we may have reason to
accept the claim that at a certain point in history God became incarnate in a particular human
being.
I anticipate that some will think that I have moved too quickly. Before we turn to consider
comparative theoretical virtues, we need to consider logical consistency. If one of WA and WT is
logically inconsistent, then we can declare a victor without any further consideration of theoretical
virtue. Historically, there has been no shortage of champions of the view that one or the other of
WA and WT is logically inconsistent or otherwise defeated on entirely a priori grounds.
Some atheists claim that the uniquely theistic ideology—“god,” “omnipotent,” “transcendent,”
“trinity,” and so on—is meaningless. Some atheists claim that the uniquely theistic theory is logi-
cally inconsistent. (Some atheists claim to show that it is logically inconsistent to suppose that
something is omnipotent. Some atheists claim to show that it is logically inconsistent to suppose
that something is omniscient. Some atheists claim to show that it is logically inconsistent to suppose
that something is perfectly just and perfectly merciful. And so on.) Some atheists claim that the total
theistic theory is logically inconsistent. (Some atheists claim to show that it is logically inconsistent
to suppose that God and evil both exist. Some atheists claim to show that it is logically inconsistent
to suppose that God and horrendous evil both exist. Some atheists claim to show that it is logically
inconsistent to suppose that God and enormous quantities of horrendous evil both exist. Some athe-
ists claim to show that it is logically inconsistent to suppose that God exists and remains hidden
from large parts of humanity. And so forth.)
Some theists claim that the uniquely atheistic theory is logically inconsistent. (Some theists
claim to show that it is logically inconsistent to suppose that there is not something than which
no greater can be conceived. Some theists claim to show that it is logically inconsistent to suppose
that there is nothing that possesses all perfections. And so on.) More theists claim that the total
atheistic theory is logically inconsistent. (Some theists claim to show that it is logically inconsistent
to suppose that there is a causal reality but no God. Some theists claim to show that it is logically
inconsistent to suppose that there is mindedness but no God. Some theists claim to show that it
is logically inconsistent to suppose that there are moral facts but no God. And so forth.)
I think that all of the extant versions of all of these claims—against WA and WT—fail. There are
no extant derivations of contradictions from sets of claims all of which plausibly belong to WA, and
there are no extant derivations of contradictions from sets of claims all of which plausibly belong to
WT. Almost always, when someone successfully derives a contradiction from a relevant set of claims,
at least one of the claims is such that it is obvious that it does not belong to the target worldview.
In other cases, where it is plausible to attribute contradiction to some believers of a particular persua-
sion, a relatively minor adjustment restores consistency to their views without taking them any dis-
tance toward the opposing camp. (I defend these views at length in Oppy 2006.) I do not believe that
these views can be fully defended except by exhaustive examination of cases, and that is no short task.
However, it is worth noting that it is prima facie completely implausible to suppose that there are
extant derivations of the inconsistency of WA or WT. It is pretty much inexplicable why there are pro-
fessional philosophers of both stripes with heavy-duty logical expertise if there are derivations that
show that the best views on one side are logically inconsistent. Of course, that is not to say that there
will not be logical inconsistences in WA or WT that are discovered in the future. Given a couple of
thousand years full of nothing but false alarms, I would not be holding my breath.
I think, then, that we should turn our attention to the data. WA has fewer commitments than
WT. Perhaps WT has greater explanatory breadth and depth than WA. If so, only a careful weighing
can disclose whether to prefer WA or WT or suspension of judgment. There is much data to con-
sider. Here are some of the things that have been claimed to be data relevant to the choice between
WA and WT: the existence of causal reality, the shape of causal reality, the fine-tuning of our part of
causal reality, the presence of organized complexity in our part of causal reality, the presence of liv-
ing organisms in our part of causal reality, the presence of conscious organisms in our part of causal
reality, the presence of rational organisms in our part of causal reality, the presence of intelligent
organisms in our part of causal reality, the applicability of science to our part of causal reality, the
applicability of mathematics to our part of causal reality, the presence of morally responsive organ-
isms in our part of causal reality, the presence of aesthetically responsive organisms in our part of
causal reality, the presence of normatively responsive organisms in our part of causal reality, the
presence of a significant body of reports of religious experience in our part of causal reality, the pres-
ence of a significant body of reports of miracles in our part of causal reality, the history of religion in
our part of causal reality, the distribution of suffering in our part of causal reality, the distribution of
virtue and flourishing in our part of causal reality, the distribution of atheistic belief in our part of
causal reality; the practical consequences of believing WA or WT, the content of theories that seek
to explain the distribution of religious belief in our part of causal reality, and so on.
Proponents of WA will suppose that, when all relevant data are weighed, WT does not have an
explanatory advantage sufficiently large to outweigh the advantage that WA has in virtue of its more
minimal theoretical commitments. Conversely, proponents of WT will suppose that, when all

relevant data are weighed, WT does have an explanatory advantage sufficiently large to outweigh the
advantage that WA has in virtue of its more minimal theoretical commitments. Of course, propo-
nents of WA and WT may agree that, for a great many of the cases mentioned, there is no explana-
tory advantage to either side; more strongly, proponents of WA and WT may agree that, in a few
cases, there is an advantage to the other side. However, it is clear that there are different ambitious
strategies available on either side. Some theists might try to argue that, on each piece of data, WT
does better than WA, and some theists might try to argue that there is no piece of data on which
WA does better than WT. Similarly, some atheists might try to argue that, on each piece of data,
WA does better than WT, and some atheists might try to argue that there is no piece of data on
which WT does better than WA.
In what follows, I shall argue that none of the data that I mentioned two paragraphs back favors
WT over WA; I shall not, however, argue that any of these data favors WA over WT. In some cases,
this leads me to disagree with what atheists and theists have said in earlier chapters of this book; in
other cases, it leads me to agree with what atheists and theists have said in earlier chapters of this
book. My arguments will, of course, leave it open that other data favor WT over WA; there is no
way that I can consider all of the relevant data in the space allotted to this chapter. (For much fuller
development of all of the coming arguments, see Oppy 2013, 2018, 2019).
THE EXISTENCE OF CAUSAL REALITY
Why is there a network of causes and effects? In principle, it seems that there are two ways that one
might answer this question: (1) There is no reason why there is a network of causes and effects
rather than no network of causes and effects (brute fact). (2) There is a network of causes and effects
because it is impossible that there not be a network of causes and effects (necessary fact).
If it is a brute fact that there is a network of causes and effects, then, a fortiori, WA and WT do
an equally good job of accommodating this fact, because neither offers any explanation for it. If it is
a necessary fact that there is a network of causes and effects, then, a fortiori, WA and WT do an
equally good job of accommodating this fact, because each claims that there is some initial part of
causal reality that exists of necessity and enters into causal relations of necessity. If we suppose that
causal reality does not involve an infinite regress, then, plausibly, WT says that God exists of neces-
sity and enters into causal relations of necessity, and WA says that the initial singularity exists of
necessity and enters into causal relations of necessity. In point of explaining why causal reality exists,
neither view has an advantage over the other.
Some may say that there is a third way in which one might seek to answer the question: one
might say that causal reality exists because it is good that it exists. I am happy to rule this suggestion
out of court. The existence of causal reality cannot be explained by the goodness of its existence.
Suppose I am wrong about that. Then, it seems, the explanation is equally available to WA and
WT; it seems to make just as much sense to suppose that the initial singularity exists because it is
good that it exists as it does to suppose that God exists because it is good that God exists. It is no
greater cost to include the relevant claim in one view rather than the other.
THE SHAPE OF CAUSAL REALITY
Why does causal reality have the shape that it does? In principle, it seems that there are just two
ways of answering this question: (1) There is no reason why causal reality has the shape that it does
(brute fact). (2) It is necessary that causal reality has the shape that it does (necessary fact). In prin-
ciple, there are two shapes that causal reality might have: infinite regress and starting point. If there
is an infinite regress, then the atheist supposes that there is an infinite natural causal regress, and the
theist supposes that there is an infinite natural causal regress and/or an infinite nonnatural causal
regress. Whether it is a brute fact or a necessary fact that there is an infinite causal regress, there
is no explanatory advantage to either side. If there is a starting point, then the atheist supposes that
causal reality begins with an initial singularity, and the theist supposes that causal reality begins with
God. Whether it is a brute fact or a necessary fact that causal reality has a starting point, there is no
explanatory advantage to either side.
It is important to recall that WT and WA are best worldviews. What stand they take on brute
fact/necessary fact and infinite regress/starting point depends on which views on these matters are
best supported by the relevant data. Effectively, we are faced with two decisions about “adjustable
parameters” in the content of WT and WA; however the chips fall, the explanation afforded by
WA is no worse than the explanation afforded by WT.
THE FINE-TUNING OF NATURAL REALITY
It is an open question whether our universe is part of a multiverse. There is no expert consensus—
for example, among cosmologists—on this matter. To simplify the coming discussion, I shall con-
tinue under the assumption—or, perhaps better, pretense—that our universe is the whole of natural
reality. Under that assumption, the topic for discussion is the fine-tuning of natural reality.
In the current standard model of particle physics, there are twenty-five freely adjustable para-
meters. Of course, the current standard model is not final physics. It includes both general relativity
and quantum mechanics, even though these theories are jointly inconsistent at sufficiently high
energy levels. It is not clear, then, how confident we should be that final physical theory will contain
so many—or even any—freely adjustable parameters.
Let us assume—or, perhaps better, pretend—that there are. Moreover, let us assume that—as is
the case for the twenty-five freely adjustable parameters in the current standard model of particle
physics—if any one of these freely adjustable parameters took a value only slightly different from
the value that it actually takes, then either our universe would have been short-lived, or it would
always have consisted of more or less nothing but empty space. That is to say, if the freely adjusted
parameters were not finely tuned to the values that they actually have, our universe would not have
contained atoms, molecules, proteins, cells, microorganisms, plants, animals, humans, planets, stars,
galaxies, galactic clusters, and so forth.
Does either WA or WT better explain the fine-tuning of freely adjustable parameters? For each
of the freely adjustable parameters, the following is true: either its value is settled at all points in the
causal order, or there is some point in the causal order at which there is a transition between the
value’s not being settled and its being settled. If the value of a freely adjustable parameter is settled
at all points in the causal order, then either it is necessary that the freely adjustable parameter takes
the value that it does, or it is a brute fact that the freely adjustable parameter takes the value that it
does. Either way, there is no advantage here for WA or WT. (Note that freely adjustable parameters
are parameters that are “put in by hand”: their values are not determined by further theoretical con-
siderations. It is consistent with a parameter’s being “freely adjustable” that it takes the value that it
actually takes as a matter of necessity.)
If, however, the value of a freely adjustable parameter is settled only at some noninitial point in
the causal order, then, at the point where the transition occurs, there is a range of possible values
that the parameter could take, and the adoption of one of those values rather than any of the others
is simply a matter of chance. In this case, it is a brute fact that the parameter takes the value that it
actually takes (rather than any of the other values that it might have taken). Again, there is no
advantage here for either WA or WT. (In the case of WA, the story might be something like this:
There is a chance distribution associated with a phase transition for the total state of the universe.
When the phase transition occurs, the “freely adjustable parameter” takes on one of the possible
values that figures in the chance distribution. In the case of WT, the story might be something like
this: There is a chance distribution associated with the choice of a particular divine action on the
universe. At the point where choice of the divine action occurs, the freely adjustable parameter is
set to one of the possible values that figures in the chance distribution. Note that neither case
requires any assumption about the shape of the chance distribution; in particular, neither case
requires us to assume that the chance distribution is flat.)

ORGANIZED COMPLEXITY
Why does our universe contain organized complexity? There are two cases to consider. On the one
hand, it might be that, at every point in the causal order, it is settled that our universe contains
organized complexity. In that case, it is either a necessary fact or a brutely contingent fact that
our universe contains organized complexity. As we have already seen, both of these positions are
available in equal measure to theists and naturalists, so there can be no explanatory advantage either
way. On the other hand, it might be that there is a point in the causal order at which there is a tran-
sition from its not being settled that our universe contains organized complexity to its being settled
that our universe contains organized complexity. In that case, it is simply a matter of chance that
our universe contains organized complexity. Again, as we have seen, this position is available in
equal measure to theists and naturalists. So the presence of organized complexity in our universe
gives no explanatory advantage to either side.
I anticipate that some might insist that, where the values are already settled in the initial state,
there is some difference in the availability of options to theist and naturalist. In this case, there is no
relevant difference between the values taken by the parameters and God’s preferences for the values
taken by the parameters. Because we are talking about the initial causal state, the only options are
brute contingency and brute necessity, and there is nothing that fits one of these options better than
one of the competing positions.
I anticipate that some might insist that, where there is a (noninitial) transition from the values
being unsettled to the values being settled, there is some difference in the options available to theist
and naturalist. How could that be? Suppose that the theist says that God settles the values by mak-
ing a free rational choice. At all prior points in the causal order, the matter is not settled: there is a
range of live possibilities. Moreover, there is a probability distribution over those possibilities: each
has a particular possibility associated with it. Then, with the making of the free rational choice,
we have a “collapse” of the possibilities: the values of the parameters are fixed to their actual values.
When we ask the question why God chose as God actually did, rather than choosing in one of the
other ways that God could have chosen, there is nothing further for the theist to say. The prior
causal state and the possibility distribution are the only relevant available factors. The key point
to note is that, had God made a different choice from among the available alternative choices that
God could have made, we would have had exactly the same relevant available explanatory factors.
Moreover, and crucially, the naturalist has equal access to those relevant available explanatory fac-
tors. Sure, the naturalist does not think that there is any free rational choice involved, but, nonethe-
less, the naturalist has equal access to prior causal state and probability distribution.
LIVING ORGANISMS
Why does our universe contain living organisms? Initially, one might be tempted to answer this
question merely by appealing to the evolutionary sciences. Whether one is a best theist or a best nat-
uralist, one accepts that there is a fully natural history that proceeds from earlier states in which
there are no living organisms in the universe to later states in which there are living organisms in
the universe. One might think that there is something incomplete about this explanation, however;
while it is true that the evolutionary sciences trace the appearance and development of life in the
universe, they say nothing about why those earlier states in which there were no living organisms
were apt for the later appearance of life in the universe. Those who take this line will insist that a
complete explanation of why the universe contains living organisms must have two parts: it must
conjoin to the fully natural history of the evolutionary sciences a further explanation of why earlier
states in which there were no living organisms were apt for the later appearance of life. Even so, it
should be clear enough how discussion of the further explanation goes: either it has always been set-
tled that there will be states apt for the later emergence of life in our universe, or else there is a tran-
sition from a state in which it has not been settled that there will be states apt for the later emer-
gence of life to a state in which it is settled that there will be states apt for the later emergence of
life. Whichever way it goes, there is no advantage to theism or naturalism that emerges.
CONSCIOUS ORGANISMS
Why does our universe contain conscious organisms? To answer this question, we need to think
about what it is in virtue of which conscious organisms are conscious. A plausible view is that for
organisms to be conscious simply is for certain kinds of neural processing to be occurring in those
organisms. Another—in my opinion, less plausible—view is that for organisms to be conscious is
for them to be in states that are emergent from certain kinds of neural processing that occurs in
those organism. A third—in my opinion, even less plausible—view is that for organisms to be con-
scious is for them to be in states that are merely correlated with certain kinds of neural processing
that occurs in those organisms. Whichever way we go, there are theistic and naturalistic versions
of the account of what makes conscious organisms conscious. Consequently, no matter what turns
out to be the best way to go, there is no advantage to theism or naturalism that emerges.
INTELLIGENT ORGANISMS
Why does our universe contain intelligent organisms? To answer this question, we need to think
about what it is in virtue of which intelligent organisms are intelligent. The right answer to this ques-
tion, I think, is that intelligent organisms are intelligent because they have central nervous systems
attached to general purpose neural processing. Organisms such as anaerobic marine bacteria are not
intelligent; they do not have central nervous systems attached to general purpose neural processing.
By contrast, organisms such as Caenorhabditis elegans—a particular kind of roundworm—are intelli-
gent; they do have central nervous systems attached to general purpose neural processing. Of course,
C. elegans is not very intelligent, but it does make decisions about which gradients in its environment
to follow on given occasions. Because the emergence of intelligent organisms in our universe falls
under the account of the emergence of life in our universe, there is nothing more that we need to add.
RATIONAL ORGANISMS
Why does our universe contain rational organisms? To answer this question, we need to think about
what it is in virtue of which rational organisms are rational. The (roughly) right answer to this ques-
tion, I think, is that rational organisms are rational because they are engaged in certain kinds of neu-
ral processing that have been appropriately shaped by local, social, and evolutionary history and that
are appropriately causally related to the environments in which those organisms are located. Given
that, for example, human reasoning capacities are a socially filtered mix of evolutionary adaptations
and exaptations, the emergence of rational organisms in our universe falls under the account of the
emergence of life in our universe. There is nothing more that we need to add.
APPLICABILITY OF SCIENCE
What accounts for the success of science in explaining natural phenomena? Initially, one might be
tempted to answer this question in the following terms. At least roughly, science is a collective
enterprise of data-driven description, prediction, and understanding in which universal expert agree-
ment functions as a regulative ideal. Given this account of what science is, we expect that (1) repro-
ducibility, parsimony, and consilience are fundamental scientific values; (2) strict protocols govern
scientific experimentation and both collection and analysis of data; and (3) significant institutions
are devoted to protecting the integrity of scientific investigation, publication, recognition, and
reward. Plausibly, then, if anything is going to produce satisfactory explanations of natural phenom-
ena, it will be science. It might be objected that this is not a complete answer to our question: we
have still not been told why science does succeed. What is it about the universe that makes it ame-
nable to scientific investigation? There are two cases to consider. Either, at every point in the causal
order, it is settled that our universe is amenable to scientific investigation, or there is a point in the

causal order at which there is a transition from our universe not being amenable to scientific inves-
tigation to its being amenable to scientific investigation. If, at every point in the causal order, it is
settled that our universe is amenable to scientific investigation, then it is either a necessary fact or
a brute fact that our universe is amenable to scientific investigation. If there is a point in the causal
order at which there is a transition from our universe not being amenable to scientific investigation
to its being amenable to scientific investigation, then it is simply a matter of chance that our uni-
verse is amenable to scientific investigation. No matter how it plays, there is no explanatory advan-
tage to either theism or naturalism.
APPLICABILITY OF MATHEMATICS
What explains the applicability of mathematics to our universe? Let us cut straight to the chase.
There are the same two cases to consider that there were in our discussion of the applicability of sci-
ence to our universe. The upshot of the discussion is exactly the same. No matter how it plays, there
is no explanatory advantage to either theism or naturalism.
MORALLY RESPONSIVE ORGANISMS
What explains the presence of morally responsive organisms in our universe? Initially, one might be
tempted to answer this question by pointing out that we should expect rational organisms to be
morally responsive. Certainly, in the case of human beings, our rational capacities emerged in tan-
dem with our moral capacities: we can see developing rational and moral capacities in our nearest
primate cousins. Moreover, moral norms—like rational norms—are universal in human societies.
For example, there is no human society that lacks a norm against killing: in every human society,
killing is prohibited except in special cases (which might include, for example, self-defense, protect-
ing kith and kin, fighting a just war, carrying out state-sanctioned executions of criminals, using ani-
mals for food, using animals for entertainment, abortion, and euthanasia). Of course, there is dis-
agreement about permissible exceptions, but that is just disagreement about special cases. The
upshot seems pretty clear that the discussion of the presence of morally responsive organisms in
our universe falls under the discussion of the presence of rational organisms in our universe.
I anticipate that some may insist that the important question about morality is not touched by
the preceding discussion. What we want to know is how moral truth is established. Suppose we grant
that it is true that it is wrong to kill (except in special cases, if there are any). What makes it the case
that it is wrong to kill? There are two cases to consider. Either, at every point in the causal order, it is
settled that it is wrong to kill, or else there is a point in the causal order at which there is a transition
from its not being settled that it is wrong to kill to its being settled that it is wrong to kill. If, at every
point in the causal order, it is settled that it is wrong to kill, then either it is a necessary fact or a brute
fact that it is wrong to kill. If there is a point in the causal order at which there is a transition from its
not being settled that it is wrong to kill to its being settled that it is wrong to kill, then it is simply a
matter of chance that it is wrong to kill. However it plays, there is no explanatory advantage to either
theism or naturalism.
AESTHETICALLY RESPONSIVE ORGANISMS
What explains the presence of aesthetically responsive organisms in our universe? Initially, one
might be tempted to answer this question by pointing out that we should expect rational organisms
to be aesthetically responsive. Certainly, in the case of human beings, our rational capacities
emerged in tandem with our aesthetic capacities. Perhaps we think that, when compared with our
moral capacities, our aesthetic capacities are more clearly exaptations. Perhaps we think that, in gen-
eral, aesthetic norms are more parochial than moral norms. While this point might affect the value
of further consideration of the question of how aesthetic truth is established, it makes no difference
to the conclusion at which we arrive. As was the case in the discussion of the presence of morally
responsive organisms in our universe, the discussion of the presence of aesthetically responsive
organisms in our universe is subsumed by the discussion of the presence of rationally responsive
organisms in our universe.
NORMATIVELY RESPONSIVE ORGANISMS
What explains the presence of normatively responsive organisms in our universe? What explains the
capacity of human beings to respond to linguistic norms, or legal norms, or cultural norms, or the
many other kinds of norms to which human beings are responsive? Here, again, I think that, in the
light of the immediately preceding discussion, we can cut straight to the chase. The discussion of
the presence of normatively responsive organisms in our universe can be subsumed under the dis-
cussion of the presence of rationally responsive organisms in our universe. There is nothing in the
explanation of the presence of normatively responsive organisms in our universe that favors either
theism or naturalism.
REPORTS OF RELIGIOUS EXPERIENCE
What explains the content and distribution of reports of religious experience in our part of the uni-
verse? To answer this question, we need to think about the range of kinds of religious experience.
(1) Some religious experience is generated by religious practice. This kind of religious experience
is fully explained by the existence of religious organizations, observances, traditions, and the like.
(2) Some religious experience involves particular kinds of responses to what is acknowledged to be
the natural world. Of course, this kind of religious experience is patterned to time and place:
religious response to the starry night sky or sunflowers or cherry blossoms or totemic animals or
feelings of remorse or shame is different in different cultures and epochs. This kind of religious
experience is fully explained in terms of prior religious belief and prior religious expectations. (3)
Some religious experience involves dreams and visions. The causes of dreams and visions are diverse.
Some dreams and visions are induced by drugs, exercise, fasting, hypnosis, meditation, mental
illness, music, near-death experiences, sex, and so on; others are not. Everyone knows that many
dreams and visions provide neither information nor insight to those who have them. Moreover,
everyone knows that many of the causes of dreams and visions produce states that are not cogni-
tively reliable: many causes of dreams and visions issue in states that impair performance even on
cognitively undemanding tasks. Given what is known about typical causes of dreams and visions,
best theists and best naturalists should agree that dreams and visions are not reliable sources of
information about their alleged contents. (4) Some religious experience is “mystical” or “sacred”
or “spiritual.” These experiences might be ecstatic (possession by the divine), numinous (theistic
encounters characterized by fear and compulsion), unitive (evanescent, ineffable, passive, tranquil),
salvific (an accompaniment of liberation, or enlightenment, or rebirth), or natural (a sense of one-
ness with nature). As with dreams and visions, much mystical experience is generated by conditions
that are negatively correlated with performance on quite simple cognitive tasks. Even when this is
not the case, mystical experience does not favor either theism or naturalism. Best theists and
best naturalists agree that mature cognitive science and evolutionary theory yields the best unified
explanation of a wider class of experiences that include religious experiences: shivers down the spine,
variations in mood and affect, feelings of being watched, intimations that we are looking at familiar
things from completely new perspectives, and so forth.
REPORTS OF MIRACLES
What explains the content and distribution of reports of miracles in our part of the universe? To
answer this question, we need to think about the range of kinds of reports of miracles, that is, of

events that allegedly involve nonnatural agents and/or the exercise of nonnatural powers. Some
reports of miracles concern the lives and deeds of central figures in particular religions. Some reports
of miracles are woven into central religious texts. Some reports of miracles belong to historical reli-
gious traditions. Some new reports of religious miracles are made every year. Of course, there are
lots of reports of other events and entities that are not supported by naturalistic science, for exam-
ple, sightings of cryptids, reports of anomalous events, productions of conspiracy theories, and affir-
mations of the virtues of alternative medicine. The main contours of the explanation of the produc-
tion and transmission of all of these kinds of reports is widely understood. Human beings are
disposed to make false attributions of agency. Human beings are disposed to believe what they are
told by those they take to be authoritative. Human beings aspire to be taken to be authoritative
in some domains.
Given all of this, there is bound to be utterance and uptake of falsehoods. While many false-
hoods are not sufficiently memorable, some falsehoods are ready candidates for transmission. In par-
ticular, minimally counterintuitive falsehoods may easily become entrenched in particular subcom-
munities or communities. By naturalist lights, nonnaturalistic worldviews are all locked into
competitive special pleading, each trying to claim greater explanatory breadth and depth on the basis
of different applications of similarly questionable techniques to claims with similar origins. In this
light, it is plausible that best naturalists and best theists will agree on uniform rejection of all reports
of miracles and relevantly similar events not supported by naturalistic science.
HISTORY OF RELIGION
What explains the historical and geographical diversity of religions in our part of the universe? To
address this question, we need an account of what religions are. Roughly following Scott Atran
and Ara Norenzayan (2004), I take it that religions are passionate communal displays of costly com-
mitments to the satisfaction of nonnatural causal beings and/or the overcoming of nonnatural causal
regulative structures resulting from evolutionary canalization and convergence of (1) widespread
belief in nonnatural causal agents and/or nonnatural causal regulative structures; (2) hard-to-fake
public expressions of costly material commitments to the satisfaction of those nonnatural agents
and/or the overcoming of, or escape from, those nonnatural causal regulative structures; (3) master-
ing of people’s existential anxieties by those agents and/or the overcoming of or escape from those
nonnatural regulative structures; and (4) ritualized, rhythmic, sensory coordination of all of the
above in communion, congregation, intimate fellowship, and the like. Following much work in
the cognitive anthropology of religion, I further suppose that the widespread belief adverted to in
this definition is readily explained in terms of natural features of human psychology and sociality.
Given this much, it is clear that there are no obstacles to purely naturalistic accounts of the histori-
cal and geographical diversity of religions. It is a plausible conjecture that best theistic and best nat-
uralistic worldviews both accept this account. If so, there is no explanatory advantage that falls to
either party. (Theists who want to give a special role to God in the foundation of their own religion
cannot improve upon the purely naturalistic explanation; at best, their explanation is no worse than
the purely naturalistic explanation.)
SUFFERING
What explains the historical and geographical distribution of suffering in our part of the universe?
Given the immense timescale for this suffering—perhaps 500 million years for animals, at least
20 million years for hominids, and at least 300,000 years for Homo sapiens—it is plausible to sup-
pose that the best explanation is purely naturalistic. It is plausible that suffering is the outcome
of naturalistic evolution. It is a plausible conjecture that best theistic and best naturalistic world-
views both accept this account, but, if so, there is no explanatory advantage that falls to either
party. (Theists who want to give a special role to God and/or comparatively recent generations
of human beings in the distribution of suffering in our part of the universe cannot improve upon
the purely naturalistic explanation; at best, their explanation is no worse than the purely natural-
istic explanation.)
VIRTUE AND FLOURISHING
What explains the historical and geographical distribution of human virtue and flourishing in our
part of the universe? Given what we know about the history of human beings, it is plausible that
there has been human virtue and flourishing for several hundred thousand years. Taking into
account what we know about the “out of Africa” migrations of Homo sapiens and the evolutionary
origins of our species, it is plausible to suppose that the best explanation of the distribution of
human virtue and flourishing is purely naturalistic. It is a plausible conjecture that best theistic
and best naturalistic worldviews both accept this account, but, if so, there is no explanatory advan-
tage that falls to either party. (Theists who want to give a special role to God in the distribution of
human virtue and flourishing cannot improve upon the purely naturalistic explanation; at best their
explanation is no worse than the purely naturalistic explanation.)
DISTRIBUTION OF NONTHEISTIC BELIEF
What explains the historical and geographical distribution of nontheistic belief in our part of the
universe? Many great human civilizations did not so much as countenance the existence of God.
Taking into account what we know about the history of those human civilizations, it is plausible
to suppose that the best explanation of the distribution of nontheistic belief is purely naturalistic.
It is a plausible conjecture that best theistic and best naturalistic worldviews accept this account,
but, if so, there is no explanatory advantage that falls to either party. (Theists who want to give a
special role to God in the distribution of theistic belief cannot improve upon the purely naturalistic
explanation of the distribution of nontheistic belief; at best, their explanation is no worse than the
purely naturalistic explanation.)
PRACTICAL CONSEQUENCES OF THEISTIC BELIEF
What explains the practical consequences of theistic belief? To answer this question, we need to
have a good grasp of the practical consequences of theistic belief. In particular, we need a compre-
hensive survey of the kind of data collected in Gregory Paul (2005)—data about relative societal
dysfunction across populations with greatly varying degrees of theistic membership. If there are
practical consequences of theistic belief, then those consequences should show up in this kind of
cross-population data about alcoholism, assault, burglary, consumption of pornography, divorce,
drug addiction, incarceration, murder, obesity, poverty, rape, smoking, software piracy, suicide, teen
pregnancy, and so forth. In fact, these kinds of cross-population data are a wash: they paint no clear
picture of any practical consequences of theistic belief. Taking into account what the evolutionary
and social sciences tell us about people, it is plausible to suppose that the best explanation of the
practical consequences of theistic belief is purely naturalistic. It is a plausible conjecture that best
theistic and best naturalistic worldviews accept this account, but, if so, there is no explanatory
advantage that falls to either party. (Theists who want to claim special practical consequences for
theistic belief cannot improve upon the purely naturalistic explanation of the data that we have; at
best, their explanation is no worse than the purely naturalistic explanation.)
In the past few pages, I have argued that none of a wide class of data favors WT over WA. As I
noted before I entered into this argumentation, there is nothing that I have said that rules it out that
some other data favor WT over WA. If it turns out that no data favor WT over WA, then the argu-
ment for the superiority of WA to WT is an argument from point-by-point dominance: WA has
more minimal theoretical commitments than WT, and yet it nowhere fares worse on the dimension
of maximization of explanatory breadth and depth. If, however, some data favor WT over WA, then

the argument for the superiority of WA to WT must claim that maximizing the trade-off between
minimization of theoretical commitments and maximization of explanatory breadth and depth
favors WA over WT.
Even if it is granted that WA is superior to WT, it might still be insisted that LT is better than
LA, that is, that the best theistic way of life is better than the best naturalistic way of life. While
there is some plausibility to the thought that the best way of life must embed the best worldview,
it is not inconceivable that some will want to argue that, even if WA is superior to WT, LT is better
than LA. In particular, some may wish to argue that, even if the superiority of WA to WT shows that
there is something better about naturalistic belief, nonetheless LT is better than LA because there is
something even better about theistic desire, hope, faith, and so forth.
Consider Blaise Pascal’s wager. Pretend that, while Pr (WA) is high, Pr (WT) is low. Even so,
provided that Pr (WT) is finite and nonzero, the wager calculation tells us that we maximize
expected utility by wagering on God. I think that, according to WA, Pr (WT) = 0. (Why? Because
I think that, according to WA, it is impossible that there are gods. But—at least on the standard
Kolmogorov account—anything that is impossible has probability zero.) Even setting aside the
many other serious objections to Pascal’s wager, then we do not have here a consideration that
speaks in favor of LT. (Perhaps some will object that it may be impossible that there are gods only
if there is logical inconsistency in the claim that there are gods. That would be an error. There are
impossibilities that are not logical inconsistencies. Consider the claim that God exists of necessity. I
do not think that there is a logical contradiction in this claim. Nonetheless, I think that it is not
possible that it is true. More generally, if you think that there is a substantive necessary a posteriori,
then you are bound to think that there are impossibilities that are something other than logical
inconsistencies.)
Consider William James’s wager. James thinks that “our passional nature not only lawfully
may, but must, decide an option between propositions, whenever it is a genuine option that cannot
by its nature be decided on intellectual grounds” ([1896] 1956, 11) In his view, given that the ques-
tion of the existence of gods cannot be decided on intellectual grounds, it is fine to accept the claim
that there are gods on passional grounds. If the argument given earlier is correct, it is wrong to sup-
pose that the question of the existence of gods cannot be decided on intellectual grounds. If WA is
superior to WT, then James’s wager does not supply a reason that favors LT over LA.
What about appeals to hope? Hope for what? There is no difficulty in the idea that naturalists
can lead virtuous, flourishing lives filled with buoyant hopes for their kin, kith, and the wider social
groups to which we all belong. When it comes to purely natural hopes—hopes that have to do
entirely with what happens in the natural world—there is no reason to suppose that any advantage
accrues to theists. (It is not uncommon for theists to have a lower estimation of the virtues and
values of the natural world than naturalists; consequently, it would not be surprising if naturalists
commonly have more buoyant purely natural hopes than theists.) What about hopes that are at least
partly about the nonnatural? I think that, according to WA, it is impossible for those who rationally
embrace WA to have desires or hopes that are partly about the nonnatural. A practically rational
agent who believes that it is impossible that p does not hope or desire that p: believing that it is
impossible that p while hoping or desiring that p is straightforwardly practically irrational. If WA
is superior to WT, then an argument for the superiority of LT to LA on grounds of hopes and desires
is an argument for forming beliefs on passional grounds even when the opposing case has been
made on intellectual grounds. Not even James was prepared to countenance this outcome.
In closing, I should make a couple of remarks about rationality. I have sketched an argument
for WA and LA. I do not suppose that this argument is rationally compelling. As the earlier discus-
sion of the weighing of worldviews and ways of life makes clear, there is no algorithm that delivers
the result that WA and LA are superior to WT and LT. Rather, the weighing of worldviews and ways
of life requires an enormous number of interrelated judgments. Moreover, it is common knowledge
that there are sensitive, reflective, intelligent, well-informed philosophers who disagree about these
interrelated judgments. It would be a mistake to take me to be arguing that, on pain of irrationality,
everyone should weigh the considerations that I have been examining in the same way that I do. If
we consider the distribution of rationality over those who think about the kinds of questions that
we have been discussing, it is plausible to hold that professional philosophers, no matter what posi-
tion they adopt, score highly on measures of sensitivity, reflectiveness, intelligence, informedness,
and so forth. It is simply common knowledge that professional philosophers occupy the full range
of opinions that it is possible to take on questions about the relative merits of WA and WT, and
LA and LT. As with most questions in philosophy, it seems that there are two ways forward: either
we all reasonably cease to make any judgments about the relative merits of WA and WT, and LA and
LT, or else we all reasonably agree to disagree about those relative merits.
Bibliography
Atran, Scott, and Ara Norenzayan. “Religion’s Evolutionary Landscape: Oppy, Graham. The Best Argument against God. Basingstoke, UK: Palgrave
Counterintuition, Commitment, Compassion, Communion.” Beha- Macmillan, 2013.
vioural and Brain Sciences 27 (2004): 713–770.
Oppy, Graham. Naturalism and Religion. London: Routledge, 2018.
James, William. “The Will to Believe.” In The Will to Believe: Human
Immortality and Other Essays on Popular Philosophy. (1896.) Reprint. Oppy, Graham. Atheism: The Basics. London: Routledge, 2019.
New York: Dover, 1956. Paul, Gregory. “Cross-National Correlations of Quantifiable Societal Health
Oppy, Graham. Arguing about Gods. Cambridge, UK: Cambridge University with Popular Religiosity and Secularism in the Prosperous Democracies.”
Press, 2006. Journal of Religion and Society 7 (2005): 1–17.

Index
The index is alphabetized in word-by-word religious experience, 138 Pascal’s wager, 640, 642, 653, 659
order. Page references in boldface indicate revelation, 174 problem of contingency, 228
chapter topics. References with the letter f spread of, 403–404 religious doctrine, 239
indicate a figure. Abstract objects science, 571, 579, 587
a priori knowledge, 308–309 statistics, 550
A Benacerraf problem, 315 varieties of, 26–28
A posteriori arguments intuitionism, 319–320 weak agnosticism, 27–28
big bang cosmology, 288–289 modal knowledge, 317 Ahern, M. B., 523
Borde-Guth-Vilenkin theorem, platonism, 319 Akrasia, 526
289–290 problems posed for theism, 322–323 Albertus Magnus (Albert the Great),
cosmological arguments, 74–78, 93 response to arguments for, 322 265, 568
defined, 65 theistic responses, 324–327 Albo, Joseph, 460–461
identity theory, 431–432, 433 Acceptance vs. belief, 648–649 Albrecht, Andreas, 342, 343, 346
modal epistemology, 318 Access internalism, 122 Allegories, miracles as, 220
suffering, 527, 528, 529, 535 Accessibility relation, 87 Alston, William
A priori: Theism, 301–312 Accidental designators, 72 a priori knowledge, 304
Anselm, 65–67 Accidentally ordered causal series, doxastic practice, 143, 480
deduction, 308–311 268–269 exclusivism, 244–245, 252–256
definitions, 301–303 Achaemenid Empire, 401 Freudian psychology, 627
empiricality vs. a priori, 38 Actual infinites, 284–288 reformed epistemology, 124–125
epistemology, 303–304 Adams, Robert, 413–414, 452–453, Alternative possibilities, principle of,
probability, 304–308 458, 459, 480 444–445
suffering, 527, 528–529, 535 Adaptation and adaptationism, 365, Ambiguity and pluralism, 245–246
A priori: Atheism, 313–329 367–371, 384–385, 628 Ambjørn, Jan, 342
abstract objects, 320–328 See also Natural selection Ampliative reasoning, 90
actual infinites, 284–288 Analogy, 10–12, 143
Aeschylus, 566
Analytic truth, 314, 453
analytic vs. synthetic truths, 314 Aesthetically responsive organisms,
Anderson, C. Anthony, 99–100
Hume’s Fork, 89–90 688–689
Animal Cognition (Reppert), 435
intuition, 315–317 Affirmative theology. See Kataphatic Animism, 401
modal knowledge, 317–319 theology “Anonymous Christians,” 235
morality, 479–480 Agape, 110–112, 115–116, 117 Anselm
problem of evil, 93–94 Age of the universe, 360, 372, 585 divine being, nature of, 9–10, 668
terminology, 313 Agency detection device, 393, 629 ontological argument, 31, 65–67,
Abduction Agents and agency 96–98
direct experiential theism, 157–158 agent causation, 445 univocity, 11
epistemic abduction, 114–115, 117 human dignity, 486 Anthropic principle, 79, 341, 347, 360,
fine-tuned universe, 332, 354–355 human freedom, 441–444 363
intuition, 41 intentional agents, 110 Anthropology
modal knowledge, 318 Agnosticism improbability of truth of theism, 587
Abraham and divine command theory, being led by God, 113 monotheism’s displacement of non-
459, 462 as epistemic view, 230 monotheism, 401
Abrahamic religions epistemology, 122 myth, function of, 191
aspects of atheism in, 24, 25–26 evolutionary religious studies, 394 natural law, 675
evolutionary study of religion, 391, as impossible, 616 religious utterances, 632
394 infinite library thought experiment, revelation, 187
God, definition of, 138 285 sacred texts, 130
religious diversity, 256 meaning, 499 theories of religion, 625–627
695
INDEX
Anti-criterialism, 440 influence on Islamic philosophy, Barnes, Luke, 333–339, 347

Antirealism, 20, 28–29, 230, 231, 452 259 Barr, S. M., 338
Antireductionism, 282 kind-name theory, 6 Barrett, Justin, 142, 256, 436, 623, 631
Antitheism, 23 laws of science, 566 Barth, Karl, 106–107, 164, 455
Apocalyptic Eschatology in the Gospel of metaphysics, 611 Bartha, Paul, 653
Matthew (Sim), 403 physics and metaphysics, 261–264 Basic belief
Apologetics soul, 419 doxastic foundations, 120, 121,
cognitive science, 590 suffering, 524, 536 123, 124–125
miracles, 223–224 sufficient reason, 266, 278 religious diversity, 233
prophecies, 188–189 teleological argument, 78 religious experience, 143
religious diversity, 244–245, 257 Arnauld, Antoine, 430 revelation, 184–185
science, 584, 589 Arrow of time and vacuum structure, “Basic goods” school of thought, 674
Apophatic attributes of God, 153–154, 343f Basicality, 233
158 Artificial intelligence, 426–427, 431, Batesian mimicry, 622
Apophatic theology, 12–15, 138, 186, 435 Bayesianism, 27–28, 432–433
500, 625 Aseity-sovereignty doctrine, 322–323, Bayes’s theorem, 305, 546–549
Aporia, 524, 528–529, 536, 537 324, 325 Bayle, Pierre, 530
Apparitions, 175–176 Aspirational theism, 45–46 Begging the question
Appeal to love, 469–470 Assertives, 632 alleged proofs God’s existence, 55
Aquinas, Thomas. See Thomas Aquinas Assigned purpose and meaning of life, Anselm’s argument, 67
Arbitrariness problem, 469, 470–471 515 evidence and belief in God, 109
Archaeology, 587 Asymptotically de Sitter endgame, 344 miracles as evidence of God, 208
Arendt, Hannah, 670–671 At the Origins of Modern Atheism Behe, Michael, 360, 369–370
Argument from design. See Intelligent (Buckley), 578–579 Belief
design Athanasius, 565 axioms, 569–570
Atonement, 661 cognitive attitudes, 22, 23
Argument from evil, 482, 544–545
Atran, Scott, 690 doxastic theism, 45
See also Problem of evil
Attributes of God. See Divine attributes ethics of, 44
Argument from religious experience,
Augustine evidentialism, 645–649
592, 637
creation story as metaphor, 379 in God vs. in existence of God,
Arguments
miracles, 212 103–104
alleged proofs, by worldview, 55–56
philosophy, role of, 665–666 meaning of life, 518–520
direct experiential theism, 155–158 naturalist vs. nonnaturalist
divine attributes, 8–10 problem of evil, 537
revelation, 177 explanations, 622–623
evidential arguments from evil, Pascal’s wager, 638–641, 649–654
533–534 Aurora borealis, 592
practical consequences of theistic
“God” as used in proofs of God’s Authenticity and the meaning of life,
belief, 691–693
existence, 5–8 517
proof, burden of, 57
indirect experiential theism, Authority, 578, 674–675
rational degree of belief, 27–28
158–160 Autonomy, 460, 461, 494–495, 515,
reason, knowledge, and truth,
by Jesus, 575, 580 556
415–416
rationality, 692–693 Averroes. See Ibn Rushd
religious experience, 149
theistic argumentation, 40–44 Avicenna. See Ibn Sina
reward for, 654–658
See also Causation and sufficient AWARE study, 595
science of, 109, 570
reason; Logic; Ontological Awareness and consciousness, 412 unreliable sources of, 585–587
arguments; Prudential/pragmatic Axial Age, 396, 633 See also Doxastic foundations
arguments Axioms, 69, 564–567, 569–570, 571 Belief-forming strategies, 656, 658
Arguments from scale, Everitt’s, Ayer, A. J., 54 Belkin, Samuel, 170
371–374 Azzouni, Jody, 326–327 Benacerraf, Paul, 308, 311, 315, 320,
Aristotelian semiotics, 3 323
Aristotle B Benedict XVI, Pope, 549
analogy, 11 The Bāb. See Mirza Ali Muhammad Bering, Jesse, 629, 631
aporia, 528 Babbage, Charles, 216 Berkeley, George, 140
atheism, foundation of, 579 Bacon, Roger, 568 Beyond Freedom and Dignity (Skinner),
causality, 282, 671 Baggett, David, 449 422–423
cosmology, 74–75, 373 Baggini, Julian, 24 Bible
existential proof through nominal Baker, Lynne Rudder, 425–426 Christianity, spread of, 403
definition, 7 Balaguer, Mark, 321–322 creation story as metaphor, 379
formal logic, 81 Balance of nature, 370, 371 interpretation and revelation, 190
God vs. god, 4 Banks, Tom, 342, 343, 347 Jesus dialogues, 577, 578, 580
individuating vs. universalizing Baptists, 167 Kierkegaard, Søren, 181–182
ideas, 353 Bargaining and morality, 477–478 miracles, 195, 203, 205, 219–220
INDEX
Bible, continued Burden of proof, 44, 57–58, 100, 233, kalām argument premise 1, 290–292
myth and religion, 633 394, 401 kalām argument premise 2, 284–288
natural causal order, 197 Byrne, Peter, 462–463 kalām argument stated, 283
religious diversity, 233 nature of causation, 281–282
revelation, 164–168, 174, 188–190, C Pascal’s wager, 659
586 Cajetan, 266 prudential/pragmatic arguments,
science and theistic retreat, 585 Calvin, John, 143, 192, 455, 612, 623 645–649
theistic narrative singularism, Calvinism, 107–108, 167 Causes, Aristotle’s table of, 282
501–502 Camus, Albert, 513, 660 Certainty and uncertainty, 130–131,
as word of God, 127–128, 129, 131 Canon of parsimony. See Ockham’s razor 563–564
Bible Belt, 568 Cantor, Georg, 537 Chalmers, David, 317–318, 429–430
Big bang Capacity-based account of human Chance
a posteriori arguments, 288–289 dignity, 485 evolutionary theory, 383
Behe, Michael, 360 Carnap, Rudolf, 54 fine-tuning argument, 685
Christian theology’s prediction of, Carr, Bernard, 339 human freedom, 442
589 Carroll, Sean, 341–342, 343, 348, quantum mechanics, 599–600
conformal cyclic cosmology, 350–351, 352 See also Randomness
343–345 Cartesian circle, 57, 58, 317 Change, 261–264, 671
gravity, 336 Chemistry and evolution, 389
Carvaka, 24
Hawking, Stephen, 571 Chesterton, G. K., 671
Catholic Church and Catholicism
inflationary cosmology, 340–343 Chicago Manual of Style, 5
Pascal’s wager, 657–658
initial conditions, 332–336 Christian mystical practice, 252,
rate of growth, 402
multiverse theory, 345, 346, 348 254–256
revelations, 164, 167–168,
physical cosmology, 597 Christian theory of religion, 616–617
170–173, 175, 176–177
three big bangs, 630 Christianity
science, 569
Big bounce, 344 assent of faith to divine revelation, 171
soul as mystery, 438 axioms and belief in God, 569
Biochemistry and evolution, 369–370,
theories of religion, 611 big bang prediction, 589
389
Causal dynamical triangulations, 342 comprehensive atheism, impact of, 21
Biological complexity and
Causal power of the mind, 410, Darwinism, 388, 390
interdependence, 365–371
Blancke, Stefaan, 619–620 420–421 direct experiential theism, 156
Blocking factors to belief, 552–553 Causal principle, 261, 283, 290 displacement of non-monotheism,
Boddy, Kimberly, 350–351, 352 Causal reality, 682, 684–685 400–404
Body-Self Dualism in Contemporary Ethics Causal set theory, 342 divine command theory, 455, 458,
and Politics (Lee and George), 669 Causal theory of knowledge, 310 460–461
Bogardus, Tomas, 229 Causality evolution and Christian revelation,
Boltzmann, Ludwig, 340–341 Augustine, 671 379
Boltzmann brains, 342, 347, 348–352 change, 671 experiential singularism, 502
BonJour, Laurence, 304 final causality, 74–75, 78, 282–283 faith, types of, 163
Boole, George, 81 human freedom, 442, 443 as historical religion, 394
Bootstrapping problem, 325 macroscopic causality, 597 interpretation and revelation,
Borde, Arvind, 342 meaning of life, 497 190–191
Borde-Guth-Vilenkin theorem, 289–290 social concept of, 596 Kierkegaard on faith, 181–183
Boudry, Maarten, 619, 622–623 Causation and sufficient reason: Theism, meaning of life, 498, 518–519
Bounded rationality theory, 435 259–279 miracles, 221, 222
Bousso, Raphael, 343, 350 a priori knowledge, 303–304, 310 motives of credibility in divine
Boyer, Pascal, 256 Aristotelian physics and metaphysics, revelation, 174
Braeckman, Johan, 619–620 261–264 myth and religion, 633
Brahe, Tycho, 568 causal perception, 276–278 natural vs. supernatural revelation,
Brief brains, 347, 348, 351–352 Enlightenment (Descartes and 164
A Brief History of Time (Hawking), 79 Kant), 269–276 Pascal’s wager, 661
Broad, C. D., 141 historical reconstruction, 259–261 perennialism, 250, 256
The Brothers Karamazov (Dostoevsky), Medieval tradition (Thomas Aquinas private vs. public revelations,
461 and Duns Scotus), 264–269 175–176
Brunner, Emil, 455 miracles, 196–197, 199 prophecy and revelation, 189–190
Bubbles, inflationary. See Inflationary Causation and sufficient reason: Atheism, realism, 565–566
cosmologies 281–300 reason and revelation, 172
Buber, Martin, 105 continued existence of the universe, reformed epistemology, 107–108
Buckley, Michael J., 578–579, 580 297–298 religious diversity, 233
Buddhism, 165, 396, 592–593 cosmological arguments from revelation, 166, 167, 168, 169, 170
Buffon, Comte de, 366 contingency, 293–297 reward for belief, 654
Bullivant, Stephen, 24 final causes, 282–283 sacred texts, 127–128, 129, 131

INDEX
Christianity, continued Cognitive blocking factors, 552 Concept-words, 5

salvation, 653 Cognitive priority of truth indicators, Conclusions, identifying, 83
science, 397, 568, 585 106–109 Concrete reality, contingent, 289, 294,
suffering, 523 Cognitive science 295–297, 313
theistic narrative singularism, 501 belief in sacred texts, 130 Concrete vs. abstract objects, 319
theories of religion, 606, 610–613, as favoring theism or atheism, Condition of reasonable epistemic
616–617 589–590 access, 543
unreliable sources of belief, as inconclusive regarding theism or Conditional likelihood, 305
586–587 atheism, 570 Confessions (Augustine), 665–666, 672
Western modernity, 399 reason, 434 Confirmation and a priori arguments,
Church fathers, 562 of religion, 142, 620, 621, 306–307
Churchland, Paul, 418 629–630, 631 Conflicting claims. See Religious
Cicero, Marcus Tullius, 243, 244, 605, religious experience, 142, 590–591 diversity
666 religious overlap, 256 Conformal cyclic cosmology, 340–345
Circular arguments Coherence and argumentation, 41–42 Confucianism, 474
causation and sufficient reason, 270 Coherence theory, 53, 120–121, 122 Conjectures and Refutations (Popper), 55
doxastic foundations, 121, 125, Coincidence and miracles, 197–198, Conjunction of all contingent entities,
129, 131, 143 224 294–295
evolutionary theory, 416 Coleman-De Luccia bubbles, 342 Conscience
miracles, 201–202, 204 Collins, Robin, 339–340, 352, 355 belief, 103, 109, 110, 112–116
sound arguments, 41 Commands, divine. See Divine ethics, 450–452, 461
Civilization command theory moral capacity, 481, 482
decline in theism and rise of Communal objectivity and religious religious conversion, 552
atheism, 398–399 experience, 149–152 Conscious intelligence, 353–354,
nontheistic belief, distribution of, Compatibilism, 422, 441, 442–444, 355–356, 383
691 584 Consciousness
origin of, 395–396 Complexity dualism and materialism, 408–410
Clarke, Norris, 271 human evolution, 383–385 dualism vs. materialism, 411–413
Clarke, Samuel, 424 intelligent design, 359, 360, God of the gaps, 426–427
Class nominalism, 322 365–371 hard problem of, 429–430
Classical foundationalism, 123, 143 reason, 434–435 meaning of life, 503–504
Classical theism values of parameters, 686 personal identity, 417–418
abstract objects, 324–327 Compossible properties, 100 reason, knowledge, and truth,
falsifiability and worldviews, 55 Comprehensibility and worldviews, 50 414–417
fine-tuning argument, 364 Comprehensive atheism, 21 theistic apologists, 590
God vs. god, 2, 3 Computers, development of, 82 theistic vs. naturalistic accounts, 687
meaning of life, 499 Conceivability principle Consequence argument, 442
norms and values, 59–60 Anselm, 66 Consistency and logic, 90–93
problem of abstract objects, causation and sufficient reason, Constants
322–323 295–296 fine-tuning argument, 79, 338, 339,
theistic personalism, 11 consciousness, 430–431, 432, 433 597, 674
worldviews, 52, 62 counteractual and counterfactual gravity, 336, 337
Classical theist kind-name theory, 3, 6, conceivability, 318 human existence, 498
8–10 logic, 87 initial conditions of the universe,
Clifford, William, 106, 183, 390–391, modal knowledge, 317 332–333
646–647 problem of evil, 537 multiverse theory, 345, 348, 352
Closed naturalism, 51 suffering, 535, 537 Constructivist ethics, 452
Closed ontological naturalism, 51 Concept nominalism, 322 Constructivist view of God, 28
Cluster definitions of god, 20, 21 Concepts Content of purpose and meaning of life,
Cogent argumentation, 41 of god, 20–21 515
Cognition morality, 478–479, 481 Contextualism and a priori knowledge,
causal perception, 276–277 ontological arguments, 94–99 310
causation and sufficient reason, 274 thinking vs. sensing and imagining, Contingency and evolutionary theory,
normative and evaluative 667–668 380
foundations of epistemology, Conceptual schemes Contingency argument, 74–78, 99,
60–61 norms and values, 60 229–230, 295
reason, 434–436 ontological commitment, 58 Contingency problem, 228–230, 234,
reason, knowledge, and truth, open naturalism, 62 235, 237, 469
414–417 prediction, 59, 61 Contingent being, God as, 323
reformed epistemology, 124–125 self-evident truths, 57 Contingent concrete reality, 289, 294,
tripartite doctrines, 22–24 suitability of, 56 295–297, 313
Cognitive argument for faith, 637–638 worldviews as, 53, 54 Contingent independent beings, 297
INDEX
Contradicting claims. See Religious science, 571 Dawkins, Richard

diversity Creation stories atheism, foundation of, 579
Conventionalism, 310, 316 global theism, 36 evolution, 380, 383, 385, 389
Conviction and belief, 25, 111–113, as metaphors, 379 “new atheism” movement, 607
148–149, 150 myth and religion, 633 scientific evidence, 109
Conway, David, 531, 534, 535 Creation theology, 3, 9 soul, 391
Copernican revolution, 260, 274, 397 Creationism theories of religion, 627
Copernicus, Nicolaus, 373, 397, 568 fallibilism, 39 De dicto vs. de re components of belief,
Core theism, 546–547, 548 science curricula, 370 104, 105, 106, 113, 117
Corollary of Bayes’s theorem, 305 in science curricula, 367–368 De Sitter, Willem, 341
Correspondence theory of truth, 53, 60, Creativity and the meaningful life, 510 De Sitter space, 341, 343–344
61 Credibility, 173–174, 175, 176, Debunking arguments, 620–622
Cosmic fine-tuning. See Fine-tuning 183–184, 588–591 Decision making and dualism, 410
argument Crimes against humanity, 483 The Decisive Treatise (Ibn Rushd), 172
Cosmological argument Crisis of meaning, 493 Declaration of the Rights of Man, 485
a priori knowledge, 303–304 Criterion of ontological commitment, 58 Deductive arguments, 155, 308–311
abduction, 41 Critique of Practical Reason (Kant), 449 Defeaters
alleged proofs of the worldview, 55 Critique of Pure Reason (Kant), 67–68, doxastic foundations, 114–117
big bang cosmology, 288–289 94–95, 260, 271 evolution and naturalism, 415
causation and sufficient reason, 275, Cross-checking religious experience, rationality, no-defeater requirements
276 138–139, 144, 255 of, 121, 123, 124, 125
from contingency, 288, 293–297 Cruelty and divine command theory, revelation, 185, 187
Eagleton, Terry, 31 459–460 Defense against argument from evil,
Epicurean arguments, 297–298 C.S. Lewis’s Dangerous Idea (Reppert), 435 544–546
God of the gaps, 425 Cudworth, Ralph, 455, 461 Definiendum, 7
kalām argument, 283, 284–288, Culture Definition: Theism, 1–17
290–292 conflicting revelations, 186 apophatic theology, 12–13
logic, 74–78, 84–85, 93 divine inscrutability, 554 divine attributes, 8–10
minimal kind-name theory, 6–7 evolution and cultural niche, 389 God vs. god, 1–5
natural law, 674 natural law, 675 kataphatic theology, 13–15
nonzero probability, 653–654 near-death experiences, 594 proofs of God’s existence, 5–8
reference fixers, 31 realism, 565 univocity, equivocation, and
simplicity, 42 religious experience, 591 analogy, 10–12
See also Universe religious overlap, 256 Definition: Atheism, 19–33
Cosmological arguments from cognitive aspects, 22–24
science, 567–569, 576–577
contingency, 288, 293–297 concepts of god, 28–32
scriptural belief, 130
Council of Trent, 167 ontological aspects, 19–22
sociological blocking factors, 552–553
Counteractual conceivability, 318 Degrees of belief, 27–28, 123
theistic belief, 550
Counteractual possibility, 318 Degrees of credibility, 183–184
worldviews, 57
Counterfactual argument and the Degrees of meaning, 509
Cumulative case arguments, 41
contingency problem, 228–230 Dei Filius (Vatican I), 164, 172
Cyclic cosmologies, 340–345, 354
Counterfactual conceivability, 318 Dei Verbum (Vatican II), 175
Counterfactual possibility, 318 Deism
Counterintuitive possibilities problem, D Darwinism and human origins, 388
469 Dalai Lama, 592 divine revelation, exclusion of, 176
Coyne, Jerry, 630 Danielsson, Ulf, 343 God vs. god, 8
Craig, William Dao and Daoism, 127, 128, 396, 673, as included in theism, 1–2
a posteriori arguments for a 675 vs. theism, 21
beginning, 288–290 Darwin, Charles Demarcation problem, 39–40
a priori argument for a finite past, adaptation, 367 Dembski, William, 368–369
287–288 biological adaptation, 359 Democracy and decline of theism, 40, 399
Hilbert’s Hotel thought experiment, cognition, 415 Democritus, 25
285–287 probability, balance of, 588 Dennett, Daniel, 53, 385, 434, 435
infinite library thought experiment, religion as evolutionary, 392 Denying and Disclosing God (Buckley),
284–285 science and theistic retreat, 585 580
kalām argument, 283, 291–292 theories of religion, 627 Deontology, 454
morality, 659 Darwinism. See Natural selection Descartes, René
theistic nominalism, 326–327 Darwin’s Black Box (Behe), 360 belief, 57
Crane, Tim, 24 Darwin’s Dangerous Idea (Dennett), 435 causation and sufficient reason,
Creation Davenport, Matthew, 351 259–260, 270–273
abstract objects, 324, 326 Davies, Brian, 15 consciousness, 411–412, 414, 430
Qur ʾanic argument, 75–76 Davies, Paul, 346, 363 definition of God, 99

INDEX
Descartes, continued univocity, equivocation, and religious diversity, 228–230, 233

divine command theory, 455, 460 analogy, 10–12 religious experience, 143
dualism, 409–410 worldviews, 681–682 responses to grounds, 45
formal logic, 81 Divine command theory self-evidencing foundations,
God vs. god, 3 arbitrariness, 470–471 109–111
immortality, 418–419 conflict thesis, 462 skepticism and defeaters, 114–117
interactionism, problem of, core principles, 454–455 Doxastic foundations: Atheism, 119–134
420–421 as epistemic thesis, 456–457 evidence and evidentialism,
ontological arguments, 67, 94 ethics, 454–462 122–123
positive dependence on God, 36 Euthyphro problem, 73–74, foundationalism, 120–122
Descent of Man (Darwin), 392 469–470 inferential justification for theism,
Descriptive statements, 2–3, 5, 29, 71–73 history of philosophy and theology, 123–124
Design. See Intelligent design 455–456 knowledge and rational belief,
Determinism modified divine command theory, 119–120
free will, 421–423 459–460 problem of disagreement, 125–126
human freedom, 441, 442, 443 nature of morality, 480 sacred texts, 127–133
nature of the universe, 380–382 normative aspects of theism, 37 See also Belief
Devil, 128, 175, 181, 186, 188 objectivitiy of moral value, Doxastic practice epistemology,
Dharma, 673, 675 452–454 480–481
Diagoras of Melos, 25 as semantic thesis, 457–459 Doxastic voluntarism, 183
Dialectical materialism, 626 similarities and differences in moral Drange, Theodore M., 129
Dialectical presence, revelation as, 170 beliefs, 474 Draper, Paul, 547, 548
Dialogue, science, 574–576 theological and moral objections to, Dualism
Dialogue Concerning Two Chief World 461–462 Augustine, 669
Systems (Galileo), 373 theological motivations for, causation and sufficient reason, 272
Dialogues concerning Natural Religion 460–461 consciousness, 411–412, 426, 432
(Hume), 93, 367, 623–624 weak dependence, 460 history of atheism, 24
Differentia, 7, 13–14 Divine conceptualism, 309, 310 interactionism, problem of,
Differentiation and Darwinism, 389 Divine grace, gift of. See Agape 420–422
Diller, Jeanine, 31–32 Divine hiddenness, 549–553, 554, mind-body problem, 408–410
Dilthey, Wilhelm, 492 661–662 personal identity and immortality,
Diogenes Laërtius, 50 Divine Hiddenness and Human Reason 418–419
Dionysius, 7, 12–13 (Schellenberg), 549 person-human distinction, 52
Direct experiential theism, 154–155 Divine inscrutability, 553–554 suffering, 549
Disagreement, problem of. See Religious Divine intervention Due good, 670
diversity causation and sufficient reason,
Duff, Anthony, 653
Disappearance of time, 344 271–272
Dulles, Avery Cardinal, 169–171
Disease and physical evils, 526–527 evolution, 389, 394–397
Duns Scotus, John, 11, 13, 14–15,
Disjunctive syllogisms, 75–76 Forrest, David, 112
268–269
Disorder. See Order/disorder miracles as, 212–214, 218, 219,
Duplication problem, 436–440
Distribution of life arguments, 360, 271–272
Durkheim, Émile, 191, 607, 617,
371–374 New Testament, 112
624–625
Distribution of suffering, 554–557 theories of religion, 623
Dynamics of Faith (Tillich), 563
Diversity. See Ethical diversity; Religious Divine perfections. See Perfect being
diversity theology
Divine attributes Divine properties. See Divine attributes E
abstract objects, 323 Divine revelation. See Faith and Eagleton, Terry, 31
classical theism, 8–10 revelation Eastern Orthodox Church, 167, 168,
consistency issues, 90–93 Divine self-manifestation and belief, 172
defining God, 20, 28–31, 32 109–111 Ecological niches and evolution, 389
deism, 2 Divine simplicity, 9–10, 42–44, 323 Economics, 400
God vs. god, 15 Divine source, 167, 168, 169, 180 Ecumenism, 186
Gödel, Kurt, 99–101 Divine sovereignty, 36, 460 Edwards, Jonathan, 463
internal/external distinction, 94–96 See also Aseity-sovereignty doctrine Edwards v. Aguillard (1987), 368
kind-name theory, 6–10 Doctrine, religious, 170, 250–251 Effective computation, 81
monotheism, 26 See also Religious diversity Effect-to-cause reasoning, 9–10
normative aspects of theism, 36–37 Dostoevsky, Fyodor, 461 Efficient causation, 290, 497
ontological arguments, 68–69, Douglas, Mary, 626 Egyptian religion, 244
70–74 Dowker, Fay, 342 Ehrman, Bart, 402
predicate logic, 86 Doxastic foundations: Theism, 103–118 Einstein, Albert, 331, 334–335, 361,
religious experience, 153, 154, convictional evidence, 111–113 509–510
155–157, 158, 159 evidence and knowledge, 104–105 Electromagnetism, 337
INDEX
Elementary particle theory, 333 revelation, 164, 177, 181, 184 doxastic foundations, 104–105
Eliade, Mircea, 609 theistic narrative singularism, epistemic rationality of belief,
Eliminative materialism, 411 501–502 122–123
Eliminativism, 282 Equality and the distribution of for faith, 579–580
The Emperor’s New Mind (Penrose), suffering, 555–556 faith and revelation, 180, 183–184
333–334 Equilibrium systems, 371 miracle claims as evidence for
Empiricism Equivocation vs. univocity, 10–12 atheism, 222–223
cosmological arguments from An Essay in Aid of a Grammar of Assent miracles, 199–200, 203–204,
contingency, 294, 295, 296 (Newman), 450–451 214–215
meaning, 27 Essence and generation argument, 76–77 Pascal’s wager, 658
modal knowledge, 318–319 Essential properties of God, 68–69 prudential/pragmatic arguments,
theistic appraisal, 38–39 Essentialism, 296, 298 645–649
The Enlightenment Essentially ordered causal series, reformed epistemology, 105–109
causation and sufficient reason, 259, 268–269 religious experience, 144
266 Eternal inflation, 341, 345–351 self-evidencing foundation for belief,
human freedom, 441 Eternity and causal order, 197 109–111
human rights, 485 Eternity of motion, 263–264 suffering, 546–549
materialism, 411 Ethical diversity, 462–463 theistic appraisal, 37–40
monotheism, 4 Ethical naturalism/nonnaturalism, 452, worldviews, 681
natural rights, 485 453, 479–480 Evidential agnostics, 27
rationalism, 171, 219, 434, 435 Ethical objectivism/nonobjectivism, Evidential argument from evil, 533–534
Ens realissimum, 274–275 462 Evidential problem of evil, 93, 542
Entheogen, 592, 593 Ethics: Theism, 449–465 Evidentialism, 122–123, 184–185
Entropy, 333–334, 340–345, 597–598 of belief, 44 Evil
Epicurean cosmological argument, 298 divine command theory, 454–462 Augustine, 669–672
Epicurus, 25 ethical diversity, 462–463 as potential defeater, 117
Epiphenomenalism, 411 God, dependence on, 672–673 trivial moral evils and inanity of evil,
Epistemic abductivism, 114–115 Kant’s argument, 449–450 525–526
Epistemic evidence of miracles, natural law, 463, 673–676 See also Problem of evil
212–214, 219, 223 Newman’s argument, 450–452 “Evil and Omnipotence” (Mackie),
Epistemic rationality normative aspects of theism, 36–37 530–531
evidence and evidentialism, 122–123 objectivity of moral value argument, Evolutionary argument against
foundationalism, 120–121 452–454 naturalism, 307–308, 434, 620
inferential justification for theism, open naturalism, 60 Evolutionary theory
123–124 virtue, 463 chance, 379–382
internalism vs. externalism, 121–122 Ethics: Atheism, 467–489 Christian revelation, 379
problem of disagreement, 125–126 faith and revelation, 181–182, Darwinism and human origins,
religious diversity, 230 183–184 387–391
simplicity as a standard of appraisal, human rights, 463–464, 482–486 ethics, 453–454
42–43 moral realism, 467–472 human virtue and flourishing, 691
Epistemic warrant, 391, 393–394 nature of morality, 477–480 improbability of truth of theism,
Epistemically possibility miracles, similarities and differences in moral 587
212–213 belief, 472–477 intelligent design arguments,
Epistemological persuasiveness, 87–89 “The Ethics of Belief” (Clifford), 365–371
Epistemology 646–647 intuition, 57, 316
a priori knowledge, 303–304 Etymology living organisms, 686
Augustine, 669 atheism, 19 moral capacity, 481–482
belief, 116–117 faith, 163 norms and values, 60–61
conceptual scheme of science, 53, 54 revelation, 163 Plantinga’s argument against
divine command theory, 456–457 Euthyphro problem, 60, 73–74, 454, naturalism, 620
evidence, knowledge, and belief, 469–470, 514 problem of evil, 382–383
104–105 Evangelical Christianity, 168, 172 reason, 434, 451
evidentialism, 645 See also Religious fundamentalism reason, knowledge, and truth,
modal knowledge, 318–319 Evans-Pritchard, Edward, 625 415–416
natural law, 675 The Event of Sinai (Maimonides), religious experience, 689
nominalism, 612 165–166 religious studies, 391–394, 398, 610
norms and values of, 60–61 Everett, Hugh, 346 suffering, 544
pluralism, 237, 238 Everitt, Nicholas, 371–374 theistic metaphysical singularism,
of receptivity, 497 Evidence 496–497
reformed epistemology, 58, a priori knowledge, 309–310 theistic retreat, 585
105–109, 124–125 convictional evidence for belief in theories of religion, 607, 627–631
religious diversity, 233–234 God, 111–113 Ewing, A. C., 453

INDEX
Exceptionalist atheists, 584 reception of revelation and role of living organisms, 686
Exclusivism, 230–233, 235–236, 239, faith, 171 mathematics, applicability of, 688
244–245, 252–257 religions that claim supernatural miracles, 689–690
Existential anxiety, 392 revelation, 165–167 morally responsive organisms, 688
Existential proof through nominal sacred texts, 174 normatively responsive organisms,
definition, 7 theories of religion, 609–614, 689
Existential quantifiers, 321, 326–327 614–615 organized complexity, 686
“Existentialism Is a Humanism” Faith and revelation: Atheism, 179–193 practical consequences of theistic
(Sartre), 515 conflicting claims, 186–187 belief, 691–693
Exoplanets, 592 Hume, David, 180–181 rational organisms, 687
Expected value calculation, 639, 640, interpretation, 190–192 religious experience, 689
642–643 judging reliability of revelations, science, applicability of, 687–688
Experience, religious. See Religious 184–185 suffering, 690–691
experience Kierkegaard, Søren, 181–183 virtue and flourishing, 691
Experience as independent of a priori meaning of life, 518–520 worldviews and ways of life, 679–684
knowledge, 302 naturalistic explanations, 187–188 Findlay, J. N., 66
Experience machine and meaning of life, prophets and probabilities, 188–190 Fine-tuning argument
509, 510 sensus divinitatis, 192–193 atheistic response to, 360–365
Explanatory filter and the design theories of religion, 622 Boltzmann brains and brief brains,
argument, 368–369 unethical faith, 183–184 348–352
Explanatory principle, 283, 298 unreliable sources, 586–587 conformal cyclic cosmology, 343–345
Explicability arguments, 293 Faith healers, 143 constants, 332
Exposition on the Apostles’ Creed (Thomas Faith in induction, 59 free parameters, 331–332, 685
Aquinas), 611 Faking belief, 655 fundamental forces, 336–338
Extensional properties and predicate Fallibilism, 39–40, 654 inflationary cosmology, 340–343,
logic, 86 False vacuum states, 348 343f, 345
Externalism vs. internalism. See Falsifiability and testability, 38 initial conditions, 332–336
Internalism vs. externalism Falsifiability and worldviews, 54–55 logic, 79
Extinct religions, 244 Family-resemblance style definition of multiverse theory, 345–352
Extraordinary events, miracles as, god, 20 natural law, 674
212–215, 217, 218–219 al-Fārābī, Ab u Nasr, 9, 68, 77 nonzero probability, 653
Extreme materialism. See Reductive Fascism, 384, 399 physical cosmology, 597–598
materialism Faustus, 671 Tegmark’s Level IV multiverse,
Eyewitnesses to miracles, 219, 221 Fear and Trembling (Kierkegaard), 352–354
181–183, 462 theory of everything, 339–340
Feder, Kenneth L., 395 transcendent intelligence theory,
F Feingold, Lawrence, 173 354–356
Facts, moral. See Moral facts Fermat’s last theorem, 307 Finite Hilbert spaces, 350
Factual content of religious beliefs, FGG inflation, 342 First Vatican Council. See Vatican I
625–626 Fiction and scriptures, 128–129 First-order logic, 84–86, 98, 320–321
Factually necessary beings, 297 Fictionalism, 322 Fischer, John Martin, 422
Faith and revelation: Theism, 163–178 Fideism, 107, 108, 171–172, 182, 612, Fisher, Ronald, 367
certainty, 563 652 Fitelson, Branden, 368–369
challenges to claims of revelation, Fides qua creditur, 163 Fitness-tracking of belief, 620, 622
177 Fiduciary theism, 45, 46 Five Ways (Thomas Aquinas), 30–31,
credibility, motives of, 173–174 Field, Hartry, 311 75, 79, 266
defining, 163 Figuralism, 327 Flew, Antony, 530, 652
evidence for, 579–580 Final causality, 74–75, 78, 282–283 Folk physics and folk psychology, 53, 57
evolution, 379 Final reckonings: Theism, 665–677 Folk religions, 401, 621
fiduciary theism, 45 Augustine, 665–672 Form theory of the soul, 438–439
interpretation, 167–168 morality’s dependence on God, Formal, final, material, and efficient
meaning of life, 502–503 672–673 causes, 77, 282
miracles, 202–203, 204 natural law theory, 673–676 Formal logics, 81–82
models of revelation, 169–171 Final reckonings: Atheism, 679–693 Formal reality, 259–260, 270
motives of credibility, 173–174 aesthetically responsive organisms, Formal validity, 84, 85, 89
natural vs. supernatural revelation, 688–689 Formalization and epistemological
163–165 causal reality, 684–685 persuasion, 88–89
possibility of revelation, 176–177 conscious organisms, 687 The Forms (Plato), 50–51, 56, 60
public vs. private revelations, distribution of nontheistic belief, 691 Forms of life and open naturalism, 58,
174–176 fine-tuning argument, 685 59, 61, 62
rational foundations of, 65 history of religion, 690 Forrest, David, 112
reason, 171–172 intelligent organisms, 687 Foundationalism, 120–122, 123, 143
INDEX
Four universal forces, 333, 336, 339 Geography myth, 633

Fragility and the meaning of life, 497, distribution of nonbelief, 550, 552, theories of religion, 606
500, 519 691 Gödel, Kurt, 68–69, 86, 99–101,
Framing propositions, 56 monotheism, 400–404 309–310
Francis of Assisi, 187 religious diversity, 228–229 God-experiences. See Religious
Frankl, Viktor, 498 virtue and flourishing, 691 experience
Fraud, 142–143, 187–188, 221 George, Robert, 669 Goetz, Stewart, 503–504
Frazer, James, 218, 607, 617, 626 German higher criticism, 182 The Golden Bough (Frazer), 626
Freak observers, 348, 350–351 Gettier, Edmund, 105, 478 Golden Rule, 473, 474, 675
Free parameters, 331–332, 339, 685 al-Ghazālī, 75, 172 Goldstein, Rebecca, 31
Free will Aristotle’s influence on, 259 Good life, 486, 659–660
Augustine, 669–672 causation and sufficient reason, Goodness
distribution of suffering, 556 271–272 Descartes, René, 270
human freedom, 441–445 natural causal order, 197 divine command theory, 459–460
meaning of life, 504, 520 representationalism, 260 Euthyphro problem, 73–74
moral agency, 419–423 theories of religion, 634 freedom, 423–424
organized complexity, 686 Gilbert, William, 224, 568 normative aspects of theism, 36–37
suffering, 533, 543, 544 Gilson, Étienne, 271, 276 Plato, 60, 74
Freedom Global problem of evil, 542 problem of contingency, 234
Augustine, 669–672 Global theism, 35 problem of evil, 530, 532
divine command theory, 461 Gnostics, 653 suffering, 541–542
of God, 612 God Gospels. See Bible; Sacred texts
metaphysics of, 669 Anselm’s definition of, 65–66 Gould, Stephen J., 370–371, 379–380,
suffering, 541 atheism and differing concepts of, 383
Frege, Gottlob, 5, 70, 81, 95 28–32 Grace, salvation by, 661
Fregean semantics, 8 as capable of reasoning, 435–436 Gratuitous suffering, 557
Freivogel, Ben, 350 Cartesian circle, 57 Grave, S. A., 451
Freud, Sigmund, 452, 607, 609, as contingent being, 323 Gravity, 336–337, 342, 546, 597
624–625 defining, 20 Greatness, maximal. See Maximal
Functional integration, biological, direct experiential theism, 155–158 greatness
365–366 as foundation for morality, 672–673 Greece, ancient
Functionalism, 316, 367, 625–626 free will, 423–425 divine explanations of empirical
Fundamental accounts, 49–50, 51–52 indirect experiential theism, phenomena, 584
Fundamental forces and the fine-tuned 158–160 history of atheism, 25
meaning of life, 513–516, 514–516
universe, 336–338 philosophy, 4, 243
miracles, 195–197, 202, 208–210,
Fundamental norms and values, 59–60 religion and democracy, 399
212–214, 218, 219
Fundamental reality and worldviews, 62 science, 397
morality of, 234, 235
Fundamentalism, religious. See Religious See also Specific Greek philosophers
negative vs. positive dependence on,
fundamentalism Greek language, 4
36
Funes, José, 577 Greenwood, Kyle, 403
norms and values, 36–37, 59–60
Futility and the meaning of life, 497, Griffin, James, 485–486
perfect-love and the definition of,
519 Griffiths, Paul, 621
137–138
Group-selection theory of religion, 628
pluralism, 237, 238
G religious experience oriented toward, Guidance control, 443
Galen, 588 153–155 Guth, Alan, 346, 347
Galen, Luke, 400 revelation, 167–169, 170–171
Galileo, 373, 568 sacred texts as word of, 127–129 H
Gaon, Saadia, 457, 460–461 thoughts of, 315, 326 Hadith, 168
Gaskin, J. C. A., 59 as used in proofs of, 5–8 Hájek, Alan, 653
Gass, William H., 508 will of, 456–457 Haldane, J. B. S., 367
Gauguin, Paul, 510 worldviews, 52, 62 Hallucinogenic drugs, 593
Gaunilo of Marmoutiers, 66–67 God, attributes of. See Divine attributes Hamiltonians, 343
Gay, Robert, 453 God of the gaps, 369, 425–427, 432, Happiness
Geach, Peter, 3 579 Kantian theory, 58, 60, 449–450
Gellman, Jerome, 139, 141, 142, 239 God talk, 11–12 meaning of life, 492, 493, 503, 504
General theory of relativity, 288, God vs. god natural law theory, 673
334–335, 361, 564, 685 atheism, 583–584 utilitarian theory, 474
Generation argument, 76–77 descriptivism vs. referentialism, 2–3 Harrington, Daniel, 403
Genetics, 587 divine attributes, 15 Harris, Marvin, 625
Genuineness and religious experience, kind-name theory, 8–10 Hart, David Bentley, 154
142–143 language differences, 1 Hartmann, William K., 586

INDEX
Hawking, Stephen, 79, 342, 347, 571, reason, knowledge, and truth, personal identity, 418, 419
577 414–417 problem of evil, 530
Hawthorne, John, 310 uniqueness, 407–408 representationalism, 260
Heaven, 310–311, 638, 639, 650–653, Human beings: Atheism, 429–448 scientific study of religion,
654 consciousness, 429–433 623–624
Hebrew Bible human freedom, 441–445 skepticism, 123
God vs. god, 3–4 personal identity and immortality, theism explained, 587
immortality, 440 436–440 worldviews, 62
personal identity and immortality, person-human distinction, 52 Hume’s Fork, 89–90, 93, 99
436 reason, 434–436 Hunter-gatherers, 392, 395, 398, 400,
prophecy and revelation, 189 Human dignity, 473, 474, 483–486 625
revelation, 166, 168, 174, 187–188 Human freedom, 441–445 Hurtado, Larry, 402
Heidegger, Martin, 260 Human history: Theism, 377–386 Huxley, Aldous, 250
Heimskringla (Sturluson), 654 chance and evolution, 379–382 Huxley, Thomas Henry, 26
Henotheism, 4, 24 Christian revelation and evolution, Hypothetical imperative, 676
Henry, Carl, 455 379
Henry VIII, 613 design, complexity, and evolution, I
Hermeneutics, scriptural, 131 383–384 Ibn al- ʿArabī, 172
Herodotus, 605 science and naturalism, 377–378 Ibn Ishaq, 586
Hick, John, 236–239, 244, 245–249, suffering and evil in nature,
Ibn Rushd, 172, 259
396 382–383
Ibn Sina (Avicenna), 9, 77–78, 259, 271
Hiddenness of God. See Divine theory of evolution, 377
Ideal subjectivism, 478
hiddenness Human history: Atheism, 387–406
Identity, personal, 417–419, 436–440
Higgs field, 351 Darwinism and human origins,
Identity stage (logic), 5
High entropy and the big bang, 387–391
Iff and the duplication problem, 436,
342–343 decline of mass theism and rise of
437–438, 439
Higher criticism, 182 popular atheism, 397–400
evidence of providential guidance, Ijjas, Anna, 347
Higher-order logic, 86
394–397 Imagination, 317–318, 667–668
Hilbert space, 350
Hilbert’s Hotel thought experiment, evolutionary study of religion, Imago dei, 485
285–287 391–394 Imitatio Dei, 111
Himma, Kenneth Einar, 237, 238 monotheism’s displacement of non- Immanence, 20
Hinduism, 165, 396 monotheism, 400–404 Immateriality, 667–668, 669
Historical testimony, 219–222 Human intelligence as adaptation, 388 Immorality. See Morality
History Human nature Immortal counterargument, 287–288
of atheism, 24–26 meaning of life, 497 Immortality, 418–419, 436–440, 504,
of philosophy, 455–456 miracles, 200 516, 565
of religion, 690 natural law, 675 Imperfect design in nature, 382
revelation as, 170 possibility of revelation, 177 Inanity of evil, 525–526
of science, 553 Human rights, 463–464, 482–486 Incarnation, 565, 569
See also Human history Human spirit, 170 Inclusivism, 231–232, 233, 235–236
Hitchens, Christopher, 23, 607 Hume, David The Incoherence of the Philosophers
Hitler, Adolf, 526 agnosticism, 27 (al-Ghazālī), 172
Hittinger, Russell, 675 causation and sufficient reason, Incompossible properties, 100
Hobbes, Thomas, 81 272–273, 276 Independent beings, 297
Hochschild, Joshua, 493 conceivability principle, 87 Indeterminacy, problem of, 92
Hogan, Craig, 338 consistency issues in theism, 93 India, 565
Holland, R. F., 197 critiques of arguments of, 201–205 Indian Rationalist Association, 224
Hope, 22, 45–46 design argument, 360 Indian religions, 24
Horizon blockers, 545 evidence and degrees of belief, Indirect experiential theism, 154,
Horizon complementarity, 350 183–184 158–160
Hortensius (Cicero), 666 functionality argument, critiques of Indispensability argument, 320–322,
Horton, Robin, 626 the, 367 326–327
Hubbard, L. Ron, 127 habits, 652 Individual meaningful lives, 509–511
Human beings: Theism, 407–428 Hume’s Fork, 89–90, 99 Individuating ideas, 353
dualism and materialism, 408–410 inference problem of evil, 93 Inductive arguments
free will, determinism, and moral laws of nature, 51 direct experiential theism, 155–157
agency, 419–423 matter-mind question, 50–51 methodology, 41
God and freedom, 423–425 miracles, 180–181, 199–201, modal knowledge, 318
God of the gaps, 425–427 205–207, 212, 214–215 noninterventionist induction, 204
personal identity and immortality, nonnaturalist explanations of belief, open naturalism and faith in, 59
417–419 622 Inequality in suffering, 555–556
INDEX
Inference rationality of theism, 43 reward for belief, 654

belief, 121, 123–124 reason, 434–435 science, 397, 569
epistemic abduction, 114–115 teleological arguments, 78–79 spread of, 399, 400, 401–404
methodology, 41 See also Fine-tuning argument teleological argument, 78–79
modal knowledge, 318 Intelligibility and causation, 276–277 theories of other religions, 606
Inference problem of evil, 93 Intensional properties and predicate unique quantification, 71–72
Inference to the best explanation, logic, 86
543–544 Intentional agents, 43, 109–114 J
Inferior and superior religions, 236, 237 Intentionality, 412–413, 433 Jaki, Stanley, 566
Infinite library thought experiment, Interactionism, problem of, 420–421 James, William
284–285 Interdependence, biological, 370 doctrinal differences, 251
Infinite past, a priori arguments against, Internal inconsistency and exclusivism, faith and revelation, 183, 184
284–288 253–254 many-contenders objection, 661
Infinite regress, 268, 269, 498, 499, Internalism vs. externalism “noetic quality” of religious
684–685 a priori knowledge, 315 experience, 140
Infinite value, 638–642, 651, 653, 654 epistemic justification for rational practical consequences of belief, 692
Infinitism, 120–121 belief, 121–122, 125 prudential/pragmatic arguments, 650
Inflationary cosmologies meaning of life, 499 religious experience as “private
Boltzmann brains and brief brains, ontological arguments, 94–96, 96–97 evidence,” 58
348–352 revelation, 181
saintliness, 248
fine-tuned universe, 340–343, 345 International Covenants on Civil and
truth and worldviews, 60, 61
multiverse theories, 346–347 Political Rights, 484
will to believe, 641–642
Q nonuniformity, 335 International law, 483–484
James’s wager, 692
Influence on others and a meaningful Interpretation
Jesus
life, 510 miracles, 217–219
afterlife, 565
Inherent anticipability, 524, 526–527, myth, 633–634
arguments, 580
535–536 revelation, 167–168, 190–192
sacred texts, 131–132 authority, 578
Initial conditions
cosmic inflation, 341–342, 345 Intersubjectivity, 38 dialogue, 574–575
fine-tuned universe, 332–336 Intrinsic goodness, 36–37 divine revelation, 166, 167, 169
multiverse, 347, 352 Intrinsic intelligibility, 524, 525–526, miracles as mythical, 221
theory of everything, 338–339 528–529, 534, 536 miracles explained, 222
Initial singularity, 289, 296, 597, 684–685 Intrinsic probability, 310 ontological atheism, 21
Inner experience, revelation as, 170 Intuition reformed epistemology, 106, 108
Inscrutable distribution of opportunity, a priori knowledge, 302, 315–317 religious diversity, 227, 232, 239
557 causation and sufficient reason, religious experience involving, 159
Insight (Lonergan), 276 274–275 suffering, 523
Instantiation, 95, 96–97, 99 meaning of life, 516 Jevons, William, 81
Institutes of the Christian Religion morality from, 474 John of the Cross, 175
(Calvin), 612 sacred texts, belief in, 132 John Paul II, 164, 391, 493
Instrumental goodness, 36 science and faith, 564, 567 Johnson, Dominic, 628
Intellectual responsibility, 44 Intuitionism, 319–320 Johnston, Mark, 31
Intellectualist theories of religion, 632–633 Ironic agnosticism, 27 Jones, Jim, 253, 254
Intelligence, 388, 687 Irrationality Joy, science as, 580
Intelligent design compelling vs. defensible arguments, Judaism
balance of probability, 588 637–638 divine command theory, 455, 458,
Bayesian analysis of, 432 evil, 525–526 460–461
biological complexity argument, Pascal’s wager, 652 as historical religion, 394
365–371 Irreducible complexity argument, 360, interpretation and revelation, 190–191
consciousness, 414, 417 369–370 natural vs. supernatural revelation,
Darwinism, 389 Isaiah, 6 164
evolution and chance, 378 Islam private vs. public revelations, 175
fallibilism, 39–40 Aristotle’s influence, 259 revelation, 165–166, 167, 168, 170,
Freudian theory, 627 cosmological argumentation, 75–78 174
human evolution, 379, 382–383, displacement of non-monotheism, saintliness, 248
383–385 400–404 science, 569
meaning of life, 497 divine attributes and unique testimony for revelation, 203
natural law, 675 quantification, 70 theories of other religions, 606
nonzero probability, 653 myth and religion, 633, 634 Jurkiewicz, Renate, 342
Pascal’s wager, 653 private vs. public revelations, 175 Just world fallacy, 555–556
physical cosmology, 597–598 reason and revelation, 172 Justification and epistemology, 120–122
quantum mechanics, 599 revelation, 166, 168–169, 170, 174 Justified belief, 478, 479, 480, 482

INDEX
K Kripke, Saul, 72, 90, 318, 478–479 Living organisms, theistic vs. naturalistic
Kahan, Dan M., 576 Kuhn, Thomas, 397 accounts, 686
Kalām argument Kvanvig, Jonathan, 143 Local atheism, 32
Craig, William Lane, 283 Local problem of evil, 542
epistemological persuasiveness, 88 L Local theism, 32
al-Ghazālī, 75 Labor and theories of religion, 607 Locke, John
predicate logic, 84–85 Lamott, Anne, 563 consciousness, 414
premise 1, 290–292 Language divine command theory, 455
premise 2, 284–290 existential proof through nominal
descriptive properties of god, 29
Kant, Immanuel definition, 7
divine command theory, 457–459
a priori argumentation critique, formal logic, 81
divine revelation, 169
67–68 natural reason, 164
God vs. god, 1, 3–5, 4, 6
causation and sufficient reason, persecution, argument against, 655
meaning of life, 508 personal identity, 418
273–276 meaningfulness of, 54 representationalist epistemology, 669
Copernican revolution, 260 speaking in tongues, 221 Loeb, Abraham, 347
ethics, 449–450 univocity, equivocation, and Logic: Theism, 65–80
human rights, 485 analogy, 3, 8, 10–12 cosmological arguments, 74–78
Hume’s Fork, challenges to, 90 Laplace, Pierre-Simon, 442, 579, 585, divine command theory, 460
individuating vs. universalizing 620, 627 fine-tuning argument, 79
ideas, 353 Latin language, 4 “God” as used in proofs of God’s
matter-mind question, 50–51 Latourelle, René, 17 existence, 5–8
moral argument for God’s existence, Lavater, Johann Kaspar, 232 ontological arguments, 65–74
384 Law, natural. See Natural law rational foundations of faith, 65
moral evils, 529 The Law of Peoples (Rawls), 484 simplicity as a standard of appraisal,
natural law, 676 Lawson, E. Thomas, 631 42–44
natural vs. supernatural revelation, Leach, Edmund, 626 suffering and the problem of evil,
164 Lee, Patrick, 669 529–533
ontological arguments, objections Lehrer, Keith, 122 teleological arguments, 78–79
to, 58, 94–96 Leibniz, Gottfried Wilhelm types of arguments, 40–41
reason and revelation, 171 Logic: Atheism, 81–101
causation and sufficient reason, 271,
self-evident truth or falsity of alleged proofs, 55–56
272
worldviews, 56–57 Anselm’s ontological argument,
definition of God, 99
Kataphatic theology, 13–15 96–98
existential proof through nominal
Keener, Craig, 223–224 Descartes and the parody objection,
definition, 7
Keil, Frank, 436 94
formal logic, 81
Kempe, Margery, 187 epistemological persuasion and
miracles, 196
Kennedy, Hugh, 402 formalization, 88–89
perfect being theology and freedom,
Kenny, Anthony, 29–30 formal logics, developments in,
423–424
Kepler, Johannes, 568 81–82
Khan, Amas, 338 Leo XIII, Pope, 172
Letter to Liebmann (Frege), 5 Gödel, Kurt, 99–101
Kierkegaard, Søren, 105, 181–183, 459, internal/external distinction, 94–96
462, 652 Leucippus, 25
Levinson, Henry-Samuel, 138 Leibniz, Gottfried Wilhelm, 99
al-Kindī, 76–77 modal logic, 86–87
Kind-name theory, 3, 6–10 Lévi-Strauss, Claude, 626
Plantinga’s modal argument, 98–99
Kinzler, Katherine, 56–57 Lévy-Bruhl, Lucien, 624, 626
positivism, 54
Kitcher, Philip, 392 Lewis, C. S., 453, 454, 671
predicate logic, 84–86
Kitzmiller V. Dover (2005), 370 Lewis, Geraint, 335–339, 347
propositional logic, 82–84
Kloet, James D., 400 Lewontin, Richard C., 370–371 validity, soundness, and
Knowledge Liber de causis, 264–265 epistemological persuasiveness,
Augustine, 666–668, 669 Life after death. See Immortality 87–89
belief, 104–105 Life stance, theism as a, 35 worldviews, 681–684
causal perception, 276–278 Life-form Logical argument from evil, 529–533
epistemic abduction, 114–115 argument from design, 434–435 Logical form, 84, 85
nature of morality, 478 bubble universes, 355 Logical positivism, 27, 54, 508
Pascal’s wager, 659–660 intelligent design, 333 Logical possibility/impossibility
rational belief, 119–120 mutations, 380 divine command theory, 457
religious diversity, 251 Pauli exclusion principle, 339, 352 doxastic foundations, 115, 116
scientism, 609–610 Lifesaving text, sacred text as, 131 faith and revelation, 182
theories of religion, 609–614 Light-year, 361–362, 587 materialism, 418
See also A priori Likelihood. See Probability methodology, 41, 50, 115
Kors, Alan, 553 Linde, Andrei, 346, 347 suffering, 533
INDEX
Logical problem of evil, 93, 198, 542 divine command theory, 460 Metaphysics
Lokāyata, 24 Gödel, Kurt, 69 Aristotle, 261–264
Loll, Jerzy, 342 phenomenological concept of Augustine, 667, 669
Lonergan, Bernard, 276, 277, 353, 356 consciousness, 431 causation and sufficient reason,
Love, appeal to, 469–470 scientism, 608 259–260, 282
The Lover of Lies (Lucian), 220 Tegmark’s Level IV multiverse, of consciousness, 430–432
Low entropy at the big bang, 333–334, 353–354 ethics, 453
340–345, 597–598 transcendence intelligence theory of of freedom, 669
Lucian of Samosata, 220 fine-tuning, 355 God and freedom, 424
Lucretius, 25, 297–298 Matter, necessity of, 297–298 knowledge, 611
Luhrmann, Tanya, 138–139, 141 Matter and mind, 50–51, 53 Leibniz, Gottfried Wilhelm, 271
Luther, Martin, 455, 462, 612 Mavrodes, George I., 216 materialism, 668–669
Mawson, T. J., 493–494 matter, 298
M Maximal greatness, 70, 96, 98–99 pluralism, 237, 238–239, 245–246
Mackie, J. L. McCauley, Robert, 621, 631 theistic metaphysical singularism,
conscience, 451–452 McClennen, Edward, 653 495–500
God as perfectly good, 55 McCloskey, H. J., 530, 532, 535 Thomas Aquinas, 265–267, 674
miracles, 214 McDowell, John, 669 Metaphysics (Aristotle), 74–75, 259,
ontological arguments, 94 McGinn, Colin, 425 261–264, 266, 528
McGrath, Alister, 497–498 Method: Theism, 35–47
suffering, 530–531, 532–533, 535,
McKim, Robert, 231 argumentation, 40–42
536
McMullin, Ernan, 379 ethics of belief question, 44
Macroevolution, 368
Meaning (language). See Semantics moral responsibility, 45–46
Macroscopic irreversibility, 598–599 simplicity as a standard of, 42–44
Meaning: Theism, 491–506
Magic, 634 thick theism, 35–40
experiential singularism, 502–503
Magical thinking theory, 626 Method: Atheism, 49–63
metaphysical singularism, 495–500
Maimonides, Moses, 11–12, 72, 138, narrative singularism, 500–502 alleged proofs of, 55–56
165–166, 197 naturalism, 491 burden of proof, 57–58
Maitzen, Stephen, 550 pluralism, 493–495 falsifiability of features of, 54–55
Malcolm, Norman, 66 subjective singularism, 503–504 fundamental accounts, 49–50
Malebranche, Nicolas, 196, 272 surrogate theory, 492–493 matter-mind question, 50–51
Malino, Jonathan, 138 Meaning: Atheism, 507–521 meaningfulness of language, 54
Malinowski, Bronislaw, 625 etiology of religious belief, 392 norms and values, 59–61
Manichaeism, 565, 666, 669, 671 God-based approach to meaning, noumenal vs. phenomenal accounts,
Many worlds theory. See Multiverse 513–516 49
Many-contenders objection, 181, 185, individual meaningful lives, private evidence, 58–59
186, 660–661 509–511 purpose of worldviews, 52–53
Margolis, Howard, 397 meaningful life on balance, 512–513 scientific study of religion,
Marian apparitions, 175 objectivism, 517–518 623–624
Maritain, Jacques, 266–267, 483 part-life and whole-life dimensions, self-evident truth or falsity, 56–57
Maritain, Raïssa, 580 511–512 stipulating two worldviews for
Marsh, Jason, 550 subjectivism, 516–517 comparison, 51–52, 61–62
Marx, Karl, 607, 609, 624, 625–626 Meaningfulness of the language of wishes and wants and the
Mass-energy density of the universe, worldviews, 54 comportment of worldviews, 61
334–335 Mécanique Céleste (Laplace), 627 Methodological materialism, 668
Material causation, 290, 296 Mediation and revelation, 168–169, Methodological naturalism
Materialism 173–174, 190–192 ethics, 479
abstract objects, 327–328 Meditation, 592–593 God of the gaps, 426
Augustine, 668–669 Meditations (Descartes), 3, 57, 94, 270, ontological naturalism vs., 51
belief, 416 410 religious experience, 141–142
free will, 420, 421 Meletus, 25 scientific method, 40
God of the gaps, 425, 426 Mencken, H. L., 26 theories of religion, 619, 630–631
history of atheism, 24, 25 Mendelssohn, Moses, 232, 234 Microadaptation, 368
intentionality and consciousness, 412 Mental concepts, 666–668 Microevolution, 368
mind-body problem, 411 Mereology, 294 Mill, John Stuart, 5, 199, 207–208, 445,
personal identity and immortality, Merits of worldviews, 61–62 536
418 Mesopotamia, 395 Miller, Geoffrey, 628
Mathematics Metaethical divine command theory, Milton, John, 540
applicability of, 687–688 469–471 Mind and Cosmos (Nagel), 669
axioms, 565 Metaethics, 230, 452–453, 480 Mind and matter, 50–51, 53
causation and sufficient reason, Metaphysical naturalism, 426, 630–631 Mind and World (McDowell), 669
273–274 Metaphysically possible miracles, 213 The Mind of God (Davies), 363

INDEX
Mind-body problem. See Dualism concepts of god, 28 Morris, Simon Conway, 389
Minimal kind-name theory of God, displacement of non-monotheism, Morrison, Toni, 510
6–8, 10 400–404 Morriston, Wes, 292
Minimal theism, 7 early Christians as quasi-atheists, 25 Motion, 261–264, 266–268
Minkowski vacuum, 350 existential proof through nominal Motivation and the meaning of life, 520
Miracles: Theism, 195–210 definition, 7 Muhammad, 127, 168, 248
alternative explanations, 197–199 God vs. god, 4, 8 Multiverse
causation and sufficient reason, miracles and religious tradition, 211 anthropic principle, 360
271–272 ontological aspects, 19–21 Boltzmann brains and brief brains,
in different religions, 205–207 as quasi-atheism, 25 348–352
evidence of God’s existence, 208–210 religious experiences, 138, 139, 150 physical laws, 339–340
firsthand experience of miracles, 201 sacred texts, 127–133 as response to the fine-tuning
Hume, David, 199–201 theories of religion, 605–606 argument, 363–364
Hume’s arguments, critiques of, unique quantification, 71–72 technical problems with, 345–347
201–205 Moore, G. E., 234, 458, 478 Tegmark’s Level IV multiverse,
natural order, 195–197 Moral agency, 420, 422–423, 461 352–354
supernatural agent, 195 Moral beliefs transcendent intelligence theory,
Miracles: Atheism, 211–226 epistemology, 480–481 354
atheism, claims as evidence for, justifications for, 479 Mutual aid, principle of, 477
222–223 similarities and differences in, My Confession (Tolstoy), 513
as divine revelation, 180 472–477, 482 Mystery, soul as, 438
four senses of miracles, 211–212 Moral evil Mystical experience. See Religious
origin of claims, 220–222 aporia, 528–529 experience
possibility of, 212–214 Augustine, 670 Mystical Theology (Dionysius), 12
quantum mechanics, 599 free will, 423 Mysticism
rational historical empiricism, natural evil vs., 382, 390 as basis for belief, 124–125
219–220 suffering, 542 religious diversity, 249–255
scientific interpretations, 217–219 trivial moral evil, 525–526 religious experience, 138, 140–143,
theistic vs. naturalistic accounts, vs. physical evil, 526–527 151, 154, 156, 157
689–690 Moral facts The Myth of Sisyphus (Camus), 513
Mirza Ali Muhammad, 175 cognitive blocking factors, 552 Mythological truth, 247, 249
Misotheism, 23 ethics, 452–453, 468–469 Mythology
Modal knowledge, 317–319 suffering, 545 biblical legends, 221
Modal logic vs. moral beliefs, 472–473 function of, 219
Anselm’s ontological argument, Moral judgments, 479–480 overlap in, 250
97–98 Moral law, 673–674
social function of, 191
cosmological arguments from Moral psychology, 667, 675
theories of religion, 626–627,
contingency, 294, 295, 296 Moral realism, 672–673
632–634
formal logics, 86–87 Moral responsibility, 422
Gödel, Kurt, 100 Moral skepticism, 545–546
Plantinga, Alvin, 69–70, 98–99 Moral worthiness, 450 N
Modal status of miracles, 214 Morality Nagel, Thomas
Model of dialectical presence, 170 Darwinism, 389 anti-theism, 23
Moderate materialism. See Nonreductive dependence on God, 672–673 ethics, 451, 454
materialism Euthyphro problem, 73–74 irreducibility of mind to matter, 669
Modernity and decline in theism, 398, as evolutionary adaptation, 384–385 naturalism, 409
399 foundational evidence for belief in problem of evil, 530
Modes of thought, 632–633 God, 109–110 subjective character of experience,
Modified divine command theory, imitatio Dei, 111–112 414
459–460 meaning of life, 510, 514–515 Nahmanides, 174
Modified theistic activism, 325 natural law theory, 673–676 Names, logic of, 28–32, 70–71, 72
Modus ponens and modus tollens, 89, 99 nonbelief, 552 Naming and Necessity (Kripke), 72
Monadology (Leibniz), 271 normative aspects of theism, 36–37 Nanak, 248
Monism Pascal’s wager, 659–660, 661 Nancy, Jean-Luc, 498–499
abstract objects, 328 pluralism, 237–238, 248 Narrative freedom, 444
metaphysics, 252 rationality and grounds, 45–46 Narrative singularism, 500–502, 505
pluralism with, 238 religious diversity, 230, 234–235, 240 Natural causal reality, 682
religious diversity, 231, 232–236 revelation, 177, 181–182, 183–184 Natural deduction, 81
vs. pluralism, 236, 239 of saints, 248 Natural evil, 382, 390, 542
Monotheism skepticism, 115 Natural faith, 614–615, 622–623
aspects of atheism in, 25–26 theistic vs. naturalistic accounts, 688 The Natural History of Religion (Hume),
cognitive aspects, 22, 23–24 Mormon Church, 402 623–624, 652
INDEX
Natural language arguments, 85 science as distinct from, 377–378, reformed epistemology, 108
Natural law 409 revelation, 164, 169, 189–190
ethics, 463 suffering, 546–547 as word of God, 127, 129
fine-tuned universe, 339 theistic metaphysical singularism, New theists, 2
human freedom, 442, 444 496–497 Newman, John Henry, 167, 450–452
miracles, 199, 200, 203–204, theories of religion, 619–623 Newton, Isaac, 267, 397, 584–585, 588
207–209, 212–219 vs. nonnaturalism and belief, Newtonian laws, 266, 273, 274
morality, 673–676 622–623 Nicholas of Cusa, 568
similarities and differences in moral worldviews, 51–52 Nickell, Joe, 224
beliefs, 474 See also Methodological naturalism Nicomachean Ethics (Aristotle), 264
theories of religion, 627 Nature Nietzsche, Friedrich, 60, 452, 492
worldviews, 51 balance of, 370, 371 Nile Valley civilization, 395
Natural order, 195–197, 261–262 as designed, 43 No-defeater requirements of rationality,
Natural revelation, 163–165 necessity, as form, 271 121, 123, 124, 125
Natural selection The Nature of Necessity (Plantinga), 70 Noetic quality of religious experiences,
Darwinism and human origins, Nazism, 399 140
387–391 Near-death experiences, 439–440, Nomen naturae. See God vs. god
epistemology, 60–61 594–596 Nominalism
human reason, 451 Necessary existents bootstrapping problem, 325, 326
human uniqueness, 407 a priori knowledge, 310 deductive arguments, 310
moral beliefs, 475–476 causation and sufficient reason, 261, existential proof through nominal
See also Adaptation and 271–272 definition, 7–8
adaptationism contingency argument, 77–78 indispensability argument, 326–327
Natural theism, 202–203 cosmological arguments, 74–75, 297 theories of religion, 612
Natural theology Gödel, Kurt, 100 types, 322
biology, 370 higher-order predicate logic, 86 Nonalethic function of belief, 622
Darwin, Charles, 627 internal/external distinction, 95–96 Nonbelief, varieties of, 552–553
doxastic foundations, 111 modal logic, 86–87 Noncognitivism and conflicting
inference to best explation, 631 Plantinga’s modal argument, 99
revelations, 187
ontological argument, 94 rigid designators, 72
Noncontradiction, principle of, 271,
revelation, 163–164, 192 Thomas Aquinas, 30, 267
273–274, 533, 674
Natural Theology (Paley), 359, 366, 585 Necessity
The Non-existence of God (Everitt), 371
Natural vs. supernatural revelation, causation and sufficient reason, 274,
Noninferential evidence, 105
163–165 294
Naturalism Non-inferential justification for belief,
of matter, 297–298
alleged proofs of, 55–56 modal knowledge, 318 121
balance of probability, 588–589 quantum mechanics, 599–600 Noninjury and noninterference,
chance, 379 Negative atheism, 22 principle of, 477
consciousness, 433 Negative dependence on God, 36 Nonintelligence proposal/intelligence
early theories of religion, 624–625 Negative theology. See Apophatic proposal, 355
ethical naturalism, 452, 453 theology Noninterventionist induction, 204
evolutionary theory, 307–308, Neil, David, 487 Nonnaturalism
628–629 Neo-Aristotelianism, 264, 673 causal reality, 682
geographic distribution of nonbelief, Neocreationism, 368, 370 ethical Nonnaturalism, 452
550 Neo-Marxist anthropology, 625–626 ethics, 479
God of the gaps, 425, 426 Neoplatonism, 12, 264–265, 670 explanations of religious belief, 622
intentionality and consciousness, Neurophenomenology, 431, 433 reason, 434
412–413 Neuroscience Nonpersonal causation, 292
intuitive knowledge, 317 improbability of truth of theism, 587 Nonreductionism and the meaning of
living organisms, 686 near-death experiences, 440 life, 496
meaning of life, 500, 513, 515–519 personal identity, 418 Nonreductive materialism, 411, 413,
mind-body problem, 409 problem of interactionism, 420–421 420, 421
miracles, 195–197, 209 religious experience, 141–142, 152 Nonrigid designators, 72
near-death experiences, 594–596 revelation, 187 Nontheistic belief, distribution of, 691
ontology, 21–22 theories of religion, 627 Nontheistic mystical experiences, 140
purpose of worldviews, 52–53 New Age, 244, 598–599 Norenzayan, Ara, 628, 690
reason, 436 “New atheism” movement, 607 Normalizability problem of the fine-
reason, knowledge, and truth, New awareness, revelation as, 170–171 tuning argument, 362
415–416 New religions, 243–244 Normative agency and human dignity,
religious experience, 141–142 New Testament 486
religious overlap, 256 convictional evidence for belief in Normative claims of the worldview,
revelation, 187–188 God, 111–112 59–61

INDEX
Normative divine command theory, “On Miracles” (Hume), 214 Orthodox conception of human rights,
470–471 On the Divine Names (Dionysius), 13 482, 484
Normative implications of theism, On the Genealogy of Morality Orthodoxy, 610
36–37 (Nietzsche), 452 Orthography, 1, 6
Normatively responsive organisms, 689 On the Gods (Protagoras), 26 Orthopraxy, 610
Norse religions, 244 On the Incarnation (Athanasius), 565 Ought-implies-can principle, 655
“Noseeum” argument, 543, 544 On the Origin of Species (Darwin), 38, Out-of-body experiences, 439–440,
Not-can-implies-not-ought principle, 359, 585, 627 594–596
655 On the Uses of the Parts of the Body Overlap, religious, 249–250, 255–256
Nothingness and the meaning of life, (Galen), 588 Overlapping consensus on human
498, 499–500 One-over-many argument, 322, 326, rights, 484
Noumena, 49, 236–239, 245–246 327 Overrider system, 253, 254
Nous, 78 Ontic divine command theory, 455, 461 Oxford English Dictionary, 1
Nowell-Smith, Patrick, 207 Ontological arguments
Nozick, Robert, 495–496 abstract objects, 323, 324, 326–327 P
Nuclear force, 337–338 Anselm, 65–67, 96–98 Page, Don, 349–350, 351
argument types, 41 Paine, Thomas, 190
O causation and sufficient reason, Paleoanthropology, 587
Object-experiences vs. God-experience, 270–271, 275 Paley, William, 359, 366–367,
135–136, 149 critiques of, 67–68 383–384, 585
Descartes, René, 94
Objective reality, 259–260, 270–271, Panentheism, 20, 21
divine attributes and unique
275, 354, 613, 673 Panpsychism, 390, 432
quantification, 70–74
Objective revelation, 177 Pantheism, 1–2, 8, 20–21
doctrine, 19–22, 24
Objective validity, 274 Paradox of the stone, 73, 91, 92
Gödel, Kurt, 68–69, 99–101
Objective vs. formal realities, 259–260 Paradoxes
indispensability argument, 321
Objectively real parameters, 339 divine attributes, 73, 90–93
Kant’s objections to, 94–96
Objectivism, 462, 517–518 mystical experience, 254
Leibniz, Gottfried Wilhelm, 99
Objectivity of moral value. See Moral Schrödinger paradox, 345
minimal kind-name theory, 6
facts Plantinga, Alvin, 69–70, 98–99 Parameters, 331, 361–363, 685, 686
Obligation problem of evil, 93–94 Parapsychology, 223, 595, 598
divine command theory, 480, 487 reference-fixing definitions of god, Parfit, Derek
moral realism, 468, 469, 470, 471, 31 morality, 478, 481, 482
478 theist worldview, 55 on pain, 540
motivation, 479 Ontological atheism, 21–22, 28 personal identity, 418
Observability, 37–38, 39, 40 Ontological commitment, criterion of, principle of sufficient reason, 293
Occasionalism, 196 58 Parmenides, 276
Ockham’s razor, 42–43, 346, 432, 433 Ontological naturalism, 51 Parody objections, 94–96
“Of Miracles” (Hume), 199–200, 205, Ontological possibility, 51 Particle physics, 685
394 Open naturalism Part-life meaning, 511–512, 513
Olaf Tryggvason, 654–655 alleged proofs of, 55–56 Pascal, Blaise, 44, 173–174, 637
Old Testament as word of God, 127, falsifiability, 55 Pascal’s wager
129 fundamentals, 51–52 many-contenders objection,
O’Leary-Hawthorne, John, 310 norms and values, 60–61 660–661
Olum, Ken, 350, 351 purpose of worldviews, 52–53 practical consequences of belief, 692
Omni-properties of God self-evident truth or falsity of, prudential/pragmatic arguments,
consistency issues, 91, 92 56–57 649–654
defining god, 20 worldviews, 54, 62 rationality, 658–660
deism, 2 Open question argument, 458, 478 reward for belief, 656, 657–658
differing concepts of god, 28, 29, 32 Open theism, 2, 36 summary of, 638–641
divine command theory, 461 Openness and miracles, 206–207 Past hypothesis, 340, 342–343, 345
freedom, 441, 612 Oppy, Graham, 6, 69, 291 Pathology, religious experience as, 141,
intuitive knowledge, 316, 317 Optimal efficiency, principle of, 43 187
paradoxes, 73 Optimal moral horizon, 545–546 Patternism, 439
problem of evil, 535 Order/disorder, 261–262, 334 Paul, 111–112
reason, 435–436 Ordinary observers, 348–352 Paul, Gregory, 691
suffering, 530–531 Oresme, Nicholas, 568 Peacocke, Arthur, 381–382, 383
theism types, 36 Orientation, objectivity of, 153 Penrose, Roger, 333–334, 340, 343,
worldviews, 681–682 Origin of civilization, 395–396 346, 347
On Divination (Cicero), 605 Original sin, 21, 390, 612 Pensées (Pascal), 649, 658
“On Genesis” (Augustine), 379 Orr, H. Allen, 370 People of the Book, 565, 654
On Human Rights (Griffin), 485 Orthodox Church, 167, 168, 172 People’s Temple, 253
INDEX
Per se vs. per accidens cause, 268–269 fine-tuning argument, 685 Politics and human rights, 483–484
Perception. See Sense and perception laws of science, 566 Polkinghorne, John, 381–382, 383
Perdurance, 668 soul, 438 Pollack, Jason, 350–351, 352
Perennial Philosophy (Huxley), 250 Thomas Aquinas, 266 Polytheism
Perennialism, 244, 249–252, 255, See also Fine-tuning argument cognitive aspects, 22, 24
256–257 Physics (Aristotle), 259, 261–264 God vs. god, 4, 8
Perfect being theology Physiology and revelation, 187 Greece, ancient, 25
abstract objects, 324 Pietism, 182 as included in theism, 1
Anselm’s ontological argument, Pike, Nelson, 532, 535 monotheism’s displacement of, 401,
65–66 Pinker, Steven, 24 402–403
apophatical theology, 13 Pius X, Pope, 176–177 ontological aspects, 19, 21
causation and sufficient reason, 271 Plan B (Lamott), 563 theories of religion, 605, 606
consistency issues, 91–92 Plantinga, Alvin vs. pluralism, 238
divine attributes, 9 a priori knowledge, 304, 315–317 Polyvalence of meanings of life, 494
divine command theory, 459–460, cognition, 434 Popoff, Peter, 143
461 evolutionary argument against Popper, Karl, 55
freedom, 423–425 naturalism, 307–308 Positive atheism, 22–23
kataphatical theology, 13–15 modal arguments, 100 Positive dependence on God, 36
logical inconsistency, 55 ontological argument, 69–70 Positive properties of God, 68–69
minimal kind-name theory of God, open naturalism, 60–61 Positivism, 27, 54, 508
6, 7, 9–10 problem of evil, 533, 535 Possibility
suffering, 551 reason, knowledge, and truth, miracles, 212–214
theism types, 35–36 415–416 modal knowledge, 318
Perfection and causation, 269 reformed epistemology, 105, probability-possibility difference,
Perfectly-loving God, 137–138 107–108, 124–125 654
Performatives, 626 religious diversity, 228–229, 233, problem of counterintuitive
Perrine, Timothy, 549 234 possibilities, 469
Perry, John, 444–445 religious experience, 143 of revelation, 176–177
Persecution, arguments against, 655 sensus divinitatis, 623 Possible existents, 77–78, 86–87
Persian Empire, 301 theories of religion, 619, 620, Possible greatness, 97–98
Personal causation, 291 621–622 Posterior probability, 305, 308, 546
Personal identity, 417–419, 436–440 Plato Post-Fregean semantics, 3, 4, 11
Person-human distinction, 52 ban of the poets, 605 Postmodern philosophers, 260
Persuasiveness, epistemological, 87–89 Euthyphro problem, 73–74, 454 Power and natural law, 674–675
Phenomena evidence and knowledge, 104–105 Practical conception of human rights,
pluralism, 238, 245–246 God vs. god, 4 482–483
scientism, 608 human freedom, 441 Practical consequences of theism,
Phenomenal conservatism, 317 matter-mind question, 50 691–693
Phenomenology norms and values, 60 Practical principles, 43
a priori knowledge, 309–310 revelation, 165 Practical reasoning, 658–660, 674, 675
causation and sufficient reason, 260, Platonism Practice-faith relationship, 610, 611,
262 a priori argumentation, 311 661–662
consciousness, 431 abstract objects, 319–320, 322–324, Pragmatic arguments. See Prudential/
meaning of life, 498–499 326–328, 327 pragmatic arguments
personal identity, 418 apophaticism, 14 Predicate logic, 81–82, 84–86, 97–98,
Philosophe, 434 atheism, 327 320–321
Philosophers, 409, 621 Augustine, 666, 667, 669 Predicate nominalism, 322
Philosophy, history of, 455–456 Field, Hartry, 311 Prediction and prophecy
Philosophy and the Mirror of Nature See also Neoplatonism conceptual schemes of science, 53,
(Rorty), 669 Plato’s heaven, 310–311 56, 59
Philosophy of the mind, 411–414, 425, Pleasure and the meaning of life, 504 evidence of, 403
426 Plotinus, 4, 670 miracles, 224
Physical evil, 526–527, 529, 670 Pluralism multiverse theory, 349
Physical reality, 283, 289–290, 291 atheistic viewpoint, 244, 245–249, normative and evaluative
Physicalism, 599–600 256 foundations of epistemology,
Physico-theological arguments, 275–276 miracles, 206–207 60–61
Physics religious diversity responses, revelation, 188–190
Aristotle, 261–264 231–232 sacred texts, 129–130
causation and sufficient reason, 273 theistic viewpoint, 233, 236–239, scientific predictions by theism, 589
cosmology, 571, 596–598 240, 493–495 scientific study of religion, 624
determinism, 381 Plutarch, 4 Premises. See Logic
evolution, 389 Polanyi, Michael, 276 Price, Richard, 455–456

INDEX
Prima facie modal knowledge, 317–318, Process theology, 36 Q

319 Proclus, 12 Q nonuniformity, 333, 334, 335
“Primitive” religions, evolutionary Prometheus Bound (Aeschylus), 566 Qualia, 414, 590
theory of, 610 Proof. See Burden of proof; Causation Quality of life, 399–400, 659–660
Principia (Newton), 397 and sufficient reason; Logic; Method; Quantifiable universal ideas, 353–354
Principle of alternative possibilities, specific argument types Quantification, unique. See Unique
444–445 Proper functionalism, 316 quantification
Principle of mutual aid, 477 Proper name God. See God vs. god
Quantifiers and abstract objects, 321,
Principle of noncontradiction, 271, Properly basic belief, 143, 184–185, 233
326–327
273–274, 533, 674 Properties of God. See Divine attributes
Quantum cosmology, 289, 597
Principle of noninjury and Property dualism, 411
Quantum gravity, 342
noninterference, 477 Prophecy. See Prediction and prophecy
Propositional connectives, 83 Quantum loopholes, 598
Principle of optimal efficiency, 43
Principle of sufficient reason, 259, 271, Propositional logic, 68, 70, 82–84, 89, Quantum many worlds model, 346
283, 293–295, 298, 303–304 530–532 Quantum mechanics, 571, 598–600,
Prior probability, 252, 303–304, 305, Propositions, unique. See Unique 685
308 quantification Quantum soul, 438
Private vs. public revelations, 174–176 Pros hen equivocation, 11, 12 Quantum theory of human freedom,
Probabilistic argument from evil, Proslogion (Anselm), 9–10, 65–66, 96 441
533–534 Protagoras, 26 Quasi-atheism, 25
Probability Protestantism Quasi-symbolist theory of religion, 632
a priori knowledge, 304–308, fideism, 172 Quine, Willard Van Orman
310–311 Reformation, 612–613 analytic vs. synthetic truths, 314
fine-tuning argument, 361–363 revelation, 168, 170 conceptual scheme of science, 53,
Pascal’s wager, 639, 650–654 reward for belief, 658 54
rational degree of belief, 27–28 Provenance, objectivity of, 153 criterion of ontological
scientific study of religion, 624 Providential guidance. See Divine commitment, 58
suffering, 546–549 intervention existential proof of God, 8
theism vs. naturalism, 588–589 Providentissimus Deus (Leo XIII), 172 Hume’s Fork, challenges to, 90
Problem of contingency, 228–230, 234, Prudential/pragmatic arguments: indispensability argument, 320–321
235, 237, 469 Theism, 637–644 open naturalism, 51
Problem of counterintuitive possibilities, cognitive vs. pragmatic arguments, science as tool for prediction, 53, 56
469 637–638 Quinn, Philip, 233
Problem of divine hiddenness, 549–553, infinite value, 642 Qur ʾan
554, 661–662 James’s will to believe, 641–642 creation argument, 75–76
Problem of divine inscrutability, Pascal’s wager, 638–641 divine revelation, 168–169
553–554 three propositions for pragmatic immortality, 437
Problem of evil faith, 643–644 myth and religion, 633, 634
a priori arguments, 93–94 Prudential/pragmatic arguments: negation, appeal to, 71–72
alleged proofs of atheism, 55 Atheism, 645–663 revelation, 166–167, 174
consistency issues in theism, 91–92 evidentialism, 645–649 as word of God, 127, 129
Darwinism, 390 Pascal’s wager, 649–654, 658–660
doxastic theism, 45 reward for belief, 654–658 R
epistemological persuasion and truth, theory of, 53
Pruss, Alexander, 99–100 Rachels, James, 461
formalization, 89
Euthyphro problem, 74 Pseudo-Dionysius. See Dionysius Radcliffe-Brown, Alfred, 625
evidential arguments, 533–534 Psionics, 223 Rahner, Karl, 177, 235–236
evolution, 378, 382–383 Psychological continuity, immortality as, Railton, Peter, 478–479, 481
God and freedom, 423–424 436–437 Ramachandran, V. S., 631
logical arguments, 529–533 Psychology, 187, 607, 624–625, 627 Randi, James, 223
meaning of life, 504 Psychosomatic cures, miracles as, Randomness
miracles, 198 221–222 cosmology, 597
Pascal’s wager, 661 Psychotropic substances, 593 evolution and chance, 379, 380
presuppositions of, 534–536 Ptolemy, 566 physical cosmology, 597
propositional logic, 83–84 Public observability, 38 quantum mechanics, 598
suffering, 523–524, 527, 539–542 Public opinion on religion-science See also Chance
vs. problem of divine hiddenness, compatibility, 607 Rasmussen, Joshua, 295–296
550–551 Public vs. private revelations, 174–176 Rational agency, 421
Problem of induction, 59 Punishment, retributivist, 450 Rational degree of belief, 27–28, 123
The Problem of Pain (Lewis), 671 Purpose and the meaning of life, 494, Rationality
Problem of the gap, 5 498, 504, 514–516, 516–517 compelling vs. defensible arguments,
Process theists, 2, 10 Putnam, Hilary, 318 637–638
INDEX
Rationality, continued natural law, 675–676 soteriological exclusivism or

evidence and evidentialism, 122–123 Pascal’s wager, 640, 658–660 inclusivism, 235–236
evil, 525–526 problem of evil, 531 theistic responses to religious
foundationalism, 120–121 Reppert, Victor, 434–435 diversity, 230–232
inferential justification for theism, revelation, 164–165, 171–172, 185 Religious diversity: Atheism, 243–257
123–124 Rist, John, 673 apologetic strategies, 244–245
intellectual responsibility, 44 theories of religion, 609–614 miracles, 222–223
internalism vs. externalism, 121–122 See also Causation and sufficient moral beliefs, 472–477
moral responsibility, 45–46 reason; Rationality Pascal’s wager, 652–653, 660–661
Pascal’s wager, 652 Rebutters, 253, 254 perennialism, 249–252
practical consequences of belief, Reductio ad absurdum arguments, 67, 87 problem of disagreement, 125–126
692–693 Reductionism, 282, 496 revelation, 181, 186–187
problem of disagreement, 125–126 Reductive materialism, 411, 421 sacred texts, 127–129, 131, 133
religious diversity, 228–230, 256 Rees, Martin, 339 skepticism between religions, 148–149
simplicity as a standard of appraisal, Referents theistic vs. naturalistic accounts, 690
42–43 intentionality and consciousness, 413 unreliable sources of belief, 586
theism, 35, 42–43 ontological argumentation, 70–71 worldviews, 682
theistic vs. naturalistic accounts, 687 proper names, 2–3, 5, 11 Religious doctrine, 170, 250–251
theories of religion, 626 reference-fixing definitions of god, See also Religious diversity
See also Causation and sufficient 29–31 Religious experience: Theism, 135–145
reason; Reason Reflective equilibrium, 479–480 availability, 136–138
Rationally compelling claims, 637–638, Reformation, 612–613 cross-checking, 138–139
641, 643 Reformed epistemology, 58, 105–109, general analogy, 143
Rationally defensible claims, 637–638 124–125, 181, 184 intersubjective agreement, 139–140
Rationes seminales, 379 Regress argument for justified belief, 121 mark of the genuine, 142–143
Ratzinger, Joseph, 616 Relation, meaning as, 499, 502 naturalism, 141–142
Ravizza, Mark, 422 Relativity theory. See General theory of object-experiences vs. God-
Rawls, John, 478, 484 relativity experience, 135–136
Real Ethics (Rist), 672–673 Reliabilism, 121–122, 310 possible conclusions, 144
Real existence, 95, 96 Reliability and religious texts, 219–220 pragmatic vs. cognitive arguments, 637
Realism Religion sense organs, 142
Augustine, 666–668 antitheism, 23 vividness, 140
causation and sufficient reason, 282 evolutionary study of, 384–385, Religious experience: Atheism, 147–161
faith and revelation, 185 391–394 atheism on religious experience, 147
god, concept of, 28–29 historical religions, 394 cognitive science, 590–591
science, 565–566, 569 human rights, 484–485 direct experiential theism arguments,
Reality moral agreement and disagreement, 155–158
causal reality, 682, 684–685 482 indirect experiential theism, 158–160
causation and sufficient reason, normative aspects of theism, 37 neuroscience, 591–594
259–260, 270–271, 275 supernatural revelation claims, openness to religious experience,
contingent concrete reality, 289, 165–167 147–149
294, 295–297, 313 theism vs., 35 as private evidence, 58–59
fundamental reality and worldviews, tripartite distinctions, 21 reformed epistemology, 124–125
62 unusual experiences as religious religion and endorsement of, 153–155
human uniqueness, 408 experiences, 150–151 religious diversity, 249–254
materialism, 668–669 See also Religious diversity religious experience as communally
meaning of life, 495–500 Religion, theories of. See Theories of objective, 149–151
objective reality, 354, 673 religion restrictions on critiques of religious
physical reality, 283, 289–290, 291 Religion within the Limits of Reason Alone experience, 151–152
pluralism, 245–247 (Kant), 171 scientific explanations, 587–588
science and naturalism, 378 Religions of the Book, 566 theistic vs. naturalistic accounts, 689
theories of religion, 613–614 Religious ambiguity, 245–246 See also Faith and revelation;
Realm of the Forms (Plato), 50–51, 56, Religious authority and tradition, Miracles
60 131–133 Religious fundamentalism
Reason Religious diversity: Theism, 227–241 failures of, 575–576
axioms, 564–567, 571 contingency problem, 228–230 sacred texts as literal truth, 127
cognitive science, 435 dual perspective, 239–240 science, 568–569, 576–578
God’s capacity for, 435–436 epistemic issues, 230 Religious Right, 576
meaning of life, 518, 520 issues, 227–228 Religious texts. See Sacred texts
metaethical divine command theory, miracles, 205–207 Religious tradition
470 pluralism, 236–239 authority, 131–133, 168
moral capacity, 481–482 revelation, 174 ethics, 460, 481

INDEX
Religious tradition, continued interpreting myth, 633–634 quantum mechanics, 598–600

miracles, 205–207, 208, 211, 215, miracles, 214, 219–220, 221, 224 religious experience, 590–594
220 reliability of, 219–220 religious overlap, 256
religious diversity, 243–244, revelation, 167–168, 172 soul, 438–439
247–249, 252, 253–254, 256 science, 574, 575, 577, 578 theism, scientific explanations of,
revelation, 186, 202–203 theories of religion, 626–627 587–588
science, 592–593, 598 “two books,” 562 as theistic retreat, 584–585
Reports and epistemic evidence, 116–117 See also specific texts theories of religion, 623–627
Representationalism, 259–260, 270, Sagi, Avi, 459, 460, 462 universal vs. exceptionalist atheists,
273, 276, 278, 669 Sai Baba, Sathya, 248 584
Res extensa, 272 Saintliness, 248–249 unreliable sources of belief,
Rescher, Nicholas, 68 Salvation. See Soteriology 585–587
Resemblance nominalism, 322, 327 Sāmkhya, 24 Science curricula, 367–368
Resurrection Sartre, Jean-Paul, 267, 417–418, 494, Scientific American, 347
miracles, 215–217, 218, 219, 220, 515 Scientific inference, 368–369
221–222, 223 Sassanid Persian Empire, 401 Scientific method, 37–40, 55, 608–609
science and theology, 565 Sauter, Gerhard, 502 Scientific revolution, 396–397
Retributivist punishment, 450 Scanlon, T. M., 479–480, 481 Scientism, 608–610
Revelation. See Faith and revelation Scheeben, Matthias Joseph, 165 Scientists
Revelation at Mount Sinai, 203 Schellenberg, J. L. motivation of scientists, 572–573
Reverenceworthiness, 37 divine hiddenness, 549, 550, 551, religious faith of, 627
Revolutions (Copernicus), 397 553, 554 Scientology, 127
Rideout, David, 342 religious experience, 137, 138 Scripture and Cosmology (Greenwood), 403
Rightness in deontology, 454 Schiefenhövel, Wulf, 628 Scriptures. See Bible; Sacred texts
Rigid designators, 72 Schmidt, Wilhelm, 401 Seachris, Joshua, 500–502
R·igveda, 633 Schopenhauer, Arthur, 21 Second Confession of the Baptists
Rist, John, 672–673 Schrödinger, Erwin, 346 (1677), 167
Rolston, Holmes, III, 630 Schrödinger paradox, 345 Second law of thermodynamics, 283,
Roman Empire, 399, 402 Science: Theism, 561–581 289, 334, 340, 345
Romance languages, 1, 5, 6 Aristotle, 261 Second Vatican Council. See Vatican II
Rome, ancient, 566 authority, 578 Second-order predicate logic, 86
axioms and belief in God, 569–570 Sectarianism and revelation, 186
Rorty, Richard, 669
as common ground, 577 Secularism, 384, 553, 617
Rota, Michael, 653, 659–660
dialogue, 574–576 Seedlike principles, 379
Rowe, William, 73, 423–424, 533, 544,
fine-tuning argument, 79 Seeming and religious experience, 144
557
foundation of atheism, 578–579 Self, 417–419, 436–440
Rule of faith, 163
foundation of theism, 579–580 Self-awareness and consciousness, 412
Runzo, Joseph, 227, 238–239
fundamentalist fallacy, 577–578 Self-creative autonomy and meaning,
Ruse, Michael, 384–385, 393, 631
God of the gaps, 425, 426 494–495
Russell, Bertrand
joy, science as, 573–574 Self-fulfilling beliefs, 647
de dicto vs. de re components of laws of science, 566 Self-interest. See Prudential/pragmatic
knowledge, 105 meaning of life, 497–498 arguments
descriptive statements, 70, 72–73 miracles, 207 Self-made persons, 445
formal logic, 81 naturalism-science distinction, Sellars, Wilfrid, 53
God vs. god, 3 377–378, 409 Semantics
instinctive belief, 140 public opinion on religion’s descriptive properties of god, 29
mystical experience, 251 compatibility with, 607 divine command theory, 457–459
naturalism and religious experience, religious experience, 142 God vs. god, 4
141 social support, 567–569 intentionality, 412–413
quality of life, 660 theories of religion, 570–571 univocity, equivocation, and
revelation, 202–203, 204–205 “two books,” science and religion as, analogy, 3, 8, 10–12
untestability of claims of god’s 562 Sense and perception
existence, 27 Science: Atheism, 583–603 Augustine, 666–668
Russell, Bruce, 487 applicability of, 687–688 causation and sufficient reason,
cognitive science, 130, 589–590 276–278
S conceptual schemes of, 53 doxastic foundations, 124
S5 system of modal logic, 87, 297 cosmology, 596–598 faith and revelation, 185
Sacred Congregation for the Doctrine of improbability of truth of theism, meaning of life, 499–500, 508
Faith, 175–176 587 miracles, 215
Sacred texts miracles, 217–218, 222 mystical perception, 254–255
conflicting claims, 174 near-death experiences, 594–596 pain, 540
doxastic foundations, 127–133 probability, balance of, 588–589 religious experience, 142, 143, 149
INDEX
Sensus divinitatis, 124, 143, 192–193, Socrates, 25, 26, 645 Substance dualism, 420, 438
586, 623 See also Euthyphro problem Subtraction argument, 294, 295
Sepkoski, John J., 388 Sorkin, Rafael, 342 Suffering: Theism, 523–538
Set of one member, 71 Sorley, W. R., 453 aporia, 528–529, 537
Shapin, Steven, 396–397 Soteriological exclusivism, 231, 232, Augustine, 670–671
Sidgwick, Henry, 454 235–236, 240 disease and other physical evils,
Signaling theory of religion, 628–629 Soteriological inclusivism, 232, 234, 526–527
Significance, objectivity of, 153 235–236, 240 evidential arguments, 533–534
Signification and meaning of life, 499 Soteriological pluralism, 237–238 evolutionary theory, 382–383
Sigurd the Stout, 654, 655 Soteriology logical arguments, 529–533
Sikhism, 248 Pascal’s wager, 650, 653, 661 meaning of life, 498, 502
Sim, David, 403 pluralism, 247–248 moral evils, 525–526
Simon, Herbert, 435 suffering, 523 presuppositions of the problem of
Simplicity, divine. See Divine simplicity Soul evil, 534–536
Singer, Peter, 556 Darwinism, 390–391 problem of evil, 523–524, 539–541
Singh, Gobind, 248 dualism, 419 Suffering: Atheism, 539–559
Singularism, theistic. See Theistic immortality, 438–439 arguments from evil, 541–546
singularism meaning of life, 504 distribution of suffering, 554–557
Singularity, 289, 296, 597, 684–685 suffering, 548–549 divine hiddenness, 549–553
Sisyphus and the meaning of life, Sound argumentation, 41 theistic vs. naturalistic accounts,
509–510 Soundness vs. validity, 87 690–691
Siyyid Ali Muhammad Shirazi. See Mirza Space-time Sufficient reason. See Causation and
Ali Muhammad big bang, 571 sufficient reason
Skeptical empiricism, 272–273 distribution of life arguments, al-Suhrawardī, 172
Skeptical strategy, 555, 557 372–373 Summa contra gentiles (Thomas
Skeptical theism, 554, 555, 557 fine-tuned universe, 334, 336, 339, Aquinas), 266
Skepticism 344, 353 Summa theologiae (Thomas Aquinas), 9,
classical foundationalism, 123 immortality, 437 31, 75, 266, 463, 673–674
doxastic foundations, 114–117 physical cosmology, 597 Sum-over-histories approach to quantum
evolutionary debunking arguments, vacua and string theory, 342–343, gravity, 342
392–393 343f Supereminent theology, 668
miracles, 217 Special objects and a priori knowledge, Superior and inferior religions, 236, 237
of religion towards other religions, 308–309 Supernatural agencies, 195, 392–393
148–149 Special theism and suffering, 547–548 Supernatural revelation. See Faith and
religious experience as God Specialization and Darwinism, 389 revelation
experience, 147, 160 Spelke, Elizabeth, 56–57 Supernaturalism
revelation, 173 Spinoza, Benedict de, 213–214, 217, 432 meaning of life, 513–516, 519
Skinner, B. F., 422–423 Spitzer, Robert, 674 physical cosmology, 596–598
Sloths and adaptation, 366 Stace, Walter, 252 quantum mechanics, 598–600
Smart, J. J. C., 390 Stark, Rodney, 402 Superstring theory, 337, 339, 346
Smith, Adam, 399 Starting points and reformed Supervenience, 468
Smith, Barry C., 136–137 epistemology, 105–109 Surds, 527, 528–529, 535, 537
Smith, Joseph, 127 Statistical principles, 355 Surrogate theory, 492–493, 505
Smith, Quentin, 533 Statman, Daniel, 459, 460, 462 Suspension of belief, 23
Snorri Sturluson, 654 Steiner, Rudolf, 589 Susskind, Leonard, 341, 342, 347
Sobel, J. Howard, 100 Steinhardt, Paul, 346, 347, 349, 352 Swann, Ingo, 595
Sober, Elliott, 368–369 Stenger, Victor, 333 Swinburne, Richard
Social cognition, 630 Stephens, Christopher, 368–369 contingent independent beings, 297
Social Darwinism, 625 Stoeger, Bill, 577 evolution, 52
Social issues Stoics, 50 habits, 652
quality of life, 399–400 Stone puzzle, 73, 91, 92 religious experience, 140, 149
revelation, 191–192 Strauss, David Friedrich, 215–216 suffering, 544–545, 548
science, 567–569 String theory landscape, 342–343, 343f, Syllogisms, 75–76, 81, 261, 528, 580
Social justice, 478 347 Symbolic logic, 81–82
Social ontology, 191 Strong agnosticism, 26, 28 Symbolist theory of religion, 632, 634
Social psychology, 630 Strong nuclear force, 337–338 Synthetic truth, 273–276, 301–302,
Social science theories of religion, Stump, Eleonore, 540–541 308–309, 311, 314
606–607, 625 Suárez, Francisco, 171
Socialization and religious experience, Subjectivism T
150, 174, 187 meaning of life, 516–518 Table of causes, Aristotle’s, 282
Sociological blocking factors, 552–553 miracles, 198–199 Tacit knowledge, 276–277
Sociology, 587, 625 religious experience, 153, 156, 157 Talmud, 166, 168

INDEX
Tanakh, 166 kataphatic theology, 13–15 Thomasson, Amie, 316

Tautological statements, 71, 149, 155 natural law, 675 Thomson, Keith, 631
Taylor, Richard, 421 supereminent theology, 668 Threat and belief, 654, 657, 658
Te Velde, Rudi, 267 See also Natural theology; Perfect Three big bangs, 630
Techno-supernaturalists, 438–439 being theology Tillich, Paul, 563
Tegmark’s Level IV multiverse, Theoretical evidence and a priori Time. See Space-time
352–354, 355 knowledge, 309–310 Timeless causation, 290–291
Teleological arguments. See Intelligent Theoretical reasoning, 658, 674, 675 Tipler, Frank, 342
design Theories of religion: Theism, 605–618 Tolerance, arguments for, 655
Temporal distribution of nonbelief, 552 aim of a theory of religion, 605 “Tolerant agnosticism,” 239
Tenzin Gyatso, 592 Christian theory of religion, 616–617 Tolstoy, Leo, 502–503, 513
Teresa, Mother, 511 development of theories of religion, Tooley, Michael, 534–535
Teresa of Avila, 188 605–606 Torah, 165–166, 167, 174, 175
Testability natural and supernatural faith, Tractatus Logicophilosophicus
cosmological theory, 589 614–615 (Wittgenstein), 260
genuineness of religious experience, “objective” theory, possibility of an, Transcendence intelligence theory of
143 615–616 fine-tuning, 354–356
multiverse, 349, 364 scientism, 608–610 Transcendence of God, 20–21
scientific study of religion, 624 Theories of religion: Atheism, 619–636 Transcendent value, 642–643
theistic appraisal, 37–38 evolutionary and cognitive theories, Transcendental idealism, 90
Testimonial evidence 627–631 Transcendentalism and a priori
faith and revelation, 180, 184 myth and magic, 632–634 knowledge, 310
miracles, 199–200, 203–204, natural and supernatural faith, Transitivity rule, 436, 623–624
214–216, 219–220 622–623 Transmutationism, 627
Texts. See Sacred texts; specific texts relevance of human sciences to Trivial moral evils, 525–526
Thales, 588 theism, 623–627 Trope nominalism, 322
Theatetus (Plato), 104–105 religion as a natural phenomenon, True beliefs
Theistic activism, 324–326 619–623 epistemic justification, 121–122
Theistic conceptualism, 325–326 Theory of everything, 339–340 knowledge, 105, 119, 478
Theistic metaphysical singularism, A Theory of Justice (Rawls), 478 orthodoxy, 610
495–500 Theory of mind, 629 Plantinga’s argument, 415–416, 620
Theistic nominalism. See Nominalism Theos, 4 pragmatic arguments, 652
Theistic personalism, 11, 13 Thermodynamic arrow, 343f religious diversity, 231, 255
Theistic platonism Thermodynamics, second law of. See unethical faith, 183
atheism, 327 Second law of thermodynamics Trust, 179–180, 451
defining abstract objects, 319 Thick theism, 35–37 Truth
intuitionism, 320 Thin theism, 35, 233 analytic vs. synthetic truths, 314
knowledge of abstract objects, 320 Thinking vs. sensation and imagining, evolutionary debunking arguments,
Plato’s heaven, 311 666–668 392–394
problem of abstract objects, 323 36 Arguments for the Existence of God: intentionality and consciousness, 413
responses to arguments for abstract A Work of Fiction (Goldstein), 31 meaning of life, 505
objects, 322 Thomas Aquinas open naturalism, 60
theistic responses to the problem of analogy, 11, 12 pluralism, 237, 245, 247, 249
abstract objects, 324, 326, 327 on Anselm’s argument, 67 religious diversity, 230–232
Theistic pluralism, 233, 236–239, 240, causal perception, 277–278 science, 573–574
493–495 divine attributes, 9, 10 worldviews, 53, 56–57, 61
Theistic responses to methodological divine command theory, 460 Truth conditions, 501–502
grounds, 45–46 ethics, 463 Truth indicators, 104, 106, 107–109
Theistic singularism existential proof of God’s existence, 7 Truth-claims and sacred texts, 127
experiential singularism, 502–503 final causes, 282 Truth-functional propositional
metaphysical singularism, 495–500 God and freedom, 424–425 connectives, 83
narrative singularism, 500–502, 505 God vs. god, 3 Truth-relativism, 186–187
subjective singularism, 503–504 kataphatical theology, 13–14 Truth-tables, 84
Themistius, 7 knowledge, 611 Truth-tracking of belief, 314, 620,
Theodicy, 117, 544, 545, 555, 556–557 miracles, 212 621–622
Theodosius I, 403 natural law, 673, 675 Turing, Alan, 81–82
Theological voluntarism, 454 normative aspects of theism, 37 Turner, Terry, 626
Theology proofs of God’s existence, 30–31 Turok, Neil, 352
apophatic theology, 12–13, 14, 15, purpose of human life, 384 Twain, Mark, 183
138, 625 revelation, 164, 171, 173, 177, 180 “Two books,” science and religion as,
creation theology, 3, 9 soul and the body, 419 562, 564, 577
history of, 455–456 teleological argument, 79 Tylor, Edward B., 607, 617, 632
INDEX
Typicality conjecture, 348 science, 589, 596–598 Walls, Jerry, 449

Tyson, Neil deGrasse, 575, 577 theistic retreat, 585 Wanderings in South America (Waterton),
theistic vs. naturalistic accounts, 366
U 685–689 War, religion as a cause of, 613
Unamuno, Miguel de, 660 Universe, cause of the. See Kalām Warranted belief, 120, 316–317
Uncertainty. See Certainty and argument Wars and religion, 399
uncertainty Univocity, 8, 10–12, 13–14, 15 Watchmaker analogy, 359, 366, 371,
Underminers, 253, 254–255 Unknowability agnostics, 26–27 384, 585
Understanding The Unlimited, 495–496 Waterton, Charles, 366
meaning as distinct from, 498 Unpredictability and evolution, 379, 381 Weak agnosticism, 27–28
thinking vs. sensing and imagining, Uthman, 168 Weak dependence and divine command
667 Utile, 650 theory, 460
Unethical teachings in sacred texts, 129 Utilitarianism, 450, 474, 479–480, 556 Weinberg, Steven, 460, 627
Unique quantification, 28–32, 70–74 Utterances and theories of religion, 626, Westminster Confession of Faith, 1643,
United Nations General Assembly, 483, 632, 634 167, 168
484 “What Does It Mean to Believe?”
United States, religion in the, 398 V (Ratzinger), 616
Unity-promoting role of religion, 628 Vacua, 342–343, 343f, 350 Whitehead, Alfred North, 81
Universal atheists, 584 Validity Whole-life meaning, 511–512, 513, 514
Universal consent argument for God, 154 Anselm’s argument, 67 Wiesel, Elie, 660
argumentation types, 41 Wigner, Eugene, 331
Universal constants. See Constants
epistemological persuasion and Wilkins, John, 621–622
Universal Declaration of Human Rights,
formalization, 88–89 The Will to Believe (James), 641–642
483–484
predicate logic, 84–85 William of Ockham, 455
Universals
propositional logic, 84 Williams, George C., 628
arguments for abstract objects, 320,
Williams, Rowan, 25–26, 32
321 vs. soundness, 87
Wilson, David Sloan, 628
Tegmark’s Level IV multiverse, Value
Wilson, Edward O., 384–385, 627
353–354 ethics, 672–673
Winzeler, Robert, 401
Universe: Theism, 331–358 meaning of life, 492, 495, 509, 515
Wippel, John, 266
Boltzmann brains and brief brains, of parameters, 686
Wisdom and kataphatical theology,
348–352 Pascal’s wager, 638–641
13–14
conformal cyclic cosmology, pragmatic arguments, 641, 642–643
Wittgenstein, Ludwig
343–345 worldviews, 59–61
ambivalent attitude towards God,
constants, 332 Van Gogh, Vincent, 510
260
determinism, 380–382 Van Inwagen, Peter, 52, 541–542, 544
antitheism, 23
free parameters, 331–332 Van Riet, Thomas, 343
comprehensibility and worldviews,
fundamental forces, 336–338 The Varieties of Religious Experience
49–50
inflationary cosmology, 340–343, (James), 58 concepts of God/god, 7, 20
343f, 345 Vatican I, 164, 171, 173 divine command theory, 455
initial conditions, 332–336 Vatican II, 167–168, 169, 175–176 language, meaningfulness of, 54
logic, 79 Vattimo, Gianni, 260 science and miracles, 217–218
meaning of life, 496–497, 498, 504 Vedas, 165 Wolterstorff, Nicholas, 463–464,
multiverse theory, 345–352 Vedic religions, 24 484–485
natural law, 674 Veil of ignorance and morality, 477–478 Word of God
science, 571 Velde, Rudi te, 267 Bible as the, 127–128, 129, 131
Tegmark’s Level IV multiverse, Verifiability of religious experience, 125 revelation, 168–169, 190–192
352–354 Verificationist theories of meaning, 5, 54 World-indexed properties, 98–99
theory of everything, 339–340 Versnel, H. S., 401 Worldviews
transcendence intelligence theory, Vicarious atonement, 661 alleged proofs of, 55–56
354–356 Vilenkin, Alexander, 347 burden of proof, 57–58
uncertainty and science, 546 Virtual reality, 509 case for atheism, 679–684
Universe: Atheism, 359–375 Virtue and virtue ethics, 449–450, 463, ethics of belief question, 44
creationism, 367–369 542–543, 691 falsifiability of features of, 54–55
design argument, 366–367 Voltaire, 211–212 matter-mind question, 50–51
distribution of life arguments, Von Bingen, Hildegard, 187 meaningfulness of, 54
371–374 merits of, 61–62
fine-tuning argument, 360–365, 685 W norms and values of, 59–61
improbability of truth of theism, 587 Waghorn, Nicholas, 498, 499–500 noumenal vs. phenomenal accounts,
irreducible complexity, 369–370 Wainwright, William, 143, 451 49
Natural Theology (Paley), 359 Waldron, Jeremy, 655 open naturalism, 51–52
physical cosmology, 596–598 Wallace, Alfred, 407 private evidence, 58–59

INDEX
Worldviews, continued Worshipworthiness, 37, 109 Yandell, Keith, 139

purpose of, 52–53 Wright, Sewall, 367 YHWH, 4
self-evident truth or falsity of, Wrongness, 648 Young earthers, 546
56–57 Wykstra, Stephen, 545, 547, 549
suffering, 546–549 Z
theistic vs. naturalistic accounts, 690 X
thin vs. thick theism, 35 Zayd, 168
wishes and wants and the Xenophanes, 243 Zero-force evolutionary law, 389, 390
comportment of, 61 Zhuangzi, 565
Wormholes, 342 Y Zoroastrianism, 165, 401
Worship and tripartite doctrines, 23–24 Yablo, Stephen, 317–318 Zwingli, Huldrych, 612

Koterski, Joseph W. - Oppy, Graham - Theism and Atheism - Opposing Arguments in Philosophy-Macmillan Reference USA (2019)

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Koterski, Joseph W. - Oppy, Graham - Theism and Atheism - Opposing Arguments in Philosophy-Macmillan Reference USA (2019)

Uploaded by

Copyright:

Available Formats

Theism and Atheism

Opposing Arguments in Philosophy

Joseph W. Koterski, S.J.

Cover photograph: Christopher Smith / EyeEm / Getty Images.

LIBRARY OF CONGRESS CATALOGING-IN-PUBLICATION DATA

Names: Koterski, Joseph W., editor.

Gale, a Cengage Company

ISBN 978-0-02-866445-3 (this volume)

This title is also available as an e-book.

Contact your Gale sales representative for ordering information.

Definition: Theism ..............................................................................................................................1

Method: Theism ................................................................................................................................35

Logic: Theism ....................................................................................................................................65

TOPIC 4: DOXASTIC FOUNDATIONS

Doxastic Foundations: Theism .........................................................................................................103

TOPIC 5: RELIGIOUS EXPERIENCE

Religious Experience: Theism ...........................................................................................................135

viii THEISM AND ATHEISM: OPPOSING ARGUMENTS IN PHILOSOPHY

TOPIC 6: FAITH AND REVELATION

Faith and Revelation: Theism ..........................................................................................................163

Miracles: Theism .............................................................................................................................195

TOPIC 8: RELIGIOUS DIVERSITY

Religious Diversity: Theism .............................................................................................................227

THEISM AND ATHEISM: OPPOSING ARGUMENTS IN PHILOSOPHY ix

TOPIC 9: CAUSATION AND SUFFICIENT REASON

Causation and Sufficient Reason: Theism ........................................................................................259

TOPIC 10: A PRIORI

TOPIC 11: OUR UNIVERSE

Our Universe: Theism .....................................................................................................................331

TOPIC 12: HUMAN HISTORY

Human History: Theism ..................................................................................................................377

x THEISM AND ATHEISM: OPPOSING ARGUMENTS IN PHILOSOPHY

Human History: Atheism .................................................................................................................387

TOPIC 13: HUMAN BEINGS

Human Beings: Theism ...................................................................................................................407

TOPIC 14: ETHICS

Ethics: Theism .................................................................................................................................449

TOPIC 15: MEANING

THEISM AND ATHEISM: OPPOSING ARGUMENTS IN PHILOSOPHY xi

TOPIC 16: SUFFERING

Suffering: Theism ............................................................................................................................523

TOPIC 17: SCIENCE

Science: Theism ...............................................................................................................................561

TOPIC 18: THEORIES OF RELIGION

Theories of Religion: Theism ...........................................................................................................605

xii THEISM AND ATHEISM: OPPOSING ARGUMENTS IN PHILOSOPHY

TOPIC 19: PRUDENTIAL/PRAGMATIC ARGUMENTS

Prudential/Pragmatic Arguments: Theism ........................................................................................637

TOPIC 20: FINAL RECKONINGS

Final Reckonings: Theism ................................................................................................................665

THEISM AND ATHEISM: OPPOSING ARGUMENTS IN PHILOSOPHY xiii

xvi THEISM AND ATHEISM: OPPOSING ARGUMENTS IN PHILOSOPHY

xviii THEISM AND ATHEISM: OPPOSING ARGUMENTS IN PHILOSOPHY

fortunate enough to have them as my colleagues. I am also indebted to my colleagues in Australia,

xx THEISM AND ATHEISM: OPPOSING ARGUMENTS IN PHILOSOPHY

Arif Ahmed Evan Fales Joseph W. Koterski, S.J.

PRUDENTIAL / PRAGMATIC ARGUMENTS :

Taner Edis Cory Juhl Siobhan Nash-Marshall

David Neil Michael Ruse Robert J. Spitzer, S.J.

Robert Nola Bruce Russell

HUMAN HISTORY : THEISM

xxii THEISM AND ATHEISM: OPPOSING ARGUMENTS IN PHILOSOPHY

WHAT THEISM, WHICH GOD?