(Cambridge Critical Guides) Michalson, Gordon E - Kant's Religion Within The Boundaries of Mere Reason - A Critical Guide-Cambridge University Press (2014)

KANT’S RELIGION WITHIN
THE BOUNDARIES OF MERE REASON
Kant’s Religion within the Boundaries of Mere Reason was written

late in his career. It presents a theory of “radical evil” in human
nature, touches on the issue of divine grace, develops a
Christology, and takes a seemingly strong interest in the issue
of scriptural interpretation. The essays in this Critical Guide
explore the reasons why this is so, and offer careful and illumi-
nating interpretations of the themes of the work. The relation-
ship of Kant’s Religion to his other writings is discussed in ways
that underscore the importance of this work for the entire
Critical philosophy, and provide a broad perspective on his
moral thought; connections are also drawn among religion,
history, and politics in Kant’s later thinking. Together the essays
offer a rich exploration of the work which will be of great
interest to those involved in Kant studies and philosophy of
religion.
gordon e. michalson is Professor of Humanities at

New College of Florida. His books include Fallen Freedom:
Kant on Radical Evil and Moral Regeneration (Cambridge,
1990) and Kant and the Problem of God (1999).
CAMBRIDGE CRITICAL GUIDES
Titles published in this series:

Hegel’s Phenomenology of Spirit
EDITED BY d ea n m o y a r an d m i c h a e l q u a n t e
Mill’s On Liberty
EDITED BY c . l. t e n
Kant’s Idea for a Universal History with a Cosmopolitan Aim

EDITED BY a m él i e ok se n b e r g r o rt y an d j a m e s s c hm i d t
Kant’s Groundwork of the Metaphysics of Morals

EDITED BY jens timmermann
Kant’s Critique of Practical Reason

EDITED BY a n d r e w s r e a t h a nd j e n s t i m m e r m a n n
Wittgenstein’s Philosophical Investigations

EDITED BY arif ahmed
Kierkegaard’s Concluding Unscientific Postscript
EDITED BY rick furtak
Plato’s Republic
EDITED BY m a r k l . mc p h e rr a n
Plato’s Laws
EDITED BY ch r i st op he r b o b o nic h
Spinoza’s Theological-Political Treatise
EDITED BY y i t z h a k y . m e l a m e d an d m i c h a e l a. r o s e n t h a l
Aristotle’s Nicomachean Ethics
EDITED BY j o n m i l le r
Kant’s Metaphysics of Morals
EDITED BY la ra d en i s
Nietzsche’s On the Genealogy of Morality

EDITED BY s i mo n m a y
Kant’s Observations and Remarks
EDITED BY su s a n m e l d s h e l l a n d ri c h a r d v e lk l e y
Augustine’s City of God
EDITED BY james wetzel
Descartes’ Meditations
EDITED BY k a r e n de t l e f s e n
Kant’s Religion within the Boundaries of Mere Reason
EDITED BY gordon e. michalson
KANT’S
Religion within the Boundaries
of Mere Reason
A Critical Guide
edited by
gordon e. michalson
New College, Florida
University Printing House, Cambridge cb2 8bs, United Kingdom
Cambridge University Press is part of the University of Cambridge.

It furthers the University’s mission by disseminating knowledge in the pursuit of
education, learning and research at the highest international levels of excellence.
www.cambridge.org
Information on this title: www.cambridge.org/9781107018525
© Cambridge University Press, 2014
This publication is in copyright. Subject to statutory exception
and to the provisions of relevant collective licensing agreements,
no reproduction of any part may take place without the written
permission of Cambridge University Press.
First published 2014
Printed in the United Kingdom by Clays, St Ives plc
A catalogue record for this publication is available from the British Library
ISBN 978-1-107-01852-5 Hardback
Cambridge University Press has no responsibility for the persistence or accuracy of
urls for external or third-party internet websites referred to in this publication,
and does not guarantee that any content on such websites is, or will remain,
accurate or appropriate.
Contents
List of contributors page ix

List of translations and abbreviations xii
Introduction
gordon e. michalson 1
1 Holy Scriptures within the boundaries of mere reason:
Kant’s reflections
o t f r i e d h öf f e 10
2 The evil in human nature
alle n w . w ood 31
3 Radical evil and human freedom
i n g o l f d a l f e rt h 58
4 Gesinnung: responsibility, moral worth, and character
alison hills 79
5 Rational hope, possibility, and divine action
a n d re w c h i g n e l l 98
6 Kant on grace
le slie ste v ens o n 118
7 Kant, miracles, and Religion, Parts One and Two
karl ameriks 137
8 Kant’s Jesus
ma nfr ed ku eh n 156
9 Pluralism in the ethical community
nicholas tampio 175
vii
viii Contents
10 Kant’s religious constructivism
pa b lo m uc h n ik 193
11 What does his Religion contribute to Kant’s conception of
practical reason?
g. felicitas munzel 214
12 Culture and the limits of practical reason in Kant’s Religion
richard velkley 233
Bibliography 250
Index 262
Contributors
karl ameriks is McMahon-Hank Professor in the Department of

Philosophy at the University of Notre Dame. He is the author of Kant’s
Theory of Mind (2000), Kant and the Fate of Autonomy (2000), Interpreting
Kant’s Critiques (2003), Kant and the Historical Turn (2006), and Kant’s
Elliptical Path (2012). He is co-editor of “Cambridge Texts in the History of
Philosophy” and has edited the Cambridge Companion to German Idealism
(2000) and Letters on the Kantian Philosophy, by Karl Reinhold (2006).
andrew chignell is Associate Professor in the Susan Linn Sage School
of Philosophy at Cornell University. He has published articles on early
modern philosophy, aesthetics, and philosophy of religion in journals such
as Noûs, Philosophical Review, Mind, Philosophical Quarterly, and Religious
Studies.
ingolf dalferth is Professor of Systematic Theology, Symbolism and
Philosophy of Religion at the University of Zürich; since 2012, he is also
Professor at the Swiss Federal Institute of Technology Zürich (Collegium
Helveticum) and, since 2008, also the Danforth Professor of Philosophy
of Religion at Claremont Graduate University in California. He has
authored and edited more than fifty books, including Die Wirklichkeit
des Möglichen. Hermeneutische Religionsphilosophie (2003), Malum.
Theologische Hermeneutik des Böse (2010), Umsonst: Eine Erinnerung an
die creative Passivität des Menschen (2011), and Selbstlose Leidenschaften.
Christlicher Glaube und menschliche Passionen (2013).
alison hills is a professor of philosophy and tutor at St. John’s College,
Oxford. Her research is in moral philosophy and epistemology, and her
publications include works on Kant’s theory of happiness and on the value
of rational nature. Her book, The Beloved Self, was published in 2010.
otfried höffe was previously Professor of Philosophy at the University
of Tübingen. He is Director of the Research Center for Political Philosophy
ix
x List of contributors
at Tübingen and Visiting Professor for Legal Philosophy at the University of
St. Gallen in Switzerland. His numerous publications on Kant and
Aristotle, as well as on political and moral philosophy, have been
translated into more than twenty languages and include Immanuel Kant
(1994), Political Justice: Foundations for a Critical Philosophy of Law and the
State (1995), Kant’s Cosmopolitical Theory of Law and State (2006), and
Kant’s Critique of Pure Reason: The Foundations of Modern Philosophy (2010).
manfred kuehn is Professor of Philosophy at Boston University. He
previously served on the faculties of Purdue University and Philipps-
Universität Marburg. He has written numerous articles on Hume, Kant,
Thomas Reid, and the Scottish, French, and German Enlightenments, and
his books include Scottish Common Sense in Germany, 1768–1800: A
Contribution to the History of Critical Philosophy (1987), Kant: A Biography
(2001), and Johann Gottlieb Fichte: Ein deutscher Philosoph (2012).
gordon e. michalson is Professor of Humanities at New College of
Florida, where he served as the college’s President from 2001 to 2012. He
was previously on the faculties of Oberlin College and Davidson College.
His books include Lessing’s “Ugly Ditch”: A Study of Theology and History
(1985), Fallen Freedom: Kant on Radical Evil and Moral Regeneration (1990),
and Kant and the Problem of God (1999).
pablo muchnik is an associate professor at Emerson College in Boston.
Educated originally in Argentina, he pursued advanced studies in Germany
and received his doctorate from the New School for Social Research. He is the
author of Kant’s Theory of Evil: An Essay on the Dangers of Self-Love and the
Aprioricity of History (2009), the editor of the first two volumes of Rethinking
Kant (2008/2010), and the co-editor of Kant’s Anatomy of Evil (2010).
g. felicitas munzel is Associate Professor in the Program of Liberal
Studies and the Department of Philosophy at the University of Notre
Dame. She is the author of Kant’s Conception of Moral Character: The
“Critical” Link of Morality, Anthropology, and Reflective Judgment (1999),
Kant’s Conception of Pedagogy: Toward Education for Freedom (2012), and
numerous articles on Kant’s moral philosophy, anthropology, and
pedagogical writings. She is the translator of Kant’s 1775/76 Friedländer
anthropology lectures in The Cambridge Edition of the Works of
Immanuel Kant (2012).
leslie stevenson was Lecturer, then Reader, at St. Andrews
University, Scotland, from 1968 to 2000, and is now Honorary Reader in
List of contributors xi
Philosophy. His publications include Twelve Theories of Human Nature
(6th edn. 2012), Inspirations from Kant (2011), and Open to New Light: An
Introduction to Quaker Spirituality in Historical and Philosophical Context
(2012).
nicholas tampio is Assistant Professor of Political Science at Fordham
University. He is the author of Kantian Courage: Advancing the
Enlightenment in Contemporary Political Theory (2012) and has published
several articles on the intersection of Enlightenment and Islamic political
theory, as well as the entry on Islamic political thought for the Wiley-
Blackwell Encyclopedia of Political Thought.
richard velkley is Celia Scott Weatherhead Distinguished Professor
of Philosophy at Tulane University. He is the author of Freedom and
the End of Reason: On the Moral Foundation of Kant’s Critical Philosophy
(1989), Being After Rousseau: Philosophy and Culture in Question (2002), and
Heidegger, Strauss and the Premises of Philosophy: On Original Forgetting
(2011). His four edited volumes include a collection of essays by Dieter
Henrich, The Unity of Reason: Essays on Kant’s Philosophy (1994), as well as a
volume in the Cambridge University Press “Critical Guides” series, Kant’s
Observations and Remarks: A Critical Guide (2012), co-edited with Susan
Meld Shell.
allen w. wood is Ruth Norman Halls Professor of Philosophy at
Indiana University and Ward W. and Priscilla B. Woods Professor
Emeritus at Stanford University. He has also taught at Cornell University
and Yale University and has held visiting positions at the University of
Michigan, the University of California at San Diego, and Oxford. He is the
author of numerous books and articles, chiefly on ethics and on the German
idealist tradition from Kant through Marx.
Translations and abbreviations
Citations of Kant’s works refer to the volume and page number in the
Akademie Edition of Immanuel Kant, Gesammelte Schriften (Berlin:
Walter de Gruyter and predecessors, 1900– ).The one exception to this
rule will be references to the Critique of Pure Reason, which simply cite the
page numbers of the A and B editions. Unless otherwise noted by the
individual author, translations are taken from Paul Guyer and Allen
W. Wood (general co-editors), The Cambridge Edition of the Works of
Immanuel Kant in English Translation (Cambridge University Press,
1992– ), hereafter “CAM.” The following abbreviations are used through-
out the book:
ApH Anthropology from a Pragmatic Point of View Ak 7

CAM: Anthropology, History, and Education, trans. Robert
B. Louden
Brief Briefwechsel Ak 10–11
CAM: Correspondence, trans. Arnulf Zweig
Dm On the Form and Principles of the Sensible and Intelligible World Ak 2
CAM: Theoretical Philosophy, 1755–1770, trans. David Walford and
Ralf Meerbote
EM The Only Possible Argument in Support of a Demonstration of the
Existence of God Ak 2
CAM: Theoretical Philosophy 1755–1770, trans. David Walford
End The End of All Things Ak 8
CAM: Religion and Rational Theology, trans. Allen W. Wood
Fort What Real Progress has Metaphysics Made in Germany since the Time
of Leibniz and Wolff? Ak 20
CAM: Theoretical Philosophy after 1781, trans. Peter Heath
G Groundwork of the Metaphysics of Morals Ak 4
CAM: Practical Philosophy, trans. Mary J. Gregor
xii
List of translations and abbreviations xiii
IC Proclamation of the Imminent Conclusion of a Treaty of Perpetual

Peace in Philosophy Ak 8
Idea Idea for a Universal History from a Pragmatic Point of View Ak 8
CAM: Anthropology, History, and Education, trans. Allen W. Wood
KpV Critique of Practical Reason Ak 5
KrV Critique of Pure Reason Ak 3–4
CAM: Critique of Pure Reason, trans. Paul Guyer and Allen
W. Wood
KU Critique of the Power of Judgment Ak 5
CAM: Critique of the Power of Judgment, trans. Paul Guyer and Eric
Matthews
L Lectures on Logic Ak 9
CAM: Lectures on Logic, trans. J. Michael Young
MA Conjectural Beginning of Human History Ak 8
CAM: Anthropology, History, and Education, trans. Allen W. Wood
MD Metaphysik Dohna Ak 28
CAM: Lectures on Metaphysics, trans. Karl Ameriks and Steve
Naragon
Metaphysik K2 Ak 28
MetM Metaphysik Mongrovius Ak 29
Naragon
ML1 Metaphysik L1 (Pölitz) Ak 28
Naragon
MM2 Moral Mongrovius Ak 29
CAM: Lectures on Ethics, trans. Peter Heath
MPC Moral Philosophie Collins Ak 27
MS Metaphysics of Morals Ak 6
O What Does it Mean to Orient Oneself in Thinking? Ak 8
OP Opus Postumum Ak 21–22
CAM: Opus Postumum, trans. Eckart Förster and Michael Rosen
xiv List of translations and abbreviations
P Prolegomena to Any Future Metaphysics that Will be Able to Come

Forward as Science Ak 4
CAM: Theoretical Philosophy after 1781, trans. Gary Hatfield
PPV Metaphysik der Sitten Vigilantius Ak 27
R Religion within the Boundaries of Mere Reason Ak 6
CAM: Religion and Rational Theology, trans. George di Giovanni
Ref Reflexionen Ak 14–23
RM Of the Different Races of Human Beings Ak 2
CAM: Anthropology, History, and Education, trans. Holly Wilson
and Günter Zöller
S The Conflict of the Faculties Ak 7
CAM: Religion and Rational Theology, trans. Mary J. Gregor and
Robert Anchor
Theod On the Miscarriage of all Philosophical Trials in Theodicy Ak 8
CAM: Religion and Rational Theology, trans. George di Giovanni
Ton On a Recently Prominent Tone of Superiority in Philosophy Ak 8
VA Lectures on Anthropology Ak 25
CAM: Lectures on Anthropology, trans. G. Felicitas Munzel and
Allen W. Wood
VpR Lectures on the Philosophical Doctrine of Religion Ak 28
WA An Answer to the Question: What is Enlightenment? Ak 8
ZeF Toward Perpetual Peace: A Philosophical Sketch Ak 8
Introduction
Gordon E. Michalson
Even by Kant’s standards, Religion within the Boundaries of Mere Reason is a

complex work, involving many layers in its depiction of issues central to his
moral and religious thought.
Its complexity is reflected in the uncommonly wide range of reactions the
work elicited in its own day, from the admonitions of the Prussian censor for
supposedly impious tendencies to the fierce reaction of Goethe, who famously
charged in a letter to Herder that Kant “had criminally smeared his philoso-
pher’s cloak . . . so that Christians too might yet be enticed to kiss its hem.”1
Insofar as one measure of a book’s richness is the range of those it displeases,
then Religion surely deserves its high standing as a key document of the late
Enlightenment. Certainly the myriad interpretive problems the book gener-
ates remain with us to this day, and the aim of this “Critical Guide” is to offer
clarification and guidance concerning some of the more prominent of these.2
Religion within the Boundaries of Mere Reason is composed of four parts,
published as a whole in 1793, well after the appearance of Kant’s best-known
major writings. A year before, Kant had published the first part (“Concerning
the Indwelling of the Evil Principle alongside the Good: or Of the Radical
Evil in Human Nature”) in the April issue of Berlinische Monatsschrift. Kant’s
submission of the first and, subsequently, second of these essays precipitated
his difficulties with the Prussian authorities, who had become much more
conservative following the enlightened era promoted by Frederick William I.3
1
Quoted in Karl Barth, Protestant Thought from Rousseau to Ritschl, trans. Brian Cozens (New York:
Simon and Schuster, 1969), p. 178.
2
A helpful summary of different contemporary interpretive approaches to Religion is provided in Chris
L. Firestone and Nathan Jacobs, In Defense of Kant’s Religion (Bloomington and Indianapolis: Indiana
University Press, 2008), Part I.
3
For background, see Allen W. Wood, “General Introduction,” and George di Giovanni, “Translator’s
Introduction,” in Immanuel Kant, Religion and Rational Theology, ed. and trans. Allen W. Wood and
George di Giovanni (Cambridge University Press, 1996), pp. xi−xxiv and pp. 41–54. See also James
J. DiCenso, Kant’s Religion within the Boundaries of Mere Reason: A Commentary (Cambridge
University Press, 2012), pp. 4–9.
1
2 GORDON E. MICHALSON
These difficulties indicate that Religion would appear at precisely the moment
when the rational inquiry into religion and religious belief would inevitably
be drawn into fresh controversy. Kant would insure a lively connection to
such controversy with his opening remark that “on its own behalf morality in
no way needs religion . . . but is rather self-sufficient by virtue of pure practical
reason” (R 6:3).
Ostensibly, the title of his book is designed to promote what Kant
himself calls a kind of “test” or “experiment” (Versuch) based on the simple
depiction of two concentric circles. Against the background of widespread
contemporary discussions of revealed and natural religion, Kant proposes
that “revealed religion” be viewed as the larger or wider circle, with the
“religion of reason” then viewed as a smaller circle within. In the examina-
tion of religion, the philosopher, “as purely a teacher of reason,” thus
remains within the smaller circle and avoids appeal to the traits of revealed
religion, such as scripture, history, and tradition. Yet because this “experi-
ment” involves concentric circles – and not two circles external to one
another – Kant has created the possibility of overlap or commonality
between revealed religion and the religion of reason. And if, as philosopher,
he were to discover such areas of overlap, “then we shall be able to say that
between reason and Scripture there is, not only compatibility but also
unity . . .” (R 6:12–13).
Obviously, this “experiment” is considerably more complicated than it
may at first seem, and not only because of its apparent artificiality.
Moreover, a telling feature of Kant’s seemingly straightforward explanation
is that it appears in the “Preface” to the second edition of Religion. Second
edition prefaces are notorious for the way they provide the author with the
opportunity to dispel confusions or reply to criticisms generated by the first
edition. In this instance, Kant is responding to reactions to “the title of this
work (since doubts have been expressed. . .regarding the intentions hidden
behind it)” (R 6:12). The long history of interpretive debate concerning the
true aims of Religion amply suggests that Kant’s efforts to provide definitive
orientation regarding the work’s content were hardly conclusive.
Kant does at least succeed in establishing that, in a general sense, his is a
book about the rational inquiry into religion, including the broad issue of
the relation between faith and reason. Since Religion was published eight
years following the publication of the first edition of the Groundwork of the
Metaphysics of Morals, and five years following the publication of the
Critique of Practical Reason, it naturally frames the issue of faith and reason
in terms defined by Kant’s moral theory and by the central role the ethical
life, as opposed to belief systems, plays in Kant’s view of religion. Indeed,
Introduction 3
Religion is where Kant develops more fully his definition of religion as “the
recognition of all our duties as divine commands” (R 6:153). Within this
context, and with reinforcing implications conveyed by the book’s title,
Religion creates the initial impression that it is a largely reductionist attempt
simply to assimilate traditional Christian orthodoxy to the more austere
terms of his rationally based ethical outlook, an apparent variation on the
deism of Kant’s time. Certainly there are many features of the book that
support this view. At the same time, however, Kant argues in Religion that
the “revealed” or “historical” side of religion can be morally useful, and
perhaps even “necessary,” as a symbolic or pictorial guide to moral improve-
ment – a “vehicle,” as he puts it, for the advancement of moral ends (R 6:106,
115, 118, 123n). Consequently, there are significant textual obstacles to inter-
preting Religion simply as reductionist with respect to the historical side of
Christianity.
Interpretive issues are made no easier by the fact that Kant’s text equiv-
ocates between obvious references to Christianity and references to “reli-
gion” in the more generic sense, implying a possible appeal to the historical
side of all religious traditions. From this standpoint, Kant appears to be
developing a theory of the historical evolution of historical faiths, with
Christianity simply farther along the trajectory of a pure “moral” faith than
other traditions. Indeed, this aspect of the book makes it a template for the
perennially attractive idea that, beneath all their differences, the world’s
religious traditions are linked by a deep commonality. In short, Kant
provides here a framework for distinguishing the “accidental” or historically
and culturally contingent features of a religious tradition from its deeper and
“real” meaning. This deeper meaning could conceivably be shared by varied
religious traditions, despite the stark differences across their historical and
scriptural claims – which is to say, their empirical aspects.
Consequently, Religion not only extends Kant’s thinking about the
relation between ethics and religion in his own thought, but also combines
commentary on obviously Christian biblical and theological topics with
commentary that could involve other religious traditions as well. Not
surprisingly, then, the aims of the book have been perennially difficult to
state in succinct terms.
As a gateway to these issues, Otfried Höffe’s “Holy scriptures within the
boundaries of mere reason: Kant’s reflections” offers a broad view of the
aims of Religion by situating Kant’s interpretive approach in the context of
the Critical philosophy as a whole. This background clarifies the overall
intentions behind the work while also offering insight into Kant’s specific
views of Christianity and the historical or empirical aspects of faith. Höffe
further connects Religion to Kant’s Conflict of the Faculties in which, five

years after the publication of Religion, he makes very explicit the difference
between a philosophical and a theological interpretive undertaking. In so
doing, Höffe argues, Kant sheds valuable light on the rational inquiry into
scripture and the historical elements of religion more generally.
Curiously, a book seemingly devoted to exploring the potential of our
rationality in fact opens with a sustained account of our deep failings.
Kant’s theory of “radical evil” is at once among the most vexing features of
the entire work and the issue that especially aroused Goethe’s ire. Unlike
his many Enlightenment counterparts who drew upon an underlying
Platonism to explain moral evil largely in terms of ignorance, Kant clearly
lodges responsibility for moral evil in the free will. This feature of his
view is challenging in and of itself, yet Kant compounds the difficulties
considerably by insisting on “the universality of the evil at issue,” to
the point where it is so “entwined with humanity itself” that we can say
that the human being is “evil by nature” (R 6:32, emphasis Kant’s). The
hint here of something like “natural necessity” would seem to be utterly
incompatible with Kant’s overall moral philosophy. Moreover, the latent
association with something very much like the Christian doctrine of
original sin – the heart of Goethe’s objection to Kant’s book – is of
course at complete odds with Kant’s thoroughgoing emphasis on human
autonomy. The matter is made no less complex by Kant’s appeal to “the
multitude of woeful examples” we have “through experience” of the deep
corruption of the human heart – as though Kant has suddenly invoked
the notion of a “noumenal eye” that can “see” into the moral qualities of
moral agents, in complete contradiction to his epistemology.
In “The evil in human nature,” Allen W.Wood clarifies Kant’s claims
that evil is “universal,” “innate,” and “inextirpable” by human powers.
Wood suggests that the discussion of radical evil as Kant’s starting point
may be no surprise at all, if we view Religion chiefly as an attempt to display
the mutually reinforcing features of Christianity and rational morality to
late eighteenth-century (largely Protestant) Christians. For such an audi-
ence, the acknowledgement of human sinfulness would be the altogether
familiar opening of any work of Christian apologetics. Similarly, in contrast
to thoroughgoing individualistic accounts of radical evil, Wood’s account
underscores the fact that Kant himself lodges the “propensity” to evil in
our social condition, though without diluting the element of personal
responsibility.
In “Radical evil and human freedom,” Ingolf Dalferth lays out the main
features of Kant’s argument by relating the specifics of his account of human
Introduction 5
nature to his theory of moral agency. This theory is largely driven by the
exercise of freedom in maxim-making, a capacity that distinguishes rational
beings by their ability to prescribe incentives to themselves opposed to those
given by nature. The terms by which moral evil becomes “radical” involve
the free choice of an underlying supreme maxim that, as a general pattern,
subordinates the incentive of duty to the incentive of self-love. While
careful to show that being evil “by nature” is not the same thing as being
evil by “natural necessity,” Dalferth locates the potentially compromising
implications that nonetheless arise from this position for Kant’s own theory
of freedom.
Due to his expanded notion of the moral “disposition” (Gesinnung) in his
effort to define an evil that is “radical,” Kant implicitly deepens his own
moral theory in ways that make Religion a crucial resource for the full
understanding of his ethics. In “Gesinnung: responsibility, moral worth,
and character,” Alison Hills examines in detail the supreme maxim account-
ing for our individual acts of maxim-making. She argues that, as the
“subjective ground” of the adoption of individual maxims, the disposition
implies a fuller and more explicit account than Kant had previously offered
of characteristic or persistent tendencies within the moral agent. Most
importantly, Hills shows how the free choice of an evil disposition accounts
for an evil that is all-pervasive and thus “radical” while also remaining
completely inexplicable. In other words, the expanded role in Religion of
the moral disposition enhances our sense of the “unity” of moral agency in
Kant while simultaneously deepening the ultimate mystery associated with
any act of freedom. Among other things, her analysis sheds light on how we
might conceive of a Kantian theory of moral “character” that is more than
simply the sum total of discrete acts of maxim-making.
Kant’s specific way of framing the idea of radical evil as a corruption of
the underlying “ground” of maxim-making exists side-by-side with the
ongoing obligation to make ourselves good again. Kant thus puts in ques-
tion the moral agent’s ability actually to meet the obligation to generate the
needed moral revolution. As he puts it, “if a human being is corrupt in the
very ground of his maxims, how can he possibly bring about this revolution
by his own forces and become a good human being on his own?” (R 6:47).
Kant’s follow-up to his own question involves challenging references to
divine “cooperation” and even “grace” – an idea seemingly antithetical to his
entire philosophical enterprise. It is in the context of these references to
divine aid in the recovery from radical evil that Kant introduces the
provocative – if epistemologically problematic – references to the idea of
“hope.” Religion considerably extends and clarifies the third of the three
questions reflecting the interests of reason that Kant originally introduced in

the Critique of Pure Reason some twelve years earlier: “What can I know?”
“What ought I to do?” and “What may I hope?” (A805/B833). Within
Religion, hope assumes fresh importance as Kant tries to work out the
terms for the regeneration of a moral agent compromised by an evil that
is radical, going to the very “roots” of human nature.
In “Rational hope, possibility, and divine action,” Andrew Chignell
reminds us that, against the background of Kant’s three questions, “hope”
and not “belief” is the most central concern of Kant’s religious thought. By
disentangling hope from belief, Chignell isolates a sort of “attitude” that, in
turn, sheds light on a Kantian position toward religious doctrines and
themes falling “outside” the limits of reason alone. Crucial here is the fact
that the kind of hope at stake is “rational” hope – as opposed, say, to wishful
thinking. Accordingly, Chignell explores in some depth the modal con-
straints structuring hope in the genuinely Kantian sense. This exploration,
in turn, illuminates in fresh ways the content and epistemological status of
the religious outlook, including the content of religious belief. Among other
things, Chignell’s account underscores the rationality of hope for social
progress embedded in Kant’s overall moral and religious thought, exempli-
fied by the idea of the ethical community in Religion. In thus endeavoring to
effect a reassessment of the balance between belief and hope in Kant,
Chignell broadens the scope of a religion “within the bounds of mere
reason” in surprising ways that go beyond more familiar readings of
Religion.
Such surprises continue in “Kant on grace,” in which Leslie Stevenson
points out that Kant creates a space for the discussion of grace by means of
attaching a “General Remark” to the end of each of the four parts of
Religion. Kant depicts the “General Remarks” as occasions to touch on
the “parerga to religion within the boundaries of pure reason; they do not
belong within it yet border on it” (R 6:52). In effect, Kant devises a means of
discussing matters falling on both sides of the “boundaries” announced in his
title. Stevenson tracks the various ways Kant includes grace among these
topics through explicit connection with a recovery from radical evil for
which finite beings seem to lack sufficient powers. Without minimizing the
difficulties Kant faces in doing so, Stevenson clarifies how Kant avoids
subverting his own theory of autonomy through a series of epistemological
limitations that allow references to grace but without establishing such
references as knowledge claims.
Similarly, in “Kant, miracles, and Religion, Parts One and Two,” Karl
Ameriks examines Kant’s treatment of miracles within the “General
Introduction 7
Remarks,” indicating how such an unexpected topic is in fact connected
with the theme of religious hope for Kant. Ameriks shows how Kant draws
on different ways of appealing to ideas that go beyond nature while
simultaneously criticizing superstitious references to miracles in other
parts of Religion. He argues that Kant’s ultimate aim is to replace references
to “miracles” (Wunder) with references to something more like an attitude
of proper “admiration” (Bewunderung) that accompanies the full commit-
ment to respect for the moral law. Ameriks suggests that the move to
Bewunderung clarifies the moral teleology at the heart of Kant’s view of
authentic religion, a teleological aim that functions simultaneously as a
Kantian theodicy.
In Kant’s day, the tensions between the universal claims of reason and the
particularities of revelation, scripture, and tradition found their clearest
expression in the figure of Jesus. While Kant never refers to Jesus by
name in Religion, the reference to him is clear in what Kant offers as the
“personified idea of the good principle” who, in his person, embodies the
rational ideal of a thoroughly good disposition, which is simultaneously a
moral disposition wholly “pleasing to God” (R 6:61). In short, Kant’s
Christology resides in the claim that Jesus is the prototype of moral
perfection, a prototype that is universally accessible through reason alone.
In “Kant’s Jesus,” Manfred Kuehn illustrates the lively nature of
Christological debates in Kant’s time, thus providing the historical back-
ground to Kant’s association of the historical figure of Jesus with this
universal rational content. By locating Kant somewhere between
Reimarus and Semler, Kuehn suggests that Kant is attempting something
of a balancing act with respect to the tensions inherent in the
Enlightenment debate about reason and revelation. The balance is struck
in the fact that, by embodying a moral perfection that is in principle
attainable by all rational beings, Kant’s Jesus remains central to a religion
within the boundaries of mere reason, though without being essential to it.
The truth Jesus embodies would still be true even if he had never lived.
Kant’s interest in Jesus is developed in Religion alongside an apparently
growing emphasis on the association of moral agents with one another.
Indeed, in Religion’s account of the ethical community ( gemeines Wesen) we
find what could be characterized as a refinement of Kant’s conception of a
realm or kingdom of ends in the Groundwork of the Metaphysics of Morals. In
contrast to the civic or political community, the ethical community is a
setting in which moral agents act out of respect for the moral law rather than
out of fear of legal repercussions or punishment. At times Kant speaks of
this ethical community in association with provocative remarks about both
the “visible” and the “invisible” church. At the very least, the ethical
community signals a shift toward a more social rather than purely individual
approach to the recovery from radical evil and the pursuit of our moral
destiny. The theme of the ethical community is also connected in important
ways with Kant’s arresting comment that “there is only one (true) religion;
but there can be several kinds of faith” (R 6:107), a remark that appears to be
an open invitation to a comparative approach to the world’s religions with a
view to their common moral core.
In “Pluralism in the ethical community,” Nicholas Tampio reminds us
that Kant’s own historical context was the Thirty Years War threatening the
stability of all of Europe. Tampio argues that, along with Leibniz and
Spinoza, Kant wants to establish a framework promoting the sort of ethical
and religious pluralism that would enable Europe to avoid such threats in
the future. By re-framing the issues in terms suggested by John Rawls’
account of justice, Tampio describes how Kant’s theory might mediate the
vexed question of which religious groups could or could not be included in
the ethical community. Tampio tests this aspect of his reformulation by
drawing from recent work on the secular state emerging from the Muslim
community.
In “Kant’s religious constructivism,” Pablo Muchnik deploys from a
different vantage point the connections between religion and politics as a
means of helping us understand the very term “religion” in Kant’s title. He
indicates how Kant would resolve the “antinomy” created by the conflicting
views of Richard Rorty and Nicholas Wolterstorff on the role of religion in
politics. Muchnik thus exploits the interplay between public and private in
contemporary thought as a means of clarifying the increasing role of shared
undertakings and common pursuits in Kant’s ethics. Properly recognizing
the social aspects of radical evil implies the need for a cooperative response,
including cooperation between religion and politics. Kant’s Religion can
thereby be viewed as the answer to the question, “What kind of religion can
promote rational and emancipatory ends?” Muchnik argues that the answer
embodied by Kant’s book is simultaneously Kant’s disclosure of a middle
ground between superstitious appeals to a transcendent view of God and the
denial of all claims presupposing God’s existence. In short, the proper
“political” reading of Religion is also the proper “religious” reading.
We have seen multiple instances of how Religion provides a more
sustained account of rational faith than Kant provides in the earlier
ethical writings, including the Groundwork and the second Critique. In
“What does his Religion contribute to Kant’s conception of practical
reason?”, G. Felicitas Munzel argues that this more sustained account finally
Introduction 9
entails a deepening of Kant’s overall theory of practical reason that carries us
beyond a reading of it in its purely formal and objective sense. For Munzel,
Religion’s aggressive contextualizing of rational faith within a broader dis-
cussion of human nature clarifies practical reason’s relation to itself, ulti-
mately suggesting profound connections among judgments of “ought,”
“can,” “hope,” and “do.” Munzel further argues that practical reason’s
relation to itself sheds significant light on the role of “conscience” in
Kant’s moral and religious thought. Obviously, a key implication of this
argument is the importance of Religion for a full understanding of Kant’s
overall philosophical project.
Similarly, in “Culture and the limits of practical reason in Kant’s
Religion,” Richard Velkley maintains that the very motif of “limits” or
“boundaries” suggests the centrality of Religion to the entire Critical enter-
prise. He argues that “boundary-drawing” is for Kant the highest legislation
of reason, exemplified by the way Kant’s strictures on theoretical cognition
dovetail with his account of the “interests” of practical reason. Kant’s genius
resides in his ability to enforce these strictures without undermining the
possibility of reason’s ends, a task accounting for the increasingly robust
teleological element in Kant’s later thought. Indeed, three years prior to the
publication of Religion, the Critique of Judgment displays reason’s interest in
an ultimate unity between nature and culture, yielding a “moral whole” as
the end of reason itself. In the face of the radical evil that is a threat to this
moral whole, Religion reveals the crucial role of rational religion in sustain-
ing reason’s interest, thereby conveying in a fresh way Kant’s famous claim
that morality “inevitably leads to religion” (R 6:6).
Given the richness of Kant’s Religion, this collection surely makes no
claims to comprehensiveness in the topics it covers. Still, the essays in this
volume highlight arresting interpretive issues raised by Kant’s remarkable
book in ways suggesting how this late work both fits into and amplifies his
philosophy as a whole. To the extent that it opens up problem areas left
unresolved within its pages, Religion is – at the very least – a telling indicator
of the creativity and restless energy of Kant’s efforts, even at such a late stage
in his life.
chapter 1
Holy Scriptures within the boundaries

of mere reason: Kant’s reflections
Otfried Höffe
1 Two Preliminary Remarks

The Enlightenment, as is well known, is not easily intimidated. Neither
religions nor their authoritative scriptures – however holy they might be –
can escape its critique. Once early Enlightenment writers and their predeces-
sors, such as Bacon, Hobbes, and Spinoza, had exposed many religious
notions as mere superstitions, at least four different types of critique came to
be widely adopted. Voltaire aimed his slogan “Ecrasez l’infâme,” directly at the
Church. According to David Hume’s treatise on religious psychology and
sociology, Natural History of Religion (1757), some peoples have no religion at
all, which is taken as evidence that religion is no anthropological constant.
Hence, religion is said to have no foundation in human reason or in human
emotions. According to d’Holbach such central religious dogmas as the
existence of God and the immortality of the soul are no more than illusions
(Système de la nature, 1770). Finally, although Rousseau’s position softens the
Enlightenment’s otherwise mostly negative attack on religion, he nevertheless
rejects any claim to truth by revealed religion and maintains that the only form
of religion that is to be advocated is the natural religion. In the “Profession of
Faith of a Savoyard Vicar” appearing in his educational novel, Emile, Rousseau
develops his concept of natural religion as the voice of the heart.
With the high point − but also the turning point − of the Enlightenment
with Kant, these four models are partially though quite incisively con-
trasted. Kant does not lay stress only on natural but also on revealed
religion. To be sure, this is not the discussion taking place in his three
Critiques, though the second one does take Christianity into account (KpV
5:23; 229ff.). Only Religion within the Boundaries of Mere Reason (= Religion)
sets out to defend its subject matter, a defense in which it also attends
explicitly to a holy scripture. The scripture in question is the Bible, though a
Bible that has been stripped of all elements of Jewish theocracy. Kant takes
up four theologoumena − that is, four basic principles − from this text which
10
Holy Scriptures within the boundaries 11
is thus reduced to its truly Christian elements. In a second step, he then
submits them to a detailed philosophical interpretation.
As is well known, philosophical theology plays a role in all three
Critiques, though, with the exception of the short passages mentioned
above, it does so while emancipating itself from any authoritative holy
scripture. The Religion text appears to reverse this emancipation. And one
might rightly ask whether in this it contradicts the three Critiques. Is there a
tension, perhaps even a divide, between Religion and the Critiques? Or is
Kant merely extending the Critiques’ theology, adding new elements to it?
This is my central question. Answering it does not, however, call for a
detailed or even a micro-analysis of small sections of the Kantian text, but
rather requires a much wider view. Put differently, we will need to peer
through a wide-angle lens. Before we proceed, though, I would like to make
two preliminary remarks, one pertaining to the historical context and the
other to the systematic place of the Religion within Kant’s oeuvre.
In contrast to his practice in the Critiques, in the Religion Kant does not
refer to any important philosophers. Nor does he refer to the thinkers of
Antiquity such as Plato, Aristotle, or Epicurus, all of whom would have been
pertinent, or to modern thinkers such as Descartes, Spinoza, or Pascal. There
is no reference even to the authors that I have quoted in the beginning of my
chapter, Voltaire, Hume, and d’Holbach. Admittedly, Kant does refer to two
of these philosophers, though not with respect to the philosophy of religion:
he writes of Rousseau but only as a moralist (R 6:20) and of Hobbes
concerning the theory of the state of nature as a state of war (R 6:97).
If we ask, nevertheless, how Kant relates himself to the history of
philosophy, the initial answer to that question will be negative, which
increases the originality of Kant’s work. Kant has no predecessors among
the great philosophers concerning the core of his work on religion: his
detailed philosophical interpretation of Christian teachings. In the end,
Kant primarily relates to his own work, namely his treatment of philosoph-
ical theology in the three Critiques.
He also deals with contemporary authors such as the influential doctor
and writer Albrecht von Haller, the philosophical writer Johann Kaspar
Lavater, Moses Mendelssohn, and especially with open-minded Protestant
theologians such as Johann Salomo Semler. This leading eighteenth-
century Protestant theologian established the Bible’s human and historical
origins, thereby rejecting the dogma of verbal inspiration. In the preface
to his main work, Zur Revision der kirchlichen Hermeneutik und Dogmatik
(1788), which appeared only a few years before Kant’s Religion, he
identifies reason and revelation as the two instances of a legitimate biblical
12 OTFRIED HÖFFE
exegesis. Kant sides with reason and learning, vehemently rejecting a third
possible instance, inner feeling (R 6:113ff.).
My second preliminary remark is this: already the content of its title, “. . .
within the boundaries of mere reason,” places the Religion within Kant’s
critical oeuvre. One could even surmise that the Religion might represent a
fourth Critique. One could presume that, after directing his critique toward
knowledge (the first critique), morality (the second), and esthetic as well as
teleological judgment (the third), Kant would here be directing his critique
toward religion.
The three Critiques, however, all investigate basic human abilities and
the strictly a-priori laws that underlie them. The Critique of Pure – and,
according to the third Critique, also “theoretical” – Reason is concerned with
“the sources of all a priori knowledge” and the laws of nature.
A religious studies scholar could easily imagine a similar elementary
faculty for religion. However, such a basic religious faculty is foreign to
Kant. In the “Introduction” (Section III) to the third Critique, he only
mentions three faculties: (1) the faculty of knowledge, for which “only the
understanding” can provide a-priori laws; (2) the “faculty of desire,” for
which “only [practical] reason” can provide such laws; and (3) the “inter-
mediary” faculty, the power of judgment (KU 5:176ff.). There is, thus, no
room left for a further faculty of the same rank, capable of generating a-
priori laws.
Kant’s basic philosophical and religious idea is also speaking against a
fourth Critique, namely, the idea that religion is obliged to morality. We
should therefore expect such a new work to elaborate further a theme that
Kant had already discussed: moral philosophy and its antecedent, philo-
sophical theology, which had been broached in the First, brought to the fore
in the Second, and pursued further in the Third, Critique. The Religion is
hence no fourth Critique. It is rather the first large publication in which
Kant turns to doctrinal philosophy after completing his three-part Critical
work, which is still a premise for the following work. What the Metaphysical
Foundations of Natural Science (1786) did (before the second Critique) for
nature, the Metaphysics of Morals (1797) after the Religion did for morality.
In the 1786 work as well as in the two-part work published in 1797 Kant
develops, on the basis of the Critiques, a possible purely rational, hence
“metaphysical,” philosophical doctrine. The Religion fulfills a transitional
task. It does not prepare the way for the Metaphysics of Morals, but rather
deals with topics that do not appear in it at all: evil, revelation, and the
fundamental elements of Christianity. That it is not the Religion but rather
the Metaphysics of Morals that is the doctrinal continuation of the critical
project is one more reason for my leading question: how does our text on
religion fit into Kant’s critical program?
We need to consider three aspects if we are to answer this leading
question. First, we must review the philosophical theology of the three
Critiques. It is a philosophy of religion that implicitly questions the impor-
tance of Holy Scriptures by not mentioning the topic at all. At the outset I
review the philosophy of religion that Kant had developed prior to the
Religion, for that is a philosophy of religion that seemingly rejects the
religious value of Holy Scripture through silence (section 2 below). This
context highlights the new elements in Religion, which can then be assessed
according to the leading question of whether they contradict or rather
expand upon the theology that can be found in the three Critiques (sections
3 and 4). This question also leads us to cast a look at the philosophical
hermeneutics of the Bible which Kant practices in Religion and first explic-
itly describes in the Conflict of the Faculties (section 5).
2 A look at the theology of the Three Critiques

Philosophical theology has traditionally been considered the highest disci-
pline of metaphysics. Its fundamental concept, God, understood as the
absolutely highest being, is considered the highest element of all knowledge.
Since Xenophanes, Plato, and Aristotle, the great philosophers have striven
to shed light upon the nature of this being by means of natural reason. They
have developed purely philosophical theories of religion that are uncon-
strained by any holy text and which usually culminate in a form of mono-
theism that is both historically and systematically independent of any of the
three religions of the Book.
Though the concept of God does appear in Kant’s texts during his pre-
critical period, it only takes its decisive turn three and a half decades later, in
the Critique of Pure Reason. The philosophical theology in this work could
perhaps even be called a revolution, and certainly it represents a paradigm
shift – a three-part shift.
First, all attempts to know God objectively, especially attempts to prove
his existence, are rejected. God remains the highest aim of thought, but
neither his existence nor his non-existence can be refuted. And so, second,
the objectively knowable transcendental idea of God is replaced with a
subjective thought, a transcendental ideal. As the principle of complete
cognition, God is not only a possible but, for a theory of experience, even a
necessary concept, one “which concludes and crowns the whole of human
cognition” (B669). But this God has little – actually nothing – to do with
14 OTFRIED HÖFFE
the God of religion, a God who listens or talks, who punishes or is graceful,
who comforts, advises, and speaks to your conscience. This theology knows
nothing of God as a partner in human dialogue or deeds. Finally, Kant’s
first Critique prepares the way for a theology of reason understood as a
moral theology – in agreement with the moral understanding of the
Enlightenment.
Much follows from this tripartite paradigm shift. For one thing, stories
about the creation of the world play a significant role in Judaism and
Christianity, as well as in other religions. Following in the steps of the
first three Critiques, in the Religion Kant does not reject the idea of the
supreme being as “Creator and Lord” of the earth (R 6:78). This thought,
however, appears in a moral context, and hence no ideas about creation can
rightly claim to be knowledge.
Second, since God can be thought without contradiction but cannot be
known objectively, the question of God should not be asked of theoretical
but rather of pure practical reason. Kant had already sketched the basic
elements of such a moral philosophy in the Critique of Pure [theoretical]
Reason (B832ff.) prior to the Critique of Practical Reason and the text on
religion, introducing the idea of a moral faith that is opposed to a purely
doctrinal one (B855).
Up to this point Kant remains in agreement with d’Holbach, though
only in part. The French Enlightenment thinker is right inasmuch as he
denies that God and the immortal soul can be known objectively. But it
does not follow that they are mere illusions. No longer matters of knowl-
edge, they become objects of philosophically grounded hope, the postulates
of pure practical reason that the second Critique investigates.
And because the third Critique itself strengthens the primacy of morality
as well as the moral postulate of God’s existence, we are brought to the
following heterodox, even heretical, reading: the three Critiques do not only
deal with a moral world, a creator of that world, and a special way of
knowing − these considerations actually represent the high mark of the
Critical philosophy. All three of the Critiques culminate in a philosophy of
religion (if only a purely natural one) that is not inspired by any holy
scriptures. So it is only reason and not canonical texts that can be said to
be holy, and even then the term should be placed in quotation marks. Kant
himself speaks of “most holy doctrines of reason” (R 6:83). One could also
call the corresponding frame of mind “holy”: the sapere aude or intellectual
courage by which one uses one’s own reason.
Despite this three-fold commonality, Kant’s treatment of religion does
evolve over the course of the three Critiques. Except for a footnote in the
first preface (Axi), the first Critique does not speak of religion, but only of
theology and God. In the second Critique, Kant speaks of religion when
dealing with the postulate of God’s existence, and defines it as the “recog-
nition of all duties as divine commands” (KpV 5:129). The third Critique
refers to religion on a number of occasions: in the theory of the (dynamic)
sublime (KU 5:260ff.), concerning the judgment of taste (KU 5:263ff.),
when discussing “the utility of the moral argument” (KU 5:459), and in
the “General Remark on Teleology” where Kant echoes the definition
offered in the second Critique, namely that religion is “the recognition of
our duties as divine commands” (KU 5:477; KpV 5:129).
3 New additions made by Religion Within

the Boundaries of Mere Reason
One could assume that the Religion simply extends Kant’s pure philosoph-
ical theology.1 This assumption would not be wrong, though again only
half-right. For the text remains true to the moral interpretation of theology
and religion, the latter being deemed neither irrational nor contrary to
reason. But Kant is no longer concerned only with religion as such.
Indeed, he extends this topic, which usually constitutes the general part of
a philosophy of religion, with a special part. He takes a closer look at one
specific religion, thereby adding a new element to the debate, with respect
both to its contents and to its methodology.
The content deals with the four building blocks of Christianity: original
sin, Christ, judgment day, and the Church. As far as methodology is
concerned, he considers the idea of a supernatural revelation, whereby
Holy Scripture − that is, a fixed authoritative text − comes into play. In
this Kant appears to abandon the three Critiques’ standpoint of a purely
natural theology free of any form of revelation. His additions are indeed
quite refined: pure natural theology looks beyond its boundaries and allows
revelation to tell which topics should be covered by a philosophy of religion
that would be true not only to philosophy but also to religion.
Here we need to ask what importance Christianity has for Kant’s phi-
losophy. Try the following thought experiment: assume that Kant had
known of neither Christianity nor any other religion, especially not of any
revealed religion. How would this have affected his thought?
1
See also Otfried Höffe (ed.), Immanuel Kant. Die Religion innerhalb der Grenzen der blossen Vernunft
(Berlin: Akademie Verlag, 2011), esp. Chs. 1 and 12.
16 OTFRIED HÖFFE
The answer is clear: Kant’s moral philosophy, that of the three Critiques
as well as the additions made by the Religion, would hardly have been any
different. His main thesis especially would have remained unaltered.
According to the Religion, morality “inevitably leads to religion,” where
religion is defined as “the idea of a mighty moral lawgiver outside the
human being” (R 6:6). Moreover, this definition corresponds to that
given in the second Critique, which was taken up in turn by the third.
Admittedly, had Christianity not existed, there would have been no
reason to examine the (moral-)philosophical value of its basic doctrines.
Kant had already dealt with the first basic doctrine’s main element, the
concept of evil, in the second Critique, in the section concerning the
“concept of an object of pure practical reason” (KpV 5:57). Simply examin-
ing this concept from a (moral-)anthropological point of view – as the first
part of the Religion does – leads to Kant’s theory of radical evil. The
Christian idea of original sin is not required.
This is followed almost seamlessly by the second part concerning the
battle of the good against the evil principle. None of the ideas introduced
here require Christian precedents – not the concept of virtue originating in
Stoic philosophy and not the idea of a “Son of God,” to be understood as an
“ideal of humanity pleasing to God” (R 6:60ff.), not even Kant’s philo-
sophical Christology. The Critique of Pure Reason already included the
concept of an idea in individuo and also that of a “divine human being in
us,” which here was the ideal of the Stoic sage (B596ff.). (However, the fact
that the Religion critiques the Stoa’s concept of virtue as inadequate
(R 6:57ff.) does mitigate attempts to equate this ideal with that of the
Stoic sage.) Neither the thought of a Son of God nor that of an inner divine
human being disrupts the propensity to evil, the central concept of Kant’s
text on religion. Neither do they affect the text’s eschatology nor the
concept of an invisible church, both of which are closely linked to Kant’s
doctrine of the postulates.
The propensity to evil and the three other main themes of Kant’s text on
religion all deal with basic human themes, in which he definitively rejects
Hume’s thesis, maintaining that there are, after all, anthropological ele-
ments, perhaps even anthropological constants, to be included in a philos-
ophy of religion. Moreover, we must consider the methodological and
systematic reasons for including these four elements (and only them) in a
religious anthropology. In any case, Kant does not privilege Christianity,
not even in the socio-cultural sense in that it was part of his social back-
ground. The only religion that is privileged is that which represents a
counterpart to autonomous morality. Accordingly, to Kant, this one true
religion (see, e.g., R 6:107ff.) remains faithful to the concept that had been
developed in the second Critique, and which was there, as well as here,
independent of Christian or any other faith. However, Christianity alone
contains the “seed of true religious faith” (R 6:131).
4 Eight particularities
Even a superficial reading of Kant’s text on religion reveals eight
particularities.
First, the Holy Scripture is considered as revelation and as holy because of
its purely divine origin. It is not holy in a weakened, comparative sense. As
concerns a merely relative form of holiness, that is as concerns secular
canonical texts, I see three possibilities that can overlap and that contain
no higher form of holiness. First, a literary culture will esteem a text for its
exceptional artistic qualities. This would apply in Greece to Homer’s epics,
to Hesiod, and to the tragedies. Second, a culture can see a text as providing
superior wisdom of life (in China this would apply to Confucius’ texts and
to the Dao De Jing). Finally, a community can consider a text to be
indispensable and mostly fixed, as is the case for the US Constitution.
A revealed text is fundamentally different from these and other examples in
that it comes from God, the one perfectly holy being. This author’s holiness
is then often carried over to the text itself, which in turn appears to be
absolutely holy – a provocative, even scandalous thought for reason and its
custodian, philosophy.
Kant takes this provocation seriously. Without taking the biblical reve-
lation to be impossible (or possible) though certainly valuable, Kant – and
this is the second particularity of the text – shows in the Religion how the four
basic Christian ideas can be established through philosophy alone, without
recourse to the Bible. This presupposes that we do not only appeal to the
principles of morality as the three Critiques did.
Kant retains these principles without any restrictions whatsoever, thereby
remaining true to the Critiques. However, he also makes use of an elemen-
tary experience that plays no role in the three Critiques – and this is a third
particularity. This experience is that of “a human nature partly laden with
good dispositions and partly with evil ones” (R 6:11).
Our text here holds on to a basic idea of the European Enlightenment,
namely that religion, of which there can be only one true instance, cannot
contradict reason. Despite its holiness, this religion must submit to critique
(Axi), whereby reason proves to be the stronger power in the long run since,
as the fully confident critical philosopher Kant declares, “a religion that
18 OTFRIED HÖFFE
thoughtlessly declares war on reason will not in the end prevail” (Bxviiiff.).
Religion must therefore submit to reason. On the other hand, Kant allows
the possibility that the doctrines of this religion might stem from “divinely
inspired men” (S 7:6). These doctrines, though initially made known
through revelation, are subsequently verified by reason. The consequence
is momentous, which is apparent from two different perspectives. As far as
its content is concerned, the one true religion, moral religion, does not
require historical revelation. Moreover, one can be religious without believ-
ing in the revelation and without sharing the Credo of a visible Church.
From a historical point of view, though, the true religion might well begin
with a revelation.
Since, according to Kant, reason cannot immediately refute the Christian
revelation’s claim to truth, he assumes that the two might be compatible.
This is a fourth particularity. Kant is able to provide new philosophically and
theologically impressive interpretations of biblical stories without abandon-
ing his earlier opposition between moral and merely doctrinal faith, and
following the hypothesis that philosophical and Christian (that is, biblical)
theologies agree. By understanding the Bible’s basic message as moral
propositions that pertain to the basic human experience of a partly good
and partly evil human nature, he thus dissolves the usual contradiction
between natural and revealed religion. In a paradoxical turn, the Christian
religion becomes a natural as well as revealed religion − that is, a religion
which “human beings could and ought to have arrived at on their own
through the mere use of their reason” (B233).
Two parallel reasons speak to the primacy of Christianity for Kant. On
the one hand, the basic Christian doctrines can be given an appropriate
philosophical and moral interpretation (i.e., through reason alone). On the
other hand, the basic ideas of moral philosophy can be found to a great
extent in Christianity. That is because the essence of Christianity lies, for
Kant, in a religion of pure practical reason, whereby he places religion, torn
to shreds in pre-Kantian Enlightenment, back in its rightful place, though
only in this form. The basic Christian doctrines are interpreted through a
moral-philosophical constellation: the opposition between good and evil
that can be described independently of any historical form as the expression
of the moral self-understanding of autonomous reason.
In order to rehabilitate religion in the form of Christianity, Kant dis-
cusses the four main Christian doctrines mentioned above: original sin (Part
One), Christology (Part Two), eschatology, or the teaching concerning the
last things (Part Three), and finally – and in opposition to Voltaire – the
doctrine of a churchly community, ecclesiology (Part Four).
In order to make this rehabilitation possible, Kant removes the “mystical
cover” from the biblical stories (R 6:83). In this process, these stories lose
their historical character, which is to say they are dis-individualized and un-
historicized. Hence, Kant does not speak of the historical person Jesus of
Nazareth, though he does speak of a Wisdom that surpasses the Stoic’s
worldly wisdom (R 6:79ff.). Kant denies also that the person living the most
pure form of morality is necessarily unique. He also avoids the terms
“Satan” and “devil”; instead he refers to an “evil spirit” and the “Prince of
this world.” They do appear in a footnote (R 6:79) but only as a quote.
After this de-mystification and de-historicizing there remains a core
based on morality and freedom which alone can lay claim to reason, thus
remaining valid throughout the world and throughout time (R 6:82).
Whereas the Bible addresses intelligible moral relations through stories
with concrete historical names, the philosophical interpretation replaces
them with concepts that are indispensable for a universal religion of
morality and reason.
The first Christian doctrine, original sin, when modified accordingly,
corresponds to the idea of radical, deep-rooted evil. In the second basic
doctrine, in Christ as the “Son of God,” Kant sees “Humanity (rational
being in general as pertaining to the world) in its full moral perfection”
(R 6:60; emphasis Kant’s): the evil principle is not completely eradicated
through the example of perfect morality, but its power is broken.
The third basic Christian doctrine, eschatology, is dealt with under the title
“The victory of the good principle over the evil principle, and the founding of a
kingdom of God on earth.” It encourages human beings to abandon that
ethical state of nature in which (as in the legal, “juridical” state of nature)
prevails a war of all against all, described as a state “in which the good principle,
which resides in each human being, is incessantly attacked by evil” (R 6:96).
This state is overcome by a community in which, contrary to coercive law, the
moral laws are freely accepted. It is not the general will, the nation, that passes
these laws, but someone “with respect to whom all true duties . . . must be
represented as at the same time his commands.” That is of course God,
understood as a “moral legislator.” Kant’s thesis is that it is for this reason
that morality inevitably leads to religion (R 6:6), and that an ethical commun-
ity “is conceivable only as a people under divine commands” (R 6:99).
Unlike a civil community, an ethical community is not a particular, but
rather a universal, unit, that includes all human beings. Contrary to a
possible world republic, it does not have, as a global republic of republics,
only secondarily a universal character, but is from the outset strictly a priori.
The reason such a community is based upon a pure religious faith is that it is
20 OTFRIED HÖFFE
a mere faith of reason which can convince everybody. By contrast, a faith

that relies on historical facts is limited. It is spatially and temporally limited,
since it is believable only as far as the news about these facts is passed on
(R 6:102ff.). A religious faith that is based on reason has the apparent
advantage that it dissolves the polemogenic potential between the denomi-
nations and religions known from history. Thus the historical faith in
revelation has only a secondary and subsidiary right. It is only required
inasmuch as human beings, according to Kant, usually require some kind
of impetus to be moral. As it is only required for didactic purposes, the
non-rational portion of revelation can only be considered a vehicle which
is – certainly today, perhaps already during Kant’s times – negatively
connoted. It is an old-fashioned and thus not particularly suitable means
of transportation.
As this unity is constituted through virtue – an inner quality that cannot
be the object of any possible experience – the corresponding community
cannot itself be experienced either. The Kingdom of God is thus a peculiar
kind of social unit, an organization that rejects any form of (visible)
organization. It is an invisible church. Nevertheless, its characteristics are
the same as those confessed in the Christian Credo. Qua being the people of
God, the invisible church is universal; as relying only on reason, it is
numerical. It is holy because it is defined as a moral community of integrity
and moral purity. It is apostolic since its constitution, its moral law giving, is
immutable. Kant does not, however, reject all forms of visible organization.
He assigns a moral-pedagogical task to the visible Church, which, because
of the “weakness of human nature” – namely its “impotence in the cogni-
tion of supersensible things” (R 6:103) – must provide a visible representa-
tion of the moral idea of a Kingdom of God.
Finally, Kant distinguishes the moral religion of a good conduct of life –
similar to Rousseau (Contrat social 4:8) – from all religions that merely seek
benevolence, the mere cult through statutes and ceremonies. All opportun-
istic religious intentions departing from a moral attitude characterizing pure
religious faith, and that expect divine good will through the grace of God,
are to be rejected because they contradict the principle of autonomy.
Kant thus discovers the basic elements of moral philosophy in
Christianity simply because he strips the Christian elements of their specif-
ically Christian character so that only an autonomous morality remains.
This remainder takes its inspiration from that basic anthropological expe-
rience, the propensity to evil. In the first part of the Religion, the Christian
dogma of original sin, the substantive understanding of evil as an inherited
characteristic of the human race, is transformed into the propensity to evil.
In the second part, the dogma of Jesus as the Son of God becomes the idea
of the good principle. Against Voltaire, Kant pleads for a visible religious
community, for a perceptible Kingdom of God, a Church. But he does
require, in the third part, that this church be cleansed of “the nonsense of
superstition and the madness of enthusiasm” (R 6:101).
In the practical conclusions in the fourth part, Kant rejects what he calls
the “counterfeit service” of the Church, the primacy of revealed faith, and its
connection to fixed statutes. Through its at times pointed critique of the
state’s religious policies, the Religion also proves to be a political manifesto –
and this is its fifth particularity.
It is not only the last part that is political in nature. The political aspect is
apparent in a new “philosophical theology” that differs in many ways from a
scripture-based “biblical theology.” Consistent with the third particularity,
this philosophical theology does not limit itself to the Bible, but takes into
account the “history, languages, the books of all peoples.” Here Kant refers
almost en passant to something he will – two years later in On Perpetual
Peace (1795) – call a task on its own, a cosmopolitan right. The sixth
particularity of Kant’s text on religion consists in a new perspective, a
cosmopolitan perspective that comes in handy in times of globalization.
In other words, Kant offers a doctrine of God that complies with every
intercultural discourse concerning religion: a truly cosmopolitan theology.
This position is characterized by intellectual openness and tolerance.
These two features enable Kant, for example, to counter contempt toward
other religions. Against the hubris of a culture that prides itself on its
monotheism (such as that of the Jews), Kant points out that most peoples
believe after all – despite their polytheism – in a universal ruler of the world.
They flank him, or her, however, with powerful subservient gods.
On the other hand, Kant’s text on religion is not only helpful in
promoting intercultural discourse, but also provoking. It helps that accord-
ing to its definition of religion there is only one religion. This one religion is
independent of any confession or Credo, which follows from autonomous
morality. Admittedly, it would be interesting to know whether there are
elements in other non-Christian religions that are analogous to the four
Christian doctrines discussed here (which Kant calls faiths: R 6:108). Since
the basic experience of what Kant calls a propensity to evil is not exclusive to
the Christian world, similar elements should exist outside its borders. If this
were not the case, two rather “inelegant,” both radical and far-reaching,
options would arise. If this anthropological basis were indeed lacking, Kant
would have to significantly restrict his book’s title: “Religion within the
boundaries of merely Christian reason.” But such a restriction would only
22 OTFRIED HÖFFE
indicate a failure on Kant’s part. Yet, if Kant is right, then the religions that
are lacking some or all of these elements must be deficient – a claim that
would immediately be countered by accusations of cultural arrogance.
Kant develops a new philosophical theology and demands “complete
freedom.” He thereby rejects all forms of dogmatic paternalism, and
is tacitly working toward a science of religion that explores foreign
religions indifferent to the claim to absoluteness of Christianity – as
history of religion since David Hume and (though in a different way)
Johann Gottfried Herder, as psychology of religion since Friedrich
Schleiermacher, and later as well as ethnology of religion.
5 A philosophical hermeneutics of the Bible

Kant himself practices this new kind of theology as a philosopher and not as
an empirical researcher. That is why he did not undertake to study all the
languages, books, and histories relevant to religion. He approaches it with a
special kind of probing, with “mere reason.”
Because he is using this probe called “reason,” he is also following the
program set forth by the three Critiques. This probing is qualified by a
“mere.” Some people even like to understand it as “pure,” as in “pure
reason.” However, when a philosopher as conceptually precise as Kant is
involved, one should ask whether “mere” is really used as synonymous with
“pure.”
According to the Deutsches Wörterbuch (II, 144–50), possible meanings of
“mere” (in German bloß) are “unclothed, naked,” plus “without weapons or
armor,” “desolate, bare, empty,” “simple, plain, ordinary,” as well as “bare,
needy, poor” and also “only, alone.” Without a question, both “pure” and
“mere” refer to things that are free of foreign elements. Their perspective
and connotation, however, differ. The word “pure” is exclusively positive; it
suggests “unmixed” and “neither contaminated nor diluted.” “Mere,” how-
ever, also connotes a deficiency. No matter whether clothes, weapons, or
foliage are concerned – “mere” indicates that something is missing; the
word encompasses a privatio, that can even become “bare, needy, poor.”
Following this interpretation, we see that the “mere reason” in the book’s
title points to a reason that is free of non-reasonable elements. This free-
dom, though, is not an entirely positive feature. Rather – and this is the
seventh particularity of Kant’s text on religion – it indicates a defect. It
indicates that the subject matter in question, religion, cannot be reduced to
reason. However, the philosopher confines herself to her profession and –
already for reasons of competence – investigates religion from the
standpoint of reason. It can also be argued that in matters of religion the
faith of reason has absolute precedence over faith in revelation.
Later, in the “preface” to the Conflict of the Faculties, Kant emphasizes that
his title does not read “Religion from mere reason (without revelation),” which
would have been presumptuous. He is rather concerned with that part of “the
religion believed to be revealed” that can be known through both a philosophy
that has been purged of all revelation and a reason without revelation (S 7:6;
see also the “Preparatory work to the text on religion,” 23:91). Kant also rejects
any “disciplinary imperialism” which claims that one’s own discipline is able
comprehensively to describe the subject matter in question.
For this reason the title, “Religion within the boundaries of mere reason,”
is both modest and immodest. It is immodest insofar as it drags religion,
despite its holiness, in front of the judge’s seat. It is modest because it allows
for religious elements that are inaccessible to reason. Reason’s perspective
on religion can be complemented, as Kant notes en passant, by a perspective
from outside of reason – a perspective that, nevertheless, is not in itself
unreasonable. Accordingly, Kant tacitly acknowledges that biblical texts
have a moral meaning, but perhaps not only a moral meaning. From the
standpoint of pure religious faith, the mere rational point of view has
absolute precedence.
Consequently, the philosophical, and at the same time highest, principle
of all exegesis corresponds to the actual goal of any rational faith − namely,
“the moral improvement of human beings.” However important the exe-
gete might be, he remains subservient to this goal. For scripture – Kant
means the Bible – is regarded as the “most worthy” and, Kant humbly adds,
it is the only means “for Europe” to unite all people in one Church.
(However, he does significantly qualify his humility by speaking of
Europe as the “most enlightened part of the world.”) The Church in turn
is necessary for people in general, who do not seem to be satisfied by
doctrines founded on mere reason (R 6:111ff.).
For this very reason, which is both anthropological and moral-
pedagogical, it is the biblical − and not just any − revelation that is given
so much importance. Kant concedes the possibility of non-natural insight,
the possibility of revelation, but he also expects anthropological and moral-
pedagogical insight from it. This provides reason with a guideline regarding
its boundaries, its limits. The nature of these boundaries is, however, not all
that easy to define: according to the stronger definition, the guideline
stresses insights that reason cannot produce itself but only re-construct –
taking re in the strict sense. Can reason only catch up with but never surpass
these insights, then?
24 OTFRIED HÖFFE
This conclusion would contradict the three Critiques since natural,

unrevealed religion would no longer suffice for their moral theologies.
Moreover, reason itself would be limited. Not least, this would cast doubt
upon Religion’s main thesis that morality leads to religion but does not
require it conceptually. Finally, moral theology would be socially and
culturally dependent upon Christianity, which would run counter to the
interpretation offered above.
In fact, a much more restricted definition suffices: reason does not need
revelation as an absolute guideline. It only requires revelation “for the
people” who lack the ability or the willingness to improve themselves for
the sake of mere reason alone, and who need instead the help of a visible
church.
Given even this restriction, revelation still does not constitute a simple
guideline. For revelation can only be thought of as a part of religion through
reason (R 6:12). Reason does not look at all of revelation, but rather operates a
double selection. On the one hand, it only chooses truly anthropological
elements. In Kant’s universal-human philosophy of religion, all historical or
quasi-historical elements are omitted – in the Old Testament, for instance, the
stories of Noah, of Joseph, and of the exodus from Egypt, the time of the
Judges and the time of the Kings as well as the stories of Ruth, Ezra, Judith,
and Esther. Above all, one basic element is omitted: the covenant between a
God and his people, that is, the exceptional relationship between Yahweh and
Israel. This selectivity is ambivalent. On the one hand, reason accepts a
guideline that is “pre-reasonable” and stems from beyond reason. On the
other hand, it lays down conditions: it establishes a criterion of anthropolog-
ical and cosmopolitan relevance. This criterion functions as a probe, searching
for the suitable elements. This procedure results, of course, in a significant
restriction. Only a fraction of the revelation finds grace in the eyes of reason.
Even though Kant does take the Judeo-Christian revelation into account,
enabling him to move beyond previous philosophical conceptions of God,
his concept nevertheless remains a “God of the philosophers” inasmuch as it
methodologically excludes as a historical element the “God of Abraham,
Isaac, and Jacob.” One could, however, claim that Kant does break out of
this methodology. In excluding the basic historical element of the Old
Testament, namely Yahweh’s covenant with Israel, he also recognizes the
basic historical element of Christianity, Jesus the Christ. Such an interpre-
tation would, however, belie Kant’s text. Just as Kant’s rational religion
discards the covenant with Yahweh because it is a historical event, so he
also denies the uniqueness and exclusivity of the “Son of God.” It is the
de-historicized element, the moral perfection, that is of concern to him.
His selection within revelation goes even further. For Kant does not
regard all universal-human elements, but only those that are genuinely
moral, thereby not rejecting the idea of humans being created in the
image of God. Interpreted from a secular point of view as the faculty to
speak and reason, this idea is clearly a precondition of the human capacity to
be moral.
In the first part of the Conflict of the Faculties Kant explains the herme-
neutics that he practices in the Religion. Critical exegesis of the Bible is
nothing new in Kant’s time. If the philosopher confines himself to his own
profession, he will think of at least two important representatives.
A son of an uneducated theologian became a notable Bible scholar, who –
without even studying theology himself – exegetes the Bible critically
rejecting all kinds of beliefs in miracles. I refer to Thomas Hobbes, author
of Leviathan.
The second philosopher, who needs be included in any history of critical
exegesis, if only because he showed the way to a critical biblical hermeneutics,
was Baruch de Spinoza whose family was forcefully converted to Christianity.
In the preface to his Theological-Political Treatise, we find the following
hermeneutical principle which remains valid to this day: “to examine scrip-
ture afresh, conscientiously and freely, and to admit nothing as its teaching
which I did not derive most clearly from it.” This independent hermeneutics
leads Spinoza to two views that anticipate Kant’s own. For one, the religion
revealed in the Holy Scriptures has a purely functional meaning, inasmuch as
it accommodates the intellectual abilities of its various addressees. For
another, this consists very simply in the fact that the scripture teaches
moral rules of life, especially that of neighborly love that aims to extinguish
hatred and conflict among human beings (Chs. 13–14).
Both pioneers of a critical hermeneutics, Hobbes and Spinoza, made way
not only for a critical hermeneutics of the Bible, but also for that seventh
particularity of a rational and critical hermeneutics in which biblical exegesis
is restricted to pure practical reason. Here Kant represents the philosophical
high mark. This kind of hermeneutics is practiced in Religion, and it is
explained in the first part of the Conflict of the Faculties.
This later text deals with the conflictual relationship among the three
higher university faculties of theology, law, and medicine and the lower
faculty of philosophy. Kant is here concerned with the antagonism between
the empirical and the a priori within the university. This issue is linked to
the opposition between heteronomy and autonomy, though not the moral
and illegitimate opposition between a heteronomous and autonomous will.
Kant does not view the former opposition as entirely illegitimate. This
26 OTFRIED HÖFFE
opposition is characterized by a government that directs the interests of the

people through external rules versus a philosophical pursuit of the truth that
has emancipated itself from any subordination, heteronomy, or external
control of its thinking.
One of the fundamental ideas of critical hermeneutics is found in Kant’s
writing as well: the idea of the multiple meanings of scripture. From a
philosophical point of view, this multiplicity is clear and narrowly restricted.
For Kant the multiple is a dual. Following his basic conceptual distinction,
Kant differentiates a pure moral religious faith (serving moral improvement
in an exclusively rational, and thus in a supra-individual as well as supra-
temporal, way) from historical faith. The latter “can indeed be useful to
religious faith as its mere sensible vehicle (for certain people and certain
eras), it is not an essential part of religious faith” (S 7:37). The idea of a
vehicle has of course already been mentioned at several points in the Religion
(R 6:106, 118, 123).
To this hierarchical assessment, Kant adds the distinction between κατ’
άνθρωπον and κατ’ άλήθειαν. In the first case, historical faith follows the “way
of thinking of the apostles’ times.” This includes the teaching method that the
apostles inherited themselves, which significantly implies that historical faith
“cannot be taken as divine revelation.” Only the second option, rational
religious faith, contains the truth, namely “doctrine itself” (S 7:37).
The conflict of the faculties naturally follows from this opposition. The
higher faculty of theology suspects the lower faculty “of philosophizing
away all the teachings that must be considered real revelation and so taken
literally, and of ascribing to them whatever sense suits it.” Conversely, the
philosophical faculty suspects that, when theology considers the faith of the
church, it “loses sight of the final end, inner religion, which must be moral
and based on reason.” Beholden to the truth and nothing but the truth,
“when conflict arises about the sense of a scriptural text,” philosophy claims
the “prerogative” to determine this sense. For Kant, this is a natural and
legitimate conflict. Understandably, the theological faculty looks to its task
from the point of view of the Church’s faith, whereas the philosophical
faculty looks to rational religious faith from the perspective of its own task.
It is within this conceptual framework that Kant lays out his “philosoph-
ical principles of scriptural exegesis.” Against the tacit suspicion that phi-
losophy might claim too many rights for itself, he explains that these
principles, though described as “philosophical,” perform no specifically
philosophical task that would be foreign to the texts. They do not demand
that exegesis be “aimed at contributing to philosophy.” Kant again rejects
any philosophical imperialism here with respect to content. He also rejects a
methodological imperialism since he does not restrict his understanding of
philosophy to its strictly academic sense. Within the context of his kat’
alētheian-exegesis, he mentions two types of critical hermeneutics, a
historical-critical and a grammatical-critical exegesis, neither of which are
philosophical in nature. And concerning them, he explains that it is not
these forms of exegesis themselves, but rather their principles, that “must
always be dictated by reason” (S 7:38).
Concerning the methodological assessment, we must also take into account
that the (broadly) philosophical exegesis functions as an “idea” (S 7:44), that
is, as a concept of reason, not of the understanding. As a consequence, these
philosophical principles cannot be empirically legitimated nor confirmed
through experience: their legitimacy lies in the idea of a rational reading of
the biblical texts. Their historical particularities are not considered.
An “idea” in Kant’s technical sense is usually too exuberant for theoretical
reason, but it does find a reality in practical reason. Hence Kant’s argument
moves in a circle, and we need to check whether it is a vicious one.
Assuming that one is looking in the Bible to find reason, one will find it,
though it will only be the pure practical, hence moral, reason one was
searching for in the first place. This moral reason, though, is only brought
about by us − namely, by our expectation to find it there. For only those
whose philosophical hermeneutics leads them to search for morality in the
Bible will find it and discover what they were looking for.
Kant establishes four hermeneutical principles. The first principle which
is elaborated most extensively concerns the difference between theoretical
and practical doctrines. Here, “theoretical” does not mean the principles of
a dogmatics that are applied “in practice.” “Practical,” on the contrary,
means moral-practical, so that “theoretical” can be negatively defined, then,
as not morally relevant, and positively as concerning questions of knowl-
edge and cognition. Biblical statements can either transcend “all rational
concepts (even moral ones)” or even contradict practical reason. When they
transcend, they can be interpreted in line with practical reason – Kant gives
the example of the doctrine of Trinity. When they contradict, they must be
interpreted in line with practical reason, for example in Paul’s teaching
concerning predestination which “cannot be reconciled with [reason’s] own
teachings on freedom . . . and so with the whole of morality” (S 7:41).
The other hermeneutical principles follow the same keynote: “the only
thing that matters in religion is deeds.” This “final aim,” according to the
second principle, “must be attributed to every Biblical dogma” (S 7:42). The
third principle requires that action “must be represented as issuing from
the human being’s own use of his moral powers.” The fourth principle,
28 OTFRIED HÖFFE
however, allows one exception: “If the human being’s own deeds are not
sufficient to justify him before his conscience (and it judges him strictly),
reason is entitled to adopt on faith a supernatural supplement to fill what is
lacking to his justification” (S 7:44).
With his characteristic thoroughness, Kant notes objections which he can
see coming both from (biblical) theology and from reason.
According to the first objection, a philosophical hermeneutics aims at “a
natural religion and not Christianity” (S 7:44). Were the objection to hold,
Kant would have bypassed his topic. The philosopher, however, diffuses the
objection by distinguishing between a naturalistic religion that “makes it
a principle not to admit supernatural revelation,” and a natural religion
that allows for the possibility of revelation. Without denying that the Bible
is a supernatural means of founding a confessing Church, natural religion
simply does not take this into account. It brackets revelation and concen-
trates on its core, rational religion. This approach strengthens the above-
mentioned, more modest, meaning that rational religion does not
necessarily need revelation.
A further objection claims that the philosophical faculty is interfering in
“the biblical theologian’s business.” Kant concedes that there is “intrusion”
but denies any “interference.” From the perspective of reason, the object of
biblical theology is the faith of the Church, the “vehicle” and hence “means”
for true faith, rational religious faith (S 7:44). Kant then switches from a
defensive to an offensive stance. Since reason is responsible for the truth,
within the philosophical faculty theology is thus being honored.
A third objection is that the philosophical interpretations are “allegorical
and mystical and therefore neither biblical nor philosophical” (S 7:45). Kant
turns this objection back on theology inasmuch as it “mistakes the husk of
religion for religion itself,” for example when it explains “the entire Old
Testament . . . as a continuous allegory . . . of the religious state still to
come” – namely Christianity (S 7:45). The mysticism he finds in people like
Swedenborg whose inner revelations lack precisely that which Kant requires
for truth and knowledge: a public touchstone (S 7:46).
Not content with external objections on the part of theology, Kant makes
four more objections of reason against its own rational interpretation of the
Bible. One objection – “As revelation, the Bible must be interpreted in its
own terms, not by reason” – is rebutted by pointing out that a revelation’s
divine origin is never established through the kind of characteristics that
experience provides. Once again the priority of reason is emphasized:
revelation must be in “harmony with what reason pronounces worthy of
[that is appropriate to] God” (ibid.).
According to the second objection by reason, practice must always be
preceded by theory, and, in the case of revelation, theoretical − which is to
say biblical and historical − propositions about the intention of God’s
inscrutable will. This, Kant retorts, might well apply to the Church’s faith
and its customs, but it does not apply to a rational religious faith directed
“solely to the morality of conduct,” since “acceptance of historical – even
biblical – teachings has in itself no positive or negative moral value and
comes under the heading of adiaphoral” (S 7:47).
The third objection is elaborated with the example of the biblical call to
“arise and walk” that is accompanied by a supernatural power. Against the
standard interpretation, Kant understands the story of Lazarus not as a
quasi-theoretical question of natural laws, but as moral and practical. The
call is made not to someone who is physically, but rather one who is
“spiritually,” dead. In that case, the call comes “through his own reason,
insofar as it contains the supersensible principle of moral life” (S 7:47), and
thus not supernaturally.
The last objection by reason states that a divine act of kindness is required
to supplement what is lacking for our justification. One cannot hope for
such an act to occur on “a stroke of luck,” but must instead know of it as a
historical event, an actual promise. Kant’s answer follows the pattern we
have come to recognize: “A direct revelation from God embodied in the
comforting statement, ‘Your sins are forgiven,’” would be a supersensible
experience which, because “experience only occurs within the limits of the
senses, is impossible.” Moreover, this impossibility, Kant continues, is “not
necessary” (S 7:48) regarding the determining moral and rational religion.
After responding to the objections, Kant draws his conclusion: inasmuch
as scriptural exegesis is of concern to rational religion (and not the Church’s
faith), it must be carried out “according to the principle of morality which
revelation has in view.” Otherwise, the interpretation of scripture is “either
empty of practical content,” and thus meaningless to morality, “or even
obstacles to the good.” But a philosophical interpretation is emphatically
authentic, and hence trustworthy and true. At the same time this position
facilitates integrating such a theological element into a secular morality. For
it is “given by the God within us.”
This elevated, at first sight presumptuous, claim is well founded in that
“we cannot understand anyone unless he speaks to us through our own
understanding and reason.” And so, “it is only by concepts of our reason . . .
that we can recognize the divinity of a teaching promulgated to us.” And
these concepts are “infallible” not in the theoretical, but only in the purely
moral, realm (R 6:70; S 7:48).
30 OTFRIED HÖFFE
I want to draw a preliminary conclusion, here: Enlightenment philoso-

phers cherished a transitive critique of religion, that is one in which one
party, reason, criticized another, religion. For Hume, religion has no general
human value. In Voltaire, this critique is raised into a furious battle against
the Church. According to d’Holbach, all religious dogmas are illusions.
Rousseau is only concerned with natural religion. And for the “late
Enlightenment” thinker Marx, communism abrogates religion.2 For “reli-
gion is the sigh of the oppressed creature . . . it is the opium of the people.”3
With Kant this transitivity becomes – though within limits – a reciprocal,
even dialogical, critique: religion makes proposals; reason selects among
them according to its criteria. It, then, undertakes a philosophical recon-
struction that culminates in the insight that reason (moral reason) needs
religion. This religion consists in a fact that the three Critiques had already
maintained for their part: “that in all our duties we look upon God as the
lawgiver to be honored universally” (S 6:103).
2
Karl Marx and Friedrich Engels, Manifest der kommunistischen Partei in Marx-Engels-Werke (Berlin,
1956– ), Vol. iv, p. 480.
3
Karl Marx, Zur Kritik der Hegelschen Rechtsphilosophie, in Marx and Engels, Manifest, Vol. i, p. 378.
chapter 2
The evil in human nature

Allen W. Wood
As its title tells us, Kant’s Religion within the Boundaries of Mere Reason
purports to be an exploration of what reason, unaided by any empirical divine
revelation, can furnish us in the way of religion. But the book’s aims are even
narrower than that, once we consider its intended audience and the assump-
tions Kant believes he shares with it. This audience is late eighteenth-century
Christians, especially Lutherans in Germany. Kant takes his audience to
accept the basic idea that human beings are capable through their own reason
of achieving scientific knowledge about nature, and also recognizing certain
principles of morality. As Christians, however, he expects them to hold certain
religious beliefs, which interact with their life as rational moral agents.
Kant’s fundamental aim in the Religion is to convince his Christian
audience that there need be no fundamental conflict between these two
aspects of their moral life. He wants to show them that the rational moral
life can be lived through the beliefs, symbols, and feelings of their Christian
religious experience. Kant represents this relation as that between two
concentric circles, the narrower one of rational religion, and the wider
one of ecclesiastical faith (R 6:12). Because we are rational beings, who
must acknowledge reason as our highest authority, the authority of the
wider circle depends on that of the narrower. But the narrower circle may be
incomplete, even in the pursuit of its own purposes, without the wider. This
means that Christianity can receive justification from rational morality, if it
is interpreted in the right way. And traditional teachings (regarded as
admirable examples of the many found in the various ecclesiastical faiths
accepted by human beings) may even be needed to complete the aims of
rational morality. But it also means that Christian teachings need to be
understood rationally if the harmony is to be preserved. As the argument
of the book progresses, we also see, from the standpoint of moral reason,
that the Christian faith ought to develop and progress in accordance
with this same rational self-understanding – a change that can be repre-
sented in Christian terms as the founding of the Kingdom of God on earth
31
32 ALLEN W. WOOD
(see R 6:115–36). A Christian church that does this, Kant says, may be
thought of as the “church militant” which strives to approximate to the ideal
of the “church triumphant” (R 6:115).
This aim determines the Religion’s point of departure, for a standard
starting point of Christian apologetics is the acknowledgement of human
sinfulness, from which Christianity offers salvation. Kant’s project in Part
One of the Religion is therefore to articulate the Christian doctrine of
original sin in terms of rational morality, so that the religion of reason can
also be seen as a saving response to it. Kant therefore begins the Religion
with the bold thesis that human nature contains a radical propensity to evil.1
1 The moral concept of evil

Evil is something for which people are morally responsible, hence something
they freely choose to do. More specifically, it is something they choose to do,
hence do for reasons, yet also something they should not do, and therefore
which they have decisive moral reasons not to do. If we do not assume people
do evil for reasons, then we could not treat evil as an exercise of their free
choice (which would annul its character as evil). At the same time, we cannot
suppose that the reasons for which they do evil are good ones, or that they
rationally justify the choice of evil. In fact, if we do not assume they have
decisive reasons not to do evil, then we would have no ground to regard evil as
something imputable and blamable, something they can be morally expected
and required not to do. This means that the concept of evil is, at a fairly basic
level, a paradoxical one. It means that evil is a form of motivated unreason –
like akrasia or self-deception, which are notoriously beset with problems as to
how they can be conceived or made intelligible.
Kant is well aware of the limits in principle on the intelligibility of evil; in
fact, this is a theme emphasized repeatedly (almost obsessively) in Part One of
the Religion (R 6:21, 25, 32, 35, 41, 43).2 Evil, he says, is “inscrutable”
1
Kant does explicitly equate the radical propensity to evil with “original sin” ( peccatum originarium)
(R 6:31). In Christian theology, “sin” is traditionally understood as human separation from God,
especially separation in the form of disobedience of God’s will. Kant follows this usage to this extent:
he understands “sin” as “the transgression of the moral law understood as a divine command”
(R 6:41–42). Since “religion” is the recognition of all duties as divine commands (R 6:153–54), this
means that the propensity to evil, insofar as it is a propensity to transgress the moral law, would count
as sinfulness when it is viewed from the standpoint of religion (including the religion of pure reason) as
distinct from merely the standpoint of morality.
2
Some commentators speak of Kant “confessing defeat” in the attempt to “explain evil.” Others
complain that his thesis of radical evil fails to “explain anything.” See William McBride,
“Liquidating the ‘Nearly Just Society’: Radical Evil’s Triumphant Return,” in A. D. Schrift (ed.)
The evil in human nature 33
(unerforschlich) (R 6:21); it must be attributed to “the free power of choice, . . .
but there cannot be any further cognition of [its] subjective ground or cause
(although we cannot avoid asking about it)” (R 6:25). Kant does think,
however, that we can form a coherent concept of evil, and even understand
it, in certain ways − at least up to a point − both as a human propensity and in
its relationship to the development of the human species in history.
Kant proposes to conceptualize evil through the relation of the two basic
kinds of rational incentives present to human action: (1) the moral incen-
tive, to conform our choices to duty or the moral law; and (2) non-moral
incentives, which Kant characterizes in general terms as those of inclination
or self-love. Both incentives are necessarily present to us as finite rational
beings; there is also a clear rational priority between them: from the stand-
point of reason, the moral incentive prevails decisively over the incentives of
inclination or self-love. Evil choice, then, as motivated yet rationally defi-
cient, can be formulated in terms of a maxim (or generalized intention):
namely, that of subordinating the (rationally stronger) incentive of morality
to some (rationally weaker) incentive of inclination or self-love.
The difference, whether the human being is good or evil, must lie not in the
difference between the incentives that he incorporates into his maxim (not in
the material of the maxim), but in their subordination (in the form of the
maxim): which of the two he makes the condition of the other. It follows that
the human being (even the best) is evil only because he reverses the moral
order of his incentives in incorporating them into his maxims. (R 6:36)
2 Two wrong ways to look at evil

Kant’s concept of evil is extremely abstract and general, but also in a way
extremely simple. We can better understand its distinctiveness if we contrast it
with two fairly common ways, both rejected by Kant, in which evil is sometimes
conceived so as to make it seem more intelligible than Kant believes it can be.
The first is to locate the ground of evil in our natural inclinations
themselves. This might be seen as an attempt to reduce evil to some
Modernity and the Problem of Evil (Bloomington: Indiana University Press, 2005), p. 38; R. Bernstein,
Radical Evil: A Philosophical Investigation (Cambridge, MA: Polity Press, 2002), p. 33. Bernstein,
Radical Evil, p. 45, then goes on to insist – in agreement with Kant, yet as though this were a discovery
Bernstein had made, and a devastating criticism of Kant – that evil, as a product of freedom, does not
admit of a complete or satisfactory explanation. To look at Kant’s discussion of radical evil as if
providing an “explanation of evil” either is or ought to be one of its intentions is to misunderstand
Kant’s views and aims, and to miss the point of Kant’s discussion. Here Robert Louden gets it right:
“Evil Everywhere: The Ordinariness of Radical Evil,” in Sharon Anderson-Gold and Pablo Muchnik
(eds.), Kant’s Anatomy of Evil (Cambridge University Press, 2010), p. 96.
34 ALLEN W. WOOD
merely natural event that is to be causally explained by the mechanism of

nature. But this could never work, because a causal explanation would
leave out the rational aspect of evil choice, and would also remove evil, as
causally necessitated, from the realm of free choice and moral responsi-
bility, which would do away with it as evil. More often, the attribution of
evil to natural inclination takes the form of treating our animal or
sensuous nature as if it were blamable or a matter for reproach. Such a
position is frequently attributed to Kant himself, on the basis of passages
in his ethical works in which he cites inclinations as offering obstacles
or a counterweight to the moral law (G 4:405, 455, KpV 5:21–28, MS 6:376,
379–80). But in the Religion, Kant’s rejection of the position that blames
evil on the senses or natural inclination is explicit, repeated and emphatic:
“The ground of evil cannot (1) be placed, as is commonly done, in the
sensuous nature of the human being, and in the natural inclinations
originating from it” (R 6:34–35, cf. 6:21, 26–27, 31, 36, 57–60). Evil,
Kant says, cannot lie in natural inclinations which “openly display them-
selves unconcealed to everyone’s consciousness, but is rather as it were an
invisible enemy, one who hides behind reason and is hence all the more
dangerous” (R 6:57). Kant’s own frequent references to inclinations as
resisting the moral law – on which the common misunderstandings are
based – must always be understood in this light. Inclinations themselves,
insofar as they do not originate from free choice, can never be evil. They
become involved in evil only when they have been taken up as incentives
in a freely chosen maxim which is evil. And then it is this maxim that is
evil, not the inclinations themselves.
The second way to give evil a false appearance of intelligibility is to
represent it as something for which a human being might have a genuine
and sufficient reason, by ascribing to the human will an original incentive
to disobey the moral law. Kant’s rejection of this alternative, of a “dia-
bolical will” or an “evil reason,” is equally emphatic: “The ground of evil
can also not be placed (2) in a corruption of the morally legislative
reason.” For, Kant argues, it would be contradictory to think of a rational
being as bound by the moral law yet either lacking any rational incentive
to obey it or possessing a rational incentive to disobey it. “Sensuous
nature,” Kant concludes, “contains too little to provide a ground of
moral evil in the human being,” since it constitutes the human being
only as an animal being, not a moral one. Yet “a reason exonerated from
the moral law, an evil reason (as it were) . . . contains too much, because
resistance to the law would thereby be elevated to an incentive,” and this
would contradict the imputability of evil (R 6:35).
Kant’s rejection of the “diabolical will” is sometimes criticized on the ground
that it excludes the possibility (as the critics like to put it) of “doing evil for evil’s
sake.” Since we know this is something of which human beings are deplorably
capable, critics then represent Kant as having been sadly oblivious to the full
depths of human evil.3 But this criticism rests on a basic misunderstanding.
Kant certainly holds that there are ‘”diabolical vices,” a name Kant gives to the
“vices of hatred”: envy, ingratitude, and malice, which set the unhappiness of
others as an end, even though it is our duty to include the happiness, never the
unhappiness, of another among our ends (R 6:27; MS 6:385–88, 458–61).
Diabolical vices should also be understood to include the aim of degrading
or humiliating rational beings in violation of our duties of respect (MS
6:465–68). These vices involve the knowing choice of an evil end, and the
choice of it as evil. Kant’s position allows, in other words, that people may have,
and act on, an inclination to pursue an evil end simply for its own sake.4
3
This is the criticism of Kant presented long ago by John Silber, “The Ethical Significance of Kant’s
Religion,” in Immanuel Kant, Religion Within the Limits of Reason Alone, trans. Theodore M. Greene
and Hoyt Hudson (New York: Harper Torchbook, 1960), and reiterated more recently in Silber,
“Kant at Auschwitz,” in G. Funke and T. M. Seebohm (eds.), Proceedings of the Sixth International
Kant Congress (Washington, DC: Center for Advanced Research in Phenomenology and University
Press of America, 1991), pp. 177–211, and in Silber’s Kant’s Ethics (Boston/Berlin: DeGruyter, 2012),
pp. 329–34. See also Claudia Card, The Atrocity Paradigm: A Theory of Evil (New York: Oxford
University Press, 2002), p. 212.
4
Kant apparently does subscribe to the scholastic adage nihil appetimus, nisi sub ratione boni; nihil
aversamur, nisi sub ratione mali (“We desire nothing except under the reason of the good; we avoid
nothing except under the reason of the bad.”) His reservations about this adage, expressed at KpV 5:59,
have to do with an ambiguity he finds in the words “good” and “bad,” not with the substantial thesis
asserted in the adage itself. The “guise of the good” thesis (as it has been called) has recently been
controverted by empiricist philosophers such as David Velleman and Kieran Setiya: Velleman, “The
Guise of the Good,” Noûs 26 (1992), 1–26; Setiya, Reasons Without Rationalism (Princeton University
Press, 2007). Those who criticize Kant at this point may have some of their arguments in mind, but
most critics of Kant’s rejection of the “diabolical will” fall foul of misunderstandings of Kant’s claims
that are more obvious and superficial than the issues involved in the controversy over the “guise of the
good”. The ‘guise of the good’ thesis has also recently been ably defended by philosophers with
Kantian leanings: see Sergio Tenenbaum, Appearances of the Good: An Essay on Practical Reason
(Cambridge University Press, 2007), and David Sussman, “For Badness’ Sake,” Journal of
Philosophy 106 (2009), pp. 613–28. Sussman’s article, which is oriented to Kant’s discussion in the
Religion, is particularly good at showing the resources available to proponents of the “guise of the
good” thesis in dealing with alleged counterexamples, such as “doing evil for evil’s sake,” and providing
an even better account of such cases than any available to critics of the thesis. One brief and fairly
obvious point here is that representing an object under the guise of the good is not at all the same as
holding that it is good, all things considered. We obviously can desire something under the aspect of
the good without really thinking it is good, all things considered. Desiring something under one aspect
is possible even when we see it at the same time under the opposite aspect, or even represent one aspect
to ourselves (perversely, or self-deceptively) as precisely the opposite aspect. Thus, doing evil for evil’s
sake necessarily involves representing evil as good. We can see this with Milton’s Lucifer (an august
authority, from whom both sides of this dispute seek support) who declares: “All good to me is lost;
Evil, be thou my good!” (Paradise Lost IV, 109–10). Having self-admittedly lost all good, Lucifer knows
he is settling for second best. His choice of evil as his good is therefore made both with the knowledge
36 ALLEN W. WOOD
This criticism is closely allied to another erroneous charge commonly

made against Kant − namely that, in characterizing evil as a choice of
inclination or self-love over morality, he displays too narrow and simplistic
an understanding of the different kinds of motivation that evil actions may
have. “Inclination” is simply the Kantian name for non-moral incentives,
whatever their content. “Self-love” is referred to here only because Kant
thinks that it consists in the most general pursuit of objects of inclination,
and therefore designates this pursuit in general. Since Kant’s aim here is to
provide the most abstract and encompassing concept of evil, ‘inclination’
and ‘self-love’ should be understood as merely placeholders for whatever
non-moral incentives might be chosen in preference to those of morality.
Kant is not imposing any limits on what one can have an inclination to
will – or representing the evil will in general in terms of some common
stereotype, such as the self-indulgent hedonist, or the shrewd self-interested
schemer.5
that evil is not good, and, at the same time, with the resolve to represent it as good, which, however, he
also knows to be perverse. There would be no such spirit of perversity lurking behind doing evil for
evil’s sake if the doer did not assume the commitments involved in the “guise of the good” thesis. To
those who deny the “guise of the good” thesis, on the other hand, willing evil, even willing something
for evil’s sake, becomes merely one innocent conative disposition alongside others. In calling it “evil,”
we could only be registering the contingent fact that it differs from our conative dispositions (or rather
from God’s). On this account, Lucifer would merely be displaying tastes different from God’s, and
God would be treating his own rationally ungrounded tastes as authoritative even for beings who did
not happen to share them. Lucifer might lend his support to opponents of the “guise of the good”
thesis, because their account makes God (and not Lucifer) the one who is outrageously prideful and in
the wrong. The only way around this would seem to be to say that God is entitled to his arrogant
presumption on the grounds of his irresistible power. He could not be held to be so on the grounds of
his goodness, unless, in a contemptibly sycophantic state of mind, we suppose that his power entitles
him to declare good and evil arbitrarily, and that we cringing slaves must accept unquestioningly his
despotic decrees.
5
Claudia Card, for instance, attributes to Kant the thesis that “self-love, a principled and unqualified
pursuit of self-interest, is the worst principled form of evil in human beings”: Claudia Card, “Kant’s
Moral Excluded Middle,” in Anderson-Gold and Muchnik (eds.), Kant’s Anatomy of Evil, p. 76. (The
same charge is made by Bernstein, Radical Evil, p. 42.) But this is belied by some of Kant’s most
conspicuous assertions about what counts as the worst forms of evil. He often suggests, for instance,
that the “diabolical” vices, or vices of hatred:− envy, ingratitude, malice, vengeance − are the worst that
human nature has to offer (R 6:458–61, 6:27). As evidence for the evil in human nature, he prefers to
cite not self-interested behavior but “never-ending cruelty . . . from which no human being derives the
least benefit” (R 6:33). Kant seems to consider lying or deception (especially self-deception in regard to
one’s conscience and the betrayal of another’s trust) to be the vilest form of human baseness (MS
6:429–31, 437–40, R 6:33–34). All these graver forms of evil can be “principled” (done deliberately, on
principle or as a matter of policy) and they can be displayed by someone who is not rationally self-
interested but even conspicuously imprudent – for instance, the person who sacrifices his own
happiness to his envious or vengeful malice, the person whose cruelty provokes an even more
monstrous vengeance on him, or the person who, by betraying a friend or deceiving himself, forfeits
his own well-being. As for the principled pursuit of happiness as the paradigm of evil, Kant thinks that,
although it can sometimes come into conflict with morality, the pursuit of happiness is always
recommended by prudential reason, and, in a moderated form, prudence often harmonizes with
What Kant denies is that a rational will could coherently be conceived as
having an original rational incentive to act contrary to moral reason. It
would be self-contradictory to conceive of a rational faculty that has an
original incentive to choose exactly the opposite of what it is rational to
choose. In any case, no will could be considered evil (blamable) for doing
precisely what it had an original rational incentive to do. Perhaps in demon-
izing those they hate, some people do frame the incoherent notion of a will
that is (as they might put it) so evil that it is incapable of good and
constituted solely by the incentive to do evil. On closer examination,
however, such a notion could be nothing but an irrational projection of
their malice, whose object they are depicting in nonsensically venomous
terms.6
Kant’s basic conception of evil needs to be highly abstract in order to
capture what is basic to all forms of evil, from the most trivial and minor
moral lapses to the most extreme forms of wickedness. It will therefore
disappoint anyone who wants from a philosophical account of evil the
deliciously lurid evocation of some uncanny state of mind that it pleases
us to imagine motivating the most monstrous evils we can think of. But
we ought to mistrust such demands. As wiser observers have pointed out,
great evils, such as the Holocaust, may actually result more often from
“total banality and prosaic triviality” than from some “satanic greatness.”7
Our prurient desire to be outraged by the extremity of evil may also
exhibit a wish to see the “truly evil” as something “other” and “alien”
(“Nazis,” “terrorists,” or whoever our bugaboo happens to be) – a fitting
object on which to unleash our hostility and resentment, but of course
having nothing in common with any petty fault we might find in our-
selves. Like the Christian doctrine of original sin, however, Kant’s thesis
duty. Contrary to false and invidious images of him, Kant is not an enemy of the pursuit of human
happiness. In the title of Kant’s Anthropology from a Pragmatic Point of View, the term “pragmatic” has
simultaneously several different meanings. (See Allen W. Wood, “Kant and the Problem of Human
Nature,” in B. Jacobs and P. Kain (eds.), Essays on Kant’s Anthropology (New York: Cambridge
University Press, 2003).) One of them, however, is “useful (for human ends)” and another is
“prudential” (conducive to the self-interest and happiness of those who take up the study). The
chief intent of the Anthropology – and of Kant’s lectures on anthropology, the most popular and most
frequently given of all his academic courses – was to acquaint his audience with information about
themselves and other human beings that could help them more effectively pursue their happiness,
through the use of their own faculties and in their worldly interactions with others. If Card were right
that Kant regarded the deliberate pursuit of one’s own happiness as “the worst principled form of evil,”
then Kant would have spent perhaps the most illustrious part of his academic career helping his
audience to practice what he himself regarded as the worst principled form of evil.
6
This point is made effectively by Matthew Caswell, “Kant on the Diabolical Will: A Neglected
Alternative?” Kantian Review 12 (2007), pp. 147–57.
7
Hannah Arendt, The Origins of Totalitarianism (San Diego: Harcourt, 1994), p. 459.
38 ALLEN W. WOOD
of radical evil is meant to direct our attention primarily to evil in

ourselves. It does not, however, aim at depicting this evil – whether in
us or outside us – in an extreme or exaggerated form. Kant’s term “radical
evil” does not refer (as this phrase most often does in our moral discourse
nowadays) to some special kind of evil that is conspicuously heinous,
odious, egregious, or (in that sense of the word) “radical.” For Kant,
“radical evil” refers only to the common root (radix) of all evil present in
human nature, whatever its degree.
3 ‘Rigorism’
Kant holds that, although human nature contains predispositions to good,
the human power of choice is by nature evil. In a Remark to his introduction
to Part One of the Religion, he considers the possibility that we might want to
say of human nature, or even of individual human beings, that they are
neither good nor evil, but either neutral between good and evil or else a
mixture of good and evil. His response to this suggestion is to avow a
predilection for a position he calls “rigorism” (“a name intended to carry
reproach, but in fact a praise”) (R 6:22). “Rigorism” “preclude[s], so far as
possible, anything morally intermediate, either in actions (adiaphora) or in
human characters” (R 6:22). If we attend closely to this formulation, however,
we can’t help realizing that rigorism could not possibly be an extreme or
controversial position; it might even be considered almost trivial. For with
many concepts, perhaps most, it seems to make sense to apply them where
one must, but to be reluctant to apply them where one need not. The concept
of the morally intermediate is surely such a concept, if its overuse is apt to
lead, as Kant fears, to the self-deceptive rationalization of bad conduct. In
favoring rigorism, however, Kant is by no means embracing a “moral
excluded middle,” subscribing to the general position that there is nothing
at all morally indifferent or ambiguous, as if on principle he were proudly and
stubbornly refusing to see “gray areas” anywhere in the moral life. In fact,
throughout his moral philosophy, Kant recognizes many cases of adiaphora
and insists on many kinds of moral intermediateness and ambiguity.8
8
Card, “Kant’s Moral Excluded Middle,” pp. 77–92, provides a long list of “moral intermediates,” as
though they were counterexamples to Kant’s rigorism. The list itself, and her discussion of it, displays a
lot of moral and philosophical insight, but the use of it to criticize Kant seems misplaced. Kant
regularly insists there are moral adiaphora both in human actions (MS 6:223, 458), and (right in the
Religion itself) in matters of religious faith and practice (R 6:44n). He condemns outright the refusal to
allow anything morally indifferent, saying that such a view would transform the dominion of virtue
Kant’s rigoristic position on human character in the Religion is fairly
narrow in scope. It is also supported by a simple and cogent argument.
The argument proceeds from the thesis that “the moral law is itself an
incentive in the judgment of reason,” so that “whoever makes it his
maxim is morally good.” It follows, Kant argues, that whenever an
agent incorporates some other incentive into his maxim ahead of the
moral law, the agent can be considered neither morally good nor indif-
ferent, but must be judged evil (R 6:24). That is the full extent of the
“rigorism” involved in Kant’s Religion.
Clearly, Kant recognizes many actions (the ones he counts as adiaphora)
where no such choice is required. Indeed, in the case of some dutiful
actions, no such choice is ever in question – namely, the kinds of action
Kant considers early in the Groundwork, where duty and some other non-
moral motive (prudence or immediate inclination) both provide sufficient
incentives to act. Since they conform to duty, such actions are good,
deserving praise and encouragement. They are in no way evil, not even
indifferent, even though they also have no authentic moral worth, and their
maxim has no moral content (G 4:397–99). They lack these precisely
because they offered no choice between the incentive of morality and
incentives of self-love or inclination.
In the Religion, however, Kant’s question is whether a position of
indifference regarding this choice is a possible one for a human being,
or for a human character, as a whole. His answer seems to be that, since
both the moral law and non-moral incentives are constantly present to
every human being in the adoption of maxims, each of us, at the deepest
level, is constantly confronted with the question which incentive is to be
preferred or made the condition of the other, and our answer to that
question can in principle never be morally indifferent (R 6:36). Each of us,
therefore, in that searching self-examination which belongs to our duty to
have a conscience (MS 6:400–1, 437–40), must judge whether, in regard
to this choice between moral and non-moral incentives, our “fundamental
maxim” is good or evil.
Kant thinks this judgment is never unproblematic, however, due to
human self-opacity and deception, and he insists that we can’t get at it
directly through the empirical observation of actions.
into a tyranny (MS 6:409). Kant is certainly no partisan of that “moral clarity” that divides people into
“good” and “bad.” He is suspicious of our tendency to see some people as “diabolical” and others as
“angelic” (MS 6:461). When we consider others whose actions we condemn, Kant would encourage us
to see how much we may have in common with them, and how our guiltlessness may be merely good
fortune in having escaped the temptations that afflicted them (MS 6:392–93).
40 ALLEN W. WOOD
We cannot observe maxims, we cannot do so unproblematically even within

ourselves; hence the judgment that an agent is an evil human being cannot
reliably be based on experience. In order, then, to call a human being evil, it
must be possible to infer a priori from a number of consciously evil actions,
or even from a single one, an underlying evil maxim, and from this, the
presence in the subject of a common ground, itself a maxim, or all particular
morally evil maxims. (R 6:20)
It may be tempting to read this passage as alluding to some sort of “a priori”

or “transcendental” argument for Kant’s thesis of radical evil in human
nature. This temptation should be firmly resisted if the real meaning of the
passage is not to escape us entirely. Kant is not yet speaking here about
human nature or the human species, but only about what it means to judge
that a single human agent is evil. So there is as yet no question about Kant’s
thesis that there is radical evil in human nature.
Kant’s claim that this judgment “cannot . . . be based on experience,”
may be puzzling, but it obviously cannot mean that the judgment is wholly
a priori – in the way that a transcendental argument would have to be. The
inference he countenances is said to proceed from one or more “consciously
evil actions,” whose occurrence could plainly be known only by experience.
Kant’s point is that we are not entirely reliable judges of our own maxims or
intentions, especially of the deeper and more pervasive ones, so we cannot
decide whether someone’s fundamental maxim is good or evil merely on the
basis of empirical reports of their good and evil actions. Rather, we need to
consider what maxim their actions ultimately presuppose, viewing this in
light of its relation to the a-priori moral law. Kant thinks that, if we engage
in rigorous self-examination, every one of us will find consciously evil
actions, and be able to infer from them a maxim involving the (at least
occasional) deviation from the moral law, and then ascribe to ourselves the
underlying evil maxim of preferring non-moral incentives to the incentive
of the moral law. That we would all find such a tendency in ourselves seems
extremely plausible, but also quite remarkable, when we consider what it
means for us as rational beings.
In the Critique of Practical Reason, Kant considers a man who “asserts of
his lustful inclination that when the desired object and opportunity are
present, it is quite irresistible to him.” Kant then suggests that we ask him
“whether, if a gallows were erected in front of the house where he finds this
opportunity” on which he would “be hanged immediately after gratifying
his lust, he would not then control his inclination. One need not conjecture
very long what he would reply” (KpV 5:30). Here, Kant is supposing, the
rational strength of the incentive to preserve his life would so far outweigh
the strength of his lascivious impulses as to make the choice an extremely
obvious and easy one. A man who chose to satisfy his lust knowing that as a
consequence he would die immediately afterward would be not only
extremely irrational, but also extremely rare.9
The moral law, however, is supposed to be decisively stronger, as a
rational incentive, than any incentive whatever arising from inclination or
self-love. It therefore ought to strike us as extremely remarkable – a sign of
something very significant about us as rational agents – that we find in
ourselves a strong tendency, at least occasionally, to make the choice of self-
love or inclination over morality. We clearly cannot infer such a tendency a
priori merely from the fact that we have non-moral incentives, since if that
were so, we could also argue validly a priori that Kant’s lustful man must
have a tendency to satisfy his lust even if he knew it would immediately cost
him his life.10 The choice of non-moral incentives over those of morality,
moreover, is even more strikingly contrary to reason than in that case, and
hence a tendency to it is even farther from holding a priori. Why, then, is
the tendency to do evil so common and familiar? What does the fact that it
is mean about our humanity?
4 Human predispositions to good

Human nature involves three basic “predispositions” (Anlagen), which Kant
insists are all entirely predispositions to good. The concept of a predispo-
sition belongs to Kant’s theory of natural teleology, and, more specifically,
of organic development. A living thing has certain tendencies or disposi-
tions to grow into the specific kind of thing it is. When these tendencies
reside in particular organs, resulting in the development to their mature
shape, Kant calls them “germs” (Keime). When the tendencies involve the
whole living thing, and their development consists in the acquisition of a
basic capacity constituting its life activity, Kant calls them “predispositions”
(Anlagen) (RM 2:434).11 In human nature, Kant identifies three original or
9
Of course, the man’s irrational conduct in facing death would be far less unexpected if we thought of
him as moved by sexual passion rather than merely by a lustful desire for sexual gratification (as Kant is
plainly thinking of him here). (We will presently have more to say about the passions.) If the man
were to face death out of love (which has a moral component), then his conduct might not even be
contrary to reason.
10
Something like this inference seems to be behind the interpretation of the doctrine of radical evil
presented by Henry Allison, Kant’s Theory of Freedom (Cambridge University Press, 1990), pp. 152–61.
11
This distinction between Keim and Anlage is a permanent part of Kant’s theory of organic develop-
ment, but it must be admitted that that precise terminology is not. Kant sometimes uses the two terms
interchangeably, and his choice of one over the other is often dictated by a desire for stylistic variation.
42 ALLEN W. WOOD
basic predispositions: animality, humanity, and personality. All are “origi-

nal” in the sense that they condition the possibility of human nature and
cannot be eradicated (R 6:28).
Animality is the source of our merely instinctive inclinations for self-
preservation, sex, and sociability (R 6:26–27). Humanity, in Kant’s technical
sense of the term, as one of the human predispositions, encompasses our
non-moral faculties of practical reason (R 6:27). These include our capacity
to set ends, to select (or devise) means to them, and also our capacity to
combine the ends based on inclination into a whole, and to pursue this
whole under the name of “happiness.” In the Anthropology, Kant therefore
distinguishes within humanity two predispositions, the “technical” and the
“pragmatic” (ApH 7:322–24). Finally, personality is the moral predisposition
that gives us our capacity to legislate and obey the moral law (R 6:27–28,
cf. ApH 7:324).
All three predispositions are predispositions to good in the sense both that
there is nothing inherently bad or evil in any of them, and that they also
“demand compliance with the good,” in the sense that their proper use,
according to principles of reason, would lead always to good and not to evil
(R 6:28). We might think that Kant holds the three predispositions are also
good in the sense that they come to us from nature, rather than involving
our own free agency – from which alone evil, as something imputable to us,
might arise. But this is only partly true, and a correct understanding of
Kant’s conception of human evil depends on appreciating the ways in which
it is not wholly true.
5 Our nature as what is given and as what we make

Kant draws a distinction, early in the Groundwork, between “temperament”
(“what nature has made of us”) and “character” (which has “a worth we give
ourselves”) (G 4:398–99). In anthropology, Kant draws a similar distinction
between our natural aptitude (Naturell) belonging to our “way of sense”
(Sinnesart) and character as belonging to the “way of thinking”
(Denkungsart); the former is what we receive passively from nature, while
the latter is what we make of it through freedom (ApH 7: 285–95). Despite
the importance of this distinction for moral philosophy we ought not to
think that Kant believes it can be drawn easily or neatly in practice. For
Kant, physiological anthropology is the study of what nature has made of the
human being, while pragmatic anthropology is the study of what human
beings have made, and ought to make, of themselves (ApH 7:119). The latter
therefore includes even our natural predispositions, insofar as they involve
reason and their development occurs through human actions.
This means that of the three original or basic predispositions, only
animality truly belongs to physiological anthropology, while humanity
and personality, involving capacities that human beings develop in them-
selves and in one another through education and social interaction, belong
to pragmatic anthropology (ApH 7:321–24). Even animality comes to be
affected by human action and social interaction, insofar as we modify and
transform our originally instinctive desires through our choice (or even the
invention) of the means to satisfy them, and through making them
the objects of our own (and others’) desires and purposive activities
(MA 8:111–13). The predisposition to personality thus includes the develop-
ment of moral feelings, including respect, conscience, and love of human
beings (MS 6:399–403), and also desires founded on reason rather than on
instinct (KpV 5: 8–9n; MS 6:212–13). Crucial to understanding the evil in
human nature is the fact that what we are is partly given by nature, and partly
our own work. Moreover, even what appears to us, superficially, as “given” to
our volitional faculties is something in which our free choice is already
complicit, so that in regard to it we bear responsibility for what we are.
6 The passions
Personality is the sole human predisposition that is incapable of corruption
by the evil maxim. The instinctual desires arising from animality are not
themselves evil, but they can have vices “grafted onto” them: the “bestial”
vices of gluttony, lust, and wanton cruelty (R 6:26–27). The predisposition
to humanity includes the development of the passions – that is, inclinations
which take the form of “mania” (or “addiction”) (Sucht) because they resist
comparison with and limitation by other desires, and consequently resist
the influence of reason (ApH 7:265). The “natural” passions are for wild,
lawless freedom, sex, and revenge, while the “social” passions − tyranny
(Herrschsucht), greed (Habsucht), and ambition (Ehrsucht) − are (respec-
tively) addictions to domination, wealth, and honor (ApH 7: 265–75;
cf. Idea 8:21, G 4:393).
All the passions, even the “natural” ones, are social in the sense that they
arise only in society and have only other human beings as their object
(ApH 7:268). And although they resist reason, the passions arise only in
connection with it, and presuppose a maxim (ApH 7:266). Because they
resist reason, the passions are also evil (ApH 7:267). These passions are
allied to the “vices of culture” or “diabolical” vices that may be “grafted
44 ALLEN W. WOOD
onto” humanity: envy, ingratitude, and gloating over harm to another

(Schadenfreude) (R 6:27).
A striking feature of the passions is also the way in which they involve
deceptiveness (self-deception) and illusion or even delusion (Wahn).
A delusion is “the inner practical deception of taking what is subjective in
a motive for something objective” (ApH 7:215). A passion “presents objects
of imagination as real ends (ways of acquiring honor, domination or
money)”; and inclinations for these things “are apt to become passionate
in the highest degree, especially when they are applied to competition
between human beings” (ApH 7:275). Thus, every passion involves a
“slavish disposition, through which another, having gained power over it,
acquires the capacity to use one’s own inclinations to serve his aims” (ApH
7:272). By self-deceptively exaggerating the value of some imagined object,
a passion makes it difficult for us to consider rationally the strength of the
incentive it offers us. Passions therefore make people imprudent – short-
sighted, impulsive, prone to take excessive risks, hence less likely to acquire
the objects they desire – but also prone to think that the gratification offered
by some desired object is greater than it is, so that even if it is acquired, the
possession of it is likely to leave the possessor disappointed and unhappy.
Where morality is at stake, the self-devised delusions involved in passion
play an essential role in the irrational preference of incentives of inclination
over those of morality. Passions are therefore always evil, and conspicuous
manifestations of the propensity to evil in human nature (ApH 7:267). That
people are prone to inflict such illusions on themselves is, of course,
fundamentally just as incomprehensible as the choice of the evil maxim
itself. But it clearly plays a role in that choice.
The basis of this self-corruption of our predisposition to humanity is
bound up both with its character as rational and its character as social, and
displays the intimate connection between human rationality and sociability
themselves:
The predisposition to humanity can be brought under the general title of a
self-love which is physical and yet involves comparison (for which reason is
required); that is, only in comparison with others does one judge oneself
happy or unhappy. Out of this self-love originates the inclination to gain
worth in the opinion of others, originally, of course, merely equal worth . . .
but from this arises gradually an unjust desire to acquire superiority for
oneself over others. (R 6:27)
We see this manifested in the passions: the natural passion for freedom is a
mania to be free of all constraints imposed on us by the existence of others
(including their rights in relation to us), while sexual passion, as Kant
understands it, is the mania to use the body of another as an object for
our gratification. Both involve an unbridled impulse to dominate. Through
the social passions, we seek superiority or control over others in specific
ways: in tyranny, through their fear; in greed, through their self-interest; in
ambition, through their opinion (ApH 7:272). The predisposition to
humanity, in other words, in its development by human beings themselves
through freedom, turns into what Kant elsewhere calls “unsociable soci-
ability” (Idea 8:20–21) or the “self-conceit” which needs to be “struck down”
or “humiliated” through respect for the moral law (KpV 5:73). This turns
out to be the source, in Kant’s account, of the radical evil in human nature.
7 The propensity to evil

Kant calls our tendency to manifest the fundamental maxim of evil a “propen-
sity” (propensio, Hang). He defines a “propensity” as “the subjective ground of
the possibility of an inclination (habitual desire, concupiscentia) insofar as this
possibility is contingent for humanity in general” (R 6:29). More specifically,
Kant thinks of a propensity as “the predisposition (Anlage) to desire an
enjoyment which, when the subject has experienced it, arouses the inclination
to it” (R 6:29n). The addiction to alcohol in some people, for instance, counts
as a propensity, when drinking it creates in them a desire to drink more. Thus
a propensity is always something acquired through action (through choice),
hence something in which our freedom is complicit, and for which we are
responsible. Kant identifies a propensity with an “inclination” (habitual
desire), for which we are responsible insofar as it involves a maxim (as a
passion does). He identifies it also with “concupiscence,” which is neither an
act of the will nor even a desire, but a “stimulus” or “enticement” (Anreiz) that
“determines desire” (MS 6:213). The propensity to evil, however, is something
we have “brought on ourselves” and may be imputed to the will (R 6:20–21).
There may be no choice, and no maxim, involved in merely having desires – at
least in having those instinctual impulses arising immediately from our
animality, and so those desires are not blamable. But there is a choice, and a
maxim, involved not only in acting on inclinations in preference to the moral
law but also in being enticed by concupiscence to do so. Not only yielding to
temptation, therefore, but also being tempted, is something for which our will
is responsible.12
12
Our will is responsible for temptation, but it would be an error to infer that we are to blame for being
tempted, since blame should attach only to an act that transgresses the moral law, and no such act is
46 ALLEN W. WOOD
The propensity to evil arises out of our use of reason, in a social context,
where we seek self-worth and status in comparison to others. Through the
attempt to achieve superiority to others, we create in ourselves an inclina-
tion to act in ways that are exempt from the universal laws we will rational
beings to follow, to treat the humanity of others as a mere means to our ends
rather than as an end in itself, and to pursue ends for ourselves that cannot
be brought into rational harmony with the ends of others in a realm of ends.
Doing this presupposes a fundamental maxim of preferring the satisfaction
of our own inclinations over the incentives of the moral law.
Not every case in which we act on this maxim is one in which we directly
pursue superiority over others, just as not every transgression of duty is
transgression of a duty to others. But the maxim of evil is grounded in a
certain way of valuing ourselves – the comparative−competitive way, where
we seek to acquire something that sets us apart from others and makes us
superior to them. This leads us to value our state or condition (Zustand) over
our person, and the happiness of the former over the virtue or morality of the
latter, since (as we have seen) “we consider ourselves happy or unhappy only
in comparison to others.” Some “bestial” vices (such as gluttony) do not
directly involve social comparison, but they do involve the excessive pref-
erence for pleasure in our state over the dignity of our person (MS 6:427).
On Kant’s account, the violation of our duty of self-preservation through
suicide involves preferring the pleasure of our state for as long as we survive,
to the worth of our person, which we are willing to sacrifice as a mere means
to it (G 4:421–22, 429, MS 6:422–24). For civilized beings like ourselves,
even “private” vices may indirectly involve social comparison: even the
simplest pleasures are more delicious when we know others cannot afford
to indulge in them.13
performed merely by our being tempted to commit it. Of the Christ ideal – the “ideal of holiness” as
the “personified idea of the good principle,” “humanity in its moral perfection,” or “the ideal of
humanity well-pleasing to God” (R 6:60–61) – Kant says that such an ideal person would be “tempted
by the greatest temptations” and although “a human being evil by nature,” he would be one who
“renounces evil on his own and raises himself up to the ideal of holiness” (R 6:61). This strongly
suggests that the Christ ideal is that of a person who is afflicted by the radical propensity to evil, for
which his own will is responsible, but who succeeds in resisting the temptations this propensity places
in his way, and avoids all actual transgression of the moral law. This human being would not have a
holy will in the sense that God does – God’s holiness consists in not having any empirical incentives at
all that might compete with those of goodness (G 4:414, KpV 5:82), but the will of the Christ ideal
would be holy in the sense that, although tempted, it never yields to temptation.
13
Here Kant’s theories are continuous with the insights of Adam Smith, which display an anything but
uncritical assessment of the motives that operate in the so-called “free market”: “It is chiefly from
regard to the sentiments of mankind that we pursue riches and avoid poverty. For to what purpose is
all the toil and bustle of this world? What is the end of avarice and ambition, the pursuit of wealth, of
power and pre-eminence? Is it to supply the necessities of nature? The wages of the meanest laborer
Other vices involving a violation of duties to ourselves more directly
involve social comparison. Sexual lust is associated (albeit sometimes indi-
rectly and by means of imagination) with the desire to subject another to
our animal impulses. Lying, whether as a means of manipulating another or,
in relation to ourselves, maintaining a flattering false image of who we are, is
manifestly social in motivation and content. The motivation, on Kant’s
account, of self-regarding vices of avarice (MS 6:432–34) and servility
(MS 6:434–36) involve social comparison and competition in equally
obvious ways. It displays a very shallow understanding of self-regarding
duties and vices to think that they constitute exceptions or objections to the
social origin of the human propensity to evil.
8 Evil and sociability

Kant’s thesis of the radical evil in human nature is intended as a representa-
tion of the Christian doctrine of human sinfulness. It is at the same time a
thoroughly modern conception of human nature, whose origins are found
in such social thinkers as Hobbes, Mandeville, Diderot, Adam Smith, and
above all Jean-Jacques Rousseau. Its import is (as one might expect from
doctrines presented as “within the boundaries of mere reason”) anti-
theological and even anti-metaphysical. Its ethical thrust is not gloomy or
misanthropic but egalitarian and, historically considered, socially progres-
sive. Kant thinks that even bewailing the evil in our nature displays a
disposition to good that is a sign of moral progress in the human species
(ApH 7:333). His thesis belongs to a characteristically Enlightenment
account of how our rational, even our moral, faculties are acquired in
history.14
To claim that for Kant the radical human propensity to evil has this social
and historical origin is not to offer an interpretation of that doctrine. It is
merely to report what Kant himself says. We have seen some explicit
statements of it already. Another occurs early in Part Three of the
Religion, where Kant intends to argue that the means of combating evil
can supply them . . . It is the vanity, not the ease, or the pleasure [of the higher ranks of life] that
interests us . . . Compared with the contempt of mankind, all other external evils are easily
supported . . . It is not ease or pleasure, but always honor, though frequently an honor very ill
understood, that the ambitious man really pursues”: Adam Smith, The Theory of Moral Sentiments
(Amherst, NY: Prometheus Books, 2000), pp. 70–71, 89, 83.
14
A very illuminating account of the way Kant thinks evil operates as a stage in every historical
development of human capacities is given by David Sussman, “Perversity of the Heart,”
Philosophical Review 114 (2005), pp. 1–24.
48 ALLEN W. WOOD
must be social (an “ethical community” or “people of God”) because the

source of evil is social:
The human being is nevertheless in this perilous state [the radical propensity to
evil] through his own fault; hence he is bound at least to apply as much force as
he can muster in order to extricate himself from it. But how? That is the
question. – If he searches for the causes and the circumstances that draw him
into this danger and keep him there, he can easily convince himself that they do
not come his way from his own raw nature, so far as he exists in isolation, but
rather from the human beings to whom he stands in relation or association. It is
not the instigation of nature that arouses what should properly be called the
passions, which wreak such great devastation in his originally good predisposi-
tion. His needs are but limited, and his state of mind in providing for them
moderate and tranquil. He is poor (or considers himself so) only to the extent
that he is anxious that other human beings will consider him poor and will
despise him for it. Envy, addiction to power, avarice, and the malignant
inclinations associated with these, assail his nature, which on its own is
undemanding, as soon as he is among human beings. Nor is it necessary to
assume that these are sunk into evil, and are examples that lead him astray: it
suffices that they are there, that they surround him, and that they are human
beings, and they will mutually corrupt one another’s moral disposition and
make one another evil. (R 6:93–94)
It is idle to dispute the claim that Kant locates our propensity to evil in the
social condition – for he does this quite explicitly. Some commentators,
however, insist that it simply can’t be right to understand Kant in this way.15
One objection is that to place evil in the context of human competitiveness
is to exempt human individuals from responsibility for it. A related objec-
tion is that attributing evil to our social condition is “reductionist,” ignoring
the “noumenal” context of evil, and suggesting that it can be made more
comprehensible in naturalistic terms than Kant allows.16 All such objec-
tions, it seems to me, proceed from a failure to appreciate the way Kant
understands human sociability and its relation to human rationality and
freedom. The assumption seems to be that the social condition must consist
of nothing but external causal interactions in the world of appearance, so
15
For example: Jeanine Grenberg, Kant and the Ethics of Humility: A Story of Dependence, Corruption
and Virtue (Cambridge University Press, 2005), pp. 31–42; Grenberg, “Social Dimensions of Kant’s
Conception of Radical Evil,” in Anderson-Gold and Muchnik (eds.), Kant’s Anatomy of Evil,
pp. 173–94; and Pablo Muchnik, “An Alternative Proof of the Universal Propensity to Evil,” in
Anderson-Gold and Muchnik (eds.), Kant’s Anatomy of Evil, pp. 128–30.
16
The charge of “reductionism” (directed at me) is made by Seiriol Morgan, “The Missing Formal
Proof of Humanity’s Radical Evil in Kant’s Religion,” Philosophical Review 114 (2005), pp. 110–11. For
my replies to such objections, see Wood, “Kant and the Intelligibility of Evil,” in Anderson-Gold and
Muchnik (eds.), Kant’s Anatomy of Evil, pp. 165–70.
that identifying evil with a propensity of our sociability must involve a
“reductive” view of it, which ignores the “noumenal” and even undermines
our responsibility for it. Also at work in these objections is a rather tradi-
tional (if nevertheless downright silly) metaphysical picture, according to
which, on the Kantian view, freedom can belong only to noumenal monads
floating about in a transcendent supernatural heaven, outside any context in
which their free choices might have any conceivable human meaning. Kant
has frequently been criticized for conceiving of human freedom and moral
autonomy as lacking any natural or social context.17 His placement of our
acquisition of the radical propensity to evil in the social context of the
natural development of human predispositions in history ought to be
sufficient to give the lie to this all too familiar but profoundly distorted
interpretation of Kant; those who deny that radical evil is unsociable
sociability have evidently bought into it.
Kant in fact views human sociability in the context of human freedom,
simply because it provides the historical context in which human reason and
freedom have developed. Our sociability is no more merely a complex of
external causal interactions than are human freedom or human history
themselves, so it is in no way “reductionist” to identify the propensity to
evil with unsociable sociability. Human nature is not only free and rational all
the way down, but also, and at the same time, social and historical all the way
down, simply because its freedom and rationality are self-made in a social and
historical context.18 Therefore, as Kant himself says, whatever noumenal
aspect, from a metaphysical standpoint, may belong to human freedom,
must belong also to human sociability and human history (Idea 8:17).
The social context of evil provides evil with the only kind of (very
limited) intelligibility Kant thinks evil can have. If evil is rooted in human
competitiveness (unsociable sociability), then it can be seen as serving an
17
For two (among almost endlessly many) examples of this misguided reproach, see Bernard Williams,
Ethics and the Limits of Philosophy (Cambridge, MA: Harvard University Press, 1985), pp. 64–65, and
Simon Blackburn, Ruling Passions (Cambridge University Press, 1998), pp. 246–48.
18
Grenberg, “Social Dimensions of Kant’s Conception of Radical Evil,” pp. 183ff., is at least on the right
track when she raises questions about Kant’s conception of “society.” But she never arrives at a
satisfactory answer to them. She is on the entirely wrong track when she attacks me for conceiving of
society as nothing more than “the presence of another person.” Apparently she does this because I
quote Kant as saying: “It suffices that [others] are there, that they surround him, and that they are
human beings, and they will mutually corrupt one another’s moral disposition and make one another
evil” (R 6:94). But neither Kant nor I would ever say that “society” refers to nothing but “the presence
of another person,” and this is not what he is saying in the quoted passage. Rather, he means that the
presence of others is sufficient to corrupt a given individual, given that this presence is sufficient to
elicit the complex propensities (including the freely chosen propensity to evil) that constitute human
sociability.
50 ALLEN W. WOOD
unconscious (natural) purpose, in promoting the historical development of

human species capacities (Idea 8:20–22). Kant explicitly invokes the same
account in the Religion:
These vices [grafted onto our predisposition to humanity] do not really issue
from nature as their root but are rather inclinations, in the face of the anxious
endeavor of others to attain a hateful superiority over us, to procure it for
ourselves over them . . . ; for nature itself wanted to use the idea of this
competitiveness (which in itself does not exclude reciprocal love) as only an
incentive to culture. (R 6:27)
This is not a causal explanation of the propensity to evil (such as might,

through natural necessity, deprive human beings of the freedom to choose it
or resist it). It does not do away with the inscrutability and incomprehen-
sibility of evil, on which Kant repeatedly insists. It belongs instead to the
natural teleology that Kant places not within the sphere of a mechanistic
science of nature, but rather within the sphere of reflective judgment, which
seeks a special kind of intelligibility in matters where the causal mechanism
can never provide one (KU 5:395–400).
9 Deception and the three degrees

of the propensity to evil
Kant conceptualizes the propensity to radical evil in the form of a highly
abstract maxim: the maxim of preferring the rationally inferior incentives of
inclination and self-love to the rationally stronger incentive of morality.
Despite its simplicity and abstractness, the manifestations of this maxim
are supposed to be as luxuriantly diverse and inventive as humans are
in manifesting evil. Kant tries to give us a sense for this subtlety and
creative opulence by distinguishing three “degrees” in which the propensity
to evil may show itself. These he names (1) “frailty” (Gebrechlichkeit,
fragilitas); (2) “impurity” (Unlauterkeit, impuritas, improbitas); and
(3) “depravity” or “perversity” (Bösartigkeit, Verderbtheit, vitiositas, pravitas,
perversitas, corruptio) (R 6:29–30). As degrees of the propensity to evil, they
are not degrees of the evil found in actions, but degrees of temptation to evil.
All three also illustrate the same deceptiveness (or self-deceptiveness) that
constitutes a fundamental feature of all evil.
The lowest degree of evil is what Kant calls the “frailty” of the human will
that adopts a good maxim but proves too weak in carrying it out. The “frail”
will deceives itself insofar as it represents its failure to live up to its good
maxim as a casual, perhaps merely “accidental,” failure, a temporary
“lapse,” whereas in fact it could not have occurred but for a systematic
choice to allow the rationally inferior incentives of self-love or inclination to
take precedence over those of reason or the moral law.
The second degree of evil is “impurity.” It is noteworthy that impurity
represents a greater temptation to evil than frailty, since frailty shows itself
in evil action, while impurity need not. With it,
although the maxim is good with respect to its object (the intended com-
pliance with the law) and perhaps even powerful enough in practice, it is not
purely moral, i.e. [has not] as it should have, adopted the law alone as its
sufficient incentive, but, on the contrary, often (and perhaps always) needs
still other incentives besides it in order to determine the power of choice for
what duty requires. (R 6:30)
What Kant means by “impurity” is not the case Kant considers early in the
Groundwork, where the honest merchant and the sympathetic friend of
humanity are considered and where the presence of non-moral incentives
for doing a dutiful action is at issue, so that there is no need for (hence also
no possibility of) acting “from duty” (G 4:394–99). Rather, impurity occurs
only where there is not only the presence of non-moral incentives, but a
systematic dependence on them, a condition of will or character where the
incentive of duty would not be sufficient (without them). This leads to the
false representation of externally dutiful conduct as reflecting a dutiful
maxim when in fact it does not, and also a tendency to represent actions
as dutiful when they may not be, simply because they accord with the
incentives of self-love or inclination on which the agent has come to depend
in the performance of duty.19
The highest degree of temptation is displayed in “depravity” or “perver-
sity.” Here there might seem to be no room for deception of any kind, since
depravity appears to be only the brazen and open choice of “maxims that
subordinate the incentives of the moral law to others (not moral ones)”
(R 6:30). But we must not ignore the essential point that the concept of evil
makes sense only if morality carries reasons with it that are overriding in
19
To read impurity into the examples in the Groundwork is sadly common, though it is unsupported by
the text, and also guarantees that these examples, as well as the point Kant is trying to make in
presenting them, will be misunderstood. For all we are told in the Groundwork, the moral incentive
might be quite sufficient for the performance of duty in the case of the merchant or the friend of
humanity – as it in fact proves to be in those variants of the examples where Kant imagines the
incentives of inclination to be removed. The role of self-deception in cases of impurity is evident,
whereas there need be no self-deception at all in realizing that the policy of honest business is prudent,
or that one has strong motives of sympathy for complying with the duty to help others.
52 ALLEN W. WOOD
relation to other reasons, such as those of self-love or inclination. Because

the rational will cannot be “diabolical,” evil can never be anything but a
failure of rational volition. And there would be no genuine evil if the
supposedly evil agent chose other incentives over those of morality only
through an innocent error about their relative rational strength – as if, for
instance, someone simply had no way of knowing that the rights of others
give him good reasons to limit his pursuit of the pleasure he takes in
harming them. (An agent who was truly ignorant about that could not be
a rational moral agent at all.)
As we saw above, the concept of evil makes sense at all only if the one
who chooses evil is well aware that the incentives of morality are rationally
superior to non-moral incentives but gives preference to the latter any-
way. Such a deliberately perverse choice, however open, brazen and
shameless it may represent itself as being on the surface, or whatever
rationalizations it may use to cloak itself, necessarily involves a deep self-
deception on the agent’s part about the structure of its own commitment
to what it knows about the relative rational force of the incentives by
which it is moved. Kant therefore regards evil as always closely bound up
with lying, especially “the inner lie” (lying to oneself, rationalization,
hypocrisy) (R 6:42n; MS 6:429–31). The familiar declaration of the villain,
when caught in the act, that he and he alone is rational and honest with
himself about his immoral aims and motives, while his accusers are all
cowards, liars, and hypocrites when they think of themselves as any
different from him – this little speech probably represents the uttermost
possible extreme in self-deception.
There are, of course, philosophical theories that would directly support
this villain’s claims. Some hold that rationality is merely a matter of
choosing the means to whatever ends you happen to have, and such choices
are equally rational whether the ends conform to morality or oppose it.
Others hold that there is a “dualism of practical reason,” making it inde-
terminate or ambiguous whether morality or self-love is the rationally
overriding incentive, or at least allowing that either preference could
count as “rational.” Still others maintain that morality is a matter only of
feeling or sentiment, implanted in us by nature or society, so that there are
not (and never could be) any reasons at all to act morally. The points I have
been making, however, imply not only that such theories must be rejected,
but also that they can never be taken at face value. They are not merely
innocent errors, but necessarily also deceptions (self-deceptions). The phil-
osophical arguments and intuitions that support them must be seen as
partaking of the deception involved in the radical evil in human nature.
10 Humanity as evil “by nature”

So far we have tried to understand what Kant understands by “evil” and by
the claim that there is in human beings a “propensity” to evil. The most far-
reaching claim of Part One of the Religion, however, is that this propensity
belongs not merely to human beings (to those, such as ourselves, who may
be conscious of evil acts and can infer from them the presence of a
fundamental maxim of evil), but to human nature, hence to the entire
human species.
In this connection, Kant makes some startlingly ambitious claims. He
insists that evil is brought upon ourselves through freedom (not inherited by
nature), and also that our moral predisposition to personality gives us the
permanent power constantly to struggle against it, which Kant presents
in Christian terms as the possibility of moral regeneration through grace
(R 6:44–53). Nevertheless, he insists that the propensity to evil must be
thought of by us as:
(1) universal, present without exception in every member of the human
species (R 6:25);
(2) “innate,” represented as preceding every deed of ours in time (R 6:31);
and finally
(3) “inextirpable,” incapable of eradication once and for all by any deed of
ours (R 6:37).
These three claims about the propensity to evil constitute what Kant means
in saying that radical evil belongs not merely to individuals but to human
nature, that every human being is “by nature evil” (R 6:32).
Some critics have charged that Kant is ultimately faced with a contra-
diction he cannot escape when he claims that evil is both imputable and also
present in human nature (some worry about the claim that it is “innate,”
others about the claim that it is “inextirpable”).20 Kant’s claim that the
propensity is “innate” does not refer to its causality (which Kant holds is
20
Innateness is the obstacle for Gordon E. Michalson, Fallen Freedom: Kant on Radical Evil and Moral
Regeneration (Cambridge University Press, 1990), pp. 46–66; while inextirpability is the stumbling
block for Bernstein, Radical Evil, pp. 11–45. No doubt the traditional Christian doctrine of original
sin is a natural target of charges of inconsistency, holding us responsible for matters for which we
could not possibly be responsible (such as the misdeeds of our distant ancestors, or what we have
biologically inherited from them). Perhaps this makes Kant’s attempt to agree with the doctrine look
to the critics like something that cannot avoid self-contradiction. But for just the same reason, Kant is
quite careful, and also (I believe) strikingly successful, in explaining how he understands the
terminology traditionally associated with the Christian doctrine of original sin, in such a way that,
so understood, the radical propensity to evil is consistent with human freedom and responsibility for
it (R 6:31–32, 38–39).
54 ALLEN W. WOOD
“inscrutable”) but to the observed fact that there is no identifiable temporal

point in our lives when it is acquired (R 6:38). But it is not contradictory to
say that we are responsible for something even though we cannot point to
the time when we chose it. People are to blame for hatred, bigotry, greed,
callousness, and other bad traits of character even though we cannot identify
a precise time at which they acquired these traits. Kant does not consider
these traits “innate” but that is presumably because, unlike the propensity to
evil (which is their root), they are not fundamental, hence not ubiquitous.
Nor is there anything contradictory about holding that we can freely and
imputably do something to ourselves that we cannot then undo. It belongs
to time’s arrow that some things, once done, cannot be undone: spilt milk
cannot be put back into the bottle, a shattered glass cannot be reassembled,
toothpaste cannot be squeezed back into the tube. People can sometimes do
harm to themselves or others which they can never fully undo or make good
by satisfactory reparations. This fact in no way lessens the responsibility of
the doer for the harm done. Kant’s claim is that a human being’s basic
choice of the evil maxim, as manifested in a radical propensity to evil,
constitutes such an irrevocable harm, self-inflicted on our own power of
choice prior to any identifiable deed. It is indeed an extremely strong claim
that all human beings, without exception, have inflicted on themselves a
propensity to evil, especially that at every point in their lives all human
beings have already done so. But it is not self-contradictory.
11 The “proof ” of radical evil

Still, it is natural for us to wonder how Kant proposes to establish that in
human nature there is a radical, innate and inextirpable propensity to evil.
It is probably misguided interpretive generosity that has led many com-
mentators to hold that Kant’s thesis of radical evil requires (and receives)
an a priori proof or “transcendental deduction” of some kind. Then they
hasten to offer various arguments, often loosely associated with well-
known Kantian doctrines or with patches of text from the Religion, in
support of their own highly inventive ideas along these lines.21 Without
21
Perhaps the best-known example of this is found in Henry Allison, Kant’s Theory of Freedom (see note
10 above). Other examples are Stephen Palmquist, Kant’s Critical Religion (Aldershot/Burlington:
Ashgate, 2000), pp. 156–58, and Palmquist, “Kant’s Quasi-Transcendental Argument for a Necessary
and Universal Evil Propensity in Human Nature,” Southern Journal of Philosophy 46 (Summer, 2008),
pp. 261–97; Seiriol Morgan, “The Missing Formal Proof of Humanity’s Radical Evil,” pp. 63–114; and
Pablo Muchnik, “An Alternative Proof of the Universal Propensity to Evil.”
trying to refute all such supposed a priori proofs (for which there is no
space here), I instead offer some explicit textual grounds for concluding
that no arguments of this kind (whatever their merits) could possibly
correspond either to Kant’s intentions or his accomplishments in the
Religion, as he understood them.
The natural place to look for Kant’s reasons for confidently asserting
the thesis of radical evil in human nature is in section III of Part One, boldly
entitled “The Human Being Is By Nature Evil” (R 6:32–39). Here he
famously asserts that “we may spare ourselves the formal proof
( förmlichen Beweis) that there must be such a corrupt propensity rooted
in the human being, in view of the multitude of woeful examples that the
experience of human deeds parades before us” (R 6:33). He then offers us an
all-too-familiar account of the notorious barbaric cruelty of so-called “nat-
ural” or “primitive” peoples, and the even worse vices of civilization – falsity,
hypocrisy, treachery to friends, ingratitude, the malicious pleasure sophis-
ticated people take in the misfortunes even of those closest to them
(R 6:33–34). This is not, as is often said in the literature, merely an argument
from induction or empirical generalization from random instances. The
idea is to consider humanity in its most extreme contrasting forms – the
“primitive” and the “civilized” – and to impress us with the overwhelming
prevalence of evil in these two types (supposing too that we will easily
recognize the same forms of evil in those specimens of humanity that lie
everywhere in between the extremes). This makes it more plausible to
consider the propensity to evil universal, but I submit that Kant realizes,
and more or less openly confesses, that it falls short of establishing either
that all human beings, without exception, display a radical propensity to evil
or that this propensity lies in human nature itself. It is not a “formal proof.”
But he seems to regard it as sufficient for his purposes in the Religion.
There is no reason, however, to think that by a “formal proof ” Kant
would have to mean an a priori or “transcendental” proof. In fact, Kant has
already told us quite explicitly at the end of the introductory paragraphs to
Part One what it would be to “prove” (beweisen) the thesis:
However, that by the “human being” of whom we say that he is good or evil
by nature we are entitled to understand not individuals (for otherwise one
human being could be assumed to be by nature good and another evil), but
the whole species – this can be proven (bewiesen) only later on (weiterhin) if it
is shown in anthropological research that the grounds that justify us in
attributing one of these two characters to a human being as innate are of
such a nature that there is no cause for exempting anyone from it, and that
the character therefore applies to the species. (R 6:25–26)
56 ALLEN W. WOOD
Here it seems quite evident that the “proof” Kant has in mind is not a
transcendental deduction or a priori demonstration of any kind. It rather
consists in the results of “anthropological research” to be completed “later
on.” The “formality” of this proof would presumably consist in a proper
scientific presentation of the grounds (partly rational and partly empirical,
as in any natural science conceived in Kantian terms) for the conclusion that
the evil propensity belongs innately to all members of the human species.
We are given the methodological hints toward such a proof (though surely
no more than that) in the teleological conjectures Kant presents in Idea for a
Universal History, where it is argued that nature provides for the full
development of human predispositions through the mechanism of unsoci-
able sociability (Idea 8:21–22). The actual proof would be provided by a
future empirical anthropologist-historian who successfully followed Kant’s
philosophical idea (Idea 8:29–31).
Some critics are bewildered that “Kant never gives – or even attempts to
give – a proof of his controversial and bold claim that man is evil by nature”;
they complain of the “absence of genuine argumentation on this crucial
point.”22 These reactions might dissipate, however, if they thought more
carefully about how ambitious a thesis radical evil is, and also recalled for
whom Kant is writing. He is not writing for them, but for an audience of
eighteenth-century Lutherans, who already accept the Christian doctrine of
original sin. He is also writing for Enlightenment philosophers who, once
they come to recognize the doctrine of radical evil as continuous with
Rousseauian philosophy of history, ought also to be quite open to it, even
if they realize that it is controversial, and also appreciate that fully establish-
ing it is necessarily going to be a daunting task for future anthropologists. It
could not have occurred to Kant, therefore, even to think of offering an a-
priori “proof” for a thesis that is so obviously an empirical one also requiring
complex theoretical underpinnings. That he demurred from this ambitious
project in the Religion is perfectly understandable in light of the fact that the
primary focus of this work is (after all) religion, not anthropology or human
history.
Regarding his Lutheran audience, Kant’s most challenging task is not to
convince them that human nature harbors a universal, innate, and
22
Michalson, Fallen Freedom, p. 46; Bernstein, Radical Evil, pp. 34–35. But Michalson is at least
approximately right when he refers at this point to Kant’s use of “biblical references to serve as a
substitute for further argument”: Michalson, “Kant, the Bible, and the Recovery from Radical Evil,”
in Anderson-Gold and Muchnik (eds.), Kant’s Anatomy of Evil, pp. 58ff. But I see Kant’s appeal to
Christian belief, at least at this point, as more implicit than explicit, and as an appeal to accepted
Christian doctrine rather than to the Bible.
inextirpable propensity to evil (that they already believe). It is rather to
persuade them that his modern, rationalistic, socially oriented, and histor-
ically progressive interpretation of original sin is one Christians should
accept. And here we must confess that Kant’s doctrine of radical evil, like
his entire project in the Religion, has had only very limited success – but not
due to weak argument on his part. That has been due rather to the fact that
too much of nineteenth- and twentieth-century religious culture has
chosen, to its irredeemable disgrace, to turn itself into a fortress of stubborn
resistance against enlightenment, freedom, science, human progress, and
anything associated with them.
A philosopher’s failure to convince an audience may be their fault rather
than his. As I hope we have now seen, Kant’s doctrine of radical evil
illustrates that truth in a variety of ways.
chapter 3
Radical evil and human freedom

Ingolf Dalferth
1 Kant’s view of human freedom
1.1 The human person

One of Kant’s most deeply held convictions was that human beings are by
nature capable of being free, able to determine what they can and ought to be
as human beings: responsible persons among persons.1 The human being “has
a character which he himself creates by being capable of perfecting himself
after the purposes chosen by himself. Through this, he, as an animal endowed
with reason (animal rationabile) can make out of himself a rational animal
(animal rationale)” (ApH 7:321), but he can do so only because he, as an animal
rationabile, is at the same time a person that is accountable to other persons.
In Religion within the Boundaries of Mere Reason, Kant describes this more
fully by distinguishing three predispositions of human beings: the predis-
position to animality (as a living being), the predisposition to humanity (as a
living, and at the same time a rational, being), and the predisposition to
personality (as a rational, and at the same time responsible, being) (R 6:26).
The first comprises the existential drive “for self-preservation,” the “sexual
drive” for “the propagation of the species” and “for the preservation of the
offspring thereby begotten through breeding,” and the “social drive” “for
community with other human beings”; the second, the “inclination to gain
worth in the opinion of others” and, based on this, “jealousy and rivalry” as “an
incentive to culture”; the third “is the susceptibility to respect for the moral
law as in itself a sufficient incentive to the power of choice” which shows human
beings to be responsible and responsive members of a community of free
persons or spirits (R 6:26–28).
1
Cf. G. Prauss, Kant über Freiheit als Autonomie (Frankfurt am Main: Vittorio Klostermann, 1983);
Henry Allison, Kant’s Theory of Freedom (Cambridge University Press, 1990); Philip J. Rossi, The
Social Authority of Reason: Kant’s Critique, Radical Evil, and the Destiny of Humankind (Albany: State
University of New York Press, 2005).
58
Radical evil and human freedom 59
1.2 Freedom and the causality of nature
Freedom is the power of finite rational beings to determine their own life
unconditionally by the moral law alone. Because the “ought” of their willing
(as persons) can in no way be derived from the “is” of their being (as rational
animals), humans by nature transcend the order of nature to which they
belong by living free lives among free persons.
For freedom to be possible, two critical differences must apply: the
practical difference between nature and freedom in our practical dealings
with the world; and the theoretical difference between phenomena and
things in themselves in our theoretical dealings with the world. Neither is a
conceptual distinction that could be used to classify phenomena of our
experience (phenomena of nature vs. phenomena of freedom, or phenom-
ena vs. noumena). Rather, they are orienting distinctions with respect to our
ways of relating to our world and experience. They are not features of the
world but ways in which we orient ourselves in the world. We are free only
by practicing our freedom, and we do so by orienting ourselves in terms of
the distinction between nature and freedom and phenomena and noumena
without mistaking these orienting distinctions for descriptive distinctions in
the world. As human beings we thus live in a web of natural events, which
determines us before we determine it; but as persons we have the capacity to
distance ourselves from the causal forces of this web and self-determine
ourselves freely by orienting toward the good in our ways of living and
acting in the world.
Kant follows tradition by calling the place in human experience of this
non-natural causality of freedom free will. For a free will to be possible, the
1st- and 2nd-person perspective of agents must in principle be irreducible to
a 3rd-person perspective on agents – free will “must be thought as altogether
independent of the natural law of phenomena. . ., namely the law of
causality” (KpV 5:29).2 This independence is not due to a metaphysical
duality of a natural world of causality (phenomena) and a supernatural
world of freedom (noumena).3 It is rather the independence of two irredu-
cibly different perspectives on our human life, the descriptive (or theoret-
ical) 3rd-person perspective of an observer, and the interactive (or practical)
1st- and 2nd-person perspective of participants. If this difference is negated,
the possibility of freedom collapses; if it is conceded, then there can be no
theoretical knowledge of the freedom of the will. The free will is no natural
2
“Phenomena” instead of “appearances.”
3
The distinction between “phenomena” and “noumena” is an orienting distinction and not a descrip-
tive distinction that could be used to describe different worlds or classify different sets of phenomena.
60 INGOLF DALFERTH
phenomenon and theoretical reason has neither anything positive nor any-
thing negative to say about freedom.
We must speak of will in order to be able to differentiate between events
and actions and causes and initiators, and of freedom in order to make the
difference between free actions and unfree actions meaningful. Both distinc-
tions are indispensable for making sense of our life-world experiences. Only
actions, not events, can meaningfully be attributed to responsible agents.
And only free but not unfree actions can reasonably be judged in moral
categories like good and evil, meritorious or blameworthy. While recourse
to causes explains something as an event, recourse to an initiator explains
something as a decision between options to act. However, it is one thing to
be an action, another to be a free action. An action is only free if it could have
been different not only in principle but also in fact, because its initiator
could have done something different than what he actually did. Actions
have initiators; free actions, free initiators.
1.3 Free will and good will

If human beings are thus seen as free initiators of actions, they come into
view not only as beings that have a will (i.e. could will one thing or another),
but are wills (i.e. can freely relate to their willing). As initiators of actions, we
are wills because we can effect changes in the world by actions that cannot
be sufficiently explained in terms of the causes of the events by means of
which they are executed. We effect not only events that have causes, but
actions that we induce as initiators.
However, this leaves the difference between free will (which assigns itself
to a purpose) and random arbitrariness (which follows any interests and
wants) unresolved. Kant further refines the ability to cause actions by
inquiring after the (second-order) ability to define one’s (first-order) ability
to act in a particular manner, i.e. to relate freely to one’s own ability to act
and to set purposes of one’s own acting. Only this second-order ability of
self-determination which Kant calls “autonomy” differentiates the will from
arbitrariness and shows it to be a free will, i.e. the ability “of producing
objects corresponding to representations or of determining itself to effect
such objects” (KpV 5:15).4 We have the ability not only to take decisions and
act (the power of arbitrariness), but to set ourselves options for decision and
act freely (the power of freedom).
4
Cf. also the corresponding definition of Begehrungsvermögen (“faculty of desire”), KpV 5:8–9.
That alone, however, still does not make the free will good. In addition, the
will must determine itself to be good. This is not the case if it wants to actualize
some specific “moral” option, but only when it wants to actualize all options
with which it concerns itself in a moral way. It is the mode of its willing, not its
object, that defines the goodness of the free will. Something is morally good
only if it is willed as the categorical imperative demands: “act only in
accordance with that maxim through which you can at the same time will
that it become a universal law” (G 4:421).5 By a “maxim,” Kant means “the
subjective principle of the volition” (G 4:401),6 i.e. that whereby someone
determines her will independent of all possible or actual objects to which it is
or could be directed so that her concrete willing and doing in various life
situations obtain a unified character. If these principles are good, then our
acting is good; if they are not good, then our acting is evil. Persons are thus
distinct in being able to live a morally good life as persons among persons, i.e.
to live their lives not merely in the animal way of striving for happiness but in
the human way of striving for the moral worthiness of being happy.
1.4 Autonomy
A will or a person7 is not autonomous by choosing between options (1st-
order freedom) but by choosing a way of choosing between options (2nd-
order freedom), i.e. of deciding for or against something in a particular
5
Compare with the differing forms of the categorical imperative: Julius Ebbinghaus, “Die Formeln des
kategorischen Imperativs und die Ableitung inhaltlich bestimmter Pflichten,” (1959) in H. Oberer
and G. Geismann (eds.), Gesammelte Schriften II: Philosophie der Freiheit (Bonn: Bouvier, 1988),
pp. 209–29; Friedrich Kaulbach, Immanuel Kants “Grundlegung zur Metaphysik der Sitten”:
Interpretation und Kommentar (Darmstadt: Wissenschaftliche Buchgesellschaft, 1996), pp. 94–100;
G. Geismann, “Die Formeln des kategorischen Imperativs nach H. J. Paton, K. Reich und
J. Ebbinghaus,” Kant-Studien 93 (2002), pp. 374–84.
6
Emphasis from the original. Cf. also KpV 5:19–21. Cf. Rüdiger Bittner, “Maximen,” in G. Funke and
J. Kopper (eds.), Akten des 4. Internationalen Kant-Kongresses (Berlin: de Gruyter, 1974), pp. 485–98.
7
Hereafter, “will” is always replaceable by “person.” The “capacity of autonomy” is not one of the
abilities of the animal rationale, which the person is, but rather evidence that this animal rationality
and rational animality lives as a person in the horizon of an Ought that cannot be derived from being a
human animal, but rather is applied to it so that the person can control living her humanity in a good
or evil manner. This point will be missed if one understands free will only as an ability for deciding,
more or less “uncoerced, intentional, and voluntary,” for or against something, as Timothy Duggan
and B. Gert, “Free Will as the Ability to Will,” Noûs 13 (1979), pp. 197–217, argue. This is also the case
when Fred Berthold Jr., God, Evil, and Human Learning. A Critique and Revision of the Free Will
Defense in Theodicy (Albany, NY: State University of New York Press, 2004), determines free will as,
following Augustine, a “mental act” (pp. 49ff.) that “must be intentional” (pp. 50–57), “uncoerced”
(pp. 57–59), “voluntary” (pp. 59–62), and “autonomous” (pp. 62ff.). This reduces the implementation
of autonomy to a mental act that actually permits “no absolute and sharp difference between the
human and the animal world” (p. 96), because he is describing a biological ability and not the human
as a person.
62 INGOLF DALFERTH
manner, and it is this manner and not that which he wants or wills that
makes him either good or evil.8 Whether a will is good or evil is a contingent
question depending upon its concrete moral decision. It must be one or the
other because one cannot concretely will anything without willing in a
morally relevant manner. Every concrete will, by its very willing something,
is at the same time determining itself as good or as not good. The practice of
the free will, therefore, does not have to be good, but is either good or bad. It
is only good if it wills what it wills as the categorical imperative prescribes.
For, according to Kant, “It is impossible to think of anything at all in the
world, or indeed even beyond it, that could be considered good without
limitation except a good will ” (G 4:393). True autonomy is not self-
determination per se, but rather self-determination to the good, and good is
quite formally everything and only that which one so wills that it furthers
the autonomy of others freely to determine themselves to promote the
autonomy of others.
This has three important implications. First, no one is free to will
simultaneously both good and evil: either the will is good or it is not, but
it cannot be both in the same respect and at the same time. Second, no one
is free to will neither good nor evil: autonomy is not a faculty one can have
without practicing it. Third – and this is the central idea of Kant’s doctrine
of radical evil9 – no one is free to decide for the good or for evil, i.e. free to
will either to will the good or not to will it (i.e. to will evil). We do not
determine ourselves in a morally neutral way to will good or evil, but we
determine ourselves to the good or we do not. Just as there is no self-
determination that is not already a self-determination to good or to evil, so
there is no moral self-determination to abstain from every moral self-
determination that was not a moral self-determination already.
Thus, we are not free to be free or not to be free. We cannot freely
determine our will not to will to be a free will. But this does not mean that
we could not abuse our autonomy by determining ourselves not to good but
rather to evil. Just as we are not abstractly free to choose between good and
evil, so we are not free not to be autonomous. For not only when we decide
against the good do we freely determine ourselves, but rather also when we do
not decide for the good: because we are destined for autonomy, we cannot not
live autonomously because when we do not decide freely for the good, then
we in fact freely decide for the opposite of the good, i.e. determine ourselves
8
Cf. Herbert Rommel, Zum Begriff des Bösen bei Augustinus und Kant. Der Wandel der ontologischen zur
autonomen Perspektive (Frankfurt am Main: Peter Lang, 1997).
9
Jacob Rogozinski, “Kant et le mal radical,” Kriterion 39 (1998), pp. 7–21.
for evil.10 For Kant, it seems impossible that we decide for evil for evil’s sake:
this is a possibility only of the devilish will11 but not for us because we always
decide for something that we take to be a good (to will something is to will a
“good”) even though we may be wrong. However, it is not only a possibility,
but a reality that we do not decide for the good but precisely thereby for the
evil: we determine ourselves as evil not by deciding for evil but rather by not
deciding for the good. There are not two decisions at stake here (for or against
the good vs. for or against the evil), but rather only one decision insofar as in
uno et eodem actu either the good will be chosen (and therefore not the evil)
or the good will not be chosen (and therefore the evil will be).
2 Kant’s account of radical evil
2.1 The moral dilemma of radical evil

This issue is the key to understanding Kant’s contentious view of radical
evil.12 Persons distinguish themselves as beings free from natural determi-
nacy and the causal nexus of events by freely willing the good. Yet that is
exactly what they do not do. We could and would be good if we only
wanted what the free will wills in its orientation to moral law. But since we
10
Mark Coeckelbergh, “Can We Choose Evil? A Discussion of the Problem of Radical Evil as a Modern
and Ancient Problem of Freedom,” in D. E. Keen and P. R. Keen (eds.), Considering Evil and Human
Wickedness (Oxford: Inter-Disciplinary Press, 2004), pp. 339–53.
11
Sharon Anderson-Gold, “Kant’s Rejection of Devilishness: the Limits of Human Volition,” Idealistic
Studies 14 (1984), pp. 35–48; Pablo Muchnik, “Radical Evil: Between the Trivial and the Diabolical,”
Contemporary Philosophy, 23 (2001), pp. 43–47; I. Rangelov, “Ideology In Between Radical and
Diabolical Evil: Kant’s Ethics of the Real,” Philosophy, Sociology and Psychology 2 (2003),
pp. 759–68; I. Samte-Porat, “Satanic Motivations,” Journal of Value Inquiry 41 (2007), pp. 77–94;
Matthew Caswell, “Kant on the Diabolical Will: A Neglected Alternative?” Kantian Review 12 (2007),
pp. 147–57.
12
Cf. T. Kadowaki, “Das radikal Böse bei Kant,” (Diss. Bonn, 1960); Olivier Reboul, Kant et le problème
du mal (Montreal: Les Presses de l’Université de Montréal, 1971); Christine Korsgaard, “Kant on
Dealing with Evil,” Philosophy and Public Affairs 15 (1986), pp. 420–40; Christine Schulte, Radikal
Böse. Die Karriere des Bösen von Kant bis Nietzsche (Munich: Walter Fink, 1988); Gordon
E. Michalson, Fallen Freedom: Kant on Radical Evil and Moral Regeneration (Cambridge University
Press, 1990); Emil L. Fackenheim, “Kant and Radical Evil,” in R. Chadwick (ed.), Immanuel Kant:
Critical Assessments. Vol. III (London: Routledge, 1992), pp. 259–73; Heinrich P. Lichtenberg, “Über
die Unerforschlichkeit des Bösen nach Kant,” Studia Philosophica 53 (1993), pp. 117–31; Otfried Höffe,
“Ein Thema widergewinnen: Kant über das Böse,” in Höffe and A. Pieper (eds.), Über das Wesen der
menschlichen Freiheit (Berlin: Akademie Verlag, 1995), pp. 11–34; S. R. Grimm, “Kant’s Argument for
Radical Evil,” European Journal of Philosophy 10 (2002), pp. 160–77; Richard Bernstein, Radical Evil:
A Philosophical Interrogation (Cambridge, MA: Polity Press, 2002); E. Willnauer, Heute das Böse
denken. Mit Immanuel Kant und Hannah Arendt zu einem Neuansatz für die Theologie (Berlin:
Rhombos, 2004), pp. 53–166; Pablo Muchnik, Kant’s Theory of Evil: An Essay on the Dangers of
Self-love and the Aprioricity of History (Plymouth: Lexington Books, 2009); J. Madore, Difficult
Freedom and Radical Evil in Kant: Deceiving Reason (London / New York: Continuum, 2011).
64 INGOLF DALFERTH
do not do this, we are, in what we will, not good, but evil. For a will that
could and should be good, but in fact is not, is evil because it neglects its
own potential for the good.
However, if we could freely choose between good and evil and do not
choose the good, then the question is why we freely decide for evil and not
for the good. If, on the other hand, we are not free either to choose between
good and evil or not to choose but must choose, and if in each concrete
situation we find ourselves to have chosen evil, then the question is how we
can be held responsible. If it is a fact that we are evil and will evil, then this is
either a result of our moral self-determination, for which we are responsible,
or it is not, and then we cannot be held morally accountable.
We are thus faced with the dilemma of either not knowing why we freely
decide against the good (and thus for evil), or not knowing whether the fact
that we are evil is the result of our moral self-determination or not (and in
that case it could not even be called “evil” in a strict sense).13 If we are evil,
then it is on the grounds of free self-determination, or we are not evil.
However, on the grounds of free self-determination we could only be evil if
we could also have been good. But not only would a situation in which we
could freely choose between being morally good and being morally evil be
pre- or extra-moral – because it itself would stand not under the determi-
nation of the good or evil will – it would also then still not be clear why we
determine ourselves not to the good but to evil. A moral determinateness
cannot be the result of a non-moral determination; rather, a moral deter-
minateness to evil can only be the result of a moral determination to evil.
For Kant, if evil can “have only originated from moral evil,” then there is
“no conceivable ground . . . from which moral evil could first have come in
us” (R 6:43).
Kant develops the dilemma in his account of radical evil. For every
moment of her life, a person can ascertain that she wills evil without
being able to explain this through appeal to a free decision in the past
between the will for good or for evil. Thus, the regression in time does not
help to explain the fact of evil will. “Every evil action must,” on the contrary,
“be so considered, whenever we seek its rational origin, as if the human
being had fallen into it directly from the state of innocence” (R 6:41). We
can clearly state wherein moral evil consists: it follows from maxims of
action that are not oriented to the moral law of practical reason, in which,
rather, ends other than willing the good decisively determine the willing and
13
Cf. Pablo Muchnik, “Kant on the Sources of Evil,” in Proceedings of the 10th International Kant
Congress (Berlin: de Gruyter, 2009), pp. 287–97.
the action. However, why this is so remains inexplicable for Kant: we can
only state the fact of evil. To be sure, we can explain it as the result of a
morally wrong determination of the maxims of our will, but why this is so
we are not capable of explaining. The fact of the morally wrong determi-
nation of our will remains as inscrutable for us as the fact of freedom14 which
makes us aware of our free will.
2.2 Incentives of animality, humanity, and personality

From a moral point of view, all human willing is either good or evil, and
there is none that is neither. Our willing is good if it orients itself by the
moral law and not by some other end of our human nature. However, the
alternative incentives of sensual humanity (rational animal) and of morality
(personality) are not a strict either/or. Kant assumes that: (1) the actual
incentives of our willing include the ends not only of our animal and
rational nature but also of morality; (2) there is no further kind of incentive;
and (3) in each concrete case our willing is determined through both kinds
of incentives. For each human being possesses three original predispositions
that belong to the very possibility of such a being, namely the predisposi-
tions to animality, humanity, and personality (R 6:26ff.).15
Kant thus understands persons as rational animals whose behavior is
determined in both a sensuous and a rational sense. We are not only what
we are in comparison with other animals (reasoning animals), but we are also
what we, as such animals, could and should be, i.e., persons. The predis-
position to personality is our aptitude for determining our willing and our
doing after the idea of humanity, i.e., in the orientation to the possibility and
obligation to be free and morally good beings.16 Human beings are not only
that which they are as reasoning animals in the wider context of life, but
14
In any case, this is the ground of the argument in KpV after the failure of efforts to deduce freedom
from the groundwork of a metaphysics of morals. The recourse to the facts of freedom supplants the
efforts of this itself as a principle deduced from a principle, all of which has led to no convincing
conclusion.
15
The “predisposition to the animality of the human being, as a living being,” comprises the urge to
“physical or merely mechanical self-love” (R 6:26), conducive of self-preservation and the preservation
of species through procreation. The predispositions “to humanity” exist in the ability of reason to
judge oneself “in comparison with others . . . [as] happy or unhappy” (R 6:27). From this arises the
desire to compete against others in the production, safeguarding, and enhancement of one’s own well-
being, but this is nothing other than the concretion of (biological) self-love that not only inclines
toward preserving life, but toward shaping life as happily as possible. The “predisposition to person-
ality,” finally, exists in the “susceptibility to respect for the moral law as of itself a sufficient incentive to
the power of choice” (R 6:27–28).
16
Cf. Matthew Caswell, “The Value of Humanity and Kant’s Conception of Evil,” Journal of the History
of Philosophy 44 (2006), pp. 635–63.
66 INGOLF DALFERTH
they are always also those who they want to be in this context. We are who we
are not independent of the way in which we understand ourselves, and of
the ends to which we determine ourselves.
Our will is therefore constantly determined through two kinds of moti-
vating forces: the biological drive of animalistic self-love aiming at self-
preservation (being, Dasein) and well-being (being happy), and the moral
drive to determine oneself through the orientation to the idea of humanity,
in willing and acting as a free person (being good).17 Only if we incorporate
respect for the moral law as a motivating force into our maxims will we
live not only biologically as rational animals, but also morally as self-
determining persons.
However, as humans, we live in such a way that even where we do not
expressly decide between good and evil, we in fact decide: whoever does
not determine himself to the good, but rather only displays his interest in
self-preservation and well-being, has thereby determined himself to evil,
because he fails to realize his predisposition to personality as it can and
should be realized in an orientation to the good. Kant’s point here is a
purely formal one: there is no possibility for human beings to live as
morally neutral. Whoever lives realizes in one and the same act his predis-
positions to animality, humanity, and personality. No person lives only as
an animal, or even as only a rational animal, but always also as a person –
be it by realizing the predisposition to personality as it should and can be,
or by neglecting it.
The idiosyncrasy of personality, as opposed to animality and ration-
ality, is that no human being is a person without living as a person of his
own accord, that no one can do this without living as a good or evil person,
and that no one who lives a human life can avoid living as a good or evil
person. No one is a person as he is a rational animal: namely, as a product
of an evolutionary history in which he became what he is. At most, one
can say that evolution made it possible for us to become persons − i.e., to
experience and express ourselves in a 1st- and 2nd-person perspective and
not merely be described in a 3rd-person perspective. One becomes a
person only by living as a person in communicating and interacting
with other persons: to be a person is to be responsible to other persons.
Everybody who is a rational animal can become a person by living freely –
that is, not only uncompelled from outside (by others) and not coerced
from inside (by desires, needs, or wishes), but also by determining the will
17
Were the will of people not always determined through both incentives, the moral law would not have
to appear in the form of the categorical imperative.
in such a way that one lives as a good or evil human person among
persons. For, just as the free will is that which makes me free when I
practice it, so the good will is the use of the free will that makes me good by
deciding freely for the good and against the evil.
2.3 Evil will

Our will is evil if respect for the moral law is not the primary incentive in the
determination of our willing and acting − i.e., if we do not give respect for
the moral law priority among our incentives in the formation of maxims,
but rather subordinate it to the incentive of self-love.
However, we are not only able to orient our wills to both incentives, but
absolutely cannot avoid doing so because as human beings we are not only
sensual and rational beings but also accountable and responsible beings.
These incentives cannot determine our willing in a balanced or parallel
manner, but rather only in a system of subordination. Either we subordi-
nate the actualization of our moral personhood to the realization of our self-
love, or we subordinate our self-love to the respect for the moral law. In the
first case, the pursuit of happiness directs our efforts to preserve our life and
improve our well-being, i.e., to live as happily as possible and to avoid
unhappiness as far as possible. In the second case, the duty to become
morally good persons obliges us to tailor our pursuit of happiness to the
right to happiness also of others, and to insist on our common obligation to
make a commitment to personhood, personal freedom, and the free deci-
sion for the good the decisive determination of our willing and acting. The
decisive moral principle is not the greatest happiness of the greatest number
or the smallest unhappiness of the smallest number, but rather the duty to
allow and enable others and ourselves to be the persons that as human
beings we ought to be and can be.
In a morally relevant sense, one cannot speak of evil within the horizon of
human self-love as the determining force of our life as rational animals, but
only within the horizon of respect for the law as the principle of human
personhood. Just as good without limitations is solely a good will, so evil
without qualifications is solely an evil will. Yet one does not have to will evil
for the sake of evil (as a devilish will does; R 6:36) in order to be evil. It is
enough, rather, not to will the good through subordinating the incentive of
duty to the incentive of self-love. For then one fails to be the person one
should and could be.
For Kant, such a failure is not only a possibility, or a merely occasional fact
of human existence, but a reality that characterizes every single human being
68 INGOLF DALFERTH
in every concrete situation of his or her life. As such, it is true of the whole
human species. “The human,” concludes Kant, “is evil by nature” (R 6:32).18
“He is evil by nature” simply means that being evil applies to him considered
in his species; not that this quality may be inferred from the concept of his
species ([i.e.] from the concept of a human being in general, for then the
quality would be necessary), but rather that, according to the cognition we
have of the human being through experience, he cannot be judged otherwise,
in other words, we may presuppose evil as subjectively necessary in every
human being, even the best. (R 6:32)
Human beings are evil not by necessity, but by nature, i.e., in fact.19 It is
not a self-contradictory idea to think human beings to be other than evil:
human beings are evil, but they do not have to be. But then one cannot
comprehend the evil of human beings from the concept of a human being,
but rather must discern it through experience: that human beings are evil
cannot be deduced from principles, but can only be established as actual fact.
In the empirical reality of human living, however, one cannot avoid
perceiving this. Wherever there is a “manifestation of the exercise of freedom
in the human being,” there we can “detect” its corruption (R 6:38).20 It
consists not only in the fact that some people always do evil, or every human
being does in one situation or another. The point of this corruption is rather
that people always and everywhere act according to an evil ordering of their
maxims, that is, determine their wills through principles that are not freely
orientated to the good, but owe their force to the incentive of self-love. People
orient themselves to a basic principle that makes not their personhood, but
the needs of their rational animality, into the guide of their willing and acting.
This pursuit has an effect not only in some, but in all situations of their acting,
and hence not only sometimes, but always. Whoever in his willing and acting
follows this immoral ordering of his maxims is, therefore, not only evil, but
cannot of his own accord stop being evil.21
18
Cf. C. H. Siegfried, "The Radical Evil in Human Nature," in Kopper and Funke (eds.), Akten des 4.
Internationalen Kant-Kongresses, pp. 605–13; E. Cherkasova, “On the Boundary of Intelligibility:
Kant’s Conception of Radical Evil and the Limits of Ethical Discourse,” Review of Metaphysics 58
(2005), pp. 571–84; C. Atkinson, “Kant on Human Nature and Radical Evil,” Philosophy and Theology
19 (2007), pp. 215–24; P. Formosa, “Kant on the Radical Evil of Human Nature,” Philosophical Forum
38 (2007), pp. 221–45.
19
Kant’s use of the phrase “by nature” is ambiguous. In the present context it does not refer to the
essence (or nature) of human beings but the reality of their existence.
20
Cf. Seiriol Morgan, “The Missing Formal Proof of Humanity’s Radical Evil in Kant’s ‘Religion,’”
Philosophical Review 114 (2005), pp. 63–114.
21
P. J. Rossi, “Kant’s ‘Metaphysics of Permanent Rapture’: Radical Evil and the Unity of Reason,” in
Sharon Anderson-Gold and Pablo Muchnik (eds.), Kant’s Anatomy of Evil (Cambridge University
Press, 2010), pp. 13–32.
Kant’s argument is complex. First, he assumes: (1) that all human willing
and acting are determined through the incentives of sensuality (the striving
for happiness) and morality (the striving for virtue); (2) that there is no other
determining ground aside from these; and (3) that there is no willing or
acting that is not thus determined. Second, he assumes with respect to
morality that: (1) every maxim of acting, i.e. every subjective principle of
willing, is “either morally good or morally evil ” (R 6:22); (2) there is no third
possibility of determining one’s willing; and (3) there is also no maxim that
would not be either morally good or morally evil – for it “is of great
consequence to ethics in general . . . to preclude, so far as possible, anything
morally intermediate, either in actions or in human characters” (R 6:22).
Whatever is not good is evil, and there is no gradation between willing good
and willing evil.
What applies for every individual maxim also applies for the whole
structure or framework of maxims of a person. Either that structure is
good because the individual maxims are good and all maxims stand in a
good order (i.e. are oriented toward the realization of moral good), or it is
evil because this is not the case. In the latter case, however, the person is not
only evil, but radically evil because he so determines his will that he not only
actually fails to do the good but can no longer act according to the good
because he accepts a general rule of willing that precludes this. Whoever
determines his willing by the principle of self-love and not of morality can
no longer undo this by his own willing and hence is not only possibly, but
actually, evil. This “evil is radical, since it corrupts the ground of all
maxims”; and “as natural propensity, it is also not to be extirpated through
human forces” (R 6:37).
Redemption from evil can, therefore, in no way be thought of as self-
redemption − for this would imply that we of our own accord choose
maxims that make the good that to which we orient our willing and acting.
However, this appears impossible, because “this could only happen through
good maxims – something that cannot take place if the subjective supreme
ground of all maxims is presupposed to be corrupted” (R 6:37). If we have
determined our will in a way that makes us evil, then we are evil, and, if we
are, then on our own accord we can at best act legally, but not live morally.
2.4 Radical evil

For Kant, therefore, human beings are characterized through the double
determination to have a predisposition to good and to be evil by nature: they
can and should be good, but in fact they are evil; however, without the
70 INGOLF DALFERTH
predisposition to good they could not be evil.22 This is Kant’s philosophical

reworking of the traditional theological view that we are creatures and
sinners and do not cease being creatures by being sinners (simul creatura
et peccator): being a sinner is different from being a creature, because it is
impossible for us to be a sinner and not to be a creature, but possible to be a
creature and not to be a sinner. The fact of being a creature and the fact of
being a sinner, therefore, require different explanations: the first is what we
are before we make ourselves (human creatures among other creatures), and
it is that which allows and requires us to make ourselves, which is to live our
human life consciously and intentionally as creatures among other creatures
in the presence of the creator. The second is what we become by making
ourselves, or, rather, by failing to make ourselves those we ought to be and
could become in relation to God’s presence. We in fact live as sinners even
though we need not have done so as creatures. In Kant’s philosophical
reworking of this position, human beings as human beings (“creatures”)
have a predisposition to good and hence a duty and capacity to become
persons by actualizing this predisposition, and yet they are evil by nature
(“sinners”) because, instead of actualizing this predisposition, they fail to do
so by not deciding to be good. Since they do so freely and of their own
accord, they are not in a position actually to become of their own accord
what they can and should be as persons.23
The factual determinateness of our will to will not good but rather evil, by
orienting ourselves essentially to self-love and not to the moral law, does not
cancel out, but depends on, our human predisposition to good. Only because
persons have this capacity – i.e., can live the life of the rational animal that
they are as the persons that they can and ought to be – can they actually be evil.
If we were not predisposed to determine ourselves to good, then we could not
be found lacking by having determined ourselves to evil. But if we were not
only actually but in principle determined to evil, then it would not merely be
impossible for us to perform the shift from evil to good ourselves, but not
possible at all: there would be no way for us to cease to be evil.
But is there a way? The situation seems to be paradoxical. On the one
hand, we must expect the turn to the good ab extra because we cannot
accomplish it ourselves in our actual situation, and this would be impossible
22
Pablo Muchnik, “An Alternative Proof of the Universal Propensity to Evil,” in Anderson-Gold and
Muchnik (eds.), Kant’s Anatomy of Evil, pp. 116–43.
23
This means conversely that the person indeed, “in principle, but in no way by nature, is good,” as
R. Wimmer, Kants kritische Religionsphilosophie (Berlin / New York: de Gruyter, 1990), p. 113, rightly
emphasizes. Cf. also Claudia Card, The Atrocity Paradigm: A Theory of Evil (New York: Oxford
University Press, 2002), pp. 73ff., which discounts the meaning of this point.
if human beings as such were not only actually evil, but could be nothing
other than evil. On the other hand, we must perform the turn to the good
ourselves for, unless we ourselves freely determine ourselves to the good, we
cannot be good: we must decide for the good ourselves because nobody can
make this decision for us. But how can we become ab extra what we must
become ourselves if we cannot turn ourselves from evil to the good?
This is the existential conflict in which persons find themselves according
to Kant. On the one hand they ought to be good and also can know that
they can be good because they ought to be so; but on the other hand they are
not able to be good because they actually are not good and also can know
that they cannot be of their own accord what they ought to be. The Kantian
difference between is and ought is thus deepened toward a divide between
the duty to be good and the inability to fulfill this duty. Persons live in the
tension not only between sensual humanity (determination by nature) and
morality (determination through respect for the moral law), but also
between their predisposition to the good in principle and their actual failure
to achieve this due to their lives as evil-doers.
It follows, first, that a person never is good, but always has some way to go
to become good, and, second, that this journey to the good is never a
journey that begins from a morally neutral place where one can decide
between an original orientation to good or to evil, nor from a place in which
one had already decided for the good. Instead, it is always a journey from a
place in which we live as we should not live if we want to do justice to our
determination to personhood. The goal of the good is thus always to be
attained from the starting position of evil. The life of moral evil results from
the corrupt orientation of our maxims of willing and acting. Therefore,
there is no way to the good that a person could reach by his own accord. No
action can make her good if she is not already good.
Kant has just as little to say about why this is so as he has about why a
person is free and predisposed toward a decision for the good. Both are facts
that can be known but not be explained in terms of something else. There is
no definite action that makes persons evil − rather, it is the general way in
which they act that attests to their orientation to self-love and not to moral
law. And it is no definite action that proves persons to be free, but it is the
fact that they are able to act at all and to orient themselves not only to what
there is but also to what ought to be that manifests their freedom. Just as the
freedom of a person cannot be deduced from any more basic principle, but
only stated as a fact, so also with the fact of the evil will. We can know the
reason why we are evil: we have subordinated the incentive of respect for the
moral law to the incentive of self-love. However, the origin of “this
72 INGOLF DALFERTH
disharmony in our power of choice with respect to the way it incorporates

lower incentives in its maxims and makes them supreme” (R 6:43) we
cannot elucidate: there is “no conceivable ground for us, therefore, from
which moral evil could first have come in us” (R 6:43).
2.5 Shadow of freedom

Kant’s discussion of evil arises from his conception of freedom. Just as for
Leibniz evil is the unavoidable shadow of contingency,24 so for Kant evil is
the inescapable shadow of freedom. If there is freedom (which does not have
to be), then there is also evil (which does not have to be), and if there is
freedom only as an unexplainable fact, then there is also the accompanying
shadow of evil only as an inconceivable fact.25 That is to say, there does not
have to be evil, but there is; and there could not be evil if there were no free
will, but there is also. The free will is the explicit form of freedom in human
life that makes both good and evil possible and conceivable. In its terms, we
can explain why there can be evil and what evil is, but not why there actually
is what can be but does not have to be.
If one proceeds from the reality of the free will, then to overcome evil is
not merely to do away with evil, but to disclose the good. Kant constructs a
complete alternative within the horizon of the free will: the will is either
good or evil, but never is it also neither one or the other. Based on our
experiences in the world we may wish to avoid speaking of evil and good and
to differentiate only between pleasant and unpleasant, happy and unhappy,
lucky and unlucky. For Kant, we would then confine ourselves to placing
human beings as rational animals in the context of life at large in which they
participate in their passions, actions, and reactions by seeking to maximize
pleasure and minimize displeasure. This approach does not go beyond
conceiving human beings as part of the biological and psychological pro-
cesses of nature. Such a view is possible because human beings possess
emotions, reason, and will − namely, the emotions to turn toward pleasure
and avoid displeasure, the reason to gauge rationally between various
options for action, and the power of choice to act one way or another.
24
Ingolf Dalferth, Malum. Theologische Hermeneutik des Bösen (Tübingen: Mohr Siebeck, 2010).
25
Fabio Ciaramelli, “Du mal radical à la banalité du mal,” Revue Philosophique de Louvain 93 (1995),
pp. 392–407; Henry Allison, “Reflections on the Banality of (Radical) Evil: A Kantian Analysis,” in
Allison, Idealism and Freedom: Essays on Kant’s Theoretical and Practical Philosophy (Cambridge
University Press, 1996), pp. 169–82; Pablo Muchnik, “On the Alleged Vacuity of Kant’s Concept
of Evil,” Kant-Studien 97 (2006), pp. 430–51.
Yet, even so, a human being still does not come into view as a person
capable of freedom who does not make her self-interest into the standard for
her actions but can instead subordinate self-interest to an orientation to the
good. She is able freely to relate to her capacity to choose between pleasure
and displeasure. It is not necessary for us to pursue that from which we
expect pleasure, but we can do it or refrain from it. And it is not necessary
for us to avoid that from which we expect no pleasure, but we can do the one
or the other. Self-love, self-interest, and usefulness can indeed, but need
not, determine us. We are free with respect to them to allow them to
become effective or not. We do not need to follow our animal drives or
rational considerations, but rather can distance ourselves from them. And if
we do one thing or another, we can assess this in one way or another,
approve and accept it or reject and avoid it.
This viewpoint reveals freedom to be more than choosing between given
options in order to maximize pleasure and to minimize displeasure. How we
live is not only conditioned through our emotions, reason, and will, but also
determined through our conscience. We can be held responsible for our
behaviors as rational animals; we can distinguish ourselves from that which
we wish and do. We can also do something that we do not want of our own
accord, not through external compulsions, but from inner insight alone,
because we ought to do it. We are in a position to co-determine that which
conditions our behavior by determining ourselves to behave in one way or
another.
These are familiar experiences which are not sufficiently accounted for by
viewing us as rational animals. Experiences such as these provide the basis
for Kant’s talk of human beings as “personalities” or “persons.” Whoever
considers human beings only from the viewpoint of their animality, emo-
tionality, and rationality does not see them as persons. However, if one
concentrates on the personhood of human beings, then one must speak of
free will, and if one speaks of free will, then – according to Kant – one
cannot avoid speaking of good and evil. Both mutually require each other:
there is no freedom of the will without the moral alternative between good
and evil, and vice versa. In short, good and evil are not opposite ends of a
single continuum and do not shade into one another or come in different
degrees, as do pleasure and displeasure.
2.6 Nevertheless and in spite of, against and instead of

Kant’s experience of freedom is thus an experience of “nevertheless” and “in
spite of”: we are not only bound to the pointless and bleak course of nature
74 INGOLF DALFERTH
to which we belong as animals capable of reason. If that were the whole

truth of human life, then people would be subjected
to all the evils of poverty, illnesses, and untimely death, just like all the other
animals on earth, and will always remain thus until one wide grave engulfs
them all together (whether honest or dishonest, it makes no difference here)
and flings them, who were capable of having believed themselves to be the
final end of creation, back into the abyss of the purposeless chaos of matter
from which they were drawn. (KU 5:452)
But we are acquainted not only with that prospect, but also with the
counterfactual experience of the “not so,” the “it could be otherwise,” the
“ought,” the conscience inviting us to live otherwise, the freedom not to be
a slave to the chain of natural processes but to be able to initiate new chains
of processes, and to do this in ways running contrary to our self-love and
self-interest as well as to purely utilitarian considerations. Despite our
inclusion in the correlation of natural processes, we are free to relate to
them in this way or another. We do not thereby leave the natural world;
rather, we are the location within nature where an order of freedom is
disclosed that is different from the natural order and cannot be reduced to it.
We experience this different order as resistance against and objection to the
natural conditioning of our lives.
This counterfactual character of freedom, to be freedom from and against
and in distinction to natural necessity, also characterizes Kant’s conception
of the moral good. We are good if we orient our wills to good, against our
natural propensity toward evil, and determine ourselves to good. No one
becomes good simply as a result of opting for the good, but rather he must
bring good to bear against evil. At no time can we directly and readily
choose the good. Instead, we can at all times obtain it only through the
“overcoming” (überwiegen: R 6:37) of actually being evil. We are not good,
but we must become so. We become so, however, not simply as a result of
willing it − rather, it must be achieved against the opposition of the reality of
evil, and this reality impedes the good not merely in others but also in
ourselves.
On this point Kant was substantially more clear-sighted than his opti-
mistic contemporaries. If being good is always the result of self-determining
our will to the good, then it is never to be obtained other than through
fighting against evil. If we were good from the beginning, we would not
have to determine ourselves to the good. If we are, however, not good from
the beginning, then we are evil from the beginning on, for not to be good is
to be evil (and not merely not good). Thus, in order to become good, we
must determine our will in a new and different way from its actual
determination, i.e. transform our evil will into a good will.
However, such a transformation is apparently impossible. A transforma-
tion of the will is not merely an alteration, but a self-alteration of the will.
Yet an evil will cannot determine itself differently of its own accord, since it
is evil precisely by not freely subordinating the incentive of self-love to the
incentive of duty. Only a free will that is not evil can become good through
self-determination. If it is evil, then it cannot transform itself into good, but
only become replaced or superseded through a good will. Thus, we must
transform ourselves by orienting our will away from evil and to the good
but, being evil – i.e., having freely decided not to be good – we cannot
transform ourselves in this way.
The transformation of the will is, therefore, not so much a conflict in one
and the same will between two different and incompatible ways, but rather a
conflict between two differently determined wills or moral identities. We
actually are evil yet it is possible for us to be good. But how can we actually
become what it is possible for us to be if we actually are what we should not
be? Solely, so it seems, by becoming another will – a will that determines
itself differently, namely to the good. Yet how this could happen remains a
mystery.
To be sure, we can describe the change that is required: where there was
previously an evil person, there is now a good person. There is a good person
only if she freely chooses to subordinate the incentive of self-love to the
incentive of the moral law. But how can she do so if she has freely chosen
not to determine herself to the good? Evil and good are incompatible self-
determinations that cannot co-exist in one and the same will or person, but
only as a conflict of different self-determinations in a succession of time in
one and the same life. But to be good in that case not only is the opposite of
being evil but is also to fight and to overcome evil. Moral goodness is a
success term: it is impossible to be morally good if one does not thereby
overcome being evil.
2.7 The right to hope

As evil persons, we cannot make ourselves good by overcoming our being
evil. All we can do is to fight evil by determining ourselves contrary to our
actual inclinations. We can do so because, as evil persons, we are both free in
principle to determine ourselves by the good (predisposition to good) and
we are not free in fact to determine ourselves in this way (natural propensity
to evil). We experience this as a conflict of moral orientation from which we
76 INGOLF DALFERTH
cannot escape. The struggle takes place within ourselves, and it is a struggle
that we cannot be sure to win.
Consequently, we can fight evil but we cannot make ourselves morally
good. We can only hope that in the long run evil will be overcome and
replaced by a re-orientation of our will toward the good. Moral self-
determination is a creative process in time in which not merely something
that was already there becomes determined or regulated in a certain way,
but something that did not previously exist comes into being. Through
moral self-determination, we freely make ourselves into that which we then
actually are: an evil person or a good person. For evil persons to become
good, they have to alter their incentive structure and give priority to the
moral incentive. They can do so as persons, but they cannot do so as evil
persons. The more we become aware of ourselves as persons (and not merely
as rational animals), the more we become aware of the moral conflict in
which we exist. We ought to become what we in fact are not. And being
what in fact we are makes it impossible for us to become what we ought to
be: we have the duty of becoming good by fighting the evil for which we
ourselves are accountable.
There thus emerge four fundamental mysteries which Kant’s standpoint
exposes but does not solve, their full extent finally exposed by his Religion
within the Boundaries of Mere Reason. First, there is the mystery that there is
anything at all, rather than nothing (ontological mystery). Second, there is
the mystery that we are not completely determined by natural causality but
capable of being free (mystery of freedom). Third, there is the mystery of
why we determine ourselves to be evil rather than good (mystery of evil).
And finally, there is the mystery of how we can overcome being evil and
become morally good (mystery of overcoming evil).
In each case Kant carefully outlines what it means to be, or to be free, or to
be evil, or to become good. But he also underlines that there is no explanation
of how it is possible that there is anything at all, or that we are free, or that
we choose to be evil, or that we can ever hope to become good. Yet, whereas
with respect to the first three mysteries Kant argues from the actuality to the
possibility of being, freedom, and evil, with respect to the fourth mystery he
can only argue from the possibility of being good to the possibility of
becoming good through overcoming evil. We are free, but we can become
good. It is not self-contradictory that we, as free beings, are good. But it is a
mystery how we, as evil beings, can determine ourselves to be good rather
than evil. It is possible for us to be free and good, but it is impossible to
explain how we, being evil, can become good. The moral end of our human
life is clear, but the way to achieve this end is shrouded in mystery. We can
be morally good, but we cannot tell how we can become morally good if we
begin in evil.
Kant saw clearly that he had no way of elucidating this mystery.
However, he argued that there is hope that what is possible to be will also
be possible to achieve even though we cannot tell how. To justify this hope
is impossible in purely anthropological terms (by reference to what we are
and can do) but requires the practical rational belief that we want there to be
a God who can make it possible for us to be morally good (i.e. worthy of
being happy) and to be happy, if not in this life, then in a life beyond.26 We
“must postulate the existence of God, as the necessary condition of the
possibility of the summum bonum” (KpV 5:124), and in this sense the
“concept of God . . . is one belonging originally not to physics, that is, to
speculative reason, but to morals” (KpV 5:140). For
admitting that the pure moral law inexorably binds every man as a command
(not as a rule of prudence), the righteous man may say: “I will that there be a
God, that my existence in this world be also an existence in a pure world of
the understanding beyond natural connections, and finally that my duration
be endless; I stand by this, without paying attention to rationalizations,
however little I may be able to answer them or to oppose them with others
more plausible, and I will not let this belief be taken from me; for this is the
only case in which my interest, because I may not give up anything of it,
unavoidably determines my judgment.” (KpV 5:143)
That is to say, the existence of God is not a theoretical belief (“God is”)
but the postulate of a practical belief (“God ought to be”). I cannot rationally
defend the theoretical claim that there is a God (nor can I rule it out
theoretically). But I have every right to the practical hope that there is a
God, because there is the moral law, the duty to be good, the possibility of
being good, and the need to overcome the actuality of evil by becoming
good. To safeguard this possibility I have a moral right to postulate the
existence of God – not as the ground of my moral obligations but as the
condition of the possibility of becoming actually what I can be potentially:
namely, a good person.27
26
Joseph P. Lawrence, “Radical Evil and Kant’s Turn to Religion,” Journal of Value Inquiry 36 (2002),
pp. 319–35.
27
According to Kant, the “postulate of the possibility of the highest derived good (the best world) is
likewise the postulate of the reality of a highest original good, namely of the existence of God. Now, it
was a duty for us to promote the highest good; hence there is in us not merely the warrant but also the
necessity, as a need connected with duty, to presuppose the possibility of this highest good, which,
since it is possible only under the condition of the existence of God, connects the presupposition of
the existence of God inseparably with duty; that is, it is morally necessary to assume the existence of
God” (KpV 5:125). However, this is not a theoretical but a practical assumption that does not furnish
78 INGOLF DALFERTH
So there is room for rational hope. For I am free – I do not have to be evil.
I know that it is possible for me to be good. I do not know how I can
become good because I am evil, yet I hope that there will be a way of
becoming actually what I can be possibly, a morally good person in the
community of morally good persons. To safeguard this possibility I postu-
late the existence of God as the condition of its possibility. I can be what I
ought to be. I cannot make myself morally good because I have chosen not
to be morally good but to be evil. However, it is possible to become what I
ought to be. And if the possibility of God is the condition of this possibility,
and the possibility of God is grounded in the actuality of God, then I have
every right to hope that there is a God because I am free. In short, the reality
of freedom justifies my rational hope in the existence of God who makes it
possible for persons to overcome their being evil by nature and orient
themselves toward the good. We can be sure that it is possible, and we
have every reason to hope that it will become actual. We do not know how.
But we do know that all we can and ought to do is to fight evil and try to
become good. For we are evil only in fact, but we have a predisposition to
good in principle. And we have every right to hope that the good will
prevail.
the basis of a natural or speculative theology. “[T]here is indeed a cognition of God but only with
practical reference, and if we attempt to extend it to a theoretical cognition we find an understanding
that does not think but intuits, a will that is directed to objects upon the existence of which its
satisfaction does not in the least depend (not to mention the transcendental predicates, as, e.g., a
magnitude of existence, i.e., duration, which, however, is not in time, the only possible means we
have of representing existence as magnitude). All of these are attributes of which we can form no
concept fit for cognition of the object, and we learn from this that they can never be used for a theory of
supersensible beings, so that on this side they are quite unable to ground speculative cognition and
their use is, instead, limited solely to the practice of the moral law” (KpV 5:137).
chapter 4
Gesinnung: responsibility, moral worth,

and character
Alison Hills
1 Introduction
In Kant’s account of morally worthy action, maxims play an important role.
Actions are based on maxims. The categorical imperative tests determine
whether a maxim is morally permissible or not. An action that is based on a
good maxim is dutiful, or, as Kant says, in accordance with duty. One which
has been chosen on that basis – because it is your duty – is morally worthy.
In the Groundwork and elsewhere, Kant also briefly mentions another
idea, that of a Gesinnung (often translated as a disposition, or “meta-
maxim”).1 He sometimes associates morally worthy action, or virtue, with
a moral Gesinnung.2 But he does not discuss in any detail what a Gesinnung
is or the role it is to play. In Religion within the Boundaries of Mere Reason,
however, the Gesinnung becomes much more prominent and Kant makes
some important and intriguing claims about it.
For instance, he says that the Gesinnung is ultimately a choice between
putting the moral law above self-love, and vice versa. An evil person puts
self-love over morality, a good person does the opposite (R 6:21–22, 25, 36).
Thus there are only two possible dispositions, and by choosing one over the
other you are either wholly good or wholly evil (R 6:22ff.). Kant also says
that we humans have some kind of propensity to choose evil: this is the
notorious claim that humans are radically evil (R 6:29ff.; 32ff.; 72).
These bold claims are a vital part of the moral philosophy of the Religion.
But before we can understand and evaluate them, we need to have a clearer
idea of what a Gesinnung is and the role it plays in Kant’s philosophy. And
these turn out to be good questions in their own right, helping us gain a
better understanding of core and fundamental Kantian claims about the
nature of action, of moral responsibility, and of morally worthy action.
1
G 406, 412, 416, 435; KpV 5:56, 75, 83–4, 99, 116, 147, 152–53; B775–78, B841–43, B857.
2
For instance, G 4:406, 416, 435; KpV 5:147.
79
80 ALISON HILLS
Finally, and perhaps most interestingly of all, whilst Kant does not of course
have a conception of character or of virtue at all like Aristotle’s familiar
ideas, the Gesinnung can become the foundation of a distinctively Kantian
conception of character.
Unfortunately, despite the fact that Kant does discuss Gesinnung in more
depth in the Religion than he does in earlier work, he leaves many issues
unsettled, and in fact raises several difficulties for his account which he does
not answer fully. Having indicated these, I will attempt to reconstruct the
best account of what a Gesinnung is and the most plausible explanation of its
ethical role, on the basis of Kant’s remarks about it in the Religion and
elsewhere.
I will begin by recapping in more detail the conception of action based on
maxims from the Groundwork and the Religion, in order to see where
Gesinnung might fit in.
2 Action on the basis of maxims

Despite the crucial importance of maxims in his account of action and of
duty and moral worth, Kant introduces them very briefly, explaining in a
footnote that a maxim is a “subjective principle of volition” (G 4.401).3 He
gives a number of examples of maxims, including:
from self-love I make it my principle to shorten my life when its longer
duration threatens more troubles than it promises agreeableness (G 4:422);
when I believe myself to be in need of money I shall borrow money and

promise to repay it, even though I know this will never happen (G 4:422);
I shall take nothing from him nor even envy him; only I do not care to
contribute anything to his welfare or to his assistance in need! (G 4:423).
The maxim on which you act can be tested by the categorical imperative
test. Maxims can be assessed as in accordance with duty, morally wrong, or
morally worthy (the right action, from duty) (G 4:402). For example, Kant
argues that we cannot will that the maxim “when I believe myself to be in
3
For helpful discussion of maxims and their role in Kant’s theory of action and of morally worthy
action, see Christine M. Korsgaard, Creating the Kingdom of Ends (Cambridge University Press, 1996);
Jens Timmermann, “Kant’s Puzzling Ethics of Maxims,” Harvard Review of Philosophy 8 (2000),
pp. 39–52; Talbot Brewer, “Maxims and Virtues,” Philosophical Review 111.4 (2002), pp. 539–72;
Richard McCarty, Kant’s Theory of Action (Oxford University Press, 2009); and Rob Gressis, “Recent
Work on Kantian Maxims,” Philosophy Compass 5.3 (2010),pp. 216–17.
Gesinnung 81
need of money I shall borrow money and promise to repay it, even though
I know this will never happen” should be a universal law, and so acting on
the maxim is contrary to duty.
From this, we can reconstruct some features of maxims. For instance,
they must be expressible as propositions. They are usually described as an
action (such as “shorten my life” or “borrow money and promise to repay
it”) together with a reason for that action (e.g. in order to avoid trouble, in
order to acquire money when I am in need). They appear to play a role in
action similar to intentions, and so it is not surprising that they have been
identified with intentions.4 But there do seem to be some important differ-
ences between intentions and maxims.
First, intentions can be highly specific or highly general. The descrip-
tion of maxims as “principles” suggests that Kant does not think of
them as being highly specific. Second, it seems possible to act on more
than one intention at a time, whereas it is implied in Kant’s discussions
that each action is based on one maxim, as the morality of the action
depends on that maxim (though he is not explicitly committed to this
claim).
Action on the basis of maxims is also importantly connected with free-
dom. In the Groundwork, Kant defines freedom as “the property of a will
that it can be efficient independently of alien causes determining it . . .
autonomy is the will’s property of being a law to itself . . .” (G 4.446–47).
And in the Religion the connection with maxims is made explicit:
Freedom of the power of choice has the characteristic . . . that it cannot be
determined to action through any incentive except so far as the human being
has incorporated it into his maxim (has made it into a universal rule for
himself, according to which he wills to conduct himself); only in this way can
an incentive, whatever it may be, coexist with the absolute spontaneity of the
power of choice (of freedom).” (R 6:24)
According to the so-called “incorporation thesis,” we act on an incentive

only when we freely choose to incorporate it into a maxim.5 Hence all
action on the basis of maxims is free, even when the maxim incorporates
a desire. When the maxim is based on the moral law, then we act
autonomously.
4
In particular, by Onora O’Neill, Constructions of Reason: Explorations of Kant’s Practical Philosophy
(Cambridge University Press, 1989), especially pp. 81–104.
5
The term “incorporation thesis” is from Henry Allison, Kant’s Theory of Freedom (Cambridge
University Press, 1990) and the idea is discussed at length there.
82 ALISON HILLS
3 What is a Gesinnung?
The introduction of Gesinnung complicates this account of action. As Kant
describes it, the individual maxims on which we act, such as “when I believe
myself to be in need of money I shall borrow money and promise to repay it,
even though I know this will never happen,” are themselves based on
another maxim (sometimes called a “metamaxim”), the Gesinnung: “a
common ground, itself a maxim, of all particular . . . maxims” (R 6:20).
Kant claims that each person’s Gesinnung is chosen – as it must be if we
are to be held morally responsible for maxims chosen on its basis – but that
it is not chosen at any one particular time (R 6:22, 25, 31). Kant suggests that
it is with us from birth (or at least from youth). Even more mysteriously,
your Gesinnung can change − that is, you can choose first an evil Gesinnung,
and then a good Gesinnung (R 6:44–52). This must be so, otherwise it would
not be possible for someone who had chosen an evil Gesinnung to become a
good person. Kant sees this, but he also makes clear that this is not some-
thing that he thinks he can explain (R6: 45).
What is the role of Gesinnung in Kant’s theory of action? One obvious
role is in the evaluation of a person. Kant is most interested in what it is to
be evil in the Religion. One natural answer is: an evil person is someone
who performs actions that are “undutiful.” But Kant is not satisfied
with this:
We call a human being evil not because he performs actions that are evil
(contrary to law) but because these are so constituted that they allow the
inference of evil maxims in him . . . In order then to call a human being evil,
it must be possible to infer a priori from a number of consciously evil actions,
or even from a single one, an underlying evil maxim, and, from this, the
presence in the subject of a common ground, itself a maxim, of all particular
morally evil maxims. (R 6.20)
Particular, “first-order” maxims can be based on an “underlying” maxim, in

this case evil. It is in virtue of having an underlying evil maxim that you are
an evil person, and, correspondingly, having an underlying good maxim
makes you a good person. The underlying maxim is the Gesinnung.
What role does the Gesinnung play in the evaluation of a person? Kant
does not give a great deal of explanation, but there are at least three clear
possibilities:
a Moral responsibility
Suppose that you borrow some money from a friend promising to pay her
back later, when you have no intention of doing so. This is a bad
Gesinnung 83
action, contrary to your duty not to make false promises. But suppose
that you made this promise only because you had been coerced to do
so. Or, more dramatically, suppose that you made the false promise
only because someone interfered with your brain ensuring that you did
so. In both examples, a good case can be made that you are not morally
responsible for what you have done. And if that is right, even though
your action is bad, this does not reflect badly on your character. That
you have performed a good or bad action is relevant to the moral
evaluation of you as a person only if you are morally responsible for
that action. To be morally responsible, you may need an underlying
maxim that grounds the maxims on which you act.
b Reasons for action
Suppose that you help someone in need. You are doing something good,
something that you have an imperfect duty to do. But suppose that your
reason for helping is that you want to get a good reputation from which
you hope to profit in the future. Though your action is good, you are
not a good person in virtue of doing that action, because your reason for
action was self-interested or at any rate was not a morally good reason. A
good or bad action is relevant to the moral evaluation of a person only
when we take into account that person’s reasons for action. To act for
good (or bad) reasons, you may need a good (or bad) underlying maxim.
c A conception of character
Suppose that you perform one undutiful action out of self-interest,
ignoring the moral reasons in favor of doing the right thing. Suppose
that you are morally responsible for your action. Are you a bad person?
Normally we would say: no. Similarly, if you perform one good action,
for the right sort of reasons, you are not necessarily a good person.
Judging a person as good or bad is not the same as judging an individual
action. In what way is it different? To judge a person is to judge that she
is the kind of person who performs good or bad actions, perhaps that she
regularly performs them, or typically does, or is disposed to performing
them. To be the kind of person who performs good (or bad) actions,
you may need a good (or bad) underlying maxim.
In the following sections, I discuss these three roles of the Gesinnung,
starting with the first.
4 Moral responsibility
Suppose that you made a false promise but only because someone
interfered with your brain. Then you have not freely chosen the maxim
84 ALISON HILLS
for which you act, and, in a sense, you have not performed an action at
all – or certainly you have not performed an action for which you are
morally responsible. A good case can be made that the same is true when
you are coerced: coercion is precisely a way of not allowing you a free
choice of maxim, hence you do not genuinely act when you are coerced –
or at least you do not perform an action for which you are morally
responsible.6
According to Kant, you are morally responsible for your action only if
your maxim was freely chosen. It may be, though, that, in addition, that
maxim must be chosen on the basis of an underlying metamaxim. There are
two slightly different arguments that could be made here.
The first argument emphasizes the choice of a maxim as an action.
According to Kant, all actions are based on maxims that are freely chosen.
But consider the action of freely choosing a maxim. It follows that this
action too must be based on a maxim. Not of course on the maxim which
you then choose, but on some underlying maxim– the Gesinnung.7
The second argument concerns the choice of a maxim as based on
reasons. Suppose that you are choosing your maxim. What is the basis of
your choice? Do you just “plump” for one rather than another? Not if this
choice is something for which you can be morally evaluated. If that is to
be true, your choice must be made on the basis of reasons. They must be
reasons to choose a certain sort of maxim, and these reasons must be
grounded in a maxim: a prior underlying maxim, a Gesinnung.
According to this theory, moral responsibility involves a hierarchy of
attitudes. There are first-order maxims, which are maxims to take particular
actions for particular reasons (“when I believe myself to be in need of money
I shall borrow money and promise to repay it, even though I know this will
never happen”), plus a metamaxim, which provides reasons for the choice of
a first-order maxim, on the basis of which some first-order maxim is
chosen.8
6
I think this is a bit more controversial, as it is not clear to me that Kant would agree that you are not
free to choose when you are coerced (it might depend on the form the coercion took). But in any case,
whether or not you are morally responsible for your action appears to depend on whether your choice
of maxim was free or not.
7
This is suggested by Kant’s reference to the metamaxim as a “ground” of individual maxims, e.g.
R 6:21, 25, 31.
8
Again, this is suggested by the references to the metamaxim as the “ground” of an individual maxim,
where the ground is understood as providing reasons (as well as, or instead of, a metaphysical
grounding), in particular where a good or bad Gesinnung provides the ground for good or bad
individual maxims, e.g. R 6:21, 31, 37.
Gesinnung 85
It is interesting to compare this theory of action with perhaps the best-
known hierarchical theory of moral responsibility, that of Harry Frankfurt.9
According to Frankfurt, free will requires an agent to have both first-order
desires and second-order volitions (which are a kind of desire), which are in
line, in the sense that you want certain first-order desires of your own to be
effective (i.e. that you want to put them into action), and those desires are in
fact effective. In Frankfurt’s famous example, an unwilling drug addict
whose desire to take the drug is put into action, but who wants that desire
not to be effective, does not have free will.
There are some obvious similarities between these two conceptions of
freely willed action. First, both involve a hierarchy of attitudes, in
Frankfurt’s case a hierarchy of kinds of desire, in Kant’s a hierarchy of
maxims. The different levels of each hierarchy are explained in different
ways, however. A second-order volition is a type of desire whose content
includes another desire. The Gesinnung does not seem to be a maxim which
contains first-order maxims as part of its content, rather it is a choice of
morality over self-love – or vice versa. Each person has only one Gesinnung
(at least, at any one time) but presumably may have a number of second-
order volitions.
In both theories, the higher-order attitudes are or should be in some sense
authoritative over the lower order. The higher-order attitude enables the
lower-order attitude to be formed on the basis of reflection and reason, and
so the lower order is supposed to comply with the higher-order attitude,
and, in a morally responsible agent, they do comply. The need for a ground
for one’s choices is present in both, for one to count as morally responsible
for one’s actions.
There is a highly significant difference between Frankfurt’s account of
moral responsibility and Kant’s, however. Though Frankfurt is not com-
mitted to the truth of causal determinism, his account of free will and of
moral responsibility is compatible with it. For Frankfurt, a free will is a will
that you want − that is, where you act on the first-order desires that you
want to be effective. It may be that you are causally determined to have a
will structured in this way. It even appears possible that your second-order
volitions or first-order desires could be manipulated by external forces – by a
curious scientist, for instance – so that you have a Frankfurtian free will.
You might have the kind of upbringing that ensures that you are not “sane,”
in Susan Wolf’s sense of the term: you have the wrong values and you are
9
Harry Frankfurt, “Freedom of the Will and the Concept of a Person,” Journal of Philosophy 68 (1971),
pp. 5–20.
86 ALISON HILLS
unable to change them.10 But you might still have a free will and be morally
responsible for your actions, according to Frankfurt.
Kant, in contrast, explicitly invokes free choice at each stage of his
account, from the choice of a Gesinnung to the choice of first-order maxims,
where this includes the freedom to choose differently. You could not be
manipulated into choosing a maxim or a Gesinnung, for if you were
manipulated, or if your “choice” were causally determined, you would not
have really chosen a maxim or metamaxim at all (R 6:25, 31, 32, 35).
Kant’s incompatibilism is both a large cost and a large benefit of his
account of moral responsibility. The benefit is in avoiding altogether the
possibility of manipulation resulting in moral responsibility. The cost is that
he is committed to our making choices that are not causally determined,
that is, noumenal choices. And these seem even more mysterious once we
have a hierarchical model. For I have to make two related choices – a
Gesinnung, and then a first-order maxim which is grounded in my choice
of morality or self-love. But what is this “grounding?” It cannot be that a
choice of Gesinnung causally determines the first-order maxim that I choose,
for then I would not have chosen that maxim freely. I will return to this
problem later.
There is another significant problem with this theory of moral responsi-
bility: the arguments sketched for it appear to lead to an infinite regress.
Suppose that, as we assumed, the choice of a maxim is an action, and all
actions must be based on maxims. Then the choice of a maxim like “when I
believe myself to be in need of money I shall borrow money and promise to
repay it” must be based on the Gesinnung. But the choice of a Gesinnung is
an action which must be based on a prior maxim – a meta-meta-maxim.
And the choice of that maxim must be based on a maxim, and so on. In
order to perform one action, we must perform infinitely many.
We can avoid the infinite regress by denying that the choice of a maxim is
a kind of action (or if it is, it is one that is quite unlike other actions, insofar
as it does not necessarily have to be based on a maxim). That seems to be a
possible response, though ideally we might like some positive account of
what choosing a maxim is. Except that, once more, we have no reason for
introducing the Gesinnung. For if the choice of a maxim need not be an
action, presumably the choice of individual maxims need not, and so there
is no need to posit a maxim for that action, and it seems that there can be
free, morally responsible action without a Gesinnung.
10
Susan Wolf, “Sanity and the Metaphysics of Responsibility,” in Ferdinand Schoeman (ed.),
Responsibility, Character and the Emotions (Cambridge University Press, 1987), pp. 46–62.
Gesinnung 87
The second argument also appears to introduce a regress. The choice of a
higher-order attitude needs a reason, and that reason must be based on an
attitude of a still higher order. This regress is acknowledged by Kant: “there
cannot be any further cognition of the subjective ground or cause of this
adoption, for otherwise we would have to adduce still another maxim into
which the disposition would have to be incorporated” (R 6:25).
Again there is an obvious difficulty. If it is possible to act for reasons
without choosing those reasons on the basis of other reasons, then presum-
ably an agent can do that when she chooses a maxim in the first place, and
there is no need for a metamaxim. If she cannot, then it is hard to see how
you could act for reasons at all, since a choice of action for reasons would
have to be supported by reasons for those reasons, and reasons for those
reasons, ad infinitum.
Other conceptions of moral responsibility that invoke a hierarchy of atti-
tudes face similar problems of course. Frankfurt’s account, for instance, has to
face the question of why freedom of the will requires that you form second-
order volitions with which your first-order desires align, but not third-order,
fourth-order, and so on. Any reason for requiring second-order volitions – for
instance, that freedom of the will requires that you reflect on your desires and
form a view about which you would like to act on, that is, a higher-order
desire – could apply equally well to the formation of second-order desires.
Frankfurt recognizes and addresses this problem. He claims that, though
it is possible to have volitions of third order or higher, it is possible for you to
identify wholeheartedly and decisively with one of your first-order desires –
this commitment “resounds” throughout the potentially endless higher
orders.
Despite clearly recognizing the potential for an infinite regress in his
argument for a Gesinnung, Kant does not fully explain how he thinks the
regress should be blocked. This is a fundamental problem, since it threatens
to undermine his entire theory of action.
He does, however, make some suggestive remarks, distinguishing the
“deed” of choice of metamaxim, from the “deed” which is the choice of an
individual maxim (R 6:31). Though this is not made explicit, one difference
between the two deeds must be that, whilst first-order maxims need to be
chosen on the basis of a further maxim, the metamaxim does not.
Why not? To explain this, we need to think again about reasons for
action. There is a difference between there being reason in favor of an
action – e.g. that it is morally required, or that it increases your happiness –
and your taking that consideration to be a reason for action. What is it for
you to take something to be a reason?
88 ALISON HILLS
It is a familiar Kantian thought that agency requires a certain amount of

consistency, and that one feature of a reason is that it is universal. So here is
one possibility: to take something to be a reason is for you to choose to
treat considerations of this kind to count in favor of (or against) action.
This both respects the universality of reasons − for these considerations
count as a reason in other circumstances as well as here and now − and
secures consistency in you as an agent, since you are committed to regarding
these considerations as reasons in other circumstances too.
It might seem as if this sort of choice could be made when you choose a
first-order maxim. For won’t that kind of maxim contain a ground (such as
“I will make a false promise to gain money”) which expresses that a
consideration will be a reason in this and other circumstances?
First-order maxims do contain grounds, but typically it makes sense to
ask of that ground: why do you take that to be a reason for action? For
instance, why take gaining money to be a consideration in favor of action?
Perhaps the answer is: so that I can buy things for myself. And why is that a
reason for action? Ultimately, Kant is committed to there being two basic
ways in which this sort of chain of reasons can end: with morality, or with
self-love. Morality is one fundamental end, because it is the law governing a
rational creature as such; self-love is another, because as creatures with
desires we always do set it as our end. Our fundamental choice is: to
which do we give priority − morality over self-love, or self-love over
morality?
It follows that, to choose a maxim like “I will make a false promise to gain
money” on the basis of reasons, you must have chosen to prioritize self-love
over morality. That is, you must have chosen a metamaxim (specifically, an
evil one).
Your choice of a metamaxim must be different. It is impossible for you to
make that choice on the basis of reasons. This is because in choosing a
metamaxim, you are choosing what will count as a practical reason for you.
Since this determines what will count as a practical reason for you, it cannot
be made on the basis of prior reasons (though this does not rule out your
recognizing that the moral law is a law of rational agents as such). Reflection
afterwards may support that choice – since I now take moral reasons to be
more significant than reasons of self-love, I endorse the decision to give the
moral law priority over self-love. That does not mean that the original
decision was made for a reason after all.
This picture makes how you can change your metamaxim seem myste-
rious. Naturally, this change cannot occur on the basis of what you at first
take to be a reason. From that perspective, a change in commitment
Gesinnung 89
therefore must appear unreasonable and perhaps ultimately inexplicable. It
is not surprising that Kant describes it as a kind of revolution, like a
“rebirth” (R 6.47). Of course, once a revolution has happened, the same
effect will occur and you will endorse your change of heart and your new
metamaxim.
A parallel argument can be made about the choice of theoretical reasons:
suppose you have to choose which physical theory to adopt. You might, or
might not, take the simplicity of the theory to be a reason in favor of it. You
can choose whether or not to do so. This is a free choice: you are not forced
or – from your own point of view at least − causally determined to choose
that simplicity either is or is not a reason in favor of a theory. But of course
there are and could not be any other reasons for the choice, since it is
precisely a choice for what is to count as a (basic theoretical) reason in favor
of a theory. On reflection, once you take simplicity to be a reason in favor of
a theory, you may endorse your decision to take simplicity to be a reason.
This does not make the original decision itself any more reasonable, but
once more the prospect of a change of mind is going to seem (from your
own point of view) unreasonable.
Returning to our basic theme, we can say that, if the choice of a Gesinnung
is not made for reasons, it is not made on the basis of a maxim. But at the
same time, “taking” something as a reason is something that you actively do.
You are bound to regard your choice of Gesinnung as made freely and actively,
hence the kind of thing for which you can be held morally responsible. It is,
therefore, a deed that is not based on a maxim.
If this is right, it explains why, in order to be morally responsible, you
need two significant levels of maxim – one which determines what counts as
a reason for you, and one where you choose actions for reasons. As we shall
see in the next section, morally worthy action also requires two levels of
maxim.
5 Reasons for action

A good person does not merely do good actions, but does them for the right
reasons. Does the Gesinnung play a role in morally worthy action?
Initially, it is not obvious that it does. The shopkeeper whose maxim is to
give the right change for the sake of his reputation performs the right action
for the wrong reasons; whereas the shopkeeper whose maxim is to give the
right change because it is his duty performs a morally worthy action. The
difference appears to be in their first-order maxims.
90 ALISON HILLS
However, Kant refers several times to the Gesinnung in the context of

morally worthy action, or the best kind of action (G 4:406, 416, 435; KpV
5:147). So there is reason to think that a morally worthy action must
ultimately flow from a good Gesinnung, not just from a first-order maxim
of the right kind.
For example, suppose that I very rarely give any money to charity. But
one day I pass a charity campaigner, listen to her speech and in response give
a generous donation. I have done the right thing because she convinced me
that doing so was right. This looks very much like a morally worthy action.
But I suggest that we cannot yet tell whether it is − for suppose that I
stopped to listen only because I was in a good mood, and I was in a good
mood only because I had unexpectedly found a coin on the pavement. Now
it is, at least, not so obvious that my action is morally worthy. There is a
sense in which I did the right thing for the right reasons, but my commit-
ment to doing so was not wholehearted, was not a deep part of my person-
ality, and was ultimately based on my having taken pleasure in unexpectedly
acquiring some money.11
If the argument in the last section is correct, then unless I had a
Gesinnung, on which I acted, I did not act for reasons at all. But even
leaving aside that argument, and supposing that, in some sense, I did act for
a reason here, there is still something morally amiss with my action. I had no
deep commitment to act for moral reasons: doing so was merely a whim,
based on my good mood. When the mood passes, I will be back to ignoring
the needs of others.
It matters not just whether I do the right action, but why I did: in Kant’s
terms, did I do it because it is my duty? A morally worthy action is the result
of a deep and wholehearted commitment to morality. In the above example,
my commitment to my duty is shallow. It can easily be overridden or simply
fail to move me.
If this is right, a morally worthy action is based on a good maxim, and
that maxim is grounded in a commitment to morality, a metamaxim
placing morality over self-love. A good maxim that was grounded in a
different metamaxim, one that allowed self-love to subordinate morality,
11
This sort of argument fits well with Arpaly’s conception of moral worth (though other aspects of her
theory are not particularly Kantian) – namely, that depth of commitment to morality influences moral
worth: Nomy Arpaly, “Moral Worth,” Journal of Philosophy 99 (2002), pp. 223–45. This view is
criticized by Julia Markovits, who argues that a theory of moral worth should not be too demanding
on agents, and that doing the right thing for the right-making reasons is sufficient for morally worthy
action: Markovits, “Acting for the Right Reasons,” Philosophical Review 119 (2010), pp. 201–42.
Gesinnung 91
would result in actions like my giving to charity because doing so was my
duty, but only when I was in a sufficiently good mood.
If this is right, does a good metamaxim need to be based on a good
higher-level maxim too, if your action is to be morally worthy? I do not
think so. Morally worthy action requires a commitment to morality over
self-love, which grounds good individual maxims. What further commit-
ment to morality could be wanted or required? Could we ask that that
whole-hearted commitment to morality must itself be based on a whole-
hearted commitment to morality? It is not clear to me what this could mean.
As we saw earlier, the choice of a metamaxim cannot be based on a prior
commitment to treat morality or self-interest as having priority, because
that is exactly what you are now choosing to do. Morally worthy action
needs a good maxim, based on a good metamaxim, and that is all.
6 Character
One good action does not make a person good; one action performed for
the wrong reasons does not make a person evil. A good person typically or
reliably performs morally worthy actions; an evil person typically or reliably
performs bad actions (or, at least, she performs actions for bad reasons).
Can we nevertheless explain the moral evaluation of persons entirely in
terms of good or bad reasons for action?
X is evil if all of the actions for which she is morally responsible are
performed for bad reasons;
or
X is evil if almost all of the actions for which she is morally responsible are
performed for bad reasons;
or
X is evil if most of the actions for which she is morally responsible are
performed for bad reasons.
These definitions would have to be refined, to take into account the
possibility of a change in character, and this may not be straightforward. But
the major drawback of this approach is that to be a bad (or good) person is
wholly explained in terms of the good or bad actions that you perform. It is
not therefore possible to give an explanation of the following form: she
made a false promise because she is an evil person. To do that, we need a
conception of the person’s character, in virtue of which she performs good
or bad actions. Since the Gesinnung is a deep commitment to morality or to
self-love, which grounds the choice of individual maxims, it seems that in
Kant’s account, your character is your Gesinnung.
92 ALISON HILLS
This of course raises many questions. I will start with an obvious prob-
lem: what is the relationship between the Gesinnung and the choice of
individual maxims?
A familiar way of understanding the relationship between character and
action is that character is a disposition to perform certain kinds of action (for
certain reasons). Can we understand the Gesinnung as a disposition, and the
individual maxims as manifestations of that disposition?
One very obvious question is whether it is possible to choose a first-order
maxim that does not conform to the Gesinnung. Can a bad person act well,
or a good person act badly? This seems to be possible, and would be possible
according to this account, provided that you can have a disposition that is
not manifested. We might, however, expect an explanation of why the
disposition is not manifested in these circumstances. It is not clear what
explanation could be given.
More seriously, it is unclear what explanation might be given of the
“straightforward” case in which a good person chooses a dutiful maxim.
The connection between a disposition and its manifestations is often under-
stood as causal. But it is hard to see how Kant could accept there being a
causal connection between Gesinnung and maxims. According to Kant,
causation is a phenomenal relation, but a Gesinnung is noumenal (R 6:31).
A causal connection is in time, and the cause must be prior to the effect. This
would imply that the Gesinnung had been adopted at a time prior to the
maxim’s being chosen, which is not consistent with Kant’s claim that the
Gesinnung is timeless (R 6:25). Finally, the idea of a causal connection
between the Gesinnung and particular maxims is problematic given Kant’s
view that both the first-order maxims and the Gesinnung are freely chosen
(R 6:21–22, 39–40).
Another way of understanding the relationship is that the Gesinnung and
first-order maxims are not really two separate things. We might deny that a
Gesinnung is really different from adopting series of maxims: having a
Gesinnung might be more like living a certain kind of life. As Sussman puts it:
Adopting an intention to walk does not have to be a precisely datable event
standing in determinate temporal and causal relations to the activity of
walking itself . . . intention is an aspect of the entire activity as a whole,
not an ingredient or a stage of it. Similarly my fundamental moral resolution
may be something that is located not in any particular episode of my
biography, but in my life as a whole.12
12
David Sussman, “Perversity of the Heart,” Philosophical Review 114 (2005), p. 173.
Gesinnung 93
Sussman offers an intriguing and appealing suggestion. It explains how a
Gesinnung can be timeless in a completely non-mysterious way: a Gesinnung
is not located at any particular time but is constituted by my life as a whole.
But it does not fit so well with other features of the Gesinnung, as described
by Kant.
First, that it is a ground of individual maxims seems to require that the
Gesinnung is separate from those maxims, but it is not clear that there is
sufficient separation here.
Second, according to Kant, in order for action based on the Gesinnung
to be imputable to you, you must have chosen your Gesinnung (R 6:25,
32). It is not clear that there is or could be a genuine choice of a life as a
whole.
I suggest that a better account does take seriously that the metamaxim is a
kind of maxim, whose content is a commitment to take certain consider-
ations as reasons for action (more precisely, to take morality and self-love to
ground reasons for action, but to take moral considerations to have weight
over self-interest, or vice versa).
This maxim leads to choices of first-order maxims involving particular
actions and particular ends or reasons for those actions. The process of
reasoning from the metamaxim to first-order maxims is similar to other
forms of reasoning, either theoretical or practical (as when we reason
instrumentally from a maxim to do some action, to a further maxim to
adopt a means to perform it).
One way of understanding this relationship is in terms of counter-
factuals. If you had not chosen to put self-love over morality, you would
not have chosen the maxim “when I believe myself to be in need of
money I shall borrow money and promise to repay it, even though I know
this will never happen.” The counterfactual dependence of the choice of
first-order maxim on the choice of metamaxim grounds an explanatory
relationship between the two. It is in virtue of your commitment to
morality over self-love (or otherwise) that you choose first-order maxims,
so it is quite appropriate to say that you did this kind of action because
you are evil, or that action because you are good. This account also leaves
room for the possibility of someone essentially bad on occasion doing a
morally worthy action, i.e. not just doing the right thing, but for the
right reasons (at least in the shallow sense identified earlier). In this
case, the first-order maxim would not be dependent on her metamaxim
and we would not be able to say: “she did that because of the kind of
person that she is.”
94 ALISON HILLS
According to certain views, counterfactual dependence is sufficient for

there to be a causal connection between the two relata.13 This would be
extremely problematic for Kant if he agreed, for it would mean that first-
order maxims that are based on a metamaxim are causally determined after
all. But he need not agree that counterfactual dependence necessitates causal
dependence. Kant is committed to there being “laws of freedom” as well as
causal laws, and an instance of counterfactual dependence could be con-
nected to the former rather than the latter (G 4:447–49, 457–58; R 6:39–40).
Laws of freedom are laws for all rational beings; they are laws of reason.
They differ from causal laws, because they concern reasons and the norma-
tive relations between them. Kant must already accept that a maxim that has
been determined by another maxim may nevertheless not be causally
determined, i.e. it can still be freely chosen. Just as an instrumental
maxim that was derived from another maxim would nevertheless be freely
chosen, action based on a maxim derived from the metamaxim would also
be freely chosen.
Is it really possible to understand the relationship between metamaxim
and first-order maxim in terms of counterfactual dependence? Doesn’t Kant
admit that those who put self-love first and those who give morality priority
will often end up performing the same action? So won’t they choose the
same maxim, whatever their metamaxim? But doing the same outward
action (e.g. helping those in need) does not imply acting on the same
maxim. Someone who puts morality first will choose a maxim of helping
those in need because doing so is her duty, whilst someone who give self-
love priority will do so for the sake of her own happiness (e.g., in the hope of
gaining a good reputation). The difference in their first-order maxims can
be explained by the difference in their metamaxims.
This explanatory connection is crucial to the question of whether we
can see Kant’s Gesinnung as a (distinctively Kantian) conception of
character.14 The idea of character has a number of different facets, but
basically it is a concept that plays an important role in theory of action
and in moral philosophy. A person’s character is a fundamental quality (or
set of qualities) that explains her actions. It is essentially connected to
13
For instance, David Lewis, “Causation,” Journal of Philosophy 70 (1973), pp. 556–67.
14
Kant’s theory of character is also developed in the Metaphysics of Morals, where he distinguishes a
number of different virtues and vices. See also Lara Denis, “Kant’s Conception of Virtue,” in
Paul Guyer (ed.), The Cambridge Companion to Kant and Modern Philosophy (Cambridge
University Press, 2006), pp. 505–37; and Allen W. Wood, Kant’s Ethical Thought (Cambridge
Gesinnung 95
the idea of unity within an agent, which as we have already seen is very
important to Kant.
For instance, we might wish to explain why Albert refrains from telling a
lie at time t, and also refrains from telling a lie at t´; that he gives to charity at
t´´ and at t´´´. And this might well be explained by his having a good
metamaxim, one that puts morality over self-love. We may also be able to
explain apparent inconsistencies in action by means of a metamaxim. For
instance, that Benedict refrains from telling a lie at t´ but not at t´´ may be
explained by his having given priority to self-love, and in both cases acting
in a way that he thinks will make him happy. Having a particular meta-
maxim gives you a certain kind of consistency in action through time – not
necessarily that you are performing the same kind of action, but that you are
acting on the basis of the same types of consideration: considerations that
you take to be reasons for action.
A second purpose of the idea of character is to explain differences in
action. For instance, suppose that Charlotte tends to treat people with
respect whilst Dorothy tends to patronize them, sometimes almost coercing
them to do what she wants. This might be explained by Charlotte giving
priority to morality and Dorothy not.
A Gesinnung can clearly play both these roles. Whether you put
morality over self-love, or vice versa, will determine the type of actions
you choose, and distinguish you from others who choose differently (you
will also have different conceptions of happiness, in virtue of your differ-
ent desires).
The concept of character also plays a vital role in the moral evaluation of a
person, and, as we have already seen, this is one of the features of Gesinnung
that is most important to Kant. To determine whether someone is a good
person or not, we cannot look to their actions – because both Charlotte and
Dorothy will often help people and refrain from telling lies – nor can we just
look at their first-order maxims, for even a bad person can, on occasion,
choose a good maxim. We need to look at their fundamental orientation: do
they give priority to morality or self-love? According to Kant, the Gesinnung
alone settles that question.
There are limits to the use of a person’s Gesinnung in explaining action
and in evaluating that person, because, according to Kant, there are
basically only two possible metamaxims: one which prioritizes morality
over self-love and one which does not. Obviously, people with the same
metamaxim will act differently. Those who prioritize self-love will have
different desires, and hence different views on what happiness is.
Moreover, some will choose self-love only when the cost to themselves
96 ALISON HILLS
of acting morally is very great; others are tempted to do wrong very easily.
Kant calls both evil, but it is surely helpful to be able to distinguish them.
Here it is useful to supplement the Gesinnung with what Kant says about
virtue and vice elsewhere, in the Metaphysics of Morals, where, for
instance, he describes how different people have different strengths of
will ( fortitudo) to resist temptation (MS 6:380).
There are obviously striking differences between Kant’s conception of
Gesinnung and the notion of character in, for instance, the Aristotelian
tradition. According to Aristotle, character traits can be classified as
virtues and vices. Virtues are habits. They involve reason but also the
emotions. They enable someone to live well and are components of
eudaimonia. The differences between an Aristotelian conception of char-
acter and Kant’s Gesinnung are many and connect to issues that run very
deep in their moral philosophy. For instance, Kant would not accept that
moral virtue can be a kind of habit, since he regards habits as essentially
unfree (MS 6:407). Aristotle’s conception of eudaimonia is very different
from Kant’s conception of happiness, which consists in the satisfaction of
desire and is not something that he thinks can play an important role in
moral theory (G 4:418–19). Kant cannot accept that feelings or emotions
play a central role in your character, as Aristotle does, because he regards
these as not freely chosen, qualities for which we are not morally respon-
sible and for which it would be inappropriate to praise or blame us
(MS 6:408–9, 457).
It is, I think, questionable whether the moral evaluation of a person can
be based entirely on the choice of a metamaxim, rather than partly on one’s
non-cognitive attitudes, emotions, and feelings. But this is a deep part of
Kant’s ethics, and not an issue that can be settled here. Despite these deep
differences between the most prominent conception of moral character and
Kant’s Gesinnung, I do not think it unreasonable to see Gesinnung as (part
of) a conception of character.
7 Conclusion
Kant describes the Gesinnung only briefly throughout his ethical and
religious writings, but there is much of interest that can be reconstructed
from his remarks, including his provocative remarks in Religion within the
Boundaries of Mere Reason. Without question, his concept of the
Gesinnung plays important roles in Kant’s conceptions of moral respon-
sibility and of morally worthy action. The questions left only partially
Gesinnung 97
answered are obviously deep ones, for not only is the free choice of a
Gesinnung mysterious, but so also is the possibility of a change of
Gesinnung from evil to good (or vice versa). Nevertheless, the Gesinnung
is clearly of great significance, not least as the foundation of a distinctively
Kantian theory of character.15
15
I am very grateful to Gordon E. Michalson and Sergio Tenenbaum for helpful comments on an
earlier draft.
chapter 5
Rational hope, possibility, and divine action

Andrew Chignell
The whole domain of the supernatural is thus removed from the

region of belief into that of simple hope, and in that, for anything
we can see, it is likely always to remain. – J. S. Mill, Theism
Voltaire said that Heaven has given us two things as a counterweight
against the many burdens of life: hope and sleep. He might well have
added laughter. . . – Kant, Critique of Judgment 5:334
1 Kant’s third question

One of the arguments for which Kant is best known (or most notorious) is the
so-called “moral proof” of the existence of God, freedom, and the immortal
soul. Versions of the proof can be found in each of the Critiques, in various
lectures, and in Religion within the Boundaries of Mere Reason. “Proof” has to
be taken loosely here, since the attitude licensed by moral considerations, for
Kant, is not knowledge but rather Belief (Glaube).1 Still, loose talk of “proof” is
appropriate insofar as the argument is supposed to motivate not mere Belief or
faith but rather “rational Belief (Vernunftglaube)” – i.e. assent that is justified
in a non-epistemic way for finite practical agents. Kant is hardly advocating an
irrationalist leap into dogmatic or mystical fancy.
Because the moral proof is so well-known, and because at least two of its
objects are broadly religious – God and the immortal soul – it comes as a
surprise when he indicates in a 1793 letter to the theologian C. F. Stäudlin
that such Belief is not the main focus of the philosophy of religion. It’s really
moral philosophy that deals with Belief; philosophy of religion, in contrast,
My thanks to Eric Watkins, Karl Ameriks, an anonymous reviewer for Cambridge University Press, and
audiences at the APA and the Humboldt Universität zu Berlin for helpful feedback on earlier drafts.
1
Glaube is an ambiguous German term that is typically translated as either “belief” or “faith.” I prefer to
avoid the latter term, since not all Glaube has to do with religious doctrine. Here I’ll use capitalized
“Belief” in order to distinguish it from contemporary notions of “belief.” See Andrew Chignell, “Belief
in Kant,” Philosophical Review 116 (2007), pp. 323–60, for more on Kant’s general notion of “Belief.”
98
Rational hope, possibility, and divine action 99
is concerned with the attitude of hope (Hoffnung). It’s even more surprising
that Kant gives hope, rather than Belief, pride of place in the list of questions
that motivate his entire critical philosophy:
all interest of my reason (the speculative as well as the practical) is united in
the following three questions:
1. What can I know?
2. What should I do?
3. What may I hope? (A806/B833)2
The third question somehow unites the first two: it is “simultaneously

practical and theoretical” – it “concerns happiness” and “finally comes
down to the inference that something is . . . because something ought to
happen.” In this way “the practical leads like a clue to a reply to the
theoretical question and, in its highest form, the speculative question”
(A805-6/B833-4, original emphasis).3
Commentators typically neglect the distinct nature and role of hope in
Kant’s system, and simply lump it together with the sort of Belief that arises
from the moral proof.4 Indeed, Kant himself is not entirely innocent of the
conflation.5 I want to suggest below, however, that, from a conceptual as well as
a textual point of view, hope should be regarded as a different kind of attitude.
It is an attitude that we can rationally adopt toward some of the doctrines that
are not able to be proved from within the bounds of mere reason – either
theoretical or practical. This does not mean that hope is unconstrained; there
are rational limits, as we shall see. In fact one of my central claims here is that a
2
See the letter to C. F. Stäudlin of May 4, 1793 (Brief 11:429), as well as the introduction to the Jäsche
Logic of 1800 (9:25).
3
In the letter to Stäudlin and the introduction to Logic, Kant adds a fourth question to the list – namely,
“What is the human being? (Was ist der Mensch? )” (L 9:25).
4
See, e.g., Hermann Cohen, Reason and Hope: Selections from the Writings of Hermann Cohen, ed. and
trans. Eva Jospe (New York: W. W. Norton, 1971); Onora O’Neill, “Kant on Reason and Religion”
(Tanner Lectures on Human Values, 1996); Philip Rossi, “Kant’s Doctrine of Hope: Reason’s Interest
and the Things of Faith,” New Scholasticism 56 (1982), pp. 228–38; Christopher McCammon,
“Overcoming Deism: Hope Incarnate in Kant’s Rational Religion,” in Chris L. Firestone and
Stephen Palmquist (eds.), Kant and the New Philosophy of Religion (Bloomington: Indiana
University Press, 2006), pp. 79–89; Katrin Flikschuh, “Hope as Prudence: Practical Faith in Kant’s
Political Thinking,” International Yearbook of German Idealism 7 (2009), pp. 95–117. Hope comes up
more often in Kant’s later writings, and some readers may conclude that Kant was weakening his
earlier and very strong claims about the necessity of firm rational Belief – perhaps, as Heine would have
it, in response to his own fading religious commitments. In fact, though, the question about hope is
already there in the 1781 A-edition. It’s just that, as Kant tells Stäudlin in the 1793 letter, he hadn’t had a
chance to discuss religion in detail prior to writing Religion.
5
See the top of R 6:62 where the ability to “hope to become pleasing to God (and thereby blessed)”
seems to be equated with someone’s being “entitled to consider himself not an unworthy object of
divine pleasure.” But being able to hope to become x is clearly a weaker state than that of being entitled to
consider oneself x.
100 ANDREW CHIGNELL
crucial difference between knowledge, rational Belief, and rational hope is that
they are governed by different modal constraints; section II discusses those
constraints and the kind of modality involved. In section III, I return to
Religion and offer what I take to be Kant’s account of the main objects of
rational hope in that text – namely, “alleged outer experiences (miracles)”; a
“supposed inner experience (effect of grace)”; and a future collective experience
(the construction of a truly ethical society) (R 6:53).6
2 The Structure of rational hope
2.1 Modal constraints

What sorts of entities or states can we rationally hope for, given our ethical
situation and psychological make-up? We have seen that Kant thinks some-
thing about “the practical” is supposed to lead “like a clue” to a commitment
regarding what is – i.e. regarding a theoretical entity or state. This doesn’t
mean that practical considerations somehow provide theoretical grounds (i.e.
empirical or a-priori reasons) for these doctrines. Rather, the talk of “theoret-
ical” or “speculative” is just Kant’s way of referring to propositions regarding
what exists – that is, things or states that we include in our ontology, broadly
speaking. In the case of hope as well as Belief, the grounds for assent regarding
“theoretical” entities will be almost wholly practical.7
6
Kant was by no means the first among Western religious thinkers to claim that hope is worth
discussing in a religious context. There is a Pauline precedent for conceiving hope along with faith
as near the top of the list of theological virtues, though not quite as important as love. Augustine has a
long discussion of hope in The Enchiridion: Faith, Hope, and Love, as well as some comments on the
virtue of hope in his tractates on the First Letter of John. Peter Lombard dedicated some of his Sentences
to the concept of hope, and as a result of this nearly everyone in the later medieval tradition remarked
on it in their Sentences-commentaries. Bonaventure is a prime example, though his account of hope is
unusual in that he construes it as a kind of meta-virtue: the sustained affective commitment that helps us
keep our faith constant and our loves properly ordered. See Rachel Lu, “Natural and Supernatural
Virtue in St. Bonaventure,” dissertation submitted to Cornell University, 2012. In most of these
thinkers, however, remarks about hope are sandwiched between much longer discussions of faith and
love. As far as I know, no one prior to Kant suggested that hope is the central attitude in religion, and
thus the central topic of the philosophy of religion. After Kant, in contrast, there have been further
efforts in that direction, typically by authors who are themselves deeply influenced by Kant – Ernst
Block, Das Prinzip Hoffnung (Frankfurt am Main: Suhrkamp, 1959); Jürgen Moltmann, Theologie der
Hoffnung: Untersuchungen zur Begründung und zu den Konsequenzen einer Christlichen Eschatologie, 3rd
edn. (Munich: C. Kaiser, 1965); James L. Muyskens, The Sufficiency of Hope: Conceptual Foundations
of Religion (Philadelphia: Temple University Press, 1979).
7
I say “almost” because there is some talk of “theoretical” grounds for Belief in Kant’s writings as well.
See Chignell, “Belief in Kant,” and Lawrence Pasternack, “Kant’s ‘Doctrinal Belief in God,’” in
Oliver Thorndike (ed.), Rethinking Kant. Vol. III (Newcastle upon Tyne: Cambridge Scholars Press,
2011), pp. 200–18, for discussion of this interesting hybrid attitude.
By comparison with our ordinary concept of “belief” or “knowledge,” there
is very little discussion of hope in recent philosophical literature. Indeed, one
philosopher – J. P. Day8 – begins his 1969 paper on the topic with the claim
that philosophers have completely abandoned hope. That’s not quite fair –
there were a few authors before Day (such as Wheatley, Bloch, and Downie)9
who had kept hope alive philosophically. If there was a consensus analysis
amongst the more analytical mid-century authors, it was something like this:
S hopes that p if and only if
(H1) S desires that p, and
(H2) S believes that p is possible.
Phillip Pettit10 refers to this as the “lowest common denominator” con-
ception of hope – first, because it is common in a lot of those earlier
discussions, and second because (H1) and (H2) are shared by more robust
analyses, including his own. Luc Bovens, Ariel Meirav, and Adrienne
Martin, likewise take this to be the “orthodox account” (Martin’s term)
before going on to offer refinements and additions.11
(H1) could be made more precise, but the idea behind it seems uncon-
troversial: if I hope that p is true, then I have some sort of desire or pro-
attitude toward p’s truth. But what about (H2)? Clearly p can describe an
event that has not yet occurred – I hope that it will be sunny in Ithaca (and
believe that this is at least possible). But the relevant p can also describe
events that have already occurred but whose details are unavailable to me. If I
wasn’t there and haven’t heard any news about the matter, I can hope that
(i.e. desire and believe possible that) it wasn’t cloudy yesterday, even though
the facts of the case have been settled. Once I learn that Ithaca has been
experiencing one of its characteristic month-long gray spells, my hope
disappears.
But even with this in mind, (H2) as stated is too strong: for surely
someone can hope for p even if he or she has no actual beliefs at all –
occurrent or dispositional – about the modal status of p. Consider, then, the
somewhat weaker
8
John P. Day, “Hope,” American Philosophical Quarterly 6 (1969), pp. 89–102.
9
J. Wheatley, “Wishing and Hoping,” Analysis 18 (1958), pp. 121–31; Bloch, Das Prinzip Hoffnung;
Robin S. Downie, “Hope,” Philosophy and Phenomenological Research 24 (1963), pp. 248–51.
10
Philip Pettit, “Hope and its Place in Mind,” Annals of the American Academy of Political and Social
Science 592 (2004), pp. 152–65.
11
Luc Bovens, “The Value of Hope,” Philosophy and Phenomenological Research 59 (1999), pp. 667–81;
Pettit, “Hope and its Place in Mind”; Ariel Meirav, “The Nature of Hope,” Ratio 22 ( 2009), pp. 216–33;
Adrienne M. Martin, “Hopes and Dreams,” Philosophy and Phenomenological Research 83 (2011),
pp. 148–73, and How We Hope (Princeton University Press, 2013).
102 ANDREW CHIGNELL
(H2rev) S is in a position to believe that p is possible.

This is an improvement, although “in a position” clauses are notoriously
tricky. For present purposes, we can simply take it to mean that a certain
counterfactual is true – namely, that S would believe that p is possible if he
were to form a belief on the matter.
In a different context it would be worth embarking on a series of thought-
experiments designed to specify further, more complicated, and jointly sufficient
conditions. Here, however, I’ll simply work with (H1) and (H2rev), and take
them to state significant necessary conditions. In other words, I’ll take desiring
something and being in a position to believe that it is possible to be two of the
main components of paradigmatic hope, even if they are not the whole story.
Note that the question Kant asks is not about what we do hope or can hope,
but about what we may hope – the German here is “Was darf ich hoffen?”
“May” connotes a concern about rational permissibility, a concern whose
source, presumably, is the recognition that the desire in (H1) or the belief in
(H2rev) might be irrational, and that irrationality might transfer to the hope as
well. If I irrationally believe that it is possible to find water that is not H2O,
and for some reason want to do so, then it looks like the irrationality of the
modal belief will infect my hope as well, even if the desire is well founded.
This suggests that, even if the conjunction of (H1) and (H2rev) provide
an account of two significant necessary conditions on hope simpliciter, we
need something more if we are to capture Kant’s talk of the hopes that we
may have. For present purposes, I propose to set aside questions about the
rationality of desire and focus on the second condition. (H2rev) is satisfied
when a subject is in a position to have a modal belief about the object of
hope, and so the addition here will presumably relate to the rationality of
that belief. We can capture both of these ideas by speaking not of justified
belief, but of being justified in believing. The idea, then, would be that
S rationally hopes that p only if
(RH1) S rationally desires that p, and
(RH2) S is justified in believing that p is possible.
To go further with respect to the second condition, we clearly need to
grasp what sort of possibility Kant is invoking.12 This will require a
quick detour through the chapter in the first Critique that takes modality
as its focus – the “Postulates of Empirical Thinking.”
12
We would also need an account of what it is to be justified in believing some proposition. In order to
avoid such a Herculean assignment here, perhaps we can simply agree on a rough-and-ready character-
ization: S must possess whatever is required for the relevant belief, if she forms it, to be justified.
2.2 The possibilities of hope
Empirical possibility
Does the modal condition on hope involve what Kant sometimes refers to
in the Critique as “empirical possibility”? This is the modal status enjoyed
by events or changes of state that are compatible with preceding states, given
the empirical laws.13 So is hope rational only when directed toward some-
thing that we are justified in taking to be empirically possible in this way?
Such a constraint, it seems to me, would be too weak in one sense, and
too strong in another. Too weak because at least some of the paradigmatic
objects of Kantian hope are not entirely empirical – Kant often speaks of
hope for a “moral world,” and such a world presumably involves our
intelligible, ethical characters. This kind of moral state may in some
minimal sense be “compatible” with empirical laws and preceding events,
since, qua moral state, it is not really part of the empirical nexus at all. But
that kind of compatibility is not what is meant when people (Kant included)
articulate modal conditions in terms of empirical possibility. In other
words, we don’t normally regard something as empirically possible just
because it is not an empirical object or state at all (e.g. an abstract object) and
so ipso facto fails to violate (or follow from) empirical laws. Rather, we
ascribe empirical possibility to something that is itself an empirical object or
event, one that, at the very least, positively coheres with the conjunction of
the empirical laws and the description of all previous events.
There is another sense, however, in which construing the constraint in
(RH2) as involving empirical possibility is too strong. For the “moral world”
does have an empirical component or upshot – Kant famously thinks that
the highest good involves not just virtue (i.e. worthiness to be happy), but
also genuine “happiness in the world proportionate to the worthiness to be
happy” (R 6:8n). Such happiness – for sensing, feeling creatures like us –
will be partly empirical (R 6:6–7n). And yet, for all we know, such happiness
may not be in keeping with the empirical laws of the actual world and thus
may be empirically impossible. And there might be other empirical anoma-
lies for which one can, in the right context, reasonably hope (see section III
below). If this is right, then a reading of the modal constraint in (RH2) as
empirical fails in at least two different ways.
13
See A220ff./B268ff., as well as Andrew Chignell and Nicholas Stang, “Postulate des empirischen
Denkens,” in G. Mohr, J. Stolzenberg, and M. Willaschek (eds.), Kant-Lexicon (Berlin: Walter
de Gruyter, 2014), n.p.
104 ANDREW CHIGNELL
Formal possibility
What other kinds of possibility are available? In the first of the three
Postulates, Kant develops the notion of a more abstract kind of modality
that invokes not the actual initial conditions and empirical laws, but rather
the general “formal conditions” of our experience, conditions that may well
be consistent with non-actual conditions and non-actual laws. He defines it
as follows: “That which agrees (übereinkommt) with the formal conditions
of experience (according to intuition and concepts), is possible” (A218/
B265).Here too there is a question about what “agreement” with the formal
conditions of experience requires. Kant’s own examples are of little help −
he cites the following as “mere figments of the brain” that do not agree with
the formal conditions of experience: “A substance that is persistently present
in space yet without filling it . . . or a special fundamental power of the
mind to intuit the future . . . or, finally, a faculty of our mind to stand in
a community of thoughts with other men (no matter how distant they
may be)” (A222-23/B270). Although the concepts of such things are logically
coherent, Kant admits, the claim that their objects are possible is “totally
groundless . . . because they cannot be founded on experience and its laws
with which we are acquainted (weil sie nicht auf Erfahrung und deren bekannte
Gesetze gegründet werden kann)” (A223/B270).
But what does “experience and its laws with which we are acquainted”
mean in this context? I’ve been simply assuming that the formal “laws” here
are the a-priori formal laws of experience – specifically, the principles
derived from the categories, such as the Causal Principle of the Second
Analogy. But the examples Kant cites (telepathy, soothsaying, ghost-like
entities) are not − or not obviously − incompatible with such highly general
and formal principles. If, on the other hand, “laws” refers to the empirical
laws, then Kant would be saying merely that we do not know that these
items are compatible with the specific empirical laws of the actual world.
But then it is not clear why this would support agnosticism about their
formal possibility. Again, Kant defines formal possibility here as compati-
bility with the formal conditions of experience, and these are supposed to be
far less determinate than specific empirical laws and thus compatible with
different sets of them.14
If we apply this point to the hope issue, formal possibility seems too weak
to be a candidate for inclusion in (RH2), given that numerous states of the
14
See Chignell and Stang, “Postulate,” for more discussion of this puzzle, and of the Postulates
generally.
intelligible realm – some of which ground the happiness of the virtuous, and
some of which don’t – are “compatible” with the formal conditions of
intuition and category-application. Such a flaccid principle is surely not
what’s intended by someone trying to provide an informative account of
rational hope.
Logical possibility
What, then, about strict logical possibility? Is the constraint on rational hope
in (RH2) that its objects must be logically possible, or at least taken by the
subject to be so? This is also too weak. It’s true that logic prevents me from
rationally hoping that 2+2 = 5 or that I will meet a married bachelor
someday. So a constraint that entails logical possibility is presumably
involved. But hope seems equally irrational, fanciful, and false when its
object is water that has a chemical constituency other than H2O, or a
zombie with empathetic feelings, or an event without a cause, or (to use
Kant’s own example in the Postulates) two straight lines that manage, on
their own, to enclose a figure (A220-21/B268). We need to find something
stronger if the condition in (RH2) is to limn the contours of the relevant
domain.
Real possibility
What philosophers now call metaphysical possibility offers, I submit, the
right joint at which to carve. It may seem anachronistic to import this
notion – lost in early analytic philosophy but rehabilitated by Kripke and
others at the end of the twentieth century – into a discussion of Kant’s
philosophy. But in fact Kant develops and uses a modal notion that is quite
similar: he calls it “absolute possibility” in the Postulates chapter, and
elsewhere simply “real possibility.” Absolute (real) possibility “goes beyond
all possible empirical use of the understanding” and relates to things and
their natures per se – it has to do with what is “possible in all respects” rather
than “possible only under conditions that are themselves merely possible”
(A232/B284).
The concepts of real possibility and real necessity are found in Kant’s
thought as early as the 1760s, and they play an important role throughout
his career. In the critical period, for instance, Kant famously argues that
we must not go beyond the domains of empirical and formal possibility
and make speculative inferences to first causes, intelligent designers, free
wills, world-wholes, and other supersensibles. Why not? At least one of his
concerns is that those speculative concepts may contain predicates that are
“really repugnant” – either alone or in combination. In other words, the
106 ANDREW CHIGNELL
objects of these thoughts may be really impossible, for all we know, and
thus for epistemic purposes we ought to regard them as mere “figments of
the brain, for the possibility of which there would be no indications at all”
(A222/B270).15
Given that Kant clearly employs a notion of real modality throughout his
career, we needn’t be too concerned about anachronism in the present
context. The proposal, then, is that the second condition in the analysis of
rational hope should be
(RH2revised) S is justified in believing that p is really possible.
Many of Kant’s explicit references to hope support something very much
like this conception. Here is one of the crucial passages in the first Critique:
it is equally necessary to accept in accordance with reason in its theoretical
use (eben so nothwendig sei es auch nach der Vernunft, in ihrem theoretischen
Gebrauch anzunehmen) that everyone has grounds (Ursache) to hope
for happiness in the same measure as he has made himself worthy of it in
his conduct, and that the system of morality is therefore inseparably
combined with the system of happiness, though only in the idea of pure
reason. (A809/B837)
Note the phrase “accept . . . that everyone has grounds to hope.” Given the
practical context of the discussion, it is clear that “acceptance” here is equiv-
alent to Belief (Glaube) in the technical Kantian sense, and indeed Kant often
treats these two terms – Annehmung and Glaube – as synonyms. What we are
Believing in this practical way, then, is that everyone has (non-epistemic)
grounds to hope for happiness in proportion to his or her own virtue.
It is crucial to see that Kant is not encouraging us baldly to accept that
there actually is a necessary connection between virtue and happiness.
Rather, the most that rational hope requires is that we Believe that such a
connection is really possible. Kant says something similar in the third
Critique: a “righteous man (like Spinoza)” who seeks to be self-consistent
must accept (annehmen) the existence of a moral author of the world, i.e. the
existence of God from a practical point of view, so that he can at least form a
concept of the possibility of the final end that is morally prescribed to him; and
15
For more discussion of real repugnance, see Andrew Chignell, “Real Repugnance and our Ignorance
of Things-in-Themselves: A Lockean Problem in Kant and Hegel,” International Yearbook of German
Idealism 7 (2009), pp. 135–59; and “Real Repugnance and Belief about Things-in-Themselves: A
Problem and Kant’s Three Solutions,” in James Krueger and Benjamin Bruxvoort Lipscomb (eds.),
Kant’s Moral Metaphysics: God, Freedom, and Immortality (Berlin/New York: de Gruyter, 2010),
pp. 177–211.
he can indeed do this, since [such an idea] is at least not intrinsically
contradictory. (KU 5:452–3, my emphasis)
God’s actual existence is accepted as a condition of the real possibility of a
moral world, which is something we also positively accept/Believe. In
Kant’s own view of rational hope, then, it looks like (RH2revised) has to
give way to
(RH2Kant) S at least rationally Believes that p is really possible.
The difference between (RH2revised) and (RH2Kant) is that the former
doesn’t require the subject to have any positive attitudes at all, while the
latter does. For the supersensible objects of hope we’re concerned with in
religious contexts, the relevant attitude will be Kantian Vernunftglaube
based on practical grounds. But the “at least” in (RH2Kant) is supposed
to indicate that different attitudes with different kinds of justification might
be appropriate in other contexts.
We have now arrived at two of the most significant necessary conditions
on Kantian rational hope:
S rationally hopes that p only if
(R1) S rationally desires that p,
(RH2Kant) S at least rationally Believes that p is really possible.
With this partial analysis as well as a sense of the modal theory in the
background, we can now turn directly to Religion in an effort to understand
its account of some of the main objects of rational hope.
3 Rational hope and Kantian religion
3.1 Hope for empirical miracles

One of the most intriguing but also baffling uses of modal concepts in
Kant’s Religion has to do with his occasional expressions of openness to the
possibility of what might be called empirical miracles (R 6:52, 84, 88n). He
consistently rejects such miracles as authenticators of central religious doc-
trine, of course – no signs, wonders, or special revelations are required for us
to understand what the moral law requires, or to understand the content of
the moral proof. In Part Two, for instance, Kant asserts that we don’t have
either a theoretical or a practical need to postulate the virgin birth, and he
also omits endorsement of the central Christian miracle – the bodily
resurrection of Christ. In the general remark at the end of Part Two, he
makes it clear that a moral religion (“the heart’s disposition to observe all
108 ANDREW CHIGNELL
human duties as divine commands”) is such that any miracles connected

with its inception are utterly dispensable (6:83ff.). Indeed, it is a pernicious
form of unbelief (Unglaube) to insist on miracles before (or instead of)
accepting reason’s practical dictates (R 6:63).
All the same, Kant is strikingly careful (for a determinist about the
empirical world) not to rule out miracles altogether. He says repeatedly
that “reason does not dispute the possibility or actuality” of historical
miracles (R 6:52; cf. 88n) and that it is “entirely conformable to the ordinary
human way of thinking,” for a new religion – even one based on “the spirit
and the truth (on moral disposition)” − to announce or “adorn” its intro-
duction with dazzling feats (R 6:84). Kant even claims to find it plausible
that the work of a “prophet” or “founder” of a new religion would be full of
miracles (thus helping to win adherents from the old religion), and that the
historical testimony to these miracles itself would be miraculously arranged
and preserved: “It may well be (es mag also sein),” he says without any
obvious Humean insincerity, that Christ’s “appearance on earth, as well as
his translation from it, his eventful life and his passion, are all miracles –
indeed that the history that should testify to the account of these miracles is
itself a miracle” (R 6:84–85).
Similar claims can be found in lectures on religion and metaphysics from
the critical period. Kant clearly has Leibniz in mind when he insists, for
instance, that
it is not at all impossible, even in the best world, that the powers of nature
may sometimes require the immediate cooperation of God in order to bring
about certain great ends. It is not impossible that the Lord of Nature might at
times communicate to it a complementum ad sufficientiam in order to carry
out his plan. (VpR 28:1112)
No world can be thought without deficiencies, without certain negations
and limitations, and thus to make up the defect of nature, miracles are
possible in the best world also, and even probable according to the concept of
God’s goodness and truth (MetM 29:871; see also Metaphysik -K2 28:732ff.).
A miracle strictly defined is called rigorous. [How] is such a thing possible?
Because there is an extramundane cause that has produced this order of
things, and thus can produce another. A miracle is thus possible in itself
internally . . . In general, an event in the world whose laws human reason
cannot at all cognize is a miracle.” (MD 28:667)
Such passages reveal at least an openness to the real possibility of anomalous

empirical events. This coheres with the claim in the Religion that the
rational person does not “dispute their possibility or actuality” even though
she also does not “sanction (statuiert)” them for use in practical or theoret-
ical reasoning (see R 6:88 and note).
But what sort of possibility could Kant have in mind here? How could a
miracle fit into his deterministic picture of the natural world? The texts are
hardly clear about this, but one option that seems open to Kant is to adopt a
variation on the Leibnizean approach to miracles, according to which there are
two different sets or levels of “laws,” only one of which is accessible to us. For
Leibniz, empirical miracles are supposed to be possible, despite the validity of
the Principle of Sufficient Reason (PSR), because the laws that we rely on in
everyday life and seek to describe in natural science are merely the subordinate
maxims of God’s will – the policies according to which the best possible world
normally operates. In certain circumstances, these policies may be suspended
so that an event in the empirical world that is not in accordance with them
occurs in accordance with higher purposes. Leibniz, devotee of orderliness and
reason, cannot allow such events to be violations of the PSR, of course, and
thus in the Discourse on Metaphysics (1686) and elsewhere, he claims that God
has a hidden set of policies which are always in effect. These policies – which
together constitute the so-called “law of the series” – entail the world’s typical
adherence to subordinate maxims but also allow for their occasional violation.
The law of the series itself, however, is too complicated for finite minds to
comprehend, and so we must aim, in scientific inquiry, to describe the
intelligible if exceptionable subordinate maxims.16
A Kantian variation on this theme could make use of the distinction
between empirical possibility and formal possibility discussed above. What
usually occurs in the empirical world is determined by the laws that we rely
on in common sense as well as scientific contexts. For all we know, however,
these laws are suspended on occasion to allow an empirical miracle – i.e.,
“that which happens contrary to the order of nature in the world” – to occur
(MetM 29:870). Such an event would still be formally possible in the sense
that it would “agree” with the formal conditions of experience – the axioms
of space and time, as well as the principles derived from the categories. And
so there would still be a fundamental lawfulness, just as there is for Leibniz.
But the event would be out of keeping with the more specific empirical laws
16
Discourse on Metaphysics sections 16–17 (1686), in Roger Ariew and D. Garber (eds.), Philosophical
Essays (Indianapolis: Hackett Publishing Co., 1989), pp. 48–49. Elsewhere in this work it becomes
clear that some events are not just above, but positively “contrary” to, the “subordinate maxims which
we call the nature of things” (Discourse section 7, p. 40). He repeats this formulation in a letter to
Arnauld of July 14, 1686 − “miracles are contrary to some subordinate maxims or laws of nature” − in
H. T. Mason (ed.), The Leibniz−Arnauld Correpondence (Manchester University Press, 1967), p. 57,
my emphasis.
110 ANDREW CHIGNELL
that typically describe what is actual and causally necessary. If this model is
coherent, then it may be rational for a practitioner of Kantian religion to
hope, in the right circumstances, for such an event to occur, even while
assuming (for all practical and theoretical purposes) that it won’t. The only
claim about miracles that we must “dispute with all our might,” Kant says, is
that they authenticate a particular religion, or that belief in them is pleasing
to God and/or religiously required (R 6:85).
There is more to be said about this,17 but since it is clear that empirical
miracles are not a central focus of Kant’s Religion, I propose to move on to two
other uses of the concept of rational hope that go to the very heart of his theory.
3.2 Hope for moral assistance

The most prominent use of the concept of hope in Religion (as well as in
Conflict of the Faculties and his lecture on religion) relates to the doctrine of
“supernatural assistance” in the moral life. Kant says repeatedly that if we do
our best to conform to the moral law, we can reasonably hope for super-
natural assistance of some sort (R 6:48, 52–55, 171; S 7:44; VpR 28:1106).
Sometimes it looks as though Kant takes the assistance to be ontological –
God actually plays a role in the conversion of our will. Other passages
suggest that it is a question of perspective – God chooses to view our efforts
at moral improvement as making us morally perfect, even though “in our
earthly life (and perhaps even in all future times and in all worlds) [our
moral state] is always only in mere becoming” (R 6:75).
There are notorious problems with both of these suggestions from the
point of view of Kantian ethics. The latter, “two perspectives” approach
ascribes to God an odd sort of overestimation or self-deception: even though
finite agents are not wholly good and perhaps never will be, those who are
making progress somehow “appear justified before their judge” (R 6:74) or
are “judged by him who scrutinizes the heart (through his pure intellectual
intuition) to be a perfected whole” (R 6:67). There is more to say about
this approach,18 but here I want to focus on the former one – the one
17
See Andrew Chignell, “Rationalism, Religion, and the Laws: Leibniz and Kant on Miracles,” in
Brandon Look (ed.), Leibniz and Kant (New York: Oxford University Press, 2014). For alternative
approaches to the topic, see A. Tuan Nuyen, “Kant on Miracles,” History of Philosophy Quarterly 19
(2002), pp. 309–23, and chapter 7 by Karl Ameriks in the present volume.
18
See Gordon E. Michalson, Jr.’s negative review of the “perspectives” account in his Fallen Freedom:
Kant on Radical Evil and Moral Regeneration (Cambridge University Press, 1990), pp. 103ff. See
Stephen R. Palmquist, Kant’s Critical Religion (Aldershot and Burlington: Ashgate, 2000), and
Christopher McCammon, “Overcoming Deism,” for more positive assessments.
according to which God is ontologically involved. This is the approach that
Kant often seems to favor, but it is deeply problematic insofar as it seems to
undermine his commitment to human autonomy. We can see this in the
following argument:
(P1) If S is morally responsible for changing the quality of his will,
then S is fully ontologically responsible19 for changing the
quality of his will.
(P2) S is not fully ontologically responsible for changing the quality
of his will.
(C) Thus, S is not morally responsible for changing the quality of
his will.
The conclusion is obviously anathema for Kant, and so something’s got to

give. (P1), however, is just a statement of what is often called Kant’s “Stoic
maxim”:
The human being must make or have made himself into whatever he is or
should become in a moral sense, good or evil. (R 6:44, emphasis Kant’s)
For he ought to become a good human being yet cannot be judged morally
good except on the basis of what can be imputed to him as done by him.
(R 6:51, emphasis Kant’s)20
(P1) thus seems unassailable. Given his theory of autonomy generally,
however, it seems clear that Kant will have to reject (P2): a moral subject
is fully ontologically responsible – is the sufficient ground – of her own
moral character. But then what room is left for supersensible assistance?
The problem here is sometimes referred to as the “conundrum” in Kant’s
theory of moral improvement.21 Here it is formulated as an allegedly
inconsistent quartet:
19
“Cause/effect” is a pure category of the understanding, and has genuine application only with respect
to phenomenal states and objects. There is an “unschematized” form of that concept, however,
that can be used to “think” about noumenal things: Kant calls it “ground and consequence.” A
ground−consequence relation can be posited between abstract entities such as premises in a proof, for
instance, but it can also be thought (if not known) to hold between things-in-themselves (see A73/
B98, A243ff./B301ff.). In order to avoid confusing talk of “causation” in this context, I’ll refer to the
grounding relation as one of “ontological responsibility.”
20
Nicholas Wolterstorff is the source of the phrase “Stoic maxim”: Wolterstorff, “Conundrums in
Kant’s Rational Religion,” in Philip J. Rossi and Michael Wreen (eds.), Kant’s Philosophy of Religion
Reconsidered (Bloomington/Indianapolis: Indiana University Press, 1991), pp. 40–53. See also
John Hare, The Moral Gap: Kantian Ethics, Human Limits, and God’s Assistance (New York:
Oxford University Press, 1996).
21
See, e.g., Michalson, Fallen Freedom; Wolterstorff, “Conundrums”; and Chris L Firestone and
Nathan Jacobs, In Defense of Kant’s Religion (Bloomington/Indianapolis: Indiana University Press,
2008).
112 ANDREW CHIGNELL
(A) S is morally responsible for making himself good (i.e. for changing
the quality of his will from evil to good) (from Kant’s ethics)
(B) S can make himself good (from [A] and Ought-Implies-Can)
(C) If S is morally responsible for making himself good, and S can
make himself good, then S is fully ontologically responsible for
making himself good as well (Stoic maxim)
(D) S requires God’s assistance in becoming good (from [A]−[C])
The conundrum here arises from the fact that the conjunction of any three
is supposed to be incompatible with the fourth. But Kant cannot reject (A),
given his overall ethical theory, and he cannot retain (A) and reject (B)
without violating Ought-Implies-Can (which is clearly endorsed in Religion
at R 6:45 and 62). So, say the conundrumists, he is forced to deny either
(C) – the Stoic maxim – or (D) the requirement of divine assistance, both of
which he seems to endorse.
I want to suggest, in contrast, that once we clarify the kind of modality at
work in this context, the incompatibility, and the conundrum, disappear.
For even if (A)−(C) are true, it might still be in some sense possible (and thus
possibly required) for S to receive some sort of assistance. The possibility
would not be empirical, of course, since we’re talking about non-empirical
states of the will. But there are at least two other options. First, consider the
noumenal analogue of empirical possibility. There are presumably princi-
ples analogous to “laws” in the intelligible world of free agents, and so
perhaps these laws allow two different agents to be fully ontologically
responsible for the same action. The picture would thus be similar to
scholastic and early modern doctrines of concurrence, though at the merely
intelligible rather than empirical level. It is revealing in this regard, then,
that in his lectures Kant explicitly refers to the doctrine of supersensible
assistance as involving a concursus divinus (VpR 28:1106) or even a concursum
morale (VpR 28:1110).
Conundrum theorists will insist here, however, that this is just obviously
a violation of Kant’s strongly held doctrines about libertarian freedom. And,
to be sure, elsewhere in his arguments about freedom, Kant does seem to
indicate that a free act is an “unconditioned” event, one that cannot involve
the assistance of another agent without violating the “laws” of freedom.
The second option simply grasps this nettle. Perhaps such a violation is
precisely what Kant has in mind when he talks about supersensible assistance:
it is a violation of the “laws” of the intelligible world of free agents, and yet it
is possible in some deeper sense, and can thus still be the object of rational
hope. Here is more of the passage just cited from the religion lectures:
Of course this idea of freedom is one which belongs to the intelligible world,
and we are acquainted with nothing of it beyond the fact that it exists, so we
also do not know the laws by which it is governed. But even if our reason
cannot deny the possibility of this concursus, it still sees that such an effect
would have to be a miracle of the moral world, just as God’s acts of cooper-
ation with occurrences in the sensible world are God’s miracles in the
physical world.22 (VpR 28:1106–7)
Kant draws an important analogy at the end of this passage: just as an empirical
miracle is possible even though it violates the particular laws of the empirical
world, he suggests, perhaps a moral concursus is possible at some absolutely
fundamental level, even though it violates the “laws” of the intelligible world.
We find complementary passages in Religion in which Kant suggests both
that our moral state is “incumbent” on us, and yet that assistance may be
possible. In the first of these, he also makes the same analogy between the
natural scientist’s ignorance (given his “occupation” of investigation) of the
deep grounds of nature’s patterns, and the moral agent’s ignorance of
whether he’s receiving some supersensible aid:
Now, to occupational affairs (Geschäften) also belongs the natural scientist’s
search for the causes of events in the natural laws governing those events; I
mean the natural laws of those events that he can thus verify through
experience, even though he must renounce acquaintance (Kenntnis) of that,
in itself, which brings about effects according to these laws, or of what these
laws for us might be relative to some other possible sense-faculty. A human
being’s moral improvement is likewise an occupational affair (Geschäfte)
incumbent upon him, and heavenly influences may indeed always cooperate
in this improvement, or be deemed necessary to explain its possibility. Yet he
himself has no understanding in the matter: neither how to distinguish with
certainty such influences from the natural ones, nor how to bring them and
thus, as it were, heaven itself down to himself; thus, since he doesn’t know
what to do with them, he does not in this case sanction miracles but rather, if he
pays heed to the call of reason, conducts himself as if all change of heart and
improvement rests solely on the application of his own effort.23 (R 6:87–88)
22
In the practical works, Kant often uses the phrase “laws of freedom” or “laws of the intelligible world”
to refer to the moral law (e.g. G 4:453–54). But in this passage he must be speaking of the “laws” that
govern the intelligible domain of freedom in some broader sense, since he says that we don’t know
them, and we do of course know the moral law. Compare, here, a passage in Religion where Kant
indicates that what we are acquainted with in practical cognition is only a subset of these laws, namely
the moral ones: “But we are at least acquainted with the (moral) laws of freedom” (R 6:191, Kant’s
parenthesis and emphasis).
23
Kant attaches a key footnote to the word “sanction” (statuirert) here, part of which I quoted earlier:
“This is the same as saying that he does not incorporate miracles into his maxims (either of theoretical
or practical reason), even though he does not dispute (anfechten) their possibility or actuality.”
114 ANDREW CHIGNELL
Reason says that whoever does, in a disposition of true devotion to duty, as

much as lies within his power to satisfy his obligation (at least in a steady
approximation toward complete conformity to the law), can legitimately
hope (hoffen dürfe) that what lies outside his power will be supplemented by
the supreme wisdom in some way or other (which can render permanent the
disposition to the steady approximation), without reason thereby presuming
to determine that way or to know (wissen) what it consists in, for God’s way
can perhaps be so mysterious that, at best, he could reveal it to us in a
symbolic representation in which the practical import alone is comprehen-
sible to us, whereas, theoretically, we could not in the least grasp what this
relation of God to the human being is in itself, or attach concepts to it, even if
God wanted to reveal such a mystery to us. (R 6:171)
To Believe that grace may have its effects, and that perhaps there must be such
effects to supplement the imperfection of our striving for virtue, is all that we
can say on the subject. (R 6:174, my emphasis)
This is precarious territory for a critical philosopher, as Kant himself recog-
nizes. In both the lectures and Religion, he frequently falls back to the weaker
point that we can’t presume to know that assistance is impossible: “as regards
a concursum morale or God’s free cooperation in the free actions of human
beings, such a thing cannot be comprehended in the nature of freedom” but
“at the same time it cannot be regarded as impossible” (VpR 28:1110).
The concept of a supernatural intervention into our moral though deficient
faculty . . . this is a transcendent concept, merely in the idea of whose reality
no experience can assure us . . . Yet its impossibility (that the two may not
occur side by side) cannot be proven either, since freedom itself, though not
containing anything supernatural in its concept, remains just as incompre-
hensible to us according to its possibility as the supernatural [assistance] we
might want to assume as surrogate for the independent yet deficient deter-
mination of freedom. (R 6:191)
This kind of disavowal of modal knowledge follows from Kant’s general

commitment to ignorance about absolute real possibility. But does denying
knowledge here leave room for rational modal Belief?
I want to suggest that it does. Kant clearly thinks there has to be room for
hope here, and as we have seen, his condition on rational hope (RHKant)
requires more than mere suspension of judgment about the modal issue.
One needs at least the positive Belief, on sufficient practical grounds, that a
“miracle of the moral world” is really possible. The two proposals I’ve
sketched above attempt to make sense of how this might work. If the
incompatibility between assistance and autonomy can be regarded as merely
apparent, or if it can be regarded as a genuine incompatibility at the
derivative level of “laws” rather than the most fundamental modal level,
then Belief in the real possibility of assistance (as well as “moral hope” for it)
would be able to withstand rational scrutiny, even if all claims to modal
knowledge must be denied.24
3.3 Hope for social progress

One of the other main controversies in the current literature on Kant’s
theory of evil has to do with how our position in society relates to our
morally evil acts – is it necessary for moral evil, or sufficient, or neither, that
we be in community with others?25 Whatever the right answer, it is clear
that Kant views communal life as morally precarious: “As soon as he is
among human beings,” a man almost inevitably falls into the kind of “envy,
addiction to power, avarice, and the malignant inclinations associated with
these” that wouldn’t afflict a lonesome Crusoe on his separate island (R
6:93–94). Kant’s catch-phrase for this thesis is “unsocial sociability”; it plays
a prominent role in his Idea for a Universal History (1784), and has prede-
cessors in the writings of Montaigne and Rousseau.
Given that isolation is impractical and that we have a propensity to
competitive nastiness and “mutual corruption,” Kant says, we can only
rationally “hope for a victory of the good principle over the evil one”
through the “setting up and diffusion of a society in accordance with, and
for the sake of, the laws of virtues” (R 6:94). This is not a political state but
rather an “ethical community” formed for moral reasons, and its possibility
is what grounds rational hope for the ultimate victory of good over evil –
both individual and collective.
But what precisely are we willing when we will to form an ethical
community? Kant says in Religion that the ethical community is “universal”
by way of involving the entire human race; later he calls it the “invisible”
albeit “true church,” and says it has the “modality” of being “unchange-
able.” So, while the ethical state of nature leaves us with “public feuding
between virtue and inner immorality” (R 6:97), the ethical community will
be an as-yet-unseen social arrangement that results in universal harmony
and mutual moral encouragement.
24
For a longer and more detailed exposition of this argument, see Chignell, “Rational Hope, Moral
Order, and the Revolution of the Will,” in Eric Watkins (ed.), Divine Order, Human Order, and
the Order of Nature (New York: Oxford University Press, 2013) pp. 198–218.
25
For recent discussions of this debate, see Jeanine Grenberg, “Social Dimensions of Kant’s Conception
of Radical Evil,” in Sharon Anderson-Gold and Pablo Muchnik (eds.), Kant’s Anatomy of Evil
(Cambridge University Press, 2010), pp. 173–94; and Allen W. Wood, “Kant and the Intelligibility
of Evil,” in Anderson-Gold and Muchnik (eds.), Kant’s Anatomy of Evil, pp. 144–72.
116 ANDREW CHIGNELL
Given that we ought to will such a community and that ought-implies-

can, the community must be really possible. But an individual cannot
generate it on his own, and he cannot assume or even hope that others
will play along: the “objective reality” of the idea of an ethical state (and thus
its real possibility), Kant says, is “entirely well-grounded” in “human reason
(in the duty to join such a state), even though we cannot subjectively ever
hope of the good will of human beings that these will work harmoniously
toward this end” (R 6:95).
More strikingly, even if all of us were to will the collective good, Kant still
thinks that this would not guarantee the possibility of its achievement. This is
because the “ethical community” (like the “moral world” discussed earlier) has
an empirical component, and the inscrutable gulf between intelligible free-
dom and empirical nature leaves it unclear whether or how it can be brought
about: “we cannot know whether as a whole it is also in our power” (R 6:98).
Thus a commitment to the real possibility of the ethical community that we
are obliged to construct must be grounded in “a higher moral being through
whose universal organization the forces of single individuals, insufficient on
their own, are united for a common effect” (R 6:98).
What we encounter in the discussion of our social duties in Religion,
then, is effectively another practical proof. We can’t know that an ethical
community is really possible, but once we see that we ought to will it, we
have to Believe on practical grounds that it is indeed possible. This in turn
licenses Belief in supersensible assistance, and also underwrites rational hope
for a this-worldly but still inconceivable goal.26
Despite this sort of appeal to providence in both the Religion and the
various critical discussions of human history (especially the third Critique),
Kant is careful to insist that one must not wait for an intervention, or
passively assume that someone else is in control. Each person must act “as if
everything depended on him. Only on this condition may he hope that
a higher wisdom will provide the fulfillment of his well-intentioned effort”
(R 6:101, my emphasis). This is an explicit effort to avoid what he regards as
the morally enervating aspects of stronger, more Calvinist doctrines of
providence. Still, Kant is clear that once we have recognized our ethical
26
O’Neil, “Kant on Reason and Religion,” goes over some of this same material in an effort to find an
account of “hope without doctrine” (p. 280). If we can manage to keep the focus entirely on this-
worldly progress, she says, perhaps we can take Kant’s argument to underwrite the Belief that the
ethical community is possible and the hope for its achievement as a result of our efforts alone, without
an appeal to providence. See Flikschuh, “Hope as Prudence,” for a critical assessment of O’Neill’s
view here.
duty in society and conformed our will to it, we are implicitly committed to
Belief that its achievement in a this-worldly context is really possible, and
able to have rational hope for this “fulfillment.”
4 Conclusion: hope in place of belief

The importance of the difference between knowledge and rational Belief
(Vernunftglaube) in Kant is clear – we have to set aside the former in order
to make room for the latter (Bxxx). Less widely recognized, however, is
the fact that hope differs from both of these – in its structure, scope, and
rationality conditions. Once we understand Kant’s conception of rational
hope, we can (I submit) make better sense of the intriguing – if peren-
nially baffling – references to empirical, moral, and social “miracles” in
Religion and related writings. In other words, given the availability of
rational hope, the “bounds of mere reason” within which Kantian religion
is contained may turn out to be quite a bit wider than the bounds of either
knowledge or rational Belief.
chapter 6
Kant on grace
Leslie Stevenson
Kant’s late work Religion within the Boundaries of Mere Reason may seem to
occupy an unstable half-way house between traditional Christianity and the
robustly humanist stance of most other Enlightenment thinkers such as Hume
and the French philosophes, and their atheist successors in the nineteenth
century, Feuerbach, Marx, and Nietzsche. Gordon E. Michalson1 has
argued that the logic of Kant’s argument takes him in the latter direction,
whereas John Hare says it points rather toward mainline Christianity.2 But
are those the only two alternatives? I will suggest that rather than Kant being
pulled toward either Christian orthodoxy or atheistic humanism, his stren-
uous wrestling with the notion of divine grace can draw both believer and
agnostic toward recognition of the ultimate inexplicability of human action
and character, and the need for forgiveness and hope.
I
The word “grace” has multiple meanings listed in the dictionaries, and is a
clear example of a family-resemblance term. Nearest to the theological
conception is the English phrase “grace and favour,” meaning the freely
chosen conferment of a benefit by a superior in status, power, or wealth,
where there is no question of recognizing a right, and not necessarily of
rewarding any particular merit. In theological usage, the superior is God,
of course. This chapter will concentrate on Kant’s treatment of the concept
of divine grace, a much-loved but much-contested concept in theology
since the beginning of Christianity.
At the end of each of the four parts of the Religion, Kant appends a
“General remark” extending to several pages. He describes these as parerga
1
Gordon E. Michalson, Fallen Freedom: Kant on Radical Evil and Moral Regeneration (Cambridge
University Press, 1990).
2
John Hare, The Moral Gap: Kantian Ethics, Human Limits, and God’s Assistance (Oxford University
Press, 1996).
118
Kant on grace 119
(R 6:52): topics not belonging within his rational conception of religion, yet
“bordering on it.” He offers as titles for his four parerga: (1) “Of Effects of
Grace”; (2) “Miracles”; (3) “Mysteries”; and (4) “Means of Grace”; and he
explains how he conceives of their borderline status.
Reason, conscious of its impotence to satisfy its moral needs, extends itself to
extravagant ideas which might make up for this lack, though it is not suited
to this enlarged domain. Reason does not contest the possibility or actuality
of the objects of these ideas: it just cannot incorporate them into its maxims
of thought and action. And if in the inscrutable field of the supernatural
there is something more than it can bring to its understanding, which may
however be necessary to make up for its moral impotence, reason even
counts on this something being made available to its good will even if
uncognized, with a faith which . . . we might call reflective, since the dogmatic
faith which announces itself to be a knowledge appears to reason dishonest or
impudent. (R 6:52)
This fits with the sharp distinction between knowledge (Wissen) and faith
(Glaube) that Kant had elaborated on in his three Critiques.3 In his view, the
religious beliefs in God and immortality can be neither proved nor disproved
by rational argument or by scientific evidence, yet they are not meaningless,
and are appropriate (and in some sense necessary) objects of faith.
In the General remark appended to Part One, Kant distinguishes
between “religion of rogation (of mere cult, connoting the courting of
favor) and moral religion, i.e. the religion of good life-conduct” (R 6:51).
About the former, he sarcastically describes devotees as expecting either that
God can make them eternally happy without any need on their part to
become better people, or that He can make them better people without
their doing anything more than asking or wishing for it. But with the latter,
Kant says we are entitled to hope for the grace of God:
it is a fundamental principle that, to become a better human being, everyone
must do as much as it is in his powers to do; and only then, if a human being
has not buried his innate talent (Luke 19:12–16), if he has made use of the
original predisposition to the good in order to become a better human being,
can he hope that what does not lie in his power will be made good by
cooperation from above. (R 6:52)
He goes on to say that it is not essential for each human being to know what
God does for his or her salvation. Alleged “effects of grace” in inner
3
I offer a systematic analysis of Kant’s epistemological vocabulary in Leslie Stevenson, “Knowledge,
Belief or Faith, and Opinion,” Kantian Review 7 (2003), pp. 72–101. A revised version of this paper is
included in Stevenson, Inspirations from Kant (New York: Oxford University Press, 2011).
120 LESLIE STEVENSON
experience could not be theoretically cognizable, because any claim about

divine causation goes beyond the limits of our experience. Moreover any
action-guiding employment of the idea of grace would be self-contradictory,
because it would mean aiming at something we know to be beyond our
power. Kant concludes Part One with the ringing statement that “we can
admit an effect of grace as something incomprehensible,” but we cannot
incorporate it into our theoretical or practical maxims (R 6:53). One may
wonder what it can be to “admit” something that one admits is literally
incomprehensible, but perhaps we should qualify this as saying that, though
we can readily comprehend the idea of a better person, we cannot understand
how God may produce such a result.
Earlier, however, Kant makes more positive mention of divine grace as
accessible to us:
Granted that some supernatural cooperation is also needed to his [the human
being’s] becoming good or better, whether this cooperation only consists in
the diminution of obstacles or be also a positive assistance, the human being
must nonetheless make himself antecedently worthy of receiving it; and he
must accept this help (which is no small matter), i.e. he must incorporate this
positive increase of force into his maxim. (R 6:44)
There appears to be a contradiction between his saying here that we must

“accept” or “lay hold of” divine cooperation and incorporate into our maxim
“this positive increase of force,” and his assertion at 6:52 that reason cannot
incorporate such ideas into its maxims of thought and action. This is a
foretaste of the antinomy he sets out in Part Three, which I will discuss below.
In Part Two, Kant invokes the concept of grace in a rather different way,
in his analysis of the tension between our imperfect moral dispositions and
our ideal of moral perfection. On the one hand, there is the “radical evil” in
human nature diagnosed in Part One − namely, our inveterate tendency to
prefer our own self-interest over the demands of morality. On the other
hand, we can set before ourselves an ideal of holiness, the idea of a person
who measures completely up to such demands and is thus “morally pleasing
to God.” Kant subverts Christian orthodoxy by suggesting that we do not
strictly need a historical example of this such as Jesus, since “the idea is
present as a model already in our reason” (R 6:62), though he concedes that
real-life exemplars may have their use. With or without actual examples of
perfection, our lives inevitably involve a “moral gap”: when judged by the
highest standard, we all fall short. There is an infinite distance between the
holiness we aspire to, or ought to aspire to, and the imperfections which
remain in us at every stage of our lives.
Kant on grace 121
According to Kant, the “change of heart” that consists in reversing the
subordination of morality to selfishness “must be possible, because it is a
duty”; but how, he asks, “can this disposition count for the deed itself, when
this deed is every time . . . defective?” He suggests that “we can think of the
infinite progression of the good toward conformity to the law as being
judged by him who scrutinizes the heart (through his pure intellectual
intuition) to be a perfected whole even with respect to the deed (the life
conduct)” (R 6:66–67). But how can we be relieved of the guilt we have
incurred for our past sins of omission and commission? In a discussion
influenced by his Lutheran culture,4 Kant offers a somewhat convoluted
rational reinterpretation of the Christian doctrines of atonement, justifica-
tion, and salvation. After conversion from the state of radical evil there is in a
sense a new person, who, in Paul’s words, “has discarded the old human
nature and the conduct that goes with it, and has put on the new nature”
(Colossians 3:9–10). As Kant puts it: “Physically . . . he is still the same
human being liable to punishments . . . Yet, in his new disposition (as an
intelligible being), in the sight of a divine judge for whom the disposition
takes the place of the deed, he is morally another being” (R 6:74). And this
involves divine grace:
Here, then, is that surplus over the merit from works for which we felt the
need earlier, one which is imputed to us by grace. For what in our earthly life
(and perhaps even in all future times and in all worlds) is always only in mere
becoming (namely, our being a human being well-pleasing to God) is
imputed to us as if we already possessed it here in full. And to this we indeed
have no rightful claim (according to the empirical cognition we have of
ourselves), so far as we know ourselves (estimate our disposition not directly
but only according to our deeds), so that the accuser within us would still be
more likely to deliver a verdict of guilty. It is always therefore only a decree of
grace when we are relieved of all responsibility for the sake of this good in
which we believe. (R 6:75–76)
God’s grace performs a different function here from in Part One. There
“co-operation from above” was hoped for (though its operations cannot
be discerned) to help make us better human beings, to bring about definite
improvements in our moral characters. But now a God’s-eye point of view
(in terms of omniscience and perfect moral standards) is invoked, firstly to
judge all our efforts as inadequate, however much moral progress we may
make, yet secondly (in terms of God’s love and forgiveness) not to impose
4
It is puzzling that Kant makes no reference to Luther (let alone Calvin). Perhaps the influence was so
pervasive in Kant’s time that it did not need to be mentioned.
the blame or punishment that we strictly deserve but to treat us, imperfect
as we are, as already “well-pleasing to God,” provided we make the requisite
change of heart and turn ourselves in the right direction. The influence of
St. Paul and Luther is obvious here, in the notion that, although we are all
guilty when judged by the highest standards, God graciously “justifies” us in
His sight, “counting the disposition for the deed” as if we already displayed
good life-conduct in full, provided only that we place our faith in Christ.
Kant’s discussion seems to replace faith in the atoning work of Christ by a
reversal of priority of our maxims of action − to put morality above self-love.
But both Kantian and Christian are unable to explain why at this most
fundamental level some people choose good, and some evil. In Religion Part
One grace is supposed to make us better (even if still imperfect), but in Part
Two grace imputes to us a status that we do not actually possess, so it seems
there is a kind of moral fiction going on. These two conceptions can be
labelled sanctifying grace and justifying (or “forensic”) grace.
In Part Three, Kant presents what he calls “a remarkable antinomy of
human reason with itself” (R 6:116).5 He says that “saving faith” contains
two elements necessarily bound together, namely “faith in satisfaction
(reparation for guilt, redemption, reconciliation with God)” together with
“faith in the ability to become well-pleasing to God in a future conduct of life”
(R 6:116), thus uniting (from the point of view of human reception or faith)
the justificatory and sanctifying conceptions of divine grace that we have
just distinguished above. Kant then formulates his antinomy about this
saving faith:
1. It is totally inconceivable, however, how a rational human being who
knows himself to deserve punishment could seriously believe that he has
only to believe the news of a satisfaction having been rendered for
him . . . in order to regard his guilt as done away with . . . No thoughtful
person can bring himself to this faith . . .
2. But if humankind is corrupt by nature, how can a human being believe
that on his own, try hard as he will, he can make a “new man” of himself,
well-pleasing to God, when conscious of the transgressions of which he
has so far been guilty, he still stands in the power of the evil principle and
finds no capacity in him sufficient to improve things in the future? . . .
Faith in a merit which is not his own, but through which he is reconciled
with God, would therefore have to precede any striving for good works,
and this contradicts the previous proposition. (R 6:116–17)
5
Each of his three Critiques contains one or more “antinomies” to which he offers a reasoned solution,
and the pattern continues into the Metaphysics of Morals.
Kant on grace 123
In line with his epistemology Kant says we cannot resolve this problem
at the theoretical level, for we can know nothing about the operation of
divine satisfaction or atonement (“for ratiocination it is an unfathomable
mystery” – R 6:143), or about the influence of divine grace on our wills. But
from the practical point of view, the question is whether we have to start
“from what God has done for our sake, or from what we ought to do in
order to become worthy of it,” and predictably Kant comes firmly down on
the side of the second (R 6:117–18). Some trenchant later statements confirm
his difference from Christian orthodoxy on this fundamental point: “true
religion is not to be placed in the knowledge or the profession of what God
does or has done for our salvation, but in what we must do to become
worthy of it” (R 6:133). He also remarks that, according to the first side of the
antinomy, faith (in “vicarious satisfaction”) is a duty, and faith in one’s own
better behavior would have to be given by grace, whereas on the second side
good life-conduct is our duty, and “satisfaction from on high” would be a
matter of grace (R 6:118). Kant thinks that only the second is acceptable, but
he tries to accommodate Christian piety by arguing that “living faith in the
prototype of humanity well-pleasing to God (the Son of God) refers, in
itself, to a moral idea of reason,” i.e. what is important in faith in Christ is
not the empirical historical details of his life and death, but the moral ideal
that we derive from the stories about him (R 6:119).
Philip Quinn has given a clear and sympathetic exposition of Kant’s
“remarkable” antinomy, while arguing for a more Christian resolution of it.6
To my mind, however, he does not do enough to explain the content of the
historical faith in Christ that he wants to cleave to, and how that content
serves to resolve the antinomy. Quinn sketches alternative models in which
human freedom cooperates with divine grace, whether in accepting God’s gift
of saving faith, or in producing subsequent good works. In a previous
generation, H. A. Hodges, writing within a catholic (small “c”) tradition,
offered a similar-sounding solution to the Kantian antinomy:
We cannot be saved without full repentance . . . But the sickness from which we
suffer is that of a diseased will, and we cannot perform this full repentance . . .
Yet . . . no one, not even Christ, can do these things for us . . . To this
problem there is only one solution. Since we cannot do it alone and He
cannot do it instead of us, it must be both together who do it, He in us and we
in Him . . . We find our salvation in our mystical union with Christ.7
6
Phillip L. Quinn, “Saving Faith from Kant’s Remarkable Antinomy,” Faith and Philosophy 7 (1990),
pp. 418–33.
7
Herbert Arthur Hodges, The Pattern of Atonement (London: SCM Press, 1955), p. 55.
This can be called cooperative grace, the human and the divine working
together to produce good results. It may sound attractive, and it could relate
to Kant’s above-noted distinction between physical and moral identity, but
this talk of Christ and human person acting together (“He in us and we in
Him”) raises perplexing questions about action, personal identity, and the
mystery of “mystical union.”8 The present chapter will also end on a note of
mystery about the springs of human action.
Jacqueline Mariña has argued that Kant has two distinct concepts of
grace,9 different subspecies of his general definition of grace as “a superior’s
decision to grant a good for which the subordinate has no more than (moral)
receptivity” (R 6:75n). One is the unknowable supernatural cooperation of
God transforming our will, desires, and motives for the better. The other,
Mariña writes, “is the kind of divine aid which must be laid hold of by the
person. It differs from the first in that such aid does not alter a person’s will
at the outset, but is, rather, some historical occurrence – a person or
situation – to which the person must respond in some way.” This fits with
Kant’s talk at R 6:44 (quoted above) of the person’s need to accept this help,
but the difficulty for this as textual elucidation is that the phrase “this help”
in that sentence refers back to the unknowable supernatural cooperation
that is Mariña’s first conception of grace. Can Kant mean to say that we
must “accept” help that is totally unknowable to us, and should incorporate
into our maxim a “positive increase of force” that is equally unknowable?
Toward the end of her paper, Mariña employs Frankfurt’s distinction
between first-order and second-order desires,10 instanced in the addict’s
desire to give up smoking although his cravings to light up remain. She
suggests that while Kant’s a-priori principle of morality can and should
guide our second-order desires or fundamental maxim, we can still stand in
need of grace in the form of divine intervention in history to help us bring
our first-order desires and hence our behavior into line with the fundamental
principles we have committed ourselves to. But that kind of grace – divine
intervention in history, at least through a person or situation to which one
can respond – is accessible to us, at least in the sense that the relevant events
are knowable in the way that any historical events can be (and the more
8
It may explain Jacqueline Mariña’s otherwise surprising remark that Kant is closer to Rome than to
the Reformation at the end of her paper “Kant on Grace: A Reply to his Critics,” Religious Studies 33
(1997), pp. 379–400.
9
Ibid. Michalson (Fallen Freedom, Chs. 4–6) pursues the issue skeptically, while Hare (The Moral Gap,
Ch. 10) develops it theologically.
10
Harry Frankfurt, The Importance of What We Care About (Cambridge University Press, 1988),
pp. 11–25.
Kant on grace 125
controversial religious claim would be that they can be recognized as in some
sense revelatory of the divine).
Apart from the thorny issue of revelation, perhaps there is a suggestion
here for an interpretation of grace that relates to other things Kant has to
say, and that might even be attractive independently of theism (see section 4
below). If we agree that a sympathetic friend or therapist or pastor or the
example of a historical person such as Jesus or Muhammad or the Buddha
(as mediated through a religious tradition) can sometimes help a person to
change for the better, we also have to agree that the person must “accept” or
somehow “lay hold” of this help – they have at least to be open to insights
and suggestions that may affect their approach to life. The believer in God
will surely want to say that this is divine aid or grace operating through
particular people or traditions – and not unknowably, though the means of
its operation may remain mysterious. Perhaps this conception can be
labeled providential grace.11
In Part Four, Kant touches on grace once more, noting that it is
“customary . . . to call nature what can be done by the human being on
the strength of the principle of virtue, and grace what only serves to supple-
ment the deficiency of all his moral capacity.” However:
The persuasion that we can distinguish the effects of grace from those of
nature (virtue), or even to produce these effects in us, is enthusiasm; for
nowhere in experience can we recognize a supersensible object, even less
exert influence to bring it down to us, though there do occur from time to
time movements that work toward morality but which we cannot explain,
and about which we are forced to admit our ignorance. (R 6:173–74)
Kant quotes John 3:8 as supporting his case: “The wind blows where it wills,
you hear the sound of it, but you do not know where it comes from or
where it is going. So it is with everyone who is born from the Spirit.”
In the “General remark” appended to Part Four, Kant says that, although
the concept of grace or “supernatural intervention” is a transcendent con-
cept which cannot be instantiated in our experience, its impossibility cannot
be proved either, since freedom itself remains just as incomprehensible
to us (R 6:191). And he insists that the conception of a means of grace,
the idea of performing some specifically religious act such as a prayer or
ritual in order to bring about divine assistance, is a self-deception and
11
See Leslie Mulholland’s “Freedom and Providence in Kant’s Account of Religion: The Problem of
Expiation,” in Philip J. Rossi and Michael Wreen (eds.), Kant’s Philosophy of Religion Reconsidered
(Bloomington/Indianapolis: Indiana University Press, 1991), pp. 77–102.
delusion (R 6:192–202). That just about wraps up what Kant has to say
about grace.12
2
It will be useful now to attempt an overview of Judeo-Christian conceptions
of grace, to compare and contrast with Kant’s treatment.13 (I am not here
defending or attacking Christian doctrine, just trying to give a fair exposi-
tion of it.)
Some references in the Hebrew scriptures to divine grace or favor suggest
that God recognized some particular virtue in Noah (Genesis 6:8) or Moses
(Exodus 33:12–17). But the original call to Abraham is presented simply as
God’s inscrutable choice (Genesis 12:1–4), and at Exodus 33:19 God majes-
tically asserts: “I shall be gracious to whom I shall be gracious, and I shall
have compassion on whom I shall have compassion” (Paul refers back to
that saying at Romans 9:14–18, though it fits ill with the conception of
God’s universal love). In the New Testament, no reason is offered for God’s
choice of Mary to be the mother of Jesus (Luke 1:26–33) or of Paul to preach
Christ to the Gentiles (Galatians 1:15–16). These cases involve yet another
conception of divine grace, for which Kant finds no use, namely God’s
selection of an individual for a function within His purposes, but without
necessarily making them better people or “imputing” righteousness to
them. We can label this electing grace.
There are anticipations of God’s sanctifying grace in the verses “I shall
give them the wit to know me” (Jeremiah 24:7) and “I shall give you a new
heart and put a new spirit within you” (Ezekiel 36:26); but, on the other
hand, at Deuteronomy 29:4 Moses says to the Israelites, “To this day the
Lord has not given you a mind to understand or eyes to see or ears to hear,”
and at Isaiah 29:10 we read “The Lord has poured on you a spirit of deep
stupor.” There are references to God working the opposite of grace (should
we call it “dis-grace?”) in hardening Pharaoh’s heart (Exodus 4:21, 7:3, 9:12,
10:27), although elsewhere we are simply told that his heart became hard
12
There is a brief reference to grace in Kant’s very late work, The Conflict of the Faculties of 1798, where
he states “grace is none other than the nature of the human being insofar as he is determined to actions
by a principle which is intrinsic to his own being, but supersensible (the thought of his duty),” and in
the same paragraph he also wrote that grace is “the hope that good will develop in us” (S 7:43). Those
definitions appear inconsistent with each other, if one can have such hope without actually obeying
the call of duty in particular cases, and vice versa. However, “determined” in the first formulation
could be read as referring to Wille, the general principles of actions that a person accepts, rather than
Willkür, the particular actions he or she actually chooses to perform.
13
My quotations are from the Oxford Study Bible: Revised English Bible with the Apocrypha,
ed. M. J. Suggs et al. (Oxford University Press, 1992).
Kant on grace 127
(8:15, 9:34–35) which leaves it open that he may have hardened it himself.
But Kant seems to have had little time for the Old Testament.
It is in the writings of Paul that we find the most developed account of
the Christian conception of grace which so influenced Kant via Luther. This
is not so much God’s election to a role (though Paul saw that as applying to
his own apostolic ministry), as rather the divine rescue plan for the human
condition of sinfulness (justificatory grace), and for infusing holiness into
individual believers (sanctifying grace):
Therefore, now that we have been justified through our faith, we are at peace
with God through our Lord Jesus Christ, who has given us access to that
grace in which we now live. (Romans 5:1–2)
For it is by grace you are saved by faith; it is not your own doing. It is God’s
gift, not a reward for work done. (Ephesians 2:8–9)
Yet, along with all such emphasis on God’s initiative, the New Testament
epistles are full of injunctions to keep to the right path and avoid wrong
ones. Hebrews 4:16 reads: “Let us therefore boldly approach the throne of
grace, in order that we may receive mercy and find grace to give us timely
hope.” The preaching of the apostles exhorted their hearers to repent (Acts
2:38). James gives a New Testament reference to dis-grace when he says that
“the spirit which God has implanted in us is filled with envious longings”
(4:5), though he immediately follows this by saying that the grace which
God gives is stronger. The implication seems to be that our human choices
have to find what free space they can, in-between the negative and positive
divine influences on us. From the very beginning, the Christian conception
of God’s grace has involved two elements that stand in uneasy tension with
each other: namely, God’s action within us, and our own responses and
choices. The two can be found within a single sentence of Paul’s: “You must
work out your own salvation in fear and trembling; for it is God who works
in you” (Philippians 2:12–13). Similar tensions can surely be found in other
religious traditions, as Kant suggests (R 6:120).14
It is not surprising then that, in the Christian theological tradition, there
have been recurring controversies about the nature and limits of human
free will. Soon after his conversion, St. Augustine wrote an extended
14
For instance, compare these two quotations: “We have placed covers over their hearts – so that they
do not understand the Qur’an – and deafness in their ears” (6:25); “God has sent down the most
beautiful of all teachings . . . then their skins and their hearts soften at the mention of God; such is
God’s guidance” (39:23) (from The Qur’an, trans. M. A. S. Abdel Haleem [Oxford University Press,
2004]).
philosophical dialogue15 defending the freedom of our human wills within

the divinely created world. Yet in later work he qualified this position by
insisting that, because we are fundamentally corrupted by sin, our free will is
weakened and incapacitated, so that we cannot save ourselves, but stand in
need of God’s gracious salvation in Christ.16 If God’s grace is not bestowed
as a reward for any human merit, it seems it is given to those whom He
elects for no reason that we can understand. St. Paul talked of God pre-
ordaining some people, even before they were born, to be justified in Christ
(Romans 8:29–30 and Ephesians 1:4–5). Hence arose the doctrine that some
people are predestined from all eternity to be saved and others to be
damned, which some theologians have felt obliged to maintain in order to
respect the complete and utter sovereignty of God, but which most people
have found repugnant to the concept of a just and loving God – including
Kant in Religion (R 6:143), and in his Lectures on the Philosophical Doctrine of
Religion at 28:1116. Most Christian thinkers have admitted that our free
responses have to play some role in the economy of salvation, but Augustine
resolutely opposed Pelagius’ assertion that we have a natural capacity to
live up to God’s requirements without any special infusion of His grace.
Pelagianism was condemned as heresy, yet most theologians have recog-
nized that that there is some important sense in which human beings retain
responsibility for their choices and actions. The controversy has continued
down the centuries, famously in the Reformation debate between Luther
and Erasmus,17 and in the nineteenth century, a couple of generations after
Kant, Kierkegaard expressed a very Augustinian view of the freedom, yet
helplessness, of the unredeemed human will.18
The tension remains at the center of Christian thought. Donald Baillie,
Professor of Systematic Theology at the University of St. Andrews in the
mid twentieth century, describes it as “the paradox of Grace”:
Its essence lies in the conviction which a Christian man possesses, that every
good thing in him, every good thing he does, is somehow not wrought by
himself but by God. This is a highly paradoxical conviction, for in ascribing
15
St. Augustine, On Free Choice of the Will, trans.Thomas Williams (Indianapolis: Hackett Publishing
Company, 1993).
16
See Eleanore Stump, “Augustine on Free Will,” in Stump and Norman Kretzmann (eds.), The
Cambridge Companion to Augustine (Cambridge University Press, 2001), pp. 124–47.
17
Erasmus–Luther, Discourse on Free Will, trans.Ernst F. Winter (New York: Frederick Ungar, 1961).
See also Gerhard O. Forde, The Captivation of the Will: Luther vs. Erasmus on Freedom and Bondage
(Grand Rapids, MI: William B. Eerdmans, 2005).
18
See Timothy P. Jackson, “Arminian Edification: Kierkegaard on Grace and Free Will,” in
Alastair Hannay and Gordon D. Marino (eds.), The Cambridge Companion to Kierkegaard
(Cambridge University Press, 1998), pp. 235–56.
Kant on grace 129
all to God it does not abrogate human personality nor disclaim personal
responsibility . . .
We can never ponder enough upon the meaning of this paradoxical
conviction which lies at the very heart of the Christian life and is the unique
secret of the Christian character. It is this that makes so wide a gulf between
the Christian way of life and any “mere morality,” so that in a sense
Christianity transcends morality altogether and there is no such thing as a
Christian ethic.19
That last sentence seems dangerous in view of our sad knowledge of priests
and pastors and cult leaders who have abused power, wealth, or sexuality:
surely nobody can be allowed to “transcend morality.” But the point is
presumably that grace enables us not merely to live up to morality, but to
surpass ordinary conceptions of it. As Paul wrote: “By God’s grace I am
what I am, and his grace to me has not proved vain; in my labours I have
outdone them all [Paul was referring to the other apostles] – not I, indeed,
but the grace of God working with me” (1 Corinthians 15:10). Here again is
the mystery of mystical union.20
3
What then is the relation between Kant’s treatment of grace and the
Christian tradition with which he wrestles with a striking combination of
sympathy and skepticism? There is an obvious difference when he appears
to align himself with Pelagianism, saying “There is absolutely no salvation
for human beings except in the innermost adoption of genuine moral
principles in their disposition” (R 6:83). But, unlike so many other
Enlightenment thinkers, Kant has a strong sense of human sinfulness, for,
as we have seen, Part One of the Religion offers an elaborate analysis of what
he calls the “radical evil” in human nature, thereby giving his interpretation
19
Donald M. Baillie, God Was in Christ (London: Faber and Faber, 1956). In his earlier paper
“Philosophers and Theologians on Freedom of the Will,” Scottish Journal of Theology 4 (1951) –
reprinted in D. M. Baillie, The Theology of the Sacraments and Other Papers (New York: Charles
Scribner’s Sons, 1957) – Baillie said that the “true saints . . . have regularly confessed that whatever was
good in their lives was not their own achievement but was due to divine grace.” Presumably the
implication is that whatever was not so good in their lives was due to their own fault – unless we are
prepared to say that all the bad stuff is due to divine dis-grace, which would remove human
responsibility. No doubt there is a corrective here for the common human tendency to make excuses
for our faults and expect praise for our virtues, yet if we are to be blamed for the bad it may seem a bit
inconsistent for us not to be allowed any credit for the good.
20
In God Was in Christ, Baillie argues that the paradox of grace is our best clue to understanding the
mystery of the Incarnation. But I will not attempt to investigate here how a paradox can illuminate a
mystery: the paradox itself provides quite enough topic for this chapter.
of the doctrine of original sin. At one point he declares that the radical evil in
us “has its source in a free power of choice, and hence is imputable,” yet in
the next sentence he says it is “not be to be extirpated though human forces”
(R 6:37). For Kant, what makes a human being evil or good is the relation
between the incentives of self-love and those of morality, and he argues
boldly (and not very plausibly) that this comes down to a single fundamen-
tal choice whether to give priority to the maxim or principle of self-love or to
the moral law (R 6:20–25, 36). It thus emerges that both Kant and
Christianity are subject to tensions about the extent of our freedom. Kant
tends to put more emphasis on human free will than some of the Christian
tradition (especially in its Protestant form), but both acknowledge an
ultimate mystery in the explanation of human behavior.
Another large difference that may in fact be less than it first appears is
that, whereas the biblical writings make some very confident assertions
about the influence of divine grace (and sometimes “dis-grace”) on certain
individuals, Kant resolutely maintains (in line with his critical epistemology)
that we can know absolutely nothing of its operation. He allows that we can
meaningfully speculate about grace (he was not a logical positivist), and we
may hope that God’s grace will supplement our own efforts to do better, but
we can never know that it has actually operated in any particular case. In
contrast, the Hebrew scriptures are full of stories about the grace of God
operating in the patriarchs and the prophets, and the New Testament is
brimming over with proclamations of the grace of God given through
Christ. Yet if we ask that awkward and un-Biblical question “How did
they know, in any particular case?” confidence may wobble.
We noted how the book of Exodus sometimes talks of God hardening
Pharaoh’s heart, but at other times of his heart merely becoming hard,
and it is tempting to suggest that these are merely stylistic variants
expressing the same state of affairs, namely that the ancient writers had no
explanation of why Pharaoh refused to accede to Moses’ demands. How,
after all, could it ever be shown that it was God who changed Pharaoh’s
mind, rather than he himself changing his mind, or that his disposition was
changed by other influences altogether? Even if a voice from the heavens
had publicly spoken to Pharaoh in resonant Egyptian tones, commanding
him to let the Israelites go (or not to do so), he would presumably have
retained enough free will to disbelieve or disobey any such putative divine
command. If we think of God influencing Pharaoh’s mind in secret ways
that were not publicly observable, we are back to the question of how we
could ever distinguish between that and the case of Pharaoh making up his
own mind.
Kant on grace 131
A similar epistemological question can be asked about how we could ever
know anything of the operations of positive grace (whether justifying or
sanctifying). At Pentecost, the Holy Spirit is said to have descended on the
apostles in the form of flame or fire, making them speak in strange tongues
(Acts 2:1–4), and the risen Christ is said to have made a belated appearance
to Paul (Acts 9:3–6). But whether or not we take those ancient reports as
historically true, the problem with any contemporary alleged public signs
of grace – such as speaking in unintelligible “tongues,” snake-handling
without being bitten, or whatever – is that such stuff need not be connected
with any transformation of mind and heart for the better. Such external
manifestations are surely neither necessary nor sufficient for an attribution
of divine grace.
Of course, we may notice a moral improvement in others, and possibly
in ourselves (though that had better be qualified with a great deal of
caution and modesty, as Kant says at R 6:67–71). Paul is confident
that divine grace is present in the early Christian communities (e.g.,
1 Corinthians 1:4), and he attributes people’s different gifts to God’s
grace (Romans 12:6). In Galatians 5, we find quite a detailed list of the
kinds of character and behavior described as the “harvest” or “fruits” of the
Spirit − i.e. the Holy Spirit, whose operations can presumably be identi-
fied with sanctifying grace. It may be rather too “behaviorist” to concen-
trate on outward ethical actions, crucial though they are, in light of the
emphasis in the gospels and epistles on the spirit in which things are done.
We can agree that it is not enough that the required actions are undertaken
if they are done grudgingly, with gritted teeth (perhaps with attention on
the end of the shift – think for example of the care of the sick, whether
in hospital or at home); what is needed is that things are done graciously,
and ideally with loving care. Kant himself said that it is preferable do our
duties cheerfully, indeed “grace-fully.”21 Some such gracious and loving
behavior may be a necessary condition for attributing divine grace, but is it
sufficient? What of the declared atheist or devotee of a non-theist religion who
exhibits as much graciousness and love of neighbor as some high-achieving
Christians: is the grace of God (or Christ) manifest in the former, despite their
lack of Christian belief?
In the face of such questions, it seems wise for the Christian (or Jew or
Muslim) to say, as Kant does, that we simply cannot know with any
21
See KpV 5:82–84; MS 6:448ff.; and the very carefully crafted footnote at R 6:23n, where Kant insists,
against Schiller, that gracefulness is not synonymous with doing one’s duty, but agrees with him that
it should accompany it in the virtuous person.
certainty about the operation of divine grace in individuals. To be sure,

there are adherents of Christianity who manifest an admirable way of life
with warmth and generosity of spirit, and it is naturally tempting to attribute
this to their religious belief and practice and to interpret it as the grace of
God operating through them. But we may find similarly admirable individ-
uals with a different religious faith or none; and even with the believers we
cannot prove that what we admire in them is produced by the influence of
God, rather than by a combination of their innate disposition, the culture in
which they have been brought up, and their own choices. Kant sums up the
situation in a footnoted comment on Exodus 33:19:
No human being can say with certainty why this human being becomes good,
that one evil (both comparatively), for we often seem to find the predisposition
that makes for the distinction already at birth, and even contingencies of
life over which nobody has any control are at times the decisive factor;
and just as little can we say what will become of either. (R 6:122n)
4
Why then does Kant feel the need to invoke the concept of divine grace at
all, if only as something unknowable that we may hope for, but can never
experience for sure? Is there any reason to follow him down this rather
peculiar path? I suggest it may be illuminating at this stage to look at ethics
and the moral life from a non-theist but morally serious perspective. After
all, Kant himself wrote in his Preface to the first edition of the Religion that
“morality in no way needs religion (whether objectively, as regards willing,
or subjectively, as regards capability) but is rather self-sufficient by virtue of
pure practical reason” (R 6:3), though he went on to say a few pages later
that morality inevitably leads to religion (R 6:6). I take it the objective bit
means that the content of our moral duties can be discerned by reason alone.
And if the subjective bit about “capability” is to be equally self-sufficient,
Kant must presumably mean by it our fundamental freedom to fulfill or to
ignore our moral duties, i.e. the “fact of reason” that he asserts in the
Critique of Practical Reason and reaffirms in Part One of the Religion
when he says that the evil in human nature, radical though it is, remains
imputable because it involves “a deed of freedom” (R 6:21). The tension
between the Pelagian assertion of unavoidable human responsibility and the
Augustinian assertion of innate human sinfulness remains.
Let us consider then a humanist ethic that agrees with Kant in thinking
that there are objectively valid moral duties, some of them negative – “Do
not kill,” “Do not steal” – and some positive – “Honor your parents,”
Kant on grace 133
“Look after your children.”22 It can be left as an open question whether
these are self-standing, or derivable from some more abstract Kantian
principle about universalizability or treating people as ends. And, on the
face of it, there is no need to accept Kant’s rigorist thesis that human beings
cannot be partly good and partly evil (R 6:24–25), which seems to fly in the
face of experience, for we hear of concentration camp guards being loving
fathers and of zealous campaigners for good causes neglecting their own
children. Perhaps most of us for most of the time can avoid outward
transgression of the negative rules of law and common morality, but we
find it more difficult to live up to the positive duties, for they are so general
and vague: there will always be some indeterminacy in just how much one
ought to do for one’s parents or children, for the forgetful old lady next door,
or for the starving babies in Africa. We may also have trouble balancing
conflicting duties to parents, children, spouses, workplace, and other com-
mitments without ignoring or hurting some of the people involved, or
losing one’s own cool. In face of the ordinary pressures and temptations of
life, we all need to learn how to deal with our own resentments, anger, fear,
envy, pride, sloth, lust, and greed.
“Living up to the demands of morality” can thus be interpreted in a less
demanding (though non-trivial) way, concentrating on not infringing the
rights of others and on fulfilling the most obvious kinds of family duty.
But all that, necessary as it is, does not tell us what we are living for – apart
from reproducing ourselves. There is a more demanding level of thought
about how we ought to live (which may deserve the label “ethics” rather than
“morality” – though not for Kant, who did not make any such distinction)
that sets before us ideals, for example of good parenthood, self-control,
compassion, and generosity to the disadvantaged. It is fair to say that none
of us manages fully to live up to these standards of perfection, with the
possible exception of some saintly figures of history – though one may
wonder if their imperfections were expunged from the record! Jesus’ teaching
as reported in the gospels clearly raises the bar to the second level, telling his
legalistic Jewish interlocutors that it was not enough to fulfill the letter of
the ancient commandments. So does Kant’s ethics, taken in the round, for
he includes “imperfect duties” in the Groundwork, and in the Metaphysics of
Morals he sets out as the two fundamental ideals or ends of human life the
perfection of one’s own capacities (including the virtues) and the happiness
of others. In the Religion he says there is always an infinite distance between
our actual lives and such ideals (R 6:72).
22
Why was the last not included in the Ten Commandments, one wonders?
We do not need to be card-carrying Kantians or devotees of any particular

religion to accept this two-level picture of morality. Moral “respectability”
at the first level is humanly achievable (though far too many people fail that
test), but we can recognize that much more is ideally demanded of us. It is
tempting to plead our incapacity to measure up to that second level, and
thus – on the principle that “ought implies can” – to try to argue away any
obligation to do better than the human average (whatever that is). But there
is ample opportunity for self-serving self-deception here, for how can we
be so certain about what we can or cannot rise to? As Kant repeatedly says,
we do not know our own hearts, so we do not know what we may find
ourselves able to do in hypothetical circumstances, or indeed what changes
of “heart” or disposition we may undergo, or bring about in ourselves. And
we do not know other people any better.
To be sure, there is all manner of evidence about physiological, psycho-
logical, and sociological factors that influence human behavior and character,
which it would be stupid to ignore. But to say that a combination of such
factors predetermines what a conscious individual will feel and say and do in
every particular situation is a claim that always goes beyond the evidence. As
I have argued elsewhere, that involves an impossible ideal of a complete and
precise knowledge of a state-description of all the relevant facts in the world,
plus a complete set of all the universal laws that apply.23 This may be a
regulative ideal that guides the progress of physical science, but it will surely
never be fulfilled. For all practical purposes, we cannot predict what we will
feel or do in every future situation, neither about ourselves nor about others.
As Kant himself said toward the end of the Religion: “freedom itself, though
not containing anything supernatural in its concept, remains just as incom-
prehensible to us according to its possibility as the supernatural [something]
we might want to assume as surrogate for the independent yet deficient
determination of freedom” (R 6:191). Thus, the unknowable exercises of free
will seem to take the place of the unknowable operations of the supernatural
in Kant’s system.
If the ethical ideals of universal love and compassion capture our imag-
ination, so that we feel an obligation at least to try to live up to them in
character and in action, then obviously we will hope that we can make
progress in the right direction; but, as has just been argued, we can never
know for certain that we will. What difference remains then between this
hope that the morally serious humanist can entertain, and Kant’s hope for
divine grace (of the sanctifying kind) of which we can know nothing at all?
23
See Leslie Stevenson, “A Kantian Defense of Freewill,” in Stevenson, Inspirations from Kant, Ch. 9.
Kant on grace 135
Is it merely putting a different label on the same situation? Does Kant’s talk
of grace amount to no more than a pious phrase, as in the saying “There but
for the grace of God go I” (which in most mouths means only: “It is, as far as
I know, a matter of sheer luck that I am not in the same situation”), or the
talk of “acts of God” in old-fashioned insurance policies (which presumably
meant events that are beyond all human control)?
It may seem difficult to make any humanist sense of the notion of
justifying grace that Kant takes so seriously in Part Two – but I am going
to try. As usually explained, justifying grace presupposes belief in God, for it
offers to explain how we can be “justified in the sight of God” despite our
sinfulness, whereas the result of sanctifying grace can be expressed in secular
terms as making us into better people. However, Kant’s appeal to the
God’s-eye point of view might be interpreted as a conceptual device for
invoking an ideal standard of moral/ethical judgment which we may
approximate to but can never completely fulfill, for he talks of “the accuser
within us” and our stern judgment upon ourselves (R 6:77). So perhaps the
problem to which justifying grace is offered as an answer might be reformu-
lated in secular terms as follows: given our painful awareness of ourselves
and each other as imperfect beings (however much progress we make), how
can we live with ourselves and each other? We may be handicapped by
feelings of guilt, not necessarily in a neurotic way worthy of psycho-analysis,
but just because we have a realistic appreciation of the gap that always
remains between the ideals we recognize and our actual attitudes and
actions. So isn’t there a need for forgiveness? It is difficult to forgive ourselves
without the mediation of another person, and without being self-indulgent,
and it is difficult to forgive others when real hurt has been done, even if
regret is expressed. We may, if we are fortunate, benefit from the attention
of a wise counsellor or therapist, a confessor, spiritual director, or guru, to
help us live better both for ourselves and for those we relate to. Such a figure
may explore with us what we feel remorseful about in ourselves, and what
we feel resentful about in others, and may be able to help free us from
guilt and anger or at least reduce them to manageable proportions. But there
is no guarantee of success, for human actions and reactions are inherently
unpredictable.
In the gospel of Luke (19:1–10), there is a dramatic story of the sudden
turnaround of a life. Zacchaeus was a tax-collector who had become very
rich at the expense of others and, like today’s bankers, was generally reviled
for it, but having heard about Jesus, he was curious to see the famous
preacher and healer for himself. When Jesus noticed Zacchaeus peering
from a tree, he straightaway invited himself to stay at his house, with the
reported result that Zacchaeus decided to give half his possessions to charity
and to repay anyone he had defrauded four times over. Here there was
a minimal movement on the part of the sinner (if only of idle curiosity), a
surprising offer of loving companionship from a person of charisma, the
acceptance of that offer, and a dramatic change in attitude to life. Such
radical conversions can and do happen, but there is no guaranteed response
to offers of compassion and love; human actions and reactions remain
inherently unpredictable, as we have repeatedly noted.
My conclusion is that all three of the positions reviewed here – that of the
thoughtful non-dogmatic Christian, and of the ethically sensitive humanist
who recognizes the reality of good and evil in human nature, with that of
Kant delicately poised somewhere in-between – have to acknowledge an
element of ultimate mystery in explaining and dealing with human behavior
and character. Our freedom and the manifold complexity of the influences
upon us mean that complete scientific explanation of human behavior
will always elude us. Yet we need, as Kant argued so deeply, to face up to
that mystery with a sense of forgiveness (justifying grace?) and hope (for
sanctifying grace?).
chapter 7
Kant, miracles, and Religion, Parts One and Two

Karl Ameriks
1. A Janus-faced achievement
Kant’s Janus-faced Religion within the Boundaries of Mere Reason is a
surprising and perplexing book. Its mere existence is highly surprising
because, after having already laid out his critical views in detail, there would
seem to have been no need for the aged Kant to publish a long volume on
very controversial religious topics. The work is Janus-faced and perplexing
because its content could only provoke hostility from both his secular and
religious acquaintances. The book’s advocacy of Vernunftglaube, the “faith”
of a moral religion of “reason,”1 combines a commitment to the secular
Enlightenment belief in human perfectibility and progress with a resolutely
non-secular insistence on a complex revolutionary and theodicical shape
to that progress. Kant’s conception of our final end borrows heavily from
non-natural Christian ideas (absolute freedom, radical evil, unconditional
value, a transcendent moral divinity), and yet it also assumes that all tradi-
tional rationalist, experiential, scriptural, and ecclesiastical routes toward
establishing that end need to be, and can be, critically undermined.
Kant did not work out the complex theodicical aspect of his conception
of Vernunftglaube all at once. Its basic idea goes back decades, and Kant’s
late works contain a series of final fine-tunings that deserve close analysis. In
addition to the four complex essays that constitute the Religion volume,
there are several closely related 1780s essays on history and at least half a
dozen very relevant later essays.2 Taken together, these works constitute a
final bold attempt on Kant’s part to reinterpret the Western teleological
tradition in a way that is, all at once, deeply religious, liberal, scientific, and
philosophically nuanced. Kant’s attempt is also riddled with theological
1
The translation “rational faith” does not do justice to Kant’s term, Vernunftglaube, which is defined by
the unconditional commitments unique to the faculty of reason, which in principle surpasses the
rationality of mere understanding. See, e.g., R 6:26n.
2
See, especially, Theod (1791), End (1794), Ton (1796), IC (1796), and S (1798).
137
138 KARL AMERIKS
“conundrums,”3 but I will leave these problems aside in order to focus on

one central philosophical component of his Janus-faced approach − namely,
his treatment of miracles at the very end of Parts One and Two of the
Religion.
In a long second-edition note added to the end of the General Remark at
the end of the first of the four Parts of the Religion, Kant makes explicit a
theme common to each General Remark that concludes a Part of the book.
Each General Remark considers one of what Kant calls the four kinds of
“extravagant ideas” that are parerga to pure moral religion but understandably
occur to reason as it reflects on its limitations in fulfilling our moral needs
(R 6:2). These four kinds of ideas are neatly organized under the headings of
“Effects of Grace,” “Miracles,” “Mysteries,” and “Means of Grace,” but the
ideas are not all on the same level, and determining exactly how they differ is
no easy task.
For our purposes it will be enough to concentrate on an issue related to
the contrast between ideas of the first two kinds. Even though Kant never
explicitly denies the possibility of miracles – which he defines early on as
items that “interrupt the order of nature”4 – he seems increasingly interested
in strongly discouraging appeal to them, especially – but not only – as
“means of grace.” With respect to mere “effects of grace,” however, it turns
out to be difficult to make sense of Vernunftglaube’s own central notion of
religious hope without appealing to effects that are at least somewhat like
miracles. As his General Remark to Part One stresses, Vernunftglaube
involves not only an initial moral effort of our own but also hope that
then “what does not lie in [our] power will be made good by cooperation
(Mitwirkung) from above” (R 6:5).5 The mention of Mitwirkung “from
above” implies reference to some kind of “effect of grace” and a providential
action upon us by a non-natural power. Passages like this generate the basic
philosophical challenge of finding a way in which Kant’s invocation of any
3
See Karl Ameriks, Kant and the Fate of Autonomy: Problems in the Appropriation of the Critical Philosophy
(Cambridge University Press, 2000), Ch. 7; Nicholas Wolterstorff, “Conundrums in Kant’s Rational
Religion,” in Philip J. Rossi and Michael Wreen (eds.), Kant’s Philosophy of Religion Reconsidered
(Bloomington/Indianapolis: Indiana University Press, 1991), pp. 40–53; Andrew Chignell, “The Devil,
the Virgin, and the Envoy: Symbols of Struggle in Religion, Part Two, Section Two,” in Otfried Höffe
(ed.), Die Religion innerhalb der Grenzen der blossen Vernunft (Berlin: Akademie Verlag, 2011), pp. 111–30;
Chignell, “Rational Hope, Moral Order, and the Revolution of the Will,” in Eric Watkins (ed.), Divine
Order, Human Order and the Order of Nature (New York: Oxford University Press, 2013), pp. 197–218; and
Robert M. Adams, “Introduction,” in Allen W. Wood and George di Giovanni (eds. and trans.), Religion
within the Boundaries of Mere Reason and Other Writings (Cambridge University Press, 1998), pp. vii−xxxii.
4
EM 2:116; cf. ML1 28:216.
5
See Karl Ameriks,”Das Schicksal von Kants Rezensionen zu Herders Ideen,” in Otfried Höffe (ed.),
Immanuel Kant: Geschichtsphilosophie (Berlin: Akademie Verlag, 2011), pp. 119–36, for a discussion of
Kant’s 1785 critique of Herder for appealing to what comes “from above.”
Kant, miracles, and Religion, Parts One and Two 139
non-naturally based effect can be understood to be sufficiently unlike the
superstitious reference to miracles that he heavily criticizes in Part Two and
other texts.6
One strategy for dispelling part – but only part – of this problem is to
note that Kant distinguishes quite different ways of invoking ideas that go
beyond ordinary nature, and he immediately adds explicit qualifications to
his own reference to help “from above.” In the body of his first General
Remark Kant stresses that Vernunftglaube, even in its most developed form,
cannot claim to amount to “conviction” (Überzeugung, R 6:51), the strong
epistemic attitude that comes with knowledge (Wissen). In addition, near
the end of a long second edition note to this Remark, Kant stresses that,
even with Vernunftglaube, we can never “summon” any effects of grace by
“incorporating” them into either a theoretical or practical “maxim of
reason” (R 6:53).
Kant may have felt a need to add this point, and perhaps the note as a
whole, as a matter of balance, because otherwise it might appear unclear
whether, in regard to mere effects of grace, he has anything at least some-
what negative to say here that would link up with the highly negative points
about references to the supernatural that are stressed in the other Parts of the
Religion. Nonetheless, essential to Vernunftglaube is religious hope, and
essential to this hope is still some kind of positive reference to some such
effects. These effects must be thought of as more than simply not (as far as
we know) impossible – for, obviously, such a weak position could be held by
all sorts of persons without faith. It is not necessary in this context to make
much more precise the status of Kantian hope as a specific epistemic
attitude.7 What matters here is simply that there must be some kind of
affirmative thought, some “holding to be true” (see the discussion of assent
at KpV 5:142; KU 5:461, 467) about some special effect of the relevant sort –
even though, as Kant repeatedly insists, this thought can never be absolutely
“certain” or come with any explanatory “insight” (Einsicht, R 6:50).
Many of the perplexities just noted in the status of Vernunftglaube are also
relevant to Kant’s more basic notion of our absolute free agency, which is
nothing less than the keystone concept of the Critical philosophy. In
the Religion, a positive reference to an “effect of [our] free power of choice”
(R 6:4) is the starting point of the first General Remark, and yet Kant
6
See also Kant’s notes, e.g., Ref 5662 (“On Miracles”) 18:320, and 5997 18:419.
7
See Andrew Chignell, “Belief in Kant,” Philosophical Review 116 (2007), pp. 323–60; and Karl Ameriks,
“The End of the Critiques: Kant’s Moral ‘Creationism,’” in Pablo Muchnik (ed.), Rethinking Kant.
Vol. I (Newcastle upon Tyne: Cambridge Scholars Press, 2008), pp. 165–90.
140 KARL AMERIKS
repeatedly stresses that here too we can have no insight into the “how,” or
even any unqualified certainty about the “that,” of our freedom. All we can
say is that the assertion of effects of our freedom supposedly follows from –
and only from – acceptance of something provocatively described as a “fact
of reason.”8 This is something taken to be, for each practical agent, a
fundamental truth involving necessities, and hence it is called a matter of
“reason”; but it is a non-demonstrable truth, and hence it, or at least our
relation to it, is also called a kind of “fact.” This acceptance can also be
described as the “deed” (Tat) of recognizing oneself as a moral being in a
strict sense. It involves what Kant even calls a kind of “moral certainty”
(A829/B857), but this is a form of certainty essentially qualified by the fact
that it requires a first-person commitment to the non-natural demands of
Kant’s strict moral perspective.
Kant’s endorsement of absolute freedom is, to be sure, not completely on
a par with his endorsement of the hope involved in Vernunftglaube. His
analysis of hope has built into it a presumption of the endorsement of
freedom, whereas the mere analysis of our endorsement of freedom does not
by itself entail the more complex commitment to religious hope and its
implications. Nonetheless, in Kant’s system the perplexing thought of our
freedom still seems relevantly enough like the perplexing thought of effects
of grace insofar as both thoughts essentially invoke a non-natural and non-
demonstrable ground of empirical effects. Hence, even prior to Kant’s
carefully worded endorsement of possible effects of an external higher
freedom acting upon us in grace in response to our devotional freedom,
there is already the more general interpretive puzzle of explaining the deep
asymmetry between, on the one hand, Kant’s negative attitude toward
affirming any specific miracles and, on the other hand, his carefully worded
but frequent affirmation of effects of our own everyday absolute freedom.
The issue of miracles dominates Kant’s General Remark to the Religion’s
second Part, and its critical tone is set in its first sentence, which proclaims
that, after the enlightened spread of Vernunftglaube, “faith in miracles” “even-
tually” will be rendered “in general dispensable” (R 6:84). This Remark still
does not declare miracles non-existent, let alone impossible (R 6:88n), but it
can be read as philosophically allowing the notion of them merely as (for all
we know) a bare logical possibility, one that has played an understandable,
but temporary, role earlier in history simply because of lack of adequate
education. A complication here is the fact that, because of issues having to
do with very sensitive church−state issues at the time, Kant’s Religion is
8
KpV 5:42; cf. R 6:50n, and PPV (1793) 27:506.
especially concerned with arguments against the notion of making public
commitment to miracles a test of faith and a matter of ministerial certifi-
cation (R 6:85n).9 This focus on the political aspect of the issue should not
distract one from noticing that Kant is also very opposed to the thought that
faith requires any sort of private belief in miracles. Because the enlightened
requirements of Vernunftglaube must be of a sort that are open to being
satisfied in principle by all human beings, simply on the basis of their
original full rational equipment, there is a kind of public condition – a
condition of real universal access – on what one can even be expected to
need to believe or affirm privately. This condition is taken to rule out
requiring, for genuine internal faith, any appeal to a special experience of the
miraculous, that is, any strictly local and historical event. Kant’s presump-
tion here is that because a miracle – whether “great” or “little,” past or
present, rare or repeated – must be thought of as a particular event said to
be accessed under unusual conditions revealing something contrary to the
whole order of “worldly events” (R 6:85n), it is not something that can be
fairly expected to be affirmed by persons as such, wherever they are, and
hence insisting on belief in such a thing would violate the principle of
rational autonomy.10 At the end of this chapter, I will argue that the
universal accessibility aspect of this presumption has significant implica-
tions for evaluating Kant’s own position.
Kant’s language in the Religion attempts to dance around the issue of
exactly how to talk about miracles and the non-natural in general without
being either offensive to his readers or untrue to his own religious concerns
and critical perspective. This perspective cannot, of course, ever absolutely
exclude non-natural sources for anything, and such sources are by no means
a far-fetched option for Kant. He points out himself that, even with respect
to phenomena obeying natural laws, the Critical philosophy must “renounce
cognition of that which brings about effects according to these laws, in
itself” (R 6:88). Given transcendental idealism, it follows that in principle
we can have no insight into the things in themselves that ultimately do
the genuine “bringing about” (what Kant calls the ultimate “causality of the
cause” at A533/B561), for this goes beyond whatever we can know in terms of
the relational principles governing spatiotemporal phenomena. Kant adds
in Religion that a similar point applies in the context of our own self and its
9
Cf. Theod 8:268n. Kant holds such tests to be not only unnecessary for genuine morality and religion
but also directly contrary to both because of the ways in which the tests can pressure people into
hypocrisy and an improper concern with effects rather than pure internal attitudes. See R 6:62ff.
10
Cf. Ton 8:393 n. 3, which totally rejects affirming testimony regarding supernatural beings or events,
despite “the subjective impossibility of my [own] inability to explain” them.
142 KARL AMERIKS
attempt at moral improvement, where we “have no understanding of how

to distinguish with certainty [non-natural] influences from natural ones,” and
so, for all we know, “heavenly influences” might “cooperate” here and be
“necessary” in an ultimate – but to us inaccessible – explanatory way (R 6:88).
In alluding to something beyond both external and internal phenomena,
Kant is, I presume, not making a merely hypothetical statement but is
referring to some non-spatiotemporal “I know not what” that he is obliged
to assert exists, at the same time that he has to say that we cannot determine
specifically what it is, beyond the very general characterization of it as
whatever ultimately allows for experience and the fulfillment of our moral
agency. For this reason, in addition to criticizing directly the main advocate
at the time of appeal to miracles,11 Kant reiterates, in a second-edition
footnote (R 6:88n), the First Part’s key claim that no appeal to miracle is
ever to be brought into our theoretical or practical maxims. Although Kant
does allow something loosely called a “theistic miracle,” it is important that
this is “only a general concept” of a “creator and ruler of the world, according
to the order of nature” (R 6:86), and so for Kant this is precisely not the
common concept of a miracle as an event that is within the world but
contrary to the order of nature.12 With respect to God’s particular effects,
beyond assuming that they would never violate theoretical or moral reason,
Kant insists that our reason cannot even “ever hope to be instructed in the
world” (R 6:87).13
In a final footnote to this Remark, Kant speaks of how events that strike
us as highly unusual are only “so-called miracles of nature” (R 6:88n).
They should not be presumed to be “genuine miracles,” for this would
only confuse and “deject” our reason, and we would then even have to
worry that all our supposedly moral thoughts may also be simply driven
into us in some miraculous way. Unusual events normally are, and should
be, taken to be occasions for “nourishing reason” and the theoretical “hope
of discovering new laws of nature” (R 6:8n). Kant dismisses worries that
we have “no cognition of the cause of gravity,” for we do have “cognition of
the laws of these forces” sufficient for “regressive employment . . . in the
11
See the discussion of Johann Caspar Lavater in a footnote (R 6:6n) that crisply expresses lines of
thought that Kant had already expressed very eloquently in a letter to him on April 28, 1775: Brief
10:175–9.
12
Cf. MD (1792–93) 28:667.
13
The same point is found already in earlier lectures: “according to the nature of our reason we have to
hold on to the universal and not try to determine how divine providence has proven itself effective in
particular cases” (VpR [1783–84] 28:114). Cf. MPC (1774–77): “Everything lies in universal providence
and it is actually better in our discourse to abstain from trying to determine anything of God’s
intentions” (27:320).
ordering of experiences under them” (R 6:88n). In his final sentence to this
footnote and Part Two as a whole, Kant responds directly to those who
“pretend to have insight” (my translation) into how the very complex
natural developments that come with every return of spring are a matter
of the “immediate influence of the creator” (R 6:89). Against this pretense,
Kant dramatically proclaims: “But these are experiences (Erfahrungen); for
us, therefore, they are nothing but effects of nature, and ought never to be
judged otherwise,” and he closes by saying that to accept this point is the
true “modesty” of reason, in contrast to the false humility of invoking
miracles (R 6:89). The appeal here to the notion of Erfahrung is, of course,
not a matter of crude representationalist empiricism but a Kantian
reminder of the common-sense Faktum underlying the entire critical
system,14 which is a structured and cognitive domain, and one that turns
out to be law-governed in a very strict way without any reference to
miraculous interventions.
The fact that Kant stresses the words “for us” and “ought” is also an
implicit reminder of another aspect of what he means by reason’s modesty −
namely, that he is speaking only about what our maxims should be for
making determinate claims, and he does not go as far as making any absolute
denials beyond these maxims. All the same, I take his vivid language to
indicate that his real point is, roughly speaking, that of course it is possible
to try to introduce miracles in nature, but a rational person should not do so,
given where we in fact stand, with our well-structured domain of Erfahrung.
Kant’s tone here indicates, albeit indirectly, that one can read as sarcasm his
earlier sentences in the text proper, which allow that “it might well be”
(es mag also sein) that Jesus’ (named only as “the teacher of the gospels”)
“appearance on earth, as well as his translation from it, his eventful life and
his passion, are all but miracles – indeed that the history that ought to testify
to the account of these miracles is itself a miracle” (R 6:85). Given its
context, I take this suggestion – that, once one starts on this path, one
might as well introduce second-order miracles – to imply that at this point
Kant means to discourage the affirmation of any particular miraculous
works. Such a stance, however, once again raises the question of whether
one might not respond similarly to Kant’s own approach, and argue that as
soon as one introduces any effects with non-natural sources, as he still does,
then one might as well be more open to something like literally miraculous
sources too.
14
See Karl Ameriks, “Introduction: The Common Ground of the Critiques,” in Ameriks, Interpreting
Kant’s Critiques (Oxford: Clarendon Press, 2003), pp. 1–48.
144 KARL AMERIKS
2 Critique of pure wonder

The issue of miracles is not a matter of casual interest for Kant, for he
repeatedly stresses another way of speaking that plays off the language of
miracles but gives it a critical twist. The first General Remark links Kant’s
central notion of a “revolution in our way of thinking,” as in the “new
creation” of a person with Vernunftglaube (R 6:47), with the attitude of what
he calls proper “admiration (Bewunderung) for virtuous actions” (R 6:48). In
the German edition (but not in the Cambridge translation, which, by using
the term “wonder,” may obscure the sharpness of the contrast intended
here) it is impossible to miss Kant’s main point here. Kant’s aim is to
substitute for reference to Wunder, that is, literal miracles, a supposedly
much more modest reference to Bewunderung, that is, a common deep sense
of amazed admiration with regard to examples of committing one’s life to
respect for the moral law. This key terminological strategy of Religion is
anticipated in Kant’s slightly earlier essay on theodicy, which uses the term
Bewunderung to describe the proper reaction to the story of Job.15 Kant takes
this story to teach us to respect the primacy of pure morality and sincerity, as
opposed to any claim to be able to discern specific actions of God in the
world as specially intended punishments. Although the essay is perhaps best
known for its title reference to a “miscarriage” or “failure” of theodicies,
Kant’s basic point is rather that, although one cannot expect speculative
philosophy to show how the world does in fact serve God’s purposes, this is
consistent with accepting what he calls “authentic theodicy,” which takes
proper service to God to rely basically on accepting the absolute primacy of
pure morality. The “end,” in the sense of the limitation or disappearance of
traditional theodicy, is thus for Kant only the beginning – and is to lead to
the goal and victory – of what he takes to be the genuinely theodicical − that
is, pure moral − attitude.16
15
Theod 8:270. This is a rare positive reference by Kant to the Old Testament. On the one hand, this
may have implications that complicate his account of the fundamental moral revolution in history
that he says comes from the gospels, but, on the other hand, it may reinforce his deeper view that
respect for the moral law is in principle possible in any age.
16
Theod 8:264. One could therefore argue that Kant’s work concerns the “end of theodicy” in a positive
historical sense in that it itself constitutes the end, in the affirmative sense of a culminating final
version, of mainline Western philosophical attempts to provide at least a close analogue to traditional
theodicies (which promise hope for a kind of personal salvation grounded in the power of a personal
God) and to make this a central aspect of one’s system. (Here I am assuming that, except for perhaps
the later Schelling, most of German Idealism is much less traditional in theology than Kant.)
Nonetheless, Kant’s work can be said to be an “‘end of theodicy” in a negative historical sense as
well because, after him, attempts to justify substantive providential claims no longer play a central role
in philosophy. There is, of course, a very significant renaissance today of apologetic work in the
That this kind of discussion of miracles is very important for Kant is also
clear from the fact that it is emphasized again in the first Part of his very last
essay, The Conflict of the Faculties. The General Remark in this Part of the
Conflict is devoted to the advocacy of our “moral metamorphosis” (S 7:55) in
a pure religious “revolution” (S 7:59) that would take us beyond all sectari-
anism. Kant goes out of his way here explicitly to reject the recourse
(“before” or “after” our change of heart) to miracles in his time by traditional
Pietists and Moravians (S 7:55ff.), and then he immediately turns to the
“something in us that we cannot cease to wonder at [admire, bewundern],”
namely, the moral law that “lies objectively in the natural order of things [NB]
as the object of pure reason” (S 7:58). Kant speaks here of our “höchsten
Bewunderung,” with italics, and he also repeatedly uses a verb form of the
term (S 7:58ff.). For significant reasons that will be discussed later, Kant
contrasts the appreciation of what he here calls the “supersensible” practical law
that is “in us” − that is, contained in our “nature” as beings of reason − with the
“greatly mistaken” move of those “who are led to consider it supernatural –
that is, to regard it as the influence of another and higher spirit.”17
The repeated use of the term “revolution” in Religion is clearly meant to
resonate on a number of authentic theodicical levels at once. In addition to
the revolution discussed first simply in terms of each individual person’s
conversion to pure morality, Kant claims repeatedly that there is a second,
related general revolution within “the human race” (R 6:63; cf. R 6:80, 81n,
and 84), one that was inaugurated, although not completed, by the gospels.
Kant goes so far as to say that the innovative moral attitude of the teacher of
the gospels fundamentally excels anything found in prior philosophy (R
6:80; cf. MPC 27:301), and that the revolutionary ideal that defines it, and is
exemplified in stories of the teacher’s life, is the major force (“that quietly
spread everywhere”:R 6:81n) behind the ultimate direction of all subsequent
history. This ideal introduces “a realm . . . in which nobody is therefore
slave” (R 6:82), for “by exemplifying that principle (in the moral ideal) that
human being [“the teacher”] opened the doors of freedom to all” (R 6:82).
The main theme of Part Two of the Religion, and especially of the subpart
concerned with “the personified good” and the “objective reality” of the
ideal present in the gospels (of being morally “pleasing to God,” which
alone can provide “an end to creation”: R 6:60), is that a relation back to this
philosophy of religion, but most of the best of this work is still limited to argumentative defenses of
general possibilities rather than the construction of ambitious theodicies that claim to offer provi-
dential explanations in the strong sense of earlier traditions up to Kant’s time. For a reminder of the
importance of this point I am indebted to participants in the conference on “God’s Order, Man’s
Order, and the Order of Nature” at the University of California, San Diego, March 4–6, 2011.
17
S 7:59. See Ton 8:396n, for a similar rejection of those who infer an external “supernatural influence”.
146 KARL AMERIKS
very ideal – as an ideal model rather than an external miraculous fact – is

what is crucial for each individual moral “revolution” (R 6:47) in modern
life. In addition, Kant links such modern individual revolutions forward to
the new ideal of an autonomous social realm, one whose general appeal he
explains in terms of the effects of a recent philosophical development. That
development is the popularization of the idea of autonomy (that is, freedom
in human action that is rationally lawful) through the “genius” of
Rousseau’s work – work that can itself be called a third kind of critical
revolution. This work literally turned around the direction of Kant’s own
thought in the 1760s,18 and so it is no accident that the very beginning of
Religion (R 6:19ff.; cf. R 6:38, 45, 54, and 66) links the theme of modern
social optimism to Rousseau’s awareness of the “seed (Keim) of goodness” in
humanity as such. According to the Collins lectures, Kant held that, “many
have maintained that in man there are no seeds of good, only of evil, and
Rousseau alone preaches the opposite.”19 Rousseau’s revolutionary philo-
sophical achievement is to begin to turn cultivated modern humanity away,
at a level that is itself reflective and literary, from the specifically modern
problem of a combined absolutization of the life of luxury and scientific
preoccupation,20 just as the gospels began to turn naïve ancient humanity
away, at a level that is itself religious and exemplary, from the obsession with
priestly trappings and superstition that Kant takes to define the surrounding
pre-Christian world.
The last two Parts of Religion, along with other late essays related to it, fill
out Kant’s revolutionary narrative by taking the “enthusiastic” affirmative
response of common people throughout Europe, in their unselfish “sym-
pathy” toward the basic anti-elitist ideal of the French Revolution, to be an
“irreversible” sign of humanity’s entrance into the political antechamber of
history’s final era21 – and thus to constitute a critical revolution in yet a
fourth and most concrete sense. This “sign” bolsters Kant’s own hope that
individual moral revolutions will be combined more and more with repub-
lican and peaceful political reformation, and that an enlightened “invisible
church” will move humanity as a species asymptotically toward an
18
See Karl Ameriks, “Kant, Human Nature, and History after Rousseau,” in Susan Shell and
Richard Velkley (eds.), Kant’s “Observations” and “Remarks”: A Critical Guide (Cambridge
19
MPC 27:317; cf. MM2 (1784–85) 29:603.
20
Kant calls such affluent times, when to many people “creation appears purposeless . . . like a play,”
“the most burdensome and dangerous for morality” (End 8:331–32).
21
S 7:85. Kant sees a somewhat similar “sign,” at the individual level, allowing a person to detect some
apparent past progress in commitment to his own moral principles, “if he has perceived the efficacy of
these principles in what he does” (R 6:68).
institutionalized realization of the theodicy of Vernunftglaube.22 Within this
last phase, works such as Kant’s own critical system and Enlightenment
essays, as well as similar works by allies such as Reinhold, can be understood
as intended to be combined with yet a fifth and final revolution, the late
modern critical turn established in Kant’s system. This turn aims at perpetually
securing, at a metalevel, the insights of Christianity’s and Rousseau’s moral
visions, as well as those of common humanity and the fans of the general
idea of the French Revolution, by solidly protecting them from future
contamination by dogmatic or naïve misunderstandings.23
Kant’s use of the term Bewunderung is therefore hardly casual. It is clearly
positioned in a place that is central to his multistage theodicical account of
how a pure moral religion is supposed to wean modern society, through
various interlocking revolutions, away from appealing to miracles and
insisting that supernatural beings are literally part of human history.24
The puzzle remains, however, that at the same time that Kant works out
a progressive Enlightenment view of society and history, he continues to
make what can now seem to be extraordinarily immodest non-natural
remarks of his own about how all human beings can and should believe
in really being able to work absolutely freely toward the highest good. In the
end, the hopeful attitude of Vernunftglaube earns its name as a kind of faith,
for a convinced Kantian must be ready to affirm nothing less than an
extraordinarily substantive three-part creed: that existence on the whole
is a teleologically unified complex of, first, a natural sphere that is fully
law-governed although not by itself moral or containing miraculous inter-
ventions, and, second, a moral sphere that is also law-governed in its own
way but not itself either sensible or literally miraculous, and yet is such that,
third, all the non-natural features just reviewed fit together marvelously, so
that the laws of morality also turn out to govern the general shape of the laws
of the natural world and world history. Kant admits right from the start that
22
Or else, if, as is also possible, freedom is misused and Christianity as a cultural institution “ceases to be
worthy of love,” then, as Kant provocatively says (in a not-so-subtle reference to the dangers of the
regime of Frederick William II), it will appear that “the Antichrist” reigns (End 8:339).
23
In notes from Kant’s pre-1788 lectures to theology students, there is Lutheran language that still
expresses this idea in terms that themselves sound dogmatic: “The foundation on which he [the
Kantian moral theist] builds his faith is unshakeable and it can never be overthrown, not even if all
human beings united to undermine it. It is a fortress in which the moral human being can have no fear
of being driven from it, because every attack on it will come to nothing. Hence his faith in God built
on this foundation is as certain as a mathematical demonstration. This foundation is morals” (VpR
[1783–84] 28:1011).
24
Kant goes as far as making the heterodox argument that Jesus’ moral effectiveness rests on his not
being thought of as literally divine, for only then is he a model that human beings can understandably
attempt to imitate. See R 6:64.
148 KARL AMERIKS
each step in his multistage revolutionary epic of freedom25 rests on non-

sensible factors that concern our own inner being but whose operation
remains “absolutely inexplicable” (schlechterdings unerklärlich) to us (R 6:9n).
This final position is so ambitious that it can hardly be given a convincing
quick defense, but one can at least try to understand its origins and status
relative to other directly relevant positions.
3 Hermeneutical hypothesis
Kant’s theodicy implies not only a very expansive teleological conception of
each of the cosmological and moral orders by themselves but also a very
strong commitment to their tight linkage. It is precisely these two orders
that are referenced in the most famous passage in Kant’s works, namely, the
second Critique’s comment about our “ever increasing” Bewunderung and
Ehrfurcht with regard to the “starry heavens above” and the “moral law
within” (KpV 5:161). What this comment should remind us of now is not so
much the individual features of these two contrasting sources of our amazed
admiration and awe, but rather the fact that this text connects them in one
grand statement, a statement that can be taken as a clue that, from the very
start in Kant’s mind, these orders are much more closely related than
contemporary readers may tend to assume.
It is true, of course, that after his early 1760s turn to Rousseau and his
dramatic realization that the distinctive fulfillment of the human species
concerns our practical rather than exclusively theoretical capacities, Kant
frequently contrasts the determined theoretical order of nature with the pure
practical order of absolute freedom. Nonetheless, after this turn Kant also
insists that reason is unified, and he eventually sees that theoretical philos-
ophy must find a systematic means – namely, transcendental idealism – that
could provide metaphysical room for our practical reason to be an absolute
free source of effects, despite all that the Critical philosophy entails about
nature as a strictly lawful spatiotemporal order. The key move here in Kant’s
practical philosophy is his Rousseauian abandonment of the notion that
freedom must come with lawlessness (cf. A446–47/B475–76), and his
development of a conception of what he calls our “nature” as free beings:
a nature that, like all natures, is governed by a law, in this case the moral law.
As has been noted above, this is an order that Kant sometimes prefers to call
“supersensible” (übersinnlich) rather than “supernatural” (übernatürlich),
25
Kant therefore is especially fond of Milton’s epic work. See Sanford Budick, Kant and Milton
(Cambridge, MA: Harvard University Press, 2010).
despite its definitely non-empirical character. This is because the super-
sensible is not in every sense non-natural. The reason why there is a sense in
which the term “nature” is very appropriate here for the supersensible as
moral, is that, for Kant the term “nature” has a most general sense that is not
limited to the physical realm and can signify the law-governed structure of a
concrete item of whatever kind (in contrast to “essence,” a term he generally
applies to abstract structures as such).26
In this broad sense, for Kant “nature” is not entirely opposed to grace, for
there is a moral nature that defines the distinct system of prescriptive laws
governing concrete rational beings in general, and that is formally similar to
the descriptive law-governed structure of physical beings as such, even
though this moral nature, and it alone, at its deepest level, is also determined
by essentially teleological rather than merely mechanical principles.27 Kant’s
notion of our lawful moral nature underlies his cosmopolitan “Idea” of the
special practical telos of the human species. According to this Idea of reason,
which is distinct from but closely connected to the purely metaphysical
Ideas of Kant’s postulates of pure practical reason, we are to believe that
there is an underlying theodicical pattern within human history, such that
legal and political developments will eventually lead to conditions that make
possible the fulfillment of our sublime moral vocation.28 Against a naïve
obsession with moralistic intentions, Kant argues that this pattern of human
history proceeds through a kind of cunning of reason that does not at first
require any moral intentions on humanity’s part, let alone any miraculous
interventions from above, and this is entirely consistent with all the laws
governing the physically natural, and even egoistic, developments of human
history.
The hermeneutical “guiding thread,” or Leitfaden, that reveals the unity
behind Kant’s vision goes back to what seems to be his very earliest
philosophical experience when, according to Jachmann’s account of Kant’s
“frequent” recollections in later life, his devout mother took her young
children out to the edge of town at night, so that they could see all the more
26
In general, Kant holds, “every nature has laws,” ML1 28:216.
27
Much of Kant’s difficult-to-follow exposition of what he calls the “antinomy” arising with teleology
rests on matters concerning the use or misuse of the distinction between the mechanical and the
“merely” mechanical.
28
This essay is Kant’s answer to the question that dominated so much discussion in the mid eighteenth
century after Johann Joachim Spalding’s work, Die Bestimmung des Menschen (1748). See
Reinhard Brandt, Die Bestimmung des Menschen bei Kant (Hamburg: Meiner, 2007); and
Karl Ameriks, “The Purposive Development of Human Capacities,” in Amélie Rorty and
James Schmidt (eds.), Kant’s “Idea for a Universal History with a Cosmopolitan Aim”: A Critical
Guide (Cambridge University Press, 2009), pp. 46–67.
150 KARL AMERIKS
clearly, from a point not very far from Copernicus’ home, the “power,
wisdom and goodness” of “the creator of all things” in the stars of the
northern European sky.29 Jachmann’s account seems genuine and especially
relevant since it links a reference to the mother’s awakening of an appreci-
ation of the “impressions of nature” with a mention of the “seed (Keim) of
goodness,” a phrase that plays a key role throughout Kant’s work on religion
and our moral nature. A very similar linking of the natural and the moral
orders can be found in the cosmology section of the 1782–83 Metaphysik
Mongrovius:
The order in nature excites wonder and respect [NB] in us at all times. But
whence does it come? From the fact that without nature [as ordered by law]
we would not have any objects of the understanding . . . We would then have
either useless understanding, or none at all. Thus, since this nobility of soul
which raises us to humanity, the worth of our understanding, rests on nature,
we thus regard this natural order as if it were something holy. (MetM 29:869)
It was only when Kant was almost forty years old, and after many years of
studying the “impressions of nature” in a scientific way, that he came back,
through the close study of Rousseau’s moral writings in 1763–64, to begin to
concentrate on the “seed” and destiny of our nature in a primarily moral
sense, with a specific law of its own. Even before that time, however, in the
first two decades of Kant’s career, it is significant that he regularly added
remarks about design, development, and divinity in his natural philosophy.
His early essays on metaphysics and the “natural history” of the universe are
composed on the presumption that there is an evident overarching teleo-
logical unity to the universe, one that redounds all the more appropriately to
the creator’s wisdom and goodness because it does not need to appeal, as
he feared Newtonianism or occasionalism does, to repeated miracle-like
interventions or acts of support by the deity.30
In the Beweisgrund of 1763, Kant still makes a twofold theoretical
teleological claim that: (1) “different natural effects are, in respect of their
beauty and usefulness, to be subsumed under the essential order of nature,
and by that means, as subsumed under God [that is, the powers of a divine
architect],” and (2) “many arrangements . . . will be subsumed under an
order of nature which is contingent in character and the product of artifice,
29
Reinhold Bernhard Jachmann, “Immanuel Kant geschildert in Briefen an einen Freund” [1804], in
Felix Gross (ed.), Immanuel Kant, Sein Leben in Darstellungen von Zeitgenossen (Borowski, Jachmann,
Wasianski) (Berlin: Deutsche Bibliothek, 1912).
30
It is in this spirit that Kant peppered his Universal Natural History (1755) with optimistic quotations
from Alexander Pope regarding God’s all-encompassing wisdom in organizing the world.
and in virtue of that subsumption they will also be subsumed under God.”31
In his critical work, Kant gives up taking this teleological claim to be a
demonstrable theoretical proposition, and he provides a new moral ground
for it, but it is striking that even in this earlier phase he is concerned, above
all, with conceiving a tight overarching order – of God, humanity, and
nature all at once – with no need for interruption anywhere by miraculous
events.32 Although at this point Kant still speaks of a system that can do
without “the assistance of frequent miracles” (EM 2:109), what is most
significant here is that he dares to go as far as explicitly challenging Newton,
whom he takes to have proposed that a “miraculous intervention” was
needed to prevent the universe from falling back into a “state of complete
stagnation” (EM 2:110n). It is typical of Kant’s firm theodicical attitude that
he goes on to argue that, whatever our local problems, we can use our
cosmological imagination to postulate that the universe can still have “great
fruitfulness elsewhere” (EM 2:110 n.).33 In this way we can still maintain a fully
teleological and yet seamlessly natural conception of the mundane sphere, as
an extraordinarily massive whole governed in a non-interventionist manner
by an all-encompassing purposeful design.
This section of the Beweisgrund also bears a close relation to Kant’s 1784
Idea for Universal History. Both discussions focus on the seeming indeter-
minacy that arises from actions that “harbour within themselves a possibil-
ity of deviating” from good (EM 2: 110), and both discussions call attention
to the statistical laws that apply even to human actions that may appear
significantly free, such as marriage choices (Em 2:111; cf. Idea 8:17). In the
Beweisgrund, to be sure, Kant still speaks sometimes without direct reser-
vations about “immediate divine intervention” and “revelation” at “specific
times and among specific nations” (EM 2:111). It is very significant, there-
fore, that in his otherwise parallel discussions in the 1780s and after, Kant
does not speak in this way, and he instead stresses that endorsing “rare”
miracles is just as questionable as referring to “frequent” ones. But although
31
EM 2:108–9, my emphases.
32
Hence this text has especially appealed to figures such as Friedrich Nietzsche, as well as Thomas
Bernhard in his play Immanuel Kant (1978). See Karl Ameriks, “Tragedy, Romanticism, and
Idealism,” in Bärbel Frischmann and Elizabeth Millán-Zaibert (eds.), Das Neue Licht der
Frühromantik: Innovation und Aktualität frühromantischer Philosophie (Paderborn: Ferdinand
Schöningh, 2008), pp. 28–38.
33
Although this is a pre-critical text, it is not to be dismissed in this context, for in PPV (1793) there is a
rare explicit reference to it, affirming its “belief in a deity which our practical reason must endorse”
(27:718). Cf. Fort (1793) which encourages “man . . . to assume therein, as object-in-itself, a morally
teleological connection, such that, by an ordering of nature beyond his comprehension, it tends to the
final purpose, as supersensible goal of his practical reason, namely, the highest good” (20:307).
152 KARL AMERIKS
this is an important change in explicit language, it can also be taken as a

natural development of strands already in Kant’s thought in the 1760s, for
even the Beweisgrund ends its discussion by saying: “I should find it amazing
if anything occurred or could occur in the course of nature . . . in need of a
miracle to improve it. And were such an event to occur [it] . . . would be
utterly incomprehensible to us” (EM 2:112). In addition, Kant already uses
roughly the same example here as in the dramatic culmination of his
discussion at the end of the second Part of the Religion − namely, the
amazing regenerative fruitfulness of plants and animals − to argue, against
the other scientific leader of the age, Buffon, that there is no need to refer to
“immediate divine action,” for “one must concede to the things of nature a
possibility, greater than that which is commonly conceded, of producing
their effects in accordance with universal laws” (EM 2:115).
Given this attitude, it is not surprising that, even prior to the critical turn
as such, Kant’s 1770 dissertation ends with a section that proposes, as a first
“rule of judging,” that “all things in the universe take place in accordance
with the order of nature,” and hence “comparative miracles, such as the
influence of spirits, are carefully excluded from the explanation of phenomena”
(Dm 2:418). In his metaphysics lectures, Kant explains that a “comparative
miracle” would be an event that seems “supernatural in relation to our
reason” but can still occur “according to certain laws unknown to us,”34 and
therefore to speak of “miracles” in this extended sense is still not to concede
that there are miracles “strictly speaking.” But the most revealing point that
Kant makes here (according to the student notes) is when he adds that a
miracle is “not something of which we do not cognize the cause, but rather
that of which we do not cognize the laws. Thus magnetic power is no
miracle, for we cognize its law (but not the cause)” (MD 28:667). I take this
to be an extremely significant qualification, because it suggests that the
distinction between “law” and “cause,” here in the sense of the ultimate
causing, may be the main reason that Kant believed he could regard our
own absolute freedom, and even all its theodicical involvement with the
moral assistance of God and nature, as not miraculous. This is because, even
if we cannot literally know the ultimate causings at work here either, we still
do grasp the law of their operation, which in this case is moral, and so in that
sense we are not going beyond nature in its most fundamental meaning,
which is simply to be lawful.
There is a second point that is crucial here. Kant conceives of our free
action as not only lawful but also internally governed in a number of
34
MD 28:667; cf. MetM 29:874, and ML1 28:219.
important but distinct senses. Insofar as our action is normatively guided by
the moral law, and insofar as Kant regards this law as internal to the
structure of our own practical reason, there is a sense in which a moral
agent is following something “within” its own self, that is, not its merely
individual and psychological self but rather its general nature as a being of
practical reason. This internal aspect is significant here because it implies for
Kant that ultimately our action is normatively, or in the sense of formal
causation, not to be thought of as guided by an external, that is, heteron-
omous and ad hoc principle – such as either mere feelings, or physical or
historical principles, or even God thought of as arbitrarily instructing us by
using miraculous power. Moreover, in a metaphysical causal sense, it is also
true that, insofar as we take ourselves to be an ultimate subject (as Kant,
against Spinoza, always thinks that we do and should35), then each of our
actions has an internal “efficient” source, but now in the very different sense
of resting precisely in our concrete individuality as such. Here too this
source, even if it is not omnipotent, is to be thought of as independent, in its
original direction, of any external − that is, ad hoc – force, such as either
mere feelings, or physical or historical forces, or even God thought of as
arbitrarily interrupting us by using miraculous power.
The most detailed indication of Kant’s view on God’s metaphysical
relation to us comes from some not clearly trustworthy notes to lectures
on philosophical theology, apparently from the 1780s.36 These notes discuss
problems in conceiving either a “natural concursus” of God with the world in
general, or a “moral concursus” with us as individual free agents. With
respect to the world, Kant takes each substance to rest entirely on God
for its existence, so in that respect God alone is sufficient, and there is no
need for the thought of a concursus (VpR 28:1105ff.). With respect to the
states of mundane things, however, Kant indicates that each previous
natural state is by itself sufficient for the effect, and so here too, although
for an opposite reason, there is no need for the thought of a concursus (VpR
28:1106, 1109). Similarly, with respect to our freedom, Kant notes that our
absolutely free choice is sufficient in its own realm, so here again it would
seem no literal concursus is to be introduced (VpR 28:1106, 1109). It is noted
35
See Karl Ameriks, “‘The Question is Whether a Purely Apparent Person is Possible’”, in
Eckart Förster and Yitzhak Melamud (eds.), Spinoza and German Idealism (Cambridge University
Press, 2012), pp. 44–58.
36
A problem with these notes in particular is that they appear to ascribe to the critical Kant an a-priori
argument for our substantiality and also – without explicit qualification – a traditional theoretical
argument for this being the best possible world (VpR 28:1097). In such places, the theological note-
taker may have failed to note that Kant was reviewing traditional positions and not presenting his own
full critical view.
154 KARL AMERIKS
that we cannot rule out that in some “not in the least conceivable way,” God
might concur with us here (VpR 28:1106), but the notes, like the Religion,
also indicate that eventually such special causings could be multiplied
arbitrarily, endlessly, and absurdly, and this would lead to the conclusion,
“what imperfection in [such] a world, totally irreconcilable with a wise
author!” (VpR 28:1110). I take this to mean that, although hypothetically we
might speak of a “miracle of the moral world, just as” – the notes go on to
say – “God’s acts of cooperation with occurrences in the sensible world are
[that is, are hypothetically to be called] God’s miracles in the physical
world” (VpR 28:1106–7), all this seems to imply that for Kant we should
not go as far as saying we know that such particular oddities are really
possible. This is not to deny, however, that with respect to the general
complex achievement of the highest good, there is still a sense in which Kant
can think of human beings and God in a kind of concursum moralem (VpR
28:1110), precisely because this highest good must involve a joint arrange-
ment. It requires free agents to fit properly together with an amenable
overall environment, independent of them, and hence whatever supreme
being is ultimately responsible for that environment must also play a role in
making the end possible.
In sum, Kant can, after all, consistently conceive his extremely elaborate
theodicical teleology as not literally miracle-involving because, insofar as it is
defined by the thought of beings – human, subhuman, and superhuman –
that are always ultimately acting in a way that fits together with a fully
purposeful set of internally determined laws, there is no assertion of special
acts or exceptions imposed on the order of nature in the broadest sense, that
is, including our nature, God’s nature, and the nature of the physical and
moral world on the whole.
4 Concluding critical caveat

There remains at least one basic philosophical problem here, one that may
be severe even from a kind of internal Kantian perspective. The problem is
that the main reason that Kant seems to prefer Vernunftglaube to the
traditional religion of miracles is that he takes Vernunftglaube alone to
appeal to considerations that can be expected to be agreed upon by all.
Any religion relying on miracles supposedly fails this kind of public con-
dition because the experience of a miracle directly, or the authentication of
one through historical and esoteric means, must rely on local and contin-
gent circumstances that cannot be presumed to be in the reach of all rational
agents as such.
The problem here is that Kant himself appears to presume items that, on
reflection, later readers may also understandably come to regard as in a sense
local, contingent, and esoteric. In particular, they may ask how it is that so
many common people and philosophers – even before and during, but
especially after, Hume’s time – have appeared to live a life of “healthy and
sound understanding” without seeing any need to assert, or even to be able
to understand, the absolute notions of freedom, morality, and eschatolog-
ical justice on which Kant relies. From a non-question-begging and general
common-sense standpoint, the specific Kantian appeal to absolutely free
and pure causings – even if as widespread as many forms of superstition and
unenlightened religion – can appear to be ultimately almost as ad hoc and
just as contingent and esoteric as the appeal to literal miracles that disturbs
Kant. Just as Kant holds that it would be unfair religiously to condemn the
scores of rational people who understandably think that they have no access
to miracles, a non-Kantian – but one who is sympathetic to Kant’s own
concern with principles that are really open to all persons – can argue that it
would be morally unfair to condemn people who understandably think that
they have no access to the demanding prerequisites of Vernunftglaube.
chapter 8
Kant’s Jesus
Manfred Kuehn
1
The claim that Jesus of Nazareth is absolutely central to the Christian faith
is not just uncontroversial, it may even be said to be so weak as to be
misleading. For Christians, Jesus is the sine qua non of all faith. He is the
Christ and thus God. As the Nicene Creed states:
We believe in one God, the Father Almighty, maker of all things seen and
unseen; and in one Lord Jesus Christ, the Son of God, begotten of the Father,
only-begotten, that is, from the essence of the Father, God of God, light of
light, very God of very God, begotten not made, of one essence with the
Father; by whom all things were made, both which are in heaven and which are
on earth, etc., and in the Holy Ghost. Those that say that there was a time
when He was not, and that He was not before He was begotten, and that He
was made of things that are not; or say that He is of a different hypostasis or
essence from the Father, or that the Son of God is created, nourished, and
capable of being changed, the Catholic Church anathematizes.
Orthodox Lutherans essentially endorse this precise creed. Thus the first
two articles of the Creed in Martin Luther’s Small Catechism read:
I believe in God, the Father Almighty, maker of heaven and earth.
And in Jesus Christ, His only Son, our Lord; who was conceived by the Holy
Ghost, born of the Virgin Mary; suffered under Pontius Pilate, was crucified,
dead, and buried; He descended into hell; the third day He rose again from the
dead; He ascended into heaven, and sitteth on the right hand of God the Father
Almighty; from thence He shall come to judge the quick and the dead.
Pietistic Lutherans take these articles of faith just as seriously. While they
often argue against mere doctrinal belief and emphasize the importance of
praxis in serving our neighbor, they place perhaps even more emphasis on
the need to accept Jesus as the Christ, and claim that even many nominal
Christians do not see “how deep they still are submerged in the old birth
and do not actively show any of the right signs of the rebirth [in Christ].
156
Kant’s Jesus 157
Thus Paul could still complain in many places”: ‘“They all seek what is
theirs, not that Christ is Jesus.’”1
To say it again, for traditional Lutherans, the belief that Jesus is the
Christ, that he died for us on the cross and saved us is a necessary condition
of the possibility of being a Christian. Calvinists or reformed theologians
would essentially agree to this view as well. No one who fails to believe that
Jesus Christ is the only way to salvation or renunciation with God has a
right to call himself Christian. Many Kant interpreters argue that this was
essentially also Kant’s view, and that his moral philosophy is not much more
than a secularized form of pietism.
It is true, of course, that Kant was born into a family of strict Lutheran
faith who lived by pietistic practices. He went to a high school that taught
and practiced pietistic principles, and he attended at the University of
Königsberg lectures, in accordance with the doctrines of Pietism. Almost
all of his colleagues were Lutherans, and even the Calvinists he knew would,
for the most part, have professed this faith.2 Accordingly, Kant would have
known this view of Jesus very well. It was all around him and it forms the
background of his mature views on religion. It is very clear that Kant
respected people who honestly held this view.
Yet it is also clear that Kant had some problems with the kind of pietistic
Lutheranism espoused by many around him. It seemed to him often more
an expression of hypocrisy than of true faith. But, more importantly, Kant
had serious problems with the traditional conception of Jesus as Christ.
Indeed, late in his life, he seems to have moved ever closer to the view of
those theologians, usually called “Neologists,” who determined much of
Christian teaching at Protestant universities in the German countries.
Neologism neither was then nor is now considered uncontroversial. It has
been characterized as the attempt to empty the concept of all historical
content and to fill it with purely rational content instead.3 Neologists
usually criticized Pietists and Pietists were as opposed to the Neologists as
they were to the Wolffians earlier on. Thus Karl Barth polemicized against
these theologians that they “partly by silence, partly by re-interpretation
eliminated [the doctrines of] the inspiration of the holy Bible, the trinity,
and especially the doctrines of the divinity of Christ, original sin, the
restitution alone through faith, and, of course, the virgin birth and ascent
1
Philipp Jacob Spener, Pia Desideria: Programm des Pietismus, ed. Erich Bayreuther (Wuppertal: Aussat
Verlag, 1964), p. 26.
2
Manfred Kuehn, Kant: A Biography (Cambridge University Press, 2001). See especially the
Introduction and Ch. 1.
3
Karl Aner, Die Theologie der Lessingzeit (Halle: Max Niemeyer, 1929), p. 4.
158 MANFRED KUEHN
to heaven.”4 Friedrich Karl von Moser (1723–98) tried to distinguish “true

enlightenment” from “false enlightenment” and claimed that “all enlight-
enment that is not grounded in and supported by religion . . . is not only the
way to destruction, immorality, and depravity, but also to the dissolution
and ruin of all civil society, and to a war of the human race within itself, that
begins with philosophy and ends with cannibalism.”5
The main themes that interested Neologists were the problems posed by
Lutheran orthodoxy and Pietism on the one hand, and by religious skep-
ticism and deism, on the other. Most of them were anti-Trinitarian.
Christological questions were pushed into the background. They were
trying to express the most essential insights of Christianity in a form
accessible to enlightened people. In fact, they were trying to advance the
project of Enlightenment by means of a reform of the Protestant church and
its teachings. Miracles were questionable. Ethics became much more central
than devotion. In this enterprise they made many more enemies than
friends, but, as long as Frederick the Great reigned, they had a great deal
of official support. With Frederick’s successor, this support not only van-
ished, but turned into hostility. In a sense, this is what the religious edict
that brought so much trouble to Kant was about. It is, however, too often
forgotten that he was by no means the only one who was negatively
impacted by it.
But there was and is a great deal of controversy as to who really belonged
among the Neologists. Johann Friedrich Wilhelm Jerusalem (1668–1726), was,
according to Aner, the most representative of the Neologists. Zoellner, Johann
Joachim Spalding, Friedrich Wilhelm Sack, and other Neologists combined
critical and historical approaches to Christianity and emphasized the moral
and practical dimensions of Christian teaching. Other names frequently
mentioned as major Neologists are Carl Friedrich Bahrdt (1741–92), Johann
August Eberhard (1739–1809), Christoph August Heumann (1681–1764),
Gotthold Ephraim Lessing (1729–81), Johann David Michaelis (1717–791),
Johann August Nösselt (1734–1807), Herman Andreas Pistorius (1730–98),
Johann Salomo Semler (1725–91), and Wilhelm Abraham Teller (1760–1813).
In addition to these, there are many more minor figures suspected of, or
praised as, belonging to this group. While Lessing is almost universally
considered to be a critic of Neologism, some of the more orthodox voices
4
Karl Barth, Die protestantische Theologie im 19. Jahrhundert. Vol. I: Die Vorgeschichte (Hamburg:
Siebenstern Verlag, 1960), p. 138.
5
James Schmidt, What is Enlightenment? Eighteenth-Century Answers and Twentieth-Century Questions.
(Berkeley: University of California Press, 1966), p. 8.
Kant’s Jesus 159
would count him among them. Reimarus, on the other hand, who might
have been counted among the Neologists during his early years, later came
to be identified as an outright deist, though some continued to blame the
Neologists for his excesses. Semler was, according to Aner, not a Neologist −
while Karl Barth characterized him as the most important among them.6
It does not matter for the purposes of this chapter who was or was not a
“Neologist” in the full sense, for I do not want to argue that Kant himself
should be counted as belonging to this group. Nor do I offer a full
discussion of the views held by the Neologists. I only want to demonstrate
that Kant’s outlook is rather close to that of some of these people. In order
to do this, I will discuss Kant’s views on Jesus in relation to two thinkers,
considered to be at the opposite ends of the Neologistic spectrum, namely
Reimarus and Semler. In particular, I will compare them on four essential
doctrines − namely, the question of the historical context in which Kant
must be seen, and the problems of miracles, mystery, and the means and
effects of grace. As might be obvious to the reader right away, the last three
problems form the content of the four “General Remarks” that Kant
appended to the four books of the Religion within the Boundaries of Mere
Reason. But this is not the main reason for using these three (or four)
problems. More importantly, they summarize the main concerns of theo-
logians like Reimarus and Semler, and the General Remarks may therefore
be said to reflect the contemporary theological discussion.
Accordingly, I will represent first the view of Reimarus; second, that of
Semler; then I will offer a discussion of the relevant similarities and differ-
ences of Kant’s view, and finally discuss some reflections on the significance
of the results. The conclusions may come as something of a surprise to those
who tend to view Kant as the philosopher of Protestantism, because they
show that Kant’s views on Jesus, at least, are closer to those of Reimarus
than they are to those of Semler.
2
Reimarus is nowadays considered a deist, though during his lifetime his
deistic tendencies were completely unknown. It was Lessing’s publication of
the Wolffenbüttel Fragments after 1774 (Apologie oder Schutzschrift der
vernünftigen Verehrer Gottes. Fragmente eines Ungenannten) that gave rise
to this reputation. But Reimarus had a great reputation before these
publications. His 1754 book on natural theology (Die vornehmsten
6
Aner, Die Thologie der Lessingzeit, pp. 98ff.; Barth, Protestantische Theologie, p. 142.
160 MANFRED KUEHN
Wahrheiten der natürlichen Religion in zehn Abhandlungen auf eine begrei-

fliche Art erkläret und gerettet), and his logic of 1756 (Vernunftlehre als
Anweisung zum richtigen Gebrauche der Vernunft), as well as his 1760 book
on the instinct of animals (Allgemeine Betrachtungen über die Triebe der
Tiere) were important contributions to the philosophical discussion of
religious issues during the post-Wolffian period. Kant himself in his The
Only Possible Argument in support of a Demonstration of the Existence of God
of 1763 was willing to “concede superiority in respect of usefulness to an
exposition of the important knowledge of God and his qualities, such as
Reimarus offers in his book on natural religion.”7
Natural religion for Reimarus consists in the knowledge of God and his
goals for the world. It teaches us who we are and that we must strive for the
highest perfection possible for us. Since this highest perfection cannot be
achieved in this world, we must assume that it will be achieved in another
world and that our lives do not end with death. This is what separates us
from all the other animals. It is therefore, as he says in the last sentence of his
Natural Religion:
religion alone which, through the confirmation of immortality makes our
entire life bearable in all circumstances, satisfactory and happy (glückselig). It
heightens the temporal pleasure through a certain hope for a still better state;
it sweetens present suffering by the greater future joy; it awaits in death, a
new hour of birth into a perfect life . . . and is sufficient for the happiness
commensurate with our nature.8
On the other hand, there cannot be any true religion without the belief in
God, the creator, and the immortality of the soul. “These two great truths
are the main support of our satisfaction and ultimately [constitute] the
influence of religion upon our happiness.”9 Reimarus’ natural theology
essentially constitutes the clear explication and defense of these two truths.
Most of his contemporaries would have suspected that natural religion
does not exhaust all of religion, and that revelation adds another important
dimension. That this hope was mistaken could have been made clear by
Reimarus’ definition of miracles. They are, as he puts it, “actions of God
which are contrary to the effects and rules of the active forces of nature,” and
they can therefore not exist, as they would “contradict . . . the orderly
7
EM 2:161.
8
Hermann Samuel Reimarus, Die vornehmsten Wahrheiten der natürlichen Religion in zehn
Abhandlungen auf eine begreifliche Art erkläret und gerettet von . . . (Professor in Hamburg)
(Hamburg: Johann Carl Bohn, 1754), Vol. II, p. 766.
9
Ibid., p. 574.
Kant’s Jesus 161
preservation of nature.”10 Miracles in any interesting sense are impossible.
Therefore, one might have concluded already in 1754 that revelation in any
significant sense is also impossible.
The fragments published by Lessing after 1774 resolved any doubt.
Natural religion was for him the only religion that we need and that we
can have. The miracles of the New Testament, just like those of the Hebrew
Bible, need to be explained away. Henry Allison summarizes Reimarus
therefore as follows: “Jesus becomes regarded as a well meaning, but
deluded fanatic, the apostles clever and self-seeking deceivers, and the
Christian religion a colossal fraud.”11 Albert Schweitzer, who may have
known better, was more positive. He called Reimarus’ account not just
one of the greatest events in the history of the critical spirit, but also a master
work in the world history of literature. The language is in general laconic and
dry, epigrammatically sharpened, like that of a man who does not so much
write as establish facts. But there are times when he rises to pathetic heights.
It is, as if the fire of a volcano paints ghostly pictures onto dark clouds.12
Schweitzer’s account focuses especially on the last installment of the frag-
ments, "The Aim (Zweck) of Jesus and his Disciples.” In it, Reimarus argues
specifically that the Biblical miracles were fraudulent and that there is no
real revelation to be found in the Bible. He characterizes Jesus’ disciples as
deceivers and Jesus himself is attacked by him as using force and cunning to
propagate his message. His baptism is a mere spectacle, staged by his cousin
John, and his supposed messianic mission is for Reimarus supported by a
number of tricks that are topped only by the resurrection stories.
As we have seen already, Reimarus thought that the doctrine of the
immortality of the soul and its happiness (Seligkeit) is absolutely essential for
any religion. Without this doctrine, we cannot speak of a religion at all. But the
Jewish religion lacked this doctrine, according to him.13 The Pharisees and
Sadducees had tried to introduce this doctrine before Jesus, but they had
combined it with an emphasis on merely external obedience to religious rules.
Jesus offered a severe critique of the emphasis on externalities that also
characterized these sects, but his goals were ultimately just political, not
religious. He wanted to change the society of the Israelites, not found a new
religion. It was the disciples of Jesus who invented a new religion in which the
10
Ibid., p. 588.
11
Henry Allison, Lessing and the Enlightenment (Ann Arbor: University of Michigan Press, 1966), p. 48.
12
Albert Schweitzer, Geschichte der Leben-Jesu-Forschung. Vol. I (Munich and Hamburg: Siebenstern
Taschenbuch, 1966), p. 58.
13
Hermann Samuel Reimarus, Von dem Zwecke Jesu und seiner Jünger. Noch ein Fragment des
Wolffenbüttelschen Ungenannten, ed. Gotthold Ephraim Lessing, 2nd edn. (Braunschweig, 1784), p. 4.
162 MANFRED KUEHN
resurrection, baptism, the Eucharist, and other means of grace played central
roles. The apostles or the authors of the gospels invented miracles out of thin
air to legitimate the claim that Jesus was the savior not just of the Jews but of all
Christians.
Reimarus’ main thesis is that we must carefully separate what Jesus himself
said and what his disciples later made him say. The message of Jesus is simply:
“Convert and have faith in the gospels!” or − which means the same thing −
“Convert, for God’s heavenly realm has come.” But, and this is of the highest
importance, “God’s heavenly realm” must be understood in accordance with
the Jewish way of talking (nach jüdischer Redensart). Jesus must be understood
as a Jewish thinker and leader. In the context of the Roman occupation of the
Jewish homeland, the call for faith in “God’s heavenly realm” could therefore
only be understood politically as a call to resist the Roman hegemony. There
was no claim that he himself was God’s very own son in the sense that he was
divine. Jesus did not want to replace the Jewish religion with something new,
but he wanted to save Judaism. In fact, Reimarus claims all of the words and
deeds of Jesus must be understood in this context.
Reimarus himself argues vehemently not just against the possibility of
miracles, but also against so-called “secrets” or “mysteries” (Geheimnisse),
by which he means anything that would go beyond reason and cannot be
proved by rational argument.14
As he himself says in section 7:
I cannot avoid uncovering a common error of Christians, who because they
confuse the doctrine of the apostles with the teaching of Jesus imagine that
Jesus had the aim of establishing some new and unknown articles of faith, of
revealing secrets and thus creating a new religious system, but that he also
wanted to abolish Judaism in accordance with its special customs of sacri-
fices, circumcision, purification, sabbaths and other Levitic rites. I know very
well that the apostles and especially Paul tried to do so and that the
subsequent teachers created ever more secrets and articles of faith and
removed more and more Jewish ceremonies until the laws of Moses were
finally completely discarded and a new religion was introduced.
But I cannot find any of this in the conversations and speeches of Jesus. He
[advocated and] lived only moral duties, the true love of God and our
neighbor. In this he placed the entire content of the law and the prophets
and placed all hope for a heavenly kingdom and blessedness.15
14
Ibid., p. 26.
15
“Ich kann nicht umhin, einen gemeinen Irrtum der Christen zu entdecken, welche aus der
Vermischung der Lehre der Apostel mit der Lehre Jesu sich einbilden, dass Jesu Absicht in seinem
Kant’s Jesus 163
The main results of Reimarus’ theory are, accordingly:
1. The teaching of Jesus must be seen in the context of Judaism alone.
2. Miracles are impossible and are also insufficient to prove the truth of
Christianity.
3. The mysteries of Christianity were invented by the apostles and form no
part of Christian teaching.
4. Insofar as the means of salvation in Christianity are based on miracles
and mysteries, they are spurious and form no part of the true natural
religion which is identical with Christianity.
The presuppositions of this view defended by Reimarus are a thorough-
going naturalistic account of the world of nature and the conviction that
religion is based on just two truths: the existence of God and the immortal-
ity of the soul. It is a rational or rationalistic form of religion in which moral
perfection is central and the historical context in which this view arose is
systematically eliminated.
3
Semler was one of the most important theologians in the period between
1740 and 1790, or between Wolff and Kant. Starting out from Wolffian
ideas as mediated by Siegmund Jakob Baumgarten − the famous brother of
Alexander Gottlieb Baumgarten, whose textbooks Kant used in many of his
lectures − Semler developed a rigorous method of historical-critical scholar-
ship with regard to the Bible. In his Abhandlung von freier Untersuchung des
Canon or Treatise of the Free Investigation of the Canon (4 vols., 1771–75), he
argued that the gospels must be viewed and discussed as historical docu-
ments just as much as they must be understood as “the word of God.”
Helmut Thielicke has called Semler “the Bultmann of his day.”16 While
Semler could not write very well and is therefore much harder to read than
Lehramte gewesen, gewisse zum Teil neue und unbekannte Glaubensartikel und Geheimnisse zu
offenbaren, und also ein neues Lehrgebäude der Religion aufzurichten, dagegen aber die Jüdische
Religion nach ihren besonderen Gebräuchen, also Opfern, Beschneidung, Reinigung, Sabbaten und
andere levitischen Zeremonien, abzuschaffen. Ich weiß wohl, dass die Apostel, und Insonderheit
Paulus, hieran gearbeitet, und dass die nachfolgenden Lehrer teils immer mehr Geheimnisse und
Glaubensartikel geschmiedet, teils auch immer mehr von den jüdischen Zeremonien zurückgezogen,
bis endlich Moses Gesetze gar abgeschafft und eine ganz andere Religion eingeführt worden ist. Allein
in allen Lehren, Reden und Gesprächen Jesu, kann ich von beiden nicht die geringste Spur finden. Er
trieb nichts als lauter sittliche Pflichten, wahre Liebe Gottes und des Nächsten; darin setzet er den
ganzen Inhalt des Gesetzes und der Propheten und darauf heisset er die Hoffnung zu seinem
Himmelreich und zur Seligkeit bauen”: Von dem Zwecke Jesu und seiner Jünger, p. 16, para. 6.
16
Helmut Thielicke, Modern Faith and Thought, trans. Geoffrey W. Bromiley (Grand Rapids, MI:
William Eerdmans Publishing Company, 1990), p. 141.
164 MANFRED KUEHN
Reimarus (or Bultmann, for that matter), he was still most influential on the
further history of biblical criticism and hermeneutics.
Semler wrote an explicit critique of the fragments by Reimarus, pub-
lished by Lessing. In it, he criticized Reimarus severely because of his
deficiencies in scholarship.17 He throws doubt both on Reimarus’ view
that immortality is an essential characteristic of any religion and the idea
that the Hebrews had no concept of it. He rejects the idea that Christianity
is based on a confusion of the doctrines of Jesus and those of his apostles,
and argues that Jesus aimed at ending the rule of the Mosaic law.18 He
criticizes Reimarus’ objection to the baptism, communion, and many
others things. But none of these attempts at refuting the details of
Reimarus’ view are as important as his refutation of the general background
of his position. In this context, he has several things to say about Reimarus’
“naturalism,” i.e. his denial of a supranatural dimension to Christianity.
He defended the view that the New Testament had to be understood first
and foremost historically. This did not mean for him that it was not also at
the same time God’s revelation. It was for him essentially both a historical
process and the revelation. His overriding goal was, in fact, to differentiate
between these two different and apparently contradictory components and
thus to determine what in the New Testament was revelation and what was
merely historical. In determining the difference between what is merely
historical and what is God’s word, he made use of the concept of a rational
reader, arguing that God’s revelation can never contradict the true criteria of
human reason.
This may sound very much like Reimarus, but it is different, for Semler is
not claiming that the New Testament should be subjected to the laws of
reason. There are some things that are above reason, that cannot be
perceived, and can only be known by revelation. It is only where reason
and revelation openly contradict each other that we must follow reason.
Physics and psychology are not dependent on the Bible, but, since the
essential message of the New Testament is, as God’s word, completely
outside time and space, it cannot be influenced by anything that our
understanding establishes. This can concern only what is external to the
Kerygma, or the essential doctrine of Christianity. Still, we cannot have the
essential message of the New Testament in isolation from history because it
is always embedded in historical claims. Thielicke has asserted, therefore,
17
Johann Salamo Semler, D. Joh. Salomo Semlers Beantwortung der Fragmente eines Ungenanten
insbesondere vom Zwecke Jesu und seiner Jünger (Halle: Verlag des Erziehungsinstituts, 1780).
18
Ibid., p. 98.
Kant’s Jesus 165
that the real revelation of the Kerygma is in some way reminiscent of Kant’s
“transcendental X”; it is like “a thing in itself.”19
This proposition goes too far, however, as Semler does make definitive
claims about what is essential to Christianity. It is not the story of the life of
Jesus per se. Rather, it is based on the teaching of Jesus, “who rose from the
dead” after being nailed to the cross − that is, he argues that Christianity is
based on both the teachings and the resurrection, and not on resurrection
alone, as he takes Reimarus to claim.20 But the “main point of Christianity”
is “a spiritual or moral salvation in contradistinction to a salvation of the
body.”21 Resurrection is just one of the most important articles of faith. It
belongs in the context of the doctrines taught by Jesus, but does not form an
essential part of it. And thus the Kerygma is about this new creation, the
moral salvation of moral beings. Semler calls Jesus the teacher of this
doctrine and claims that both he and Jesus live in God: that is, their
common father.
He confesses that he believes himself to have a much more secure ground
for his own belief that Jesus rose from the dead, and would believe, “even if
he had no knowledge of external circumstances, like time and space.”22 The
reason to believe in the resurrection of Jesus is for him internal. Thus he has
Jesus speak to him in the book as follows: “As little as you know the origin
and directions of the wind in the visible world, so little can you explain the
effects of God which are in the invisible realm of the truth and growth of all
spiritual perfection just as certain and indispensable for the internal perfec-
tion of all rational creatures.”23 Indeed, “the spirit of God produces a new
creation among human beings.”24 Furthermore, the
resurrection of Jesus is not a mere physical fact or event (physikalische
Begebenheit); it is a supra-natural event, a possibility that even the
Pharisees and others admit as in itself possible. But from this fact, it does
not follow that the resurrected Jesus can be seen by all human beings, like a
tree, a mountain, or a bird. Rather, it depends on his own will whom he will
give permission to see him; and he is visible only to those he allows to see
him.25
19
Thielicke, Modern Faith and Thought, p. 154. 20 Semler, Beantwortung, p. 263.
21
Ibid., p. 256; see also p. 257. 22 Ibid., p. 263.
23
Ibid., p. 266: “So wenig Du in der sichtbaren Welt den Ursprung und den Gang des Windes weißt,
eben so wenig kannst du weiter über diese Wirkungen Gottes erklären, die in dem unsichtbaren
Reich der Wahrheit und des Wachsthums in aller geistigen Vollkommenheit, ebenso gewis und allen
vernünftigen Geschöpfen so unentbehrlich sind, zu ihrer innerlichen Vollkommenheit.”
24
Ibid. 25 Ibid., p. 274.
166 MANFRED KUEHN
Accordingly, the resurrection was no natural event which was “subject to

the laws of motion and the senses, but a free effect of Jesus which he
determined and directed in accordance with his final goal.”26
The deists, Semler points out, “presuppose entirely different principles
and that is how they can deny this event [i.e., the resurrection]. They can do
this, but in doing so they do not refute Christianity,” or so Semler
believes.27 For him, “the resurrection of Jesus is connected with the life
and final goal of Jesus. Anyone who has [really] experienced his doctrine,
will also admit that God woke him from the dead.”28 For Semler, as for
Reimarus, religion has therefore ultimately to do with morality and perfect-
ibility. It is “the new, independent and certain knowledge (Erkenntnis) that
God is to human beings the forgiving, holy and infinite spirit who is
[concerned] only with their actions and omissions insofar as they know
what is good” that characterizes such a “perfect worship of God.”29 Moral
actions mean serving God. But it is important to understand that, in
contrast to the standpoint of Reimarus, for Semler “moral” ideas must
not be understood as being entirely different from dogmatic ideas.30 The
moral for him remains dependent on the religious. But the moral is for him
also always internal. External or public religion should not be made absolute
and rule over internal religion, which is always relative and is the true
religion. It is, according to Semler, also rational.31 Differentiating on this
basis, like Kant, between public (or external) and private (or internal)
religion, he demanded tolerance for different theological systems.32
To conclude the discussion of Semler, let us compare the main results of
Semler’s view with those of Reimarus:
1. Whereas Reimarus claimed that Jesus’ teaching must be seen in the
context of Judaism alone, Semler thinks this is false. Jesus propounded
a new moral theory that was addressed to all of humanity. It had, per se,
little to do with Judaism. Christianity is a radically new and revolutionary
doctrine that preaches rational perfection. Like Reimarus, he considered
26
Ibid. 27 Ibid., p. 279. 28 Ibid.
29
Semler, Letztes Glaubensbekenntnis über natürliche und christliche Religion (Königsberg: Nicolovius,
1792), p. 24.
30
Josef Bohatec, Die Religionsphilosophie Kants in der “Religion innerhalb der Grenzen der blossen
Vernunft” (Hildesheim: Georg Olms, 1966), p. 423.
31
Semler, Letztes Glaubensbekenntnis, pp. 61f.
32
Bohatec, Die Religionsphilosophie Kants, p. 460, claims that Kant is dependent on Semler in his
distinction between private and public religion, and in particular on the claim that Jewish religion has
no “unity in accordance with concepts.” For Semler, Christianity is “more worthy in its principle of
intention and its internal nature of the mind (mit seinem Gesinnungsprinzip und seiner inneren
Gemütsverfassung).”
Kant’s Jesus 167
the Old Testament as no more than ancient Jewish prejudice, which was
ultimately very different from the message of the New Testament. Thus
he claimed: “The Jews do not in any way transcend the external history of
their predecessors. For this reason, the Jewish and the Christian religion
cannot ever be unified.” Nevertheless, the Christian religion presupposes
“the previous Jewish religion as its imperfect predecessor and really
originated from new and greater concepts of God and their more liberal
and internal application.”33 What is Jewish can therefore concern only
what is external to the Kerygma. His view of the relationship between
Judaism and Christianity is thus radically different from that of
Reimarus.
2. Reimarus argued that miracles are absolutely impossible as well as
insufficient to prove the truth of Christianity. Semler admits that they
are by themselves insufficient to establish the Christian religion, but he
argues that they are neither impossible nor useless. He criticizes
Reimarus for giving an insufficient definition of miracles: “Above all,
the anonymous author should have made clear first of all what he means
by the term ‘miracle.’”34 He admits that he did not accept Christianity
on the strength of so-called “miracles,” but they are means of “prepara-
tion” for Christianity. In any case, they were meant for the contempo-
raries of Jesus not for believers of the eighteenth century. Miracles do
not, however, belong to the central tenets of Christianity.
3. For Reimarus, the mysteries of Christianity were invented by the apostles
and form no part of Christian teaching. This is clearly not Semler’s view.
In fact, his claims about why he believes in the resurrection show that
there are for him mysteries of faith. God has effects in this world. Jesus
can be experienced by those who truly experience his teachings. While
external miracles leave him relatively cold, “internal miracles,” like “our
own faith, fides divina, are based on God’s present assistance which
brings forth and fixes truths through their content in our soul. The
deist laughs about us – are we concerned and uncertain for this reason?”35
The answer is clearly “No!”
4. Insofar as Semler believes in a “fides divina,” he also rejected Reimarus’
claim that mysteries are spurious and form no part of the true natural
religion which is identical with Christianity. He holds on to the belief that
33
Semler, Letztes Glaubensbekenntnis, pp. 126, 41f., 116. 34 Semler, Beantwortung, p. 247.
35
Ibid., p. 380: “unser eigener Glaube, fides divina, ist auf Gottes jetziger Mitwirkung gegründet, welche
die Wahrheiten durch ihren Inhalt in unserer Seele hervorbringen und festsetzen. Der Deist lacht
über uns, sind wir deshalb unruhig und ungewiß?”
168 MANFRED KUEHN
there are mysteries, and he rejects Reimarus’ naturalism for precisely this
reason. There are, in other words faint echoes of the kind of Christology in
Semler that we noted to be central for traditional Christianity in general
and Pietism in particular. This, or any kind of, Christology is not just
completely absent from Reimarus’ work, but the main target of his
criticism. While Semler also assimilates the teaching of Christ and the
rationalist doctrine of moral perfection, he does want to hold on to a thin
Christology, in which God has real effects on our conduct. He may reject
the sacraments and other religious exercises as belonging to external
religion and not to inner or private religion, but he does not think that
they contradict each other.
4
Kant’s Religion must be seen as situated between the respective positions of
Reimarus and Semler, or, perhaps better, as Kant’s attempt to find a middle
way between these two positions. There can be no doubt that he knew the
two positions intimately. In fact, he referred to Reimarus directly in his
work, referring to him as the “author of the Wolffenbüttel Fragments.”36
There are also many oblique references to Semler and some of his terminol-
ogy clearly comes from Semler.37 Bohatec claims that Semler’s influence on
Kant was “very strong,” and he in general argues that Kant is closer to Semler
than to Reimarus.38
The question is therefore precisely how Kant’s view relates to the
thoughts of his predecessors. Let me follow the schema used in the preced-
ing sections:
1. Kant fully agrees with Semler’s rejection of Reimarus’ claim that Jesus
really had no genuinely moral purpose and had only political goals or that
he wanted to overthrow the rules of the priests and perhaps establish
36
Since it was generally known at the time that the Wolffenbüttel Fragmentist was identical with
Reimarus, he understood that he was referring to his favorite logician and religious philosopher. He
not only owned Reimarus’ Logic and referred to it very positively, he also willingly conceded in The
Only Possible Argument its “superiority in respect of usefulness to an exposition of the important
knowledge of God and his qualities, such as Reimarus offers in his book on natural religion” (EM
2:161). He goes on, however, to insist on having paid “greater attention . . . to logical rigor.” Gottfried
Hasse, Kant’s Königsberg contemporary, wrote that “among philosophers, for him [Kant] Reimarus
stays above all”: J. G. Hasse, Der alte Kant, Hasses Schrift: Letzte Äusserungen Kants und persönliche
Notizen aus dem Opus Postumum (Berlin/Leipzig: A. Buchenau und G. Lehmann, 1925), p. 30.
37
See Bohatec, Die Religionsphilosophie Kants, pp. 423, 427, 433–35, 472–77, 499–501.
38
Ibid., p. 27n (original emphasis).
Kant’s Jesus 169
himself as a ruler (R 6:161).39 Very much like Reimarus and Semler who
do not seem to differ in this from most contemporary theologians, he
claims: “The Jewish faith, as originally established, was only a collection
of merely statutory laws supporting a political state; for whatever moral
additions were appended to it, whether originally or only later, do not in
any way belong to Judaism at all” (R 6:125). Christianity, for him, is
accordingly a complete abandonment of Judaism in which it originated,
grounded on an entirely new principle. It “effected a total revolution in
doctrines of faith” (R 6:127). As regards the first point of comparison,
there seems to be little difference between Kant and Semler.
2. Turning to the problem of miracles, we can see, however, that Kant is
ultimately closer to Reimarus, though some of the things he says may
sound more like Semler. Thus he claims that belief in miracles contra-
dicts moral religion which consists “in the heart’s disposition to observe
all human duties as divine commands” (R 6:84). Belief in miracles is not
just dispensable (as Semler would have agreed), but betrays “a culpable
degree of moral unbelief” because it shows that we do not put enough
trust in the precepts of duty (R 6:84). This does not mean that he does
not grant miracles a historical function and allows for “veneration” of
them as “external” or “historical” cover (R 6:85). Indeed, he seems to be
talking about Semler when he admits that “there are rational human
beings who, though not disposed to renounce belief in them, never allow
this belief to intervene in practical matters; and this is as much as to say
that, in theory, they do indeed believe that there are miracles, but avow
none in their practical affairs” (R 6:85).
He also seems to want to answer Semler’s question about the defi-
nition of miracles on Reimarus’ behalf. Miracles are for him “events in
the world, the causes and effects of which are absolutely unknown to us
and so must remain” (R 6:86). Differentiating between theistic and
demonic miracles, he completely dismisses demonic miracles and claims
that we possess only a negative criterion with regard to theistic miracles −
that is, we can most definitely say that something cannot be a miracle, if
it is in conflict with morality. Thus, if a father were presumably ordered
by God to kill his innocent son and then is miraculously hindered from
doing so, we may discount this miracle. In any case, “in practical affairs
we cannot possibly count on miracles, or in any way take them into
consideration in our employment of our reason (which is necessary in all
39
See also Ibid., pp. 460–68.
170 MANFRED KUEHN
circumstances of life)” (R 6:87). Later in the Religion he calls faith in

miracles “delusory faith” (R 6:194).40 This is also Reimarus’ view.
3. If anything, Kant is still more negative on the problem with mysteries.
He claims: “investigation into all forms of faith that relate to religion
invariably runs across mystery behind their inner nature, i.e. something
holy, which can indeed be cognized by every individual, yet cannot be
professed publicly, i.e. cannot be communicated universally” (R 6:237).
Semler’s “fides divina” come to mind. While we cannot determine a
priori “whether there are such mysteries or not,” Kant entirely rules them
out by means of an analysis of the “inner, subjective, part of our moral
disposition” (R 6:138). Our moral faith “really contains no mystery, since
it expresses solely God’s moral bearing toward the human race. It is also
by nature available to all human reason and is therefore to be met with in
the religion of most civilized people” (R 6:140). Kant does seem to find it
there in the doctrine bound up with our duty to promote the highest
good or with the doctrine discussed in the second Critique as the
“Postulates of Pure Practical Reason.” He claims “that here opens up
before [us] the abyss of a mystery regarding what God may do, whether
anything at all is to be attributed to him and what this something might
be” (R 6:139). But the idea of a ruler and our faith in it “really contains no
mystery, since it expresses solely God’s moral bearing toward the human
race” (R 6:140). While it may be called revelation insofar as the Christian
faith made it first known under this name, it is really expressing pure
moral religion and not an anthropomorphic servile faith. Mysteries
ultimately do not exist and the supposed revelations about such mys-
teries cannot exist either, as “God has revealed nothing to us [about
them], nor can he reveal anything, for we could not understand it”
(R 6:144). Kant goes on to reject the particular mysteries of (i) “the call
of human beings to be citizens of an ethical state,” (ii) the mystery of
satisfaction, and (iii) the mystery of election, and makes clear that for
him religion consists in our call to live a good life through the moral law,
i.e. by having respect for this law that lies within us (R 6:144). Here, the
language is very different from Reimarus and, in fact, very Kantian.
40
Thus, miracles amount to the belief that we can know something through experience which is not in
accordance with the objective laws of experience. But miracles constitute just one form of the three
kinds of “delusory faith” or of “overstepping the boundaries of our reason with respect to the
supernatural.” The other two are: (i) that we must include in our concepts of reason something of
which we can have no rational concepts (i.e. mysteries); and (ii) that we can, through “purely natural
means,” bring about God’s assistance in moral matters (“faith in means of grace”). All three forms of
faith are mere delusions for Kant, just as they were for Reimarus.
Kant’s Jesus 171
Kant’s moral religion is most definitely not Reimarus’ natural religion.
But, and this is the most important point, religious mystery has just as
little place in Kant’s moral world as in Reimarus’ mechanistic universe.
4. Kant, just like Reimarus, rejects: (i) prayer; (ii) church-going; and
(iii) baptism or initiation into the church-community; as well as
(iv) communion. They belong to “priestcraft” or the dominion of the
clergy “over minds by pretending to have exclusive possession of the
means of grace” (R 6:200).
We must make ourselves “into whatever [we are] or should become in
a moral sense, good or evil” (R 6:44). For this reason it is neither
necessary nor essential that we know what God does or has done for
our salvation. What is essential is what we have to do to ourselves in
order to become worthy of God’s assistance (R 6:52).
This means that, for Kant as for Reimarus, Jesus has no role to play in our
atonement. We may need saving, but we cannot be saved by Jesus or even
God himself. We need to save ourselves. But even the idea that we must
“save ourselves” is misleading, for in the end this meant for Kant − as for
Reimarus, Semler, and all the other so-called “Neologists” − no more than
to “perfect ourselves morally.”
Accordingly, the name “Jesus” does not appear many times in Kant’s
published writings. The name “Christ” or “Christus,” if we discount the
calendar expression “nach Christus,” appears even less often. Insofar as
Kant spoke of Jesus, he preferred to talk about him as “the teacher,” the
“son of God,” the “divinely inspired human being” that we represent as an
“original ideal” for ourselves (R 6:61).41 This idea of the son of God is
closely connected for Kant with the “personified idea of the good,” the
“ideal of moral perfection,” the “ideal of a humanity pleasing to God,”
and the idea of a human being that alone pleases God (R 6:63f.). Jesus
represents this idea in human form, just as the Stoic wise man represents
the idea of wisdom. They are both ideals because they represent their idea
in human form.
But the idea of moral perfection does not really need any individual
instantiation. It “has complete reality in itself,” insofar as it is practical. It
“resides in our morally-legislative reason” (R 6:62). We need no examples
from experience to make this idea into an ideal. On the other hand, Kant
argues that the experience of an individual that instantiates the idea must be
possible, if only because every one of us should act in the required way. And
41
I reject the translations of the last two expressions in the Cambridge edition. “God-like” for “göttlich
gesinnt” is clearly a mistranslation, and “prototype” seems too mechanical for Kant’s purposes.
172 MANFRED KUEHN
Jesus, the “divinely inspired teacher,” is “in fact totally human” (R 6:65). He
came as “an actual human being as example for all others” (R 6:82). Yet
his coming only fractured the power of evil, whose “kingdom still endures”
(R 6:82). We must still wait for a new epoch in which evil is finally defeated.
The only way this can happen is if we “adopt,” in our innermost being,
“genuine moral principles” as our disposition (R 6:84).
Jesus actually appeared among the Jews who were prepared for his
message by “the Greek sages’ moral doctrines on freedom,” and who felt
at the same time “the full measure of all the evils of a hierarchical
constitution” (R 6:79). They were “ripe for a revolution” (R 6:80). Jesus
was “a person whose wisdom [was] even purer than that of the previous
philosophers,” and though he seemed “descended from the heavens,” he
announced himself as a true human being. He claimed that his doctrines
and his example were in fact truly human. At the same time he also
claimed to be “an envoy of heavenly origin” who had no part in original
sin. He promulgated the rational idea of “the possibility of a person free
from innate propensity to evil” which for Kant is consistent with an
undeniable “moral instinct.” In the tradition, this view became entangled
with the idea of virgin birth which is comprehensible as well, but also
difficult to explain. Kant believes that the many arguments for and against
virgin birth that have been offered in theology are unnecessary just
because it is sufficient for our practical purposes “to hold the idea
itself before us as a model, as symbol of humankind raising itself
above temptation to evil (and withstanding it victoriously)” (R 6:80n).
Put differently, Jesus is for Kant both a philosophical teacher who
taught by example and doctrine, and a “model” or “symbol of human-
kind”: an idea.
Clearly, the ideal Jesus is much more important to Kant than the
historical Jesus, and this colors his picture of the life of Jesus. Still, he
also relates the bare outlines of his ideas about the “historical” Jesus. Since
Jesus was a human being of “original innocence,” having no part of the
original sin of the rest of humanity, he posed a special threat to the realm
of evil or the rule of “the prince of this world” (R 6:80). Accordingly, he
was first subjected to temptations. When this did not work, the lord of the
realm of evil took away from him “anything that could make his life
agreeable” and reduced him to poverty. He “also provoked against him
every persecution by which evil human beings could embitter him –
sufferings that only one well-disposed can truly feel with depth [such
as] the slandering of his teaching’s pure intention” (R 6:81). And finally he
was made to suffer the most “ignominious death” (R 6:81). Looking at his
Kant’s Jesus 173
physical life, therefore, we can only say that he lost to the forces of evil.
But just because he never betrayed the moral principle, we can also say
that he won in moral or “legal” terms − or, perhaps better, in the realm of
freedom, where human beings can be controlled only insofar as they
adopt the right or the wrong principles.
The story of Jesus is the story of a moral hero who resisted temptation
and persecution to the death. It is not a story about someone who died on
the cross for our sins. The blood of Jesus does not cleanse us from these sins:
“So the moral outcome of this conflict, on the part of the hero of the story
(up to the death), is not really the conquering of the evil principle” (R 6:82).
The kingdom of evil still endures. Overcoming it requires our moral faith
“in the universal religion of reason,” which involves the “basis for a con-
tinual approximation to the ultimate perfection” (R 6:123).
Therefore, Kant could not have subscribed to the idea that “Jesus
Christ, His only Son, our Lord . . . descended into hell; the third day
He rose again from the dead; He ascended into heaven, and sitteth on the
right hand of God the Father Almighty; from thence He shall come to
judge the quick and the dead.” Kant’s Jesus is no Christ.42 Just like
Reimarus, Kant claims that the existence of God and the immortality of
the soul are the only essential ideas of moral religion or, as he calls them
“postulates of pure practical reason.” He avoids Reimarus’ naturalism
because, as he says in The Conflict of the Faculties, he does not adopt as a
principle the claim that “supranatural revelation” is impossible. The book
simply takes no notice of this source where religious doctrine is concerned
(R 6:44).43
To believe in Jesus as the son of God who has taken up human nature can
mean for Kant only that we should try to live in accordance with the
genuine moral principles he revealed (R 6:62). If we believe in this way,
we foster in ourselves the kind of disposition that makes the categorical
imperative our rule of action (R 6:66). And if we do so, we may hope to be
pleasing to God. This is what practical faith is and what believing in Jesus
means. It is a purely moral disposition. Following Jesus thus means for Kant
doing what Jesus taught and what he lived − namely, following the moral
42
This should also be clear from Kant’s discussion of the trinity as the mere “representation of a
practical idea” or the designation of “three specifically different ways” in which “God wills to be
served” (R 6:142).
43
Put differently, Kant does not feel the need that Johann Georg Hermann expressed in his attempt to
derive a faith in Jesus, the savior, from the historical record and his moral teachings. See Johann
Georg Herrmann, Ethik (Tübingen: J. C. B. Mohr, 1904), p. 115.
174 MANFRED KUEHN
law out of respect for the moral law: no more, no less. There is, accordingly,
just as little room for a Christology in Kant’s religious philosophy as there
was in Reimarus’.44 We may doubt that Kant personally viewed Jesus as
Christ, but we cannot know this. Luckily, perhaps, this is absolutely
irrelevant to the philosophical concerns of this chapter.
44
I agree with Gordon E. Michalson that, “strictly speaking,” there is “no reason why a christology
should appear in the Religion” (see his Fallen Freedom: Kant on Radical Evil and Moral Regeneration
[Cambridge University Press, 1990], p. 109). I also agree with him that Kant discusses atonement in a
Christian framework, but I am not sure that we need to call these considerations “christological,” but
perhaps the differences are merely semantic, because his criteria for christology seem minimal whereas
mine are more substantively theological, as they are for me defined by the sentences of the Nicene
Creed with which this chapter begins.
chapter 9
Pluralism in the ethical community

Nicholas Tampio
A major dilemma confronting contemporary liberal democratic political

theory is how to envision and actualize just societies that both presuppose
and nurture deep diversity. In Political Liberalism, John Rawls frames the
problem thus: “how is it possible for there to exist over time a just and stable
society of free and equal citizens, who remain profoundly divided by
reasonable religious, philosophical, and moral struggles?”1 To address this
question, Rawls uses aspects of Kant’s meta-ethical procedure (“constructi-
vism”) but distances himself from the comprehensive nature of Kant’s
moral doctrine.2 For Kant, the doctrine of right was part of an architectonic
system that also included a philosophy of religion. Rawls, however, would
abjure Kant’s thesis that “There is only one (true) religion, but there can be
several kinds of faith” (R 6:107–8). Political liberalism holds that it is both
imprudent and unethical to determine which moral and philosophical
doctrines reside within the one true religion. And yet Kant raises an
interesting point in the Religion that practical philosophy may raise “a
banner of virtue” to inspire people to go beyond the narrow confines of right
(R 6:94). Is there a way for political liberals to recover Kant’s insight that ethical
communities require a coalition of faiths? In other words, is there a way for
political liberals to incorporate Kant’s vision of religious pluralism into Rawls’
conception of a well-ordered society?
The aim of this chapter is to propose an affirmative answer to this question,
to show that political liberals may salvage more of Kant’s comprehensive
doctrine than Rawls realized. Initially, I historicize Kant’s philosophy of
religion by showing how it seeks a middle ground between Spinoza’s deni-
gration of traditional forms of religiosity and Leibniz’s dream of a new
European respublica christiana. Next, I show that Kant’s conception of the
1
John Rawls, Political Liberalism (New York: Columbia University Press, 2005), p. 4.
2
See Robert S. Taylor, Reconstructing Rawls: The Kantian Foundations of Justice as Fairness (University
Park: The Pennsylvania State University Press, 2011).
175
176 NICHOLAS TAMPIO
ethical community problematically excludes Jews, Muslims, and atheists.

Third, I describe Rawls’ insight behind, and means of, transforming aspects
of Kant’s comprehensive doctrine into a political conception of justice.
Finally, I consider how a politicized conception of the ethical community
may appeal to Muslim reformers such as Abdullahi Ahmed An-Na‘im. In
addition to recovering the idea of the ethical community for political liberal-
ism, the chapter seeks to show how contemporary political theorists may
approach the eighteenth-century European Enlightenment as a mine of ideas
that we should feel free to recast.
1 Reason, religion, and politics in the early Enlightenment

Kant’s Religion entered a conversation about reason, religion, and politics
that had been going on in Europe for over a century. Kant, like many of the
other great Enlightenment philosophers, was born, raised, and educated,
and worked, in a Christian milieu.3 Kant’s parents were Pietists, as were his
teachers at the University of Königsberg, and many of his students would go
on to serve as Lutheran pastors, teachers, and academics.4 A recurrent theme
in Kant’s writings is that his philosophy merely clarifies (Christian) com-
mon sense (KpV 6:8; G 4:404). At the same time, Kant shared the sentiment
common among Enlightenment philosophers that the medieval Christian
worldview had collapsed as a viable intellectual or political project. To
contextualize Kant’s attempt to advance Christian ideals in a philosophical
manner, I first describe the religious politics of Spinoza and Leibniz.5
Spinoza’s Theological-Political Treatise (1670) was an opening salvo in
what has come to be known as the radical Enlightenment.6 What makes this
conception radical is its attempt to extirpate, at the root, Christian theology
and theocracy. The first fifteen chapters of the Theological-Political Treatise
lay out a method of scriptural interpretation that (implicitly) denies the
intellectual plausibility of biblical stories. On its surface, Spinoza’s method
3
Peter Gay, The Enlightenment: The Rise of Modern Paganism (New York: Knopf, 1966), pp. 59–60.
4
Ian Hunter, Rival Enlightenments: Civil and Metaphysical Philosophy in Early Modern Germany
5
On the Spinoza−Leibniz relationship, see Matthew Stewart, The Courtier and the Heretic: Leibniz,
Spinoza, and the Fate of God in the Modern World (New York: Norton, 2006). On how these two
philosophers form the backdrop to Kant’s Critical philosophy, see Susan Neiman, The Unity of Reason:
Rereading Kant (New York: Oxford University Press, 1994), particularly pp. 147–56 on the “Pantheism
Controversy” of the 1780s.
6
Jonathan Israel, Radical Enlightenment: Philosophy and the Making of Modernity, 1650–1750 (New York:
Oxford University Press, 2001). On Kant’s ambiguous relationship to the radical Enlightenment, see
Jonathan Israel, A Revolution of the Mind: Radical Enlightenment and the Intellectual Origins of
Modernity (Princeton University Press, 2010), pp. 12–14.
Pluralism in the ethical community 177
of Biblical exegesis corresponds to Luther’s strategy of reading scripture
unmediated by subsequent traditions or impositions: “Provided we admit
no other criteria or data for interpreting Scripture and discussing its con-
tents than what is drawn from Scripture itself and its history, we will always
proceed without any danger of going astray.”7 Yet whereas Luther intended
to reinforce faith through returning to the Bible, Spinoza highlights textual
inconsistencies and implausibilities to dissolve religious orthodoxy.
Spinoza’s rebuttal to this charge is revealing: “my methodology works in
favor of Scripture by preventing passages which are clear and pure from
being corrupted to fit defective passages.”8 Yet Spinoza’s biblical herme-
neutics renders a great deal of the Bible “defective,” including its pervasive
accounts of miracles: “If anything is found [in Scripture] which can be
demonstrated conclusively to contradict the laws of nature or which could
not possibly follow from them, we must accept in every case that it was
interpolated into the Bible by blasphemous persons.”9 Spinoza, then, denies
the intellectual respectability of how the vast majority of Jews and
Christians have hitherto read the Bible.
After challenging the intellectual foundations of Christian theology,
Spinoza turns to laying the foundation for a purely secular politics. He
builds his political theory upon the idea of natural right, that is, “the
supreme law of nature that each thing strives to persist in its own state so
far as it can.”10 Spinoza’s conception of natural right does not differentiate
between human beings and other animals: every being naturally and rightly
strives to increase and preserve its vitality (conatus). Spinoza thus intensifies
the secularization process in earlier works of modern political philosophy
such as Hobbes’ Leviathan, which still retain traces of the classical natural
law tradition.11 According to Spinoza, the most natural political regime –
“that which approaches most closely to the freedom nature bestows on every
person”12 – is democracy, and the task of the state is to guarantee that
theocrats do not acquire a disproportionate amount of power for them-
selves.13 Spinoza thus provides the philosophical foundation for the secular
state that guarantees collective governance and individual rights by, in the
main, relegating religion to the private sphere, where it may still be
7
Benedictus de Spinoza, Theological-Political Treatise, trans. Jonathan I. Israel (Cambridge University
Press, 2007), p. 98.
8 9
Ibid., p. 153. Ibid., p. 91. 10 Spinoza, Theological-Political Treatise, pp. 195–96.
11
Steven B. Smith, Spinoza, Liberalism, and the Question of Jewish Identity (New Haven: Yale University
Press, 1997), p. 131.
12
Spinoza, Theological-Political Treatise, p. 202.
13
“The supreme right of deciding about religion belongs to the sovereign power”: ibid., p. 207.
178 NICHOLAS TAMPIO
regulated by the sovereign power.14 Just as reason trumps revelation in

Biblical interpretation, for Spinoza the secular power has incontrovertible
authority over religion.
Spinoza has an ambivalent relationship to religious pluralism. On the one
hand, Spinoza takes pains to emphasize “the usefulness and necessity of
Holy Scripture or revelation.”15 Scripture teaches, or can be re-interpreted as
teaching, basic moral lessons such as that “there exists a supreme being who
loves justice and charity, and that, to be saved, all people must obey and
venerate Him by practicing justice and charity towards their neighbor.”16
Spinoza thus seems to offer a peace branch to religious believers including,
presumably, the tolerant Christians (e.g. Collegiants and Socinians) who
befriended him in seventeenth-century Holland. On the other hand,
Spinoza presents a vision of the universe and humanity’s place in it that
he wishes all humans, to the best of their intellectual abilities, to share.
Spinoza’s religious terminology in the Theological-Political Treatise expresses
a naturalistic philosophy – developed with great rigor in the Ethics – that has
no need for the Abrahamic god of transcendence. Spinoza announces a
“universal or catholic religion (religio catholica) which might serve as a civil
theology for citizens of the modern democratic-republican state.”17 In the
short term, there might be no choice for Spinozists but to accept religious
diversity in the secular state. In the long term, however, Spinoza wages a
spiritual war against the claims of revelation and theocracy. Spinoza’s new
religio catholica is simply a civic ethos of tolerance that all citizens under a
democratic state share even as they pursue their own individually chosen
paths to happiness.
Leibniz dedicated his political philosophy to reviving the idea of a
respublica christiana to combat Spinoza’s championing of a secular meta-
physics and political order. Leibniz’ political philosophy is encapsulated in
his famous formulation, “justice is the charity of the wise.” Leibniz views
caritas sapientis as the glue that can bind together Christendom after the
unfortunate schism between Catholics and Protestants. Leibniz trusts that
Christian love, properly understood, can heal political−religious divisions
within Germany and the rest of Europe.18 Yet Leibniz thinks that Spinoza’s
work, among other developments, has rendered a return to scripture
untenable as a strategy in the project of doctrinal reunification. Leibniz’
14
Israel, A Revolution of the Mind, pp. vii−viii. 15 Spinoza, Theological-Political Treatise, p. 194.
16
Ibid., p. 182. 17 Smith, Spinoza, p. 131.
18
Patrick Riley, Leibniz’ Universal Jurisprudence: Justice as the Charity of the Wise (Cambridge, MA:
Harvard University Press, 1996), p. 236.
political philosophy endeavors to provide a philosophical foundation to
unite Catholics and Protestants in charitable relations and to fight common
enemies.19
Leibniz lashes out at the ethical consequences of a purely secular order as
envisioned by modern philosophers following in the footsteps of Spinoza.
To those who think that the purpose of statecraft is to attend to this life
alone, Leibniz charges, “more sublime and perfect is the theory of natural
law according to Christine doctrine . . . that not everything should be
measured by the goods of this life.”20 To those who think that the law
may only touch the human form, and not the soul, Leibniz contends that
“he who has control of the education or instruction of others is obligated, by
natural law, to form minds with eminent precepts, and to take care that the
practice of virtue, almost like a second nature, guides the will toward the
good.”21 What Leibniz is against is perfectly clear: a world drained of
concord among men (humanae tranquilitas), or, more precisely, a world
in which Christians do not trust each other or God. Leibniz fears the secular
state that relinquishes the task of soulcraft. And yet Leibniz does not cite
scripture to buttress his argument, for that would lead to disputations that
could prevent the whole project from getting off the ground.
Leibniz provides an argument that might provide Catholics and
Protestants with a basic rule to decide moral and political questions. In
the face of skeptics or cynics who deny that justice means anything other
than the rule of the strong, Leibniz contends that there are “necessary
and eternal truths which must be the same everywhere.”22 To determine
the content of justice, Leibniz suggests, perform a few simple thought
experiments. Say that someone could easily throw you a rope to save
you from drowning but doesn’t: wouldn’t you say that that person is
unjust? Or say that you could easily remove an impediment from someone’s
way to prevent him from suffering pain, but you don’t: are you too being
unjust?
If you refuse the request, he has reason to complain, since he can judge that
you would make the same request if you were in the place of him who makes
it. And it is the principle of equity, or what is the same thing, of the equality
19
See Ian Almond, History of Islam in German Thought from Leibniz to Nietzsche (New York: Routledge,
2010), Ch. 1.
20
Gottfried Wilhelm Leibniz, “Opinion on the Principles of Pufendorf,” in Leibniz, Political Writings,
ed. and trans. Patrick Riley (Cambridge University Press, 2001), p. 67.
21
Ibid., p. 69.
22
Gottfried Wilhelm Leibniz, “Meditation on the Common Concept of Justice,” in Leibniz, Political
Writings, p. 49.
180 NICHOLAS TAMPIO
or of the identity of reasons, which holds that one should grant [to others]
whatever one would wish in a similar situation, without claiming to be
privileged, against reason, or [without claiming] to be able to allege one’s
will as a reason.23
Unlike Kant in the Groundwork (G 4:430), Leibniz freely admits that “the
rule of reason” reformulates the Biblical passage quod tibi non vis vieri, etc.,
and, furthermore, Leibniz’s argument is heteronomous from a Kantian
point of view by holding that moral truths are discerned by theoretical (as
opposed to practical) reason. Yet Leibniz anticipates Kant’s aspiration to lay
the foundation for a practical philosophy that different ecclesiastical faiths
may endorse upon reflection.
Leibniz proposes several ideas for reconciling Catholics and Protestants.
Philosophers could exposit doctrines with sufficient rigor so that all parties
recognize their fundamental agreement. Lutherans could defer to Catholics,
in the interest of reestablishing the hierarchy and government of the visible
church, on the conditions that Lutherans not be humiliated and forced to
recant their views and that Catholics promise to rethink some of their
abusive practices. Finally, all Christians could abstract from or suspend
doctrinal exposition on controversial matters in order to facilitate the all-
important task of reunification: “each, on his side, must make the most
extreme effort which is possible without injuring his conscience, by showing
the greatest obligingness for the others that he can have without offending
God.”24 Near the end of his life – partly as a result of interlocutors such as
Boussuet striving for conversion rather than reconciliation – Leibniz low-
ered his expectations for a revival of the respublica christiana. Yet he con-
sistently hoped that “fanaticism and bloodshed could be ended throughout
Europe if caritas sapientis replaced theological hair-splitting.”25
Spinoza and Leibniz represent contending poles in the early
Enlightenment debate about the relationship among reason, religion, and
politics. Spinoza articulates a secular, modern politics in which individual
human beings realize that they best serve their natural rights by bonding
together into a democracy that protects the freedom to philosophize.
Leibniz yearns to recover a Christian, medieval politics that envisions the
Pope, the Holy Roman Emperor, and religious philosophers leading
the reunification of Christendom. It is into this battle – between the
23
Ibid., pp. 55–56.
24
Gottfried Wilhelm Leibniz, “Excerpts from Two Letters to Bossuet Concerning the Re-Unification
of Christendom (1692–99),” in Leibniz, Political Writings, p. 190.
25
Riley, Leibniz’ Universal Jurisprudence, p. 238.
metaphysical and radical Enlightenment, or between those who look back
to the medieval order shattered by the Thirty Years War and those who
yearn for an as-yet unimaginable secular era – that Kant enters with his
Religion.
2 Pluralism in the ethical community

Kant’s Religion incorporates elements of both Spinoza’s and Leibniz’
responses to the theological-political problem. Kant’s strategy of reading
the Bible – as a text that offers, or may be interpreted to offer, practical
rather than theoretical guidance – resembles Spinoza’s: both philosophers
historicize the Bible, and render its cognitive claims null, in order to subvert
the historical rule of revealed religion. At the same time, Kant, like Leibniz,
cherishes Christian ethics and wishes to see a flourishing religious culture
replete with churches and ministers (or scriptural scholars) (R 6:112–13).26 In
this section, we consider the vision of pluralism encapsulated in Kant’s
statement that “There is only one (true) religion; but there can be several
kinds of faith” (R 6:107–8).
According to Kant, establishing political right is a necessary but insuffi-
cient step for self-reflective practical agents. Beings who both require the
assistance of others and possess understanding may recognize the need to
establish authorities and institute the principle of right “limiting the free-
dom of each to the conditions under which it can coexist with the freedom
of everyone else” (R 6:98). The principle of right precludes the state from
enforcing any ecclesiastical faith because that imposition seems a contra-
diction in terms (or a gross violation of piety) and the state fails to fulfill its
duty if it stirs up religious strife rather than adheres to its duty of guarantee-
ing external right. In a Spinozist vein, Kant exclaims, “woe to the legislator
who would want to bring about through coercion a polity directed to ethical
ends” (R 6:96)! Even a “nation of devils” could establish a state on such a
secular foundation (ZeF 8:366), but this fact simply highlights the insuffi-
ciency of it as an aspiration for self-reflective practical agents. Human beings
“mutually corrupt each other’s moral disposition (Gesinnung) and make one
another evil” (R 6:94). The way to remedy the moral corruption induced by
other people is to collaborate with the right kind of people in the right way:
“an association of human beings merely under the laws of virtue, ruled by
this idea, can be called . . . an ethical community (ethische gemeine Wesen)”
26
On the tension between Kant’s religious sensibility and philosophy of autonomy, see Gordon
E. Michalson, Kant and the Problem of God (Oxford / Malden, MA: Blackwell, 1999).
182 NICHOLAS TAMPIO
(R 6:94). An ethical community differs from a state, insofar as it lacks

physically coercive mechanisms, but it is like a state, in that it systematizes
the maxims of self-reflective practical agents. Kant does not appeal to caritas
sapientis, as does Leibniz, but he does think that human beings need to forge
ethical relationships with other human beings that transcend the ius strictum
of the principle of right.
Kant reinforces this point by stating that self-reflective practical agents
need to join an ethical community to escape the ethical state of nature. An
ethical state of nature is one “in which the good principle, which resides in
each human being, is incessantly attacked by the evil which is found in him
and in every other as well” (R 6:96–97). An ethical state of nature is one in
which other human beings follow the strict interpretation of the principle of
right but hesitate to perform imperfect duties that cannot be legislated but
are necessary for a full expression of our moral nature. That is why Kant’s
Religion may be properly read as a work of political philosophy: Kant
acknowledges that a moral agent always needs to follow the moral law,
but it would be perverse not to recognize the social dimensions of moral
action. That is, Kant thinks that we have a moral responsibility to work
alongside others to minimize the social pressures that tempt us to “will not
to resist the inclinations when they invite transgression” (R 6:58). We may
not use peer pressure to justify our propensity to radical evil, but nor may we
use the moral responsibility of the individual will to turn a blind eye to the
political dimension of moral agency. An ethical community is one in which
self-reflective practical agents help one another embody moral virtue
and attain the highest good: a real-world approximation of the realm of
ends (G 4:433–36).
The Religion fleshes out the abstract speculations of the Groundwork by
specifying what human beings ought to do to advance the cause of the highest
good: “The sublime, never fully attainable idea of an ethical community is
greatly scaled down under human hands, namely to an institution which, at
best capable of representing with purity only the form of such a community,
with respect to the means for establishing a whole of this kind is greatly
restricted under the conditions of sensuous human nature” (R 6:100). A
human being cannot form an ethical community on her own: she needs
other people. Nor can a human being simply join an invisible church, which
would attain purity at the cost of being merely an intellectual abstraction. For
a human being who has both a moral principle and the need for human
communion, there is no other alternative than to join a social body that we
call a church: “the idea of a people of God cannot be realized (by human
organization) except in the form of a church” (R 6:100).
Ideally, human beings could participate in a universal church founded on
“pure religious faith,” that is, one that honors only the moral principle and
removes all sensual representations or incentives (R 6:102–3). Yet, given our
sensuous (sinnlichen) nature, we need some earthly stories, traditions, cus-
toms, songs, buildings, and so forth to house rational religion’s pure moral
teaching, and there may be no other alternative than a “historical (revealed)
faith, which we can call ecclesiastical faith” (R 6:102). An ecclesiastical faith
satisfies “the natural need of all human beings to demand for even the
highest concepts and grounds of reason something that the senses can hold on
to” (R 6:109). An ecclesiastical faith instructs and animates with basic
principles for action; it helps us see and move toward the right thing to
do (R 6:112). In practice, then, human beings have no choice but to join
already-established churches with their own traditions and interpretations
of scripture. More so than Spinoza, Kant recognizes the ineradicability of
churches – or at least the profound danger of abandoning the institution of
churches without collective ethical bodies to replace them.
Kant, however, agrees with Spinoza that philosophy dictates the terms of
cooperation between faith and reason. Pure practical reason presents the
moral law and the religious postulates that confirm our faith in the possi-
bility of its actualization. Historical revelation occurs at certain times and
places and may express passions and superstitions. Given that pure practical
reason needs the assistance of revealed religions, and given that revealed
religions are as likely to corrupt as to honor moral incentives, then pure
practical reason may need to force scripture to reveal moral principles that
otherwise are not there: “To unite the foundation of a moral faith . . . with
such an empirical faith which, to all appearances, chance has dealt to us, we
require an interpretation of the revelation we happen to have, i.e. a thor-
oughgoing understanding of it in a sense that harmonizes with the universal
practical rules of a pure religion of reason” (R 6:110). What happens in the
realm of ideas also transpires in the social realm: interpreters of scripture
(such as Kant) present the teachings that are popularized by scriptural
scholars (who follow Kant’s lead) against religious enthusiasts (whose
morality is heteronomous from a Kantian point of view) (R 6:112–14).
Kant’s long-term goal might be nothing less than the capture of the religious
establishment to propagate the teachings of pure rational morality.27
En route to the formation of a universal church, however, diverse ecclesi-
astical faiths may join the ethical community here and now. Kant provides a
27
Susan Meld Shell, Kant and the Limits of Autonomy (Cambridge, MA: Harvard University Press,
2009), p. 315.
184 NICHOLAS TAMPIO
distinction to determine who may or may not join this social union. A faith of
divine service views human beings as obligated to serve God through rituals
“however morally indifferent the actions might be in themselves” (R 6:103).
This type of faith is “a slavish and mercenary faith (fides mercenaria, servilis)
and cannot be considered as saving, because it is not moral” (R 6:115).
Religious mercenaries clearly are not fit to join the universal church or the
ethical community. A faith of moral religion, on the contrary, promotes and
propagates “good life conduct” (R 6:105). This type of faith recognizes that
“whenever [human beings] fulfill their duties toward human beings (them-
selves and others), by that very fact they also conform to God’s commands”
(R 6:103). A faith of moral religion knows that the object of human striving
ought to be a good will, not an action or quality that may or may not be moral
(G 4:393). Kant concedes that there may be many ways to graft this teaching
onto sensual human beings – which is why there may legitimately be several
kinds of faith. But Kant also thinks that any ecclesiastical faith in the ethical
community must ultimately endorse the one religion of reason (R 6:123).
Kant’s paradigm of a faith of moral religion is Christianity. In his
historical account of the gradual establishment of the “dominion of the
good principle on earth,” Kant begins with Christianity, the religion that
“effected a total revolution in doctrines of faith” (R 6:127). There is more
than a grain of salt in the dictum that Kant tries to secularize Christianity −
that is, to provide Christian ethics with a new philosophical foundation.28
How, though, does Kant square his respect for Christianity with the
previous century’s evidence of bloody intra-religious warfare? “The terrible
voice of orthodoxy . . . split the Christian world into bitter parties over
opinions in matters of faith (upon which, without recourse to pure reason
as the expositor, no universal agreement can possibly be attained)” (R 6:130).
For Kant, as for Leibniz, Christendom can reunify only if diverse ecclesias-
tical faiths recognize that what binds them (pure moral religion) is much
greater than what divides them (ceremonies, etc.). Kant makes this point
forcefully in The Conflict of the Faculties: “Enlightened Catholics and
Protestants, while still holding to their own dogmas, could thus look
upon each other as brothers in faith, in expectation (and striving toward
this end): that, with the government’s favor, time will gradually bring the
formalities of faith closer to the dignity of their end, religion itself” (S 7:52).
The historical referent in the title of the Religion within the Boundaries of
Mere Reason is Christianity, pure and simple.
28
This argument goes back at least to the young Hegel. See Georg Wilhelm Friedrich Hegel, Early
Theological Writings, trans. T. M. Knox (University of Chicago Press, 1948).
Jews as such, Kant emphasizes, are not welcome to join the ethical
community: “The Jewish faith, as originally established, was only a collec-
tion of merely statutory laws supporting a political state” (R 6:125). Judaism
exposits coercive laws rather than moral principles; it ignores the moral need
for a belief in a future life; and it excludes nearly the entire human race from
its communion (R 6:125–27). Catholics and Protestants may disagree on
peripheral matters while still sharing the same final moral end, but Jews
“need to exchange their old garment for a new one.”29
Muslims also belong to a “fetish-faith (Fetischglaube)” that denies them
entry into the ethical community (R 6:193). A fetish-faith seeks to circum-
vent the arduous work of morality by finding an “escape route” that will
reward customs or formalities: “In every type of public faith the human
being has devised certain practices for himself, as means of grace, even
though such practices are not related in all faiths, as in the Christian, to
practical concepts and to dispositions conformable to them.” Kant notes
that four of the five pillars of the “Mohammedan faith” – washing, praying,
fasting, and the pilgrimage to Mecca – contain no intrinsic worth, and the
last one – almsgiving – may be performed by extorting others (R 6:194). In
light of Kant’s rare references to Islam, one scholar speculates that Kant
viewed Islam either as historically insignificant or as a potential threat to
rational philosophy hitherto at home only in Europe.30
Kant’s approach to the theological-political problem in the Religion
mirrors Spinoza’s in the Theological-Political Treatise.31 Both philosophers
think that reason may torture revelation, so to speak, to confess moral truths
(R 6:110–11). Both philosophers yearn for the day when historical or eccle-
siastical faiths may disappear in the universal recognition of rational moral
principles (R 6:135). And both philosophers temper their hopes for a
universal rational religion with a recognition that sensual human beings
with vivid imaginations are likely to form schisms and sects that prohibit the
formation of a single church (R 6:123). Yet the content of Kant’s moral
religion is very different from Spinoza’s religious ethics.32 To take only one
of many examples: Spinoza’s one-substance doctrine that denies the exis-
tence of an intelligible or noumenal soul (and thus its freedom or immortal-
ity) leads directly, in Kant’s view, to morality-killing fatalism (KpV 5:101–2).
29
See Shell, Kant and the Limits of Autonomy, p. 325.
30
Almond, History of Islam in German Thought, p. 29.
31
Yirmiahu Yovel, “Bible Interpretation as Philosophical Praxis: A Study of Spinoza and Kant,” Journal
of the History of Philosophy 11 (1973), pp. 189–212.
32
Gilles Deleuze, Spinoza: Practical Philosophy (San Francisco: City Lights Books, 1988), pp. 17–29.
186 NICHOLAS TAMPIO
Kant seems to prohibit Spinozists – or, more colloquially, atheists – from

joining the ethical community.
What, then, is Kant’s vision of religious pluralism in the Religion? Kant is
much more committed to the secular state than was Leibniz – particularly in
early works such as the Caesarinus Fürstenerius (1677) in which Leibniz pined
for a revival of the Holy Roman Empire. Kant was also more religiously
ecumenical than were many of the leading figures in the Reformation or the
Counter-Reformation: Kant feared any “catholic” church that tried to make
its ecclesiastical faith universally binding (R 6:109). From the vantage point of
political theorists living in 21st-century liberal democracies, however, Kant’s
breadth of vision is remarkably narrow, allowing Jews, Muslims, and atheists
(perhaps) to live securely in society but not to participate in the full life of the
ethical community. Rather than discard the insights of Kant’s Religion,
however, I propose to take a cue from Rawls on how Kant’s practical
philosophy may be recast for pluralistic societies.
3 Recasting the ethical community as a political conception

Kant presents his argument in Part Three of the Religion as the disciplined
following of an a priori argument available to anyone willing to reflect on
the duty of the human race toward itself: “we must following up the leading
thread (Leitfaden) of that moral need and see where it will lead us” (R 6:98).
And yet, Allen W. Wood notes, Kant’s conclusion is “obviously an idealized
Enlightenment version of the Christian church.”33 Perhaps there is an
a priori route to Christian ideals.34 But this will strike many citizens in
pluralistic societies as dubious, as long as Kantians are in the minority, or as
dangerous, if somehow Kantians set the terms of ethical coalitions. Rawls
knew that Kant’s practical philosophy provided resources for contemporary
liberal democratic political theorists.35 And yet Rawls also saw – and
increasingly emphasized in his later work – that trying to impose, through
physical or peer pressure, Kant’s doctrines on a diverse society would
constitute a form of illiberal oppression. Thus Rawls proposed to politicize
Kant’s ideas – such as the kingdom of ends36 – to make them speak to a
33
Allen W. Wood, Kant’s Ethical Thought (Cambridge University Press, 1999), pp. 316–17.
34
“Kant’s project bears a striking similarity to the Christian gospel; and Kant ultimately affirms
Christianity (at least as taught by Jesus himself) as a rational religion”: Chris L. Firestone and
Nathan Jacobs, In Defense of Kant’s Religion (Bloomington: Indiana University Press, 2008), pp. 5–6.
35
See Nicholas Tampio, “Rawls and the Kantian Ethos,” Polity 39.1 (2007), pp. 79–102.
36
For a critique of this appropriation, see Katrin Flikschuh, “Kant’s Kingdom of Ends: Metaphysical,
Not Political,” in Jens Timmermann (ed.), Kant’s Groundwork of the Metaphysics of Morals: A Critical
Guide (Cambridge University Press, 2009), pp. 119–39.
wider spectrum of citizens.37 Clearly, Rawls’ approach violates the letter of
Kant’s doctrines, but it may point to a way to revive their relevance in
contemporary circumstances.
How does one go about politicizing a Kantian idea? The first step is to
assemble “basic intuitive ideas found in the public culture of a constitutional
democracy.”38 This empirical procedure drops Kant’s claims to universality or
necessity, but it does ensure that the subsequent argument has traction with its
audience. Then, the philosopher attempts to reconstruct these intuitions in
propitious ways, including by bringing in ideas from the history of philosophy
or other cultures, to forge theories that may act as an “ambulance service” for
democratic common sense in moments of crisis.39 For the purposes of political
liberalism, Rawls recommends that citizens exercise moderation when
expounding their metaphysics in democratic fora. That is to say, citizens
who endorse Kant’s practical or religious philosophies in their non-public
lives may choose to ponder the implications of Kant’s thesis − for instance,
that “the concept of the Divinity actually originates solely from the conscious-
ness of [moral laws]” (R 6:104) − but politicizing Kant’s ideas means not
demanding adherence to a perspective that will strike many citizens of faith
as heretical. Two points must be made immediately. A politicized conception
may still be a moral conception. The ideas in a public political culture may be
just, but the diversity of comprehensive moral and philosophical doctrines
precludes deep agreement on existential questions. Furthermore, a politicized
conception is, in principle, reachable through different routes. In an over-
lapping consensus each comprehensive doctrine may endorse the politicized
conception in its own way.40 In sum, a politicized Kantian idea does not
demand assent to Kant’s metaphysics but reworks the idea to capture and
reshape a shared intuition among citizens.
Kant’s great insight in Part Three of the Religion is that progressive
politics needs to aspire to more than a just legal order. The core idea of
the left is equality among all human beings. Leftists often try to use the
legislature or the courts to establish just laws that guarantee the appropriate
37
To politicize a conception means to find arguments that may appeal to a wide range of reasonable
citizens who propose and honor fair principles of justice and recognize the burdens of judgment.
Ideally, there are many such arguments, not just “secular” ones.
38
John Rawls, “Justice as Fairness: Political Not Metaphysical,” in Rawls, Collected Papers, ed.
Samuel Freeman (Cambridge, MA: Harvard University Press, 1999), p. 410.
39
Jerome B. Schneewind, Essays on the History of Moral Philosophy (New York: Oxford University Press,
2010), p. 7.
40
Rawls, “Justice as Fairness,” p. 411. On political liberalism’s appeal to Sunni Muslims, see Andrew
F. March, Islam and Liberal Citizenship: The Search for an Overlapping Consensus (New York: Oxford
188 NICHOLAS TAMPIO
distribution of rights and duties. Left legalism, however, poses numerous

problems: it reifies certain conceptions of who or what should be equal; it
breeds resentment among those who think that reasonable people may
disagree on what has been instituted through law; and it inhibits other,
perhaps better, strategies to foster human fraternity.41 One of the great
questions of contemporary political theory is how to conceptualize an ethos
that transcends an obsession with right or justice.42
How might we politicize Kant’s conception of the ethical community? The
purpose of the idea of the ethical community is to unfurl a “banner of virtue”
that may rally a wide range of people (R 6:94). Human beings often tempt and
pressure each other to do the wrong thing. One way to combat the “dominion
of evil” is to collaborate with other people to establish ethical arrangements
that reinforce the good principle in each of us (R 6:93). Citizens of many
different faiths, I conjecture, can agree that ethical (or reasonable or decent or
thoughtful or Christian or Muslim or . . .) behavior requires that we work
alongside others to advance common ends. A politicized conception of the
ethical community refrains from demanding universal assent to a compre-
hensive moral doctrine. Thus, Kantians should not assert that all citizens must
endorse Kant’s conception of radical evil that plays a germinal role in the
Religion.43 Kantians, of course, would be valuable members of any ethical
community – Kant provides compelling reasons (for many people) why we
should treat all others as ends-in-themselves. Yet Kantians would temper their
expectations for full philosophical consensus by recognizing the danger posed
by any “ecclesiastical faith which rules despotically” (R 6:131). Let us now turn
to perhaps the most controversial test case today of whether an ethical
community is possible.
4 Muslims and the ethical community

Rawls invented the term “overlapping consensus” to describe a situation in
which most citizens subscribe to reasonable comprehensive doctrines that
affirm “a political conception of justice underwriting a constitutional dem-
ocratic society.”44 An overlapping consensus is formed in a pluralistic
41
See Wendy Brown and Janet E. Halley (eds.), Left Legalism / Left Critique (Durham: Duke University
Press, 2002).
42
See, for instance, Stephen K. White, The Ethos of a Late-Modern Citizen (Cambridge, MA: Harvard
43
See Sharon Anderson-Gold and Pablo Muchnik (eds.), Kant’s Anatomy of Evil (Cambridge University
Press, 2010).
44
John Rawls, Political Liberalism (New York: Columbia University Press, 2005), p. 482.
society when reasonable citizens endorse minimal standards of justice from
the vantage points of their own comprehensive worldviews. The overlap-
ping consensus agrees to ius strictum but not to the metaphysical reasons
that support it. A politicized conception of the ethical community, how-
ever, envisions coalitions to advance common ends that surpass the minimal
requirements of justice. Many Islamists – that is, those who think the
Qur’an and the normative example of the Prophet (sunna, hadith) ought
to govern the entirety of the social and political world45 – would find the
idea of forming ethical coalitions with atheists absurd, akin to retreating to
the age of pagan ignorance of Islam (jahiliyya).46 Yet it is worth considering
whether some Muslims would favor joining the ethical community. If the
answer is yes, then those of us emerging from the Kantian tradition could
set about doing the hard work of instantiating this idea in practice. It is with
that intent in mind that we should read Abdullahi Ahmed An-Na‘im’s
recent book, Islam and the Secular State: Negotiating the Future of Shari‘a.
An-Na‘im opens the book with a declaration that true faith may not be
coerced: “In order to be a Muslim by conviction and free choice, which is
the only way one can be a Muslim, I need a secular state.”47 An-Na‘im’s
argument against religious coercion is explicitly Islamic: “Religious com-
pliance must be completely voluntary according to personal pious intention
(niyah), which is necessarily invalidated by coercive enforcement of those
obligations. In fact, coercive enforcement promotes hypocrisy (nifaq),
which is categorically and repeatedly condemned by the Qur’an.”48
We see, here, a gesture toward refuting Kant’s assertion that Islam is a
“fetish-faith” unconcerned with purity of conscience. We also see An-
Na‘im aspiring to refute Islamists who wish to use the state, for instance,
to punish those they deem apostates.
An-Na‘im’s second claim is that Muslims have the resources from within
their own history and texts to endorse the principle of right underlying the
secular state. The Ottoman Empire indicates that Muslim authorities have
often been willing in practice to differentiate the state and religion, and
postcolonial Muslim rulers have clearly imported the model of the totalitarian
state from Europe.49 In addition to policy reasons why a secular state is more
effective than an Islamic state, there are also Islamic reasons to favor
45
Roxanne L. Euben and Muhammad Qasim Zaman (eds.), Princeton Readings in Islamist Thought:
Texts and Contexts from Al-Banna to Bin Laden (Princeton University Press, 2009), p. 4.
46
See Sayyid Qutb, Milestones (Cedar Rapids: Mother Mosque Foundation, 1995).
47
Abdullahi Ahmed An-Na‘im, Islam and the Secular State: Negotiating the Future of Shari‘a
(Cambridge, MA: Harvard University Press, 2008), p. 1.
48
Ibid., p. 4. 49 Ibid., pp. 16, 20.
190 NICHOLAS TAMPIO
differentiating Islam and the state. According to An-Na‘im, the Islamic

principle of reciprocity (mu‘awada), or the Golden Rule, supplies the rationale
for why Muslims should not expect the state to favor one religion over
another.50 What is mu‘awada? In The Second Message of Islam, Mahmoud
Mohamed Taha – An-Na‘im’s mentor in Sudan – explains how this principle,
alongside the principle of retribution (al-qasas), forms “the essence and
foundation of both the Shari‘a and the truth (haqiqah).”51 Taha presents
this idea by explicating the Qur’anic verse justifying an “eye for an eye” (5:45).
Anyone who pulls out the eye of another person, in a fit of anger, for example,
does not do so while fully realizing the degree of pain and the magnitude of the
injury he is thereby inflicting upon his victim. If he received retribution by
being placed in the same position as his victims, and his eye is pulled out in
reciprocity (mu‘awada) for what he had done, then two purposes would have
been served at the same time. Firstly, the interest of the community would be
preserved by deterring the aggressor himself, as well as deterring others by his
example. Secondly, the aggressor deepens his sensitivity, by himself experienc-
ing the pain he inflicts upon others, and thus realizes the severity of the pain and
the magnitude of the loss he has caused.52
This passage will not be particularly persuasive to Christians who believe in

turning the other cheek to an evil person.53 Nor does it mean that Muslims
must enact eye-for-an-eye justice; in fact, the principle of “doing good to
others” (al-ihsan) may promote forgiving aggressors.54 For our purposes,
though, it shows that Muslims may formulate their own rationale for the
secular state.55 An-Na‘im agrees with Spinoza, Kant, Locke, and Rawls that
the state should not practice soulcraft.
An-Na‘im’s third claim is that Muslims may and must connect Islam and
politics even as they distinguish Islam and the state: “The state can serve the
ideals of an Islamic society for social justice, peace, goodness, and virtue by
enabling and facilitating their realization through civic discourse and the
fabric of political life.”56 Islamic ethics promotes certain objectives for social
life, and Muslims may and must use political means to advance them.57
50
Ibid., pp. 34, 95.
51
Mohamed Mahmoud Taha, The Second Message of Islam, trans. Abdullahi Ahmed An-Na‘im
(Syracuse University Press, 1987), p. 72.
52
Ibid., p. 73. 53 Matthew 5:39. 54 Taha, The Second Message of Islam, p. 77.
55
Cf. Kant’s explication of Psalm 59 that, on its face, is a “prayer for revenge,” but that may be read to
enjoin that “one should seek satisfaction for insults in the court of justice” (R 6:110).
56
An-Na‘im, Islam and the Secular State, p. 293.
57
Surprisingly, An-Na‘im eschews appealing to the “objectives of Shari‘a” (Maqasid al-Shari‘a) to
formulate policy objectives that may be shared with non-Muslims. Cf. Tariq Ramadan, Radical
Reform: Islamic Ethics and Liberation (New York: Oxford University Press, 2009).
Yet Muslims cannot simply expect non-Muslims to accept the rationale
“because it is in the Qur’an or Prophetic tradition.” Therefore, An-Na‘im
demands that Muslims honor human rights, follow constitutional proce-
dures, and employ public reason when presenting policy arguments to non-
Muslims in order to honor the principle of reciprocity. For An-Na‘im,
Muslims can be ethical and, in Rawls’ terms, reasonable, respecting the need
in pluralistic societies to formulate fair terms of justice and to recognize the
existence of reasonable disagreement about the ultimate ends of life.
One of An-Na‘im’s goals in Islam and the Secular State is to present an
Islamic argument for pluralism. Historical Shari‘a perpetuates a dhimma
system that grants full political rights to Muslims, some political rights to
“people of the book” (Ahl al-Kitab) such as Jews and Christians, and no rights
to unbelievers.58 Though few countries institute the dhimma system, An-
Na‘im thinks that its idealization for many Muslims corrupts civil relations
between people around the world. For An-Na‘im, it is not enough for
Muslims to accept the (unfortunate) fact of religious, ethnic, and demographic
diversity. Muslims must find a way to endorse pluralism, “an ideology and
system that accepts diversity as a positive value and facilitates constant
negotiations and adjustments among varieties of difference without seeking
or expecting to terminate any or all of them permanently.”59 An-Na‘im
recurrently appeals to Muslims to join an overlapping consensus supporting
the secular state, even if Muslims may adjust Rawls’ terminology to suit local
circumstances.60 He would also, I believe, be receptive to the idea of Muslims
joining an ethical community committed to promoting the good such as
distributing resources to the poor in what Jews call tzedakah, Christians call
caritas, Muslims call zakat, and Kantians call a duty of beneficence to others
(MS 6:432).61 An-Na‘im thus points to groups such as the Liberal Islam
Network, based in Indonesia, that promote a pluralistic vision of Islam.62
An urgent task facing Muslims today, according to An-Na‘im, is to lock arms
with non-Muslims to work for both political right and ethical ends.
5 Conclusion
The thesis of this chapter – that contemporary political theorists should
politicize Kant’s notion of the ethical community – will draw fire from at
least two sides. Enlightenment liberals may think that this approach makes
58
Abdullahi Ahmed An-Na‘im, “Translator’s Introduction,” in Taha, The Second Message of Islam, p. 22.
59
An-Na‘im, Islam and the Secular State, p. 225. 60 Ibid., p. 101. 61 Ibid., p. 95.
62
Ibid., p. 257.
192 NICHOLAS TAMPIO
too many compromises with illiberal forces. Robert S. Taylor, for instance,
thinks that Rawls’ political turn was a mistake and that there may be certain
comprehensive doctrines with which one simply cannot negotiate.
Muslims, on Taylor’s account, view Shari‘a as binding law and thus could
not accept the idea of political autonomy that makes possible cooperation in
pluralistic societies.63 Far better, for Taylor, for liberals to advocate Kant’s
practical philosophy in the present war of ideas against illiberal doctrines
and regimes.64 From another angle, Islamists such as President Numerei in
the Sudan view any critique of Shari‘a or the dhimma system as fomenting
religious turmoil ( fitnah) – which is why he had Taha executed on January
17, 1985.65 Clearly we could not expect to find universal, or maybe even
wide, assent to the idea of a pluralistic ethical community today.
And yet there are good reasons, I contend, why Euro-American liberals
situating themselves in the Enlightenment tradition may be receptive to the
idea of politicizing the idea of the ethical community. Kant’s religious philos-
ophy in its original form contravenes what Rawls called the “fact of oppres-
sion” – namely, that a comprehensive philosophical or religious doctrine can
only maintain political or social hegemony through oppressing dissidents.66
Kantians may promote a vision of a secularized kingdom of God but they
should not coerce or stigmatize Jews, Muslims, atheists, or other groups that
doubt that pure practical reason is the ground of moral principles. This
position is consonant with admiration for Kant’s insight that we need to
form ethical assemblages that go beyond strict right and that incorporate a
wide range of voices: “All societies are in fact negotiating the relationship
between religion and the state [and politics] over many issues at different
times.”67 An-Na‘im’s point, I think, is valid and urges us – inspired by Kant’s
writings – to do our part in raising a banner of virtue that may rally people to
pursue ethical ends in tandem.68
63
Taylor, Reconstructing Rawls, p. 263.
64
“Liberals should dedicate themselves chiefly to perfecting, extending, and popularizing the canonical
comprehensive liberalisms,” especially Kant’s: ibid., p. 315.
65
An-Na‘im, “Translator’s Introduction,” in Taha, The Second Message of Islam, p. 16.
66
Rawls, Political Liberalism, p. 37. 67 An-Na‘im, Islam and the Secular State, pp. 36–37.
68
Pluralists may retain an ideal of the ethical community, I think, while recognizing in practice the
ineliminability of many interlocking and contending ethical communities. This idea abandons Kant’s
ideal of a universal church triumphant (R 6:135), but it may provide better guidance for human beings
who thoughtfully disagree on the basis of ethics. The United Nations, with all its flaws, may provide
an approximation of what the ethical community might look like today.
chapter 10
Kant’s religious constructivism

Pablo Muchnik
I want to suggest a general interpretative strategy for reading Religion within

the Boundaries of Mere Reason − namely, as an attempt to find a middle
ground between what Kant considers two forms of excess: the appeal to a
transcendent conception of God, and the denial of any claim that presup-
poses God’s existence.
To make my case, I will start by presenting side by side the conflicting
views of two contemporary philosophers, Richard Rorty and Nicholas
Wolterstorff. Their opposing claims on the role of religion in politics gives
rise to a situation comparable to a Kantian antinomy. Putting things this
way underscores the originality of Kant’s view. For, as Kant saw it, this
kind of impasse is a “remarkable phenomenon [which] works most
strongly of all to awaken philosophy from its dogmatic slumber, and to
prompt it toward the difficult business of the critique of reason itself”
(P 4:338). The wake-up call of an antinomy will serve to motivate Kant’s
solution to the problem raised by dogmatic religious claims, as well as to
capture what – in my view – is the distinctive ethical function he reserved
for religion in the critical system: the support of the non-individualistic
virtues involved in shared undertakings and common pursuits. These are
the sociable virtues necessary to overcome the destructive effects of our
unsociable sociability – virtues which individuals cannot cultivate on their
own, because they require bonds of mutual affection and affiliation to a
community based on trust.1 Kantian religion is tailored to encourage this
kind of affiliation and restrain those aspects of religious claims that set
people at odds – Kant’s God is made so that religion can promote the
conversation of humankind.
I want to thank Lawrence Pasternack, Gordon E. Michalson, and Lauren Barthold for their helpful
comments on an earlier version of this chapter, and the audiences of Oklahoma State University and
Rochester Institute of Technology for a lively discussion of my views.
1
The idea of singling out these sociable virtues is inspired by Nancy Sherman. See Nancy Sherman,
“The Virtues of Common Pursuit,” Philosophy and Phenomenological Research 53 (1993), pp. 277–99.
193
194 PABLO MUCHNIK
1
Liberalism, however, has accustomed us to seeing religion in a very different
light. Take, for instance, the case of Richard Rorty, who (at least in his
earlier writings) considers religion as an eminently private matter, some-
thing to be kept within the confines of one’s own conscience and aban-
doned when entering the public square. As he pithily puts it: “The main
reason religion needs to be privatized is that, in political discussion with
those outside the relevant religious community, it is a conversation-
stopper.”2 According to this view, those who bring religion into politics
have “bad taste,” because religion, very much like our family and love lives,
pertains to the search for private perfection and has no role to play in
deciding matters of common concern. Such matters, Rorty thinks, are best
settled “by public discussion in which voices claiming to be God’s, or
reason’s, or science’s, are put on a par with everybody else’s” (RCS 172).
In the free exchange of ideas, what matters is the ability “of a political
proposal . . . to gain assent from people who retain radically diverse ideas
about the point and meaning of human life, about the path to private
perfection. The more such consensus becomes the test of a belief, the less
important is the belief’s source” (RCS 173). To the extent that religion draws
on premises meaningful only for those who share a certain creed, its appeals
have no chance to achieve widespread consensus. Thus, in the extreme
liberal view the early Rorty advocates, avoiding reference to the religious
source of one’s vocabulary is necessary to preserve the “Jeffersonian com-
promise that the Enlightenment reached with the religious” (RCS 169) −
namely, the privatization of one’s religious views in exchange for liberty and
toleration. “It does me no injury,” Jefferson famously claimed, “for my
neighbor to say that there are twenty Gods or no God” – for what one’s
neighbor thinks about questions of ultimate importance, provided she
keeps her views sufficiently private, does not interfere with our common
project of building a just society together.3
Rorty eventually softened this view after reading Nicholas Wolterstorff’s
essay “Why We Should Reject What Liberalism Tells Us about Speaking and
2
See Richard Rorty, “Religion as a Conversation-Stopper” (hereafter, RCS), in Philosophy and Social
Hope (London: Penguin Books, 1999), p. 171. This paper was first published in 1994 (see Common
Knowledge 3.1 [Spring 1994], pp. 1–6). Page numbers are from the Penguin edition. Rorty later revises,
and considerably weakens, this extreme liberal view (see note 4).
3
See Richard Rorty, “The Priority of Democracy to Philosophy,” in Objectivity, Relativism, and Truth:
Philosophical Papers I (Cambridge University Press, 1991), pp. 175–96.
Kant’s religious constructivism 195
Acting for Religious Reasons.”4 I will ignore Rorty’s reformulation, because it
is in the starker version that Kant’s ingenuity becomes clearer. In his influential
essay, Wolterstorff takes issue with what he calls the “neutrality postulate” of
liberalism.5 This postulate contains two strands: the separation position, which
claims “that government is to do nothing to advance or hinder any religion”
(RLRR 165), and the independent-basis position, which claims that the legiti-
macy of a public utterance or decision must rest on “some source independent
of any and all of the religious perspectives to be found in the society”
(RLRR 166). The insistence on such an independent basis, Wolterstorff argues,
is a remnant of the wars of religion, but is no longer necessary in contemporary
societies. The “slaughter, torture, and generalized brutality of our century
has mainly been conducted in the name of one and another secularism”
(RLRR 167), not by invoking God or canonical scriptures. Contemporary
liberals, Wolterstorff reckons, misplace the source of their fears: the danger for
liberalism does not lie in religion, but in the secular ideologies that have
replaced it. Furthermore, the effort to find an independent criterion among
liberal theorists, old and new, has been futile. Neither Locke’s appeal to a
universal human nature nor Rawls’ “consensus populi” could do the job
(RLRR 168–76). Thus, in Wolterstorff’s critical assessment, contemporary
liberals have also misplaced the direction of their theoretical efforts.
The combined effect of these failures is devastating. The separation
position, Wolterstorff believes, has led to the split between the private
and public selves of the religious, and the independent-basis position, to
the impoverishment of public discourse. “[W]hat has rushed to fill the void
[left by silencing religion] is mainly considerations of economic self-interest,
4
See Richard Rorty, “Religion in the Public Square: A Reconsideration,” Journal of Religious Ethics 31.1
(2003), pp. 141–49. Here Rorty “back-paddles” (his words) and distinguishes between two different
functions of religion. At the level of the parish, a “congregation of believers ministered to by pastors”
can “help individuals find meaning in their lives, and . . . help individuals in their times of trouble”
(p. 142). This supportive role of religion, however, must be distinguished from the divisive function of
“ecclesiastical organizations – organizations that accredit pastors and claim to offer authoritative
guidance to believers” (p. 141). It is the encroachment of these organizations that needs to be
forestalled: to advance toward a secular utopia, one has to “prune back” religion to the parish level,
for ecclesiastical organizations aim at “promulgating orthodoxy and acquiring economic and political
clout” (p. 141), and “typically maintain their existence by deliberately creating ill-will toward people
who belong to other such organizations, and toward people whose behavior they presume to be
immoral. They thereby create unnecessary human misery” (p. 142). For the late Rorty, then, the
pertinent distinction is no longer between religion and secular reason, but between good and bad
religion, and good and bad secular reason – “good” understood as involving the diminution of
suffering and humiliation in both cases.
5
See Nicholas Wolterstorff, “Why We Should Reject What Liberalism Tells Us about Speaking and
Acting for Religious Reasons” (hereafter RLRR), in Paul J. Weithman (ed), Religion and Contemporary
Liberalism (University of Notre Dame Press, 1997), pp. 162–81. In my quotations, I provide the page
numbers from this text.
196 PABLO MUCHNIK
of privatism, and of nationalism” (RLRR 177). These selfish demands

religion has always kept in check: “in all the great religions of the world
there are strands of conviction which tell us that pocketbook, privacy, and
nation are not of first importance” (RLRR 178). To neglect this uplifting
aspect of religion, to bury its moral authority in the depths of our privacy,
deprives politics of its capacity to inspire and motivate us.
2
As I see it, the dispute between defenders and resisters of religious privati-
zation hinges on whether the so-called “Jeffersonian compromise,” i.e., the
conception of politics as a strictly secular enterprise, is sufficient to over-
come the disruptive effects our selfish tendencies have on the political – the
same effects and tendencies which make politics necessary in the first place.
Rorty endorses a self-sufficient and self-correcting conception of political
deliberation that accepts no authority higher than the consensus we may
reach – thus, he construes any appeal to a transcendent authority (be it God,
reason, or science) as a dubious attempt to escape our freedom and be told
by another what to do with ourselves. Interpreted this way, the privatization
of religion is part of “plugging away at the familiar tasks set for us by the
Enlightenment . . . [namely], getting our fellow citizens to rely less on
tradition, and be more willing to experiment with new customs and
institutions” (RCS 168). Wolterstorff, on the other hand, is skeptical about
the capacity of politics to achieve, all by itself, our highest aspirations – hence,
he resorts to God and scripture to guide (and correct) the political deliber-
ation process, breathing into it a moral life it would otherwise lack. For
Wolterstorff, religious appeals do not betray our freedom – they nudge us to
realize it. Interpreted this way, the privatization of religion does not contrib-
ute toward the goals of the Enlightenment – it undermines those goals.
Instead of the universal spread of freedom, peace, equality, and justice,
which were supposed to ensue from fighting God’s shadows, banning religion
from the public square has resulted in the impoverishment of our public
discourse and the hegemony of selfishness everywhere one turns.
The impasse resulting from these opposing views presents the features of
an antinomy of what, in Kantian spirit, one might call “religious reason” –
the variant of practical reason that plays itself out in religious matters.6
6
Although talk of God runs throughout the first Critique and is the centerpiece of the “Ideal of Pure
Reason,” what I refer to here by “religious reason” is the use that emerges in connection with Kant’s
doctrines of the highest good and radical evil. The “God” invoked here pertains to morality, not to
Kantian antinomies present opposed, yet equally plausible, arguments.
While their clash threatens to tear human reason apart, their resolution
turns on altering the scope of the assumptions upon which the conflicting
parties build their case.7 The effectiveness of this alteration resides in the
fact that, at the root of all antinomies, Kant finds an illegitimate extension of
human reason beyond its proper boundaries. Once reason is made aware of
its own limits, the clash loses its destructive force: although the questions
persist, “since they are given . . . as problems by the nature of reason itself”
(Avii), one can adjudicate between them by circumscribing each line of
argument to its proper domain.
Kant drew the arguments of his antinomies from the history of philosophy.
The assumption driving this procedure is that “reason itself does not operate
instinctively, but rather needs attempts, practice, and instruction in order to
gradually progress from one stage of insight into another” (Idea 8:18). That is,
human reason has a history, driven primarily by false starts and widespread
disagreement, at the end of which comes Kant’s Critical philosophy to offer
the olive branch of peace to all contenders.8 This philosophical self-
conception led Kant to “flatter [himself] that . . . [he had] succeeded in
removing all those errors that have so far put reason into dissension with
itself” (Axii). Thus, Kant would have seen the persistence of the conflict
between advocates of religious privatization and proponents of open religious
entrance into the public square as a sure sign of boundary transgression, as a
symptom of insufficient critical awareness. I suggest we take Kant’s diagnosis
seriously, for the first to suffer the ailment was Kant himself.
3
Our antinomy arises because it is not prima facie clear whether the
Jeffersonian compromise is in itself sufficient to guarantee a thriving and
speculation. In both cases, Kant believes, the idea of God “proceeds entirely from our own reason and
we ourselves make it, whether for the theoretical purpose of explaining to ourselves the purposiveness
in the universe as a whole or also for the purpose of serving as the incentive in our conduct” (MS
6:443–44). But it is in connection with morality that God’s existence gains a real grip in the Kantian
system – and it is this sense of “God” that will occupy us here.
7
I am referring, of course, to the so-called “dynamical” antinomies, for the resolution of “mathematical”
antinomies requires a different strategy − namely, declaring the falseness of both dialectically opposed
assertions (A531/B559). This technical distinction, however, is unimportant for our goals here.
8
For a similar reading of Kant’s Critical philosophy, see Philip J. Rossi, The Social Authority of Reason:
Kant’s Critique, Radical Evil, and The Destiny of Humankind (Albany: State University of New York
Press, 2005), particularly Chs. 1 and 5. Also, see Onora O’Neill, Constructions of Reason: Explorations of
Kant’s Practical Philosophy (Cambridge University Press, 1994), particularly Chs. 1 and 2.
198 PABLO MUCHNIK
peaceful polity. Kant struggled with this problem throughout his philo-
sophical career. One can see him gradually drifting away from a position
closer to Rorty’s in texts like What is Enlightenment? (1784), to a position
closer to Wolterstorff’s in texts like Religion (1793). For, during the 1780s,
Kant became increasingly skeptical about his earlier, optimistic claim that
“freedom to make public use of reason in all matters” (WA 8:37) was all that
human progress needed. This claim committed him to a melioristic view of
history, according to which
when nature has unwrapped, from under this hard shell the seed for which
she cares most tenderly, namely, the propensity (Hang) and calling (Beruf) to
think freely, the latter gradually works back upon the mentality of the people
(which thereby gradually becomes capable of freedom in acting) and even-
tually even upon the principles of government, which finds it profitable to
itself to treat the human being, who is now more than a machine, in keeping
with his dignity. (WA 8:41)
Kant was forced to question this gradualist view when he realized that at
the center of that progressive tendency to use reason freely lurked a self-
imposed propensity to evil (Hang zum Bösen), from which not even the best
of us is exempt – a propensity which corrupts rationality at its core and
“throws dust in our own eyes” (R 6:38). This realization made Kant more
wary about the self-sufficiency of the secularizing tendencies of the
Enlightenment, more prepared to admit a dark side to human reason, and
hence more reliant on God’s assistance to remove the dust from our eyes.9
In short, it made Kant more willing to have a porous relation between
religion and politics, rather than the impregnable wall the extreme liberal
view wants us to build. This wall, Kant thought, was necessary to protect us
from two equally mistaken religious attitudes − namely, the attitude he calls
“naturalistic unbelief,” “which combines indifference or, indeed, even
antagonism to all revelation with an otherwise perhaps exemplary conduct
of life” (R 6:119), and the attitude he calls “ritual superstition,” which
requires faith in contingent aspects of a creed as a pre-condition to a good
life-conduct. Kant’s “moral religion” avoided both extremes – and hence
made the rigid liberal wall outlast its function.
To put the point differently, in the run-up to Religion, Kant’s reflections
on radical evil made him draw a wedge between the calling (Beruf) to think
freely, which he identified in the Enlightenment essay with our moral
destiny (Bestimmung), and the propensity (Hang) that was supposed to
9
See Joseph P. Lawrence, “Radical Evil and Kant’s Turn to Religion,” Journal of Value Inquiry 36.2–3
(2002), pp. 319–35.
lead us there. This propensity Kant had linked in the 1780s to the invisible
hand of providence operating in history through the mechanism of unsocial
sociability (Idea 8:20).10 According to this view, the plurality of agents and
points of view would converge, in spite of their selfish and private agendas,
into increasingly more comprehensive political agreements – first at the
level of a single juridical state, then in a whole federation of republican
nations that would put an end to war in international relations. In the 1790s,
however, Kant began to suspect that those self-interested agreements would
remain forever unstable, and, therefore, that the progressive outcome of our
unsocial sociability was not to be trusted: the mechanical play of human
inclinations, abandoned to its own devices, left human beings stranded in
the imminence of continuous conflict, if not in open war.11
Religion, in Kant’s mind, thus became increasingly important to trans-
forming, from the ground up, the comparing and competitive tendencies that
count on self-love to secure a lasting peace. Between the destiny of human
beings and the historical mechanisms that promised its achievement, Kant
now realized, there was an abyss which only faith in God could help us cross.
Given the radicalism of evil, political reform had to be accompanied by a moral
revolution in our mode of thinking (Denkungsart) – and this revolution itself
seemed to require “some supernatural cooperation” (R 6:44), for “[h]ow it is
possible that a naturally evil human being should make himself a good human
being surpasses every concept of ours” (R 6:45). Politics could not deliver
reliable results without religion, because “out of such crooked wood as the
human being is made, nothing entirely straight can be fabricated” (Idea 8:23).
The question for Kant henceforth became: what kind of religion could
support, and even advance, the emancipatory goals of the Enlightenment?
What kind of God is compatible with human autonomy?
4
Before tackling this question, we must first discuss the role evil played in the
design of Kant’s answer. By “evil,” Kant means the form of volition that
underlies culpable wrongdoing. This form results from an inversion of the
moral order of priority between the incentives in the human will: “self-love and
10
The most lucid exposition of the relation between the propensity to evil and unsocial sociability
appears in Allen W. Wood, Kant’s Ethical Thought (Cambridge University Press, 1999), 286–90. I
criticized the naturalizing tendencies of Wood’s reading in Pablo Muchnik, Kant’s Theory of Evil: An
Essay on The Dangers of Self-Love and the Aprioricity of History (Plymouth: Lexington Books, 2009),
particularly Ch. 2.
11
See Howard Williams, Kant’s Political Philosophy (Oxford: Basic Blackwell, 1983), pp. 244–71), and
Paul Guyer, “Nature, Morality, and the Possibility of Peace,” in Guyer, Kant on Freedom, Law and
Happiness (Cambridge University Press, 2000), pp. 408–34.
200 PABLO MUCHNIK
their inclinations [are made] the condition of compliance with the moral law –
whereas it is the latter that, as the supreme condition of the satisfaction of the
former, should have been incorporated into the universal maxim of the power
of choice as the sole incentive” (R 6:36).
What is most disconcerting about the Kantian view is that evil is
compatible with good conduct:
So far as the agreement of actions with the law goes . . . there is no difference
(or at least there ought to be none) between a human being of good
morals . . . and a morally good human being . . . except that the actions of
the former do not always have, perhaps never have, the law as their sole and
supreme incentive, whereas those of the latter always do. (R 6:30)
Just as the good will is good not for “what it effects or accomplishes”
(G 4:394), the immorality of evil does not consist in its empirical manifes-
tations (violence, cruelty, greed, etc.). Evil is “an invisible enemy . . . who
hides behind reason and [is] hence all the more dangerous” (R 6:57).
Kant reached the conclusion about the invisibility of evil by reflecting on
the pattern of human development throughout history.12 This pattern
reveals a tendency in inclinations to overcome themselves and produce
concord by means of discord (Idea 8:20). Since unlawful inclinations are
patently destructive, Kant thought, they force human beings to submit
them to rational discipline. Nature thus comes to the aid of reason,
“precisely through those self-seeking inclinations, so that it is a matter
only . . . of arranging those forces of nature in opposition to one another
in such a way that one checks the destructive effect of the other or cancels it”
(ZeF 8:366). The value of nature’s aid, however, is ambiguous. For, Kant
realized, although nature can make us good citizens, it allows us to remain
evil people: nature teaches us to hide, behind a semblance of good conduct,
our unwillingness to place the moral law as “sole and supreme incentive”
(R 6:30). Having learned to channel unsociable desires toward socially
permissible goals, our reason has also managed to hide those motives
under the cover of politeness and respectability.
Kant’s expectation was that, by “checking the outbreak of unlawful
inclinations,” the legal order would gradually generate the conditions for
“the development of the moral predisposition to immediate respect for
right” (ZeF 8:376n). But, in the 1790s, he came to admit that legal and
political progress only represented “a great step . . . toward morality
12
I argued for this view in Muchnik, Kant’s Theory of Evil, introduction and Ch. 4. What I present here
is a snapshot of that argument.
(though . . . not yet a moral step)” (ibid.). The protection of the state
deprives agents of excuses for further violence, but since this arrangement
concerns only the externality of actions and leaves the “crooked wood”
intact, it does not amount to moral improvement. The illusion of virtue
(Tugendschein) reaches no further than the power of coercion − hence, it
cannot guarantee that a change in the balance of forces would not again
reveal the deleterious effects of our “invisible enemy”: “We are cultivated in
high degree by art and science. We are civilized, perhaps to the point of
being overburdened, by all sorts of social decorum and propriety. But very
much is still lacking before we can be held to be already moralized” (Idea
8:26). To the extent that our antagonism leads us to create a lawful civil
order out of the very self-love that gives rise to it (Idea 8:21), Kant con-
cluded, the moral battle merely begins with political victory. Although
pacification is attained through the power of the state, the value of such
achievement is uncertain, because it allows a “nation of intelligent devils” to
remain as evil as they were in the state of nature:
The problem of establishing a state, no matter how hard it may sound, is
soluble even for a nation of devils (if only they have understanding) and goes
like this: ”Given a multitude of rational beings all of whom need universal
laws for their preservation but each of whom is inclined covertly to exempt
himself from them, so to order this multitude and establish their constitution
that, although in their private dispositions they strive against one another,
these yet so check one another that in their public conduct the result is the
same as if they had no such evil dispositions.” Such a problem must be
soluble. For the problem is not the moral improvement of human beings but
only the mechanism of nature, and what the task requires one to know is
how this can be put to use in human beings in order so to arrange the conflict
of their unpeaceable dispositions within a people that they themselves have
to constrain one another to submit to coercive law and so bring about a
condition of peace in which laws have force. (ZeF 8:366)
This type of peace, however, amounts to no more than a truce, a temporary
cessation of hostilities. It thus neglects a fundamental Hobbesian principle:
War consisteth not in battle only, or the act of fighting, but in a tract of time
wherein the will to contend by battle is sufficiently known. And therefore, the
notion of time is to be considered in the nature of war, as it is in the nature of
weather. For as the nature of foul weather lieth not in a shower or two of
rain . . . , so the nature of war consisteth not in actual fighting, but in the known
disposition thereto during all the time there is no assurance to the contrary.13
13
Thomas Hobbes, Leviathan, ed. Edwin Curley (Indianapolis: Hackett Publishing Company, 1994),
Ch. xiii, 8.
202 PABLO MUCHNIK
Unlike Hobbes, however, Kant thought that the assurance of peace could not
be merely political. The Leviathan, wielding in one hand the public sword and
in the other the religious scepter, represents the temptation to eliminate the
difference between external and internal freedom. This gives rise to the most
dangerous form of despotism, because coercion through the public sword
ranges over the legality of actions but cannot determine their morality (MS
6:220). Juridical and ethical laws have distinctive spheres of influence.
Hobbes’ sovereign, however, tries to ignore this distinction, no matter how
ethically minimalist the political intervention on the religious sphere turns out
to be. For Kant, this intervention represents an invasion of our “internal
freedom,” where protection from external constraint opens up a space for
setting ends that give content to our moral vocabulary. As Kant puts it in the
Metaphysics of Morals: “I can indeed be constrained by others to perform
actions that are directed as means to an end, but I can never be constrained by
others to have an end: only I myself can make something my end” (MS 6:381).14
Since coercion to have ends is self-contradictory, Kant recognized that the
effectiveness of politics in pacifying a multitude of self-seeking individuals
depends on a source that is not itself political. For, unless peace is adopted as
an end in itself by each one of the intelligent demons, their belligerent
disposition is simply awaiting the chance to show its destructiveness again.
Such destructiveness could only be prevented if morality were to provide the
incentive to follow the law. But this is precisely what juridical laws cannot
possibly do, for they do not determine our motivation at all (MS 6:219).
Although politics may manage to transmute private vices into public
virtues, each of the demons remains in his conscience as inclined to dodge
the law as if he were still in the state of nature. Thus, the “unpeaceable
disposition” which led the demons to form a political compact marks, at the
same time, the limit of their political bond: although coercion can prevent
the outbursts of hostility, it cannot impose peace as an end in itself.
Peace could only result from the coordinated exercise of the use of internal
freedom by individual agents. This coordination, however, creates a reality
irreducible to the aggregative sum of their particular wills − for, once it is
adopted through separate and discrete acts of volition, peace becomes a shared
end. The pursuit of this kind of end represents a new type of virtue, different
in kind from the virtues Kant discusses in the Metaphysics of Morals.15
14
See Rossi, The Social Authority of Reason, pp. 101–4.
15
I take this to be the gist of Nancy Sherman’s interpretation of “common pursuits.” I part ways from
her analysis, because she does not seem to think that religion is essential to achieving a common end.
See Sherman, “The Virtues of Common Pursuit.”
For, while one’s own perfection (physical or moral) and the happiness of
others (MS 6:385) are ends whose adoption depends exclusively on the use of
my freedom, the end of peace is relatively independent of my will. As an end
we hold together, it is not totally my own – and this imposes constraints on
my freedom different from the self-constraint which I had to exert, as a single
individual, in order to adopt the end in the first place. In the “we” constituted
by holding a common end, there is more than the autonomous “I”s which have
made it possible: there are other end-setting selves whose freedom directly
affects mine, for the pursuit of the end in question depends on their will no
less than on mine. This mutuality of pursuit goes deeper than the standard
Kantian demand of non-instrumentalization: the requirement that another
may contain in herself my end (G 4:429–30) is expanded here to include her
adoption of an end that is ours – an end that can exist as neither hers nor mine.
Non-instrumentalization leaves our wills separate; mutual pursuits bind them
together in a single fate. To distinguish this type of end from the individualist
ends that depend on my freedom alone, let me call the obligation to adopt the
end of peaceful union a “sociable virtue.”
In order to adopt such a sociable virtue, the parties must have overcome
the effects of unsocial sociability in their respective wills, i.e., they must have
dislodged from their moral dispositions the vices of “ambition, tyranny, and
greed” which heretofore had made them mutually dependent but equally
hateful (Idea 8:21). Kant describes the situation thus:
He is poor (or considers himself so) only to the extent that he is anxious that
other human beings will consider him poor and will despise him for it. Envy,
addiction to power, avarice, and the malignant inclinations associated with
these, assail his nature, which on its own is undemanding (genügsame), as
soon as he is among human beings. Nor is it necessary to assume that these are
sunk into evil and are examples that lead him astray: it suffices that they are
there, that they surround him, and that they are human beings, and they will
mutually corrupt each other’s moral disposition and make one another evil.
If no means could be found to establish a union which has for its end the
prevention of this evil and the promotion of the good in the human being . . .
however much the individual human being might do to escape from the
dominion of evil, he would still be held in incessant danger of relapsing.
(R 6:93–94)16
16
Kant’s claim that the isolated individual has an “undemanding” nature, Jeanine Grenberg has rightly
pointed out, must be interpreted as part of a strategy of self-deception and rationalization character-
istic of radical evil. See Jeanine Grenberg, “Social Dimensions of Immanuel Kant’s Conception of
Radical Evil,” in Sharon Anderson-Gold and Pablo Muchnik (eds.), Kant’s Anatomy of Evil
204 PABLO MUCHNIK
That the pervasiveness of corruption does not require people to be “sunk into
evil” makes sense in light of our prior discussion: after the demons have
abandoned the “juridical state of nature” and established a sovereign, their
vices do not really disappear – they are concealed behind the appearance of
good conduct. Holding the monopoly of coercion, the state creates, at most,
the conditions for securing “freedom from the dominion (Herrschaft) of evil,”
but such freedom leaves citizens still “exposed to the assaults (Angriffen) of the
evil principle” (R 6:93). Kant calls this ambiguous situation the “ethical state
of nature” (R 6:95). Like in its political counterpart, the ethical state of nature
is one in which “each individual prescribes the law to himself,” does not
recognize an “effective public authority,” and “is his own judge” (R 6:95):
“Just as the juridical state of nature is a state of war of every human being
against every other . . . so is the ethical state of nature a public feuding between
the principles of virtue and the state of inner immorality which the natural
human being ought to endeavor to leave behind as soon as possible” (R 6:97).
The public sword is of no assistance at this juncture, for external legislation
has no “dominion over minds” (R 6:95) and cannot transform our moral
disposition: “woe to the legislator who would want to bring about through
coercion a polity directed to ethical ends! For he would thereby not only
achieve the very opposite of ethical ends, but also undermine his political ends
and render them insecure” (R 6:96). It is impossible to impose virtue by
force – only the individual can, of her own accord, make the end of peace an
end for herself.
Thus, Kant recognizes the need to keep apart the two kinds of lawgiving
to which a legislator might want to resort: the juridical and the ethical – but
the separation does not proceed along the lines of the Jeffersonian com-
promise. “Justice” and “virtue” do not divide in the same way as “public”
and “private” do in liberalism. The Kantian sphere of internal freedom
cannot be completely privatized. “What we do with our solitude” might
(under some libertarian interpretation) comprise permissible ends, and
perhaps even include the obligatory ends that depend on my freedom
alone.17 But the ends associated with shared undertakings directly affect
our collective project of building a common moral world. It cannot hence
be a matter of indifference whether my neighbor believes in “twenty Gods
or no God” at all − for, where my neighbor places her ultimate conception
of the good (i.e., the meanings that govern what she does with her freedom)
affects the way she goes about her more prosaic political business, how far
17
“What we do with our solitude” is a phrase Rorty takes from Whitehead. He uses it to describe the
eminently private role religion should play in a liberal polity (see RCS 169).
she is willing to uphold justice and respect the rule of law. To the extent that
the use of my internal freedom in adopting common ends depends on others,
sociable virtues have an inextricably “public” dimension. Peace is an end I
cannot sustain alone – mutual dependence and cooperation are essential to
the endeavor. Thus, like some contemporary liberals, Kant leaves the choice
of forming “a union which has for its end the prevention of evil and the
promotion of the good” (R 6:94) up to the individual agent. Yet, unlike
those thinkers, Kant refuses to consider the choice “private” in the liberal
sense of the term. Liberalism, in Kant’s mind, slices agents through the
wrong joints: even if, as a demon, my indifference to common ends does not
ensue in any harm to others, the moral disposition it conceals is nonetheless
a matter of public concern, for my attitude undermines the collective
chances of establishing an ethical community and securing a lasting peace.
It is true that the ethical community “can exist in the midst of a political
community and even be made up of all the members of the latter” (R 6:94).
But it cannot possibly be confused with it, since it has a “special unifying
principle of its own (virtue) and hence a form and constitution essentially
distinct from those of the other” (ibid.). The battle for communal virtue
starts where politics ends: since the demons’ commitment to civil laws
remains utterly contingent without sociable virtues, only good persons
can make good citizens. The task of Kant’s religion is thus determined by
the inability of politics to moralize human beings – or, to put it positively,
Kantian religion is designed to account for how the demons could shed their
selfishness and become good people.
5
The contribution of religion in this process is best understood by distin-
guishing between two different levels of moral analysis. First, religion is
supposed to help the single individual with a morally evil disposition to
become a “new man,” acquire a “good heart” (Herz) and be “well-pleasing
to God.” Second, religion is supposed to allow all virtuous persons who have
been released from the dominion of evil to gain victory over the universal
propensity to evil (Hang zum Bösen).18 Victory requires them to unite forces
in an ethical community. This division of labor between units of moral
analysis is reflected in the structure of Kant’s book: the possibility of
18
This group contains the newly converted and those who, in spite of unsocial sociability, have adopted
a good disposition to begin with, but nonetheless remain, as isolated individuals, “in incessant danger
of relapsing into [evil]” (R 6:94).
206 PABLO MUCHNIK
individual conversion is the concern of Religion Part Two, while the

communal project of overcoming evil is the topic of Part Three. I have
argued elsewhere about the importance of this distinction, and will not
rehearse those arguments here.19 What I want to do instead is to discuss
Kant’s views in light of the conflict between advocates and resisters of
religious privatization.
For Kant, “Religion is (subjectively considered) the recognition of all our
duties as divine commands” (R 6:153). This is not to be confused, however,
with a duty toward another being called God – there are no such duties for
Kant (MS 6:443–44). It is, rather, a perspective we are led to assume when
considering the overall effect of our moral endeavors. Were we to recognize
God’s will as the objective ground of obligation − “objective” in the sense of
something external and different from the “subjective” stance we adopt in
observing his commands − complying with those duties would count as no
more than heteronomy on our part.20
Kantian religion is, then, a “subjective matter,” for it is entwined with our
moral disposition (Gesinnung), not with an alien, divine will existing
independently from us. Yet, for all its subjectivity, religion is not something
“private” or “idiosyncratic” in the liberal sense – that is the kernel of truth
Kant would recognize in Wolterstorff’s position. Faith in God for Kant is
not contingent – it manifests a natural predisposition (Naturanlage) in our
reason, deeply connected with our moral aspirations, and which cannot be
eradicated (R 6:111). Such a predisposition is certainly in need of criticism
and restraint. Yet, to want to extirpate it is as pointless as is the attempt to
remove the equally original predispositions that generate metaphysics and
morality. All we can do with these tendencies is to educate them − i.e., guide
them to enter “the secure path of science.”
In the case of traditional religion, this entails purging it from “all
empirical grounds of determination, all statutes that rest on history and
unite human beings provisionally for the promotion of the good through
the intermediary of an ecclesiastical faith” (R 6:121). Such is the kernel of
truth Kant would recognize in Rorty’s position: although morality hallows
all that follows from it, not everything that passes for religious is moral.21
19
See Muchnik, Kant’s Theory of Evil, Chs. 2 and 4.
20
See Allen W. Wood, “Religion, Ethical Community, and the Struggle Against Evil,” in
Charlton Payne and Lucas Thorpe (eds.), Kant and the Concept of Community (University of
Rochester Press, 2011), p. 122.
21
The later Rorty captures this difference by distinguishing between ecclesiastical and faith-based
religion, and excluding only the former from the public square. See Rorty, “Religion in the Public
Square: A Reconsideration.”
For instance, Wolterstorff’s appeal to Psalm 72, with its defense of “the
widows, the orphans, the aliens, and the poor” (RLRR 162) Kant would
consider compatible with our duties and hence a legitimate candidate to
be part of divine commands. On the other hand, those homophobic
passages of Leviticus Rorty so dislikes, or Psalm 59, “where we find a
prayer for revenge that borders on the horrific” (R 6:110n), give us “cause
to consider [the alleged divine statutory law] as spurious, for [they]
contradict a clear duty, whereas that it is itself a divine command can
never be certified sufficiently on empirical evidence to warrant violating
on its account an otherwise established duty” (R 6:100n). In cases like
this, Kant suggests, I must “either fit the passage to those of my moral
principles which stand on their own” or, if this proves to be impossible,
assume that the passage is a historical contingency, not really part of
the rational core of religion (R 6:110n). Kant would wholeheartedly
endorse Rorty’s appeal to privatize or discard such “spurious” claims.
“Because the Bible says so” is a conversation-stopper – but those religious
enthusiasts who hinder the democratic dialogue in liberal societies make
the same mistake that Kant criticized in his own contemporaries. As he
argues in What does it Mean to Orient Oneself in Thinking?, Mendelssohn’s
alleged insight into God’s will and Jacobi’s alleged inspiration
equally undermine the principles of “rational faith” (Vernunftglaube)
(O 8:140–41) – the first by ignoring the boundaries of cognition, the
second by pitting faith against reason. Even more disturbing, Kant
thought, are the views of Johann David Michaelis, who claimed: “The
psalms are inspired; if they pray for revenge, they cannot be wrong: We
should have no holier morality than the Bible” (R 6:110n). The proper
response to ascribing moral infallibility to the Bible is to “ask whether
morality must be interpreted in accordance with the Bible, or the Bible,
on the contrary, in accordance with morality” (R 6:110n).
This question introduces a Copernican turn into religious matters.
Kant’s method here is analogous to the one he had employed in the
Groundwork, where the design of the book followed a twofold move: first,
to analyze the assumptions implicit in common moral consciousness in
order to discover its supreme principle (the regressive argument that
organizes Groundwork I and II), and then to engage in a progressive/
synthetic style of argument that allows us to move from “the examination
of this principle and its sources back to the common cognition in which
we find it used” (G 4:392) (section III). The “two experiments” Kant
refers to in the Preface to the second edition of Religion replicate these
moves:
208 PABLO MUCHNIK
Since, after all, revelation can at least comprise also the pure religion of
reason, whereas, conversely, the latter cannot do the same for what is
historical in revelation, I shall be able to consider the first as a wider sphere
of faith that includes the other, a narrower one, within itself (not as two
circles external to one another but as concentric circles); the philosopher,
as purely a teacher of reason (from mere principles a priori), must keep
within the inner circle and, thereby, also abstract from all experience.
From this standpoint I can also make this second experiment, namely, to
start from some alleged revelation or other and, abstracting from the pure
religion of reason (so far as it constitutes a system of its own), to hold
fragments of this revelation, as a historical system, up to moral concepts,
and see whether it does not lead back to the same pure rational system of
religion from which I have abstracted. The latter, though not from the
theoretical point of view . . . may yet, from the practical point of view, be
independent and sufficient to genuine religion, which, as a rational con-
cept a priori (remaining after everything empirical has been removed), only
obtains in this relation. If this is the case, then we shall be able to say that
between reason and Scripture there is, not only compatibility but also
unity, so that whoever follows the one (under the guidance of moral
concepts) will not fail to come across the other as well. (R 6:12–13)
One must keep in mind, however, two important qualifications: first, the
experience in question in Religion is not moral but religious, and hence the
starting point of Kant’s analysis is “revelation,” not the alleged common
conception we have of a good will. Yet, here, as much as in morality, religious
experience contains empirical impurities and contingencies that threaten
its a-priori core with “all sorts of corruption as long as [it is] without that
clue and supreme norm by which to appraise them correctly” (G 4:390).
Second, the supreme principle Kant discovers to appraise religious experi-
ence is not a competing moral principle, offering an alternative form of
justification to the categorical imperative. For Kant, there is one morality
just as there is one religion (R 6:107–8) – and the former inevitably leads to the
latter, “and through religion it extends itself to the idea of a mighty
moral lawgiver outside the human being, in whose will the ultimate end (of
the creation of the world) is what can and at the same time ought to be the
ultimate human end” (R 6:6). Since, according to Kant, the belief in God
supervenes and completes morality, we can move, in the second experi-
ment, from the supreme moral principle we discovered by abstraction to
the religious consciousness from which we started the analysis. There is thus a
benign circularity in Kant’s Copernican method – the same circularity that
governs all our other rational undertakings, where we find in the objects what
our a priori concepts have put into them (Bxii).
6
The two experiments in Religion, then, are not substantially different from
the experiments Kant performed in speculation and morality, disciplines
which underwent a similar purge from their empirical grounds of determi-
nation. From this purge, pure a-priori metaphysics and morality emerged,
preserving what was rational in their pre-critical ancestors. This means that
an unspoken thought accompanies the title of Kant’s work: to reduce
traditional religion to the bounds of mere reason also implies that critical
religion is not a transitional genre, a concession to superstition that will
eventually vanish once the Enlightenment has done its secularizing job.
Religion, like everything that wants to procure “unfeigned respect” in the
“age of criticism,” must enter the tribunal of reason in order “to withstand
free and public examination” (Axiiin). But it will not vanish after it leaves
the court, because religion expresses a fundamental interest of human
rationality.22 As Kant puts it in the Canon:
All interest of my reason (the speculative as well as the practical) is united in
the following three questions:
1- What can I know?
2- What should I do?
3- What may I hope?
The first question is merely speculative . . . The second question is merely
practical . . . The third question, namely, “If I do what I should, what may I
then hope” is simultaneously practical and theoretical, so that the practical
leads like a clue to reply to the theoretical question and, in its highest form,
the speculative question. (A804/B832 − A 806/B834)
The question of hope pertains to religion and cannot be satisfactorily
answered by either metaphysics or morality, for it involves the problem of
the combination of the causality of nature and the causality of freedom.
These causalities belong to incommensurably different domains in the
Kantian system, and hence cannot be thought of as deriving analytically
from one another (KpV 5:111). While metaphysics can account for the
causality of nature, it assures us only of the possibility of freedom; and
while morality proves the reality of freedom, it is impotent to determine
22
I disagree with Yirmiahu Yovel’s analysis in “Bible Interpretation as Philosophical Praxis: A Study of
Spinoza and Kant,” Journal of the History of Philosophy 11.2 (1973), pp. 189–212. Yovel considers
religion as a ladder to be disposed of as soon as we have climbed to the (moral) top. In my view, hope
raises questions reason cannot pass over in silence but which morality is not prepared to answer. This
secures a distinctive place for religion in Kant’s critical system.
210 PABLO MUCHNIK
the causality of nature.23 This is why, in the dialectic of the Critique of

Practical Reason (1788), Kant introduces the idea of God as a moral creator of
the world, i.e., as a being in whose will the two causal orders are united.
Kant returns to this strategy in Religion. But the idea of God has in the
meantime undergone a significant transformation. In the second Critique,
God served as a postulate necessary to ensure the possibility of achieving “the
exact correspondence of happiness with morality” (KpV 5:125), i.e., the
combination Kant calls the “highest good.” At that time, however, Kant
conceived of the highest good as a goal every virtuous agent is devoted to
bringing about in isolation. This goal required endless moral progress in order
to attain full “conformity of dispositions with the moral law” (i.e., virtue), and
hence entailed the assumption of the immortality of our soul as a second
postulate (KpV 5:122). The God of Religion is no longer exclusively linked to
the moral needs of the individual agent. God becomes the head of the “ethical
community.” In that capacity, He provides a focal point of reference to all
moral agents and relations, allowing them to build a system of mutual support
based on the laws of virtue (R 6:95). This new God assumes an unmistakable
communal role: his ethical laws “unfurl a banner of virtue as rallying point for
all those who love the good” (R 6:94). This banner allows separate agents to
converge under a common goal, binding their wills in a moral undertaking
whose success requires mutuality of pursuit.
Kant conceives of the ethical community as an alternative model of inter-
subjective relation, exempt from the unsociable tendencies ruling all other
aspects of our lives. God’s legislative function is important, because it gives
public sanction, valid for all agents, to the same ethical commands each
individual can find in her own reason. This element of publicity extricates
religion from the privacy of conscience: worship of God allows isolated
individuals to abandon the ethical state of nature, in which each prescribes
the law of virtue to herself and is her own judge (R 6:95). The God they
worship, however, does not express the will of another being. His laws are
anchored in the will of each agent: divine legislation is identical with the
demands of our own reason and compatible with our autonomy. The same
23
In the third Critique (1790), Kant resorted to reflecting judgment to bridge the “incalculable gulf fixed
between the domain of the concept of nature, as the sensible, and the domain of the concept of
freedom, as the supersensible” (KU 5:176). This mediation, however, could satisfy only the reflecting
needs of judgment, not the determining demands of practical reason. See Paul Guyer, “Bridging the
Gulf: Kant’s Project in the Third Critique,” in Graham Bird (ed.), A Companion To Kant (Hoboken,
NJ: Wiley Blackwell, 2009), p. 424, and Kristi Sweet, “The Moral Import of the Critique of
Judgment,” in Pablo Muchnik (ed.), Rethinking Kant. Vol. II (Newcastle upon Tyne: Cambridge
Scholars Press, 2010), pp. 222–37.
duality that explains the emergence of individualist virtues in the
Metaphysics of Morals − namely, the duality between the obligating I (das
verpflichtende Ich) that imposes a duty and the obligated I (das verpflichtete
Ich) who endorses it (MS 6:417) − Kant uses in Religion to explain the
emergence of sociable virtues. In the new context, however, the I which
gives rise to the obligation assumes the guise of God. As Kant puts it:
There must therefore be someone other than the people whom we can
declare the public lawgiver of an ethical community . . . [Someone] with
respect to whom all true duties, hence also the ethical, must be represented as
at the same time his commands; consequently, he must also be one who
knows the heart, in order to penetrate to the most intimate parts of the
disposition of each and every one, and, as must be in every community, give
to each according to the worth of his actions. (R 6:99)
God is the obligating I in its legislative role of sociable virtues – the source
of a “we” that requires more than my agency, but which cannot exist
without my freedom.
God’s public dimension is the appropriate response to what Kant con-
ceives of as a collective moral problem, i.e., the unsocial sociability he
identifies with the propensity to evil. Yet, in direct contrast with traditional
religious views, the Kantian God does not entail a commitment to a tran-
scendent object, independent from our own reason. “Each individual,”
Kant claims, “can recognize by himself, through his own reason, the will
of God which lies at the basis of religion; for the concept of the Divinity
originates solely from the consciousness of [purely moral] laws and from
reason’s need to assume a power capable of procuring for them the full effect
possible in this world in conformity with the moral end” (R 6:104). Only the
god within, and the all-too-human need to assume his existence in order to
fulfill our collective moral destiny, can be compatible with Kantian
autonomy. Only a divine legislator, who gives at the same time (zugleich)
expression to my own will, can be the source of the kind of solidarity
necessary to moralize the demons.
7
We have seen that, like Wolterstorff, Kant refuses to accept the liberal
assumption that religion is a merely private matter, only shared by those
who happen to be so inclined. Properly understood, religion opens up a
space, beyond the political, where agents can pursue peace as a common goal.
Far from disrupting the dialogue, religion is a conversation-trigger: it
212 PABLO MUCHNIK
enables citizens to unite in building a peaceful world they can all share. It
does so, because it contains an essential moral core that underlies and shapes
all forms of ecclesiastical organization: “There is only one (true) religion; but
there can be several kinds of faith. – We can say, further, that in the various
churches divided from one another because of the difference in their kinds
of faith, one and the same true religion can nevertheless be met with” (R
6:108). This is the true religion of reason, hidden within but also structuring
the narrative form of historical revelations. The latter represent “a wider
sphere of faith that includes the other, a narrower one, within itself (not as
two circles external to one another but as concentric circles)” (R 6:12). Just as
Kant discovered the forms of intuition by abstracting from the matter of
sensations (A20/B34), or the categorical imperative by abstracting from the
matter of desires (KpV 5:27), he also discovers rational religion by abstract-
ing from the manifold of religious stories. This is the gist of the “first
experiment” in the Preface (R 6:12) – its importance cannot be under-
estimated: with it, Kant can discard religious dogmas but save religion.
Yet, like Rorty, Kant refuses to accept claims that require faith in God as a
transcendent authority. Such claims are a sign of obsequiousness, imma-
turity (Unmündigkeit), and “superstition” (Aberglaube) on our part. They
signify an attitude of mind (Denkungsart) that construes God like “a great
lord of this world,” revealing thus a kind of false consciousness which turns
religion upside-down. As Kant describes the process, the intention of those
who serve God through “passive obedience” is to perform some
service (Dienst) or other . . . for God . . . It does not enter their heads that,
whenever they fulfill their duties toward human beings (themselves and
others), by that very fact they also conform to God’s commands; hence, that
in all their doings and not doings, so far as they have reference to morality, they
are constantly in the service of God; and that it is absolutely impossible to serve
him more intimately in some other way (for they can act and exercise their
influence on no other than earthly beings, not on God). (R 6:103)
The urge to escape our freedom and be told by another what to do with
ourselves – the transcendent/metaphysical urge which so worries Rorty and
leads him to privatize religion − Kant turns to strictly immanent uses.24
24
This move replicates the strategy Kant used in the first Critique to deflect the speculative use of reason
to strictly regulative purposes, turning the unavoidable, yet self-destructive, search for the uncon-
ditioned, to the service of expanding human cognition. As Kant saw it, the critical project did not
consist in the wholesale destruction of traditional metaphysics, morality, and religion; rather, it
consists in their radical transformation to preserve what is rational in them. This is captured by the
dictum Kant uses to sum up his philosophical enterprise: “I had to deny knowledge in order to make
room for faith” (Bxxx).
The true service to God (not the counterfeit service which is a sure sign of
heteronomy) is always, at the same time, service to human beings: to be
“well-pleasing to God” requires no more than good life-conduct. Kant’s
“second experiment” shows that appeals to faith that undermine human
autonomy do not belong to rational religion – they are “conversation-
stoppers,” hurdles in our moral progress. With this experiment in hand,
Kant can construct a religion that furthers human freedom while avoiding
dogmatism.
As in the rest of his dynamical antinomies, Kant found the solution for
religious reason in limiting the scope of the opposing claims. This limitation
allows him to avoid, with a single stroke, the transcendent tendencies to
base religion in knowledge of God, as well as the atheistic dismissal of what
Kant considers a fundamental need of our reason. I have tried to show how
much both moves owe to Kant’s reflections on radical evil. The Kantian
God is made in the semblance of human reason. His kingdom is thus
protected from the encroachment of the public sword, the assaults of the
bigots, and the doubts of the skeptics. Such making of God is what I loosely
refer to as “Kant’s religious constructivism” in the title of this chapter.
chapter 11
What does his Religion contribute to Kant’s

conception of practical reason?
G. Felicitas Munzel
Morality thus inevitably leads to religion, and through religion it

extends itself to the idea of a mighty moral lawgiver outside the
human being, in whose will the ultimate end (of the creation of the
world) is what can and at the same ought to be the ultimate human
end. (R 6:6 )
If the strictest observance of the moral laws is to be thought of as
the cause of the ushering in of the highest good (as end), then, since
human capacity does not suffice to effect happiness in the world
proportionate to the worthiness to be happy, an omnipotent
moral being must be assumed as ruler of the world, under
whose care this would come about, i.e., morality leads inevitably
to religion. (R 6:7–8n)
A remarkable parallel obtains between these passages from the first Preface
to Kant’s Religion Within the Boundaries of Mere Reason and his discussion in
the Dialectic of Pure Practical Reason in his Critique of Practical Reason.
Kant’s account in this section (which makes up a quarter of the second
Critique) is much more detailed, but for example in one of its summary
paragraphs we read that:
in this manner, through the concept of the highest good as the object and
end of pure practical reason, the moral law leads to religion . . . The moral
law commands us to make the highest possible good in a world the final
object of all our conduct. This I cannot hope to effect except through the
agreement of my will with that of a holy and beneficent Author of the
world . . . Morals is not really the doctrine of how to make ourselves happy
but of how we are to be worthy of happiness. Only if religion is added to it
can the hope arise of someday participating in happiness in proportion as we
endeavored not to be unworthy of it. (KpV 5:129–30)
While commentators have attended seriously both to Kant’s idea of the
highest good and to his Religion, the connection (indeed “inevitable”
connection) between morality and religion is typically not addressed in
interpretations of Kant’s formal moral philosophy, even among more recent
214
What does his Religion contribute to Kant’s practical reason? 215
readings of it as a virtue ethics. Where the emphasis is on objectively
practical reason, on reason as legislative, as author of the moral law, and
so on reason requiring nothing else for the execution of its function, it has
been difficult to understand what role religion or faith (even conceived as
moral or rational faith) can have. It is a problem similar to the question of
how Kant’s account of practical reason can be reconciled with a call for the
education (or cultivation) of reason. After all it is Kant himself who in both
texts (and elsewhere) reiterates that:
on its own behalf morality in no way needs religion (whether objectively, as
regards willing, or subjectively, as regards capability) but is rather self-
sufficient by virtue of pure practical reason. . . . Morality needs absolutely
no material determining ground of the free power of choice, that is no end,
either in order to recognize what duty is or to impel its performance; on the
contrary, when duty is the issue, morality can perfectly well abstract from
ends altogether, and ought to do so. (R 6:3–4)
Thus Kant’s moral philosophy is also incongruous for traditional moral

discourse that begins from an end, especially the good as the end, and seeks
avenues to achieving it, or for a theological perspective that likewise begins
with scriptural teachings as the basis for ascertaining the ethical principles of
a morally good human life.
The story gets complicated further when one also takes into account
Kant’s affirmation in his lectures on anthropology that anthropology, as
knowledge of the nature of humanity, is indispensable for the complete
story of the fulfillment of moral life and judging. “Knowledge of the human
being must be combined” with “morality and religion,” writes Kant, in
order for these to “obtain their final purpose” (VA 25:472). Specifically,
anthropology is a “pragmatic knowledge of what results from our nature”; it
is a “knowledge of humanity” that is not “bound to time and place,” but it
does belong more broadly to and constitutes the “second part of knowledge
of the world” (VA 25:471, 470). In his Religion, Kant states that the purpose
of his “four essays” consists in making “apparent the relation of religion to a
human nature partly laden with good dispositions and partly with evil ones”
(R 6:11) and so affirms here too the connection with anthropology. How
then do all these elements fit together: what we ought to, and therefore can,
do (morality), what we may hope to achieve (religion, faith), and what we
can and actually do (anthropology, history, biography)?
The challenge is to read the Religion in the context of Kant’s unfolding
thought and indeed in its connection with the Critical philosophy as such.
Kant poses the issue from the outset of the Critique of Pure Reason as an
216 G. FELICITAS MUNZEL
“inquiry whose object cannot be indifferent to human nature” and claims

that we must “assume God, freedom, and immortality for the sake of the
necessary practical use of my reason” (Ax−xi, Bxxx). In the Canon of the
first Critique, he turns to: (1) the primacy of practical reason in regard to
these three objects, to the question of “if, now, I do what I ought to do, what
may I then hope?” (A805/B833); (2) the idea of the highest good; and (3) the
notion of a moral world. He concludes again with the affirmation that “in
these questions no human being is free of all interest” (A830/B858). It is a
theme then that pervades Kant’s writings, and in fact occurrences of the
notion of rational faith (Vernunftglaube) are found in more of Kant’s texts
than just Religion.1
The focus in this chapter is on what we learn about Kant’s conception of
practical reason beyond its familiar, formal, objective sense. How are we to
understand the connection between practical reason and faith, the relation
of practical reason to human nature, while still preserving the independence
of morality from the conditions under which and the ends to which it is
concretely exercised in the world? We will begin with this issue of an end,
purpose, or consequence of human choice making. As Kant puts it in his
Religion,
the proposition, ‘Make the highest possible good in this world your own
ultimate end,’ is a synthetic proposition a priori which is introduced by the
moral law itself, and yet through it practical reason reaches beyond the law.
And this is possible because the moral law is taken with reference to the
characteristic, natural to the human being, of having to consider in every
action, besides the law, also an end (this characteristic of the human being
makes him an object of experience). (R 6:7n)
Next it is necessary to clarify the sense of “human nature” that is entailed

in Kant’s account of the relation of morality (or practical reason) and
religion. Kant is explicit about his understanding of human nature in this
regard, both in his Religion and in other texts. Third we will turn to the
notion of an inner judge, conscience as a function of practical reason, or as
Kant puts it in his anthropology lectures, as the Deity’s vicar. Here again,
relevant passages are found both in the Religion and in other writings.
1
There are, of course, three terms involved – Glaube (which may be translated as belief or faith,
depending on the context); moralische Glaube (moral faith); and Vernunftglaube (rational faith) – and
commentators have written on the differences and connections among them. Of Kant’s relatively few
usages of the term, Vernunftglaube (thirty-one according to a concordance check), only eight occur in
the Religion; eleven appear in Kant’s 1786 essay “What Does it Mean to Orient Oneself in Thinking?,”
four are found in the Critique of Practical Reason, four in the Conflict of the Faculties and three in the
Jäsche Logic.
Ultimately, we will see that the relation of practical reason to human nature
means the question of the relation of practical reason to itself in its various
roles. It entails an inner trust relation (a matter of faith) between the
judgments of “ought,” “can,” “hope,” and “do.”
Distinction between basis or ground and consequence

The need for a reference to an end (as spelled out in the passage above from
the Religion) is established by Kant in his Critical philosophy. That the
practical or moral as such includes not only the question of its own basis or
ground objectively speaking, but also what follows from such objective
moral determination, is expressly stated in the Critique of Pure Reason:
“Practical is everything that is possible through freedom” (A800/B828);
“The power of choice that can be determined independently of sensible
impulses and hence through motivating causes that are presented only by
reason is called the free power of choice (freie Willkür) (arbitrium liberum);
and everything connected with this free power of choice, whether as basis
(Grund ) or consequence (Folge), is called practical ” (A802/B830).
Moreover, where the consequence in question is a matter of the end
enjoined by the moral law and the conditions of the possibility of that end,
necessity is entailed. This necessity also relates to the modality of the kind of
faith that is involved. The “word faith” in general, writes Kant, applies “only
to the guidance that an idea gives to me, and to the idea’s subjective
influence on the furtherance of my acts of reason – the furtherance that
keeps me attached to the idea even though I am unable to account for it
from a speculative point of view”; however, in the case of “moral faith,” such
guidance comes with certitude and steadfastness (A827–29/B855–57):
Here there is an absolute necessity that something must occur, namely that I
comply in all points with the moral law. Here the purpose is inescapably
established, and – according to all the insight I have – only a single condition
is possible under which this purpose coheres with the entirety of all purposes
and thereby has practical validity, namely the condition that there is a God
and a future world. (A828/B856)
As Kant reiterates in the Critique of Practical Reason, while the “moral law
is the sole motive of the pure will,” it in turn “alone must be seen as the
ground for making the highest good and its realization or promotion the
object of the pure will” (KpV 5:109). Hence too Kant makes a repeated call
to make the moral concepts “subjectively practical”: the concepts “must not
stop short with the objective laws of morality, to be admired and esteemed
in relation to humanity, but their representation must be considered in

relation to human beings and to the individual” (KpV 5:157). From the
standpoint of individuals and their particular choices, the question of the
end is both unavoidable and necessary.
Kant begins his Religion with this observation and spells out once more
the distinction between basis and consequence. It is worth citing the passage
because it is, on the one hand, a succinct restatement of what one finds in
the first two Critiques, while, on the other hand, it highlights the transition
in emphasis to the natural interest rational human beings take in the ends to
be achieved and to practical reason’s role in that regard.
In the absence of all reference to an end no determination of the will can take
place in human beings at all, since no such determination can occur without
an effect, and its representation, though not as the determining ground of the
power of choice nor as an end that comes first in intention, must nonetheless
be admissible as the consequence of that power’s determination to an end
through the law (finis in consequentiam veniens); without this end, a power of
choice which does not [thus] add to a contemplated action the thought of
either an objectively or subjectively determined object (which it has or should
have), instructed indeed as to how to operate but not as to the whither, can
itself obtain no satisfaction. So morality really has no need of an end for right
conduct; on the contrary, the law that contains the formal condition of the use
of freedom in general suffices to it. Yet an end proceeds from morality just the
same; for it cannot possibly be a matter of indifference to reason how to answer
the question, What is then the result of this right conduct of ours? nor to what we
are to direct our doings or nondoings, even granted this is not fully in our
control, at least as something with which they are to harmonize. And this is
indeed only the idea of an object that unites within itself the formal condition
of all such ends as we ought to have (duty) with everything which is condi-
tional upon ends we [do] have and which conforms to duty (happiness
proportioned to its observance), that is, the idea of highest good in the
world, for whose possibility we must assume a higher, moral, most holy, and
omnipotent being who alone can unite the two elements of this good. This
idea is not (practically considered) an empty one; for it meets our natural need,
which would otherwise be a hindrance to moral resolve, to think for all our
doings and nondoings taken as a whole some sort of ultimate end which reason
can justify. What is most important here, however, is that this idea rises out of
morality and is not its foundation; that it is an end which to make one’s own
already presupposes ethical principles. It cannot be a matter of indifference to
morality, therefore, whether it does or does not fashion for itself the concept of
an ultimate end of all things. (R 6:4–5)
Just as in Kant’s understanding of human reason as such we cannot be

indifferent to the questions of God, freedom, and immortality, so in the
case of moral choices, we cannot be indifferent to what we ought to, can,
and do achieve. It lies in the nature of human practical reasoning to take an
inevitable interest in the end or ends we realize. This turn to ends brings
with it, however, a number of additional issues.
Consideration of the ends of our choices and actions leads in the first
place to the issue of the human agent being able to deliver on ends that reach
beyond the moral law. (We will return to this point in the discussion of the
sense of human nature that is involved here.) It also leads to the issues of an
inner unity of that human nature and of the exercise of reason in its
theoretical and practical functions. It is again in the Canon of the
Critique of Pure Reason that Kant broaches the issue of a systematic unity
of purposes. In its practical or moral use, pure reason contains
principles of the possibility of experience, namely of the experience of such
actions as could be encountered in accordance with moral precepts in the
history of the human being. For since pure reason commands that such
actions ought to occur, they must also be able to occur. And hence a
particular kind of systematic unity must be possible, namely a moral unity.
(A807/B835)
The issue of moral accountability too arises not only with regard to one’s
duties, formally speaking, but with regard to one’s deeds (the realization of
the ends enjoined by the moral law). To conceive of oneself as thus
accountable, writes Kant in his Metaphysics of Morals, is a “moral self-
consciousness,” or “conscientiousness (also called religio),” for “conscience”
is here “thought of as a subjective principle of being rendered accountable
before God for one’s deeds” (MS 6:439).
That such conscientiousness is nonetheless a judgment of practical
reason in regard to its own exercise is a notion on which we will elaborate
in the discussion of conscience. Its role in the interconnection, however, of
morality and religion had already been raised by Kant in his 1775–76
Friedländer Anthropology Lectures. In the course of his reflections on the
character of humanity in general, Kant discusses civil constraint, the con-
straint of propriety, and the “final constraint,” the “constraint of conscience,
where everyone would judge in accordance with his conscience about his
actions” instead of looking to the opinions of others (VA 25:694–95). This
constraint “cannot, however, be achieved without religion, but religion
cannot have any effect without morality; hence religion aims at the highest
perfection of human beings. This would be the domination of conscience,
and since conscience is the Deity’s vicar, this would thus be the kingdom of
God on earth” (VA 25:695). Kant goes on to ask:
Is this state of the perfection of humanity possible, and when is it to be hoped

for? Since the germs for this are actually innate in humanity, it is thus
possible that they will be developed through cultivation and can achieve
perfection. But when is it to be hoped for, and how should it happen, and
what can one do in connection with this in order to bring such about?
Should one begin from the education of children, or from the education of
the entire [political] state? (VA 25:695–96)
It is a question still very much in play in his 1793 Religion.

With this overview, then, of the formal moral condition leading to the
question of moral ends, and then in turn to issues of their realizability, of
human accountability for the same, and of reason’s self-assessment with
regard to the entire enterprise, we can take a closer look at each of these
subsequent components. Before doing so, however, it is worth making a
note about Kant’s use of what seems on the face of it to be religious language
from traditional Christian literature, because the possible source of that
language in the classical literature also bears on the conception of practical
reason operative in his account. The point is not to discount the connection
with Christian concepts, but to widen our understanding of the meaning of
practical reason. Granted, Kant himself begins his Religion (in the Preface to
the second edition) by explaining how “not only compatibility but also
unity” between “reason and Scripture” can be understood (R 6:13). He also
explicitly draws parallels between his moral conceptions and the New
Testament. In his second Critique discussion of ourselves as both legislative
members and subjects in a moral realm, he observes that “the possibility of
such a command as, ‘Love God above all and they neighbor as thyself,’
agrees very well with this. For, as a command it requires esteem for a law
which orders love,” and “to love God means in this sense to do His
commandments gladly, and to love one’s neighbor means to practice all
duties toward him gladly . . . That law of all laws, like every moral prescrip-
tion of the Gospel, thus presents the moral disposition in its complete
perfection” (KpV 5:82–83).
Still, such language as Kant’s frequent reference to Tun und Lassen
(“doings and nondoings,” or “what is done and left undone,” as it appears
in the traditional formulation of the confession of faith) is just the sort of
language that Cicero uses in his account of practical reason, an account with
which Kant was very familiar and which he greatly admired (explicitly
noting that what he means by his notion of duty is Cicero’s sense of officiis
and naming Cicero as an example of the genius who is the model for the
rule). Tun und Lassen is an important phrase, definitive of practical reason’s
activity; as Kant puts it in the Critique of Pure Reason, in reason’s practical
use our “intention is directed to what is to be done and left undone” (A803/
B831). In a similar vein, we find that Cicero, in his De Officiis, writes that
reason “teaches and explains what should be done and what should be left
undone.”2 Reason’s call for the fulfillment of the highest good (the key
concept that leads to Kant’s notion of, and need for, moral or rational faith)
also has its precedence in Cicero, as does the notion that neither happiness
nor moral conduct (the worthiness to be happy) on its own constitutes the
complete human good.3 Cicero’s self-characterization of his project in De
finibus bonorum et malorum includes the question of the highest good as the
paramount “problem of life” treated in the “topics of philosophy”: namely,
the “question raised in these volumes – What is the End, the final and
ultimate aim, which gives the standard for all principles of well-being and of
right conduct?” Cicero’s stated aim for his investigation into “the highest
good and the truest rule in every relationship of life” is a thorough treatment
of “the whole subject of the Ends of Goods and Evils.”4 In the course of it,
he identifies the role of reason as follows: “when reason has been super-
added” in an animal organism having sensation, reason “is placed in such a
position of dominance that all those primary gifts of nature are placed under
its protection. Accordingly Reason never abandons its task of safeguarding
the earlier elements; its business is by controlling these to steer the whole
course of life.”5 As I have shown elsewhere, in Kant’s account of all the
human aptitudes (Naturanlagen) ordered to humanity’s final purpose (the
highest good), it is only under reason’s influence that the other natural
aptitudes are developed to the full realization of their own inherent poten-
tialities.6 Arguably, then, Kant’s account of practical reason is also a recov-
ery of a classical, philosophical understanding of its essential nature and role
in human life. An awareness of this can help us appreciate the philosophical
character (thus “within the boundaries of mere reason”) of what might
otherwise seem to be a foray into theology. What Kant is concerned about is
that there be a unity and not “two religions in one person” in which the
“purely moral religion (the religion of reason)” is separated from the teach-
ings of scripture (R 13).
2
Cicero, De officiis, trans. Walter Miller (Cambridge, MA: Harvard University Press, 1975), I.xxviii,
101.
3
Cicero, De finibus bonorum et malorum, trans. H. Rackham (Cambridge, MA: Harvard University
Press, 1931), IV.xvi.
4
Ibid., I.iv, 11. 5 Ibid., IV.xiv, 38–39.
6
See G. Felicitas Munzel, Kant’s Conception of Moral Character: The “Critical” Link of Morality,
Anthropology, and Reflective Judgment (University of Chicago Press, 1999), pp. 107–26.
Reason, human nature, and obstacles to the fulfillment

of the human moral vocation
The specific hallmark of human nature (entailed in Kant’s discussion of its
role in both the realization of the human moral calling and the hindrances to
the latter) is the free power of choice. As he puts it in the Religion, we are to
understand by “human nature only the subjective ground of the employment
of freedom as such,” which is the “first basis of the adoption” of maxims, a
basis found in our capacity of choice (R 6:21); our resulting state of being is an
“effect (Wirkung) of a free power of choice” (R 6:44). In this text Kant
describes the “entire determination of the human being” or entire set of
purposes making up the whole purpose of human existence, as a three-part
hierarchy under the rubric of the “original aptitude for good in human
nature”: our aptitudes (Anlagen) as (1) living, animal beings (our animality);
(2) living, rational beings (our humanity); and (3) rational, morally account-
able beings (our personality) (R 6:26). Several passages later, he further
qualifies these “elements”: “we are here speaking only of those aptitudes
that relate immediately to the capacity of desire and the exercise of the
power of choice”; in other words, all of these levels are understood in terms
of their relation to the rational capacities of setting purposes and making
choices (Wille and Willkür, R 6:28). It is again not the first time Kant has made
these distinctions. In the Critique of Practical Reason he locates the root or
origin of duty in
personality, the freedom and independence from the mechanism of nature
regarded as a capacity of a being subject to special laws (pure practical laws
given by its own reason), so that the person belonging to the world of sense is
subject to his own personality so far as he belongs to the intelligible world.
For it is then not to be wondered at that man, as belonging to two worlds,
must regard his own being in relation to his second and higher vocation
(Bestimmung) with reverence, and the laws of this vocation with the deepest
respect. (KpV 5:86–87)7
With regard to the conditions of the possibility of these aptitudes, Kant
writes in the Religion that the
first does not have reason at its root at all; that the second is rooted in a reason
which is indeed practical, but only as subservient to other incentives; and
that the third alone is rooted in reason practical of itself, i.e. in reason
7
For the importance and meaning of the notion of Bestimmung, both in Kant’s writings and in the
wider eighteenth-century context, see G. Felicitas Munzel, Kant’s Conception of Pedagogy: Toward
Education for Freedom (Evanston, IL: Northwestern University Press, 2012), pp. 70–81.
legislating unconditionally. All these aptitudes in the human being are not
only (negatively) good (they do not resist the moral law) but they are also
predispositions to the good (they demand compliance with it).
“They are original,” meaning that they “belong with necessity to the
possibility of this being,” to the “possibility of human nature” (R 6:28).
Reason’s role in relation to these levels is both to attend to their specific
interests and to maintain the requisite subordination of lower to higher so
that the moral vocation of the entire being can be realized. As Kant puts it in
the Critique of Practical Reason:
man is a being of needs, so far as he belongs to the world of sense, and to this
extent his reason certainly has an inescapable responsibility from the side of
his sensuous nature to attend to its interest and to form practical maxims
with a view to the happiness of this and, where possible, of a future life. But
still he is not so completely animal as to be indifferent to everything reason
says on its own . . . He needs reason, to consider at all times his weal and woe.
But he has reason for a yet higher purpose, namely to consider also what is in
itself good or evil, which pure and sensuously disinterested reason alone can
judge, and furthermore, to distinguish this estimation from a sensuous
estimation and to make the former the supreme condition of the latter.
(KpV 5:61–62)
This hierarchy is, of course, familiar to the reader of Kant’s moral
philosophy, but the important point to underscore is the definition of all
aspects of human nature with reference to reason and specifically the power
of choice. Thereby, all aspects of human nature and human interest are
connected to and brought under the umbrella of the human moral vocation.
Indeed, even the essence of self-preservation is redefined from a natural to a
moral sense. If we are to talk of it at all, writes Kant in the Critique of
Judgment, it is “a self-preservation of a completely different kind” from its
conventional meaning, namely that of the individual and the species as
moral beings (KU 5:261). In the defining context of the relation to reason
and our moral vocation, the obstacles that arise in regard to the latter are also
defined in relation to reason, in relation to its judgments of “ought,” “can,”
“hope,” and “do.” This point is important, in turn (as we will see), for
reason’s authorship of and appeal to faith in addressing them. The appeal is
made in the name of the self-preservation of reason itself, its effort to achieve
unity in its judgments.
The pivotal issue is whether and how the human rational subject can
deliver on what is expected, on the “ought.” It is an issue that Kant spells out
explicitly in his anthropology lectures in connection with his claim about
the need for knowledge of human nature. For example, in the Friedländer
Lectures we read that “the human being however, the subject, must be
studied whether he can even fulfill what we require that he ought to do.
Lack of knowledge of human beings is the reason that morality and
sermons, which are full of admonitions of which we never tire, have little
effect. Morality must be combined with knowledge of humanity” (VA
25:471–72). The issue in all of its facets is understood by Kant in a more
complex way than the conventional opposition of reason and the appetites
that pervades so much of moral philosophy. Ultimately it centers on reason
being in danger of being a house divided upon itself. Parts of the problem
relate to the inherent nature of reason and parts relate to the conditions in
relation to and under which it must judge as a human rational power. We
will give an overview of the problems that are encountered and then turn to
the role that faith plays in reason’s effort to retain an inner unity and to hold
steadfast to the striving to realize the human moral vocation. Many of the
problems are identified already in the Critical philosophy, as is the turn to
rational faith as the avenue for addressing them (and so, in recounting them,
we will draw on all the texts in which they appear). When read in this
context, the integral relation of Kant’s Religion to the Critical philosophy
also becomes manifest.
That reason is a capacity that in general requires and seeks the totality
of conditions is a familiar characteristic from Kant’s Critique of Pure
Reason. He returns to this point and its consequences in the Critique of
Practical Reason: “In both its speculative and its practical employment
pure reason always has its dialectic, for it demands the absolute totality
of conditions for a given conditioned thing, and this can be reached only
in things in themselves” (KpV 5:106–7). So, too, as “pure practical
reason,” it seeks the “unconditioned totality of the object of pure practical
reason, under the name of the highest good” (KpV 5:108). Such a unity of
complete happiness and perfect virtue as a goal, end, consequence to the
moral law to be realized by human rational subjects runs into trouble in a
number of ways. In his Religion Kant identifies the corresponding moral
comportment of mind to such an unconditioned totality as one com-
mensurate with the archetype of moral perfection, namely the Son of God
(R 6:61, 74). In the section “Difficulties That Stand in the Way of the
Reality of this Idea, and their Solution,” he addresses the problem of
reconciling this ideal of holiness with the effect which human beings are
able to bring about, an effect that is to be nothing less than their conduct
of life (Lebenswandel) as the deed that corresponds to this idea as the
“good in appearance” (R 6:66–78). The “first difficulty,” Kant writes here,
is that
the distance between the goodness which we ought to effect in ourselves and
the evil from which we start is, however infinite, and, so far as the deed is
concerned – i.e. the conformity of the conduct of one’s life to the holiness of
the law – it is not exhaustible in any time. Nevertheless, the human being’s
moral constitution ought to agree with this holiness. (R 6:66)
Later in the Religion, Kant returns to the point that,

since by himself the human being cannot realize the idea of the supreme
good inseparably bound up with the pure moral disposition, either with
respect to the happiness which is part of that good or with respect to the
union of the human beings necessary to the fulfillment of the end, and yet
there is also in him the duty to promote the idea, he finds himself driven to
believe in the cooperation or the management of a moral ruler of the world,
through which alone this end is possible. (R 6:139)
This passage echoes the conclusions in the Critique of Practical Reason. In

his identification there (in the discussion of the antinomy of practical
reason) of the immediate difficulties, the conflict between the judgments
of “ought” and “can” are readily evident: “No necessary connection, suffi-
cient to the highest good, between happiness and virtue in the world can be
expected from the most meticulous observance of the moral law” (KpV
5:113). “The possibility of such a connection of the conditioned with its
condition belongs wholly to the supersensible relations of things and cannot
be given under the laws of the world of sense,” but “the actions which are
devoted to realizing the highest good, do belong to this world” (KpV 5:119).
Given that the human rational subject is not “nature’s cause, his will cannot
by its own strength bring nature, as it touches on his happiness, into perfect
harmony with his practical principles” (KpV 5:124). Moreover, the “moral
law does not of itself promise happiness, for happiness is not, according to
concepts of any order of nature, necessarily connected with obedience to the
law” (KpV 5:128). As is well known, the resolution to this and other
difficulties (for example, the duration of existence required to achieve
perfect virtue) are the postulates of the existence of God and immortality
of the soul. As Kant notes, the “righteous man” will say that he “stands by
this and will not give up this belief, for this is the only case where my interest
inevitably determines my judgment because I will not yield anything of this
interest” (KpV 5:143). The interest at stake is the interest of reason, not only
in the postulates as such, but in securing the inner unity between what
ought to happen (what the moral law prescribes) and what lies in the power
of the human rational agent to realize. Kant is clear about what the
alternative would be: “If, therefore, the highest good is impossible according
to practical rules, then the moral law which commands that it be furthered
must be fantastic, directed to empty imaginary ends, and consequently
inherently false” (KpV 5:114).
The difficulties do not end with completing the logical move, the
affirmation of the conditions of the possibility of the highest good. There
is the on-going issue of sustaining the steadfastness of the moral conduct of
thought (Denkungsart) and moral comportment of mind (Gesinnung). It is
the “second difficulty” Kant identifies in his discussion in the Religion: it has
to do with “the assurance of the reality and constancy of a disposition that
always advances in goodness (and never falters from it)” (R 6:67). To have
merely doctrinal faith, a theoretical cognition of the existence of God as a
condition for the unity of the moral and the natural “has something shaky
about it,” writes Kant in the Critique of Pure Reason; “for the difficulties
encountered in speculation often drive one away from this faith, although
inevitably one always returns to it again” (A827–28/B856). Hence the
conviction needed is “not a logical but a moral certainty”; a “faith in a
God and in another world so interwoven with my moral comportment of
mind that, as little as I am in danger of losing my moral comportment of
mind, so little am I worried that my faith could ever be torn from me”
(A829/B857). Kant returns to this issue in his discussion of teleological
judgment in the Critique of Judgment. Without moral faith, “upon the
contravention of theoretical reason’s demand for a proof (of the possibility
of the objects of morality), the moral conduct of thought” enjoys “no firm
steadfastness and vacillates instead between practical commands and theo-
retical doubts” (KU 5:472). A few lines earlier, Kant identifies faith or belief
(Glaube), considered “not as an act” but as “habitus,” as the “moral conduct
of thought” on the part of “reason in regarding as true what is inaccessible to
theoretical cognition” (KU 5:471). Again Kant is clear on the alternative: if
moral purpose (the highest good) were given up as impossible, it brings with
it the unavoidable impairment of the morally good comportment of mind
(KU 5:452). Faith also addresses the issue of the “intelligibility” or “com-
prehensibility” of the human final purpose conceived as the ideal of the
highest good (KpV 5:126). This conceptual possibility of grasping it as
meaningful in and for the world is a further help in overcoming the
vacillation between theoretical doubt and moral command.
There remain at least two more threats to the realization of the human
moral vocation: the inherent weakness of the power of choice itself − the
propensity for evil − and the sheer fatigue of maintaining moral striving in
the face of the vicissitudes of life. As Kant articulates it in the Religion, the
“root (Grund) of this evil” can lie neither in “human sensibility
(Sinnlichkeit)” and its “inclinations,” nor in “a corruption of morally
legislative reason” (R 6:32–34). To attribute it to the former would be
“too little,” for the level of sensibility is only that of our aptitudes for
animality, while a reason that releases itself from the moral law is too
much: it would make of the subject a “diabolical being” (R 6:34). Thus
neither our inclinations, nor our moral capacity objectively considered − as
author of the moral imperative and our power of practical desire determined
by reason − is at issue. Rather, this propensity is “rooted in” and “inter-
woven with” the “subjective highest basis” of the adoption of “all maxims,”
our power of choice (R 6:32). The core of the issue is as follows. We begin
from an inherent capacity for setting purposes (and so indeed a germ of
good) and thereby from a capacity for orienting all human choice and action
in accordance with our final purpose and vocation. This capacity is at the
same time in itself subject to the propensity to take its guiding principle of
choice from the inclinations; this rooted-in propensity (and so “radical”
evil), the “true enemy” of the good, is a mental state (Gesinnung) so oriented
that when an “inclination entices one to the commission of a transgression”
against the law, one “does not want to resist” the inclination (R 6:58n).
Therefore, that act of resolve, the firm and resolute adoption of the moral
law from a subjective point of view as the unshakable highest maxim of the
free power of choice, is so important, and, in turn, so too is the elimination
of the hindrances to that resolve.8 Nonetheless, Kant names it as the “third
and apparently the greatest difficulty,” namely, “however steadfastly a
human being may have persevered in such a disposition in a life conduct
conformable to it, he nevertheless started from evil, and this is a debt which is
impossible for him to wipe out” (R 6:72). His conclusion ultimately is that
“it is always therefore only a decree of grace when we are relieved of all
responsibility for the sake of this good in which we believe” (R 6:76).
At the outset of his Religion, Kant notes that “we simply cannot do
without” the “combination” of “the purposiveness [deriving] from freedom
and the purposiveness of nature” (R 6:5). In his third Critique discussion of
the teleological judgment that would effect such a combination, Kant
identifies further impairments to the requisite steadfastness. It must also
be ensured that our moral “striving not be seen as utterly futile in its effects”
and that “the danger of it thereby growing weary [or flagging]” in its efforts
be avoided (KU 5:447): “Compelled by the moral law to strive for a
universal highest purpose, we nonetheless feel that we and all of nature
are incapable of achieving it” (KU 5:446). The “flagging” that Kant is
8
For an earlier discussion, see Munzel, Kant’s Conception of Moral Character, pp. 133–83.
ultimately concerned about goes beyond physical fatigue; the issue is

maintaining mental strength, conviction or fortitudo moralis in a lifetime
of effort, in the face of the perceived human inadequacy to fulfill the task.
The threat is again ultimately within − namely, in human rationality (as a
whole) becoming a house divided against itself in relation to the “practical
employment of reason,” to the “feasibility of carrying out” its purpose (KU
5:470–71). In a parallel discussion to the one here in the Critique of
Judgment, in the Logic one finds a description of rational faith, of regarding
as true those conditions under which the highest good is realizable, as the
“casus extraordinarius without which practical reason cannot maintain [or
preserve, erhalten] itself in regard to its necessary purpose, and which stands
it in good stead favor necessitatis in its own judgment.” Practical reason
“can . . . only oppose whatever hinders it in the use of this idea” (L 9:68n).
What can threaten the self-preservation of reason is self-contradiction or per
se impossibility. In this case, the antinomy is not simply a theoretical one;
reason cannot both author an idea and enjoin the human to actualize it as a
purpose and, at the same time, recognize this purpose as “nothing but a
chimera” or pipe dream (Hirngespinst) (KU 5:472). The rational, logical
consequence of holding the latter viewpoint is, as Kant repeats in both the
second and third Critiques, to call the moral law itself into question (KpV
5:114; KU 5:452–43, 471n). “Dogmatic unbelief” is hence completely incom-
patible with the Denkungsart in which the moral maxim prevails, while faith
or belief mixed with doubt raises the obstacle (albeit surmountable) of “lack
of conviction”; to the latter, “critical insight into the limits of speculative
reason” may yet successfully respond on the behalf of practical interest
(KU 5:472).
The flagging in one’s efforts can also arise in another way. Kant recog-
nizes that the thinking person who reflects upon all the ills of humanity is
vulnerable to suffering yet one more kind of conflict, a dissatisfaction
(Unzufriedenheit) with the entire, providential course of the world, a kind
of sorrow Kant calls “inimical to morality,” for it undermines the very
courage and spirit required to face the need for “self improvement”
and thereby to take on “the in all likelihood sole cause of all these ills”
(MA 8:120–21). In the Critique of Judgment, he presents the case of the
individual who, “partly due to the weakness of all the so highly praised
speculative arguments, and partly due to the many irregularities found
[both] in nature and the moral world,” has become persuaded that there
is no God (KU 5:451). “How,” asks Kant, will such individuals “judge their
own inner determination to a purpose,” following as it does “from the moral
law which they actively revere” (KU 5:452)? The individual in question is
one of good character and moral comportment of mind, one who esteems
the good for its own sake and not for personal advantage, and “toward
which the holy law has” in the first place “directed all his powers,” but who
quickly confronts the limits to human efforts and the ills endemic to the
human situation: one can expect, at most, nature’s contingent cooperation
in one’s efforts, while vice abounds on the one hand, and the ills of
deprivation, disease and untimely death visit the righteous with utter
disregard for their virtue, on the other (KU 5:452). In fact, observes Kant,
over the course of human history, once people began to reflect on justice
and injustice (or right and wrong), moral consciousness became manifest −
that is, people inevitably came to the “judgment that in the end it must
make a difference whether a person has acted honestly or deceitfully, fairly
or violently, even if to the end of his life he has received” no visible
corresponding reward or punishment (KU 5:458). From there, reasoning
proceeded to the notion of a supreme cause that rules the world according to
moral laws, a “moral relation” in the governance of the world which Kant
affirms “even the most uncultivated reason can grasp, provided it considers
itself as practical reason” (KU 5:458).
In sum, as Kant reviews the issues in his Religion, human rational agents
recognize that, if and when left solely to the judgment of reason from a
practical point of view, neither the whole of one’s conduct of life as the good
in appearance in time, nor its basis, the finite, human comportment of mind
(whose propensity for evil cannot be extirpated), can be brought by human
effort into complete commensurability with reason’s idea of the highest
good; in regard to this idea rational faith mediates the gulf between “ought”
and “can” by affirming the conditions of the realizability of the idea by such
agents (R 6:67ff., 139ff., 163). This faith thus effectively brings about an
inner trust relationship between objective and subjective practical reason.9
To have faith (or belief) purely and simply speaking, writes Kant, is to have
“trust (Vertrauen) in the attainment of an aim we are duty [bound] to
advance, but for which we lack insight in regard to the possibility of carrying
it out” (KU 5:472). It is a “free assent” (or “regarding as true”) which means
also that the individual who maintains it “cannot compel another” thus to
assent “on the basis of [given] grounds” (KU 5:472; L 9:69 n.; see also A829/
B857). More specifically, it is an inner promissory relationship, as Kant
characterizes it in his Religion Within the Boundaries of Mere Reason, Critique
of Judgment, and Logic. In connection with an explicit discussion of rational
faith, he writes:
9
For an earlier presentation of this interpretation see ibid., pp. 202–15.
Faith (fides) is actually keeping faith (Treue) in an agreement (pacto), or a

subjective mutual confidence (Zutrauen), that the one will keep the promise
[made] to the other . . . According to the analogy, practical reason, so to
speak, is the promisor (Promittent), the good anticipated from the act, the
promised [object] (Promissum), and the human being [is] the recipient of and
respondent to the promise (Promissarius) [the one bound by it and who must
concretely deliver on it]. (L 9:69n)
In the third Critique, as a footnote to his identification of faith or belief
with reason’s moral conduct of thought (in regarding as true what is
inaccessible to theoretical cognition), Kant’s description of what fides
expresses parallels the statement in the Logic; here, too, reason (through
the moral law) is portrayed as making the promise (Verheißung) that the
final purpose is attainable for the human subject (KU 5:47n). Again, in the
Religion, Kant observes that human beings are not only “called to a good
conduct of life (Lebenswandel) by the moral law,” but “find within the
promise” that they will be able to fulfill this calling (R 6:144). Or, as Kant
also puts it in his essay on orienting oneself in thinking, “rational faith” is a
subjective principle, a maxim, or more specifically a “source of judgment”
(O 8:140). Such a “pure rational belief (Vernunftglaube) is thus the guide
(Wegweiser) or compass whereby” not only the “speculative thinker orients
himself,” but also “human beings of common, but (morally) sound reason
are able to map out their path, both in regard to their theoretical and
practical aims, completely commensurate with the entire purpose of their
vocation” (O 8:142).
Conscience
For maintaining the moral journey over a lifetime, reason requires yet one
more subjective principle of direction or orientation, namely conscience.10
As we saw above, in his anthropology lectures Kant already connected
conscience with religion, while in other texts it is held to be of a piece
with practical reason. For example, we read in the Metaphysics of Morals that
“conscience is practical reason” in its role of “holding before human beings
in every case their duty under law, for acquittal or condemnation,” and its
relation is “solely to the subject” (whose “moral feeling” is thus “affected by”
reason’s “act”) (MS 6:400). As an “authority (Gewalt) watching over the
laws within,” conscience is (as it were) “incorporated into the [very] being”
of human nature (MS 6:438). It is thus effectively an “innate judge” of
10
For an earlier discussion of conscience, see ibid., pp. 215–23.
humans over themselves, a “court” before which they stand for sentencing
(MS 6:437ff.; R 6:77, 146n). As such a “presentation of duty,” conscience is
an “original intellectual moral aptitude,” whose “business” is one of “human
beings with themselves” (MS 6:438). Most pertinent for our discussion here
is the exercise of reason, in its function as conscience, of an on-going self-
assessment to ensure that its guiding activity, the bringing of all contem-
plated and executed actions before it for moral judgment, is not allowed to
lapse. It is a role of conscience that Kant defines in his Religion: one “could
define conscience as the power of moral judgment passing judgment on
itself.” That is to say, “conscience does not judge actions as cases that fall
under the law” − this is the work of “reason insofar as it is subjectively
practical. Rather, here reason judges itself, as to whether it has really under-
taken in all diligence such judgment of actions (as to whether they are just or
unjust), and it calls upon individuals to be witnesses for or against them-
selves as to whether or not” such judgment has been exercised (R 6:186, my
emphasis). It is precisely in this “subjective judgment, as to whether or not I
have compared” a given case with “my practical (here judging) reason, for
the purpose” of making the objective judgment as to whether or not the
thing considered is a matter of duty, “that I cannot be mistaken,” writes
Kant in his Metaphysics of Morals (MS 6:401) − for, to be mistaken in that is
“not to have exercised practical judgment at all,” and so, too, “unscrupu-
lousness is not a lack of conscience, but the propensity to ignore such
judgment” (MS 6:401).
Conscience is thus intimately bound up with our moral self-
consciousness and ensuring that its activity not lapse. Its role relates to all
three of our spheres: what we ought to and therefore can do, what we
actually do, and what we may hope to achieve. In his Religion, in consid-
eration of the question of what “at [life’s] end, human beings may either
promise themselves, or have [reason] to fear, based on their conduct of life
(Lebenswandel),” Kant explicitly identifies the outer judge with the inner
one (R 6:76). The “verdict of the future judge” is thought of as one’s “own
awakening conscience together with empirical self-knowledge summoned” to
one’s aid; the basis for passing this judgment must be thought of further as
“having one’s entire life placed before one’s eyes at that time, not merely a
segment of it,” such as “perhaps the last and for oneself the most advanta-
geous part” (R 6:77, emphasis added). Later, in his interpretation of the
three Persons of the Godhead, Kant describes the Holy Spirit as humanity’s
judge who “speaks to our conscience according to the holy law known by
us” and in terms of “our own reckoning” (R 6:140n, 145n). In a parallel
discussion in the Metaphysics of Morals (MS 6:439–40), conscience is said to
“lead inevitably” to the idea of an “ideal person,” to that of “an authorized

judge” who is an “infallible interpreter of hearts.” The idea of the actuality
of “such a supreme being outside of oneself” is “given subjectively by
practical reason which obligates itself to act in keeping with this idea.” In
this light, “conscience” itself is “thought of as a subjective principle of being
rendered accountable before God for one’s deeds”; it is this “concept” which
is “always contained in every moral self-consciousness” and whereby human
beings receive “direction” (or guidance, Leitung).
Conclusion
On the basis of this presentation of the intertwined relation one finds in
Kant’s writings on morality (what we ought to and therefore can do),
religion (what we may hope to achieve), and anthropology (what we can
and actually do), one comes to see a much fuller account of the human
moral vocation than is typically considered. In this regard, Kant’s Religion
serves an integral role in completing this moral account. As we have seen,
the various themes of our discussion are found in the Critical philosophy (in
all three Critiques). Whether or not one is persuaded by Kant’s appeal to
faith, understanding the role of the appeal is central to completing the
account of Kant’s conception of practical reason. In this context, one can
perhaps best understand Kant’s notion of faith in terms of his third Critique
conception of reflective judgment. In fact, in the Religion, Kant notes that
“one could call” such faith “reflective” (R 6:52).
chapter 12
Culture and the limits of practical reason

in Kant’s Religion
Richard Velkley
1
The place of the essays of Religion within the Boundaries of Mere Reason
within Kant’s Critical philosophy cannot be duly appreciated unless one
considers how religion figures centrally in the critical enterprise. For this
purpose a preliminary account of the critical enterprise itself is indispen-
sable. One can start with a statement from near the end of the Critique of
Pure Reason:
From the whole course of our critique we will have been sufficiently
convinced that even though metaphysics cannot be the foundation of
religion, yet it must always remain its bulwark, and that human reason,
which is already dialectical on account of the tendency of its nature, could
never dispense with such a science, which reins it in and, by means of a
scientific and fully illuminating self-knowledge, prevents the devastations
that a lawless speculative reason would otherwise inevitably perpetrate in
both morality and religion. We can therefore be sure that however obstinate
or disdainful they may be who know how to judge a science not in accord
with its nature, but only from its contingent effects, we will always return to
metaphysics as to a beloved from whom we have been estranged, since
reason, because essential ends are at issue here, must work without respite
either for sound insight or for the destruction of good insights that are
already at hand. (A849–50/B877–78)
The fate of metaphysics is inextricably linked to the fates of morality and
religion, in that speculative reason proceeding in “lawless” fashion endan-
gers both the practical realm or “the interests of human beings” and
speculative inquiry itself, to which the philosopher is attached as to a
beloved. The investigation of the “rights of speculative reason” alone can
sever the roots of anti-religious and immoral doctrines (materialism, fatal-
ism, atheism) as well as hold reason back from “enthusiastic” claims to know
the supersensible, which bring it into theoretical conflict with itself
(Bxxxiv). The “natural dialectic of reason” that devastates both practice
233
234 RICHARD VELKLEY
and theory makes the self-criticism of reason the necessary first concern of
philosophy, for reason, if not subjected to criticism, is exposed to self-
destruction.1 The principal lesson of criticism is the existence of a boundary
that divides the striving of free rationality to realize itself according to ideas
of the unconditioned from the determination of the laws of natural phe-
nomena according to the categories of the understanding. Since this
boundary-drawing – the highest legislation of reason – has the character
of free self-restraint, its essence is moral; and Kant says it is with ancient
precedent that he regards moral philosophy as the highest part of philoso-
phy (A840/B868; KpV 5:107–10).
As Kant argues most fully in the Critique of Practical Reason, practical
reason has a constitutive principle within its own sphere, since the pure
moral law is the objective ground of an unconditional determination of the
will. It has no constitutive power beyond that, and the postulations con-
cerning God, freedom, and immortality that reason requires for the sake of
conceiving the realization of the highest practical end of reason are not
theoretical assertions but subjectively necessary assumptions made by prac-
tical agents so as to conceive the possibility of carrying out their duty, and
they admit neither theoretical confirmation nor theoretical disproof. In this
connection, Kant writes of the “primacy of practical reason,” in the sense
that the practical requirements of reason lead to the assertion of positive
claims (concerning God, freedom, and continued existence beyond this life)
that are not theoretically demonstrable. Practical reason has its own sphere,
independent of the theoretical, in which its requirements are realized. But
there is a further aspect of Kant’s position, until recently mostly ignored by
commentators − namely, that the highest practical end of reason has a role
in the account of theoretical reason as well as of practical reason, as the
organizing and justifying end of all uses of reason. In other words, the
practical provides the end in relation to which all uses of reason are oriented.
In the “Canon of Pure Reason” in the first Critique, Kant writes: “What
sort of use can we make of our understanding, even in regard to experience,
if we do not set ends before ourselves? The highest ends, however, are those
of morality, and only pure reason can grant us cognition of these” (A816/
B844). In the same passage, Kant states that reason’s unavoidable propensity
to go beyond the bounds of cognition by means of mere ideas, whereby it
1
See O 8:144–46, where Kant argues that freedom of thinking in the form of lawless “flights of genius”
results in the destruction of civic order and ultimately in the destruction of speculative reason itself.
Kant proposes the orientation of reason by pure practical reason as necessary for the self-preservation of
reason.
Culture and the limits of practical reason 235
falls into the dialectical errors of speculative reason, cannot be explained
solely by the speculative interest in these ideas, which he calls “very small.”
It is far outweighed by reason’s practical interest in them. Accordingly, the
effort to attain knowledge of these ideas for practical purposes is what sets
speculative reason in motion (A797–98/B825–26). Kant puts forth his
critical strictures, such that speculative knowledge of these ideas is shown
to be impossible, but use of them as postulates grounded in knowledge of
our moral end is permissible. But beyond this he also claims that the
practical use of the ideas, as the outcome of the critical inquiry, is what
ultimately justifies that inquiry. More precisely and fully, reason as moral
necessarily gives itself a highest object to be realized, the highest good as a
moral world in which perfect virtue exists in just proportion with happiness
− striving toward this goal then brings into play the ideas of God, freedom,
and immortality. More strikingly, in the “Canon,” Kant introduces the
notion which is the germ of the Critique of Judgment, namely that the ideal
goal of realizing the highest good leads to “a special kind of systematic unity,
namely, the moral” (A807/B835) which in turn “leads inexorably to the
purposive unity of all things that constitute this great whole, in accordance
with the universal laws of nature.” In other words, it leads to the idea of a
highest unity which
unifies practical with speculative reason. The world must be represented as
having arisen out of an idea if it is to be in agreement with that use of reason
without which we would hold ourselves unworthy of reason, namely the
moral use, which depends throughout on the idea of the highest good. All
research into nature is thereby directed toward the form of a system of ends,
and becomes, in its fullest extension, physico-theology. (A815–16/B843–44)2
Kant goes on to say that the moral order encompassing nature is a “unity
grounded in the essence of freedom and not contingently founded through
external commands.” Kant already has the elements of the program of the
third Critique wherein pure morality grounds the approach to the totality of
reason, both theoretical and practical, as reason seeks out signs that the
natural order is purposively directed toward the promotion of the end of
pure practical reason. Kant certainly denies that the practical interest can
2
Philip Rossi, following Susan Neiman, views Kant’s concern with system as incompatible with Kant’s
account of the fracture within reason between the rational ideal and practical realization. But Kant’s
notion of system does not abandon this distinction, since it still maintains the theoretical limits of
reason and the primacy of the practical, in which the practical gives orientation to all uses of reason.
See Philip J. Rossi, “Kant’s ‘Metaphysics of Permanent Rupture’: Radical Evil and the Unity of
Reason,” in Sharon Anderson-Gold and Pablo Muchnik (eds.), Kant’s Anatomy of Evil (Cambridge
236 RICHARD VELKLEY
justify making constitutive theoretical claims regarding nature’s support of

the moral, and thus his statements in the “Canon” point toward the view of
the purposiveness of nature as a regulative principle, or principle of reflec-
tive judging. Even so, the supreme practical interest leads toward a pro-
gressive stance toward nature in two respects. In the first place, human
beings must strive to bring about a world in which progress toward moral
perfection is realized as much as possible by their own efforts, and thereby
they must strive to achieve the condition wherein perfection is confirmed
and reflected in nature through the corresponding happiness of the self-
perfecting moral agents. This condition must be conceived as an endless
project for the striving of the human species as a whole. In the second place,
the inquiry into nature is guided by the practical interest in uncovering
evidence of natural assistance for human efforts, not in the sense that hope
for the natural satisfactions of happiness motivate the inquiries but, on the
contrary, in the sense that the inquiry seeks evidence that the human is
amenable to a reform of its character as both sensible and rational, in which
the inclinations become more governable by the pure moral disposition.3
Kant regards the capacity for disinterested pleasure in aesthetic judging as
one crucial form of evidence of such natural purposiveness in the human
character.
The primary practical interest of reason gives birth to a way of relating to
the world-whole as a realm of moral projects. Alternatively one might say
that reason’s essence is to be world-transformative, such that the realization
of moral projects expresses its deepest need, far outweighing purely theo-
retical interests. Furthermore, in pursuit of its deepest interest, reason
reflects on the world-whole, encompassing both freedom and nature, as
originating in an idea as its unifying ground. One must again underline that
this idea, or the supersensible substrate as Kant calls it in the third Critique,
is not a possible object of theoretical cognition for Kant, but it serves a
necessary regulative function.
Kant’s view of reason’s highest interest in purposive unity grounded in
freedom relates directly to some of his definitions of philosophy and the
philosopher. Kant writes: “Philosophy is the idea of a complete wisdom,
which shows us the final ends of human reason”; and similarly he describes
philosophy as “the science of the highest maxims of the use of our reason”
(L 9:24). That the final end is determined from a moral standpoint is very
3
For the best account of the acquisition and development of character in Kant, see G. Felicitas Munzel,
Kant’s Conception of Moral Character: The “Critical” Link of Morality, Anthropology and Reflective
Judgment (University of Chicago Press, 1999).
clear from an extended passage defining the philosopher in the Critique of
Pure Reason. In this context Kant is discussing the Weltbegriff (cosmopolitan
or world concept) of philosophy, which he distinguishes from the
Schulbegriff (scholastic concept), which has no end in view other than the
logical perfection of knowledge as a system of cognition (A839/B867). Kant
says one can refer to the ancient idea of the philosopher as first and foremost
a moral philosopher, due to the preeminence which moral philosophy has
over other uses of reason, and it is this notion that grounds the Weltbegriff
(A840/B868): “From this point of view philosophy is the science of the
relation of all cognition to the essential ends of human reason (teleologia
rationis humanae) and the philosopher is not an artist of reason but the
legislator of human reason.” The statement does not mean that the philos-
opher produces a legislation simply expressing his own arbitrary will, but
that the philosopher looks to the essential end that reason itself prescribes,
and whose idea “is found in every human reason.” From the standpoint of
that highest end (which one can relate to the highest good and its fulfillment
through the pursuits that realize the purposive unity it requires) the philos-
opher evaluates the roles of the various sciences and uses of reason in
promoting the essential end – an end which concerns all human beings,
and hence philosophy, as promoting it, is called the Weltbegriff of philos-
ophy. “The mathematician, the naturalist, the logician are only artists of
reason, however eminent the former may be in rational cognitions and
however much progress the latter may have made in philosophical cogni-
tions. There is still a teacher in the ideal who controls all of these and uses
them as tools to advance the essential ends of human reason” (A839/B867).
Separated from the third Critique by only a few years, Religion exposes a
remarkable state of affairs. The account of rational faith in this writing has
still the same basic structure as the approaches presented in the three
Critiques: pure morality alone is the valid foundation for religion as a set
of beliefs supporting the aim of realizing the highest good in the world, now
called the “the ethical commonwealth” which gives the highest good
communal embodiment, inherently elusive, as an “invisible church.” But
the notorious reflection on “radical evil” seems to call into question the
intelligibility of the structure of the practical project, insofar as it places end-
setting itself in a problematic light. Thereby philosophy itself, as science of
the highest ends, falls under the shadow of this question-mark. At the same
time, the notions of culture as nature’s support of the project, featured
especially in the third Critique but also in the historical and political essays,
have a precarious and almost nugatory role in Religion. Yet, as I will argue, at
crucial points in Religion the role of poetic figures, the products of “genius”
238 RICHARD VELKLEY
in the terms of the third Critique, underlines the presence of the concerns of
culture.4 Moreover, they point to the thought that philosophy itself can
exist or have a home in the world only because of the persisting gap between
autonomous lawgiving and realization of ends that characterizes the human
as finite being. Paradoxically, then, radical evil has a hidden beneficence as
the condition for metaphysics as the “beloved from whom we have been
estranged” and to which “we will always return.” As I shall argue, the
necessary incompleteness of the practical project is conveyed by poetic
figures coming to the assistance of philosophical reason, in an unexpected
twist in Kant’s conception of culture as nature’s promotion of reason’s
highest end.
2
The problem of ends is announced immediately in the preface to the first
edition. The highest end the human being can have is grounded on reason
as a self-legislative power that conceives for itself the object of its will, as an
end it places before itself in order to formulate the course of right conduct.
As Kant says, in the absence of all reference to an end no determination of
the will can take place, for no such determination can occur without an
effect (R 6:4). To think of the most comprehensive end of conduct is to
think of an object that unites within it the formal condition of all such ends
as we ought to have (namely, duty as compliance with the moral law) with
everything that is conditional upon ends we naturally have, yet in accord-
ance with duty, namely happiness proportioned to observance of duty. But
Kant insists that morality as autonomy in conceiving its duty alone abstracts
from all ends, for the moral law binds the will through the mere form of
universal lawfulness. The will which determines itself by a material ground
ceases to be moral (R 6:3–4). The human will’s need for the idea of a
comprehensive end only reflects the dual character of the human, as needing
to realize moral self-determination in the realm of nature where human
action must achieve the effects of willing.5 Human reason has need of a
particular point of reference for the unification of all ends, and to conceive it
4
See especially Sections 49, 53 and Comment I to Section 57 of the Critique of Judgment (KU 7) on spirit
(Geist) as an animating principle in the soul, found above all in the talent (genius as a natural gift) for
poetry, or in the capacity to produce imaginative presentations (aesthetic ideas) which give sensible
expression to rational ideas of the supersensible.
5
The human will as determining itself under the moral law for a particular material maxim as an end is
acting as a particular will or, one could say, as the particularization (or individuation) of pure practical
the human who honors the moral law asks himself this question: what sort
of world would I create were this in my power, under the guidance of
practical reason (R 6:5)? Kant says the moral agent arrives at this judgment
with complete impartiality, without regard to his own happiness, and in this
way he evinces a need of adding to the thought of his duties an ultimate end.
Thus morality inevitably leads to religion, the idea of a world that finite
moral beings strive to realize but know they cannot bring about wholly
through their own power (R 6:5–6).
Hereby the book on Religion begins by declaring that religion as such is
problematic, insofar as all end-setting is problematic, evincing a limitation
of human reason as finite.6 The limitation is intrinsic to the defect of
historical religion in which humans seek relief from the difficulties of
realizing their highest end by appealing to superhuman powers and their
acts of grace to lighten the human burden. Historical religion accordingly
focuses on practices of praising supposed superhuman powers and inter-
prets virtue as being pleasing to them, and thereby it is a prime manifes-
tation of moral evil, as the subordination of compliance with the pure moral
law to maxims of self-love. The project of overcoming moral evil is insep-
arable from that of transforming historical or ecclesiastical religion into the
religion of pure practical faith. But insofar as humans in this project must be
guided by conceptions of ends and never cease to face challenges to their
actualization, humans remain susceptible to the attraction of deliverance by
higher beings; they remain vulnerable to the appeal of miracles and other
signs and promises of direct participation in supernatural powers.
Moreover, Kant poses in this work the most disturbing questions about
the realizability of the critical rational project itself, even as he forbids any
retreat from the strictest standard of autonomous self-improvement. Thus
Kant’s proposals in his other works of a “hidden plan” of nature to further
the attainment of the highest good seem to fall by the wayside.
Since the task of overcoming evil involves addressing the teleological
tendency of reason, the nature of evil cannot be placed anywhere outside
reason itself. Kant claims that he wants to find a middle ground between the
dark pessimism of the complaint that “the world lieth in evil,” as having
fallen from a Golden Age – a complaint as old as history and the even older
reason. In effect the problem of radical evil is inherent in the human particularization of the pure will,
which reason represents as an event (choice) although clearly there is no possible conceivable will that
as pre-particularized chooses to be subject to the conditions of a particularized (self-loving) will.
6
See Susan Shell’s account of the central problem of Religion: the moral agent must assume responsi-
bility for the natural flaw in human reason, its need to represent an end, but at the same time human
beings can never make present to themselves whether they have made progress in this: Susan Shell,
Kant and the Limits of Autonomy (Cambridge, MA: Harvard University Press, 2009), Ch. 5.
240 RICHARD VELKLEY
art of poetic fiction (Dichtkunst) – and the “heroic opinion” found only
among philosophers (Seneca and Rousseau are cited) that the world pro-
gresses from bad to better since the natural foundation of goodness in the
human being permits cultivation (R 6:19–20). Experience supports neither
position, and in fact the impossibility of observing maxims (as contrasted
with lawful and unlawful actions) means that no judgment of human evil
is reliably based on experience (but see R 6:32). Kant then makes a risky
a priori inference from consciously evil actions to an underlying evil maxim
as the subjective ground of all particular evil maxims. The underlying
propensity to evil can be called innate in the sense that it is “posited as
the ground antecedent to every use of freedom given in experience,” but it is
not a ground that determines as a natural impulse, which would contradict
freedom. The human being alone is its author in an act that, however, does
not fall within experience (R 6:21–22). To say that the propensity is innate
does not mean that it has not been earned by the human being who harbors
it; it means that it has not been earned in time (R 6:25). A propensity (Hang)
differs from a predisposition (Anlage) in that it can indeed be innate like a
predisposition, “yet may be represented as not being such: it can rather be
thought of (if it is good) as acquired, or (if evil) as brought by the human
being upon himself” (R 6:29). In its status as “posited” and “represented,”
the propensity to evil has neither empirical confirmation nor transcendental
legitimation.7 It is oddly both contingent with respect to humanity and
“woven into human nature,” “entwined with humanity itself” (R 6:29, 6:30,
6:32). The “middle ground” supports both pessimism and optimism and
provides no definite support, it seems, for inclining one way or the other.
Kant admits that his way is not easy to follow, as he notes that moral
philosophy has commonly supposed that the presence of moral evil in the
human can be “very easily explained” by appeal to the power of sensible
incentives and by the impotence of the incentive of reason (R 6:59n). The
true culprit is “an invisible enemy, one who hides behind reason and hence
all the more dangerous” (R 6:57). The ground of evil is not a natural error
of weakness or failure of will but a positive principle of free choice, one
that is “absolutely inexplicable,” as reason’s ability to become master of
the inclinations to the good and its propensity to hand over its authority
to the inclinations to evil are equally incomprehensible. Philosophers
7
On the lack of proof that the human is by nature evil, see Peter Fenves, Late Kant: Towards Another
Law of the Earth (New York: Routledge, 2003),Ch. 4; also see Pablo Muchnik, “An Alternative Proof
of the Universal Propensity to Evil,” in Anderson-Gold and Muchnik (eds.), Kant’s Anatomy of Evil,
pp. 116–43, for an account of the proof as “quasi-transcendental . . . neither reducible to empirical
observation nor totally severed from it.”
understandably recoil from an explanatory basis “forever shrouded in dark-
ness,” and yet the problem of evil (or freedom) can be seen as the genuine
provocation to philosophy: “For if all the world proceeded in accordance
with the precept of the law, we would say that everything occurred accord-
ing to the order of nature, and no one would even think of inquiring after
the cause” (R 6:59n). The argument of Religion is as much a critique of
earlier philosophy as of historical religion, and it draws on biblical sources in
the correction of philosophy. For when the apostle represented the invisible
enemy as a being outside us (“principalities and powers . . . evil spirits”) the
intent was not to extend cognition to the supersensible but “to make
intuitive, for practical use, the concept of something to us unfathomable”
(R 6:59). More remarkably, Kant praises the Christian representation of
moral goodness and moral evil in terms of the distinction between heaven
and hell, which is “not any the less philosophically correct in meaning” for
being figurative and stirring, and which “though containing an element of
horror, is nonetheless sublime” (R 6:60n). It was not philosophers who first
saw correctly the “immeasurable gap” between good and evil, and the
practically effective way of conveying this truth requires a poetic gift. At
the same time, self-critical philosophy does not mistake the poetic image for
supersensible insight, and while facing the unfathomable mystery of the
cause of evil it also remains provoked to inquire about it. Perhaps philos-
ophy is indebted to discoveries of religion for the disclosure of the mystery
of freedom (enabling it to correct a blindness in the Stoics, the ancient
moralists, and their modern heirs), as it must also criticize the tendency of
historical religion to obfuscate the discoveries with literal readings of the
poetic representations as beings outside the human will.
But Religion also clearly pronounces that critical philosophers themselves
will have the tendency to self-obfuscation, as reason as such in striving for
realizable ends will tend to schematize the non-sensible character of the
highest moral end grounded in freedom through sensible representations.
“It is plainly a limitation in human reason, one which is ever inseparable
from it,” that matters of moral worth will be anthropomorphically
expressed, “for we always need a certain analogy with natural being in
order to make supersensible characteristics comprehensible to us.” Thus
proceeds the “philosophic poet” who, as we have seen, can be aware of his
own devices (R 6:65n). Kant here refers to his account in the third Critique
of reason’s need to make the cause of organic life comprehensible to itself by
means of an analogy of an artificer in relation to his work, i.e., through
anthropomorphizing nature by attributing understanding to its causality.
Kant thus refers directly to his own use of poetic schemata as indispensable
242 RICHARD VELKLEY
to the presentation of Critical philosophy.8 His observation can be extended

to all the characterizations of nature as promoting with a “hidden plan” the
rational and moral improvement of the human species, as such anthropo-
morphic−teleological schemata provide a way for human moral agents to
conceive their highest end as actualizable and thereby to entertain consoling
conceptions of the future course of human history. Just as certainly as with
the poetic images of biblical religion, such schemata are misread if their
limits are disregarded. In Religion, where Kant is most thoroughly critical of
conceptions of grace, or the lightening of the human burden of self-
improvement by external powers, and where he is most wary of the con-
sequences of the teleological need of reason, he seems to put himself at
greater distance from the language of nature’s support of human culture.
One might say that Religion proposes a self-critique of the religiosity of the
Critical philosophy, or, rather, it exposes the danger in this religiosity
without abandoning the practical project that gives birth to it.
Furthermore, this takes to a new level the critical reflection on philosophy
itself as understood according to its teleological−cosmopolitan concept.
3
It could be said that Kant engages in another philosophic−poetic use of
analogy to illuminate the highest mysteries when he characterizes the
propensity to evil as the “intelligible deed,” for here he applies the temporal
language of deed to a ground that cannot be temporal. Nothing can be
morally evil except what is our own deed, “yet by the concept of propensity
is understood a subjective determining ground of the power of choice that
precedes every deed, and hence is itself not yet a deed” (R 6:31). The evil
propensity corrupts the very highest maxim in us, which determines
whether the will is disposed toward or against the law. As we have seen,
this propensity must be represented as acquired, and hence as the outcome
of a free deed. On the other hand, this subjective condition corrupting all
actions materially considered, as unfolding in time, cannot itself be a
material−temporal event. Kant says one must take the term “deed” as
having two meanings: one as the non-temporal propensity or “formal
ground” of material deeds, and the other as the material−temporal resistance
8
See Gordon E. Michalson, “Kant, the Bible, and the Recovery from Radical Evil,” in Anderson-Gold
and Muchnik (eds.), Kant’s Anatomy of Evil, pp. 57–73, for Kant’s use of narrative elements from the
Bible as “stand-in” for philosophic reasoning, when he is unable to provide the required philosophic
argument (notably in the motif of “revolution” in moral disposition), and the relation of this use to
“schematism of analogy” in Kant.
to law. He says here, as elsewhere, that the propensity can be called innate,
but adds “because it cannot be eradicated,” for this would require that the
will, following the supreme maxim for the good, perform the eradication.
This is impossible, since the will’s supreme maxim has been corrupted and
there is no higher will above the corrupted will to carry out this conversion.
In effect the highest “deed” has no “doer,” just as it has no temporal
“doing”: “We are just as incapable of assigning a further cause for why
evil has corrupted the very highest maxim in us, though this is our own
deed, as we are for a fundamental property that belongs to our nature.” The
propensity to evil must be regarded both as a fundamental property and as
the effect of a deed. Kant asserts that the intelligible deed is “cognizable
through reason alone apart from any temporal condition,” unlike a deed
that is “sensible, empirical, given in time (factum phenomenon).” But what
kind of cognition is this? It seems to be only an intuitive schematizing of a
ground that Kant elsewhere acknowledges is “absolutely unfathomable.”
Only this schematizing allows us to think of the propensity as the work of
freedom, hence imputable to the will and thus in principle corrigible by the
will. But this schematism is unlike an intuitive−analogic representation of
what is simply unknown, since here the original ground must be conceived
both as unchanging and as the effect of a free will. At one point Kant openly
states the conflict of regarding the propensity both as unchangeable and as
the effect of freedom:
This evil is radical, since it corrupts the ground of all maxims; as natural
propensity, it also is not to be extirpated through human forces, for this could
only happen through good maxims, something that cannot take place if the
subjective supreme ground of all maxims is presupposed to be corrupted. Yet
it must be equally possible to overcome this evil, for it is found in the human
being as acting freely. (R 6:59)
This is nearly a contradiction if not literally one. What is the intent? The
causality of freedom as a power to affect sensibility is unknowable, yet a
representation of it must be posited for practical purposes. It would seem
that avoiding a contradictory positing is a minimal requirement – unless
something is gained, something disclosed, by contradictory positing. This
may be the case if something is learned by trying to carry out the thought of
overcoming the evil propensity and then encountering the impossibility of
wholly doing so. If evil is in fact inextirpable, what remains possible and
worthwhile in the human condition? The effort to improve the human
condition remains a worthy end. Yet this effort to improve, and with it
something else of utmost human importance, remains a possible good for
244 RICHARD VELKLEY
human beings only if the complete overcoming of evil is never achieved.

The other thing is philosophy, for philosophy as the inquiry into the
ultimate end rests on the persisting perplexity of the end. Indeed philosophy
as inquiry into the ultimate end presupposes the finitude of the human as a
divided being: on the one hand aware of an unconditioned good as object of
striving, and on the other limited in its power to achieve that good. Without
that division, all would be “according to the order of nature, and no one
would even think of inquiring after the cause.” Therefore, to think back to
the “intelligible deed” as the non-temporal deed that corrupts the will is to
think back to the necessary condition for philosophy and all human
excellence: humans as rational beings necessarily strive beyond the natural
but cannot complete that striving.9 But why should we think of this
condition as having a beginning in our own will? The simple but naïve
answer is that by thinking of ourselves as the origin of the evil we also
suppose ourselves able to overcome it. But a deeper understanding of the
point is that it underlines that we are both responsible for evil and at the
same time unable to overcome it. For in this case taking responsibility
means affirming the evil propensity (or affirming human finitude) and not
viewing it as merely a natural fate that befalls the human and as something
one can hope could be otherwise – a mere unfortunate contingency. It is to
acknowledge that evil−finitude belongs to the constitution of the human as
human, and that paradoxically the humanity of evil must be affirmed, that
is, evil−finitude is understood as having this essential character of making
the higher endeavors of the human possible.
Thus Kant conceives that the exercise of the higher human rational
possibilities depends on having a religion supporting an unattainable object,
and therefore it is necessary to reject all “easy” resolutions of human efforts
offered by traditional religion or by enlightened philosophical optimism.
Kant calls for courage and good cheer in the dedication to an endless task,
sustained by the hope of continuing progress.10 At the same time, he allows
9
In The End of All Things (8: 334–36), Kant remarks that the cessation of all striving in a state of eternal
tranquility is an idea “very closely related to reason in its practical reference,” although surpassing our
powers of comprehension, since “the representation of an infinite progression toward the final end is
nevertheless at the same time a prospect of an infinite series of ills.” But this idea as equivalent to the
end of all alteration, and with it time itself, must outrage the imagination, “for then the whole of
nature will be rigid and as it were petrified.” Indeed, this condition must spell the annihilation of
thought itself for a being like ourselves, which can become conscious of its existence only in time:
“Thinking contains a reflecting, which can occur only in time.” Practical reason’s demand for
achieving the highest end therefore tends toward a conception that undermines all reason, including
philosophy.
10
See the note at R 6:23, where Kant replies to Schiller’s criticism that Kant’s notion of duty as
unconditionally necessary stands opposed to gracefulness (Anmut). Kant attests that virtue as a firmly
that some readers will more deeply embrace the idea of unattainability,
understanding its beneficial consequences, and be sustained not only by the
thought of progress but also by its endlessness. These will in the full sense
re-enact the “intelligible deed” and affirm evil−finitude as the necessary
condition of the higher uses of reason.
4
It is contrary to prevailing readings of Kant to view him as a philosopher of
the beneficence of evil, and it might even be for some readers a shocking
heresy. Yet one aspect of this reflection in Kant is widely known and
acknowledged as central to his thought, and this is his account of human
unsociable sociability. According to Kant’s hypothesis about a hidden order
in human historical development, the natural inclinations of seeking supe-
riority in honor, wealth, and power over others have the unintended
benefits of furthering the development of the rational powers, disciplining
and refining natural impulses, and forcing human beings into a law-
governed way of life in which they can more freely expand their powers.
This “hidden plan of nature” culminates in a federation of sovereign states
maintaining peace. Such historical developments, Kant claims, lay the basis
for the moral progress of the species in the internal legislation of maxims
that no external order can enforce but which despotic and warlike orders can
hinder. The role that Kant then gives to nature as furthering civilization (the
external and legal advances in rationality) and culture (the discipline and
refinement of natural dispositions leading toward moral autonomy)
describes nature as self-limiting and self-effacing (see Idea, MA, and KU,
especially section 83).11 It could be said that such historical conjectures
present evils relating to natural inclinations as nullifying themselves over
time. But if there is a doctrine of the beneficence of evil in these accounts it
seems a conditional one, in that moral evils can be seen as good only insofar
as they provide the basis for their final overcoming, as human beings acquire
through them the disposition needed for living in full autonomy and
acquiring complete mastery over their natural inclinations.
grounded disposition to fulfill duty gives rise to a courageous and joyous frame of mind, hence to an
aesthetic benefit of virtue, allowing for a graciousness not in duty itself but in its consequences: “The
morally oriented reason (through the imagination) calls sensibility into play.” Kant illustrates his
point with a poetic figure, whereby he exemplifies the virtue described: “Hercules becomes Musagetes
only after subduing monsters, a labor at which those good sisters shrink back in fear and trembling.”
11
For “culture as a site of the self-effacement of nature in its influence on the will,” see Kristi Sweet,
“Kant and the Culture of Discipline: Rethinking the Nature of Nature,” Epoche: A Journal of the
History of Philosophy, 15 (Fall 2010), pp. 121–38.
246 RICHARD VELKLEY
Kant in such accounts does not claim that the natural inclinations them-
selves are evil, but rather that humans acquire the vicious aspects of social life as
rational beings by employing their natural inclinations for purposes not given
by nature. Scholars have commented on the similarity of this account to
Rousseau’s views on human perfectibility and the transformation of amour
de soi into amour propre.12 Kant’s social-historical analyses suggest a speculative
projection that the rational increase of human power (entailing new miseries as
well as new satisfactions) will ultimately result in a moral rationality triumphing
over rationally instituted evils, in an optimistic correction of Rousseau’s lower
estimate of reason. So the question arises whether Kant’s accounts constitute a
social and historical analysis of the ultimate origin of evil. It may seem that
Kant proposes such a genesis in order to envision the social-historical over-
coming of evil. On close examination, one sees that this is not his thought.
In Religion, Kant introduces unsociable sociability under the heading of
predispositions to the good, and describes it as “self-love which is physical and
yet involves comparison (for which reason is required)” (R 6:27). This predis-
position gives rise to the inclination to gain worth in the opinion of others,
which originally means only equal worth, but becomes a desire to acquire
superiority over others which (at some point) takes unjust forms in the
passions of jealousy, envy, ingratitude, etc. Kant asserts that “nature itself
wanted to use the idea of such a competitiveness (which in itself does not
exclude reciprocal love) as only an incentive of culture” and so even the desire
for superiority is, in itself, not a vice. But onto this good inclination there are
“grafted the greatest vices of open or secret hostility” or “vices of culture”
which do not really issue from nature. Reason is active in the first beneficial
form of this predisposition and becomes the corrupting factor. The natural
inclination is not the cause of the corruption. Therefore sociability as such, the
tendency to judge one’s worth only in the opinion of others, is not the cause of
vice, for this tendency is in itself good and has a natural purpose. Although the
greatest vices occur in the sphere of human social life, the social inclinations,
including competitiveness, are not, as such, vicious. No causal account of evil
is offered in this passage, but one draws from it the idea that something within
reason or affecting reason, but not nature, transforms the good predisposition
into vice. Also it should be said that Kant offers no suggestion here that
nature’s purpose of promoting culture can, by itself, bring about progress in
moral self-legislation. Thus far, Religion’s account is consistent with accounts
of nature, history, and culture in other critical writings treating these themes.
12
See, for example, Allen W. Wood, “Kant and the Intelligibility of Evil,” in Anderson-Gold and
Muchnik (eds.)., Kant’s Anatomy of Evil, pp. 144–72.
Kant follows the discussion of natural predispositions to the good (which
include animality or “mechanical self love” and personality or susceptibility
to respect for the moral law, as well as humanity or unsociable sociability)
with a sobering account of the propensity to evil and the claim that the
human is by nature evil. Here enters the grounding of evil in the intelligible
deed, and as this deed precedes all empirical determination of the will by
humans Kant claims that one never finds an empirical state of nature in
which humans are free of evil. Reason’s corruption of the good dispositions
has always already occurred. Kant does not suggest that humans have ever
existed, or ever can exist, in a noumenal or intelligible condition in which
they have purely moral wills. It is not an implication of the notion of
the intelligible deed that the human will fell from a pure condition in a
genetic−temporal story, although the language of “deed” temporally sche-
matizes a non-temporal state of affairs. Kant catalogues some of the “vices of
culture and civilization” which he calls the “most offensive of all evils,”
capable of inducing the vice of misanthropy. Above all, he cites the evil of
war, a constant condition between civilized peoples living in states that
stand in relations of raw nature with each other (R 6:34). “No philosopher
has yet been able to bring into agreement with morality” the principles that
states follow in their mutual relations. The passage has a footnote which
might seem puzzling if one thinks of Kant as arguing in the historical
writings that war is the greatest evil. It claims that there is an evil yet greater
than war between states, namely universal monarchy, “a state constitution
in which all freedom would necessarily expire, and together with it, virtue,
taste and science (which follow upon freedom).” Kant speaks once more of
nature having ends which are not the conscious aims of human actors, for
the tendency of states to subjugate their neighbors in striving for universal
rule results in “monsters” which disintegrate of themselves as their compo-
nents rebel. The liberated smaller states fail to form a republic of free
federated peoples and “begin the same game all over” (R 6:34). But Kant’s
message here is not that nature’s plan is to compel states to form a federation
of free peoples (as in the essays on history), although he speaks in praise of
scorned “philosophical chiliasm which hopes for a state of perpetual peace.”
In this passage, nature, as operative in the human as both sensible and
rational, only aims at preventing the death of reason in universal monarchy,
and indeed by means of war (internal rebellion).13 Kant pens the rather
13
See also R 6:123n on a “design of providence” averting a premature and dangerous (since it would
come before the moral improvement of the species) fusion of states through two causes: the difference
of languages and the difference of religions.
248 RICHARD VELKLEY
startling assertion that war is “not so evil as the grave of universal despotism
(or even as a federation of nations pitted against the relaxation of despotism
in any state).” The parenthetical federation, one has to suppose, is not the
same as “the republic of free federated peoples” but one wonders what kind
of alliance formed against a despotic state causes such great injuries to
freedom.
The passage which seems to be the most suggestive of a historical and
genetic account of human evil begins the third part of Religion on the
victory of the good principle over evil. Speaking of the dominion of evil over
the human, Kant writes that “the human being is nevertheless in this
perilous state through his own fault; hence he is bound at least to apply as
much force as he can muster in order to extricate himself from it” (R 6:93).
What follow sounds very much like a wholly social account of the origin of
evil, in which association with other humans is mutually corrupting. Living
in society arouses the passions, for the human concern with the opinion of
others necessarily draws one away from a life of limited needs: “It suffices
that they [other humans] are there, that they surround him.” But again one
does not have a genuine genetic account, as there is no beginning to this
condition, which belongs to the human as such. Thus Kant writes that if a
person “searches for the causes and the circumstances that draw him into
this danger and keep him there, he can easily convince himself that they do
not come his way from his own raw nature so far as he exists in isolation, but
rather from the human beings to whom he stands in relation of association.”
One can “easily convince onself” (sich leicht überzeugen) through the abun-
dant experience of mutual corruption that a condition of solitary innocence
precedes the corruption and that only entry into social life brings about evil.
Yet one knows from the account of the good predisposition to sociability
(humanity) that the rational concern with the opinion of others belongs,
like the other good predispositions, “to the possibility of human nature”
and is original (R 6:28). Concern with the other members of the species
belongs to the human as such, and therefore the transition from innocent
forms of that concern to vicious forms is not explicable through becoming a
member of the social order. Kant’s effort at explication – the intelligible
deed – is a poetic schematism that posits imputability to the human for
what must remain a mystery. Even so, Kant invites one to take encourage-
ment from the notion that human cooperation in the formation of a moral
society − that is, progress toward an ethical commonwealth, an idea made in
analogy to the juridico-civil state − will gain the upper hand over evil (R
6:94). But as soon as this idea is introduced, it is dissolved: “The idea of such
a state has an entirely well-grounded, objective reality in human reason (in
the duty to join such a state), even though we cannot subjectively ever hope
of the good will of human beings that these will work harmoniously toward
this end” (R 6:95).
5
The argument of Religion has a profoundly perplexing character, as it
proposes the most stringent and austere version of Kant’s conception of
human progress toward the highest good in the world as the necessary
object of practical reason and the central teleological idea of philosophy. So
austere is the presentation that Kant removes the comforting props, well-
known from other critical writings, of support from nature as promoting
culture and preparing the human species for moral perfection. In fact, this
work shows a few instances of nature’s hidden intentions, but these relate
ambiguously to the ultimate end of the ethical commonwealth. Thus nature
has a hidden mechanism for destroying universal monarchies, but not for
the creation of the universal federation of states; unsociable sociability
provides the arena for the unfolding of vices but almost nothing is said
about how it also furthers the moral end of the species. At times Kant openly
puts forward the moral project as incompletable: evil cannot be eradicated
yet we must think that it can be overcome. The unbridgeable gap is
exhibited in the figure of the “intelligible deed.” Beyond this, in general
Kant emphasizes the unavailability of any causal or genetic account (as
contrasted with “positing”) of the universal propensity for evil. At the same
time, Kant employs here, as in other works, figurative language intended to
help make intuitable for practical purposes fundamental concepts that lie
beyond sensibility. But the poetic figures seem to serve a paradoxical
purpose in Religion or, to put the matter another way and more precisely,
they make present the necessity of contradiction as the condition for human
striving toward the good. In other terms, Kant resorts to the products of
poetic genius to expose this paradox as the deepest philosophic meaning of
the project of rational religion.
Bibliography
Adams, Robert M., “Introduction,” in Robert M. Adams and George di Giovanni

(eds.), di Giovanni and Allen W. Wood (trans.), Religion within the Boundaries
of Mere Reason and Other Writings (Cambridge University Press, 1999), vii–xxxii
Allison, Henry, Kant’s Theory of Freedom (Cambridge University Press, 1990)
Lessing and the Enlightenment (Ann Arbor: University of Michigan Press, 1966)
“On the Very Idea of a Propensity to Evil,” in Allison, Essays on Kant (Oxford
University Press, 2012), 99–109
“Reflections on the Banality of (Radical) Evil: A Kantian Analysis,” in Allison,
Idealism and Freedom: Essays on Kant’s Theoretical and Practical Philosophy
(Cambridge University Press, 1996), 169–82
Almond, Ian, History of Islam in German Thought from Leibniz to Nietzsche (New
York: Routledge, 2010)
Ameriks, Karl, “The End of the Critiques: Kant’s Moral ‘Creationism,’” in
P. Muchnik (ed.), Rethinking Kant, Vol. I, 165–90
Interpreting Kant’s Critiques (Oxford: Clarendon Press, 2003)
“Kant, Human Nature, and History after Rousseau,” in Susan Shell and
Richard Velkley (eds.), Kant’s “Observations” and “Remarks”: A Critical
Guide (Cambridge University Press, 2012), 247–65
Kant and the Fate of Autonomy: Problems in the Appropriation of the Critical
Philosophy (Cambridge University Press, 2000)
“The Purposive Development of Human Capacities,” in Amélie Rorty and
James Schmidt (eds.), Kant’s “Idea for a Universal History with a Cosmopolitan
Aim”: A Critical Guide (Cambridge University Press, 2009), 46–67
“‘The Question is Whether a Purely Apparent Person is Possible,’” in
Eckart Förster and Yitzhak Melamud (eds.), Spinoza and German Idealism
“Das Schicksal von Kants Rezensionen zu Herders Ideen,” in Otfried Höffe (ed.),
Immanuel Kant: Geschichtsphilosophie (Berlin: Akademie Verlag, 2011), 119–36
“Tragedy, Romanticism, and Idealism,” in Bärbel Frischmann and
Elizabeth Millán-Zaibert (eds.), Das Neue Licht der Frühromantik:
Innovation und Aktualität Frühromantischer Philosophie (Paderborn:
Ferdinand Schöningh, 2008), 28–38
Ameriks, Karl (ed.), The Cambridge Companion to German Idealism (Cambridge
University Press, 2000)
250
Bibliography 251
Anderson, Pamela Sue, and Jordan Bell, Kant and Theology (New York and
London: Continuum, 2010)
Anderson-Gold, Sharon, “God and Community: An Inquiry into the Religious
Implications of the Highest Good,” in Rossi and Wreen (eds.), Kant’s
Philosophy of Religion Reconsidered, 113–31
“Kant’s Ethical Commonwealth: The Highest Good as a Social Goal,”
International Philosophical Quarterly 26 (1986), 23–32
“Kant’s Rejection of Devilishness: the Limits of Human Volition,” Idealistic
Studies 14 (1984), 35–48
Unnecessary Evil: History and Moral Progress in the Philosophy of Immanuel Kant
(Albany: State University of New York Press, 2001)
Anderson-Gold, Sharon, and Pablo Muchnik (eds.), Kant’s Anatomy of Evil
(Cambridge University Press, 2010)
Aner, Karl, Die Theologie der Lessingzeit (Halle: Max Niemeyer, 1929)
An-Na‘im, Abdullahi Ahmed, Islam and the Secular State: Negotiating the Future of
Shari‘a (Cambridge, MA: Harvard University Press, 2008)
Arendt, Hannah, The Origins of Totalitarianism (San Diego: Harcourt, 1994)
Arpaly, Nomy, “Moral Worth,” Journal of Philosophy 99 (2002), 223–45
Atkinson, C., “Kant on Human Nature and Radical Evil,” Philosophy and Theology
19 (2007), 215–24
Augustine, St., On Free Choice of the Will, trans. Thomas Williams (Indianapolis:
Hackett Publishing Company, 1993)
Barth, Karl, Protestant Thought from Rousseau to Ritschl, trans. Brian Cozens (New
York: Simon and Schuster, 1969)
Baumeister, Roy, Evil: Inside Human Violence and Cruelty (New York: Freeman, 1999)
Bernstein, Richard, Radical Evil: A Philosophical Investigation (Cambridge, MA:
Polity Press, 2002)
“Radical Evil: Kant at War with Himself,” in María Pía Lara (ed.), Rethinking
Evil: Contemporary Perspectives (Berkeley: University of California Press,
2001), 55–85
Berthold, Fred, Jr., God, Evil, and Human Learning: A Critique and Revision of the Free
Will Defense in Theodicy (Albany: State University of New York Press, 2004)
Bittner, Rűdiger, “Maximen,” in G. Funke and J. Klopper (eds.), Akten des 4.
Internationalen Kant-Kongresses (Berlin: de Gruyter, 1974), 485–98
Blackburn, Simon, Ruling Passions (Cambridge University Press, 1998)
Bloch, Ernst, Das Prinzip Hoffnung (Frankfurt am Main: Suhrkamp, 1959)
Bohatec, Josef, Die Religionsphilosophie Kants in der “Religion innerhalb der Grenzen
der blossen Vernunft” (Hildesheim: Georg Olms, 1966)
Bovens, Luc, “The Value of Hope,” Philosophy and Phenomenological Research 59
(1999), 667–81
Brandt, Reinhard, Die Bestimmung des Menschen bei Kant (Hamburg: Meiner,
2007)
Brewer, Talbot, “Maxims and Virtues,” Philosophical Review 111, 4 (2002), 539–72
Brown, Wendy, and Janet E. Halley (eds.), Left Legalism / Left Critique (Durham,
NC: Duke University Press, 2002)
252 Bibliography
Bruch, Jean-Louis, La Philosophie religieuse de Kant (Paris: Aubier, 1968)
Budick, Sanford, Kant and Milton (Cambridge, MA: Harvard University Press,
2010)
Byrne, Peter, Kant on God (Farnham: Ashgate, 2007)
Card, Claudia, The Atrocity Paradigm: A Theory of Evil (New York: Oxford
“Kant’s Moral Excluded Middle,” in Anderson-Gold and Muchnik (eds.), Kant’s
Anatomy of Evil, 74–92
Caswell, Matthew, “Kant on the Diabolical Will: A Neglected Alternative?”
Kantian Review 12 (2007), 147–57
“The Value of Humanity and Kant’s Conception of Evil,” Journal of the History
of Philosophy 49 (2006), 635–63
Cherkasova, E., “On the Boundary of Intelligibility: Kant’s Conception of Radical
Evil and the Limits of Ethical Discourse,” Review of Metaphysics 58 (2005),
571–84
Chignell, Andrew, “Belief in Kant,” Philosophical Review 116 (2007), 323–60
“The Devil, the Virgin, and the Envoy: Symbols of Struggle in Religion, Part
Two, Section Two,” in Höffe (ed.), Die Religion innerhalb der Grenzen der
blossen Vernunft, 111–30
“Rational Hope, Moral Order, and the Revolution of the Will,” in Eric Watkins
(ed.), Divine Order, Human Order, and the Order of Nature (New York:
Oxford University Press, 2013), 197–218
“Rationalism, Religion, and the Laws: Leibniz and Kant on Miracles,” in
Brandon Look (ed.), Leibniz and Kant (New York: Oxford University Press,
2014)
“Real Repugnance and Belief about Things-in-Themselves: A Problem and Kant’s
Three Solutions,” in James Krueger and Benjamin Bruxvoort Lipscomb (eds.),
Kant’s Moral Metaphysics: God, Freedom, and Immortality (Berlin / New York: de
Gruyter, 2010), 177–211
“Real Repugnance and Our Ignorance of Things-in-Themselves: A Lockean
Problem in Kant and Hegel,” International Yearbook of German Idealism 7
(2009), 135–59
Chignell, Andrew, and Nicholas Stang, “Postulate des empirischen Denkens,” in
G. Mohr, J. Stolzenberg, and M. Willaschek (eds.), KANT-LEXICON
(Berlin: Walter de Gruyter, forthcoming), n. p.
Chignell, Andrew, Pamela Anderson, Chris Firestone, et al., “Symposium: Kant’s
Philosophy of Religion,” Faith and Philosophy 29 (2012), 144–228
Ciaramelli, Fabio, “Du mal radical à la banalité du mal,” Revue Philosophique de
Louvain 9 (1995), 392–407
Cicero, De finibus bonorum et malorum, trans. H. Rackham (Cambridge, MA:
Harvard University Press, 1931)
De officiis, trans. Walter Miller (Cambridge, MA: Harvard University Press, 1975)
Coeckelbergh, Mark, “Can We Choose Evil? A Discussion of the Problem of
Radical Evil as a Modern and Ancient Problem of Freedom,” in D. E. Keen
Bibliography 253
and P. R. Keen (eds.), Considering Evil and Human Wickedness (Oxford:
Inter-Disciplinary Press, 2004), 339–53
Cohen, Hermann, Reason and Hope: Selections from the Writings of Hermann
Cohen, ed. and trans. Eva Jospe (New York: W.W. Norton, 1971)
Dalferth, Ingolf, Malum: Theologische Hermeneutik des Bösen (Tübingen: Mohr
Siebeck, 2010)
Day, John P., “Hope,” American Philosophical Quarterly 6 (1969), 89–102
Deleuze, Gilles, Spinoza: Practical Philosophy (San Francisco: City Lights Books,
1988)
Denis, Lara, “Kant’s Conception of Virtue,” in Guyer (ed.), The Cambridge
Companion to Kant and Modern Philosophy, 505–37
Despland, Michel, Kant on History and Religion (Montreal and London: McGill-
Queen’s University Press, 1973)
DiCenso, James J., Kant, Religion, and Politics (Cambridge University Press, 2011)
Kant’s Religion within the Boundaries of Mere Reason: A Commentary (Cambridge
di Giovanni, George, Freedom and Religion in Kant and his Immediate Successors:
The Vocation of Humankind, 1774–1800 (Cambridge University Press, 2005)
“Translator’s Introduction,” in Allen W. Wood and George di Giovanni (eds.
and trans.), Religion and Rational Theology (Cambridge University Press,
1996), 41–54
Downie, Robin S., “Hope,” Philosophy and Phenomenological Research 24 (1963), 248–51
Duggan, Timothy, and B. Gert, “Free Will as the Ability to Will,” Noûs 13 (1979),
197–217
Ebbinghaus, Julius, “Die Formeln des kategorischen Imperativs und die Ableitung
inhaltlich bestimmter Pflichten,” in H. Oberer and G. Geismann (eds.),
Gesammelte Schriften II: Philosophie der Freiheit (Bonn: Bouvier, 1988), 209–29
Erasmus–Luther, Discourse on Free Will, trans. Ernst F. Winter (New York:
Frederick Ungar, 1961)
Euben, Roxanne L., and Muhammad Qasim Zaman (eds.), Princeton Readings in
Islamist Thought: Texts and Contexts from Al-Banna to Bin Laden (Princeton
Fackenheim, Emil L., “Kant and Radical Evil,” in R. Chadwick (ed.), Immanuel
Kant: Critical Assessments, Vol. III (London: Routledge, 1992), 259–73
Fenves, Peter, Late Kant: Towards Another Law of the Earth (New York: Routledge,
2003)
Firestone, Chris L., and Nathan Jacobs, In Defense of Kant’s Religion
(Bloomington/Indianapolis: Indiana University Press, 2008)
Firestone, Chris L., and Stephen Palmquist (eds.), Kant and the New Philosophy of
Religion (Bloomington/Indianapolis: Indiana University Press, 2006)
Flikschuh, Katrin, “Hope as Prudence: Practical Faith in Kant’s Political
Thinking,” International Yearbook of German Idealism 7 (2009), 95–117
“Kant’s Kingdom of Ends: Metaphysical, Not Political,” in Jens Timmermann
(ed.), Kant’s Groundwork of the Metaphysics of Morals: A Critical Guide
254 Bibliography
Forde, Gerhard O., The Captivation of the Will: Luther vs. Erasmus on Freedom and
Bondage (Grand Rapids: William B. Eerdmans, 2005)
Formosa, P., “Kant on the Radical Evil of Human Nature,” Philosophical Forum 38
(2007), 221–45
Frankfurt, Harry, “Freedom of the Will and the Concept of a Person,” Journal of
Philosophy 68 (1971), 5–20
The Importance of What We Care About (Cambridge University Press, 1988)
Gay, Peter, The Enlightenment: The Rise of Modern Paganism (New York: Knopf, 1966)
Geissmann, G., “Die Formeln des kategorischen Imperativs nach H. J. Paton,
K. Reich, und J. Ebbinghaus,” Kant-Studien 93 (2002), 374–84
Greene, Theodore M., “The Historical Context and Religious Significance of
Kant’s Religion,” introduction to Kant, Religion within the Limits of Reason
Alone, trans. Greene and Hoyt H. Hudson (New York: Harper Torchbook,
1960), ix–xxviii
Grenberg, Jeanine, Kant and the Ethics of Humility: A Story of Dependence,
Corruption and Virtue (Cambridge University Press, 2005)
“Social Dimensions of Kant’s Conception of Radical Evil,” in Anderson-Gold
and Muchnik (eds.), Kant’s Anatomy of Evil, 173–94
Gressis, Rob, “Recent Work on Kantian Maxims,” Philosophy Compass 5 (2010),
pp. 216–22
Grimm, St. R., “Kant’s Argument for Radical Evil,” European Journal of Philosophy
10 (2002), 160–77
Guyer, Paul, “Bridging the Gulf: Kant’s Project in the Third Critique,” in
Graham Bird (ed.), A Companion to Kant (Hoboken, NJ: Wiley Blackwell,
2009), 423–40
Kant on Freedom, Law and Happiness (Cambridge University Press, 2000)
Guyer, Paul (ed.), The Cambridge Companion to Kant and Modern Philosophy
Hare, John, The Moral Gap: Kantian Ethics, Human Limits, and God’s Assistance
(Oxford / New York: Oxford University Press, 1996)
Hasse, J. G., Der alte Kant, Hasses Schrift: Letzte Äusserungen Kants und persönliche
Notizen aus dem Opus Postumum (Berlin/Leipzig: A. Buchenau und
G. Lehmann, 1925)
Herman, Barbara, Moral Literacy (Cambridge, MA: Harvard University Press, 2007)
The Practice of Moral Judgment (Cambridge, MA: Harvard University Press, 1993)
Herrmann, Johann Georg, Ethik (Tübingen: J. C. B. Mohr, 1904)
Hobbes, Thomas, Leviathan, ed. Edwin Curley (Indianapolis: Hackett Publishing
Company, 1994)
Hodges, Herbert Arthur, The Pattern of Atonement (London: SCM Press, 1955)
Höffe, Otfried, “Ein Thema Wiedergewinnen: Kant über das Böse,” in Höffe and
A. Pieper (eds.), Über das Wesen der menschlichen Freiheit (Berlin: Akademie
Verlag, 1995), 11–34
Höffe, Otfried (ed.), Immanuel Kant. Die Religion innerhalb der Grenzen der blossen
Vernunft (Berlin: Akademie Verlag, 2011)
Bibliography 255
Hunter, Ian, Rival Enlightenments: Civil and Metaphysical Philosophy in Early
Modern Germany (Cambridge University Press, 2001)
Israel, Jonathan, Radical Enlightenment: Philosophy and the Making of Modernity,
1650–1750 (New York: Oxford University Press, 2001)
A Revolution of the Mind: Radical Enlightenment and the Intellectual Origins of
Modernity (Princeton University Press, 2010)
Jachmann, Reinhold Bernhard, “Immanuel Kant geschildert in Briefen an einen
Freund” [1804], in Felix Gross (ed.), Immanuel Kant, Sein Leben in
Darstellungen von Zeitgenossen (Borowski, Jachmann, Wasianski) (Berlin:
Deutsches Bibliothek, 1912), 117–212
Jackson, Timothy P., “Arminian Edification: Kierkegaard on Grace and Free
Will,” in Alastair Hannay and Gordon D. Marino (eds.), The Cambridge
Companion to Kierkegaard (Cambridge University Press, 1998), 235–56
Jaspers, Karl, “Das Radikal Böse bei Kant,” in Jaspers, Rechenschaft und Ausblick
(Munich: R. Piper, 1951), 90–113
Kadowaki, T., “Das radikal Böse bei Kant” (dissertation: Bonn, 1960)
Kaulbach, Friedrich, Immanuel Kants “Grundlegung zur Metaphysik der Sitten.”
Interpretation und Kommentar (Darmstadt: Wissenschaftliche Buchgesellschaft,
1996)
Korsgaard, Christine, Creating the Kingdom of Ends (Cambridge University Press,
1996)
“Kant on Dealing with Evil,” Philosophy and Public Affairs 15 (1986), 420–40
Kuehn, Manfred, Kant: A Biography (Cambridge University Press, 2001)
Lawrence, Joseph P., “Radical Evil and Kant’s Turn to Religion,” Journal of Value
Inquiry 36 (2002), 319–35
Leibniz, Gottfried Wilhelm, Political Writings, ed. and trans. Patrick Riley
Lewis, David, “Causation,” Journal of Philosophy 70 (1973), 556–67
Lichtenberg, Heinrich P., “Über die Unerforschlichkeit des Bösen nach Kant,”
Studia Philosophica 53 (1993), 117–31
Longuenesse, Béatrice, Kant on the Human Standpoint (Cambridge University
Press, 2005)
Louden, Robert, “Evil Everywhere: The Ordinariness of Radical Evil,” in
Anderson-Gold and Muchnik (eds.), Kant’s Anatomy of Evil, 93–115
Lu, Rachel, “Natural and Supernatural Virtue in St. Bonaventure,” Cornell University
dissertation, 2012
Madore, J., Difficult Freedom and Radical Evil in Kant: Deceiving Reason (London /
New York: Continuum, 2011)
March, Andrew F., Islam and Liberal Citizenship: The Search for an Overlapping
Consensus (New York: Oxford University Press, 2009)
Mariña, Jacqueline, “Kant on Grace: A Reply to his Critics,” Religious Studies 33
(1997), 379–400
Markovits, Julia, “Acting for the Right Reasons,” Philosophical Review 119 (2010),
201–42
256 Bibliography
Martin, Adrienne M. “Hopes and Dreams,” Philosophy and Phenomenological
Research 83 (2011), 148–73
How We Hope (Princeton University Press, 2013)
Mason, H. T. (ed.), The Leibniz–Arnauld Correspondence (Manchester University
Press, 1967)
McBride, William, “Liquidating the ‘Nearly Just Society’: Radical Evil’s
Triumphant Return,” in A.D. Schrift (ed.), Modernity and the Problem of
Evil (Bloomington/Indianapolis: Indiana University Press, 2005), 28–38
McCammon, Christopher, “Overcoming Deism: Hope Incarnate in Kant’s
Rational Religion,” in Firestone and Palmquist (eds.), Kant and the New
Philosophy of Religion, 79–89
McCarty, Richard, Kant’s Theory of Action (Oxford University Press, 2009)
Meirav, Ariel, “The Nature of Hope,” Ratio 22 (2009), 216–33
Michalson, Gordon E., Fallen Freedom: Kant on Radical Evil and Moral
Regeneration (Cambridge University Press, 1990)
“God and Kant’s Ethical Commonwealth,” The Thomist 65 (2001), 67–92
“In Defense of Not Defending Kant’s Religion,” Faith and Philosophy 29 (2012),
181–92
“The Inscrutability of Moral Evil in Kant,” The Thomist 51 (1987), 246–69
“Kant, the Bible, and the Recovery from Radical Evil,” in Anderson-Gold and
Muchnik (eds.), Kant’s Anatomy of Evil, 57–73
Kant and the Problem of God (Oxford / Malden, MA: Blackwell Publishers, 1999)
“Moral Regeneration and Divine Aid in Kant,” Religious Studies 25 (1989),
259–70
“The Problem of Salvation in Kant’s Religion within the Limits of Reason Alone,”
International Philosophical Quarterly 37 (1997), 319–28
Moltmann, Jürgen, Theologie der Hoffnung: Untersuchungen zur Begründung
und zu den Konsequenzen einer Christlichen Eschatologie, 3rd edn. (Munich:
C. Kaiser, 1965)
Moore, AdrianW., Noble in Reason, Infinite in Faculty: Themes and Variations in
Kant’s Moral and Religious Philosophy (London: Routledge, 2003)
Morgan, Seiriol, “The Missing Formal Proof of Humanity’s Radical Evil in Kant’s
Religion,” Philosophical Review 114 (2005), 63–114
Muchnik, Pablo, “An Alternative Proof of the Universal Propensity to Evil,” in
Anderson-Gold and Muchnik (eds.), Kant’s Anatomy of Evil, 116–43
“Kant on the Sources of Evil,” in V. Rohden, R. R. Terra, and G. A. de Almeida
(eds.), Proceedings of the 10th International Kant Congress (Berlin: de Gruyter,
2009), 287–97
Kant’s Theory of Evil: An Essay on the Dangers of Self-Love and the Aprioricity of
History (Plymouth: Lexington Books, 2009)
“On the Alleged Vacuity of Kant’s Concept of Evil,” Kant-Studien 97 (2006),
430–51
“Radical Evil: Between the Trivial and the Diabolical,” Contemporary Philosophy
23 (2001), 43–47
Bibliography 257
Muchnik, Pablo (ed.), Rethinking Kant, Vol. I (Newcastle upon Tyne: Cambridge
Scholars Press, 2008)
Rethinking Kant, Vol. II (Newcastle upon Tyne: Cambridge Scholars Press,
2010)
Mulholland, Leslie, “Freedom and Providence in Kant’s Account of Religion: The
Problem of Expiation,” in Rossi and Wreen (eds.), Kant’s Philosophy of
Religion Reconsidered, 77–102
Munzel, G. Felicitas, Kant’s Conception of Moral Character: The “Critical” Link
of Morality, Anthropology, and Reflective Judgment (University of Chicago
Press, 1999)
Kant’s Conception of Pedagogy: Toward Education for Freedom (Evanston, IL:
Northwestern University Press, 2012)
Muyskens, James L., The Sufficiency of Hope: Conceptual Foundations of Religion
(Philadelphia: Temple University Press, 1979)
Neiman, Susan, Evil in Modern Thought (Princeton University Press, 2002)
The Unity of Reason: Rereading Kant (New York: Oxford University Press, 1994)
Nuyen, A.Tuan, “Kant on Miracles,” History of Philosophy Quarterly 19 (2002),
309–23
O’Neill, Onora, Constructions of Reason: Explorations of Kant’s Practical Philosophy
Oppenheimer, Paul, Evil and the Demonic: A New Theory of Monstrous Behavior
(New York University Press, 1996)
Palmquist, Stephen R., “Introduction,” in Kant, Religion within the Bounds of Bare
Reason, trans. Werner S. Pluhar (Indianapolis: Hackett Publishing Company,
2009), xv–xlix
Kant’s Critical Religion (Aldershot / Burlington: Ashgate, 2000)
Pasternack, Lawrence, “Kant’s ‘Doctrinal Belief in God,’” in Oliver Thorndike
(ed.), Rethinking Kant, Vol. III (Newcastle upon Tyne: Cambridge Scholars
Press, 2011), 200–18
Pettit, Philip, “Hope and its Place in Mind,” Annals of the American Academy of
Political and Social Science 592 (2004), 152–65
Pinkard, Terry, German Philosophy 1760–1860: The Legacy of Idealism (Cambridge
Pojman, Louis, Philosophy of Religion (Houston: Mayfield Publishing Company)
Prauss, G., Kant űber Freiheit als Autonomie (Frankfurt am Main: Vittorio
Klostermann, 1983)
Quinn, Philip L., “Christian Atonement and Kantian Justification,” Faith and
Philosophy 3 (1986), 440–62
“Original Sin, Radical Evil and Moral Identity,” Faith and Philosophy 1 (1984),
188–202
“Saving Faith from Kant’s Remarkable Antinomy,” Faith and Philosophy 7
(1990), 418–33
Qutb, Sayyid, Milestones (Cedar Rapids: Mother Mosque Foundation, 1995)
Ramadan, Tariq, Radical Reform: Islamic Ethics and Liberation (New York: Oxford
258 Bibliography
Rangelov, I. “Ideology in between Radical and Diabolical Evil: Kant’s Ethics of the
Real,” Philosophy, Sociology and Psychology 2 (2003), 759–68
Rawls, John, Collected Papers, ed. Samuel Freeman (Cambridge, MA: Harvard
Political Liberalism (New York: Columbia University Press, 2005)
Reboul, Olivier, Kant et le problème du mal (Montreal: Les Presses de l’Université
de Montréal, 1971)
Reimarus, Hermann Samuel, Von dem Zwecke Jesu und seiner Jünger. Noch ein
Fragment des Wolffenbüttelschen Ungenannten, ed. Gotthold Ephraim Lessing,
2nd edn. (Braunschweig, 1784)
Die vornehmsten Wahrheiten der natürlichen Religion in Zehn Abhandlungen auf
eine begreifliche Art erkläret und gerettet von . . . (Professor in Hamburg)
(Hamburg: Johann Carl Bohn, 1754)
Riley, Patrick, Leibniz’ Universal Jurisprudence: Justice as the Charity of the Wise
(Cambridge, MA: Harvard University Press, 1996)
Rogozinski, Jacob, “Kant et le mal radical,” Kriterion 39 (1998), 7–21
Rommel, Herbert, Zum Begriff des Bösen bei Augustinus und Kant. Der Wandel
der ontologischen zur autonomen Perspektive (Frankfurt am Main: Peter
Lang, 1997)
Rorty, Richard, “The Priority of Democracy to Philosophy,” in Rorty, Objectivity,
Relativism, and Truth: Philosophical Papers I (Cambridge University Press,
1991), 175–96
“Religion as a Conversation-Stopper,” Common Knowledge 3 (Spring 1994), 1–6
“Religion as a Conversation-Stopper,” in Philosophy and Social Hope (London:
Penguin Books, 1999)
“Religion in the Public Square: A Reconsideration,” Journal of Religious Ethics 31,
1 (2003), 141–49
Rosenthal, Abigail L., A Good Look at Evil (Philadelphia: Temple University
Press, 1987)
Rossi, Philip J., “Kant’s Doctrine of Hope: Reason’s Interest and the Things of
Faith,” New Scholasticism 56 (1982), 228–38
“Kant’s ‘Metaphysics of Permanent Rupture’: Radical Evil and the Unity of
Reason,” in Anderson-Gold and Muchnik (eds.), Kant’s Anatomy of Evil,
13–32
The Social Authority of Reason: Kant’s Critique, Radical Evil, and the Destiny of
Humankind (Albany: State University of New York Press, 2005)
Rossi, Philip J., and Michael Wreen (eds.), Kant’s Philosophy of Religion
Reconsidered (Bloomington/Indianapolis: Indiana University Press, 1991)
Samte-Porat, I., “Satanic Motivations,” Journal of Value Inquiry 41 (2007), 77–94
Schmidt, James, What is Enlightenment? Eighteenth-Century Answers and
Twentieth-Century Questions (Berkeley: University of California Press, 1966)
Schneewind, Jerome B., Essays on the History of Moral Philosophy (New York:
Oxford University Press, 2010)
The Invention of Autonomy: A History of Modern Moral Philosophy (Cambridge
Bibliography 259
Schulte, Christine, Radikal Böse. Die Karriere des Bösen von Kant biz Nietzsche
(Munich: Walter Fink, 1988)
Schweitzer, Albert, Geschichte der Leben-Jesu-Forschung, Vol. I (Munich/Hamburg:
Siebenstern Taschenbuch, 1966)
Semler, Johann Salomo, D. Joh. Salomo Semlers Beantwortung der Fragmente eines
Ungennanten insbesondere vom Zwecke Jesu und seiner Jünger (Halle: Verlag
des Erziehungsinstituts, 1780)
Letztes Glaubensbekenntnis über natürliche und christliche Religion (Königsberg:
Nicolovius, 1792)
Setiya, Kieran, Reasons without Rationalism (Princeton University Press, 2007)
Shell, Susan Meld, Kant and the Limits of Autonomy (Cambridge, MA: Harvard
Sherman, Nancy, “The Virtues of Common Pursuit,” Philosophy and
Phenomenological Research 53 (1993), 277–99
Siegfried, C. H., “The Radical Evil in Human Nature,” in G. Funke (ed.), Akten
des 4. Internationalen Kant-Kongresses (Berlin: de Gruyter, 1974), 605–13
Silber, John, “The Ethical Significance of Kant’s Religion,” in Greene and Hudson
(trans.), Religion within the Limits of Reason Alone, lxxix–cxxxiv
“Kant at Auschwitz,” in G. Funke and T. M. Seebohm (eds.), Proceedings of the
Sixth International Kant Congress (Washington, DC: Center for Advanced
Research in Phenomenology and University Press of America, 1991), 177–211
Smith, Adam, The Theory of Moral Sentiments (Amherst, NY: Prometheus Books,
2000)
Smith, Steven B., Spinoza, Liberalism, and the Question of Jewish Identity (New
Haven: Yale University Press, 1997)
Spener, Philipp Jacob, Pia Desideria: Programm des Pietismus, ed. Erich Bayreuther
(Wuppertal: Aussat Verlag, 1964)
Spinoza, Benedictus de, Theological-Political Treatise, trans. Jonathan I. Israel
Stevenson, Leslie, Inspirations from Kant (New York: Oxford University Press,
2011)
“Knowledge, Belief or Faith, and Opinion,” Kantian Review 7 (2003), 72–101
Stewart, Matthew, The Courtier and the Heretic: Leibniz, Spinoza, and the Fate of
God in the Modern World (New York: Norton, 2006)
Stump, Eleanore, “Augustine on Free Will,” in Stump and Norman Kretzmann
(eds.), The Cambridge Companion to Augustine (Cambridge University Press,
2001), 124–47
Sussman, David, “For Badness’ Sake,” Journal of Philosophy 106 (2009), 613–28
“Perversity of the Heart,” Philosophical Review 114 (2005), 1–24
Sweet, Kristi, “Kant and the Culture of Discipline: Rethinking the Nature of
Nature,” Epoche: A Journal of the History of Philosophy 15 (Fall 2010), 121–38
“The Moral Import of the Critique of Judgment,” in Muchnik (ed.), Rethinking
Kant, Vol. II, 222–37
Taha, Mohamed Mahmoud, The Second Message of Islam, trans. Abdullahi
Ahmed An-Na‘im (Syracuse University Press, 1987)
260 Bibliography
Tampio, Nicholas, Kantian Courage: Advancing the Enlightenment in Contemporary
Political Theory (New York: Fordham University Press, 2012)
“Rawls and the Kantian Ethos,” Polity 39, 1 (2007), 79–102
Taylor, Robert S., Reconstructing Rawls: The Kantian Foundations of Justice as
Fairness (University Park: The Pennsylvania State University Press, 2011)
Tenenbaum, Sergio, Appearances of the Good: An Essay on Practical Reason
Thielicke, Helmut, Modern Faith and Thought, trans. Geoffrey W. Bromiley
(Grand Rapids, MI: William Eerdmans Publishing Company, 1990)
Timmermann, Jens, “Kant’s Puzzling Ethics of Maxims,” Harvard Review of
Philosophy 8 (2000), 39–52
Velkley, Richard, Freedom and the End of Reason: On the Moral Foundation of Kant’s
Critical Philosophy (University of Chicago Press, 1989)
Velleman, J. David, “The Guise of the Good,” Noûs 26 (1992), 1–26
Wheatley, J., “Wishing and Hoping,” Analysis 18 (1958), 121–31
White, Stephen K., The Ethos of a Late-Modern Citizen (Cambridge, MA: Harvard
Williams, Bernard, Ethics and the Limits of Philosophy (Cambridge, MA: Harvard
Williams, Howard, Kant’s Political Philosophy (Oxford: Basil Blackwell, 1983)
Willnauer, E., Heute das Böse denken. Mit Immanuel Kant und Hannah Arendt zu
einem Neuansatz für die Theologie (Berlin: Rhombos, 2004), pp. 53–166
Wimmer, R., Kant’s kritische Religionsphilosophie (Berlin / New York: de
Gruyter, 1990)
Wolf, Susan, “Sanity and the Metaphysics of Responsibility,” in Ferdinand Schoeman
(ed.), Responsibility, Character and the Emotions (Cambridge University Press,
1987), 46–62
Wolterstorff, Nicholas, “Conundrums in Kant’s Rational Religion,” in Rossi and
Wreen (eds.), Kant’s Philosophy of Religion Reconsidered, 40–53
“Why We Should Reject what Liberalism Tells Us about Speaking and Acting
for Religious Reasons,” in Paul J. Weithman (ed.), Religion and Contemporary
Liberalism (University of Notre Dame Press, 1997), 162–81
Wood, Allen W., “General Introduction,” in Allen W. Wood and George di
Giovanni (eds. and trans.), Religion and Rational Theology (Cambridge
University Press, 1996) xi–xxiv
“Kant and the Intelligibility of Evil,” in Anderson-Gold and Muchnik (eds.),
Kant’s Anatomy of Evil, 144–72
“Kant and the Problem of Human Nature,” in B. Jacobs and P. Kain (eds.),
Essays on Kant’s Anthropology (New York: Cambridge University Press, 2003),
40–42
Kantian Ethics (Cambridge University Press, 2008)
Kant’s Ethical Thought (Cambridge University Press, 1999)
Kant’s Moral Religion (Ithaca and London: Cornell University Press, 1970)
“Kant’s Practical Philosophy,” in Ameriks (ed.), The Cambridge Companion to
German Idealism, 57–75
Bibliography 261
Kant’s Rational Theology (Ithaca and London: Cornell University Press, 1978)
“Religion, Ethical Community, and the Struggle Against Evil,” in Charlton Payne
and Lucas Thorpe (eds.), Kant and the Concept of Community (University of
Rochester Press, 2011), 138–49
Wood, Allen W., and George di Giovanni (eds. and trans.), Religion within the
Boundaries of Mere Reason and Other Writings (Cambridge University Press,
1998)
Yovel, Yirmiahu, “Bible Interpretation as Philosophical Praxis: A Study of Spinoza
and Kant,” Journal of the History of Philosophy 11, 2 (1973), 189–212
Kant and the Philosophy of History (Princeton University Press, 1980)
Index
Abraham, 126 Christianity, 3, 4, 10, 15–17, 18–22, 53, 118, 120, 130,
Allison, Henry, 161 147, 158, 164, 184, 220
Aner, Karl, 158, 159 and natural religion, 28
animality, 42, 43, 45, 65–67, 68, 222–23, 227, 247 Christology, 7, 16, 18
An-Na‘im, Abdullahi Ahmed, 176, 192 church, 8, 18, 23, 28, 115, 183
and coercion, 189 invisible, 16, 20, 146, 182, 237
and pluralism, 191 universal, 183–84
and secularity, 189 Cicero, 220–21
connection between Islam and politics, 190 Conflict of the Faculties, 4, 13, 23, 110
antinomy, 122, 123, 193, 197, 225, 228 and Kant’s religious outlook, 25–30
Aristotle, 11, 13, 80, 96 Confucius, 17
Augustine, St., 127, 128 conscience, 9, 28, 74, 210, 216, 219–20,
autonomy, 49, 61–63, 146, 213, 238 230–32
and divine will, 210 and God, 232
and radical evil, 4 and practical reason, 230
relation to Christianity, 20 defined, 231
Copernicus, 150
Bacon, Francis, 10 Critique of Judgment, 9, 223, 226, 228, 229, 235
Bahrdt, Carl Friedrich, 158 Critique of Practical Reason, 2, 8, 40, 132,
Baillie, Donald, 128 148, 170, 210, 214, 217, 222, 223, 224,
Barth, Karl, 157, 159 225, 234
Baumgarten, Alexander Gottlieb, 163 Critique of Pure Reason, 6, 12, 16, 215, 217, 219,
Baumgarten, Siegmund Jakob, 163 220, 224, 226, 233–37, 241
belief, 98, 117 religious implications of, 13
relation to hope, 6, 98–100, 106 culture, 237, 242, 246, 247
Bible, 17, 19, 21, 157, 181, 220, 241 and moral improvement, 245, 249
and practical reason, 183
and Semler, 163 Day, J. P., 101
and Spinoza, 177 depravity, 35, 50, 51
rational interpretation of, 10, 13, 19, 23, 26–29 Descartes, 11
Bloch, Ernst, 101 diabolical will, 34, 36, 227
Bohatec, Josef, 168 disposition, 5, 134, 181, 204, 205, 206, 210, 225,
Bossuet, Bishop, 180 226, 227, 236, 245
Bovens, Luc, 101 and character, 91–96
Buddha, 125 and infinite regress, 86–87
Buffon, Georges Louis Leclerc, 152 and moral law, 79
Bultmann, Rudolf, 163 meaning of, 79–80, 82–83
relation to action, 89–91, 95–96
categorical imperative, 79, 80, 173, 212 relation to maxims, 82–83, 84, 86, 89, 93
character, 42, 80, 229 Downie, R. S., 101
and the disposition, 91–96 duty, 51, 67, 75, 79
262
Index 263
Eberhard, Johann August, 158 as possible object of knowledge, 13–15, 29, 213,
ecclesiastical faith, 183, 184, 188, 206 216, 226
Epicurus, 11 as postulate of practical reason, 77–78, 173,
Erasmus, 128 225, 234
ethical commonwealth. See ethical community relation to ethical community, 48, 116, 184,
ethical community, 7, 19, 48, 115–16, 175, 205, 210
210–11, 237, 248 relation to the world, 153–54, 225, 229
and coercion, 182 Goethe, Johann Wolfgang, 1, 4
and ethical state of nature, 182 grace, 5, 6, 53, 100, 114, 138, 227, 239, 242
and Islam, 188–91 and nature, 125, 149
and pluralism, 181–86 as incomprehensible, 119, 130, 132
and the individual, 182, 204–05, 210 as providential, 125
as political, 186–88, 205 as supernatural cooperation, 120, 124, 138, 199
ethical state of nature, 182, 204 traditional view of, 118, 126–29
evil, 199–200, 226, 239–41, 249 Groundwork of the Metaphysics of Morals, 2, 7, 8,
and free choice, 32, 240 42, 51, 79, 81, 133, 180, 182, 207
and human nature, 68
and will, 62–63, 67–69, 243 Haller, Albrecht von, 11
compatible with good conduct, 200 happiness, 35, 36, 46, 61, 67, 87, 95, 96, 99, 103,
imputability of, 42 106, 133, 160, 161, 210, 214, 218, 223, 224, 225,
incentives to, 33 235, 238
intelligibility of, 32, 33–34, 49, 51, 240, 243 Hare, John, 118
social context, 45, 46–50, 115, 181, 188, 203, 206, Herder, Johann Gottfried, 1, 22
245–49 Hesiod, 17
Heumann, Christoph August, 158
fact of reason, 140 highest good, 77, 103, 154, 170, 210, 214–15, 216,
faith. See historical faith; rational faith 217, 218, 221, 225, 226, 229, 249
Feuerbach, Ludwig, 118 and Cicero, 221
frailty, 50 and the ethical community, 182
Frankfurt, Harry, 85–86, 87, 124 and the moral law, 225
Frederick William I, “Frederick the Great,” 1, 158 as end of reason, 235
freedom, 5, 49, 75, 78, 85, 87, 112, 114, 128, 130, 134, historical faith, 20, 183, 239, 241
137, 139, 148, 151, 196, 212, 217, 218, 227, 235, considered as a vehicle, 3, 20, 26
240, 241, 247 relation to moral faith, 26
and causality, 59–60, 74, 76, 209, 243 historical religion. See historical faith
and ethical community, 181–82 Hobbes, Thomas, 10, 11, 25, 47, 177, 202
and evil, 72–73 Hodges, H. A., 123
and God, 153–54 Holbach, Baron d’, 10, 11, 14, 30
and good will, 60–61 Homer, 17
and maxims, 81, 222 hope, 5, 6, 77–78, 138, 139, 140, 209, 216, 236
and radical evil, 53 and divine aid, 110–15, 134
as absolute, 140, 148, 152 and miracles, 107–10
external/internal, 202–03, 204–05 and moral progress, 115–17
empirical possibility of, 103
Gesinnung. See disposition formal possibility of, 104–05
God, 8, 10, 17, 98, 110, 119, 150, 170, 193, 196, 208, logical possibility of, 105
212 modal status of, 100–02
and grace, 127 real possibility of, 105–07
and moral discouragement, 228 humanity, 42, 43, 44, 65–67, 222–23
and moral duties, 206 Hume, David, 10, 11, 16, 22, 30, 118, 155
and moral regeneration, 110–15, 121–23
as moral author of the world, 106, 210, 214, 218 Idea for a Universal History, 56, 115, 151
as moral legislator, 19, 210 ideas of reason, 27, 149
264 Index
immortality of the soul and the categorical imperative, 80
as postulate of practical reason, 210, 225, 234 relation to freedom, 81
impurity, 50, 51 relation to intentions, 81
incentives, 5, 40, 51, 52, 58, 65–67, 69, 75, 76, 130, Meirav, Ariel, 101
183, 199 Mendelssohn, Moses, 11, 207
and the passions, 44 Metaphysical Foundations of Natural Science, 12
inclinations, 36, 40, 44, 45, 199, 200, 236, 240, Metaphysics of Morals, 12, 96, 133, 202, 211, 219,
245, 246 230, 231
and evil, 34, 227 Michaelis, Johann David, 158, 207
interests of reason, 9, 99, 225, 235, 236 Michalson, Gordon, 118
Islam, 185 Mill, John Stuart, 98
and the ethical community, 188–91 miracles, 6, 7, 100, 107–10, 113, 117, 162, 167, 169,
Kant’s view of, 185 239
and natural laws, 141–43, 151–52
Jachmann, Reinhold Bernhard, 149 and Neologists, 158
Jacobi, Friedrich Heinrich, 207 and Reimarus, 160
Jerusalem, Johann Friedrich Wilhelm, 158 Kant’s definition of, 169
Jesus, 7, 19, 21, 24, 120, 125, 135, 143 political aspects, 141
and moral perfection, 171 possibility of, 138
and Neologism, 159 relation to “wonder,” “admiration,” 144–48
as an ideal, 172, 224 theistic, 142
as viewed by Reimarus, 161–63 Montaigne, Michel, 115
breaks the power of evil, 172 moral argument for existence of God, 15, 98–99
context for Kant’s views, 168–71 moral feeling, 43, 230
humanity of, 172 moral law, 34, 40, 41, 45, 51, 58, 67, 77, 148, 153,
Kant’s use of name, 171 174, 183, 200, 214, 216, 217, 219, 223, 225
relation to Christian tradition, 157 and the will, 234, 238
Job, 144 called into question, 228
Judaism, 161–63, 166, 185 moral perfection, 46, 133, 171, 219, 236, 249
Kant’s view of, 169 moral progress, 47, 210, 213, 240, 244, 245, 249
relation to hope, 115–17
Kierkegaard, Soren, 128 moral regeneration, 53, 69, 75, 76–78, 239, 243
Kripke, Saul, 105 and religion, 205
as a revolution, 145, 199
Lavater, Johann Kaspar, 11 relation to hope, 110–15
Lectures on the Philosophical Doctrine of Religion, Moser, Friedrich Karl von, 158
128 Moses, 126, 130
Leibniz, Gottfried Wilhelm, 8, 72, 108–09, 175, Muhammad, 125
176, 178–81, 184, 186
Lessing, Gotthold Ephraim, 158, 159, 161, 164 natural religion, 2, 18, 30, 160, 171
Locke, John, 190, 195 necessity, 4, 5, 50, 68, 105, 140, 187, 217
Luther, Martin, 127, 128, 156, 177 Neologists, Neologism, 157–59
view of Chistian teachings, 157–59
Mandeville, Bernard de, 47 Newton, Isaac, 151
Mariña, Jacqueline, 124 Nicene Creed, 156
Martin, Adrienne, 101 Nietzsche, Friedrich, 118
Marx, Karl, 30, 118 Nösselt, Johann August, 158
maxims, maxim making, 5, 40, 43, 45, 46, 50, 54, noumenal, 4, 48, 59, 86, 247
61, 79, 143
and action, 80–81, 83, 84, 86 On Perpetual Peace, 21
and character, 83 original sin. See radical evil, relation to original sin
and evil, 68–69, 71, 130, 227, 240, 242–43
and freedom, 94, 222 Pascal, Blaise, 11
and grace, 120 passions, 43–45, 248
and miracles, 142 Paul, St., 121, 126, 127–29, 131, 157
Index 265
Pelagianism, 128, 129 and search for totality, 224
personality, 42, 43, 53, 65–67, 73, 222–23, 247 as having ends, 218, 235, 236–37, 238–40, 241
Pettit, Philip, 101 in title of book, 23, 207–09
Pietism, 145, 157, 168, 176 redemption. See moral regeneration
Pistorius, Herman Andreas, 158 reflective judgment, 50, 232
Plato, 11, 13 Reimarus, Hermann Samuel, 7, 159–65, 170, 173
postulate, 14, 77–78, 104, 105, 149, 170, 173, 183, and Kant, 168–71
210, 225, 234, 235 compared with Semler, 163–68
Practical Reason, 8, 14, 18, 27, 52, 148, 153, 183, 196, his view of Jesus, 161–63
215, 228, 239, 249 summary of teachings, 163
and classical thought, 221 Reinhold, Karl Leonhard, 147
and search for totality, 224 revealed religion, 2, 3, 15, 183
and trust, 229 relation to morality, 16, 18
multiple sense of, 216–17 revelation, 7, 11, 12, 15, 17, 18, 20, 23, 24, 28, 151,
primacy of, 216, 234 160, 183, 185, 198, 207–08, 212
relation to theoretical, 235 rigorism, 38–41
relation to theoretical reason, 234 Rorty, Richard, 8, 193–96, 198, 206–07, 212
predispositions, 45, 48, 53, 56, 69–70, 75, 200, Rousseau, Jean-Jacques, 10, 11, 20, 30, 47, 115, 146,
206, 246–47 147, 148, 150, 240, 246
and freedom, 42
as aptitudes, 222–23 Sack, Friedrich Wilhelm, 158
contrast with propensity, 240 schemata, 243, 247, 248
propensity to evil, 4, 16, 20, 21, 44, 55, 75, 79, 172, and visibile representation, 241–42
182, 198, 205, 211, 226, 227, 229, 240, 242, Schleiermacher, Friedrich, 22
243, 247–49 Schweitzer, Albert, 161
and social context, 48, 49 Semler, Johann Salomo, 7, 11, 158, 159
and the will, 243–45 and Kant, 168–70
degrees of, 50–52 comparison with Reimarus, 163–68
propensity and freedom, 45 view of Jesus, 165–66
propensity defined, 45 Seneca, 240
pure religious faith. See rational faith Smith, Adam, 46
Spalding, Johann Joachim, 158
Quinn, Philip, 123 Spinoza, Baruch, 8, 10, 11, 25, 106, 153, 175–79, 181,
Qur’an, 189–91 183, 185, 190
and religious pluralism, 178
radical evil, 5, 38, 120, 132, 137, 188, 198, 213, 227, Stoicism, 16, 19
237, 243 Sussman, David, 92
and human nature, 53–57, 69–70 Swedenborg, Emanuel, 28
and moral obligation, 5
and the disposition, 5 Taha, Mahmoud Mohamed, 190
and the Enlightenment, 47, 56 Taylor, Robert S., 192
hidden beneficence, 238, 245 teleology, 7, 9, 41, 50, 137, 147, 149–51, 154, 227,
innateness of, 53–54 249
relation to original sin, 16, 20, 32, 38, 47, 53, 56, Teller, Wilhelm Abraham, 158
69–70, 129 The Conflict of the Faculties, 145, 173, 184
social aspects, 8, 48 theodicy, 7, 144, 147, 148, 154
rational faith, 9, 14, 20, 137–40, 147, 154, 207, 216, The Only Possible Argument in support of a
224, 229, 237 Demonstration of the Existence of God, 150,
as inadequate translation, 137, 216 151, 160
relation to knowledge, 139 Thielecke, Helmut, 163, 164
Rawls, John, 8, 175–76, 186–87, 188, 190–92, title explained, 2, 21, 22–23
195
reason, 22, 23, 24, 29, 58, 223, 224, 246 unsociable sociability. See evil, social context
and its limits, 119, 183, 197, 228, 233, 234, 239
and propensity to evil, 46 Voltaire, 10, 11, 18, 21, 30, 98
266 Index
What does it Mean to Orient Oneself in Thinking?, Wolf, Susan, 85
207 Wolff, Christian von, 163
What is Enlightenment?, 198 Wolterstorff, Nicholas, 8, 193, 194–96, 198, 206,
Wheatley, J., 101 207, 211
will, 52, 64, 74, 199, 217–19, 238, 247 Wood, Allen W., 186
and evil, 217–19, 243
and moral regeneration, 75, 243–45 Xenophanes, 13
and practical reason, 217–19
as power of choice, 222–23 Zoellner, Johann Karl Friedrich, 158

(Cambridge Critical Guides) Michalson, Gordon E - Kant's Religion Within The Boundaries of Mere Reason - A Critical Guide-Cambridge University Press (2014)

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

(Cambridge Critical Guides) Michalson, Gordon E - Kant's Religion Within The Boundaries of Mere Reason - A Critical Guide-Cambridge University Press (2014)

Uploaded by

Copyright:

Available Formats

KANT’S RELIGION WITHIN

THE BOUNDARIES OF MERE REASON

Kant’s Religion within the Boundaries of Mere Reason was written

gordon e. michalson is Professor of Humanities at

Titles published in this series:

Kant’s Idea for a Universal History with a Cosmopolitan Aim

Kant’s Groundwork of the Metaphysics of Morals

Kant’s Critique of Practical Reason

Wittgenstein’s Philosophical Investigations

Nietzsche’s On the Genealogy of Morality

Cambridge University Press is part of the University of Cambridge.

List of contributors page ix

karl ameriks is McMahon-Hank Professor in the Department of

ApH Anthropology from a Pragmatic Point of View Ak 7

IC Proclamation of the Imminent Conclusion of a Treaty of Perpetual

P Prolegomena to Any Future Metaphysics that Will be Able to Come

Even by Kant’s standards, Religion within the Boundaries of Mere Reason is a

further connects Religion to Kant’s Conﬂict of the Faculties in which, ﬁve

questions reﬂecting the interests of reason that Kant originally introduced in

Holy Scriptures within the boundaries

1 Two Preliminary Remarks

2 A look at the theology of the Three Critiques

3 New additions made by Religion Within

a mere faith of reason which can convince everybody. By contrast, a faith

5 A philosophical hermeneutics of the Bible

This conclusion would contradict the three Critiques since natural,

opposition is characterized by a government that directs the interests of the

I want to draw a preliminary conclusion, here: Enlightenment philoso-

The evil in human nature

1 The moral concept of evil

2 Two wrong ways to look at evil

merely natural event that is to be causally explained by the mechanism of

This criticism is closely allied to another erroneous charge commonly

of radical evil is meant to direct our attention primarily to evil in

We cannot observe maxims, we cannot do so unproblematically even within

It may be tempting to read this passage as alluding to some sort of “a priori”

4 Human predispositions to good

basic predispositions: animality, humanity, and personality. All are “origi-

5 Our nature as what is given and as what we make

onto” humanity: envy, ingratitude, and gloating over harm to another

7 The propensity to evil

8 Evil and sociability

must be social (an “ethical community” or “people of God”) because the

unconscious (natural) purpose, in promoting the historical development of

This is not a causal explanation of the propensity to evil (such as might,

9 Deception and the three degrees

relation to other reasons, such as those of self-love or inclination. Because

10 Humanity as evil “by nature”

“inscrutable”) but to the observed fact that there is no identiﬁable temporal

11 The “proof ” of radical evil

Radical evil and human freedom

1 Kant’s view of human freedom

1.1 The human person

1.3 Free will and good will

2 Kant’s account of radical evil

2.1 The moral dilemma of radical evil

2.2 Incentives of animality, humanity, and personality

2.3 Evil will

2.4 Radical evil

predisposition to good they could not be evil.22 This is Kant’s philosophical

disharmony in our power of choice with respect to the way it incorporates

2.5 Shadow of freedom

2.6 Nevertheless and in spite of, against and instead of