Buenting J. (Ed.) - The Problem of Hell. A Philosophical Anthology

THE PROBLEM OF HELL
For Mom and Dad

The Problem of Hell
A Philosophical Anthology
Edited by
JOEL BUENTING
University of Alberta, Canada
First published 2010 by Ashgate Publishing
Published 2016 by Routledge

2 Park Square, Milton Park, Abingdon, Oxon OX14 4RN
711 Third Avenue, New York, NY 10017, USA
Routledge is an imprint of the Taylor & Francis Group, an informa business
Copyright © Talor & Francis, Joel Buenting and the contributors 2010
All rights reserved. No part of this book may be reprinted or reproduced or utilised in any
form or by any electronic, mechanical, or other means, now known or hereafter invented,
including photocopying and recording, or in any information storage or retrieval system,
without permission in writing from the publishers.
Notice:
Product or corporate names may be trademarks or registered trademarks, and are used only
for identification and explanation without intent to infringe.
Joel Buenting has asserted his right under the Copyright, Designs and Patents Act, 1988, to
be identified as the editor of this work.
British Library Cataloguing in Publication Data

The problem of hell : a philosophical anthology.
1. Hell—Christianity. 2. Judgment of God.
I. Buenting, Joel.
236.2’5–dc22
Library of Congress Cataloging-in-Publication Data

Buenting, Joel, 1977–
The problem of hell : a philosophical anthology / Joel Buenting.
p. cm.
1. Hell. 2. Future punishment.
I. Title.
BL545.B84 2009
236’.25—dc22
2009030052
ISBN 9780754667636 (hbk)

Contents
List of Contributors vii
Introduction 1
1 Grace, Character Formation, and Predestination unto Glory 7

Thomas Talbott
2 Is it Possible to Freely Reject God Forever? 29

Raymond J. VanArragon
3 Annihilationism: A Philosophical Dead End? 45

Claire Brown and Jerry L. Walls
4 Compatibilism, “Wantons,” and the Natural Consequence

Model of Hell 65
Justin D. Barnard
5 Value, Finality, and Frustration: Problems for Escapism? 77

Andrei A. Buckareff and Allen Plug
6 Hell, Wrath, and the Grace of God 91

Stephen T. Davis
7 Molinism and Hell 103

Gordon Knight
8 Hell and Punishment 115

Stephen Kershnar
9 Why I Am Unconvinced by Arguments against the Existence

of Hell 133
James Cain
10 Hell and Natural Atheology 145

Keith E. Yandell
11 Infernal Voluntarism and “The Courtesy of Deep Heaven” 163

Bradley L. Sickler
vi The Problem of Hell
12 Birth as a Grave Misfortune: The Traditional Doctrine of Hell and

Christian Salvific Exclusivism 179
Kenneth Einar Himma
13 Species of Hell 199

John Kronen and Eric Reitan
Index 219
List of Contributors
Justin D. Barnard is Associate Professor of Philosophy and director of the Carl.

F. H. Henry Institute for Intellectual Discipleship at Union University in Jackson,
Tennessee. He holds an M.A. and Ph.D. in philosophy from the Florida State
University.
Claire Brown is a Ph.D. candidate and lecturer in philosophy at the University

of Notre Dame. She has research interests in ethics, bioethics, and philosophy of
religion. Her dissertation is an attempt to provide an account of supererogation
that a virtue ethicist can endorse.
Joel Buenting is lecturer of philosophy at the University of Alberta. He has

research interests in epistemology and philosophy of religion. He has published
on epistemology and is developing a monograph about contrastive knowledge.
Andrei A. Buckareff is Assistant Professor of Philosophy at Marist College,

New York state. He is co-editor with Jesús H. Aguilar of Philosophy of Action: 5
Questions (Copenhagen, 2009) and Causing Human Action: New Perspectives on
the Casual Theory of Action (Cambridge, forthcoming 2010). He has published
numerous articles on topics in epistemology, philosophy of mind and action, and
philosophy of religion.
James Cain is Associate Professor of Philosophy at Oklahoma State University.

He has written articles in various areas, including philosophy of religion, freedom
of will, ethics, and logic.
Stephen T. Davis is the Russell K. Pitzer Professor of Philosophy at Claremont

McKenna College, California. His degrees are from Whitworth University (B.A.),
Princeton Theological Seminary (M.Div.), and the Claremont Graduate University
(Ph.D.). He is the author or over 80 scholarly articles and 14 books, including
Encountering Evil (Louisville, 2001), Christian Philosophical Theology (New
York, 2006), and Disputed Issues (Waco, 2009).
Kenneth Einar Himma is an Associate Professor of Philosophy at Seattle Pacific

University and formerly taught in the philosophy department, information school,
and law school at the University of Washington. He specializes in philosophy of
law, information and computer ethics, and philosophy of religion. He has published
numerous scholarly articles in legal philosophy, information ethics, applied ethics,
and philosophy of religion.
viii The Problem of Hell
Stephen Kershnar is a Professor of Philosophy at the State University of New

York at Fredonia and an attorney. He focuses on applied ethics and political
philosophy. He has written three books: Desert, Retribution, and Torture (Lanham,
2001), Justice for the Past (New York, 2004), and Desert and Virtue: A Theory of
Intrinsic Value (Lanham, 2009).
Gordon Knight teaches philosophy at Iowa State University. His publications and
research interests center on metaphysics and the philosophy of religion.
John Kronen received his B.A. from Marquette University in 1985 and his Ph.D.
from State University of New York Buffalo in 1990. He has taught philosophy at
the University of St. Thomas in St. Paul, Minnesota, since 1990. He is the author
of several articles on metaphysics and the philosophy of religion, and is the co-
translator, with Jeremiah Reedy, of Suarez’s Metaphysical Disputation XV, On the
Formal Cause of Substance, published by Marquette University Press (2000).
Allen Plug is Assistant Professor of Philosophy at Malone University. His areas

of research specialization include epistemology, metaphysics, and philosophy of
religion. He has published articles in Religious Studies.
Eric Reitan is a Professor of Philosophy at Oklahoma State University. Dr.

Reitan has published extensively in the areas of ethics and philosophy of religion,
including a number of articles critically assessing the coherence of the Christian
doctrine of hell. His book Is God a Delusion? A Reply to Religion’s Cultured
Despisers was released in 2008 by Wiley-Blackwell.
Brad L. Sickler received his Ph.D. in philosophy from Purdue University

after earning a B.S. in physics from the University of Minnesota and an M.A.
in philosophy of religion from Trinity Evangelical Divinity School. He lives in
Minnesota with his wife and two children, and teaches philosophy of religion and
apologetics at Northwestern College.
Thomas Talbott is Emeritus Professor of Philosophy at Willamette University,

Salem, Oregon. Dr. Talbott has argued in several places, including “The Doctrine
of Everlasting Punishment” (Faith and Philosophy, 1990) and an entry on
universalism in The Oxford Handbook of Eschatology (2007), that the traditional
understanding of hell is inconsistent with the Christian concept of God.
Raymond J. VanArragon is Associate Professor of Philosophy at Bethel

University in St. Paul, Minnesota. He has published articles on philosophy of
religion and also has interests in epistemology and metaphysics. He is co-editor of
Contemporary Debates in Philosophy of Religion (Oxford, 2003).
List of Contributors ix
Jerry L. Walls is Professor of Philosophy at Asbury Theological Seminary. Dr.

Walls is the author of Hell: The Logic of Damnation (Notre Dame, 1992) and
Heaven: The Logic Of Eternal Joy (Oxford, 2002) as well as the editor of The
Oxford Handbook of Eschatology (2008). He is currently working on a book on
purgatory.
Keith E. Yandell is the Julius R. Weinberg Professor of Philosophy at the University

of Wisconsin-Madison and Adjunct Professor of Philosophy of Religion at Trinity
International University TEDS in Deerfield, Illinois. He has authored six books
and over 70 articles and book chapters.
This page has been left blank intentionally
Introduction
The Problem of Hell: A Philosophical Debate focuses on philosophical problems

raised by reflection upon the theological doctrine of everlasting punishment.
Contributions to this volume concern how a perfectly good God can send a person
to hell. Whether it is ever fair or just for God to send a person to hell is one version
of the problem of hell. The problem of hell thus poses a challenge for anyone who
believes (a) there is life after death and (b) some people suffer for eternity.
In framing the problem in terms of fairness and justice, I am following the
early work of Marilyn McCord Adams.1 Others frame the problem differently.
John Hick, for example, thinks hell is a problem because the existence of sinners in
hell is an affront to God’s sovereignty.2 Likewise, hell is a problem for Peter Geach
because the existence of sinners in hell permanently mires God’s creation.3
Despite these variations, many commentators identify the suffering of sinners
as the core problem. So, for instance, Jonathan Kvanvig thinks hell is a problem
because there is no greater purpose or point to suffering in hell. Since a person
in hell has no hope of leaving, hell is merely unending pointless misery.4 Others
characterize the relationship between hell and suffering by modeling the problem
of hell after the logical problem of evil.5 Discussions about the logical problem of
evil begin with claims about the goodness and power of God, proceed to a claim
about the existence of evil (“suffering”), then conclude with a question about their
compatibility. In discussions about hell, one replaces the premise “evil exists” with
the premise “there is at least one person suffering in hell for eternity.” Relying on
pre-theoretical intuitions about the badness of hell, one then generates a problem
by asking, “How is the infinite goodness and mercy of God compatible with a
sinner’s infinite punishment and suffering?” It would be fair to say, then, that the
“problem of hell” denotes a family of closely related topics that develop when one
takes seriously the doctrine of everlasting punishment.
Predictably, key disputes about hell concern how best to solve these problems.
Some argue that an acceptable doctrine of hell can be preserved despite the reasons
1
Marilyn McCord Adams, “Hell and the God of Justice,” Religious Studies, 11/4
(1975): 433–47.
2
John Hick, Evil and the Love of God (New York, 1966), pp. 377–8.
3
Peter Geach, Providence and Evil (Cambridge, 1977), p. 140.
4
Jonathan Kvanvig, The Problem of Hell (Oxford, 1993), p. 3.
5
See, for example, Andrei Buckareff and Allen Plug, “Escaping Hell: Divine
Motivation and the Problem of Hell,” Religious Studies, 41/1 (2005): 39–54; Thomas Talbott,
“No Hell,” in Michael L. Peterson and Raymond J. VanArragon (eds.), Contemporary
Debates in Philosophy of Religion (Oxford, 2004).
2 The Problem of Hell
we find it objectionable. Others argue for the possibility of post-mortem salvation,

whether the opportunity is accepted by hell’s denizens or not. Still others argue
that how we conceptualize the notion of personhood, freedom of the will, and
divine foreknowledge bears significantly on these issues. Apart from these topics,
other contributions to this volume mark out ways in which thinking about hell
creates other problems in mainstream Christianity.
To begin with, one response to the problem of hell is to argue that hell is empty.
This is the universalist response to the problem of hell. Universalists say everyone
is eventually reconciled to God in heaven. In the opening chapter of this volume,
Thomas Talbott discusses the relationships between universalism and grace. Talbott
argues that the notion of grace makes better sense of our choices in the formation
of a good character than the libertarian idea that people are responsible for the
formation of their own moral characters. This is because our free choices (both
good and bad) afford God the opportunity to demonstrate his true character and the
true nature of his love. So after criticizing Robert Kane’s idea of a “self-forming”
action, Talbott sketches out the ways it makes sense to think about God’s role in
response to the decisions we make.6 Talbott thus offers an account of how God
actively participates in the moral development of a person such that “a glorious
end is ultimately inescapable.”
Universalists do not deny a post-mortem existence in a state or place other
than heaven, but they do deny any person will be in hell forever. In Chapter 2, Ray
VanArragon argues that it is possible for a sinner to remain in hell for eternity.
Moreover, God might be justified in allowing this to happen. VanArragon develops
this argument by articulating what it means for a person to freely reject God
forever. On VanArragon’s view, freely rejecting God amounts to (i) continually
acting contrary to God’s will; and (ii) any person acting according to (i) is at least
minimally rational and aware of what she is choosing. VanArragon then defends
this account against an argument offered by Talbott according to which no one can
freely reject God forever.7
An increasingly popular reply to hell is to opt for the annihilation of the
damned. “Annihilationism” denotes a family of views according to which God
actively destroys (or simply allows) sinners to terminate their existence in hell.
According to one version of this view, physical death represents the cessation of
conscious life and the termination (or “annihilation”) of existence. According to
a second version, sinners experience hell but at some point can terminate their
existence and commit “metaphysical suicide.” In the third chapter, Claire Brown
and Jerry Walls critically discuss historical and contemporary motivations for
annihilationist alternatives to hell.
According to what Brown and Walls call the “natural consequence” motivation,
considerations about sin and its consequences entail or make probable the
annihilation of the damned. A second motivation is that it is preferable to have one’s
6
Robert Kane, The Significance of Free Will (London, 1996), p. 72.
7
Thomas Talbott, The Inescapable Love of God (Boca Raton, 1999).
Introduction 3
existence extinguished than to endure conscious eternal suffering. Given that God
is a particularly good being concerned with the welfare of the people he creates,
God annihilates sinners as a means of preventing their suffering in hell. A third
motivation for annihilationist views is that God’s purpose requires the annihilation
of the dammed over their continued existence in hell. This is because in the absence
of annihilation God’s purpose for humanity is eternally frustrated, and to that extent,
defeated. Annihilation is God’s means of achieving divine supremacy. Brown and
Walls maintain that each view is philosophically unmotivated. Consequently, a
heavy burden of proof remains on those who defend them.
A fourth response denies that hell is a problem on the general and thoroughgoing
grounds that people are not consigned to hell by God. People are in hell because
they have freely chosen it. Michael Murray calls this the “Natural Consequence”
model of damnation.8 Together with the “Penalty Model” (the view that punishment
amounts to an awareness of loss), the Natural Consequence model has a number of
virtues as a response to the problem of hell. Chief among these is that there is no
legitimate sense in which God is culpable for the suffering of the damned. People
in hell have damned themselves, so to speak. In Chapter 4, Justin Barnard argues
that this model suffers a small defect that leaves it open to an important objection.
Barnard’s worry is that this model retains its plausibility to the extent that sinners
are in hell of their own free will. But after introducing Harry Frankfurt’s influential
discussion of free will, Barnard makes the argument that sinners are in hell against
their will.9 If so, the Natural Consequence model loses a measure of its plausibility.
Barnard then argues that Frankfurt’s notion of a “wanton” provides the conceptual
resources necessary to address this worry, thus defending Murray’s proposal and
describing hell’s inhabitants.
Issuant concepts of hell can be understood as attempts to explain hell by appealing
to God’s loving nature. At first blush, attempting to explain hell by appealing to
God’s loving nature seems counterintuitive. The issuant rationale is that God’s love
for created persons motivates creating a place for people who genuinely do not want
communion with him. The most popular contemporary issuant view is represented
in Chapter 5, the position Andrei Buckareff and Allen Plug call “escapism”.
As Buckareff and Plug articulate it, escapism is the view that (i) hell exists and
might be populated for eternity and (ii) if there are people in hell, they can accept
God’s grace and leave. One can therefore “escape” hell when ready.
Thinking about escapism raises a question about the value of being in hell:
given that hell is not conceptualized in retributive terms, is hell a good place
for people? Thinking about escapism also raises a question about eschatology:
is escapism consistent with the view that heaven and hell represent “finalities”?
A third question raised by escapism concerns God’s plan for humanity. Since
8
Michael Murray “Heaven and Hell,” in Michael Murray (ed.), Reason for the Hope
Within (Grand Rapids, 1999), pp. 289–317.
9
Harry Frankfurt, “Freedom of the Will and the Concept of a Person,” The Journal
of Philosophy, 68/1 (1971): 5–20.
escaping hell implies that one can remain in hell, can someone remain in hell
forever, thereby frustrating God’s purpose for humanity? Buckareff and Plug
discuss these issues in turn, developing their account of escapism.
The topic of escaping hell is again brought up in Chapter 6. Here, Stephen Davis
discusses escapism with a view towards what type of support might be acquired
scripturally. Crucial to Davis’s position is that God’s wrath (his opposition to evil
instantiated in human activities) and God’s grace (his willingness to treat us better
than we deserve) are not opposing aspects of God’s nature. Davis argues that both
aspects are triggered by human disobedience. So after presenting the case that
grace allows for possibility of “post-mortem evangelism,” Davis distinguishes his
view from universalism, arguing that some will resolutely and continually reject
God forever.
As indicated, one way to think about the problem of hell is to think about the
joint incompatibility of a set of propositions. These propositions include “God is
unsurpassibly good and powerful” and “there is at least one person suffering in
hell.” Discussions about hell are then motivated by this apparent inconsistency. In
Chapter 7, Gordon Knight discusses the connections between hell and the ways it
makes sense to think about God’s foreknowledge. Of particular interest to Knight
is the Molinist tradition that God possesses “middle knowledge” (knowledge of the
counterfactuals of creaturely freedom). If God possesses this type of knowledge,
he does not merely know what you will do in the circumstances you are actually in;
God knows what you (or anyone else) will do in any possible circumstance. Since
God possesses this knowledge prior to creation, God can fine tune the universe
to his liking before creation. Given that God cannot control the free choice of the
creatures he creates, he can only create worlds with an optimal ratio of the blessed
in heaven to the damned in hell. So, after introducing Alvin Plantinga’s application
of middle knowledge to the problem of evil, Knight critically discusses William
Craig’s application of middle knowledge to the problem of hell.10 Knight argues
that in creating people God knows will be damned, God can never be completely
exonerated from the moral culpability of damnation.
Pre-theoretically at least, hell is closely associated with punishment. In Chapter
8, Stephen Kershnar argues that God would not (and perhaps cannot) send sinners
to hell. Kershnar’s argument rests on the premise that if persons go to hell, then God
sends them as punishment. But if so, the duration of punishment involved (infinite
punishment) must be justified. Kershnar then argues that no person warrants infinite
punishment because no person has an infinitely bad character or acts in infinitely
blameworthy ways. Since there is no reason for God to override the demands of
justice, a perfectly good and just God would not impose such a punishment.
The connections between hell and punishment are again taken up in Chapters 9
and 10. In Chapter 9, James Cain proposes a strategy for defending the traditional
10
William Lane Craig, “‘No Other Name’: A Middle Knowledge Perspective on the
Exclusivity of Salvation Through Christ,” Faith and Philosophy, 6/2 (1989): 172–87; Alvin
Plantinga, The Nature of Necessity (Oxford, 1974).
Introduction 5
doctrine of hell against the types of modifications surveyed so far. The strategy
consists of proposing a set of conditions that an acceptable account of hell ought to
satisfy. According to Cain, an acceptable doctrine of hell must meet a “grounding
condition”: it must be based on revelation (scripture, religious tradition, or a
revelatory experience). It must also meet the “consistency condition”: it must
not be inconsistent with other claims we have good reason to believe. Finally,
an acceptable account of hell must meet the “McTaggart condition”: it must not
undermine the trustworthiness of the being that is the source of the revelation
concerning hell. Cain then sketches out the ways in which these conditions might
be satisfied yet remain reasonably faithful to the traditional doctrine of hell,
including the notion that hell is fundamentally a place of retribution.
Keith Yandell’s contribution can also be seen as a discussion about punishment.
Yandell begins Chapter 10 by noting that one could use premises about hell for an
argument that God does not exist. Yandell calls this the “atheological argument”.
Part of the reason Yandell rejects this argument is that the existence of God requires
the existence of hell. This is because God’s justice requires that unrepentant sinners
are punished. Given that punishment does not occur pre-mortem, it must occur
post-mortem. Yandell then argues that how we conceptualize the value of persons
ought to inform how we think about punishment and how we ought to think about
universalist and annihilationist alternatives to hell.
Bradley Sickler begins Chapter 11 by articulating the problem of religious
diversity in connection to hell. Assuming Christian exclusivism, the doctrine of
hell entails the stark conclusion that all non-Christians are hell-bound, so to speak.
Sickler’s contribution argues that this conclusion is a non-sequitur. Sickler develops
his argument by introducing the theology of C.S. Lewis. Lewis’s position, Sickler
argues, provides the framework for viewing Christianity as the consummation of
other religions. The rationale is that God has been active among all people at all
times, leading them through different religious traditions to see important truths
fully revealed in Christianity through the incarnation. Coupled with the Lewisean
view that heaven and hell are ultimately places we choose, Christian exclusivism
entails that pious non-Christians shall have communion with God.
Kenneth Himma begins Chapter 12 by observing that many discussions about
hell are motivated by a standard objection. This objection is that nothing a person
could do in a finite period of time warrants infinite punishment. For reasons
that are obvious, this is sometimes called the “proportionality objection.” The
proportionality objection alleges that God treats sinners unjustly if he sends them
to hell. But thinking seriously about hell motivates other worries in mainstream
Christianity, and Himma’s contribution is to detail one worry in particular. Himma
argues that bringing a child into the world when there is a morally significant
chance the child will end up in hell conflicts with our moral intuitions and the
general Christian view that having children is either a good thing or a moral duty.
In the final Chapter of this volume, John Kronen and Eric Reitan offer a
taxonomy of “species” of hell—or different versions of the doctrine of hell—with
a view towards what reasons God might have to will or allow the suffering of the
damned. So, after discussing elements any version of hell must satisfy (i.e., the
cause of damnation and the types of evils endured by sinners in hell), Kronen
and Reitan distinguish six separable views of hell. Kronen and Reitan then argue
that the reasons God might will or permit damnation on any such view cannot be
reconciled with God’s moral character. Consequently, there is no non-problematic
version of the doctrine of hell.
Chapter 1
Grace, Character Formation, and
Predestination unto Glory
Thomas Talbott
Christians have traditionally held that, because they are saved by grace, they can
take no credit for their own salvation, or even for a virtuous character (where
such exists). All credit of this kind goes to God. As St. Paul himself put it in his
letter to the Ephesians: “For by grace you have been saved through faith, and this
[the faith] is not your own doing; it is the gift of God—not the result of works, so
that no one may boast.”1 Indeed, as I interpret him, Paul taught that God’s grace
is utterly irresistible in this sense: However free its recipients might be to resist
it in certain contexts, or even to resist it for a substantial period of time, they
are not free to resist it forever. For the end, at least, is foreordained. In Paul’s
own words, “For those God foreknew [that is, loved from the beginning] he also
predestined to be conformed to the likeness of his Son.”2 But if some end, such as
a person’s eventually being conformed to the likeness of God’s Son, is predestined
or foreordained, then that end cannot be avoided forever; and even if one should
insist, as some have, that such a predestined end rests upon God’s foreknowledge
of certain human choices (something that, so far as I can tell, Paul himself never
claimed3), this would be of no help to the large number of Christians who believe,
as I do not, that divine foreknowledge is itself incompatible with human freedom.
In Paul’s scheme of things, moreover, acquiring a good moral character just is
conforming to the likeness of God’s Son. So it looks as if a good moral character
is, according to Paul, wholly a work of God within and not something for which
the morally virtuous are entitled to credit themselves. And perhaps that is why
1
Ephesians 2:8–9.
2
Romans 8:29—New International Version (NIV), my emphasis.
3
Observe that, in the text just quoted, it is persons who are foreknown, not their
free choices. Elsewhere Paul used the same term when he wrote: “God has not rejected
his people [i.e., the people of Israel] whom he foreknew [i.e., whom he has loved from the
beginning]” (Romans 11:2). Here those foreknown are all the people of Israel, including
those disobedient ones who had rejected Christ, who had been blinded and hardened (11:7),
but who had not stumbled so as to fall (11:11) and whose full inclusion (11:7) would
eventually result in the salvation of all Israel (11:26). However disobedient they may have
been, “as regards election they are beloved, for the sake of their ancestors; for the gifts and
the calling of God are irrevocable” (11:28–9).
Paul consistently praised God, not the individuals themselves, for the faithfulness
of his Christian co-workers.
Now the first thing to observe about this Pauline doctrine of grace is how well
it accords with the actual attitudes of the morally virtuous themselves. Are not
the most virtuous among us typically the last to credit themselves for their own
moral virtues? A loving mother, for example, will not credit herself for the love
that controls her, however thankful she may be for the opportunity to care for (or
even to sacrifice on behalf of) her children; and a faithful husband, who would
never dream of a sexual indiscretion, will not credit himself merely because he
wants to maintain, without jeopardizing it, his valued relationship with his wife.
Such faithfulness, he may feel, is a product of clear vision, not profound moral
effort. And if pressed to explain how they came to be the kind of people they are,
those who consistently display the highest moral virtues may point to their own
parents who brought them up in a certain way, or to plain good fortune, or (if they
are religious) to the grace of God. Even where an intense moral struggle leads
to a more virtuous character in the end, as it sometimes does, the strengthened
character may not seem to be a product of one’s own moral effort to overcome
temptation. To the contrary, it may seem more like the product of a wholly new
perspective, such as we sometimes acquire only after experiencing first hand the
disastrous consequences of succumbing to temptation in the first place.
It is hardly false modesty, then, but instead clear moral vision that prevents
the truly virtuous from crediting themselves—that is, from crediting their own
free choices and moral efforts—for their own good character. For although the
religious expression “There but for the grace of God go I” seems to me quite
problematic if taken to imply that some other person is not an object of God’s
grace, it nonetheless remains a nice way of affirming that one’s own free choices
do not suffice to make one any better, or any more worthy of God’s grace, than
anyone else. It is even a way, perhaps, of saying something like the following:
“Had I been in Hitler’s shoes, facing his demons, my free choices may not have
been any better than his were; and had Hitler benefited from the advantages that I
have enjoyed, his free choices may not have been any worse than mine have been.”
I do not claim that I (or anyone else) could give a clear and coherent sense to such
a remark. But the point, once again, is merely to acknowledge that a good moral
character is something for which one should be thankful, not something for which
one should try to take credit. For a good character, like salvation itself, ultimately
“depends,” according to Paul, “not on human will or exertion, but on God who
shows mercy”4 and on the clear moral vision he will eventually impart to all.
Accordingly, in this chapter I shall challenge the idea, so widely accepted
among libertarians, that free agents “make themselves into the kinds of persons
they are”5 and that they are, for this very reason, morally responsible for their
own character. Then, after examining (and criticizing) the idea of a “self-forming
4
Romans 9:16.
5
See Robert Kane, The Significance of Free Will (London, 1996), p. 72.
Grace, Character Formation, and Predestination unto Glory 9
action,” as Robert Kane calls it, I shall argue that St. Paul’s pre-philosophical
understanding of God’s all-pervasive grace in fact makes far better sense of the
role that our free choices, the bad ones no less than the good ones, play in the
formation of a good character. It also helps to clarify how libertarian freedom,
indeterminism, and even sheer chance, if you will, could fit into a predestinarian
scheme in which a glorious end is ultimately inescapable.
Free Choice and Character Formation
Many libertarians now concede to the compatibilists, as I believe they should, that
an action can be free even when determined by an appropriately formed character,
and their intuition seems to be that an agent’s character is appropriately formed
only when the agent is at least partly responsible for it. James F. Sennett thus writes
as if we sometimes choose our own character: “A character that is libertarian freely
chosen is the only kind of character that can determine compatibilist free choices.”6
Laura Ekstrom likewise suggests that our judgment that an action is praiseworthy
“may presuppose the idea that the agent’s good character is ultimately of his own
making.”7 And Robert Kane explores the idea of a “self-forming action” in great
detail and with considerable insight.8
But just what might it mean, in the first place, to say that someone has made,
or formed, or produced his or her own character? Robert Kane speaks of certain
“voluntary ‘self-creating’ or ‘self-forming’ actions (including refrainings) in the
life histories of agents for which the agents are personally responsible.”9 These self-
forming actions (or SFAs), says Kane, are “both undetermined … and such that the
agents willingly performed them and ‘could have voluntarily (or willingly) done
otherwise’.”10 Although undetermined—and, as some might say, self-generated or
self-originated—they are also self-forming in the sense that they help to determine
or shape the agent’s present motives, purposes, and character traits: “Agents with
free will … must be such that they could have done otherwise on some occasions
of their life histories with respect to some character- or motive-forming acts by
which they make themselves into the kinds of persons they are.”11
Now given my own libertarian proclivities, I have no objection to the idea
that a good character is appropriately formed only when an agent’s life history
includes some undetermined choices that could have gone the other way. But as
soon as we try to puzzle out the precise relationship between these undetermined
6
James F. Sennett, “Is There Freedom in Heaven?” Faith and Philosophy, 16/1
(1999), p. 74.
7
Laura Ekstrom, Free Will: A Philosophical Study (Boulder, 2000), p. 165.
8
See Kane, The Significance of Free Will, Ch. 5.
9
Ibid., p. 75.
10
Ibid.
11
Ibid., p. 72.
choices in an agent’s life history and the agent’s present moral character, a host
of difficulties begin to emerge. The root idea to which Ekstrom and Kane both
appeal is that of a partial causal explanation, or a contributing cause—as when, for
example, Ekstrom suggests that, if a person S performs a determined action A at a
time t, then S is morally responsible for doing A at t only if S’s present character
and resulting inability to act otherwise “is causally explicable at least in part [my
emphasis] by S’s own act(s) at some time(s) other than t, such that S could have
done otherwise at that (those) other time(s).”12 Or, as Kane puts it in one place,
a self-forming action must actually make “a difference in what you are (or in the
character and motives you now have).”13
So now we must ask: Just what might count as a relevant difference in the
present context? Where “UA” is shorthand for “an undetermined action such that
the agent who performed it categorically could have done otherwise,” suppose that
a woman has only one UA in her life history, namely her decision as a youngster
to spend her allowance on swimming lessons rather than on violin lessons. If that
single UA partly explains why she later became an expert swimmer, indeed an
Olympic champion rather than a concert violinist, and if her swimming expertise
partly explains why she found it unthinkable and therefore psychologically
impossible to stand by as a child was drowning—why she leapt into a dangerous
river in an effort to save the child—then she evidently meets the Ekstrom necessary
condition of being morally responsible for a determined action. As we have just
described the case, moreover, this single UA buried in the woman’s past made a
huge difference to the kind of person she now is, that is, to her present character
and to the motives she now has. I doubt, however, that many libertarians would
see this difference, however significant it may be in the woman’s life history, as a
morally relevant difference. The decision to take swimming lessons presumably
had no great moral significance at the time it was made, and, beyond that, it was
not the woman’s intention as a young girl to make herself into a crack swimmer
or to prepare herself for saving the child later in life; she just enjoyed swimming
as a recreation. She nonetheless illustrates how easily one can meet the Ekstrom
necessary condition of moral responsibility and how little clarity it provides in the
present context.
Of course, Ekstrom never intended for anyone to treat her necessary condition
as if it were a sufficient condition. But even if we restrict our attention to UAs that
express morally significant choices, a serious problem remains. For if we should
examine carefully the life history of some virtuous person S, we would likely find,
I suspect, that S’s immoral choices had an even greater causal impact upon the
development of S’s virtuous character than S’s virtuous choices did. Suppose, by
way of illustration, that a young and somewhat irresponsible married man should
12
Ekstrom, Free Will: A Philosophical Study, p. 210. Ekstrom uses the expression “at
some time(s) other than t” rather than “at some time(s) prior to t” so as not to exclude by
assumption certain time travel cases involving backwards causation.
13
Kane, The Significance of Free Will, p. 72 (his italics).
succumb to temptation and should fall into a rather frivolous affair; suppose also
that his wife should subsequently find out about the affair and should seriously
consider divorcing him on account of this and other irresponsible actions on his
part; and suppose, finally, that the young man should then come to appreciate what
he is about to lose and, terrified by the prospect of losing the wife he genuinely
loves, should feel utterly compelled to re-establish a relationship of trust. So
once his wife finds out about the affair, it is fully determined, let us suppose, that
he will change his wayward ways; never again does he even consider an affair,
lest it undermine the very relationship that he now values so highly. If the man’s
decision to have an affair qualifies as a UA, then this UA may not only have made
a difference, but also a morally significant difference, to the kind of person he
eventually becomes. It also seems to qualify as a contributing cause. For had he
not made his foolish choice at this precise time and in circumstances where he
would eventually be caught, perhaps he would have gone through his entire adult
life sneaking around and taking his wife for granted.
So do we have here a case where the free decision to act in an unfaithful way
and to have an affair helped to shape a more trustworthy and faithful character?
And do we also have a case where a man acquires his clear vision and therefore his
faithfulness in an appropriate way? I think we do. We are not here talking about a
man being “zapped,” to borrow an expression from Michael Murray, and simply
being reconstituted with a more virtuous character; we are instead talking about
a man experiencing the consequences of his own free decision to act unfaithfully
and about how he learns an important lesson in the process. The man also acted
freely, or at least so I would argue. For not even a libertarian would deny that a
determined action can sometimes be voluntary; and if an action is both voluntary
and determined by one’s own fully rational judgment concerning the best thing
to do, then it remains a paradigm, so I have argued elsewhere,14 of free action. Its
being voluntary rules out what Kane calls “constraining control,” such as being
held at gunpoint, and its being determined by one’s own fully rational judgment
concerning the best thing to do rules out what Kane calls “nonconstraining control,”
such as might be “exemplified by … cases of behavioral conditioning and behind
the scenes manipulation.”15
Now some will no doubt find counter-intuitive the idea that our immoral choices
are sometimes more helpful than our morally proper choices are in producing
a virtuous character. For libertarians almost always seem to adopt, as a kind of
unexamined metaphysical assumption, a picture similar to what Kane sketches in
the following passage:
The probabilities for strong—or weak—willed behavior are often the results
of agents’ own past choices and actions, as Aristotle and other thinkers have
14
See Thomas Talbott, “God, Freedom, and Human Agency,” Faith and Philosophy,
26/4 (2009): 376–95.
15
See Kane, The Significance of Free Will, p. 64.
insisted. Agents can be responsible for building their moral characters over time
by their (moral or prudential) choices or actions, and the character building will
be reflected by changes in the probabilities for strong- or weak-willed behavior
in future situations. Each time the [alcoholic] engineer resists taking a drink
in difficult circumstances, he may strengthen his will to resist in the future;
and conversely, when he succumbs, his will to resist may lessen (or crumble
altogether, as sometimes happens with alcoholics).16
But even if such a picture reflects accurately some of our experience in some
contexts—very limited ones, I believe—the way in which UAs in a life history,
assuming there are such, affect one’s character and motives may be just the
opposite of what Kane has imagined; worse yet, the effect is apt to depend upon
intervening factors utterly outside the agent’s control.
Kane is right, of course, about the alcoholic engineer, at least partly. One
biochemical effect of alcohol on the brain, at least in the case of alcoholics, seems
to be that it undermines the will to resist another drink.17 But that is not even
close to the whole story. For as an alcoholic friend of mine once pointed out, the
longer she stayed off the alcohol, the easier it became during times of stress to
deceive herself into believing that this time a couple of drinks would do no harm;
so curiously, the longer she resisted the temptation, the stronger her temptation
became. Indeed, it was not until she had succumbed to temptation and had binged
terribly on a good many occasions that she finally learned to recognize such
deception for what it was. So in that sense, her experience was just the opposite
of what Kane describes: the more often she successfully resisted temptation, the
harder it became to resist such temptation in the future; and the more often she
succumbed to it and experienced the destructive consequences of doing so, the
easier it became to resist such temptation in the future.
So here is an obvious case where some bad choices helped to undermine a bad
(or at least a weak) character. Experience also provides examples where freely
resisting temptation, particularly in difficult situations, seems to weaken the will
over time rather than to strengthen it. I daresay that many men—and this would
include some Christian ministers I know—have sincerely (even fervently) resisted
sexual temptation for many years, only to succumb to it, finally, in middle age.
For it may happen that the harder a man tries, for the most earnest of reasons,
to suppress his childish yearnings and unrealistic fantasies, the more intense his
temptations become and the more likely he is to succumb to them in an explosion
of destructive behavior. Perhaps it would be misleading, however, to describe
this as a case where good choices help to undermine a good character. For if we
16
Ibid., p. 180.
17
Nor should one make the mistake of supposing that the effect of alcohol on the
alcoholic lasts only for the duration of the high. There are many long-term effects as well,
including an ever-increasing craving for the drug and the gradual destruction of the very
anti-anxiety centers of the brain upon which the alcohol works to provide temporary relief.
suppose that the described behavior really is destructive and really is the product
of childish yearnings and unrealistic fantasies, then it is also, perhaps, the product
of deeper character flaws of which the agent is unaware—character flaws that first
need to be exposed before they can be dealt with effectively.
Is my point, then, merely that the ultimate springs of human action are
mysterious and incredibly complex, so that only God could assess moral
responsibility with any degree of accuracy? Not at all. My point is that no one
has yet given a coherent account of what it might even mean to say that free
agents “make themselves into the kinds of persons they are”; at the very least,
we need something more than a requirement for a life history to include some
UAs. The relevant UAs must also qualify, in Kane’s own words, as “self-forming
actions” (SFAs), and this in turn requires that an agent be personally responsible
not only for the relevant UAs themselves, but also for the effect that the UAs
have, in conjunction with a complex variety of other circumstances, on the agent’s
character. But the problem, as Manuel Vargas has recently noted, is that “even
freely chosen features of our lives and ourselves can, because of our epistemic
limitations, yield unanticipated consequences.”18 One person may lie and cheat in
pursuit of wealth and fame, only to discover that the result is emptiness and misery;
and the circumstances surrounding this discovery may causally determine (even
compel) a life transformation. Another may sincerely cultivate moral integrity
and inadvertently produce some of the worst character traits: moral rigidity, self-
righteousness, and a lack of compassion. As Bernard Williams once observed,
“One’s history as an agent is a web in which anything that is the product of the will
is surrounded and held up and partly formed by things that are not” products of the
will.19 Indeed, the assumption that even God could consider how people exercise
their libertarian freedom and, on that basis alone, divide them into the good and
the bad, or into those who deserve a reward and those who deserve punishment,
now seems to me radically confused.
Grace Verses Works in Pauline Theology
It seems evident that St. Paul was acutely aware of the point just made, which he no
doubt believed to have been confirmed in his own experience. For whether or not he
actually wrote (in his own hand) the letter known as I Timothy, the self-description
attributed to him there—namely, that he had been “the foremost” or “the worst”
of sinners20—surely did reflect accurately the converted Paul’s understanding of
his former life. He clearly numbered himself, in other words, among those whose
sincere efforts at cultivating a more virtuous character had contributed to, or at
18
Manuel Vargas, “The Trouble with Tracing,” Midwest Studies in Philosophy, 29/1
(2005), p. 282.
19
Bernard Williams, Moral Luck (Cambridge, 1981), p. 29.
20
I Timothy 1:15 and 17.
least had revealed, even deeper character flaws. They had revealed, in particular,
the heart of a religious terrorist who was, prior to his conversion on the road to
Damascus, “a blasphemer, a persecutor, and a man of violence.”21 If his actions
were less destructive on the whole than were those of a Hitler or a Stalin, this
is only because he did not have twentieth-century technology or the power of a
modern state at his fingertips. So no wonder he opposed so adamantly any hint
of salvation by good works, which is essentially the idea that, as Laura Ekstrom
put it in the above quotation, “the agent’s good character is ultimately of his own
making.” Having discovered in his own life how easily moral seriousness and
genuine religious fervor can betray one into a pattern of destructive behavior and
even into acts of terror,22 he had no confidence in his own ability either to generate
moral virtue in himself or to pull himself up by his own bootstraps, so to speak.
No less important than the New Testament rejection of the libertarian idea
that an “agent’s good character is ultimately of his own making” is the implied
diagnosis of where Paul had gone wrong in the past. He went wrong, so we
read in the text, precisely because he had “acted ignorantly in unbelief,”23 and
this underscores the essential role that ignorance plays in even the worst of sins.
Although Christians sometimes seem suspicious of the Socratic idea that the
essence of virtue is a certain kind of knowledge, insight, and clarity of vision,
we find ample support for such an idea in the Bible itself. Did not Jesus himself
declare from the Cross: “Father, forgive them; for they do not know what they
are doing”?24 And we find Peter expressing a similar attitude when he charged an
audience with killing “the Author of life.” “I know that you acted in ignorance, as
did also your rulers.”25 The clear implication here is that those who crucified the
Lord had no idea that they were acting wrongly and may even have presumed that
they were doing the right thing; in that respect, they were no different from those
who drowned Anabaptists in Zurich, or those who burned Servetus at the stake in
Geneva, or those who burned young women as witches in Salem, Massachusetts.
I do not mean to minimize the evil implicit in such acts of terror; far from it. But
those who commit such acts of terror often count themselves among the righteous
doing battle against evil, and they are, more often than not, utterly oblivious of
their own self-righteous motives and attitudes. More generally, even our everyday
sins and indiscretions may express deeply rooted fears, jealousies, animosities,
and feelings of personal inadequacy, of which again we may be less than fully
21
I Timothy 1:13a.
22
In general, we Westerners find it relatively easy to appreciate this point when
considering certain aspects of the Muslim culture, but we find it much more difficult, it
seems, to appreciate the same point with regard to the history of the Christian Church.
23
I Timothy 1:13b.
24
Luke 13:34. Even though this well-known prayer from the Cross does not appear in
some of the best manuscripts, it does, I presume, represent a reliable tradition.
25
Acts 3:17.
aware. According to Robert Adams,26 therefore, some of our most important and
most pervasive sins are involuntary, because “voluntary consent, as ordinarily
understood, implies knowledge.”27 With respect to the sin of ingratitude, for
example, Adams concludes that “the search for voluntary actions and omissions
by which you may have caused your ingratitude keeps leading to other involuntary
sins [or moral weaknesses] that lie behind your voluntary behavior.”28
But that is only half the story. For quite apart from involuntary sins of which
we may be unaware, Paul held that we are powerless to prevent ourselves from
sinning even in cases where we are able to discern right from wrong. Our earliest
moral experience, he contended, arises from an emerging ability to understand the
moral law (or moral rules, if you will), and it is from the beginning an experience
of the will in bondage to sin. In Paul’s own words, “If it had not been for the law, I
would not have known sin … But sin, seizing an opportunity in the commandment
… deceived me and through it killed me.”29 He even went on to write: “I do not
understand my own actions. For I do not do what I want, but I do the very thing
I hate … Now if I do what I do not want, it is no longer I that do it, but sin that
dwells within me.”30 All of which seems to accord better with the idea that sin
is something that happens to us early in life rather than something we do freely
from the beginning of our lives. Like death itself, sin here seems more like an
enemy from which we need to be rescued than a perfectly free choice for which
we deserve some sort of retributive punishment. It is essentially anything in us that
alienates us from others and from God, that is, anything in us that undermines our
capacity to love perfectly; as such, it is also, according to the Christian faith, the
principal source of human misery. Paul thus exclaimed, “Wretched man that I am!
Who will rescue me from this body of death [or sin]?”31
Accordingly, in Pauline theology, so I would argue, salvation from sin is not
an escape from deserved punishment; nor is it, as some Christians have made it
out to be, the removal of an inherited moral taint. It is instead more like being
rescued from a kind of slavery or bondage that we are powerless to escape on our
own—sort of like being rescued from alcoholism or a drug addiction. For even
as an alcoholic might judge it best to refuse another drink and nonetheless find it
psychologically impossible to do so, Paul declared himself to be “captive to the
law of sin that dwells in my members.” Indeed, for all of his talk about the wrath
of God (in the early part of Romans, for example32), Paul did not seem to regard
26
See Robert Adams, “Involuntary Sins,” Philosophical Review, 94/1 (1985): 3–31.
27
Ibid., p. 13.
28
Ibid.
29
Romans 7:7b–8 and 11.
30
Romans 7:15 and 20.
31
Romans 7:24.
32
And, of course, as is made abundantly clear in Romans 11 and elsewhere, Paul
understood God’s wrath, his severity towards sin, and even his hardening of a heart as itself
an expression of his mercy (or compassion) toward sinners. On this point, see Thomas
sin as essentially a matter of personal guilt at all;33 instead, he held that we are
already sinners, already “dead” in our “trespasses and sins,”34 even before our
moral consciousness fully emerges, before we become rational enough to qualify
as free moral agents, and before we are fully aware of our own selfish motives and
destructive desires. This is not to say, of course, that the concept of personal guilt
had no role at all to play in Paul’s thinking. But so far as I can tell, not one word
in his letters implies that we somehow deserve retributive punishment either for
our inherited character weaknesses (and imperfections) or for the initial bondage
of our will to sin; neither does anything there so much as hint that, in the words
of Jonathan Edwards, we “are ten thousand times more abominable in his [God’s]
eyes than the most hateful venomous serpent is in ours.”35 It seems to me, at any
rate, that Paul’s own writings were remarkably free from such neurotic appeals to
fear and guilt.
Paul nonetheless offered, I believe, a profound insight into the nature of moral
corruption and into the way in which a bad moral character differs from a good
one. Like alcoholism and drug addiction, a bad moral character will inevitably
enslave a person in one of two ways: Either it will undermine over time one’s
power to follow one’s own judgment concerning the best course of action, or it
will eventually undermine altogether one’s ability to learn from experience and to
make rational judgments concerning the best course of action. I contend that this
is just what makes a bad moral character objectively bad: it will tend to undermine
over time one’s rational control over one’s own actions. But a good moral character
Talbott, “A Pauline Interpretation of Divine Judgment,” in Robin A. Parry and Christopher H.

Partridge (eds.), Universal Salvation? The Current Debate (Grand Rapids, 2003), pp. 32–4.
33
As I have written elsewhere: “Both Jesus and Paul consistently rejected as
inappropriate the very reactive attitudes upon which so many rest their understanding of
moral guilt. Personally, I doubt that the ideas of intrinsic desert and ‘metaphysical guilt’
played a substantial role, if any at all, in their thinking. Yes, Paul explicitly stated in Romans
1:32 that those who commit certain sins are ‘worthy of death,’ and this may initially appear
to imply that death is the intrinsically fitting punishment for sin. But the appearance is in fact
misleading. For within the context of Pauline theology as a whole, the relationship between
sin and death is clearly non-contingent. First, the relevant death, which Paul elsewhere
described as ‘the wages’ (or the price) of sin (Rom. 6:23) and also as ‘the end’ of sin (Rom.
6:21), is spiritual death; it is separation from God and from the ultimate source of human
happiness. Nor could it have been otherwise, because in sinning one precisely chooses death
over life, separation over reconciliation. In Paul’s own words, ‘To set the mind on the flesh
is death, but to set the mind on the Spirit is life and peace’ (Romans 8:6). So death, which
is the unavoidable consequence of sin, is its intrinsically fitting punishment only in the
sense that a painful burn is the intrinsically fitting punishment for intentionally thrusting
one’s hand into a fire” (Thomas Talbott, “Why Christians Should Not Be Determinists:
Reflections on the Origin of Human Sin”, Faith and Philosophy, 25/3 (2008), pp. 305–6).
34
Ephesians 2:1.
35
See Jonathan Edwards, “Sinners in the Hands of an Angry God,” reprinted in Ola
Elizabeth Winslow, Jonathan Edwards: Basic Writings (New York, 1966), p. 159.
is just the opposite. For whereas a bad character leaves the will in bondage, a good
character does not; to the contrary, a good character will expand one’s rational
control over one’s actions and will therefore liberate the will. That, at least, seems
to have been Paul’s view of the matter, and he therefore spoke of salvation as if it
were a release from bondage, a means by which our very wills are set “free from
the law of sin and death.”36
In any event, we have now identified two reasons why, according to Paul, we
are powerless to prevent ourselves from sinning and from falling into error and why
we are also powerless to save ourselves once we have fallen into error. First, our
most sincere efforts at cultivating moral virtue may inadvertently produce some of
the worst character traits, as Paul clearly believed was true of himself, and such
efforts will inevitably reveal, in any case, even deeper character weaknesses (or
imperfections) in ourselves. For the fact is that we come into this earthly life with
many flaws, imperfections, and moral weaknesses of which we may be unaware.
And second, the context in which our moral consciousness emerges during our
early childhood is one in which our wills are already in bondage to sin. So even
when we know what is right and what is wrong, we too often find ourselves unable
to avoid “missing the mark” and doing what we know to be wrong. This does not
mean, however, that we never act freely and are never morally responsible for any
of our actions while the will is still in bondage to sin. An alcoholic, whose will
is in bondage to alcohol, may nonetheless make many free choices, such as the
decision to seek treatment after a destructive binge, and may be responsible for
these choices without being morally responsible for the genetic predisposition to
alcoholism. And similarly, those whose wills, according to Christian theology, are
in bondage to sin may nonetheless make many free choices (and many good choices
born of love for their children, for example) without being morally responsible for
having generated their own imperfections and moral weaknesses in the first place.
It is just that, without outside help, these imperfections and moral weaknesses will
continue to have destructive consequences in their lives.
“Felix Culpa” or Creation in Two Stages
But why, one may wonder at this point, would God start us out with so many
imperfections and moral weaknesses and in a context in which our wills are already
in bondage to sin? Why bring us into being as sinners and then go to the trouble of
saving us from our sin? Why not simply bypass all the misery and suffering along
the way and bring us into being as perfected saints in the first place?
The assumption behind such questions is that, if he so desired, God could
have created each of us (or perhaps a different set of persons) instantaneously as
self-aware, language using, fully rational, and morally mature individuals who
are from the beginning perfectly fit for intimacy with God. But why suppose that
36
Romans 8:2.
to be metaphysically possible at all? For my own part, I seriously doubt that God
could have created any persons at all without satisfying certain metaphysically
necessary conditions of their coming into being, and the most important of these
would be “an initial separation from God,” which I have elsewhere described in
the following way:
By this admittedly vague expression, I mean to imply, among other things, a

severance from God’s direct causal control on the metaphysical level and an
experience of frustrated desire and frustrated will—the sort of thing that naturally
leads to a sense of estrangement and alienation—on the psychological level. If
these should be metaphysically necessary conditions of our creation, then our
very creation would virtually guarantee the occurrence of error and misguided
choices.37
If God had no choice, provided he wanted to create any persons at all, “but to
permit their embryonic minds to emerge and to begin functioning on their own
in a context of ambiguity, ignorance, and indeterminism,”38 then the creation of
a person is, of necessity, a much more complicated and time-consuming process,
even for an omnipotent being, than one might have imagined. And if the required
context is one that virtually guarantees erroneous judgments and misguided choices
(perhaps even an initial bondage of the will to sin, as Paul understood it), then God
faces the following dilemma in creation: Some of the very conditions essential to
our emergence as rational individuals distinct from God are themselves obstacles
to perfect fellowship (or union) with him, and these cannot be overcome until after
we have already emerged as a center of consciousness distinct from God’s own
consciousness.
Of course, I might be mistaken in my conception of what is, and is not,
metaphysically possible in the matter of God’s creating persons distinct from
himself. But even if I am mistaken, the process by which we humans in fact emerge
in this earthly life and develop into rational agents is indeed both complicated and
time-consuming. So if one supposes that God exists at all, then one must also
suppose, at the very least, that God had good reasons to permit our embryonic
minds to emerge in a context of ambiguity and ignorance. And Paul clearly
embraced that idea in any case. For he clearly taught that God employs a two-
stage process, or two Adams as he calls them, in creating Sons and Daughters. As
I have put it elsewhere:
The first Adam, according to Paul, “was from the earth, a man of dust” and
“became a living being”; the second was not from the earth, but “from heaven” and
“became a life-giving spirit” (I Cor. 15:45 & 47). The first Adam thus represents
37
Talbott, “Why Christians Should Not Be Determinists,” p. 307; but see the entire
section, pp. 306–10.
38
Ibid.
the first stage in the creation of God’s children: the emergence of individual
human consciousness in a context of ambiguity, illusion, sin, and death; the
second Adam, or Jesus Christ, represents the second stage: the divine power that
successfully overcomes all sin and death and therefore all separation from God,
so that the true Sons and Daughters, or the true creations of God, can emerge.39
Paul also wrote: “it is not the spiritual that is first, but the physical [i.e., that which
pertains to our animalistic and sensuous nature], and then the spiritual.”40 And
though he nowhere used the language of necessary and sufficient conditions, he
seems clearly to have held that the first stage of creation—namely, our emergence
from the dust of the earth in a context of ambiguity, illusion, sin, and death—is
a necessary condition of the second, wherein God reconciles us to himself and
perfects as saints.
So interpreted, Paul’s vision of creation also carries an important implication
for Alvin Plantinga’s recently formulated Felix Culpa theodicy.41 According
to Plantinga, human sinfulness is a “fortunate fault” in the sense that it makes
possible the great goods of redemption and atonement; so, because God wanted
to actualize a world that includes these great goods, he chose to actualize one that
includes human sin, indeed lots of it. For sin is obviously a necessary condition
of redemption from sin or of an atonement for it. Plantinga also anticipated the
objection, which others have subsequently raised,42 that such a theodicy makes
God seem “too much like a father who throws his children into the river so that he
can then heroically rescue them, or a doctor who first spreads a horrifying disease
so that he can then display enormous virtue in fighting it in heroic disregard of his
own safety and fatigue.”43 But if our sinful condition, or even an initial bondage
of the will to sin, is an unavoidable consequence of conditions essential to our
creation, then our Creator need be nothing like the father or the doctor in the
above examples. He is not, first of all, the direct cause of our sin; hence, he is
nothing like a father who throws his children into a river or a doctor who spreads a
horrifying disease. And if, as I have suggested, conditions that virtually guarantee
sin, error, and spiritual death are essential to the emergence of distinct persons,
then it seems overwhelmingly probable that any worthwhile world within God’s
power to actualize will include these great enemies as well as a rescue of God’s
loved ones from them.
39
Thomas Talbott, “Christ Victorious,” in Robin A. Parry and Christopher H.
Partridge, Universal Salvation? The Current Debate (Grand Rapids, 2003), p. 18.
40
I Corinthians 15:46.
41
See Alvin Plantinga, “Supralapsarianism, or ‘O Felix Culpa,’” in Peter van Inwagen
(ed.), Christian Faith and the Problem of Evil (Grand Rapids, 2004), pp. 1–25.
42
See, for example, William Hasker, The Triumph of God over Evil (Downers Grove,
2008), pp. 167–70 and Marilyn McCord Adams, “Plantinga on ‘Felix Culpa’: Analysis and
Critique,” Faith and Philosophy, 25/3 (2008): 123–40.
43
Plantinga, “Supralapsarianism,” pp. 21–2.
Beyond that, Paul also insisted upon the glorious truth that all of those who
participate in the first stage of creation will likewise participate in the second and
will thus experience in the end the “towering goods” of redemption and atonement,
as Plantinga calls them. Nor do I see how Paul might have expressed himself any
more plainly than this: “For God has imprisoned all [humans] in disobedience so
that he may be merciful to [them] all.”44
Divine Grace: Its Universal Scope and Unconditional Character
All Christians believe in divine grace. Within the Western theological tradition,
however, one encounters two “respectably orthodox” traditions that interpret
God’s saving grace in two very different ways. According to a long tradition that
stretches back through the Protestant Reformers and ultimately has its roots in the
thought of St. Augustine—call it the Augustinian tradition—God’s saving grace,
though utterly unconditional and irresistible, is nonetheless limited in scope. For
although God bestows his grace or special favor on some sinners, he does not
bestow it equally upon all of them; in that respect, he is very much, contrary to
repeated statements in the New Testament, a “respecter of persons.” But according
to a competing theological tradition, sometimes called the Arminian tradition,45
God’s saving grace, though universal in scope, is nonetheless limited in its power
and efficacy. For although God at least offers saving grace to all sinners, some will
irrationally continue to reject it throughout all of eternity and thereby prevent God
from ever achieving a complete victory over sin and death.
But neither the Augustinians nor the Arminians, I shall now argue, have
properly understood the Pauline doctrine of grace, and it is ironic, perhaps, that
both parties are quite correct in their criticisms of each other. The Augustinians are
certainly right about this: Our being the object of God’s grace in no way depends,
according to Paul, upon anything we have (or have not) done, freely or otherwise;
nor is it something we could ever purchase or earn by keeping the moral law or by
doing good works. But despite Paul’s explicit statement (quoted above) that God
is merciful to all, the Augustinians draw the further inference that God’s grace is
utterly gratuitous and supererogatory rather than an essential expression of his
own justice or righteousness. The assumption here is that, even as our Creator,
God has no intrinsic responsibility for our moral and spiritual welfare. Because
our first parents somehow polluted the entire human race, God owes us nothing
further; in particular, nothing in his nature—neither his justice, nor his love, nor
his mercy—constrains him to extend his grace to a single sinner. As Augustine
44
Romans 11:32. See also I Corinthians 15:22: “for as all die in Adam, so all will be
made alive in Christ.” It is clear that Paul’s vision of creation in two stages lies at the very
heart of his theology.
45
A tradition named after Jacobus Arminius (1560–1609) for his opposition to the
Calvinistic doctrine of predestination and limited election.
himself put it, “the whole human race was condemned in its apostate head by a
divine judgment so just that even if not a single member of the race were ever
saved from it, no one could rail against God’s justice.”46
But even if one should accept the dubious supposition here that, as Adam’s
descendents, we have all inherited his guilt, why make the further assumption that
our inherited guilt relieves God of all responsibility for our moral and spiritual
welfare? As James B. Gould points out in a recent paper,47 this further assumption
is simply confused. You might as well argue that a child’s disobedience relieves its
parents of all responsibility for the child’s future welfare as well—which is absurd.
Just as the decision to have a child creates an obligation to promote the child’s
welfare, however disobedient the child might happen to become, so God’s decision
to create us entails a freely accepted obligation to promote our welfare, however
disobedient we might have become. In fact, this is precisely why, according to
Paul, we can do nothing to earn God’s grace (or favor): It is already and always
present, whether we know it or not, from the very beginning of our earthly lives.
We can hardly earn something through good works that will always be present—
albeit in different forms, perhaps—regardless of what we do.
By way of a reply to this, the Augustinians sometimes argue that it is the very
nature of mercy that it must be supererogatory. Insisting that mercy is simply
undeserved love (as if it were possible for someone created in God’s image to be
undeserving of God’s love), Paul Helm thus writes:
What is essential to such love is it could, consistently with all else that God is,
be withheld by him. If God cannot but exercise mercy as he cannot but exercise
justice then its character as mercy vanishes. If God has to exercise mercy as he
has to exercise justice then such “mercy” would not be mercy [i.e., would not
be undeserved love] … A justice that could be unilaterally waved would not be
justice, and a mercy which could not be unilaterally waved would not be mercy.48
But suppose now that we replace the word “mercy” in this quotation with any one
of the following: “beneficence,” “kindness,” “compassion,” or even “pity.” Helm
would not, I presume, argue as follows: “If, given his essential attributes, God
cannot but exercise beneficence [kindness, compassion, or pity] as he cannot but
exercise justice, then its character as beneficence vanishes.” Why is this important?
46
Enchiridion, section 99. For the sake of accuracy, I have altered the position of
“not” in Albert C. Outler’s translation, which reads: “not even if a single member of the
race were ever saved from it, no one could rail against God’s justice.” See Augustine:
Confessions and Enchiridion, Library of Christian Classics, vol. VII (Philadelphia, 1955),
pp. 398–9.
47
James B. Gould, “The Grace We Are Owed: Human Rights and Divine Duties,”
Faith and Philosophy, 25/3 (2008): 261–75.
48
Paul Helm, “The Logic of Limited Atonement”, Scottish Bulletin of Evangelical
Theology, 3 (1985), p. 50.
Because the central Pauline concept, sometimes translated in our English Bibles
with the word “mercy,” is not that of undeserved love at all. It is instead that of
beneficence, kindness, compassion, or pity. It has in view not the setting aside
of a just punishment, as Helm supposes, but the relief of misery or distress. One
could therefore accurately translate a text such as Romans 11:32, which I quoted
above, as follows: “For God has imprisoned all in disobedience so that he may
be beneficent to them all.”49 Accordingly, Helm’s point about “mercy,” however
appropriate it may be in his own context, has no relevance, so far as I can tell,
either to Paul’s claim that God is beneficent to all or to my own philosophical
assumption that this beneficence flows from the inner necessity of God’s own
righteousness. Why suppose that a conception of divine mercy, according to which
God might not ever be merciful to a single created person, is even relevant to
Paul’s own understanding of salvation?
In fact, although a doctrine of grace may appear to lie at the heart of Augustinian
theology, the appearance is quite illusory. For once you try to combine a doctrine
of free and irresistible grace with the Augustinian understanding of limited
election—the pernicious idea that God, being limited in compassion, restricts his
mercy to a limited elect—the very idea of grace evaporates altogether. Is God
being gracious to an elect mother when he makes the baby she loves with all her
heart an object of his “sovereign hatred”50 and supposedly does so, as in the case
of Esau, even before the child has done anything good or bad? What really lies at
the heart of Augustinian theology, I believe, is a logical impossibility: the idea that
God could extend his love and compassion to one person even as he withholds it
from some of that person’s loved ones. It is hardly surprising, therefore, that most
Christian philosophers writing today—Arminians, Catholics, and other freewill
theists—rightly reject any hint of limited election and understandably appeal to
libertarian freewill in an effort to explain why God’s perfecting love, which he
extends equally to all, successfully transforms some sinners but not others.
It seems to me, however, that Arminian theology ultimately places a burden
upon so-called libertarian freedom that it cannot coherently bear. If we all start out
in a context of ambiguity, ignorance, and illusion, then it stands to reason that our
salvation from this condition (and that our eventual perfection) would require, as
the Christian faith implies, belief of a certain kind, faith, or (as I like to think of it)
clarity of vision. And according to Paul in particular, these are gifts from God, the
product of his providential control of our lives, rather than cognitive states that we
somehow manufacture in ourselves simply by deciding to do so. But despite Paul’s
49
Helm admits that, even on his own view, God could be merciful to all. So one
wonders why he cannot take seriously Paul’s explicit statement that God is merciful (or
beneficent) to all?
50
According to G.C. Berkouwer, the Dutch Reformed theologian Herman Hoeksema
described God’s attitude towards the non-elect as the “sovereign hatred of his good
pleasure.” For the quotation from Het Evangelie, see Berkouwer, Divine Election (Grand
Rapids, 1960), p. 224.
clear teaching on this point, Arminian theologians typically speak of our deciding
to believe something, as if our religious beliefs were properly under the control of
our wills. In rightly opposing the Reformed understanding of limited election, for
example, the Arminian theologian Jack Cottrell insists that “every sinner is able
to make his own decision of whether to believe or not.”51 So just how are we to
understand such frequently encountered religious language?
It is utterly non-controversial, I presume, that a very simple empirical belief,
such as the belief that fire can burn and cause terrible pain, is not properly a matter
of the will at all. Someone might choose to walk near a fire, or to place a hand on
a hot coal, or to experiment with fire in some other way, and relevantly similar
choices might play an important role in someone’s discovering the true nature
of fire. But once the consequences of such choices are experienced, the resulting
belief that fire can burn and cause terrible pain is not itself the product of some
further choice, much less of some libertarian free choice. For discovering the truth
about something is very different from manufacturing a belief in oneself by an act
of will—which is not even psychologically possible in many cases.
Certainly religious beliefs are typically more complex than simple empirical
beliefs, and some of them could, perhaps, involve the will in a host of subtle ways.
As religious people typically understand it, moreover, belief in God goes far beyond
a mere intellectual assent to the proposition that God exists; it also includes such
attitudes as love, trust, and gratitude. So are these properly any more the product of
choice or will than simple empirical beliefs are? I doubt it. I learned at a very early
age to trust my mother implicitly—not because I decided to trust her, but because I
discovered her to be altogether trustworthy. I also learned to love her—not because
I decided to love her, but because she first loved me and demonstrated her love in
thousands of ways. I have no doubt that certain free choices, if you will, were an
important part of the process whereby I discovered my mother’s true character.
For I was just as disobedient and snotty at times as any other child and just as
rebellious during my teen years as many others are. But the free choices I made,
both the good ones and the bad ones, merely provided my parents with additional
opportunities to demonstrate their true character, and at no time in my life could
I have freely chosen, so I believe, not to love them and at no time could I have
freely chosen to separate myself from them altogether. There was simply never any
motive to spurn the love of someone who always put my own interests first. And
similarly for God, our supremely perfect Mother and Father: We learn to love him
because he first loved us and will continue to demonstrate throughout all eternity,
if necessary, his faithfulness in meeting our true spiritual needs and in satisfying
our heart’s desire in the end. Accordingly, our free choices, whichever way they go,
merely provide God with additional opportunities to demonstrate his true character
and the true nature of his love for us, even as he continues to shatter our illusions
and to correct our erroneous ways of thinking.
51
Jack Cottrell, “The Classical Arminian View of Election”, in Chad Owen Brand
(ed.), Perspectives on Election (Nashville, 2006), p. 121.
So perhaps the sum of the matter is this: In view of his explicit statement
that God is merciful to all, Paul would have rejected, it seems, the Augustinian
understanding of limited election; and in view of his repeated statement that faith
itself is a gift from God, Paul would also have rejected, it seems, the Arminian
understanding of conditional election. For in Pauline theology, at least, God’s
saving grace is both universal in scope and unconditional in nature.
Predestination unto Glory
So far, I have challenged the assumption, widely shared by libertarians, that an

“agent’s good character is ultimately of his own making.” I have also noted that
the most virtuous among us are typically the last to credit themselves and the first
to credit good fortune—or perhaps the grace of God, if they are religious—for
their own moral virtues; they are wise enough, in other words, not to attribute
their moral virtues, whatever these might be, to the virtuous character of certain
free choices buried in their causal history. For as St. Paul would be the first to
acknowledge, the difference between a Hitler or a Mussolini, on the one hand, and
himself, on the other, does not lie in the more virtuous character of his own free
choices. But having said that, I also hold that free choice, indeterminism, and even
sheer chance have an important role to play both in the emergence of independent
rational agents and in the process whereby they are finally reconciled to God. So
how do I propose to put all of this together? Four observations will have to suffice
for the present.
First, a necessary condition of both moral freedom and saving faith is, I
presume, a minimal degree of rationality, including an ability to discern reasons
for acting, an ability to learn important lessons from experience and from the
consequences of one’s actions, and a capacity for moral improvement. Not even
God, after all, could reveal himself to a stone, and neither could he both leave a
newborn infant in a state of undeveloped rationality and, at the same time, reveal
himself to the infant. So in the case of those who fall below the relevant threshold of
rationality—small children, the mentally challenged, the severely brain damaged,
paranoid schizophrenics, the criminally insane, and the like—the question of how
God might honor their free choices or utilize such choices as a means of saving
them does not even arise. Neither does the concept of saving faith have a relevant
application to them. In no way does this imply, of course, that such individuals are
not objects of God’s grace. It is just that God must first permit the newborn infant
to develop into a minimally rational agent, either in this life or the next, and must
also restore the paranoid schizophrenic to some semblance of rationality before
the concept of saving faith can have any relevant application to them.
Second, with respect to those who emerge as independent rational agents in
a context of ambiguity and ignorance, God can surely correct them and even
foreordain a destiny for them without directly controlling their individual choices.
God had no need, for example, to control individual human choices, not even
someone’s decision to experiment with fire, in order to guarantee that the human
race would eventually discover the power of fire to burn and to cause pain; he
needed only to allow minimally rational people to emerge in an environment in
which they would encounter fire with some degree of frequency. God can also
employ the consequences of our free choices as a means of revelation, that is,
as a means of shattering our illusions and of correcting the false assumptions
that underlie our bad choices in particular. If I freely act on the illusion that I
have the skill to ski down a treacherous slope, a fall and a broken leg may, quite
unexpectedly, shatter that illusion to pieces; and if, because I have misconstrued
the conditions of my own happiness, I repeatedly pursue my perceived interests
at the expense of others, I may eventually discover, again quite unexpectedly,
the error of my ways. Indeed, because their consequences can be so effective in
correcting our misguided judgments, our immoral and destructive choices may
sometimes, as we have already seen, be more useful to God in transforming us
than a more virtuous choice might have been. So, just as a grandmaster in chess
need not control, or even predict, the moves of a novice in order to checkmate a
novice every time, neither would God be required to control a sinner’s individual
choices in order effectively to checkmate the sinner over time and to eliminate
every possible motive the sinner might have for rejecting fellowship with God.
Third, what is essential to the formation of a good character and to the gift of
saving faith is not that a rational agent should choose rightly rather than wrongly,
but that the agent should choose freely one way or the other. For God never simply
bypasses our own reasoning processes, however fallible and imperfect they may
be; neither does he violate our unique personality through manipulation, or by
simply implanting beliefs in us, or by artificially reconstituting us. Instead, our
own reasoning processes and the choices we make help to determine how God can
respond most appropriately, given the lessons we still need to learn, in bringing the
second stage of our creation to its glorious completion.
Still—and this is my fourth and final point—once we have emerged as individual
centers of consciousness and rational agents, God can nonetheless transform our
perspective, perhaps even instantaneously, in a perfectly rational way; he need
only grant us a direct “face-to-face” encounter with himself, thereby providing
compelling evidence for both his existence and the bliss of union with him. By
“compelling evidence” I mean (roughly) evidence that both (a) justifies one in
believing a given proposition and (b) renders one powerless in the face of this
evidence not to believe it. If an alien spaceship should unexpectedly land in full
view on the White House lawn, then this would no doubt alter the perspective of
many people almost instantaneously and would do so in a perfectly rational way;
and similarly, if Saul of Tarsus (or Paul) really did encounter the risen Lord on the
road to Damascus, as Christians believe he did, then it is hardly surprising that
such an encounter should likewise have altered his anti-Christian perspective in a
perfectly rational way. More generally, for any person S who is rational enough to
qualify as a free moral agent, if S should have a direct encounter of the relevant
kind with God, S would then possess compelling experiential evidence, I suggest,
for both the existence and the unsurpassable goodness of God.52 And that is why,
with respect to anyone who is rational enough to qualify as a free moral agent,
God always has a trump card to play, namely the revelation of his own being, that
guarantees from the outset his ultimate victory over sin and death.
Some will no doubt ask at this point: “Well, if God has such a trump card up
his sleeve, so to speak, why not play it sooner rather than later?” But I would ask
just the opposite question: “If God has a guarantee of ultimate victory, why not
play his trump card later—at the moment of each person’s death, if necessary,
or even later than that—rather than sooner?” Why not, in other words, allow the
drama of human history to play itself out on its own and in a context where parents
have the privilege of raising and caring for their children, where one person’s
choices can have a direct bearing upon the temporary welfare of others, where real
dangers and real threats to one’s temporary happiness exist to struggle against,
and where one’s personal failures and sins give real meaning to repentance,
forgiveness, reconciliation, and atonement? Imagine a world without any of this.
Imagine first a world with no created order at all, a world consisting of nothing
but an eternal Trinity, where the Father’s extravagant artistic skills and creative
powers lie eternally dormant and unexercised, where his infinite grace has no role
to play, and where his unbounded capacity to perfect the unperfected and to care
for the weak and the helpless has no means of expression. Are we to suppose that
such a world, even if possible, would be anything like as desirable from God’s
perspective as a world like ours in which everyone has a story to tell, indeed lots
of stories, but no one is finally excluded from eternal bliss? For my own part, I find
such a supposition utterly implausible.53
But now try to imagine a world in which God creates billions upon billions of
people over time, not one of whom has a real live story to tell, except this: Once a
distinct center of consciousness emerges, it is immediately brought into a mystical
union with God where it remains forevermore, sort of like someone experiencing
an eternal high, perhaps even quivering forever with intense pleasure, but without
anything further to do. In such a static world (without meaningful progress) there
would be no adventure, no quest for truth, no new discoveries to be made about
the wonders of God’s creation, no moral struggles of any kind to be won, and no
need for God to repair or to cancel out the harm we have done either to others or
to ourselves. Such a world would not only be very different from the actual world,
but would also be, in my opinion, altogether inferior to it as well. For it is simply
52
As Marilyn McCord Adams points out, “What accounts for our refusal, our
miscalculations of what might be reasonable to accept is ignorance of a special kind … Not
propositional ignorance of what we might read in textbooks. But experiential ignorance of
the immeasurable goodness that God is.” See “Plantinga on ‘Felix Culpa’: Analysis and
Critique”, Faith and Philosophy, 25/3 (2008), p. 138.
53
I therefore belong to the camp that thinks it necessarily true that God creates
additional persons to love and, for all we know, may never stop creating additional persons
to love.
a mistake, as I see it, to view the bliss of union with God as if it were logically
separable from the things we do in this earthly life, the things that happen either
to us or to our loved ones, and the grace imparted to us over time and in many
different contexts. It is no less a mistake to view such bliss as logically separable
from the tasks we shall continue to perform as God reveals the riches of his grace
through us in future ages (see Ephesians 2:7).
Put it this way: As the most creative artist conceivable, God loves a good story,
and he has granted each of us the privilege of being a part of many good stories,
perhaps even infinitely many of them—stories that will never end but will instead
merge gradually into one great ever-expanding story in which, as C.S. Lewis put it
at the end of The Last Battle, “every chapter is better than the one before.”
Chapter 2
Is it Possible to Freely Reject God Forever?
Raymond J. VanArragon
Many Christians believe that some people will spend the afterlife in hell, alienated
from God forever. This belief is a majority position in the Christian tradition,
primarily because it seems to be taught in the Bible and in particular by Jesus
himself. But recently among Christians there has been a certain anxiety about
the doctrine of hell, an anxiety that has developed in some quarters into outright
rebellion. Powerful philosophical and biblical objections have been put forward
against the doctrine, claiming that it would be morally unacceptable for God to
allow hell to exist and that scriptural passages traditionally cited in support of it
can be interpreted differently. Discontent with hell is growing.
Despite this, many Christians continue to maintain that the doctrine of hell is
defensible. Much of their defense remains biblically based; but for some it is also
rooted in the conviction that human beings have freedom, freedom understood in
the libertarian sense where to perform a free action requires that it not be caused
or determined by anything other than the agent. If human beings are free, so the
reasoning goes, then they are able to freely choose to separate themselves from
their Creator or perform actions that merit the punishment that hell might include.
And it would be fitting, even loving, for God to allow people to make such choices,
respecting their autonomy even if they should use it to damn themselves forever.
Thus, some defenders of the doctrine of hell suggest, the Bible dictates that some
people will be eternally damned, and philosophical reflection on the nature of human
freedom gives us a plausible way to fill in the story of how this will happen.
But this libertarian gambit has not ended the debate either. For how could
anyone really choose to reject God forever? Would not such a person have to be
crazy, or desperately misinformed? And would not a loving God actively prevent
anyone from doing something so catastrophic? Some opponents of the doctrine
of hell argue that the libertarian picture just presented cannot possibly be true,
because no one could make the free choices imagined, and God would not let
people destroy themselves in that fashion anyway. If this is correct, then that
closes the door to one promising defense of this mainstream Christian doctrine—
and perhaps the most promising, since hell without libertarian freedom presents
philosophical problems that are even more severe.
For that reason, in this chapter I want to defend a libertarian conception of hell
and damnation from the objections just sketched out. I will argue that it is in fact
possible to freely reject God and hence to damn oneself forever. After explaining
what it means to freely reject God forever, I will defend the possibility of doing so
by responding to some arguments against it. As I go along I will also sketch out a
proposal for how a person could freely reject God forever and how God could be
entirely justified in letting it happen, and even in a sense facilitating it.
The reader should be aware that in what follows I will not be assuming a
medieval picture of hell, according to which residents there must spend every
moment in unspeakable agony.1 To reject God forever and so to spend eternity
alienated from God does mean tragic and endless failure to reach the end for which
one was created, but that need not imply a state of being akin to living in an eternal
torture chamber.
Freely Rejecting God Forever
Let us begin, then, by considering what it might mean to freely reject God and
what it might mean to do so forever. There are numerous things that a person
can do that could constitute rejecting God. For instance, he might respond with
a firm, “No, thank you” to a clear and explicit invitation from God to follow him
and abandon some unhappy way of life. Or he might choose to turn around and
head in the other direction in response to a personal welcome from God at the
pearly gates, with a good view of what is inside. These are examples of blatant,
brazen ways to reject God. But there are also more subtle ones. Deciding to pursue
a policy of selfish behavior and to henceforth put oneself ahead of God and all
others constitutes rejecting God. Performing some sinful action while being aware
that one is displeasing God (or by ignoring God or God’s will) is also a common
way of doing so. Indeed, simply performing an action that one knows one should
not perform also could be understood as rejecting God, even if the agent has no
clear awareness of God or God’s displeasure. Doing so does go against God’s will
and command, after all, while also rendering the agent blameworthy and deserving
of punishment.
In what follows, then, I propose that we understand rejecting God in a very broad
sense, so that to reject God means to choose to act in a way that runs contrary to God’s
will.2 To freely reject God in this fashion requires more than that, especially since
1
For a discussion of the medieval picture see Kelly James Clark, “God is Great, God
is Good: Medieval Conceptions of Divine Goodness and the Problem of Hell,” Religious
Studies, 37/1 (2001): 15–31.
2
Readers might object that freely sinning without having any awareness of God
should not be called “rejecting God,” and that by defining rejection of God so broadly I am
draining the term of any sensible meaning. Please note, in response, that I define the term
this broadly for the sake of simplicity, so that I do not need constantly to distinguish in the
text between sinning with awareness of God and God’s will and sinning only with awareness
of right and wrong. And it is not essential that I distinguish between them because, in my
view, persistently sinning in either way can merit equally the consequences that I shall
discuss later in the chapter.
Is it Possible to Freely Reject God Forever? 31
acting freely entails being morally responsible for at least some of the consequences
of one’s action. To freely reject God requires a certain degree of rationality and
some knowledge and awareness about what one is choosing, what the consequences
might be, and what other options there are. Moreover (and here our account of free
action becomes explicitly libertarian), there can be nothing that causally determines
the choice that one makes, so that one’s free action could not be caused by God,
dictated by the laws of physics, or entailed by one’s own psychological states. This
last condition is worth some elaboration. It is plausible to think that one cannot
freely perform an action that one has absolutely no desire to perform, or freely
refrain from an action that one has strong desire to perform and no desire not to.
In such cases, one’s action (or failure to act) is psychologically determined and
therefore not free. If a choice is to be free, then, the agent must have options and
must have some desire for more than one of them, where more than one is such that
performing it is consistent with the laws of physics and with God’s causal activity.
We should supplement this simple picture of the conditions for free action in
two ways. First, a person can be blameworthy for having performed an action
when the action itself was not a free one, if his lack of freedom was appropriately
linked to his previous free actions. For example, if a person acts non-freely due to
ignorance of his options and their consequences, but that ignorance is itself a direct
consequence of his past free actions (perhaps he earlier chose not to pay attention
when the relevant information was broadcast), then in so far as he is blameworthy
for his ignorance he is also responsible for his action. Second, in some cases a
person may actually desire only to perform one particular action among the options
before him, but it may nonetheless be (or have been) within his power to summon
up the desire to perform one of the others. Perhaps this is common when people
succumb to sinful temptations: at the moment of acting they lack the desire for
anything except for the option they choose, but earlier they did have the freedom
to stop and ponder the moral character of that option, and had they freely done so
they would have had at least some desire not to sin. In such cases, while the action
may not strictly speaking be a free one (given the presence of no countervailing
desires), it is nonetheless “close enough” to free for our purposes and definitely an
action for which we would be inclined to blame the agent.
To sum up, then, to freely reject God is to act in a way that goes against
God’s will, where performance of that action is sufficiently rational, sufficiently
informed, and not determined by God, nature, or desire. Moreover, given the
broad way that we will understand the notion of rejecting God, freely doing so is
possible for a person who is of a non-Christian religious persuasion, or is actively
ignoring God, or does not even believe that God exists. It is sufficient that the
agent has some sense of right and wrong, a sense that matches up at points with
what God commands and forbids, and that at a point of matching up the agent
freely does what she recognizes is wrong. And if she performs such an action non-
freely but is blameworthy for its non-free character (due to ignorance or a lack of
countervailing desires), she may still be blameworthy for performing it and, we
should add, for some of its consequences as well.
Now, what would it be to freely reject God forever? There are two possibilities:
first, to do so might involve making one critical free choice (or a finite series of
choices); and second, it might involve making an endless series of choices. In the
simplest example of the first case, a person could be presented with two options:
to forever close the door on God, so that throughout her continued unending
existence she will never again get the chance to experience the divine presence,
or to refrain from doing that. If she freely chooses the first option and closes the
door, then we can properly say that she has freely rejected God forever. Or, in the
second case, a person might just continually sin, continually and consistently (and
with the required knowledge and awareness) do what is wrong, what in fact goes
against God’s will—where this carries on not just for a lifetime, but endlessly. If
such a person’s sins are free at least some of the time and there is no point after
which none of the person’s sins are free, then we can properly say that she too has
freely rejected God forever.3
Is it Possible to Freely Reject God Forever?
Is it possible in either of these ways for a person to freely reject God forever? I
grant that it is difficult to defend the possibility of the first kind of rejection, where
only one free choice (or a finite series) is made. The main obstacle to the freedom
of such a choice, as I see it, is ignorance. An appropriate degree of knowledge and
awareness is a necessary condition for free choice; but it seems doubtful that anyone
who chooses to close the door on God forever will have a sufficient grasp of what
she is rejecting, and even more doubtful that she could appropriately comprehend
the monumental character of her choice. Indeed, it seems likely that no human
being can freely choose something where the consequences extend forever, where
the “forever” is part of the content of the choice—simply because none of us can
adequately grasp what that would mean. If so, then anyone who chooses something
forever is not making a free choice, in so far as she lacks adequate knowledge of
what she is choosing. The choice to reject God forever in the first way, then, seems
non-free: such an agent could not have the requisite knowledge of what she is
choosing. For that reason, in what follows I shall leave this possibility aside.
(It is worth noting, however, that even if it may not be possible for a person
to make one free choice to reject God forever, it may be possible for her to freely
make some choice which has the consequence of rendering her forever unable
to freely accept God. For instance, a person may freely choose to perform some
evil action; this may be the choice that finally solidifies her character for evil and
prevents her from desiring what is good. Such a situation may be similar to that
of a drug user who crosses the threshold into uncontrollable addiction. By freely
3
Perhaps we could only say this from an eternal or timeless perspective, since it will
never be true, at any point in time, that she has freely rejected God forever (though it may
be true that she will).
choosing to partake of the drug one more time, the addict enslaves himself to it.
Presumably he did not choose enslavement; that was an unforeseen consequence
of his choice. Still, he could bear some blame for his situation, in so far as he
should have known that such a lack of freedom could result from careless drug
use, and he should have refrained from starting out on that path in the first place.
Similarly, the person who essentially enslaves herself to sin—and even does so
permanently—may not have chosen such enslavement, but she may nonetheless
bear some blame for her predicament in virtue of the fact that she should have
known that by liberally doing what was wrong she was courting catastrophe, and
that she simply should have refrained from freely sinning. This possibility too I
shall mostly set aside in what follows, though some of the discussion in the next
two sections will be relevant to it.)
What about the second kind of rejection? Can a person freely sin forever,
continually and consistently doing what is wrong? I believe that such a thing is
possible, but first I want to defend this possibility indirectly by considering an
important argument that it is not. The argument comes courtesy of Thomas Talbott,
a prominent contemporary defender of universalism. In arguing that no one can
continue rejecting God without end, Talbott says this:
Once one has learned, perhaps through bitter experience, that evil is always
destructive, always contrary to one’s own interest as well as to the interest of
others; and once one sees clearly that God is the ultimate source of human
happiness and that rebellion can bring only greater and greater misery into
one’s own life as well as into the lives of others, an intelligible motive for such
rebellion no longer seems even possible.4
His contention is that any person’s motivation for sinning will eventually be
stamped out: sin will bring painful consequences to the sinner (and others) to such
a degree that she will be left with absolutely no motive to carry on with it. And
since having no motive seems to entail having no desire—having the desire does
give one at least the motive of fulfilling it—this would mean that the desire to sin
is eliminated, too. And if so, then she can no longer freely sin. The choice to sin
under those conditions would be patently irrational. As Talbott says, it would be
like a boy who, with no motive for doing so and strong motive not to, continually
thrusts his hands into a fire.5 Thus the consequences of sin will eventually make it
impossible for the persistent sinner to continue sinning, and she will inevitably and
non-freely turn instead to God.
Is Talbott correct about this? It seems to me that at least two claims are
questionable. First, he claims that evil is always destructive and inevitably brings
greater and greater misery into the life of the perpetrator. He is not saying that
4
Thomas Talbott, The Inescapable Love of God (Boca Raton, 1999), p. 186.
5
Ibid., p. 184. See also Talbott’s essay, “Craig on the Possibility of Eternal
Damnation,” Religious Studies, 28/4 (1992): 500–502.
evil has these features in virtue of the fact that God actively punishes evildoers.
Instead, evil has these features essentially, by its very nature. God does not have
to step in and punish the sinner so that doing evil leads to misery; evil does that all
by itself. Second, Talbott believes that all persistent sinners will eventually come
to recognize that evil is in fact destructive for them, and this recognition will rob
them of any motive (any desire) to continue sinning. This recognition too comes
along with the very nature of evil—that is the way evil is. Let us consider why
both these claims may be mistaken.
First, it is doubtful that evil is always destructive for perpetrators in the way
that Talbott says it is. It is certainly true that some kinds of evil—some kinds of
actions that go against God’s will—seem to bring greater and greater misery into
the lives of the perpetrators. The behavior exhibited by drug addicts fits this mold.
But there are countless other less dramatic ways one might choose to act that run
contrary to God’s will, and it is not obvious that they are all similarly destructive.
Think, for instance, of sins involved in mistreating the weak, the disadvantaged,
the outsider. This sort of behavior is common, and it does not seem especially
destructive to those who participate in it, especially if they are surrounded by
people who make them feel loved and valued. Think also of the easily ignored
sins involved in spending money on status symbols and luxuries when one could
instead use that money to help people who desperately need it. Behavior of that
sort is also widespread and yet clearly runs contrary to God’s will.6 And consider
finally the sort of negligence Jesus mentions in the famous New Testament passage
about the sheep and the goats, where some (the goats) are accused of failing to feed
the hungry, give the thirsty something to drink, clothe the naked, care for the sick,
and visit the prisoner.7 Again, these failures are sinful, but it is a bit of a stretch to
call them “always destructive, always contrary to one’s own interest” (emphasis
added). Many people who engage in them seem well able to carry on even when
they have some awareness of the moral status of the choices they are making.
(Of course, that does not mean that there is no cost to this sort of sinful living.
So-called eudaimonistic moral theories, developed by the ancient Greeks and
attractive to many Christians, tell how immoral behavior leads to an impoverished
existence and disorder in the soul, and that it prevents people from attaining true
happiness and fulfillment. But these costs can be much less noticeable than Talbott
suggests.)
Second, is Talbott correct that all persistent sinners must eventually recognize
the destructive nature of evil in such a way that they are robbed of any desire to
continue sinning? Clearly, anyone who rejects Talbott’s first point will also reject
this one, since if evil is not necessarily destructive in the way he specified, then
it may not be destructive for every persistent sinner and hence some of them may
fail to recognize its destructive power. But even if evil is damaging to a person, it
6
Powerful arguments for the immorality of such behavior have been presented, for
example, by Peter Unger in his book Living High and Letting Die (New York, 1996).
7
Matthew 25:31–45.
is not clear that knowledge of this will necessarily drain from persistent evildoers
the desire for evil or eliminate their ability to freely choose it. It may be that while
they are properly aware of what they know about the consequences of evil it is
impossible that the desire to choose it be present; but our freedom clearly extends
to what facts we are properly aware of. Many of us know that certain kinds of sins
are not really in our best interests, and yet under certain conditions we willfully
render ourselves unaware of what we know (or perhaps purposefully subject it to
doubt) and choose to perform them anyway.
So it is not obvious that full knowledge of the destructive effects of sin (whatever
they may be) will of necessity eliminate the freedom to do it. Indeed, knowledge of
both the destructive effects of sin and benefits of turning to God may not eliminate
the freedom to sin either. Talbott claims that “if God is the ultimate source of
human happiness … then anyone in a position to make a fully informed decision
[to reject God] would also seem to have the strongest conceivable motive not to
reject God,”8 but this too is questionable. Presumably a “strongest conceivable
motive” is a motive that demolishes opposing motives, so that if one has such
a motive to do A, then one is not able even to awaken a desire to do not-A. And
how would being fully informed give one a motive like that? Again, many of us
already have a great deal of knowledge about the deleterious effects (to others and
ourselves) of sinful behavior and the various goods that come with doing the right
thing, and yet we freely sin anyway.9 Would comprehensive knowledge of this sort
change things? I do not see why it would have to.
Well, why not? The problem, I think, is that mere knowledge is often not
sufficient to eliminate or control strong desires. Knowledge of what is good
and right often needs a supplement, a boost from experience before we can no
longer shrug it off and ignore it when we make decisions. Nearly all of us know
that it is dangerous to drink and drive, and yet many people fail to act on this
knowledge until something—perhaps a loved one getting injured in an accident
with a drunk driver—jolts them and makes the knowledge come alive, so to speak.
And it may be the same for the knowledgeable sinner: being fully informed about
the consequences of sinful behavior (rejecting God) may provide the “strongest
possible motive” to turn around only when the information comes with some sort
of powerful and jarring experience. Perhaps a tour through a refugee camp, an
encounter with an angel, or some personal tragedy might do the trick. Yet such
experiences may not be sufficient either. People who vow off drinking and driving
do slip up after all, as thoughts of their loved ones fade from view. Similarly, after
8
Talbott, “Craig on the Possibility of Eternal Damnation,” p. 500.
9
It is actually quite difficult to characterize the epistemic status of our beliefs about
the real consequences of sin. Perhaps when we freely choose to sin, we at the same time
do not even hold those beliefs, even if we do hold them at other times. Perhaps the beliefs
come and go, so to speak. But surely it is the case that if they do come and go, and if when
we sin we fail to know that sinning is bad for us (because at the time we do not believe it),
that can be our own fault, something for which we are blameworthy.
episodes of violence, abusers often feel so remorseful that the desire to lash out
is eliminated in them; but as time passes such desires tend to return. With this in
mind, I should think that the only sort of experience that would necessarily and
permanently stamp out the desire to sin would have to be both overwhelming
and long-lasting. And it is not obvious that having that sort of experience is the
inevitable consequence of perpetual sinning. Thus, even if such an experience
were effective in permanently turning the sinner around, there is no guarantee that
the sinner would encounter it—unless God would of necessity intervene and force
such an encounter in order to prevent anyone from freely rejecting him forever. We
will consider that possibility in the next section.
For now, we should conclude that the arguments we have considered do not
establish that it is impossible to freely reject God forever. Persistent sinning need
not on its own have such obviously destructive consequences that the sinner loses
any desire or motive to carry on doing it. Moreover, the sinner’s being very well
and even powerfully informed about what is truly good for her seems consistent
with her maintaining (or soon recovering) the desire to sin. And provided she
at the same time has some desire to do what is right (to follow God’s will) and
the other conditions for free choice are met, in maintaining her desire to sin she
maintains the capacity to do so freely.
I conclude this section by briefly indicating what it might look like for a person
to freely sin, and hence reject God, without end. Many of the ingredients have
been implicit in my responses to Talbott. Such a person could live an ordinary
life, only one which carries on forever. (Perhaps after earthly death, his existence
continues more or less as it had before.) His mental life could be fairly shallow
and his memories may not run especially long. His actions could reflect the sort
of sinful choices that many of us are prone to, choices where he consistently puts
himself ahead of others or does not look outside his own needs, where he fails to
feed the hungry, give the thirsty something to drink, clothe the naked, care for the
sick, and visit the prisoner. These choices could be rooted in a conscious decision
to elevate his own perceived good above the good of others, or they could be less
reflective but nonetheless consistent. He does not look to change his lifestyle, or in
a heartfelt way to request divine assistance to do so, because his situation does not
appear to him to be all that bad overall. Again, in so far as it is obviously possible
for a person to manage this throughout his earthly life, it seems possible, under
similar conditions at least, for such behavior to carry on indefinitely.
Some may object that this picture of the afterlife is overly tame. But remember
that my aim is only to defend the thesis that it is possible for a person to freely
reject God forever. Perhaps a person could do so in much harsher postmortem
conditions, but the possibility is easier to defend on the plausible assumption
that postmortem conditions could be similar to ante-mortem ones.10 And it is also
10
Charles Seymour, in his book A Theodicy of Hell (Dordrecht, 2000), adopts what he
calls the “freedom view of hell.” On his view, those in hell stay there because they “eternally
choose to reject God through their sin” (161). That is consistent with the possibility I am
important to note that in the picture I am developing, the state of those not saved is
monumentally less happy than that of those in heaven, even if they are not aware
of this. In what follows I shall fill in this picture somewhat when I first consider
whether God should intervene to ensure the person’s salvation and prevent such
an unsatisfying existence from continuing, and then return to the concern that the
persistent sinner’s actions ultimately would not be free.
Must God Intervene to Prevent Unending Rejection?
I have suggested that it is possible for a person to freely sin (and hence freely
reject God) forever; but would God even let this happen? I want to consider and
reject an argument, again from Talbott, that God would of necessity intervene and
prevent anyone from doing so; then I shall argue that God could in fact have good
reason not to.
Talbott’s argument that God must intervene makes use of an analogy. Suppose
a loving parent is in a position to prevent her child from committing suicide. To
commit suicide is to do oneself irreparable harm (or at least to bring harm on
oneself that no human being can repair). For that reason, a parent ought to interfere
with her child’s actions, even if doing so means compromising her child’s freedom.
Now God is in a similar position with respect to a person who will freely reject
him forever. To reject God forever is to do oneself irreparable harm. Therefore,
like the parent, God ought to intervene and prevent it, even at the cost of violating
the person’s freedom. As Talbott puts it, there are limits to the freedom God can
allow a person to have! A loving God should not and would not allow a person to
bring upon herself harm that is truly irreparable.11
As compelling as this argument is, it is important to see that it is ineffective
against the possibility I am defending. I have not argued that it is possible for a
person with one free choice to forever close the door on God. By making a choice
like that, a person would indeed be doing herself irreparable harm; and Talbott
makes a plausible case that God would not let that happen.12 But I have instead
defending. But he also defines hell as “an eternal existence, all of whose moments are on the
whole bad” (6)—bad both subjectively, because it feels bad to the person, and objectively,
because the person is not achieving happiness of an Aristotelian sort (7). It may be possible
under those conditions for the person to freely reject God forever, but that is a more difficult
case to make.
11
See Talbott, “The Doctrine of Everlasting Punishment,” Faith and Philosophy, 7/1
(1990): 19–42, pp. 38–9, and The Inescapable Love of God, pp. 189–92.
12
This observation also applies to the possibility, noted in the second section above, of
a person performing an action that permanently enslaves him to sin. Such a person would be
doing himself irreparable harm, and thus Talbott has provided a good argument for thinking
that God would intervene to prevent him from doing this or to ensure that the enslavement
is not permanent.
defended the possibility of a person forever continuing to reject God freely. And
a person who does that at no time brings irreparable harm on herself: there will
always be a time in the future when she is free to do what is right and thereby
turn toward God.13 (There may of course be many moments where she is not free
to turn toward God; but there will always be future moments where she is.) So
Talbott’s analogy does not apply: it does not give us reason to think that God must
dramatically intervene in the life of someone who continually rejects him without
ever bringing irreparable harm on herself.14
Let us turn now to reasons God might have to refrain from intervening and
actively preventing a person from freely rejecting him forever. One way God
might intervene would be to simply eliminate the person’s freedom to sin. God
may have some options here too, but before we consider them we should note that
eliminating this freedom is not something to be taken lightly. The freedom to sin (or
do right), after all, is a valuable thing. Christians have long argued as much, often in
response to the problem of evil. There is something deeply significant about being
able to choose one’s path and face the appropriate consequences of those choices.
Moreover, many contend, freedom makes possible genuine relationships with God
and other people. If God were to give people morally significant freedom with the
condition that the ability to go wrong would be eliminated if it were persistently
used, that would seem to undercut its value. One thinks that God would, other
things being equal, be disinclined to forego the goods that come with morally
significant freedom, especially against the will of the person who continually and
freely chooses to disobey him.15
So, eliminating the persistent sinner’s freedom would in many ways be a bad
thing; but how might God do it, anyway? One way might be “internal”: God might
manipulate the brain of the sinner and amputate his desire to sin. This could be
problematic: not only would it violate the person’s autonomy by eliminating his
13
She also does not at any time “undermine the very possibility of supreme happiness
… in everyone else [i.e., her loved ones in heaven] as well” (Talbott, The Inescapable Love
of God, p. 189).
14
My description of how a person could freely reject God forever escapes a related
objection of Talbott’s. In “Freedom, Damnation, and the Power to Sin with Impunity,”
Religious Studies, 37/4 (2001): 417–34, Talbott criticizes those who propose a free-will
theodicy of hell but fail to offer a clear explanation of what it might mean to “embrace
an eternal destiny freely” (pp. 418–21). I am not defending the possibility that one might
embrace an eternal destiny freely—that sounds like making one free choice (or a finite
number of free choices) to close the door on God forever. I am defending the possibility that
a person might forever continually make choices that run contrary to the ones God wants
her to make. Such a person, as I am envisioning her, is not properly described as embracing
an eternal destiny; she could instead be mostly living in the moment.
15
We should note the common view that people in heaven do not have the freedom to
sin; but their state is due at least partly to their own choices to do what is right, and to their
heartfelt decision to ask God for help. For more on this issue, see James F. Sennett, “Is there
Freedom in Heaven?” Faith and Philosophy, 16/1 (1999): 69–82.
freedom without his consent, it is also not obvious that God could do this without
significantly damaging the person or causing untoward side effects. Desires, after
all, are complex things, deeply intertwined with other features of a person’s psyche,
and not easily erased, perhaps even by God. And it is unclear in any case whether
such a desire extraction would take, in a person who is not fully on board with
the procedure. Rehabilitation programs tend not to work for addicts who are not
whole-heartedly committed to their success; perhaps desires eliminated against or
independent of an agent’s own will would over time rear their ugly heads again
without constant autonomy-depriving attention from God. And of course if those
desires return, with them comes the freedom to sin again.
Another way that the desire to reject God could be eliminated might be
“external”: for example, God could forcefully reveal to the sinner the nature of
his own predicament, the consequences of his actions on himself and others,
and the merits of a different way of life. As we have said before, it could take a
truly spectacular and long-lasting experience of this sort to guarantee elimination
of that person’s desire and freedom to sin; anything less than that may leave
open the possibility of continued free rejection. And God might be excused for
refraining from resorting to such drastic measures in order to rescue a persistent
and unrepentant sinner.
But it is worth pondering further what character such a revelation must take
in order that it might effect the necessary changes in the person receiving it.
The persistent sinner we are imagining is one who has followed a policy of self-
centered behavior, putting her own needs first and setting herself above everyone
else. It would certainly be unsatisfactory for God to attempt to win her over simply
by showing her the glories of heaven and the throne she might sit in there: such
a revelation by itself could leave her self-centered inclinations (and her lack of
concern for others) intact while merely providing new information on how to
fulfill them. That would be like revealing to a greedy thief that he need not engage
anymore in petty theft, because he can help himself to all the money in a loaded
bank vault that has been opened for him. The thief would find this appealing, but
only because it would speak to his over-riding greed. Such a revelation would
not change the thief’s fundamental and sinful desires. And what sort of revelation
could be guaranteed to do that? Giving the thief insight into the lives of people who
are poor but deeply happy and content may be ineffective because the thief just
might not see it. He cannot feel their happiness himself, and the notion that their
lot is desirable may strike him as bizarre. And of course something similar may
be true of the persistent sinner. She may fail to see the appeal of the lives of those
who follow God’s will. (The appeal may be clear only to those who do it.) Perhaps
such lives would be attractive if her own lifestyle had proven disastrous to her, but
as we have seen that is not inevitable. Suffice it to say that these sorts of reflections
indicate at the very least that it is not obvious what sort of revelation might be
guaranteed to effect the necessary realignment of the persistent sinner’s desires.
We have been discussing methods that God might employ to eliminate the
sinner’s ability to carry on rejecting God. We should recognize, however, that
taking such a drastic step need not be God’s only avenue to saving the sinner who
would, without intervention, continue freely rejecting him. For surely God could
take steps simply to weaken her desire to sin, perhaps by removing some of the
ignorance that clouds her decision-making, and in so doing start her on a path that
would ultimately lead to the complete transformation of her will and to union with
God in heaven. Of course, such steps may prove unsuccessful since they leave the
sinner’s freedom intact. Still, I want to conclude this section by arguing that God
could be perfectly justified in for the most part letting the sinner be and refraining
from forceful interference. The argument I give will also provide one more reason
that God might refrain from engaging in the more dramatic, and problematic,
interventions just discussed.
Now on the face of it, it seems like God would have an over-riding reason to
remove the ignorance of the persistent sinner: doing so is necessary for her to have
and maintain the ability to choose freely. We said earlier that in order for an act to
be free, the agent must have an appropriate degree of knowledge and awareness
of what she is choosing, what the options are, and what the consequences are.
And our persistent sinner seems to lack this. She continues to choose in a self-
centered fashion, unaware of the joy that she is missing out on and ignorant of
the comparative misery of her own lot. Does not her ignorance threaten both her
freedom and responsibility for what she does? If so, then one thinks that God
ought to intervene and enlighten her.
We can begin to see that God need not to do this if we focus a moment on what
our sinner does know as she makes her choices. She knows that there are others
suffering, that she has the power to help, that it would be morally right or good
to do so and wrong to neglect it, and perhaps that God commands that she do so.
There are all sorts of variables here: the knowledge may grow dim over time; she
may find it easier as she goes along to ignore the plight of others and ignore the
commands of God (and ignore that they are God’s commands); her awareness of
these alternatives to self-centered behavior may be altogether eliminated as that
behavior becomes habitual. (We shall return to the habitual nature of her actions
in the next section.) But in any case, what she does know is sufficient to make her
sinful choices free, at least some of the time. Free action does not require anything
like comprehensive knowledge, after all.
It is true, though, that non-culpable ignorance may limit responsibility for
some of the consequences of sinful actions. If a person’s failure to know of some
terrible consequence does not stem from some previous moral failure of his, then
his responsibility for that consequence is reduced should his actions bring it about.
But the important point here is that our persistent sinner is blameworthy for her
ignorance, and indeed her ignorance may be the significant consequence, to her, of
her pattern of sinful behavior. She does not know the satisfaction of a life of service;
she does not know the peace that attends such behavior (in the short or long term);
she does not know that moral behavior is essential to genuine human flourishing
and that in fact her sinful choices harm her in subtle but very real ways—including
by extending her ignorance and making her behavioral alternatives less obvious
to her. (Or if she has some knowledge along these lines, it is easy to ignore come
decision time.) And neither does she know, in any compelling way, that the end
result of moral living is salvation,16 glorious union with God, and the fulfillment
of her deepest desires. But the fact is that she would at least begin to know these
things if she were to freely do what is right. She can do that, and indeed she ought
to. Thus her ignorance, which clouds her judgment and leaves her in a truly woeful
position in comparison with that of the residents of heaven, is a consequence of her
own free action and a consequence for which she is blameworthy. For God to allow
her ignorance to remain would respect the value of her freedom and the seriousness
of her sins, and indeed would serve as an entirely just punishment for those sins.
But if so, then God would be justified in failing to remove it.
To sum up, it seems that it is not the case that God must intervene to prevent a
person from freely rejecting him forever. The choices of the persistent sinner under
consideration are not akin to suicide since there is no point at which the person
does irreparable harm to herself. Moreover, foolproof measures that God might
use to stop the sinner, like elimination of her freedom, are plagued with difficulty.
Less drastic measures leave open the possibility of failure, since a person who
maintains the freedom to sin can continue to do it. And God may be justified in
withholding those measures in any case, since the unrepentant sinner deserves the
ignorance in which she wallows, and that ignorance may serve as an appropriate
punishment for the sins she continually commits.
But Could the Sinner’s Freedom be Maintained?
So far I have argued that it is possible to freely reject God forever and that God
could have good reason not to intervene and prevent it. In this final section I
consider and reject one more reason for thinking that any person’s ability to sin
freely must eventually be extinguished.
We have seen that Talbott believes that the ability to sin freely could not
be maintained because the desire to do so would eventually and necessarily be
eliminated when the person realizes that the sin is destroying her. I argued that
this is doubtful. What may be of more concern is that the person will become
addicted to sin in such a way that she is unable to refrain from committing it.
Selfish behavior, turning away from those in need, and being motivated solely by
the aim of fulfilling one’s own self-centered desires could become habitual to such
a degree that the agent loses any desire to do what is right, with the result that that
desire disappears and so does the freedom to act on it. If this result is necessary
and inevitable, then it is impossible to freely reject God forever, because at some
point the sinner must permanently lose the freedom to turn to God without freely
16
The moral living and its results are not entirely the product of her own power, of
course. A full Christian picture of this will appeal to (requested) divine assistance and the
work of Christ which makes salvation possible in the first place.
choosing to lose it. Such a person’s existence would be that of an irredeemable

sin-junkie, and even if her subjective state were not one of unrelenting misery
and anguish (since that would seem to give her some motivation to search for
a different way of living), there would still be a degree of pointlessness to her
existence that might make one wonder why God allows it to go on.
We should recall first in response to this concern that to freely reject God
forever does not require maintaining the ability to sin freely at every moment
of one’s existence. It requires only that there be no point after which none of the
person’s sins are free; and this is entirely consistent with there being periods of
time—even long periods—in which the person is effectively enslaved to sin. And
there are also a number of ways that such enslavement might at least temporarily
be ended.
First, the person might end it herself by mustering up the desire to serve
someone other than herself, a desire strong enough that she is capable of acting on
it. Of course, if she is always able (that is, free) to muster up such a desire, then her
failure to have it initially only eliminated her freedom in a strict sense; throughout
her sinning she was still in a state that earlier we called “close enough” to having
freedom. But if she does not always have this ability, she may nonetheless have it
some of the time; and if she acts on it and generates in herself such a desire, then
for that moment anyway she is free, and if she responds again by setting aside that
desire and sinning, that sin is a free one.
Second, the person might in the ordinary course of things encounter some
situation or experience that awakens in her the desire to help, as when one sees on
television the plight of the poor and suffering in other parts of the world. It is easy
to become deadened to these sorts of experiences because of their commonness
(and the ease of changing the channel), but if the desire, however short-lived, is
strong enough to motivate action, then the enslavement is temporarily lifted and
the person’s subsequent failure to act on it is free.
Third, God might intervene, either internally by tinkering with the person’s
psyche or externally by placing experiences in her path, to liberate her from her
sinful desires, not by eliminating them but by giving her the freedom to act against
them. This liberation, like the others, may or may not end up motivating right
action—and if it does then in the long run it may lead to the person’s salvation—
but the freedom it engenders may help to preserve the point of the person’s
continuing existence. And that could give God reason to do it, enabling the sinner
to freely reject God forever rather than drift into eternity with her ability to do so
completely extinguished.
And here we have a way that God could actually facilitate a sinner’s perpetual
rejection of him—not a malicious way, but instead a way that enables a person to
carry on a meaningful existence—by preserving the freedom and autonomy that
could otherwise be eliminated by sin. It would be a terrible thing for a person to
freely reject God forever, but it could be that only the grace of God stands between
that and a fate that would be far worse.17
Conclusion
In this chapter I have tried to demonstrate that it is possible to freely reject God
forever. To reject God forever is to alienate oneself from God forever, and thus
to defend this possibility is to defend the Christian doctrine of eternal hell. As I
have said, however, it has been no part of my aim to establish what hell is actually
like. Readers who accept the conception of hell sketched here—a hell in which
the sinner’s existence is much like an earthly one and the ability to sin freely
is partnered with the ability to begin to turn around—may be able to use what I
have written as a theodicy, an account of why God in fact allows hell to exist. But
many readers will not be satisfied with that conception. For some of them, what
I have said might serve as a starting point for an argument that a person could
freely reject God forever under much harsher postmortem conditions, or for an
argument that a person who does not actually freely reject God forever might
nonetheless justifiably be damned. Or, on the other hand, they may conclude that
a mere defense of the possibility of freely rejecting God forever is the best we can
do, and that much about the doctrine of hell is bound to remain a mystery to us.
It is worth turning in this connection to the writings of C.S. Lewis. Readers
familiar with them may have noticed that much of what I have said here fits quite
well with the account of hell he provides, in particular in his book The Great
Divorce.18 That is no accident, since I have often found Lewis’s writings on this
subject to be powerful and illuminating. But at the same time, it is important to
recall the cautionary words that end the preface to that book, where he says about
the fable he has spelled out, “the transmortal conditions are solely an imaginative
supposal: they are not even a guess or a speculation at what may actually await
us. The last thing I wish is to arouse factual curiosity about the details of the after-
world.”19
17
Indeed God may give freedom to the persistent sinner both to grant continued
meaning to her existence and to give her the opportunity to begin the process of turning
around and accepting him. (God could withhold this gift for long periods where the person’s
ignorance and lack of freedom serve as punishment for sin.) Theists differ on whether God
would know how the gift would be used before giving it; suffice it to say, again, that if the
gift grants the sinner freedom, the sinner may still freely choose not to use it as God would
want her to.
18
New York, 1946.
19
My thanks to my Philosophy Department colleagues at Bethel University, an
audience at the Society of Christian Philosophers meeting at Messiah College, Jerry Walls,
David Vander Laan, and Kelly Clark for helpful comments on earlier drafts of this chapter.
Chapter 3
Annihilationism: A Philosophical Dead End?
Claire Brown and Jerry L. Walls
While adherents to some religious traditions and most naturalists hold that, after
death, all human persons cease to exist, or at the least, that all human persons
permanently cease to be conscious,1 the traditional Christian position on death is
that it is followed by an afterlife. Disputes about various aspects of this afterlife
notwithstanding, the tradition has affirmed at least this: all humans who die will
eventually undergo a bodily resurrection, face divine judgment, and thereafter
continue forever to have conscious existence. The future of those judged righteous
will involve blessed existence in the presence of God, but the unrighteous will
exist forever in hell. While the precise nature of hell is contested, the tradition
does agree that, at least when compared to the fate of the blessed, hell promises a
miserable existence and, in some important sense, separation from God. Let us call
this view that some people after death will face continued conscious existence in
hell the “doctrine of hell” (DH).
While DH is part of traditional Christian teaching on the afterlife, it is not
without its detractors, even within Christianity. One way Christians deny DH is by
accepting universalism, the view that no one will suffer unending existence in hell
because, at least eventually, all persons will repent and be saved. A second way of
rejecting DH has been less popular traditionally but is today gaining an increasing
number of adherents. This view, typically called “annihilationism,” holds that the
unrepentant wicked do not face an eternal lived hell.2 Rather, at some point those
who reject God will be annihilated and cease to exist. This position is an attractive
1
On some forms of materialism with respect to the nature of human persons, human
persons are identical to their bodies, whether those bodies are living or not. On those views,
one does not cease to exist upon death. Rather, one continues to exist as a corpse. Because it is
in principle possible for a corpse to be preserved forever thereafter, a materialist of this stripe
could allow for the possibility that some people, once they come into existence, exist forever
thereafter. At the least, she will say that a great many people continue to exist after their death
even though most of them will (after sufficient decay of their bodies) cease to exist.
2
It is also sometimes called “conditionalism” to underscore the view that human
beings are not intrinsically immortal. Eternal life is thus a gift of God, and conditional upon
our acceptance of it. The term annihilation applies most strictly to the position that souls
are naturally immortal, and persist in existence unless something is done to destroy them.
See Clark H. Pinnock, “Annihilationism,” in Jerry L. Walls (ed.), The Oxford Handbook of
Eschatology (Oxford, 2008), p. 462.
option for those who are troubled by the traditional doctrine of hell, but remain
unconvinced by universalism.
Whether this annihilation is something that God inflicts on the unjust or whether
it is something they do to themselves (whether intentionally or unintentionally) is
a point of dispute among annihilationism’s adherents. A likewise disputed point
concerns whether the unjust are annihilated immediately upon death or whether,
instead, some time passes prior to such annihilation, perhaps during which time
they exist in (a non-eternal) hell.
Because defenses of annihilationism, unlike defenses of universalism, are
relatively recent phenomena, critical assessment of the reasons offered in support of
annihilationism are accordingly relatively rare. Most of the literature on this issue,
moreover, has focused on issues of biblical interpretation rather than philosophical
considerations. In what follows, we aim to help remedy this deficiency by
subjecting to critical analysis, and finally rejecting, three broad philosophical
motivations that a Christian might have for affirming annihilationism. Motivations
of the first sort turn on the notion that non-existence is the natural consequence of
sin or rejection of God, so we should not be surprised that annihilation is the fate
of the damned. The second and third broad categories of motivations are supposed
to succeed if one grants at the outset (perhaps on the basis of scripture or on some
other basis) that, when it comes to theories of the afterlife, DH and annihilationism
are the only live options. Motivations in the second category appeal to God’s
moral perfection, claiming that it is deeply at odds with DH, from which it follows
that annihilationism must be true. The third type of motivation is grounded in the
conviction that the continued existence of the sinful in hell is incompatible with
the final supremacy of Christ. So, if we are committed to the final supremacy of
Christ and if the only viable alternative to DH is annihilationism, annihilationism
follows.
These three types of motivation represent the only philosophical reasons for
adopting annihilationism of which we are aware. And the only non-philosophical
reasons for adopting annihilationism of which we are aware are rooted in certain
(non-traditional) interpretations of scripture. If our arguments are successful, then,
until new arguments in favor of annihilationism are developed, the theory should be
considered philosophically unmotivated. What is more, until such new arguments
for the annihilation of the unsaved are developed, the success of our chapter means
that even scripture-based arguments for annihilationism, while not obviously
incorrect, nonetheless have a high burden of proof to meet: if the tradition has
not licensed the annihilationist interpretations of scripture (as it has not), and if
(as our chapter maintains) we lack substantial philosophic reasons independent of
scripture for being predisposed to favor those annihilationist interpretations, then
we should be quite leery of annihilationist interpretations of scripture. Finally,
although we are quite explicitly addressing reasons Christians might seem to have
for adopting annihilationism, much that we say could be useful both to adherents
of other religions and to some non-religious non-naturalists.
Annihilationism: A Philosophical Dead End? 47
Natural Consequence Motivations
According to what that we will call (broadly) “natural consequence” motivations,

certain facts about the essence of sin and its natural consequences either entail or
make probable annihilation as the ultimate fate of the wicked. One argument that
draws on natural consequence motivations picks up on a traditional conception
of sin (or evil) as non-existence. According to certain branches of the Christian
philosophical tradition, most famously the Thomistic tradition, evil is not a positive
thing that exists but is simply a privation, i.e., a lack, an absence, or non-existence.
Evil’s opposite, goodness, is the opposite of such privation, namely, being. The
terms “goodness” and “being” are thus materially equivalent. It follows, according
to this argument, that if someone becomes increasingly evil, she exists to a less
and less degree. So, if she continues on this descent, she will (or at least may)
eventually cease to exist at all. Thus non-existence is the fate of the wicked in
hell: they become increasingly depraved until they finally become so evil that
they become extinct. We shall call this particular argument that trades on natural
consequence motivations “the privation argument for annihilationism.”3
Notably, while many have adhered to the traditional conception of good-
as-being and evil-as-privation, very few have drawn the conclusion that, if one
is sufficiently evil, one will cease to exist. Indeed, in the history of western
philosophy, the good-as-being view has been dominant among Christians while
seldom leading any who have held this view to infer annihilationism. While this
fact should not preclude us from considering whether or not annihilationism is the
appropriate conclusion to draw from the view that evil is a privation, it should give
us pause before leaping to that conclusion. In short, if we begin to think that the
view that evil is a privation entails that some will be so evil as to cease to exist, we
should at least ask whether we are misreading the privation view.
When we do investigate the view that evil is a privation, we find that, as Aquinas
understood it, it does not entail the possibility of the wicked ceasing to exist as a
direct result or natural consequence of their own sin. The heart of the doctrine that
evil is a privation, for Aquinas, is that “[g]oodness and being are really the same,
and differ only in idea.”4 Or, to use Eleonore Stump and Norman Kretzmann’s
translation of this line, typically, “the terms ‘being’ and ‘goodness’ are the same in
3
For discussion of the privation argument and some of its close relatives see Jonathan
Kvanvig, The Problem of Hell (New York, 1993), pp. 146–7, and Paul J. Griffiths, “Self-
Annihilation or Damnation?: A Disputable Question in Christian Eschatology,” in Paul J.
Weithman (ed.), Liberal Faith: Essays in Honor of Phillip Quinn (Notre Dame, 2008),
pp. 83–117. What we call “the privation argument” should not be confused with responses
to the problem of evil that turn on the claim that evil as such does not exist but is merely a
privation.
4
Summa Theologica, trans. the Fathers of the English Dominican Province (London,
1920), PP, Q. 5, A. 1.
reference, differing only in sense.”5 Armed with this piece of knowledge, we may
indeed be tempted to assume that the doctrine implies that those who are evil have
less being than those who are not, so that in becoming increasingly evil one runs a
genuine risk of losing all of one’s being and thereby ceasing to exist.
Crucially, though, when a Thomist says that “being” and “goodness” typically
share a referent but not a sense, he means something different than a contemporary
analytic philosopher would mean if she were to say the same thing, and this
disparity occurs because the medieval conception of being extends beyond the
contemporary idea of actual existence.6 In contemporary discussions, “being” is
conceived of in a single way: as something that a possible object either has (when
it actually exists) or lacks (when it does not actually exist) and that does not come
in degrees.7 In Aquinas, however, one can distinguish two conceptions of “being”:
absolute existence and existence in a certain respect. Absolute existence is simply
the instantiation of a substantial form, and it comes close to our contemporary
notion of “being” as actual existence. Conceived of absolutely, being does not come
in degrees: it is rather something that all of the objects in the actual world simply
possess and must possess if they are to exist. The second way of conceiving of
being, as existence in a certain respect, is less familiar to present-day sensibilities.
This second sense does not consist in the mere instantiation of a substantial form
but in the degree of actualization of that form.8 So, if one of two birds more fully
fulfills its potential as a bird than the other does—as might happen if, say, the
first bird flies very well but the second has a broken wing and hence only has an
unfulfilled potential for flying—the bird that more fully fulfills its potential and
hence actualizes its form to a higher degree has more being, conceived of in the
second way, than does the injured bird.9
5
Eleonore Stump and Norman Kretzmann, “Being and Goodness,” in Thomas V.
Morris (ed.), Divine and Human Action: Essays in the Metaphysics of Theism (New York,
1988), p. 282.
6
Ibid., pp. 282–3.
7
Hence Peter van Inwagen’s insistence on the univocity of “being” and the equivalence
of “being” and “existence” in “Meta-Ontology,” Erkenntnis, 48/2 (1998): 235–7.
8
Stump and Kretzmann, “Being and Goodness,” pp. 283–4, 288–9.
9
Compare the absolute existence/existence in a certain respect distinction with a
distinction that Derek Parfit makes between two ways of conceiving of the relation, “is
a relative of” in “Later Selves and Moral Principles,” in A. Motefiore (ed.), Philosophy
and Personal Relations (London, 1973), reprinted in Steven M. Cahn and Joram G. Haber
(eds.), 20th Century Ethical Theory (Upper Saddle River, 1995), p. 477. Parfit observes:
“We can use the phrase ‘related to’ so that what it means has no degrees; on this use, parents
and remote cousins are as much relatives. It is obvious, though, that kinship has degrees.
This is shown in the phrase ‘closely related to’: remote cousins are, as relatives, less close.
I shall summarize such remarks in the following way. On the above use, the fact of being
someone’s relative has in its logic no degrees. But in its nature—what it involves—it does
have degrees. So the fact’s logic hides its nature. Hence the triviality of the claim that all
our relatives are equally our relatives.” The upshot of the absolute existence/existence in a
This elaboration on the two ways of conceiving of being helps us to make

sense of the claim that being and goodness are (materially) equivalent. If we are
conceiving of being in the second way, as degree of actualization of form, then it
is easy to see how, for Aquinas, “being” is materially equivalent to “goodness.” A
thing’s perfection consists in its fulfillment of its form: the more fully something
actualizes its form, the better that thing is. So, if we conceive of “being” as degree
of actualization of form, the more being a thing has, the more goodness it has,
and vice versa. “Being” and “goodness” will thus have the same referent: degree
of actualization of substantial form.10 Notably, though, the fact that “being” and
“goodness” are equivalent in this way does not imply that the natural consequence
of evil is non-existence. It merely implies that those who are evil do not have full
being, that they suffer from some sort of privation and are not fully realizing their
potential. As one becomes more and more evil, the lack becomes more and more
acute, but (crucially) even though the slide from good to evil results in a decrease
in one’s being in a certain respect, we have no reason to suppose that it likewise
results in an eradication of the instantiation of one’s substantial form, which is
what is required for being conceived of as absolute existence to be lost. Indeed, we
know from observation that people (and other creatures) frequently actualize their
substantial forms to lesser and lesser degrees without ever ceasing to exist.11 The
privation argument for annihilation thus fails because, even granting the long-held
view that “being” and “goodness” are materially equivalent, we lack sufficient
reason for thinking that being evil, i.e., failing to live up to one’s substantial
form/suffering from privations, naturally leads to non-existence, the complete
eradication of one’s substantial form, as opposed to a (perhaps ever-increasing)
loss in the extent to which one actualizes it.12
certain respect distinction may be that, for Aquinas, the fact of something’s having being
has no degrees in its logic but does have degrees in its nature.
10
Stump and Kretzmann, “Being and Goodness,” pp. 282–6.
11
Sometimes creatures even preserve their absolute existence because they fail to
live up to their forms. The freeloading wolf that never joins the pack in the hunt fails to
fulfill its function in an important way, but doing so may allow it to survive by helping it
to avoid some fatal situations. See Rosalind Hursthouse, On Virtue Ethics (Oxford, 1999),
p. 196. Conversely, a number of flourishing creatures lose their existence without ever first
suffering a decline in being in the second sense. Thus, when a nuclear bomb strikes an area
and all of the birds in that area die, even the most flourishing of the birds—i.e., even the
birds that have the most being in the second sense—do not first lose some of their being (in
the second sense) and then cease to exist. In the case of a nuclear bomb, having so much
being does not help to save the flourishing birds; nor does lacking so much being hasten the
deaths of the less-flourishing birds.
12
Eleonore Stump gives a reason for thinking that the view that evil is a privation
actually detracts from the plausibility of annihilationism: if being and goodness are
materially equivalent, then annihilating a creature, which requires “eradicat[ing] being”
is a “prima facie evil, which an essentially good God could not do unless there were an
We can find a different argument for the belief that annihilation (non-existence)
is a natural consequence of sin by drawing on Paul Griffiths’s work on the
particular ways in which sin corrupts humans.13 Notably, Griffiths does not insist
that sin leads to non-existence; he merely says that the idea that it does so lead
is worth entertaining.14 Following Augustine, Griffiths holds that because we are
created in the image of God, we are essentially beings who reflect that image by
having certain capacities—namely, volitional and cognitive capacities. Griffiths
also holds that sin corrupts these capacities to such an extent that it is worth asking
whether some humans will, in sinning, so corrupt themselves that they lose these
capacities altogether. Because human beings have these capacities essentially, to
lose them is to cease to exist.15 In such a situation, whatever traces of the (former)
humans might remain—perhaps a grumble, a rant, or what Griffiths calls “psychic
detritus”—would be just that: traces.16
Despite having some appealing points, this argument, which we will call the
“corruption argument,” faces problems if it is thought to provide motivation for
the annihilationist picture. Few, if any, Christians will deny that we were created in
the image of God or that sin deeply corrupts us even to the point of doing serious
damage to our will and cognitive capacities. But drawing the annihilationist
conclusion from this is much more controversial.
In the first place, despite the fact that most Christians have reason to believe that
sin can significantly damage the sinner’s volitional and cognitive abilities, they lack
reason to believe that sin destroys those abilities completely, as it seems that it must
if a person is completely to cease to bear the image of God. Regarding the will, sin
would have to be capable of damaging the will so severely that the person would be
incapable of willing ever again. Such a person’s volitional state, then, would have
to be worse than even the pre-salvation volitional state of Paul, the “foremost” of
sinners who, recall, continued to exist while he was in his sinful state.17
overriding good which justified it,” “Dante’s Hell, Aquinas’s Moral Theory, and the Love
of God,” Canadian Journal of Philosophy, 16/2 (1986): 196.
13
Griffiths, “Self-Annihilation or Damnation?” esp. pp. 89–90. Annihilationist Clark
Pinnock anticipates Griffiths’s general idea without ever developing the particulars of how
sin corrupts the individual to lead to extinction, calling hell, which Pinnock interprets as
annihilation, “a terrifying possibility, the possibility of using our freedom to lose God and
destroy ourselves,” “The Conditional View,” in William Crockett (ed.), Four Views on Hell
(Grand Rapids, 1992), p. 165.
14
Griffiths, “Self-Annihilation or Damnation?” p. 107.
15
Ibid., pp. 89–90.
16
Ibid., pp. 85, 93, and 116–17 n61, in which Griffiths recalls a passage from C.S.
Lewis’s The Great Divorce (New York, 1946), pp. 20–21.
17
See 1 Timothy 1:15 (RSV). Even if Paul’s claim in this passage is rhetorical, the
passage is often taken as implying that no one need fear of being so sinful as to be irredeemable.
If God can save Paul, “the chief of sinners,” the reasoning goes, then no sin is so great as to
preclude the sinner’s salvation. As long as this implication holds, the fact that Paul, under
the inspiration of the Holy Spirit, called himself the “foremost of sinners” supports the point
Now, many Christians will admit that sin could so damage a person’s will that,
on his own, he could will only evil, but that is very different from saying that such
damage renders him unable to will at all. And if he still has some sort of will, even
a corrupt one only capable of willing evil, then he still somewhat—albeit quite
weakly—reflects the image of God in his volitional capacity.
Of course, if sin damages the will to this extent, then it is only by grace that
humans can do good. Many Christians will readily admit this, indeed, insist upon
it, but this hardly commits them to the claim that anyone born with original sin
is not a real human or that Adam and Eve ceased to exist when their wills were
bound. The reason they are not so committed is that being capable of willing good
on one’s own without grace surely is not essential to human personhood.18
Let us turn now to our cognitive abilities, which present their own distinctive
challenge. Here, Christians typically agree that sin compromises our intellectual
capacities, blinding us from some truths and diminishing our ability to think
clearly. But it is difficult to see how sin should ever bring us to a point at which we
do not have any knowledge including, say, knowledge of incorrigible beliefs. We
lack an account of how, for instance, sin could render a creature unable to know
that she is in pain. Now, perhaps the appropriate response is that merely having
any old knowledge (knowledge of one’s own pain, for instance) is not sufficient
for reflecting the image of God: to reflect the image of God, one must have certain
kinds of knowledge—ethical or theological knowledge, perhaps. The problem with
this suggestion is that infants and the severely mentally disabled lack such higher
forms of knowledge. Thus, a more plausible suggestion is that, in order to reflect
the image of God, one must have a capacity for certain types of knowledge.
The capacity suggestion, however, is not without problems. First, some will
complain that the capacity distinction is artificial. The worry is not that there is
no distinction between, say, knowing something and being capable of knowing
it, or willing something and being capable of willing it; surely there is such a
distinction. Rather, the problem arises because the capacity suggestion is introduced
precisely for the purpose of allowing infants and the severely mentally disabled,
but not those who are sufficiently sin-damaged, to reflect the image of God. The
suggestion, then, says that infants and the mentally disabled have volitional and
cognitive capacities that the sufficiently sin-damaged lack. The obvious rejoinder
is to point out that whatever cognitive and volitional capacities infants and the
severely mentally disabled have are thin enough that the sin-damaged—even the
very-sin-damaged—have those capacities as well.
The capacity suggestion also faces a difficulty analogous to the one we saw
above in the discussion of volitional capacity. The suggestion works only if sin
made here that sin does not utterly destroy the person and all volitional capacity. If sin were
to so destroy volitional capacity, then sin could put some beyond the reaches of salvation.
18
Indeed, some will say that avoiding Pelagianism requires admitting that being
incapable of willing good on one’s own without grace is essential to non-divine human
personhood.
is capable of completely and irreparably destroying a creature’s capacity to know

particular sorts of truths. On such a view, sin is capable not only of leading us to
false beliefs, of making us bad thinkers, or even of damaging our cognitive abilities
to such an extent that we cannot know God or good without grace; sin must also
be capable of damaging us so severely that we lose our epistemic capacities for
the relevant knowledge.
Moreover, it is not clear that our capacities to will and to know ethical and
theological truths are essential to us. Plausibly, the self could, despite losing
those capacities, continue to exist. To see the plausibility of this claim, consider
a thought experiment. Suppose that Hitler is in a post-mortem place of suffering.
Sin has corrupted him severely, so that his volitional and cognitive capacities are
deeply damaged. Now suppose that he continues to follow evil. Suppose that, as a
result, his volitional and cognitive capacities become as severely and irreparably
damaged as possible. What remains is a creature that is unable to will or to know
crucial ethical and theological truths. Suppose, however, that this creature that
remains is suffering in the same manner that Hitler was. Suppose also that there is
an uninterrupted continuity of consciousness between Hitler and this creature. If
such were the case, we would be strongly inclined to say that Hitler has survived.19
19
In speaking of unity of consciousness, we are gesturing at what we take to be a
very intuitive belief—the belief that there is a close tie between personal survival and
consciousness or continuity of consciousness. Versions of this belief show up throughout
the history of philosophy. Thus, Augustine reasons that, so long as the mind has memory
and expectation, it must have life. “On the Immortality of the Soul,” in George G. Leckie
(trans.), Concerning the Teacher and On the Immortality of the Soul (New York, 1938), 3.3–
3.4. Elsewhere Augustine says, “[t]he soul is called immortal, then, because, at least to some
extent, it never ceases to live and feel.” See R. W. Dyson (trans.), The City of God Against
the Pagans (Cambridge, 1998), sec. 13.1 (emphasis ours). In Aquinas, Summa Theologica,
PP, Q. 75, A. 6, reply to objection 3 and PP, Q. 89, A. 1, reply to objection 3, we see the
idea that one’s soul, to exist, must always understand, whether by understanding corporeal
phantasms when united with the body or by (via divine aid) understanding some other
species (post-mortem and pre-resurrection), which is what would allow for consciousness.
In the modern period, Locke of course famously tied personal identity to continuity of
consciousness. More recently, philosophers—including Christian philosophers—have
supposed that personal identity consists in having the same first-person perspective.
Consider, for instance, Sydney Shoemaker, “The Unity of Consciousness and Consciousness
of Unity,” in Sydney Shoemaker (ed.), The First-Person Perspective and Other Essays
(New York, 1996), and Lynne Rudder Baker, Persons and Bodies: A Constitution View
(New York, 2000), esp. chapter 3. We should say that Baker’s understanding of the first-
person perspective differs from Shoemaker’s and our own inclinations in that, for Baker,
mere animals (dogs, for instance) do not have first-person perspectives.
Now, if a person maintains that a deep tie exists between personal survival and
continuity of consciousness, that claim certainly needs to be worked out carefully in order
to avoid objections. Continuity of consciousness, for instance, if it is said to be necessary
for personal identity—and we are not sure that it should be—cannot be seen to require
memory of one’s past, for people can certainly survive loss of memory. In addition, as many
And if Hitler would survive, then either reflecting the image of God is not essential
to us or sin is not capable of utterly destroying each of the ways in which we reflect
the image of God.
The corruption argument thus faces serious problems. It works only if sin is
capable of corrupting persons so much that their cognitive and volitional capacities
become irreparably damaged to an extreme degree. Furthermore, the argument
assumes that a human person would cease to exist were his volitional and cognitive
capacities to become so thoroughly damaged, and we have seen reason to think that
a person could survive the loss of the relevant capacities simply by maintaining
consciousness.
Interestingly, both the privation argument for annihilation and the corruption
argument face an additional problem. On both of these arguments, over time, a
person becomes increasingly evil until, eventually, having reached some threshold,
he ceases to exist. If this progression toward non-existence occurs in hell and
if hell is (at least partially) a (non-eternal) place of punishment, then we might
worry that some in hell, namely, those that are becoming evil at the fastest rates,
face better overall fates than those whose progression toward evil/annihilation is
slower. The slower one’s descent, the longer one will spend in hell. If one will
not go to heaven, then the way to make one’s stay in hell as short as possible is to
make oneself as depraved as possible as quickly as possible. We might question,
however, whether doing so should enable one to minimize one’s total stay in hell.
Motivations Rooted in God’s Moral Perfection
A second class of motivations for adopting an annihilationist view involves

the worry that God’s moral perfection either requires that the unrepentant face
annihilation or, at the least, makes annihilationism more plausible than DH,
its most prominent alternative. It is worth noting that the annihilationist who
takes the latter route and offers an argument not for annihilationism as such but
simply for annihilationism’s superiority over DH cannot rest assured of having
offered sufficient motivation for annihilationism unless she also has sufficient
reasons (whether based on the divine attributes or on other considerations) for
annihilationism’s being preferable to universalism. For now, we will bracket the
question of whether the annihilationist can so demonstrate the preferability of
annihilationism over universalism and investigate only whether considerations
of God’s moral attributes push the plausibility of annihilationism ahead of the
plausibility of DH.
have pointed out, problems arise if one takes continuity of consciousness to provide a non-
circular criterion of personal identity. Nonetheless, we think there is much intuitive support
for the idea that continuity of consciousness is (in fact) sufficient for personal identity. And
if unity of consciousness is sufficient for personal identity, then, in our example, Hitler
survives despite any damage sin exerts on his cognitive and volitional capacities.
An initial attraction of the annihilationist’s position is that it avoids a problem

that DH is said to face with regard to God’s justice. The problem, the thought
goes, arises because DH depicts God as condemning people to something that they
cannot possibly deserve: everlasting misery and punishment. Such condemnation
is thus incompatible with the just nature of a perfectly good God. Annihilationism
is an attractive alternative because it seems to avoid the obvious problems posed
by DH for God’s perfect goodness since it denies that God condemns anyone to
everlasting misery.
Upon closer examination, however, it is much more doubtful that annihilationism
is the solution that it appears to be. We can see this by getting clearer on why DH
is thought to be in conflict with God’s justice. The first (and most popular) reason
for thinking this is that humans are finite creatures and hence incapable of doing
anything deserving of an infinite punishment. But hell, qua everlasting, constitutes
an infinite punishment. So, no human could possibly deserve hell. The second
reason commonly given for thinking that no human deserves hell has to do not
with hell’s duration but with its nature. No matter how evil a person might be, the
worry goes, no one could possibly deserve an existence as miserable and tortured
as hell is said to be.
Responses to these two objections are well known, and we will not here
detail them all.20 What most interests us is that, whether or not these two justice
objections are successful against DH, annihilationism fares no better against
similar considerations than do at least some of the available versions of DH
itself. To see this, recall that the source of the descriptions of the damned facing
a punishment that is everlasting (or eternal) is scripture itself, the authority of
which is common ground for both sides.21 For those who accept DH, the meaning
of these passages is straightforward and obvious, namely, that some people will
face an everlasting punishment, and that everlasting punishment is life in hell. The
annihilationist interpretation of such passages is less obvious: that interpretation
states that the passages in question teach that some people will indeed face an
everlasting punishment, but that everlasting punishment is, in fact, annihilation.
The punishment of annihilation counts as everlasting not because the recipients
will consciously experience it forever but because its consequences, while not
20
A quick look at the most common responses reveals that the following strategies
are prevalent: (1) attempting to show that hell, while everlasting, need not be a place of
everlasting punishment; (2) attempting to show that hell is a place of everlasting punishment
but not a place of infinite punishment; or (3) attempting to show that humans are capable
of deserving eternal or infinite punishment. See Richard Swinburne, Responsibility and
Atonement (Oxford, 1989), p. 182; James Cain, “On the Problem of Hell,” Religious Studies,
38/3 (2002): 355–62; and especially Charles Seymour, A Theodicy of Hell (Dordrecht,
2000), pp. 37–94.
21
See Matthew 25:46 and 2 Thessalonians 1:9.
experienced, nonetheless last forever.22 Crucially, then, annihilationists classify

(correctly or not) annihilation as everlasting punishment. So, annihaltionists,
no less than defenders of DH, hold that some humans will face everlasting
punishment, a punishment that is all the more poignant since it involves the loss of
the infinite good of eternal life with God. Consequently they are saddled with their
own version of the problem that defenders of DH face with respect to the objection
that humans, qua finite, could never deserve punishment that, qua everlasting, is
infinite. Once this is recognized, part of the initial attraction of annihilationism
evaporates into thin air.23
Similar reasoning comes into play when assessing the issue of whether
annihilationism dodges the objection (which DH is said to face) that the nature
of a lived hell is such that no human being, however depraved, could deserve it.
In contemporary discussions, defenders of DH have become increasingly inclined
to eschew traditional descriptions of hell-as-torture-chamber (literal fire and
22
See, for instance, what annihilationist Edward Fudge says in his and Robert
Peterson’s Two Views of Hell: A Biblical and Theological Dialogue (Downers Grove,
2000), pp. 59–60. In discussing the Thessalonians passage, Fudge uses the translation
“eternal” rather than “everlasting,” which allows him to suggest an additional way in which
the punishment of annihilation counts as eternal: “it belongs to the age to come and not to
the present order of created space and time.” Of course, this suggestion cannot stand on its
own (and Fudge does not say that it does): if it stood on its own, to say that God is eternal
would only imply that God does not belong to the present order of space and time, and
presumably saying that God is eternal implies more than that.
23
Interestingly, annihilationism may actually be in worse shape with regard to the issue
of “everlasting punishment” than is DH, and that in two respects. First, defenders of DH
who take scripture as claiming that the wicked face unending punishment can say that this
punishment is deserved because the sin of the wicked is unending: while in hell the unjust
continually sin, at the least by maintaining an unrepentant and defiant attitude. It is only on
annihilationism that the sins of the wicked might be finite and hence fail to merit everlasting
punishment. This problem is related to the “continuing sin” response to the problem of
hell’s justice. For more on that topic see Seymour, A Theodicy of Hell, pp. 81, 84, 161–2;
Michael Murray, “Heaven and Hell,” in Michael Murray (ed.), Reason for the Hope Within
(Grand Rapids, 1999), pp. 293 and 296; and Kenneth Himma, “Eternally Incorrigible: The
Continuing-Sin Response to the Proportionality Problem of Hell,” Religious Studies, 39/1
(2003): 61–78.
A second potential problem for annihilationism is related. It is one that Anselm
raises, and it arises even if one does not take scripture to claim that some people will face
everlasting punishment. If God were to annihilate the unrepentant, then these people, who
will cease to exist while maintaining a contemptuous attitude toward God, will face the
same fate as do uncreated souls who are never born. So, says Anselm, “the guiltiest soul
would be in the same state as the most guiltless,” which may constitute a violation of
justice. Anselm, The Major Works, eds., Brian Davies and G.R. Evans (New York, 1998),
p. 76. Seymour discusses Anselm’s position in A Theodicy of Hell, pp. 183–4. Although we
hesitate to endorse Anselm’s argument, here, we do hold that Anselm’s point is one worth
considering.
brimstone optional) in favor of more moderate views of the nature of hell. These
views may or may not depict hell as involving positive punishment (as opposed
to, say, the punishment of separation from the divine life of the Trinity or the
self-inflicted punishment that consists in being a person who has certain character
defects). On some views, hell is simply the natural outcome of the facts that God
takes our choices seriously, that some people continually reject God, and that God
will never stop loving even those people by ceasing to offer himself to them. As
such, hell is unending for those who continually reject God, but no more can be
said about its nature than that those in hell are, in some important way, missing out
on God. Other so-called “mild” and “moderate” views of hell are more specific
about hell’s nature, but even these more specific views are not in agreement on
whether or not, say, all of those in hell experience at least some physical pains.
On some of the moderate and mild views, the damned in hell will even experience
some (rather twisted) pleasures.24
Mild and moderate views of hell are significant for our purposes because, by
providing alternatives to stronger views of hell, they deprive annihilationism of the
claim to being the only legitimate afterlife option whose very nature is consistent
with what humans, even the most wicked ones, could deserve.25 Mild hell views
show us that DH does not, as such, entail that the damned face an eternal torture
chamber. If one wants to support the annihilationist view by claiming that humans
could deserve annihilation but not hell, one needs to compare annihilation not
only to torture-chamber hells but also to mild and moderate hells. If any mild
or moderate hells are not obviously worse fates for the wicked than the fate
that the annihilationist defends—and we think, at the very least, that we cannot
know that the wicked would fare worse on the mildest views than they would
on annihilation26—then annihilationism is not the only game in town when the
question of a fate that a human could deserve arises.
One might here object that mild hell views are too mild, but to such an
objection the defender of DH is not without a response. He can point out that either
annihilation is likewise too mild or annihilation is not too mild. In the case of the
former, annihilationism faces its own problem. In the case of the latter, given that
24
They might, for instance, mimic Mr. Crawley, a character in Anthony Trollope’s
The Last Chronicle of Barset, who, finding himself in a decidedly difficult situation, resists
help from those who offer it and instead takes positive pleasure in his situation by brooding
over the many ways in which he perceives himself to be ill-used. He particularly enjoys this
exercise in self-pity because it allows him to bask in the idea that he is a martyr virtuously
bearing his cross while the many whom he holds in contempt idly sit by.
25
Recall that, in this section, we are excluding universalism as an option.
26
On this point, we should keep in mind that humans are traditionally poor judges
both of what will make them happy and of what will make them miserable. Thus, people
routinely assume that wealth brings happiness and disability (blindness, for instance) brings
misery when interviews with the wealthy, the non-wealthy, the able-bodied, and those with
disabilities often suggest otherwise.
some views of hell are, as best as we can tell, as mild as annihilation, a fate can be
as severe as continued existence in (an at least mild) hell without raising a justice
problem. So, at least some (mild) views of hell are available that are not so severe
as to fall prey to the justice problem. Thus, annihilationism is not the only way out
of the justice problem, and the justice problem does not sufficiently motivate it.
Jonathan Kvanvig has offered different reasons for thinking that God’s moral
perfection requires the annihilation of the wicked in at least some cases. For
Kvanvig, the key is not that no human could deserve an eternal lived hell. Rather,
the key is that at least some of the people who find themselves in hell are likely
to attempt metaphysical suicide. When these suicide attempts are rational—and
Kvanvig argues that there is no in principle reason why at least some of them could
not be—God’s moral perfection requires that he respect the choices of the wicked.
Not so to respect their choices would be paternalistic on God’s part.
Although Kvanvig is unwilling to draw the conclusion that some people
will definitively face annihilation, we can draw on Kvanvig’s work to formulate
an argument on behalf of the annihilationist for that conclusion. We will call
this argument “the rational suicide argument for annihilationism,” and it is as
follows:
K1. Some of the wicked will attempt post-mortem suicide.

K2. In some of the instances in which the wicked so attempt post-mortem
suicide, they will be acting rationally.
K3. If someone attempts post-mortem suicide and is acting rationally, it would
be paternalistic for anyone, including God, to interfere.
K4. God would never act in a paternalistic manner.
K5. Hence, God would never attempt to interfere with someone attempting
post-mortem suicide if that person were acting rationally. (From K3 and
K4.)
K6. Hence, God will not interfere with some people who will attempt post-
mortem suicide. (From K2 and K5.)
K7. If God does not interfere with a person’s attempted post-mortem suicide,
that person will be annihilated.
K8. Hence, some people will be annihilated. (From K6 and K7.)27
The conclusion that (at least) some people will be annihilated follows given K1,
K2, K3, K4, and K7.28 While we wish to flag the fact that “paternalistic” is a thick
(some would say loaded) ethical term that implies impropriety, we readily grant
that K4 is true: God is the creator of the cosmos, has a plan for the world, and acts in
providential ways, but never does God’s providential activity fall under the rubric
27
See Kvanvig, The Problem of Hell, pp. 139–48.
28
Note that while the rational suicide argument is, as we present it, an argument for
annihilationism, it is not an argument for the claim that all of those who never receive
salvation will eventually be annihilated, which is what most annihilationists hold.
of paternalism. We have some reservations about K7, but we are willing to table
those reservations for the present. The premises that we find most problematic and
on which we would like to focus are K1, K2, and K3.
At first glance, K1 is an obvious judgment of common sense reflecting on
human nature and what life post-mortem sans divine grace would be like. Every
year, thousands of people in this world commit suicide. If significant numbers of
people in this world are that quick to opt for non-existence, it stands to reason that
many in the next will, if deprived of God, opt for the same, given sufficient time.
Kvanvig offers an additional reason for thinking that some of the wicked will
desire non-existence: their hearts will be hardened, and they will want to reject
God forever, even if doing so costs them existence. Since everything that exists
depends on God for its existence, those who thus thoroughly reject God come
what may are choosing annihilation.29
Despite the plausibility of these claims, however, neither the fact that many
people commit suicide while on earth nor the fact, if it is a fact, that some of
the damned will likely take themselves to prefer non-existence to existence in a
hell-like state30 shows that K1 is true. K1 speaks not of what sorts of situations
the damned will prefer but of how they will act. But we simply do not know
what sorts of actions are open to people in the next life. People in this world
who commit suicide are able to do so due in part to their access to items such as
handguns, sleeping pills, and tall bridges. We may well wonder whether there will
be afterlife analogues to our instruments of death that will be effective when used
on the resurrected bodies of the unjust. If the unjust do not have access to such
instruments, it would be surprising if K1 were true. Many of the unsaved may well
wish to be annihilated, but if they have no recourse to methods that are perceived
to result in non-existence, they will be unlikely to attempt to annihilate themselves,
just as even the most adventurous in this world are unlikely to attempt to jump over
the planet Jupiter, given that they see no way to do so.31 So, without an argument
29
Kvanvig, The Problem of Hell, pp. 146–7.
30
Walls disputes this point in Hell: The Logic of Damnation (Notre Dame, 1992),
p. 137. He contends that the damned may be so attached to their sin and the twisted pleasures
of hell that they prefer continued existence in hell to non-existence.
31
We could replace K1 with K1*: “Some of the wicked will want to annihilate
themselves.” To make the argument valid, though, we would then have to replace other
premises accordingly. Perhaps most significantly, we would have to say that God would
be paternalistic were he to fail to grant rational desires for annihilation. Such a claim is
implausible. There is nothing paternalistic about a doctor’s refusing to kill someone who
wants to be killed, and there is likewise nothing paternalistic about God’s failing to grant
a person’s desire for her own annihilation, no matter how rational such a desire may
be (assuming that desires are the sorts of things capable of manifesting the property of
rationality). No doubt some of the people who commit suicide in this life desire to cease to
exist (as do some who do not, for whatever reason, attempt suicide). But even if we grant
that some such people are rational, it does not follow that God is paternalistic in not catering
to their fancies and annihilating them.
for the claim that those in hell will have access to methods of annihilation—and
it is hard to see how such an argument would go—the rational suicide argument
fails from its first premise.
Now, Kvanvig does suggest a response. To Kvanvig, because God’s sustaining
activity is required for existence, to choose to reject God is de facto to choose
non-existence, i.e., to attempt post-mortem suicide. Or, at the least, for those who
recognize that all creatures are sustained by God, choosing to reject God is choosing
non-existence.32 The annihilationist could follow Kvanvig’s lead and say that God’s
allowing people in hell a way of committing post-mortem suicide is not a matter of
his providing them with the afterlife equivalent of our handgun; it is simply a matter
of him allowing them to reject him forever. K1 is thus true, the annihilationist may
say, if God gives some of the wicked the opportunity to reject him, and at least
some of them, knowing that they cannot exist without God, do so.
The problem with this way of supporting K1 is that we still lack reason for
thinking that God gives the wicked the opportunity to reject him in the way that
the reply specifies. The reply assumes that God, by virtue of allowing his creatures
to reject a loving relationship with himself, commits himself to ceasing to sustain
those creatures in existence if they come to hate him so much that they do not
even want him to sustain them in existence. But God’s allowing people to reject
his love by no means commits him to satisfying their every preference, including
whims for their own non-existence. In allowing people to reject him, God, to be
sure, grants them a remarkable degree of control over the kind and quality of
relationship they bear to him.
But it is another matter altogether to think he should give his creatures the
prerogative to decide whether they shall be related to him at all. God settled that
matter when he chose to create us in his image. God did not need our permission (in
any sense of “need,” including moral senses) to create us, and he does not need our
permission to sustain us in existence. The “freedom” to decide whether or not we
exist is simply not a freedom that we have any good reason for believing God ever
does or will offer anyone. It is not, after all, as if, say, a genuine love relationship
with God is possible only if he allows us to cease to exist at will. Moreover, we do
have a reason that God might not want to allow us such a freedom: to ensure one’s
own non-existence is to ensure that one will never again have any other freedoms.
For the sake of the first-order freedoms, God may not want to allow us the second-
order freedoms.
We can put the above point in another way: according to our Kvanvig-inspired
defense of K1, some people in the afterlife will effectively say to God, “I’d rather
be dead than serve you.” If saying this were, ipso facto, to constitute attempting
post-mortem suicide, then the case for K1 would be good. But attempting post-
mortem suicide, like attempting ordinary suicide, involves more than having
certain desires (e.g., the desire to cease existing). And we have no strong reason
32
Kvanvig, The Problem of Hell, pp. 146–7.
to think God would grant such a wish or set things up in such a way that merely
having such a wish guarantees annihilation.
Moving on from K1, the rational suicide argument encounters a further
difficulty as soon as we consider K2, which endorses the idea that post-mortem
suicide can, in some cases, be rational. Kvanvig supports K2 first by drawing an
analogy to some cases of suicide in this life. While not all suicide cases are rational,
he says, some are. In particular, in cases in which “a person faces a significantly
painful and protracted end to his or her life,” suicide can be rational.33 If such cases
of suicide are rational in this life, Kvanvig reasons that certain situations in the
afterlife can make (post-mortem) suicide (i.e., self-annihilation) rational.34 These
certain situations are ones in which the agent rationally views herself as having
no tolerable alternatives to suicide. To such an agent, the prospect of continued
existence is “insufferable.”35
What impresses us about Kvanvig’s contention about what a case of rational
suicide might look like is that, as Kvanvig himself admits, so much depends upon
what one means by “rational.” On some conceptions of hell, it is quite obvious
that no person will ever be in a post-mortem situation in which suicide is in fact
that person’s only tolerable option. Specifically, on so-called second-chance
conceptions of hell, the wicked always have the option of repenting and turning
to God. If they do so, they will find a life that is not only tolerable; it is supremely
blessed! According to Kvanvig’s account of rationality, though, suicide would still
be rational on second-chance views if a person in such a situation perceived suicide
to be his only tolerable option and if reflection would not change his perception.36
Whether his perception is the result of radically disordered affections and/or a lack
of knowledge of his options would not be relevant to the rationality of his action.
The notion of rationality with which Kvanvig is working, then, is one on which
rationality is determined by the stability of one’s perception of one’s options. But
one’s perception of one’s options may be stable even if shaped both by false beliefs
about which options are available and by ill-formed affections. On conceptions of
rationality that require either full information or well-formed affections, K2 is, if
second-chance views are allowed, false. Since we cannot rule out the possibility of
second-chance views at the outset (to do so would beg the question against certain
versions of DH), K2 is acceptable only if we are using the term “rational” to refer
to a weak sort of rationality like Kvanvig’s that is compatible with both ill-formed
affections and a lack of full information, including full information about one’s
available options.
We are willing to grant that one conception of rationality is this weak sort
that allows for the truth of K2. We must, however, keep in mind that whatever
notion of rationality is at work in K2 must likewise be at work in K3 if the rational
33
Ibid., p. 140.
34
Ibid., pp. 141–7.
35
Ibid., pp. 140–41.
36
Ibid., p. 143.
suicide argument is to be valid. Recall that K3 says that anyone interfering with
rational post-mortem suicide is acting paternalistically, even if God is the one
doing the interfering. Significantly, K3 is implausible if we import the weak notion
of rationality that we need to make K2 acceptable. In this life, if a person desires
things that are deeply antagonistic to her best interests, and if these desires lead her
wrongly to perceive upon reflection her own continued existence as intolerable,
there is nothing wrong with preventing her from committing suicide and attempting
to change her disordered affections. Indeed, if a clinically depressed person thinks
about her situation, despairs of her future, continues to despair even after reflection,
and attempts suicide, we plausibly have an obligation to interfere with the attempt.
By analogy, at the very least, God would not be acting paternalistically if he were
to prevent a person from committing post-mortem suicide if that person were
exercising only the weak sense of rationality at work in K2 if K2 is true and those
in hell always have the option of submitting to God.
Motivations Rooted in Divine Supremacy
The final motivation for annihilation that we will consider is the view that
annihilationism is required if God (perhaps in the person of Christ) is to exercise
final authority.37 Like the motivation rooted in God’s justice, this motivation for
annihilationism works indirectly, by raising problems for DH that annihilationism
does not seem to face. Thus, Clark Pinnock argues for his annihilationist view
by insisting that views on which hell is unending imply a metaphysically and
theologically problematic cosmic dualism. Without annihilation, says Pinnock,
“the disloyal opposition [of the unsaved] would eternally exist alongside God in
a corner of unredeemed reality in the new creation.”38 Pinnock thus assumes that
defenders of DH are right to say that some people will never submit to God’s
authority, but Pinnock adds that this fact implies annihilationism: the only way
for God to reign finally supreme, given that some people will never accept his
authority, is for those people to cease to exist so that, at some point, no one will be
in rebellion against God.39
37
Clark Pinnock, “The Destruction of the Finally Impenitent,” Criswell Theological
Review, 4/2 (1990): 243–59. See also Edward Fudge in Fudge and Peterson, Two Views of
Hell. John Wenham uses the language of the “final supremacy of Christ” to make this point
in “The Case for Conditional Immortality,” in Nigel M. de S. Cameron (ed.), Universalism
and the Doctrine of Hell (Grand Rapids, 1992), p. 189.
38
Pinnock, “The Destruction of the Finally Impenitent,” p. 255. Pinnock makes the
same point in “The Conditional View,” pp. 154–5.
39
If the issue is not supremacy as such but the Biblical picture of the form such
supremacy will take, Pinnock does little to show how a problem arises for DH given the
biblical picture. He mentions two biblical passages concerning the end times: 1 Corinthians
15:28, which says that God will be “all in all” and Revelation 21:5, which speaks of God’s
The weaknesses of the divine supremacy motivation become apparent when

we inquire into what makes the rebellion of the unjust problematic. Pinnock never
explains this point, and looking carefully at what is involved when a person rebels
against God does not help his cause. First, we should note that a person can rebel
against God while recognizing God’s authority. Such a person will recognize
divine authority but resent it.40 Second, if we then look at why rebellion against
God is a bad thing, we will notice that it is bad because (among other reasons) it
is ungrateful, it hurts the agent, and it involves severely disordered affections. Yet,
for all of that, rebellion as such poses no worrisome challenge to God’s authority.
Neither the fact that a particular person resents God’s authority nor the fact that
that person will never cease so to resent God’s authority reduces the power that
God has over that person, or the actual and legitimate grounds on which that
person’s creator may claim authority over that person. To put the point somewhat
differently, if the rebellious in hell continually resent God and reject his love, it
does not follow that they have defeated either God or his love: they have simply
rejected his love in a way that allows God to demonstrate his unending mercy in
continuing to offer love to them.
At this point, the reader may have noticed a striking similarity between
annihilationist claims about the requirements for divine supremacy and Calvinist
claims about limited atonement. Both see something problematic in the idea of a
God who lavishes love and grace on those who ultimately reject it. For the Calvinist,
God’s love is wasted if rejected, whereas for the annihilationist, God’s love (or,
worse, God himself) is defeated if rejected.41 To see the profound distortion in these
two pictures of rejection of God requires us to take a closer look at the Christian
story and what that story teaches us about God and his perfections. Notably, the
God of Christianity is decidedly lavish with love. Thus, the Psalmist muses:
When I look at thy heavens, the work of thy fingers,

the moon and the stars which thou hast established;
making “all things new,” “The Destruction of the Finally Impenitent,” p. 255. It is a very
strong claim—and one that Pinnock does not defend—that the existence of an everlasting,
non-empty hell is incompatible with God’s making all things new and being all in all.
Notably, the 1 Corinthians 15:28 discussion of God’s being all in all occurs within a
discussion about everything being in subjection under Christ, which appears compatible
with the existence of hell so long as Christ is in some way reigning over hell, which would
occur if the requirements of divine justice were being satisfied in hell even if those in hell
were to resent Christ’s reign.
40
Consider James 2:19: “You believe that God is one; you do well. Even the demons
believe—and shudder” (RSV), which suggests that demons recognize who God is but resent
it rather than rejoice in it.
41
It is worth noting that universalists advance a similar argument that God is defeated
if all are not saved. See Thomas Talbott, “Freedom, Damnation and the Power to Sin with
Impunity,” Religious Studies, 37/4 (2001): 432ff. For criticism of Talbott on this point, see Jerry
L. Walls, “A Hell of a Choice: Reply to Talbott,” Religious Studies, 40/2 (2004): 214–15.
what is man that thou art mindful of him,

and the son of man that thou dost care for him?42
And Paul tells us that the fact that Christ died for us while we were sinners teaches
us about the nature and extent of God’s love.43 The God whose very nature is to
pour out extravagant love on even the foremost of sinners is not a God who is in
any way threatened or diminished if that love is rejected. So, we should not worry
for God’s honor, authority, or supremacy if he continues to pour out love in this
way ad infinitum.
In any case, if the rebellion of the lost does challenge God’s supremacy, it is
a problem that annihilationism does not solve. If annihilationism is true, some
people will never repent and submit to God’s authority. According to the divine
supremacy motivation, this would be a problem for God’s authority if the people
in question were to continue to exist and rebel forever. But why should the mere
fact that the people in question cease to exist, without ever submitting to God,
solve this supposed problem for divine authority? If, instead of annihilating the
people in question, God freely continues to love them, sustain them in existence,
and even exercise judgment over them, how is God’s reign any less supreme
than it would be if the rebellious were to cease to rebel only by ceasing to exist?
When one thinks that, in order to reign supreme, God must annihilate some of his
creatures, one’s conception of supremacy has gone terribly wrong. To draw an
analogy, suggesting that God can get out of a challenge that the unrepentant pose
to divine authority by killing them off is like suggesting that a child can ensure
his own “supremacy” over a video game by unplugging the system when he finds
himself unable to master some of the levels.44 Unplugging the video game does
not solve any problem of supremacy that the game may have posed for the child,
and annihilating the wicked does not solve any problem of supremacy that the
unrepentant may pose for God.
42
Psalms 8:3–4 (RSV).
43
Romans 5:8 (RSV).
44
Of course, on our view, the analogy is not perfect. The video game, while
undefeated, does in some ways preclude the child’s supremacy over it; the same is not true
of the unrepentant with regard to God. Unlike the child, God’s goal need not be to “defeat”
the unrepentant by ensuring their love of him. In Hell: The Logic of Damnation, Walls says
“God’s perfect happiness, like his perfect goodness, does not depend in any way on upon
human choice. God views rejection with an attitude of regret because he wants human
beings to be happy for their own sake. But he does not need them to be happy for his sake”
(p. 109). We might add that God’s supremacy likewise does not depend upon human choice,
that he does not need humans to choose him for the sake of his own authority.
Closing Remarks
By now it should be clear that the major philosophical arguments for annihilationism
do not begin to carry sufficient conviction to motivate adopting that position. Given
the fact that annihilationism is a distinct minority position, as well as something
of a novelty in theology, its proponents bear a heavy burden of proof. Having
formidable philosophical arguments to buttress their minority report in biblical
exegesis would go a significant way in reinforcing that exegesis. Unfortunately, it
appears that such philosophical arguments are yet to be found.
Annihilationism may seem to avoid problems that DH faces with regard to
God’s moral perfection and the final supremacy of Christ, but if we recognize
the wide variety of views on hell that allow one to endorse DH, it becomes clear
that annihilationism is by no means the only way to resolve the problems in
question. For any of the problems with respect to divine moral perfection or the
final supremacy of Christ that have been raised thus far, there are versions of DH
available that are equally satisfactory in addressing them, if not better, than the
annihilationist alternative. So, neither motivations rooted in God’s moral perfection
nor ones rooted in divine supremacy sufficiently motivate annihilationism. The
only other philosophical supports for annihilationism of which we are aware are
what we have called natural consequence motivations, which hold that the natural
consequences of sin are so destructive that sin in the long run that it eventually
results in annihilation. As we have shown, though, these natural consequence
motivations fail. The doctrine that sin is a privation, properly understood, does not
even suggest that evil is capable of annihilating a person, and we lack reason to
believe it ever fully eliminates even what is essential to the agent. Annihilationism
is thus philosophically unmotivated and unless powerful new arguments emerge,
or annihilationists can convincingly show that the majority report in biblical
interpretation has been misguided, their position is at a dead end.
Chapter 4
Compatibilism, “Wantons,” and the Natural
Consequence Model of Hell
Justin D. Barnard
Hell is neither so certain nor so hot as it used to be.

(Bertrand Russell)
Engaging in informed speculation about the nature of hell is rather like trying to
characterize the experience of being sucked through a wormhole. Both tasks are
plagued by two important difficulties. First, and perhaps most importantly, just as
we have no first-hand experience of being pulled through a wormhole, no one (at
least of whom we can be certain) has ever been to hell and back and lived to tell
about it. Quite simply, there is a shortage of eyewitness accounts through which
our speculation might be informed. Yet even if we managed to overcome this
somewhat daunting obstacle, the further difficulty remains that our conception of
the nature of hell, like our conception of traveling through wormholes, is perhaps
irreparably tainted by the forces of popular imagination. Just as what we believe
about wormholes will forever be influenced by what we have seen on Star Trek,
so also what we believe about the nature of hell is inseparably linked to such
archetypal sources as Dante’s Inferno or Jonathan Edwards’s “Sinners in the Hands
of an Angry God.” Consider just one of many representative samples of Edwards’s
depiction of the anguish that awaits unrepentant sinners.
Let us consider how great misery would it be to be always burning and roasting
in a fire and yet never be able to die, yea, to have the senses preserved in every
part of the body in their usual quickness, the feeling not to be at all dulled by
the fire: this would be a vastly greater torment than a man to [be] burnt alive,
because the fire presently sears his flesh and weakens the life and dulls the sense;
but in hell they shall have sense in an exquisite degree, and they shall have it to
no other end but to bear torment.1
Undoubtedly, Edwards viewed his remarks as offering an occasion for repentance

against the backdrop of theological precision. Yet the ubiquity of such imagery in
public discourse since the Great Awakening has resulted in hell’s being a subject
for comedy as much as an occasion for fear. One notable example of the former
1
Jonathan Edwards, “The Torments of Hell Are Exceeding Great,” in Kenneth P.
Minkema (ed.), The Works of Jonathan Edwards: Sermons and Discourses 1723–1729 (26
vols, New Haven, 1997), vol. 14, p. 311.
comes in a memorable scene from the sitcom Seinfeld in an episode entitled “The
Burning.” In it, Elaine tries to persuade her Christian boyfriend, David Puddy,
to steal the morning newspaper from the tenant across the hall. Filled with self-
righteousness, Puddy refuses, piously citing the eighth commandment and insisting
that Elaine steal it herself. After all, having already established her eternal destiny
earlier in the episode, Puddy blandly remarks to Elaine, “What do you care, you
know where you’re going.” At this juncture, Elaine reaches the boiling point in
her internal consternation over her impending date with, as Jerry had wryly put
it earlier in the episode, “the prince of darkness.” She lashes out at Puddy in a
moment of hilarious despair. “David, I’m going to hell! The worst place in the
world! With devils and those caves and the ragged clothing! And the heat! My
god, the heat! I mean, what do you think about all that?”2 David Puddy responds
with a characteristically droll remark filled with irony, “Gonna be rough.”3
While we may chuckle at Elaine’s uncharacteristic catharsis over the state of
her soul, there are those for whom the images that inform our imagination of
Divine punishment are no laughing matter. This is why, as I said at the outset,
informed discussions about the nature of hell are so challenging. On the one
hand, there are those who assert, with all seriousness, that the very idea of hell,
or least a flaming one, is fundamentally incompatible with a Divine Being whose
essence is love or justice. While on the other hand, the objection itself is rooted
in a particular conception of the nature of hell that has certainly been influenced
to one degree or another by a collective historical imagination about a theoretical
entity with which we have no first-hand experience. Thus, one can neither make
nor can one make go away the objections about hell’s incompatibility with Divine
goodness in the absence of a clear-headed account of what hell is really like. Yet
one cannot provide a clear-headed account of what hell is really like because our
current conception of hell is so heavily shaped by popular imagination and because
we lack an exhaustive and authoritative depiction of hell by which the errors of
popular imagination, if any, may be corrected.
To the extent that I have accurately characterized the dialectical dilemmas
posed by speculative discussions about the nature of hell, one would think it the
better part of wisdom to avoid serious inquiry in such matters altogether—heeding
the counsel that C.S. Lewis puts in the mouth of George MacDonald, who appears
as a character in The Great Divorce, Lewis’s fictional account of heaven and hell,
and sternly cautions, “Do not fash yourself with such questions.”4 But ironically,
despite the etymology of their titles, contemporary professional philosophers
(especially those of the analytic stripe) have long since given up on wisdom as a
desideratum. Consequently, the apparent foolishness of attempting to engage in
informed speculation about the nature of hell is of no concern to a professional
philosopher. After all, as long as the foolishness of the so-called “problem of hell”
2
“The Burning,” http://www.seinfeldscripts.com/TheBurning.html.
3
Ibid.
4
C.S. Lewis, The Great Divorce (New York, 1996), p. 69.
Compatibilism, “Wantons,” and the Natural Consequence Model of Hell 67
continues to captivate the moral imaginations of those who, like Elaine, can’t
take the heat, fools like me will continue to craft ever-increasingly, finer-spun
speculations that help get God out of the kitchen, so to speak. This chapter is my
contribution to the latter.
The Problem of Hell and the Natural Consequence Model
The problem of hell arises in connection with a traditional conception of the nature
of hell. In his book, The Problem of Hell, Jonathan Kvanvig characterizes this
traditional conception (what he calls “the strong view of hell”) by articulating
four theses to which the traditional view is purportedly committed. They are as
follows.
(H1) The Anti-Universalism Thesis: Some persons are consigned to hell.

(H2) The Existence Thesis: Hell is a place where people exist, if they are
consigned there.
(H3) The No Escape Thesis: There is no possibility of leaving hell and nothing
one can do, change, or become in order to get out of hell, once one is
consigned there.
(H4) The Retribution Thesis: The justification for hell is retributive in nature,
hell being constituted to mete out punishment to those whose earthly
lives and behavior warrant it.5
To the extent that these four theses jointly capture the essence of a traditional view
of hell, one version of the problem of hell, grossly oversimplified, goes something
like this.
1. Really nice beings do not consign people to places where they will receive
punishment (of a rather nasty sort) forever—without the possibility of
parole.
2. God is really (awfully!) nice.
3. Therefore either hell is not real or no one is actually consigned to hell, or at
the very least not forever (or just for good measure, there is no God).
At its heart, the problem of hell is an attempt to show that the traditional conception
of hell, as characterized by Kvanvig’s four theses, ultimately forms an inconsistent
set given certain plausible assumptions about the nature of God and how God’s
nature might constrain his actions. Thus, the traditional view of hell requires
modification or rejection of one or more of its key assumptions.
5
Jonathan Kvanvig, The Problem of Hell (Oxford, 1993), p. 25.
In an essay entitled “Heaven and Hell,”6 Michael Murray responds to this

problem by proposing a model of hell that essentially accepts Kvanvig’s four
theses but attempts to avoid the implication of the main argument by suggesting
that hell’s punishment is, in effect, self-inflicted. Somewhat crudely, those in hell
want to be there or, at the very least, lack the will to leave. Murray’s solution
borrows heavily from what he calls a “Natural Consequence” (NC) view of hell.
This view is perhaps most succinctly summarized in a famous line from C.S.
Lewis’s The Problem of Pain, “the doors of hell are locked on the inside.”7 Central
to the NC model of hell is a view of human nature and an eschatology that assumes
a disjunctive teleology. In the end, human beings either become maximally
confirmed in their orientation as lovers-of-self or maximally confirmed in their
orientation as lovers-of-God. The latter enjoy heaven; the former occupy hell.
The virtue of the NC model is, for Murray, simultaneously its principal vice
with respect to Murray’s overall response to the problem of hell. For while perhaps
absolving God of the responsibility of “consigning” people to hell, the NC model
of hell stands in tension with the retributive dimensions of hell as traditionally
conceived—a dimension that Murray accepts in defending what he calls the “Penalty
Model” of hell.8 Murray seems aware of the tension and attempts to alleviate it in
discussing the sense in which the NC model of hell constitutes “punishment.”
Still, Murray’s treatment of this tension does not adequately remove the worry that
the hybrid model of hell he defends is ultimately tenable. In this chapter, I hope to
make this worry explicit based on Harry Frankfurt’s compatibilist account of free
will and subsequently argue that Frankfurt’s concept of a “wanton” provides the
resources to respond to the concern, thereby defending Murray’s proposal.
Compatibilist Freedom and the Natural Consequence Model
As has been noted, one of the key virtues of the NC model of hell is that the
denizens of hell are there of their own accord, so to speak. Thus, God’s love or
justice is at least apparently absolved in light of the fact that he does not send or
consign anyone to hell against their will. Rather, some people choose hell by virtue
of choosing self-love over the love of God. The natural consequence of a lifetime
(or perhaps eternity) of self-love over the love of God is to become a creature who
6
Michael J. Murray, “Heaven and Hell,” in Michael J. Murray (ed.), Reason for the
Hope Within (Grand Rapids, 1999), pp. 287–317.
7
C.S. Lewis, The Problem of Pain (New York, 2001), p. 130. For an excellent recent
discussion of Lewis’s response to the problem of hell, see Matthew Lee, “To Reign in Hell
or to Serve in Heaven: C.S. Lewis on the Problem of Hell and Enjoyment of the Good,” in
David Baggett, Gary Habermas, and Jerry Walls (eds.), C.S. Lewis as Philosopher: Truth,
Goodness and Beauty (Downer’s Grove, 2008), pp. 159–74. For a critique of Lewis’s
account see Kvanvig, The Problem of Hell, pp. 120–23.
8
Murray, “Heaven and Hell,” pp. 291–4.
is maximally fixated in self-love. Thus, hell is chosen because God is ultimately,

finally, and irreversibly rejected.
As Murray correctly points out, one who adopts this view of hell must conceive
of hell’s punitive or retributive dimensions in slightly revised ways. For example,
the Bible’s portrayal of hell as being a place of “fire” and torment, of “weeping
and gnashing of teeth,” cannot literally be taken to imply that such suffering is
being meted out against the wills of hell’s residents. For if it were the case that
God was, in fact, imposing punishment in the form of various kinds of suffering
on those who were maximally confirmed in their orientation as self-lovers against
their wills, then the NC model would lose its appeal as a response to the problem
of hell.
Perhaps aware of this difficulty, Murray suggests that on the NC model, hell’s
punishment consists not in literal fire and brimstone imposed by Divine fiat, but
in “an agonizing and conscious awareness of loss”—a loss of the supreme good
(i.e., fellowship with God) for which human beings were made. “Thus,” Murray
explains, “a deep, eternal regret nags at the person who becomes a lover of self.”9
Murray wonders whether it is plausible to believe that those in hell can
simultaneously be conscious of the pain and regret associated with the loss of
such a surpassing good and “still never seek reconciliation with God.”10 Murray’s
response to this concern is worth quoting at length.
Since they are maximally disposed to be self-lovers, i.e., they have become set
in their ways, they might intellectually recognize how bad off they are in their
condition, but still not desire to change it. NC theorists often liken this state to
the state of an unwilling drug addict. The addict recognizes his ruined condition,
and wishes that he no longer wanted to take drugs. But nonetheless, he does
want to take them and thus continues to do so. Similarly, the one in hell, though
recognizing that he would be better off if he loved God, still refuses to do so. And
we need not resort to the drastic examples of the unwilling drug addict to illustrate
this phenomenon. People who are addicted to smoking, or who simply love foods
that are devastating to their health are not situated much differently. They see full
well that the behaviors they are engaging in are harmful and destructive for them.
They may even wish that they didn’t desire to engage in these behaviors. But
they nonetheless do desire to continue to engage in these behaviors.11
On the face of it, Murray’s response to his own query makes good sense. It is at
least conceptually coherent to imagine a person who (a) cognitively apprehends the
relatively undesirable nature of her present conditions, yet (b) lacks the capacity (i.e.,
will or desire) to change those conditions. However, Murray’s resolution here leaves
his view open to an important objection that undermines the plausibility of the NC
9
Ibid., p. 300.
10
Ibid.
11
Ibid.
model as a response to the problem of hell. In short, the NC model of hell retains
its plausibility only to the extent that those in hell are not there against their wills,
yet Murray’s description of the NC model as a form of eternal punishment makes
it appear that those in hell are not there willingly. To see why this is so, we need
briefly to consider an influential compatibilist account of free will articulated in Harry
Frankfurt’s seminal paper, “Freedom of the Will and the Concept of a Person.”12
In articulating the notion of freedom of the will, Frankfurt distinguishes between
first- and second-order desires. First-order desires are desires whose object is
some action. Thus, a first-order desire can be captured by sentences of the form,
“S desires to X” where X is some action. Second-order desires are desires whose
object is another desire. Beings with the capacity for second-order desires can
have desires about their first-order desires. For example, I might have a first-order
desire to eat a piece of chocolate cake, accompanied by a second-order desire that
my (first-order) desire to eat cake not be effective in compelling the fork toward
my mouth (perhaps because I am on a diet).
It is this distinction between first- and second-order desires that enables
Frankfurt both to define his concept of “the will” and to articulate what constitutes
“free will” or to distinguish those cases in which one has acted freely from those
in which one has not. Frankfurt defines the will as effective first-order desires, i.e.,
those desires that actually cause one to act or desires that are the efficient causes
of actions.13 Thus, whether the will is free, in Frankfurt’s view, is a function of
whether it aligns with what Frankfurt calls a second-order volition. A second-order
volition is a specific kind of second-order desire. Specifically, it is a desire to the
effect that a specific first-order desire be one’s will (i.e., be effective in causing one
to act). Importantly, it is the having of second-order volitions (not merely second-
order desires) that is a necessary condition for being a person and for freedom of
the will.14 In those cases in which a person’s second-order volitions are congruent
with his effective first-order desires, that individual may be said to have acted
freely. By contrast, a person is not free when his effective first-order desires are
inconsistent with his second-order volitions. In short, having free will amounts to
having the will one wants.15
When applied to Murray’s analysis of the condition of the maximally confirmed
self-lover in hell, Frankfurt’s compatibilist account of free will makes the problem
evident. Those in hell have first-order desires that are oriented toward self-love.
They want to love themselves. And they continue to choose (or to desire) to love self
over God. Still, the deep regret and pain that is associated with the overwhelming
loss of the supreme good for which they were made produce in them a second-
order volition that is not congruent with their effective first-order desire. They
12
Harry Frankfurt, “Freedom of the Will and the Concept of a Person,” The Journal
of Philosophy, 68/1 (1971): 5–20.
13
Ibid., p. 8.
14
Ibid., pp. 10–11.
15
Ibid., p. 14.
wish that they did love God. Or, perhaps more to the point, they wish (i.e., desire)
that their first-order desire to love self was not effective (i.e., that it was not their
will). Consequently, just like the unwilling addicts whose first-order desires to take
drugs do not align with their second-order volition (i.e., that their first-order desire
to take drugs not be effective), the residents of hell are not there freely since their
first-order desire to love self does not align with their second-order volition (i.e.,
that their first-order desire to love self not be effective). In short, as described by
Murray, those in hell—even on the NC model—are there against their wills (or at
the very least, they are not there of their own free will). Thus, the NC model loses
some measure of plausibility insofar as it appears that those who dwell in hell have
been consigned there against their will.
At this juncture it is important to delineate a few possible moves generally
available to the NC theorist, but unlikely to be embraced by those, like Murray,
who wish to retain the traditional punitive nature of hell. First, it is worth noting
that what generates this particular problem for Murray is a conception of hell as
a place of punishment—minimally, an undesirable state of affairs. For the NC
theorist, one way out of this predicament is to adopt a model of hell according
to which being maximally confirmed in one’s orientation toward self-love is not
an undesirable state of affairs. That is, it is not a state of affairs that would be
accompanied by an overwhelming sense of regret and pain for having eternally
lost the supreme good. The doors of this hell would be freely locked from the
inside because the residents there would have both a first-order desire to love
themselves along with a second-order volition that such a first-order desire be
effective in moving them to act accordingly. Note, however, that this move is
restricted to those willing to concede Kvanvig’s retribution thesis, an option not
open to Murray.
Alternatively, one might argue that while Murray’s proposal entails that those
in hell are ultimately there against their wills, their so being in that condition is a
direct (i.e., natural) consequence of a lifetime of decisions, choices, and actions
that were themselves free. In other words, a lifetime of decisions, choices, and
actions in which they were free to have the will that they wanted resulted in being
maximally “set in their ways” such that they are no longer free to have the will they
want. As with the previous move, this response to the criticism that I have raised
for Murray’s proposal makes good sense in the absence of a punitive conception
of hell. As long as the denizens of hell are not psychologically cognizant of the
overwhelming loss of the enjoyment of God’s love, it seems consistent with God’s
love and justice (assuming that freedom is a sufficiently valuable good) that God
would permit such persons to become maximally confirmed in self-love through
a lifetime of free decisions, choices, and actions, even if it resulted in the loss of
free will for that person. However, since the person we are here envisaging is not
cognizant of the fact that she has lost her freedom of will, she effectively has no
complaint against God’s love or justice. By contrast, Murray asks us to imagine a
person who has lost free will (granted through a series of free decisions, choices,
and actions), is aware of having lost free will, and wishes he or she could get it
back. This exposes Murray’s position to the typical charges advanced against all
views of hell that attempt to retain Kvanvig’s retribution thesis—an unsympathetic
gloss on which goes something like this: “If God were really as awfully nice (and
oh, by the way, super strong) as you say he is, then he would either stave off
the maximal confirmation in self-love (and hence, loss of free will) indefinitely
or he would reverse the natural consequences at the point where the maximally
confirmed self-lover became cognizant of the regret. After all, really nice people
do not leave poor souls helplessly floundering in eternal misery once those souls
have realized they have made a big mistake.”
For now, I will leave aside the question of whether Murray’s view has the
resources to respond to this kind of charge. For purposes of this chapter, I will
assume that it does not. In light of this assumption, it appears that Murray cannot
have his retributive cake and eat it with a natural consequences fork. However,
I hope to show that this gustatory quandary is merely apparent, as Frankfurt’s
compatibilist account of free will contains the resources to address the criticism
that I have raised.
Wantons as Occupants of Hell
Recall that on Frankfurt’s compatibilist account of free will it is the presence of

second-order volitions that constitutes not only a necessary condition for free
will, but for personhood as well. Frankfurt muses about the logical possibility
of creatures who have second-order desires while lacking second-order volitions.
“Such a creature, in my view,” Frankfurt explains, “would not be a person.”16
Frankfurt designates creatures “who have first-order desires but who are not
persons because, whether or not they have desires of the second order, they have
no second-order volitions” with the term “wantons.”17
The essential characteristic of a wanton is that he does not care about his will. His
desires move him to do certain things, without its being true of him either that he
wants to moved by those desires or that he prefers to be moved by other desires
… Nothing in the concept of a wanton implies that he cannot reason or that he
cannot deliberate concerning how to do what he wants to do. What distinguishes
the rational wanton from other rational agents is that he is not concerned with
the desirability of his desires themselves. He ignores the question of what his
will is to be. Not only does he pursue whatever course of action he is most
strongly inclined to pursue, but he does not care which of his inclinations is the
strongest.18
16
Ibid., p. 11.
17
Ibid.
18
Ibid.
In short, a wanton is a creature for whom the very question of free will does
not arise. This is because wantons lack the requisite type of second-order desires
altogether—namely, that some particular first-order desire of theirs (or another) be
effective in moving them to act.
My proposal is that Frankfurt’s concept of a wanton can be employed so as to
avoid the worry that I have raised for Murray’s hybrid view of hell—a view that
attempts to embrace the NC model while retaining the punitive nature of hell.
If the occupants of hell are wantons, in Frankfurt’s sense, then there is a sense
in which they are there because they have chosen to be. Having lived a lifetime
as self-lovers, making decisions, choices, and actions in which they were free to
have the will that they wanted, the wantons in hell have, as Murray puts it, become
maximally “set in their ways.” However, the sense in which they have become
maximally set in their ways does not entail that they are in hell, against their
wills. As wantons, these creatures still have first-order desires, and since they are
maximally confirmed lovers of self, they persist in their first-order desires to love
self for eternity. What they have lost are second-order volitions. They no longer
care about which of their first-order desires are effective. Thus, they are neither
in hell against their wills, nor are they there freely—at least not after the point
where their capacity for second-order volitions is irreparably lost. Still, because
they have not lost the capacity for reason, they are in an intellectual position to
grasp their status of having eternally lost the supreme good. One might further
hypothesize that such an apprehension carries with it the kind of deep regret and
pain to which Murray alludes. It is in this sense that hell is punitive.
Beyond this, we could even imagine that the denizens of hell experience the
deep regret and pain to a degree that it produces in them a first-order desire to
leave. However, because they are wantons, they lack a second-order volition to the
effect that this first-order desire to leave be effective in moving them away from
the love of self and toward the love of God.19 In short, they are ambivalent about
their first-order desire to leave. They have it; they recognize they have it; but they
do not care whether it moves them in one way or another.20
Still, the proposed amendment to Murray’s position is susceptible to at least
one important concern that needs to be addressed. Specifically, the suggestion that
the denizens of hell are wantons does not help in defending against the criticism I
have raised for Murray if people are made to become wantons against their wills at
death. In other words, the NC view which, under the revision I am proposing entails
an eventual transition from person to wanton, seems to require something like an,
in principle, postmortem opportunity to change. And perhaps more importantly, it
requires that the transition from person to wanton occur in keeping with the natural
19
I thank Garrett Pendergraft for pointing this out to me in conversation.
20
Note the difference between the psychological state of a wanton in hell and the
psychological state of the person in hell, as originally described by Murray. The latter
arguably has second-order volitions (i.e., desires about their first-order desires to leave);
the former does not.
consequences of one’s free decisions, choices, and actions rather than by Divine
fiat at an arbitrary point in time (e.g., one’s bodily death).21
In response to this concern, it seems perfectly consistent with the revisions
I have offered to Murray’s proposal to think that it is neither the case that God
brings about the transition from person to wanton in a manner that compromises
the unfolding of the natural consequences, nor that death represents the point at
which that transition necessarily occurs for human beings. In effect, the proponent
of Murray’s view is committed to something like the picture of the afterlife that
C.S. Lewis imagines in The Great Divorce. Nothing in my proposed revisions
to Murray’s view requires either that damned human beings become wantons at
death or that they are made to become wantons by God.
Conclusion
If both my critical appraisal and friendly amendment of Murray’s hybrid model of

hell is correct, it carries with it at least one interesting consequence worth noting
in closing. Specifically, if the denizens of hell are all (and perhaps only) wantons,
then strictly speaking there are no persons in hell. At a minimum, this entails a
slightly new reading of both Kvanvig’s anti-universalism and existence theses,
each of which refers to “people.” However, envisioning hell as being populated by
wantons does not run counter to the substance of either thesis. The revised theses
would go something like this.
H1*: The Anti-Universalism Thesis: Some (wantons) creatures that are

psychologically continuous with the persons they once were are
consigned to hell.
H2*: The Existence Thesis: Hell is a place where (wantons) creatures that are
psychologically continuous with the persons they once were exist, if
they are consigned there.
I take it that the principal import of the first two theses in Kvanvig’s traditional
conception of hell is the affirmation that hell is a real place (H2*) and it is not
empty (H1*). Together with the no escape (H3) and retribution (H4) theses, these
first two have the effect of constituting a psychologically undesirable state of
affairs. This is part of what gives the traditional view its lack of appeal among
revisionists about the nature of hell. My point here is merely that the revision of
the first two theses to accommodate my claim that hell is occupied by wantons
does not undermine the substance of Kvanvig’s four original theses. Thus, the
proponent of Murray’s hybrid model can accept the revised version of Kvanvig’s
21
I am grateful to an anonymous participant at the 2008 Annual Meeting of the
Evangelical Philosophical Society for bringing this point to my attention during a
presentation of an earlier draft of this chapter.
theses and still argue that hell’s punishment is not inconsistent with either God’s
justice or love because the wantons in hell are there as a result of the natural
consequences of a lifetime of decisions, choices, and actions directed toward self-
love rather than the love of God.22
The revision of Murray’s proposal that I am suggesting makes sense of
something that C.S. Lewis wrote about hell’s occupants in The Problem of Pain.
What is cast (or casts itself) into hell is not a man: it is “remains”. To be a
complete man means to have the passions obedient to the will and the will
offered to God: to have been a man – to be an ex-man or “damned ghost” – would
presumably mean to consist of a will utterly centered in its self and passions
utterly uncontrolled by the will. It is, of course, impossible to imagine what
the consciousness of such a creature – already a loose congeries of mutually
antagonistic sins rather than a sinner – would be like.23
To be a wanton in hell is to be the remains of a person. As Lewis points out,

it is perhaps impossible for us to imagine what it is like to be a wanton. Still,
Frankfurt’s analysis provides some hint. To be a wanton in hell is to be a creature
who is cognitively aware of the infinite loss of God’s goodness (and perhaps the
pain that attends that loss), but who nonetheless persists in the (first-order) desire
to love self, and perhaps most sadly, altogether lacks a (second-order) concern
as to whether that (first-order) desire to love self persists in its effectiveness for
eternity or not. Thus, the wanton in hell both wants to be there and does not want
to be there, does not like it, but at the same time neither wants to want to be there
nor wants to not-want to be there. This seems the very essence of damnation.
22
Of course, whether Murray’s proposal succeeds in this larger task is beyond the
scope of this chapter. For whatever it is worth, the reader is correct in assuming that my
defense of Murray’s proposal reflects an underlying agreement with the substance of
Murray’s view as a response to the problem of hell.
23
Lewis, Problem of Pain, p. 128.
Chapter 5
Value, Finality, and Frustration:
Problems for Escapism?
Andrei A. Buckareff and Allen Plug
In recent papers, we have focused on the problem of hell from the perspective of
Christian theism.1 We have argued for a theory of hell we claimed provides an
adequate response to the problem of hell without rejecting hell altogether, thus
not amounting to a radical departure from the dominant Christian tradition. We
christened our view of hell “escapism.” We defined escapism as the conjunction of
the following two claims:
1. Hell exists and might be populated for eternity.

2. If there are any denizens of hell, then at any time they have the ability to
accept God’s grace and leave hell and enter heaven.2
Additionally, we specifically endorsed an issuant view of hell, wherein hell is a

place that God, out of his love for all of his creatures, has provided for those who
do not wish to be in communion with God; as opposed to a retributive view of hell,
wherein hell is a place of punishment.
In this chapter we wish to extend and clarify escapism further. In particular we
will address the following three questions:
(1) According to escapism, is hell an unmitigated good thing for those that
are in hell?
(2) Is escapism consistent with a Christian view of eschatology that requires
an element of finality or consummation?
(3) Does escapism allow for the possibility that God’s redemptive plans may,
ultimately, be frustrated?
1
In Andrei A. Buckareff and Allen Plug, “Escaping Hell: Divine Motivation and the
Problem of Hell,” Religious Studies, 41/1 (2005): 39–54; and “Escapism, Religious Luck,
and Divine Reasons for Action,” Religious Studies, 45/1 (2009): 63:72.
2
Buckareff and Plug, “Escaping Hell,” p. 46. An additional note on 2: we allow
that it might be psychologically challenging for some agents to accept God’s offer if their
characters have settled into a position where they will not accept God’s grace. However,
according to escapism, God never gives up on any individual. See note 21 in “Escaping
Hell” for further clarification on this point.
We will argue (1) that escapism is not committed to the view that hell is an
unqualifiedly good thing; (2) that escapism has the resources to account for the
Christian requirement of finality and consummation; and (3) that while it is possible,
according to escapism, that God’s redemptive plans may be frustrated, this is in
fact an asset of the view and not a liability. We will address each of the questions in
turn. But first we will briefly review the original argument for escapism.
The argument3 for escapism begins with an assumption regarding divine
action: namely, that all of God’s actions are just and loving. We argue that if that
is true, then God’s motivation regarding his sotereological activity will be driven
by God’s desire for the most just and loving outcome. The most just and loving
outcome, we argue, is for everyone to freely choose to be in communion with God.
Thus, we argue, we should expect that God would make provisions for people to
convert in the eschaton and that the opportunities for persons to convert should not
be exhausted by a single post-mortem opportunity. In more detail, the argument
is as follows:4
3. All of God’s actions are just and loving

4. If all of God’s actions are just and loving, then no action of God’s is
motivated by an unjust or unloving pro-attitude.
5. If no action of God’s is motivated by an unjust or unloving pro-attitude,
then God’s soteriological activity is motivated by God’s just and loving
pro-attitudes.
6. If God’s soteriological activity is motivated by God’s just and loving pro-
attitudes, then God’s provision for separation from God is motivated by
God’s desire for the most just and loving state of affairs to be realized in
the eschaton.
7. If God’s provision for separation from Him is motivated by God’s desire
for the most just and loving state of affairs to be realized in the eschaton,
then God will provide opportunities for people in hell to receive the gift of
salvation and such persons can decide to receive the gift.
8. Therefore, God will provide opportunities for people in hell to receive the
gift of salvation and such persons can decide to receive the gift.
We will now move on to a discussion of escapism and value, finality and divine
frustration. We begin with escapism and value.
3
What follows is a brief summary of our theory of hell as outlined on pages 40–42
of “Escaping Hell.”
4
This is the argument that is spread out over pages 42–5 of “Escaping Hell.”
Value, Finality, and Frustration: Problems for Escapism? 79
Escapism and Value
It has been suggested that a feature of escapism that distinguishes it from other
views of hell is that, on escapism, hell is good for the persons who reside there.5
This is not an objection, per se. It is an observation made by some that is worth
addressing. Our own concern is with whether escapism commits one to the claim
that hell is good for persons. We believe it does not commit one to such a claim.
The motivation that, on escapism, hell is a good comes from the assumption
of issuantism built into escapism and the view that those that are in hell are there
because that is where they choose to be. Given issuantism, hell is a manifestation
of God’s love for His creatures, so those in hell are not being punished according
to escapism. Further, given that being in hell is a choice on the part of those who
reside in hell, being in hell at the minimum satisfies a desire on the part of those
who are there. Indeed, if they were not in hell some desire of theirs would not be
fulfilled.
While on escapism it is the case that the denizens of hell enjoy positive
(quantitative) well-being and so there is a sense in which hell is not bad, we are
hesitant to say that hell is good for persons who reside there without qualification.
This is the case if for no other reason than there are competing conceptions of
good and we are not sure that aggregative theories of good accurately account
for all of the possible good-making features of a state of affairs. If they did, then
persons in heaven and in hell, since their experiences last for an infinite amount
of time, would enjoy the same amount of good since the units of well-being that
are aggregated constitute an infinite amount of good. This is the case even if for
at any time t the net quantity of well-being a person in heaven enjoys is infinite
and the net well-being of a person in hell is +1.6 But even if one’s theory of value
is such that one can grant that those in hell enjoy positive well-being (even an
infinite amount of well-being), where the good in question is quantifiable, one
need not be committed to the claim that hell is unqualifiedly good. The lack of
certain qualitative goods may render the overall state of being in hell something
bad relative to the good of heaven, or at least not preferable to heaven.
Harry Frankfurt’s work on autonomous agency may help us understand
the difference between the quality of certain goods. On Frankfurt’s account of
autonomous agency, a willing agent is someone who endorses a particular
motivational state over other competing motivational states he may have. 7 Such an
5
In correspondence and in a recent paper, Stephen Kershnar suggested this further
characteristic of escapism. See Stephen Kershnar, “The Injustice of Hell,” International
Journal for Philosophy of Religion, 58/2 (2005): 103–23, at p. 19, n5.
6
This is a result we get if appeal is made to Cantorian infinities; but on standard
mathematics the result would be that neither sum would be well defined, since infinite
numbers and infinitesimals are not recognized for addition and multiplication.
7
See Harry Frankfurt, “Freedom of the will and the concept of a person,” Journal of
Philosophy, 68/1 (1971): 5–20.
agent “wants a certain desire to be his will.”8 Frankfurt calls such desires “second-
order volitions.”9 Such an agent identifies with the desire that is the intentional
object of his second-order desire. In later writings, Frankfurt adds that an agent who
is fully integrated and autonomous “wholeheartedly” identifies with the relevant
psychological items that constitute the person as he truly is or wants to be.
Consider a person who believes something wholeheartedly, who is wholehearted

in some feeling or attitude, or who intends wholeheartedly to perform a certain
action. In what does his wholeheartedness with respect to these psychic elements
consist? It consists in his being fully satisfied that they, rather than others that
inherently (i.e., non-contingently) conflict with them, should be among the
causes and considerations that determine his cognitive, affective, attitudinal, and
behavioral processes.10
An unwilling agent has “conflicting first-order desires” and has a second-order

volition, wanting a certain desire to be effective; only he finds himself acting on
the desire that he does not want to be his will.11 A “wanton,” on the other hand,
according to Frankfurt, is someone who finds herself, in effect, carried along by
her desires. Such an agent may have second-order desires, but she lacks second-
order volitions. The wanton is unreflective, lacking any concern regarding whether
the motivational states that move her to act are the ones she wants to be moved
by. Children and non-human animals are characteristically wantons. “They do
whatever their impulses move them most insistently to do, without any self-
regarding interest in what sort of creature that makes them to be.”12
Compare a willing heroin addict to an unwilling addict and to someone who is
more like Harry Frankfurt’s wanton. All of them want to take heroin. The unwilling
addict experiences a deep conflict within herself. She is not unlike St. Paul who,
in his letter to the Romans (7:14–25), expresses his frustration over failing to act
in accordance with the law, which he wants to act on. Paul’s reference to what he
wants to do, viz., act in accordance with the law, seems to be like a second-order
volition. His actions are obviously motivated, only they are motivated by sinful
first-order desires. He does not act in accordance with the law that he delights in
within his “inmost self” (Romans 7:22). The unwilling addict, like St. Paul, has
a divided self. She does not identify with her desire to take heroin and certainly
fails to be wholehearted with respect to her actions when she shoots up. That
such a state of affairs is somehow good strains credulity; it is a description that
8
Ibid., p. 10.
9
Ibid., p. 7.
10
Harry Frankfurt, “The Faintest Passion,” Necessity, Volition, and Love (New York,
1999), pp. 95–107, at p. 103.
11
Frankfurt, “Freedom of the will,” p. 12.
12
Harry Frankfurt, “Taking Ourselves Seriously,” Taking Ourselves Seriously and
Getting it Right (Stanford, 2006), pp. 1–26, at p. 6.
rightly deserves our disapprobation. Such an agent enjoys little well-being. She is
tormented and fragmented.
The wanton, unlike the unwilling addict, is not a tormented agent, but she is
hardly worth calling an autonomous agent. She does not evaluate her desires and
has no concern for what motivational states she acts upon. If we suppose that she
is simply driven by her appetites, and if her appetites are always satisfied, there is
certainly a sense in which she has a good life. From her standpoint, things could not
be better. The quantity of her well-being is higher than that of the unwilling addict.
But the overall quantity of well-being she enjoys and the quality of it is certainly
lower than that of someone not addicted to heroin. For that matter, it is fair to say
that there is something qualitatively better about the life of the unwilling addict.
The unwilling addict is a reflective agent. The unwilling addict is akratic, but she
is aware of this fact about herself and is not satisfied with this. She recognizes that
some states of affairs are better than others for reasons to which the wanton seems
wholly unresponsive.
Finally, the willing addict prefers to be on heroin, endorsing her desire to take
heroin and enjoying the sensation, having a second-order volition to take heroin.
Suppose that the willing addict is a successful, productive member of society.
She clearly enjoys some positive well-being. But it is only good from her first-
personal and perhaps a quantitative standpoint. Arguably, the state of being a
heroin addict is still bad, although the willing addict is different from the unwilling
addict who prefers not to take heroin, yet finds that she acts on her desire to shoot
up. Moreover, the willing addict is also a reflective agent. We may even suppose
that she wholeheartedly identifies with her first-order desire to take heroin. Her
addiction does not compel her to shoot up in the same way it does the unwilling
addict. We should say that it may compel her in the same way, but she does not feel
bound by such compulsion. The desire to satisfy her desire to be on heroin is like
a person with a normal appetite’s desire to satisfy her desire to eat when hungry.
While the willing addict enjoys some positive well-being, being an addict is not an
overall good state of affairs. There is something qualitatively inferior about being
an addict versus not being an addict. Specifically, there is some loss of autonomous
control one suffers as a result of an addiction. Withdrawal from heroin is far more
stressful and puts more strain on one’s body than, say, withdrawal from eating
cheesecake. Furthermore, the destructive effects of the drug taken in the quantities
necessary to satisfy the agent’s addiction count against willing addiction being
an overall good state of affairs, even if the quantity of well-being does not differ
between the willing addict and someone who willfully abstains from heroin use.
The lesson of the foregoing summary of Frankfurt’s account of autonomous
agency and its application to the case of addiction for the goodness of hell is, we
hope, pretty straightforward. On escapism, those in hell are obviously not like the
unwilling addict. It would be odd, indeed, for a loving God not to enable the wills
of those who do not identify with a desire to be separated from God but have a
second-order volition to be with God. Such agents are, as we noted, like St. Paul
describes himself. Not to render their second-order volitions effective would be
out of character for a loving God who extends prevenient grace towards all and
redeems those who wish to enter into communion with God.
It would be odd if the inhabitants of hell were like the wanton. The wanton
is not an autonomous agent. He is unreflective. But it is not inconceivable that
inhabitants of hell are like wantons. In fact, if hell is always populated, wantons
may populate it. If they only reflected on their desires, over time, they would realize
that a more desirable state of affairs awaits those who enter into communion with
God. This is not to say, however, that a rational agent would obviously choose
communion with God. Otherwise rational agents calculate the costs and benefits
of smoking and willfully continue to smoke, wholeheartedly endorsing the desire
to smoke. Things could be similar with rational persons who elect never to leave
hell. Such persons are like the willing addict. In any case, the object of the willing
agent’s desire with which he identifies as his second-order volition and acts upon
is at least qualitatively inferior to that of the person who wholeheartedly identifies
with a desire to be with God in heaven. If we assume that being with God is
qualitatively superior to being separated from God, then the wantons and willing
inhabitants of hell take as the intentional object of their desires a state of affairs
that is qualitatively inferior to the state of affairs that would be realized by being
with God. Things may go well for them by one measure, specifically, if we just
aggregate goods, quantifying over units of well-being. But just by failing to be
in communion with God, each unit of well-being is qualitatively inferior to what
is experienced by those in heaven. This is just like how each unit of well-being
enjoyed by an addict high on heroin is qualitatively inferior to each unit of well-
being enjoyed by someone not addicted to heroin enjoying the same quantity of
well-being by engaging in some other activity.
For those who are skeptical about such qualitative distinctions between states
of affairs, in the case of comparable well-being at an infinite number of temporal
locations (e.g., between heaven and hell) one could get the result that heaven is
still better. Depending upon how we carve things up, persons in heaven enjoy a
greater amount of well-being even if non-aggregative theories of value are false.13
The problems only arise when we expand our view to consider the total well-
being. Suppose that heaven and hell have the same number of temporal locations.
Suppose further that the amount of well-being is finite at each temporal location
in each place. The quantity of well-being at each temporal location is greater in
heaven than in hell. This assumption is reasonable if for no other reason than
heaven is described in Christian tradition as a better place than hell. The problem
for escapism comes from aggregating the total well-being in heaven and hell,
both having infinite quantities of well-being since there is an infinite number of
temporal locations in each. But if we consider a finite subset of temporal locations
13
See Peter Vallentyne and Shelly Kagan, “Infinite value and finitely additive value
theory,” Journal of Philosophy, 94/1 (1997): 5–26 for a defense of the claim that where w1
and w2 are two sets of infinite temporal locations with different values at each location, one is
better than the other that makes use of comparisons of finite sets of locations in each world.
and aggregate the well-being, the well-being enjoyed by persons in heaven is

greater than those in hell. So for any proper finite subset of temporal locations
in heaven and hell, the quantity of well-being enjoyed by a person in heaven is
greater than that of a person in hell. The sets of temporal locations in question can
be quite large. So long as they are not infinite, the amount of well-being enjoyed
by those in heaven is greater than that enjoyed by those in hell. So for any single
temporal location or finite set of multiple temporal locations there is a quantitative
difference between the well-being enjoyed by persons in heaven versus those in
hell. Any putative problems only arise from considering the entire infinite set of
temporal locations. While this does matter when quantifying the total well-being
of persons in either location, the facts “on the ground,” so to speak, would betray a
difference in the amount of well-being enjoyed by persons in each location at any
single moment of time.
While we have labored the point that escapism does not commit one to the claim
that hell is unqualifiedly good, the foregoing problem by itself does not constitute
a problem for escapism. Escapism does not commit one to any particular view
regarding the goodness of hell beyond that some positive well-being is enjoyed
by persons in hell. They have chosen to be there. They have not chosen separation
from God exclusively under another description. They may choose hell under
another description than choosing separation from God; but they also choose not
to be with God. Moreover, escapism’s issuant commitments render hell a place
provided by God out of love; it is not a place to exact retribution. These features
of escapism allow us to say that being in hell affords some well-being to agents
who would reside there. Variations of escapism may deviate from the formulation
of the view’s commitments on the goodness of hell offered here. Our goal in this
section was simply to argue that one can still explain how hell is bad in some sense
vis-à-vis heaven on escapism. We believe this should be apparent at this juncture.
Finality and Escapism
Many traditional Christian theists regard the finality of hell as central to the concept
of hell. Such theists regard the failure of a view of hell to account for its finality
to count against it. The idea here, we take it, is that, according to such Christian
theists, heaven and hell are final—there is no never-ending revolving door between
the two. This idea is expressed in the following quote from Jonathan Kvanvig:
Other second chance views claim that consignment to hell cannot be postponed,
but that escape from it is not impossible; all that is needed to get out is the same
change of heart, mind, and will required in one’s earthly life to be fit for heaven.
One difficulty for such a view is theological rather than philosophical, for such
views fail to be truly eschatological accounts of heaven and hell. Eschatology
is the doctrine of the last things, and one feature of this idea of culmination or
consummation is that there is a finality to it. In Christian thought, this idea is
expressed vividly in the idea of a final judgment, and any conception of the
afterlife that treats residence in heaven and hell in the geographic way in which
we think of residence in, say, Texas or California, simply does not fall into the
category of an eschatological doctrine at all. If heaven and hell are conceived
of as mere extensions of an earthly life, where people can pack up and move at
will, such a conception has more affinity to religious perspectives that espouse
endless cycles of rebirth than to religions with a substantive eschatology.14
So, according to Kvanvig, for an eschatological view to count as a Christian

eschatological view the view must, at minimum, contain a sense of finality—of
consummation. And this finality extends to both heaven and hell. The worry here
is that, if this is in fact a requirement of Christianity, then escapism is inconsistent
with Christianity. This threatens escapism in two ways. First, escapism claims that
those in hell have, at any time, the ability to leave hell and enter into communion
with God (i.e. enter heaven). This aspect of escapism seems, at first, to eliminate
any possibility of finality or consummation for hell. Second, it is possible that our
argument in “Escaping Hell” implies that God also maintains an open door policy
with respect to those in heaven. If so, then escapists would also be committed
to the view that there is no finality in heaven. We will not address the question
of whether Christianity does require the sense of finality that Kvanvig endorses.
Rather we will argue that escapism is consistent with the demands for finality
that Kvanvig issues above. We will first argue that the considerations that support
escapism for hell do not extend to heaven—i.e., escapists need not hold that God
maintains an open door policy for heaven. We will then move on to a discussion
of finality and hell.
We begin then with a discussion of escapism and heaven. Specifically, we will
consider whether our argument in “Escaping Hell” implies that God also maintains
an open door policy with respect to those in heaven. The idea, we take it, is that
heaven and hell are symmetrical. And if God maintains an open door policy with
respect to hell, then the same considerations would lead God to maintain an open
door policy with respect to heaven. But this, the objection goes, runs counter to a
venerable Christian tradition that claims that one cannot lose one’s salvation once
in heaven.15 We want to claim that the considerations that lead to escapism do not
lead to the position that God maintains an open door policy in heaven. Rather, we
claim that those considerations in fact lead to the position that the choice to enter
into full communion with God means that one can never choose to leave heaven.
We believe that an agent who intends to enter into communion with God must
have the character necessary to form such an intention. Such a character change
results in a shift in an agent’s motivational states, particularly her preferences, and
14
Jonathan Kvanvig, “Hell,” in Jerry Walls (ed.), The Oxford Handbook of Eschatology
(New York, 2008), pp. 413–26, at p. 418.
15
Additionally, persons from the Reformed tradition of Christianity may be concerned
that it results in a denial of the doctrine of the perseverance of the saints.
her emotional life. Such a change is necessary for agents to fulfill their purpose
as bearers of the imago dei. Regarding the purpose of human agents, in our
original paper we cited the answer to the first question of the Westminster Shorter
Catechism approvingly, which states that our purpose is “to glorify God and to
enjoy him forever.” We expressed agreement with Cornelius Plantinga’s claim
that this requires that we share God’s intentions and the purposes they represent.
Plantinga writes that, “To enjoy God forever is to cultivate a taste for this project,
to become more and more the sort of person for whom eternal life with God would
be sheer heaven.”16 The sort of change in an agent that must take effect for her to
be one who has cultivated a taste for God’s project and for whom heaven would be
eternal life with God is radical, to say the least.
Nothing short of either an immediate person-transforming change upon
entering heaven or a fair amount of time in purgatory developing a taste for the
divine project can bring about the change needed in an agent to be fit for heaven.
We take it that whatever the case may be, an agent who has undergone such a
transformation would have to be the sort of person for whom any sort of turning
from God would be psychologically impossible. So, if escapism is true, then the
change necessary for an agent to enter into complete communion with God—i.e.,
heaven—would be such that it would result in an agent being such that she could
never leave heaven.
A concern that may emerge at this point may come from those sympathetic to
our emphasis on God’s respect for the autonomy of created persons. Autonomy
appears to have been tossed out the window in heaven. This is not true.
An agent in heaven, in either developing his or her character or submitting
to an instant transformation, willingly gives up the ability to choose not to be
with God. But such an agent is now an agent who performs what Michael Smith
calls “orthonomous actions.”17 An orthonomous action is one that is performed
for right reasons—i.e., a right action. We take it that an agent in heaven would be
an orthonomous agent. The trade-off in becoming a citizen of heaven is that one
becomes someone who always acts for right reasons. There is still an obvious sense
in which such an agent is still autonomous. She would be self-governed. Only her
behavior is entirely aimed at ends that are consistent with God’s purposes and
continued communion with God, the final object of such an agent’s motivational
states and actions. Such an agent is not one we expect would be able to simply
pack up and move to hell (and maybe back to heaven later); nor would she be
capable of becoming someone who is able to move to hell—since in becoming
such a person she would have needed to act for wrong reasons, which is now
impossible for her.
16
Cornelius Plantinga, Not the Way It’s Supposed to Be: A Breviary of Sin (Grand
Rapids, 1995), p. 37.
17
Michael Smith, “The Structure of Orthonomy,” in John Hyman and Helen Steward
(eds.), Action and Agency (Cambridge, 2004), pp. 165–93.
Defenders of libertarianism may find orthonomous agency to be inadequate

for moral agency.18 But orthonomy is consistent with libertarianism. Specifically,
orthonomous agency is consistent with wide source incompatibilism.19 Wide source
incompatibilists endorse an actual-sequence theory of moral responsibility for
actions that issue from a character for which an agent has ultimate responsibility.
So on wide source incompatibilism, an agent is morally responsible for some action
A at some time t2 that issues from a character formed at some earlier time t1 as a
consequence of performing what Robert Kane calls a “self-forming action” (SFA)
at which time libertarian free agency was exercised.20 When the agent performed
the SFA, she had alternate possibilities and so could have acted otherwise. An
action issuing from the character formed by such an SFA is one that an agent can
be morally responsible for on this view. If this view is right, orthonomous agency
in heaven is also consistent with source incompatibilism. At the time an agent
made the choice to be reconciled with God the agent began forming the character
of an orthonomous agent who, when finally in heaven, could never turn from God.
Such an agent has chosen to be someone who will eventually act orthonomously.
So the argument in “Escaping Hell” does not extend to heaven. We now want
to argue that while escapism does claim that those in hell have the ability, at any
time, to leave hell and enter heaven, this claim is not inconsistent with the demand
of finality.
There are two different ways we can understand the finality of hell, as illustrated
by the following two claims (a de re modal claim and a de dicto modal claim,
respectively):
9. There is some time t, such that, necessarily, all those who are in hell at t will
remain in hell for all time.
10. Necessarily, there is some time t, such that, all those who are in hell at t will
remain in hell for all time.
(9) is inconsistent with escapism. If (9) is true, then there is some moment of time
such that from that time forward no one will get out of hell. And it is the case that
that moment of time has that property in every possible world.21 So there is a time
18
Stephen Kershnar raised this worry.
19
For a defense of wide source incompatibilism, see Kevin Timpe, “Source
Incompatibilism and Its Alternatives,” American Philosophical Quarterly, 44/2 (2007):
143–55.
20
See Robert Kane, The Significance of Free Will (Oxford, 1996).
21
We are assuming a substantival view of time. So moments of time are real entities
that would have transworld identity. If we dispense with substantivalism about time, we can
simply refer to states of affairs and events as having transworld identity and as indexed to
times in those worlds when the proper triggering events that bring about the relevant event
or state of affairs under consideration—in our case, the state of affairs that obtains when no
one will leave hell.
when, for all those currently in hell, it is no longer possible for them to leave hell
and enter into communion with God. And this is inconsistent with escapism. So
if (9) is the proper way to understand the finality of hell, then escapism cannot
account for the Christian requirement of finality.
But (9) is false. (9) places an unjustified limit on God qua creator as
characterized by traditional versions of theism. If (9) is true, then there is some
time which is necessarily the “last time.” But we take it that it is uncontroversial to
claim that God, if He so desired, could extend the time available for those in hell
to be reconciled with God. That is, God could choose to have the last time occur
sooner or later.
Of course, God could declare in advance when that last time will occur. It
could occur at the moment God foreknew that everyone in hell has settled on
their choice to remain estranged from God and all of those whom God foreknew
would decide to be reconciled have left hell (leaving it open as to whether or not
hell empties out or fails to lose any of its residents). But if (9) is true, then there is
some time which is, necessarily, the last time. This would mean that God would
not have been free to declare some other time to be the last time. This, we take it,
is too grave a violation of God’s sovereignty to accept for traditional theists.
(9) is inconsistent with escapism but, fortunately, is false. (10), however, is
true if escapism is true. Recall that we deny that God maintains an open door
policy with respect to heaven. Thus, if escapism is true there are only two possible
outcomes:22
A. Eventually everyone enters into communion with God (and so leaves

hell).
B. There are some persons who never enter into communion with God (and so
hell is eternally populated).
On either possibility there is a time at which all those in hell at that time will
remain there for eternity. Thus (10) is true if escapism is true. Suppose possibility
(A) holds. Consider the time when the last person leaves hell and is reconciled
with God. It is trivially true then that after that time all those in hell will be in hell
for all eternity. Similarly with possibility (B), take the time when the last person
leaves hell and enters into communion with God. It is true that after that time
all those in hell will be in hell eternally. So if escapism is true, then (10) is true.
Escapism, then, can account for the traditional Christian demand of finality.
22
It is important to note here that the following only holds with the additional assumption
that those in heaven cannot leave and (re)enter hell. Without that additional assumption
there would be a third possibility: that there are some individuals who continually leave hell
and enter heaven (i.e., enter into communion with God) and then again leave heaven (break
the communion with God) and then continue the cycle. On that possibility there is no time
such that at that time all those in hell will remain in hell for all eternity. So technically (10)
follows from escapism and the assumption that those in heaven cannot leave.
The above will not satisfy everyone however. While (10) is consistent with
escapism it does not obviously capture what Kvanvig and others mean by “finality.”
Kvanvig uses the language found in the Christian scriptures when discussing
finality. Kvanvig writes, “In Christian thought, this idea is expressed vividly in
the idea of a final judgment.”23 We do not pretend to understand perfectly the
eschatological language found in the Christian scriptures. The language in the
Christian scriptures that refers to the eschaton, beyond the general statements about
God’s redemptive intentions, is abstruse at best. If we take some hermeneutical
liberties when interpreting what is meant by “final judgment,” we are not doing
anything most other non-literalists about hell (including most traditionalists) are
not doing. The explanation we provide by understanding finality in terms of (10)
allows us to explain how there can be some finality in the eschaton. Granted, the
finality, we assume, is not because God made a supralapsarian declaration that at
some arbitrary moment of time God’s policy would shift to not allowing persons to
be reconciled with God. Rather, if some remain in hell, God foreknows that a time
will come after which those in hell will have decisively shut the door on God.24
Of course, some might want to claim that the eschatological language found
in Christian scriptures requires an event that marks the final judgment, or the last
day. We do not see that this is required; however, escapism is consistent with
such a requirement. However, the timing of such an event, according to escapism,
would be determined by God’s foreknowledge of when all those who eventually
accept God’s grace and enter into communion with him will have done so. So
God foreknows when the last person will accept the offer of grace and enter into
communion with God. That foreknowledge then allows God to determine the
timing of the final event—be it the last day or the final judgment.
One might be worried that the above requires an unacceptable loss of divine
sovereignty. The worry is as follows. On the above view the timing of the last days
is out of God’s control. The timing of such an event is determined not by God, but
by those who are in hell. For the end days will not occur until the last person who
will accept God’s offer of grace does so accept.
The above worry is mistaken. For on escapism the timing of the final event is
determined solely by God. While it is true that if escapism is true (and assuming
there is some final event), that the final event will not take place until all who
will accept God’s offer of grace have done so, that is because that is what God
so chooses. God could choose some other time. All that escapism shows is that,
assuming a traditional understanding of God’s motivational states, God would not
wish to choose any other time—but, presumably, if his desires were different, he
23
Kvanvig, “Hell,” p. 418. Some often cited, popular passages that describe the
finality of separation from God include Matthew 25:31–46; Acts 17:31; Romans 2:16; and
2 Thessalonians 1:6–9.
24
Such agents may be the sorts of agents who could fit the description of having
finally blasphemed against the Holy Spirit by being completely shut off from divine grace
and God’s persuasive activity in their lives (Luke 12:10).
would choose some other time. So escapism can meet the Christian demand of
finality and in so meeting that demand escapism does not unacceptably diminish
God’s sovereign control.
Escapism and Divine Frustration
The last concern comes from the perspective of universalism. One of the motivations
Thomas Talbott offers for his sophisticated universalism is that God’s purposes
would be frustrated if some persons are not redeemed. This divine purpose issues
from God’s being a supremely loving agent who intends to bring it about that all
persons are freely reconciled with him.25 If not everyone is saved, then God’s
overriding redemptive purpose would be frustrated. But if God’s purposes are
frustrated, then God’s sovereignty over the created order seems to be at risk. The
problem for escapism, then, is that by leaving it open whether or not all persons
will finally be redeemed escapism treats the frustration of God’s redemptive plans
as a live possibility in the actual world. We do not believe that escapism treats the
frustration of God’s redemptive plan as an actual possibility.
We understand God’s soteriological purpose to be to redeem all created
persons who are estranged from God. We take it to be the case, however, that
God does not merely desire that all persons would be in heaven. Rather, God
prefers having all those who prefer to be with God to enjoy the fullness of God’s
presence. This requires that God respect their autonomy and not compromise it.
The choices of agents may be objectively irrational when you catalog all of their
preferences and what is actually in their best interest. But their lack of knowledge
of this fact and their failure to love God makes their choice at least subjectively
rational. Thomas Talbott suggests that a removal of all the impediments to making
an objectively rational, autonomous choice would result in an agent’s choosing
communion with God.26 But it is not obvious that such a change can be effected
without actually doing violence to the character and will of the agent. Moreover,
it assumes that persons would still choose rationally once the change has been
effected. At least in the case of humans, our moral psychology is complex and it
is difficult to accurately predict how agents will act (whether or not libertarianism
or compatibilism is true). We do not doubt that God knows how persons will react.
But the scriptural data is opaque enough for us to be comfortable in asserting that
we cannot be confident about whether or not all persons will be reconciled with
God. In any case, if some persons fail to be reconciled to God, God’s purposes
would not be frustrated given that all of those who prefer to be with God enjoy the
25
Thomas Talbott, “No Hell,” in Michael L Peterson and Raymond J. VanArragon
(eds.), Contemporary Debates in Philosophy of Religion (Oxford: 2004), pp. 278–87, at pp.
279–81. For a classic statement of Talbott’s version of sophisticated universalism, see his
paper, “The doctrine of everlasting punishment,” Faith and Philosophy, 7/1 (1990): 19–42.
26
Talbott, “No Hell,” p. 283.
fullness of the divine presence, while those who prefer not to be with God remain
in hell. If escapism is true, God’s plans have been successfully executed—viz.,
all who wish to be reconciled with God are finally reconciled with God, but the
others know that there is always room at the table. Such a state of affairs strikes
us as tenable.
In concluding this section, it is worth noting a particular strength of escapism
we have noted in our original paper defending escapism. Escapism has the
resources built into it to satisfy the universalist, so long as the universalist is not
committed to a predestinarian soteriology. We claimed that “a strength of escapism
is that universal reconciliation without divine coercion is not merely a logical
possibility but may be a likely state of affairs in the eschaton.”27 If we assume
that the universalistic passages in the Christian scriptures should be taken at face
value, we can hold that it is a contingent fact about the actual world that all of
the denizens of hell will actually be reconciled with God after some time. The
scriptures report a fact about how things will finally turn out.
This approach allows one to make sense of some of the biblical data that counts
in favor of universalism. Escapism is still necessarily true; and there are possible
worlds where no one in hell is reconciled with God, possible words where only
some are, and others where all created persons go to hell. But in the actual world
God foreknew that everyone would finally be reconciled with God and so God
revealed this fact to humankind in scripture. God does, as a matter of fact, triumph
in the end. But it is by persuading and not coercing us. An added bonus of such an
approach is that we also get finality in the afterlife.
Escapism, of course, does not commit one either way to universalism or
its rejection. One can be an escapist and be a universalist. But a commitment
to one does not entail commitment to the other. In either case, whether we are
universalists or not, escapism has the resources built into it to explain how God’s
divine purposes are not frustrated.
Conclusion
In conclusion, we have tried to extend and clarify the escapist view of hell. In
particular, we have argued that escapism is not committed to the view that hell is
unqualifiedly good for those in hell; second, that escapism can acceptably account
for the Christian demand of eschatological finality; finally we argued that though
escapism does allow for the possibility that God’s sotereological plans may be
frustrated, this, in fact, counts as a virtue.28
27
Buckareff and Plug, “Escaping Hell,” p. 50.
28
We wish to thank Stephen Kershnar and Michael Murray for their comments on
portions of earlier drafts of this chapter.
Chapter 6
Hell, Wrath, and the Grace of God
Stephen T. Davis
The wrath of God is revealed from heaven against all ungodliness and wickedness
of those who by their wickedness suppress the truth.
(Romans 1:18)1
But God, who is rich in mercy, out of the great love with which he loved us even
when we were dead through our trespasses, made us alive together with Christ—by
grace you have been saved.
(Ephesians 2: 4–5)
The traditional Christian picture of hell is problematical. It is said to be a place of

punishment where sinners are sent, against their wills, into eternal fiery torment.
Many Christians have rejected the notion and even those who accept it, or parts
of it, seem only rarely to want to talk about it. And this is for the obvious reason
that the traditional picture seems to many people today to be inconsistent with the
claim that God is morally good, perfectly just, and infinitely loving.
As a fairly conservative Christian, I believe in the existence of hell, and indeed
in its eternity. In the present chapter I want to argue that hell, properly understood,
is not only consistent with but is entailed by God’s loving and gracious nature. Part
of my argument will involve amending certain aspects of the traditional picture.
Naturally, I cannot claim to know that my opinions about hell are true. This chapter
simply constitutes my best effort to reach clarity on a dark and difficult theological
and philosophical topic.2
1
All biblical quotations in this chapter are taken from the New Revised Standard
Version (NRSV).
2
I should note that little in my chapter is original; much of what I say has been said
by others. I have been influenced by C.S. Lewis’s The Problem of Pain (New York, 1993).
And although they would not agree with everything that I have said in this chapter, I would
like to express my gratitude to Jonathan Kvanvig, The Problem of Hell (Oxford, 1993) and
Jerry L. Walls, Hell: the Logic of Damnation (Notre Dame, 1992).
II
In Christianity, God is indeed said to be loving, gracious, and merciful. But God is
also said to be one who judges, punishes, and has wrath. Are these two aspects of
God’s nature consistent? Must we choose one or the other?
What exactly is God’s wrath? It is simply God’s opposition to, hatred of, and
dissatisfaction with human disobedience. What God wants, of course, is obedience
to his commands: “I am the Lord your God: sanctify yourselves therefore, and
be holy, for I am holy” (Lev. 11:14). Just as most of us resent and regret terrible
injustice when we see it, so God, in an infinitely morally superior way, resents
and regrets human sinfulness. God’s usual reaction, so it seems, is to punish our
disobedience so that we are moved to repent. A morally perfect being like God will
hate and oppose evil. And that is what God’s wrath is. Thus Ephesians 5:6: “The
wrath of God comes on those who are disobedient.”
But then what is God’s grace? It is the willingness on the part of God to treat
us better than we deserve. God loves us even though we are unlovable; God
accepts us even though we are unacceptable; God forgives us even though we are
unforgivable. If God were to treat us as justice strictly demands, we would all be
condemned. So God treats us better than justice requires. That is the grace of God.
Thus Ephesians 2:8: “For by grace you have been saved through faith, and this is
not your own doing; it is the gift of God.”
Christians are used to the idea that the grace of God is our only hope. We know
that we have been systematically disobedient to God and that justice requires our
condemnation. So we can only be reconciled to God if God takes the initiative and
graciously forgives us. But it is equally true that the wrath of God is our only hope.
God’s wrath shows the human race that some acts are morally right and some are
morally wrong, that we are obligated to perform the first and eschew the second,
and that our acts count, have consequences. If it were not for the wrath of God, we
might sink into the pit of moral and religious relativism, which in my opinion is
one of the many roads to hell. The wrath of God is what keeps our world, at least
most of the time, from deteriorating into something like Hobbes’s state of nature.3
Religious and moral relativism is the view that whatever you think is morally
or religiously correct is morally or religiously correct, for you. So as long as you
are sincere, it does not matter what you believe or do: it is okay. The wrath of God
helps us here. The wrath of God is both a facet and an expression of God’s just
and holy nature. Justice is not a temporary whim of God but an eternal necessity,
except when God graciously treats us better than justice would demand. God’s
justice—so we might say—constitutes the moral equilibrium of the world. Of
course Christians know full well that the world, in many ways, is not in moral
equilibrium. Like human beings, the world has been corrupted by evil and needs to
be redeemed (Rom. 8:18–23). But it is important to note that most human beings,
3
See Chapter 12 of my Christian Philosophical Theology (Oxford, 2006), where I
explore these ideas further.
Hell, Wrath, and the Grace of God 93
most of the time, know what is right and what is wrong. And people know that
normally there are consequences for wrongdoing; we reap what we sow. We also
know that when one person harms another, some sort of apology or reparation
should be made. And most people react negatively—with righteous indignation—
to egregious wrongdoing or unjust suffering.
It is certainly possible for us to imagine worlds—like the one Hobbes
described—where these facts do not hold, where there is no or virtually no
moral equilibrium. This would be a world in which people do what they want,
unencumbered by communal or ethical considerations and in which actions are
determined only by considerations of self-interest and power. I am arguing that
what keeps us from such a world is the wrath of God. That wrath is revealed in
various ways, such as in God’s providential governance of the world, in scripture,
and even in the revelation of God’s will in the individual conscience. God’s wrath
reminds us that there are morally right actions and morally wrong actions, that that
distinction does not crucially depend on what anybody thinks is morally right and
morally wrong, and that there are consequences for doing wrong.
I do not hold to the view that human sin is infinite in its gravity (whatever
“infinite” means here) because it is an affront to an infinite being, God. This
argument was once common in Christian theological circles as a way of justifying
the traditional picture of hell, especially the eternality (infinite time span) of
hell. But I do not see how any human sin can be infinite in any important sense.4
However, sin certainly is a rejection of the sovereignty of an infinitely loving God.
Sin is serious; it cannot be just passed over as if it did not really matter.
Accordingly, in order to think about God correctly, we need to affirm both
God’s wrath and God’s grace. To emphasize the one without the other results in
radically misleading and dangerous conceptions of God. Christians do not hold
that God does not care about sin, coddles it (so to speak); nor do we hold that God
is a maximally severe, unforgiving judge. How then are God’s wrath and grace
related? The answer is that both are aspects of God’s nature. But they are not—as
is often presupposed—conflicting or opposed attributes. God does not have a split
personality; God does not (for example) in the atonement pay off with the loving
side of his character a debt incurred against the wrathful side. Love and kindness
are intrinsic and essential properties of God (they are essential aspects of God’s
inter-Trinitarian relations), while God’s wrath only emerges as a result of human
disobedience. But since grace amounts to treating the disobedient better than they
deserve, it, too, emerges as a reaction to human sinfulness.
Paul argues, in the form of a rhetorical question, that God patiently withholds
his wrath against us so that his mercy can be shown: “What if God, desiring to
show his wrath and to make known his power, has endured with much patience the
objects of wrath that are made for destruction; and what if he has done so in order
4
It is possible to argue that the effects of a given sin might be virtually endless,
maybe thus infinite in at least some sense, and thus the sinner might be morally responsible
in part for an almost endless train of consequences. But that is another matter.
to make known the riches of his glory for the objects of mercy?” (Rom. 9:22–3; cf.
I Thess. 1:9–10). Indeed, it is precisely because of God’s love and mercy toward
human beings that our sinfulness, and the suffering that it produces, can provoke
God’s wrath.5
Paul also asks this question: “Do you not realize that God’s kindness is meant
to lead you to repentance?” (Rom. 2:4). This point at first glance seems odd. We
naturally think it is God’s wrath that is meant to produce human repentance. And
that notion certainly seems true. But the very fact that Paul insists it is God’s
mercy that is designed to do so is significant. It shows that the divine wrath and
the divine mercy are not opposed to each other after all. Both aim at the same
result, viz. human repentance and divine forgiveness. Indeed, a completely moral
person will react positively to right-doing and negatively to wrongdoing. And such
a person who is also full of grace will not only react negatively to wrongdoing, but
give wrongdoers a chance to receive mercy.
Thus God’s wrath and grace are simply two aspects of God’s nature. Along
with God’s wrath, God’s justice and grace equally constitute the moral equilibrium
of the world. Only if we take God’s righteous judgment of sin seriously can we
understand the grace of God.
III
What is hell like? Naturally, the answer is that I do not know (and hope never to
know). Much of what the Bible says about hell is clearly metaphorical or symbolic.
For example, the New Testament uses the metaphor of fire to portray the suffering
of the denizens of hell. But this does not mean that the damned literally suffer the
pain of burns. Indeed, I do not believe that they do. Note that Mark 9:48 describes
hell as a place in which “the worm never dies, and the fire is never quenched.”
Why take the second image literally but not the first? Both metaphors seem to me
to point toward the horror and eternality of hell.6
Here is Jesus’ story of the Rich Man and Lazarus:
There was a rich man who was dressed in purple and who feasted sumptuously
every day. And at his gate lay a poor man named Lazarus, covered with sores,
who longed to satisfy his hunger with what fell from the rich man’s table; even
the dogs would come and lick his sores. The poor man died and was carried
5
As is argued by Robert Oakes, “The Wrath of God,” in David Shatz (ed.), Philosophy
and Faith: A Philosophy of Religion Reader (Boston, 2002), p. 10.
6
Thus John Calvin says, “We may conclude from many passages of Scripture, that
[eternal fire] is a metaphorical expression.” John Calvin, Commentary on a Harmony of the
Evangelists, Matthew, Mark, and Luke, trans. William Pringle (Grand Rapids, 1949), cited
in William Crockett, “The Metaphorical View,” in William Crockett (ed.), Four Views on
Hell (Grand Rapids, 1996), p. 44.
away by the angels to be with Abraham. The rich man also died and was buried.
In Hades, where he was being tormented, he looked up and saw Abraham far
away with Lazarus by his side. He called out, “Father Abraham, have mercy on
me, and send Lazarus to dip the tip of his finger in water and cool my tongue;
for I am in agony in these flames.” But Abraham said, “Child, remember that
during your lifetime you received your good things, and Lazarus in like manner
evil things; but now he is comforted here, and you are in agony. Besides all this,
between you and us a great chasm has been fixed, so that those who might want
to pass from here to you cannot do so, and no one can cross from there to us.”
He said, “Then father, I beg you to send him to my father’s house—for I have
five brothers—that he may warn them, so that they will not also come into this
place of torment.” Abraham replied, “They have Moses and the prophets; they
should listen to them.” He said, “No, father Abraham, but if someone goes to
them from the dead, they will repent.” He said to him, “If they do not listen to
Moses and the prophets, neither will they be convinced even if someone rises
from the dead” (Luke 16:19–31).
This powerful story is sometimes taken to be a literal picture of the afterlife, but
that interpretation seems to me incorrect. The story is one of the parables of Jesus;
that is, it is a made-up tale with a real life or recognizable setting that is designed
to make a religious point. (Nobody thinks the stories of the Good Samaritan or the
Prodigal Son describe real historical events.) Jesus’ parable of the Rich Man and
Lazarus may possibly have reflected common ideas of the afterlife in certain first-
century Jewish circles.7 But it is difficult (at least for me) to believe that heaven
and hell, as in the parable, are separated by a “great chasm” which cannot be
crossed but across which communication is possible.8
There are, in fact, many metaphors for hell in the New Testament: for example,
eternal fire (Matt. 25:41); bottomless pit (Rev. 9:2); outer darkness (Matt. 8:12,
Jude 13); place of weeping and gnashing of teeth (Matt. 8:12); place of no rest
(Rev. 14:11); place where the last penny must be paid (Matt. 5:26); and place
of destruction (Matt. 7:13; II Thess. 1:9). None of them, I believe, is a literal
description.
In some sense, hell can be spoken of as a place of punishment. It will
undoubtedly be experienced by its denizens as punishment. But I deny that hell is
primarily an arena of retribution, where God gets even with God’s enemies. The
7
It is at least possible that hell is not the same thing as the Hades that Jesus was
speaking about and that the latter was seen as the temporary abode of the dead until the
last judgment. But that idea does not seem to me to increase the probability that the parable
amounts to a literal picture of the afterlife.
8
There is, however, a great deal that we can learn from this deep parable and one
such point is relevant to the present chapter: sin will continue in the afterlife. Notice the
rich man’s implicit self-justificatory claim that he was not given enough warning during his
lifetime.
central truth about hell is that it is a place of separation from God. This is not total
separation, of course, for that would mean that hell could not exist. Moreover, the
biblical tradition affirms that nothing can ever be totally separated from God (Pss.
139:7–12). But hell is separation from God in the sense of being cut off from the
source of all love, joy, peace, and light. There is no deep or ultimate joy in hell and
I believe its denizens are miserable. To be apart from the source of love, joy, peace,
and light, is to live miserably.
But why are the damned in hell? As noted, I am not much attracted to the idea
of retribution, although there are biblical texts that might be taken to support the
idea. To put the point bluntly, I believe they are in hell because they choose to
be there. People are not sent to hell, kicking and screaming, against their wills.
Unfortunately, some people choose to live their lives apart from God, harden their
hearts, and will continue to say no to God after death; some will doubtless do
so forever. For such people, the prospect of living in the presence of God will
seem worse than living apart from God. Allowing them to live forever in hell is
simply God’s continuing to grant them the freedom that they experienced in this
life to say yes or no to God. Nevertheless, I suspect the people in hell are deeply
remorseful.
Can people both freely choose hell over heaven, knowing that they would be
unable to endure heaven, but still regret the fact that they cannot happily choose
heaven? Yes, I believe that this is quite possible. There are people who can carry
a grudge against an enemy for years despite recognizing the fact that it would be
better for all concerned if they forgave the enemy and moved on with their lives.
Moreover, the evidence that we have from this life is that some people will go on
denying God forever. Their hostility will grow; their hearts will grow ever harder.
People in hell will surely suffer, but their suffering will be largely self-inflicted.
Moreover, I suspect that they will cause each other to suffer; that is, they will
inflict pain on others who are within reach.
It might be objected that the notion that some people will voluntarily choose
hell might make sense initially (i.e., immediately after death) but that the idea
that they will do so everlastingly does not. And it is certainly true that heaven will
always be the most sensible option. But, again, I hold that people who continue
voluntarily to choose hell (even if we grant that they will always be offered the
option of repenting and being promoted to hell) will not be sensible. Their hatred
of God will have overcome them.
Will there be gradations in hell—more suffering for moral monsters and less for
more run-of-the-mill sinners? I do not claim to know the answer to that question.9
Doubtless something within us cries out for increased suffering in hell for people
like Hitler and Stalin. But people are not in hell, so to speak, for their sins but for
their sin. That is, they are in a self-chosen and unrepentant state of separation from
God and God’s law. And that will be true of all the denizens of hell.
9
Jesus seems to hint as much in Matthew 11:24.
Is the existence of hell consistent with the loving and gracious nature of
God? Yes, it is. Some Christians try to justify the existence of hell by viewing
it as the “natural consequence” of a life of sin. I accept the notion that hell is a
natural consequence of a life of rebellion against God (and in that sense, hell is a
punishment). But this point does not in itself justify God in condemning people to
hell, because it does not justify the divinely ordained laws of natural necessity that
make hell sin’s natural consequence. My claim, then, is that the denizens of hell
are in hell because they freely choose to be there. That is, they freely choose not
to live in the presence of God. If so, then hell is not only an expression of divine
justice but of divine love as well. And hell will exist eternally because the hearts
of some of the condemned will grow ever and ever harder against God. They will
never choose anything other than hell.
Do the denizens of hell get a “second chance,” i.e., do they have the opportunity
to decide freely to repent and instead choose heaven (and then be transferred to
heaven)? I do not know the answer to that question either—nor, in my opinion, does
any earthly person. But let me develop a proposal. It is based on three assumptions:
(1) the Bible does not tell us everything we might want to know about God, God’s
will, and our future state; (2) all people who are saved or reconciled to God are
saved or reconciled because of the life, teachings, death, and resurrection of Christ
(John 14:6); and (3) it would be unjust on the part of God to condemn people to
hell because, for reasons beyond their control, they never had faith in Christ.
So perhaps there are ways—unknown to us—by which those who die in
infancy, or who are mentally impaired, or who live and die in ignorance of Christ
can be saved through Christ. In other words, if redemption is to be found only
through Christ, and if it is God’s will that everyone be saved (I Timothy 2:4), and
if God is both loving and just, then it seems to follow that it must be possible for
all people, wherever or whenever they live or however ignorant of Christ they may
be, to come to God through Christ. Christian salvation is accordingly universally
available. God gives everyone the grace necessary for faith.
But in precisely what way is it universally available? Here I offer a proposal
(or, better, conjecture)—postmortem evangelism. (This is indeed a conjecture or
perhaps merely a hope; it is not a dogma or teaching or even a firm belief of
mine.) There is at least some support for the idea in the nearly universal Christian
consensus that allowance is made for the salvation of infants who die despite
their ignorance of Christ. It is quite true that the two cases—dead infants and
unevangelized pagans—are not precisely parallel. But the central point of similarity
is that members of both groups die in inculpable ignorance of Christ. And if it is
possible for members of one group to be saved, why not members of the other?
Does anything in the New Testament support this conjecture? The truth is, not
much. But there is, for example, the tradition of the harrowing of hell. It arises
from a few texts:
Therefore it is said, “When he ascended on high he made captivity itself a

captive; he gave gifts to his people.” (When it says, “He ascended,” what does
it mean but that he had also descended into the lower parts of the earth? He who
descended is the same one who ascended far above all the heavens, so that he
might fill all things.) (Eph. 4:8–10)
For Christ also suffered for sins once for all, the righteous for the unrighteous,
in order to bring you to God. He was put to death in the flesh, but made alive in
the spirit, in which also he went and made a proclamation to the spirits in prison,
who in former times did not obey, when God’s waited patiently in the days of
Noah, during the building of the ark, in which a few, that is, eight persons, were
saved through water. (I Pet. 3:18–20)
But they [the Gentiles] will give an accounting to him who stands ready to judge
the living and the dead. For this is the reason the gospel was proclaimed even to
the dead, so that, though they had been judged in the flesh as everyone is judged,
they might live in the spirit as God does. (I Pet. 4:5–6)
Christian tradition has often interpreted these texts (see also Matthew 12:40;
Acts 2:24–31; Romans 10:6–8) to mean that after his crucifixion and before his
resurrection appearances, Christ descended into Hades in order to rescue the Old
Testament or antediluvian righteous, who were unable to ascend into heaven until
Christ had done his atoning work. This is doubtless the biblical basis for the assertion
in the Apostles’ Creed that Christ “descended into hell.” The New Testament can
accordingly be taken at least to suggest the possibility of postmortem salvation.
(Notice also Paul’s cryptic but apparently approving comment about baptism for
the dead in I Corinthians 15:29.)
It seems to have been common teaching in the Christian Church from the time
of the Apostolic Fathers onward that Jesus, between his death and resurrection
appearances, visited Hades. Hades was the abode of the dead (roughly equivalent
to the Old Testament Sheol); by New Testament times it was said to be divided
into two sections, the abyss (or Gehenna) for evil people and paradise for the
righteous. The Fathers generally held that Christ’s descent into Hades was for
the purpose of redeeming righteous people of Old Testament times. And despite
the existence of scores of interpretations of the puzzling texts just cited, this still
seems a possible exegesis of them.
Beginning with Clement of Alexandria (AD ca. 150–ca. 213), several of the
Church Fathers (e.g., Gregory of Nazianzus in his Orations, 45.23, and Cyril
of Alexandria in his Pascal Homily, 7) argued that the descent into Hades had
the effect of rescuing righteous pagans as well—people who lived moral lives
according to their lights but never had the opportunity to be exposed to Christian
teachings. In Hades, Christ (or, Clement suggested, perhaps the apostles) preached
the gospel to them; some accepted it and so were rescued.
Clement made this move in part because he was sensitive to the charge that
God’s condemnation of ignorant pagans might otherwise be considered unjust.
Thus he asked: “Who in his senses can suppose the souls of the righteous and
those of sinners in the same condemnation, charging Providence with injustice?”10

Speaking of those who die before the incarnation of Christ, he said, “It is not right
that these should be condemned without trial, and that those alone who lived after
the advent should have the advantage of the divine righteousness.”
Clement claimed that the denizens of Hades, both Jews and Gentiles, heard
the preaching of the gospel and then either gladly accepted it or else admitted
that their punishment was just. Clement suggested two reasons why some of them
repented and believed, only the first of which I can endorse. First, he said, God’s
punishments are not retributive, but rather are “saving and disciplinary, leading
to conversion.” The second, an odd, Platonic argument, is that the disembodied
state of the citizens of Hades may have increased their susceptibility to hearing
the gospel.11
Most interestingly, Clement suggested that God’s redemptive power can even
now recall postmortem souls: “I think it is demonstrated that the God being good,
and the Lord powerful, they save with a righteousness and equality which extend
to all that turn to Him, whether here or elsewhere. For it is not here alone that the
active power of God is beforehand, but it is everywhere and always at work.”
But if the gospel was once preached to the dead, maybe this practice continues.
If so, perhaps the ignorant receive after death the chance that they never had
before to turn in faith to Christ. Perhaps they live in the postmortem (but pre-last
judgment) state that Paul seems to speak of in II Corinthians 5:8 (“we would rather
be away from the body and at home with the Lord”) and Philippians 1:23–4 (“I
am hard pressed between the two [life and death]. My desire is to depart and be
with Christ, for that is far better. But to remain in the flesh is more necessary on
your account”). And in John 5:28–9, Jesus is even reported to have said that the
dead will hear the message of the Son of God, and that when they do they will be
bodily resurrected—“those who have done good, to the resurrection of life, and
those who have done evil, to the resurrection of judgment.”12
As long as it is recognized that these are conjectures or expressions of hope
without clear or systematic biblical warrant, we might even suggest that Christ
has the power to save human beings wherever they are, even in hell.13 Some, of
course, will resist this suggestion. It is one thing—so they might say—to suggest
10
This and subsequent citations from Clement are taken from The Stromata, 6.6. See
Alexander Roberts and James Donaldson (eds), The Ante-Nicene Fathers (2 vols, Grand
Rapids, 1989), vol. 2. pp. 490–92.
11
He said, “Souls, although darkened by passions, when released from their bodies,
are able to perceive more clearly, because of their being no longer obstructed by the paltry
flesh.”
12
For a good discussion of whether the New Testament texts that I have cited should
be interpreted along the lines that I am suggesting, see John Sanders, No Other Name: An
Investigation into the Destiny of the Unevalgelized (Grand Rapids, 1992), pp. 207–8.
13
C.S. Lewis seems to suggest as much (in literary form) in The Great Divorce (New
York, 1946). This book has been a helpful myth to many Christians, including me.
that the ignorant after death receive a chance (their first) to respond positively to
the gospel, but it is quite another to suggest that those who have been condemned
receive other chances to respond positively.
But here I have a question: Is it possible that there are persons who would
respond positively to God’s love after death even though they have not responded
positively to it before death? I believe this is quite possible. Indeed, one reason for
this latest conjecture is the observation that some who hear the gospel hear it in
such a way that they are psychologically unable to respond positively. Perhaps they
heard the gospel for the first time from a scoundrel, or bigot, or fool. Or perhaps
religiously skeptical parents or teachers influenced them to reject Christianity.
Whatever the reason, I believe it would be unjust of God to condemn those who
did indeed hear the good news but were unable to accept it. This is why I suggest,
or perhaps hope, that even in hell, people can be rescued. Everyone must have a
genuine opportunity to respond positively to God.
Speaking about people who commit the unforgivable sin, Jesus said, they “will
not be forgiven, either in this age or in the age to come” (Matt. 12:32, emphasis
added). This text might well be taken to imply that people who commit lesser
sins can be forgiven of them in the age to come. And although precise translation
of this next text is difficult, the NRSV renders Hebrews 7:25 as: “Consequently
he is able for all time to save those who approach God through him.” Does the
expression “for all time” mean that once one is saved one is saved for all time
(what Calvinists call eternal security) or does it mean that one can be saved at any
time, even after death? If the second, we have here another hint of postmortem
evangelism. Finally, note Revelation 21:15, where the city of God is described as
follows: “Its gates shall never be shut by day—and there shall be no night there.”
IV
Jonathan Kvanvig had raised two interesting objections to the kind of view of hell
that I am suggesting. Let us consider them.14 Kvanvig’s first point is that choice
theories of hell (as opposed to retribution theories), especially when combined
with the possibility of postmortem escape to heaven, remove the finality that
Christians have always expected of the eschaton. He argues that if a second chance
is deserved, it is hard to see why a third would not be deserved, and then a fourth,
and so on. There could be, he says, “an infinite sequence of delays of consignment
to hell” (if escape to heaven from some sort of purgatory or holding place is
contemplated). Or if escape from hell itself is contemplated, an infinite series of
delays of permanent consignment to hell is on offer. And so we have an account of
hell that is not truly eschatological.
14
They are found in his essay “Hell,” in Jerry Walls (ed.), The Oxford Handbook of
Eschatology (Oxford, 2008); see especially pp. 418, 421–3.
Second, Kvanvig argues that what we have here is more akin to reincarnational
theories, with their endless cycles of rebirths, than Christian eschatology. Hell is
merely an extension of ordinary life in the sense that after death one is still allowed
to decide what one’s ultimate fate is to be. Progress and moral development can
continue from one life to the next.
The second criticism is easier to answer than the first. The only similarity to
reincarnation is that on the choice-plus-possible-postmortem-promotion theory,
it is possible after death to improve one’s ultimate lot by growing spiritually,
repenting, and choosing heaven. But the differences are striking:
(1) Nothing like an almost endless cycle of lives, deaths, and rebirths is part of
the theory. There are only two lives—ordinary earthly life and postmortem
life. There is never any rebirth into a wholly new earthly body.
(2) The opportunities for promotion may not be endless—perhaps there is
only one per person. Once one has heard and understood the invitation of
heaven and turned it down, there are no more opportunities.
(3) Most of all, there is no karma doctrine (an inescapable part of standard
reincarnational theories).
Religions that posit reincarnation almost always claim that one’s salvation or
liberation or enlightenment (or whatever term is preferred) is a matter of one’s
own effort. What is basically wrong with human beings is ignorance or false
consciousness; they need to understand reality correctly; and for most human
beings, it takes many, many lives to build up their karma to the point of reaching
the desired spiritual end state. In Christian theories, on the other hand, the basic
problem is not ignorance but guilt; and no matter how hard we try, we are unable
to save ourselves. There is no concept of improving one’s karma; one must simply
accept the grace and forgiveness of God.
But Kvanvig’s first point is essentially correct. If people can make postmortem
decisions that improve their ultimate lot, I accept that the finality of death is not
quite the same as it is for those Christians who do not envision the possibility of
such decisions. I believe this is, however, more a description of the theory rather
than a criticism of it.
In conclusion, let me make some brief remarks about two alternatives to hell that
I reject, viz., annihilationism and universalism.
Annihilationism is the idea is that instead of sending the reprobate to hell,
God permanently destroys them; they no longer exist. And although some biblical
texts can be interpreted along those lines (e.g., Matthew 7:13; Romans 9:22;
Philippians 1:28, 3:19; II Thessalonians 1:9), I do not think that scripture supports
annihilationism. Moreover, the theory makes moral sense only in juxtaposition
with the traditional view of hell. That is, it does seem better for God to destroy the
wicked than subject them to eternal torture. But given the view of hell that I am
working with, that point is not nearly so obvious. Would the wicked themselves
prefer annihilation to hell? I doubt it.
Universalism is the view that if hell exists at all, it is only temporary; in the
end all human beings will be with God in heaven. Does the idea of possible
postmortem repentance bring in universalism? Definitely not. I have little doubt
that some will say No to God eternally (as the Bible predicts), nor do I see any
need for a “second chance” for those who have freely and knowingly chosen in
this life to live apart from God. Perhaps God never gives up on people, but some
folks seem to have hardened their hearts to such a degree that they will never
repent and turn to God. For such people, hell as separation from communion with
God exists forever, just as it is a reality for them in this life. But perhaps some who
die in ignorance of Christ will hear the good news, repent, and be rescued. Perhaps
even some denizens of hell will do so too. Again, the key word here is perhaps.
There are no grounds to dogmatize here. I do not think we know much about the
future life. All that I am certain of is that God’s scheme for the salvation of human
beings will turn out to have been both gracious and just, probably in ways that we
do not now understand.
Let us return to the idea that God’s grace is our only hope. We deserve to be
condemned, but out of love for us, God forgives us and saves us. If our salvation is
a matter of grace alone, then one implication of that idea is worth noting. The point
is this: if hell is inconsistent with God’s love, as universalists always maintain, then
our salvation, i.e., our rescue from hell, is no longer a matter of grace. It becomes
a matter of our justly being freed from a penalty that we do not truly deserve. So
in the end, the argument that hell is inconsistent with God’s love overturns the
Christian notion of grace.15
15
I would like to thank Professors James Bradley and Dale Tuggy for their helpful
comments on an earlier draft of this chapter.
Chapter 7
Molinism and Hell
Gordon Knight
At least since the time of the early Church Fathers, there have been universalist
Christians who held that all human beings will eventually be redeemed. There
have also been those who, while denying universal salvation, nevertheless modify
the concept of hell in various ways, either by construing it as a less horrendous
separation from God or as simple annihilation. I will not survey each of these
options, but rather focus on the traditional concept of hell as a place, or perhaps a
state of being, in which the unredeemed undergo extreme and everlasting suffering.
Indeed, it is striking how such a ghoulish and surreal conception of the afterlife has
been taken for granted in some Christian circles. But ghoulish and surreal it is, and
before engaging in a limited evaluation of this doctrine, it might be well to stop
and consider the character and extent of the horrors that are thought to constitute
the eternal fate of the damned. At least, it would be useful to do this if we could,
but I doubt human imagination is sufficient to grasp an eternity of suffering. This,
of course, has not stopped preachers from engaging in lurid descriptions. Consider
this nugget from Jonathan Edwards:
It would be dreadful to suffer this fierceness and wrath for even one moment; but
you must suffer it to all eternity: there will be no end to this exquisite, horrible
misery: when you look forward, you shall see a long forever a boundless duration
before you, which will swallow up your thoughts, and amaze your soul; and you
will absolutely despair of ever having any deliverance, any end, any mitigation,
any rest at all …1
Even allowing for hyperbole (and it is not clear Edwards intends any hyperbole),
the place described by Edwards is a very bad place indeed.
But while hell presents a moral problem for all Christians who endorse the
doctrine, there are also particular problems that arise depending on how one
conceives of God’s relationship to the world. For example, anyone who holds that
God predetermines the post-mortem fate of the saved and the lost will have a much
more difficult time defending the moral legitimacy of hell than, for example, those
1
Jonathan Edwards, “Sinners in the Hands of an Angry God,” in Harold P. Simonson
(ed.), Selected Writings of Jonathan Edwards (New York, 1970), p. 112.
who advocate open theism.2 In this chapter I want to consider how the traditional
conception of hell fares if we approach it from the standpoint of a currently popular
view of God’s foreknowledge and providence, the Molinist account.
Molinism
According to many Christians, God is in complete providential control of

everything that happens in the world. Such control is understood as absolute,
including both deterministic natural phenomena and creaturely choices. Because
God has such control, we can be certain that God’s purposes for the world will be
fulfilled, not just in general terms, but with respect to each and every detail of the
entire history of the universe. This conception of divine providence has been aptly
characterized by Thomas Flint:
Being omniscient God, God has complete and detailed knowledge of this
world—its history, its current state, and its future. Being omnipotent, God has
complete and specific control over that world, a world which has developed
and will continue to evolve in accord with his sovereign and never-failing will.
Being omnibenevolent, God has used his knowledge and power to fashion and
execute a plan for his world that manifests his own moral perfection and the
inexhaustible love he bears for his creation. According to this traditional picture,
then, to see God as provident is to see him as knowingly and lovingly directing
each and every event involving each and every creature toward the ends he has
ordained for them.3
An immediate worry that comes to mind when we consider this strong view
of divine providence is that it seems to conflict with the belief in free will, at
least if we understand such freedom in a libertarian, as opposed to compatibilist,
sense. By “libertarian freedom,” I mean the sort of freedom that involves not just
freedom from external control, but also freedom to control one’s own choices
and decisions. While compatibilists have typically held that a person is free if
they can act on the basis of their choices without external restraint, libertarians
hold that a necessary condition for choice being free is that the choice itself is not
completely determined. As the name implies, compatibilists hold a person can
be free and yet have their decisions totally determined (be it by God or natural
causation), whereas libertarians hold that true freedom requires that one’s choices
are themselves undetermined. Given the assumption of libertarian freedom, the
problem can be crystallized as this: if God has control over the entire history of
the world, including, for example, my decision to drink another cup of coffee right
2
For a discussion of how an open theist may approach the doctrine of hell, see my
“Universalism for Open Theists,” Religious Studies, 42/2 (2006): 213–23.
3
Thomas P. Flint, Divine Providence: The Molinist Account (New York, 1998), p. 12.
Molinism and Hell 105
now, then it seems it is God, not me, that is really in control of my choice and I
am not really free. Some Christians, notably Calvinists, are happy to endorse this
consequence of a strong conception of divine providence for free will. But for the
majority of Christians such theological determinism is deeply troubling. For one
thing, the Calvinist doctrine contains within it the seeds of a devastating argument
from evil against the existence of God. For if God is responsible for each of our
choices, it follows that God is directly responsible for both the huge number of evil
actions that human beings have committed over the years, and, more troubling still,
the eternal fate of each creature. While proponents of Calvinistic predestination
are happy to defend this consequence of their view of divine sovereignty, many
others find the God imagined by such Calvinists a barbaric caricature of the
biblical God of love.4 Even if we set aside this theological difficulty, theological
determinism shares with its naturalistic cousin the difficulty of seeing how any
creature can be really morally responsible for his or her actions. This is not the
place to discuss this controversy in detail. In what follows I will simply assume
the position that wherever the truth may lie, it cannot be found in theological
determinism. Instead I propose to consider in depth one way philosophers have
tried to reconcile strong divine providence with libertarian free will. While this
approach, known as Molinism, has its roots in the scholastic philosophy of the
sixteenth century, it also has many recent defenders. In particular, I will examine
whether Molinism provides a conception of divine sovereignty that is morally and
logically consistent with the belief that God’s plan includes the eternal suffering
of the damned.
How is this reconciliation to be accomplished? It is here that Flint and others
refer to the distinctive contribution of the sixteenth-century Jesuit theologian,
Luis de Molina.5 Molina’s understanding of divine providence is complex, but
we may begin to understand his view by means of a contrast with another view of
God’s relationship to the total history of the universe, a view we can call simple
foreknowledge. According to simple foreknowledge, God, in virtue of being
omniscient, has exhaustive knowledge of everything that happens, past, present,
and future. My choice to drink another cup of coffee today is, to the mind of
God, just as well known as the two cups of coffee I have drunk previously. Thus
the simple view of foreknowledge preserves the traditional belief that God has
exhaustive knowledge of future events. Nothing is hidden from the mind of God.
But on this view God’s knowledge is limited to what will actually occur. God
knows what will actually come about in the actual world that God creates. In
contrast to this Molina argued that God’s knowledge, prior to creation, is much
more extensive. God does not just know what I will, in fact, choose to do; God
4
For a defense of the Calvinist position, see Daniel Strange, “A Calvinist Response
to Talbott’s Universalism,” in Robin Parry and Chris Partridge (eds.), Universal Salvation?
The Current Debate (Grand Rapids, 2003), pp. 145–68.
5
Luis de Molina, On Divine Foreknowledge, trans. Alfred Freddoso (New York,
2004).
knows what I would choose to do in any circumstance. Thus, God knows that were
I writing a paper in a coffee house (as I am now doing), I would freely choose to
drink coffee. He also knows that in other circumstances, circumstances that will
not in fact obtain, I would make other choices. In fact God knows what I would
choose in any possible circumstance I am placed in. Nor is this knowledge limited
to persons that God actually chooses to create. According to Molina, God knows
what any possible person would freely choose to do in every possible circumstance
in which such a person could be placed.
It is clear that from the standpoint of strong divine providence, this Molinist
view has several advantages over simple foreknowledge. For while the simple
foreknowledge view gives God exhaustive knowledge of how the history of the
world will play out, it is unclear how such knowledge gives God any insights that
may be useful in providentially governing the world. In brief, God’s knowledge of
what will take place happens “too late” for God to do anything about it. Suppose
it is part of God’s providential plan that I do not drink coffee this afternoon. If
this is so then how does God’s knowledge that I will drink coffee help God fulfill
God’s purposes? For if God knows that I will drink coffee, then I will drink coffee.
There is nothing God can do about it. Simple foreknowledge, far from giving God
resources for controlling how the world will turn out, rather presents God with
a fait accompli that God can do nothing about.6 Even if this worry about simple
foreknowledge can be countered, it is still clear that Molinism provides God with
much greater resources for fine tuning the history of the world. If God knows
the counterfactual truth, “were I placed in a coffee house on 5 January 2009, I
would choose to drink coffee” as well as all the other counterfactual truths about
what I would do in other circumstances, then God can place me in situations that
are best suited to fulfill God’s plans. If God wants me to drink coffee, God can
place me in a situation in which I would choose to drink coffee. If God does not
want me to drink coffee, God can place me in other circumstances in which I
do not make that choice. And so on with respect to any decision I, or any other
possible creature, may make. On the other hand, God’s knowledge of these so-
called “counterfactuals of freedom” limits the sorts of worlds that God can create.
For on the Molinist account what a possible creature would freely choose to do is
totally independent of God’s will. God knows what I would choose to do, but God
cannot change the fact that, in a certain circumstance, I would make a particular
choice (assuming God does not violate my freedom). So if God wants me to drink
coffee and there are no possible situations in which I would choose to drink coffee,
God is out of luck. Likewise, it may be that there is a situation in which I drink
coffee, but that this situation includes other factors that are inimical to God’s
plan. Maybe the only situation in which I drink coffee is one in which I am also
being tortured, for example. In this way counterfactuals of freedom are similar to
logically necessary truths. Just as God cannot create a world in which there is a
6
See John Sanders, “Why Simple Foreknowledge Offers No More Providential
Control Than the Openness of God,” Faith and Philosophy, 14/1 (1997): 26–40.
round square, so too God cannot create a world in which I freely drink coffee if the
counterfactuals of freedom are such that I never would make such a choice. This
is why Molina called knowledge of these counterfactuals “middle knowledge.”
Logically speaking, these truths that God knows are known after his knowledge
of necessary truths, yet prior to his free knowledge of how the world will actually
turn out (this last knowledge is “free” because it depends on God’s free decisions
in creation).
The Molinist view, if coherent, therefore provides a way of retaining a
libertarian theory of the will along with strong divine providence. This is one of
the advantages of the theory. Another advantage, first noted by Alvin Plantinga, is
that Molinism allows us an elegant defense against the anti-theistic argument from
evil. This argument claims that there is logical inconsistency between:
(1) God is omnipotent and wholly good;
and
(2) There is evil.
The reasoning is familiar. If God is good God would want to create a world without
evil. If God is omnipotent then God is able to create a world without evil. But there
is evil, ergo, there is no God. A traditional response to this argument has been to
appeal to free will. God cannot be blamed for the evil done by creatures because
it is us fallen mortals, not God, who bears the ultimate responsibility for evil. To
this defense J.L Mackie has retorted that God, being omnipotent, can create any
possible world, including the possible world in which free creatures freely choose
not to do evil. But, again, there is evil. So God does not exist.7 It is at this point that
Plantinga, who had not yet heard of Molina, invoked a remarkable Molinist-style
defense. Very briefly, Plantinga’s approach is to simply point out that if the truth
value of counterfactuals of freedom are not dependent on God’ will, there is no way
for even an omnipotent God to make a creature freely choose one way or another.
The best God can do is place free creatures in circumstances in which their choices
will be most conducive to God’s plan. One cannot, in other words, claim that
God’s omnipotence and goodness necessitates that God create the best possible
world for, on the Molinist account, there are possible worlds that are not feasible
for God to create. These are worlds that violate no necessary truths, but which
God cannot create because of certain brute truths about the choices free creatures
would make. In fact, Plantinga argues it may be that no free creature is such that
it will never freely commit a serious moral wrong no matter what situation it is
placed in. Thus, the goodness of free will provides the theist with a counter to the
anti-theistic claim that evil and God’s existence are logically incompatible.8
7
J.L. Mackie, “Evil and Omnipotence,” Mind, 64/244 (1955), p. 200.
8
Alvin Plantinga, The Nature of Necessity (Oxford, 1974), pp. 164–95.
Hell
As we have seen, one prime reason for rejecting the Calvinist conception of
providence is that it seems to present an intractable problem of evil. For while one
may present a serious moral challenge to any variety of Christianity that allows
for hell, the challenge becomes even stronger if the decisions on which God
judges creatures are themselves dependent on the will of God. We have also seen
that, unlike Calvinism, Molinism provides an apparent way to allow for strong
providential control without denying libertarian free will. Molinism has also
been thought useful by many in developing ways of countering the atheological
argument from evil. Now the question comes: how does Molinism square with
what is, for those who endorse it, the most difficult problem of evil, the problem
of reconciling the belief in a good and loving God with the eternal suffering of
the unredeemed in hell? We can easily see why someone may think Molinism
is preferable to Calvinism on this score, for the Molinist explicitly endorses a
robust libertarian view of free will. Those who go to hell do so because of their
own choices and decisions, not because God has forced them to do so. This is
the approach explicitly endorsed by William Lane Craig.9 Craig takes Plantinga’s
Molinistic free will defense one step further, arguing that not only finite evil, but
also the horrendous suffering of hell can be accounted for in a morally acceptable
manner if we accept the Molinist account of providence.
To understand Craig’s argument, we need to be clear about the specific moral
problem he is concerned with. Craig has no concern for the fate of those who
“reject Christ,” for he sees these people as sealing their own fate through their free
choices. What concerns Craig are only those who through no fault of their own
have never heard the gospel. While Craig allows that some of these individuals
may be redeemed through their response to “the light of general revelation” such
people are few and far between.10 The problem, as Craig sees it, lies in the belief
that there are some true counterfactuals of freedom. Take any individual who has
not heard the gospel in his lifetime, call him George. George has not heard the
gospel, yet it seems on the face of it possibly true that had he heard it, George
would have been redeemed. Yet, Craig insists that unless George is one of the
very special people who respond to general revelation, George will be damned,
even though he would have been saved if he had heard the gospel message. Here
is the “soteriological problem of evil.” A good and loving God would want to save
George. An all-powerful God would be able to save George. Yet, by hypothesis,
George is damned. In this way either the existence of a good and loving God, or
Craig’s strong variety of Christian exclusivism, is called into question.11
9
William Lane Craig, “No Other Name: A Middle Knowledge Perspective on the
Exclusivity of Salvation through Christ,” Faith and Philosophy, 6/2 (1989): 172–88.
10
Ibid., p. 176.
11
Unlike some exclusivists, Craig does not seem to allow for the possibility of post-
mortem salvation. Nor does he allow that some may be saved through Christ “anonymously,”
It is worthwhile to emphasize that Craig understands this problem in an extremely

narrow light. The problem only rises for those who have heard the gospel. He has no
concern at all for those who do not agree with the truth of Christianity after having
heard of it. For these unbelievers, eternal damnation is only to be expected from
a just and holy God. This limited understanding of the scope of the problem has
several peculiarities. First, it ignores the apparent fact that there are people who are
sincere in their non-belief and also people who are sincere in their commitment to
alternative religious traditions. Gandhi, for instance, knew of Christianity but chose
to remain committed to Hinduism. For Craig, this is enough to show that Gandhi
“rejected” Christ and thus deserves damnation. But in what sense does a sincere
Hindu reject Christ? It would seem that in order to reject Christ, one must first
believe that there is a Christ (that is, a risen Lord, not merely an historical Jesus).
But the most likely explanation of Gandhi and others who reject of Christianity is
not that they believe in the cross and resurrection and then deny it. They are not
like the demons, who, in the words of James, believe and tremble. They simply
do not believe at all. What Gandhi and others do is not reject God or Christ, but
reject the truth of the central tenets of Christianity. Perhaps sometimes such non-
belief is a matter of self-deception. One may pridefully reject what one deep down
knows to be true. This may be the case sometimes, but it is hardly plausible as a
blanket explanation for every case of a person deciding that they do not believe
that Christianity is true. Suppose further that a thoughtful Christian is persuaded
by an examination of the argument from evil that the Christian God does not exist.
We may disagree with the reasoning, but it is hardly plausible that God damns this
individual for their thoughtfulness and yet blesses the naïve Christian for whom
evil was never seen to be a problem. Or consider the example of the person who
hears the genuine gospel from preachers who themselves do not live it, who may
even be physically or sexually abusive. It is a logical error to conclude, from the
character of the messenger, that the message is wrong. But such a logical lapse is
not a moral error, and certainly not the sort of error a loving God could not forgive.
To this we may add, in agreement with Marilyn McCord Adams, that concretely
lived human life is full of innumerable contingent obstacles and dysfunctions that
render human beings incapable of an informed choice that seals our eternal fate.12
But even if Craig is mistaken about the scope of the problem, his solution, if
plausible at all, will work for many, if not all, of these cases.13 This solution is
that is, without explicit conscious acceptance.

12
Marilyn McCord Adams, “The Problem of Hell: A Problem of Evil for Christians,”
in Eleonore Stump (ed.), Reasoned Faith (Ithaca, 1993), p. 313.
13
One might worry about the case of the wavering Christian considering the problem
of evil. Is such a person saved if they were a sincere believer, and then later rejected such
belief? What if after further reflection they would have returned to the faith? I suppose a
Craig-like answer would be that if such a person dies before their faith is revitalized, then
God knows they never would return to the Church. But would it not be better, given Craig’s
eschatology, for such a person to die before their apostasy?
to attempt to apply Plantinga’s free will defense to the problem of hell. Recall
that Plantinga claimed that counterfactuals of creaturely freedom limited the
kind of world that God could create. While there is a logically possible world
in which everyone is both free and yet does not sin, these counterfactuals may,
for all we know, render such a world not feasible for God. It may even be that
there are counterfactuals of freedom that guarantee that whatever free creature
God creates, they will inevitably sin; the inevitability here is not of God’s doing.
God does not make creatures sin. It is rather a brute fact about the free choices
creatures would make. Armed with this Molinist conception of divine providence,
Craig argues in a similar vein. Succinctly put, his argument is that it is possible
that none of those who do not in their mortal lives hear the gospel would, in any
possible circumstance, become converts to Christianity. We cannot blame God
for rigging the world so that many do not hear of Christ, because God knows that
these people, no matter how much grace is bestowed on them, will never turn
to God. Just as Plantinga claimed it possible that all free creatures suffer from
transworld depravity (they all would sin in any possible circumstance), so too
Craig argues that it is possible that those who never hear the gospel suffer from
transworld damnation. And since these lost souls are lost through their free choices
and not because of divine decree, God cannot be held morally accountable for
their damnation. Craig insists that God loves each individual and wants each to be
saved; it just so happens that many of these people that God loves he also knew,
from the beginning, would end up in hell.14
The Cannon Fodder View of the Unredeemed
The first thing to notice in evaluation of Craig’s position is that the analogy
between his approach and that of Plantinga is incomplete. For Plantinga, transworld
depravity is a possible feature that all possible free creatures possess. Transworld
damnation, on the other hand, is only possessed by the many who through the
centuries have never had Christianity explained to them. Those who hear the
gospel and respond to it are, obviously, not transworldly damned. Likewise for
those few who respond to “general revelation.” Thus while Plantinga can argue
that God could not create any world that both has free creatures and also is sinless,
it appears this option is not possible for Craig. If God could create a world in
which all are saved, why didn’t he?
Craig is aware of this difficulty and tries to respond it to it. Craig allows that
it is “obvious that, all things being equal, an omnibenevolent God prefers a world
in which all persons are saved to a world containing those same persons some
of whom are lost.”15 But, Craig insists, all things are not equal. For it may be
that in order to receive an “optimal balance” between the saved and lost, God
14
Craig, “No Other Name,” p. 176.
15
Ibid., p. 182.
must create many who are lost. As Craig boldly puts it: “An omnibenevolent God
might want as many creatures as possible to share salvation; but given certain true
counterfactuals of creaturely freedom, God, in order to have a multitude in heaven,
might have to accept a number in hell.”16
Craig’s point is that whether or not a certain person responds to Christ will
depend on contingent features of the world that person finds herself in. These
features will include facts such as that so and so is or is not a believer in Christianity,
and whether he or she has heard of Christ. Thus one cannot legitimately infer
from the fact that there are, for example, one billion saved persons in this world
along with huge numbers of unredeemed, that God could create a world which
contained these redeemed persons alone, without the company of the transworldly
damned. As a matter of logic, Craig is surely correct. Just because A is feasible in
circumstance B, it does not follow that A is feasible in circumstance C. Furthermore,
we may even allow that the claim has certain plausibility in some cases. It may
be that only by observing the tragic life of a contemporary that a person can be
jarred out of the dogmatic slumber of petty materialism and respond to Christ.
But as a general claim about the relationship between saved and lost we must
conclude that it is totally implausible. For the vast majority of those who Craig
believes to be transworldly damned have, historically, had little contact with those
who hear the gospel and have the opportunity to accept or reject it. It is surely
natural to suppose that if another person’s life is going to affect my fundamental
life choices, I should at least be aware of such persons. Yet throughout much of
history, such mutual awareness between would-be Christians and those who have
never heard of Christ just did not exist. Craig may reply that there is nevertheless
an indirect connection between, for example, the population of China in 300 BCE
and later Christian conversions; but without some plausible indication how such a
connection can not only exist but also be so significant as to alter substantially the
ratio of saved to lost, the claim is incredulous.
But there is a deeper problem with Craig’s proposed eschatology. Recall that
a chief reason for rejecting theological determinism is that such a belief, when
combined with the belief that all are not saved, seriously calls into question
the goodness and loving character of God. Craig believes that by affirming the
Molinist commitment to libertarian freedom, he is able to avoid this sort of
devastating critique. But is this really so? Does the bare allowance of libertarian
free will, when coupled with Molinist providence, really reconcile the existence of
a good and loving God with the existence of hell? Let us compare the situation as
Craig understands it with the analogous position of a theological determinist. The
determinist holds that God creates human beings of various sorts and in creating
them determines that some will be damned, others redeemed. Prior to the creation
of the world, God knows what the end result will be. God knows that many will
spend eternity in hell because of God’s decree, and is, nevertheless, happy with
this situation. According to Craig’s Molinist alternative God also knows, prior to
16
Ibid.
creation, that there will be a certain ratio between saved and lost. God knows that
in creating many of God’s creatures, he is ensuring that they will spend eternity in
hell. True, God does not causally determine the choices that lead to the damnation
of these persons. Nevertheless, the counterfactuals of freedom ensure that if
these individuals are created, they will be damned. Therefore, merely in the act
of creation God is guaranteeing for many of God’s creatures an eternal life that
is much worse than never having existed at all. As Craig insists, God engages in
elaborate Molinist fine tuning to ensure the best ratio of saved to lost. On this
Molinist scheme God does not force anyone’s decision, but this does not prevent
God from being utterly manipulative in his attitude towards creation. Let us allow
that Bob would freely reject Christ in any possible circumstance. Let us further
allow, as Craig insists, that in such rejection, freely made, Bob is responsible for
his own fate. Neither of these allowances justify Craig’s claim that a Molinist,
damnation-friendly deity can also consistently be viewed as good and loving
towards each of God’s creatures. For, while in this example Bob is responsible,
so too is God. One does not, in general, lose all responsibility for an action just
because it involves another person’s free choice. If I knowingly give the keys of a
car to Sally, who happens to be drunk, I share a responsibility in the outcome, even
though it was Sally’s free decision to drive while inebriated. Likewise God, in
creating persons whom he knows will never accept Christ, shares a responsibility
for the resultant suffering in hell.
Consider also the strikingly consequentialist character of Craig’s conception of
the deity. God willingly creates a certain number of people whom he knows will be
damned so that in the end the result would consist in a greater number of damned.
How exactly this cost/benefit analysis works in the divine mind is naturally left a
little vague. It cannot be just that God aims at the best possible ratio of saved to
lost, since Craig allows that it is likely there is some world that God could create
which only contains the redeemed. Yet Craig seems willing to admit that there is
some sacrifice involved in creating people who will spend eternity in hell. What is
important is that on Craig’s view God creates some persons who will be damned
solely in order to allow that other people will be saved.
For many people this sort of approach to human lives, in which some are
sacrificed for the greater good, is unacceptable even in the case of human choices
and decisions. It is this sort of attitude towards others, an attitude which sees other
rational beings as tools to be used rather than as bearers of intrinsic value, that
underlies Kant’s famous moral dictum that we ought to treat others not merely as
a means, but also as an end. Yet if we focus simply on the human case, it is not
obvious that sometimes we may not have to treat people in just this way. While
most of us cringe at the thought of a surgeon who kills one of his patients in order
to give organ transplants to five others, those of us who are not pacifists must
also allow that sometimes it may be necessary to sacrifice some for the sake of a
greater good (e.g. the defeat of Hitler). But even here it is a part of standard just
war theory that such “collateral damage” is never to be specifically sought out.
One may allow that non-combatants will die, but one cannot pick out specific
civilians to directly kill as a means for achieving victory. And yet, according to this
Molinistic soteriology, this is exactly what God does in creation.
In the human case, perhaps such an absolutist conception of the legitimate
ends of action is naïve. Maybe there are cases in which it is permissible to directly
cause the death of another for the sake of the greater good. But even if we allow
for this, it is not clear that such consequentialist justifications transfer over to the
divine. If I am in a situation in which I must kill one person to save the lives of
100, there is no way I can prevent the death of at least one person. No matter what
I do, a dreadful event will occur. Like many less dramatic cases in human life, I
am forced to choose the lesser of two evils. But God, prior to creation, is in a very
different situation. God does not simply find himself placed, at no fault of his own,
in a situation in which the circumstances demand that some be damned eternally
to hell. God specifically chooses to create a world in which there not only is a hell,
but one that is full to the brim with suffering people. Craig, to be sure, replies to
this that these damned in hell are there through their own free choices. But, again,
this does not eliminate the cold and manipulative character that Craig is compelled
to impute to the Christian God of love. We may say to a greedy person who loses
a fortune to a con artist, that it was their own fault. But this does not exculpate the
criminal from responsibility. A factory owner who pays workers less than a living
wage is exploitative, even if each of the workers freely chooses to be employed in
the factory. When we think through all the cases in which a human being may be
morally culpable for using another human being, the vast majority of such cases
do not involve violating another person’s free will, but in treating a person’s free
choices as a way of satisfying some other end. The crucial ethical insight of Kant
is not that it is wrong to make other people do something against their will, but that
in our social relations with others we ought to treat them as having value for their
own sake, and not merely as instruments to be used for ulterior ends. Even if we
allow that we must in some tragic circumstances use another merely as a means
because the cost of not doing so is so great, we cannot apply that rationale to God.
The moral analog to Craig’s God is not that of a person stuck in a situation in
which there are no good options. God, on Craig’s view, intentionally puts himself
in such a no-win situation.
Nevertheless, one might think it is a mistake to saddle our creator with Kantian
moral concerns. God is our creator, and what is wrong for us may not be applicable
to the actions of that being whose responsibility involves the entirety of creation.
But we do not have to rely on the controversial claim that God is restrained by
moral duties to show the incoherence of Craig’s soteriology. For the same point
can be made with even more force if we focus not how God ought to act towards
us, but on God’s character as a loving God, a God who cares about the world and
each of the creatures in it. God’s love is understood to be akin to, yet infinitely
greater than, the love human parents have for their children. Furthermore,
God’s love is not merely a generalized love for the totality of creation; it is a
particular love for each individual.17 While one can argue that a general concern
for the goodness of creation justifies the creation of many destined to be lost, it
is impossible to reconcile this claim with the belief that God loves, individually,
each of the people he knows will go to hell. From the standpoint of scripture,
divine love is likened to parental love, only greater. Even if we admit that human
parental love has limitations, those limitations have little application to the much
greater love of God. No marginally decent human parent would ever intentionally
conceive a child so that its organs can be used to save the lives of others. To love
is to value an individual for their own sake, to wish for the good of that individual.
A loving God does not merely aim to create a good world; such a God is in a deep
and significant loving relationship with each person, perhaps even each conscious
creature, in that world. God does not merely want to make a good world; God
desires the good of each person in that world. It may be controversial in particular
circumstances what it is that really constitutes my or your good. Certainly each
of us can be mistaken about what the good is. But whatever my good may be,
an eternal existence consisting primarily of massive and unmitigated suffering,
suffering that serves no remedial purpose, is not it.
To conclude, while the doctrine of hell is difficult to defend on any view of
divine providence, Molinists are especially vulnerable to the charge that their
construal of God, if it is not a universalist one, forces upon us a conception of
deity as manipulative and unloving towards creatures. While Molinistic hell-
defenders are right to insist that their view is some improvement over Calvinistic
predestination, the insistence on libertarian free will by itself does not remove the
clear aura of cold manipulation from their eschatology. The God of the Molinists
is indeed one who has great providential control over creation. But far from
allowing an escape from the problem of hell, this providential control only makes
the problem more intense. God’s fine tuning of creation presents us with the image
of a deity who intentionally creates persons who God knows will suffer eternally
in hell. This deity is like the general who sends out many of his troops as cannon
fodder, as sacrificial casualties in the name of ultimate victory over the enemy.
Such a heartless strategy may lead to victory in battle, but it is hardly one befitting
the God of overflowing love and perfect goodness.
17
This point about significance of God’s particular love to individuals has been well
articulated by Marilyn McCord Adams. See her Horrendous Evils and the Goodness of God
(New York, 1999) esp. pp. 17–31.
Chapter 8
Hell and Punishment
Stephen Kershnar
In this chapter, I argue that God will not send human beings to hell. My argument
is that God would send someone to hell only if justice permits it as a means of
punishing them. Justice permits such a punishment only if someone does an
infinitely wrong act or has an infinitely bad character. Because human beings do
not meet either condition, God will not send them to hell. In the first part of this
chapter, I argue that God would not, and perhaps cannot, impose hell on human
beings. In the second part, I explore how my argument intersects with different
conceptions of hell. Before going to the argument, we need to look at the nature of
hell and just punishment.
The Nature of Hell
This chapter makes the following assumptions about hell. More specifically, it
assumes that if hell exists then it has the following features.1
1. Some Inhabitants Thesis: There are some human beings in hell. The
purpose of this assumption is to avoid the notion that hell is a location
without inhabitants.
2. No Escape Thesis: If a human being enters hell, he cannot leave. The idea
here is that once a person goes to hell, he does not leave. Purgatory might
be similar to hell in that it involves a period of suffering, but its temporary
nature distinguishes it from hell. I leave aside the issue whether hell
and purgatory are constituted by psychological conditions (for example,
suffering), geographical regions, or something else.
3. Infinite Negative Well-Being Thesis: Hell results in a person having an
infinitely negative amount of well-being.2
1
My conditions differ from the strong view of Jonathan Kvanvig because his account
does not assume infinite negative suffering and does assume the purpose of hell is punishment.
For his account, see Jonathan Kvanvig, The Problem of Hell (Oxford, 1993), 25.
2
This is the view found in such historically significant authors as Augustine, The
City of God, trans. John Healey, ed. R. V. G. Tasker (Dutton, 1972), Book 21, ch. 17 and
Aquinas, Summa Theologica, trans. Fathers of the English Dominican Province (New York,
1946). The traditional Christian view of hell is that God punishes some human beings by
The idea behind this third assumption is that persons’ lives go poorly for each
significant period of time in hell and there are an infinite number of such periods.
The “significant” locution allows that someone in hell might experience momentary
pleasures. Someone’s life goes poorly if it is worse than non-existence.
An objector might claim that an individual who does not exist has no level
of well-being rather than a zero level, the idea being that a level of well-being
is an intrinsic feature of an individual and a thing must exist in order to have an
intrinsic property. However, if one thinks that an individual should be rationally
indifferent between not existing and having a zero level of well-being, then we
can say that the value of non-existence to the individual has a value equivalent
to zero. Alternatively, negative well-being might be seen as occurring when the
things that make an individual’s life go poorly outweigh the things that make his
life go well.
I am also assuming here that suffering over time does not approach a limit.
For example, assume that well-being is measured in well-being units and that a
person’s well-being level for each successive year is as follows: 1, ½, ¼, etc. Here
even though the human being’s life goes poorly for each significant period of time
and even though there are an infinite number of such periods, he still does not have
an infinite negative well-being level because the total amount approaches a limit.3
On some accounts, hell involves an everlasting amount of suffering. Because
each additional unit of suffering (for example, -10 well-being units) merely adds a
finite number to an already finite amount of suffering, an individual never has an
infinitely negative amount of well-being. The assumption here is that the amount
of suffering a person has undergone is a cumulative property had at a time and
there is no time at which a person has had an infinite amount of it. However,
there do seem to be timeless propositions about the future (for example, Jones
will suffer each year for an infinite number of years) that entail that persons will
receive an infinitely negative amount of well-being, and hell might involve such
propositions. In any case, I shall assume this is correct. If not, my arguments can
be rephrased in terms of everlasting suffering.
Other versions of post-earthly existence involve universalism (everyone
eventually goes to heaven), escapism (everyone is able to leave hell), annihilation
(God causes persons who do not go to heaven to go out of existence), and the
view that hell is a good place (persons’ lives go well in hell, although perhaps not
as well as they go in heaven).4 The first one denies the conjunction of the Some
sending them to hell. Those sent exist there and cannot leave. Kvanvig, The Problem of
Hell, 19, 25. My analysis of hell overlaps with this traditional thesis only in so far as it
assumes that human beings exist in hell.
3
This notion is discussed in James Cain, “On the Problem of Hell,” Religious Studies,
38/3 (2002): 355–62; Charles Seymour, “Hell, Justice, and Freedom,” International Journal
of Philosophy, 43/2 (1998): 84 n5, citing Thomas Flint.
4
The universalist view can be seen in Thomas Talbott, “The Doctrine of Everlasting
Punishment,” Faith and Philosophy, 7/1 (1990): 19–43; Daniel Howard-Snyder, “In
Hell and Punishment 117
Inhabitants Thesis and the No Escape Thesis, the second denies the No Escape
Thesis, and the third and fourth deny the Infinite Negative Well-Being Thesis.
The belief in permanent hell or annihilation and that this is bad for those who
go there or are annihilated is part of the Catholic and many Protestant traditions.5
In the King James version, biblical reference to post-earthly existence can be
seen in Matthew 13:49-50; 25:41; 25:46. Post-earthly existence is found in three
famous creeds: Apostles’, Nicene, and Athanasian. Mormonism also accepts it.
The notion that many will not be saved can be seen in Luke 13:23 and Matthew
7:13-14. In addition, the New Testament appears to refer to hell. For example,
there are references to “everlasting destruction” (Thessalonians 1:9), “eternal fire”
(Jude 7), “tormented day and night for ever and ever” (Revelation 20:10).
Just Punishment
Punishment occurs when one person intentionally harms a second because of what
the second did.6 Conventional punishment occurs when an authority intentionally
harms a second because the second violated a rule. Punishment is just when it
satisfies the demands of justice. We now look at the notion of just punishment.
Defense of Naïve Universalism,” Faith and Philosophy, 20/3 (2003): 343–63. For a nice
discussion of universalism, see Michael Murray, “Three Versions of Universalism,” Faith
and Philosophy, 16/1 (1999): 55–68. The case for annihilation can be seen in Richard
Swinburne, Responsibility and Atonement (Oxford, 1989), pp. 180–84; John W. Wenham,
“The Case for Conditional Immortality,” in N. Cameron (ed.), Universalism and the Doctrine
of Hell (Carlisle, 1992), pp. 161–91; Clark Pinnock, “The Conditional View,” in W. Crocket
(ed.), Four Views on Hell (Grand Rapids, 1992), pp. 135–66; John Robinson, In then End,
God (New York, 1968); Oscar Cullman, Immortality of the Soul or Resurrection of the
Dead? (Epworth, 1958). The notion that inhabitants of hell have a positive level of well-
being can be seen in Andrei Buckareff and Allen Plug, “Escaping Hell: Divine Motivation
and the Problem of Hell,” Religious Studies, 41/1 (2005): 39–54; Andrei Buckareff and
Allen Plug, “Escapism, Religious Luck, and Divine Reasons for Action,” Religious Studies,
45/1 (2009): 63–72.
5
For the Catholic tradition, see the Catechism of the Catholic Church (paragraph
1033): We cannot be united with God unless we freely choose to love him. But we cannot
love God if we sin gravely against him, against our neighbor or against ourselves: “He
who does not love remains in death. Anyone who hates his brother is a murderer, and you
know that no murderer has eternal life abiding in him.” Our Lord warns us that we shall
be separated from him if we fail to meet the serious needs of the poor and the little ones
who are his brethren. To die in mortal sin without repenting and accepting God's merciful
love means remaining separated from him forever by our own free choice. This state of
definitive self-exclusion from communion with God and the blessed is called “hell.”
6
By “harm,” I mean “a setback to an interest.” This notion comes from Joel Feinberg,
Harm to Self (Oxford, 1984), ch. 1.
Feature 1: Right to Punish
Consider the following account of just punishment. One person justly punishes a
second if and only if the first has a right to punish the second and the punishment
severity is no greater than the proportionality ceiling.
The idea here is that just punishment is one that respects the rights of the
person who is punished. This occurs when punishment is imposed on someone
who has no claim against punishment and when the punishment takes place
within the boundaries of the right to punish, that is, it does not exceed the justice-
based ceiling on punishment severity. A proportionality ceiling is the maximum
punishment that justice allows. A just punishment might be less than the ceiling
because the person with the right to punish may waive part or all of his right to
punish. This is similar to the way in which a person who is owed money may
waive part or all of a debt.
The notion that a just punishment can only be imposed by the victim or her
authorized agent can be seen if we consider punishment in the state of nature.7
In the state of nature if Al savagely beats Betty, then we intuitively think that
Betty has a right to punish Al. However, if roving agents have already punished
Al for what he did, then Betty loses her right to punish Al, or she has a right to
impose a disproportionate punishment. The former is incorrect, because it is hard
to see what Betty has done that makes her forfeit or waive her right to punishment.
The latter is incorrect because in the absence of such a proportionality constraint,
many people could punish Al, resulting in a vastly disproportionate amount of
punishment being imposed and this intuitively seems incorrect.
Feature 2: Proportionality
The second feature focuses on the ceiling on just punishment. A punishment is

proportionate to the ground if and only if the severity of the punishment equals the
seriousness of the ground.
It is this notion that explains why we think that just punishments have a ceiling.
For example, we think that a ten-year sentence for stealing a candy bar is too harsh
and a $10 fine for a brutal rape is too light. The seriousness of the ground is the
7
The notion that the right to punish is an individual right that must be transferred
to others can be seen in Robert Nozick, Anarchy, State, and Utopia (New York, 1974), ch.
3. The underlying idea is that the right to punishment occurs because responsible agents
who wrong others forfeit their right against punishment. This forfeiture account can be
seen in Vinit Haksar, “Excuses and Voluntary Conduct,” Ethics, 96/2 (1986): 317–29; Alan
Goldman, “The Paradox of Punishment,” Philosophy & Public Affairs, 9/1 (1979): 43; A.
John Simmons, “Locke and the Right to Punish,” in A. John Simmons, Marshall Cohen,
Joshua Cohen, and Charles Beitz (eds.), Punishment (Princeton, 1995), pp. 238–52; Judith
Jarvis Thomson, The Realm of Rights (Cambridge, 1990), pp. 365–6.
product of the agent’s responsibility for an act and, perhaps also, the significance
of the right on which the act infringed or was intended to infringe.
This account of the seriousness of the ground captures two notions. First, other
things being equal, diminished responsibility entails diminished proportionate
punishment. For example, provocation, duress, and other responsibility-
diminishing features lessen the punishment that intuitively may be imposed on a
wrongdoer. We also think that insanity and infancy completely excuse the actor
from punishment. The same is true for ignorance of the relevant facts when the
agent is blameless for his ignorance.
Second, other things being equal, a person who attempts to infringe on a more
significant right should receive greater punishment than one who attempts to infringe
on a less significant right. For example, if one person tries to cut up a model’s face
and a second tries to steal her car, we intuitively think that justice permits a greater
punishment of the first. There is an issue as to whether just punishment is grounded
by attempts or outcomes. For simplicity, I shall focus on outcomes, but justice
actually focuses on attempts. This can be seen if we consider two doppelgängers
(identical persons with identical mental states—specifically motives, intentions,
desires, etc.), both of whom attempt to shoot someone. The first succeeds. The
second fails due to a defective bullet, which was something that the second could
not have predicted. If punishment rests on things for which the agent is responsible
and if the doppelgängers are responsible for the same things (same intention, mental
act, physical movement, etc.) then punishment cannot track outcome.
With these assumptions in mind, let us turn to the argument against hell.
God Will Not Send Human Beings to Hell
In this part, I provide an argument for the claim that no human beings go to hell,
and then defend the premises.
(P1) If some human beings go to hell, then God imposes it as punishment.

(P2) If God imposes hell as punishment, then justice permits God to impose an
infinite punishment on some human beings.
(C1) Hence, if some human beings go to hell, then justice permits God to
impose an infinite punishment on some human beings. [(P1), (P2)]
(P3) Justice does not permit God to impose an infinite punishment on some
human beings.
(C2) Hence, no human beings go to hell. [(C1), (P3)]
The argument probably supports a stronger thesis, namely that God cannot send
human beings to hell. The idea here is that a person can do something only if
he can choose to do it. A person can choose to do something only if he can be
motivated to choose it. If a motivation would conflict with an individual’s essential
nature, then it is not one that he can have. Because God is essentially all good, he
cannot be motivated to choose to do evil things, such as creating hell or sending

someone there.8 While I think this argument is sound, I shall focus on the more
limited claim that God will not send anyone to hell even if he could do so.
Defense of Premise (P1): If some human beings go to hell, then God imposes it
as just punishment
The idea behind this premise is that if some human beings go to hell, then
God imposes a negative choice-consequence (either directly or indirectly). If
God imposes a negative choice-consequence, then the consequence is either a
disincentive or punishment. Hell is not a disincentive. This is because there is
nothing that hell can deter that cannot also be deterred by a finite punishment,
albeit, perhaps, a very severe one.
One objection here is that God does not impose hell (either directly or
indirectly). Rather it is a choice of the persons who choose to separate themselves
from God.9 However, God intentionally makes the consequences of the choice
harsh and this makes it a punishment. Consider the analogy: Janitor-Punishment.
A school principal sets up the punishment for student fighting. If he sets up
a system whereby the janitor forcibly sodomizes fighters, then the principal is
responsible for the fighters’ suffering even if they have made themselves liable
for it.10
Similarly, if God sets up a system where persons suffer greatly for refusing to
accept him in their lives, then he punishes them.
A second objection is that God imposes hell, but it is a disincentive. A finite
disincentive will work as well for morally responsible individuals, and that is who
should be the focus here. This is because the value of freedom does not provide
a reason for God to avoid interfering with those who are insane or have another
responsibility-undermining condition. The only thing that a finite punishment
cannot deter is an infinite gain. Outside of heaven, which cannot be achieved through
wrongdoing, it is hard to see what infinite gains human beings hope to receive.
8
The notion that God is essentially all good follows from the notion that he is a
maximally great being. Maximal greatness is filled out, at least in part, in terms of maximal
intrinsic goodness. In addition, God cannot be contingently all good, because then his
goodness would depend on external factors or random and arbitrary forces. This dependence
is inconsistent with maximal greatness. The above point comes from Stephen Kershnar,
“Moral Responsibility in a Maximally Great Being,” Philo, 7/1 (2004): 97–113.
9
The notion that persons choose hell can be seen in C. S. Lewis, The Great Divorce
(New York, 1946); Jerry Walls, Hell: The Logic of Damnation (South Bend, 1992), esp. p.
13; Kvanvig, The Problem of Hell; Thomas Talbott, “Providence, Freedom, and Human
Destiny,” Religious Studies, 26 (1990): 239–41.
10
This example comes from Stephen Kershnar, “The Injustice of Hell,” International
Journal for Philosophy of Religion, 58/2 (2005): 106.
A third objection is that in allowing people to go to hell, God refuses to provide

them with a benefit rather than harming them. The idea is that hell is the separation
from God and with it comes the loss of love (including God’s love), purpose, and
community. Because there is no duty to provide such a benefit, those sent to hell
have not been unjustly punished. However, if someone can provide a benefit to
another and can do so at no cost to himself, failure to do so indicates too little
beneficence. Sending persons to heaven is a benefit that God can provide at no
cost to himself and hence his failure to do so would indicate too little beneficence,
which is impossible in a perfect being. Here is the applicable principle.
Beneficent-character principle If someone could provide a benefit to another at

no cost to anyone, and without making the world a worse place, then refusal to do
so reflects a defect in beneficence.
The objector might respond that life in heaven is only possible for a person
who willingly wants to join God. Alternatively, he might argue that life there
would not be good for someone who does not accept God or some aspect of his
love and, perhaps, also morality. The idea here is that a human being who does
not warrant heaven would suffer there because he is terribly unsuited to join God.
However, in accord with beneficence, God would then provide a life that is as
good as possible, or at least one that does not involve negative infinite well-being.
If this is not possible, then a beneficent being like God would annihilate someone
rather than consign him to hell.
Another objector might respond that a person who warrants hell is evil and that
making an evil person’s life go well makes the world a worse place.11 An example
that illustrates this claim is the common intuition that if Hitler were isolated on a
tropical island, the world would be in itself better if he were to suffer rather than
flourish, even though his condition would not affect anyone else. One concern here
is that it is not clear if everyone who warrants hell is evil. In addition, if a person’s
suffering would in itself make the world a better place, the beneficent thing for
God to do would be to annihilate him rather than send him to hell.
A different objector might claim allowing people who do not warrant heaven
to go there would impose a cost on God. Hence, beneficence does not dictate that
he send such an individual there. Even if true, a beneficent being would again opt
for annihilation.
11
The notion that additional well-being to an evil person makes the world a worse place
or at least does not improve it can be seen in Fred Feldman, “Adjusting Utility for Justice: A
Consequentialist Reply to the Objection from Justice,” Philosophy and Phenomenological
Research, 55/3 (1995): 567–85; Thomas Hurka, “The Common Structure of Virtue and
Desert,” Ethics, 112/1 (2001): 6–31; Neil Feit and Stephen Kershnar, “Explaining the
Geometry of Desert,” Public Affairs Quarterly, 18/4 (2004): 273–98.
Defense of Premise (P2): If God imposes hell as just punishment, then justice
permits God to impose an infinite punishment on some human beings
The idea here is that if there is no consequentialist override, then God acts in
accord with justice. Here there is no consequentialist override. A consequentialist
override is a reason that trumps other moral reasons and focuses on bringing about
the best outcome. If there were a consequentialist override, then God would have
the duty to create a maximally great state of affairs, but such a duty would be
impossible to satisfy because there is an infinite sequence of increasingly better
states of affairs. For example, God could have created a world with 1 happy person,
2 happy people, 3 happy people, etc. Alternatively, God could have created a world
in which the people had an average of 1 well-being unit/life, 2 well-being units/
life, 3 well-being units/life, etc. If there is an infinite sequence of increasingly
better worlds, then there would never be a consequentialist reason to bring about
any one world because this reason would always give God a reason to create a still
better world.12 Whether we focus on the total or average amount of well-being,
each world is better than the previous one and there is no limit. If God created a
world with an infinite amount of good in it, then it is unclear whether subtracting
or adding an infinite amount of good (for example, a person in heaven) makes
the world have zero value or some other value.13 Again, then, there would be no
particular state of affairs that would be maximally good.
An objector might claim that there is a consequential override in this context.
The maximally great state of affairs is the one with an infinite amount of intrinsic
goodness. He might claim that some infinities are better than others. For example,
he might claim a life that had the following infinite number of well-being level/
year 2, 2, 2, 2, … is better than this life, 1, 1, 1, 1, …, even though the total
amounts are the same. This is also true if we are talking about total well-being/life.
One principle that might explain this is the Weak Pareto principle. It states that if
each value location has more goodness in one world compared to another, then the
first is better.14 On some accounts, a stronger principle is available. On one such
principle, one world with infinite value locations can be better than a second with
infinite value locations even if the first is not better at every location. The idea is
that if relative to one finite subset the first is better, and no matter how that set is
12
I am assuming here that person-affecting principle in which one state of affairs is
better than a second only if it makes the persons who exist in both states better off. A nice
criticism of this theory can be seen in Gustaf Arrhenius, “The Person-Affecting Restriction,
Comparativism, and the Moral Status of Potential People,” Ethical Perspectives, 10/3 (2003):
185–95. Among the problems with such a theory is that it does not satisfy transitivity.
13
There are some infinite sets (power sets) that are larger than other sets if larger is
understood in terms of higher cardinality (the number of elements in a set). It is unclear how
such sets are relevant to the comparison between infinite sets of objects.
14
This principle comes from Luc Lauwers and Peter Vallentyne, “Infinite
Utilitarianism: More is Always Better,” Economics and Philosophy, 20/2 (2004): 307–30.
finitely expanded the first remains better, then the first is better than the second.15
There still, however, is no limit to goodness because there could be successive
additions to the happiness of an infinite population. For example, in the first world,
the well-being/life is 8, 8, 8, … After addition it becomes, 9, 9, 9, … Hence, there
is no maximally great state of affairs and hence no consequentialist duty for God
to override justice.
The claim that there is a consequential override can be seen in an argument by
William Lane Craig. Craig argues that God must allow some individuals to go to
hell in order to make the world a better place.16 He claims that this is because a
large number of individuals can go to heaven only if some go to hell. The reason
for this is that God has actualized a world in which there is an optimal balance
of saved and unsaved individuals. On his account, there are two reasons for this.
First, there might be individuals who suffer transworld damnation, that is, they
are lost in every world that is feasible to God. They might be lost in every world
in which they exist, or every world in which they exist and a large number are
saved. Second, they might suffer from contingent damnation. Here who is lost
and who is saved varies between possible worlds. On Craig’s account, God has
middle knowledge (knowledge of what human beings will choose under different
conditions) and hence knows, given what world he actualizes, who will be saved
and who will not.
Craig’s argument fails. If one individual’s decision were necessarily linked to
another, then the linkage would undermine responsibility and hence the basis for
sending or allowing someone to go to hell. If instead individuals’ decisions are not
so linked, then it is logically possible for everyone to be saved. More specifically,
there is a particular possible world that lacks the transworld damned and for which
it is true that all of its inhabitants are saved. If such a world is possible and if
God has middle knowledge, then he can make such a world actual and thus save
everyone. It is hard to see how one would have either more well-being units or
higher-order goods by creating a world in which some are damned. One good
is higher-order than a second if the first has more moral weight than an equal
quantity of the second.
Defense of Premise (P3): Justice does not permit God to impose an infinite
punishment on some human beings
If justice permits God to impose an infinite punishment on some human beings,

then human beings warrant an infinite punishment. If human beings warrant an
infinite punishment, then either their character or acts warrant it.
15
For defense of these principles, see Lauwers and Vallentyne, “Infinite Utilitarianism:
More is Always Better”; Peter Vallentyne and Shelly Kagan, “Infinite Value and Finitely
Additive Value Theory,” The Journal of Philosophy, 94/1 (1997): 5–26.
16
See William Lane Craig, “‘No Other Name’: A Middle Knowledge Perspective on
the Exclusivity of Salvation Through Christ,” Faith and Philosophy, 6/2 (1989): 172–88.
Character-Based Punishment
A human being’s character does not warrant infinite punishment. The idea here is
that punishment tracks desert and desert tracks character. Here is the argument for
the notion that desert tracks character. Acts and character are the most plausible
grounds of desert because these are the two things for which a person is most
responsible. Desert does not track acts. This is because if desert tracks some
feature of an act, then it tracks the motive from which the act is done. Because a
person’s character is the collection of his mental states and the relations between
them, motive is part of that collection. If desert tracks motive and motive is part of
character, then it tracks character, or at least part of it. The notion that desert tracks
motive can be seen when we consider the following examples.
4. Good motive and right act: Alice tries to save Betty’s life because she loves
Betty.
5. Bad motive and right act: Carla tries to save David’s life because she
believes that David will kill her ex-husband and Carla wants him dead.
6. Good motive and wrong act: Erica is a surgeon who tries to save Francis’s
life through a life-saving operation. Unfortunately, she is relying on another
physician’s, Gerry’s, diagnosis and it is incorrect. The combination of
incorrect diagnosis and surgery brings about Francis’s death.
7. Bad motive and wrong act: Hank kills Inez because he finds it arousing to
watch women die.
Our intuitions support the conclusion that negative desert (that is, deserved
suffering) tracks the motive from which an act is done and not its deontic status.
Positive desert also intuitively seems to track good motives. Because desert tracks
the motive from which an act is done rather than the deontic status of the act, the
former is what grounds desert.
The badness of character is a function (for example, a sum) of the virtue and
vice of particular attitudes.17 A human being has an infinitely bad character only
if he has an infinite number of mental states or one of his mental states has an
infinite intensity or duration. This rests on the following account of goodness of
a character:
17
Attitudinal theories of virtue hold that virtue consists in part or whole in having
certain attitudes. The notion that it consists in taking a pro-attitude toward the right sort
of object can be seen in Aristotle, Nichomachean Ethics 1099a7–21, 1104b3–1105a17,
1106b16–23, 1152b5–6, 1172a19–27; G. E. Moore, Principia Ethica (Cambridge, 1988),
pp. 208–11, 216–17; W. D. Ross, The Right and the Good (Indianapolis, 1980), p. 160
(one type of virtuous action is action done from a desire to bring about something good);
Roderick Chisolm, Brentano and Intrinsic Value (Cambridge, 1986), pp. 62–7; Noah
Lemos, Intrinsic Value (Cambridge, 1974), pp. 34–7, 73–7; Robert Nozick, Philosophical
Explanations (Cambridge, 1981), pp. 429–33.
Goodness of character = [(#good mental states)(intensity/good mental

state)(duration/good mental state)] − [(#bad mental states)(intensity/bad
mental state)(duration/bad mental state)].
If a character is infinitely bad, then a human being has an infinite number of

attitudes or some of his attitudes have infinite intensity. Because of human beings’
finite mental capacity, neither is true, at least during earthly life. Remember that
on the conventional account of hell, it has to be warranted at the end of earthly life.
Hence, bad character does not warrant infinite punishment.
An objector might assert that we do have infinite attitudes (for example, 1 is
less than 2, 3, 4, …). Even if this is true, it is probably not true of virtuous and
vicious attitudes. Even if there were an infinite number of virtuous and vicious
attitudes, there would still be a basic virtuous or vicious state that would be the
bearer of value.18 For example, the mental state Jones’s pleasure at the thought of
Smith’s pleasure would be basic virtue-state; and other states, for example, Jones’s
pleasure at the thought of Smith’s pleasure while in Aruba, would have value only
in so far as it contained the basic value-state.
Even if bad character grounds deserved suffering, it likely does not warrant
punishment. After all, we do not normally think that mean-spirited hermits should
be punished. In any case, we can sidestep this issue because human beings cannot
have bad enough characters to warrant hell.
Here is a summary of objections to character-based infinite punishment.
8. No punishment: Even if bad-character grounds deserves suffering, it does

not justify punishment.
9. No infinitely bad character: Human beings do not have infinitely bad
characters.
One additional argument that appears to fit in here is by Thomas Talbott. Talbott
argues that the existence of suffering beings in hell prevents the existence of
heaven because persons in heaven would feel pain at the thought of the suffering
of those in hell.19 This is unclear because it is not clear that one should feel pain
at the thought of another receiving his just deserts or punishment; and even if one
should, it is not clear that it should prevent the flourishing of those in heaven.
When a virtuous soul in heaven realizes the multitude saved and the few damned,
it might, for example, reduce his well-being level from +50 well-being units/year
to +45 well-being units/year.20 When it comes to virtuous attitudes, pain at the
18
This idea for this point comes from Fred Feldman, “Basic Intrinsic Value,”
Philosophical Studies, 99/3 (2000): 319–46.
19
Thomas Talbott, “Providence, Freedom, and Human Destiny,” 239–41.
20
The idea here, that the intensity of a pleasure or pain should be proportionate to the
intrinsic goodness or badness of the attitudinal object, can be seen in Thomas Hurka, Virtue,
Vice, and Value (Oxford, 2001), ch. 3.
thought of those suffering in hell should be weighed against the pleasure of the
thought of others in heaven.
Act-Based Punishment
A human being’s acts do not warrant infinite punishment. Just punishment tracks
seriousness of wrongdoing and human beings do not perform an infinitely serious
wrongdoing or an infinite number of wrongdoings that are in the aggregate
infinitely serious.21
Finite acts If human beings’ acts warrant infinite punishment, then either finite
or infinite acts do so. In this section, I argue that human beings’ finite acts do
not warrant infinite punishment. The first assumption is that finite acts against
human beings do not warrant infinite punishment. One reason is that an agent can
cause or attempt to cause an infinite loss only if he tries to send someone to hell.
This presupposes that hell exists and people get sent there. In the absence of an
independent reason for hell to exist, this begs the question. A second reason is that
a person can be responsible for a result (for example, Jones going to hell) only if
he intends it, that is, only if this is his goal or means. My guess is that it is rare to
find someone who intends to bring this about. Even if the agent in question were
indifferent about whether he was ending a person’s earthly or eternal life, if he did
not intend the latter he still lacks the relevant intention.
The second assumption is that finite acts against God do not warrant infinite
punishment. God does not have the right to punish. This is because human beings
do not victimize him. He might gain the right to punish on behalf of victims if the
victims transfer their right to punish to him. This is similar to the way in which,
in business, a principal can authorize an agent to act on his behalf. However,
a transfer of right requires uptake in order for the right to transfer.22 Uptake is
a speech act that constitutes acceptance. For example, a buyer must accept the
seller’s offer in order to gain the right to a good. As far as I can tell, many people
have not transferred their right to punish to God and even if others have, I see no
evidence of his uptake.
In addition, if the people who have transferred their right to punish to God are
saved, perhaps because they accept God in their lives and act accordingly, then no
one who transfers the right to punish suffers an infinite loss. Hence, even if there is
transfer and uptake, it must be done on behalf of someone who suffers an infinite
loss because he ends up in hell. Depending on what is necessary to be saved, this
is an unlikely scenario.
21
The general claim that perfect justice does not allow human beings to suffer an
infinite punishment can be seen in Marilyn McCord Adams, “Hell and the God of Justice,”
Religious Studies, 11/4 (1975): 433–47.
22
The notion that a right transfer requires uptake can be seen in Judith Jarvis Thomson,
The Realm of Rights (Cambridge, 1990), pp. 322–3.
Human beings might be thought to wrong God because they harm his creations
(analogy: property), disobey (analogy: state authority), or are ungrateful (analogy:
child and parent). In addition, because his well-being is infinitely important,
harming him warrants an infinite punishment. None of these arguments work.
First, the analogies do not apply. Humans are not property; otherwise they would
lack moral claims. The assumption here is that an individual owns something just
in case he is owed all the claims with regard to that thing. Nor do humans owe
him obedience if they have not promised it. In general, the greater wisdom of one
person does not result in his having authority over slower persons. Nor do persons
owe him gratitude if they have not actively accepted his benefits and, even if owed,
the duty may be satisfied in other ways. For example, it might be satisfied through
charity, prayer, or spreading the word about salvation. If no such duties exist, then
human beings rarely wrong God. Second, even if they do wrong God, it is unlikely
that they act with the requisite intent. Hence, God rarely, if ever, has a right to
punish them. Third, even if a human being wrongs God and even if he has the
relevant intent, it is not clear the wrong is infinitely serious. This depends on which
of God’s rights human beings infringe or attempt to infringe on. In the absence of
such an account, we cannot be confident that the wrong is infinitely serious.
One objection is that human beings wrong God when they mistreat his creations.
Jonathan Kvanvig discusses the notion that this wrong is similar to one that occurs
when a person wrongs a parent by wronging her child. Kvanvig’s idea is that
when two people stand in an intimate, immediate, and connected relationship, it is
possible to wrong one by wronging the other.23
There are a couple of problems with this objection. One problem is that it is not
clear that God stands in an intimate, immediate, and connected relationship with
human beings. Kvanvig admits that one does not wrong a parent when he harms
her adult child, and the distance between God and a human being appears to be as
great as that between a parent and her adult child. Deciding whether God has such
a relationship with humans requires a better understanding of intimate, immediate,
and connected relationships than we currently have.
A second problem is that the basic idea conflicts with what is probably the best
view of rights. If one adopts a property-rights theory of moral rights (that is, all
rights are property rights), then it is false that one wrongs a parent by wronging her
child. On this theory, at least at some point, a child owns herself and no one else
owns her. Hence, wronging the child does not infringe on her parent’s right. The
property-rights theory asserts that rights are claims to particular things, negative,
and do not conflict. This theory coheres nicely with intuitions about the rights to
life, liberty, and property.24 It also provides a clean explanation of various political
23
See Kvanvig, The Problem of Hell, pp. 35–40.
24
A clean statement of this theory is set out in Robert Nozick, Anarchy, State, and
Utopia. On some interpretations, this theory of rights can also be seen in John Locke’s
classic theories of property and government. See John Locke, Two Treatises of Government
(Cambridge, 1988), pp. 285–302.
freedoms, such as those concerning religion and speech, and how the state gains
authority. If the property-rights model is correct, then one does not wrong a parent
by wronging her child, and hence the objection does not get off the ground.
Infinite acts If human beings do not do acts that in the aggregate are infinitely
serious, then their acts do not add up in such a way as to warrant infinite
punishment.25 The notion that human beings do not do acts that in the aggregate
are infinitely serious rests on the following assumptions. A human being cannot
do an infinite number of acts during his life on earth.26 He might be able to do an
infinite number in the afterlife. However, it is not clear what acts he could do that
wrong other human beings or God and that are done an infinite number of times.
Even if there were some acts that wrong other human beings or God and that
could be done an infinite number of times, it is not clear the individual could be
responsible for doing them. If human beings do an infinite number of acts and
these acts ensure that he goes to or stays in hell, then the refusal to stop doing them
reflects a mental defect. Specifically, it reflects a defect in knowledge, desire, or
willing. For example, the defect might be ignorance (for example, not knowing to
avoid such acts), intrinsically irrational desires (for example, wanting to suffer),
or weakness of the will (for example, intending to refrain from such acts, but still
doing them). After a while, the persistence of such a defect reflects a loss of moral
responsibility, and it is unjust to punish someone under this condition.27
25
The notion that human beings do an infinite number of acts that together warrant
hell is discussed in Seymour, “Hell, Justice, and Freedom,” 78–79 and Adams, “Hell and the
God of Justice,” 433. Also, the severity of punishment might not be a sum of the seriousness
of the wrongdoings that an agent committed (perhaps measured in harm, or well-being
setback, units) because cumulative harms to one person might cause greater suffering than
the sum of individual harms to different individuals. For example, it is worse for one person
to lose all of his teeth than for 32 different people to each lose one tooth. This point and
example can be seen in Marilyn McCord Adams, “Divine Justice, Divine Love, and the Life
to Come,” Crux, 13/1 (1976–77): 14–16.
26
Note I am assuming that a fine-grained account of acts is false. Such an account
asserts that through one intentional bodily movement (for example, moving my finger)
I instantiate other properties and thereby do other acts (for example, flipping the switch
and scaring a burglar). If such an account were correct, then a person might be able to do
an infinite number of acts. For a fine-grained account, see Alvin Goldman, A Theory of
Human Action (Englewood Cliffs, 1970), pp. 1–19. For a denial of this account, see G. E.
M. Anscombe, Intention (Ithaca, 1958), p. 45; Donald Davidson, “Actions, Reasons, and
Causes,” in his Essays on Actions and Events (Oxford, 1980), pp. 4–5.
27
Marilyn McCord Adams makes a different but related point. She argues that human
beings have diminished responsibility because of their psychological flaws and as a result
it would be cruel to send them to hell. Marilyn McCord Adams, “The Problem of Hell: A
Problem of Evil of Christians,” in Eleonore Stump (ed.), A Reasoned Faith (Ithaca, 1993),
pp. 313–14. This is particularly true given the large number of psychological traits that
are influenced by genetic factors. See, for example, Steven Pinker, The Blank Slate (New
There is an issue as to whether human beings can ever do an infinite number of

acts. Every act they do adds one more to a finite number. If a human being were to
count endlessly (0, 1, 2, 3, …), he would never reach infinity.28 Acts seem to work
in the same manner. On the other hand, if an individual were to wrong another in
every significant finite period (for example, one wrong a year) and he persists for an
infinite number of years, then he would perform an infinite number of wrongs. This
is true even though there is no time at which he has completed an infinite number
of wrongs. If a just punishment can only be imposed at a particular time, then hell
is unjust. Also, if a just punishment must follow the act or acts that warrant it, and if
hell is warranted by infinite acts, then hell is unjust.29 It is worth remembering here
that on the account of hell we are assuming, no one in hell can escape it.
Summary of Act-Based Punishment
The following chart summarizes the findings about act-based punishment.
Number Wrongs Human Beings Wrongs God

of Acts
Finite Begs the question: This begs the Do not wrong God: Human beings do
question in assuming hell exists. not wrong God or attempt to do so.
Lack intention: Human beings rarely Do not intentionally wrong God:

intend to cause others infinite harm. Even if human beings wrong God,
they do not intentionally do so.
No right transfer: Some human No infinitely serious wrong: Even

beings do not transfer their right to if human beings wrong God and do
punish to God. Those that do, do so intentionally, the wrong is not
not suffer an infinite loss. infinitely serious.
No uptake: Even if human beings

transfer their right to punish to God,
he does not give uptake.
York, 2002), esp. pp. 372–8. Similarly, Thomas Talbott argues that the notion that a fully
informed agent might freely choose eternal suffering for himself is incoherent. Talbott,
“Providence, Freedom, and Human Destiny,” 228.
28
The notion that infinity cannot be achieved through successive addition can be seen
in William Lane Craig, “The Existence of God and the Beginning of the Universe,” Truth:
A Journal of Modern Thought, 3/1 (1991): 85–96.
29
For an argument that the act that grounds punishment must precede it, see Saul
Smilansky, “The Time to Punish,” Analysis, 54/1 (1994): 50–53.
Infinite No infinite wrongs: Human beings No infinite wrongs: Human beings

do not infinitely wrong other human do not infinitely wrong God.
beings.
No responsibility: Even if there No responsibility: Even if there

are infinite wrongs, the agent is are infinite wrongs, the agent is
eventually not responsible for doing eventually not responsible for doing
them. them.
Infinities: Even if human beings can Infinities: Even if human beings can
do an infinite number of wrongs, do an infinite number of wrongs,
and this is not clear, there is no time and this is not clear, there is no time
at which hell may be imposed. at which hell may be imposed.
Same as above: There is no right Same as above: Human beings do

transfer to God or, if there is, he not wrong God or attempt to do so.
does not give uptake. Even if they do, they do not do so
intentionally.
Other Theories of Hell
Theory 1: Annihilation
Hell does not fare any better if we drop one or more of the assumptions we made
about the nature of hell. Consider annihilation. This occurs when, after earthly
existence, God causes some human beings to cease to exist. This conflicts with the
Infinite Negative Well-Being Thesis (hell results in a person having an infinitely
negative amount of well-being). Annihilation is just only if God has a right to
punish human beings and, as argued above, he does not. In addition, an individual
suffers an infinite loss when he is annihilated rather than sent to heaven. If an
infinite punishment of human being is unjust, and for the reason above I think it is,
then such a punishment is unjust.
A proponent of annihilation might object that it is a natural end to human
existence, except when God intervenes to ensure that someone lives on in heaven.
This is similar to the way in which a physician who allows a person who is not his
patient to die of a curable cancer has not killed her, but merely allowed her to die.
The idea, then, is that God would not wrong anyone because no one has a claim
against him that she be kept alive for eternity.
A response to this objection is that God created persons who have this
natural end. He could have created them with a different end because mortality
is not essential to free being, good being, or any other type of being he might
create. Hence, God would be blameworthy for creating less than optimal beings.
However, if there is an infinite sequence of increasingly better beings, this might
not indicate a defect in God any more than creating a world with an infinite
amount of good signals a defect just because it is always possible for God to tack
on more good things. The real problem with this account is that it conflicts with
the Beneficent-Character Principle. That is, the refusal to benefit someone when
it has no cost to the benefactor reflects a defective character, in particular a lack
of beneficence.
Theories 2 and 3: Escapism and Universalism
The escapist theory of hell asserts that everyone is able to leave hell. It denies
the No Escape Thesis (if a human enters hell, he cannot leave) and the weaker
thesis that some people who enter hell are unable to leave it. Again, there is a
concern whether God has the right to punish people. In addition, if people can
escape, then the refusal to do so in the long run can only be explained by factors
that undermine responsibility. The only reasons an individual would not want to
join God over time is his failure to recognize, desire, or will being with God. Such
failures over vast stretches of time can only be explained by factors that undermine
moral responsibility.
Universalism asserts that everyone eventually goes to heaven. It conflicts with
the conjunction of the Some Inhabitants Thesis and No Escape Thesis. If it is
combined with the notion of purgatory, a temporary period of suffering that occurs
before some persons can go to heaven, then it assumes God has the right to punish.
In the absence of an assumption about purgatory, it is an appealing theory.
Theory 4: Hell is a Good Place
A fourth theory is that hell is a good place. It is worse than heaven in that it lacks
something that makes individuals’ lives go better, probably a personal relationship
with God or something along these lines. The motivation for this theory is that
because God is a loving being, he would not want to cause anyone to suffer
unnecessarily. This is true whether God does so directly or indirectly. One concern
about this theory is that it makes hell look like part of heaven. This issue gets
murky in the absence of an account of whether heaven and hell are locations,
states of being, or something else. In addition, if persons did not eventually leave
hell, this would likely be explained by factors that reflect a lack of responsibility.
Again, the idea is that over vast stretches of time, the only reason for an individual
to refuse to make his life go better, and do so in a way that connects him to the true
and the good, is that he suffers from a defect. This defect might be a problem with
his knowledge, motivation, intention, or volition. Any of these defects eventually
constitute a responsibility-undermining condition.
Here is a summary of the annihilation and alternative theories of hell:
Theory Rejected Assumption(s) Objections
Annihilation (God causes Infinite Negative Well- No right to punish: God does
persons who do not go Being Thesis not have (or rarely has) a right
to heaven to go out of to punish human beings.
existence)
No infinite ground: Human
beings do not warrant an infinite
punishment.
Escapism (Everyone is No Escape Thesis No right to punish: See above.
able to leave hell)
No responsibility: If persons
did not escape, this results
from factors that undermine
responsibility.
Universalism (Everyone Conjunction No right to punish: If combined
eventually goes to No Escape Thesis with purgatory, then it assumes
heaven) Some Inhabitants Thesis God has the right to punish.
Hell is a good place Infinite Negative Well- No responsibility: If persons did
(In hell, persons’ lives go Being Thesis not eventually leave, this results
well) from factors that undermine
responsibility thesis
Part of heaven: This makes hell

look like a sub-optimal part of
heaven.
Conclusion
In this chapter, I argued that God would not send someone to hell. My argument
rests on the claim that if persons go to hell, then God sends them as punishment.
If God sends someone to hell as punishment, then an infinite punishment must be
just. I then argued that human beings do not warrant an infinite punishment. They
do not have an infinitely bad character or do an infinitely blameworthy act or set
of acts. I then argued that because there is no reason to override the demands of
justice, God would not impose such a punishment. I also briefly argued that he
cannot impose such a punishment. Two other theories of post-earthly existence,
annihilation and escapism, are also false. Hence, I conclude that God does not
send anyone to hell. Now there is some good news.30
30
I am grateful to Andrei Buckareff, Andrew Cullison, Neil Feit, and Dale Tuggy for
the extremely helpful comments and criticisms.
Chapter 9
Why I Am Unconvinced by Arguments
against the Existence of Hell
James Cain
There are a number of very strong philosophical objections to the view that hell
exists, and yet I do not find them convincing. I want to try to explain why I am
unconvinced. I will begin by considering what I take to be a bad account of hell
and use that account to help single out some constraints that must be placed on any
reasonably acceptable account. In terms of those constraints I will lay out some
of the standard philosophical objections to the existence of hell and then consider
whether the objections preclude the development of an acceptable account that
incorporates certain standard features of the main traditional view. The features
I have in mind include endlessness, punishment, suffering, unhappiness, and
misery.
Consider the following claim, (C), which many people have believed:
(C) An infant who dies unbaptized will experience an endless sequence of

days in torment in flames as a form of punishment by God for having died
in a state of original sin.
The first constraint on acceptable claims about hell is what I will call the grounding
condition: any substantive doctrine about hell that asserts its existence or describes
the conditions in hell must be appropriately grounded in revelation—either in
scripture, religious tradition, or some sort of revelatory experience.1 I cannot
attempt here to explicate how the grounding condition may be met, or even how
one should decide on which scripture, tradition, or experience may be used to
ground beliefs concerning hell. One certainly needs to be careful not to be overly
hasty in claiming that a doctrine is grounded in scripture, since scripture (at least
if we take the Bible to be suitable scripture) employs exaggeration, analogy,
symbolic use of language, and other devices that can make interpretation difficult.
So without going into detail I will simply claim that (C) is not properly grounded
in scripture and tradition.
1
Of course the non-existence of hell might be taken to follow from arguments against
the existence of God. I will not be concerned with such arguments in this chapter.
The second constraint is the consistency condition.2 If a claim about hell,

claim (C) for example, is inconsistent with known facts it must of course be false.
Furthermore, if (C) is inconsistent with other claims that we have good reason
to believe or if (C) is inconsistent with such claims together with claims that we
take to be grounded in scripture, tradition, or experience that will to some extent
cast doubt on (C). If the latter claims are better grounded than (C), then (C) will
not be a tenable belief. We know that it would be unjust and morally repugnant to
punish unbaptized newborn infants as set out in (C). Furthermore the claim that
God is just and morally upright is certainly better grounded than (C), so (C) is
untenable.
The third constraint, which will be called the McTaggart condition (in honor
of John McTaggart3), goes as follows: Suppose for the moment that claim (C)
was asserted as a part of a revelation put forward by a being powerful enough
to make (C) be the case. Only a being who is extremely wicked would treat an
unbaptized infant the way that is set out in (C). But we have little reason to trust
the word of such a being. So if (C) were part of a “revelation” put forward by one
with the power to implement (C), we would have little reason to believe either (C)
or the “revelation” containing (C). In general if a purported revelation implies a
wicked character on the part of the being who is the source of the revelation, the
trustworthiness of the purported revelation is undermined. This seems to me a
strong argument.
So I hold that the grounding condition, the consistency condition, and the
McTaggart condition are all violated by (C). My primary concern, of course, is
not with (C) but with whether there might be a substantive doctrine of hell that
does not clearly violate any of the three constraints and yet reasonably fulfills the
biblical account of hell. I will not argue for the truth of such a doctrine; my purpose
is to consider whether all such doctrines must be susceptible to the main standard
objections against the existence of hell. What might such a doctrine look like? I
will sketch out what I take to be a fairly reasonable version of such a doctrine—call
it (D). (D) will then be subjected to some of the standard philosophical objections
that are found in the literature. I will lay out (D) as a series of subclaims with
comments interspersed.
(D1) There will be a final judgment after which every human being will fall into
one of two non-empty classes: (1) those that will enjoy an everlasting life
of supernatural happiness in which they experience the beatific vision of
God; and (2) all others. I will count all those in the second class as being
in hell.
2
The grounding and consistency conditions need not be independent in their
applications. One may, for example, have to appeal to considerations of consistency in
order to determine what claims are grounded in scripture.
3
See John McTaggart, Some Dogmas of Religion (New York, 1968), pp. 215–16
(section 177). See also Peter T. Geach, Providence and Evil (Cambridge, 1977), pp. 134–6.
Why I Am Unconvinced by Arguments against the Existence of Hell 135
Note that I am using the term “hell” broadly here and it is not a matter of definition
that all those in hell are miserable. I will leave it an open question whether that is
the case, or whether some of hell’s inhabitants, like Virgil and the philosophers in
Dante’s Inferno, are not in misery. If there exists a limbo this will count as a part
of hell even if those in limbo do not suffer in misery, and even if those in limbo
find a sort of happiness suitable to the human nature—though not the supernatural
happiness of those in heaven.
(D2) Those who are in hell will always remain in hell and will have a conscious
existence.
(D3) Hell is an arena in which punishment takes place. The punishments found
in hell include pains of loss as well as experiential suffering. Many will
experience inescapable endless misery there.
(D3), as well as (D1), leaves it open whether all in hell suffer punishment and
misery. Traditionally punishment in hell is divided into pains of the senses and
pains of loss. I will speak of experiential sufferings and sufferings of loss. The
former encompasses conscious states of suffering such as physical pain, anguish,
boredom, and the like. One may suffer loss without an awareness of the loss. It
is even possible to suffer loss without any corresponding experiential suffering;
for example a person who is in a vegetative state suffers a great loss but has no
experiential suffering. To the extent that one suffers anguish at a loss that will be
experiential suffering. I will not assume that all in hell are punished with experiential
suffering. So, for example, if the view that unbaptized infants do not go to heaven
is correct, then it would seem reasonable to expect, following Aquinas, that they
do not suffer pains of the senses or interior anguish as punishment.4 Among those
who suffer experientially, the nature and amount of the suffering may differ. If one
thought that an unbaptized infant does not experience suffering, it would seem
odd to think that one who died as an older child and had only committed minor
sins would experience grievous suffering. I will take it that hell does include some
who do experience grievous suffering (which may include frustration, regret, and
sorrow) but it will be left open just what form this takes. The notion of punishment
may be used analogically here; for instance it might be counted as a form of
punishment that one who has cultivated vicious character traits has to endure a life
in which those traits manifest themselves in characteristic ways—so for example
one who cultivates hatred and anger might experience unrelieved rage.
(D4) A future of hell and misery is not a mere logical or metaphysical

possibility, but a genuine threat for anyone (with one or two possible
exception cases) who in this lifetime is capable of fully functioning as a
genuine moral agent.
4
See Thomas Aquinas, “Quaestiones disputatae de malo,” trans. Richard Regan, On
Evil (Oxford, 2003), q. 5, aa. 2–3.
I will not be concerned with the question of whether most people—or even a large
percentage of people—will suffer hell. But we should not think of hell as reserved
just for the likes of Adolph Hitler and Ted Bundy. (The reason that I mention
possible exception cases is that I do not want to say that hell was a genuine threat
for Jesus and I want to leave to the side the status of Mary.)
So far (D) may seem to be an excessively weak doctrine. Most people associate
the term “hell” with a condition of inescapable endless misery. I have simply left
it an open question whether all who are excluded from heaven endure endless
misery. (D3) asserts that some will endure inescapable endless misery.
Let us turn now to objections that might be raised against (D). If the New
Testament (NT) and mainstream tradition are suitable sources for grounding
religious doctrine, then it is hard at least initially to see that the grounding condition
could be violated. On a first reading the New Testament certainly seems to endorse
the existence of hell. I do not want to state categorically here that the doctrine is
grounded in the NT. Determining what claims are grounded in the NT can be a
difficult matter. In judging whether a given doctrine is grounded in scripture one
may have to consider how it meshes with what else scripture has to say, and to do
so one may well need to appeal to considerations of consistency. I will relegate
such objections to (D) to our consideration of the consistency condition. It goes
beyond the scope of this chapter, however, to deal with the exegesis of scriptural
passages—my concern will be with broadly philosophical difficulties.
A number of objections might be raised against (D) by appeal to the consistency
condition. Some important objections go as follows:
(1) Scripture and tradition depict God as just and morally upright. The doctrine
of hell contradicts this characterization of God, for the punishment God is
said to impose exceeds what justice permits. This is especially the case
when the punishment involves unending misery and is thus infinite in
magnitude. The finite sins of a person’s lifetime cannot deserve infinite
punishment.
(2) Scripture and tradition depict God as loving. This is contradicted by the
view that God permanently excludes some from full union with him in
love.
(3) Scripture and tradition depict God as almighty and loving, and as having a
plan for the salvation of all, but this is contradicted by the doctrine of hell
which implies that the Almighty’s plan of salvation will be frustrated.
(4) Scripture and tradition pictures those in heaven as enjoying unadulterated
happiness, but the existence of hell precludes this, for those in heaven
would either be in ignorance about those in hell (and this ignorance would
adulterate their happiness) or their happiness would be marred by their
knowledge of the conditions of those in hell.
The McTaggart condition might also be used to raise objections against the doctrine
of hell. Some of these overlap substantially with objections already raised. For
example, if the infliction of hell as set out in (D) involves a gross violation of
justice as objection (1) claims, then the putative being who is the source of the
“revelation” that asserts the existence of hell would have to be wicked and thus
not a trustworthy source of revelation. So, according to this objection, doctrine (D)
lacks credibility. If objection (1) could be answered then so could this McTaggart-
style objection; thus I will not list it separately.
One might anticipate that my response to the line of objection found here and
in (1) will focus on the weak characterization of hell to which (D) is committed,
and raise the possibility that a hell meeting the conditions of (D) might be
possible without a violation of justice. Even if this rejoinder were to succeed,
there is another, indirect, version of the McTaggart-style objection that we need
to consider. It is directed at those who take the NT to be a fundamental source of
revelation grounding (D), and it goes as follows:
(5) If one accepts (D) as grounded in the NT, then one will have to recognize
that a sterner doctrine of hell is also grounded in the NT; this sterner
doctrine holds that an eternity of agony in fire awaits the great masses
of humankind. This latter doctrine falls prey to the McTaggart condition,
and thus the NT, or at least the parts of the NT that pertain to hell, are
disqualified as a source for grounding views on hell. But with this ground
removed (D) becomes ungrounded as well.
I am inclined to agree with the part of this objection that holds that it would be
unjust to inflict on those in hell an infinite future experience of torment in flames.5
So I am inclined to think that the doctrine of hell is only tenable if it does not
include such torment. But, as objection (5) suggests, we cannot simply set aside
biblical references to hellfire. One answer to this objection is to hold that biblical
references to fire are to be taken symbolically and need not imply that people will
actually suffer an infinite future experience of being burned. Let us add a new
clause to doctrine (D) which makes this explicit:
(D5) Humans stand in danger of a kind of suffering in hell that can be aptly
symbolized in terms of burning in an unending fire.
It would of course be desirable to explain how (D5) could be true without raising
a new form of the McTaggart-style objection; viz., if the experience of hell with
which we are threatened can be symbolized in this way, would not that show that its
infliction is grievously wicked, and thus that it cannot come from a credible source?
5
Here and elsewhere when I speak of suffering pain or torment in fire, I of course
mean suffering painful sensation immediately caused by the fire, where the pain is like, or
worse than, the pain we normally feel when being burned.
But, once again, if a satisfactory answer to objection (1) were forthcoming it could
be used to answer this objection, and so there is no need to list it separately.
Let us now consider whether these sorts of objections can be answered. If
(D) is in fact true then perhaps the most satisfactory response would be to give
a somewhat full account of the afterlife and show how God’s justice, power,
and love is manifest there in a way that avoids the difficulties of the objections.
Unfortunately, we do not know in sufficient detail what the afterlife (if it exists)
is like and so cannot provide such a response. A poorer, but philosophically
legitimate, substitute for such an answer to the objections would be to do the
following: first set out a (possibly fictitious) scenario in which a hell exists that
satisfies the conditions in (D); and then show that the objections do not apply to
the hell in this scenario. If that could be done it would show that the objections
are not sufficient in themselves to show that (D) is untenable. A great difficulty
confronting this method is that it requires us to make judgments about how God
might or might not be able to manifest justice, power, and love in our hypothetical
scenario, and it is hard to feel at all confident about our abilities to make such
judgments.
I am afraid that I must approach the objections in an even less satisfying way. If
the objections are sound then they will apply to any description of hell that meets
the conditions of (D). I will lay out a scenario which fulfills the conditions set out
in (D) and then explain why I am not convinced that the objections apply in this
case. In doing so I will thereby offer reasons why I do not find the objections to be
decisive refutations of (D). To make my case the scenario meeting condition (D)
need not actually be an accurate depiction of hell.
I will begin by focusing on objection (1) which deals with justice. I suspect
that our intuitions are clearer on this issue than those that pertain to divine love.
Objection (1) states that if God were to impose a hell of endless misery on a person
then God would act unjustly, for it is wrong to punish a finite sin with infinite
misery. A couple of preliminary points about the notion of punishment are in order.
I mentioned earlier that though doctrine (D) holds that God punishes people in
hell, I am allowing that the notion of punishment may be used analogically. Plato,
in the Republic, paints a picture of the unjust soul that has thrown off reason to be
ruled by the passions and the spirited part of the soul; the result is a many-headed
monster, a chaotic soul out of harmony with itself. The unhappy state of such a
soul is not normally thought of as a form of punishment, but by analogy the notion
might be applicable. God made us in such a way that a misuse of the will brings
with it a disordered state of the soul. That state of the soul can be thought of as a
“natural” punishment. If a person suffering such a punishment were immortal, it
is not clear that the punishment need be unjust on the part of God, especially if the
disorder arises from a state of the will the agent continues to embrace.
Suffering the loss of heaven—even if this is an infinite loss—also does not
seem to involve an injustice if what one has lost is not something one has a right
to in the first place. But what about the experiential suffering that results from
suffering this loss? If one feels misery at the (partial or complete) realization of
one’s losses, that (by analogy) might also be counted as punishment in cases in
which one has lost, by immoral action, what one could have had otherwise. Here
again it is not clear that the punishment need be unjust.
One generally associates hell with the idea of punishments inflicted by God
upon the recipient. One way to inflict punishment would be to bring about
disagreeable sensations; another would be to intentionally put the recipient into
situations that evoke inner states of suffering, for example states of anguish. If
objection (1) is correct, this kind of punishment would violate divine justice if the
inflicted suffering were endless. But even here I do not think that it would be an
obvious violation of justice. If the suffering resulted from the natural punishments
mentioned above or it arose as a natural reaction to one’s awareness of loss, then
even suffering of infinite duration might not be unfair if it was not wrong to give
such a being an immortal life of infinite duration. But perhaps it might be thought
that it would be immoral for God to keep a miserable being in existence endlessly,
especially if the being had no hope of its misery coming to an end and the being
had no desire to continue in existence: one might argue that God should in such
a case either have the creature’s existence come to an end or give it a meaningful
chance of recovery. For God to do otherwise, one might object, would be to
unjustly make the creature experience an infinite amount of suffering for a finite
amount of wrong on the creature’s part. But to hold that God is under such an
obligation would seem to undermine (D), for according to (D) many in hell will
experience inescapable endless misery. If God were required to provide those who
desire non-existence an escape (at least through annihilation), then it might seem
that God could not ensure that a hell conforming to (D) exists.
A lot could be said in response to this objection, but I want to focus on a crucial
feature of the objection that needs to be questioned. It is assumed that if God were
to inflict endless misery on a creature then God would be inflicting an infinite
amount of misery by making the being experience an infinite amount of suffering.
This supposition needs to be questioned. I suspect that it may be false.6
To see why I say so it will be helpful to consider a famous thought-experiment
that is often used in explaining Einstein’s theory of relativity. We are told to
imagine a pair of twin brothers, one of whom takes a voyage in a spaceship while
the other stays on earth. From the perspective of the brother on earth, the brother
in the spaceship travels at very high speeds, approaching the speed of light before
returning to earth. When the space traveler returns he has hardly aged, whereas
the brother who stayed behind is many years older. Let us add a little to this story.
Suppose that before leaving earth the space traveler experiences a pain of a short
duration, for example he stubs his toe and it hurts for one minute as timed by his
watch. During his trip he also stubs his toe and feels a pain of equal intensity
which also lasts for one minute as timed by his watch. We may even imagine
that he would describe the two experiences as involving the same amount of
6
For a more detailed discussion of this point see my “On the Problem of Hell,”
Religious Studies, 38/3 (2002): 355–62.
experiential suffering. However, if these two experiences of the space traveler’s

pain were measured by the watch of the brother on earth, we might find that the first
experience was clocked at one minute while the second was clocked at one year.
We would be wrong to think that the space traveler experienced more suffering
the second time he stubbed his toe. Rather we need to measure the amount of
experiential suffering that takes place from the point of view of the sufferer.
Now consider a further variation on the example. Imagine that the space
traveler takes off and continuously accelerates to higher and higher speeds. From
the point of view of those on earth he travels in large loops through space; each
year (by earth time) his spaceship swings by earth. He does this forever, circling
past the earth an infinite number of times, in an unending series of loops. On the
first pass he has aged ½ minute since takeoff. On the second pass (now traveling
much faster) he has aged an extra ¼ minute; on the third pass, ⅛ minute, and so
on—during each year of earth time he only ages one half the amount he aged the
year before. Throughout all of infinite (earth) time the space traveler ages just
one minute, the sum of ½ + ¼ + ⅛ + ¹/₁₆ + …. To finish out the example, imagine
that at takeoff both the space traveler and his earth-bound brother stub their toes
and suffer pain of equal intensity for the span of time which continues for the
next minute as measured out by their respective watches. That is, the earth-bound
brother feels a pain that ends when his watch says that one minute has passed, and
the space traveler feels his pain as his watch measures off (what he experiences as)
the minute after takeoff. Experientially the brothers suffer the same amount of pain.
Clearly the amount suffered is finite since the earth-bound brother’s pain ends one
earth minute after takeoff. The space traveler’s pain, though experientially finite,
is endless since his watch never reaches the one minute mark.
The point of the example is that it seems conceivable that God could impose
endless suffering on a person without the experiential amount of suffering being
infinite: God could do this by giving the person an endless but experientially finite
future. So even if, as objection (1) states, it is unfair to impose infinite punishment
on finite sin, it would not follow from that that God could not impose endless
suffering, even endless intense suffering. If a finite amount of experiential suffering
could be justly inflicted in such a way as to cause misery during the affliction, then
it would seem that the same amount could be inflicted in a way that causes misery
and yet is endless.
I want to explore a bit further the idea of endlessly enduring a finite experience.
If there is an afterlife, for all we know there may be modes of consciousness
there that are quite different from the way in which we now experience things.
Over time our conscious experience in this life consists of a flow or succession of
diverse perceptions.7 But on occasion the flow, to some extent at least, seems to
stop for a while; we might for example find ourselves focused on a single thought
or experience for a short duration of time. Consider the possibility that the afterlife
might be experienced by some as having a quality somewhat like a moment in
7
I am using the notion of perception in a broad Humean sense here.
which we do not feel a flow of perceptions. If such an afterlife were endless one
might think that it would nonetheless be experientially infinite—it would just be
one infinitely long uniform experience—and if one were experiencing suffering,
then over the course of one’s afterlife one would experience infinite suffering if
the afterlife were endless. I think that the example with the space traveler raises
a difficulty about coming to such a conclusion. Recall that in the example after
takeoff the space traveler orbiting the earth has only one minute of experiential
time but that time period never comes to an end. Imagine that the space traveler
has an experience in which the flow of perceptions seems to have stopped and
that this takes place during the last two seconds (as measured by his watch) of the
unending minute after takeoff. Here we may have an experiential approximation
of what a qualitatively uniform, experientially finite, endless afterlife might be
like. Perhaps for some hell may be a sempiternal, finitely experienced moment of
loss that stands in a simultaneity relation with the infinite progression of time.
There is something especially frightening in these scenarios in which hell is
experienced by some people as endless, yet of finite experiential duration. It would
seem that a point would be reached at which so little experiential time is left that
one would be, so to speak, endlessly at the point of death. If the suffering that
took place in that last moment were intense—whether it was, say, pain, remorse,
frustration, or a feeling of doom—then it might fittingly be symbolized in terms of
being thrown into a fire from which there is no escape, and it might be compared
to a living death in which there will be no new future thoughts or acts of will (and
no time for repentance). And yet, it would seem that the amount of experiential
suffering endured need not be more than one deserves.
I have looked at ways in which temporal features of the afterlife might play
into our assessment of the objections. I want to briefly consider some other ways
existence in hell could differ from our current mode of existence and consider
how they too might play into our assessment of the objections. A common
way of thinking of hell takes the experience of hell to be, in important ways,
a straightforward continuation of our current experience: In hell we might have
more intense experiences and perhaps new sensations, but consciousness there is
to be understood as basically similar to our current consciousness. When thinking
of hell on this view we might imagine that if someone from hell came to us and
described it we could grasp the description and get something like a concrete feel
for what hell is like. But perhaps that is an unreasonable assumption; perhaps
we can at best only get a highly abstract conception of hell. Perhaps descriptions
of hell as, for example, being like a furnace are at best figurative attempts to
convey an aspect of hell that we are currently unable to grasp in terms of what the
experience would be like. This might even be expected if the nature of the afterlife
state were dramatically different than that of our present life. Suppose that we tried
to give a rough sense of the comparison between our current life and our state in
the afterlife in terms of a ratio: our current life is to the afterlife state as X is to Y.
How might we fill in X and Y? Think of the following series of comparisons. Our
current life (in important relevant respects) is to the afterlife state as:
the mental life of a five-year-old is to that of an adult;

the mental life of an infant is to that of an adult;
the mental life of a mouse is to that of an adult human;
the mental life of goldfish is to that of an adult human;
the mental life of a slug is to that of an adult human.
As we move down the list and the comparison becomes more dramatic it becomes
plausible that if such a comparison gave us an indication of the difference between
our current life and the afterlife, then only a very abstract conception of the future
life might be available to us, and it might best be given through figurative language.
There might even be a richness and some goodness for (some of) those in hell that
is lacking in this life, though there would not be a full supernatural union with God
in the beatific vision. And this might even be the case for some who undergo quite
significant suffering.
Consider too what it might be like if our intellectual powers were far more
developed than they currently are. Given the interaction of intellect and will,
this could lead to a striking difference in the way the will is exercised. Consider
two ways in which we may form intentions with regard to future action. One
way is simply to decide to perform a given action (a simple decision). Another
is to conditionally decide that if so-and-so occurs then I will do such-and-such
(a conditional decision). If the intellect were powerful enough to at once survey
all the options which might face one in the future then it might be possible at
once to form one’s intentions so thoroughly that the will’s role in deciding future
actions would in effect already have been completed; that is, if one’s simple and
conditional decisions managed to cover all contingencies, then no new decisions
would need to be made in light of future conditions. The dynamics of changing
one’s mind and of repentance might be dramatically different in such a scenario.
Perhaps in some such scenario escape from suffering through repentance would
not be available because the necessary change of mind involved in repentance
would not be available.
Now it is time to assess: First we need to consider whether a picture of hell
can be drawn from our discussion that accommodates the doctrine of hell as set
out in (D1)–(D5). Then we need to ask how it stands in light of the five objections
presented earlier. On our picture of hell there is a final judgment in which all
humans are permanently separated into two classes, those in heaven and those in
hell. Those in hell have an endless conscious existence, though it is to be left open
whether the conscious existence of some (or all) is of a finite experiential duration.
Furthermore it may be the case that some of those in hell, even those who suffer
endlessly, have a form of life that is rich in ways that we may currently be unable
to comprehend in a way that gives us a concrete feel for their form of existence.
Many of those in hell suffer in misery endlessly. This picture of hell would seem
to accommodate (D1) and (D2). (D3) says that there is punishment in hell, some of
which involves inescapable endless misery; we have seen how this condition can
be accommodated in a way in which the suffering experienced in the punishment

is finite. (D4) says that a future of misery in hell is a genuine threat for humans who
in this lifetime are capable of fully functioning as genuine moral agents. One way
to accommodate (D4) is to hold that those who are capable of fully functioning as
moral agents face the following two dangers: (1) they capable of acting so wickedly
that they would justly deserve to have misery imposed on them in a form that is
endless, though perhaps experientially finite, and if they do act that wickedly then
God might punish them in this way; and (2) they are capable of becoming morally
corrupt to such an extent that were they to exist endlessly under certain conditions
they would be miserable as a natural consequence of their moral state, and if they
become so corrupt God may with justice place them in such an unending state
(though perhaps their time there will be experientially finite). We have already
addressed (D5), which states that we stand in danger of a kind of suffering in hell
that can be aptly symbolized in terms of burning in an unending fire.
So we must turn to the objections. The first objection says that the doctrine of hell
contradicts the moral uprightness of God since the punishments are infinite and thus
unjust because excessive. Once we allow that the experiential suffering may be finite
this objection loses much of its force. It is hard to see that the punishments outlined
above need be excessive. But perhaps, it may be said, the mere fact that the misery is
endless is sufficient to rule it out as unjust. Other things being equal, endless misery
does seem worse than misery of the same amount that comes to an end. But I do not
see why endlessness need always rule a punishment out as unjust.
The second objection says that scripture and tradition depict God as loving
and it claims that this contradicts the view that God permanently excludes some
from full union with him. This raises difficult issues I cannot pursue here. So I will
confine myself to a few remarks. I find it hard to make confident judgments about
how a being of such a different nature from us would express its love in the matters
before us. If someone asked me to state a priori how an all-perfect, omniscient,
almighty, first cause of the universe who is loving might act, I would not be able
to say much. After all, who would have expected that a loving God would create a
world like this one? There are some things I would say almost a priori about how
a loving God would not act; for example, a loving God would not send unbaptized
infants to everlasting agony in hellfire. But the problem before us now involves
much more subtle issues that make it hard to know what to say. There are issues
of freedom of the will, of God’s respect for a sinner’s freedom, of the role of
free response in love, and of original sin and grace, to name just a few. This is
further complicated by the problem noted above that we do not know what mode
of existence awaits those in the afterlife and so it is hard to fathom how God’s love
might be expressed towards those who are not in full union with him.
One might think that we could turn to revelation to help ground our claims
about how a loving God would act in this situation. I find it difficult, however, to
see how the NT could provide us strong evidence that God’s loving nature would
not permit the existence of hell, when a doctrine of hell seems embedded in the
NT. Similar remarks apply to the third objection, which holds that God’s plans for
salvation would be frustrated if there was a hell. The belief that God has a plan
for universal salvation that is incompatible with the existence of hell needs to be
grounded in revelation. But if the NT is used as a primary source for grounding
this belief, then we are faced with the problem that the doctrine of hell seems
embedded in the NT, where God’s plan appears to include the existence of hell.8
The fifth objection has already been considered, so only the fourth remains.
The fourth objection states that the existence of hell would preclude the possibility
of those in heaven enjoying unadulterated happiness, for those in heaven would
either be in ignorance about those in hell (and this ignorance would adulterate
their happiness) or their happiness would be marred by their knowledge of the
conditions of hell. This objection may seem to depend solely on familiar facts
of human psychology. But of course it really goes far beyond such facts. We
should expect the mental life of those in full union with God in heaven to be quite
different from ours now. Those in heaven may, for instance, know those in hell
with a thoroughness that far surpasses our current knowledge of ourselves. And
they may have a God-like respect for the free will of others, a God-like sense of
justice, and a God-like love of others. Perhaps under those conditions they may be
at peace with the fate of those in hell.
So I remain unconvinced by the objections to the existence of hell, though I do
not pretend to have complete answers to them. To me the possibility of hell, and of
permanent separation from full union with God, remains a genuine worry.9
8
But I do not want to press the issue, as I claim no special expertise in biblical
scholarship. Furthermore, as William Wainwright has pointed out to me, just as I treat
the notion of hellfire symbolically, it is open to those who reject hell to interpret biblical
references to hell symbolically (e.g., an annihilationist might treat reference to endless
punishment as a symbol for permanent loss of God). A full discussion of these issues is
beyond the purview of this chapter.
9
I wish to thank William Wainwright for his insightful comments on an earlier
version of this chapter.
Chapter 10
Hell and Natural Atheology 1
Keith E. Yandell
Preface
In the academy, belief in hell seems rather rare. In many seminaries and churches,
the same is true. The closest that many in the academy will come to tolerating a
discussion of hell is that it may allow the question to be raised whether religious
monotheism requires that hell exist. Within many seminaries and churches, it is
assumed that if there is a good and loving God, there is no hell. Questioning that
assumption may be a quite unpopular activity. Whether or not this is so, I do
question that assumption. My manner of doing so comes by way of considering
an argument against the existence of God. My purpose does not include giving an
exegetical study of what either the Hebrew Bible or Old Testament, or the New
Testament, says about hell. Others, better equipped than I to take on the task,
have done this.2 My focus is on the question as to whether those who believe in
God can, without inconsistency, also believe in hell—a question which some have
explicitly, and many implicitly,3 answered in the negative.
A brief word should be said about how “God” and “hell” are to be understood
here. By “God” I mean “that omnipotent, omniscient, just and loving being who
created, sustains, and providentially governs the world.” While each property
that God has, according to this concept, could itself be a subject of considerable
attention, I will assume here that the concepts of these properties are clear enough
1
Natural atheology includes attempts to prove that God does not exist by beginning
with some phenomenon generally admitted to exist and claiming that this phenomenon
would not exist if God existed. It also includes endeavors to show that the concept of God
is logically inconsistent or that religious monotheism is contradictory in some fashion. The
present instance of natural atheology endeavors to show that religious monotheistic beliefs
are logically incompatible, and that neither of the incompatible beliefs can be properly
abandoned by religious monotheists.
2
Compare, for example, David Powys, ‘Hell’: A Hard Look at a Hard Question
(Carlisle, 1997); Philip S. Johnston, Shades of Sheol: Death and Afterlife in the Old
Testament (Downer’s Grove, 2002); W.G.T. Shedd, The Doctrine of Endless Punishment
(Edinburgh, 1990); Robert A. Patterson. Hell on Fire (Philadelphia, 1995). Powys has an
extensive bibliography.
3
At least, I suspect, this is true of many of those theists who have simply not spoken
about the topic in their sermons, writings, and teaching. Neglect is not the same as denial.
But denial is often the explanation of neglect.
to bear the weight of the present discussion.4 By “hell” I mean “that condition in
which God determines that a created person who has sinned, but not repented of
her sins and asked for forgiveness, and whose sins are thus unforgiven, receives
just recompense for her sins, including her sin of not having repented and asked
for forgiveness.” Here, too, I assume that the relevant concepts are sufficiently
clear for present purposes. What I will be asking here is whether God (a being of
whom the relevant above description is true) and hell (a condition of which the
relevant above description is true) are related in such a way that “God exists and
hell exists” can express a true proposition. In a word, is the proposition expressed
by this sentence self-contradictory?
The essay that follows has four sections. The first considers a bit of natural
atheology—an argument against the existence of God—in this case an objection
centrally focused on the idea of there being a hell. The second section offers a brief
review of some of the salient history of Christian views concerning hell, making
no pretense of being either exhaustive or original. A third section considers a
little of the conceptual neighborhood in which the views presented in section two
make their homes. A final section considers the relevance of the content of section
three to the bit of atheology with which we began, and subjects that atheology to
analysis.
Section 1: Hell and Atheology
That an omnicompetent loving God would allow hell to exist is simply

unthinkable—involves a logical inconsistency—is not a new idea. It is a central
part of what motivates universalism, the view that all will be saved. It is also
salient to the rationale for annihilationism, the idea that the unredeemed cease to
exist. Here, I want to consider a specific version of an argument that there is the
aforementioned inconsistency.
Here is the bit of natural atheology in question, in briefest version.
1. Either hell exists or it does not [necessary truth].

2. If hell exists, then God does not exist.
3. If hell does not exist, then God does not exist.
4. God does not exist [from 1–3].
4
I should mention that God is omnipotent—by way of a beginning of an analysis—
does not entail If a proposition P is necessarily false, then God can make P true or If “God
makes P true” is necessarily false, then God can make P true. There is no such thing as
making a necessary falsehood true, so God not being able to do “that” is not a matter of God
not being able to do “something.”
Premise 1 is a necessary truth. The argument is valid,5 thus if the premises are true,
so is the conclusion. Things are easy this far. But why accept premises 2 and 3? In
each case, consider a sub-proof. An argument for 2 goes as follows.
Argument for premise 2: (1) Hell exists (assumption for conditional proof);
(N1) Necessarily, if hell exists, then God created persons such that their existence
overall is so undesirable that it is a bad thing for them that they exist; (N2)
Necessarily, if God created persons, none would be such that their existence is so
undesirable that it is a bad thing for them that they exist; (2) God created persons
such that their existence overall is so undesirable that it is a bad thing for them that
they exist [from 1 and N1]; (3) God did not create persons [from 2 and N2]; (N3)
Necessarily, if hell exists, then there are persons; (4) There are persons [from 1
and N3]; (5) Persons were not created by God [from 3 and 4]; (N4) Necessarily, if
God exists (and given that there are persons), then persons were created by God;
(6) God does not exist [from 4 and 5]; (7) If hell exists, then God does not exist
[1–6, close conditional proof].
An argument for premise 3 goes as follows.
(1*) Hell does not exist (assumption for conditional proof); (2*) There are
unrepentant sinners (sinners who never repent of their sins before they die)
[empirical fact];6 (N5) If there are unrepentant sinners, then if a just God exists,
God will require that justice be done regarding them; (3*) If a just God exists, God
will require that justice is done regarding unrepentant sinners [from 2* and N5];
(4*) If justice is done regarding unrepentant sinners, it occurs post-mortem (and
thus there is hell) [justice is not done pre-mortem: empirical fact—so if it is done,
it is done post-mortem]; (5*) If there are unrepentant sinners, then if a just God
exists, then there is hell [from 3* and 4*]; (6*) If a just God exists, then there is
hell [from 2* and 5*]; (7*) A just God does not exist [from 1* and 6*]; (8*) If hell
does not exist, a just God does not exist [close conditional proof, 1*–7*]; (N6) If
a just God does not exist, God does not exist; (9*) If hell does not exist, then God
does not exist [from 8 and N6].
If this line of reasoning is correct, then any variety of theism that includes or
entails the view that hell does not exist (and admits that there are sinners who die
unrepentant) is logically inconsistent.
One obvious objection comes from annihilationsim, the view that unrepentant
sinners are annihilated. On this view, hell does not exist, but God does exist. A
reply to the objection is also at hand. It is that an omnicompetent, loving God
would not create persons whose continued existence is worse for them than their
ceasing to exist would be. If persons are of the high worth that theism ascribes to
them—made in the image of God—then God will not create them knowing that
they will have to be destroyed. There is a response to this, namely that future-tense
statements (or at least those concerning the free actions of creatures) have no truth
value now; they are neither true nor false until the action occurs. This being so, it is
5
Being of the form: p or not-p; if p then q; if not-p then q; therefore q.
6
Strictly, it is something that the theist will grant to be an empirical fact.
logically impossible for God to know the truth about such matters, there being no
such truth (beforehand) to be known. This is the position that Open Theists take.
Since I take future-tense statements, even those about the free actions of persons,
to have truth value, I will not pursue this perspective.
The sort of claims specified can be viewed in different ways. One could simply
press the point that the doctrines of God and hell are logically incompatible, period,
thus contending that orthodox Christian theology, by virtue of embracing both
doctrines, has a built-in contradiction. One could argue that a just God will create
hell, given the conditions that prevail in the world (a host of unrepentant sinners), but
a loving God would not create persons that he knew would be unrepentant sinners.
One could argue that a God who was loving would find some way or other to bring
any person to freely accept His grace. All of these views have been offered and will
be briefly discussed later. But the distinctive character of the present argument is its
opening premise and its claim to deduce that God does not exist from that premise
plus the two others given, and its claim that the deduction in question contains only
necessary truths and propositions accepted as part of religious monotheism, some
of which are (with the qualification noted) empirical facts.
We now have at least the gist of the bit of natural atheology to be considered
here. It is time to consider some historical views concerning the nature of hell.
Section 2: A Brief Historical Sketch
Annihilationism
Arnobius, while not one of the greatest thinkers of the Church, rejected the Platonic
doctrine of the natural immortality of the soul and the (widely accepted) Christian
doctrine of immortality by grace. Concerning the lost and their being consigned
to hell, he writes:
They are cast in, and being annihilated, pass away vainly in everlasting
destruction … this is man’s real death, this which leaves nothing behind. For
that which is seen by the eyes is a separation of soul from body, not the last
evil—annihilation: this, I say, is man’s real death, when souls which know not
God shall be consumed in long-protracted torment with raging fire7
Here, then, is an early statement of annihilationism: the soul lasts long enough to
receive its just punishment, and then is destroyed.
7
The Seven Books of Arnobius Adversus Gentes, trans. Hamilton Bryce and Hugh
Campbell (Edinburgh, 1871), vol. 19, pp. 79–81: 2:14.
Universalism
Origen, an early universalist, tells us that God:
So … enters “like the fire of a smelting furnace” to mold the rational nature
which has been filled by the lead of evil and other impure substances which
adulterate the soul’s golden or silver nature, so to speak … since he makes the
evil which has penetrated the whole soul to disappear.8
God is a consuming fire in the sense that God refines the soul and makes it fit for
heaven, prepared for its new and far better environment. He reads the consuming
fire as purifying.
Origen also appeals to what he finds an attractive line of reasoning, saying:
For the end is always like the beginning; as therefore there is one end of all
things, so we must understand that there is one beginning of all things, and as
there is one end of many things, so from one beginning arise many differences
and varieties, which in their turn are restored, through God’s goodness, through
their subjection to Christ and their unity with the Holy Spirit, to one end, which
is like their beginning.9
While the force of the reasoning is not exactly overwhelming, this gives us one
universalist perspective.
Unending Retribution
Tertullian, remarking on Matthew 10:28, says:
If, therefore, any one shall violently suppose that the destruction of the soul
and flesh in hell amounts to a final annihilation of the two substances [body
and soul] and not to their penal treatment (as if they were to be consumed, not
punished) let him recollect that the fire of hell is eternal—expressly announced
as an everlasting penalty; and let him then admit that it is from this circumstance
that this never-ending “killing” is more formidable then merely human murder,
which is merely temporal.10
Augustine agrees:
8
Origen, Contra Celsus, trans. Henry Chadwick (Cambridge, 1965), pp. 191–2;
4:13.
9
Ibid., p. 53; 6:2.
10
The Ante-Nicene Fathers, eds Alexander Roberts and James Donaldson (Grand
Rapids, 1973), 3:570.
what God said though the mouth of his prophet, about the eternal punishment
of the damned, will come true; it will most certainly come true that “their worm
will never die and their fire will never go out”11
He adds his own line of reasoning concerning Matthew 25:41 and Revelation
20:10:
“Eternal” in the first passage is expressed in the second by “for ever and ever,”
and those words have only one meaning in Scriptural usage: the exclusion of
a temporal end. And this is why there cannot conceivably be found any reason
better founded or more evident for the fixed and immutable conviction of true
religion that the Devil and his angels will never attain to justification and to life
of the saints. There can be, I say, no stronger reason than this: that the Scriptures,
which never deceive, say that God has not spared them.12
We have, then, significant support for the view that hell is endless. While no review
will be complete, it seems appropriate to add some Reformation perspectives.
Luther, commenting on Psalm 21, offers this view:
The fiery oven is ignited merely by the unbearable appearance of God, and
endures forever. For the Day of Judgment will not last for a moment only, but
will stand in throughout eternity and thereafter never come to an end. Constantly
the damned will be judged, constantly they will suffer pain, and constantly they
will be a fiery oven, that is, they will be tortured within by supreme distress and
tribulation.13
Here eternal punishment is asserted but literal interpretation rejected.

Calvin too regards the language of fire and worm as figurative.
Now, because no description can deal adequately with the gravity of God’s
vengeance against the wicked, their torments and tortures are figuratively
expressed to us by physical things, that is, by darkness, weeping, and gnashing
of teeth, unquenchable fire, an undying worm gnawing at the heart. By such
expressions the Holy Spirit certainly intended to confound our senses with
dread.14
11
Augustine, The City of God, ed. David Knowles (Harmondsworth, 1972), p. 988;
21.12.
12
Ibid., p. 1001; 2.23.
13
Edwin M. Plass, What Luther Says, 2 vols (St. Louis, 1959), vol. 2, p. 628.
14
John Calvin, Institutes of the Christian Religion, trans. Ford Lewis Battles and ed.
John T. McNeill (Philadelphia, 1960), vol. 20, p. 1008; 3.25.22.
The “Temporal Sin Does Not Deserve Endless Punishment” Objection
To the standard objection that no sin can deserve infinite (unending) punishment
Aquinas replies:
The magnitude of the punishment matches the magnitude of the sin … Now
a sin that is against God is infinite; the higher the person against whom it is
committed, the graver the sin—it is more criminal to strike a head of state
than a private citizen—and God is of infinite greatness. Therefore an infinite
punishment is deserved for a sin committed against him.15
It is by no means clear that striking the head of a king or president is “more

criminal” (as opposed to more likely to be punished)—at least, that morally it
is worse. Further, since any sin is “against God,” it follows that any sin deserves
infinite punishment.
Augustine has his own reply:
Now the reason why eternal punishment appears harsh and unjust to human
sensibilities, is that in the feeble condition of those sensibilities under their
condition of mortality man lacks the sensibility of the highest and purest wisdom,
the sense which should enable him to feel the gravity of the wickedness in the
first act of disobedience.16
On Augustine’s view, our sense of the seriousness of sin is profoundly inadequate

and thus we find its appropriate punishment excessive.
Their replies, of course, are complementary rather that competitive. There is
another reply that is complementary to the other two. It is suggested by Luther:
Since God is a just Judge we must love and laud His justice and thus rejoice in
God even when He miserably destroys the wicked in body and soul; for in all
this His high and inexpressible justice shines forth. And so even hell, no less
than heaven, is full of God and the highest Good. For the justice of God is God
Himself; and God is the highest Good.17
The state of affairs of deserved punishment being received is itself a good state of
affairs, and thus one that God properly institutes. This comment directly concerns
whether a good God could allow hell; the reply is that God’s love is a holy love, not
a “love” that ignores sin, and a holy God will require retribution unless forgiveness
15
Thomas Aquinas, On the Truth of the Catholic Faith, Summa Contra Gentiles,
trans. Vernon Burke (Garden City, 1956), p. 216; 144.8.
16
Augustine, City of God, pp. 977, 983; 21:7, 9.
17
Plass, What Luther Says, vol. 2, p. 627.
is sought. But, in combination with the other two points just mentioned, it provides
a significant supplement.
There is at least one more (again complementary) reason. Suppose someone
clings fiercely to the illusion that his life, rather than being a gift from God, simply
naturally arose. It is his to do with as he pleases, no matter what that may be. He
may so commit himself to the principle “I am my own” that worship of God ceases
to be an option. He may, so to say, reaffirm himself in rejection of grace again and
again. In that case, the sin is one of total rejection of anything short of, or higher
than, a self-direction that knows no accepted constraint. At the best, it knows no
gratitude to God for the gift of life. Here, if (libertarian) freedom is essential to
agency, and being an agent is of great worth even if that agency is badly misused,
continued presence “in” hell may indeed be the best that God can do for him, it
being the best that he will allow to be done.
It seems quite possible to hold that the perpetually unrepentant sinner who has,
in Kant’s terms, made the principle of evil (self-love unconstrained by love for
others and for God) the ground of his choices and behavior, and who receives in
his own person the just desert of his choices, yet receive the best that he will allow
God to give.18
At this point, we have a brief survey of some theistic views regarding hell—
one on which there is no such “place” because the unredeemed are annihilated;
one on which there is no such “place” because all are redeemed; and one on which
hell is retributive and endless.
Section 3: Some Alternatives Developed
Population
There are various views regarding various things about hell. One concerns the
number of its inhabitants, more carefully as to the percentage of candidates who
will be “there.” The candidates are human beings and, often, (fallen) angels.
There is little basis for percentage estimates regarding the latter class, but there
are people who at least take it that we have strong basis for percentage estimates
regarding the former. While the estimates or predictions are not precise—say, 90
percent of the humans who ever exist—the percentage is often taken to be high.
Such terms as “most” or “many” come into play. If the alternatives are hell and
heaven, the view is that, at least concerning humans, the candidates will be quite
a lot less represented within the pearly gates than they will be represented in the
fiery pit. If it is sufficient for going to hell that one not have specifically responded
18
Those who deny libertarian freedom may want to appeal to an alleged responsibility
that attaches to compatibilist freedom, or simply embrace an incompatibilist determinist
view and appeal to God’s being able to do as God likes. I will let compatibilists, and
incompatibilist determininsts, develop their own lines of reasoning.
favorably to the Christian Gospel, then the sheer fact that more have not heard than
have heard it substantiates this broad estimate. Of course many who have heard
the Gospel message have not favorably responded to it—they have not believed
it and trusted God on its basis. They have heard, but hearing has not saved them.
So, along with those who have not heard, they also populate hell. On this view,
there is a hell and it is well filled. On this view, one can, without inconsistency,
hold that even though for the person who is “in” hell, it would be better that she
be annihilated than that she linger on in hell, it is better that she be in hell receiving
the just punishment she is receiving. It is also possible, without inconsistency, to
hold that, even for the person in hell, on the whole it is better still that he exist
even there than that he be annihilated. Here, the issue has to do with the kinds
and greatness of worth a person has, and how bad hell is. It makes sense, then, to
describe these views of hell in the following terms. On the former account, hell
is populated because there are persons who deserve the punishment being there
involves (actually, is), and their desert never ends; perhaps they continue to sin
even in hell. Contrary to a common assumption, then, one who holds this view
need not hold that the sins of this lifetime are ever enough to justify unending
retributive punishment beyond the separation from God that hell’s occupants have
chosen, but continued sin justifies continued retribution. This view we will call the
“extrinsic view” of hell: the person in hell may be such that it would be better for
him that he be annihilated, but that the state of affairs of his getting his just deserts
has a positive worth that trumps the negative worth that attaches to the person
being annihilated and so not getting his just deserts as well as his having a life that
would be better for him to have than not to have. The contrasting view we can call
the “intrinsic view,” which holds that the positive worth of the person’s existing,
even in hell is great enough to make it a bad thing that he not exist at all.
Monotheistic Denials of Hell
There are two varieties of religious monotheism on which no one is in hell. One
is universalism, the view that, in the long run, God sees to it that everyone is
saved and goes to heaven. A loving and omnipotent God will let no one perish.
The other is annihilationism, the view that those who do not go to heaven simply
cease to exist. Here, too, the idea is that a loving and omnipotent God will let no
one be in hell. It adds that being in hell is worse than not existing—any rational
and relevantly informed human who made a really defensible judgment would
prefer obliteration to occupancy in hell. On these accounts, there is—at least in
effect—no hell. “Hell” presumably refers to what would be a miserable condition,
but to a condition under which no one exists. It is not logically inconsistent for a
universalist not to think that being in hell would be a worse fate than non-existence.
For an annihilationist, it will be inconsistent to hold that continuing to exist would
be better than non-existence (or perhaps even tied in badness) when compared to
existing in hell.
One could, of course, hold that a person in hell lasts only so long as it takes for
her just deserts to have been meted out, and then she is annihilated, or that, once
they are all meted out, she goes to heaven. Then there is the notion of purgatory
as a sort of finishing school for those who will move on to heaven when they are
ready. But I do not think that considering this further fine-tuning will materially
aid us in our reflections.
Among these views, there is agreement that being in hell is not desirable. Quite
the contrary—it is very bad indeed. But that leaves open the question as to what,
in the broadest terms, makes it so bad. Very roughly: is it what those in hell could
have had but lack, or what they do have (or both) that makes hell so unattractive?
These questions are relevant to a discussion of which of the alternatives we have
canvassed is most defensible. In particular, they are relevant to the question as to
whether God can be loving and omnipotent and permit hell. A doctrine of hell will
contain at least these elements: (i) an account of the nature of hell—what is true of
May if May is “in” hell?; (ii) who, if anyone, is “there?”; (iii) if there are people
there, why are they there?
Loss, Retribution, and Literal Interpretation
A traditional view of hell sees it in terms of loss and retribution—loss of everlasting

life with God and believers, retribution for sins committed. The latter is typically
taken to involve severe suffering—perpetual pain and agony. It is hard not to
believe that some considerable degree of satisfaction is taken in detailing the
misery of the lost. What Scripture says on the topic is expounded and expanded
far beyond anything sheer exegesis would allow. The Scriptural “where the worm
turneth not and the fire is not quenched” becomes a lurid picture of unending
torture.
While “the worm” is sometimes taken to refer to literal worms eating everlasting
bodies, the reference is sometimes construed non-literally, as we saw in Calvin and
Luther. Consider also, this passage from Matthew Henry’s Concise Commentary:
It is repeatedly said of the wicked, Their worm dieth not, as well as, The fire
is never quenched. Doubtless, remorse of conscience and keen self reflection
are this never dying worm. Surely it is beyond compare better to undergo all
possible pain, hardship, and self denial here, and to be happy for ever hereafter,
than to enjoy all kinds of worldly pleasure for a season, and to be miserable
for ever. Like the sacrifices, we must be salted with salt; our corrupt affections
must be subdued and mortified by the Holy Spirit. Those that have the salt of
grace, must show they have a living principle of grace in their hearts, which
works out corrupt dispositions in the soul that would offend God, or our own
consciences.19
19
Matthew Henry, Matthew Henry’s Concise Commentary on the Whole Bible, 9 vols
(Chicago, 1981), vol. 9, pp. 41–50.
It is possible to view hell without more to retribution than loss. This is, in effect,
the treatment that hell receives in The Great Divorce by C.S. Lewis, in which
the lost can take a bus to the outskirts of heaven but go back to their dreary lives
where, among other things, there are learned disquisitions on the possibilities of
post-mortem life, should there be such a thing, and the goal of living is always
to search but never find (that would be too definite). Here, people are allowed
to confirm themselves in errant ways, receiving in themselves the recompense
for their folly. Living without God and pursuing their own interests in their own
manner is punishment enough. The retribution, then, is the loss, neither more nor
less. The reference to unquenched fire and undying worm is a warning stated in
vivid but hyperbolic terms. Hell is to be as much avoided in the next life as is
consuming fire in this. There is no need to detail how hot the fire will be.
On the loss = retribution account, it is far less clear that a good and loving God
would not allow hell, at least so long as it was better for the person to exist rather
than not, her being in hell. If her being “there” is the result of a firm free choice
she has made—her being there arises from her own doings in a condition in which
she carries on her own projects, becoming ever more confirmed in her life without
God—God has “given her up” in the light (or the darkness) of her convictions. She
follows a path of her own choosing, and is left to do so.
Two Perspectives on Hell
Two rather different views of hell go as follows.

1. Hell is the place chosen by God for those who are not among the elect,
the elect being those whom God decided, before creation, would be in heaven.
Sometimes the distinction is made between double election (God chose exactly
who would be in heaven, and who would be in hell) and single election (God only
chose exactly who would be in heaven). Either way, however, the result is the
same. Suppose there are 100 people—50 whose last names begin with “A” and
50 whose last names begin with “B”—clinging to parts that are all that remains
of their ship. Another ship comes along whose captain decides to rescue all the A-
people and does not decide to save any of the B-people. Thus the As are on board
the ship as it leaves the Bs to their unhappy fate. When the As object to the Bs
being left in the water—there was room for them on the ship as well—the captain
replies that he did not decide not to save the Bs, but rather just did not decide to
do so. After all, he did not have to rescue any, and it was gracious of him to do
what he did. Since, he says, it would not have been wrong of him to rescue none,
it was not wrong (and so not unjust) for him to rescue only some. It may be added
that while the saved are trophies of God’s love, the lost are trophies of God’s holy
wrath. Along with this line there typically goes the view that those in hell deserve
to be there even though they were chosen, before creation, to go to hell. This often
is defended in terms of what philosophers call compatibilist freedom, where one
is free to perform an action at a time if one did perform it at that time, wanted to
do so, and did so because of what one wanted, it mattering not at all if one could
have not wanted or could have quelled the desire and refrained from the action.
Put very roughly, so long as the causal chain that leads to the action runs through
one’s consciousness in the did/wanted to do/did due to the want manner, one is
free with regard to what one did. Thus, we are told, Jim can be guilty of murder,
and properly held responsible for murder, even though the causal chain leading to
the murdering began just after creation and inevitably led to the murderous action.
The did/wanted to do/did due to the want provides a barrier to the responsibility
going back in the causal chain to God, or indeed beyond Jim. We can call this
view, put in terms of either single or double election, the election view of hell. God
graciously saves the chosen, and either chooses who will be lost or just does not
choose to save them (a distinction whose worth escapes me).
2. Hell is the “place” where people are who have freely rejected God’s grace—
have not responded, or have responded poorly, to the religious knowledge provided
them. They are “there” due to their own choices, or their own neglect to choose.
After all, the results of one’s ignoring, and those of one’s rejecting, an offer are
identical so far as accepting the offer goes; neither way does one accept.
The final section develops the second alternative.
Section 4: Atheology Revisited
It is time to return to the argument with which we began. The argument was valid,
and the first premise was a necessary truth. So if premises 2 and 3 are true, then
the conclusion is true. For each premise, there was a sub-proof—an argument,
distinct from the original but in its support. The sub-proof for premise 2 contains
four propositions said to be necessary truths:
(N1) Necessarily, if hell exists, then God created persons such that their
existence overall is so undesirable that it is a bad thing for them that they
exist.
(N2) Necessarily, if God created persons, none would be such that their
existence is so undesirable that it is a bad thing for them that they exist.
(N3) Necessarily, if hell exists, then there are persons.
(N4) Necessarily, if God exists (and given that there are persons), then persons
were created by God.
One premise—that hell exists—is assumed for a conditional proof. The strategy of
a conditional proof is to assume some proposition p and to show that, given p, you
can validly traverse true premises until you reach proposition q. Then you close the
argument with a statement of what you (claim to) have proved, namely, If p then
q. This “dismisses the assumption.” The other premises follow from preceding
steps in the argument, and so need not be considered separately. The question here
is: Are (N1) through (N4) necessary truths? Each must be true in order for the
sub-proof to succeed. (N3) and (N4) are safe ground for the critic. Given that, if
hell exists, a condition exists in which there are persons in a highly unsatisfactory
state, (N3) is indeed a necessary truth. (N4) is at least typically held by religious
monotheists, who take it that everything that exists, but might not have existed,
is created, and sustained in existence, by God. So the critic can appeal to (N4) as
typically granted by theists. We are left with (N1) and (N2).
(N1) says that any person who is in hell would have a life that it would be
better for her not to have—a life such that having it is so bad that, if she were fully
relevantly informed, fully clear-headed, and fully rational, she would choose that
it (and hence she) cease. It would be better for her that her continued existence
end. (N2) says that if God exists, God will not create any person whose life is so
unbearable that it would be better for her that she be annihilated than that she
continue.
Both claims will be disputed. One can say that the state of affairs of, say,
Tom’s receiving the just deserts of his sins trumps it being better for Tom that
he be annihilated. The former has positive worth that justifies the claim that,
overall, things are better if the former state of affairs obtains, and the latter does
not, rather than the reverse. If this is true, then perhaps God can create someone
whose existence God knows will be so bad that it would become better for him to
be annihilated than for him to live on. On balance, on this view, this might well
be the right choice. Alternatively, perhaps God does not know the future—perhaps
statements that are tensed to the future have no truth value, since what will make
them true or false has not yet arrived. So God might create unfortunate Tom and
not know that Tom will be unfortunate. Perhaps God would not have created Tom
had God known, but God did not know. No doubt there are other replies, but these
will have to suffice here.
Neither response convinces me. I can explain why by sketching a view that
there is not space to fully develop, much less defend. According to religious
monotheism in the Semitic tradition, God created Tom in God’s image. According
to Christianity, God paid humanity the ultimate compliment of becoming incarnate
in Jesus of Nazareth. One philosophical manner of expressing at least something
very like this is to say that persons have dignity rather than price. At least part of
loving someone in friendship, let alone agape, is respecting this dignity.20
A concrete consequence of our having this status is that if a parent loses a chair
or a parrot, it is possible to replace it with another. If it is your fault, an apology
and a replacement chair or parrot is appropriate. If a parent loses a child due to
your actions, an apology and a replacement child is not appropriate. Here, “equals
for equals” does not make things right. Persons have intrinsic worth; buying you
a new one when you accidentally kill the original is simply a new immoral action.
Another concrete consequence is that even if Tom is a rotten fellow, it is not right
20
That Paul in Romans says that we have no business in judging our Maker does not
entail that God does not love those made in God’s image in a way that includes recognition
of the intrinsic dignity—dignity based on the nature of persons, the imago dei—given to
created persons. Unfortunately, this distinction seems not often made.
that he be taken out to the forest and recreationally hunted. Even rotten Tom cannot
rightly be only used for the pleasure, or even the benefit, of others.
It seems to me proper to distinguish between two sorts of worth or value that
attach to persons. One is metaphysical worth. The environmental movement has
reintroduced the idea of such worth into our thought. Their claim is that the spotted
owl, and the virgin forest, has this feature: it is good that they exist. Thus it is
intrinsically wrong to wantonly destroy them. The movement does not typically
rest on the claim that spotted owls have the moral law written on their hearts.
Nonetheless they claim that it is wrong to get rid of the spotted owl. Sometimes
the curious view is held that it is the species that has intrinsic worth, but not
any member thereof, but we need not get into that here. I take that any value
“the species” has is entirely vested in the actual owls, and I grant them natural
value. Natural value is non-moral worth—metaphysical worth. A person has very
considerable metaphysical value in virtue of being a person.
The other relevant sort of worth is moral. In contrast to metaphysical, this
worth is not “built into” a person by being created. It develops over time and is
a function of a person’s choices and actions. There is no guarantee that a person
having metaphysical value will come to have great moral worth—become a really
good person. There is a deep connection between metaphysical and moral worth.
Coming to have (positive or negative) moral worth is constituted by developing
into maturity the nature one has as a person. The Ten Commandments are not rules
set down arbitrarily by a cosmic policeman out to give tickets. They are guidelines
for a life that is fully human. What matters most morally is the sort of person one
becomes, the goal being to use those capacities that make one distinctively human
in such a way that one flourishes as a person. This requires intentionally helping
others to flourish and treating them with the dignity their nature justifies. So acting
toward other persons is a great deal of what loving God includes.21
This sketch gives the barest outline of a perspective on the value of persons.
Even so, it raises a relevant question: What happens if we include a person’s
metaphysical worth into our consideration of whether being in hell might not still
be better for the person so “located” than not existing at all? There are two aspects
of hell according to traditional doctrine. One concerns loss—the person in hell
lacks a right relation to God, the forgiveness of sins, the new life of the redeemed.
She is cut off from anything like the flourishing she might have had. The other is
retribution. Spurning mercy, she gets justice. In a word, she both gets what she
deserves, and lacks what grace offers. The depth and breadth of the loss is sufficient
to justify the (to put it mildly) strong warnings concerning hell that one finds in
Christian Scripture. But there is also talk of retribution. One biblical suggestion
as to what this might include is Paul’s bleak description of the unrepentant living
“life without God, in the world” and in which God has “given them up to their own
desires.” At least part of the retribution is that God gives them what they want,
21
The remarks here are not intended to cover the whole scope of flourishing as
person—of realizing the potential that is in the imago dei.
thereby allowing tragic self-harm. George McDonald’s sentiments in C.S. Lewis’s

The Great Divorce are apt: The principle of hell is “I am my own” and “hell is
the best God can do for those who allow God to do no more.”22 This prospect of
being left to one’s own resources and devices is enough to justify the Scriptural
language without there being literal fires burning, and undying worms eating, the
citizen of hell.
Ironically, protests that God allowing hell would be wicked gain in apparent
plausibility where greater worth is ascribed to persons. The higher the being, the
worse his fall. But the higher the worth of a being, the worse his annihilation.
The higher the being, the worse her being allowed to come as close to destroying
herself as she can. Hence the higher the being, the more appropriate that he be
in hell, given the alternatives. A high view of persons then, given unrepentant
sinners, argues not against hell, but for it.
It is time to turn to the sub-proof for premise 3. Here, I believe, things are more
promising. The sub-proof for premise 3 also contains putative necessary truths:
(N5) If a just God exists, God will require that justice is done regarding
unrepentant sinners.
(N6) If a just God does not exist, then God does not exist.
The remaining premises are an assumption for a conditional proof (and so is

“dismissed”), a couple of “empirical facts” that theists typically will grant, and
premises that follow from other premises and so need not be considered separately.
(N6) says that being just is an essential property of God; no being lacking it can be
God. (N5) offers one consequence of God’s being just. Both seem correct. Suppose
that it will be the consequence of divine justice that any unrepentant sinners receive
justice. Then, given that there are unrepentant sinners, it does follow that there is
a hell in the sense defined earlier. What its intensity and duration will be does not
follow from the sub-proof for premise 3, but there is no reason why it should.
Assuming that God does not just “let the unrepentant sinner off scot-free”—that
doing so is neither just nor loving—the sub-proof for premise 3 seems sound and
valid. It is a successful argument. This, of course, is little comfort for the critic if
premise 2 is unfounded.
There are, then, such considerations as these:
1. There is great positive worth in being a person.

2. There is great positive worth in being a “realized” person—loving God,
and others as oneself.
3. One cannot have any of the latter worth without having the former.
4. One can have the former worth while having very little of the latter.
5. Having little of the latter worth does not cancel out the former worth.
22
These claims are relevant to our fundamental question as follows:
A. Given 1, it is possible that a person have little by way of “realization” as a

person, due to his own choices and actions, and yet it be better, for intrinsic
reasons, that he continue to exist rather than being annihilated—better for
him.
B. Given 1, it is possible that a person in hell who has significant “realization”
has more worth, so there are more grounds, for intrinsic reasons, that she
exist rather than being annihilated—better for her.
If persons have dignity rather than price in virtue of being created in the image
of God, A and B make basic points. Persons are not merely to be used, even for
the sake of other persons. This is why A and B are central to reflecting about hell.
They raise questions about whether the state of affairs being a person receiving
just deserts can trump its being better for a person himself that he continue to exist
rather than to be annihilated. In fact, they deny this. That the worth of persons
be preserved puts a constraint on what the nature of retribution can be and how
much of it can be exacted, at any given time and over time. (The “at any given
time” constraint is perhaps relevant to the “no everlasting punishment” objection.)
Points A and B are easily overlooked in discussions of hell, but seem to me to be
of central importance, for both theological and philosophical reasons.
Presumably “realization” comes in degrees over a large range that is difficult
to measure. The degree of “realization” a person has with regard to one area or
use of capacity can differ greatly from what he has with regard to another. Much
“realization” seems to be available, given common grace. What degree and sort
might be consistent with the loss, and\or the retribution, involved in hell may
well be beyond our ability to say. The point is that, whatever the degree of it
that someone possesses combines with his metaphysical worth in terms of how
good his life can be. But that an omnipotent, just, and loving God render even life
in hell something that is good enough for annihilation not to be better does not
seem to me implausible. This is compatible with the loss being enormous and the
retribution being strong.
Things go at least one cut deeper. There would be no value in being a “realized”
person were there no value in being a person. The perfect realization of a worthless
thing is worthless. The point is not simply that if a person did not exist, she could
not be a “realized” person—there would be no “she.” The point is also that were
there not worth in being a person, there would be none in being a “realized”
person.23
The suggested constraints on what the nature of hell can be that have been
considered here leave room for what we might call a “very substantive view” of
hell. There are limits to what even a holy and just love, or a holy and loving justice,
23
The concern here is only with intrinsic worth, since the worth of persons is not
merely extrinsic.
or a just and loving holiness, can do. But that seems to me exactly what one should
expect. It does follow from our considerations that if one could prove that there
is no hell, and that there are unrepentant sinners24 who deserve retribution and do
not get it in this lifetime, one would have an argument of some force against the
existence of God. This should be a sobering consideration for annihilationists.
We have argued that far from there being a hell, understood along the lines that
are outlined here—a hell that is worthy of the name—being incompatible with the
existence of a good, just, and loving God, it (given the presence of unrepentant
sinners) is required by there being just such a God. Divine love will not be soft and
flabby, but serious and demanding. The dignity of persons, and the nature of a holy
God, requires nothing less.
24
More carefully, unrepentant wicked people—there are sinners only if there is a God
to sin against.
Chapter 11
Infernal Voluntarism and
“The Courtesy of Deep Heaven”
Bradley L. Sickler
The halcyon days of hell are over. Once enormously popular, the idea of an
eternal burning inferno for the damned is no longer vogue; it has been passed
by in favor of a kinder, gentler way of thinking about divine justice. Given the
present paucity of fire and brimstone sermons, and given the increasing religious
diversity in our culture, it is considered impolite and boorish to raise the question,
“Who—if anyone—is going to hell?” One of the biggest challenges raised by
religious diversity pertains to heaven, hell, and the possibility of salvation outside
Christianity.
Infernal Voluntarism and Universalism
The traditional Christian view,1 as we may call it, is that hell will be occupied by all
and only those who do not consciously repent of their sins and confess Jesus Christ
as Lord (ignoring for now the troublesome cases of infants, mental incompetents,
and the like). It is relatively easy, both socially and psychologically, to discuss hell
for non-Christians when those non-Christians are thousands of miles removed.
It is much more difficult when those non-Christians are one’s neighbors, friends,
and students. That is the challenge raised by religious diversity for the traditional
doctrine of hell: should we endorse a blanket condemnation of all those who
do not claim to follow Jesus Christ, even though they are people of great piety,
dedication, and sincerity within their own religions? The traditional view seems to
offer an unambiguous “yes” to this question. But, the objection goes, through no
fault of their own, many pious people cannot fulfill the necessary conditions for
entering into heaven because they have never heard or truly understood the Gospel.
But would God not be very unfair to condemn people for failing to perform an
impossible task? Many who have never heard or understood the Gospel are, by
all accounts, as pious and righteous as Cornelius. Why should they be condemned
because of an historical peculiarity that isolated them from effectively receiving
the Good News?
1
I write as a Christian and deal with the topic of hell from a Christian perspective.
Currently, a popular rival to the traditional view of hell is universalism, which

seems to offer a preferable position regarding hell and the fate of non-Christians.
According to universalism, the fate of every human soul is to be with God in
Paradise forever. There will be no lake of fire—at least, not permanently. It is fitting
that God be ascribed as much love and grace as possible, and universalism trades
on that by saying God will generously withhold his hand of judgment in order to
enjoy fellowship with all of his children eternally. The appeal universalism holds
is understandable. It is frequently claimed that a good and loving God would not
throw anyone into hell, especially those who never had a chance to respond to the
Gospel in the first place.
But notice that the traditional view and the universalist view share a common
but, I argue, false assumption. Each of them presumes that no one would choose
to go to hell and everyone would choose to go to heaven. The assumption is this:
people in hell are there against their will.
But what if the occupants of hell are not thrown there by God, but, as it were,
they freely jump? That view, which I will elaborate on in this chapter, can be
called “infernal voluntarism” (IV). Instead of assuming that everyone will want
to be in heaven, IV takes the position that there are very many people who would
sacrifice the joys of a life lived in submission to God in exchange for something
else. Odd as it may sound, IV acknowledges that there is something that people
may prefer over true flourishing and happiness—namely, rebellion. If the infernal
voluntarist’s position is right, then hell is chosen by its occupants, not foisted on
them against their will. Conversely, heaven would be entered by choice too: no
one who seriously desires heaven will be denied it. As we shall see, a corollary
to at least one famous infernal voluntarist’s position is that Christ has made the
way to heaven, and it is only through him that anyone enters—but that does not
entail that conscious belief in the person and work of Jesus Christ is a necessary
condition for being saved by and through him.
The challenge raised by religious diversity is multifaceted. The question
we asked earlier was, “Who is going to hell?” It is deeply troubling to many
contemporary Christians to answer, “Everyone who is not a Christian.” After all,
St. Paul seemed to find some religious common ground with the worshippers in
Athens, despite their complete lack of knowledge regarding Jesus Christ. God, it
seems, had given the Greeks enough revelation of himself to pave the way for the
Gospel; other biblical passages may indicate that God has done so with everyone.
The relevant problem raised by pluralism here is this: anyone who believes heaven
is only accessible to those who have heard and responded to the Gospel of Jesus
Christ must also believe that untold millions—even billions—of people will die
without even a chance at salvation. But notice again what gives the objection force
to begin with: the underlying assumption that people in hell would want heaven if
only God would let them have it.
Yet we must pay attention to eschatology a little more closely. According to
Christianity, heaven is not just a place of happiness, but happiness centered on
submission to God in Jesus Christ. If we abandon the assumption that everyone
Infernal Voluntarism and “The Courtesy of Deep Heaven” 165
wants that sort of happiness, universalism becomes not only unnecessary but
downright unappealing. Why should we want everyone to be in heaven if some
people want to live apart from God? Is it true that everyone desires what is
objectively best for them over everything else? Suppose that people are capable
of desiring self-governance above true flourishing under God. If they are, we
should be cautious about assuming that everyone would choose heaven over
hell if they had the chance. The infernal voluntarist argues that, contrary to first
impressions, universalism is not a generous orthodoxy but a frightening usurpation
of the freedom and dignity God grants even to his enemies. To force someone
to live in heaven—or even to force them to want to live in heaven—may be as
grievous as denying heaven to someone who longed for it. The apparent appeal of
universalism is the promise that everyone will get what they want in the end. That
is a premise that infernal voluntarists can agree with. The disagreement comes
over the proposition that everyone really wants heaven.
So, universalism may offer a deceptive hope. It may be that we should not want
everyone to be in heaven, because some people do not want that for themselves and
would resist it even though their rebellion costs them their ultimate happiness.2 The
driving principle of IV is that both heaven and hell are selected by their occupants.
What, then, is left of the challenge from religious diversity? That challenge was
that God unfairly locks righteous people out of heaven who would otherwise
choose to be there, throwing them instead into hell against their wishes. But God,
as we touched on briefly, seems to have been at work among all the peoples of
the world since the beginning. He seems to have given everyone, even the most
remote tribes, enough light to see that he exists, and to see (however dimly) the
sort of life he calls them to. What they have is short of the full revelation of God
in Jesus Christ, but it is enough to make an informed choice between living with
God or apart from him. Understanding heaven and hell in this way—as places
we ultimately choose for ourselves—could open the doors of Paradise to non-
Christians, as long as everlasting life in submission to God in Jesus Christ is what
they really want. It vitiates the challenge from religious pluralism by showing that
no one gets unfairly thrown into hell when heaven is their true desire. In so doing,
it also diminishes the appeal of universalism, which rests on the questionable
assumption that what we will all really want is to live with God forever.
The most influential proponent of infernal voluntarism is almost certainly
C.S. Lewis. His views can be summed up neatly in his claim that, “All find what
they truly seek.”3 In that vein, we will turn to him for an elaboration on the views
discussed above. Lewis believed that admission to heaven depended almost totally
on the volition of the one admitted; his view of hell was that people who occupied
it did so because they chose to. The objective here is not to defend Lewis, but to
2
Should we at least want everyone to want salvation? Perhaps. But such an end could
only come about through divine coercion of a variety so strong that no incompatibilist about
freedom could accept it.
3
The full quotation, from The Last Battle (New York, 2000), follows below.
stroll through his writings to piece together an argument for IV that he, as IV’s most
prominent advocate, would accept. Thus, this part of our mission is primarily one
of exploration and reconnaissance, though we will apply his writings to the broader
discussion of IV, pluralism, and universalism in the final section of this chapter.
The philosophically interesting material in defense of IV has to be cobbled
together from disparate parts of Lewis’s corpus, making this equally a mission
of reconstruction. Lewis quite intentionally couched much of his theology and
philosophy in the language of fiction and allegory, thinking those modes of
presentation would impact both the intellectual and affective aspects of his readers.
While those genres serve his purpose, and imitate the pattern he saw in the Bible
of presenting truth through story, they also serve to obscure some of the reasons
he had for the views so presented. Thankfully there are, in addition to the fictional
elements, claims he makes in his non-fiction work that can be illuminating here
too. I will draw on Lewis’s writings of both styles as the investigation proceeds.
Religion, Myth, and Reality
It has often been the case that Christians want to draw a sharp and absolute
distinction between Christianity and all other religions. Clearly Christianity has at
least one individuating characteristic; all religions necessarily do. But the amount of
overlap that paganism, Hinduism, Native American religions, Islam, Christianity,
and many other religions have is what is in question. Some people have insisted
that Christianity is thoroughly unlike every other religion. Lewis demurs.
Before proceeding, we should pause to think a bit about what is meant by
“religion.” Unless we answer this question first, we cannot understand the way
in which Lewis thinks that God has been speaking his truths to humans across
all cultures and times. And unless we understand that universal revelation of God
through the various stories and traditions of all religions, we cannot understand
how Lewis can get IV off the ground. It is only in the context of understanding
what lies at the heart of all religions that we can understand how everyone—
Christian and pagan alike—can be in a position to make an informed decision
about, to put it crudely, which side they are on.
Lewis does not take the popular line that religion is whatever deals with God
and immortality. Instead, he says, “The essence of religion … is the thirst for an
end higher than natural ends; the finite self’s desire for, and acquiescence in, and
self-rejection in favour of, an object wholly good and wholly good for it.”4 This
essence is expressed primarily in terms of the “numinous” and the “ethical.” Lewis
claims that every religion has numinous and ethical traits; that is what ties them
together, and those elements make them congruent with what Lewis saw as the
full revelation of God in the person of Jesus Christ. To be more specific, religions
4
C.S. Lewis, “Religion Without Dogma?” in Walter Hooper (ed.), God in the Dock
(Grand Rapids, 2001), p. 131.
display a certain pattern of what Lewis happily describes as divine revelation. In

the early stages, there is a feeling of awe inspired by something recognized as
being beyond nature and oneself. This is the first or numinous stage. Christians
understand the numinous as the triune God of the Bible; but the numinous does
not need to be understood by religious devotees as a single being, or indeed, even
as something personal. Following the numinous stage comes the second stage, in
which awareness arises that one has transgressed a moral law. An encounter with
the numinous breeds the feeling that one is simply not morally up to snuff. This
feeling comes from the perceived holiness of the numinous, and the recognition
that humans are decidedly unholy in virtue of having broken the moral law. That
leads to the third stage: seeing that the numinous itself is the wellspring of the
moral law. Thus far, Lewis thinks, nearly all religions basically agree, and all those
religions are basically right: the numinous exists, humanity is guilty of violating
the moral code, and the moral code humanity has universally violated finds its
source in the numinous itself.
According to Lewis, there is a fourth stage that only one religion has entered,
however: in the last stage of religious development, a man is born and “claims to
be the Numinous.”5 This, of course, is the stage uniquely entered by Christianity.
But it is important to see that there is a very substantial core of agreement between
Christianity and other religions in the marketplace. Lewis would fervently reject any
religious taxonomy that classified two immiscible kinds of religion—Christianity
and everything else. Christianity is not the antithesis of all other religions: it is
their fulfillment.
In Lewis’s science fiction book Perelandra, the character Dr. Ransom visits the
sinless planet Venus shortly after the advent of its human-like inhabitants. While
there, he observes things that are uncannily like the things told of in our myths and
legends. Ransom wakes at one point from a deep sleep and describes a scene that
made him think he was still dreaming.
He opened his eyes and saw a strange heraldically coloured tree loaded with
yellow fruits and silver leaves. Round the base of the indigo stem was coiled a
small dragon covered with scales of red gold. He recognized the garden of the
Hesperides at once … He remembered how in the very different world called
Malacandra—that cold, archaic world, as it now seemed to him—he had met the
original of the Cyclops, a giant in a cave and a shepherd. Were all things which
appear as mythology on earth scattered through other worlds as realities?6
Later, after seeing what appear to be mermaids and mermen, we are told “he
remembered his old suspicion that what was myth in one world might always be
fact in some other.”7
5
C.S. Lewis, The Problem of Pain (New York, 1986), p. 23.
6
C.S. Lewis, Perelandra (New York, 2003), p. 45.
7
Ibid., p. 88.
It was Lewis’s view that these ancient fables were not mere chimeras. They
spoke of something lurking beneath the common consciousness of humanity;
something that was a whisper of what was to come; something that God planted
deep in the hopes, dreams, and fears of all people everywhere. So the mythical
elements in other religions should not, as has too often been the case, be dismissed
by Christians as either demonically deranged fancy or the products of human
imagination only. They should be seen as vehicles of revelation, tenuous but
tendentious. Lewis says plainly,
I believe that in the huge mass of mythology which has come down to us a
good many different sources are mixed—true history, allegory, ritual, the human
delight in story—telling, etc. But among these sources I include the supernatural
… If my religion is erroneous then occurrences of similar motifs in pagan stories
are, of course, instances of the same, or a similar error. But if my religion is
true, then these stories may well be a preparatio evangelica, a divine hinting
in poetic and ritual form at the central truth which was later focused and (so to
speak) historicized in the Incarnation … My conversion, very largely, depended
on recognizing Christianity as the completion, the actualization, the entelechy,
of something that had never been wholly absent from the mind of man8
And later,
[Religious] traditions conflict, yet the longer and more sympathetically we study
them the more we become aware of a common element in many of them: the
theme of sacrifice, of mystical communion through shed blood, of death and
rebirth, of redemption, is too clear to escape notice … Rather in that tradition
which is at once more completely ethical and most transcends mere ethics—in
which the old themes of sacrifice and rebirth recur in a form which transcends,
though it no longer revolts, our conscience and our reason—we may still most
reasonably believe that we have the consummation of all religion, the fullest
message of the wholly other, the living creator, who, if He is at all, must be
the God not only of the philosophers, but of mystics and savages, not only of
the head and heart, but also of the primitive emotions and the spiritual heights
beyond all emotion. We may still reasonably attach ourselves to the Church, to
the only concrete organization which has preserved down to this present time
the core of all the messages, pagan and pre-pagan, that have ever come from
beyond the world9
Lewis clearly believed that there was much of value in religious traditions outside
Christianity, and that God had very assuredly been revealing divine truth through
them.
8
Lewis, “Religion Without Dogma?” p. 132.
9
Ibid., p. 144.
The “Frighteningly Unfair” Hypothesis
So let us return to the theme of infernal voluntarism, armed with Lewis’s premise
that God has been active among all people at all times, leading them through all
the various religious traditions to see important truths fully and finally revealed
to the Church. We just saw that, for Lewis, the work of Jesus Christ focuses and
fulfills something that has “never been wholly absent from the mind of man.” But
it does more than that too. It digs the channel into which the meritorious aspects
of all religious tributaries will be redirected; where all true acts of worship offered
to the numinous, even though he be obscured and misunderstood, are credited as
proper acts of fealty to the One True God.
Suppose that infernal voluntarism is true and that hell is only occupied by
people who choose to be there. This obviously implies that no one in hell would
choose to be in heaven if they were allowed. From this it follows that if some pagan
worshiper is so constituted that she would choose to live under the benevolent rule
of Jesus Christ if given the chance, then she is not going to choose hell. But if
hell is only occupied by those who choose it, and she would not choose to live
apart from God, then she simply cannot go to hell if IV is right. Keeping in mind
that hell and heaven are the only two eschatological options, then if she would
not choose hell she must join the happy throngs of like-minded worshipers who
choose heaven. So if everyone who dwells in hell chooses to be there and is there
as a result of their will to rebel, then everyone—unevangelized pagan or not—will
dwell in heaven if they so choose. This is the importance of Lewis’s position. He
recognizes that the doctrine of infernal voluntarism leads to the conclusion that all
people can go to heaven if they want to, irrespective of their ignorance of Jesus
Christ on earth. Because of this, he is strongly inclined to believe that God has
been revealing himself universally so that people can have the foundation they
need to make that fateful choice.
In The Last Battle, the final installment of Lewis’s Narnia series, we read of the
soldier Emeth. Emeth was a Calormene warrior who had worshiped the deplorable
god Tash all of his life, but had done so with honorable love and righteous
devotion. Tash is waiting on the other side of a magical portal into which all the
Calormenes are being compelled by the victorious Narnians, and Tash devours the
Calormene soldiers as they pass through. Tash was a disgusting demon, but Emeth
had rendered praiseworthy acts of service unto him in response to his inward,
divinely inspired convictions. Emeth is surprised, however, to pass through the
door and find not Tash, but Aslan—who, surely, needs no introduction—waiting
on the other side. This surprised both him and the Narnians in Aslan’s land, but
Emeth explains it as follows. Upon seeing Aslan,
The Glorious One bent down his golden head and touched my forehead with his
tongue and said, Son, thou art welcome. But I said, Alas, Lord, I am no son of
Thine but the servant Tash. He answered, Child, all the service thou has done
to Tash, I account as service done to me. Then by reason of my great desire for
wisdom and understanding, I overcame my fear and questioned the Glorious

One and said, Lord, is it then true, as the Ape said, that thou and Tash are one?
The Lion growled so that the earth shook (but his wrath was not against me)
and said, It is false. Not because he and I are one, but because we are opposites,
I take to me the services which thou hast done to him, for I and he are of such
different kinds that no service which is vile can be done to me, and none which
is not vile can be done to him. Therefore, if any man swear by Tash and keep
his oath for the oath’s sake, it is by me that he has truly sworn, though he know
it not, and it is I who reward him. And if any man do a cruelty in my name,
then though he says the name of Aslan, it is Tash whom he serves … But I said
also (for the truth constrained me), Yet I have been seeking Tash all my days.
Beloved, said the Glorious One, unless thy desire had been for me thou wouldst
not have sought so long and so truly. For all find what they truly seek.10
All good acts and pure motivations are, on Lewis’s view, properly offered as
worship to the One True God. As St. John says, “Everyone who loves is born
of God.” Despite the fact that his Calormene countrymen worshiped Tash in
despicable ways, Emeth knew that such worship was unseemly for the numinous.
The point must be clear here: it is most emphatically not that there is no ultimate
difference between religions, as John Hick and other pluralists would say; it is not
that Tash was just another name for the One True God; all religions are not equally
veracious. But it is possible on Lewis’s telling to worship God without knowing
it—even in such a state that one would positively deny it if asked. In an interview,
Lewis was asked, “Supposing a factory worker asked you: ‘How can I find God?’
How would you reply? … [Answer:] People will find God if they consciously seek
from Him the right attitude towards all unpleasant things.”11 Lewis did not believe
that explicit knowledge of and belief in Jesus Christ is a necessary condition for
finding God. Just as Emeth found Aslan even though he thought he was following
Tash—and, indeed, would have denied in his ignorance that he was really searching
for Aslan—Lewis believes Christ will be found by many who do not know it is
him they are after. “All find what they truly seek.”
Lewis would say that the general revelation of God in the myths and precepts
of other religions is one of the ways that Christ has, as it were, been preached to all
nations already. He says that any other view amounts to the claim that some have
been abandoned by God, wandering around in the dark, waiting and yearning for
some shred of knowledge of him and his ways but not receiving it. But God has
never abandoned anyone. In one of his letters to Malcolm on prayer, Lewis says
that God has been governing by providence the path of each and every human
soul.
10
Lewis, The Last Battle, pp. 161–5.
11
C.S. Lewis, “Answers to Questions on Christianity,” in Walter Hooper (ed.), God in
the Dock (Grand Rapids, 2001), p. 50.
It is an old and pious saying that Christ died not only for Man but for each man,
just as much as if each had been the only man there was. Can I not believe the
same of [Providence]—which, as spread out in time, we call destiny or history?
It is for the sake of each human soul. Each is an end.12
In The Great Divorce, the narrator’s guide in the in-between land says, “All
moments that have been or shall be were, or are, present in the moment of His
descending. There is no spirit in prison to Whom He did not preach.”13 There are,
in a very robust sense, no “unevangelized” people—God has already and always
ministered to each of them; he has preached to all the “spirits in prison.”
“Christianity,” Lewis says, “is primarily the fulfillment of the Jewish religion,
but also the fulfillment of what was vaguely hinted in all the religions at their best.
What was vaguely seen in them all comes into focus in Christianity—just as God
Himself comes into focus by becoming a Man.”14 It is because God has already
been showing himself through all the religions—not just Christianity—that he can
apportion to himself the fitting acts of service and piety that are offered therein,
for they are in response to that true revelation of the One True God that God has
presented and preserved within those religions. Lewis says, “There are people in
other religions who are being led by God’s secret influence to concentrate on those
parts of their religion which are in agreement with Christianity, and who thus
belong to Christ without knowing it.”15
For Lewis, it would be an injustice for God to operate any other way. God has
made himself manifest in all religions (though obviously to varying degrees and
only completely in Christianity) because it would be a moral failing not to. Lewis
says,
Is it not frighteningly unfair that this new life should be confined to people who
have heard of Christ and been able to believe in Him? … We do know that no
man can be saved except through Christ; we do not know that only those who
know Him can be saved through Him.16
Here, then, is Lewis’s “frighteningly unfair” hypothesis: God will only save those
who have explicit knowledge of Jesus Christ. Lewis says that the fair hypothesis,
on the other hand, is that salvation is possible—always through Christ, it must be
emphatically noted—even for those who are adherents of other religions besides
Christianity. So God will take as unto himself the worship offered to others, as
long as that worship is in response to the true revelation he has given of himself.
Christ has, in a meaningful sense, been preached to all men already, though not
12
C.S. Lewis, Letters to Malcolm (New York, 1992), p. 55.
13
14
Lewis, “Answers to Questions on Christianity,” p. 54.
15
C.S. Lewis, Mere Christianity (New York, 1984), p. 64.
16
Ibid., p. 217.
always under his true name. Thus, for Lewis, whoever is saved is saved through
Christ, but the name of Jesus may not be known to the recipient of salvation. For
God to do anything else would be unfair and out of keeping with his love and
providence, which he extends not just to mankind, but to each and every man.
The Disjunction
The purpose of all these ruminations about the degree of divine revelation outside
Christianity is to guide our thoughts about the final state of our souls. In his
prolonged struggle against the tempter in Perelandra, to his utter horror Ransom
sees the devil smile. The sight was so miserable that he nearly fainted from despair.
He says, “And though there seemed to be, and indeed were, a thousand roads by
which a man could walk through the world, there was not a single one which did
not lead sooner or later either to the Beatific or the Miserific Vision.”17 Here, then,
is the exhaustive disjunction: the final destination of every soul is either heaven
or hell. That much, at least, is thoroughly unremarkable to Christians and many
other religious adherents. It is a tenet of orthodox Christian theology that heaven
and hell are real, that they are the only two ultimate options, and that both will be
populated.
Thus far, as I said, there is nothing remarkable here. But in keeping with
infernal voluntarism it would be more accurate, per Lewis’s treatment, to speak not
of people being thrown into hell, but of them jumping into hell. Jonathan Edwards
(no infernal voluntarist, certainly), in his famous Sinners in the Hands of an Angry
God, says that those who will inhabit hell
are liable to fall of themselves, without being thrown down by the hand of
another; as he that stands or walks on slippery ground needs nothing but his own
weight to cast him down. [The] reason why they are not fallen already, and do
not fall now, is only that God’s appointed time is not come. For it is said, that
when that due time, or appointed time comes, their foot shall slide. Then they
shall be left to fall, as they are inclined by their own weight. God will not hold
them up in these slippery places any longer, but will let them go18
It is critical to Lewis’s soteriology that hell is a place people choose to go. In

The Great Divorce, George MacDonald, the author’s guide to heaven, tells him,
“There are only two kinds of people in the end: those who say to God, ‘Thy will
be done’, and those to whom God says, in the end, ‘Thy will be done’ … Without
that self-choice there could be no Hell.”19 Those who go to hell get exactly what
17
Lewis, Perelandra, p. 111.
18
Jonathan Edwards, “Sinners in the Hands of an Angry God,” in John E. Smith (ed.),
A Jonathan Edwards Reader (New Haven, 1995), p. 89.
19
Lewis, The Great Divorce, p. 72.
they want. They are not coerced or forced, but act by their own wills. No one
is outwardly compelled to hell: they take for themselves the judge’s judgment
willingly. I have sometimes responded to the unruly behavior of my children by
sending them to their rooms, which always leads to weeping and the gnashing of
teeth. When given the opportunity to either repent or continue their incarceration,
though, my children will often choose to obstinately resist penitence. They will
say they want to come out, and I believe them. But they would prefer to remain
where they do not want to be if it means they can persist in their rebellion. So it is
with hell according to infernal voluntarism. In The Problem of Pain, Lewis says,
“I willingly believe that the damned are, in one sense, successful rebels to the end;
that the doors of hell are locked on the inside.”20
But how could anyone choose such a thing? How could anyone want to be
separated from a divine being considered by all major religious accounts to be the
ultimate in Love and Beauty and Goodness? To Lewis, as to any Christian, this
choice must have seemed nearly incomprehensible. After all, the greatest hope of
a Christian is to one day be united with the numinous, to “see him face to face.”21
But it is not so for everyone, Lewis claims.
Again in The Great Divorce, the narrator asks MacDonald to explain the
puzzling fact that people would exchange happiness for misery.
What do they choose, these souls who go back [to hell]? … And how can they
choose it? “Milton was right,” said my Teacher. “The choice of every lost soul
can be expressed in the words ‘Better to reign in Hell than serve in Heaven’.
There is always something they prefer to joy—that is, to reality.22
In a letter to his brother in 1940, Lewis wrote, “I begin to suspect that the world is
divided not only into the happy and the unhappy, but into those who like happiness
and those who, odd as it seems, really don’t.”23
Those who go to hell go because they choose it. Likewise, those who go to
heaven get their choice too. While standing near the omnibus that is shuttling
people back and forth between heaven and hell, the narrator of The Great Divorce
asks what happens to the “poor Ghosts who never get into the omnibus at all?”
The answer is, “Everyone who wishes it does. Never fear … No soul that seriously
and constantly desires joy will ever miss it. Those who seek find. To those who
knock it is opened.”24 Here we have again the refrain that, to repeat what Emeth
said above, “All find what they truly seek.” Anyone who wants heaven shall have
it. Anyone who wants hell shall have it. But to all, God grants what they seek to
find.
20
Lewis, The Problem of Pain, Ch. 13.
21
I Corinthians 13:12.
22
Lewis, The Great Divorce, p. 69.
23
Walter Hooper, C.S. Lewis: A Companion and Guide (New York, 1996), p. 281.
24
Lewis, The Great Divorce, pp. 72–3.
So while he never makes all of his reasons for holding infernal voluntarism
explicit, Lewis offers us some premises that might be used to reconstruct an
argument for IV that he would approve of. At the heart of Lewis’s argument is the
conviction that God is fair. Because of his fairness, God would not treat people
differently simply because of historical and geographical accidents outside their
control. Presumably this notion of fairness is a special application of the principle
that God loves humanity, and not just humanity in the abstract. God loves every
individual human. Divine love is not dependent on our merit, but flows from God
equably and universally. It is not that we first loved him, as St. John points out,
but that he first loved us—all of us, regardless of our status as sinners (a status that
we all share).
Lewis says that God has universally revealed himself through the myths told by
people of nearly every tradition. This is a result of his fairness to all people. Lewis
sees in the stories of humanity several common themes that are anticipations of
the Christian story, which completes those shadowy and incomplete tales. Fairness
explains this because God was not willing to leave anyone without a significant
understanding of the truth about spiritual matters, because he treats all people
equally. Lewis would likely see the commonality of the first three steps in religious
progress mentioned above as further evidence for his doctrine of God’s fairness.
This universal self-revelation also allows God to deal with us based on our
response to him without showing favoritism. No one will be thrown into hell
involuntarily, nor will anyone enter heaven involuntarily—both destinations are
freely chosen by their inhabitants and no one will be denied what they ultimately
desire. God will honor our decision either way. Lewis says, “Any man may choose
eternal death. Those who choose it will have it.”25 Everyone—both Christian and
pagan—has enough of God’s revelation to make an informed decision about their
final destiny, and God will honor their choice. Such are the contours of Lewis’s
argument for infernal voluntarism.
Objections to Infernal Voluntarism (IV)
As attractive as infernal voluntarism may be to some, and despite the persuasive

power of Lewis’s defense, it is not without challenges. In the final section of this
chapter we will consider three of the more powerful objections to IV.
The first objection to IV is that it fails to capture the picture of hell offered
in the Bible by denying that God actively judges human sin and punishes it
accordingly. The voluntarism element would seem to deny any role for God as
judge. What about the passages where Jesus says that the unsaved “will be thrown
into outer darkness”?26 Or when he will say “depart from me—I never knew you”
25
Ibid., p. 124.
26
Matthew 8:11.
to people pleading for mercy?27 If God is storing up his wrath for the final Day of
Judgment, when the sheep are separated from the goats and the wretched are cast
out of God’s presence forever, then IV misrepresents these events by saying that
God does not judge us, but rather simply stands aside and lets us judge ourselves.
Naturally anyone interested in an orthodox Christian understanding of hell cannot
neglect the important role given to God as judge. It is essential to Christianity to
acknowledge that God has been offended by our sin, and that we have stirred up
his wrath through our rebellion. Lewis agreed with all of this too—it was common
to all religions to say this much in the first three stages outlined above.
But this objection seems to presume that having us choose our destination is
incompatible with God exercising judgment over his creation. That is, the objector
assumes that S is separated from God in hell as a result of S’s choice and God
separates the righteous from the wicked in the Final Judgment cannot both be true.
But why not?
Consider again the parallel case of a rebellious child whose behavior calls for
discipline from a loving parent. Suppose, for instance, that a little girl has taken
her younger brother’s toy after pushing him to the ground (for some of us this
merely calls for recollection, not imagination). The girl is instructed by her father
to return the toy and apologize to her brother. She refuses. The father warns her
that failure to comply with his commands will result in something unpleasant for
her: she will be confined to her room until she stops resisting and apologizes for
her wrongs. She makes it clear that she does not want to be sent to her room, but
she makes it equally clear that she is unwilling to repent. The father puts her in her
room and closes the door. There is much wailing as the girl “receives in herself the
due penalty for her sins,” but she persists in her refusal to avail herself of the one
way out—repentance and compliance with the will of her father.28 She chooses
rebellion and misery over submission and peace. Is it not true in this scenario
that (i) the father has judged and disciplined the girl, and (ii) she has chosen to be
confined to her room rather than to enjoy the blessings of liberty? And so it seems
that something similar might be said about those who have chosen to live in hell.
It is true that they have been judged by God, and equally true that they are only
in hell because they choose to be. We often have conflicting desires, but the will
can only choose one thing. In the case of the rebel who chooses hell, her desire
for happiness conflicts with her desire for self-rule. Her will has chosen the latter
over the former.
The second objection to IV grows out of this response to the first, claiming
that we have described something that could not really take place. According to
this objection, no one could make an informed decision to live apart from God.
The case of the young girl is not a good example, the objection might go, because
the girl will eventually choose to comply. Surely she would not choose to stay in
her room forever, especially if the father continued to woo her unto repentance as
27
Ibid., 7:23.
28
Romans 1:27.
any loving parent would. In the same way, the objector claims that no soul could
choose hell for very long, seeing what misery it entails and the bliss available in
heaven. When stripped of distraction and encountered face-to-face with God, the
argument goes, God’s goodness will overcome the rebel and they will want to
embrace their true Father.
Along these lines, Thomas Talbott has argued that no one could make a fully
informed, free choice to live a life of continued rebellion and rejection of God’s
salvation. Talbott suggests that all cases of rejecting God, and therefore the
summum bonum or Highest Good, are cases of choosing in ignorance.29 Nobody
who is properly informed could freely choose to reject God. But since we will
all ultimately be properly informed through a combination of beholding God and
being relentlessly pursued by him in love, it is inconceivable that anyone would
stay in hell.
But however appealing this objection might be, Christianity in particular
provides us with clear counterexamples. We are given at least two examples of
agents who lived in sinless perfection in the very presence of God, enjoying his
company and favor without impairment. They were fully informed, having beheld
God directly, and they were fully aware of God’s loving character. In the Garden
of Eden, Adam and Eve walked with God in perfect peace. Nevertheless, despite
lacking sufficient rational reason for doing so, both of them chose rebellion over
conformity to God’s will, thereby choosing to be banished from God’s side. And
an even clearer violation of Talbott’s position can be seen in the case of the chief
of all rebels: Satan himself. Before anyone had sinned, and absent any external
source of temptation (as Adam and Eve had through the serpent), Lucifer—a being
endowed with even greater power, knowledge, and wisdom than humans—quite
freely chose to reject God in favor of self-governance. It is hard to imagine a
clearer refutation of the argument that a rational being could not make a free
choice to permanently turn from God and to persist in that rebellion. And even
if one supposes that the stories of Satan and Eden are mythical (and I do not), it
seems abundantly clear that they are meant to underscore the propensity we have
to go against our best interest, choosing sin and self over God.
In the Platonic dialogue Protagoras, Socrates argues that no one knowingly
chooses a lesser good over a greater good, or an evil over a good. Early in Church
history, however, Augustine goes to great lengths in the Confessions to show
that an incident in his life was a clear case of choosing an evil over a good. To
while away the hours, a young (pre-conversion) Augustine and his chums raided a
neighbor’s pear tree. Listen to him tell the story.
There was a pear tree close to our own vineyard, heavily laden with fruit, which
was not tempting either for its color or for its flavor. Late one night—having
29
In addition to his chapter in this volume, see his “The Doctrine of Everlasting
Punishment,” Faith and Philosophy, 7/1 (1990): 19–42; “Providence, Freedom, and Human
Destiny,” Religious Studies, 26/2 (1990): 227–45.
prolonged our games in the streets until then, as our bad habit was—a group
of young scoundrels, and I among them, went to shake and rob this tree. We
carried off a huge load of pears, not to eat ourselves, but to dump out to the hogs,
after barely tasting some of them ourselves. Doing this pleased us all the more
because it was forbidden. Such was my heart, O God, such was my heart—which
thou didst pity even in that bottomless pit. Behold, now let my heart confess to
thee what it was seeking there, when I was being gratuitously wanton, having
no inducement to evil but the evil itself. It was foul, and I loved it. I loved my
own undoing. I loved my error—not that for which I erred but the error itself. A
depraved soul, falling away from security in thee to destruction in itself, seeking
nothing from the shameful deed but shame itself.30
In this passage and in what follows, Augustine makes it clear that he is rejecting
the Socratic claim revived by Talbott: he chose evil freely, knowingly opting for
the lesser good. He was “being gratuitously wanton, having no inducement to
evil but the evil itself.” In light of these counterexamples, it is hard to see how
traditional Christianity can deny the possibility of free and informed rebellion
as argued for by infernal voluntarists. It is just simply not true that no one could
persist in freely and knowingly rejecting God.
But the infernal voluntarist is not out of the woods yet. The final objection
to IV we will consider is that God would be unjust in allowing a person to
permanently persist in a choice so obviously self-destructive. One might grant
that it is possible for a person to choose evil, given the discussion above, but that
a good and omnipotent God would not allow a person he loved to actually carry
out their chosen course of action. Returning to the analogy of the willful child,
no loving parent would let their child persist in a course of self-destruction even
if the child were free and fully informed. If God truly loves someone and wants
their good, how could he allow them to choose destruction? One proposal is that
God could, and should, override our free will when it comes to rejecting him. Free
will is a good thing, but it is not so good that God could never justifiably override
it when it is in our best interest. A parent might respect a child’s freedom, but not
to the point of letting him play in the knife drawer or gun rack. For his own good,
the child will be restrained and his choices will not be honored. Any loving parent
will do this with some regularity.
But it seems that this objection might oversimplify our situation. What would
it look like for God to “override our freedom” when it comes to choosing to
submit to him? This would be no mere tweaking of our preferences, like changing
a person’s favorite ice cream from strawberry to chocolate. No, the way we choose
to respond to God is at the very heart of our personalities, and to change it would
be to change something essential to our selfhood, not something merely accidental
or peripheral. That is, it seems unlikely that God could turn us from rebels to lovers
without fundamentally altering who we are and making us into new creations.
30
St. Augustine, Confessions, trans. Albert C. Outler (Philadelphia, 1955), p. 128.
In some sense this is Christian teaching for what happens at salvation, but non-
Calvinists insist that it is our submission to God that precedes and prompts God’s
initiation of the “new birth.” If there is any choice I must be allowed to make
freely, it is how I respond to God. God could not override our freedom in this
essential area, contradicting our will rather than responding to it, without in some
sense destroying us and replacing us with someone else. But then it is not I who
gets saved—I am done away with and some doppelgänger takes my place.
Conclusion
In a pluralistic age like ours, the question of hell has taken on a new face. Instead
of speculating about the fate of far-off pagans, today’s western Christians have to
deal with the presence of many competing religions in the marketplace of ideas.
As a response, some have opted for a universal view of salvation, according to
which Christianity has no special place and all people will ultimately be saved.
Others have maintained the traditional Christian teaching on hell and insisted that
explicitly acknowledging Jesus Christ as necessary to be saved, and everyone who
does not do so will be condemned—despite their longing for heaven.
Infernal voluntarism responds to these positions by affirming the reality of
hell, but rejecting the assumption that those who are consigned there will be going
against their will. By doing so, IV undercuts the appeal of universalism, which
seems to presume that everyone will want to live with God. It also diminishes
what may appear as the unnecessarily harsh exclusivity and vindictiveness of
Christianity by affirming that God has been working to reveal himself in all
religions, and that the choice of our eternal destiny ultimately rests with us. God
will honor our choice, whether it is to follow him or to persist in rejecting him. To
pave the road to heaven through the work of Jesus Christ, but to leave the choice
of whether to take that road or venture out on our own—this is what Lewis calls
“the courtesy of Deep Heaven.”
Chapter 12
Birth as a Grave Misfortune:
The Traditional Doctrine of Hell and
Christian Salvific Exclusivism
Kenneth Einar Himma
Although the traditional doctrine of hell is a subject of much philosophical debate,

the debates fail to capture the range of issues that arise for this perplexing and
disturbing fundament of mainstream Christianity. Most of the focus is on the so-
called “proportionality problem,” which is concerned with the issue of whether
and how finite beings can be deserving of eternal punishment. But the doctrine
of hell, especially in conjunction with Christian exclusivism, raises many other
issues. In this chapter, I would like to discuss what I take to be a novel issue
regarding the morality of having a child if there is a morally significant chance that
his or her ultimate fate will be eternal suffering in hell.
It is typically thought obvious by Christians and non-Christians alike that, as a
general matter, it is morally permissible for two married parents to have children.
Indeed, many persons go further than that, viewing the having of a child as both
morally good and as a moral right (in the sense that it is wrong for others, including
the state, to interfere with their doing so). Many Christians, however, take the
position that it is a moral duty to have children, basing this view on Genesis 1:28,
which states that “God said unto them, Be fruitful, and multiply, and replenish
the earth.” Although some denominations take this, falsely I think, to imply a ban
on contraception, many Christians believe it imputes a positive duty on married
couples to attempt to have some children during their marriage, if possible. On
this view, it is morally wrong for two married persons who can have children to
deliberately attempt to avoid doing so for the entire duration of their marriage.
In this chapter, I wish to argue from the standpoint of ordinary moral intuitions
that if Christian exclusivism and the traditional doctrine of hell are true, then
this view is mistaken. In particular, I argue that it is morally wrong, given these
traditional Christian doctrines, to bring a child in the world when the odds that he or
she will spend an eternal afterlife suffering the torments of hell are as significantly
high as they would be if these two doctrines are true. For example, only one-third
of the world’s 6 billion people claim to be Christians, consigning the other two-
thirds immediately to hell for eternity. Notwithstanding the command of Genesis
1:28, it is intuitively wrong—and gravely so—to bring a child into this world
facing such high odds of a terrible fate if these traditional doctrines are true.
In effect, the argument is against these traditional doctrines, which have seemed
to me difficult to reconcile with God’s moral perfection and omnibenevolence.1 It
seems clear to me that it is not, other things being equal, morally wrong to have
children and that a morally perfect God would want for us to have children (at
least up to a point). Although I think that exclusivism and that the idea that hell
involves an eternity of torment unmatched by anything we could experience on
earth should both be rejected, only one of these claims needs to be rejected to
block the implication that having children is morally wrong. Since the conclusion
is so counterintuitive, I think it important for the reader to keep in mind that it is
not a claim that I believe is true, and attempt only to show it is implied by certain
doctrines of Christianity in an attempt to show at least one of those doctrines
should be rejected.
Having Children as Moral Good
It is initially tempting to think that bringing new human life into existence is an
unqualified moral good no matter whose point of view one takes. First, it is good
from the standpoint of the parents when they want a child and have a new baby.
Although I am not a parent, I am an uncle and have had my life enriched by my
two nieces, Angela and Maria, in ways I could never have predicted. The love I
feel for them, and the joy that accompanies this is utterly unprecedented in my life.
While I remain deeply in love with my wife after 14 years of marriage, my love
for her remains self-regarding and hence conditional in a way that my love for my
nieces is not. Infidelity on my wife’s part would, I suspect, end our relationship
and at least constitute the beginning of the end of my love for her. But I have a hard
time imagining what my nieces could do that would end my love for them; I would
love and support them even if they did something bad enough to merit a prison
sentence. I have experienced many kinds of love, but the love that a parent feels
toward her child must, if my experience is any indication, be something especially
intense—and hence as great a moral good as any emotion could be.
Second, bringing a child into the world is considered good from the standpoint
of the child, who did not exist until the pregnancy as a conscious subject of a
life that has its shares of ups and downs but is on the whole a good one. The
goods that make life worth living—love, sex, art, philosophy, food, friendship, the
beauty of nature, comedy, music, dancing—arguably (for most of us in the affluent
world, anyway) outweigh the sufferings that occasionally afflict us all (sadness,
anger, sickness, being victimized by a moral wrong, fear of death). This is why
we commonly think of being alive as a wonderful gift and hence of new life as the
recipient of such a gift, and hence as a great moral good.
1
See Kenneth Einar Himma, “Finding a High Road: The Moral Case for Salvific
Pluralism,” International Journal for Philosophy of Religion, 52/1 (2002): 1–33.
Birth as a Grave Misfortune 181
These are not uncommon views. I think it is safe to hypothesize that most
people, Christians and non-Christians alike, regard news of a new child, other
things being equal, as being an unqualified moral good. There is nothing more
joyous than the occasion of a new human life in the world—or so goes the common
intuition. Having a child is an important moral good.
The view that having children is a great moral good also has Scriptural support.
Psalm 127:3, for example, states that “children are a gift of the Lord.” It is true,
of course, that gift-giving need not result in moral good, or even be considered
a moral good: if the content of the gift is immoral (e.g., a free subscription to
a pornographic magazine) or the motives improper (e.g., a desire to manipulate
another person by creating a feeling of indebtedness), then it is reasonable to think
that the gift neither results in nor constitutes a moral good. But a gift from the
Lord is surely a moral good, if anything is. Accordingly, Christians have another
reason, beyond widely shared intuitions or reactions, to think of new human life
as a great moral good.
Having Children as Moral Duty
Some Christians take a somewhat stronger position. The birth to married parents
of a new child is not only wonderful news; it is also the satisfaction of a moral
duty. There are a number of Scriptural verses that seem to suggest this. Genesis
1:28 states “And God blessed them, and God said unto them, Be fruitful, and
multiply, and replenish the earth.” Genesis 9:7 repeats this injunction to Noah after
the flood subsides: “And you, be ye fruitful, and multiply; bring forth abundantly
in the earth, and multiply therein.” According to 1 Timothy 5:14, “So I counsel
younger widows to marry, to have children, to manage their homes and to give the
enemy no opportunity for slander.”
Many, but not all, Christians interpret these verses as implying that we have
a moral duty to have children when we are married and able to do so. Some
Christians take this duty as being both absolute and a part of a larger argument
against contraception.2 Consider the following words from a Christian website:
The very first recorded words of the Creator to the man and woman he had made
in his image were, Be fruitful and increase in number; fill the earth and subdue it
(Gen. 1:28). This precept was repeated after the flood to Noah and his sons: As
for you, be fruitful and increase in number; multiply on the earth and increase
upon it (9:7). God’s plan is that married couples multiply descendants to fill the
whole earth … When and if the earth ever actually becomes full, we can trust
God to deal with the situation his own way. Our job is to obey his commands …
Having a number of children is the normal fruit of marriage, and it is God’s will
2
The other argument is that contraception interferes and changes the character of an
act that God intended to result in procreation.
for marriage … It seems to this writer that God’s Word is very clear. Christians
should not partake of the world’s birth-control mindset. They should embrace
God’s plan for marriage, including the procreative purpose of sex, and joyfully
accept as blessings all the children that God sends them. Further, they should
develop the long range vision that sees children as the means to advance the
kingdom of Christ and defeat his enemies. The more children he gives, the better.
They know that God is a loving Father who will provide for every child of his.3
The idea here is that it is God’s will that all married persons have children, regardless
of whether the earth seems overpopulated to us, and presumably regardless of the
health or quality of life into which the child is born.
Other Christians take the position that married persons have a duty to have
children, but this duty is not unlimited. Many Christians believe that we have a
duty to have at least one but no more than two children, believing that the world is
in danger of being overpopulated and hence of threatening many elements of God’s
creation by polluting the environment and threatening ecosystems in a variety of
ways. Some Christians share with a number of atheist philosophers the view that it is
wrong to bring a child into the world if one knows that child’s life will be short and
painful. Either way, for these Christians, married persons have a duty, under these
verses in Scripture, to have some children if they have reason to think these children
will be born healthy. Having children, on this view, satisfies this duty—as long as it
does not exceed any moral limit on the number of children one should have.
Not all Christians take this strong position. Some ordinary Christians interpret
Genesis 1:28 as asserting no more than just that children in a sacramental marriage
are a blessing, instead of the stronger claim that having children in a sacramental
marriage (even up to some limit) is a moral duty. According to Raymond C. Van
Leeuwen:
Many [Christians] … argue on the basis of the created order (sometimes called
natural law) and Scripture that God has actually commanded married people
to have children … God does not command humans to be fruitful. Rather, he
himself will bless his creatures and see to it that they are fruitful. He has provided
for this by making us male and female, by investing our humanness with sexual
desire and love, and by ordaining marriage as the place for, among other things,
joyful lovemaking. Marriage is also the God-given matrix from which family
naturally springs, the place where children may be born and reared with love
and wisdom, “in the fear of the Lord.” … If Genesis 1:28 were a “command”
that applied to every individual, then Paul would have been disobedient in his
apostolic singleness.4
3
“Being Fruitful: A Biblical View of Birth Control,” Life and Liberty Ministries, 8
January 2005; available at http://www.lifeandlibertyministries.com/archives/000172.php.
4
Raymond C. Van Leeuwen, “Be Fruitful and Multiply,” Christianity Today, 12 November
2001; available at http://www.christianitytoday.com/ct/2001/november12/4.58.html.
This presumably common view among many Christians expresses the view that,
while not a duty, having children is a blessing and hence a moral good from the
standpoint of Christian ethics.
A Blessing to Whom?
It appears, from the last section, that the view of having children that can most
accurately be attributed to the largest number of Christians is this: having children
is a blessing. And it is blessing to both parents and newborn children for all the
reasons described above: parents get the joy associated with child-rearing, while
newborn children get the blessing of a sentient life with access to the goods that
make life so worth living.
At least part of this view, on closer look, seems problematic. Certainly, the
ultimate motivation for most parents—and the one of which they are most aware—
for having children is not to benefit the children who will come into existence; it
is reasonable to hypothesize that the motivation for having children is generally
self-regarding. Parents want children for a variety of reasons—most of which have
to do with certain desires having to do with lifestyle.
And it is not implausible to think such self-regarding and God-regarding
motivations, such as the desire to do God’s will, are the only coherent motivations.
As David Benatar explains:
Children cannot be brought into existence for their own sakes. People have
children for other reasons, most of which serve their own interests. Parents
satisfy biological desires to procreate. They find fulfillment in nurturing and
raising children. Children are often an insurance policy for old age. Progeny
provide parents with some form of immortality, through the genetic material,
values, and ideas that parents pass on to their children and which survive in their
children and grandchildren after the parents themselves are dead. These are all
good reasons for people to want to have children.5
The first sentence states a problem without explaining it; the problem here, on
Benatar’s view, is that the absence of a benefit is not a moral evil, but the absence
of harm is always a moral good. The idea that one can benefit a child by bringing
it into existence presupposes that the absence of a benefit is a moral evil that is
eliminated by bringing into existence a child who will somehow be benefited.
Indeed, Benatar defends a stronger, somewhat counterintuitive, claim that it
is always better not to have been brought into existence than to be brought into
existence. On his view, there is a decisive asymmetry in comparing harms and
benefits when it comes to comparing them as they affect existent individuals and
5
David Benatar, “Why It Is Better Never to Come into Existence,” in David Benatar
(ed.), Life Death, and Meaning (Oxford, 2004), p. 164.
non-existent but possible individuals. As Benatar points out, everyone suffers in

life: for example, old age, disease, fear of death, dying, and loss of companionship
are events causing suffering that befall all of us; there is simply no escaping
suffering. For an existent individual x, pleasure is always a moral good, while
suffering is always a moral bad. If x does not exist, then the absence of suffering
x would have experienced is always a moral good, but the absence of pleasure x
would have experienced is not a moral bad.
He offers a number of intuitive reasons for this latter judgment. First, people
rarely have as a primary or even secondary motivation a desire to benefit the child
who has not yet been conceived. In contrast, it is not unusual for people who decide
not to have a child to do so on the basis of medical information that indicates a
high probability that they would have a child with a serious medical condition that
would cause the child great suffering.
Second, we might regret having brought a suffering child into the world. But
if we regret not having had a happy child, it will not be for the sake of the child;
rather, it will be for the sake of interests we have in raising a child that went
unsatisfied. As he puts the matter:
Bringing people into existence as well as failing to bring people into existence
can be regretted. However, only bringing people into existence can be regretted
for the sake of the person whose existence was contingent on our decision. One
might grieve about not having had children, but not because the children which
one could have had have been deprived of existence. Remorse about not having
children is remorse for ourselves, sorrow about having missed child-bearing and
child-rearing experiences. However, we do regret having brought into existence
a child with an unhappy life, and we regret it for the child’s sake, even if also
for our sakes. The reason we do not lament our failure to bring somebody into
existence is because absent pleasures are not bad.6
“Absent pleasures are not,” as he puts it, “bad,” while actual suffering is bad.
Third, I submit that it helps to explain why the problem of evil has the emotional
and intellectual force it has for people who take it seriously. Consider, perhaps,
the most famous passage from The Brothers Karamazov known for raising the
problem of evil in the most poignant terms ever expressed. Ivan, an irreligious
man, challenges his brother Alyosha, a monk, with a famous question:
“Tell me yourself, I challenge you—answer. Imagine that you are creating a

fabric of human destiny with the object of making men happy in the end, giving
them peace and rest at last, but that it was essential and inevitable to torture
to death only one tiny creature—that baby beating its breast with its fist, for
instance—and to found that edifice on its unavenged tears, would you consent to
be the architect on those conditions? Tell me, and tell the truth.”
6
Benatar, “Why it is Better Never to Come into Existence,” 157–8.
“No, I wouldn’t consent,” said Alyosha softly.7
The point here is that Alyosha’s answer conveys the moral force of the problem
of evil precisely because of the asymmetry between the absence of pleasure and
the absence of pain. Alyosha declines to bring a world into existence in which an
infant suffers such torments, despite the loss of all of the pleasures that would also
be lost, precisely because the absence of pain is regarded as a moral good even if
the absence of pain is explained by there being no sentient creatures capable of
experiencing pain, while the absence of pleasure is not regarded as a moral bad.
Of course, a contestable judgment is being made here about how to weigh the
moral goods and evils, and one might reasonably take issue with the idea that the
absence of the infant’s suffering has so much weight in the deliberation. But it
should be clear that the asymmetry Benatar describes is playing a role in Alyosha’s
thinking—and that the asymmetry is both plausible and properly considered in
such deliberations.
Benatar seems correct about this—even if he is incorrect in thinking that it
implies or even supports the admittedly counterintuitive claim that it is better
for everyone not to have been born. The cases mentioned above, if (as seems
plausible) not enough to support his claim that it is always better not to have been
born than to have been born, do a fine job of explaining a variety of widely shared
judgments that seem to lack an adequate alternative explanation.
Of course, not everyone will accept Benatar’s view or the supporting examples
given above. Utilitarians are committed to regarding acts that culminate in the
creation of pleasure as a moral good and acts that culminate in the absence of
pleasure as a moral evil. Since pleasure and pain are two ends of the same scale, and
increasing pleasure is a moral good and decreasing pleasure and increasing pain
a moral evil, they are committed to denying Benatar’s views and the supporting
examples.
Christians, of course, cannot accept an act utilitarian theory of morality for
a number of reasons. Most important among them is that act utilitarianism is
vulnerable to refutation by what are regarded as counterexamples among both
Christians and many non-Christians. It simply cannot be right, from the standpoint
of Christian ethics, to kill an innocent person simply because it maximally increases
pleasure in the relevant community.
Intriguingly, Benatar does not draw a conclusion I will draw later in this
chapter on the assumption certain traditional Christian doctrines about hell and
the conditions for salvation are true. In particular, he does not draw the conclusion
that having children is wrong. Although he thinks the considerations adduced
above show that there cannot be a moral duty to have children, he thinks that
having children can be morally justified if one has adequate reason to believe that
the benefits to the child will outweigh the harms:
7
Fyodor Dostoevsky, The Brothers Karamazov, trans. C. Garnett (New York, 1955),
Book V, Chapter 4.
Because most people who live comfortable lives are happy to have come into
existence, prospective parents of such people are justified in assuming that, if they
have children, their children too will feel this way. Given that it is not possible to
obtain consent from people prior to their existence to bring them into existence,
this presumption might play a key role in a justification for having children.8
Still, he concludes that it is always better not to exist than to be brought into
existence even if we can sometimes justify imposing the harm of being brought
into existence on someone.
While I accept the asymmetry that exists between the absence of a harm and
the absence of a benefit and will rely on this below, I do not accept Benatar’s view
that, bracketing the concerns I raise arising from certain Christian doctrines, it is
always better not to have been brought into existence. If, from a purely secular
perspective, a person has a sufficiently good life, I see no reason to think that it is
for that person worse to have been brought into existence than not to have been
brought into existence. From a purely secular standpoint, Benatar simply cannot
validly infer that it is necessarily better not to have been brought into existence
from the claim that being created involves a harm and the asymmetry between
absence of benefits and absence of suffering. He needs much more to make that
inference.
However, once you bring Christian exclusivism, the traditional doctrine of
hell, and some contestable views about what constitutes authentic Christian faith
into the picture, there is enough to make that inference based on what Benatar
seems to have gotten right. Indeed, I will go even further below, arguing that the
harm caused to every child by having been brought into existence is, assuming
certain traditional doctrines of Christianity are true, sufficiently great in most, if
not all, cases to entail that it is morally wrong to have children—notwithstanding
the Scriptural passages to the contrary—in order to show that either Christian
exclusivism or the traditional doctrine of hell should be abandoned or modified.
In other words, I assume, without argument, that the claim that it is nearly always
morally wrong to have children is inconsistent with doctrines more fundamental
to Christianity, including the doctrine that God is all-loving, than either the
exclusivist doctrine or the traditional doctrine of hell. Again, it bears emphasizing
that my point here is not that I really believe it is wrong to have children from
the standpoint of Christianity; it is to use this view as a reductio against at least
one of the two doctrines that help to imply this counterintuitive result—Christian
exclusivism and certain elements of the traditional doctrine of hell, namely that
hell is a non-empty place of eternal torment unmatched in severity to any suffering
we can experience on earth.
8
Benatar, “Why It Is Better Never to Come into Existence,” pp. 164–5.
Some Exceptional Cases
Both of the two positions discussed in the first two sections rest on a certain
conception of the moral quality of the act of procreation—in particular as being
either a morally good act or a morally neutral act, but never morally wrongful. As
Seana Shiffrin elegantly points out, this view cannot be justified, as procreation is,
as a matter of nomological necessity, a morally hazardous act. Since she explains
the point far more eloquently than I could, I let her speak for herself on this score:
I suggest a different moral perspective toward routine procreation, what I will

call the “equivocal view.” The view regards procreation as an intrinsically and
not just epistemically hard case … that ineliminably involves serious moral
hazards … Even though procreators may benefit progeny by creating them, they
also impose substantial burdens on them. By being caused to exist as persons,
children are forced to assume moral agency, to face various demanding and
sometimes wrenching moral questions, and to discharge taxing moral duties.
They must endure the fairly substantial amount of pain, suffering, disability,
significant disappointment, distress, and significant loss that occur within the
typical life. They must face and undergo the fear and harm of death. Finally they
must bear the results of imposed risks that their lives may go terribly wrong in
a variety of ways … all … without the child’s consent … Hence, procreation
is a morally hazardous activity because in all cases it imposes significant
risks and burdens upon the children who result. The imposition of significant
burdens and risks is not a feature of exceptional or aberrant procreation, but of
all procreation.9
I doubt that most people will agree that all procreation is morally hazardous or
that being brought into existence is, as Benatar believes, necessarily a morally
significant harm; after all, many of us live lives that are quite happy on the whole,
and procreation in cases where one has adequate reason to believe this is the case,
according to this intuition, is morally innocuous at worst and morally good at best.
I am not entirely sure this position is right. Although I would characterize my life
as being on the whole a “happy” one, I must confess that the grind of earning a
living and the increasing fear of my own mortality and what my ultimate fate might
be frequently tempt me with the thought that I might have been better off never
having been born. This is a thought that occurred to me during rough periods in my
youth as well, but growing older and facing the challenges that go with growing
older elicits that thought far more frequently in me—despite the fact that, on the
whole, I am enjoying my life greatly. Had I not been born, it is true that I would not
have experienced the goods that make life worth living; it is also true I would not
have experienced the suffering I have. Sometimes (but not always)—and notice
9
Seana Shiffrin, “Wrongful Life, Procreative Responsibility, and the Significance of
Harm,” Legal Theory, 5/2 (1999): 117–48, at 136–7.
this is a point different from the one made by Benatar—the latter seems to me to
outweigh the former. Although the former more frequently outweigh the latter, the
frequency of the thought that sufferings outweigh the benefits and that I would
have been better off not having been born is enough to convince me, at least, of the
intuitive plausibility of Shiffrin’s view that all procreation is morally hazardous.
The same is true for another view I think is not uncommon, at least among
people of this generation. Sometimes I just look at the state of the world and am
overwhelmed by the problems: crime, environmental degradation, a consumerism
that seems almost malignant in its passion, global poverty and the comparative
indifference of the affluent, wars being fought everywhere, and (in the case of
the U.S.) a national debt that has to be serviced no matter what. Despite my great
love for my nieces, I especially fear for the next and coming generations who face
unprecedented problems. This generation of 18-year-olds, for example, is the first
in U.S. history that is expected to have a shorter life span and be economically
worse off than its predecessor. My thought in response to this when deliberating
about having a child is this: “Is this a world I really want to bring a child into?”
Of course, one might think these latter thoughts are unique to my generation,
but I doubt this. This is just a gesture in the direction of an argument—and not an
argument that can do much work—but I imagine life has been pretty tough for all
preceding generations, even those that had the most reason for optimism. Crime
has always been with us. Up until quite recently, people needed to perform even
in the wealthiest nations very hard physical and menial labor just to survive—
and sometimes needed the help of their children to do so, long before they were
physically or mentally mature enough to do so without harm to themselves. The
disincentives were all there, whether noticed or not, but the incentives to have
children in conditions of lesser affluence have been stronger: people had to have
children to provide for their own economic security, especially during times of
infirmity that prevented sufficient productive activity to meet one’s basic needs.
In any event, nothing in the argument I want to make here really turns on these
intuitions. Shiffrin’s position that all procreation is morally hazardous and hence
problematic to some extent might very well overstate things a bit as far as ordinary
intuition is concerned, but what is important for our purposes is this: she surely
captures a widely shared view that what she calls “exceptional” or “aberrant”
procreation is morally hazardous.
Although most people regard the arrival of new child as a moral good, this
is not always true. There are a number of situations in which we may judge the
act of bringing a new life into the world, or specifically, the procreative act itself,
as being wrong. Consider, for starters, a couple who decides to conceive a child
knowing there is a high probability that the child will be born with a terrible
condition that will result in a short and terribly painful life—pain that is so bad
that the child cannot even be picked up without exacerbating it. Assume also that
this same couple would face a morally insignificant risk of giving birth to a child
with this condition if the couple simply delays conception by a few years. Whether
or not the child is born with the condition (but especially if he or she is), it is quite
plausible to think (and nearly everyone I asked about this case took this position
even while exhibiting sympathy for the parents) the parents committed a moral
wrong—and one against the child—if they elected not to delay conception.
As Richard Brandt observes in a case where the probability of conceiving a
child with such a condition is temporary:
Obviously, we think it would be outrageous morally for the mother not to delay.
Of course, if she delays, she will not have the same child as the one she would
have had if she had not delayed; but we do not think we need worry about any
rights of the child she might have had, in view of the fact that the later-conceived
child will have a better life.10
Brandt incorrectly thinks this judgment is grounded in a moral judgment that

is broad enough to support euthanizing such infants without their consent. This
intuitive reaction to the thought experiment is quite common: it would be wrong
not to delay conception if doing so would result in a normal child—and people
think the act of procreation is a wrong under these circumstances that ultimately
results in a wrong against the child even though the child does not exist prior to
the act of conception.
As far as I can tell in talking to a small sample, the intuition extends to cases
where there is a permanently high probability that conception will result in a child
with such a condition. Even when they have no reason to think they will conceive a
child with a terminal painful condition, parents who do so usually feel tremendous
guilt and remorse—as if they had done something wrong. Similarly, people utterly
desperate to have a child will regard information that there is a high probability
that they will conceive a child with a terminal painful condition as a good reason
not to have a child— not just for their sake as a matter of prudential rationality,
but for the sake of the child who would be born as a matter of morality. If there
are people who think it morally permissible all things considered to have a child
who is likely to be born with such a condition, they surely recognize that the fact
that the terrible quality of this child’s life is at least a prima facie moral reason to
abstain from conception.
Here the intuition might not be quite as strong because the mother and father may
feel that they have some sort of right to procreate that might partly counterbalance
the wrong that is done by conceiving a newborn with such a condition in cases
where they know the probability of doing so is high. If so, they misunderstand
what it means to have a moral right to procreate. The idea that one has a right
to procreate, if correct, is consistent with the idea that procreating under certain
circumstances is wrong. As odd as it may seem, we have a moral right to commit
some moral wrongs; in other words, it would be wrong for anyone to coercively
interfere with our commission of a morally wrongful act in order to prevent or
10
Richard Brandt, “Defective Newborns and the Morality of Termination,” in John
Arthur (ed.), Morality and Moral Controversies, 7th edn (Upper Saddle River, 2004), p. 232.
punish it. Thus, for example, it seems clear that it would be wrong for the state to
criminalize an ordinary unilateral lie that is not likely to induce harmful behavior
on the part of the victim—unlike fraud and commercial deception where untruths
are used to induce consumers to make purchases they would otherwise not make.
In any case, the intuition is common here that such parents have a moral duty to
abstain from conception and that this duty is owed to any future children they
might have with the condition.
One might reasonably think that the parents would be committing a wrong
against the child even if the child were born without the predicted condition.
Reckless or negligent risk-creation is itself considered a wrong. For example,
we consider driving while under the influence of an intoxicant a moral wrong
regardless of whether anyone is hurt precisely because such behavior results in
wrongful risk-creation. If I am driving on the same road as the intoxicated driver,
my risk of death or grievous bodily injury is increased to a morally significant
extent. Thus, the driver’s behavior is wrongful regardless of whether this risk is
realized. Again, here, the wrong would be the procreative act and the wrong would
be committed against the child.
Indeed, driving while intoxicated is prosecuted, like all crimes, by the state,
instead of citizens, because it is regarded as a breach of the peace and hence a
wrong against the community regardless of whether the driver has an accident or
otherwise physically injures someone. Creation of a substantial risk of harm to
innocent persons without adequate justification is generally considered, by itself,
a moral wrong to all those persons subject to the risk. In the case of a driving while
intoxicated (DWI), this includes not only everyone who is on the road at the same
time, but everyone who might have been there.
At this point, a couple of objections may occur to the reader. The first objection
is that it is simply not possible to assert that an act is morally wrong based on a
probability of risk-creation of a significant harm that is less than 1. But this is
inconsistent with intuitions that are widely shared in our culture. Driving under
the influence is morally wrong because there is a morally significant probability
that it will result in significant harm to others; indeed, it is for this reason that it
is legitimately criminalized. Acts that are highly likely to result in risk-creation
of significant harms to other people that cannot be justified by some more
important moral benefit are typically characterized as morally wrong and often, in
consequence of the probable harm, justifiably criminalized.
Indeed, it may well be that the threshold probability level for wrongful risk-
creation might be much lower than one might initially think. In the state of
Washington, one can be arrested for driving with a blood alcohol level of .08—and
presumably legitimately so in virtue of its wrongful risk-creation. But it is not at all
clear how high the probability of an accident is when one of the drivers has a blood
alcohol level of .08. While the probability of an accident increases as blood alcohol
level increases, it is just not clear within a certain range of blood alcohol levels
beginning with .08 exactly what the probability is that the driver’s intoxication will
cause an accident. Driving while under the influence is probably much more common
than anyone is comfortable believing; however, accidents involving drunk drivers

are comparatively rare—and usually involve much higher blood alcohol levels.
The second objection is that a procreative act simply could not be a wrong
against a person who is not yet existing, the underlying intuition being we cannot
wrong presently non-existent but future persons; but there are a number of cases in
which we seem to have a contrary intuition. Environmentalists frequently argue—
and not counterintuitively—that we owe future generations an obligation to protect
the environment; if so, then present acts that pollute the environment now wrong
those persons, even though they do not presently exist. Similarly, Joel Feinberg
argues that if a time bomb is placed in a kindergarten that is guaranteed to go off
seven years later at a time when all the children are present is a wrong—and one
that is committed against the children even though they do not exist at the time the
bomb is activated.11 Indeed, one might think that these acts violate the child’s right
to life even though they do not exist prior to the existence of the child. As Seana
Shriffin explains, “Our moral duties emanate from the force these future rights exert
on us now, not from any right predicated to be held by nonexistent persons.”12
Indeed, it is worth noting that wrongful life that occurs in cases of what
Shiffrin has termed exceptional or aberrant cases of wrongful procreation is legally
actionable in a few states. These suits are comparatively unusual, but they are filed
in cases where a child with a severe disability seeks damages from parents who
conceived and gave birth to her knowing of the risks. The suits have sometimes
succeeded and sometimes not, but the underlying rationale is that parents who
know that procreation runs a significant risk of resulting in a child with special
disabilities or a painful condition but engage in procreative activity that proximately
results in such a child are engaging in behavior that is sufficiently wrongful from a
moral point of view to be justifiably actionable in tort. Accordingly, while law and
morality are conceptually distinct, the underlying rationale for liability in wrongful
life lawsuits is that engaging in procreative activity under these conditions is
(1) morally wrongful behavior that (2) foreseeably and (3) proximately causes
significant harm to another person. Of course, legislators and judges are not
morally infallible, but these cases when they occur do not precipitate the kind of
moral outrage that occurred when McDonald’s was ordered to pay $3 million to a
woman who suffered burns after spilling hot coffee that she held between her legs
while driving. This suggests that the wrongful life cases are in line with widely
shared moral views.
These are not the only cases in which the birth of a new child is regarded
as potentially morally problematic and, indeed, possibly the violation of a moral
duty. Here is an example interesting for its complexity. We tend to regard news of
a pregnancy in an unmarried woman who is too young to have a child as morally
problematic for a number of reasons. Christians believe it wrong to engage in
11
Joel Feinberg, Harm to Others (Oxford, 1987), p. 97.
12
Shiffrin, “Wrongful Life, Procreative Responsibility, and the Significance of Harm,”
138.
sexual relations outside of marriage; however, even liberal atheists hold that men
and women should not become sexually active until they reach a certain level
of maturity and should take certain precautions to prevent pregnancy if they do
become sexually active before they can provide support for a child.
These, of course, are moral judgments that some moral obligation has been
violated—although, apart from God, it is not entirely clear to whom these
obligations might be owed. But this much also seems clear: we intuitively believe
that the child has been wronged for a number of reasons. First, it is highly likely that
a young unmarried woman is unable to provide adequate care for the child—both
in terms of what material resources a child needs to thrive and in terms of what
kind of guidance and care a child needs. Young unmarried women simply lack
many of the kinds of resources needed to be good parents who can prepare a child
for a happy life. Second, even if a young mother does what is in the best interest
of the child by giving her up for adoption, that child will eventually become aware
of that fact and have to come to terms with it, which can be a very difficult matter
requiring the help of professional counseling. Either way, the judgment is that the
young man’s and young woman’s irresponsible sexual behavior culminated in a
harm to a child who should not, as a moral matter, have been conceived. Here,
again, a moral wrong seems to have been done to the child by having brought her
into the world. This is clearly a case in which, as Shiffrin might put it, procreation
is a morally hazardous act.
Many of us in affluent nations with comparatively low birth rates that can
be sustained given the amount of material resources immediately at our disposal
tend to regard new births in countries that cannot support a larger population as
potentially being problematic. I say “potentially” here because this case involves
factors that complicate the judgment. For example, a woman might not have
the requisite knowledge of alternatives, like contraception or (if she is really
undereducated) abstinence. But where she has such knowledge and knows that
she and her mate, if there is one, cannot support that child, subjecting him or her
to a high probability of malnutrition and its attendant health problems, which can
ultimately lead to death, many of us share the intuition that a wrong has been
committed and that wrong has been committed against the child who has to face
burdens that he or she ought not to have faced because he or she ought not to have
been conceived. Procreation here, as in the other cases, seems to have resulted in
a moral wrong against the child.
Capturing these widely shared views is a general principle defining any
moral obligations we might have regarding procreation where morally hazardous
procreation results in a wrong against the child. This principle is vague and can
roughly be stated as follows:
New Life Principle (NLP): It is morally impermissible to bring a new child

into the world when there is a sufficiently high probability when doing
so will create a substantial risk that the child will invariably suffer severe
harm as a direct consequence of being born.
Something akin to NLP is both intuitively plausible and explains our intuitions
that a child has been wronged by being brought into the world in each of the cases
above. Of course, this principle is vague in three important respects that reflect
some vagueness with respect to the cases in which it applies. The first source of
vagueness occurs with the term “sufficient” in “sufficiently high probability.” It
is simply not possible to quantify exactly where the line is between a probability
of risk-creation that is sufficiently high and one that is not. I would be surprised if
anyone had reliable statistics with respect to the probability that driving under the
influence of an intoxicant might result in a harm; after all, many DWIs are missed
by police who simply cannot be everywhere at the same time.
The second source of vagueness is with the term “substantial.” It is probably
not possible to quantify this notion by drawing a line between what counts as a
probability constituting a substantial risk and what does not. For example, it seems
to me that even a probability of .5 that a child will be born with a condition that
will kill him or her within a short and extremely painful period of time when his or
her pain cannot be alleviated in any way constitutes a substantial risk—one that is,
from a moral standpoint, unacceptable to take. But I suspect at least some people
would disagree on this case.
The third source occurs with the term “severe.” Exactly what counts as a severe
harm is just not clear. Suppose we know to a moral certainty that conception will
result in a child with such severe cognitive disabilities that he or she will never be
able to learn much more than the rudiments of a language and will never be able
to learn even the most basic arithmetic. For my part, I do not know whether this
constitutes a sufficiently severe harm that would require parents to abstain from
conceiving, and I suspect intuitions among ordinary folk and perhaps even among
Christians will cover the entire spectrum. The same is likely true for severe cases
of Down’s syndrome or cerebral palsy. As far as I am concerned, I do not have
sufficient confidence about my intuitions in any of these cases to be able to state
NLP with more precision than is stated above.
Even so, such vagueness is not sufficiently problematic to warrant rejecting it.
After all, many moral principles that seem uncontroversial have areas of vagueness.
Certainly, most people would agree that the state should criminalize any wrongful
act that results in severe harm to others—“severe” having the same vagueness
here that it does in NLP. As for the term “substantial risk,” we criminalize driving
under the influence of an intoxicant precisely because it creates a substantial risk
of harm to others. Many states have arrived, perhaps somewhat arbitrarily, at a
threshold number measuring the amount of alcohol in the bloodstream that triggers
automatic liability; however, it is just not clear what the odds are, as a general
matter, that driving with an amount of alcohol in one’s system that meets, but does
not exceed, the threshold triggering liability will culminate in an accident harming
other persons. The moral life is difficult precisely because we have to try to rely on
principles that we are all too often unable to state with ideal precision.
A Sufficiently High Probability of a Substantial Risk of Severe and Eternal

Harm
In this section, I assume that certain traditional doctrines of Christianity are true.
First, I assume that Christian inclusivism is false: that is, I assume that it is a
necessary, although not sufficient, condition for salvation that one believe that the
core doctrines of Christianity are true. I say here that it is not a sufficient condition
because I assume that more than propositional belief is required. At the very least,
a personal relationship with Jesus requires that we express genuine remorse for
our sins and ask forgiveness for them. A genuinely saving faith simply cannot
involve mere propositional belief without any change in one’s dispositions to act
or one’s attitudes toward one’s wrongful acts.
Second, I assume that the traditional doctrine of hell is true. In particular, I
assume that universalism is false and that there are people in hell for eternity
suffering torment unmatched in severity to anything they could possibly experience
in this world. Putting these two doctrines together, I assume that the fate of people
who lack authentic Christian faith—however this turns out to be defined beyond
its being a necessary condition for salvation that one believe Christianity’s core
doctrines as summarized in one of the creeds—is eternal torment in hell. It should
be clear that not only is eternal torment in hell a “severe harm” if anything is
a severe harm, but also the most severe harm any human being can face. One
necessary condition for the application on NLP is satisfied.
Now if Christian inclusivism is false, then anyone lacking authentic Christian
faith will inevitably suffer eternal torment in hell. I think it is fair to characterize
the inevitability of such a fate as a “substantial risk of a severe harm.” In terms
of probability, there is no possibility a person can escape such a fate if Christian
inclusivism is false and the traditional doctrine of hell is true. I cannot think of a
more substantial risk than one that is realized as a matter of logical or metaphysical
necessity. Two conditions for the application of NLP are satisfied.
The third is also satisfied for much of the world’s population if not for all of it.
Consider, first, that only 2 billion of the world’s population claim to be Christian
at least in the minimal sense of believing that the core doctrines are true. The
remaining more than 4 billion people would deny at least one of these claims. Jews
and Muslims would, for example, deny that Jesus is Lord and Savior; in addition,
Jews would deny that Jesus is the Messiah. Buddhists and many Hindus would
deny not only the first of these claims, but even that God is a personal being. And,
of course, there is a growing number of agnostics and atheists around the world—
many of whom are quite ethically conscientious. If Christian inclusivism is false
and the traditional doctrine of hell is true, the probability that any child brought
into the world, calculated from a global perspective, will suffer the most severe
harm imaginable is greater than 2 in 3. Although it is, of course, not clear where
exactly to draw the line between a sufficiently high probability of risk creation and
one that is not sufficiently high to trigger application of NLP, ordinary intuitions
and practices suggest 2 in 3 is clearly high enough; after all, it would be very
surprising if the risk of harm caused by an accident instigated by someone driving

under the influence is anywhere near that high. Considered from the standpoint of
a global perspective, NLP implies it is wrong to have children.
Nevertheless, one might think this is simply not the right way to apply NLP.
The reasoning here is that the probability of authentic Christian faith differs
from one geographical location to the next. Although it is prohibitively low in
the Middle East, it is comparatively high in the United States. Indeed, the most
reliable sociology we have on the issue is that the very best predictor of a child’s
religious beliefs and her ethical commitments towards those beliefs are those of
her parents (which, of course, creates other problems by introducing salvific luck
into the picture since one does not choose one’s parents).13
This should not be thought either surprising or limited to children in the U.S.
Parents undoubtedly play a profound role in determining early on which among
contestable views will form part of a person’s views about the world. Social
psychologists have found a strong statistical correlation between a parent’s views
on religious and moral matters and those that the child will come to embrace.
So strong are these correlations that Michael Argyle and Benjamin Beit-Hallahmi
conclude that “[t]here can be no doubt that the attitudes of parents are among the
most important factors in the formation of religious attitudes.”14
There is, of course, nothing particularly controversial or mysterious about
any of this. As C. Daniel Batson points out, “[since p]arents are a major source
of social rewards and punishments … the child experiences strong pressure to
conform to their wishes.”15 Given a child’s strong desire to please her parents, it
is quite natural that parents exercise a strong influence on the development of a
child’s moral and religious views. Indeed, the assumption that parents exercise a
profound and lasting influence on a child’s moral and religious beliefs is at the
foundation of most, if not all, mainstream normative views on rearing children.
Anyone who takes the time to try to bring a child up to love God or to respect
morality must believe that her efforts will have more than a transient influence on
the child’s views and personality.16
13
See Christian Smith and Melinda Denton, Soul Searching: The Religious and
Spiritual Life of American Teenagers (Oxford, 2005).
14
Argyle and Beit-Hallahmi, The Social Psychology of Religion (London, 1975), p.
30. The interested reader is invited to consult Chapter 3 of this volume for a useful survey
of the relevant statistical evidence. It is worth noting, however, that the evidence tracks only
parent and child Christian affiliation.
15
C. Daniel Batson, Patricia Schoenrade, and W. Larry Ventis, Religion and the
Individual: A Social-Psychological Perspective (Oxford, 1993), p. 43.
16
This is not, of course, to say that parental efforts “cause” the child’s beliefs—at
least, not in any sense that is obviously inconsistent with free will. It should be clear that the
claim that parents can exercise a lasting influence on a child’s development does not entail
accepting a deterministic view of human behavior. This important point should be kept in
mind for the remainder of the chapter.
Of course, the correlation is not perfect. Parents are not, as one might expect,
the only persons to have an impact on a child’s views. Other groups also exercise
profound social influence on the development of a child’s views. As the child
grows older, the influence of peers and teachers becomes increasingly important
relative to that of parents. In discussing the formation of religious views, for
example, Raymond F. Paloutzian observes:
During childhood, two main factors, family and church, work to mold the child
in varying degrees toward (or against, in the case of a nonreligious family)
religion. During adolescence, on the other hand, new social factors enter the
picture … The first new influence is peers. Although peers influence younger
children, they are an especially potent influence during adolescence … The
second critical influence is school.17
Peer and teacher influence, of course, can cut both ways: it can either reinforce
or challenge what parents have taught. For this reason, there is no guarantee
that early parental training will have the desired long-term impact on the child’s
views; widespread exposure to contrary influences can help to facilitate a person’s
rejection of her parents’ religious or moral views. But, either way, it is clear that
belief formation and maintenance is subject to ongoing social pressure throughout
the maturation of the child.
What this fact buys us, if it is a fact, is still not satisfactory. It implies that
we should calculate probabilities by region in applying NLP. This would appear
to entail that the probability that someone who is brought into the world in
fundamentalist Islamic communities will acquire authentic Christian faith is
so low that NLP implies it is morally wrong for parents in this region to have
children. On the traditional views about inclusivism and hell, this, to put it bluntly,
amounts to saying fundamentalist Muslims have a moral duty, albeit one of which
they are unaware, to allow themselves to become extinct. Perhaps some Christians
are comfortable with this conclusion; I certainly am not—and am beginning to see
such implications as entailing that at least one of the two traditional doctrines that,
together with NLP, have such counterintuitive results ought to be rejected.
But any optimism that these facts will show that, under NLP, it is morally
permissible, at the very least, for the 2 billion people characterizing themselves as
Christian to have children is premature. The problem here consists precisely in the
fact that belief that the core doctrines of Christianity are true is necessary but not
sufficient for authentic Christian faith.
There are two related epistemic problems here that get in the way of our being
able to apply NLP to these cases. First, it is just not clear how many of these
17
Raymond F. Paloutzian, Invitation to the Psychology of Religion, 2nd edn (Boston,
1996), p. 126. I choose a textbook to underscore the uncontroversial nature of the claims
being made. See also Argyle and Beit-Hallahmi, The Social Psychology of Religion,
pp. 33–46, for a discussion of the relevant empirical evidence.
2 billion people in the world who claim to be Christian in virtue of having the
requisite beliefs have a saving authentic Christian faith. Many Christian students
with whom I have spoken are tempted to put the percentage of nominal Christians
who are authentic Christians at a surprisingly low .2 (or 1 in 5). If they are correct,
then it would appear that NLP applies, as a general matter, to Christians as well
as non-Christians. Even if the number were 1 in 2, I would be tempted to think
that the application of NLP has the same result: a .5 risk of eternal suffering is an
unacceptably high risk for any person to be subjected to without their consent.
Of course, if the more optimistic number is correct, there is a 1 in 2 chance
that they experience eternal bliss. But here is where what David Benatar calls the
asymmetry problem plays an important role. The avoidance of suffering—even
if it is avoided because the possible being who would have experienced it does
not exist—is considered a moral good. In contrast, pleasure that does not occur
because of the non-existence of the possible being who would have experienced it
is not considered a moral bad. NLP implicitly reflects the asymmetry problem in
comparing pleasures and pains that do not occur because the possible being who
would experience them does not exist.
Second, and related to the first problem, it is simply not possible to be
epistemically justified with respect to any theory of what constitutes authentic
saving Christian faith. There is a striking range of disagreement on this issue,
which should be enough to show that we do not have enough information on the
question that would clearly show which view is correct, and hence that no one
is epistemically justified in whatever view they take on the issue (and I say this
with great reluctance believing some doctrines of easy salvation seem obviously
inconsistent with Jesus’ words in the Gospels). Some Christians believe that all
that is needed is remorse for one’s sin and a prayer asking for forgiveness. Others,
like myself, believe that a more demanding standard defines authentic Christian
faith: on this standard, which most nominal Christians would fail, one would have
to make a good faith effort to love one’s neighbors as one loves oneself, which
requires a great deal by way of personal sacrifice. In particular, it requires that
spending of discretionary income on luxury goods be diverted to saving lives in
the developing world where malnutrition and AIDS are pandemic. On no plausible
interpretation of this imperative can someone who spends $30 on a shirt they do not
need when that money can be used to sustain the life of a child in absolute poverty
for a month be plausibly characterized as loving that child as herself. After all,
the child’s life is at stake, while nothing more important than the affluent person’s
fashionability is at stake. And there are presumably a wide range of beliefs on this
issue in between these two extremes.
The problem here is that we lack sufficient information to determine what
counts as authentic saving faith and so we cannot even begin to estimate the most
relevant probability in applying NLP. Moral concern dictates that, under conditions
where we lack such important information, we guard against the most catastrophic
of outcomes—which, of course, would be eternal torment in hell. Even where
nominal Christians are concerned, it appears that it is morally wrong to have
children not because the conditions of NLP are clearly satisfied, but because we
are not in any epistemic position to determine whether they are—and are morally
required to adopt a more conservative strategy in protecting the interests of potential
future beings. And this we Christians can do only by refraining from bringing
new life into the world. It appears that, given ordinary intuitions about morality
and the exclusivist doctrine and the traditional doctrine of hell, we Christians,
like Muslims, should allow ourselves to become extinct because what we believe
constitutes authentic saving Christian faith is the most reliable predictor of what
our children believe—and none of us are epistemically justified in our views about
this. No one can give an argument that successfully refutes any of the mainstream
views in play. That is why disagreement continues on this issue.
Conclusions
The reader might be relieved to know that I do not believe that it is morally wrong,
as a general matter, for anyone to bring children in the world—unless we are
in danger of overpopulation, or the mother is young and cannot provide for the
child, or the child has a substantial risk of some severe disease or disability. But
I do believe that NLP and the requirement that we adopt a conservative strategy
for protecting people against catastrophic results are, for the most part, correct
statements of moral principles. Since I believe these two principles, together with
Christian exclusivism and the traditional doctrine of hell, entail that it is morally
wrong for anyone to have children, I believe that either Christian exclusivism or
the traditional doctrine of hell (or both) should be rejected as core doctrines of
Christianity in favor of views that seem far more consistent with God’s moral
perfection and omnibenevolence. It seems morally arbitrary that what explains why
God shows mercy to an ethical devout Christian with eternal bliss and condemns an
equally ethical devout Muslim to eternal torment is that the former has the correct
view about God’s nature and the latter has an incorrect view (i.e., the disagreement
about the Trinitarian nature of God is, after all, just a disagreement about God’s
nature). If Christian exclusivism, the traditional doctrine of hell, and ordinary
moral intuitions imply that it is always morally wrong to have children, then one of
these views should be rejected. Of course, one can always bite the bullet and give
up the ordinary moral intuitions that are captured in NLP, but the more reasonable
response, I think, is to give up one of the other views, which, quite honestly, cause
so many philosophical problems that they seem far less reliable than the intuitions
expressed in NLP. But I cannot even begin to defend that point here.
Chapter 13
Species of Hell
John Kronen and Eric Reitan
In recent years, numerous scholars have challenged the rationality of the traditional
Christian doctrine of hell (DH)—by which we mean the doctrine that some created
persons will endure eternal suffering in alienation from God, or damnation. Critics
of DH typically argue that, assuming the broad framework of Christian doctrine,
God would have no reasons compatible with His moral character (what we will call
“God-justifying reasons”) for permitting anyone to suffer eternally.1 Defenders of
DH, by contrast, think there may be such reasons.2
These debates are complicated by the fact that, historically, DH has been
explicated in different ways. More significantly, DH admits of variations that may
have bearing on whether there could be God-justifying reasons for permitting
or bringing about damnation, and if so, what those reasons might be. Our main
purpose in this chapter is to offer a “taxonomy” of species of DH, distinguished
in terms of what kind of answer they provide to two questions: First, what is the
1
For examples of articles developing this line of argument in various ways, see
Eric Reitan, “A Guarantee of Universal Salvation?,” Faith and Philosophy, 24/4 (2007):
413–32; John Kronen and Eric Reitan, “Talbott’s Universalism, Divine Justice, and the
Atonement,” Religious Studies, 40/3 (2004): 249–68; Eric Reitan, “Eternal Damnation and
Blessed Ignorance: Is the Damnation of Some Compatible with the Salvation of Any?,”
Religious Studies, 38/4 (2002): 429–50; Thomas Talbott, “Three Pictures of God in Western
Theology,” Faith and Philosophy, 12/1 (1995): 79–94; Marilyn McCord Adams, “The
Problem of Hell: A Problem of Evil for Christians,” in Eleanor Stump (ed.), Reasoned Faith
(Ithaca, 1993), pp. 301–27; Thomas Talbott, “Providence, Freedom, and Human Destiny,”
Religious Studies, 26/2 (1990): 227–45; Thomas Talbott, “The Doctrine of Everlasting
Punishment,” Faith and Philosophy, 7/1 (1990): 19–42; and John Kronen, “The Idea of
Hell and the Classical Doctrine of God,” The Modern Schoolman, 77/1 (1999): 13–34.
2
Book-length defenses of the doctrine include Charles Seymour, A Theodicy of
Hell (Dordrecht, 2000); Jonathan L. Kvanvig, The Problem of Hell (Oxford, 1993); and
Jerry L. Walls, Hell: The Logic of Damnation (Notre Dame, 1992). Philosophical articles
include William Lane Craig, “Talbott’s Universalism,” Religious Studies, 27/3 (1991):
297–308; William Lane Craig, “‘No Other Name’: A Middle Knowledge Perspective on the
Exclusivity of Salvation Through Christ,” Faith and Philosophy, 6/2 (1989): 172–8; Eleanor
Stump, “Dante’s Hell, Aquinas’s Moral Theory, and the Love of God,” Canadian Journal of
Philosophy, 16/2 (1986): 181–98; Richard Swinburne, “A Theodicy of Heaven and Hell,” in
Alfred J. Freddoso (ed.), The Existence and Nature of God (Notre Dame, 1983), pp. 37–54.
nature of the evils endured by the damned? Second, what is the impelling cause
of damnation?3
These questions help to distinguish species of DH in a manner relevant for
determining what God-justifying reasons, if any, there might be for damnation.
Hence, once we have identified species of DH in these terms, we will be positioned
to determine the kinds of God-justifying reasons that God would need to have in
order to justify each species. In each case, we will briefly argue that there is a
prima facie basis for thinking either that the identified reasons are ultimately not
God-justifying (they cannot be reconciled with God’s moral character) or that,
while they are motives God could have, they are not ones that would motivate Him
to permit damnation.
These will be only sketches, because our purpose here is not to provide an
exhaustive case against DH but to identify hurdles that defenders of any version of
DH will need to overcome. Our conclusion will be that, at least from a philosophical
standpoint, the burden of proof rests with those who think there is some species of
DH Christians should embrace.
God-Justifying Reasons for Damnation
It is usually agreed in current debates that the Christian God would prevent
damnation in the absence of God-justifying reasons not to. Put more precisely, the
assumption is that God’s moral character is such that He possesses a prima facie
compelling motive to save from damnation any created person that He can save.
This assumption is plausible under a broadly Christian understanding of God,
insofar as it seems to follow from the doctrine that God is love—usually taken to
imply that God loves all created persons. Such love would seem to entail that God
has at least some motive to save the damned if He can, since the damned are created
persons and hence objects of divine love, since His love for them would lead God to
desire their welfare, and since damnation is clearly contrary to their welfare.4
3
In most contemporary discussions of the doctrine of eternal hell, two broad species
are generally acknowledged: those according to which the sufferings of hell are imposed
by God as a punishment for sin, and those according to which the sufferings of hell are a
consequence of the damned’s free decision to reject God and His offer of salvation. Thus,
for example, Kvanvig argues for the replacement of the “Retribution Thesis” (which he
identifies with the classical “strong” view of hell) with what he calls the “Self-Determination
Thesis”—according to which the damned are in hell “because of their determination to
avoid the company of the redeemed and the God who redeems” (Kvanvig, Problem of Hell,
p. 158). While we do not reject this distinction (it being the central difference in relation
to the question of damnation’s cause), our aim in this chapter is to offer a more nuanced
taxonomy, in part by focusing on different views concerning the suffering of the damned.
4
Both Kvanvig and Walls begin their arguments for the possibility of hell from
the assumption that God is perfectly good, and they both construe this to mean that, in
Kvanvig’s words, “God’s fundamental attitude towards human beings is one of love”
Species of Hell 201
Obviously, one God-justifying reason for His failure to save all might be
that it is simply not in His power to do so. Since God cannot be obligated to
do what He cannot do, His failure to save the damned would then be justified.
More commonly, however, a God-justifying reason for permitting damnation
posits some kind of clash between God’s prima facie motive to save all and other
motives which, if acted on, would preclude Him from doing so, and holds that it
is more consistent with God’s moral character to act on the latter motive(s) than
the former. In classical theological terms, it is God’s antecedent will to save all but
not His consequent will.5
In speaking here of God’s motives consistent with His moral character, rather
than about God’s duties, we are acknowledging that some thinkers believe God
has no duties—God, as the ultimate authority, has the right to do as He pleases.
While we do not accept this view ourselves, we do not want to exclude it in
advance. But even those who, like Marilyn McCord Adams, believe that God has
no duties, nevertheless believe that God has a definite moral character and that
there are some actions which it would be incompatible with that character for
God to perform (e.g, acts of arbitrary torture).6 Our own view is that God has a
duty to refrain from arbitrary torture, and that God in fact refrains because His
moral perfection will not permit Him to violate a duty. But even those who find
the language of duty inappropriate with respect to God would typically admit that
God’s moral character precludes arbitrary torture.
And to admit this is to accept that moral character influences, often decisively,
what one does or does not do. In other words, one’s moral character gives rise to
motives for action, the totality of which excludes some actions, permits others,
and necessitates still others. To say that one’s motives exclude or necessitate
certain actions is to say there is a kind of contradiction between possessing a
certain motive-set and behaving in certain ways. The sense of “contradiction” at
work here is difficult to spell out precisely but seems an inescapable feature of our
ordinary thinking about psychology and behavior.
(Kvanvig, Problem of Hell, p. 119). Both admit that this fundamental attitude implies that
God, in Walls’s words, “wants all persons to accept salvation” (Walls, Logic of Damnation,
p. 84). Their views here are fairly typical, at least of the current discussion, and they imply
that, at least prima facie, God would save anyone He could save.
5
On this distinction, see Leibniz, “A Vindication of God’s Justice,” paragraphs 24–6,
which can be found in Monadology and Other Philosophical Essays, trans. Paul Schrecker
and Anne Martin Schrecker (Indianapolis, 1965), pp. 118–19. Leibniz notes here that he is
using the notions of God’s antecedent and consequent will much in the way that Scotus and
Aquinas did.
6
Marilyn McCord Adams maintains, in the spirit of Anselm and Duns Scotus, that
God has no obligations with respect to creatures (Adams, “The Problem of Hell,” p. 308),
but she also maintains that there are things God would not do to creatures because it would
be cruel—and God, in her view of His moral character, is not cruel (ibid., p. 311).
But even if it is difficult to precisely characterize the sense in which behavior can
contradict motives, we can say a bit more about what a motive is, and what it means
to say that the totality of one’s motives excludes or necessitates some actions.
By a “motive,” we mean, roughly, either a reason for action that is taken to be
a good one by the agent (whether it is or not), or a desire for an anticipated end
of the act. Each disjunct identifies a distinct species of motive, what we might
call intellectual and affective respectively. These may interact in various ways.
I might desire an end and judge its attainment a good reason to act—in which
case my intellectual and affective motives reinforce one another. Alternatively,
I might desire an end which I judge neither worthy nor unworthy of pursuit on
any grounds distinct from the mere fact of my desire—in which case my affective
motive is independent of but compatible with my intellectual motives. Finally, I
might desire that which, in my judgment, I have good reasons to avoid, in which
case my intellectual and affective motives conflict.
In the case of God, there will presumably not be cases of the final sort, since
moral perfection precludes such clashes.7 But conflicts among intellectual motives
may be possible, at least on the assumption that some such motives are merely
prima facie ones, in the sense that they would be operative—that is, they would
inspire actions of the relevant kind—only in the absence of other motives that
override or defeat them. The idea here is that God may possess hierarchically
ordered intellectual motives, with lower-order motives generating action only
when they do not conflict with higher-order ones.8 Given the assumption that God
is both morally perfect and all-knowing, we can conclude that all such hierarchical
orderings in God’s motivational structure are the ideal ones—that is, they fully
reflect God’s judgment about what is best, and God’s judgment about what is best
is impeccable.
At least in the case of God, then, the totality of God’s motives would necessitate
(or exclude) an action A just in case God possesses at least one undefeated motive
to do (not do) A. Such a motive would have to be either (a) an intellectual one
7
It is an interesting question, one we will not explore here, whether God could just
so happen to desire certain things which there are no reasons not to desire but also no
reasons to desire (apart from the mere fact that one does desire them). In other words, we
will not here explore whether God might have affective motives that are not in some sense
an outflow of divine intellectual motives. All we need note here is that, if there are such
motives, they will not conflict with the totality of God’s intellectual motives for action.
8
For several classical articulations of this idea, see Heinrich Schmid, The Doctrinal
Theology of the Evangelical Lutheran Church (Minneapolis, 1960), pp. 282–3. Especially
useful is the quotation from Hollaz, p. 283. It should be noted that while the distinction
between God’s antecedent and consequent will characterizes much medieval, as well as
classical Lutheran theology, the theological determinism of the Calvinists rendered the
distinction virtually impossible for them to make and seemed to drive them, logically if not
admittedly, to the conclusion that God wills sin, not per se, but in order to manifest the glory
of His justice and mercy. See Heinrich Heppe, Reformed Dogmatics: Set Out and Illustrated
from the Sources, trans. G. T. Thomson (Grand Rapids, 1978), pp. 90–92, 143–9.
Species of Hell 203
which is such that there is no other intellectual motive that both conflicts with it
and has a higher place in the hierarchical ordering of God’s intellectual motives,
or (b) an affective motive which does not conflict with any intellectual motives.
God’s motive for saving each creature that can be saved would seem to us to be
an intellectual one, insofar as each creature has intrinsic value, and insofar as
damnation amounts to a vitiation of this value.
If we are right about this, then we can see in general terms what a God-
justifying reason for permitting a creature’s damnation would look like. Assuming
that God’s moral character implies a prima facie intellectual motive for saving
each creature (a prima facie saving motive), a God-justifying reason not to do so
would be some other intellectual motive that conflicts with the prima facie saving
motive and has a higher place in God’s motivational structure. In short, a God-
justifying reason not to save all would be a reason for action that entails not saving
all and which God judges to be not merely a good reason to act, but a better one
than God’s reasons for saving all.
So far, we have expressed all of these ideas in terms that explicitly avoid the
contestable assumption that God has duties. But while we have done so in this
section, we find it stylistically clumsy to continue doing so. And we believe that
nothing of substance hinges on the decision, for ease of expression, to speak instead
in terms of God’s actual and prima facie duties. And so we will do so when it is
stylistically clumsy to do otherwise, and we simply invite those who are so inclined
theologically to make the following translations: “God has a duty (that is, an actual
duty) to do (not do) X” can be taken to mean “X is necessitated (excluded) by the
totality of motives generated by God’s moral character”; “God has a prima facie
duty to do (not do) X” can be taken to mean “God has a prima facie intellectual
motive for doing (not doing) X, which is such that God would do (not do) X in the
absence of any higher motives that defeat the prima facie motive.”
With these introductory ideas in place, we are in a position to offer our
taxonomy of DH species.
A Generic Definition of the Doctrine of Hell (DH)
In order to situate the species of DH, we need a definition of DH as a genus. Such a

definition can be simply laid out as follows: Every doctrine of hell holds that some
created persons, while preserved forever in being, will never be saved.9
This general account needs clarification. First, we need some notion of what
the Christian faith means by “salvation.” That salvation involves perfect and
9
The clause, “while preserved forever in being,” is introduced into this general
account of DH to exclude the doctrine of annihilation (DA) from its scope. DA holds that
some persons will be utterly and irretrievably destroyed—and hence, a fortiori, will never
be saved. Both DH and DA may be viewed as versions of the broader doctrine of limited
salvation (DLS).
unending bliss is generally agreed, but to define salvation as eternal bliss is far
too superficial. In light of both Scripture and tradition, it should be clear that
the primary element in salvation is the clear or direct ongoing experience of the
divine essence—in other words, the beatific vision.10 Bliss is one consequence of
experiencing the beatific vision, but not the only one. Moral sanctification—the
purging of all sinful dispositions—is at least as significant.11
This final point deserves some attention. For Christianity, God is taken to be
the highest good. This is a claim about what is objectively valuable, and as such
imposes an obligation on our subjective values: if we fail to value God above
all things, our subjective values are defective—and our moral character is more
broadly compromised.12 When we fail to order our values appropriately, we
inevitably fail to behave in ways that display appropriate respect for the inherent
worth of things. Rational creatures are thus fully moral only when they make God
the object of their highest devotion—in other words, only when they love God
above all else. Loving God above all else completes the moral nature of rational
creatures. By clinging to God—the sovereign good—and loving Him because of
His perfection, creatures are perfected in both intellect and will, so that all their
inclinations and actions are in accord with right reason.
But this kind of rightly ordered love of God can only be fully achieved by a
direct vision of the divine essence. Thus, Melanchthon asserts that
although the law points out what God is like, such righteousness cannot be in
anyone unless God himself dwells in him and gives him his light and glory. Thus
the law is entirely fulfilled in us only in eternal life, in eternal righteousness,
when we have eternal joy in God, and God has become all in all.13
In sum, salvation in the Christian tradition means a spiritual union with God that
brings about both eternal moral perfection and perfect happiness. With this notion
of salvation in mind, it should be clear what it means to say that some created
persons are never saved. It means that some are never granted the beatific vision,
which is the only thing that will fully complete them as rational creatures. Their
intellects are thus eternally darkened by false notions of what is true and good,
leading to disordered desires that overvalue some things and undervalue others.
10
See, for example, Aquinas, Summa theologica, trans. Fathers of the English
Dominican Province (New York, 1946), I-II, q. 3, art. 8.
11
On this point see St. Augustine, The City of God, trans. John Healey, ed. R.V.G.
Tasker (Dutton, 1972), Book 22, chapter 30; Schmid, The Doctrinal Theology, p. 661,
especially the excerpt from Hollaz; Heppe, Reformed Dogmatics, p. 707.
12
Cf. Augustine, City of God, Book 14, chapters 3–6; Philip Melanchthon, Loci
communes¸ 1555, Locus VI, translated as On Christian Doctrine¸ Clyde Manschreck
(Grand Rapids, 1965), pp. 75–7.
13
Melanchthon, Loci communes, Locus VI, p. 76.
Species of Hell 205
Not only will they never love God with their whole hearts, they will never properly
love their neighbors or themselves.
The tradition has referred to this disorder in the desires as “concupiscence,” as
a selfishness that seeks above all else the gratification of carnal desires.14 But, in a
sense, this selfishness manifests a lack of self-love. As Aristotle would put it, those
who are good love what is best for themselves, namely virtue, while the wicked
reject virtue (which is to their soul what health is to their body), preferring instead
such lower goods as fame, wealth, power, sensual pleasure, self-righteousness,
etc. The wicked thus eschew, in favor of lower goods, the very thing their souls
need in order to be healthy.
All species of DH agree, then, that the essence of hell is the eternal privation
of the beatific vision and the consequent sickness of the soul. But some species
of DH hold that hell is also characterized by further ills of body and/or soul.
Among the ills of the soul that have been historically listed are such things as
never-ending pangs of conscience, hatred of God (all the more intense because the
damned know that such hatred is unmerited), continual frustration at the inability
to satisfy perverse desires, etc.15 Among the ills of the body (also called external
pains) have been listed torment at the hands of other fallen creatures (particularly
devils), never-ending burning produced by an infernal fire, etc.16 But, though some
versions of DH list these as punishments which accompany damnation, not all do,
and the core of DH in all its forms can thus be taken to consist in the everlasting
deprivation of the beatific vision and the unending moral and spiritual vitiation
which necessarily attends it.
The Nature of the Sufferings of Hell
Species of DH can be distinguished according to at least two parameters: the

nature of the sufferings of the damned (NH), and the cause of damnation (CH).
We begin with NH. Here, there are two general possibilities:
14
See, for example, Augustine, City of God, Book 14, chapters 2–9; Aquinas, Summa
theologica, I-II, q. 82, art. 3; Melanchthon, Loci communes, Locus VI, pp. 75–6. It should
be noted that when the Christian tradition speaks of “carnal desires” it does not mean to
solely refer to the excessive love of sensual pleasures, but rather to any excessive devotion
to any created thing, the most subtle of which appears in the devotion of the Pharisee to his
own self-righteousness. For particularly incisive and beautiful statements of this point see
Hans Martensen’s classic Christian Ethics, trans. C. Spence (Edinburgh, 1871), pp. 102–8,
and Luther’s celebrated commentary on the first commandment in his Large Catechism.
15
See Aquinas, Summa theologica, Suppl. q. 98; and the passage from Gerhard’s Loci
theologicci in Schmid, The Doctrinal Theology, pp. 634–5.
16
See Augustine, City of God, Book 21, chapter 9; Aquinas, Summa theologica, Suppl.
q. 97; the catena of quotes from the older Lutheran divines in Schmid, The Doctrinal Theology,
pp. 658–9; and the older Reformed divines in Heppe, Reformed Dogmatics, pp. 710–11.
NH1 The evils suffered by the damned include not only the privation of
the beatific vision and whatever positive evils result necessarily from
that privation, but also certain other positive evils that do not so result
(hereafter ancillary evils).
NH2 The evils suffered by the damned do not include any evils other than
the privation of the beatific vision and whatever positive evils result
necessarily from that privation.
NH1 holds that the damned suffer not only the deprivation of the beatific vision,
but also certain further evils that are not a necessary result of such deprivation.
In order to understand this doctrine better, we need to remember that Christians
have historically believed that the life and welfare of any creature is constituted
by God’s presence to that creature. Since the time of St. Augustine, the biblical
notion that death is a punishment for sin has been interpreted to mean that sinners,
by turning from God, cut off the “spiritual light” they need in order to flourish.17 In
being deprived of this light, their intellects are darkened, not only in the sense that
they do not know certain things they ought to, but in the positive sense of falling
into error concerning what is true and right. Furthermore, as already noted, in
being deprived of the light of God, they come to love some things more than they
ought (especially themselves and carnal goods), and other things less than they
ought (especially God and their neighbors).18
But the positive evils humans suffer from as a result of losing the light of God
may not be limited to evils of the soul. According to some thinkers, they also
include evils of the body. The idea is that since the soul itself is weakened by being
deprived of the light of God, it experiences a diminution of all of its vital powers.
There results from this all sorts of illness and, finally, physical death.19
Whatever one takes to be the positive evils that result necessarily from privation
of the beatific vision, the more important question is whether the sufferings of hell
are limited to such evils or also include ancillary evils. It is how one answers this
question that establishes a more important distinction among species of DH, and
hence serves as the basis for our distinction between NH1 and NH2. As such, it is
important to think about which evils are really ancillary to the loss of the beatific
vision, and which are really essential. To make this determination, we think it
17
See Augustine, City of God, Book 13, chapters 13–15, and Schmid, The Doctrinal
Theology, pp. 237–8 (esp. the passage from Hollaz).
18
On this matter see Isaak Dorner’s magisterial discussion of sin, supported by
numerous biblical references, in volume 2, Part II of his great A System of Christian
Doctrine, trans. A. Cave and J. S. Banks (Edinburgh, 1885), no. 73, pp. 313–24. As with
all the greatest Christian divines, Dorner emphasizes that the worst of moral evils is
“spiritual arrogance, the conceit of self-righteousness,” which shuts the self off from loving
communion with God and neighbor. Cf. Augustine, City of God, Book 14, chapter 3.
19
See Augustine, City of God, Book 13, chapter 15; Schmid, The Doctrinal Theology,
p. 238.
Species of Hell 207
helpful to make a further distinction between what we will call “objective” and
“experiential” evils.
By an objective evil, we mean an evil whose reality does not depend on the
awareness of the subject afflicted by it. Cancer that has yet to be discovered
or to manifest overt symptoms is nevertheless an affliction of the body, and is
an objective evil in our sense. A sinful disposition remains a part of a person’s
character, and hence is an objective evil, even while the person is asleep and is not
actively experiencing the urgings of that disposition or its implications. It surely
remains an evil even though the person is ignorant of its intrinsic wickedness and
hence lacks the experience of being afflicted by sin.
With most objective evils of this sort, it is possible for the subject to become
conscious of these evils as evils: to experience them actively for what they are.
When this happens, the conscious experience of the objective evils becomes an
additional evil that afflicts the subject, insofar as awareness of one’s own sinful
nature or knowledge of a deadly cancer generates an intrinsically undesirable
conscious state. This is what we have in mind when we speak of experiential
evils. They are, in the simplest terms, pains. Broadly speaking, experiential evil is
equivalent to actual conscious suffering. Physical pain is an immediate experience
of an objective evil of the body, although it is not necessarily mediated by a
conscious awareness of the nature of the objective evil and its significance. Such
awareness, if the cause of the physical pain is a degenerative illness, might well
generate a further psychological pain to supplement the merely physical. But both
pains are experiential evils. Morphine might eliminate, for a time, the physical
pain, but not the underlying objective evil that is its cause. Unconsciousness might
remove the psychological distress, but not the disease.
The point of drawing this distinction is simply this: while someone might well
hold that privation of the beatific vision necessarily generates objective evils, it
is harder to maintain that it necessarily generates experiential evils. Conscious
awareness does not seem to be a necessary concomitant of being denied the
beatific vision. Hence, while alienation from God may well give rise to objective
evils that afflict the soul and even the body, it does not necessarily give rise to
suffering. The kinds of conscious suffering that are generated by an awareness of
being deprived of the beatific vision, or by an awareness of the various positive
evils that immediately result from being so deprived, are ancillary. They are not
necessary concomitants of being denied the beatific vision, for the simple reason
that those denied the beatific vision need not be conscious at all.
What this means is that those who ascribe to NH2, and who therefore deny that
the damned are subjected to ancillary evils, must thereby deny that the damned
are afflicted by any experiential evils, that is, by any evils that are constituted at
least in part by conscious awareness. They must hold, in other words, either that
the damned are plunged into a state of everlasting unconsciousness or, if they are
conscious, that none of their conscious states will in any way add to the evils that
they suffer from (e.g. they will have no awareness of being separated from God, if
such awareness causes them pain; they will not be tormented by grief for their past
sins, at least if such torment is properly conceived as an evil in light of the fact that
it cannot lead to repentance).
Adherents to NH1, on the other hand, hold that the damned do suffer ancillary
evils—most significantly, the experiential evils that we call pain or suffering. But
the pains in question here need not go beyond what immediately accompanies
a conscious awareness of the privation of the beatific vision and its attendant
objective evils. That is, the ancillary evils affirmed in NH1 might well be limited
to the evils that result from a conscious awareness of those evils that are not
ancillary. In other words, the ancillary evils affirmed in NH1 need not be taken to
include pains inflicted by God for retributive reasons. But, then again, they might.
And these alternatives are significant. Hence, we need to draw a further distinction
between two different sub-species of NH1, as follows:
NH1a The evils suffered by the damned include not only the privation of the
beatific vision and whatever positive evils necessarily result from that,
but also certain ancillary evils that are inflicted by God as retribution
for sin.
NH1b The evils suffered by the damned include not only the privation of the
beatific vision and whatever positive evils necessarily result from that,
but also certain ancillary evils that, while willed or permitted by God,
are not inflicted as retribution for sin.
NH1a holds that the ancillary evils of damnation have their ultimate cause in
the retributive justice of God. According to this species of NH1, God wills that
the damned suffer certain ancillary evils because it is intrinsically fitting that the
unregenerate should suffer in such a way.
NH1b, on the other hand, does not commit itself to any particular motive God
might have for willing or allowing the damned to suffer from ancillary evils;
it simply denies that He positively wills such pains for retributive reasons. We
suspect that in most cases, allegiance to NH1b would entail the view that the
ancillary evils suffered by the damned are limited to the suffering that comes from
consciously experiencing those evils that are not ancillary. This is because, once
retributive motives are taken off the table, it is hard to imagine what else might
motivate God to allow or inflict evils beyond this. Conscious awareness of the
objective evils one is subject to might be justified, however, on the grounds that
awareness of the truth is better in itself than the lack of such awareness.
The Causes of Damnation
To fully characterize a species of hell, it is not enough to identify the nature of

the suffering endured by the damned. It is also necessary to offer an account of
why they suffer damnation at all. Put in simple terms, one needs an account of the
causes of damnation (CH). But when we refer to the causes of damnation, we are
Species of Hell 209
not concerned with describing mechanistic forces or laws of the sort that scientists
might offer as causal explanations of empirical events. Rather, we are interested
in agent-oriented causes. It is generally agreed in the Christian tradition that either
the damned suffer their fate because of what they choose (and God merely permits
them to endure the consequences of their choices), or the damned suffer their
fate because of what God chooses (perhaps in response to choices made by the
damned). It is also possible, of course, that the sufferings of hell result from some
combination of the two.
Focus on such agent-based causes makes sense if our interest is in identifying
the God-justifying reasons for damnation. Such God-justifying reasons will take
one of two forms: (a) they will give an account of why God cannot save the
damned; or (b) they will give an account of why God decides not to save the
damned even though it is in His power to do so (that is, why God either wills
or permits their damnation). Both of these alternatives, as we shall see, must
presuppose that damnation hinges up the choices of agents: either the damned are
responsible for their fate through their own choices, or God is responsible for their
fate through His choices, or some combination of the two.
That this is so is clearest in the case of (b), in which case the free choices of
God will have to be viewed as playing the decisive role. But a little reflection will
show that, in the case of (a), we must assume that the decisive impelling cause of
damnation lies in the free choices of the damned. Since Christianity affirms that
some persons are saved and that the damned are not different in their fundamental
nature from the saved, the view that God cannot save them will not be explicable
by reference to a difference in their nature. And so where else could the relevant
difference lie, except in the will of the damned? In other words, if God cannot
save the damned, it will be because the damned make choices which somehow
take them out of the class of those God is capable of saving. The most obvious
candidate for such a choice is the decision to reject God, in the sense of refusing to
willingly experience the beatific vision. But given God’s presumed omnipotence,
to say that God cannot save those who make such a choice is coherent only if the
experience of the beatific vision is taken to involve the free participation of the
creature in some essential way. Perhaps “the beatific vision” describes a relational
state involving God and the creature, and perhaps the creature’s contribution to
the relation can only exist through an exercise of libertarian freedom which God
cannot bring about even in Plantinga’s weak sense. If so, then God would be
incapable of saving those who willfully reject Him. This, then, would serve as His
God-justifying reason for not acting on His prima facie motive to save all.
Notice also, however, that even in the case of (a) there is something God actively
does that contributes to the condition of the damned: He preserves them in being. On
orthodox Christian assumptions, everything that exists does so by virtue of being
sustained in existence by God. If God ceases at any moment to sustain something in
existence, it ceases to exist.20 And so, even in the case of (a), God could presumably
bring about the non-existence of the damned (and thereby “spare” them the evils of
hell) even if it is not in God’s power to save the damned in the sense of bringing them
into communion with the divine. And so God still contributes actively to the state of
damnation, and there will be an impelling cause of God’s decision to preserve the
damned in being. Most likely, this cause would be that, given their essential nature
as creatures of God, the being of the damned retains positive value, and God has
a disposition to preserve whatever has positive value (even if that means that the
damned will endure subjective suffering as a result).
In case (b), the assumption is that God can save the damned but chooses not to.
But since God has a prima facie intellectual motive to save all, the damnation of
some must be explained by the presence of a higher motive which conflicts with
this prima facie saving motive, and thus defeats it. This higher motive would have
to be intellectual (not merely affective) in keeping with God’s moral character.
In the simpler language of duties, it would have to be an overriding duty which
defeats God’s prima facie reasons for saving all. Put simply, option (b) implies
that, in order to save all, God would have to do something that it is impermissible
for God to do.
This moral impediment to saving all could manifest either at the level of means
or at the level of ends. In the first case, God would not have any reasons not to
pursue the salvation of each created person, but He would find that sometimes the
actions He would need to perform in order to save a creature would be morally
impermissible ones. In this case, the moral impediment would manifest at the level
of means. Alternatively, however, God might have duties which preclude Him
from even pursuing the salvation of some creatures. That is, it would be wrong,
all things considered, to seek their salvation (despite His prima facie intellectual
reason to do so), even were there morally permissible means of doing so.
Consider the following analogy. Suppose I have good prima facie reasons to
financially support my son through college, that is, to bring it about that my son
gets to go to college without having to work full time in order to pay for it himself.
But suppose the only way I could do this would be by stealing money. In this case,
there would be a moral impediment at the level of means. But suppose that my son
has proven to be irresponsible with the use of opportunities given him by others,
and displays responsibility only when the opportunity is made available through
his own efforts. In that case, I may have compelling moral reasons not to bring
it about that my son gets to go to college without having to work for it. Were my
son’s character different in important ways, my prima facie reasons for bringing
about this state might be decisive (he would be able to focus more fully on his
studies and pursue more extracurricular educational opportunities). But, given his
character, it would be wrong for me to pursue this end even if I could do so without
20
See Aquinas, Summa theologica, 1, q. 104, art. 1; Melanchthon, Loci communes,
Locus III, p. 41.
Species of Hell 211
stealing money (or doing something else impermissible). In this case, we have a
moral impediment at the level of ends.
If God were morally barred from saving the damned at the level of means,
the most likely reason would be because, with respect to at least some of the
damned, saving them would require action that failed to show proper respect or
love for them. But, clearly, saving them would contribute to their welfare, and
so saving them would never display a lack of concern for their welfare. So in
what would the lack of respect lie? The common answer is that it would display
a lack of respect for freedom or autonomy. In other words, were God barred from
saving the damned at the level of means, it would be because He could only save
them by violating their free choices or autonomy in morally impermissible ways.
While God could save the damned, and while God could respect the autonomy of
the damned, He could not do both—and respect for the latter is a more pressing
obligation than respect for the former.
In other words, if God is morally barred from saving the damned at the level of
means, we find ourselves with a theory about the cause of damnation very similar
to what we encounter in the case in which God is taken to be unable to save some:
In both cases, the free rejection of God by the damned takes them out of the pool
of creatures that God can legitimately save. Either He cannot save them at all, or
He cannot save them in a manner that is morally legitimate. In either case, it makes
sense to say that it is the damned’s free choices that put them in hell—and God is
simply left with no legitimate choice but to let them suffer.
In the case in which God fails to save the damned because He no longer wills
their salvation as an end, the situation is different. In this case, if God willed the
salvation of the damned, He would have available to Him a means of saving them
that, in itself, was not morally objectionable. What most directly keeps God from
saving the damned under this view, then, is His own rejection of them, in the sense
that He simply does not will (all things considered) that they be saved. While their
rejection of Him may be what motivates God to reject them—while their rejection
of Him may constitute either a morally good or morally compelling reason for
Him to reject them—God’s rejection of the unregenerate is the more immediate
cause of damnation.
It is one thing to say that, because the damned reject God, God’s hands are
tied and He cannot save them even though He still desires their salvation. In such
a case, God is not willing their damnation. He certainly is not inflicting their
damnation on them. They are damned, we might say, by what they will, not by
what God wills—even if God wills His own obedience to certain moral principles
which so happen to preclude Him from performing acts that might have saved
these damned souls. In the latter case, however, God’s will intervenes in a more
direct way between what the damned will and their state of damnation, even if (as
may be the case) what motivates God to will that the damned not be saved is the
fact that they freely rejected God.
But it should also be noted that something other than the sinner’s rejection of
God may be what motivates God to reject the damned. Consider the Calvinist view
that God rejects some of those determined by their fallen natures to reject Him,
and saves others, so as to manifest the glory of His justice and mercy.21
For all these reasons, we think it is most helpful to distinguish the causes of
hell according to the following alternatives:
CH1 The damned freely reject God, and although God never stops willing
their salvation as an end, He cannot save them either because (a) it is
logically impossible for Him to do so or (b) all of the logically possible
means of securing their salvation are morally impermissible.
CH2 If God willed the salvation of the damned as an end, it would be logically
possible for Him to achieve this end through means that were morally
permissible; but He does not will their salvation as an end, either for
(a) morally good but not compelling reasons or (b) morally compelling
reasons.
Several observations should be made about this division. First, there are really
four alternative accounts of the causes of damnation laid out here. However, the
first two accounts and the last two are so closely connected that they can and will
be typically dealt with together. As such, we find the present division particularly
suitable.
Second, it is possible to imagine a hybrid theory of DH according to which both
the end of saving the damned and the available means were morally objectionable.
To our knowledge no one has embraced such a view, but more importantly, the
existence of such a hybrid view does not much influence the critical assessment of
DH. If neither CH1 nor CH2 is defensible, then a hybrid will not be either. And if
one of these alternatives is defensible, then a hybrid will be defensible a fortiori.
For the purposes of philosophical assessment, then, this hybrid is superfluous.
Finally, we should say a few words about the distinction between CH2a and
CH2b. While both stress that God does not will the salvation of the damned as
an end—and it is this fact, rather than any inability to achieve that end, that stays
God’s hand—the nature of God’s motive in each case is different. In the case of
CH2b, the motive is a morally compelling one: if God saved the damned, He would
be achieving an immoral end. Put another way, His moral character compels Him
to reject the sinner despite the suffering that thereby results. In CH2a, however,
the causes for God’s refusing to grant salvation to the damned must be conceived
in terms of motives upon which He might act but need not. Under this view, then,
God would be doing nothing strictly wrong or impermissible if He did, in fact,
save all. Universal salvation would be perfectly compatible with the demands of
justice or any other moral standards employed to judge the fittingness of that end.
But even so, God does not will this end. While God does have reasons for rejecting
21
See Heppe, Reformed Dogmatics, pp. 165, 187.
Species of Hell 213
the damned, these reasons do not compel Him to act, and He would be perfectly
just and moral were He to save all.
This view, while it strikes us as very puzzling, seems to be the view of Aquinas,
which is the main reason we include it here. Put in terms of God’s intellectual
motives, we might articulate CH2a in the following terms: while God has a prima
facie reason to save all, He also possesses God-justifying reasons against saving
all. But these latter reasons are neither more compelling nor less compelling than
His reasons for saving all. So God is free to act on them or not. In effect, the
totality of motives consistent with God’s moral character is such that they neither
necessitate nor preclude saving all. And God just so happens to decide to act on
those reasons which dictate against saving all.
The most common reason why, historically, it has been thought that God would
cease to will the salvation of the damned as an end is this: the damned deserve the
evils of hell as a matter of retributive justice. On this view, God does not merely
have a prima facie motive to see that the demands of retributive justice are met,
but His motive to this effect is as pressing (CH2a) or more pressing (CH2b) in the
hierarchy of divine motives than whatever prima facie reasons God has for willing
the salvation of His creatures. So, while God would will the salvation of the
damned were it not for the fact that they do not deserve to be saved (but do deserve
to suffer the punishments of hell); given what they deserve He ceases to will their
salvation—either because it would conflict with His perfect moral character to do
otherwise (CH2b), or because it is consistent with His moral character to go either
way, and He just so happens to choose to cease willing their salvation (CH2a).
Complete Species of DH
We now need to say something about the ways in which the species of NH are
logically related to the species of CH. Any complete theological defense of DH
must spell out both the nature of the sufferings of the damned and the causes of it,
and so must combine some species of NH with some species of CH.22 Hence, any
complete account of DH would involve the conjunction of both these elements.
First, NH2 can be combined with any form of CH to compose a complete
doctrine of hell. Even if God is morally required to reject the damned (CH2b) on
22
One might hold that this does not apply to a purely scriptural defense of DH. For
it might seem that, based on Scripture, one might simply hold that some will forever suffer
eternal torment without spelling out the precise nature of that torment or even giving a
complete account of the causes of damnation. However, it seems that anyone who bases
their belief in DH on Scripture must hold that the forma of damnation at least includes
separation from God and that its causes must have something to do with creaturely sin and
divine justice. No Christian theologian who has treated these matters has held otherwise.
For this reason alone it seems that even a purely scriptural doctrine of hell will involve
embracing, at a minimum, NH1, even if it prescinds from embracing any form of CH.
account of justice, justice might demand nothing more than the infliction of those
evils which naturally flow from being deprived of the beatific vision. In fact, one
might hold that God’s willingness to forever deprive a rational creature of the
vision of Himself, on account of the creature’s sin, is a very harsh punishment
indeed. But if NH2 is compatible with CH2b, then it is a fortiori compatible with
CH2a (according to which there are morally good but not compelling reasons why
God does not save the damned). This is so because, if God is not morally required
to punish the damned at all, it is hardly plausible to think the punishments cannot
be merely privative. But then it also seems that NH2 is compatible with CH1,
which holds that, although God may wish to save the damned, He cannot do so for
logical or moral reasons.
NH1a and b are also compatible with every form of CH. It might at first appear,
however, as if NH1a is not compatible with CH1. If God wills the salvation of
the damned but is barred from realizing this end by the creature’s rejection of
Him, it may seem that He would not then will on the damned evils which are
not a necessary result of their rejection of Him (especially if we suppose that the
reason God continues to will the salvation of the damned is because He continues
to love them despite their rejection of Him). Nevertheless, some have held that,
though God does continually desire to save the damned, He nevertheless continues
to inflict certain pains on them out of retributive justice, pains which are not a
necessary result of their alienation from Him.23 In short, while He desires their
salvation and would save them if He could do so in a morally permissible way,
He also desires that retributive justice be done—and so, insofar as the damned are
deserving of additional pains beyond the merely privative, He inflicts them.
If NH1a can be combined with CH1, then it seems it can be combined with
CH2. After all, if it is possible for God, even though He loves the damned and
desires their salvation, to will on them suffering that does them no good, then
it seems a fortiori conceivable that He could will such suffering were He to no
longer will their salvation. If God, on account of the wickedness of the damned,
no longer desires for them the only thing that will make them happy and morally
good, it is easy to suppose He might also be willing to inflict further pains on them
as punishment for sin.
Finally, if NH1a can be combined with either CH1 or CH2 it seems clear that
NH1b can. First, if NH2 could be held to be consistent with CH2, it also seems
that NH1b could be for similar reasons. NH1b also seems to be compatible with
CH1. One might hold, for instance, that the only way to shield the damned from
23
Isaak Dorner seems to have taught that this is at least a possible state for some
created beings, since he held, first, that human freedom makes it impossible even for God
to ensure universal salvation (A System of Christian Doctrine, trans. Alfred Cave and J.
S. Banks (4 vols, Edinbugh, 1880), vol. 4, Part II, no. 154, pp. 420–28); second, that the
wicked deserve to suffer what we have called experiential evils (vol. 3, Part II, no. 88, pp.
120–32); and third, that it would not be fitting of God to apply the merits of Christ to those
who have no living faith in Christ (vol. 4, Part II, no. 132, pp. 215–17).
Species of Hell 215
any actual pain would be for God to actively cast them into a state of perpetual
sleep, or to cause them to be forever deluded about their unhappy state. But these
things might be thought to be inconsistent with either the love or respect God has
for them as rational beings. So it seems that someone who holds to CH1 could very
easily endorse NH1b, even if she denies that God inflicts any pains on the damned
out of retributive justice.
In sum, six complete versions of DH emerge from the foregoing analysis:
(DH1) The conjunction of NH1a and CH2;

(DH2) The conjunction of NH1b and CH2;
(DH3) The conjunction of NH2 and CH2;
(DH4) The conjunction of NH1a and CH1;
(DH5) The conjunction of NH1b and CH1;
(DH6) The conjunction of NH2 and CH1.
DH1 is the doctrine of hell most commonly embraced in the popular imagination:
God not only casts the unregenerate from His presence, but also heaps torments
upon them for all eternity. DH1, DH2, and DH3 together constitute variants of
what can be called the “classical” doctrine of hell—that is, the version of DH most
commonly expressed in medieval Catholic and early Protestant theology. What all
variants of the classical doctrine have in common is their insistence on CH2—that
is, on the view that God does not will the salvation of the damned as an end, such
that even if there were a morally permissible means of saving them, God would
continue to damn them.
DH5 and DH6 have emerged as the most popular contemporary expressions
of DH among more learned theologians and philosophers, and might be dubbed
variants of the “liberal” doctrine of hell. What each has in common is a rejection of
the idea that retributive motives play any role in God’s failure to save the damned.
What the damned suffer, they suffer because their own free choices bring about
consequences that God cannot negate, or cannot negate without doing something
morally wrong.
DH4, which might be called the “retributive” doctrine of hell, amounts to a
via media between the classical and the liberal versions. On this view, the damned
do not suffer their fate because God has ceased to will their salvation. He still
does so but, given their rejection of Him, has no (legitimate) means of saving
them. However, on this view the unregenerate deserve to suffer not merely the
privative pains that accompany being denied the beatific vision, but also certain
additional pains. Were God able to save them, they would cease to be unregenerate
and would no longer deserve these afflictions. But, insofar as God cannot save
them, they continue to be unregenerate in ways that invite a retributive response.
And so, while perhaps anguished by His inability to save them, God nevertheless
heaps upon them additional retributive afflictions.
Challenges Faced by the Various Versions of DH
While the most significant problems for any species of DH arise in relation to the
causes of hell, there are some interesting concerns that arise when we reflect on
the nature of the suffering endured by the damned. These are worth reflecting on
before turning to the deeper difficulties.
We start with those species of DH that endorse NH2—in other words, DH3 and
DH6, both of which deny that the damned endure any ancillary evils, that is, any evils
beyond those that necessarily accompany being deprived of the beatific vision.
As we already noted, those deprived of the beatific vision need not be
conscious—and so the evils of damnation that require consciousness are all
ancillary. It follows that DH3 and DH6 must deny that the damned have any
conscious experience of the evils they are subject to, and so must deny that the
damned suffer.
If this is right, we might well ask whether DH3 and DH6 are really species
of DH at all, insofar as suffering is typically understood as an integral part of
damnation. Perhaps DH3 and DH6 describe a kind of via media between the
doctrine of hell as ordinarily conceived and some doctrine of annihilation. But
even if we set this worry aside, we still confront additional worries.
First, since consciousness seems to be essential to our nature as rational beings
(we cannot exercise any of our intellectual or spiritual powers in its absence),
perpetual unconsciousness would be a decisive vitiation of that nature. And so
it seems as if the loss of consciousness needed to prevent active suffering would
itself be an evil, an objective one. But it does not seem to be the kind of evil that
would result necessarily from being deprived of the beatific vision. Hence, it is an
objective evil that God would presumably have to inflict on the damned, perhaps
out of a motive of mercy. If this is right, then God might be seen as having to
choose among ancillary evils: He must either permit suffering that is not strictly
necessary, or inflict oblivion.
If this is right, then no species of DH that included NH2 would be possible
at all, since whatever God did, the damned would endure one kind of ancillary
evil or another. The only way to avoid this outcome, it seems, would be for the
defender of DH3 or DH6 to make the case that perpetual unconsciousness would
be a necessary consequence of being cut off from the beatific vision, and hence
would not be an ancillary evil after all.
Whatever we think of this issue, there is another that is at least as important.
As we have already seen, being deprived of the beatific vision entails being cut off
from a clear understanding of the objective order of values, resulting in disordered
desires that overvalue some things and undervalue others. Put more simply,
damnation confirms the damned in moral degeneracy for eternity. And this will
be true even if they are put into a state of perpetual unconsciousness. They may
be sleeping villains, but they remain, in character and disposition, villainous. In
short, even if suffering is not a necessary consequence of losing the beatific vision,
perpetual moral degeneracy is.
Species of Hell 217
And if it is true given NH2 that the damned are forever mired in wickedness,
it is true a fortiori for all species of DH. And so, this fact highlights a global
feature of DH: for every species of DH, one of the evils that necessarily
accompanies damnation is eternal moral degeneracy. What is troubling here is
that this degeneracy is not the rationale for damnation—that is, it is not some
fact about the creature which justifies God in either permitting or bringing about
their damnation. Rather, it is an effect of being deprived of the beatific vision.
God either cannot or will not provide what amounts to a necessary condition for
overcoming wickedness.
In short, any version of DH seems to commit one to the view that God is
either defeated by sin or complicit in its perpetuation. Defenders of DH need to
face these worries head-on if they want to reconcile the doctrine of hell with the
Christian idea of a sovereign God who loves all creatures and hates sin.
This line of thinking is most readily explicated in connection with difficulties that
arise from alternative proposed causes for damnation—an issue we turn to now.
Proponents of CH1 and CH2 face different sorts of challenges. Those who
defend a version of DH that falls under CH2 must maintain that, despite God’s
love for every sinner, there are moral considerations that would inspire God not
to will the salvation of at least some of His creatures. But insofar as love for
creatures involves, at least prima facie, willing what is best for them, it is hard to
see what moral considerations could motivate a God whose fundamental moral
characteristic is love to consign some sinners to eternal misery. Defenders of CH2
therefore face the challenge of offering an account of such considerations.
The magnitude of this challenge, it seems to us, is routinely overlooked by
defenders of CH2. But the challenge is clearly enormous. Defenders of CH2
must overcome the prima facie absurdity of the following conjunction: God loves
every one of His creatures with a profound and unwavering benevolence and He
wills upon some of these creatures whom He loves the very worst kind of evil
conceivable; and He wills that they suffer it for all eternity, even though it cannot
possibly do them any good at all, since it never culminates in anything other than
more of the same.
Defenders of a version of DH falling under CH1 face different challenges.
Because they hold that damnation originates in the creature’s own free rejection
of God, they must accept two claims: first, that some creatures freely reject God
forever; and second, that God cannot—either logically or morally—overcome
this free rejection of Him (despite an infinite time frame in which to work on
overcoming their intransigence).
That someone created in the divine image, and hence naturally ordered
towards the good, should eternally reject God, the perfect good, strikes us as prima
facie unlikely, especially if God has not given up on the creature and continues
unremittingly to seek the creature’s repentance. Furthermore, that an omnipotent
and omniscient God should eternally fail to find a morally legitimate way to
transform an unwilling creature’s heart strikes us as prima facie dubious.
But the problem runs deeper, as can been seen when we focus on the fact
that a necessary feature of every species of DH is that the damned are eternally
confirmed in moral degeneracy. This is troubling in its own right, but it becomes
even more troubling when combined with CH1. CH1 entails that, in God’s war
against sin, in the souls of the damned God confronts ultimate defeat. Despite all
of His infinite resources, despite infinite time in which to work, despite His perfect
knowledge of every nuance of the souls of the damned, despite His unrelenting
love, His efforts will be for naught. At least in some human souls, sin will prove
more powerful than God.
This seems an unavoidable implication of any of the “liberal” versions of DH,
and it is an implication that should give every defender of DH pause. In fact, it
strikes us as verging on blasphemy.
And this may be a main reason why the classical doctrine of hell has not entirely
gone away despite its drawbacks. If God’s salvific aims simply do not include the
damned, then we are not driven to the unsettling conclusion that God’s aims are, in
some human souls, ultimately defeated. In the various forms of the classical version
of DH, the eternal alienation of the damned is directly intended by God, and so
cannot be viewed as God’s failure or defeat. On this view, God prevails over sin in
different ways: in the saved, through their attainment of blessedness, which includes
sanctification; in the damned, through punitive expulsion from the goods of heaven.
But this way of thinking obscures deep problems that, once again, become
evident when we recall that being confirmed in wickedness is a necessary
consequence (perhaps the only necessary one) of being deprived of the beatific
vision. On the classical doctrine of hell, the damned are punished for their
wickedness at least in part by being confirmed in wickedness for eternity.
To see the full magnitude of the difficulty here, it may help to reflect for a
moment on exactly what is so bad about sin. Sin at its heart is a failure to value
things according to their objective degree of value. It is a failure to appropriately
express, in actions and dispositions, due reverence for the inherent worth of things.
The most significant element of sin, on classical theology, is the failure to do this
with respect to God. God has infinite inherent worth, and thus ought to be valued
above all things. To fail to do so is an objective affront to the divine majesty, akin
to the sociopath’s failure to properly value his victim but magnified in severity by
the infinite worthiness of God.
According to the classical doctrine of hell, God responds to this infinite affront
against His dignity by deliberately acting to ensure that this affront to His dignity
continues for all eternity. While He could stop it from continuing, He chooses
instead to make sure that this most intolerable of all evils persists forever in the
souls of the damned by deliberately withholding the necessary condition for
bringing it to an end.
And so the defender of any form of the classical version of DH must explain
why it would be a demand of justice to bring it about that a criminal never stop
committing his crime. We, at least, cannot conceive of any coherent conception of
justice under which this would make any sense at all.
Index
a fortiori 212, 214, 214, 217 suicide in 58

a priori 143 theories of 46; see also eternal life;
absence 183, 184, 185, 186, 197 heaven; hell
actions 9, 24, 30, 31, 36, 70, 71, 74, 92, agency 79, 81, 86, 152, 187
112, 113, 155–6, 158 agents 8, 9, 10, 13, 24, 118, 119, 176
in afterlife 58 autonomous 79–81, 89
complexity of 13 and causes of damnation 209
and desert 124 free choice of 25, 40, 209
determined 10, 11 freely rejecting God 30–2ff.
finite 126–8 and heaven 84, 85
future 142 intentions of 84
of God 201, 202, 210, 212, 213 motives of 202
infinite 128–9 own behaviour of 14
morally hazardous 187ff. reflective 81
motives for 201–2 unwilling 80; see also free agents;
non-free 31, 33 moral agents; rational agents
right 85, 93 agony 154
and risk-creation 190 alcoholics 12, 15, 16, 17
undetermined 9, 10, 11, 13 alienation 15, 18, 29, 30, 42, 207
unwilling 80 allegory 166, 168
voluntary 11, 80 ambiguity 18, 19, 22, 24
warranting infinite punishment 123, ambivalence 73, 75
126–30 analogies 127, 133, 135, 138, 139, 175,
wrong 93; see also divine action; free 177
action; good actions; self-forming ancillary evils 206, 207, 208, 216
action angels 152
Acts, Book of 98 anger 135, 180
actual-sequence theory 86 anguish 135, 139
actualization 48, 49, 123 animals 80
Adam 18, 19, 21, 51, 176 annhilation 2–3, 5, 46, 47, 121, 139, 152,
Adams, Marilyn McCord 1, 109, 201 153, 157, 159
Adams, Robert 15 and divine supremacy 61–3, 64
adults 142 and eternal punishment 54–5ff., 130,
affections 60, 62 132
affective motives 202, 203 mildness of 56–7
afterlife 1, 36, 45, 116, 155, 179 rates of 53; see also annihilationism
account of 138 annihilationism 101–2, 116, 117, 146, 147,
in Bible 117 148, 161, 216
experience of 140–41, 142 defences of 45, 46, 47ff., 53ff., 61ff.
free choice in 96, 97 and denial of hell 153–4
disputes about 46 badness 83, 115, 154, 156, 157, 184; see
and doctrine of hell 53–6ff., 64 also bad character; evil; moral bad;
and God’s moral perfection 54, 57–8, sin; wickedness
64 Barnard, Justin 3
and natural consequence motivations Batson, C. Daniel 195
47–52, 64 beatific vision 134, 142, 172, 204
objections to 49ff., 54–5ff., 64 privation of 204, 205, 206, 207, 214,
privation argument for 47–8, 53 215, 216
rational suicide argument for 57–61 consequences of 217, 218
and scripture 46 rejection of 209
antecedent will 201 beginning 149
anti-universalism thesis 67, 74 behaviour 11, 14, 15, 34, 40, 85, 201–2
Apostolic Fathers 98 being 130
appetites 81 and existence 48, 49, 209–10
Aquinas 47, 48, 49, 135, 151, 213 and goodness 47–9, 130
Argyle, Michael and Beit-Hallahmi, medieval concept 47, 48–9
Benjamin 195 modern concept 48
Aristotle 11, 205 Beit- Hllahmi, Benjamin see Argyle,
Arminian tradition 20, 22–3, 24 Michael and Beit-Hallahmi,
Arnobius 148 Benjamin
assumptions 11, 13, 17, 20, 21, 24, 36, 67, beliefs 51, 52, 60, 133, 145, 164, 170, 194,
78, 79, 82, 87n., 97, 115ff., 126ff., 197, 198
145, 147, 153, 159, 185, 200, 202, formation of 195–6; see also religious
210; see also false assumptions beliefs
assymetry 183, 185, 186, 197 beneficence 21, 22, 121
atheists 182, 194 Beneficent-Character Principle 121, 131
atheology 5, 146–8, 156–61 benefits 35, 121, 131, 183, 184, 185, 187,
atonement 19, 20, 26 188
limited 62 absence of 186
attitudes 14, 23, 124, 125, 195 Benatar, David 183–4, 185–6, 188, 197
infinite 125 Bible 29, 46, 69, 94, 96, 97, 101, 117,
vicious 124, 125 133, 134, 158, 166; see also
virtuous 124, 125 New Testament; Old Testament;
Augustine, St 149–50, 151, 176–7, 206 scripture
on divine grace 20–21, 22 birth 179–98
authority 54, 62, 127, 128 blessing of 183
autonomy 42, 79, 85, 89, 211 as moral duty 179, 181–3, 185, 190
awareness 31, 32, 40, 69, 75, 80, 81, 111, as moral good 179, 180–81, 187
135 as moral hazard 187, 188, 191, 192–3
of suffering 207–8 as moral right 189
awe 167 as moral wrong 182, 185, 186, 188–93,
195, 196, 198
bad character 16, 17 motives for 183, 184, 188
finite 124–5 and severe risk of eternal harm 195
infinite 115 blame 30, 31, 33, 40, 41
bad choices 12, 24 bliss 25, 26, 27, 176, 197, 198, 204
bad motives 124 body 99, 101, 149, 205, 206, 207
bondage 16, 17, 18
Index 221
boredom 135 to leave heaven 84, 85

brain-damaged people 24 to leave hell 86
Brandt, Richard 189 of rejecting God 29, 32ff., 173
Brown, Claire 2, 3 theories 100; see also free choices; free
Buckareff, Andrei 3, 4 will
Buddhists 194 Christ 7, 14, 29, 46, 63, 64, 94, 95, 97, 98,
burden of proof 46, 200 99, 136, 157, 169, 170, 174, 194,
burdens 187 197
belief in 163, 164
Cain, James 4–5 rejection of 109
Calvin 150 and unforgivable sin 100
Calvinism 62, 100, 105, 108, 212 Christianity 5, 15, 45, 47, 62, 98, 101, 157,
cancer 207 164, 176, 194, 209
capacities 50–52, 69 and escapism 77, 83, 84
Catholicism 22, 117, 215 and finality 88, 89
causal chain 156 and other religions 166–8, 171;
causes 10, 19, 31, 112, 207 see also Christians; doctrines;
agent-oriented 209 exclusivism, Christian
contributing 10, 11 Christians 5, 8, 12, 15, 29, 46, 50, 51, 92,
of damnation 200, 208–13ff. 93, 111, 185, 194, 196–7
problems of 216ff. and belief in hell 145
chance 9, 24, 100 and marriage 182
change 69, 73, 84–5, 89, 142, 177, 194 nominal 197; see also faith
character 124–6 circumstances 106, 107
flaws 13, 14, 17, 131 city of God 100
traits 135; see also moral character Clement of Alexandria 98–9
children 5, 8, 10, 17, 21, 22, 24, 26, 37, 63, cognitive abilities 51, 53, 69
80, 113, 127, 128, 135, 157, 175, comedy 65–6
177, 183, 191 communion with God 82, 85, 87
as moral agents 187 compassion 21, 22
in poverty 197 compatibilism 9, 72, 104
religious beliefs of 195–6, 198 and freedom 155
suffering of 179, 184–5, 188–9ff. compliance 175
wrongs against 192–3; see also birth; conception, act of 189; see also birth
parents concupiscence 205
choices 2, 3, 5, 7, 18, 23, 68, 74, 82, 101, conditional election 24
111, 112, 119, 120, 152, 157, 158, conditional proof 147, 156, 159
175 conditions 5, 10, 31; see also necessary
by agents 209 condition; sufficient condition
consequences of 32–4, 71 conscience 93, 205
by damned 209, 211, 212 consciousness 18, 19, 25, 45, 52, 53, 69,
determination of 104–5 135, 140, 141, 155, 207, 208, 216;
by God 113, 155, 209, 210, 211, 216 see also moral consciousness
God’s knowledge of 105–6 consequences 11, 12, 13, 24, 25, 31, 32–5,
of heaven 164, 165, 169, 173 38, 92, 93, 120, 204
of hell 79, 96, 97, 169 everlasting 54–5, 217
of non-existence 59 of freely rejecting God forever 32
number of 32
of loss of beatific vision 217, 218; see confirmed in moral degeneracy

also natural consequence model 216–17, 218
consequent will 201 God’s rejection of 211, 212, 213ff.
consequentialist override 122, 123 will of 209; see also sufferings of the
consistency 5, 88, 91, 146 damned
as grounding condition 134, 136 Dante 65, 135
consummation 77, 78, 83–9 Davis, Stephen 4
and heaven 83, 84–6 Day of Judgement see Last Judgement
and hell 77, 78, 83, 86–9 de facto 59
context 12, 17, 18, 24, 26 dead, the 98, 99
contingency 90, 111, 123, 142 death 2, 15, 19, 45, 46, 74, 98, 99, 100,
contraception 179, 181, 192 130, 141, 187, 206
contradictions 136, 137, 143, 148 deception 12
control 11, 12, 23, 59, 81, 104–5 decisions 71, 74, 104, 123, 142, 175
conversion 111 conditional 142
Corinthians 98, 99 by God 209
corruption 92, 143 deduction 148
argument 50, 53 defect 121, 128, 130, 131
Cottrell, Jack 23 degree 48, 49
counterexamples 176, 185 denizens ( of hell) 71, 79, 95, 96, 135ff.,
counterfactuals 4, 106–7, 108, 110, 111, 142–4, 152–3
112 free choice of 77, 97, 172–3; see also
Craig, William 4, 108–13, 123 damned; sinners
criticism of 111–12 deontic status 124
creation 4, 17–20, 26, 59, 107, 111, 112, depression 61
130, 147, 155, 156 desert 124, 125, 151, 153, 157, 158, 160,
and existence 156, 157 213
good of 114 negative 124
necessary conditions of 18, 19 positive 124
two-stage 18–20 desires 16, 31, 59, 69, 79–80, 158, 183
creatures 49, 59, 74, 75, 107, 110–11, 206, to be in hell 79
209 defective 128
redeemed 111, 112, 123 disordered 204–5, 216
unredeemed 111, 112, 123; see also elimination of 39, 41
persons; wantons ends of 202
creeds 117 first-order 70, 71, 72, 73, 75, 80
crime 188, 190, 193 for heaven 165
Cyril of Alexandria 98 and knowledge 35
second-order 70, 72, 73, 75, 80
damnation 4, 6, 29, 75, 112, 155, 199 to sin 33, 34, 35, 39, 40
causes of 200, 208–13ff. weakening of 40
problems with 216–18 destiny 24
evils of 206–8 destructive behaviour 12, 13, 14, 25, 33–4,
God justifying reasons for 200–203 81
prevention of 200ff. determinism 104–5, 111
transworld 110, 111, 123 difference 10
damned 46, 111, 123 morally significant 11
choices of 209, 211, 212 dignity 157, 160, 218
Index 223
disability 187, 191, 193, 198 infinite 124, 139

disabled children 188–9, 191, 193 duties 113, 122, 123, 127, 191
disagreements 165, 197, 198 of God 201, 203, 210
discovery 23, 26, 27 to have children 179, 181–3, 185, 190
disease 207
disincentive 120 earthly life 17, 18, 24, 27, 36, 83, 84, 101,
disobedience 4, 21, 92, 93, 127, 151 125, 126, 140, 141, 142, 152
disordered soul 138 Edwards, Jonathan 16, 65, 103, 172
divine action 78, 155 Ekstrom, Laura 9, 10, 14
divine attributes 53 elect 155
divine essence 204; see also beatific vision empirical beliefs 23
divine foreknowledge 2, 12, 14, 17, 88, emptiness 2, 74
104, 105, 106 encounter 25
divine frustration 18, 77, 78, 89–90, 136, end 7, 9, 30, 85, 130, 149, 166
144, 217, 218 ends 112, 113, 202, 210–11, 212, 215
divine intervention 37–41 enemies 96
divine judgement 45; see also God’s enslavement 33, 42
justice; judgement environment 158, 188, 191
divine providence 170, 171 Ephesians 7, 27, 91, 92, 98
strong view of 104, 105–6, 107, 108 epistemic problems 13, 35n., 52, 187, 196,
divine supremacy 61–3, 64 197, 198
doctrine of hell (DH) 1–2, 67, 133, 163, error 17, 18, 19, 23, 25, 109, 168, 177, 206
179, 180, 186, 194 escapism 3–4, 67, 74, 77–90, 100, 115,
and annihilationism 53–6ff., 61, 64 116, 117, 139, 142
constraints on 133–4ff. and Christian finality 77, 78, 83–9
defence of 29, 43, 151–2, 159, 161, defence of 77, 78, 90
213, 218 and God’s plans 89–90
generic definition 203–5 and heaven 83, 84–6
and God’s justice 54–5, 57 and hell as a good 77, 78, 79–83
and God’s moral perfection 53–63 objections to 83, 84, 87, 89
objections to 1, 29, 45–6, 54ff., 67ff., and punishment 129, 131, 132
136, 137–8ff., 143–4, 153–4, 196, and universalism 89, 90
198, 199, 200, 216ff. eschatology 3, 68, 77, 84, 87, 88, 111, 164
solutions to 2–6 denial of 100
species of 199–200, 206, 213–18; see eschaton 78, 88, 90
also God justifying reasons; hell; eternal damnation 29, 109
problem of hell eternal life 36, 55, 85, 112, 126, 134, 165,
doctrines 134, 148, 179, 180, 185, 186, 204
194, 196, 197, 198 eternal punishment 1, 2, 54–5, 136,
Dostoevsky 184–5 149–50, 197
double election 155 defence of 151–2
driving while intoxicated 190–91, 193 eternity 1, 2, 3, 77, 87, 94, 96, 97, 102,
drug addicts 15, 16, 34, 39, 80–81 115, 116, 133
choice of 32–3, 69 meaning of 150
unwilling 80 of moral degeneracy 216–17, 218
willing 81, 82 rejection of God for 32–7, 100
dualism, cosmic 61 salvation for 100
duration 125, 141
ethics 166, 167, 168, 183, 185; see also facts 47, 111, 134, 147, 148, 159
morality failure 30
eudaimonism 34 faith 7, 22, 24
euthanasia 189 authentic Christian 186, 194, 195,
Eve 176 197–8
everlasting punishment see eternal false assumptions 25, 164
punishment false beliefs 52, 60
evidence 25 false consciousness 101
evil 4, 32, 51, 92, 121, 149, 152, 206–8 families 196
anti-theistic argument from 105, 107, fantasies 12, 13
108 fathers 175–6; see also parents
choice of 176–7 fears 14, 16
consequences of 33–5, 47 Feinberg, Joel 191
damage by 52 figurative language 94, 141, 142, 150
degrees of 48, 49, 53 finality 77, 78ff., 100, 101; see also
existence of 1 consummation
inflicted by God 139, 140, 208, 216 finite actions 126–9
kinds of 34 finite creatures 54, 55
nature of 34, 49 finite experience 141, 143
problem of 1, 4, 38, 108, 184; see also finite punishment 120
ancillary evils; experiential evils; finite series 32
objective evils; positive evils finite sin 138
evil action 32 finite suffering 116, 140, 143
exaggeration 133 fire 55, 65, 69, 94, 117, 133, 137, 149, 159,
exception cases 135, 136 163, 205
exclusivism, Christian 15, 108, 109, 178, as metaphor 150, 155
179, 180, 186 symbolism of 137, 141, 143
problems of 196, 198 Flint, Thomas 104, 105
existence 5, 25, 29, 58, 67, 116, 125, 141, flourishing 158
146–7, 156, 183–4 foolishness 66
absolute 48, 49 forgiveness 26, 94, 96, 100, 146, 151
and being 48, 49, 209–10 form 48, 49
in a certain respect 48 Frankfurt, Harry 3, 68, 70, 72–3, 75,
of God 105, 107, 146, 147, 156ff. 79–80, 81
in hell 153; see also hell, existence of free action 11, 29, 31, 40, 41
Existence Thesis 67, 74 concept of 70; see also free choices
experience 11, 12, 15, 16, 23, 24, 25, 33, free agents 8, 13, 16, 29
35, 36, 39, 42, 65, 66, 207 free choices 8, 9, 17, 23, 24, 25, 96, 97,
of divine essence 204 112, 113, 155, 156, 164, 173, 174
of eternal suffering 135, 138, 139 and character formation 9–13, 25
finite 140, 141, 143 conditions of 104
as grounding condition 133, 134 of damned 211, 215
of hell 141–2 endless series of 32, 33
infinite 141 and God’s actions 78, 209
of misery 139–40 and inevitability of sin 110
experiential evils 207, 208 necessary condition of 32
ex-persons 75 number of 32
‘extrinsic view’ 153
Index 225
and rejection of God 29–30ff., 36, free decisions of 107

37–8, 41ff., 176 goodness of 1, 3, 54, 57–8, 67, 107,
removal of 38–9; see also infernal 111, 151
voluntarism hatred of 205
free will 2, 3, 22, 23, 29, 70, 104–5, 107, intervention of 37–41, 42
108, 110, 143, 144 meaning of 145, 146
loss of 71, 72 moral character of 2, 6, 200, 201, 210,
necessary condition of 72 212, 213
override of 177; see also free action; moral perfection of 46, 54, 57–8, 64
free choices motivation of 78, 88, 120, 200, 201–3,
freedom 7, 13, 22, 29, 32, 42, 120, 128, 208, 211, 212, 216; see also God-
152, 177, 178, 211 justifying reasons
counterfactuals of 108, 110, 111, 112 nature of 62–3, 67, 93, 94, 119, 134,
first-order 59 136
and ignorance 40, 41 and non-Christians 170, 171
morally significant 38 punishment inflicted by 139, 208, 216
second-order 59 purposes of 3, 4, 85, 89, 104, 106, 218
to sin 38, 143 rebellion against 62, 63
weak sense 209; see also free rejection of damned by 211, 212,
action; free choices; free will; 213ff., 217–18
libertarianism responsibility of 20, 21, 105, 112, 156
frustration see divine frustration right to punish 118, 126, 131
future 7, 97, 105, 116, 140, 142, 157 and sin 217, 218
persons 191, 198 and time limits 87
statements about 147–8, 157 unable to save damned 209, 210, 211,
212
Gandhi 109 unable to send persons to hell 119ff.
Geach, Peter 1 will of 2, 30, 31, 34, 108, 201, 209,
Genesis 179, 181, 182 211, 212, 213, 214, 215, 217
Gentiles 98, 99 wronging of 127, 129; see also God,
geography 115, 195, 196 existence of; God’s justice; God’s
gift 7, 78, 92, 152, 180, 181 love; image of God; rejection of
global perspective 194–5, 217 God; separation from God; union
God 13, 24, 26, 30, 182, 198 with God; wrath of God
actions of 201, 202, 210, 212, 213 God, existence of 5, 25, 146, 147, 156, 157
affront to 218 arguments against 105, 107, 145, 146,
attributes of 93, 94, 145 147, 148, 159, 161
choices by 113, 155, 209, 210, 211, God justifying reasons 199, 200–203, 208,
216 209ff., 213ff.
creation of persons by 17–20 God’s justice 54–5, 57, 78, 97, 115ff., 134,
culpability of 3, 4 138, 148, 159
decisions of 209 contradiction of 136, 137, 143
defeat of 62, 210, 217, 218 and eternal punishment 151
duties of 201, 203, 210 and grace 92
and evils 208, 216 and unrepentant sinners 147
fairness of 174 God’s love 3, 59, 62, 63, 68, 71, 91, 92, 93,
foreknowledge of see divine fore- 102, 108, 111, 113–14, 121, 138,
knowledge 155, 174
contradiction of 136, 138–9, 143, 148, greater good 112, 113

186, 211 Gregory of Nazianzus 98
and damnation 200ff., 215, 217 grounding condition 5, 118, 119, 133, 134,
and God’s actions 78 136
defeat of 62 guilt 16, 21, 101, 189
and hell 77, 79, 97, 151
post-mortem response to 100 Hades 98, 99
God’s plans 77, 78, 89–90, 136, 144 happiness 33, 35, 37, 39, 134, 187, 204
defeat of 218 Christian view of 164
good 3, 4, 19, 35, 36, 38, 51, 69, 130, 151, in heaven 135, 136, 144
184, 2004 in hell 135
aggregative theory 79, 82–3 rejection of 135; see also beatific
of creation 114 vision
greater 176 harms 93, 127, 186
of having children 179, 180–81, 187 absence of 183, 186
and hell 77, 78, 79–83, 116, 131, 132 and having children 185, 187, 192–5
lesser 176 and risk-creation 190
of persons 114 severe 193, 194ff.
qualitative 79–80, 82 harrowing of hell 97–8
quantitative 79, 81, 122–3; see also hatred 96, 135, 205
supreme good hearts 96, 97, 102
good actions 202 heaven 3, 4, 37, 39, 66, 79, 120, 121, 123,
good character 9, 11, 12, 16–17, 24, 25, 130, 131, 155
125 choice of 164, 165, 169, 173
source of 14 consummation of 83, 84–6
good choices 12, 17 existence of 125
good motives 124, 202 happiness in 135, 136, 144
good works 14, 20, 21 and hell, 111, 123, 125, 132, 136, 144
goodness 107, 176 rejection of 96
and being 47–9; see also good; good transfer to 97
character well-being in 82
gospel 98, 99, 100, 153 Hebrews, book of 100
ignorance of 108–9, 163 hell
Gould, James B. 21 badness of 154
grace 2, 3, 4, 51, 82, 88, 102, 143 ‘being in’ 153, 154, 158
definition of 92 belief in 145
and justice 92 choice of 79, 96, 97, 164, 169, 172–3,
and libertarianism 102 175; see also infernal voluntarism
limits on 20, 23 delays in consignment to 100
and mercy 21–2 deserving of 54, 55, 56
and non-believers 97 duration of 53, 54, 93
operation of 24–7 emptiness of 2
resistance to 7 eternity of 93, 94, 97, 102, 134
traditions 20–24 experience of 141–2
and wrath 93, 94 finality of 77, 78, 83, 86–9
gratitude 23, 127 as genuine threat 135, 136, 143
Great Divorce, The (Lewis) 43, 66, 74, good of 77, 78, 79–83, 116, 131, 132,
155, 159, 171, 172, 173 151
Index 227
gradations of 96 ignorance 14, 18, 22, 24, 31, 97, 101, 102,
and heaven 111, 112, 123, 125 108–9, 110, 111, 128, 176, 207
hybrid view of 73, 74 and freedom 40, 41
imposition of 115, 119, 120 and punishment 119
inhabitants of 152–3 removal of 40
life in 157 of those in heaven 136, 144
meaning of 146 illness 188, 205, 206, 207
mild views of 55, 56 illusion 19, 22, 23, 25
nature of 54, 55–6, 65–6, 94–6, 115– image of God 51, 53, 59, 85, 147, 157, 160
17, 133, 135, 138, 141, 154, 160 imagery 65, 66
opportunities in 78 imagination 65, 66, 168, 215
reasons for 43, 151–2159 immoral behaviour 34, 35, 139
strong doctrine of 137 immoral choices 10, 12
versions of 5–6, 136, 137 beneficial effects of 11, 25; see also
weak doctrine of 136, 137; see also bad character; bad choices
doctrine of hell (DH); hell, imposition 69
existence of; problem of hell; incarnation 5
sufferings of the damned inclusivism 194
hell, existence of 3, 5, 29, 74, 77, 97, 125, incompatibilism
126, 129, 133–44, 146, 147, 156 wide-source 86
in New Testament 136, 137, 143–4 inconsistency 67, 86, 102, 107, 134
objections to 133, 134, 137–8ff., logical 146, 147, 148, 153
143–4, 148 indeterminism 9, 18, 24
Helm, Paul 21–2 individuals 24, 93, 114, 116, 123
Henry, Matthew 154 existent 183
Hick, John 11, 170 possible 184; see also creatures;
hierarchy (of motives) 202, 203, 213 persons
higher-order goods 123 infants 51, 97, 119, 142
higher-order motives 202 suffering of 184–5, 188–9
Himma, Kenneth 5 unbaptized 133, 134, 135, 143
Hinduism 109, 166, 194 inference 186
history 104, 105, 106, 148–52, 168, 171 infernal voluntarism 164, 165–6, 169–74,
Hitler 8, 14, 24, 52–3, 96, 112, 121, 136 178
Hobbes, Thomas 92, 93 objections to 174–8
holiness 151, 160, 161, 167 infinite acts 128–9
hope 92, 102 infinite attitudes 125
horrors 94, 103 infinite gain 120
human beings 51, 54, 74, 115, 127, 134, infinite intensity 124, 125
142, 152, 167 infinite loss 126, 130, 138
infinity of acts 129 infinity 4, 93, 115, 116, 122, 129, 130
punishment of 119ff. finite experience of 140–41
purpose of 85 of punishment 54, 55, 121, 123–4,
sacrifice of 112–13 126–30, 136, 138, 139
saved and unsaved 111, 112, 123; see character-based 124–5
also creatures; persons of well-being 79
human nature 68, 92 information 60, 175, 176, 197
hybrid theory 212 ingratitude 15
injustice 5, 97, 98, 99, 130, 138, 139
insanity 119, 120 Kretzmann, Norman 47

instantiation 48, 49 Kronen, John 5, 6
intellect 142, 204, 206, 216 Kvanvig, Jonathan 1, 57–61, 67, 68, 71,
intellectual motives 202–3, 210, 213 74–5, 83–4, 88, 100–101, 127,
higher-order 202, 210 200n.
lower-order 202
intensity 124, 125, 139 language 88, 133, 150
intentions 80, 82, 84, 85, 126, 129, 142 Last Battle, The (Lewis) 169–70
defect in 131 Last Judgement 88, 134, 142, 150, 175
interference 57 law 191; see also moral law
interpretation 46, 54, 133 Lazarus 94–5
intervention 36, 37, 38, 40, 41, 42, 130, learning 11, 25, 33
211 Leibniz 201n.
intrinsic property 116 Lewis, C. S. 5, 27, 43, 66, 68, 74, 75, 155,
intrinsic value 203 159, 165–6, 178
‘intrinsic view’ 153 and infernal voluntarism 164, 169,
intuitions 124, 127, 179, 184, 188, 189, 172–4
190, 192, 193, 198 and religion 166–8ff.
ipso facto 59 liability 193
irreparable harm 37, 38, 41 libertarianism 2, 8, 9, 10, 11, 14, 22, 29,
issuantism 3, 77, 79, 83 31, 86, 209, 215, 218
defence of 29
Jews 99, 171, 194 and grace 102
John 97, 99 and Molinism 104, 107, 108, 111
joy 40, 96, 173, 180, 183, 204 lies 190
judgement 84, 91, 92, 94, 174, 175, 202; life 36, 39, 40, 96, 99, 101, 109, 125, 141,
see also Last Judgement 142, 153, 181, 183, 206
just war theory 112 gift of 152, 180
justice 1, 4, 5, 21, 57, 71, 147, 157, 159, in hell 112, 139, 155, 157
160, 212, 214, 218 sufferings of 187–8; see also afterlife;
and infinite punishment 115, 117–19ff., earthly life; eternal life
132, 136, 138–9, 143 life after death see afterlife
post-mortem 147 life histories 9, 12, 13
retributive 213, 214 light 206
violation of 136, 137; see also God’s limbo 135
justice limitations 20, 23, 37, 62, 86, 114, 160
on suffering 116
Kane Robert 2, 9, 10, 11–12, 86 limited election 20, 22, 24
Kant 112, 113, 152 locations 82, 131
karma 101 logic 27, 90, 106, 107, 109, 111, 123, 146,
Kershnar, Stephen 4 147, 148, 153, 194, 212, 214
kindness 21, 22, 93, 94 loss 49, 70, 71, 72, 135, 154–5, 158, 160,
Knight, Gordon 4 187, 206
knowledge 4, 14, 15, 31, 32, 35, 40, 41, awareness of 3, 69, 75, 135
123, 143, 144, 176, 192 of consciousness 216
defect in 128, 131 eternal 205
God’s 104, 105, 157 misery of 138–9; see also infinite loss
kinds of 51–2
Index 229
love 2, 3, 8, 15, 20, 21, 23, 59, 62, 83, 152, and counterfactuals 106–7, 108, 110,
157, 160, 180 111, 112
absence of 205 criticism of 104–5, 114
and free will 23 defence of 108
of God 204 and God’s knowledge 105–6
parental 114 and libertarianism 104, 107, 108
undeserved 22; see also God’s love; saved and lost souls 111–12
self-love and suffering 108
lower-order motives 202 moments 26, 31, 38, 42, 83, 86, 87, 88,
Lucifer 176 116, 140, 141, 171, 209
Luke 95, 117 monotheism 145, 148, 157
Luther 150, 151 and denial of hell 153–4
moral agents 16, 85, 86, 135, 143
Mackie, J. L. 107 children as 187
‘McTargett condition’ 5, 134, 137 moral bad 184, 185, 186, 197
manipulation 11, 25, 38, 112, 113, 114, 181 moral character 2, 6, 7, 56, 201, 204, 210
Mark 94 and change 84–5
marriage 179, 181–2 formation of 9–13
Matthew 95, 98, 100, 101, 117, 149, 150 and infinite punishment 124–6
meaning 26, 42 Paul on 13–15, 16
means 112, 113, 126, 210, 211, 212, 215 responsibility for 8
Melanchthon 204 moral choices 2, 3, 5, 8, 32
mental defect 128 moral consciousness 17, 31
mental life 142, 144 moral corruption 16
mental states 124, 125 moral degeneracy, eternal 216–17, 218
mentally ill people 24, 51, 97 moral development 101
mercy 20, 21–2, 62, 92, 216 moral duties 5, 187, 191
and God’s wrath 93–4 moral equilibrium 92, 93, 94
metaphor 94, 95, 150 moral error 109
metaphysical necessity 194 moral experience 15
metaphysical suicide 57–61 moral good 180–81, 183, 184, 185, 188,
metaphysical worth 158, 160 197, 212
metaphysics 11, 18, 61 moral hazards 187–8ff., 192
Middle East 195 moral impediment 210, 211
middle knowledge 4, 107, 123 moral improvement 24
misery 33, 34, 40, 45, 54, 96, 103, 154 moral law 15, 20, 80, 158, 167
choice of 173, 175 moral perfection 53–63, 92, 201, 202
as constraint 133 moral principles 193, 198, 211
degrees of 135, 136, 143 moral reasons 122, 210, 212
endless 138, 139, 142, 143 moral responsibility 8, 10, 13, 31, 86, 105,
and loss 138–9 120, 124
modal claims 86 loss of 128
de dicto 86 moral rights 127, 189
de re 86 moral sanctification 204
Molina, Luis de 105, 106 moral struggle 8, 26
Molinism 4, 104–14 moral vision 8, 11, 14, 22
advantages of 106, 107 moral worth 158
morality 92, 94, 98, 182, 191, 195, 197, non-existence 46, 47, 49, 53, 58, 116, 139,
198, 204, 212 146, 147, 158, 197, 210
Mormonism 117 non-existent persons 183, 184, 185, 186
mothers 8, 189 wrongs against 191
unmarried 191–2 non-sequitur 5
motivations 2–3, 9, 10, 14, 35, 42, 78, 79, numinous, the 166, 167, 169, 170
84, 119–20, 131, 181, 183, 201–3
and actions 201–2 obedience 92, 127, 211
and annihilationism 46ff., 53ff. object 70, 80, 82, 204
and desert 124 objective evils 207, 208, 216
and desires 80–81 objective value 204
and divine supremacy 61–3 obligation 21, 61, 139, 192
and God’s moral perfection 53–63, oblivion 216; see also annhilation
202, 203, 213 Old Testament 98
hierarchy of 202–3 omnipotence 104, 107, 143, 145, 209, 217
‘natural consequence’ 2, 3, 47–52, 64 omniscience 104, 105, 143, 145, 217
‘penalty model’ 3 open door policy 84, 87
and retribution 208, 215 Open Theists 148
to sin 33, 34; see also affective options 31, 32, 38, 39, 60, 96, 110, 113,
motives; bad motives; good 142
motives; intellectual motives Origen 149
Murray, Michael 3, 11, 68, 69–72, 73, 74 original sin 133, 143
Muslims 194, 196, 198 orthonomous actions 85–6
mystery 43 others 26, 34, 40, 96, 112, 113, 125, 152,
myths 167–8, 170, 174, 176 158
wronging of 127
natural consequence model 2, 3, 47–52, 64, outcomes 78, 87, 119, 122, 197, 216
68–72, 97 overpopulation 182, 198
naturalism 45, 151
nature 92, 118, 158 pagans 97, 98, 166, 168, 169; see also non-
necessary condition 10, 18, 19, 32, 72, 164, Christians
170, 194, 196 pains 51, 69, 70, 71, 73, 96, 125, 135, 154,
necessary truths 90, 106, 107, 147, 148, 184–5, 187, 188, 205, 214
156, 157, 159 absence of 185, 197
necessity 22, 35, 36, 92, 97, 187, 194 amount of 139–40
negative choice-consequence 120 and conscious suffering 207
New Life Principle (NLP) 192–3, 194ff., of unregenerate 215
198 Paloutzian, Raymond F. 196
New Testament 14, 20, 34, 94, 95, 97–8, parables 95
101, 117 paradise 98, 165
and hell 136, 137, 143–4 parents 37, 113, 114, 127, 128, 157, 175–6,
No-Escape Thesis 67, 74, 115, 117, 131, 177, 179, 180, 183, 186, 189, 191,
132 192
non-belief 109, 111 religious beliefs of 195, 196
non-Christians 5, 31, 46, 97, 98, 108–9, passions 75, 138
111, 163, 185, 194, 196, 198 paternalism 57, 58, 61
and God 170, 171 Paul, St 24, 25, 63, 93, 94, 98, 99, 158, 164
and universalism 164 on creation 17–20
Index 231
on grace 7–8, 9, 20, 24–7 possible worlds 86, 90, 107, 110, 112, 122,
on involuntary sin 15–17 123
and mercy 22 post-mortem conditions 36
on moral character 13–15, 16 post-mortem evangelism 4, 97, 98, 100
as sinner 13–14, 17 post-mortem justice 147
unwilling actions of 80 post-mortem punishment 5
‘penalty model’ 3 post-mortem salvation 2, 12, 98, 99–100
perceptions 60, 140–41 post-mortem suicide 57–61
Perelandra (Lewis) 167, 172 poverty 197
perfection 49, 198, 201, 202, 204 power 1, 62, 93, 138, 201
personhood 2, 51, 72 prayer 170, 197
persons 11, 13, 39, 45, 49, 80, 85, 200 predestination 7, 9, 24, 105
creation of 17–18, 147 predetermination 103, 104
and evils 207–8 preferences 84, 89, 177
existence of 147, 156 pregnancy outside marriage 191–2
good of 114 premises 11, 14, 58, 59, 119ff.,147, 148,
in hell 74, 75, 115, 153 156, 159, 165
non-existent 183ff., 191 preservation 209, 210
realized 158, 159, 160 prima facie 189, 200, 201, 202, 203, 209,
survival of 52–3 210, 213, 217
value of 5, 157, 158 principles 122, 192, 193
and wantons 73–4 privation 47, 53, 64, 204, 205, 214, 215,
worth of 157–60; see also creatures; 216
human beings probability 190, 192, 193, 194ff.
persuasion 90 problem of hell 1–2, 3, 4, 6, 67, 68, 69,
Peter 14, 98 70, 77, 91, 103, 110, 114; see also
Philippians 99, 101 doctrine of hell (DH)
philosophers 66, 135, 176, 182, 215 Problem of Pain, The (Lewis) 173
philosophy 22, 29, 46, 47, 48, 133, 134, procreation see birth
136, 138, 157, 160, 166, 198, 200, progress, moral 26
212 property rights 127, 128
physical nature 19 proportionality 5, 118–19, 179
Pinnock, Clark 61, 62 propositions 25, 116, 146, 148, 156, 165
pity 21, 22 inconsistency of 4
place 74, 82, 83, 84, 103, 152 Protestantism 20, 23, 117, 150
Plantinga, Alvin 4, 19, 20, 85, 107, 110, providence 104, 108, 114; see also divine
209 providence
Plato 99, 138, 148, 176 Psalms 62, 150, 181
pleasure 116, 126, 184, 185 psychology 18, 23, 31, 71, 74, 80, 85, 89,
absence of 184, 185, 197 100, 115, 201, 207
Plug, Allen 3, 4 punishment 3, 4–5, 15, 16, 29, 30, 71, 73,
pluralism see religious diversity; see also 91, 95, 97, 174, 206
non-Christians as constraint 133
population 82, 87, 152–3 degrees of 56, 135
positive evils 206, 207, 208 deserved 15, 41, 68, 124
possibilities 110, 184, 212 duration of 4, 53
possible circumstances 106 eternal 150, 160, 205
possible persons 191, 198 finite 120
infinite 54, 67, 119, 122, 136, 151 Reitan, Eric 5, 6

and moral character 124–6 rejection of God 2, 4, 29–30ff., 152, 156
inflicted by God 139 by the damned 209, 211–12, 214
and justice 115, 117–19ff., 132, 136, problems of 217–18
138–9, 143 choice of 30, 176–7
ceiling of 118–19 and Christ 99–100
justification of 92 forever 32–7, 56, 96, 97, 102
as loss of God 69, 70 forms of 30–31
notion of 138 freely rejecting God 30–32ff.
right to 118, 126, 131, 132 God’s facilitation of 42
self-inflicted 68, 69ff. God’s refusal of 37–41
severity of 118, 119 and God’s supremacy 62–3
undeserved 54, 55, 136, 151 and non-Christians 109
unjust 130, 136, 143 and non-existence 59
purgatory 85, 100, 115, 131, 154 and self-love 69
purposes 9, 85, 89, 104, 106, 218 relationships 11, 38, 59, 114, 127, 194, 209
relativism 92
qualitative good 79–80, 82 religions 109, 166–8, 171
quantitative good 79, 81, 82, 83, 122 common elements 168, 174
question-begging 126, 129 religious beliefs 23, 195–6
religious diversity 109
rational agents 18, 24, 25, 72, 82, 89 problem of 5, 163, 164–5
rational creatures 204, 215, 216 remorse 189, 194, 197
rational suicide argument 59, 60 reparation 93
rationality 2, 11, 16, 17, 18, 24, 31 repentance 26, 45, 94, 97, 102, 142, 146,
and earthly suicide 60, 61 163
notions of 60–61 refusal of 175
and post-mortem suicide 57–8, 60 resistance 7, 12, 165, 173, 175
weak form 60, 61 respect 211, 215
ratios 112 responsibility 9, 10, 20, 21, 105, 112, 123,
reason 138 130, 132, 156
reasons 122, 134, 151, 152, 160, 184, diminishing factors 119, 128
199ff. resurrection 45, 99
compelling 212, 214 retribution 5, 15, 16, 67, 74, 83, 95, 96, 99,
logical 212, 214 149–50, 151, 158
morally good 212, 214 and ancillary evils 208
prima facie 189, 202, 210, 213; see doctrine 215
also God justifying reasons; moral and justice 213, 214
reasons and loss 154, 155
rebellion 62, 63, 164, 173, 175, 176, 177 and worth of persons 160
reconciliation 26, 69, 89, 90, 92 Retribution Thesis 67, 71, 72, 74
redemption 19, 20, 77, 111 revelation 5, 39, 90, 93, 133, 134, 137,
post-mortem 99; see also divine 143, 167
frustration general 108, 170
reductio 186 and myth 168
reflection 39, 60, 61, 81, 82, 154 Revelation, Book of 100, 117, 150
regret 69, 70, 71, 72, 73, 96, 135, 184 right 31, 35, 38, 85, 92, 93, 94
reincarnation 101 right acts 124
Index 233
righteous people 98, 204 selfishness 16, 30, 39, 40, 41, 205
righteousness self-love 68–9, 70, 71, 72, 73, 75, 152, 206
evil effects of 14, 17 lack of 205
rights 118, 119, 126, 127–8, 189 self-righteousness 14, 66
future 191 Sennett, James F. 9
of God 127 senses 135
risk 187, 195 separation from God 18, 45, 121
creation 190, 194ff. awareness of 207
of severe harm 193, 194–8 eternal 102
substantial 193 and God’s plan 78
Romans 15, 22, 80, 91, 92, 94, 98, 101 meaning of 96
sexual relations 192
sacrament 182 sexual temptation 12
sacrifice 112, 197 Shiffrin, Seana 187–8, 191, 192
saints 19 Sickler, Bradley 5
salvation 14, 15, 17, 22, 41, 84, 89, 155 ‘significant’ 116, 119
of all 210, 213 sin 16, 18, 19, 34–6, 47, 97, 146, 151, 153
chances of 99–100 addiction to 41
conditions of 194, 196 awareness of 207
easy 197 destructive effects of 51, 52
and God’s will 212, 213 endless 32, 33
meaning of 203–4 enslavement to 33, 42
and non-Christians 171–2 God’s reaction to 92, 217
opportunities for 78 and ignorance 14
post-mortem 2, 98, 99–100 infinite 93
and punishment 99 involuntary 15–17
and will of the damned 209, 211, 212 motives to 33, 34
sanctification 204, 218 nature of 218
Satan 176 and rejection of God 30, 32ff., 41ff.,
saved creatures 111, 112, 123, 126, 209, 100
218 unforgivable 100
schizophrenics 24 victory of 218; see also bondage; evil;
schools 196 sinners
scripture 4, 5, 29, 54, 88, 90, 93, 154, 158 sincerity 92, 109
and children 181–2 single election 155
as grounding condition 133, 134, sinners 1, 13, 16, 20, 23, 33–4, 98, 99, 102,
136, 143; see also Bible; New 205, 212
Testament attitudes 33–6
second-chance concepts 60, 97, 100, 102 free choice of 2, 3, 215
secular viewpoint 186 freedom of 38
Seinfeld 66 ignorance of 40–41
self 52, 166, 177 persistent 34, 36, 39, 41, 52, 147, 152,
divided 80 153, 177
self-awareness 82 suffering of 1, 3, 206
self-destruction 177 unrepentant 5, 45, 152, 158, 159
self-forming action 9, 10, 13, 86 will of 69, 70; see also damned; sin
self-governance 176 Smith, Michael 85
self-interest 93, 155, 159, 183 sociology 195
Socrates 14, 176, 177 ancillary evils of 206, 207, 208, 216
Some Inhabitants Thesis 115, 117, 131, doubts about 216
132 permitted by God 209, 216
sorrow 135, 184 positive evils of 206, 207, 208
soul 138, 148, 149, 172, 205, 206 and privation of beatific vision 206,
sovereignty 1, 87, 88 207
space traveler 139–40, 141 sufficient condition 10, 19, 31, 194, 196
spiritual welfare 20, 21 sufficient probability 193
spirituality 19 suicide 41, 60, 61; see also metaphysical
state 190, 193 suicide
state authority 127, 128 supererogation 20, 21
statements 147–8, 157 supernatural 168
states of affairs 71, 74, 78, 80, 81, 82, 122, supreme good 70, 73
123, 125, 151, 153, 157, 160 survival 52–3
states of being 131 symbolism 133, 137, 141, 143
statistics 195 symmetry 84
stories 26, 27, 95, 166, 168, 174
strong view 67 Talbott, Thomas 2, 33–5, 36, 37, 38, 41,
strong-willed behaviour 11, 12 89, 125, 176, 177
Stump, Eleonore 47 teleology 68
subject 207 temporal locations 82
subjective value 204 temptation 8, 12, 31
submission 164 terrorism, religious 14
sub-proof 147, 156, 159 Tertullian 149
substantial form 48, 49 theodicy 19, 43
suffering 1, 3, 52, 103, 114, 125, 126, 154, theism 86, 107, 147
184, 186, 187–8 theologians 215
absence of 184, 185 theology 160, 201, 218
amount of 139–40 Thessalonians 94, 95, 101, 117
avoidance of 197 Thomism 47, 48
awareness of 207–8 thought-experiment 139
of children 179, 184ff. ; see also birth time 18, 38, 42, 46, 53, 82, 83, 86, 87, 116,
conscious 207, 208 129, 131, 140, 141, 160
as constraint 133 ‘last’ 87, 88
deserved 124, 125 I Timothy 13, 97, 181
eternal 116, 135, 140, 197 torment 65, 69, 95, 117, 133, 137, 150,
experiential 135, 138, 139–40, 207, 179, 180, 186, 194, 197, 205, 207,
208 208, 215
finite 116, 140, 143 torture 30, 55, 56, 150, 154, 184, 201
infinite 139 traditional concepts 103
inflicted by God 139, 140, 208 traditions 20, 45, 47, 109, 117, 133, 134,
limits on 116 136
and Molinism 108 tragedy 30, 113, 159
and procreation 187 transfer 126, 129
self-inflicted 96 transformation 13, 25, 40, 85
unjust 139; see also punishment; transition 73, 74
sufferings of the damned transworld damnation 110, 111, 123
sufferings of the damned 205–8, 213–18 transworld depravity 110
Index 235
trust 11, 23 second-order 70, 71, 72, 73, 80, 81, 82;
trustworthiness 5, 134 see also infernal voluntarism; will
truth 52, 90, 106, 107, 147, 157, 168, 194, volitional capacities 53
196, 208 voluntary actions 11
unbelievers 109 Walls, Jerry 2, 3, 200n.

unconsciousness 207, 216 wantons 3, 72–5, 80, 81, 82
unfaithful behaviour 11 and persons 73–4
union with God 18, 25, 26, 27, 40, 41, 136, Weak Pareto principle 122
142, 143, 144, 204 weakness 17
United States 188, 195 weak-willed behaviour 11, 12, 69, 128
universalism 2, 4, 5, 33, 45, 46, 53, 97, wealth 197
102, 116, 146, 149, 153, 170, 194, welfare 200, 206, 211
210, 213 well-being 127
defect of 165 in hell 79, 121
and divine frustration 89, 144 infinitely negative 115–16, 117, 130,
and escapism 89, 90 132
and non-Christians 164 levels of 116
and punishment 131, 132 quantity of 82, 83, 122–3, 125
universe 4, 104, 105 of wantons 81
unloving actions 78 wicked being 134, 137
unregenerate sinners 215 wickedness 91, 151, 207, 214
uptake 126, 129 confirmed in 217, 218; see also bad
utilitarianism 185 character; moral bad
will 12, 13, 17, 23, 38, 39, 69–70, 72, 75,
vagueness 193 79, 80, 173, 204
value 13, 38, 116, 203, 210, 218 against people’s 69, 70, 71, 73, 91, 96,
and escapism, 79–83 164, 178
of freedom 41, 120 antecedent 201
intrinsic 112, 113 concept of 70
objective 204, 216 consequent 201
of persons 5, 114 damage to 51
subjective 204; see also values of damned 209
value theories 79, 82 defect in 128, 138
values 125, 158, 160, 183, 204, 216 and future actions 142
location of 82, 122 lack of 69; see also infernal
VanArragon, Ray 2 voluntarism
Vargas, Manuel 13 Williams, Bernard 13
via media 215, 216 wisdom 66, 127, 151, 170, 176, 182
vice 124, 125, 135 women, unmarried 191–2
victims 118, 126, 190, 218 works 7; see also actions; good works
violence 14 world 92, 93, 103, 104, 106, 107, 114, 121,
virtue 14, 24, 124, 125, 205 188; see also possible worlds
virtuous persons 7, 8, 11; see also good worm 94, 150, 154, 155, 159
character; moral character worth 153, 204, 218
vision 204; see also beatific vision higher 159
volition 131, 165 intrinsic 157, 160
metaphysical 158, 160
moral 158 acts 124, 189, 190, 192, 193

non-moral 158 to have children 186, 188–93, 195,
positive 153, 157, 159 196, 198
wrath of God 4, 91, 92, 155, 175 infinite 115, 128, 129, 130
and grace 93, 94 wrongful life lawsuits 191
and mercy 93–4
purpose of 92, 93 Yandell, Keith 5
wrong 31, 32, 92, 93, 94, 107, 127, 128

Buenting J. (Ed.) - The Problem of Hell. A Philosophical Anthology

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

Buenting J. (Ed.) - The Problem of Hell. A Philosophical Anthology

Uploaded by

Copyright:

Available Formats

THE PROBLEM OF HELL

For Mom and Dad

Published 2016 by Routledge

Routledge is an imprint of the Taylor & Francis Group, an informa business

British Library Cataloguing in Publication Data

Library of Congress Cataloging-in-Publication Data

ISBN 9780754667636 (hbk)

List of Contributors vii

1 Grace, Character Formation, and Predestination unto Glory 7

2 Is it Possible to Freely Reject God Forever? 29

3 Annihilationism: A Philosophical Dead End? 45

4 Compatibilism, “Wantons,” and the Natural Consequence

5 Value, Finality, and Frustration: Problems for Escapism? 77

6 Hell, Wrath, and the Grace of God 91

7 Molinism and Hell 103

8 Hell and Punishment 115

9 Why I Am Unconvinced by Arguments against the Existence

10 Hell and Natural Atheology 145

11 Infernal Voluntarism and “The Courtesy of Deep Heaven” 163

12 Birth as a Grave Misfortune: The Traditional Doctrine of Hell and

13 Species of Hell 199

Justin D. Barnard is Associate Professor of Philosophy and director of the Carl.

Claire Brown is a Ph.D. candidate and lecturer in philosophy at the University

Joel Buenting is lecturer of philosophy at the University of Alberta. He has

Andrei A. Buckareff is Assistant Professor of Philosophy at Marist College,

James Cain is Associate Professor of Philosophy at Oklahoma State University.

Stephen T. Davis is the Russell K. Pitzer Professor of Philosophy at Claremont

Kenneth Einar Himma is an Associate Professor of Philosophy at Seattle Pacific

Stephen Kershnar is a Professor of Philosophy at the State University of New

Allen Plug is Assistant Professor of Philosophy at Malone University. His areas

Eric Reitan is a Professor of Philosophy at Oklahoma State University. Dr.

Brad L. Sickler received his Ph.D. in philosophy from Purdue University

Thomas Talbott is Emeritus Professor of Philosophy at Willamette University,

Raymond J. VanArragon is Associate Professor of Philosophy at Bethel

Jerry L. Walls is Professor of Philosophy at Asbury Theological Seminary. Dr.

Keith E. Yandell is the Julius R. Weinberg Professor of Philosophy at the University

The Problem of Hell: A Philosophical Debate focuses on philosophical problems

we find it objectionable. Others argue for the possibility of post-mortem salvation,

Free Choice and Character Formation

Grace Verses Works in Pauline Theology

Talbott, “A Pauline Interpretation of Divine Judgment,” in Robin A. Parry and Christopher H.

“Felix Culpa” or Creation in Two Stages

By this admittedly vague expression, I mean to imply, among other things, a

Divine Grace: Its Universal Scope and Unconditional Character

Predestination unto Glory

So far, I have challenged the assumption, widely shared by libertarians, that an

Freely Rejecting God Forever

Is it Possible to Freely Reject God Forever?

Must God Intervene to Prevent Unending Rejection?

But Could the Sinner’s Freedom be Maintained?

choosing to lose it. Such a person’s existence would be that of an irredeemable

Natural Consequence Motivations

According to what that we will call (broadly) “natural consequence” motivations,

This elaboration on the two ways of conceiving of being helps us to make

is capable of completely and irreparably destroying a creature’s capacity to know

Motivations Rooted in God’s Moral Perfection

A second class of motivations for adopting an annihilationist view involves

An initial attraction of the annihilationist’s position is that it avoids a problem

experienced, nonetheless last forever.22 Crucially, then, annihilationists classify

K1. Some of the wicked will attempt post-mortem suicide.

Motivations Rooted in Divine Supremacy

The weaknesses of the divine supremacy motivation become apparent when

When I look at thy heavens, the work of thy fingers,