Understanding Mor al Obligation

In many histories of modern ethics, Kant is supposed to have ushered in

an anti-realist or constructivist turn by holding that unless we ourselves
‘author’ or lay down moral norms and values for ourselves, our auton-
omy as agents will be threatened. In this book, Robert Stern challenges
the cogency of this ‘argument from autonomy’, and claims that Kant
never subscribed to it. Rather, it is not value realism but the apparent
obligatoriness of morality that really poses a challenge to our autonomy:
how can this be accounted for without taking away our freedom? The
debate the book focuses on therefore concerns whether this obligato-
riness should be located in ourselves (Kant), in others (Hegel) or in
God (Kierkegaard). Stern traces the historical dialectic that drove the
development of these respective theories, and clearly and sympathetic-
ally considers their merits and disadvantages; he concludes by arguing
that the choice between them remains open.

robe rt st e r n is Professor of Philosophy at the University of

Sheffield. He is the author of Hegel, Kant and the Structure of the Object
(1990), Transcendental Arguments and Scepticism: Answering the Question
of Justification (2000), Hegel and the ‘Phenomenology of Spirit’ (2002), and
Hegelian Metaphysics (2009). He is editor of G.W.F. Hegel: Critical Assessments
(1993) and Transcendental Arguments: Problems and Prospects (1999).
 mode r n eu rope a n ph i lo s oph y
General Editor
WAYNE M ARTIN, University of Essex

Advisory Board
SEBASTIAN GARDNER , University College, London
BEATRIC E H AN - PILE , University of Essex
H ANS SLUGA , University of California, Berkeley

Some recent titles

Michelle Grier, Kant’s Doctrine of Transcendental Illusion
Henry Allison, Kant’s Theory of Taste
Allen Speight, Hegel, Literature, and the Problem of Agency
J. M. Bernstein, Adorno
Will Dudley, Hegel, Nietzsche, and Philosophy
Taylor Carman, Heidegger’s Analytic
Douglas Moggach, The Philosophy and Politics of Bruno Bauer
Rüdiger Bubner, The Innovations of Idealism
Jon Stewart, Kierkegaard’s Relations to Hegel Reconsidered
Michael Quante, Hegel’s Concept of Action
Wolfgang Detel, Foucault and Classical Antiquity
Robert M. Wallace, Hegel’s Philosophy of Reality, Freedom, and God
Johanna Oksala, Foucault on Freedom
Béatrice Longuenesse, Kant on the Human Standpoint
Wayne Martin, Theories of Judgment
Heinrich Meier, Leo Strauss and the Theologico-Political Problem
Otfried Höffe, Kant’s Cosmopolitan Theory of the Law and Peace
Béatrice Longuenesse, Hegel’s Critique of Metaphysics
Rachel Zuckert, Kant on Beauty and Biology
Andrew Bowie, Music, Philosophy and Modernity
Paul Redding, Analytic Philosophy and the Return of Hegelian Thought
Kristin Gjesdal, Gadamer and the Legacy of German Idealism
Jean-Christophe Merle, German Idealism and the Concept of Punishment
Sharon Krishek, Kierkegaard on Faith and Love
Nicolas de Warren, Husserl and the Promise of Time
Benjamin Rutter, Hegel on the Modern Arts
Anne Margaret Baxley, Kant’s Theory of Virtue
David James, Fichte’s Social and Political Philosophy
Espen Hammer, Philosophy and Temporality from Kant to Critical Theory
U n de r sta n di ng mor a l
obl ig at ion
Kant, Hegel, Kierkegaard

Robert Stern
University of Sheffield
 c a mbr idge u ni v er sit y pr ess
Cambridge, New York, Melbourne, Madrid, Cape Town,
Singapore, São Paulo, Delhi, Tokyo, Mexico City
Cambridge University Press
The Edinburgh Building, Cambridge c b2 8ru, UK

Published in the United States of America by Cambridge University Press, New York

www.cambridge.org
Information on this title: www.cambridge.org/9781107012073

© Robert Stern 2012

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 2012

Printed in the United Kingdom at the University Press, Cambridge

A catalogue record for this publication is available from the British Library

Library of Congress Cataloguing in Publication data

Stern, Robert, 1962–
Understanding moral obligation : Kant, Hegel, Kierkegaard / Robert Stern.
p. cm. – (Modern European philosophy)
Includes bibliographical references and index.
i sb n 978-1-107-01207-3
1. Duty. 2. Responsibility. 3. Kant, Immanuel, 1724–1804.
4. Hegel, Georg Wilhelm Friedrich, 1770–1831.
5. Kierkegaard, Søren, 1813–1855. I. Title.
bj1451.s69 2012
170.92′2–dc23
2011039382

i sb n 978-1-107-01207-3 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.
Ode to Duty
William Wordsworth
Stern Daughter of the Voice of God!
O Duty! if that name thou love,
Who art a Light to guide, a Rod
To check the erring, and reprove;
Thou, who art victory and law
When empty terrors overawe;
From vain temptations dost set free;
And calm’st the weary strife of frail humanity!

Kind Of An Ode To Duty

Ogden Nash
O Duty,
Why hast thou not the visage of a sweetie or a cutie?
Why displayest thou the countenance of the kind of conscien-
tious organizing spinster
That the minute you see her you are aginster?
Why glitter thy spectacles so ominously?
Why art thou clad so abominously?
Why are thou so different from Venus
And why do thou and I have so few interests mutually
in common between us?
Why art thou fifty per cent martyr
And fifty-one per cent Tartar?
 C on t e n t s

Acknowledgements page ix
List of references and abbreviations xi

Introduction 1

Pa r t i K a n t 5
1 Kant, moral realism, and the argument from autonomy 7
2 The argument from autonomy and the problem of 41
moral obligation
3 Kant’s solution to the problem of moral obligation 68

Pa r t i i H e ge l  101
4 Hegel’s critique of Kant (via Schiller) 103
5 Hegel’s solution to the problem of moral obligation 148

Pa r t i i i K i e r k e g a a r d 171
6 Kierkegaard’s critique of Hegel 173
7 Kierkegaard’s solution to the problem of moral obligation 204
Conclusion: From Kant to Kierkegaard – and back again? 220

Bibliography 255
Index 273

vii
 Ack now l e d ge m e n t s

I have incurred many debts in writing this book, both institutional and
personal.
At an institutional level, I am especially grateful to the Leverhulme
Trust, who kindly awarded me a Major Research Fellowship for
2008–10, during which time most of this book was written. I am also
grateful to the University of Sheffield for allowing me a further year
of research in which to complete it. To have such amounts of time at
one’s disposal is a rare privilege; I can only hope to have made good
use of it.
This book has also benefitted hugely from conversations (face-to-
face and virtual) with others working in the field. I would particu-
larly like to thank: Karl Ameriks, Sorin Baiasu, Anne Margaret Baxley,
Chris Bennett, Sarah Broadie, Katerina Deligiorgi, Paul Formosa,
Paul Franks, Fabian Freyenhagen, John Haldane, Béatrice Han-Pile,
Rob Hopkins, Stephen Houlgate, Ian Hunter, Terry Irwin, Sue James,
Petter Korkman, Pauline Kleingeld, Charles Larmore, Jimmy Lenman,
John McDowell, Bill Mander, David Owens, Steve Makin, George
Pattison, Tom Pink, Adrian Piper, Terry Pinkard, Robert Pippin, Mike
Ridge, Joe Saunders, Ulrich Schlösser, Clemens Schwaiger, Oliver
Sensen, Yonatan Shemmer, John Skorupksi, Sergio Tenenbaum, Oliver
Thorndike, Jens Timmermann, Daniel Viehoff, Andrew Vincent, Dan
Watts, Marcus Willaschek, Ken Westphal, and anonymous referees. I
have also benefitted from comments from participants in various sem-
inars and conferences.
I am grateful to Carlton Books, Curtis Brown, and Frances Nash
Smith, for their permission to reproduce the fragment of Ogden Nash’s
poem ‘Kind of an Ode to Duty’, from the book Candy is Dandy: The
Best of Ogden Nash by Ogden Nash, with an Introduction by Anthony

ix
x Acknowledgements

Burgess, published by Andre Deutsch (Methuen, 1st edition 1985),

originally appearing in the publication in 1938, by Little, Brown &
Company. I am also grateful to be able to reproduce a fragment of
William Wordsworth’s ‘Ode to Duty’, from The Complete Poetical Works
of William Wordsworth, edited by Henry Reed (Philadelphia: James Kay,
1837).
Wayne Martin has been particularly supportive and encouraging in
the publication process for the book, as have Hilary Gaskin and the
rest of the staff at Cambridge University Press.
Finally, of course, a book of this sort is in no position really to repay
the obligations I owe to my family in writing it, however those obliga-
tions are to be understood.
 R e f e r e nc e s a n d a bbr e v i at ion s

In general, I have used the author–date system of referencing.

However, in the case of frequently cited authors – Aristotle, Hegel,
Kant, Kierkegaard, and Schiller – I have used a system of abbreviations
(see below), where details of the works are given in the Bibliography.
Also, for pre-Kantian texts where there is no standard edition or trans-
lation to refer to, as well as giving page numbers, I have included infor-
mation on where in the work the reference occurs (e.g. chapter or
section numbers), where the work is structured in such a way as to
make this helpful. The hope is that this will enable readers with differ-
ent editions to find the reference reasonably easily.

Works by Aristotle
References are given using the standard ‘Bekker numbers’, and to the
book, section, and page numbers in the following translations:
EE Eudemian Ethics, translated by J. Solomon, in Aristotle 1984: ii,
1922–81
NE Nicomachean Ethics, translated by W. D. Ross, revised by J. O.
Urmson, in Aristotle 1984: ii, 1729–1867

Works by Hegel
References are given first to the volume and page number of the Theorie
Werkausgabe edition of Hegel’s works (Hegel 1969–71), except in the
case of some references to Hegel’s lectures on the history of philosophy,
which are given to Hegel 1986 and Hegel 1989, and in the case of refer-
ences to Hegel’s lectures on the philosophy of religion, which are given

xi
xii list of References and abbreviations

to Hegel 1984. Where appropriate, references are also given to section

numbers (where ‘Z’ indicates a reference to a ‘Zusatz’ or student note).
References are then given to one of the following translations:
DFS ‘The Difference Between Fichte’s and Schelling’s System of
Philosophy’, translated by H. S. Harris and Walter Cerf, in
Hegel 1977a
EL Logic: Part i of the Encyclopaedia of the Philosophical Sciences,
translated by T. F. Geraets, W. A. Suchting, and H. S. Harris,
in Hegel 1991b
EN Philosophy of Nature: Part i i of the Encyclopaedia of the Philosophical
Sciences, translated by M. J. Petry, in Hegel 1970
LA Aesthetics: Lectures on Fine Art, translated by T. M. Knox, in
Hegel 1975
LHP Lectures on the History of Philosophy, translated by E. S.
Haldane and Frances H. Simson, in Hegel 1892–96
LHPii Lectures on the History of Philosophy 1825–6, volume i i : Greek
Philosophy, translated by Robert F. Brown and J. M. Stewart,
in Hegel 2006
LHPiii Lectures on the History of Philosophy 1825–6, volume i i i : Medieval
and Modern Philosophy, translated by Robert F. Brown and J.
M. Stewart, in Hegel 2009
LPRiii Lectures on the Philosophy of Religion, volume i i i : The Consummate
Religion, translated by R. F. Brown, Peter C. Hodgson, and
J. M. Stewart, in Hegel 2007
PR Elements of the Philosophy of Right, translated by H. B. Nisbet,
in Hegel 1991a
PS Phenomenology of Spirit, translated by A. V. Miller, in Hegel
1977b
SC ‘The Spirit of Christianity and its Fate’, translated by T. M.
Knox, in Hegel 1948: 182–301
SL Science of Logic, translated by A. V. Miller, in Hegel 1969

Works by Kant
References are given first to the volume and page number of the
Akademie edition of Kant’s works (Kant 1900–), except for references
to CPR, which are given to the pagination of the first (A) and second
(B) editions in the standard manner. References are then given to one
of the following translations:
 list of References and abbreviations xiii

CF The Conflict of the Faculties, translated by Mary J. Gregor and

Robert Anchor, in Kant 1996b: 233–327
CJ Critique of the Power of Judgment, translated by Paul Guyer and
Eric Matthews, in Kant 2000
CPR Critique of Pure Reason, translated by Paul Guyer and Allen
Wood, in Kant 1998
CPrR Critique of Practical Reason, translated by Mary J. Gregor, in
Kant 1996a: 133–272
GMM Groundwork of the Metaphysics of Morals, translated by Mary J.
Gregor, in Kant 1996a: 37–108
LE Lectures on Ethics, translated by Peter Heath, in Kant 1997
LR Lectures on the Philosophical Doctrine of Religion, translated by
Allen W. Wood, in Kant 1996b: 335–452
MM The Metaphysics of Morals, translated by Mary J. Gregor, in
Kant 1996a: 353–604
NM ‘Notes on Moral Philosophy’, translated by Curtis Bowman,
Paul Guyer and Frederick Rauscher, in Kant 2005: 405–78
PE ‘Inquiry Concerning the Distinctness of the Principles of
Natural Theology and Morality’ [the ‘Prize Essay’], trans-
lated by David Walford, in Kant 2002a: 243–86
Relig Religion within the Boundaries of Mere Reason, translated by
George di Giovanni, in Kant 1996b: 39–216
TSP ‘On a Recently Prominent Tone of Superiority in Philosophy’,
translated by Henry Allison, in Kant 2002b: 429–45

Works by Kierkegaard
For Kierkegaard’s published writings, references are given first to
the volume and page number of Søren Kierkegaard’s Samlede Værker
(Kierkegaard 1901–06), and then to one of the following translations:
CUP Concluding Unscientific Postscript to ‘Philosophical Fragments’,
translated by Howard V. Hong and Edna H. Hong, in
Kierkegaard 1992
EO Either/Or, Part ii, translated by Howard V. Hong and Edna H.
Hong, in Kierkegaard 1987
FT Fear and Trembling: A Dialectical Lyric, translated by Sylvia
Walsh, in Kierkegaard 2006
PC Practice in Christianity, translated by Howard V. Hong and
Edna H. Hong, in Kierkegaard 1991
xiv list of References and abbreviations

SLW Stages on Life’s Way, translated by Howard V. Hong and Edna H.

Hong, in Kierkegaard 1988
WL Works of Love, translated by Howard V. Hong and Edna H.
Hong, in Kierkegaard 1995
For Kierkegaard’s journals and papers, references are given first to the
volume and page number of Søren Kierkegaard’s Papirer (Kierkegaard
1909–78), and then to the following translation:
JP  Søren Kierkegaard’s Journals and Papers, translated by Howard V.
Hong and Edna H. Hong, in Kierkegaard 1999.

Works by Schiller
References are given first to the volume and page number of Schiller’s
Werke: Nationalausgabe (Schiller 1943–), and then to one of the follow-
ing translations:
AE On the Aesthetic Education of Man in a Series of Letters, trans-
lated by Elizabeth M. Wilkinson and L. A. Willoughby, in
Schiller 1967
GD On Grace and Dignity, translated by Jane V. Curran, in
Schiller 2005
KL ‘Kallias or Concerning Beauty: Letters to Gottfried
Körner’, translated by Stefan Bird-Pollan, in Schiller 2003
MUAM ‘The Moral Utility of Aesthetic Manners’, translated
anonymously, in Schiller 2006: 94–100
NLBF ‘On the Necessary Limitations in the Use of Beautiful
Forms’, translated anonymously, in Schiller 2006: 168–84
OS ‘On the Sublime’, translated anonymously, in Schiller
2006: 101–9
 I n t roduc t ion

My hope for this book is that it will shed light on the issues discussed at
two levels: at the level of the history of ideas, in showing the role these
issues have played in the thought of Kant, Hegel, and Kierkegaard, and
their period more generally; and at the philosophical level, in helping
us to understand these issues more clearly in a systematic way.
As regards the first, more historical, level, my aim is to offer an
account of a central strand in the history of modern ethics from the
mid eighteenth to mid nineteenth centuries, an account which dif-
fers from what I think has become the standard story. According to
this story, a new turn in ethics is taken when Kant (in part foreshad-
owed by other figures such as Rousseau) introduces a radical notion
of autonomy into ethical thinking, whereby autonomy is seen to
require that all forms of moral realism are rejected; this ‘argument
from autonomy’ (as I will call it) is then said to lead Kant to replace
this realist conception with one whereby ethics is now grounded in
the self-legislating moral subject. However, despite its appeal to the
modern mind, this picture of self-legislation is seen to raise certain
fundamental difficulties, particularly the threat of emptiness: if no
prior set of moral values obtain, what is to guide the legislating sub-
ject, and to prevent the act of legislation from becoming groundless?
It is this problem and related ones that are said on the standard story
to constitute what is sometimes called the ‘Kantian paradox’, where
this paradox is supposed to set the agenda for Kant’s successors, such
as Hegel and Kierkegaard.
More will be said about this standard story in Chapter 1, where I will
also argue that it is mistaken. This will involve looking at the argu­ment
from autonomy itself in some detail. I will claim that this argument
is harder to make plausible than it may seem, and that there is little

1
2 Understanding Mor al Obligation

reason to think that Kant himself would have endorsed it in any radical
form, given that his own position retains important elements of moral
realism (or so I maintain). In fact, I will suggest, while clearly giving
considerations of autonomy a central role within his ethics, that Kant
saw these considerations in a much narrower way than his anti-real-
ist and constructivist interpreters have assumed; for, it is only when it
comes to accounting for the obligatoriness of certain actions, rather than
their moral goodness or rightness, that the concern about autonomy
really leads towards self-legislation for Kant, with the idea that other-
wise the ground of this obligatoriness might be some external lawgiver
or authority, as on certain sorts of divine command theory. It is this nar-
rower concern, I will argue, that frames what I will call the ‘problem of
moral obligation’ for Kant – namely, the problem of accounting for the
imperatival or binding force of morality in a way that makes this com-
patible with our autonomy – rather than moral values as such, which
on their own pose no such threat. In response to this problem, I argue,
Kant offers what I call a ‘hybrid’ theory, which treats the obligatoriness
of morality as a function of our limited moral nature and the fact that
our non-moral desires need to be constrained by reason in order to do
what is right in a self-legislative manner: for us, therefore, the right and
the good appear to be necessitating, in contrast to a holy will, for whom
morality does not take the form of commands.
My aim in the rest of the historical narrative is to follow out the
development from Kant to Hegel to Kierkegaard once this problem of
moral obligation is taken as our starting point, rather than the prob-
lems posed by the Kantian paradox which forms the starting point on
the standard story, but which, on my picture of Kant’s position, is not
the key issue for Kant or for his successors. So, instead of following the
standard story which sees Hegel as being preoccupied with difficulties
created by Kant’s own way of dealing with the Kantian paradox, I will
claim that, instead, Hegel faced the difficulties created by Kant’s way
of dealing with the problem of moral obligation: for this was seen by
Hegel to rely on a dualistic view of the will and an alienating picture of
our relation to morality, as requiring an element of constant struggle.
I will therefore suggest that this dissatisfaction with Kant led Hegel to
offer a different solution to the problem of moral obligation, by put-
ting forward a ‘social command’ account, which treats duty as arising
from the constraints imposed on us by others. For Kierkegaard, how-
ever, because Hegel’s solution to the problem of moral obligation was
intended to avoid any Kantian tension between duty and inclination,
 Introduction 3

this meant that the social command account could not treat moral-
ity as asking too much of us as individuals; it thus threatened to ren-
der our moral lives too complacent by reducing the moral demand.
Kierkegaard held that only by returning to something more like the
divine command theory that Kant had rejected could this demand be
restored to the right level, where it again makes sense to think of mor-
ality as presenting us with a challenge that we must struggle to fulfil.
This return to a divine command theory, however, brings us back to
face the argument from autonomy in the way I claim Kant understood
it: namely, that any such divine command theory will introduce an
insupportable element of heteronomy into ethics.
This, then, sets up a kind of dialectical ‘circle’ of positions, each
with their respective advantages and disadvantages, and the aim of
my discussion will be to explore these positions further, and show how
they can be compared in more detail. The attraction of Kant’s pos-
ition might be that it accounts for the puzzling obligatoriness of mor-
ality, while doing away with the need for any external lawgiver who
has authority and power over us. However, the price of this Kantian
account (as we shall see) is that it appears to rely on a dualistic pic-
ture of the human will torn between reason and desire, where acting
morally requires a kind of battle with our non-moral inclinations. The
attraction of Hegel’s position is that it does away with this dualism
and sense of struggle, but the price is that to overcome this dualism it
seems forced to reduce the level of the moral demand, and to make
our moral requirements too easy to satisfy. And the attraction of the
Kierkegaardian position is that this strenuousness is restored, but at
the apparent price of introducing a divine command model, which in
turn leads us back to Kantian concerns over autonomy.
The book therefore sets up a comparative study of these three
thinkers, not as regards their ethical outlooks in their entirety, but
as regards what I am focusing on in talking about the ‘problem of
moral obligation’, namely ‘what gives moral obligations their binding
or constraining character?’ – ourselves, others, or God? These think-
ers of course have many other points of agreement and contestation,
and the problem of moral obligation can doubtless be understood
more broadly:1 but these are the disagreements and the issue that will
1
For example, it is often a focus of discussions of moral obligation that the reasons to
act in accordance with them are said to be categorical, overriding, and universal; but
while this may or may not be the case, as Darwall points out, there is arguably more
to them than that, as the same is plausibly true of the reasons to follow other norms,
4 Understanding Mor al Obligation

concern us here, namely how it is, as Pufendorf put it, that ‘obligation
places a kind of bridle on our liberty’.2 While other features of such
obligations will come up, it is this that will be the main issue, as it is
this that most obviously raises the question of autonomy as it arises for
the obligatoriness of the moral.
As well as offering a new perspective on this set of historical ques-
tions and relations, I hope also that my discussion will cast some light
on the philosophical issues underlying them concerning autonomy,
moral realism, moral obligation, divine command theories, and so on.
For it is often by seeing new philosophical options that one can see the
historical story differently, and vice versa. It will be argued that these
issues lie at the heart of the problem of moral obligation, in a way
that makes it so intractable: there are many competing pressures on a
satisfactory solution, and the considerations that tell for and against
each side run deep. The problem, therefore, is of more than merely
historical interest, as the questions that it raises remain at the heart of
current philosophical debate.

such as those of logic and scientific reasoning, where it is in having the force of a
demand that such differences can be said to consist. See Darwall 2004: 110–11 and
2006: 13–14, 26–7, and 10: ‘I argue that moral requirements are connected conceptu-
ally to an authority to demand compliance’.
2
Pufendorf 1682: Book i, Chapter 2, §3, p. 13/1991: 27. While it is probably unwise to
place too much weight on etymology, it is perhaps notable that ‘obligation’ comes from
‘ob [to] + ligare [bind]’. See also Brandt 1964: 386 and 391, and Crisp 2006: 34.
Pa rt i

K ant
 1

K a n t, mor a l r e a l i sm , a n d t h e
a rgu m e n t f rom au tonom y

My aim in this chapter is to differentiate the account I want to give

of Kant’s ethics from what, in the Introduction, I called ‘the stand-
ard story’. I will therefore begin by setting out that story, and the way
in which it interprets Kant and then the history of modern ethics
that comes after him. Crucial to my re-telling will be the question of
whether or not Kant subscribed to the argument from autonomy as
that is put forward today by various constructivists and anti-realists in
Kant’s name; my claim will be that he did not. I will then argue that
as a result, therefore, we should recognise that the standard story has
misidentified what Kant saw as the major issue, which was not to pre-
serve our autonomy by finding an alternative to moral realism as an
account of moral values, but to do so by finding an alternative to div-
ine command theories as an account of moral obligation. The nature of
this alternative account will then be considered in the next chapter.

The history of modern ethics: the standard story

I believe that very few would disagree with J. B. Schneewind’s obser-
vation that ‘[t]he conception of morality as autonomy was Kant’s fun-
damental innovation in moral philosophy’.1 But more contentious is
exactly what that conception of autonomy consisted in, and what it
committed Kant to in ethics and meta-ethics. According to most con-
temporary readers of Kant, the answer is that it committed him to
a form of constructivism, and thus to a form of anti-realism in meta-
ethics; and it further committed Kant and his successors to working
out the issues and problems posed by such a position. The reason this

 Schneewind 2002: 88/2010: 245.

1

7
8 Understanding Mor al Obligation

is so, according to these interpreters, is that autonomy as Kant con-

ceived it is incompatible with moral realism, so that Kant was obliged
as a result to reject moral realism and to move towards constructivism.
This connection is particularly prominent in the work of John Rawls:
just as the position of ‘Kantian constructivism’ was one that was fam-
ously first identified as such by Rawls,2 so the reading that treats Kant
as committed to the argument from autonomy goes back to his work
also, where his reading has then been extremely influential on others.3
To understand the standard story of the history of modern ethics, we
must therefore look at this connection more closely, beginning with a
consideration of constructivism.
The distinction between constructivism and realism is a complex
one, and will be considered further in what follows, but a helpful pre-
liminary characterisation of it is offered by Sharon Street:
[T]he key point at issue between realists and antirealists is the answer to
the central question of Plato’s Euthyphro (in roughly secular paraphrase),
namely whether things are valuable ultimately because we value them
(antirealism), or whether we value things ultimately because they pos-
sess a value independently of us (realism). In the final analysis, in other
words, is normativity best understood as conferred or recognized?
Metaethical constructivism falls squarely on the antirealist side of
this divide … [M]etaethical constructivism asserts a counterfactual
dependence of value on the attitudes of valuing creatures; it under-
stands reason-giving status as conferred upon things by us. According

2
Most notably in Rawls 1980.
3
A similar reading of the history of modern ethics which (as far as I know) is independ-
ent of Rawls’s, but which lacks any comparable influence, can be found in Olafson
1967, especially pp. 38–47. Olafson both attributes the argument from autonomy to
Kant, and sees him as adopting the constructivist’s response to it: ‘Freedom as auton-
omy means that the principle of our action must not itself be derived from any exter-
nal source whatsoever, and that all action under principles that have such an external
origin must be viewed as being under a special kind of constraint … Instead of its
being our duty to will what is good, the morally good is that which can be willed in a
certain way; and it is the will itself that by willing establishes the duty to which it is then
subject … On this interpretation, the will is rational not by virtue of accepting and
translating into action moral truth that the intellect apprehends, but by actualizing
its own peculiar virtue of consistency. The novelty of this view resides in the fact that
the rational or logical controls over the will have been introjected into the will itself,
so that any maxim of conduct that the will can accept while remaining faithful to its
essential nature becomes ipso facto morally right’ (pp. 39–41). On the basis of this read-
ing, Olafson sees Kant as the source of a ‘philosophical voluntarist’ tradition in ethics
that leads to existentialism, in opposition to the earlier more realist ‘intellectualist
tradition’. See also Silber 1959.
 Kant, mor al realism, and autonomy 9

to metaethical constructivism, there are no facts about what is valuable

apart from facts about a certain point of view on the world and what is
entailed from within that point of view.4

On this account, the recognitional view to which constructivism is

opposed counts as realist because the activity of practical reason in
telling us how to act is to be measured against a prior order of values,
whereas the constructivist view counts as anti-realist because the order
of dependence is reversed. A constructivist Kantian such as Christine
Korsgaard therefore commits herself to this kind of reversal on Kant’s
behalf when she writes:
The point I want to emphasize here is that the Kantian approach frees
us from assessing the rationality of a choice by means of the apparently
ontological task of assessing the thing chosen: we do not need to iden-
tify especially rational ends. Instead, it is the reasoning that goes into
the choice itself – the procedure of full justification – that determines
the rationality of the choice and so certifies the goodness of the object.
Thus the goodness of rationally chosen ends is a matter of the demands
of practical reason rather than a matter of ontology.5

Or, as Rawls has put the point even more succinctly, on the construct-
ivist view ‘practical reason constructs for the will its own object out of
itself and does not rely on a prior and antecedent order of values’.6

4
Street 2010: 370–1. See also Cullity and Gaut 1997: 4, and Hills 2008: 182–3.
5
Korsgaard 1983: 183/1996a: 261. See also Korsgaard 1996a: 407: ‘Does Kant think,
or should a Kantian think, that human beings simply have unconditional or intrinsic
value, or is there a sense in which we must confer value even upon ourselves? … I now
hold [the latter view]’; and Korsgaard 1996c: 19: ‘According to [realism], moral claims
are normative if they are true, and true if there are intrinsically normative entities or
facts which they correctly describe … [By contrast] Kantians believe that the source of
the normativity of moral claims must be found in the agent’s own will, in particular in
the fact that the laws of morality are the laws of the agent’s own will and that its claims
are ones that she is prepared to make on herself’; and Korsgaard 1998: xxiii: ‘[A]s
rational beings we make the law, we legislate it. Suppose, for instance, I undertake a
program of scientific research … [M]y choice is an act of legislation: I lay it down, for
myself and others, that this research is good, and shall be pursued. We may say that I
confer a value upon scientific research, when I choose to pursue it’.
6
Rawls 2000: 230, also p. 241: ‘The observation about constructivism concerns the rela-
tion of priority between the order of values and the conceptions implicit in our prac-
tical reason. By contrast with rational intuitionism, constructivism sees the substantive
principles that express the order of moral values as constructed by a procedure the
form and structure of which are taken from the conceptions and principles implicit in
our practical reasoning’. Cf. Herman 1993: 215, who glosses Rawls’s Kantian construct-
ivism as offering a positive answer to the question: ‘Can formal rational constraints be
or constitute a conception of value?’.
10 Understanding Mor al Obligation

Now there are a number of arguments that can be given in favour of

constructivism and against realism. For example, it can be argued that
constructivism is more consistent with naturalism; that it fits better
with a motivational internalism; and that it provides a better answer
to the sceptical question of why one should act morally. In response,
realists can adopt various replies. For example, it can be argued that
realism can be made consistent with naturalism when this is properly
understood, or alternatively that naturalism is itself questionable; real-
ists can provide their own accounts of moral motivation; and they can
claim to be able to answer the question of moral scepticism, or that the
question itself is ill-conceived and so does not require an answer.
As well as these considerations, however, constructivists have also
offered the argument from autonomy in favour of their position, claim-
ing that moral realism is a threat to our autonomy as agents, so that if
the former were true, the latter would be undermined. By contrast, it
is claimed that this problem does not plague the constructivist, in so
far as on their account, the moral realm is not constituted by anything
outside our will, which therefore remains autonomous.
Now, for those who read Kant as this kind of constructivist,7 it is the
argument from autonomy that has been treated as the predominant
motivation for his rejection of realism. It is easy to see why this should
be seen as the decisive issue for Kant. His own commitment to tran-
scendental idealism makes it unlikely that naturalistic considerations
should play a major role, while it can be claimed that Kant did not
take the threat of moral scepticism as seriously (or in the same way)
as many modern philosophers.8 Equally, in so far as internalism that
rejects realism often rests on Humean assumptions, which are ones
that Kant himself did not share, it may also seem unlikely that this
would be the basis for him to turn against the realist position.9

7
Not everyone who thinks of themselves as adopting a Kantian constructivism takes
this to involve a stance on meta-ethical issues, and thus any implication either way
regarding realism; and some have claimed that constructivism is ‘neutral’ or agnos-
tic on meta-ethical questions, while others have taken it to be realist at some level but
anti-realist at others. For a helpful taxonomy of such different approaches with further
references, see Galvin forthcoming.
8
This is a complex issue, of course, that cannot be discussed fully here; but for some use-
ful remarks on Kant’s attitude to moral scepticism, see Timmermann 2007: 129–30. I
consider the matter further in Stern 2010.
9
Cf. Darwall 1995: 331, who presents the issue of autonomy as the key issue in Kant’s
turn towards internalism.
 Kant, mor al realism, and autonomy 11

The issue of autonomy, however, has a considerably greater prima

facie claim to plausibility as what might have led Kant to become a con-
structivist. For, of course, Kant was much concerned with autonomy as
a topic, so that it seems plausible to think that reflection on it would
have drawn him away from realism, motivated by something like the
argument from autonomy. In this way, it is claimed, Kant arrived at
his distinctive doctrine of ‘self-legislation’, which is interpreted in con-
structivist terms, and based on the desire to preserve our freedom in
the face of the threat posed to it by moral realism. Kant is thus said to
relate autonomy to self-legislation in this way when he writes: ‘Hence
the will is not merely subject to the law but subject to it in such a way
that it must be viewed as also giving the law to itself [als selbstgesetzge-
bend] and just because of this as first subject to the law (of which it can
regard itself as its author [Urheber])’.10
This, indeed, is the way Rawls himself presents the situation:
Another and deeper meaning of autonomy [than ‘doctrinal auton-
omy’] says that the order of moral and political values must be made, or
itself constituted, by the principles and conceptions of practical reason.
Let us refer to this as constitutive autonomy. In contrast with rational
intuitionism, constitutive autonomy says that the so-called independent
order of values does not constitute itself but is constituted by the activ-
ity, actual or ideal, of practical (human) reason itself. I believe this, or
something like it, is Kant’s view. His constructivism is deeper and goes
to the very existence and constitution of the order of values. This is part
of his transcendental idealism. The intuitionist’s independently given
order of values is part of the transcendental realism Kant takes his tran-
scendental idealism to oppose.11

10
GMM 4:431 (p. 81).
Rawls 1993: 99–100. Cf. Neiman 1994: 33: ‘Having declared that reason is in the
11

world, Leibniz is stuck with the fact that reason is in the world – to be read off of, rather
than put into, the objects of experience. Naturally, those objects are not the every-
day ones to which empiricists appeal but the supersensible truths of an intelligible
world. For Kant, however, the determination of reason by eternal truths is as funda-
mentally heteronomous as its determination by any other object’; and Lafont 2004: 28:
‘[Kantians] agree with the anti-realist that our normative judgments do not purport
to describe a pre-given moral order, heteronomously imposed on us independently of
our practical reason’. Although it cannot be discussed further here, somewhat similar
issues arise in other areas of Kant’s reception history, for example, in the contrast that
is drawn between Kantian autonomy and the supposedly more heteronomous position
of someone like Emmanuel Levinas, who speaks of the ethical demand as arising from
our encounter with other people rather than being imposed on ourselves by reason:
see e.g. Chalier 2002: Chapter 4.
12 Understanding Mor al Obligation

Thanks to Rawls’s considerable influence, the understanding of Kant’s

position that he offers here has gone on to achieve a kind of ortho-
doxy, particularly amongst those who studied and worked with him,
such as Schneewind, Korsgaard, Stephen Darwall, Andrews Reath,
and others;12 and this understanding, in turn, has shaped the way that
the rest of Kant’s ethical philosophy has been read, and the ethics of
his major successors.13
For, as constructivist interpreters of Kant fully recognise, if Kant
is read as a constructivist and committed to this conception of ‘self-
­legislation’, then he must be seen as having to deal with certain fun-
damental philosophical difficulties in articulating his position and
rendering it coherent, difficulties that were then bequeathed to those

12
See Schneewind 1986: 66: ‘The defining feature of an autonomous agent, in Kant’s
view, is its ability to guide its own action by the choices of a will that is such that what-
ever it wills is good simply because it is willed by it … The point is … that when some-
thing is chosen or pursued by such a will, that very fact makes the object of the will
good. An agent so guided is not led by anything outside himself’; and Schneewind
2002: 85/2010: 241: ‘[Kant] said that morality is a human creation. It is the legisla-
tion that comes from our own rational will’. Darwall 1997: 310: ‘For a Kantian such as
Korsgaard, the idea of an independent order of normative fact is inconsistent with the
autonomy of the will. Practical reasoning is not a matter of orientating oneself properly in
relation to some external source of value. Rather it is a self-government or autonomy –
the agent determining herself in accord with principles she can prescribe for herself as
one rational person among others’. Reath 2006: 164 note 17: ‘A constructivist account
of Kant’s moral theory offers one way to spell out what it is for the “rational will” to
have autonomy’. Rauscher 2002: 496: ‘Kant’s conception of autonomy precludes such
a [realist] conception of morality. Human beings cannot be dependent upon anything
distinct from their will for the moral law that binds them … If the distinctive feature of
Kant’s moral theory is autonomy, and if autonomy requires the dependence of moral
principles upon the human will, and if this dependence on the human will is idealist,
then the distinctive feature of Kant’s moral theory is its idealism. Kantians ought to
embrace moral idealism as the distinctive feature of Kant’s moral theory’. LeBar 2008:
185: ‘Kant’s primary motivation for constructivism, as I understand him, arises from
his conviction that our autonomy as moral agents can be preserved only if we give to
ourselves the laws governing our conduct’. Beiser 1992: 30–1: ‘Although Kant was not
fully aware of it, his Rousseauian ideas [of freedom] mark a profound break with the
natural law tradition of Pufendorf and the Wolffian school … Compared to this trad-
ition, Kant’s new ethics are revolutionary. The source of moral value is the rational will
inside us, not the providential order outside us. Here lies the real depth and impact
of Kant’s Copernican revolution. This took place not only in epistemology but also
in ethics. Just as the natural world depends on the laws of the understanding, so the
moral world depends on the laws of the will … [T]he human will creates moral values,
so that it is obliged to obey only the laws of its own making’.
13
This Rawlsian view of Kant also shapes the way in which pre-Kantian moral philosophy
has been read, as leading up to a Kantian model of autonomy as self-legislation: see,
for example, Korsgaard 1996c: 1–6, and for a classic study Schneewind 1998.
 Kant, mor al realism, and autonomy 13

who came after him. This set of difficulties that Kant is seen as hav-
ing to face up to in this way is sometimes referred to as the Kantian
paradox,14 which has several distinct aspects.
The first, and perhaps most obvious, might be called the problem of
emptiness: if there is nothing prior to the act of legislation by the subject
that is itself normative, how are any responsible or rational law-making
decisions to be made? Unless the process of legislation is grounded in
some values or norms that obtain prior to the process, won’t that leave
the legislating will in a void, unable to take any legislative decisions at
all – or at least, unable to take any rational ones that are more than a
mere motiveless plumping for one thing rather than another? Thus, as
Charles Larmore puts it, in the course of an argument for realism:
[W]hen we do impose principles on ourselves, we presumably do so for
reasons: we suppose that it is fitting for us to adopt them, or that adopt-
ing them will advance certain of our interests. Self-legislation, when it
does occur, is an activity that takes place in the light of reasons that we
must antecedently recognize, and whose own authority we therefore do
not institute but rather find ourselves called upon to acknowledge.15

The realist’s concern is that the Kantian gives the agent no way to act
except through a kind of ‘arbitrary self-launching’,16 and it is hard to
see that as either an accurate picture of our agency, or as a promising
account of the way in which normativity can be explained. The anti-
realist must therefore show how these difficulties can be avoided if the
constructivist account of self-legislation is to be maintained.
As well as this problem of emptiness, two further difficulties for the
constructivist relate to the authority of norms that are self-legislated in
this way. The first of these difficulties concerns whether self-legislating
can ever really amount to a genuine kind of legislating or binding – for
if I bind myself, can’t I unbind myself at will, leaving it unclear how
I was ever really bound in the first place?17 And the second, related,

14
Cf. Pinkard 2002 and 2003, who makes the Kantian paradox of self-legislation cen-
tral to his account of post-Kantian German Idealism. See also Pippin 2000a: 192, who
refers to the Kantian position as involving a ‘paradox’ of self-subjection.
15
Larmore 2008: 44. It is not only realists like Larmore who raise this difficulty, but
it is one frequently commented on by constructivists themselves. See, for example,
Korsgaard 1996c: 98.
16
Regan 2002: 278.
17
Cf. G. A. Cohen, who raises this as an objection to Korsgaard: ‘how can the sub-
ject be responsible to a law that it makes and can therefore unmake?’ (Cohen 1996:
170). Cf. Thomas Aquinas, Summa Theologiae IaIIae 93.5 (cited by Schneewind 2002:
14 Understanding Mor al Obligation

difficulty is raised by Elizabeth Anscombe in the claim that ‘the con-

cept of obligation requires superior power in the legislator’.18 But if it is
right that some such hierarchical relation is needed here, how can that
be the case if I am legislating over myself, where it would seem that both
sides are on a par, so making the act of legislation impossible?
Taken together, these issues represent the core of the Kantian
­paradox, which is said to arise for Kant out of his commitment to the
argument from autonomy and the picture of self-legislation that this
leads to; constructivist readers of Kant have therefore then gone on to
try to find in Kant some way in which to overcome the paradox. Thus,
in response to the problem of emptiness, it is argued that it is a mis-
take to see the Kantian self-legislating subject as ‘spinning in the void’,
as some constraints are constitutive of the will or practical agency as
such, so that there is some content built into the activity of legislation
by such a will from the outset.19 Secondly, when it comes to the prob-
lem of ‘unbinding’, appeal is again made to the idea of the norms in
question being constitutive of reason or of practical agency in some
way, so that as a result it is not clear that the subject could ever ‘repeal’
or withdraw them; but nonetheless, it is argued, it still makes sense
to think of the norms as self-legislated, as they are constitutive of our
agency and hence not imposed from outside. And thirdly, in response
to Anscombe’s challenge concerning the ‘superior power’ seemingly
required in any talk of legislating, constructivists have made use of
the hierarchical structure that Kant gives to the self, which seemingly
gives the rational aspect of the person a special position in relation to
other aspects in a way that enables it to have this legislating role. Thus,
in this vein Korsgaard writes:
We might say that the acting self concedes to the thinking self its right
to government. And the thinking self, in turn, tries to govern as well as
it can. So the reflexive structure of human consciousness establishes a
relation here, a relation which we have to ourselves. And it is a relation
not of mere power but rather of authority. And that is the authority that
is the source of obligation.20

84/2010: 240): ‘Law directs the actions of those who are subject to the government of
someone. Hence, properly speaking, none imposes a law on his own actions’.
18
Anscombe 1958: 2/1981: 27. As others have noted, Anscombe’s position somewhat
resembles Schopenhauer’s: see 1962/1965 §4.
19
Cf. Korsgaard 1996c: 235–6. For an account of constructivism that attempts to correct
for its apparent voluntarism, see Formosa forthcoming.
20
Korsgaard 1996c: 104. Cf. Kant MM 6:417–8 (pp. 543–4), to which Korsgaard refers.
 Kant, mor al realism, and autonomy 15

In this way, then, constructivist readers of Kant have developed

an ambitious programme on his behalf, built up as a response to
the problematic Kant is said to be led into as a result of his notion of
autonomy, and his resulting conception of self-legislation. And these
responses have themselves provoked further debate, concerning, for
example, the idea of practical reason having such constitutive princi-
ples, and whether, even if it does, they could ever plausibly be claimed
to amount to the norms we associate with morality, and thus whether
anything like this Kantian project in ethics can be expected to succeed,
or whether it must give way to something more modest.21 Questions
have also been raised about whether any aspect of the self can be given
normative authority over another aspect, unless there is already some
prior normative order on which that authority can be based.22
Now, the assumption that Kant has an anti-realist outlook has not
only influenced the reading of Kant himself, it has also shaped the
accounts that are given of post-Kantian ethics. For, it is argued, those
who came after Kant inherited the difficulties associated with the
Kantian paradox, because they too were swayed in their thinking by
the argument from autonomy, and thus drawn into a constructivist
outlook.
So, according to an influential reading of Hegel put forward by
Robert Pippin and others, Hegel offered a kind of ‘social’ solution
to the paradox that differed from the Kantian one. Thus, on Pippin’s
account, Hegel holds that ‘we do not face normative claims as singular,
unattached, noumenal beings, capable of acting as uncaused causes,
but as subjects located in historical time (as modern subjects) in vari-
ous non-detachable social and ethical relations to each other’.23 What
this means, Pippin argues, is that (contra the Kantian) ‘the legislation
of such a law does not consist in some paradoxical single moment of
election’, where the subject has no context for their legislating because
everything must be fixed at once; rather, each individual finds them-
selves ‘in medias res’ (so to speak), with values and norms that belong

21
Cf. the so-called ‘Humean constructivism’ proposed by Street, which, while sharing
some common ground with the Kantian, holds that the upshot of the constructiv-
ist position must be more limited: see Street 2008, esp. pp. 243–5 and 2010, esp.
pp. 369–70.
22
Cf. Cohen 1996: 171, who argues against a constructivist interpretation of Kant along
these lines: ‘Kant’s person indeed makes the law, but he cannot unmake it, for he is
designed by nature to make it as he does, and what he is designed to make has the
inherent authority of reason as such’.
23
Pippin 2000a: 194–5.
16 Understanding Mor al Obligation

to the community of which they are part.24 However, (contra the real-
ist) these values and norms are not things that the community can
claim to have ‘found’ or ‘discovered’ in some realist manner, but are
themselves the result of a collective and historical process of value cre-
ation, driven by the problems in previous attempts to take up values
and norms of other kinds, which themselves have proved unstable.25
On Pippin’s account, therefore, Hegel completes the Kantian self-leg-
islation project in a way that aims to avoid the worries raised by the
realist, by introducing the characteristic Hegelian themes of sociality
and historicity.26
Likewise, it has become commonplace to read Kierkegaard as also
inheriting the Kantian idea of the self-legislating subject, but as follow-
ing it through to its logical conclusion, so that the apparent emptiness
and arbitrariness of this subject’s position becomes fully clear. This

24
Cf. Pippin 1991: 72–3: ‘Just as when we attempt to “ judge objectively” or “determine
the truth,” we inherit an extensive set of rule-governed, historically concrete practices,
so when we attempt to “act rightly,” and attempt to determine our action spontan-
eously, we must see ourselves as situated in a complex collective and historical setting,
a dependence on setting very much like that implicitly asserted by the narrative form
of the modern novel … Thus it could be said that, in a way much like the classical ideal
of freedom as “realization within the whole,” Hegel too tries to show how the attempt
at self-determination requires (at least at some, often very implicit, level) an under-
standing of oneself as occupying a “place” within a larger whole, except in his view that
whole is not nature or the cosmos, but the history of a collectively self-determining
subject’.
25
Cf. Pippin 1991: 69–70: ‘On [Hegel’s] account, the question of what comes to count
as, in general, an authoritative explanation of objects and events, the decisive classifi-
catory procedure, or evaluative criterion, can never itself be resolved by appeal to an
ultimate explanatory principle, or general regulative ideal, or basic argument strat-
egy. There are, finally, no rules to tell us which rules we ought to follow in regulating
our discursive practices, no institutions certifying the axioms out of which such rules
should be constructed, and no transcendental argument for the necessary conditions
for any experience. What we always require is a narrative account of why we have come
to regard some set of rules or a practice as authoritative. In Hegel’s “phenomeno-
logical” version, such an account must always appeal to a pre-discursive context or
historical experience (sometimes simply called “life”) as the origin of such authori-
tative procedures and rules (even while Hegel also maintains that such a context or
experience is itself the “product” of a kind of prior reflective principle, now become
implicit, taken for granted, in everyday social life). Our account of our basic sense-
making practices is thus tied to an account of the aporiai “experienced in the life of
Spirit,” and so such a justification is everywhere, to use the famous word, “dialectical,”
and not “logical.”’ Pippin has continued his development of the idea of collective self-
legislation in Hegel in his more recent book: see e.g. Pippin 2008: 17–22 and 65–91.
26
Similar themes are also emphasised by Pinkard: see e.g. Pinkard 2002, 2003, and
2007. See also Krasnoff 2008, especially Chapters 3 and 4. For a helpful overview of
this approach, see Gordon 2005.
 Kant, mor al realism, and autonomy 17

then leads to Alasdair MacIntyre’s famous account of Kierkegaard in

A Short History of Ethics and After Virtue, as facing a situation of radical
(because groundless) choice: for Kierkegaard, he writes, ‘[t]he individ-
ual utters his moral precepts to himself in a far stronger sense than the
Kantian individual did; for their only sanction and authority is that
he has chosen to utter them’.27 And Iris Murdoch, in The Sovereignty
of Good, puts forward a picture of existentialism more generally that
chimes with MacIntyre’s conception of Kierkegaard:
The centre of this type of post-Kantian moral philosophy is the notion
of the will as the creator of value. Values which were previously in some
sense inscribed in the heavens and guaranteed by God collapse into
the human will. There is no transcendent reality. The idea of the good
remains indefinable and empty so that human choice may fill it. The
sovereign moral concept is freedom, or possibly courage in a sense
which identifies it with freedom, will, power. This concept inhabits a
quite separate top level of human activity since it is the guarantor of the
secondary values created by choice. Act, choice, decision, responsibility,
independence are emphasized in this philosophy of puritanical origin
and apparent austerity. It must be said in its favour that this image of
human nature has been the inspiration of political liberalism. However,
as Hume once wisely observed, good political philosophy is not neces-
sarily good moral philosophy.28

MacIntyre 1967: 215. See also MacIntyre 1985: 39–47.

27

Murdoch 1970: 80–1. For a more recent treatment along the same lines, see Pinkard
28

2002: 347–8: ‘Kierkegaard had fully absorbed the modernist and therefore Kantian
stress on autonomy. For Kierkegaard, the Kantian lesson – that in both experience
and practice the meaning of things for us could not simply be given but had to be
supplied by our own activity, our own self-direction – seemed almost self-evidently
true, and the shock was how much it seemed by the 1840s to have been forgotten.
That we are called to be self-directing, to lead our own lives, to be subject only to a law
we impose on ourselves, is, as Kant originally saw, quasi-paradoxical. If nothing else,
it means that we are called (or determined, to capture the dual connotations of the
German term, Bestimmung) to choose what we are to make of ourselves, and, curiously,
this calling to radical choice is both not itself something that is subject to choice, and
involves the paradox of demanding reasons for choice while ruling them out. We can
be subject only to those laws that we author for ourselves; but, as authors, we must have
reasons for the laws we author, since otherwise they cannot be “laws” (reasons) but
only contingent events; and, as even Kant had seen, that seemed to be paradoxical’.
Cf. also Krasnoff 2008: 158–9, on what he sees as Kierkegaard’s critique of Hegel: ‘For
Kierkegaard, the narrative satisfaction of seeing oneself as the product of the past and
one’s society was a pale substitute for an authentically lived life … For Kierkegaard …
the very idea of a reconciliation to the present is a distasteful affront to the autonomy
of the self, which always retains its prerogative to break with the past and set out on
its own path’.
18 Understanding Mor al Obligation

As Murdoch’s final observations indicate, she sees a clear link here

with the notion of autonomy developed by Kant and others within the
liberal tradition, one that (on this account) has led moral philosophy
into the kind of self-legislative story that culminates in the existential-
ism of a figure like Kierkegaard.
We have seen, therefore, how the argument from autonomy is held
to play such a crucial role in shaping the thought not only of Kant, but
also of his successors, as he and they sought to come to terms with the
‘paradox’ of self-legislation that this argument seemed to lead them
into – at least according to the standard account. I will now suggest,
however, that there are telling grounds on which to claim that this
standard account is mistaken and deeply misleading, and that a differ-
ent problematic is what mattered to Kant and his successors.

Difficulties with the standard story: Rawls’s account

of Kant’s argument from autonomy
The difficulties in the standard story can be seen clearly in Rawls’s
own treatment of the autonomy argument that he attributes to Kant,
which we can now look at in more detail.
In the fifth of his lectures on Kant in Lectures on the History of Moral
Philosophy, Rawls focuses on Kant’s discussion of autonomy and heter-
onomy at the end of Section ii of the Groundwork, and in related pas-
sages from the second Critique. It is here, Rawls suggests, that one can
find Kant objecting to non-constructivist positions on the grounds
that they result in a heteronomous conception of the will. However,
as we shall see, Rawls faces telling problems in making this argument
out: on the one hand, given Kant’s own conception of the autonomy/
heteronomy distinction, Rawls acknowledges that it is hard to mount
a plausible argument from autonomy against the realist; but on the
other hand, once Rawls drops this distinction from the argument, he
seems to offer no other grounds on which Kant might make it work.
Given these difficulties with arriving at a charitable view of Kant’s texts
here as amounting to an argument from autonomy against realism, I
will suggest that we should not see them as having this purpose.
We can begin, then, by considering Kant’s account of the autonomy/
heteronomy distinction, and seeing why Rawls does not think this can
be the basis for his argument against a realist moral theory such as per-
fectionism. According to Kant’s account, heteronomy results when the
will is constrained to act not by reason, but by desire, where then reason
 Kant, mor al realism, and autonomy 19

only plays a more subordinate role in helping us to act in such a way as

to get that desire satisfied (by helping us to decide on the best means to
this end, for example, or weighing up between competing desires). In
these latter circumstances, the agent is brought to act by the attractive
or repulsive force he feels for a certain object, and reason has a merely
assisting part to play – a role that, if this were all we needed reason for,
might have made it dispensable.29 However, used in this manner rea-
son serves ‘merely to administer an interest not belonging to it’,30 in
helping to bring about the satisfaction of our desires, where these can
range from the desire for pleasure (as on hedonistic accounts) to the
desire for the realisation of our capacities (as on perfectionist ones);
but then, in ‘administering this interest’, reason fails to ‘show its own
commanding authority as supreme lawgiving’,31 because it is placed
in this merely subservient position, even if on some of these accounts,
we must use reason to help us determine what the realisation of these
desires could possibly amount to.32 Thus, because reason fails to be
in command here, and is itself reduced to working in the service of
desire, it is no longer autonomous; and because reason is no longer
autonomous, the same is true of the will as a whole. So, for example, if
I resist telling a lie because I am concerned about my reputation, what
determines my will is the ‘object’ of desire, in this case my long-term
happiness; and because of this, reason has no ‘commanding’ role to
play, rendering my action heteronomous. If, however, I resist telling a
lie because my maxim fails the universalisability test, here reason is

29
See GMM 4:395–6 (pp. 50–1). 30
GMM 4:441 (p. 90).
31
GMM 4:441 (p. 90). Cf. also: CPrR 5:61 (p. 189): ‘[The human being] is nevertheless
not so completely an animal as to be indifferent to all that reason says on its own and
to use reason merely as a tool for the satisfaction of his needs as a sensible being’; LE
27:499 (p. 266): ‘[I]f, for example, the principle of universal happiness were to be the
basis for determination of the moral law, it would be a question of how far our needs
were satisfied in their entire totality by following these laws; but here laws of nature are
involved, and the moral laws would have to be subject to them, so that reason would
have to obey the laws of nature and sensibility, and that in a necessary fashion (for in
the physical order this is so anyway). But this would obviously put an end to the auton-
omy of reason, and thus be heteronomy’; LE 29: 625–6 (p. 263): ‘Reason attends either
to the interest of the inclinations, or to its own interest. In the first case it is subservi-
ent, but in the other, legislative’.
32
It is because this is a feature of perfectionist and theistic positions that Kant classifies
them as rationalist material principles, ‘for perfection as a characteristic of things, and
the supreme perfection represented in substance, i.e. God, are both to be thought only
by means of rational concepts’ (CPrR 5:41 (p. 172)).
20 Understanding Mor al Obligation

playing a governing role in my actions, in such a way as to make reason

authoritative, and so to render the action autonomous.33
Now, clearly, given this Kantian way of drawing the autonomy/heter-
onomy distinction, we can ask whether moral realism would result in
heteronomy in this Kantian sense. That this is so has been suggested
by some constructivists, for example Christine Lafont when she writes:
‘The standard reason that Kantian constructivists adduce against any
kind of moral realism is always the concern that any concession to
realism unavoidably involves introducing heteronomous consider-
ations about what human beings happen to want or desire which are
incompatible with the crucial role that the notion of autonomy plays
in Kantian moral theories’.34 Thus, we might take Kant’s conception
of autonomy and use this as the basis for a constructivist argument
against realism along these lines.
However, as Rawls himself seems to recognise, if we do operate in
this way, it would seem that we are attributing a position to Kant that is
vulnerable to an easy response from the realist. For it is not clear why

33
Cf. GMM 4:441 (pp. 89–90). Cf. Rousseau 1964: Book i, Chapter viii, p. 365/1997: 54:
‘for the impulsion of mere appetite is slavery, and obedience to the law one has pre-
scribed to onseself is freedom’. Constructivist readers of Kant invariably take Rousseau
to be an important influence in leading Kant to adopt a self-legislation model of
autonomy, where this passage is then usually cited in support: see e.g. Rawls 1971:
251; Korsgaard 1989: 237–8 note 31/1996a: 40 note 28; Schneewind 1992: 314/2010:
254–5; Reath 1994: 437/2006: 94; Rawls 2000: 204. However, it is less frequently noted
that the context for this comment strongly suggests that Rousseau took freedom to be
the result of obeying a law, not because that law is a law that one has made for one-
self, but because in obeying the law one is no longer acting out of desire but reason,
where Rousseau says that ‘moral freedom’ is ‘that which alone makes man truly mas-
ter of himself’, as this means that we are then no longer driven by our appetites. So,
although Rousseau may well be an influence on Kant here in linking morality with
autonomy, this influence arguably has less to do with the idea that morality involves
self-­legislation as obedience to a self-made law than with the idea that freedom requires
the exercise of reason and not desire, where this can come about through the moral
quality that our actions have for the first time in ‘the civil state’, where ‘man, who until
then had looked only to himself, see[s] himself forced to act on other principles, and
to consult his reason before listening to his inclinations’ (1964: 364/1997: 53). The
law that one follows here then counts as ‘self-imposed’ not because one is making it for
oneself, but because in following it, one is following what reason tells one to do, where
reason is a part of the self in a way that desire is not.
34
Lafont 2004: 35. Cf. also Korsgaard 1996b: 228–9 note 4/2008: 178 note 3: ‘There are
two elements to Kant’s notion of heteronomy: (a) the law is not the will’s own law, but
rather is given to it from outside, and (b) the will therefore can be bound by that law
only through an inclination or an interest, which renders the imperative to follow the
law hypothetical’. Korsgaard, however, seeks to downplay this second element, insist-
ing that ‘the real essence of heteronomy lies in the first element’.
 Kant, mor al realism, and autonomy 21

the realist cannot simply accept Kant’s claim about the need to give
reason and not desire the ‘commanding’ role, but argue that nothing
in value realism rules this out, where on many such realist accounts,
it is through reason that the agent discerns and is motivated to act by
those values.35 Rawls himself points out just this difficulty in reading
Kant’s argument against perfectionism, if we are to take it along the
lines outlined above:
Why can’t we know the given order of moral values and then be moved
by the thought of a certain action as expressing that order? The text
reads as if Kant thinks that there are only two alternatives: either
pure practical reason as itself the supreme maker of law determines
the object for the will, or else, if the will has an object, even an object
specified by rational ideas given to our reason, the will is determined
by an impulsion arising from the expected effect of doing the action,
in which case nature would make law. Yet this is incorrect, for there
is a third alternative. If we were to act from the principles of a moral
order lying in divine reason, affirming those principles as such as the
supreme law, then not nature but our acceptance of the rational ideas
lying in God’s reason would make law. This is just Leibniz’s view: we
should act on these ideas as far as we can know them.36

Thus, because Rawls thinks that Kant fails as things stand to rule out
this third alternative, and thus show why a position like Leibniz’s is
heteronomous in these terms, he tries instead to attribute to Kant
an argument from autonomy against moral realism that does not
rely on the false dichotomy that this view of heteronomy apparently
generates.

35
Cf. Regan 2002: 290, who argues that the Kantian position ‘suggests a false dichot-
omy’: ‘Generating the content of the law out of our own reason is not a necessary pre-
condition of nondetermination by sensuous motives’.
36
Rawls 2000: 229; also Rawls 2000: 235–6: ‘One main weakness is that Kant supposes
there to be only two possibilities: either the moral law is founded on an object given to
it, in which case it depends on our susceptibility and the pleasure we anticipate from
realizing that object, or the moral law as pure practical reason determines (constructs)
its own object out of itself … What is missing here is the recognition that intuitionism
says that knowledge of the order of values can arouse moral feelings and the desire to
act accordingly’. Reath registers the same worry, when he writes: ‘[I]t is not clear that
Kant has offered any reason to reject a version of rational intuitionism which holds
that: (a) the first principles of morals state truths about right action that obtain in vir-
tue of intrinsic values or relationships between objects that exist independently of the
will; (b) these principles lay down unconditionally valid obligations, whose authority
is independent of any contingent interests; and (c) our grasp of the validity of these
principles has motivating force’ (Reath 2006: 137).
22 Understanding Mor al Obligation

However, as a result, Rawls himself finds it very difficult to say any-

thing to flesh out the argument from autonomy he offers on Kant’s
behalf, other than to make the simple claim that any position which
sets up a ‘prior and antecedent order of values’ ipso facto renders us
heteronomous: ‘Besides holding that we cannot discern these ideas,
Kant’s point must be that even if we could, our pure practical reason
must be the supreme maker of its own principles’,37 if autonomy is to
be preserved. But the question now is, if this is Kant’s position, why
should the realist accept it? What makes realism heteronomous, if it is
not that it makes morality a matter of desire rather than reason?
One suggestion here, offered by Robert Johnson, is that Kant’s
position should be modelled on that of the theistic voluntarists, who
argue that unless God’s will were itself responsible for determining the
value of things, then his omnipotence would be undermined. Johnson
writes as follows:
First, let me give you an intuitive sense of why autonomy commits Kantian
ethics to denying that value is a source of reasons. Consider a parallel
example: the Divine Command theory’s resolution of a Euthyphro-style
dilemma. That dilemma begins with the assertion that God loves (or
responds in some appropriate way to) all and only good things. This
raises the question, Why? Is it because their value provides a reason for
God to give them their due, His love? But if God loves a good thing only
because its goodness gives him a reason to love it, then its goodness
explains the appropriateness of God’s love for it, and this is incompat-
ible with God’s omnipotence. The value that provides a reason for God
to love it would be a constraint on God’s love in the sense that God must
respond to reasons provided by the value of things or else fail to have

37
Rawls 2000: 229–30. See also Darwall 2006: 221–2, where Darwall introduces a ‘deonto-
logical intuitionist form of heteronomy’ as follows: ‘Can acceptance of this principle
[of promise-keeping] after a deontological intuitionist fashion manifest Kantian
autonomy? I think the answer has to be “no”. To be sure, such an acceptance is not
based on features of any object of the agent’s desire or on the independent desirabil-
ity of any possible state of the world … [But the agent] accepts the principle because
(as she believes) promise-keeping is (self-evidently) intrinsically right (pro tanto). She
doesn’t accept the principle of keeping promises because it follows from some formal
principle of the will like the CI [categorical imperative]. Her acceptance derives rather
from intrinsic features of promise-keeping and the property of rightness that she takes
to supervene upon these’. And cf. also Korsgaard 1996b: 229 note 4/2008: 178 note
3, who allows that it could be said that ‘[o]ne is not bound to considerations of honor
by inclination or interest: one is honorable for its own sake, moved by a conception of
how one ought to act’; but, she insists, ‘this kind of action is still not fully autonomous
because the laws of honor are not the will’s own law’.
 Kant, mor al realism, and autonomy 23

the requisite response. The alternative is to say that those things are
good because of God’s love for them. God’s love explains their value.
But then goodness looks like an arbitrarily distributed shadow cast by
God’s attention. For if God loved some entirely different set of things,
then those other things would have been good. The Divine Command
theorist opts for the second horn, and then is saddled with the problem
of explaining why goodness isn’t arbitrarily distributed after all.
Kantians must resolve a similar Euthyphro-style dilemma in the
same way as the Divine Command theory. What possess value on the
Kantian view are all and only the objects of rational agency. Now if
value is the source of the reasons for the pursuits of rational agents,
then the authority governing rational agency is external to that agency
itself, in the value of things that are its objects. But on the Kantian view,
rational agency must be autonomous, in the sense that the requirements
binding it are wholly self-generated and self-imposed. The autonomy of
reason, the central guiding idea behind Kantian moral theory, is thus
the very foundation of the case against the claim that there is some
value that provides reason to conform to moral obligation. Autonomy
requires that value not be a source of reasons.38

The central claim seems to be, then, that just as value realism would
clearly undermine God’s omnipotence in the theistic case, so it would
clearly undermine our autonomy as human beings.
However, if Rawls rejected the earlier interpretation of Kant’s pos-
ition for its philosophical weakness, so the same would seem to be
true of this interpretation. For, the realist can argue in response to
this challenge that there is a significant difference between omnipotence
and autonomy, however we exactly spell out the latter, and that this is
apparently being overlooked by Johnson. Thus, while it may indeed
be plausible to hold that God’s omnipotence would be undermined if
values obtained independently of his will (though of course even this
could be denied), given that autonomy is a much weaker notion than
omnipotence, nothing so far shows that the parallel argument has any
such plausibility and that to model the argument on the theistic case is
highly misleading for that reason. The realist may therefore grant that
Johnson is right that realism would undermine omnipotence in the
theistic case, but still feel she has been given no grounds on which to
think it undermines autonomy in ours. For, suppose the goodness of
something did indeed provide God with a reason to love it, such that
he would be rationally criticisable were he not to do so; there would

38
Johnson 2007: 140–1.
24 Understanding Mor al Obligation

then be some sense in which he ‘must’ respond to that thing with love,
based on this reason. Perhaps Johnson is right that then God would
have to fulfil certain requirements in a way that would limit his ‘pow-
ers’ in some way and thus threaten his omnipotence; but (the realist
will argue) how can it be that having to follow a reason in this manner
in his actions would make him less autonomous, any more than having
to follow what there is reason to think does so? It would seem, then,
that the argument from autonomy as Johnson presents it needs fur-
ther development if it is to be made convincing.
Nonetheless, leaving considerations of charity aside, it might still
be said that something like this argument must have been Kant’s, on
textual grounds – where one such text in particular is frequently cited
by constructivist interpreters, from the Groundwork:39
If the will seeks the law that is to determine it anywhere else than in the
fitness of its maxims for its own giving of universal law – consequently
if, in going beyond itself, it seeks this law in a property of any of its
objects – heteronomy always results. The will in that case does not give
itself the law; instead the object, by means of its relation to the will, gives
the law to it … [The categorical imperative] must therefore abstract
from all objects to this extent: that they have no influence at all on the
will, so that practical reason (the will) may not merely administer an
interest not belonging to it [fremdes Interesse], but may simply show its
own commanding authority as supreme lawgiving.40

Constructivists typically take this passage to be targeted against moral

realism, as a position which holds that the will should be determined
by the moral properties possessed of certain objects, where Kant is
then read as claiming that any such position is heteronomous, because
then the moral law comes from such objects rather than the will itself,
where the latter form of lawgiving is said to be needed for autonomy.
Understood in this fashion, this important passage may seem to be a
clear attempt by Kant to offer an argument from autonomy against
realism, just as Rawls and others have supposed.
However, I think that a closer look at this text suggests that such
constructivist interpretations are misplaced. For, as we have seen, at
the centre of Kant’s conception of the autonomy/heteronomy distinc-
tion is a contrast between reason controlling the will on the one hand,
and reason acting as a ‘handmaid’ to achieve some end given to it by

39
 See, for example, Rawls 2000: 226–7; Reath 2006: 128. Cf. also Darwall 1995: 323–4.
40
GMM 4:441 (pp. 89–90).
 Kant, mor al realism, and autonomy 25

desire on the other. Kant’s claim in this passage, therefore, is that the
will is only autonomous if ‘practical reason (the will)’ does not ‘merely
administer an interest not belonging to it’ – that is, if it does not simply
help to obtain some object or end in which we have an interest, as this
results in heteronomy on Kant’s picture.41 Thus, Kant claims, auton-
omy is not possible if I follow merely hypothetical imperatives, as here
reason is serving merely as a handmaiden in this manner; it can only
be in control when following a categorical imperative, where it then
shows its ‘commanding authority as supreme lawgiving’, because it is
no longer just engaged in pursuing such an end.42 As we have seen,
however, one can accept this form of Kantian autonomy/heteronomy
distinction, without taking that to tell against realism in the way that
constructivist interpreters suggest.43
Moreover, this way of understanding this passage explains some-
thing that Rawls finds puzzling, which is that while Kant offers a cri-
tique of perfectionism for its heteronomy, he does not offer a critique
of rational intuitionism as such on similar grounds. This is perplex-
ing for Rawls because he takes the realism of perfectionism to be the
basis for its heteronomous status, which might therefore be expected

41
Cf. Irwin’s reading of this passage: ‘In Kant’s view, an autonomous will does not “go
outside itself” to seek the law that will determine it “in the property of any of its
objects” … Hence the law that determines the will is the law proper to the will, and not
introduced from any object of inclination. This aspect of autonomy does not preclude
objectivity, if the law proper to the will is the law recognized by reason’ (Irwin 2009:
156). See also p. 164: ‘We go outside the rational will in appealing to other sorts of
incentives that are not essential to a rational will; [Kant] does not say that we go “out-
side” it if we introduce an objective reality that guides the will’.
42
Cf. CPrR 5:118 (p. 235): ‘Inclination is blind and servile, whether it is kindly or not;
and when morality is in question, reason must not play the part of mere guardian to
inclination but, disregarding it altogether, must attend solely to its own interest as
pure practical reason’.
43
Relatedly, we can therefore see that Korsgaard also goes too far when she writes: ‘I think
Kant’s view is that the rational mind is by its very nature under its own authority, and
a will that simply accepted the authority of state-of-the-world regarding reasons would
be governed by something alien to itself. It would be, in Kant’s words, “consciously
receiving direction from any other quarter” – something Kant says reason cannot do
(4:448)’ (Korsgaard 2007: 15). Taking Kant’s remark in context, it is clear that ‘the
other quarter’ that would be ‘alien to itself’ is not some ‘state-of-the-world’ that reason
has not itself constructed, but the faculty of desire, which if this is in control would
indeed take away from the authority of the agent’s rational mind in judging how to
act: ‘Now, one cannot possibly think of a reason that would consciously receive direc-
tion from any other quarter with respect to its judgments, since the subject would then
attribute the determination of his judgment not to reason but to an impulse’ (GMM
4:448 (p. 96)).
26 Understanding Mor al Obligation

to extend naturally enough to rational intuitionism as well. Rawls is

uncomfortably aware, however, that Kant never seems to expand his
argument from the heteronomy of perfectionism to the heteronomy
of realist intuitionism in this Rawlsian manner. Thus, while he insists
that ‘it is not sufficiently appreciated that Kant would reject rational
intuitionism as a form of heteronomy just as firmly (or so I think) as
he would reject Hume’s psychological naturalism’, he nonetheless con-
cedes that ‘part of the explanation for the failure to appreciate this
fact about Kant’s view may lie in Kant’s own exposition: he nowhere
clearly describes rational intuitionism and there shows why he thinks
that it is heteronomous’.44 Now this, I think, is a remarkable admis-
sion, which should give us pause; and it is made even more remarkable
by Rawls’s further explanation for why Kant nowhere gives this clear
description of rational intuitionism, which is that ‘[p]ossibly he lacks a
clear conception of it’.45 We can now see, however, why it is no accident
that Kant offers no such critique of rational intuitionism: for, given the
distinction between autonomy and heteronomy that Kant is working
with here, any critique of this sort would be out of place.
What I am proposing thus far, therefore, is a dilemma for the con-
structivist: on the one hand they could use the Kantian notion of
heteronomy as related to desire and not to reason, but accept that the
realist cannot easily be convicted of heteronomy in this sense.46 On the
other hand, they might just adopt the simple claim that moral realism
just is incompatible with autonomy, but without any real resources to
explain why. Either way, it appears, we have not found a way of develop-
ing an argument from autonomy in a manner that would make Kant’s
position compelling, if he is interpreted along these lines.

Difficulties with the standard story: Kant’s value realism

As well as these difficulties faced by the ‘standard story’ in finding
a cogent argument from autonomy in Kant’s texts, another problem
with the constructivist’s account of Kant’s position is that there are
good grounds for thinking that Kant himself was a value realist of
sorts – which of course makes it unlikely that he took such realism to
be a threat to our autonomy.

Rawls 2000: 235.

44 45
 Rawls 2000: 235.
Oliver Sensen takes this option, and so admits that Kant’s argument is ‘not watertight’
46

(see Sensen 2011: 270).

 Kant, mor al realism, and autonomy 27

These value realist readings of Kant have often grown out of dissat-
isfaction with attempts to treat the categorical imperative as norma-
tively prior and self-standing, and have instead claimed that underlying
it there is a value that makes it binding and explains our commitment
to it, and which gives it content. That value is generally taken to res-
ide in our status as free and rational agents, and because our worth as
such agents is not itself constructed, so neither is the moral law that
is grounded upon it. Thus, as Allen Wood puts this more realist view:
‘The content of the [moral] law is not a creation of my will, or the out-
come of any constructive procedure on my part. The law of autonomy
is objectively valid for rational volition because it is based on an object-
ive end – the dignity of rational nature as an end in itself’.47
Proponents of the value realist account of Kant have largely pointed
to two main pieces of textual evidence in favour of their view, both
found in the Groundwork. The first is from the opening section, which
identifies the good will as good without qualification, in a way that
appears to make a clear attribution of value to it:
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 …
A good will is not good because of what it effects or accomplishes,
because of its fitness to attain some proposed end, but only because of
its volition, that is, it is good in itself and, regarded for itself, is to be val-
ued incomparably higher than all that could merely be brought about
by it in favor of some inclination and indeed, if you will, of the sum of
all inclinations.48

Kant admits that to some, his value claim here may appear peculiar,
particularly if they think the aim of reason is just to help us to become

47
Wood 2008: 108, also 1999, especially pp. 46–7, 111–55. As well as Wood, proponents
of the value-realist approach include also Schönecker 1999: 387–9; Guyer, particularly
in his collection of essays Guyer 2000; Langton 2007; and Hills 2008. A form of ‘teleo-
logical’ reading that emphasises consideration of value in Kant’s work is also proposed
by Barbara Herman in Herman 1993, esp. pp. 208–40; but Herman explicitly states
that ‘[t]he sense in which I think we must leave deontology behind does not require
that we argue for an independent conception of value as the foundation for practical
reason’ (p. 239). For critical reviews of Wood and Guyer respectively, see Pippin 2000b
and Reath 2003.
48
GMM 4:393–4 (pp. 49–50). Cf. also LE 29:608 (p. 231): ‘Everyone knows that nothing
in the world is absolutely good without restriction, save a good will … An imperative is
categorical if it is the rule of a will itself intrinsically good’.
28 Understanding Mor al Obligation

happy and look after our well-being, rather than to attain moral
goodness:
There is, however, something so strange in this idea of the absolute
worth of a mere will, in the estimation of which no allowance is made
for any usefulness, that, despite all the agreement of even common
understanding with this idea, a suspicion must yet arise that its covert
basis is perhaps mere high-flown fantasy and that we may have misun-
derstood the purpose of nature in assigning reason to our will as its
governor.49

Kant responds to this worry, however, by turning the tables on his

opponent, and arguing that in fact reason is a pretty poor and dispens-
able guide to happiness that nature could have done without if that
was its purpose; so reason’s real role clearly lies elsewhere, in helping
us to become good wills, where ‘the true vocation of reason must be
to produce a will that is good, not perhaps as a means to other pur-
poses, but good in itself, for which reason was absolutely necessary’.50
The value realist’s claim, therefore, is that these early passages from
the Groundwork show that Kant was working with a conception of the
value of the good will (where this good will is later identified with the
rational will that acts in accordance with the categorical imperative).
A second part of the Groundwork that is also appealed to by the
value realist is Kant’s discussion of the Formula of Humanity in the
middle of Section ii. Kant begins that discussion by distinguishing
between objective and subjective ends for action, in relation to which
the rational agent follows law-like principles. Subjective ends acquire
their status as ends thanks to the desires of the agent, while object-
ive ends serve as ends for all rational beings. A material principle is
one that is aimed at the attainment of a subjective end, whereas a for-
mal principle is related to an objective one. Material principles can
therefore only supply hypothetical imperatives, and so are not suited
to serve as moral principles. The question is, therefore, whether there
are any formal principles, and thus whether there is any objective end
which might underlie them.
It is at this point that Kant writes in a way that has attracted the
attention of the value realist, in seeming to claim that this objective
end needs to be something of ‘absolute worth’, which could then serve
as an end in itself, and thus as a ground of the formal principle: ‘So act

49
GMM 4:394 (p. 50). 50
GMM 4:396 (p. 52).
 Kant, mor al realism, and autonomy 29

that you use humanity, whether in your own person or in the person of
any other, always at the same time as an end, never as a means’51 – that
is, the so-called Formula of Humanity. Kant writes as follows:
But suppose there were something the existence of which in itself has an
absolute worth, something which as an end in itself could be a ground of
determinate laws; then in it, and in it alone, would lie the ground of a
possible categorical imperative, that is, of a practical law.
Now I say that the human being and in general every rational being
exists as an end in itself, not merely as a means to be used as this or that
will at its discretion; instead he must in all his actions, whether directed
to himself or also to other rational beings, always be regarded at the
same time as an end… [Rational beings] are not merely subjective ends,
the existence of which as an effect of our action has a worth for us, but
rather objective ends, that is, beings the existence of which is in itself an
end, and indeed one such that no other end, to which they would serve
merely as means, can be put in its place, since without it nothing of abso-
lute worth would be found anywhere; but if all worth were conditional
and therefore contingent, then no supreme practical principle for rea-
son could be found anywhere.52

As one value realist commentator has remarked: ‘Kant could hardly

make his views any clearer: the existence of rational beings has in itself
an absolute value’.53
There is, moreover, a third discussion in the Groundwork that I
believe value realists could also appeal to, though this is less often
remarked upon.54 This occurs in Section iii where Kant’s intentions
and argument are notoriously opaque. In general, it appears, Kant
is trying here to deal with certain sceptical worries that might arise
in respect of the account of morality he has offered in the first two
sections, where the concern is that morality could still be claimed to
be a ‘chimerical idea’ or a ‘phantom’.55 These sceptical worries, how-
ever, are not best seen in the manner of the amoralist’s challenge to
the ethics of the ‘why be moral?’ variety, but more like the worries
that might plague someone who is well-disposed towards morality, and
yet who is puzzled by it in certain fundamental respects.56 In this last

51
GMM 4:429 (p. 80).
52
GMM 4:428 (pp. 78–9).
53
Hills 2008: 190.
54
But cf. Schönecker 1999: 387–9.
55
GMM 4:445 (p. 93).
56
Cf. Timmermann 2007: 129–30. For further discussion, see Stern 2010.
30 Understanding Mor al Obligation

section of the Groundwork, Kant focuses on three interrelated aspects

that may appear to make morality problematic and hence ‘chimerical’:
the extent that we can claim to be the free agents that moral actions
require; the difficulty of accounting for the binding nature of moral-
ity; and the ground and nature of the special ‘interest’ that we take in
acting morally, when this does not seem to reduce to any self-interest,
and indeed can go against our desires and individual concerns. The
latter can seem particularly puzzling: for if morality is not something
that satisfies those desires and concerns, what can explain our caring
about it in the way that we do? Kant expresses this puzzlement as fol-
lows, where he puts the issue in terms of needing to show that moral
action has a kind of worth or value, independently of serving our sub-
jective ends:
[F]or, if someone asked us why the universal validity of our maxims as a
law must be the limiting condition of our actions, and on what we base
the worth we assign to this way of acting – a worth so great that there
can be no higher interest anywhere – and asked us how it happens that
a human being believes that only through this does he feel his personal
worth, in comparison with which that of an agreeable or disagreeable
condition is to be held as nothing, we could [so far] give him no sat-
isfactory answer … [W]e cannot yet see, in this way, that we ought to
detach ourselves from such interest, that is, to regard ourselves as free
in acting and so to hold ourselves yet subject to certain laws in order to
find merely in our own person a worth than can compensate us for the
loss of everything that provides a worth to our condition; and we can-
not yet see how this is possible, and hence on what grounds the moral law
is binding.57

Kant’s puzzle here, therefore, seems to be this: what can explain why
we think it is more valuable to act as moral agents, even though we
know that on occasion doing so will go against the very pleasures and
satisfactions the having of which we take to bring value to our lives, and
hence in a way that apparently robs our lives of value in these respects?
Unless this puzzlement is dispelled, Kant fears, we may fall into doubts
about the intelligibility of morality.
His response, then, seems to be to try to give some sort of account
of where the distinctive and special value of moral agency resides, and
why we find it so compelling to treat this as having greater worth than
the worth that comes into our lives from having non-moral pleasures

57
GMM 4:449–50 (p. 97).
 Kant, mor al realism, and autonomy 31

and satisfactions. To do so, Kant appeals to the transcendental dis-

tinction he draws in the first Critique between the level of appearance
(in which our desiring selves reside) and the noumenal level (which
contains our moral selves), using this transcendental distinction of
levels, and the fact that the latter is prior to the former, to underpin
the value claim made on behalf of the good will, and to show why this
matters to us so much more. In this way, Kant suggests, he can explain
why even ‘the most hardened scoundrel’ will admit that he would be a
‘better person’ (i.e. a person of greater worth) by bringing himself to
act morally, where he would then be acting in the ‘higher’ world of the
understanding and not the ‘lower’ world of sense.58 Of course, Kant
says, the details of how exactly this works, and how these two ‘worlds’
interrelate will always be murky to us:59 but something along these
lines must be right, he argues, if the value we place on being moral
agents is not to remain unfathomable, and render morality itself ques-
tionable as a result.
We have seen, then, that there appear to be good textual grounds
on which to attribute value realism to Kant himself;60 and if so, it of
course becomes implausible to think that he would have taken such
realism to be a threat to our autonomy.
However, as is perhaps to be expected, this value realist reading of
Kant can and has been challenged in a number of ways, so that before
seeing this as settling the issue, these challenges require some con-
sideration. We can first consider some responses to the realist’s read-
ing of the three parts of the Groundwork that we have gone through,
and then some other textual evidence used against the value realist’s
approach.
Beginning, then, with Kant’s discussion at the start of Section i
of the Groundwork: in her defence of the realist reading, Alison Hills
imagines a possible reply, which is to claim that ‘the first paragraphs
set out a version of value realism that Kant will later repudiate as philo-
sophically naïve’, on the grounds that the structure of the Groundwork

58
GMM 4:454–5 (p. 101). Cf. also CJ 5:451 (p. 317), where Kant argues that a person
would feel himself ‘worthless in his own eyes’ were he to become amoral and decide to
‘hold the laws of duty to be merely imaginary, invalid, and nonobligatory, and were to
decide to transgress them without fear’. And cf. CPrR 5:86–9 (pp. 209–11).
59
See the section ‘On the Extreme Boundary of all Practical Philosophy’, GMM 4:455–63
(pp. 101–8).
60
Although I will not consider them further here, Guyer also appeals to other writings
such as the lectures Naturrecht Feyerabend (27: 1319–22) in support of a realist reading:
see Guyer 2000: 152–3, 156–8, and 170–1.
32 Understanding Mor al Obligation

is such that Kant ‘begins with “ordinary rational knowledge of moral-

ity”, and in the course of each chapter, moves to a more philosophic-
ally sophisticated theory’.61 In response, however, Hills points out that
Kant never repudiates such ‘ordinary rational knowledge’, while (as
we have seen) the realist can claim that evidence for their reading can
equally be found in later parts of the text, so that there is no reason to
think that this earlier material is anomalous in any way. Nonetheless,
it could then be said that while Kant does not later repudiate such
knowledge, in the later sections he gives an account of this knowledge,
and of what the value of the good will consists in, that is not as realist
as these remarks suggest, so that it is the later discussions that must be
considered as the important ones in this dispute.
Turning now to Kant’s treatment of the Formula of Humanity, it
has been accepted by critics of the realist account that Kant does treat
humanity as having a value; but it has then been argued that ‘[t]he
value of the good will and humanity … comes from their being objects
of rational choice, and not the other way around’,62 as the realist
claims. The suggestion, then, is that the Formula of Humanity discus-
sion is no exception to Kant’s general anti-realism about value, even
if it may superficially appear otherwise. However, I think the realist
can respond by noting that even their critics seem to concede that the
realist reading of the text is the more natural one,63 and can argue

61
Hills 2008: 190. Guyer also considers and rejects this response in Guyer 2000:
139–47.
62
Johnson 2007: 144. Cf. Dean 2006: 12, and 151: ‘Kant … undeniably attributes spe-
cial value to the end in itself [and so to humanity], saying it has an incomparably
high value, a value independent of inclinations, and the like. So the point of dispute
cannot be whether he thinks the Categorical Imperative is based on some end, or
whether he ever talks about the value of that end. Instead, the question is which part
of Kant’s theory is conceptually prior and fundamental – the claim that there is an
end that necessarily provides a reason for acting in certain ways, or the attribution of
value to such an end. I have proposed that an argument can be provided for treating
humanity in special ways, an argument that does not rely on any prior claims about
value. Only after these practical requirements are established, I have argued, can we
express the practical requirements in terms of value’. For a similar view, cf. also Sensen
2009 and 2011: 271, where Sensen suggests that for Kant ‘value is dependent on the
Categorical Imperative’, and not vice versa; and Schneewind 1996a: 286/2010: 278,
where Schneewind writes that ‘[g]oodness and value, on this reading of Kant, are
always explained in terms of rational willing’; and Formosa forthcoming.
63
Cf. Dean 2006: 12, where Dean admits that ‘Kant himself misleadingly presents the
argument, in Groundwork 427–9, as if it may depend upon prior claims about the
“absolute value” of humanity’. And Sensen allows that Kant ‘refers to value in raising
the issue’ of the ground of the Formula of Humanity (Sensen 2011: 270).
 Kant, mor al realism, and autonomy 33

that they only refuse to accept this reading because it runs contrary
to what they take to be Kant’s overall position; but this, of course, is to
beg the question against the realist, pending further textual support.
This is not to deny that, using Kantian materials, a conception of the
value of rational beings could be offered that is more constructivist, as
arising from the way practical reason requires us to apply the univer-
salisability test to our maxim or to act on reasons that others can share
(for example), which will mean that we will accord rational agents a
special status. But the fact remains that this is not how Kant appears to
proceed in the talk about the value of such agents in the passages we
have cited, where instead he argues that unless there were something
of absolute value the moral law could not have a properly categorical
structure, where he argues for rational agents as having such value
by elimination.64 A remaining interpretative card to be played at this
point is to stress Kant’s insistence on the intimate relation between the
different formulations of the categorical imperative, where prioritis-
ing the more formalistic Formula of Universal Law may seem to favour
a constructivist approach.65 But such is the range of options in con-
necting the formulae here that it is unlikely that anything conclusive
will emerge that has not already been covered.66
Finally, in relation to Kant’s discussion of value in Section iii of the
Groundwork, this is a point that realists themselves have not empha-
sised so greatly, and that has therefore received less critical debate
from this perspective than the other texts.67 Perhaps, given the sheer
complexity of Kant’s discussion here, it could just be claimed that the
realist account I have given cannot be established with any certainty,
and other accounts of Kant’s solution to the problem of our ‘inter-
est’ in morality can instead be offered that do not depend on our
moral agency having a special value based on its position in the ‘world
of understanding’ rather than the ‘world of sense’.68 Moreover, the
response may well be that any such realist reading must be discounted,
given the other textual evidence against it. To try to settle this issue,
therefore, we need now to look at this material.

64
For a defence of this way of taking Kant’s argument here, in opposition to Korsgaard’s
more constructivist one, see Timmermann 2006.
65
GMM 4:436–7 (pp. 85–6). This point is made by Sensen 2009 and Formosa
forthcoming.
66
See e.g. Guyer’s account in Guyer 2000: 159–61.
67
But cf. Sensen 2009: 106.
68
Korsgaard 1996c could be understood along these lines.
34 Understanding Mor al Obligation

The first set of such textual evidence comes from the Groundwork
itself, where it is argued that Kant says things in this work that count
against the realist reading. We have already considered in the previous
section how Kant’s discussion of the heteronomy/autonomy distinc-
tion may be said to show that Kant is not a value realist. As we have
seen, however, this way of taking the discussion is disputable. We have
also mentioned the passage in which Kant talks of the subject ‘giv-
ing the law to itself’;69 but realist readers have argued that this can-
not be taken at face value in constructivist terms, in so far as Kant
also states that the content of this law has no author,70 while there is
a tentativeness in the way that Kant puts things here which should
also perhaps give rise to doubts (‘must be viewed’; ‘can regard itself’).
Another piece of text that can be cited from the Groundwork is Kant’s
claim that ‘[n]othing can have a worth other than that which the law
determines for it’,71 where it is then argued on this basis that Kant can-
not have wanted to have the Formula of Humanity (as a formulation of

69
GMM 4:431 (p. 81).
70
Cf. LE 27:282–83 (p. 76): ‘The lawgiver is not always simultaneously an originator of
the law; he is only that if the laws are contingent. But if the laws are practically neces-
sary, and he merely declares that they conform to his will, then he is a lawgiver. So
nobody, not even the deity, is an originator of moral laws, since they have not arisen
from choice, but are practically necessary; if they were not so, it might even be the case
that lying was a virtue. But moral laws can still be subject to a lawgiver; there may be a
being who is omnipotent and has power to execute these laws, and to declare that this
moral law is at the same time a law of His will and obliges everyone to act accordingly.
Such a being is then a lawgiver, though not an originator; just as God is no originator
of the fact that a triangle has three corners’, and MM 6:228 (p. 381): ‘A (morally prac-
tical) law is a proposition that contains a categorial imperative (a command). One
who commands (imperans) through a law is the lawgiver (legislator). He is the author
(autor) of the obligation in accordance with the law, but not always the author of the
law. In the latter case the law would be a positive (contingent) and chosen law. A law
that binds us a priori and unconditionally by our own reason can also be expressed as
proceeding from the will of a supreme lawgiver, that is, one who has only rights and
no duties (hence from the divine will); but this signifies only the idea of a moral being
whose will is a law for everyone, without his being thought of as the author of the law’,
and cf. also LE 27:544 (p. 302): ‘Were we to conceive of the legislator as auctor legis,
this would have reference only to statutory laws. But if we ascribe an auctor to laws that
are known, through reason, from the nature of the case, he can only be an author of
the obligation that is contained in the law. Thus God, too, by the declared divine will,
is auctor legis, and precisely because natural laws were already in existence, and are
ordained by Him’. For further discussion of this material and related passages, see
Acton 1970: 38–9; Kain 1999: 177–99, 2004, 2006; Hare 2001: 94–7; Ameriks 2003;
Irwin 2004; Timmermann 2007: 106–7; and Wood 2008: 106–16.
71
GMM 4:436 (p. 85). This claim is emphasised by Johnson 2007: 143–7 and Dean
2006: 45.
 Kant, mor al realism, and autonomy 35

the moral law) be ‘determined’ or grounded in what has a prior value.

However, in response, Rae Langton has pointed out that ‘ “Determine”
is a notoriously slippery verb, one that can be pushed either way on the
Euthyphro question. Do the gods “determine” what is good? Two ques-
tions really: Do they decide it? Do they discover it?’72 Langton suggests,
in defence of the value realist approach, that we should interpret Kant
as meaning the latter. Alternatively, if (as someone like Guyer claims)
freedom is of fundamental value for Kant, but that freedom requires
us to follow self-imposed laws, our worth can be said to be determined
by this lawgiving, without being derived from it:73 our worth comes
from a freedom that is constrained in the right way, where this free-
dom can only be realised through such lawgiving and so is determined
by it in this sense.
Now, some opponents of the value realist reading have accepted
that the Groundwork is equivocal on this sort of issue, but they have
argued that Kant’s considered position is much clearer elsewhere in
his later works, particularly in the Critique of Practical Reason, and espe-
cially in Kant’s discussion there of the relation between the good and
the moral law. It is this discussion, it is widely claimed, that shows that
Kant finally renounced any lingering value realism that may be con-
fusingly present in the Groundwork, and put forward his settled view.74
It is to this text, therefore, that we need finally to turn, as a last test for
the plausibility of the value realist interpretation.
The discussion in question occurs in the second chapter of ‘The
Analytic of Practical Reason’, entitled ‘On the Concept of an Object
of Pure Practical Reason’. According to opponents of the value realist
interpretation of Kant, what this chapter shows is that Kant rejects any
attempt to treat the good as prior or foundational in ethics, as doing
so would inevitably make moral action heteronomous, as then the
motivation to act would come through our desire for that good, and
not through reason. Kant’s ‘Copernican revolution’ in ethics is there-
fore said to treat the categorical imperative as prior thanks to its status
as a formal principle of practical rationality, and to hold that the good
must be constructed out of that; and, just as in the first Critique, Kant
is supposed to accept the appearance of paradox that such a reversal

72
 Langton 2007: 183. Cf. also Timmermann 2006: 78.
73
Cf. LE 27:344 (p. 125); NM 19:288 (p. 473).
74
Cf. Dean 2006: 45–9, 117, 151.
36 Understanding Mor al Obligation

will have to those wedded to the old, pre-Copernican, picture. Rawls

puts this line of interpretation as follows:
We might say that a moral conception is not to revolve around the good
as an independent object, but around a (formal) conception of the right
(together with a [formal] conception of the good will), as constructed
by our pure practical reason, which any permissible good must fit. Or as
Kant puts it when he explains what he calls ‘the paradox of method’ in
critically examining practical reason (KP 5:62f): ‘The paradox is that the
concept of good and evil is not defined prior to the moral law, to which, it would
seem, the former would have to serve as a foundation; rather the concept of good
and evil must be defined after and by means of the law.’ (Kant puts the entire
sentence in italics.)75

The claim is, then, that Kant’s discussion here supports the construct-
ivist turn, and so shows that the value realist’s understanding of his pos-
ition is wrong-headed. I will now argue, however, that Kant’s position
here is rather different from the one attributed to him by Rawls and
others, as can be seen when Kant’s discussion is placed in context.
As Kant makes clear in the Preface to the Critique of Practical Reason,
and as I think must be agreed on all sides, the context for Chapter ii
of the Analytic is given by Herman Andreas Pistorius’s critical review
of the Groundwork, where here he is complimented by Kant as some-
one ‘who is devoted to truth and astute and therefore always worthy
of respect’76 – where it is rare for his opponents to receive any such
accolade from Kant. He summarises Pistorius’s objection as being that
in the Groundwork, ‘the concept of the good was not established before the
moral principle (as, in his opinion, was necessary)’.77 Kant expresses the
‘hope’ in his Preface that he has ‘dealt adequately’ with this criticism
in Chapter ii.
As his respectful remarks about Pistorius indicate, Kant accepts that
Pistorius’s objection can easily appear compelling: for, it may indeed

75
Rawls 2000: 227. Rawls is using the translation of this passage in Kant 1956: 65. Cf.
also Rawls 1989: 93/1999: 509. For similar interpretations, see Silber 1959; Gaut 1997:
162–3; Dean 2006: 151; Sensen 2011: 268–9; and Schneewind 1992: 316/2010: 256–7,
where Schneewind argues that Kant rejected the idea that ‘a right act can be one that
brings about good states of affairs’, because ‘it makes autonomy in his sense impos-
sible. Suppose that a kind of state of affairs is intrinsically good because of the very
nature of that kind of state of affairs. Then the goodness occurs independently of
the will of any finite moral agent, and if she must will to pursue it, she is not self-
legislating’.
76
CPrR 5:8 (p. 143).
77
CPrR 5:8 (p. 143). See Pistorius 1975, especially 145–6.
 Kant, mor al realism, and autonomy 37

seem plausible to think that in order to establish any moral principle,

one should start by establishing what is good and bad, and then frame
one’s moral principles in terms of that, as helping us to attain the
former and avoid the latter. Thus, it may well seem that Kant gets his
method of moral inquiry in the Groundwork the wrong way round, by
first setting out what he takes the principles of morality to be, and then
determining what is good accordingly, rather than vice versa, in the
way that Pistorius recommends, and which (Kant allows) has an obvi-
ous intuitive appeal. This, then, is Pistorius’s challenge as I think Kant
sees it.
Moreover, there is an important further issue lying behind Pistorius’s
objection, which makes it even more pressing for Kant. For, accord-
ing to Pistorius, it is no accident that because of the order in which
Kant proceeds in the Groundwork, of beginning with moral principles,
he then ends up with a conception of the good that is at odds with
a certain more traditional conception, which holds that moral good-
ness and the good for the individual form complementary aspects of
the summum bonum, in which such individuals can therefore take an
interest and find some satisfaction. Pistorius therefore argues that his
method rather than Kant’s will result in a conception of the good that
is better able to explain what (as we have seen) Kant himself struggles
to explain in the Groundwork, namely our ‘interest’ in morality, and
how this relates to our well-being as agents.
Now, as I understand Kant’s strategy in responding to Pistorius in
Chapter ii of the Analytic, his approach is to agree with Pistorius that
by trying to think about the good prior to determining any moral
principles concerning what is right and wrong, the conception of the
good that will emerge is precisely going to be one that sees the good in
terms of happiness and well-being; but, far from that being an advan-
tage for Pistorius’s proposal (as he himself supposes), it in fact shows
what is wrong with it, as the resulting outcome is a distortion in our
moral thinking – or so Kant maintains.
Kant thus begins the chapter by characterising the position taken
by Pistorius, and tracing out what he takes to be its consequences:
If the concept of the good is not to be derived from an antecedent
practical law but, instead, is to serve as its basis, it can be only the
concept of something whose existence promises pleasure and thus
determines the causality of the subject, that is, the faculty of desire, to
produce it. Now because it is impossible to see a priori which represen-
tations will be accompanied with pleasure and which with displeasure,
38 Understanding Mor al Obligation

it would be up to experience alone to make out what is immediately

good or evil.78

Kant’s claim, then, is that his opponent will end up conceiving of the
good as pleasure, and that therefore his position will then have to rely
on experience to determine what is good and bad and hence right and
wrong, because it is experience that tells us what brings pleasure and
what does not.
In response, Kant argues that goodness as well-being and moral
goodness are clearly distinct notions, although in languages that fail
to have words to mark this distinction (such as Latin, which has bonum
for both) there may be a resulting confusion on this point.79 And he
argues that reason rather than desire forms the basis for our moral
judgements, but if goodness were pleasure or well-being, then reason
would be reduced to a merely instrumental role, while what is good
would always be a conditional judgement, based on how people hap-
pened to be affected.80
Now, it is precisely at this point that Kant brings up ‘the paradox of
method in a Critique of Practical Reason’, in terms that echo those used
in the Preface and which were quoted by Rawls: ‘namely, that the con-
cept of good and evil must not be determined before the moral law (for which, as
it would seem, this concept would have to be made the basis) but only (as was
done here) after it and by means of it’.81 Kant clearly thinks he has shown
why, after all, it was worth adopting this ‘paradoxical’ method of first
trying to identify the principles that make up the moral law, and only
then attempting to form a conception of good and evil – for otherwise
we will end up making Pistorius’s mistake of arriving at an ethics that
is too eudaimonistic and empiricist. And, Kant claims, because most
moral theorists have in fact proceeded in Pistorius’s way (even ‘ration-
alist’ ones), this has also led them to a heteronomous conception of
ethics in Kant’s sense, because the individual has been seen as drawn
towards the good and hence moral action more by his inclinations
than by reason.82

78
CPrR 5:58 (p. 186). 79
CPrR 5:58–62 (pp. 187–90).
80
CPrR 5:62 (p. 190). 81 CPrR 5:62–3 (p. 190).
82
CPrR 5:64–5 (pp. 191–2): ‘Now, whether they placed this object of pleasure, which was
to yield the supreme concept of good, in happiness, in perfection, in moral feeling,
or in the will of God, their principle was in every case heteronomy and they had to
come unavoidably upon empirical conditions for a moral law, since they could call
their object, as the immediate determining ground of the will, good or evil only by
its immediate relation to feeling, which is always empirical. Only a formal law, that is,
 Kant, mor al realism, and autonomy 39

We have seen, then, how it is that Pistorius’s objection to the

Groundwork shapes Kant’s discussion here, where that objection relates
to the method of inquiry into what the principles of morality are, and
whether one should conduct that inquiry by beginning from a con-
ception of the good. In defending his alternative procedure, Kant is
claiming that Pistorius’s strategy (despite its undoubted plausibility)
means that the conception of the good with which one starts will then
inevitably be a eudaimonistic conception, so that then only a distorted
account of the principles can be the result. As Kant writes:
[S]ince this concept [of the good] had no practical a priori law for its
standard, the criterion of good or evil could be placed in nothing other
than the agreement of the object with our feeling of pleasure or dis-
pleasure, and the use of reason could only consist partly in determining
this pleasure or displeasure in connection with all the feelings of my
existence and partly in determining the means for providing myself
with the object of such feelings.83

Now, when seen in this light, it should therefore be clear that Kant’s
claim is that is a mistaken method of inquiry that leads to error here and
which he therefore rejects, not a mistaken meta-ethical position. The
difficulty arises, that is, because the theorist sets out to think about
moral principles by beginning from a conception of the good that is
as yet unconstrained by any such principles, and so is inevitably insuffi-
ciently ‘moralised’ (as it were). But this is no part of the realist’s position,
in claiming that a principle like the Formula of Humanity must have
some value underlying it as what grounds its validity; for this good is not
the good of well-being, and may well be one that could only be seen to
be a good by someone who first identifies the Formula of Humanity as a
moral principle, and then asks what value could make it so.
Indeed, something very like this latter view is something we could
plausibly attribute to Kant himself, in a way that shows there to be

one that prescribes to reason nothing more than the form of its universal lawgiving as
the supreme condition of maxims, can be a priori a determining ground of practical
reason. The ancients revealed this error openly by directing their moral investigation
entirely to the determination of the concept of the highest good … The moderns, with
whom the question of the highest good seems to have gone out of use or at least to
have become a secondary matter, hide the above error (as in many other cases) behind
indeterminate words; but one can still see it showing through their systems, since it
always reveals heteronomy of practical reason, from which an a priori moral law com-
manding universally can never arise’. For a further discussion of the accuracy of Kant’s
criticism of ‘the ancients’ here, see Engstrom 1996.
83
CPrR 5:63 (p. 191).
40 Understanding Mor al Obligation

no great change of outlook between the Groundwork and the Critique

of Practical Reason. That is, in the Groundwork, on this account, Kant
identifies the value that underlies the Formula of Humanity but in
a way that proceeds in accordance with his method of beginning his
inquiry by attempting to identify such principles without starting out
from any ‘pre-moral’ conception of value; and then in the Critique of
Practical Reason, in the face of Pistorius’s accusation that this method
gets things back to front, and so ends up with a misconception of how
moral principles relate to the good, Kant here defends himself. This
reading of this chapter from the second Critique, therefore, not only
shows how Kant can here be read as a value realist, but also shows how
this later work can therefore be made compatible with the elements of
value realism which even some constructivists recognise are present in
that earlier text.

Conclusion
In this chapter, I have attempted to set out the ‘standard story’ of the
history of modern ethics as a rejection of value realism on the basis of
heteronomy, where Kant is seen as the starting point for this narrative.
I have argued, however, that it is in fact harder than this story sup-
poses to identify any such concern in Kant’s work, and so it is wrong
to see him as subscribing to the kind of argument from autonomy put
forward in his name by contemporary constructivists. I will now go on
to show, in the next chapter, that Kant’s concerns were more narrowly
focused on the problem of moral obligation, and thus how the obligatoriness
of morality could be accounted for without heteronomy, where seeing
things in this light will give a different trajectory to the story to be told
about ethics after Kant.
 2

T h e a rgu m e n t f rom
au tonom y a n d t h e probl e m
of mor a l obl ig at ion

In the previous chapter, we presented reasons for thinking that Kant

did not intend to offer an argument from autonomy against value real-
ism, so that the standard story of the history of modern ethics that
begins from this assumption is misleading; that history will therefore
be retold in later parts of the book. As a new starting point for that
retelling, in this chapter I will suggest that the concerns about auton-
omy which take Kant in an anti-realist direction are raised not by the
realist conceptions of value, but by accounts of the nature of moral obli-
gation, particularly those divine command accounts which understand
that obligatoriness in terms of God’s will. I will claim, therefore, that
it is against this target, and not moral realism, that Kant’s argument
from autonomy should be conceived. I will suggest that when viewed
in this way, the argument itself can be made more plausible, and also
better fits the textual evidence in Kant; in later chapters we will then
see how Kant’s own account of obligation works, and how this account
was challenged by his successors, showing how Kant’s legacy is differ-
ent from the one standardly proposed.

The problem of moral obligation

As we saw in the previous chapter, one difficulty constructivist inter-
preters have had in attributing the argument from autonomy to Kant
is in finding a way to make it compelling as a critique of value real-
ism: why, if there is such an ‘independent order of value’ (in Rawls’s
phrase), should this thereby render us heteronomous? We saw how
Rawls says remarkably little to spell out this worry, while attempts to
do so by others in Kant’s name seemed to be unsuccessful, both philo-
sophically and interpretatively.

41
42 Understanding Mor al Obligation

However, if we turn from the issue of value to the issue of the appar-
ently obligatory nature of morality, worries about heteronomy seem to
arise much more easily and plausibly. For it is widely held that morality
is a matter of obligation, and to that extent a matter of laws or princi-
ples or duties that bind or command us, with a special kind of imperatival
force. But then, it may appear natural to think that any such system of
laws or principles will require a legislator to account for this, so that mor-
ality can only come about as a result of some act of legislation that the
legislator undertakes – thus, as Korsgaard has put this view, ‘Obligation
must come from law, and law from the will of a legislating sovereign;
morality only comes into the world when laws are made’.1 It may then
seem, however, that we are left with only two options: either morality is
legislated by us, or by some other sovereign authority, such as God. Now,
if we opt for the second alternative, concerns about autonomy can then
arise very naturally: for it will now appear that in following the dictates
of morality as these are laid down independently of us, we are follow-
ing the directives of another will with commanding authority over us,
as the institutor of the moral law. Faced with this alternative, the first
option can seem to be the only way of preserving our autonomy, where
we become the legislators of the moral law ourselves. Thus, taken in
this form, as directed against divine command theories of moral obli-
gation, it is much easier to give an argument from autonomy some real
intuitive bite, whether or not it turns out to be ultimately compelling.2
Now, in what follows we will explore the way in which this might pro-
vide a better context for understanding Kant’s own concerns, and thus
a more convincing account of his argument from autonomy. The claim
will be that Kant’s position is not to be understood precisely along the
lines sketched above, as his worry about autonomy arises somewhat dif-
ferently; nonetheless, this provides a better approximation to Kant’s
position than the argument considered in the previous chapter, in
terms of its target and motivations. We will also explore the way in
which Kant’s resulting position then differs from that attributed to him

1
Korsgaard 1996c: 23. Cf. also Anscombe’s discussion of the ‘special moral sense’ of
‘ought’ and its connection with the ‘law conception of ethics’ in Anscombe 1958.
2
Cf. Nowell-Smith 1961: 13–14: ‘The idea of heteronomy is also strongly marked in
Christian morality. “Not as I will, but as thou wilt.” The demand made by Christianity
is that of surrendering self, not in the ordinary sense of being unselfish, of loving our
neighbour and even our enemy. It is the total surrender of the will that is required’.
Proponents of divine command theories have often taken on the task of responding to
Nowell-Smith’s objection: see e.g. Adams 1979a; Mouw 1990: 10–21; Hare 2009: 267.
See Conclusion for further discussion of this issue.
 Autonomy and mor al obligation 43

by the constructivist, once it is shown that Kant’s aim was not to avoid
value realism, but rather divine command accounts of obligation.
We will consider what I take Kant’s alternative account of obligation
to be in the next chapter. In this chapter, I will lay out the context for
that account, and show how Kant came to see the problem of moral
obligation relating to the issue of autonomy. I will also consider an
important challenge to what I say about this, based on the thought
that my treatment of Kant on this question overlooks or ignores the
place of theology and God in his ethics. The challenge is that when
this is properly appreciated, it shows that Kant would not have sub-
scribed to anything like the argument from autonomy outlined above,
so that in this case I am no better off than the constructivist in giving
that argument a genuine place in Kant’s ethics.

Approaches to the problem of moral obligation

Interest in the nature of moral obligation itself, of course, is in no
way original to Kant, and has a considerable history, leading to a wide
variety of approaches. As it arose within medieval Christian thought
onwards, the central issue raised by the apparent obligatoriness of mor-
ality concerned whether this had to depend on the authority of some
divine commander or sovereign, and whether or not that authority had
the freedom and power to make any act obligatory by so commanding
(as the voluntarists claimed), or whether it was limited in what it could
legitimately command by the prior rightness of certain actions or prin-
ciples. Thus, on some alternative accounts, the idea was that morality
constituted a ‘natural law’ in which God played a more indirect role,
where an act is made right and hence obligatory because it conforms
to the nature of things, and where God is the source of that nature as
creator, but not the source of obligatoriness as commander.3 And there

3
Cf. Haakonssen 1996: 6: ‘At issue was the old problem of whether natural law had
moral force for humanity solely because it was God’s will or whether in addition it has
independent moral authority with us. Few disputed that natural law existed because of
God’s will; the question was rather whether or not there were moral values shared by
God and humanity which entailed the moral obligations of natural law independently
of our regard for God’s willing this to be so’. Cf. also Haldane 1991: 137: ‘It is important
to appreciate that in the foregoing [natural law] account the role of God in relation to
the moral law is an indirect one. An action is good because fitting, given the nature of
things – a nature owing to God’s design and manufacture. But according to the second
view mentioned above [i.e. voluntarism], God’s role is wholly direct, for the natural law
is nothing other than a body of legislation willed into existence by God for the govern-
ance of human affairs. And this law need have no relation to the design of the created
44 Understanding Mor al Obligation

were also what have been called ‘intermediate’ positions4 that held that
what is right becomes an obligation through God’s command that it
be done (hence opposing some in the natural law tradition), but that
rightness itself is prior to and independent of obligation and hence
of God’s will (hence opposing voluntarists). Moreover, on this view, in
being commanded God adds a new moral dimension to what was there
before (so, for example, because it violates God’s will, telling lies is not
merely bad or wrong, but sinful or sacrilegious). We might therefore
represent the basic alternatives here as follows, along with some of their
representative proponents:5

Does obligation require command?

Yes: divine command theory No

Are there any constraints on what can be commanded? Moral rightness implies obligation, so
obligation without command
(natural law: e.g. Aquinas, Vasquez,
Clarke, Cudworth, Price)

No: so no intrinsic rightness Yes: constrained by what is right, which has no obligatory force
(voluntarism: e.g. Ockham, Luther, Calvin) until commanded, where that command adds a further
normative dimension
(intermediate position: e.g. Suarez, Scotus, Culverwell)

world’; and Finnis 2002: 34: ‘[W]hen Aquinas, following Augustine, says that the nat-
ural moral law (and thus all just human, positive law) has its obligatoriness “from the
eternal law” he is referring not to a divine command but rather to the intelligibility,
goodness, beauty, and rational attractiveness of the great scheme of things chosen, in
creating, by divine wisdom’.
4
Cf. Irwin 2004: 148, and 2008: 3–4. Irwin identifies this ‘intermediate’ position par-
ticularly with Suarez, who in his Tractatus de Legibus ac Deo Legislatore [On Law and God
the Lawgiver], characterises his account of natural law as ‘a middle course’ because he
holds that while the natural law is indicative of good and bad in itself, it is also prescrip-
tive, in a way that in addition requires prohibitions by a divine superior to account for
this (cf. Suarez 1612: Book ii, Chapter vi, §5, p. 121/1944: 191).
5
Of course, such classifications are always a complex and disputable matter, so e.g.
Aquinas might perhaps be placed by some in the ‘intermediate’ position (as indeed
 Autonomy and mor al obligation 45

In the debate between the proponents of these positions, five dif-

ferent aspects to the issue of moral obligation can be identified, which
can be interrelated in a variety of ways:
(1) Motivational aspect: In acting morally, it is widely held, we should
not be motivated to act in a narrowly self-interested and pruden-
tial way, with a view to our own happiness or desire-satisfaction.
One issue of dispute between natural law theorists and divine
command theorists concerned which position best captured this
aspect of what it is to follow a moral obligation. On the one hand,
natural law theorists connected morality to the good, while div-
ine command theorists connected it to respect or obedience or
gratitude owed to a superior. The former may therefore appear
worse off on this matter than the latter, in so far as the good here
may be understood in terms of happiness and well-being, and thus
merely as relating to the interests of the moral agent. On the other
hand, it can be argued that the good appealed to by natural law
theorists is not best conceived of in these hedonistic or egoistical
terms, as the good in question may be said to break down any dis-
tinction between the agent’s good and the good for others, and
thus involve a significant altruistic dimension. At the same time, if
the ground on which the divine will is said to exercise its authority
over us is related too closely to his power to punish and reward,
then the divine command account can appear to reduce moral-
ity to expediency in its own way, where if this authority is not so
explained, then it may appear unclear how God’s will can motivate
us to be moral or give us any reason (unless it is somehow related
to the good) why we should act morally.
(2) Normativity aspect: It seems plausible to distinguish between the
moral sense of ought and other senses, such as the ‘advice’ sense
(‘you ought to pay more into your pension’), or the sense it has
in certain non-moral contexts (‘you ought not to speak with your
mouth full’; ‘after three years in the post, you ought to now step
down’). Morally obligatory acts therefore seem to have a different
moral dimension from other actions, so that failing to do them

he is by Suarez), whilst others might deny that Scotus is a divine command theorist at
all, while the voluntarism of figures such as Pufendorf and Hobbes can also be hard to
classify. Given that I obviously lack the space to enter fully into such scholarly debates
here, my placings are therefore best considered as indicative only: my main interest is
in the ‘conceptual map’ itself, rather than defending accounts of where exactly to place
particular philosophers upon it.
46 Understanding Mor al Obligation

is wrong or wicked or evil, not merely foolish or impolite or uncon-

stitutional. One area of dispute between divine command theo-
rists and their opponents, therefore, is whether this normative
dimension can be captured satisfactorily if the morally right is not
thought of as being commanded by God – just as one might plaus-
ibly say that to conceive of an action as sinful and so wrong in this
special sense is not really intelligible unless one relates this action
to what God asks of us, and how his expectations have been disap-
pointed. Thus, proponents of the ‘intermediate’ position such as
Suarez argued that while there might be things that are good or
even right for us to do without God commanding them, they could
not have the normative dimension that is distinctive of morality
unless this were so;6 on their part, by contrast, this was denied by
natural law theorists such as Vasquez, who argued that no lawgiver
is required in order to give certain actions the special normative
character that belongs to the moral ‘ought’.7
(3) Epistemological aspect: Another element of dispute was how far
reason could be relied upon to give us guidance on moral mat-
ters, without depending on the assistance of any higher authority
on such questions. It was characteristic of natural law theorists to
be more optimistically rationalistic on this score; for many divine
command theorists, by contrast, such reliance on our own reason
was misguided, and instead it was argued that we must depend on

6
Cf. Suarez 1612: Book ii, Chapter vi, §17, p. 126/1944: 202: ‘Therefore, my own [view]
is that in any human act there dwells some goodness or evil, in view of its object, consid-
ered separately in so far as that object is in harmony or disharmony with right reason;
and that, in its relation to right reason, such an act may be termed an evil, and a sin,
and a source of guilt, in view of the considerations above mentioned, and apart from its
relation to law, strictly speaking. In addition to this [objective goodness or wickedness],
human actions possess a special good or wicked character in their relation to God, in
cases which furthermore involve a divine law, whether prohibitory or preceptive; and
in accordance with such laws, these acts may in a special sense to be said to be sins or
to involve guilt in the sight of God, by reason of the fact that they transgress a true
law of God Himself. It was to this special form of wickedness that Paul [Romans 4:15]
apparently referred in the term “transgression”, when he said: “For where there is no
law, neither is there transgression” ’.
7
Cf. Vasquez 1606: 150 c. 3 §22, p. 7: ‘[S]ome things are evils and sins from themselves
in such a way that that this prohibition depends on no will, even the will of God’. For
a more recent defence of this natural law position, which questions the role of any
‘acts of will’ in creating obligatoriness, see Finnis 1980, esp. Chapter xi. Finnis in fact
interprets Vasquez as an opponent rather than an ally, and puts him close to Suarez on
this matter: see pp. 44–6. But for a contrary and more plausible reading of Vasquez, see
Pink 2004a and Irwin 2008: 6.
 Autonomy and mor al obligation 47

God to communicate such knowledge to us through the demands

he makes of us, in a way that can then give us confidence that we
are acting rightly. On the other hand, of course, questions could
be raised about the reliability of the channels of this communica-
tion (such as the Bible or direct revelation), where it then seemed
that the divine command theorist might have to rely on reason for
its proper interpretation, so giving their position no obvious epi-
stemic advantage over the natural law model.8
(4) Phenomenological aspect: It has often been noted that it is charac-
teristic of our experience of morality that we feel coerced or bound
to act in accordance with it, and thus that it is required of us. It was
then claimed by divine command theorists that this phenomeno-
logical feature of our moral lives could not be explained by appeal
to our own reason, as this is a capacity for judgement and so holds
no power over us, and even if it did, as this power comes from
within, it is too easy for us to set aside and treat as nugatory; it must
therefore come from the force exercised over us by another will.9
In response, natural law theorists argued that reason is not merely
a capacity for making judgements, but can also give those judge-
ments motivational effect, while claiming that obligation need not
be experienced merely negatively as a power to which we must sub-
mit, but instead as a good towards which we are inevitably drawn.
(5) Legislative aspect: In so far as all three positions connected moral
obligation to the idea of law, they all in some sense connected it
to the thought that it involved the taking away from the agent of
his freedom to act otherwise, and thus such obligations could not
be self-imposed but must come from an external will, where God
was the obvious candidate to serve this role. However, as we have
seen, in natural law theory that connection to God’s will is indir-
ect, related to his plans for the creation, while in voluntaristic and
intermediate theories, it is more direct. The question raised for
natural law theories, therefore, was whether this indirectness left
it unclear how a loss of liberty was in fact involved, particularly
when such laws related to the good of the individual concerned,
who might then be expected to act spontaneously without any con-
straint at all, and thus without any obligation in this sense. But the
8
For an illuminating discussion of how this epistemological dispute played out in the
work of Hobbes, Selden and Cumberland, see Parkin 1999, especially pp. 56–72.
9
For an account of Suarez that emphasises this phenomenological aspect, see Crosby
2004.
48 Understanding Mor al Obligation

question raised for theories that saw this loss of liberty as coming
about through the imposition of God’s will, was how to distinguish
such imposition from mere coercion and thus an illegitimate tak-
ing away of the agent’s freedom, and how to explain the motiv-
ations involved in obeying this will without reducing them to mere
fear or terror. Put simply, the problem may be seen as a dilemma:
on the one hand, if a moral obligation asks us to do what we are
already inclined to do, how can it be said to have a binding force on
us, as we are going to do it anyway? But if it does bind us and so to
some extent takes away our freedom, how can it do so in a way that
is not then rendered objectionable?
Many of these issues will come up again for discussion in what follows.
In the debate over them, the primary considerations until the modern
period were theological, rather than a concern over autonomy and so
whether this can be accommodated within a divine command theory.
Thus, for example, voluntarists argued that their view best made room
for God’s omnipotence,10 while anti-voluntarists challenged them with
rendering God’s will empty and arbitrary. In a world where subordin-
ation to authority was commonplace, and our relation to God was
naturally seen as hierarchical, it was scarcely surprising that issues of
autonomy were slow to arise.
Nonetheless, those who rejected the divine command model did
see that it had important implications in the account it offered of the
will and moral motivation, of a broadly anti-rationalist kind. Thus, it
was argued that such models assume that the obligatory force of mor-
ality is based on the threat of punishment or the promise of reward as
instituted by God, where it is this that makes an action compulsory for
us; but critics argued that this view of obligation demeaned the moral
‘must’ while also debasing us in relation to it. It was argued instead
that obligations can obtain without any need of sanctions or incentives
in this way, and that we can grasp the obligatory force of morality in its
own right, and so without the need for a divine commander. Thus, for
example, Samuel Clarke argues that while ‘the sanction of rewards and
punishments’ may be ‘absolutely necessary to the government of frail
and fallible creatures, and truly the most effective means of keeping
them in their duty’, nonetheless this ‘is yet really in itself only a second-
ary and additional obligation or enforcement of the first’ or ‘original’

10
As Idziak notes, however, one should be careful not to exaggerate this issue, at least
among the scholastics: see Idziak 1979: 10–13.
 Autonomy and mor al obligation 49

obligation, which is ‘the eternal reason of things’.11 Clarke therefore

concludes: ‘Further yet: as this law of nature is infinitely superior to all
authority of men and independent upon it; so its obligation, primar-
ily and originally, is antecedent also even to this consideration, of its
being the positive will or command of God himself’.12 While not yet an
argument from autonomy, Clarke is clearly concerned here to avoid a
picture of the will where sanctions are always supposed to be required
in order for morality to be made obligatory for us, rather than our
being able to find it obligatory on its own.
Similar issues arise in Leibniz’s brief critique of Pufendorf, which
was originally written in the form of a letter in 1706, but which became
well known in the early eighteenth century through Jean Barbeyrac’s
translation and response to it, which Barbeyrac appended to his trans-
lation into French of Pufendorf’s De officio hominis et civis juxta legem
naturalem [On the Duty of Man and Citizen According to Natural Law]
and published in 1718.13 Pufendorf’s position itself is quite hard to
classify, and while often seen as voluntarist, this voluntarism is moder-
ated in various ways that move it somewhat closer to a more intermedi-
ate position such as Suarez’s – an ambiguity that Barbeyrac exploits in
his reply to Leibniz’s criticisms, which tends to emphasise these more
moderate elements. At the centre of Leibniz’s critique are concerns
about the arbitrariness of God’s will on Pufendorf’s position, and also
that if we allow Pufendorf to define law ‘as a command by which the
superior obliges the subject to conform his action to what the law itself
prescribes’,14 then ‘no one will do his duty spontaneously’.15 Leibniz’s
own position is close to the natural law tradition, so he argues that
‘reasons restrain even by themselves’,16 without the need of fear: ‘To

11
Clarke 1711: 54/2003: 301. Cf. also Clarke 1711: 89/2003: 307: ‘Lastly, this law of
nature has its full obligatory power, antecedent to all considerations of any particular
private and personal reward or punishment, annexed either by natural consequence
or by positive appointment to the observance or neglect of it. This is also very evident,
because, if good and evil, right and wrong, fitness and unfitness of being practiced,
be (as has been shown originally) eternally and necessarily in the nature of the things
themselves, ’tis plain that the view of particular rewards or punishments, which is only
an after-consideration and does not at all alter the nature of things, cannot be the ori-
ginal cause of the obligation of the law, but is only an additional weight to enforce the
practice of what men were before obliged to by right reason’.
12
Clarke 1711: 86/2003: 306. 13 See Leibniz 1706.
14
Pufendorf 1673: Book i, Chapter ii, §2, p. 12, as quoted in Leibniz 1706: 279/1988b:
70. In Pufendorf 1991: 27 this is translated as ‘Law is a decree by which a superior
obliges one who is subject to him to conform his actions to the superior’s prescript
[praescriptum]’.
15
Leibniz 1706: 279/1988b: 70. 16
 Leibniz 1706: 282/1988b: 74.
50 Understanding Mor al Obligation

such a degree is it repugnant to reason to say that only the law or

constraint make a man just’.17 As with Clarke, therefore, the issue was
whether the divine command account could properly accommodate
our motivation to be moral, without distorting this into a purely pru-
dential matter.
Likewise, as Schneewind has noted, in his Theodicy Leibniz pro-
tests explicitly against voluntarists such as Hobbes and William King,
fearing that this position puts us in a relation of obedience to God,
rather than love.18 He thus sets out ‘to banish from men the false ideas
that represent God to them as an absolute prince employing a des-
potic power, unfitted to be loved and unworthy of being loved’.19 As
Schneewind has also noted, similar concerns had been raised previ-
ously by the Cambridge Platonists, where ‘John Smith contrasted the
“servility” required by voluntarism with the “freedom and liberty of
soul within us” that comes from a “right knowledge of God” ’.20
As well as these historical debates on the issue of obligation, con-
troversy shaped by them was also reflected in the philosophical dis-
putes of Kant’s own period and milieu, particularly in the work of
Christian August Crusius, Christian Wolff, and Alexander Gottlieb
Baumgarten, and their followers.21 Crusius defended a voluntarist pos-
ition, according to which ‘ just this necessary will of God is the ground
of the morally good, from which we see that the morally good, like all
other goods, has its grounds in a will’.22 He also saw obligation as aris-
ing out of obedience to God, to whom we are indebted: ‘A law is a uni-
versal will of a more powerful being who does not have another more
powerful being over him, through which an indebtedness springing
from this will is imposed on those subordinate to him to do or omit
something’.23 Crusius was here in part reacting against the influential

17
Leibniz 1706: 280/1988b: 72.
18
See Schneewind 1996b: 26–7/2010: 204–5.
19
Leibniz 1710: Part i, §6, p. 106/1998: 127.
20
Schneewind 1996b: 27/2010: 205, where Schneewind is citing from Smith 1859: 28.
For a statement of Smith’s antivoluntaristic rationalism, see p. 137: ‘We must not con-
ceive God to be the freest agent, because He can do and prescribe what He pleaseth,
and so set up an absolute will which shall make both law and reason, as some imagine’.
Cf. also Schneewind 2000: 214/2010: 225–6.
21
For a classic discussion of this background to Kant’s ethics, see Schmucker 1961. For a
more recent study, see Schwaiger 2009.
22
Crusius 1767: §173, p. 248/2003: 579.
23
Crusius 1767: §165, p. 238/2003: 578. Cf. Kant, LE 27:262 (p. 55): ‘Crusius believes
that all obligation is related to the will of another. So in his view all obligation would
 Autonomy and mor al obligation 51

position proposed by Wolff. While Wolff began as a voluntarist under

the sway of Pufendorf, he was led by Leibniz’s critique of the latter
to abandon the position, and to adopt instead a more perfectionist
account that was therefore closer to a natural law theory, according
to which ‘human acts are good or evil, or better or worse, before one
can say that God wills or does not will them’.24 Wolff therefore gave a
rather minimal place to God within his account,25 according to which
the natural law is also divine not because God stands over us as a com-
manding authority, but because he is the author of the world and has
structured things in such a way that we are motivated to act rightly
in pursuit of our perfection and that of others. For, Wolff held, ‘[t]o
obligate someone to do or omit something is only to connect a motive
of willing or not willing to it’.26 Wolff can therefore say that ‘[i]n this
way simultaneously, natural obligation is divine obligation, and the
law of nature is a divine law’, in the sense that God has ‘connected
the motives with the actions, and accordingly he obligates men to do
what the law of nature requires’.27 Thus, Wolff argues, ‘Men of under-
standing and reason need no obligations beyond the natural’; and
while some may require ‘[t]he servile fear of the force and might of
a superior [to] restrain them from doing what they would like to do’,
to guide a person in this way ‘through compulsion’ is to treat him as
an animal or beast [Vieh].28 Finally, Baumgarten’s approach is broadly
similar to Wolff’s. He therefore treats seeking perfection as one of
the ‘obligating propositions’ of morality in so far as we are motivated
by nature to follow it, making it a natural law in this sense;29 but he
also treats such propositions as divine laws, in so far as God is the cre-
ator of nature and thus their ultimate source, who can therefore be

be a necessitation per arbitrium alterius [by the choice of another]. It may indeed seem
that in obligation we are necessitated per arbitrium alterius; but in fact I am necessitated
by an arbitrium internum, not externum’.
24
Wolff 1733: §35, p. 27.
25
Cf. Wolff 1733: §38, pp. 28–9/2003: 337: ‘Because a reasonable man is a law unto
himself and, besides natural obligation, needs no other, neither rewards nor punish-
ments are, for him, motives to do good acts and avoid bad ones. So the reasonable man
does good acts because they are good and does not do wicked ones because they are
wicked. In this case he becomes like God, who has no superior who can obligate him
to do what is good or not do what is wicked, but does the one and not the other simply
because of the perfection of his nature’.
26
Wolff 1733: §8, p.8/2003: 335. 27
Wolff 1733: §28, p. 20/2003: 337.
28
Foreword to the second edition (which has no page numbers) of Wolff
1736/2003: 334.
29
Cf. Baumgarten 1760: §39, p. 17; §70, p. 34/Kant 1900– : 19:23 and 19:34–5.
52 Understanding Mor al Obligation

called their legislator.30 In his lectures on moral philosophy, Kant was

to use Baumgarten’s main works on ethics as his textbooks, including
the Initia philosophiae practicae primae [Introduction to Practical First
Philosophy], in which discussions of obligation are central.31
Given this background, then, it is unsurprising to find that one of
the most significant issues in Kant’s ethics is how to account for obli-
gation and the imperatival force of morality. As we have already seen,
Kant fears that unless certain fundamental questions of this sort can
be answered, we may come to think of morality as a ‘phantom’ or as
‘chimerical’: so as well as seeking to explain the interest that we take
in moral action, and the freedom of the will that this action involves,
Kant therefore also feels the need to explain how moral obligation is
possible. Thus, just as in the theoretical case, where Kant is concerned
that worries about metaphysical necessity might lead us (as it did
Hume) into doubting that there is any synthetic a priori knowledge, so
in the practical case, Kant is concerned that doubts about the necessity
of the moral ‘must’ might lead us into doubt concerning the moral law.
As Kant puts it: ‘This question does not inquire how the performance
of the action that the imperative commands can be thought, but only
how the necessitation of the will, which the imperative expresses in the
problem, can be thought [or conceived, or made sense of: gedacht]’.32
If the ‘must’ in ‘you must not tell lies’ is not explained, therefore, this
can leave us wondering how there can be any such necessity33 – just as
in the case of metaphysical necessity, we can be left wondering how it
can be the case that every event must have a cause.

30
Cf. Baumgarten 1760: §100, pp. 61–2/Kant 1900– : 19:48: ‘The author of the obli-
gation announced by the law is said to declare that law, and he who has the right to
declare the laws is said, in an extended sense, to be a legislator, and he is the legislator
of that law which he has declared. Now God is the author of the nature of the universe,
and of all realities that come to pass, and moreover all natural obligations are some-
thing real and positive, and have sufficient reason in the same. Therefore God is the
author of obligation, and thence also of the natural law. Since with respect to which
things he has the highest right, he is called, broadly, the legislator of the natural law
and the whole natural law’.
31
For an edition of Kant’s lectures that sets out their relation to Baumgarten very clearly,
see Kant 2004.
32
GMM 4:417 (pp. 69–70), my emphasis.
33
Cf. Garner 1990: 141 and 143: ‘How could any feature of something outside us make
it the case that we are objectively required to do something? … It is the peculiar com-
bination of objectivity and prescriptivity…that makes moral facts and properties queer
… It is hard to believe in objective prescriptivity because it is hard to make sense of
a demand without a demander, and hard to find a place for demands or demanders
 Autonomy and mor al obligation 53

Kant, divine command theories, and the

argument from autonomy
Now, as we have seen, one way to account for this necessitation is to
adopt a divine command theory of morality, which grounds the obliga-
tory force of the ethical demand in the superior power and authority
of a divine legislator. And we have also seen how objections were raised
to such accounts on broadly theological grounds, particularly when
such theories took a voluntaristic form. For Kant, however, the prob-
lem with solving the question of obligation in this way is not primar-
ily theological, but is closer to the concerns of someone like Clarke:
namely, that an account of obligation which takes this route will end
up explaining obligation in terms of sanctions. Kant then adds a new
dimension to this worry by relating it to his conception of autonomy,
in arguing that to understand obligation along these lines is to make
acting under it heteronomous; for the obligatory force of morality is
then seen to rest in the threat of punishment or the hope of reward,
so that the will is thereby governed by the workings of our desires and
interests, to which reason is then subordinated. We can therefore find
an argument from autonomy in Kant, but one that is directed not at
value realism, but at the divine command model of obligation. At the
same time, because it employs Kant’s distinctive understanding of the
autonomy/heteronomy distinction, it differs from the simpler argu-
ment from autonomy sketched at the beginning of this chapter.
Kant’s position can be seen most clearly in the discussion at the end
of Section ii of the Groundwork, which we have already considered. As
discussed previously, Kant here differentiates between autonomy and
heteronomy, where this difference turns on whether reason or desire
is the primary determinant of the will: that is, whether reason merely
‘serves’ our inclinations in a subordinate role (resulting in heteron-
omy), or rather ‘commands’ the will in its own right (resulting in auton-
omy). Now, the purpose of the first two sections of the Groundwork
that lead up to Kant’s introduction of this distinction is to identify the
‘supreme principle of morality’,34 where Kant has argued that the best

apart from human interests and conventions. We know what it is for our friends, our
job, and our projects to make demands on us, but we do not know what it is for reality to
do so’. Garner is of course here explicating one aspect of J. L. Mackie’s famous ‘argu-
ment from queerness’ (cf. Mackie 1977, especially pp. 38–42).
34
Cf. GMM 4:392 (p. 47): ‘The present groundwork is, however, nothing more than the
search for and establishment of the supreme principle of morality’, where the ‘search’ is
conducted in Sections i and ii, and its ‘establishment’ in Section iii.
54 Understanding Mor al Obligation

candidate for this principle is the Formula of Universal Law, and the
related formulae. Then, at the end of Section ii,35 he attempts to use
his autonomy/heteronomy distinction to diagnose where previous the-
orists who have searched for this ‘supreme principle’ have gone wrong;
and his diagnosis is that they have done so because they have treated
the will in terms that are heteronomous rather than autonomous. This
has had the result, Kant argues, that they have produced moral princi-
ples that relate to our desires and interests as their ‘object’, rendering
them material and not formal and hence hypothetical and not cat-
egorical, where the latter is required if they are to fit with our concep-
tion of what it is for something to be a duty.36
Kant makes clear his diagnostic ambitions here in the following
passage:
If we look back upon all previous efforts that have ever been made to
discover the principle of morality, we need not wonder now why all of
them had to fail. It was seen that the human being is bound to laws by
his duty, but it never occurred to them that he is subject only to laws given
by himself but still universal and that he is bound only to act in conformity
with his own will, which, however, in accordance with nature’s end is a
will giving universal law. For, if one thought of him only as subject to a
law (whatever it may be), this law had to carry with it some interest by
way of attraction or constraint, since it did not as a law arise from his
will; in order to conform with the law, his will has instead to be con-
strained by something else to act in a certain way. By this quite necessary
consequence, however, all the labor to find a supreme ground of duty
was irretrievably lost. For, one never arrived at duty but instead at the
necessity of an action from a certain interest. This might be one’s own
or another’s interest. But then the imperative had to turn out always
conditional and could not be fit for a moral command. I will therefore
call this basic principle the principle of the autonomy of the will in con-
trast with every other, which I accordingly count as heteronomy.37

These diagnostic ambitions are also evident in the titles of the final
sub-sections of Section ii: ‘Heteronomy of the Will as the Source of
all Spurious Principles of Morality’,38 and ‘Division of all Possible
Principles of Morality Taken from Heteronomy as the Basic Concept’.39
35
Related discussions can be found at CPrR 5:40–2 (pp. 172–3), and at various points in
his lectures on ethics (see e.g. LE 29:621–9 (pp. 239–46)).
36
For a helpful general discussion of Kant’s position here, see Kerstein 2002, especially
Chapter 7.
37
GMM 4:432–3 (pp. 82–3).
38
GMM 4:441 (p. 89). 39 GMM 4:441 (p. 90).
 Autonomy and mor al obligation 55

As these subtitles make clear, Kant thinks that once one begins from
a heteronomous conception of the will, this then acts as a ‘source’ for
the concoction of ‘spurious’ principles of morality, where he believes
that he can show systematically the kind of wrong-headed principles
that will arise as a result. Thus, Kant claims, until a new conception of
the will emerged out of his own critical philosophy, moral theorising
concerning the fundamental moral principle was bound to take the
wrong path: ‘Here, as everywhere else, human reason in its pure use,
as long as it lacks a critique, first tries all possible wrong ways before it
succeeds in finding the only true way’.40
Now, it is in this context that Kant brings up divine command the-
ories of morality, claiming that they too conceive of the will in heter-
onomous terms, along with the other theories he mentions here, such
as moral sense theories and perfectionism. Moreover, Kant also claims
that these divine command theories are significantly worse than these
other two positions, because while all three are heteronomous, these
at least do not ‘infringe morality’41 in the way that (he claims) divine
command theories do. We need to see, therefore, what it is about the
‘theological concept’ of morality which shows it to have been ‘taken
from heteronomy’, and also what makes it potentially ‘anti-moral’.
This theological position, Kant argues, attempts to derive morality
from a ‘divine, all-perfect will’, and so makes conforming to this will
into its principle (‘do what God commands’). However, Kant claims, it
then faces a dilemma. On the one hand, we could then see our moral
duties as arising from God’s perfection and hence the rightness of
what he wills, in which case the appeal to his willing plays no explana-
tory role in grounding those duties;42 and the accompanying supreme
principle would then collapse into telling us to do what is right, in a
way that is unhelpful and empty.43 On the other hand, we could see

40
GMM 4:441 (p. 90). 41 GMM 4:443 (p. 91; translation modified).
42
Cf. LE 27:9 (pp. 5–6): ‘Supposing the arbitrium [choice] of God to be known to me,
where is the necessity that I should do it, if I have not already derived the obligation
from the nature of the case? God wills it – why should I? He will punish me; in that
case it is injurious, but not in itself wicked; that is how we obey a despot; in that case
the act is no sin, in the strict sense, but politically imprudent; and why does God will it?
Why does He punish it? Because I am obligated to do it, not because He has the power
to punish. The very application of the arbitrium divinum to the factum, as a ground,
presupposes the concept of obligation’. Cf. also LE 27:1426 (p. 69).
43
Cf. Kant’s remarks in his lectures on Baumgarten’s principle Fac bonum et omitte malum
[Do good and omit evil]: ‘Do what is morally good. But in that case there ought to have
been another rule, to tell us what moral goodness consists in’ (LE 27:264 (p. 57)).
56 Understanding Mor al Obligation

our duties as arising from God’s will once his moral qualities are set
to one side. But then, Kant argues, ‘the concept of his will still left to
us’ would be ‘made up of the attributes of desire for glory and domin-
ion combined with dreadful representations of power and vengeful-
ness’, and so would then be ‘the foundation for a system of morals that
would be directly opposed to morality’.44
Now, it is important to recognise that in what he says about the
‘theological concept’ here, Kant is criticising this position at two lev-
els. The first and most general one is that like all the other theories
discussed in this sub-section of the Groundwork, it is heteronomous.
For on this account, what would make morality obligatory would not
lie in its rightness, but in the absolute and supreme power over us of
the divine commander; and this would then render our relation to
morality heteronomous, as our ‘interest’ in so acting would be to avoid
punishment and to gain reward, and not in the rightness of our duty
as such.45 We can see, then, how the ‘theological concept’ of morality
fits the general pattern that Kant claims to identify in all the positions
he discusses here, of thinking of us ‘as subject to a law (whatever it may
be)’, where as a result ‘this law had to carry with it some interest by way
of attraction or constraint’, an interest that in the divine command
case comes from our relation to an all-powerful God.

44
GMM 4:443 (p. 91). Cf. also LE 27:10 (p. 6): ‘The quarrel between reformers and
Lutherans over arbitrium divinum and decretus absolutus [absolute decree] is based on
the fact that even in God, morality must exist; and every conception of the divine
arbitrium itself vanishes, if morality is not presupposed … How dreadful, though, is a
God without morality. The jus naturae divinum, and even positivum, vanishes, if there be
no morality as ground of the relation and conformity of my arbitrium and that of God.
Without the prior assumption of obligation, punishments come to nothing; what God
displays is merely ill-will; the physical consequences I can avoid, and thus the action
is no longer a transgression. Morality is more general than the arbitrium divinum’. Cf.
Leibniz 1706: 280/1988b: 71: ‘Justice, indeed, would not be an essential attribute of
God, if he himself established justice and law by his free will’.
45
Cf. LE 27:307 (p. 96): ‘If we imagine a religion prior to all morality, it would still have to
have a relation to God, and then it would amount to this, that I picture God as a mighty
lord who would need to be placated’. Cf. also LE 29:627–8 (pp. 244–5): ‘People think
that morality should not take precedence over the divine will, so that I cannot say that
God tells us to perform actions because they are duties; they are duties, rather, because
God tells us to do them. But in that case the moral laws would be arbitrary, and we
should not perceive the slightest necessity in them. They would be statuta, having no
power to bind on their own account, but acquiring it through the will of another …
[But] If actions are not grounded upon duty, the cause of their performance must be
the authority of the overlord; yet that is not moral, but merely legal. Actions, in that case,
will be based upon fear and hope. A created law is called sanctio [decree]. Moral laws as
sanctiones seem at first to be very desirable for morality; but if religion is made prior to
 Autonomy and mor al obligation 57

But, in addition, Kant is here also making a narrower point, which

is to claim that the ‘theological concept’ is significantly worse than per-
fectionism (which is the alternative rationalist position), even though
both involve heteronomy; for, he argues, the theological position
can result in a kind of anti-morality, by attempting to ground mor-
ality in a conception of God’s will that is prior to and hence distinct
from morality itself, leaving nothing but ‘attributes of desire for glory
and dominion combined with dreadful representations of power and
vengefulness’ to serve as its basis, with dire results for its ethical prin-
ciples. So, Kant is also arguing here, while perfectionism is no better
than the ‘theological concept’ at serving ‘to support [morality] as its
foundation’46 because it too involves a heteronomous conception of
the will, it at least does not threaten to ‘infringe morality’, and so is at
least to be preferred to that extent.
I have argued, therefore, that when it comes to Kant’s more general
critique of the ‘theological concept’ of morality, it is indeed right to
claim that Kant has an argument from autonomy, but that the target for
that argument is not value realism, but the divine command account of
moral obligation.47 This approach is to be preferred to the one adopted
by the standard story, I have claimed, because it can be given clear text-
ual support and also be made philosophically defensible, in a way that
(I suggested in the previous chapter) the latter approach cannot.

The place of God in Kant’s ethics

However, I now want to consider an important objection to my position,
which is to suggest that my treatment of Kant’s argument from autonomy

morality, the first principle becomes: Obey the divine will, and with that the whole of
morality is destroyed’. And LE 27:262 (p. 56): ‘If I do a thing because it is ordained or
brings advantage, and omit a thing because it is forbidden or brings harm, that is not a
moral disposition. But if I do it because it is absolutely good in itself, that is a moral dis-
position. So an action must be done, not because God wills it, but because it is righteous
or good in itself; it is because of this that God wills it and demands it of us’.
46
GMM 4:443 (p. 91).
47
Cf. Relig 6:3 (p. 57): ‘So far as morality is based on the conception of the human being
as one who is free but who also, just because of that, binds himself through his reason
to unconditional laws, it is in need neither of another being above him in order that
he recognise his duty, nor, that he observe it, of an incentive other than the law itself’.
And cf. also TSP 8:405 (p. 444), where Kant objects to conceiving of the commanding
force of the moral law as something which ‘proceeds from another being’, rather than
‘the absolute authority of [man’s] own reason’, and thus to ‘personifying this law and
making out of morally commanding reason a veiled Isis’.
58 Understanding Mor al Obligation

has no greater textual support than the constructivist’s, as Kant was no

more opposed to divine command theory than he was to value realism,
and that therefore my reading of this part of the Groundwork is mistaken.
This interpretation of Kant as a divine command theorist concerning obli-
gation, and this alternative view of the Groundwork, has been defended in
a series of works by John E. Hare.48 Hare’s strategy is to cite a number of
passages as evidence that Kant was a supporter of the divine command
position, and so to argue that his target in the Groundwork cannot be this
position in general, but a more specific form of it, where Hare proposes
Crusius’s extreme voluntarism as the most likely candidate. If Hare were
correct, therefore, I would be wrong to claim that Kant had a divine com-
mand approach as the target for his argument from autonomy; indeed,
Hare thinks autonomy is perfectly compatible with this approach, and
suggests that his reading of Kant shows that Kant thought so as well.49
Given what I have said above, therefore, it is important for me to consider
Hare’s alternative view in some detail.
We might begin, then, with Hare’s claim that Kant himself was a
divine command theorist of some kind. Hare offers a variety of evi-
dence in support of this interpretation,50 such as Kant’s talk of duties
as divine commands,51 and his remarks in the Groundwork itself about
God being the head of the kingdom of ends.52 He also cites Kant’s
comments in one of his lectures on ethics, that ‘[o]ur conduct towards
God is of three kinds. We may reverence, fear and love Him. We revere
God as a holy law-giver, love Him as a benevolent ruler, and fear Him
as a just judge’.53 As Hare notes, Kant says similar-sounding things else-
where; he argues, therefore, that ‘[b]ecause this is a sustained theme
in Kant, we are better off regarding his attack in the Groundwork as
directed not at divine-command theories in general, but at some spe-
cific variety’,54 where Crusius’s position is his preferred candidate.

48
See, for example, Hare 2000a, 2000b; 2001 especially pp. 87–119; and 2009 especially
pp. 122–75. Because there is a good deal of overlap between the articles and books, I will
only cite the latter in what follows. For a helpful discussion of the issues raised by Hare,
see Kain 2005.
49
Cf. Hare 2001: 114–9, and 2009: 268–9.
50
See Hare 2001: 105, and 2009: 152–3.
51
CPR A811/B839 (p. 680), CPrR 5:129 (p. 244), and Relig 6:153 (p. 177).
52
GMM 4:433–4 (pp. 83–4).
53
LE 27:322 (p. 107). (In Hare 2001, Hare refers to an older translation of Kant’s lec-
tures on ethics, and so gives this quotation in a slightly different form – see p. 105.)
54
Hare 2009: 153. Cf. also Hare 2001: 105.
 Autonomy and mor al obligation 59

However, given the subtleties in Kant’s attitude to religion, it is of

course not possible just to take these comments at face value.55 Kant’s
position is made particularly complex here by the distinction he draws
between a view that treats religion as prior to morality, and one that
treats it as morality’s ‘completion’.56 Kant is clear in rejecting the former
option, fearing that it will lead to a subsequent distortion of morality,
as this religious basis will then itself be insufficiently ‘moralised’.57 By
contrast, Kant is happy to conceive of religion as ‘completing’ morality

55
I should perhaps emphasise here that I agree with Hare’s insistence that it is a mistake
to ‘secularise’ Kant’s position, or to treat his conception of God as merely ‘fictional’ or
hypothetical in some way (an ‘as if’). See Hare 2001: 103–4. I do not take anything that
I say below as ‘weakening’ Kant’s theological commitments in this manner – although
of course, as Hare would himself allow, none of those commitments can amount to
established theoretical knowledge.
56
Of relevance here too is Kant’s related distinction between ‘theological morality’ and
‘moral theology’ respectively. See CPR A632/B660 note (p. 584).
57
See Relig 6:183 (p. 201), where Kant argues that were a person to begin with ‘the con-
cept of a world ruler, who makes of this duty a commandment for us … he would run
the risk of dashing his courage (which is an essential component of virtue) and of
transforming divine blessedness into a fawning slavish subjection to the commands
of a despotic might’. As well as those cited above in note 45, see also the following
passages: CPR A818–9/B846–7 (p. 684): ‘For it was these [moral] laws alone whose
inner practical necessity led us to the presupposition of a self-sufficient cause or a
wise world-regent, in order to give effect to these laws, and hence we cannot in turn
regard these as contingent and derived from a mere will, especially from a will of
which we would have had no concept at all had we not formed it in accordance with
those laws. So far as practical reason has the right to lead us, we will not hold actions
to be obligatory because they are God’s commands, but will rather regard them as
divine commands because we are internally obligated to them’; LE 29:628 (p. 245):
‘Religion is nothing else but morality and theology combined. Prior to morality, the-
ology is not possible … All religion, if morality is built upon it, rests on nothing but a
currying of favours. If the law has been arbitrarily instituted by God, He can also, to
be sure, dispense us from it’; LE 27:10 (p. 6): ‘How dreadful, though, is a God without
morality. The jus naturae divinum, and even positivum, vanishes, if there be no morality
as ground of the relation and conformity of my arbitrium and that of God. Without the
prior assumption of obligation, punishments come to nothing; what God displays is
mere ill-will; the physical consequences I can avoid, and thus the action is no longer
a transgression. Morality is more general than the arbitrium divinum’; LE 27:15 (p. 9):
‘Should the Christian ethic be given priority over philosophical ethics, or vice versa?
One must certainly be explained by the other as theoretical physics is explained from
experimental physics; but the natural ethic must rightly be given priority’; LE 27:18
(p. 11), where Kant criticises ‘our author’ (i.e. Baumgarten) for having an incorrect
method, ‘since it begins from religion, whereas it ought to have started from morality,
which then would be increasingly purified’; LE 27:166 (not translated): ‘All ethics is
chimerical … which puts religion before morality’; LR 28:1002–3 (pp. 348–9): ‘But
moral theology is something wholly different from theological morality, namely, a morality
in which the concept of obligation presupposes the concept of God. Such a theological
morality has no principle; or if it does have one, this is nothing but the fact that the will
60 Understanding Mor al Obligation

in important respects; as he puts it in his lectures on ethics: ‘How does

theology arise? If morality is expounded, the very concept of it brings us
to belief in God’.58
Kant thus argues that while morality cannot be grounded on reli-
gion, we would nonetheless find morality very hard to sustain and make
sense of without combining it with a religious perspective, including
a belief in God’s existence, a belief that is then justified on practical
grounds. Moreover, Kant argues, once this perspective has been taken
up, it then becomes appropriate to think of God as a kind of com-
mander. However, because we are not now using this conception of
God to ground morality but to complete it, we should not think that in
this role it is God who gives validity and obligatory force to the moral
law which otherwise would lack it, as this would make religion prior
to morality. As a result, then, our conception of God’s place as com-
mander or ‘moral ruler of the world’ must primarily be to act as the
judge of those who transgress the law and as executor of it by making
sure the law is followed, by ensuring that virtues are rewarded and the
wicked punished, while our moral efforts are coordinated and we are
given the help we need to act as morality requires.59 Kant therefore
gives God a clear judicial and executive function, which he thinks is

of God has been revealed and discovered. Morality, however, must not be grounded on
theology, but must have in itself the principle which is to be the ground of our good
conduct. Afterwards it can be combined with theology, and then our morality will
obtain more incentives and a morally moving power’; MM 6:434 (pp. 596–7), where
Kant also insisted on this order of priority when it comes to moral education, empha-
sising that the ‘moral catechism’ must come before the religious one, ‘[f]or otherwise
the religion that [the pupil] afterwards professes will be nothing but hypocrisy; he will
acknowledge duties out of fear and feign an interest in them that is not in his heart’.
58
LE 27:306 (p. 96). Cf. also LE 27:715 (p. 438): ‘Religion is thus founded on mor-
ality’. A similar emphasis can be found in the famous Preface to Relig (Relig 6:3–7,
pp. 57–60).
59
Cf. the following passages: Relig 6:139 (p. 165): ‘Since by himself the human being can-
not realise 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 this 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’; Relig 6:99 (pp. 133–4): ‘There must therefore be
someone other than the people whom we can declare the public lawgiver of an eth-
ical community. But neither can ethical laws be thought of as proceeding originally
merely from the will of this superior (as statutes that would not be binding without his
prior sanction), for then they would not be ethical laws, and the duty commensurate
with them would not be a free virtue but an externally enforceable legal duty’; CPrR
5:128–9 (pp. 243–4): ‘The Christian doctrine of morals now supplements this lack
(of the second indispensable component of the highest good) by representing the
world in which rational beings devote themselves with their whole soul to the moral
 Autonomy and mor al obligation 61

required to ‘complete’ our picture of morality. Moreover, Kant allows,

it is also legitimate to think that God’s special concern to act as judge
and executor in this way stems from the fact that these are his laws that
he is upholding. Thus, the idea of God as lawgiver is also appropriate
here, but again only in a way that conforms to the priority of morality
over religion, which then qualifies this conception in certain crucial
ways. Kant makes this point clear in the following passage from his
lectures on ethics:
It is assumed, indeed, that the binding force of moral laws lies in the
divine will, and God has been viewed as the moral legislator. But this
cannot be construed to mean that the divine laws, and the binding

law as a kingdom of God, in which nature and morals come into a harmony, foreign to
each of them of itself, through a holy author who makes the derived highest good pos-
sible … Nevertheless, the Christian principle of morals itself is not theological (and
so heteronomy); it is instead autonomy of pure practical reason by itself, since it does
not make cognition of God and his will the basis of these laws but only of the attain-
ment of the highest good subject to the condition of observing these laws, and since
it places even the proper incentive to observing them not in the results wished for but
in the representation of duty alone, faithful observance of which alone constitutes
worthiness to acquire the latter’; LE 27:277–8 (p. 68): ‘The cause of this derivation of
morality from the divine will is as follows: Because moral laws run, Thou shalt not, it
is supposed that there must be a third being, who has forbidden it. It is true that any
moral law is an order, and they may be commands of the divine will, but they do not
flow from such a command. God has commanded it because it is a moral law, and His
will coincides with the moral law. It also seems that all obligation has a relation to
one who obliges. In performance, to be sure, there must indeed be a third being, who
constrains us to do what is morally good. But for the making of moral judgements we
have no need of any third being. All moral laws can be correct without such a being.
But in execution they would be empty if no third being could constrain us to them.
It has therefore been rightly perceived that without a supreme judge all moral laws
would be without effect, since in that case there would be no inner motive, no reward
and no punishment. Hence the knowledge of God is needed in the execution of moral
laws’; LR 28:1116–7 (p. 442): ‘God’s government of the world in accordance with moral
principles is an assumption without which all morality would have to break down. For
if morality cannot provide me with the prospect of satisfying my needs, then it cannot
command anything of me either. Hence it is also necessary that God’s will should not
be made the principle of rational morality; for in this way we could never be sure what
God had in mind for the world. How can I know by reason and speculation what God’s
will is, and what it consists in? Without morality to help me here, I would be on a slip-
pery path, surrounded by mountains which afford me no prospect. How much danger
I would be in of having my foot slip, or, because no clear horizon ever meets my eyes,
of wandering lost in a labyrinth! The cognition of God must therefore complete mor-
ality, but it must not first determine whether something is morally good or a duty for
me! This I must judge from the nature of things in accordance with a possible system
of ends; and I must be just as certain of it as I am that a triangle has three angles. But
in order to provide my heart with conviction, weight and emphasis, I have need of a
God who will make me participate in happiness in accordance with these eternal and
unchangeable laws, if I am worthy of it’.
62 Understanding Mor al Obligation

force of the divine will, can be discerned and known no otherwise, than
through the positive divine will; this is impossible. On the contrary,
we must first have discerned through reason that the divine will is in
accordance with the concept of a moral law, i.e., that it coincides with
the universal laws of nature, before we acknowledge the universal will
as a binding law and subject ourselves to it. The binding force of the law
lies, therefore, in the principle as it is known to reason; on the other
hand, we can and must attach to this hypothesis the sense that God,
as a moral and omnipotent being, is the supreme executor of all inner
and outer moral laws, that He adds to their force the efficacy that is
needed to manifest it, and that we, therefore, when we observe or trans-
gress the laws, are subject to God’s judgment-seat, in that we have acted
according to His will, or against it, and must expect the consequences.
By reason alone, therefore, the moral law can be demonstrated and
known, and doubtless followed as well; yet there is no denying that if,
apart from the binding efficacy of the law itself, our free choice is fur-
ther supported by the idea that our action conforms to the will of a
higher external cause, this hypothesis is a good, one might even say a
necessary, accompaniment to human nature.60

We might say, therefore, that Kant saw the judicial and executive func-
tions of God as required by morality, where our conception of those
functions can also bring along with it the idea that it is God’s own law
that he is judging us by and enforcing, given the special nature of his
judicial and executive roles; but at the same time, Kant does not want
this legitimate belief in God as lawgiver to invert the relative priority of
morality over religion, which is what happens on the divine command
model where God is seen as the prior source of the content and bind-
ingness of morality. This bindingness must be seen to come from our
reason, which at most God’s role as lawgiver can be said to reinforce,
where reason’s control over our will proves weak.
It can be seen, then, how it is that in the passages Hare cites, Kant
can talk of God as a commander, or lawgiver, or world ruler and so
on, without that implying that Kant is committing himself to a divine
command account of obligation, which treats the former as the ground

60
LE 27:530 (pp. 290–1). Cf. also LR 28:1011–2 (p. 356): ‘[The human being] tries to act
according to the duties he finds grounded in his own nature; but he also has senses
which present the opposite to him with a blinding bedazzlement, and if he had no
further incentives and power to resist it, then he would in the end be blinded by their
dazzle. Hence in order that he may not act against his own powers, he is set by his own
reason to think of a being whose will is those very commands which he recognizes to
be given by themselves a priori with apodictic certainty’.
 Autonomy and mor al obligation 63

for the latter, where Kant is clear that the obligatory force of morality
must first be conceived as arising from reason, as I shall discuss further
in the next chapter. And of course this can also explain why in other
passages, Kant can forcefully reject such accounts, as in the one from
the Groundwork we discussed. Thus, for Kant, when we move from mor-
ality to religion, as the former requires us to do, we will arrive at a reli-
gious outlook that presents God as a lawgiver; but this is consistent with
rejecting the divine command model, which Kant takes to move from
religion to morality, and so conceives of God as lawgiver, commander
and so on in a much stronger and more problematic way.
Having shown, then, that Hare’s claims regarding Kant’s position as
a divine command theorist are questionable, we can now turn to the
second part of his case, which is that it is a mistake to think of Kant’s
critique of the ‘theological concept’ of morality in the Groundwork as
an attack on such theories in general, but that it is rather an attack on
Crusius’s position in particular. Hare tries to substantiate this claim
by pointing out what he sees as specific parallels between what Crusius
holds and the critical points in Kant’s argument, of which Hare picks
out three in particular: first, that Kant starts the critique by saying that
we cannot intuit God’s perfections, where ‘this starting point makes
sense if it is Crusius he has in mind’;61 second, Kant criticises the pos-
ition for being crudely circular, in a way that fits Crusius’s position;
and third, that Kant’s final objection is that ‘if we think we understand
what God is telling us to do without using moral concepts, we will be
left without morality at all’,62 where (Hare claims) the best explanation
for Kant saying this is that he has Crusius as his target.
Now, in response to the first two points, I would argue that it is
important to recall the context of Kant’s discussion, where (as we have
noted previously) part of his aim is to compare the ‘theological concept’
of morality unfavourably to perfectionism. To do this, Kant therefore
tries to claim that in two respects where perfectionism is problem-
atic, the theological position is on a par: namely, both involve con-
cepts and not intuitions, and both involve a kind of circularity, in that
while perfectionism ‘cannot avoid presupposing the morality which
it is supposed to explain’,63 so in the end the same will be true of the
theological position. This suggests, I would claim, that Kant mentions

61
 Hare 2001: 105, 2009: 153.
62
 Hare 2001: 107, 2009: 154.
63
GMM 4:443 (p. 91).
64 Understanding Mor al Obligation

our lack of intuitive knowledge of God’s will, and the requirement to

think of that will in moral terms, not in order to pick out Crusius as a
particular target, but in order to make the parallel clear that he wants
to draw between the theological position and perfectionism.
On the third point, things are more complicated. For, on the one
hand, I agree with Hare that the conclusion of the paragraph is not
best read as directly criticising the theological position for making
morality a matter of punishment or reward;64 rather, he is right to say
that it is criticising that position for removing moral concepts from our
understanding of God, and so for potentially sanctifying a kind of anti-
morality. This is why, as we have seen, Kant takes the theological pos-
ition to be more problematic than the perfectionist one, as the former
but not the latter can end up ‘infringing’ morality in a dangerous way.
So I agree with Hare when he says that Kant’s immediate point here
is not that ‘his opponent bases morality on hope of reward or fear of
punishment’,65 as some commentators perhaps too easily assume.
On the other hand, Hare’s reading does not take into account the
broader context in which this discussion takes place, where (again as
we have seen) Kant is identifying a whole series of positions for their
heteronomous starting points, and where he does not then specifically
criticise them for this when he deals with them individually, but takes
this as read.66 Thus, when he spells out this issue more explicitly else-
where, as we have seen, he does make plain that it is the issue of fear
and reward that renders divine command theories heteronomous, and
that this therefore is a significant criticism of them – it is just somewhat
masked by the context of the discussion here. What truth there is in
Hare’s reading, therefore, does not imply what he thinks it does, and
is not sufficient to make his identification of Crusius as Kant’s target
into a compelling claim.67
64
But cf. CPrR 5:41 (p. 173): ‘the will of God if agreement with it is taken as the object
of the will without an antecedent practical principle independent of this idea, can
become motives of the will only by means of the happiness we expect from them’.
65
Hare 2001: 107, 2009: 154.
66
So, for example, when it comes to perfectionism, Kant criticises it for its specific fail-
ures as against the other positions, and does not spell out the way in which it is heter-
onomous – though he clearly thinks it is, as he states explicitly where the context is
somewhat different, when it is not signalled by the ‘frame’ of the discussion as a whole.
See, for example, LE 27:16 (p. 10), where he argues that Baumgarten and Wolff ‘con-
tinually based perfection on the relation between cause and effect, and thus treated it
as a means to ends grounded in desire and aversion’.
67
It is perhaps worth noting in this context that in the Critique of Practical Reason, where
Kant presents his table of different ‘Practical Material Determining Grounds’ of
 Autonomy and mor al obligation 65

Finally, we may turn to the third aspect of Hare’s case, which is not
just that Kant was a divine command theorist, and that Kant’s critical
comments are directed only at Crusius’s version of such a theory, but
also that Kant held that autonomy can be make perfectly consistent
with the kind of authority over us that the divine command theorist
gives to God, provided that the theory is understood in the right way.
Thus, according to Hare, ‘Kant thinks submission [to a divine will] is
compatible with autonomy’.68
In order to support this view, Hare turns to Kant’s treatment of polit-
ical submission, where Kant accepts a role for the legislative, executive,
and judicial role of the state in order to prevent coercion by individ-
uals and so permit freedom. From this, Hare concludes that ‘Kant can-
not mean to construct an argument from autonomy against all forms
of external authority. The opposite is true. He thinks that autonomy
requires submission to at least one kind of external authority, namely
the authority of the state’.69 Of course, Hare accepts that there are
constraints on how the state must relate to its citizens for this status
of autonomy to be preserved, just as there are in the case of God.
In particular, ‘there is nothing heteronomous about willing to obey
a superior’s prescription because the superior has prescribed it, in a
discretionary way, as long as the final end is shared between us, and we
have trust also about the route’.70
Now, it may seem curious for Hare to stress the political analogy
here; for, as he notes, it is clear that Kant’s own political sympathies
were republican,71 and so might be felt to run against the account of
submission to an ‘external will’ that Hare offers here, as on this repub-
lican view authority is ultimately held to derive from the people them-
selves. However, Hare’s claim is that whatever might be true of the
political case, in his account of the political structure appropriate to
ethics, Kant is a royalist, who puts God at the head of a kingdom of ends,
so that ‘God has combined in one person, in his kingdom, the legisla-
tive, executive, and judicial functions that Kant thinks should be sepa-
rated in a well-run earthly republic’.72

morality, and lists ‘the will of God’ as one of the options, that he explicitly states
that this is ‘according to Crusius and other theological moralists’ (CPrR 5:40 (p. 172), my
emphasis), which hardly suggests that he considered Crusius to be some sort of special
case.
68
Hare 2001: 109. 69 Hare 2001: 110. 70 Hare 2001: 115.
71
See Hare 2001: 111 note 46. 72
 Hare 2001: 111.
66 Understanding Mor al Obligation

However, as we have already seen, Kant’s talk of God’s functions in

these respects needs to be handled with some caution. In particular,
when it comes to the discussion in the Groundwork to which Hare is
referring here, although Kant does speak of God in terms of sover-
eignty, it can be argued that this is not because he has a supreme
legislative role; rather, it is because in the communal system of moral
laws in which we stand together with God, he has his place as a holy
will and is thus never subject to these laws, in the way that we are made
subject to them through our non-moral needs and desires. Much more
will be said about this contrast in the next chapter; but for now, the
claim is just that this means that for Kant, while we human beings are
both members and sovereigns in this kingdom, God is only a sovereign
and in this sense supreme, because the law is not anything to which
he is subjected qua duty.73 To claim this place of supremacy alongside
God, we would have to be as God is, for then we would never feel that
our inclinations are subordinated to morality as a mere member of
the kingdom of ends, in the way that in fact is inevitable for creatures
like us. But if this interpretation is correct, nothing in Kant’s talk
of God as sovereign suggests that ‘other members [of the kingdom]
are subject to his will, while he is not subject to the will of any other
member’;74 rather, it means merely that he is not subjected to the law
in the way we are, not that he imposes it upon us. So, nothing in what
he says here suggests that Kant would have seen the latter relation as
compatible with our autonomy.
We have seen, then, how a plausible interpretative case can be made
for the view that Kant saw a threat to our autonomy in divine com-
mand theories of moral obligation, stemming from the way in which
they account for the force of that obligation through an appeal to
punishments and rewards, and thus through fear and inclination, in
what for Kant is a heteronomous manner. But perhaps a more effective
response to Hare’s position even than this, is to make clear how Kant
offers an alternative to the divine command account of obligation; for,
if I can show that in fact he does so, this will also support my suggestion

73
Cf. GMM 4:434 (p. 84): ‘Now, if maxims are not already of their nature in agreement
with this objective principle of rational beings as givers of universal law, the necessita-
tion of an action in accordance with this principle is called practical necessitation,
that is, duty. Duty does not apply to the sovereign in the kingdom of ends, but it does
apply to every member of it and indeed to all in equal measure’.
74
 Hare 2009: 156.
 Autonomy and mor al obligation 67

that Kant was dissatisfied by the latter in important respects, along the
lines we have indicated. It is to this alternative account of obligation,
therefore, that I will turn in the next chapter. What will remain to be
discussed, of course, is the cogency of Kant’s critique, and how success-
ful it is as an argument from autonomy against the position of the
divine command theorist; this issue will be considered in more detail
later in the book, in the Conclusion, once the historical influence of
Kant’s alternative to such theories has been explored.
 3

K a n t’s s olu t ion to t h e probl e m

of mor a l obl ig at ion

In the previous chapters, I have suggested that the focus of Kant’s

concerns about autonomy was not value realism, but the account of
moral obligation offered by the divine command theorist. In this
chapter, I will consider what I take to be Kant’s alternative account,
designed to avoid the objections he raised against the ‘theological
concept’ of morality and so to offer a better solution to the problem
of moral obligation. The key to Kant’s solution, I will suggest, is the
contrast he draws between our will and a divine or holy will, where he
takes it that the notion of obligation does not apply to the latter; in
explaining why, he comes to his distinctive account of the obligatory
force of morality as it applies to us, which differs markedly from what
has gone before.
This is so, as we shall see, because while Kant’s account is opposed
to the divine command approach to the problem, it is equally at odds
with a more traditional natural law approach, while at the same time
also combining elements of both. Having already discussed in the pre-
vious chapter what Kant objected to in the former, we may begin here
by seeing what is problematic about the latter, and thus what intellec-
tual pressures led Kant to adopt an alternative strategy, and thus to
what I will call his ‘hybrid’ account.

The problem of moral obligation and the

natural law tradition
In terms of the traditional debate over moral obligation, as we have
seen, the major alternative to the divine command theory was the nat-
ural law theory, which holds that obligatoriness is inherent to certain
actions, principles or states of affairs, and so does not see any direct

68
 Kant’s hybrid account of obligation 69

role here for any divine sovereign or commander.1 A view of this sort is
proposed by Clarke when he writes:
[T]hese eternal and necessary differences of things make it fit and rea-
sonable for creatures so to act; they cause it to be their duty, or lay an
obligation on them, so to do; even separate from the consideration of
these rules being the positive will or command of God; and also antecedent
to any … particular private and personal advantage or disadvantage, reward
or punishment, either present or future.2

Similar views can be found in Vasquez, Cudworth, Price and (on some
interpretations) Aquinas,3 as well as many others at least to some
degree.

1
As others have noted, it is a curious lacuna in Anscombe’s influential treatment of
these issues in Anscombe 1958, that she does not see that this kind of ‘no commander’
view of obligation was present as a live option amongst thinkers who were not secular
in other respects, so that alternatives to the divine command position existed even
prior to the rise of modern secularism and the ‘death of God’. Cf. Crisp 2004: 86 note
33: ‘Oddly, Anscombe appears never to consider the view that claims that we have such
obligations might be self-standing, requiring no justification from elsewhere, though
she does consider, as alternatives to divine legislation, the norms of society, self-leg-
islation, the laws of nature, Hobbesian contractualism, and the virtues. Perhaps, like
the early Greeks, she also felt that a nomos had to be nomizetai (“dispensed”)’. Cf. also
Pidgen 1988, esp. pp. 34–5, and Irwin 2006: esp. pp. 325–30. For a diagnosis of why
Anscombe is blind to this option, see Pink 2004b.
2
Clarke 1711: 35/2003: 295. As Korsgaard has noted, passages such as this can be con-
trasted with other passages, in which Clarke suggests that ‘the normative force derives
not from the intrinsic reasonableness of the action alone, but from the fact that the
agent determines herself to do what is reasonable’ (Korsgaard 1996c: 32), where she
cites the following: ‘For the judgement and conscience of a man’s own mind, concern-
ing the reasonableness and fitness of the thing, that his actions should be conformed to
such and such a rule or law, is the truest and formallest obligation … For no man will-
ingly and deliberately transgresses this rule in any great and considerable instance, but
he acts contrary to the judgement and reason of his own mind, and secretly reproaches
himself for so doing. And no man observes and obeys it steadily … but his own mind
commends and applauds him for his resolution, in executing what his conscience could
not forbear giving assent to, as just and right’ (Clarke 1711: 53–55/2003: 301).
3
For Vasquez, see above Chapter 2, note 7. For Cudworth, see Cudworth 1996: Book i,
Chapter ii, §4, p. 20: ‘For the will of another doth no more oblige in commands, than
our own will in promises and covenants. To conclude, therefore, things called natur-
ally good and due are such things as the intellectual nature obliges to immediately,
absolutely, and perpetually, and upon no condition of any voluntary action that may be
done or omitted intervening’. For Price, see Price 1948: Chapter vi, pp. 105–6: ‘virtue,
as such, has a real obligatory power antecedently to all positive laws, and independently
of all will; for obligation, we see, is involved in the very nature of it … Whatever it is
wrong to do, that it is our duty not to do, whether enjoined or not by any positive law’.
For a discussion of Aquinas on these issues, see Irwin 2007: 545–56.
70 Understanding Mor al Obligation

Now, of course, proponents of both approaches looked for argu-

ments against their opponents, so that defenders of the natural law
theory objected to the divine command account along the lines dis-
cussed in the previous chapter, while in their turn proponents of div-
ine command theory raised significant objections against the natural
law theory. A helpful place to get a sense of those objections is by con-
sidering a debate we have already mentioned, namely by looking at
Barbeyrac’s critique of Leibniz and the latter’s attack on Pufendorf’s
divine command theory.
In that attack, Leibniz contrasted Pufendorf’s view with his own,
which was that law and obligation is to be found ‘in the nature of
things and in the precepts of right reason that conform to it’, and so
not ‘in the command of a superior’:4 God’s role was thus to be the
source from which these things ‘emanate’ rather than to lay down the
moral law, which is as fixed and necessary as the laws of logic, arith-
metic and geometry. Leibniz therefore repeats a familiar trope from
the natural law tradition when he writes:
Neither the norm of conduct itself, nor the essence of the just, depends
on his free decision, but rather on eternal truths, objects of the divine
intellect, which constitute, so to speak, the essence of divinity itself …
[J]ustice follows certain rules of equality and proportion [which are]
no less founded in the immutable nature of things, and in the divine
ideas, than are the principles of arithmetic and geometry. So that no
one will maintain that justice and goodness originate in the divine will,
without at the same time maintaining that truth originates in it as well:
an unheard-of paradox by which Descartes showed how great can be
the errors of great men; as if the reason that a triangle has three sides,
or that two contrary propositions are incompatible, or that God himself
exists, is that God has willed it so.5

In his response to Pufendorf, therefore, Leibniz writes as a natural law

theorist, in holding that ‘ justice and goodness … belong to the neces-
sary and eternal truths about the nature of things, as do numbers and
proportions’.6

4
Leibniz 1706: 279/1988b: 70.
5
Leibniz 1706: 280/1988b: 71–2. Leibniz’s reference to Descartes is to the sixth set of
replies to objections to the Meditations: see Descartes 1984: ii, 291–2. Cf. also Cudworth
1996: Book i, chapter ii, §1, pp. 16–17; and Clarke 1711: 36–7/2003: 295.
6
Leibniz c 1702: 56/1988a: 45.
 Kant’s hybrid account of obligation 71

Now, in offering his able support to Pufendorf,7 Barbeyrac’s strategy

is in part to claim that Leibniz has exaggerated and misrepresented the
implications of Pufendorf’s position, and so rendered it more problem-
atic than it really is, particularly when it comes to the supposed arbi-
trariness of God’s willing; far from this being a difficulty, Barbeyrac
argues that according to Pufendorf, ‘He follows inviolably the rules of
justice that conform to His infinite perfection, such that He neither
wills nor could will to act otherwise’.8 But as well as responding defen-
sively, Barbeyrac also goes on the attack, and raises some fundamental
objections to the natural law approach adopted by Leibniz.
His central claim is that in obligation, our freedom of action is
curbed or limited.9 But, Barbeyrac argues, nothing in the nature of
things on their own could achieve this; rather, the command of a
superior is required in order to account for it, as only such a superior
can prevent an agent from acting otherwise, in a way that renders us
obligated:
There is thus no middle point: either obligation to the rules of justice
among men is absolutely independent of the divinity, and grounded
solely in the very nature of things, like the ‘principles of arithmetic
and geometry’; or it is no way grounded in the nature of things. Now,
of itself, the nature of things could not impose an obligation upon us,
properly speaking. That there is such and such a relation of equality
or proportion, of propriety or impropriety, in the nature of things, of
itself commits us only to recognizing that relation. Something more is
required in order to constrain our liberty of action, in order to com-
mand us to govern our conduct in a certain manner. Nor can reason,
considered in itself and independently of the Creator who granted it
to us, absolutely compel us to follow these ideas, although endorsed by
them, as founded in the nature of things.10

7
For further discussion of this dispute, see Schneewind 1998: 250–9; Korkman 2001,
2003; Hunter 2003; Irwin 2008: 321–30.
8
Barbeyrac 1735: 413/2003: 291.
9
Cf. Pufendorf 1672: Book i, Definition xii, §1, p. 81/1931: 72: ‘[Obligation] places, as
it were, a kind of moral bridle [vinculum juris] on our liberty of action, so that we are
unable rightly to turn in a direction different from that to which obligation leads’;
and Pufendorf 1682: Book i, Chapter 2, §3, p. 13/1991: 27: ‘Obligation is commonly
defined as a bond of right [vinculum juris] by which we are constrained by the neces-
sity of making some performance. That is, obligation places a kind of bridle on our
liberty’; and Hobbes 1998: Chapter ii, Section 10, p. 36: ‘[F]or obligation begins where
liberty ends’.
10
Barbeyrac 1735: 415/2003: 292.
72 Understanding Mor al Obligation

In support of this position, Barbeyrac makes two main points.

First, he contends that while it may be the case that reason can
lead us to act on grounds of ‘propriety or impropriety’, as the nat-
ural law theorist says, we can also be led to act by our passions and
inclinations in a contrary direction. But, Barbeyrac asks, ‘[w]hy then
would we follow the former rather than the latter, if there is no exter-
nal principle, no superior being that compels us?’.11 The response
cannot be that it is somehow more ‘natural’ to follow reason, as ‘is
not the inclination of the heart as natural as the ideas of the mind?’12
The reply might be instead that following reason is in our interests.
But then, Barbeyrac objects, it would only be considerations of pru-
dence that would lead us to act morally, and not duty, while we are
‘each of us free to renounce our advantage’,13 so that the constraint
imposed by obligation has not really been accounted for here in an
adequate manner.
Secondly, Barbeyrac claims, it is impossible to make sense of obliga-
tion as coming through the use of our reason alone, as we would then
be imposing an obligation on ourselves, and so be at liberty to free our-
selves from it, making it really no obligation at all:
But what must be addressed above all, and what is enough to destroy the
thought I am fighting, is the fact that our reason, considered aside from
any dependence upon the Creator from whom we received it, is finally
nothing other than ourselves. Now no one can impose on himself an
unavoidable necessity to act or not to act in such and such a manner. For
if necessity is truly to apply, there must be absolutely no possibility of
it being suspended at the wish of him who is subjected to it. Otherwise
it reduces to nothing. If, then, he upon whom necessity is imposed is
the same as he who imposes it, he will be able to avoid it each time he
chooses; in other words, there will be no true obligation, just as when
a debtor comes into the property and rights of his creditor, there is no
longer a debt.14

11
Barbeyrac 1735: 416/2003: 293. 12
 Barbeyrac 1735: 416/2003: 293.
13
Barbeyrac 1735: 416/2003: 293. Cf. Pufendorf 1688: Book ii, Chapter iii, §20,
p. 149/1934: 217: ‘But if these dictates of reason are to have the force of law, there is
need of a higher principle; for although their advantage is most manifest, still it alone
could never lay so firm a restraint upon the spirit of men that they could not forsake
such dictates if they should find satisfaction in disregarding this advantage, or believe
that they could better consult their own advantage in some other way’.
14
Barbeyrac 1735: 416–7/2003: 293. For a discussion of these ideas in a Hobbesian con-
text, cf. Cohen 1996: 167–70.
 Kant’s hybrid account of obligation 73

Thus, against the natural law theorist, who treats obligatoriness as

inherent in the order of things, Barbeyrac argues that it is only by
appeal to the will of God that we can really do justice to the demand-
ingness of morality. Barbeyrac’s basic point is that being obligated
takes away our freedom, and that reason does not have the capacity
to do that by informing us about how things stand in the world, for
we can always go against reason or change our minds if we don’t like
what it is telling us;15 only if obligation comes from an authority with
controlling power over us can we explain how our liberty genuinely
comes to be curbed, and thus make sense of the obligatoriness of the
moral.16
Closer to Kant’s own time, we can find Crusius raising similar con-
cerns to Barbeyrac and thus echoing the latter’s critique of Leibniz
in his own critique of the Leibnizian position popularised by Wolff.
Crusius’s basic objection is that because this position does not treat
the bindingness of morality as arising from the will of God, it in the
end must appeal to prudential considerations to explain why the agent
feels the force of the moral ‘must’; but that leaves the obligatoriness
of morality too contingent and makes its motivations too instrumen-
tal, and so blurs the fundamental distinction between the conditional
necessity of prudence and the absolute necessity of morality.17 As is
well known, Kant accepts Crusius’s emphasis on this distinction, writ-
ing as early as the ‘Prize Essay’ of 1764 that this kind of conditional
necessity (which he will later associated with hypothetical imperatives)

15
Cf. Selden 1892: Chapter lxxvii, p. 101: ‘I cannot fancy to myself what the law of
nature means, but the law of God. How should I know I ought not to steal, I ought
not to commit adultery, unless somebody had told me, or why are these things against
nature? Surely ’tis because I have been told so? ’Tis not because I think I ought not to
do them, nor whether you think I ought not. If so, our minds might change. Whence
then comes the restraint? From a higher power; nothing else can bind. I cannot bind
myself, for I may untie myself again; nor an equal cannot bind me: we may untie one
another. It must be a superior, even God Almighty’.
16
Cf. Richard Cumberland’s summary of one of the fundamental objections to natural
law theory made by critics like John Selden and Jeremy Taylor: ‘The first Objection
against this Proof of the Law of Nature, is, That it supposes, without Proof, the
Legislative Power of Reason, which is not to be suppos’d. “Reason is not the Law, or
its Measure; neither can any Man be sure, that any thing is a Law of Nature, because
it seems to him hugely reasonable, neither, if it be so indeed, is it therefore a Law. For
Reason can demonstrate, and it can persuade, and invite, but not compel any thing but
Assent, not Obedience, and therefore it is no Law.” ’ (Cumberland 2005: Chapter ii,
§12, p. 914).
17
Cf. Crusius 1767: §372, pp. 454–8/2003: 584.
74 Understanding Mor al Obligation

‘does not indicate any necessity at all’.18 At the same time, however, as
we have seen, Kant could scarcely be happy with the divine command
account Crusius himself offered instead, as for Kant that too either
ultimately rendered the obligatoriness of morality a purely prudential
matter, based on the fear of God,19 or if it did not, seemingly left it
unexplained how else his commands could come to constrain us.20
As this brief discussion shows, therefore, from Kant’s perspective,
both the traditional natural law position and the divine command the-
ory had their problems. We can see why in this context, while Kant may
have wanted to reject the divine command approach as a result of his
concerns about autonomy, at the same time there were also grounds
for being dissatisfied with the orthodox natural law approach. It is not
surprising, then, to find Kant taking a new direction on these issues,
and so arriving at a position that does not fit easily into either camp.
The complexity of Kant’s attitude to both sides in this debate can
be gauged in an important passage from his Metaphysics of Morals. In
this passage, Kant states his agreement in part with the natural law
approach of someone like Leibniz, that the moral law is independent
of a lawgiver, so that the wrongness of lying is indeed not contin-
gent and chosen but rather fixed and necessary, and thus requires
no ‘author’.21 On the other hand, Kant still holds that some account

18
PE 2:298 (p. 272).
19
Cf. Hutcheson’s parallel complaint against Barbeyrac: ‘Mr. Barbeyrac, in his
Annotations upon Grotius, makes Obligation denote an indispensable Necessity to act in
a certain manner. Whoever observes his Explication of this Necessity, (which is not nat-
ural, otherwise no Man could act against his Obligation) will find that it denotes only
“such a Constitution of a powerful Superior, as will make it impossible for any Being to
obtain Happiness, or avoid Misery, but by such a Course of Action”’ (Hutcheson 2002:
Treatise ii, Section i, p. 146).
20
Crusius himself treats the desire to obey God as ‘fundamental’ or primitive
(a Grundtriebe); but then, it is not clear why the natural law theorist could not appeal
to an equally primitive desire to do what is right. For Crusius’s view, see Crusius 1766:
§452, pp. 929–30.
21
MM 6:227 (p. 381). Cf. LE 27:282–3 (p. 76) and LE 27:544 (p. 302): ‘But if we ascribe
an auctor to laws that are known through reason from the nature of the case, he can
only be an author of the obligation that is contained in the law’. Cf. also LE 29:616 (p.
236): ‘Can we, however, regard moralitas objectiva as a morality that arises from the div-
ine will? Yes [says Baumgarten], for since the divine will is the idea of the most perfect
will, we can say that such a will commands it. But from this it does not follow that it
would have to be derived from the divine will; we could not, in that case, perceive it to
be necessary … If I can discern, from the nature of the matter, that an action is moral,
then I do not need the divine will’; LE 29:633–34 (not translated): ‘Legem ferre dicitur,
qui dat obligationem; quam lex pronunciat et qui potestatem habet legem ferre est legislator. [He
is said to make law, who gives an obligation; the law pronounces this obligation and
 Kant’s hybrid account of obligation 75

must be given of the obligatoriness of such moral principles, and

how they appear binding to us, which the divine command ­t heory
tries to give by treating God as ‘the author (auctor) of the obliga-
tion in accordance with the law’. We have seen, however, what makes
such accounts problematic in Kant’s eyes, so that while ‘[a] law that
binds us a priori and unconditionally by our own reason can also be
expressed as proceeding from the will of a supreme lawgiver’,22 the
former way of thinking must be given priority over the latter; Kant
therefore tries to make sense of this idea of bindingness, in a way
that neither the natural law nor divine command approaches had
managed to achieve.

Moral obligation, the holy will, and the human will

To do so, I will argue, Kant makes important use of a distinction he
draws between a holy or divine will, and one such as ours.
Compared to many other aspects of his rich and complex practical
philosophy, Kant’s discussion of the holy or divine will, and the dis-
tinction that he draws between that will and one such as our own, has
been little discussed. In some ways this is not surprising – for in a con-
text where the predominant mood of ethical theorising is secular, talk
of ‘the holy will’ may in itself cause misgivings; and at the same time,
at the core of Kant’s discussion seems to lie one of those notorious
‘Kantian dualisms’ (here between reason and desire), where sympa-
thetic commentators have perhaps been understandably wary of mak-
ing too much of the notion as a result. It may seem best, then, to treat
this distinction as of no great significance both to Kant and to philoso-
phy in general, and to pass over it in somewhat embarrassed silence.
On the other hand, it is a distinction that Kant draws frequently, and
with great emphasis. Commentators have not made much of it, how-
ever, perhaps because they have not seen what work Kant uses the dis-
tinction to do, and so what insights can be gained from reflecting on

he who has the power to make the law is the legislator.] Whoever is the legislator is not
the author of the law; he is rather the author of the obligation of the law. The two can
be different. God is to be regarded as the moral legislator; but he is not the author of
the laws, for these lie in the nature of things and what is added from God is only a new
obligation … God is not the author of morality, since otherwise it would come to be
through his will and we would come to know them through nature as well. It lies in the
essence of things. In the same way, God is not the author of the relation of mathemat-
ical figures through his will’.
22
MM 6:227 (p. 381).
76 Understanding Mor al Obligation

it.23 My aim in this chapter, by contrast, is to show that this attitude is a

mistake, so that when it comes to Kant’s practical philosophy, it should
be given as much prominence as the related distinction between the
intuitive and discursive intellects.24 Thus, I will try to show that, far
from being an element in his thought that we would do best to ignore,
Kant’s conception of the holy will is therefore one that deserves to be
taken seriously, as well as being integral to a proper understanding of
his ethical views.
Kant draws a distinction between the holy will and a will such as
ours throughout his ethical writings and in his lectures on ethics.
The actual difference he points to is in essence a simple one, and
obviously relates to standard theological conceptions of our ‘fallen’
nature: whereas a divine will acts only in line with the good, and has
no inclinations to do otherwise, we have immoral desires and inclina-
tions, that mean we find ourselves drawn to adopt immoral courses
of action. As Kant puts it: ‘[t]he dispositions of the deity are morally
good, and those of man are not. The dispositions or subjective mor-
ality of the divine are therefore coincident with objective morality’,25
but ours are not.
While the contrast Kant draws is perhaps not unusual, however,
the way Kant uses it is considerably more distinctive. For, he deploys
it precisely in order to resolve (and in part dissolve) the problem of
moral obligation; namely, in order to explain the particular force that
morality has for us, which takes the form of a command or imperative,
as telling us that there are things that we must or must not do – what

23
An exception is Terence Irwin, who has also related Kant’s treatment of this distinc-
tion to the questions of obligation that concern us here. See Irwin 2004 and 2009,
especially pp. 157–63. See also Alston 1989a: 310–15/1989b: 259–65.
24
Lewis White Beck makes this comparison in Beck 1960: 50.
25
LE 27:263 (p. 56). Cf. also LE 27:1425 (p. 68): ‘[T]he divine will is in accordance with
the moral law, and that is why His will is holiest and most perfect … God wills every-
thing that is morally good and appropriate, and that is why His will is holy and most
perfect’; and LE 29:604 (p. 229): ‘In the Gospel we also find an ideal, namely that of
holiness. It is that state of mind from which an evil desire never arises. God alone is
holy, and man can never become so, but the ideal is good. The understanding often
has to contend with the inclinations. We cannot prevent them, but we can prevent
them from determining the will’; and LR 28:1075 (p. 409): ‘A holy being must not be
affected by the least inclination contrary to morality. It must be impossible for it to will
something that is contrary to the moral law. So understood, no being but God is holy.
For every creature always has some needs, and if it wills to satisfy them, it also has incli-
nations which do not always agree with morality. Thus the human being can never be
holy, but of course [he can be] virtuous. For virtue consists precisely in self-overcoming’.
 Kant’s hybrid account of obligation 77

Kant calls ‘constraint’ or ‘necessitation’ (Nötigung).26 In many pas-

sages, Kant explains this obligatoriness in terms of the distinction
between the holy will and our own, arguing that it is because we have
dispositions to do things other than what is right that the right for us
involves a moral ‘must’; but for a holy will, which has no inclination to
do anything other than what is right, no such ‘must’ applies. A typical
statement of Kant’s view is the following from the Groundwork of the
Metaphysics of Morals:
A perfectly good will would, therefore, equally stand under objective
laws (of the good), but it could not on this account be represented as
necessitated to actions in conformity with law since of itself, by its subject-
ive constitution, it can be determined only through the representation
of the good. Hence no imperatives hold for the divine will and in gen-
eral for a holy will: the ‘ought’ is out of place here, because volition is of
itself necessarily in accord with the law. Therefore imperatives are only
formulae expressing the relation of objective laws of volition in general
to the subjective imperfection of the will of this or that rational being,
for example, of the human will.27

Thus, the principles that determine what it is good and bad to do

apply to the holy will, where these principles are moral laws because
they hold of all agents universally, and of such agents independently
of the contingencies of their desires and goals, and thus necessarily.28
However, because the holy will is morally perfect, these laws lack any
commanding or imperatival force for it, whereas our lack of moral
perfection means that they possess such force for us.29
26
Schwaiger 2009: 69–70 helpfully explains the background of this terminology in
Baumgarten.
27
GMM 4:414 (p. 67). As we shall soon see, the ‘good’ here can be broader than the
moral good as it may include things that are good qua means; on the other hand, as
Timmermann observes (2007: 62–3 note 27), ‘[i]t is most likely that Kant first and
foremost has the moral law in mind when he says that the same objective laws of the
good hold for both kinds of will [viz. human and holy] alike’.
28
On universality, see: GMM 4:412 (p. 65): ‘moral laws are to hold for every rational
being as such’. On necessity, see Kant’s distinction between principles and laws, where
the former are what govern ‘what it is necessary to do merely for achieving a discre-
tionary purpose’, and so can be ‘regard[ed] as in [themselves] contingent and we can
always be released from the precept if we give up the purpose’, whereas a moral law
‘leaves the will no discretion with respect to the opposite, so that it alone brings with
it that necessity which we require of a law’ (GMM, 4:420 (p. 72)). Cf. also GMM 4:389
(pp. 44–5).
29
Cf. MM 6:222 (p. 377): ‘An imperative is a practical rule by which an action in itself
contingent is made necessary. An imperative differs from a practical law in that a law
indeed represents an action as necessary but takes no account of whether this action
78 Understanding Mor al Obligation

Kant’s position here is part of a general account of imperatives, not

just the imperatives of morality. The latter are categorical, but Kant
distinguishes them from hypothetical imperatives which can be either
‘problematic’ or ‘assertoric’: that is, imperatives that say what we ought
to do in so far as this is a means to some end, where that purpose is
discretionary and not shared by all (such as winning a race, or getting
a high paying job), or an end that everyone naturally wills (such as
happiness).30 Categorical imperatives, by contrast, tell us what we must
do independently of prior ends, as in the commandments of morality.
But despite this difference between the categorical and hypothetical
cases, Kant holds that both types get their imperatival force for us
because unlike a holy will, we are not simply geared up to do what we
see to be good (whether conditionally or unconditionally), but are also
tempted to act otherwise rather than simply following the latter. So,
when it comes to hypothetical imperatives, while I may know that exer-
cising has conditional value as a means to winning the race, I none-
theless want to stay in bed; or while I may know that to make myself
happy I should live more simply, I am still reluctant to do so because of
the pleasures that my materialist lifestyle brings. Thus, Kant says, ‘[a]ll
imperatives are expressed by an ought and indicate by this the relation
of an objective law of reason to a will that by its subjective constitution
is not necessarily determined by it (a necessitation). They say that to
do or to omit something would be good, but they say it to a will that
does not always do something just because it is represented to it that it
would be good to do that thing’.31
Now, if imperatival force is best understood in this way, Kant none-
theless thinks that there is a special problem with respect to the categor-
ical imperatives of morality, as against the problematic and assertoric
hypothetical imperatives. For while in both cases, imperatives apply
to us because we possesses contrary inclinations that need to be con-
strained by reason in its representation to us of what is good, Kant
holds that how this process works is clearer in the case of hypothetical

already inheres by inner necessity in the acting subject (as in a holy being) or whether
it is contingent (as in a human being); for where the former is the case there is no
imperative. Hence an imperative is a rule the representation of which makes necessary
an action that is subjectively contingent and thus represents the subject as one that
must be constrained (necessitated) to conform with the rule’.
30
See GMM 4:414–5 (pp. 67–69). For a slightly different classification, see CPrR 5:20–1
(pp. 154–5).
31
GMM 4:413 (p. 66).
 Kant’s hybrid account of obligation 79

imperatives than categorical ones. However, while Kant makes this

much plain, he is unfortunately less transparent and consistent on
where exactly he takes the difficulty to lie, so that a range of interpret-
ative options have been canvassed: for example, that the means/end
reasoning behind hypothetical imperatives is self-evidently or analytic-
ally rational, whereas the reasoning behind categorical imperatives is
not; that the end in following a hypothetical imperative is self-evident
or analytically contained in the imperative, whereas the end in follow-
ing a categorical imperative is not; or that the motivation in following
a hypothetical imperative is self-evident or analytic, whereas in a cat-
egorical imperative it is not – and where, of course, all these options
can be interrelated in various ways.32 Thus, while some such difficulty
prompts Kant to ask a transcendental or ‘how possible?’ question of
categorical but not hypothetical imperatives, because the problem
behind the question is not made entirely lucid, so too it is not entirely
clear what is at stake in answering it.
Fortunately, for our purposes it is not necessary to enter this mine-
field, as we are less concerned with what it is in the case of a categorical
imperative that makes moral action into one that reason follows, and
more with why that action appears to us as commanded or necessitated,
as it is this aspect of the ‘how possible?’ question that needs settling if
the obligatory force of morality is to be explained without comprom-
ising our autonomy. Thus, what matters here for us is that when Kant
raises the ‘how possible?’ question in Section ii of the Groundwork and
tells us that he will postpone his resolution of it to Section iii,33 and
that when that resolution turns out to involve the machinery of tran-
scendental idealism,34 he still comes back to characterising that obliga-
toriness in terms of the distinction between the holy will and ours,
but this time mapped onto the transcendental distinction between the
noumenal and phenomenal realms (or the ‘intelligible world’ and ‘the
world of sense’) that he has now introduced:
And so categorical imperatives are possible by this: that the idea of free-
dom makes me a member of an intelligible world and consequently,
if I were only this, all my actions would always be in conformity with
the autonomy of the will; but since at the same time I intuit myself as

32
For example, the first option is emphasised in Paton 1967: 120–8, esp. p. 128; the
second in Guyer 2007: 80–1; and the third in Timmermann 2007: 69. I do not claim
that these options are exhaustive.
33
See GMM 4:420 (p. 73). 34 See GMM 4:450–1 (p. 98).
80 Understanding Mor al Obligation

a member of the world of sense, they ought to be in conformity with it;

and this categorical ought represents a synthetic proposition a priori,
since to my will affected by sensible desires there is added the idea of
the same will but belonging to the world of the understanding – a will
pure and practical of itself, which contains the supreme condition, in
accordance with reason, of the former will … The moral ‘ought’ is then
[the person’s] own necessary ‘will’ as a member of an intelligible world,
and is thought by him as ‘ought’ only insofar as he regards himself at
the same time as a member of the world of sense. 35

Kant thus uses his transcendental idealism, and his dualistic picture
of the will, to elaborate on the distinction between the holy will and
ours, a dualism that is designed to help solve the problem of ‘how
the necessitation of the will, which the imperative expresses … can be
thought’.36 Kant’s distinction between the holy will and ours therefore
forms a crucial part of his answer to the problem of accounting for the
moral ‘must’, in a way that explains its possibility (unlike a view that
simply treats the ‘must’ as a feature of the world), but without recourse
to the problematic notion of a divine legislator as the source of that
‘must’ (thus avoiding any need to adopt a divine command theory).

Analysing the distinction between the holy

will and the human will
Having shown the way Kant sets out the distinction, and indicated the
role he gives it within his practical philosophy, I now want to examine
it in a little more detail. For, although in some ways Kant’s position is
fairly straightforward, it is nonetheless not without its ambiguities.

35
GMM 4:454–5 (pp. 100–101). Cf. also LE 27:510 (pp. 274–5): ‘Although the obligation
is established by reason, it is nevertheless assumed that in the performance of our duty
we have to regard ourselves as passive beings, and that another person must be pre-
sent, who necessitates us to duty. Crusius found this necessitating person in God, and
Baumgarten likewise in the divine will, albeit known through reason, and not posi-
tively, and on this principle a particular moral system has been erected. If, however,
we pay heed to self-regarding duties, then man is presented in his physical nature, i.e.,
insofar as he is subject to the laws of nature, as the obligated, and rightly so; but if the
obligator is personified as an ideal being or moral person, it can be none other than
the legislation of reason; this, then, is man considered solely as an intelligible being,
who here obligates man as a sensory being, and we thus have a relationship of man qua
phenomenon towards himself qua noumenon. The situation is similar in obligations
towards others’. For a closely related passage, see MM 6:417–8 (pp. 543–4).
36
GMM 4:417 (pp. 69–70). We thus have Kant’s answer to Ogden Nash: If duty did have the
visage of a sweetie or a cutie for us, it wouldn’t be duty – just as it isn’t for the holy will.
 Kant’s hybrid account of obligation 81

One issue concerns what it is about the holy will, exactly, that makes
it the case that there is no moral obligation for a will of this kind.
The simplest answer to this question, which I have largely adopted
in sketching Kant’s position above, is to think that morality lacks any
obligatory force for the holy will because it has no resistance to moral-
ity: unencumbered as it is by any non-moral desires and inclinations, it
feels no constraint in acting morality, because nothing in its will fights
against the moral course of action as determined by reason, and so
no part of its will has to be restrained or held back in any way. This
picture fits with many passages from Kant, such as when he says that
if we had a holy will ‘the [moral] law would finally cease to be a com-
mand to us, since we would never be tempted to be unfaithful to it’;37
or characterises a ‘holy (superhuman) being’ as one ‘in whom no hin-
dering impulses would impede the law of its will and who would thus
gladly do everything in conformity with the law’;38 or states that ‘[God
is] unlimited only in this, that no moral necessitation can be supposed
in Him, in regard to the determination of His will, since He lacks the
limitations imposed on human nature, of an inclination to contravene
the laws’.39
However, as well as this way of characterising what is distinctive about
the holy will as being a will that lacks any non-moral inclinations, Kant
also characterised the holy will in terms that deploy more of his tech-
nical machinery, particularly the idea of a maxim. Unfortunately, how-
ever, there is some ambiguity in Kant’s account of what maxims are,
where this is then reflected in apparently contradictory claims Kant
makes about how the holy will stands in relation to maxims – Kant
sometimes states that the holy will (unlike us) has no maxims,40 and
sometimes that it has maxims, but unlike us, its maxims always coin-
cide with the moral law.41
The difficulty here arises because of an underlying unclarity in
what Kant means in characterising maxims as ‘subjective principles

37
CPrR 5:82 (p. 206). 38
MM 6:405 (p. 533). 39
LE 27:547 (p. 304).
40
Cf. CPrR 5:79 (p. 204): ‘All three concepts, however – that of an incentive, of an interest
and of a maxim – can be applied only to finite beings. For they all presuppose a limi-
tation on the nature of a being, in that the subjective constitution of its choice does
not of itself accord with the objective law of a practical reason; they presuppose a need
to be impelled to activity by something because an internal obstacle is opposed to it.
Thus they cannot be applied to the divine will’.
41
Cf. GMM 4:439 (p. 88): ‘A will whose maxims necessarily conform with the laws of
autonomy is a holy, absolutely good will’; and CPrR 5:32 (p. 165): ‘a holy will … would
not be capable of any maxim conflicting with the moral law’.
82 Understanding Mor al Obligation

of acting’.42 At times he seems to mean by this nothing more than

the idea that a maxim is a principle on which an agent acts or pro-
poses to act, in which case there is no difficulty for him in attributing
maxims to the holy will, and claiming that the principles on which it
acts will always be ones that conform to the moral law. However, Kant
also characterises the ‘subjective’ nature of maxims in a further way,
not just as principles employed by subjects in acting, but as principles
that have merely subjective validity, in contrast to the objective valid-
ity of the practical law, where the subjectivity of a maxim in this sense
means that it holds only for the subject in so far as it relates merely to
his or her ‘conditions’, such as what he or she is inclined to do, and
thus does not apply to those whose ‘conditions’ are different. So, out
of concern for my health, I might make it my maxim to drink less
coffee because I think drinking coffee leads to insomnia; but this is
merely a subjective rule, as it does not profess to be valid for anyone
else, relating as it does to my particular inclinations, and my views
about the effects of coffee on my sleep and insomnia on my health.
By contrast, on this picture, a moral principle has an objectivity that
maxims lack, because they do indeed apply to others, where here the
relevant grounds or conditions are not confined to the individual sub-
ject.43 Thus, while in both cases, because they are general principles,
maxims and practical laws are determined by reason, in the former
case it is reason working on the basis of the particular agent’s particu-
lar preferences and so framing principles with limited applicability,
while in the latter it is reason arriving at genuine laws that apply to
all, independent of these circumstances. Now, given this conception
of maxims, as having only subjective validity in this sense, it is under-
standable why Kant might come to say that the holy will lacks them

42
Cf. GMM 4:420 note (p. 73), and also 4:401 note (p. 56).
43
Cf. CPrR 5:19 (p. 153): ‘Practical principles are propositions that contain a general
determination of the will, having under it several practical rules. They are subjective,
or maxims, when the condition is regarded by the subject as holding only for his will;
but they are objective, or practical laws, when the condition is cognized as objective,
that is, as holding for the will of every rational being’. Cf. also LE 27:495 (p. 263): ‘The
maxim of an action differs, that is, from an objective principle in this, that the latter
occurs only insofar as we consider the possibility of the action on certain rational
grounds, whereas the former includes all subjective grounds of action whatsoever,
insofar as they are taken to be real. N.B. The principle is always objective, and is called
a maxim quoad subjectum [as to the subject]. It is understood as the rule universally
acknowledged by reason, while the maxim is the subjectively practical principle, inso-
far as the subject makes the rule by which he is to act into the motive of his action as
well’.
 Kant’s hybrid account of obligation 83

altogether: for, if none of the ‘conditions’ that give merely subjective

validity to maxims apply to the holy will, and if instead the only prin-
ciples that guide it are the principles of the objectively valid practical
law, then it becomes clear why Kant might claim that we cannot think
of the holy will as having maxims. But it also becomes clear why, given
the other way in which the notion of a maxim is also used by Kant, he
might allow that the holy will can have maxims, and speak of the holy
will in these terms.44
Having resolved the apparent contradictions in Kant’s position with
regard to the holy will on the issue of maxims, we can now turn to another
area where there also might appear to be some ambiguity: namely, on
what basis does Kant claim that the holy will cannot be obligated to act,
or stand under a duty? In the section above, I suggested that Kant’s rea-
son for this claim is that he thought that for these notions to apply to an
agent, what it is right for them to do must exert some necessitating force,
which is impossible in the case of a holy will, as it lacks the inclination
to do anything other than the moral action; but there is another way of
taking Kant’s position here that should be considered.
On this alternative view, the reason why the holy will has no duties
or obligations is that duties and obligations require the agent to be
able to fail to act as they are obligated to do, so that in this sense ‘ought
implies might not’. Kant’s position has been presented along these
lines by Samuel Kerstein, who writes that
[a]ccording to Kant, one can be obliged to do something only if there is
some possibility that he will fail to do it. Yet some beings, for example,
God, might be such that they cannot fail to obey the supreme principle
of morality. It would thus make no sense to say that they have an obli-
gation to obey it.45

Kerstein cites evidence that Kant believed in the ‘ought implies might
not’ principle by pointing to a passage in the Metaphysics of Morals,
where Kant is discussing whether we could have a duty to pursue our
own happiness, which Kerstein reads as saying that ‘an agent cannot

44
Cf. Paton 1967: 61: ‘Kant speaks at times as if all maxims are grounded on sensuous
inclinations, and consequently as if a divine or holy will could have no maxims. A holy
will would have no maxims which were not also objective principles; but to say this is
not to deny that it acts in accordance with maxims, if we interpret “maxims” to mean
principles manifested in action. It is all-important to recognize that while maxims are
commonly based on inclinations … it may nevertheless be possible to act on maxims
which are not so based’.
45
Kerstein 2002: 2.
84 Understanding Mor al Obligation

have an obligation to promote the end of his own happiness, since

each agent unavoidably has this end’.46
Now, the principle ‘ought implies might not’ is perhaps plaus-
ible, and there is some additional textual evidence that Kant held it
as a necessary condition for obligatoriness.47 But in fact the passage
Kerstein cites seems to suggest that Kant also believed that more was
required to account for obligatoriness as it applies to us; for here Kant
says that we cannot have the duty to pursue our happiness because this
happiness accords with our desires, and thus there is no experience of
resistance in aiming at happiness as an end (in contrast, say, to our own
perfection, where we do experience resistance from desire, and where
therefore it makes sense to speak of a duty to self):

46
Kerstein 2002: 193 note 4.
47
See, for example, LE 27:486 (p. 256): ‘In God the nature of action is likewise that it
accords with the moral laws which are formed by the concepts of the highest reason;
save only that since no subjective possibility of contravening such laws is possible in
His case, His actions being morally necessary both objectively and subjectively, no
imperative is appropriate to Him either, since however He acts, He does so in accord-
ance with the moral laws, and will at all times act freely and unconditionally’; GMM
4:413 (pp. 66–7): ‘[Imperatives] say that to do or to omit something would be good,
but they say it to a will that does not always do something just because it is represented
to it that it would be good to do that thing’; GMM 4:414 (p. 67): ‘Hence no impera-
tives hold for the divine will and in general for the holy will: the “ought” is out of place
here, because volition is of itself necessarily in accord with the law’; GMM 4:449 (pp.
96–7): ‘this “ought” is strictly speaking a “will” that holds for every rational being
under the condition that reason in him is practical without hindrance; but for beings
like us – who are also affected by sensibility, by incentives of a different kind, and
in whose case that which reason by itself would do is not always done – that neces-
sity of action is called also an “ought,” and the subjective necessity is distinguished
from the objective’; MM 6:222 (p. 377): ‘An imperative differs from a practical law in
that a law indeed represents an action as necessary but takes no account of whether
this action already inheres by an inner necessity in the acting subject (as in a holy
being) or whether it is contingent (as in the human being); for where the former
is the case there is no imperative’; CJ 5:403–4 (p. 403): ‘Now since [in the practical
case] … the action which is morally absolutely necessary can be regarded physically
as entirely contingent (i.e. what necessarily should happen often does not), it is clear
that it depends only on the subjective constitution of our practical faculty that the
moral laws must be represented as commands (and the actions which are in accord
with them as duties), and that reason expresses this necessity not through a be [Seyn]
(happening) but through a should-be [Seyn-Sollen]: which would not be the case if
reason without sensibility (as the subjective condition of its application to objects of
nature) were considered, as far as its causality is concerned, as a cause in the intelli-
gible world, corresponding completely with the moral law, where there would be no
distinction between what should be done and what is done, between a practical law
concerning what is possible through us and the theoretical law concerning what is
actual through us’.
 Kant’s hybrid account of obligation 85

[H]is own happiness is an end that every human being has (by virtue
of the impulses of his nature), but this end can never without self-
­contradiction be regarded as a duty. What everybody already wants
unavoidably, of his own accord, does not come under the concept of
duty, which is constraint to an end adopted reluctantly. Hence it is self-
contradictory to say that he is under obligation to promote his own hap-
piness with all his powers.48

Here Kant clearly seems to make the conceptual claim that I am inter-
ested in, and not just the one attributed to him by Kerstein, namely
that ‘duty … is constraint to an end adopted reluctantly’, where it is
precisely this lack of ‘reluctance’ and hence constraint that seems to
make the concept of duty ‘self-contradictory’ in this context – and
thus, equally contradictory in the case of the holy will.49 In the light of
this textual evidence, I think it is reasonable to conclude that there is
more to Kant’s position here than just an appeal to the ‘ought implies

48
MM 6:386 (p. 517). Cf. MM 6:417 (p. 543): ‘For the concept of duty contains the con-
cept of being passively constrained (I am bound)’. Cf. also CPrR 5:37 (p. 170): ‘A com-
mand that everyone should seek to make himself happy would be foolish, for one never
commands of someone what he unavoidably wants already’. Of course, ‘foolishness’ is
a weaker notion than self-contradictoriness, but is a negative mark against such a com-
mand nonetheless.
49
Kant seems to have this view in several other passages, e.g. MM 6:379 (p. 512): ‘The
very concept of duty is already the concept of a necessitation (constraint) of free choice
through the law. This constraint may be an external constraint or a self-constraint. The
moral imperative makes this constraint known through the categorical nature of its
pronouncements (the unconditional ought). Such constraint, therefore, does not
apply to rational beings as such (there could also be holy ones) but rather to human
beings, rational natural beings, who are unholy enough that pleasure can induce them
to break the moral law, even though they recognize its authority; and even when they
do obey the law, they do so reluctantly (in the face of opposition from their inclina-
tions), and it is in this that such constraint properly consists’; and LE 29:616–7 (p. 236):
‘Necessitas actionis invitae [necessity of action against one’s will] is a compulsion. For
this it is required, not only that our will be not morally good, but also that it have
hindrances. A compulsion always presupposes a hindrance in the will. A man often
has inclinations which conflict with the moral law. So duty we regard as a compulsion.
A compulsion occurs when we have an inclination to the opposite of an action. The
necessitation to an action, such that we have an inclination to its opposite, is therefore
compulsion’. Cf. also CPrR 5:32 (pp. 165–6); 5:83 (pp. 206–7); LE 29:605 (pp. 229–30);
LE 29:611 (pp. 234–5); LE 27:519 (p. 282); 27:623 (p. 365); and Vorarbeiten zur Religion
23:100 (not translated): ‘If all men gladly and willingly followed the moral law in so far
as it contains reason as the rule, so it would give no duty, just as one cannot think of
this law which determines the divine will as obligating it [this will]. If therefore there
are duties laid down if the moral law in us is a command for us (categorical impera-
tive), then we must be considered as necessitated [genöthigt] without pleasure or inclin-
ation. A duty to do something gladly and from inclination is a contradiction’.
86 Understanding Mor al Obligation

might not’ principle, and that the account I offered in the previous
section should be allowed to stand.
A worry might be raised, however, that there is something prob-
lematic in Kant’s position here, and whether Kant can in fact make
the idea of the holy will consistent with his own philosophical out-
look, and in particular with his conception of the good will. For, in the
Groundwork and elsewhere, Kant famously characterises the good will
as a will that acts out of duty; but if the holy will cannot act in this way,
how can it be good?
I think this worry underlies H. A. Prichard’s claim, in his unpub-
lished lectures on the Groundwork, that ‘[Kant’s] idea of a holy will
is untenable’.50 Prichard argues that Kant must explain how the holy
will is moved to action, where (Prichard states) the mere goodness or
rightness of some state of affairs cannot explain this in itself. However,
Kant has ruled out using a ‘sense of obligation’ as the explanation for
action by the holy will, as this will is not supposed to have any such
sense. Thus, Prichard argues, Kant is forced to say that the holy will
acts out of a ‘good desire’. But then, it seems, Kant has contradicted
his own analysis of the good will, which is a will that acts out of a sense
of duty, not inclination, no matter how beneficent: ‘[f]or though it is
possible to perform duties from some good motive other than a sense
of obligation, e.g. a desire arising from affection or public spirit, and
though such an act will manifest goodness, the goodness manifested
will not be moral goodness’.51 The question Prichard is pressing, there-
fore, is the internal coherence of Kant’s position: ‘having formulated
the spirit in which a moral being will act, viz. the sense of obligation’,52
but having deprived the holy will of any such sense, how can the holy
will be a moral being?
While interesting and prima facie plausible, I think, however, that
Prichard’s concern can be allayed.
A first point to note regarding Kant’s conception of the good will
itself is that it is a mistake to think that Kant identified the idea of the
good will with the will that acts from duty, as if goodness must always
involve dutifulness. In fact, when introducing his conception of the
good will at the beginning of the Groundwork, Kant says that dutiful-
ness only characterises the good will when thought of ‘under certain
subjective limitations and hindrances’; he just wants to stress that these

Prichard 2002: 55.

50
 Prichard 2002: 55–6.
51
 Prichard 2002: 56.
52
 Kant’s hybrid account of obligation 87

‘subjective limitations and hindrances’ do not take away its goodness

altogether – indeed, ‘far from concealing it and making it unrecogniz-
able, [they] rather bring it out by contrast and make it shine forth all
the more brightly’.53 These ‘subjective limitations and hindrances’ are
obviously the non-moral inclinations that prevent us from being holy
wills, but where we can still display goodness through acting out of
duty: but it is clear from the outset of the Groundwork that Kant did not
identify being a good will with acting out of duty, but merely thought of
this as a way of being good, under special constraints and conditions.
In other words, the notion of the good will is broader than the concept
of the will to which the concept of duty applies.54 There thus seems to
be no incoherence, from a Kantian point of view, in holding that the
holy will acts from the thought that doing this act would be good, and
not from the thought that it is its duty.
Second, there are two possible responses one might give to Prichard’s
worry concerning the actions of a holy will, where Prichard thinks this
is a worrisome issue because such actions cannot be explained as aris-
ing from a sense of duty (as a holy will has no such sense) or from
desires (because a holy will has none). One response is to accept that
Prichard is right to be puzzled here, but to argue that Kant would have
not have seen anything problematic in such puzzlement; rather, it is
just what we would expect, given our limited understanding of any
such will (where, of course, this kind of response would fit with Kant’s
more general emphasis on our cognitive limitations regarding ques-
tions of this sort). Even if this kind of reply is in some ways unsatisfy-
ing, it is not clear why this still doesn’t allow Kant’s conception of the
holy will to do the work he requires of it, namely to provide a contrast
class to a will such as our own.
A more positive response, however, would be to challenge Prichard’s
claim that the holy will is a ‘being without desire’55 in any sense. For,
of course, while holy wills must lack any desire to do what is wrong,
we might still think of the holy will as having inclinations that are in
accord with what is right, in a way that would explain the contrast with
our own case, while making the agency of the holy will less inchoate
than on the first response. So, if we first consider what moral action
involves in our own case, Kant’s account seems to be that we have vari-
ous inclinations (such as the desire to commit suicide, or keep some

53
GMM 4:397 (p. 52). 54
Cf. Wood 1999: 26–7. 55
 Prichard 2002: 55.
88 Understanding Mor al Obligation

money that has been borrowed), but conscience then leads us to ques-
tion such inclinations and test their associated maxims, where the vari-
ous formulae that Kant proposes are the tests we use to determine the
rightness or wrongness of the actions we feel inclined to perform.56 We
can therefore think of the holy will as also having inclinations,57 which
are what lead it to act, but only inclinations that are moral;58 but when
it acts on them, it does so because they have been assessed in accord-
ance with the formulae and thus in moral terms, so it is not acting
merely out of feeling or on the basis of inclinations. However, unlike us,
because the holy will has no non-moral inclinations, these formulae do
not constitute imperatives, or what it ought to do, but rather what it will
do, in a way that makes the holy will exempt from duty and obligation,
as Kant claims.
I will not attempt to adjudicate between these responses to
Prichard here, but both seem available to Kant in ways that suggest
the Prichardian worry can be defused. In fact, although Kant does
not introduce this distinction very often or with much emphasis,59
he might be said to draw the contrast between us and the holy will
in both these ways, when he distinguishes between the human will on
the one hand, and two types of holy will – namely finite holy wills and
infinite ones, where the latter are divine and the former are not. The
finite holy wills seem to have inclinations arising from their sensuous
natures, but these inclinations (unlike ours) are always in harmony
with reason or at least offer no temptation to it, while the infinite holy
will then appears to contrast with the finite holy will in having no incli-
nations at all. One might then worry, in a Prichardian manner, how
such an infinite divine or holy will could operate – but again it may
seem plausible to respond that such understanding is beyond us, while
pointing to the finite holy will as a perhaps more intelligible but no
less significant contrast class to the human case, on whom the impera-
tival force of morality still falls.

56
GMM 4:421–4 (pp. 73–5).
57
For a defence of the view that we should see the holy will as having inclinations, see
Willaschek 2006: 130–2.
58
Cf. Relig 6:58 (p. 102), where Kant says that even natural inclinations do not necessarily
tend towards the bad, but can be ‘good, i.e. not reprehensible’, so that we do not need
to ‘extirpate’ them altogether in order to avoid evil, as the Stoics claimed.
59
MM 6:383 (p. 515).
 Kant’s hybrid account of obligation 89

Kant’s hybrid solution to the problem of moral obligation

I have thus considered Kant’s distinction between the holy will and the
human will in some detail, and shown how it allows Kant to offer an
account of how the obligatoriness of ethics is possible. I will now sug-
gest that what makes this account attractive is that it enables Kant to
offer a kind of ‘hybrid’ view that enables him to do justice to elements
from competing positions and thus steer a way between realism and
anti-realism; externalism and internalism; and divine command and
natural law models.

Realism and anti-realism. It is scarcely surprising that the dispute

between realists and anti-realists in ethics is long-standing and
ongoing, as both sides have their clear attractions, while also having
their drawbacks, where each is often mirrored in the views of the
other. And in the recent literature on Kant, as we have seen, these
respective advantages and disadvantages have been highlighted in
competing interpretations of Kant himself, where both realist and
anti-realist constructivist accounts have emerged. The dialectical
back-and-forth between realism and anti-realism is familiar. On
the one side, the realist will argue that the constructivist cannot
avoid endorsing realism at some point, for unless some moral facts
are taken to obtain prior to the choices we make as agents, where
these choices are then governed by these facts, then the constructive
procedure will be utterly unconstrained and its results morally
arbitrary and relative. On the other side, the constructivist will
argue that any such independent moral facts are too mysterious to
be treated as explanatorily basic but are not in any way explained by
the realist, while their independence from us makes them a threat to
our autonomy as agents. Each side thus raises concerns (of emptiness
and relativism on the one hand; and of ‘queerness’ and heteronomy
on the other) that appear genuine and serious, but which neither
can wholly address left to itself.
At this point, therefore, it is natural to look for a way out of the
impasse: and I would argue that Kant’s account of obligatoriness using
his position on the holy will provides us with such a way forward. For
this account can be viewed as having two levels, and thus as a kind of
‘hybrid’ position that combines elements of realism with elements of
anti-realism, to the advantage of both.
90 Understanding Mor al Obligation

The realist level concerns the content of morality, what is right and
wrong, and the value of rational nature on which this rests. As we have
seen, Kant goes along with the natural law theorists in insisting that
not even God can determine by an act of will or choice what this con-
tent is to be, as this is as fixed and necessary as the fact that a triangle
has three corners, and so obtains independently of any relation to any
legislator.60 As we have also seen, Kant speaks as if he conceives of the
value of the free rational agent in realist terms.61 However, Kant’s dis-
tinction between the holy will and our own means that he does not
treat the obligatoriness of what is right and wrong as independent in
this way, for we give the content of morality its obligatory form, in so
far as this depends on our limitedness as finite creatures. So, on Kant’s
account, this obligatoriness is just the way in which what is right and
wrong presents itself to us, from our human (all too human) perspec-
tive; from the perspective of a divine will, and so from the ‘absolute
standpoint’, there is no duty and obligation, but only what is right and
wrong, because the divine will has none of the non-moral inclinations
which (as we have seen) means that what is right is represented to us
in the guise of duties and obligations, and thus as the moral ‘must’.62
Kant is thus able to side with the realist about the right, and thus avoid
the spectre of emptiness and arbitrariness that threatens construct-
ivism; but he is able to side with the anti-realist about the obligatory,
and thus avoid much of the ‘queerness’ associated with the idea that
the world in itself makes demands on us, and also avoid the threat to
our autonomy that any such purely ‘external’ demand might seem to
imply. Kant thus offers us a hybrid view that is neither fully realist nor
fully anti-realist ‘all the way down’, but combines elements of both to
the advantage of his position as a whole.63

60
See Chapter 1 note 70.
61
See Chapter 1, third section.
62
Cf. CJ 5:403–4 (p. 273): ‘it is clear that it depends only on the subjective constitution
of our practical faculty that the moral laws must be represented as commands (and
the actions which are in accord with them as duties), and that reason expressed this
necessity not through a be (happening) but through a should-be’.
63
I think Irwin also sees Kant as adopting this sort of hybrid view: ‘[Kant] recognises
intrinsic rightness without any acts of commanding or obligating … In Kant’s view,
commands and act of binding are [only] relevant to finite rational agents, who are
also subject to other incentives and so have to be instructed and urged to follow the
moral law’ (Irwin 2004: 149). Cf. also Irwin 2009: 161–2; Korsgaard 1996c: 5: ‘If the
real and the good are no longer one, value must find its way into the world some-
where. Form must be imposed on the world of matter. This is the work of art, the work
of obligation, and it brings us back to Kant. And this is what we should expect. For it
 Kant’s hybrid account of obligation 91

To a significant degree, therefore, I would argue that the basic

structure of Kant’s ethics resembles that of his theoretical philosophy,
which equally employs a form/content distinction to combine realism
with idealism. Thus, while Kant rejects an idealism that goes ‘all the
way down’ and leads to a Berkeleyan subjectivism that would be uncon-
strained by the world, he also rejects a full-blown realism that aims
to treat all our experience of that world as if it conformed to reality
as it exists in itself, wholly independent of our perspective on it. And
equally, of course, Kant hoped that this would enable us to escape
the fruitless oscillation between empirical idealism on the one hand,
and transcendental realism on the other. In ethics, it can therefore be
argued, Kant could use his distinction between the divine will and our
own to settle a similar kind of impasse, in a way that promises some
of the similar dividends of his ‘Copernican revolution’ in theoretical
philosophy.64

Externalism and internalism. I now want to turn to a second area where

again I believe that Kant’s account of obligation also enables him to
take a distinctive ‘middle path’, this time between externalism and
internalism concerning the relation between the normative status of
certain actions and the motivations of agents.65

was Kant who completed the revolution, when he said that reason – which is form –
isn’t in the world, but is something that we impose upon it’. On the hybrid reading of
Kant that I am proposing, Korsgaard is only half right – right about obligation, but
not about value.
64
Kant’s position here, regarding the moral ‘must’, may also be compared to what he says
regarding the modal distinction between possibility and actuality, which he claims
arises from our perspective as discursive intellects, and is therefore ‘merely subject-
ively valid for the human understanding’, as it does not hold for an intuitive intel-
lect such as God’s, who cannot think of something without it existing. Cf. CJ 5:401–3
(pp. 272–3), where Kant goes on to discuss the moral case immediately afterwards:
5:403–4 (p. 273).
65
This is to be distinguished from a different debate in ethics where the vocabulary of
externalism and internalism is also used, which centres on reasons, where the internal-
ist argues that for an agent to have a reason to act, that action must in some way relate
to the agent’s desires, interests, and concerns, in a way that the externalist denies.
The debate I will discuss is also to be distinguished from another more closely related
internalism/externalism dispute, which concerns motivations, but which focuses on
whether making a positive moral judgement about an action can in itself give one a
motivation to so act (internalism), or whether it can only do so in conjunction with
some additional factor (such as a desire to act morally). Following Darwall, this pos-
ition is often called judgement internalism, whereas the one I will discuss is called exist-
ence internalism: see Darwall 1992.
92 Understanding Mor al Obligation

This terminology is used in a number of different ways, but at the

heart of the dispute I am interested in here is the question whether
(as the internalist claims), for someone’s actions to possess their moral
status (of being right or wrong, or good or bad, or a duty, or an obliga-
tion, or something they ought to do, and so on), it is necessary for the
agent to have (actually or dispositionally) a motivation so to act, so that
the former depends on the ‘internal states’ of the latter; or whether on
the other hand there is no such link or dependence, as the externalist
claims. Thus, on the internalist view, the moral features of actions are
said to be essentially related to considerations concerning motivation,
thereby bringing in the motivational states of agents, not just features
of the action that stand outside the agents concerned and are ‘exter-
nal’ to them (for example, whether the action would maximise happi-
ness, or is divinely commanded, or is ‘fitting’, or whatever).
One’s stance on this dispute can be influenced by a variety of consid-
erations, but perhaps three are central: the relation between cases of
moral judgement and action; the threat of moral scepticism and how
to deal with it; and the relation between this issue and meta-­ethical
disputes concerning realism and various forms of anti-realism.
On the first issue, internalists have been struck by the fact that in
making a positive moral judgement concerning some action, it then
seems very curious not to act or to at least admit to some inclination
to doing so: surely, the internalist argues (absent weakness of will and
other complicating factors),66 in these circumstances we would be
forced to conclude that the agent did not really have the moral con-
cept in the first place or had not genuinely made a judgement involv-
ing it.67 This then suggests (the internalist will claim) that what makes
the moral judgement true partly concerns the agent’s motivational
state and not just some fact ‘external’ to this. By contrast, externalists
have thought that such cases, while in fact (thankfully) rare in practice
(perhaps because most of us possess an antecedent disposition to act
on our moral judgements, or that these judgements bring about such
motivations in themselves), nonetheless do not involve any conceptual
incoherence as such, as they would if to make a first-personal moral
judgement about an action is to already attribute to oneself some
motivation to so act.

66
 For example, that there is a greater obligation to do something else.
67
Cf. Smith 1994: 67–71.
 Kant’s hybrid account of obligation 93

On the issue of moral scepticism, internalists have taken it as an

advantage of their view that a certain sort of moral sceptic is ruled
out, namely a sceptic who makes a moral judgement concerning some
action (‘this is right’, ‘this is what I ought to do’), but denies any motiv-
ational connection to the action, and so still asks to be given some
motivating reason to act in this way. In so far as dispensing with the
threat of scepticism is always desirable in a philosophical position,
internalists have therefore claimed it as an advantage of theirs that
it does so.68 Against this, however, externalists have claimed that the
internalist position is in fact implausibly strong when it comes to scep-
ticism, in seeming to make amoralism of this kind incoherent, when
in fact it is not – the amoralist, the externalist argues, is a perfectly
conceivable creature, and we are not guilty of any conceptual error in
taking him seriously as a threat.69
And on the background meta-ethical issue, philosophers have
often been attracted to internalism because it seems most naturally to
go along with an anti-realist position in ethics, since this most easily
explains why the truth of a moral judgement depends on the motiv-
ational status of the agent: for example, on an expressivist view, moral
facts are themselves grounded in the attitudes, desires and passions
of agents, where these are states that motivate the agent to act, so it is
clear why the internalist claim might hold. On the other side, there-
fore, realists have generally been wary of accepting internalism in this
form, fearing that to do so will push them into an anti-realist stance.
Here, then, we again have a debate where it is possible to be pulled
both ways. For, even if the externalist is right to claim that there is not
something wholly inconceivable in a person forming a moral judge-
ment about an action and yet not being moved to so act, on the other
hand the internalist seems right to point out that there is something
curious in this situation that needs to be explained. And even if the
externalist is right that amoralist moral scepticism is not incoherent in
every respect, on the other hand it would be good to have something
to say to such a sceptic along internalist lines. And while the realist
might be right to be wary of internalism, this does not in itself count
as an argument against it, so that to anyone who is not a realist, it might
instead count in its favour.

68
For a thorough discussion of the relation between internalism and moral scepticism,
see Superson 2009: 127–59.
69
Cf. Brink 1989: 46–50.
94 Understanding Mor al Obligation

Now, from the perspective of Kant’s position as I have characterised

it, I think we are in a position to diagnose this debate, and see how it is
that there is something plausible in both sides. This diagnosis hinges
on the distinction we have drawn between rightness or goodness on
the one hand, and obligatoriness or duty on the other, a distinction
which (as we have seen) relates closely to Kant’s contrast between the
holy will and a will such as ours. For, I will argue, this enables Kant
to be an internalist at the level of obligation and duty, while being an
externalist about the right and the good, and thus once again allows
him to combine elements from both positions, and so resolve the
impasse between them.
When it comes to obligation or duty, as we have seen, Kant holds
that these arise for wills such as ours because our non-moral inclina-
tions are in some sort of tension with our moral motivations, in a way
that does not happen for a holy will, which is only ever motivated to
act morally. It follows from this account, therefore, that for a person to
judge or see that they have an obligation or duty to φ, they must have
at least some motivation to φ, as otherwise no such tension could arise.
Kant’s position on obligation and duty is therefore at least weakly
internalist, in the sense that it is committed to there being at least
some degree of motivation to φ in the agent who has φ-ing as their duty
or obligation, whether or not that motivation will always ultimately be
the one that wins out. To this extent, therefore, the Kantian can agree
that there is something indeed incoherent in acknowledging a duty or
obligation and failing to acknowledge any motivation to do it, which
fits with the internalist’s intuitions about some of the cases; and he can
agree about the absurdity of the sceptic who claims to see he ought to
do φ, or that φ is his duty, but not to have any motivation to φ: to grasp
the former, he must have the latter, for the former involves a tension
between that motivation and some other.
On the other hand, Kant can also accommodate externalist con-
cerns that embracing internalism can take us too far: for, as we have
seen, Kant’s hybrid account allows him to distinguish between moral
notions like duty and obligation on the one hand, and notions like the
right and the good on the other, and so to hold that while the inter-
nalist may be right about the former, the externalist is right about the
latter. For, while the internalist’s position might well fit with impera-
tival concepts like duty and obligation, for Kant it is possible also to
conceive of moral concepts that do not take this imperatival form, so
that judgements involving such concepts need assert or imply nothing
 Kant’s hybrid account of obligation 95

about the agent’s motivations, but can be employed and made true in
an externalist manner. This, then, allows room for an externalist treat-
ment of actions that are right and good for the agent to do, while com-
bining this with an internalist treatment of actions that are obligations
or duties for the agent. Moreover, this means that Kant can respect
internalist intuitions about the sceptic who ascribes moral properties
like duty and obligatoriness to her actions, as such properties would
seem to have motivational considerations ‘built in’. At the same time,
however, it can also make space for externalist intuitions concern-
ing the sceptic who only ascribes to them properties like rightness or
goodness: for nothing in the moral properties of the right or the good
involves this internal relation to motivational forces in the way that it
does for obligatoriness and duty, on Kant’s picture. And Kant need not
therefore claim that the amoralist about the right and the good, who
accepts that an act has these features but lacks any motivation to so
act, can be ruled out on the grounds of misusing the relevant concepts
or failing to make a judgement.
Moreover, we can also see how Kant can offer a more complex treat-
ment of the relation between internalism and externalism on the one
hand, and anti-realism and realism on the other. For, just as his anti-
realism operates at the level of obligatoriness but not at the level of
the right and the good, so too his internalism operates only at that
level and not the second; he therefore shows how the realist can safely
embrace internalism about some aspects of the moral, in so far as this
does not in itself force him to become an anti-realist all the way down.
In these respects, it is interesting to compare Kant’s position with
that of W. D. Falk on such issues. Falk, of course, is widely credited
with first crystallising the whole externalism/internalism debate, both
in terms of the distinction itself, and also how it relates to wider meta-
ethical issues, such as the contrast between realism and anti-realism.
However, it has less often been noted that Falk seems to adopt a view
that, like Kant’s, operates at different levels and so is equally hybrid in
defending an internalism and anti-realism about obligation and duty,
and an externalism and realism about the right and the good. This
has been obscured, I think, because the subsequent debate has often
focused on moral properties or judgements in general rather than on
duty and obligation in particular, in a way that means this distinction
of levels gets lost – so that it is not properly appreciated that when Falk
talks about internalism, it is really an internalism about obligation that
he is speaking about, not about all moral properties.
96 Understanding Mor al Obligation

This feature of Falk’s position comes out particularly clearly in his

article ‘Obligation and Rightness’.70 In this paper, Falk’s basic aim is
to question whether Ross (and to a lesser degree Prichard), in pro-
viding an account of rightness, can also claim to have accounted for
moral obligation. Falk does so by contrasting rightness and goodness
as states of affairs external to the subject, with duties and obligations,
which involve the subject feeling ‘internally constrained to do the act in
question’.71 It is for this reason that Falk thinks that ‘when we try to
convince another that he ought to pay his bills, we expect our argu-
ment if accepted to affect a change of heart in him, though it may still
not change his outward actions’.72 Falk therefore argues that the good
and the right can be no more than the ground of what is obligatory,
and should not be confused with obligatoriness as such:
The nature of the things which we are obliged to do contains only the
grounds, but not yet the essence of moral obligation. What alone can ren-
der a prospective action obligatory is that an agent is in some man-
ner impelled to do it, or that he thinks he would be so impelled if he
reflected … Hence, what makes the good act a duty is not the bare fact
that it would be good when done, but the fact that the thought of it is
related to ourselves in a special manner; and even if it were the case that
ultimately none but good acts were obligatory, their goodness would be
no more than the ground of a separate obligation to do them … The
same argument applies to rightness or fittingness.73

In distinguishing the right and the good from duty and obligation in
this way, Falk, like Kant, appears to adopt a hybrid view; and this means
his position can contain both internalist and externalist elements.
Thus, Falk is willing to accept an externalist position like Ross’s
and Prichard’s when it comes to the right and the good. He argues,
however, that this should not be confused with obligation and duty,
which has a different logical form from the right and the good, which
involves taking into account the motivations of the agent when attrib-
uting to them any obligation and duty:
Here it is interesting to note that the belief that obligations are inde-
pendent existents is in some manner fostered by the suggestiveness of
language. ‘Having an obligation’ or ‘being under an obligation’ sug-
gests a state of affairs existing for an agent, yet not merely in relation to

70
Falk 1945. This is one of Falk’s earliest papers, and perhaps for this reason it was not
included in his collection of papers (Falk 1986) – and therefore is correspondingly
little discussed.
71
Falk 1945: 138. 72
 Falk 1945:141. 73 Falk 1945: 145.
 Kant’s hybrid account of obligation 97

him. But to ‘have an obligation’ is not like ‘having money in the bank’;
to be ‘under an obligation’ is not like being ‘under a shower’ or ‘in the
water’. If anything it is like ‘having an impulse’, ‘having an obsession’,
or ‘being in trouble’. For the second set of expressions we can substitute
assertions about individual states of mind, like ‘being impelled’, ‘being
obsessed’, or ‘being troubled’, for the first we cannot. I have no doubt
that ‘having an obligation’ ranks with the second. The possibility of
substituting for it the expression ‘being obliged’ is a clear clue to this.
Strictly speaking, there is nothing that can be called an obligation. What
we think of when we use the term is that agents are obliged by the thought
of them, or that the thought of actions obliges agents to do them. We think not
of an entity, but of a relation between agents, the thought of actions, and
the doing of actions.74

Falk therefore argues that ‘[t]o oblige is to affect, to be obliged is to

be affected’.75 But it is important to note that he is talking only about
obligation here, not the right and the good as such, where he raises no
objection to the realist accounts of this offered by Ross and Prichard;
his objection is just to treating the two normative levels in the same
manner, and so treating law and duty as if they were inherent ‘in the
nature of things’.76 Falk rightly (in my view) characterises his pos-
ition as Kantian, both to the extent that ‘[t]o Kant the very existence
of a duty was inseparable from the existence of a motive’, as provid-
ing ‘some real check on [a person’s] freedom to act otherwise’,77 and
because like Kant he traces the ‘must’ of obligation back to the con-
flict between reason and desire, and the limitations of the human
will.78 Thus, when Falk comes to argue in his later papers that there is
a certain incoherence in moral scepticism that the externalist cannot
acknowledge, and that the externalist cannot account for the absurd-
ity of being faced with someone who accepts that φ-ing is their duty but
questions whether they have any motivation for doing so, it is import-
ant to remember that Falk (like Kant) is pressing the force of these
internalist considerations just at the level of duty and obligation, so
that nothing here commits him to internalism about the ‘ground’ of
obligation, which is the right and the good; and in so far as internal-
ism has any affinity with anti-realism, nothing commits him to anti-
realism about this ‘ground’ either.

74
Falk 1945: 143. 75
 Falk 1945: 143.
76
Falk 1950: 69/1986: 180 77 Falk 1947–8: 121/1986: 29.
78
Cf. Falk 1950: 74/1986: 184, and 1944: 8–9/1986: 168–9.
98 Understanding Mor al Obligation

Divine command and natural law theories of obligation. Finally, I think

we can now see how it is that Kant occupies an unusual position in
relation to the traditional debate between divine command and
natural law theories of obligation, and thus why his account has been
hard to classify.
Where his position resembles that of the natural law theorist is in
the way that he treats what is right as independent of any direct act of
will, and so to this extent he can agree with Leibniz that the moral law
is not authored and is as fixed and necessary as the fundamental prin-
ciples of mathematics or geometry. On the other hand, he can also
agree with the divine command theorist that some account is needed
of what it is that makes this law a constraint and something that binds
us. If the natural law theorist treats this as stemming from a concern
with our own good, the position threatens to become too eudaimonis-
tic and to make the imperatives of morality merely hypothetical; but
if it just posits that obligatoriness as inherent in things, the position
threatens to become too mysterious, while also putting God under
such obligations in a way that the voluntarists abhorred. Kant, there-
fore, can offer an explanation of the nature of this obligatoriness
which neither traces it back to a concern with our own interests, nor
posits it as inherent in things in a puzzling way, or in a manner that
puts God under moral obligations just like us; he can therefore pre-
serve a crucial distinction between the human and the divine on the
one hand, without making us subordinate to the divine or rendering
God’s will arbitrary on the other.
Thus, to the extent that Kant’s position separates out obligatoriness
from what it is that is obligatory, it most resembles the kind of ‘inter-
mediate’ theory of figures such as Suarez and Culverwell. However, his
position also differs from theirs, because they still offered a divine com-
mand account of how the good or right comes to have an obligatory
force, whereas for Kant it comes not from an external will, but from the
structure of our own wills, and the way in which our non-moral inclina-
tions set up a resistance to acting morally that then makes it appear to
us that the moral action is required or binding. Thus, whereas Suarez
and others offer what might be called a ‘positive’ account by ground-
ing obligatoriness in a divine will, Kant gives a ‘negative’ one, arguing
that this obligatoriness is an appearance (though of course none the
less real to us for that), and so a function of our perspective on moral-
ity. For Kant, therefore, it would be misguided and misleading to offer
a positive account by appealing to the divine will, however much we
 Kant’s hybrid account of obligation 99

may then need to introduce such a will in order to ‘complete’ morality

in other respects, out of which (as we have seen in the previous chap-
ter) a different and harmless picture of God as the author of obliga-
tion may then emerge.
This hybrid position, then, enables Kant to overcome the tension
between obligatoriness on the one hand, and autonomy on the other.
For, as we have seen, his concern was that by treating God as the source
of obligatoriness, we will render the will heteronomous, so that both
the voluntarist and intermediate divine command positions are prob-
lematic in this respect. To this extent, a natural law account is to be
preferred; but then it fails to explain what it is that makes the right
obligatory to us. Kant’s position, therefore, tries to offer an explan-
ation, but because it does not rely on any external will to do so, it
dissolves the threat to our autonomy that arises when that obligatori-
ness is given a source beyond ourselves, as on the divine command
approach. To this extent, then, constructivist interpreters of Kant are
right to see him as moved by the argument from autonomy, and as
offering a way out that can be characterised as involving a notion of
‘self-legislation’. But, as we have also seen, this is only an anti-realist
account of obligatoriness, combined with a realism about what it is
that is obligatory, so to that extent they are wrong to hold that Kant’s
anti-realism goes all the way down, and that the argument from auton-
omy led him into a constructivism about values.
Interpreters such as Schneewind and others are therefore perfectly
correct to stress the importance of autonomy to Kant’s ethical thought;
they are just mistaken in the implications they draw from it. But this is
not to minimise the significance and originality of Kant’s position in
any way: as we have seen, it remains a position that is hard to assimilate
to any of the traditional options in the debate, and stands out as an
extremely elegant response to the competing demands that any such
position needs to meet, given the assumptions with which Kant was
working. Thus, it remains the case that Kant’s ethical thought takes the
tradition in a new direction, even if it is not the same as the one it is
said to follow on the standard story, as on my account the starting point
raises a different set of issues for Kant’s successors from the one that
this story assumes. It is those issues that we will now go on to explore,
where the hope is that this new reading of Kant will enable us to reach
a better understanding of the direction of post-Kantian ethics.
Pa rt i i

H ege l
 4

H ege l’s c r i t iqu e of K a n t

( v i a S c h i l l e r)

We turn now from Kant to his successors. We have already discussed

how on the standard account, the central difficulty he is said to have
bequeathed them is the Kantian paradox of self-legislation: namely,
how to make sense of a legislative will that relies on no prior order of
values, without rendering that willing empty from a normative point
of view. On the account of Kant’s position that I have offered above,
however, this is no longer the central difficulty: for, I have argued,
Kant’s position allows for just such a prior order of values. The stand-
ard story, of course, holds that Kant’s concerns about autonomy rule
out an appeal to any such values; but, I have also argued, it would be
wrong to follow the standard story here as well.
Aside from avoiding the problem of emptiness, the other problem-
atic aspects of the self-legislation picture also look different on the
account I have offered. The first of these concerns how self-legislation
could amount to genuine constraint, where it would seem to remain
in the power of the self to declare itself unbound at any time. On
the hybrid account, however, this difficulty can be avoided, for the
will is not in the Euthyphro-situation whereby it is free to make any-
thing obligatory; and the only way to free itself from the constraining
force of the moral law would be for it to become like the holy will,
rather than to just set that law aside in a voluntaristic manner. The
other problem for the self-legislation picture concerned how there
could be such legislation without some part of the self having ‘super-
ior power’ over another part, but where that power cannot be mere
coercive force, but must be some sort of normative authority; but then,
the question is, how can it get that normative authority if there is no
prior set of values on which this authority can be based? Again, this is
less of a difficulty on the hybrid account, as in constraining the will’s

103
104 Understanding Mor al Obligation

inclinations to act, the rational self can derive its authority from its
capacity to discern what has moral value and to lead the subject to act
on it, thereby explaining how the desiring self can be put under its
control in a legitimate way.
However, while the hybrid account may avoid the difficulties associ-
ated with the standard way of understanding Kant’s position, this does
not mean that it is without problems of its own. For, as we have seen, the
hybrid picture relies on the distinction Kant draws between the holy
will and the human will, based on the idea that for us, there is always
a pull against the moral course of action, which renders that action
morally obligatory from the human perspective in a way that it would
not be for the holy will. The account of obligation therefore rests on
the idea that we are ‘fallen’ creatures, torn between an awareness of
what is right on the one hand and inclinations to act differently on the
other, for whom morality always therefore involves an element of resist-
ance or struggle. The question for many of Kant’s successors, then, and
particularly for those in the romantic or idealist tradition, was whether
this account of how we stand in relation to morality was acceptable, or
whether it represents an unattractively dualistic and conflicted picture
of our nature as moral agents – one that rendered nugatory Kant’s way
of dealing with the problem of moral obligation, whatever its advan-
tages over the competing natural law or divine command accounts.
In this chapter, I will explore this concern further, by looking at
the response to Kant in the work of Schiller and Hegel. In the next
­chapter, I will consider what position Hegel was led to in his attempt to
go beyond Kant’s treatment of obligatoriness and avoid what he took
to be its problematic implications.

Schiller on moral beauty

Schiller serves as an important transitional figure here, I will argue,
because while in many respects he follows Kant, and accepts a good
deal of Kant’s way of framing his solution to the problem of moral
obligation, Schiller nonetheless made very vivid the kind of dualistic
picture on which Kant relied, while being more despairing about that
dualism than Kant was prepared to be. To use the familiar distinction
that Schiller himself applied to Kant, while the letter of Schiller’s posi-
tion remains closer to Kant than many have supposed, the spirit inform-
ing that position is very different – and it is by imbibing that spirit that
Hegel comes to challenge the letter of Kant’s position as well.
 Hegel’s critique of Kant (via Schiller) 105

Until relatively recently, the most significant contribution by

Schiller to debates surrounding Kant’s ethics was generally held to be
his well-known and amusing epigram, poking fun at ‘The Scruples of
Conscience’:
Gladly I serve my friends, but alas I do so with inclination
And I am therefore troubled that I am not really virtuous.
To which the reply is:
For this there is no other advice: you must seek to despise them
And so do with disgust what your duty commands.1

Not unreasonably, Kantians have taken this as an objection to their

hero, dismissing it as based on a gross misreading and distortion of
his position; while those more sympathetic to Schiller have accepted
the latter response, but argued that it was never intended as anything
more than a jest, whose only serious point was to correct the com-
mon misreading of Kant by others, which Schiller is here caricatur-
ing.2 The Kantians are surely right to be dismissive of this as a serious
criticism of Kant; and the Schillerians are also surely right to insist
that the position of their hero should not be judged merely on the
basis of this, but rather by looking at his main writings on ethical
issues: namely, the so-called Kallias Letters; the short essay On Grace
and Dignity; and of course Schiller’s main philosophical work, On the
Aesthetic Education of Man.3 It is to further consideration of these that
we therefore now turn.

The Kallias Letters. In 1793, Schiller wrote a series of letters to his

friend, Christian Gottfried Körner. These are known as the Kallias
Letters [Kalliasbriefe], after the title of the essay on beauty that he
told Körner he would write, but never did. The letters were published
posthumously in 1848, in the Weiner Jahrbücher der Literatur. The letters
are strongly marked by Schiller’s encounter with Kant, whose work he
had begun to read in 1787, engaging with it more closely in the spring

1
Schiller (1943–): 1:357.
2
On the Kantian side, see Beck 1960: 231; Paton 1967: 48–50; Prauss 1983: 240–59;
Allison 1990: 180–4; Timmermann 2007: 152–4. On the Schillerian side, see Reiner
1951/1983; Wood 1999: 28–9; Beiser 2005: 169–90. For other recent discussions of the
Kant/Schiller relation, see Gauthier 1997; Baxley 2003 and 2010; Roehr 2003; and
Deligiorgi 2006.
3
Although I will only discuss Schiller’s prose works, his poems are also of course relevant
(particularly ‘Das Ideal und das Leben’ [1795], ‘Die Künstler’ [1789], and ‘Die Götter
Griechenlandes’ [1788]), as well as his other writings.
106 Understanding Mor al Obligation

of 1791. While beauty provides the focus of the letters, as always with
Schiller this is strongly linked also with ethical themes.4
Not surprisingly, given the forceful impression made on him by the
works of Kant that were available at this time,5 much in the letters
is orthodoxly Kantian. Schiller thus follows Kant in distinguishing
between formal and material principles, and in treating the latter as
heteronomous, while also accepting Kant’s transcendental conception
of freedom: ‘The only existing thing which determines itself and exists
for itself must be sought outside of appearances in the intelligible
world’.6 However, Schiller also introduces another notion of freedom,
which is ‘the autonomy of the sense-world [Sinnlichen]’,7 or autonomy
as this relates to the phenomenal and not the noumenal realm. Here,
Schiller argues:
Judging an object as free in appearance depends simply on completely
abstracting it from its grounds of determination (since not-being-
determined-from-the-outside is a negative representation of being-
determined-through-oneself, which is its only possible representation,
because one can only think freedom and not recognize it, and even the
philosopher of morals must make do with this negative representation
of freedom). Thus a form appears as free as soon as we are neither able
nor inclined to search for its ground outside it.8

Schiller thus turns from considering freedom at the noumenal level

to freedom at the phenomenal level, or what he calls ‘the autonomy of
appearance [Erscheinung]’,9 while arguing that ‘only if it explains itself’
in this manner is a form beautiful.10
Having taken this turn in the discussion, Schiller then makes a
fateful step. For, in line with what he has said above, he now argues
that when it comes to a moral act, this lacks autonomy when the agent
must use her reason to restrain herself in some way; for now, in viewing
that act, we will be inclined to think of the action as having a ground
or some process that brought it about, rather than ‘explaining itself’
or ‘being-determined-through-itself’ – for the self has been divided
into parts that stand in an explanatory relation to one another. Thus,

4
For a general discussion of the tradition of ‘aesthetic morality’ that Schiller is drawing
on here, and his place within it, see Norton 1995, especially Chapter 6.
5
For an account of Schiller’s reading of Kant in this period, see Beiser 2005: 37–46.
6
KL 26:193 (p. 155). 7
KL 26:194 (p. 156).
8
KL 26:193 (p. 155). 9
KL 26:194 (p. 156). 10 KL 26:194–5 (p. 156).
 Hegel’s critique of Kant (via Schiller) 107

Schiller comments: ‘When the form of the non-reasonable [nicht-

­vernünftig] is determined by reason (theoretical or practical, both are
the same here), its natural determination is constrained, and beauty
cannot arise’.11 On the other hand, Schiller suggests, when the whole
self is expressed and engaged in moral action, then it can be experi-
enced as beautiful.
Now, in order to show Körner more clearly what he is getting at
here, Schiller undertakes to ‘tell [him] a story’.12 The story concerns
a man who has been robbed, and left naked and injured on the street
in the cold. He is passed by travellers on five occasions, and on each
occasion the travellers react differently to his plight. The first traveller
responds with empathy to his suffering (‘I suffer with you’), but can-
not bring himself to help on his own behalf as the appearance of the
injured man revolts him, so he offers him some money so that he can
pay some other people to help him instead. This traveller is criticised
for meaning well, but for not being prepared to overcome his ‘tender
senses’ and do his ‘duty to humanity’ [Menschenpflicht], where he only
acted to the extent that he was emotionally moved by the plight of the
man. Schiller comments: ‘What was this action? It was neither useful,
morally generous nor beautiful. It was merely impulsive, kind-hearted
out of affect’. A second traveller then appears, who complains that sav-
ing the injured man will cost him time and money, but who offers to
do so if he is compensated. Of this traveller Schiller writes: ‘And what
was this action? It was neither generous nor dutiful, neither magnani-
mous nor beautiful. It was merely useful’. A third traveller comes by.
He stands there ‘battling with himself’. On the one hand, he needs his
coat and horse himself, as he too is cold and ill; on the other hand, he
sees that ‘duty commands that I serve you’, and so reluctantly agrees to
help. Schiller says of this third traveller: ‘This action was purely moral
(but also no more than that), because it occurred against the interests
of the senses, out of pure respect for the law’. Now two travellers arrive,
who recognise the injured man as an enemy of theirs that they had
been looking for, to seek revenge for a misfortune he has brought on
them. The injured man expects them to kill him; instead they offer to
help him, but without forgiving him. Their help is therefore rejected
by the injured man, who says: ‘I would rather die a miserable death

11
KL 26:195 (p. 156).
12
KL 26:195 (p. 157). Subsequent quotations in this paragraph come from KL 26:195–7
(pp. 157–8).
108 Understanding Mor al Obligation

than to owe such an enemy my life’. Finally, the injured man sees yet
another traveller carrying a heavy load, who he resolves to let go past,
having been disappointed thus far. However, on this fifth occasion,
the traveller drops his burden and rushes to help ‘as soon as the wan-
derer sees him’, offering to take the injured man to the next village, at
the risk of having his own belongings stolen by abandoning them on
the street. ‘But what will become of your load which you leave here on
the open road?’, the injured man asks him. This traveller replies: ‘That
I don’t know, and it concerns me little … I do know, however, that you
need help and that I owe it to you to give it to you’.13 Schiller concludes
this letter to Körner by saying ‘Greetings from all of us here. In the
meantime, think about why the action of the carrier was beautiful’.
In the letter that he writes to Körner the next day, Schiller spells out
what he takes to be the moral of the story:
The beauty of the fifth action must lie in that characteristic which sets
it apart from all the previous ones.
(1) All five wanted to help; (2) most of them chose an adequate
means for the job; (3) several of them were willing to have it cost them
something; (4) some overcome their own self-interest in order to help.
One of them acted out of the purest moral purpose. But only the fifth
acted without solicitation, without considering the action, and disregard-
ing the cost to himself. Only the fifth forgot himself in his action and
‘fulfilled his duty with the ease of someone acting out of mere instinct’.
Thus, a moral action would be a beautiful action only if it appears as an
immediate [sich selbst ergebenden] outcome of nature. In a word: a free
action is a beautiful action, if the autonomy of the mind and the auton-
omy of appearance coincide.
For this reason the highest perfection of character in a person is
moral beauty brought about by the fact that duty has become its nature.14

Schiller goes on:

Clearly the violence against our drives which practical reason brings to
bear on our moral determination of will appears as something insult-
ing and embarrassing. We never want to see coercion, even if it is reason itself
which exercises it; we want even nature’s freedom to be respected because
‘we regard every being in aesthetic judgement as an end in itself’ and
it disgusts (outrages) us, for whom freedom is the highest thing, that

13
Translation modified. The original reads: ‘Ich weiβ aber, daβ du Hilfe brauchst und
daβ ich schuldig bin, sie dir zu geben’.
14
KL 26:198 (p. 159).
 Hegel’s critique of Kant (via Schiller) 109

something should be sacrificed for something else, and used as a means.

That is why a moral action can never be beautiful if we observe the
operation through which it is won from the sensory-world. Our sensory
nature must thus appear free, where morality is concerned, although it
is really not free, and it must appear as if nature were merely fulfilling
the commission of our drives by subjugating itself to the mastery of the
pure will, at the expense of its own drives.15

In this way, Schiller makes transparent the intended lessons of his tale
of the travellers and the injured man which he had told Körner in the
previous letter.
It should be obvious from what he says that he offers here an implied
criticism of Kant, but also a position that still tries to remain close to
Kantian orthodoxy, and where the criticism is meant to be an internal
one. For, Schiller had earlier written to Körner that ‘[i]t is certain that
no mortal has spoken a greater word than this Kantian word, which
also encapsulates his whole philosophy: determine yourself from
within yourself’.16 This, Schiller says, represents an important break
from those ‘religiously orientated thinkers of moral philosophy’,17 who
put the source of morality in God. However, Schiller is also criticising
Kant for not going far enough in the conception of autonomy that
he had pioneered; for although he has avoided placing God over us
as an external source of constraint, he has instead introduced the
internal constraint of reason in its place, which takes away from our
autonomy in its own way, and in a manner that is equally insulting and
demeaning.
Kant adopts this position, Schiller holds, because he does not take
seriously enough the possibility of the kind of beautiful action exem-
plified by the fifth encounter with the traveller: that is, he does not
accept that in a moral action, the agent may act fully in accord with
his ‘sensory nature’, and so achieve an ‘appearance of freedom’ – even
though, Schiller accepts, Kant is right that considered transcenden-
tally this can never be anything more than an appearance and not real
freedom. The case here, it is worth emphasising, is different from the
well-known Kantian examples in Section i of the Groundwork, of the
philanthropic lover of mankind and of the cold-hearted follower of
duty, to which Schiller’s epigram contentiously refers: for the issue

15
KL 26:198 (p. 159) (my emphasis). Cf. also ‘Die Künstler’, lines 83–3: ‘The heart that
she guides with gentle bonds disdains to be led in servility by duties’.
16
KL 26:191 (p. 153). 17 KL 26:191 (p. 153).
110 Understanding Mor al Obligation

here is not whether or not it takes away from the moral value of an
action that it is done from the pleasure or satisfaction of so acting,18
or (conversely) that a person is only truly moral if they can find no
pleasure or satisfaction in the moral action when it is done. Schiller
is thus neither defending a certain hedonistic view of moral action
contra Kant, nor accusing him of an extreme rigorism which sees no
connection at all between morality and any sense of well-being that
the agent may derive from it. Thus, it is important to the telling of his
story that Schiller does not inform us at all about how much inner sat-
isfaction the final traveller does or does not derive from acting well,
and whether the anticipation of such satisfaction provides him with
his motivations: for, unlike the first traveller, there is no motivational
role here given to ‘affect’. Instead, the key difference between the final
traveller and the others, and of course particularly the third, is that
on seeing the situation of the injured man, he realises that the right
thing to do is to provide help as this is what is needed, and nothing in
him speaks against acting in this way. The issue this raises for Schiller,
therefore, is whether as moral agents we can be more like the holy will,
which does not act morally because it thereby gains satisfaction, or out
of inclination, but also does not have to resist any non-moral inclina-
tions in doing so, and so does not act against them: both sides of the
agent’s nature are therefore in alignment in this respect.
Now, although Schiller does not mention him explicitly, it is hard
not to hear echoes here of Aristotle’s conception of the virtuous agent,
both in the story Schiller tells, and in the contrast he draws with Kant.
As is well known, in the Nicomachean Ethics, Aristotle makes a distinc-
tion between six types of moral character, in ascending order of value:
brutishness, vice, incontinence, continence, virtue, and superhuman
virtue.19 The extreme ends of the scale scarcely apply to us and so
receive little discussion, where the brute is more like an animal than
a human being, and the supremely virtuous being is more like a god;
the former are wholly within the grip of their depraved desires with

18
As Kant characterises the cases he discusses, the contrast is between someone who
does what is right because they are ‘so sympathetically attuned’ that ‘they find an
inner satisfaction in spreading joy around them and can take delight in the satisfaction of
others so far as it is their own work’, and a person who does what is right without this
bringing them any sense of happiness of this kind, and thus without any inclination in
its favour (GMM 4:398, pp. 53–4 (my emphasis)).
19
NE Book vii, §1, 1145a15–1145b21, pp. 1808–9. (I have largely used the translation
found in Aristotle 1984, with some modifications (most obviously, the use of ‘virtue’
as the translation of arête, rather than ‘excellence’)).
 Hegel’s critique of Kant (via Schiller) 111

no real capacity for ethical training, while the latter are the opposite,
in being so good as not to require any such training and hence seem-
ing to be born of the gods rather than of man. Between these two, lie
four states of moral character to which almost all humans are prone,
and which thus interest Aristotle the most. The worst of these four is
vice, where reason and appetite are in harmony in pursuing the wrong
ends. For the incontinent individual, by contrast, the rational part of
their nature aims at the right end, but the non-rational part does not,
and so there is a conflict in which the non-rational part generally wins.
For the continent person, there is also conflict of the same kind, but
here the rational part triumphs over the non-rational part. Finally,
for the virtuous person, these two parts of the soul are in harmony,
where both are directed at the right end, and so speak with the ‘same
voice’.20
Now, as we shall see, Aristotle’s classification here and his position
as a whole are not without their complexities. But for the moment, it
is natural to read Schiller’s story and its moral in Aristotelian terms,
and see the third traveller who proposes to act ‘against the interests
of the senses, and out of pure respect for the law’ as merely contin-
ent, and the final one who acts with no sense of struggle against his
non-moral inclinations as properly virtuous. And other aspects of
Schiller’s position here are also recognisably Aristotelian, particularly
his emphasis on considerations of character, and his suggestion that
duty or right action has ‘become its nature’ for the virtuous person, echo-
ing Aristotle’s conception of the second nature that can be instilled in us
through habit and training,21 overlaying our ‘first’ nature but in a way
that enables us to act morally without effort or struggle, with the two
parts of our soul aligned.

On Grace and Dignity. While such themes are aired in the letters to
Körner, they are made even more explicit in Schiller’s short treatise
On Grace and Dignity [Über Anmuth und Würde], which appeared
in the Neue Thalia later in the same year, in June 1793. As in the
Kallias Letters, ethical and aesthetic issues are intertwined, both in
Schiller’s characterisation of grace and dignity themselves, and in

20
NE Book i, §13, 1102b28, p. 1742. Cf. also Book i, §8, 1099a10–21, p. 1737; Book i, §13,
1102b13–28, p. 1737; Book iii, §11, 1119a11–20, p. 1766; Book vii, §2, 1146a10–12,
p. 1810; Book vii, §9, 1151b34–1152a7, p. 1820.
21
NE Book ii, §1, 1103a14–1103b26, pp. 1742–3.
112 Understanding Mor al Obligation

his characterisation of the ‘beautiful soul’ [die schöne Seele]. The work
itself is divided into two halves, the first of which deals with grace and
the second with dignity, while in aesthetic terms the former is linked
to the beautiful and the latter to the sublime. As we shall see, while
Schiller’s conception of grace echoes the critique of Kant that we
found in the Kallias Letters, his introduction of the notion of dignity
considerably complicates and may seem to soften that critique; and it
is certainly a mistake to judge Schiller’s position in relation to Kant
here only on the basis of what he says about the former and thus about
the ‘beautiful soul’.22
The basic contrast behind Schiller’s distinction between grace
and dignity reflects the distinction between the third traveller in
the story he sent to Körner and the final one: that is, in dignity
there is a conflict between reason and the non-rational parts of the
agent, whereas in grace there is harmony, where again it is natural
to map this distinction onto Aristotle’s contrast between continence
and virtue respectively. Schiller thus considers there to be three
ways in which the rational and non-rational aspects of a person may
interrelate:
Humans either suppress the demands of their sensuous nature in order
to have a proper relation to the higher demands of their rational nature;
or they reverse this and subjugate the rational part of their being to the
sensuous, and thus respond only to the prod that natural necessity uses
to mobilize them as well as other appearances; or the impulses of the

I think this is a weakness of Anne Margaret Baxley’s otherwise helpful essay (Baxley
22

2003), which focuses mainly on what Schiller has to say about grace, where what he
then says about dignity is then noted briefly towards the end, but not really taken up
(see p. 510); however, this discussion is developed further in Baxley 2010. Kantian
critics of Schiller have also not always given sufficient weight to the second part of the
essay; so, for example, Jens Timmermann declares that ‘Kant was a dualist, Schiller
a monist’ (Timmermann 2003: 154) on the basis of Schiller’s discussion of grace,
without seeing that Schiller’s discussion of dignity shows that he treats this as an ideal
(much like the holy will), and so is as arguably as much of a dualist about our natures
as Kant ever was. As David Pugh has rightly warned, ‘In their reading of Über Anmut
und Würde, scholars have focused on the history of physical grace and moral beauty at
the expense of their opposites, dignity and moral sublimity. This approach gives rise
to a one-sided picture of Schiller as a devotee of beauty and harmony, a somewhat
mannered picture that used to feature prominently in standard accounts of Weimar
Classicism but that is hard to square with Schiller’s accomplishments as a tragedian.
The elements of indecision and logical strain in the text of Über Anmut und Würde are
glossed over in this selective reading, causing it to appear a more serene and inte-
grated piece of writing than it really is’ (Pugh 1996: 240).
 Hegel’s critique of Kant (via Schiller) 113

sensuous settle into harmony with the rules of the rational and human
beings are at one with themselves.23

In a Platonic manner, Schiller compares these three possibilities to

three ways in which a state may be governed:
The first of these relationships between the two natures in the human
is reminiscent of a monarchy in which the ruler’s strict surveillance bri-
dles every free stirring; the second is like a wild ochlocracy, in which the
citizen, by refusing obedience to the rightful supremacy, fails to gain
freedom, just as the human constitution, by suppressing moral self-mo-
tivation, fails to become beautiful; instead, he becomes a victim of the
more brutal despotism of the lowest classes, just as form falls victim to
mass. Just as freedom lies in the centre between anarchy and the suppres-
sion of law, we now find beauty in the middle between dignity, that is to
say, the expression of dominant mind, and lust, as the expression of
dominant impulse.24

Schiller first explores this middle ground, before going on to dis-

cuss dignity as the more autocratic position in the second part of the
essay.
Schiller starts by stating his agreement with Kant, that while we
can feel pleasure in following our reason, in acting morally we do not
follow it in order to gain any such pleasure, but act morally for its
own sake.25 He also agrees that Kant is right to hold that in assessing
whether a person has performed a moral action, any consideration of
how far they were inclined by their desires to act in the way that they
did should be set aside.26 Thus far, therefore, Schiller shows himself

23
GD 20: 280 (p. 147).
24
GD 20:281–2 (p. 148). Cf. also GD 20:278–9 (pp. 145–6).
25
Cf. GD 20:282 (p. 148): ‘In order to become the object of inclination, obedience to
reason must deliver a source of enjoyment, because the impulse is only set in motion
by pleasure or pain. In common experience it is the other way round, and pleasure is
the reason for acting rationally. We have to thank the immortal author of the Critique
[i.e. Kant], to whom fame is due for having reestablished healthy reason out of philo-
sophical reason, for the fact that morality has finally stopped using this language’.
26
Cf. GD 20:283 (p. 149): ‘For, because ethical behavior is not about the legality of the
deeds but only about the dutiful nature of the attitude, it is right not to set any store
by the observation that in the first case it is usually advantageous if the inclination is
aligned with duty. This much seems certain, that sensuousness, in bestowing approval,
may not actually make the dutiful nature of the will suspicious, but it is at least not in a
position to vouch for it. The sensuous expression of this approval in grace will never bear
adequate and valid witness to the ethical nature of the action in which it appears, and
one will never learn of the moral worth of an attitude or action from the beautiful way it
114 Understanding Mor al Obligation

to be in every sympathy with Kant’s discussion of the relation between

duty and inclination in Section i of the Groundwork.27
However, the discussion now takes a less Kantian turn, from what
makes an action dutiful to what makes a person virtuous. In order for
virtue to be achieved, Schiller argues, a greater degree of harmony
between reason and inclination is required, and only then will the
‘ethical spirit’ of the individual be truly realised:
Firmly persuaded as I am – and just because I am persuaded – that the
part played by inclination in a free action demonstrates nothing about
the purely dutiful nature of this action, I believe I can conclude from this
that the ethical perfection of the human being can only become clear
precisely because of the part played by inclination in moral actions.
The human being is not destined to perform individual ethical actions
but to be an ethical being. Virtue is prescribed for him, rather than vir-
tues, and virtue is nothing other than ‘an inclination for duty’. However
much actions from inclination and actions out of duty stand opposed to
one another in an objective sense, this is not the case in the subjective
sense, and humans not only may, but should combine enjoyment with
duty; they should obey their reason with joy. Not in order to discard
it like a burden or shrug it off like a course covering, no, their purely

is presented’. As Baxley 2010: 88 points out, Schiller here appears to foreshadow Paton’s
emphasis on Kant’s ‘method of isolation’ in the Groundwork: see Paton 1967: 47.
27
Schiller’s also shows his sympathy with Kant in his essay ‘On the Sublime’, where
Schiller contrasts the person who ‘finds his delight in practicing justice, beneficence,
moderation, constancy, and good faith’ and the same person who through misfor-
tune finds this delight has evaporated, but who still continues to act virtuously. Of the
former Schiller asks ‘though he may inspire us with affection, are we quite sure that
he is really virtuous?’, while of the latter he says: ‘It is this revelation of the absolute
moral power which is subjected to no condition of nature, it is this which gives to
the melancholy feeling that seizes our heart at the sight of such a man that peculiar,
inexpressible charm, which no delight of the senses, however refined, could arouse
in us to the same extent as the sublime’ (OS 21:43–5 (p. 105)). These latter comments
are very reminiscent of Kant’s remarks in Section iii of the Groundwork. Cf. also this
passage from ‘On the Necessary Limitations in the Use of Beautiful Forms’: ‘The man
who enjoys a continual prosperity never sees moral duty face to face, because his incli-
nations, naturally regular and moderate, always anticipate the mandate of reason,
and because no temptation to violate the law recalls to his mind the idea of the law.
Entirely guided by the sense of the beautiful, which represents reason in the world of
sense, he will reach the tomb without having known by experience the dignity of his
destiny. On the other hand, the unfortunate man, if he be at the same time a virtuous
man, enjoys the sublime privilege of being in immediate intercourse with the divine
majesty of the moral law; and as his virtue is not seconded by any inclination, he bears
witness in this lower world, and as a human being, of the freedom of pure spirits!’
(NLBF 21:27 (p. 179)).
 Hegel’s critique of Kant (via Schiller) 115

intellectual nature is accompanied by a sensuous one so that the sen-

suous can agree as closely as possible with the higher self. In that it
made him a rational, sensitive being, that is, a human, nature gave the
human being notice of his obligation not to separate what it had bound
together and, even in the purest expressions of his divine part, not to
neglect the sensuous, and not to base the triumph of the one on the
subjugation of the other. Only when it gushes forth from mankind as
a whole as the combined effect of both principles, when it has become his
nature, is the ethical spirit secure, since, as long as the ethical spirit still
employs force, the natural impulse has to respond to it with strength. The
enemy who has merely been laid low can get up again, but the one who
is reconciled has been truly overcome.28

In claiming here that virtuous human beings should go along hap-

pily or joyfully with their moral actions, Schiller is not rejecting what
he takes to be Kant’s central point, that moral actions should not be
guided by one’s inclinations or feelings; rather, he is objecting to the
suggestion that in being guided by reason in acting morally, the vir-
tuous person will always undergo some experience of struggle and
so will go against their sensuous self, which means that they cannot
undertake the action naturally and with ease or joy – where this makes
it hard to see that the pleasure that even Kant allows can come from
acting morally can indeed obtain in this case (except, perhaps, the
strenuous pleasure of a really hard task having been successfully com-
pleted, and the satisfaction of extirpating thereby a troublesome part
of oneself).29
Having opened up this point of disagreement with Kant, which
largely reflects his position in the Kallias Letters, Schiller suggests
that it only arises from the rather extreme way in which Kant himself
presented his position, which (Schiller fully accepts) was perfectly jus-
tified by the dialectical situation of the time, where it was necessary
to respond forcefully to the threat of eudaimonism and hedonism in
ethics.30 Thus, Schiller writes, Kant was ‘the Draco of his time, because

28
GD 20:283–4 (pp. 149–50).
29
Cf. CPrR 5:117–19 (pp. 234–5), where Kant remarks in a Stoic manner on the ‘con-
tentment’ that can come from ‘mastery over one’s inclinations, hence of independ-
ence from them and so too from the discontent that always accompanies them’ (5:119
(p. 235)).
30
Cf. also OS 21:52 (p. 109): ‘Away then with that false theory which supposes falsely a
harmony binding well being and well doing’.
116 Understanding Mor al Obligation

he did not regard his time as worthy of a Solon or yet able to receive
one’.31 And yet, Schiller goes on:
How were the children of the house at fault, if he was only concerned about
the servants? Must an unselfish emotion in the noblest of breasts come
under suspicion just because impure inclinations often usurp the name
of virtue? Just because the moral weakling would like to introduce a
certain laxity into the law of reason, to make it a toy for his own conveni-
ence, does this mean that a rigidity has to set in, transforming the most
powerful expression of moral freedom into merely an honorable kind
of servitude? Does the truly ethical human have a freer choice between
self-regard and self-reproach than the slave of the senses between pleas-
ure and pain? Is there perhaps less pressure on a pure will than on a
depraved one? Does mankind have to be accused and humiliated simply
by the imperative form of the moral law, and does the most sublime docu-
ment of its greatness also have to be a certification of its frailty? Could
one indeed, in this imperative form, have avoided a situation where a
prescription given by humans to themselves as rational beings and there-
fore binding only on them and compatible only with their feeling of
freedom took on the appearance of an unfamiliar and positive law – an
appearance that could be reduced only with difficulty, because of their
radical tendency (of which they stand accused) to work against it?32

In this complex series of rhetorical questions, Schiller is clearly chal-

lenging the Kantian to avoid the complaint that he has exaggerated
the difficulties we face in living the moral life, and the effort that this
requires, albeit as part of Kant’s laudable campaign to defeat the false
impression of the time, that morality consists in nothing more than act-
ing in accordance with our desires and interests. In seeking to uphold
the imperatival force and obligatory nature of the moral law, the ques-
tion is, has Kant overdone the dualism that he introduces to explain it?
Having commented that ‘[h]uman nature is a more coherent whole
in reality that a philosopher, who can only achieve results through
separation, is permitted to reveal’,33 Schiller moves on to give his

31
GD 20:285 (p. 151). Schiller adopts a similar attitude in AE 6th Letter, §13, 20:327
(pp. 41,43), praising Kant’s achievements as stemming from a necessary rationalism,
while suggesting that this rationalism now needs to be replaced by a more balanced
position, if it is not to do more harm than good.
32
GD 20:285–6 (p. 151).
33
GD 20:286 (p. 152). Cf. also AE 1st Letter, §4, 20:310 (p. 5): ‘Like the analytical
chemist, the philosopher can only discover how things are combined by analyzing
them, only lay bare the workings of spontaneous Nature by subjecting them to the
torment of his own techniques’. This concern that an ‘analytic’ method falsifies the
 Hegel’s critique of Kant (via Schiller) 117

­ ell-known account of the ‘beautiful soul’, in which rational and non-

w
rational elements are in such harmony that the agent can allow herself
to be led entirely by her sensibility and inclination: ‘One refers to a
beautiful soul when the ethical sense has at last so taken control of all
of a person’s feelings that it can leave affect to guide the will without
hesitation and is never in danger of standing in contradiction to its
decisions’.34 While conceding to Kant that ‘the actions of the beautiful
soul are not themselves ethical’, because not carried out through duty,
Schiller claims that ‘the character as a whole is so’, because it is truly
virtuous:
The beautiful soul has no other merit besides being. It carries out
humankind’s most exacting duties with such ease that they might sim-
ply be the actions of its inner instinct, and the most heroic sacrifice that
it exacts from natural impulse appears to the eye as a free operation
of this impulse. This is why the beautiful soul itself never knows about
the beauty of its actions and it does not even occur to it that one could
act and feel differently. A pupil well-schooled in the rules of ethics, as
required by his master’s instructions, on the other hand, will always
be ready to give a precise account of the relation between his actions
and the law. His life will resemble a drawing in which one can see the
rules behind the precise lines and from which at most an apprentice
could perhaps learn the principles of art. In a beautiful life, however,
all the sharp dividing lines disappear, as in a painting by Titian, and yet
the whole impression comes across more faithfully, more harmoniously,
and truer to life.35

Finally, Schiller argues, it is just this harmony that gives the beauti-
ful soul grace: ‘It is in a beautiful soul that sensuousness and reason,
duty and inclination are in harmony, and grace is their expression as

unity of the phenomena it studies was shared by Goethe and others: cf. Faust Part
One lines 1936–41, which were quoted by Hegel on several occasions (EL §38Z 8:110
(p. 63), and EN §246Z 9:21 (ii, p. 202)): ‘Who would study and describe the living,
starts / By driving the spirit out of the parts: / In the palm of his hands he holds all
the sections, / Lacks nothing, except the spirit’s connections. / Encheirisis naturae
[‘manipulating nature’] the chemists baptize it, / Mock themselves and don’t realise
it’ (Goethe 1963: 199).
34
GD 20:287 (p. 152).
35
GD 20:287 (pp. 152–3). Cf. also KL 26:217 (p. 174): ‘But a character, an action, is not
beautiful if it shows the sensual nature of the person who is its recipient under the
coercion of law, or constrains the senses of the viewer. In this case the actions will
produce mere respect, not favour or a good disposition; mere respect humbles the per-
son who receives it’.
118 Understanding Mor al Obligation

appearance’.36 After some musings about the way in which grace is

more particularly associated with women than men, this section of the
work then closes.
However, having seemed perhaps to offer the beautiful soul as his
alternative to the Kantian position, Schiller’s discussion now takes
what may then appear to be a surprising course, which apparently
brings him back closer to Kant again, as he moves from grace to dig-
nity in the second part of the work.
He begins by arguing that while human beings have the task of
achieving the kind of harmony required by grace, and thus this beauty
of character, it is nonetheless ‘only an idea that they can vigilantly
strive to live up to, yet, despite all efforts, can never fully attain’.37 This
is because, Schiller claims, our natural selves are primarily driven by
the desires to attain pleasure and to avoid pain, which will therefore
mean that this aspect of ourselves will always find itself running up
against the requirements of morality, no matter how much we can do
to bring these desires under our control. It is in exercising control in
this way that the human being displays dignity rather than grace, and
the aesthetic quality of sublimity rather than beauty:
And since nature, for ethical reasons, never withdraws her demands,
and therefore everything on her side remains the same, no matter
how the will behaves in relation to her, so here there is no agreement

36
GD 20:288 (p. 153).
37
GD 20:289 (p. 154). Cf. also ‘Das Ideal und das Leben’: ‘If human sin confronts the rigid
law / Of perfect truth and virtue, awe / Seizes and saddens thee to see how far / Beyond
thy reach, perfection; – if we test / By the ideal of the good, the best, / How mean our
efforts and our action are! / This space between the ideal of man’s soul / And man’s
achievement, who hath ever past? / An ocean spreads between us and that goal, / Where
anchor ne’er was cast! / But fly the boundary of the senses – live / The ideal life free
thought can give; / And, lo, the gulf shall vanish, and the chill / Of the soul’s impotent
despair be gone! / And with divinity thou sharest the throne, / Let but divinity become
thy will! / Scorn not the law – permit its iron band / The senses (it cannot chain the
soul) to thrall / Let man no more the will of Jove withstand, / And Jove the bolts let fall!’
Given Schiller’s position here, it is curious to find Henry Allison drawing such a sharp
contrast with Kant, on the grounds that for the former the beautiful soul is obtainable
by human beings, whereas for the latter the holy will is not: see Allison 1990: 183:
‘This emphasis on human finitude likewise accounts for the radical gulf separating
Kant’s problematic conception of holiness from Schiller’s superficially similar con-
ception of a beautiful soul. Reduced to its essentials, the difference consists in the
fact that whereas the latter is conceived as a moral condition obtainable by human
beings through a moralization of the inclinations (the development of an inclination
to duty), holiness is a purely regulative idea to be approached asymptotically but never
obtained by a finite rational agent’.
 Hegel’s critique of Kant (via Schiller) 119

possible between inclination and duty, between reason and sensuous-

ness; so humans cannot here act with their whole nature in harmony,
but only with their reason. In such cases, then, they also do not act with
moral beauty, because inclination necessarily has to participate in the
beauty of action as well, but here it stands in conflict. But they do not
act with moral greatness because all those things and only those things
are great that give evidence of superiority of the higher faculties over
the sensuous … Control of impulses through moral strength is spiritual
freedom, and its expression in appearance is called dignity.38

Thus, unlike Kant’s holy will, we as human beings cannot count on

being naturally good, or having a ‘virtue born of temperament’: for in us
this would be a virtue that relied on the fortunate alignment of moral-
ity and our interests, which can come apart at any time. Reason might
be willing to cede control to sensuousness when this alignment occurs;
but it will always need to be prepared to take back this control when it
does not, thus reflecting a disposition which is sublime and dignified,
rather than beautiful and graceful. Returning, then, to the analogy
with the state that he had offered earlier, Schiller contrasts the liberal-
ity of moral beauty with the autocracy required of dignity.39
Now, at this point it may seem that Schiller has gone back entirely to
an orthodox Kantian position. For he may now appear to be agreeing
with Kant that while human beings can achieve moral action and a
virtuous disposition, this must always involve a struggle by reason with
their non-moral inclinations; and while he has introduced the contrast
between this and what is required for grace, by treating the beautiful
soul who possesses this quality as no more than an ideal, it may also
appear that Kant has no reason to disagree with him, as Kant can
likewise claim that his ideal of the holy will has a similar harmony that
we can only admire and try to model ourselves on, without every really
hoping that it can be attained. It may thus seem that the discussion

38
GD 20:293–4 (p. 158). Cf. OS 21:42 (p. 103): ‘In the presence of beauty we feel our-
selves free, because the sensuous instincts are in harmony with the laws of reason. In
presence of the sublime we feel ourselves sublime, because the sensuous instincts have
no influence over the jurisdiction of reason, because it is then the pure spirit that acts
in us as if it were not absolutely subject to any other laws than its own’.
39
Cf. GD 20:296–7 (p. 160): ‘In dignity, then, the mind conducts itself in the body as
ruler, because there it has to assert its independence against domineering instinct,
which proceeds without it, and would like to escape from its yoke. In grace, on the
other hand, the mind governs with liberality, because it is the mind that sets nature in
action here, and it finds no opposition to quell. Leniency is due only to the obedient,
however, and only opposition justifies severity’.
120 Understanding Mor al Obligation

here in On Grace and Dignity is much less clearly opposed to Kant than
the one in the Kallias Letters, whereby the critical power of the open-
ing section on grace is softened and brought more into line with Kant
by the subsequent discussion of dignity.
However, it is important to remember that the beautiful soul is a
rather special case, which does not just have a harmony between ‘sen-
suousness and reason, duty and inclination’, but has it to such a degree
that all control of the will can in this being be given over to its sen-
suous and natural side, and reason can stand back entirely, knowing
here that inclinations can be trusted to be on the side of right action
without its authority needing to be imposed. In treating the beautiful
soul as an ideal, then, Schiller can be read as merely claiming that for
creatures like us, such trust would always be misplaced and foolish,
as non-moral inclinations can always arise; as a result, reason must
always be vigilant and remain in control.40 It may of course be true,
Schiller notes, that we would expect a virtuous individual to be able to
put faith in his sensuous side to some extent, and to be able to rely on it
not to run amok entirely;41 but we would also not expect a responsible
human being to put faith in it altogether, at the expense of abandon-
ing all rational vigilance over his inclinations and desires.42

40
Cf. NLBF 21:27 (p. 179): ‘These are the dangers that threaten the morality of the char-
acter when too intimate an association is attempted between sensuous instincts and
moral instincts, which can never agree in real life, but only in the ideal’. And cf. OS
21:53 (p. 109): ‘Without the beautiful there would be an eternal strife between our
natural and rational destiny. If we only thought of our vocation as spirits we should be
strangers to this sphere of life. [But] Without the sublime, beauty would make us for-
get our dignity. Enervated – wedded to this transient state, we should lose sight of our
true country. We are only perfect citizens of nature when the sublime is wedded to
the beautiful’.
41
Cf. GD 20:287 (p. 152): ‘A person does not make a good impression on me if he can
trust the voice of his impulse so little as to feel obliged to test its tone against that of
moral principles; one respects him more highly if he confidently trusts it and is not in
danger of being misled by it’.
42
As Rosalind Hursthouse has rightly stressed, there is much common ground between
the Aristotelian and Kantian pictures in this respect: ‘In short, [for Aristotle] the emo-
tions of sympathy, compassion, and love, viewed simply as psychological phenomena,
are no guarantee of right action, or acting well. There is nothing about them, qua
natural inclinations, which guarantees that they occur “in complete harmony with
reason”, that is, that they occur when, and only when, they should, towards the people
whose circumstances should occasion them, consistently, on reasonable grounds and
to an appropriate degree, as Aristotelian virtue requires … Kantians and Aristotelians
agree on the fact that [Kant’s happy philanthropist, who acts out of inclination and
not from duty] … cannot be relied upon to act well’ (Hursthouse 1999: 102).
 Hegel’s critique of Kant (via Schiller) 121

Thus, because from a human perspective, the model of the beau-

tiful soul is really rather extreme, this explains something that has
puzzled many commentators43 – which is that Schiller now brings his
discussion of grace and dignity together, to suggest on the one hand
that grace is at least achievable for us in part, while on the other hand
this means that dignity can be incorporated into a picture of virtue, to
which otherwise it would remain alien. The implied criticism of Kant,
therefore, is not that he saw dignity where there is only grace, but that
he saw only dignity where there is also grace.
Thus, having considered a character that is either all about grace
or all about dignity, Schiller then proceeds subtly to explore the way
in which both aspects of character can and should be displayed by
human beings, who cannot (given the nature of their desires) always
be expected to act morally without any sense of control over those
desires being required, but who also can be expected to so act on occa-
sions; and while Kant does full justice to the first aspect of our moral
lives, Schiller can still be viewed as criticising him for not properly
accommodating the latter, and so not seeing that considerations of
duty are not always appropriate for virtue, even in limited moral crea-
tures such as ourselves.
So, Schiller argues, there can be circumstances in which, if a person
displays no struggle with their inclinations in acting ethically, and thus
no dignity, then this would suggest to us that he is taking the matter too
lightly, as we would expect anyone to find the situation challenging; on
the other hand, there are cases where we would expect a person not
to undergo any struggle in opting for the ethical course of action, so
here we would criticise him for lacking grace. There are, then, norms
of human behaviour that guide us in these judgements, where those
norms reflect an awareness of what human desires and inclinations
standardly are, and how far they do and do not generally align with
the moral course of action. We would therefore expect a person in
a situation of great danger to feel some resistance in acting to save
others, and so would not criticise them for displaying dignity rather
than grace; equally, we would not expect a person who was asked by
someone for their help, who is perhaps their friend or lover, or who
can deliver that help with no great cost to themselves, to find that they

Cf. Norton 1995: 243, who comments on Schiller’s claim that grace and dignity must
43

be united: ‘With that statement, Schiller quite simply abdicated the domain governed
by the laws of logic and rational argument and fled into a world of his own creation’.
122 Understanding Mor al Obligation

feel a desire to do otherwise that pulls against the moral action, so that
in this sort of case we would criticise them for lacking grace. As Schiller
puts this idea: ‘In general, the law is valid here that humans should do
everything with grace that can be carried out within humanity, and
everything with dignity that requires going beyond humanity’.44
By introducing this complexity, and thus by complicating the simple
picture of grace in the first part of his essay, Schiller may here seem to
be compromising the Aristotelian picture with which the essay began
and which is implicit also in the Kallias Letters. But in fact, it is argu-
able that this subsequent discussion merely deepens the connection
to Aristotle, who himself introduces a similar kind of complexity into
his account of continence and virtue. For, while Aristotle sometimes
presents the contrast along the lines outlined above, where virtue
consists in desire and reason being in straightforward alignment and
continence consists in their being in some sort of tension, in fact the
cases that Aristotle discusses are rarely that simple, and are closer to
Schiller’s approach.45 Thus, while in the case of temperance, Aristotle
does not expect the temperate individual to have to resist the desires he
has to act intemperately,46 in the case of the brave man who is risking
44
GD 20:298 (p. 162).
45
For a discussion of some of these complexities of Aristotle’s position, see Foot 1978:
1–18. She introduces her discussion of those complexities as follows: ‘[W]e both are
and are not inclined to think that the harder a man finds it to act virtuously the more
virtue he shows if he does act well. For on the one hand great virtue is needed where it
is particularly hard to act virtuously; yet on the other it could be argued that difficulty
in acting virtuously shows that the agent is imperfect in virtue: according to Aristotle,
to take pleasure in virtuous action is the mark of true virtue, with the self-mastery of
the one who finds virtue difficult only a second best. How then is this conflict to be
decided? Who shows most courage, the one who wants to run away but does not, or the
one who does not even want to run away? Who shows most charity, the one who finds it
easy to make the good of others his object, or the one who finds it hard?’ (Foot 1978:
10). Cf. also Hursthouse 1999: 91–107, and Broadie 1994.
46
Cf. NE Book vii, §2, 1146a10–17, p. 1810: ‘Further, if continence involves having strong
and bad appetites, the temperate man will not be continent nor the continent man
temperate; for a temperate man will have neither excessive nor bad appetites. But
the continent man must; for if the appetites are good, the state that restrains us from
following them is bad, so that not all continence will be good; while if they are weak
and not bad, there is nothing admirable in resisting them, and if they are weak and
bad, there is nothing great in resisting these either’. Cf. also Book iii, §11, 1119a11–20,
p. 1766: ‘[The temperate man] neither enjoys the things that the self-indulgent man
enjoys most – but rather dislikes them – nor in general the things that he should not,
nor anything of this sort to excess, nor does he feel pain or craving when they are
absent, or does so only to a moderate degree, and not more than he should, nor when
he should not, and so on; but the things that, being pleasant, make for health or for
good condition, he will desire moderately and as he should, and also other pleasant
things if they are not hindrances to these ends, or contrary to what is noble, or beyond
 Hegel’s critique of Kant (via Schiller) 123

his life or serious injury, Aristotle thinks he remains fully brave and
virtuous, despite not wanting to be killed or wounded and having to
compel himself to fight regardless of these desires47 – so to do what he
sees as his duty, we might say.
Thus, much as Aristotle also seems to allow in these cases, Schiller
stresses that while grace is a kind of ideal, we will as human beings
combine elements of both: ‘If grace, supported by architectonic
beauty, and dignity, supported by strength, are united in the same
person, then the expression of humanity is complete in that person,
and he stands there, justified in the world of spirit and affirmed in
appearance. The two legislations are in such close contact here that
their boundaries flow together’.48 Schiller thus runs through a num-
ber of cases where grace and dignity should both find a place, where
‘it is only from grace that dignity acquires recognition and only from
­dignity that grace acquires value’.49 Thus, for example, he claims that
‘[o]ne demands grace from a person who places obligations, and dig-
nity from someone who is placed under an obligation’,50 and that
‘[o]ne should criticize a mistake with grace, and confess one with
dignity’,51 because otherwise, if you confess it with grace it will look
like you take correcting it to be to your advantage, while if you make
the criticism with dignity, it will look like you don’t care enough about
the wrongdoing yourself.

his means. For he who neglects these conditions loves such pleasures more than they
are worth, but the temperate man is not that sort of person, but the sort of person that
right reason prescribes’, and Book iii, §12, 1119b13–17, p. 1767: ‘Hence the appetitive
element in a temperate man should harmonize with reason; for the noble is the mark
at which both aim, and the temperate man craves for the things he ought, as he ought,
and when he ought; and this is what reason directs’.
47
Cf. NE Book iii, §9, 1117b7–10, p. 1764: ‘death and wounds will be painful to the brave
man and against his will, but he will face them because it is noble to do so or because it
is base not to do so’. Cf. also NE, Book iii, §7, 1115b11–13 and 1115b17–19, p. 1761: ‘Now
the brave man is not as dauntless as man may be. Therefore, while he will fear even the
things that are not beyond human strength, he will fear them as he ought and as reason
directs, and he will face them for the sake of what is noble; for this is the end of virtue
… The man, then, who faces and who fears the right things and with the right aim, in
the right way and at the right time, and who feels confidence under the corresponding
conditions, is brave’. See also EE Book iii, §1, 1128b39–1129a7, p. 1946.
48
GD 20:300–1 (p. 163). Cf. also OS 21:53 (p. 109): ‘Without the beautiful there would
be an eternal strife between our natural and rational destiny. If we only thought of our
vocation as spirits we should be strangers to this sphere of life. Without the sublime,
beauty would make us forget our dignity. Enervated – wedded to this transient state,
we should lose sight of our true country. We are only perfect citizens of nature when
the sublime is wedded to the beautiful’.
49
GD 20:300 (p. 163). 50
GD 20:299 (p. 162). 51
GD 20:299 (p. 162).
124 Understanding Mor al Obligation

A final and particularly interesting example of where Schiller

stresses the complementary nature of grace and dignity concerns the
relation of love on the one hand, and respect on the other – where
of course respect is a notion that Kant famously made much use of,
while seeming to be comparatively dismissive of the claims of love.52
As one might expect, Schiller particularly associates love with grace,
and certainly praises it highly as a result. However, Schiller empha-
sises at the same time that love for us can also be a selfish emotion,
and one that can involve self-deception about the true worth of the
beloved – which is why love should also be tempered by respect.53 And
at the same time, respect without love can become servility and fear,
so that both are in the end required for us. As Schiller summarises
his position: ‘Dignity prevents love from becoming desire. Grace pro-
tects respect from becoming fear’. Here, Schiller’s claim is not just
that the different attitudes of love and respect will be required in
different situations and circumstances,54 but that they will need to
be combined at all times (as it were), to prevent the other lapsing into
desire on the one hand or servility on the other – or at least, they
will need to be so combined by us human beings, for in a higher crea-
ture, no such lapsing of love into desire would occur, and so they
could remain graceful without dignity, beautiful without sublimity.55
Moreover, when it comes to the comparison with Kant, it is notable
that he too characterises a relation like friendship in terms of this

52
 Cf. CPrR 5:76 (p. 202).
53
Schiller had previously discussed the relation between love and egoism in ‘The
Theosophy of Julius’, which was published in 1786 as part of his Philosophische Briefe.
54
This is Frederick Beiser’s suggestion on how grace and dignity can both be displayed
by a single character. While that is certainly plausible in some cases, I think the
example of love (which Beiser does not discuss) suggests that Schiller wanted the
integration to be even closer, where both may be aspects of one and the same situ-
ation. For Beiser puts his view as follows: ‘What Schiller has to say about the com-
plementary nature of grace and dignity makes perfect sense once we see that he is
talking about different occasions or contexts of acting according to a single dispos-
ition … [Schiller] thinks that we must act with dignity only in tragic circumstances,
with grace in non-tragic ones’ (Beiser 2005: 115–6). A similar view is held by Lesley
Sharpe: ‘In Über Anmut und Würde Schiller attempts a synthesis of the beautiful soul
and the sublime by postulating the need for the beautiful soul, when under pressure,
to evince dignity. Grace and dignity should, thus, alternate in the same person. This
attempt to synthesize two arguably incompatible models, with the resulting logical
problems, occurs again in the Ästhetische Briefe and Über naïve und sentimentalische
Dichtung’ (Sharpe 1995: 3).
55
For more comments by Schiller on the dangers of a one-sided emphasis on love, see
also NLBF 21:24–5 (pp. 177–8).
 Hegel’s critique of Kant (via Schiller) 125

balance, speaking of it as ‘the union of two persons through equal

mutual love and respect’.56
Thus, in so far as Schiller introduces no such complement to beauty
in the Kallias Letters, and in so far as dignity involves a move in a
more Kantian direction, On Grace and Dignity may appear to be a more
Kantian work than the Letters, in acknowledging that for us, it is unreal-
istic to think that we might ever achieve a complete and stable harmony
between the rational and sensuous sides of our natures, and thus that
we might ever get beyond the experience of duty, even though this may
on occasion be the case. We have seen, then, that Schiller presents both
a powerful alternative to Kant’s position in one way, but seemingly faith-
fully recapitulates it in another. That is, it may at first appear (and has
so appeared to many) that in On Grace and Dignity he offers a critique of
Kant, in setting up virtue as more admirable than duty; treating grace
as more beautiful than dignity; in insisting that the relation of reason
to inclination should be one of harmony and not discord; and allowing
his moral ideal of the beautiful soul to retain a place for desires and
inclinations, whereas Kant’s ideal of the holy will appears to do with-
out them altogether, at least in its ‘infinite’ form. On the other hand,
Schiller makes clear that these are claims that only apply to what is for
us a kind of eternal ‘beyond’, where as a result in reality the moral life
for a human being must involve duty, dignity, and the autocracy of rea-
son over inclination. And, of course, Kant himself said as much, for he
too treated duty not as a characteristic of the good will in itself, but of
the will ‘under subjective limitations and hindrances’,57 while agreeing
that the holy will is clearly superior to ours as an ideal. There is thus no
dispute between Kant and Schiller that the very best moral life would
be without such ‘hindrances’ – or that this differs fundamentally from
the best moral life for us, and what differences that might involve in
the nature of our moral experience. To this extent, then, On Grace and
Dignity may seem to be a considerably more Kantian document than
the Kallias Letters, even while containing elements similar to the latter,
but treated now as a moral ideal.
However, much as in the Kallias Letters, in On Grace and Dignity,
Schiller still insists on a degree of achievable harmony between rea-
son and sensuousness, duty and inclination, even if that harmony can-
not be made complete, to the extent that (like the beautiful soul) we

56
MM 6:469 (p. 584).
57
GMM 4:397 (p. 52).
126 Understanding Mor al Obligation

can safely give control over to our ‘natural’ selves altogether. And of
course, as the story of the Kallias Letters only presents us with the
agent on one occasion, it is not clear that Schiller wanted to go that far
here either, and so whether he really changed his mind in On Grace and
Duty. In so far as it is only the more modest claim that Schiller needs
in order to make his case against Kant (that grace as well as dignity can
characterise our moral lives), and in so far as the Kallias Letters need
involve no more than this in the contrast they draw between the third
and final travellers, Schiller can consistently treat the beautiful soul as
an ideal in On Grace and Dignity (as such a being acts only with grace),
without abandoning the anti-Kantian impact of the essay entirely.

Aesthetic Letters. We may turn finally to Schiller’s most significant

philosophical writing, On the Aesthetic Education of Man in a Series
of Letters of 1795. Although this work has a wider scope than those
previously discussed, many of the themes that concern us still emerge
strongly.
Adopting the language of drives (Triebe) which he took from
Reinhold, in the twelfth letter Schiller distinguishes between the sen-
suous drive that encompasses sensations, feelings, and desire, and the
form drive which relates to reason. His conception of these drives in the
Aesthetic Letters is broader than the largely ethical distinction between
inclination and reason suggested in the previous works; nonetheless,
they still retain an ethical aspect, as does Schiller’s insistence that they
be harmonised, in a way that he again presents as going against the
letter of Kant’s system, if not its spirit.58 Thus, in introducing the play
drive [Spieltrieb] as a synthesis of the other two, Schiller writes that ‘[t]o
the extent that it deprives feelings and passions of their dynamic power,
[the play drive] will bring them into harmony with the ideas of reason;
58
Cf. AE 13th Letter, §2 note, 20:348: (p. 87): ‘In the Transcendental method of phil-
osophizing, where everything depends on clearing form of content, and obtaining
Necessity in its pure state, free of all admixture with the contingent, one easily falls
into thinking of material things as nothing but an obstacle, and of imagining that our
sensuous nature, just because it happens to be a hindrance in this operation, must of
necessity be in conflict with reason. Such a way of thinking is, it is true, wholly alien to
the spirit of the Kantian system, but it may well be found in the letter of it’. Cf. also AE
13th Letter, §2 note, 20:347–8 (p. 85): ‘Once you postulate a primary, and therefore
necessary, antagonism between these two drives, there is, of course, no other means of
maintaining a unity in man than by unconditionally subordinating the sensuous drive
to the rational. From this, however, only uniformity can result, never harmony, and
man goes on for ever being divided. Subordination there must, of course, be; but it
must be reciprocal’.
 Hegel’s critique of Kant (via Schiller) 127

to the extent that it deprives the laws of reason of their moral neces-
sitation [Nötigung], it will unite them with the interests of the senses’.59
Moreover, as in On Grace and Dignity, Schiller treats the full reconcili-
ation of the two drives as a kind of infinite task that cannot actually be
realised by human beings:
Such reciprocal relation between the two drives is, admittedly, but a
task enjoined upon us by Reason, a problem which man is only capable
of solving completely in the perfect consummation of his existence. It is,
in the most precise sense of the word, the Idea of his Human Nature, hence
something Infinite, to which in the course of time he can approximate
ever more closely, but without ever being able to reach it.60

Schiller also associates the harmony between the drives with beauty,61
but rather than linking that strongly with the notion of grace and then
contrasting this with dignity, he here says little about either notion;62

59
AE 14th Letter, §6, 20:355 (p. 99; translation modified). Cf. also AE 14th Letter, §5,
20:354 (p. 97): ‘Both drives [i.e. the sense-drive and the form-drive], therefore, exert
constraint upon the psyche; the former through the laws of nature, the latter through
the laws of reason. The play-drive, in consequence, as the one in which both the others
act in concert, will exert upon the psyche at once a moral and a physical constraint; it
will, therefore, since it annuls all contingency, annul all constraint too, and set man
free both physically and morally. When we embrace with passion someone who deserves
our contempt, we are painfully aware of the compulsion of nature. When we feel hostile
towards another who compels our esteem, we are painfully aware of the compulsion of
reason. But once he has at the same time engaged our affection and won our esteem,
then both the compulsion of feeling and the compulsion of reason disappear and we
begin to love him, i.e., we begin to play with both our affection and our esteem’.
60
AE 14th Letter, §2, 20:352–3 (p. 95). Cf. also AE 16th Letter, §1, 20:360 (p. 111): ‘This
equilibrium [between the sense-drive and the form-drive], however, remains no more
than an Idea, which can never be fully realized in actuality. For in actuality we shall
always be left with a preponderance of the one element over the other, and the utmost
that experience can achieve will consist of an oscillation between the two principles, in
which now reality, now form, will predominate’.
61
Cf. AE 15th Letter, §2, 20:355 (p. 101): ‘The object of the play-drive, represented in a
general schema, may therefore be called living form: a concept serving to designate all
the aesthetic qualities of phenomena and, in a word, what in the widest sense of the
term we call beauty’.
62
One of the few places where grace and dignity are mentioned is at the end of the fif-
teenth letter, where Schiller is discussing the Greeks, and where he emphasises the way
in which they combined the two: ‘It is not Grace, nor is it yet Dignity, which speaks to
us from the superb countenance of Juno Ludovisi; it is neither the one nor the other
because it is both at once’ (AE 15th Letter, §9, 20:359 (p. 109)). The two notions also
make a sonorous appearance at the very end of the book, where we are told that in the
Aesthetic State, people will be as ‘free alike of the compulsion to infringe the freedom
of others in order to assert their own, as of the necessity to shed their Dignity in order
to manifest Grace’ (AE 27th Letter, §12, 20:412 (p. 219)).
128 Understanding Mor al Obligation

and as a consequence, perhaps, there is also hardly any discussion in

this work of the sublime.63 And although Schiller in this work also says
little here about virtue, he does distinguish between the ‘moral con-
dition’ as involving a sense of duty and constraint, from the ‘aesthetic
condition’ which does not, even though the latter is not arbitrary.64
Two features of the Aesthetic Letters add an extra dimension to his
discussion here, however. One is the comparison he makes between
Greece and his contemporary world in the sixth letter, in an attempt
to diagnose the ills of that world which Schiller had enunciated in the
previous letters. Here he famously presents the Greeks as enjoying a
greater harmony than us in the way in which the parts of the soul were
related to one another, claiming that ‘[a]t that first fair awakening of
the power of the mind, sense and intellect did not as yet rule over
strictly separate domains; for no dissension had as yet provoked them
into hostile partition and mutual demarcation of their frontiers’.65
While in other works the comparison with the Greeks is also made,66
Schiller presents his most explicit and elaborated connection here,
although of course the contrast he draws between the Greeks and ‘us
Moderns’ goes more widely than the merely ethical case.
A second feature that is new to the Aesthetic Letters is the political
dimension Schiller gives to the discussion. This is present in the open-
ing letters, where Schiller focuses on the social and spiritual crisis of
the times, and considers whether the action of the state might resolve
that crisis; Schiller claims clearly in letter seven, however, that this is
impossible until a deeper spiritual reform has occurred, for which he
thinks we must look to aesthetic education. Having developed his con-
ception of that education in the main body of the work, he then briefly
but significantly returns to its political implications at the end, and
thus tells us what this new form of society or ‘Aesthetic State’ might
be like. Here, he argues, the individual will no longer feel under the

63
For further discussion of the relation between the Aesthetic Letters and Schiller’s treat-
ment of the sublime, see Pugh 1996: 304–17.
64
Cf. AE 20th Letter, §4 note, 20:376 (p. 143; translation modified): ‘our psyche in the
aesthetic condition does indeed act freely, is in the highest degree free from all com-
pulsion, but is in no wise free from laws; and … this aesthetic freedom is distinguish-
able from logical necessity in thinking, or moral necessity in willing, only by the fact
that the laws according to which the psyche then behaves do not become apparent as such,
and since they encounter no resistance, never appear as a constraint’.
65
AE 6th Letter, §3, 20:321 (p. 31).
66
See, for example, the opening GD 20:251–6 (pp. 123–7), and ‘Die Götter
Griechenlandes’.
 Hegel’s critique of Kant (via Schiller) 129

command of duty, in so far as his non-moral nature will no longer

require any imposition or constraint:
No privilege, no autocracy of any kind, is tolerated where taste rules,
and the realm of aesthetic semblance extends its sway. This realm
stretches upwards to the point where reason governs with uncondi-
tioned necessity, and all that is mere matter ceases to be. It stretches
downwards to the point where natural impulse reigns with blind com-
pulsion, and form has not yet begun to appear. And even at these fur-
thermost confines, where taste is deprived of all legislative power, it
still does not allow the executive power to be vested from it. A-social
appetite must renounce its self-seeking, and the Agreeable, whose
normal function is to seduce the senses, must cast toils of Grace over
the mind as well. Duty, stern voice of Necessity, must moderate the
censorious tone of its precepts – a tone only justified by the resist-
ance they encounter – and show greater respect for Nature through a
nobler confidence in her willingness to obey them … In the Aesthetic
State everything – even the tool which serves – is a free citizen, having
equal rights with the noblest; and the mind, which would force the
patient mass beneath the yoke of its purposes, must here first obtain
its assent.67

Thus, Schiller links his search for a harmony between reason and
inclination with a kind of political programme, through which
that harmony (along with many others) will be realised by the citi-
zens of a community in which a process of aesthetic education has
occurred – although, as ever, Schiller emphasises how difficult this
may be for us to achieve, even while not quite making it an impos-
sible ideal.68

Schiller’s position and its implications. In the previous discussion, we

have traced the complex way in which Schiller appears both to depart
from Kant and criticise him, but also to return to a Kantian position
and largely follow it. It is therefore not surprising that it has proved
hard for commentators to gauge the exact nature of their relation,

67
AE 27th Letter, §11, 20:411–2 (pp. 217,219).
Cf. AE 27th Letter, §11, 20:412 (p. 219): ‘But does such a State of Aesthetic Semblance
68

really exist? And if so, where is it to be found? As a need, it exists in every finely attuned
soul; as a realized fact, we are likely to find it, like the pure Church and the pure
Republic, only in some few chosen circles, where conduct is governed, not by some
soulless imitation of the manners and morals of others, but by the aesthetic nature we
have made our own’.
130 Understanding Mor al Obligation

where some have emphasised their antipathy,69 while others have

stressed their common ground.70
However, although Schiller does indeed draw back from presenting
a radical alternative to Kant’s position in the form of the ‘beautiful
soul’, nonetheless some of the difficulties that he creates for Kant can
be seen in Kant’s own efforts to respond to Schiller, and the way in
which that response appears to miss the mark. Kant is known to have
replied to Schiller in three places: in a long note added to the second
edition of Religion Within the Limits of Reason Alone; in a brief discussion
in the Vorarbeiten to that work which was not itself published; and in a
reference to Schiller in the Vigilantius lectures on ethics – where all
three are a reaction to On Grace and Dignity.71
In the footnote that appears in Religion Within the Limits of Reason
Alone, Kant clearly shows himself to have been stung by Schiller’s com-
ments on his account of obligation, and the apparent suggestion that
‘it carries with it the frame of mind of a Carthusian’.72 In response,
Kant takes up what Schiller has to say about dignity as well as grace,
and emphasises that while he cannot find a place for grace when it
comes to duty, he is nonetheless happy to associate it with virtue. Thus,
on the one hand, he insists that ‘the concept of duty includes uncondi-
tional necessitation, to which gracefulness stands in direct contradic-
tion’, so that it is only appropriate to think of it in terms of dignity. On
the other hand, while he treats virtue as ‘the firmly grounded dispos-
ition to fulfill one’s duty strictly’, he allows that the virtuous character
will take an attitude to the demands being placed upon them that is

69
Cf. Reiner 1983: 41–9; Henrich 1965; and Baxley 2010, esp. pp. 85–123.
70
Cf. Guyer 1996: 354: ‘Schiller clearly shares with Kant the belief that moral appro-
bation is directed most of all precisely to the activity of reason by which we elevate
ourselves above passive nature. In the end, surely, the point of Schiller’s twofold ideal
of grace and dignity is not primarily to criticize Kant’s ethics but to use his artistic
powers to defend Kant’s view, perhaps from the more scornful tendency within him-
self and certainly from the many critics who had ridiculed Kant’s separation between
happiness and virtue from the moment the Groundwork was published’.
71
Baxley plausibly argues that the section in Metaphysics of Morals on ‘Ethical Ascetics’
(MM 6:484–5 (pp. 597–8)) may also be taken to be addressed implicitly to Schiller,
although he is not mentioned (see Baxley 2010: 104–6).
72
Relig 6:23 note (p. 72). Further references in this paragraph are to this note. See also
MM 6:484–5 (pp. 597–8) and Vorarbeiten zur Religion 23:99. Cf. Schiller GD 20:284
(p. 150): ‘In Kant’s moral philosophy, the idea of duty is presented with a severity that
repels all graces and might tempt a weak intellect to seek moral perfection by taking
the path of a somber and monkish asceticism’.
 Hegel’s critique of Kant (via Schiller) 131

‘courageous and hence joyous’; so, rather than being ‘weighed down
by fear and dejected’, the virtuous individual will relish the challenge
posed to her by the demands of the moral life, much as Hercules rel-
ished the challenge posed to him by the monsters he must subdue,
instead of ‘shrinking back in fear and trembling’ like the muses.73 As a
result, Kant suggests, while duty must be all about dignity, nonetheless
‘the glorious picture of humanity, as portrayed in the figure of virtue,
does allow the attitude of the graces’. Kant expresses the hope that by
making his position clear to Schiller in this way, given their conver-
gence ‘upon the most important principles’, this should be enough to
secure his agreement concerning Kant’s treatment of obligation, and
defuse the accusation that he expects the virtuous individual to dis-
play a ‘slavish frame of mind’ bowed down by a resentment against the
demands that the moral law places upon them.
Now, whether or not this expressed hope for concord was disin-
genuous, it seems clear that Kant has missed much of the force of
Schiller’s differences with him, both with respect to duty and with
respect to virtue. For, although Schiller would indeed accept that
duty qua duty cannot involve grace, the deeper question for him is
whether our experience of morality must always take this form, or
whether (like the last traveller in the story from the Kallias Letters),
we might be such as to act rightly without any sense of ‘unconditional
necessitation’ much like the holy will, and thus in accordance with
grace. And although Schiller would indeed accept that the virtuous
person can and should view the struggle involved in acting morally in

73
Kant also uses Herculean imagery in relation to virtue elsewhere: cf. MM 6:405
(pp. 533–4): ‘strength is required [for virtue] in a degree which we can assess only
by the magnitude of the obstacles that the human being himself furnishes through
his inclinations. The vices, the brood of dispositions opposing the law, are the mon-
sters he has to fight. Accordingly this moral strength, as courage (fortitudo moralis), also
constitutes the greatest and the only true honor that man can win in war’. Cf. also LE
27:490 (pp. 259–60), where Kant accounts for the satisfaction we derive from having
acted rightly by comparing it with what we experience after the accomplishment of a
challenging task: ‘the attraction we feel, after fulfilment of a duty, for the action itself
… is derivable … from the cheer experienced on getting through work that has cost
us trouble, and is evidence … of the burdensomeness of duty’. Baxley takes Kant’s
emphasis on the joy and cheerfulness of the virtuous individual to show that ‘there
need be no actual (phenomenological) feeling of constraint at the time of action,
because [the agent] does her duty from duty with a cheerful heart’ (Baxley 2010: 133;
cf. also 103). But I think Kant’s Herculean model of virtue shows that this is too sim-
plistic, as the joy comes precisely from overcoming these constraints. This issue will be
discussed further in the Conclusion.
132 Understanding Mor al Obligation

a positive light, the fundamental issue for him is whether the virtuous
person need always be involved in such a struggle in the first place,
and thus whether Kant’s model of virtue amounts to anything more
than continence.74
It is perhaps no accident, however, that Kant does not identify these
deeper challenges here, given his conciliatory strategy, as they make
it much harder for him to claim common ground with Schiller – as
he makes clear in his other, unpublished, responses to On Grace and
Dignity, where he is more frank about their differences. Thus, Kant
seems to have thought, the only way to prevent beings like us having
to struggle against our inclinations in acting morally is for the moral
action to be seen as attractive to our inclinations in some way;75 but
then, as we have seen, Kant holds that our actions would be lacking in
autonomy (because guided by inclination and not reason), while also
lacking moral worth. And likewise, when it comes to virtue, while Kant
accepts a role for developing some inclinations (such as sympathetic
feelings) to serve as allies for reason in overcoming our non-moral
inclinations,76 this does not make the conflict any less real for the vir-
tuous agent. Thus, for Kant, Schiller’s position involves a fundamental

74
This charge is commonplace amongst virtue theorists and classicists: see, e.g. Nussbaum
2001: 172, where she writes that ‘Kant thought that virtue must always be a matter of
strength, as the will learns to keep a lid on inappropriate inclinations, rather like a
good cook holding down a boiling pot’. Cf. also Annas 1995: 53. Schneewind gives
voice to this view, though perhaps without endorsing it, when he writes: ‘Kant sees vir-
tue in a most un-Aristotelian way, as always a struggle, never a settled principle. Kant’s
vision of the divided self is the villain here, with morality springing from an impossibly
pure reason in conflict with reprobate passions forever calling for discipline. Virtue is
not so much the expression of our nature at its most developed as it is the triumph of
one part of it over another’ (Schneewind 1990: 61/2010: 199). For attempts to correct
this picture, see Baxley 2008 and 2010. See also Rosler 2005: 125–9. Some of the issues
raised here are discussed at greater length in the Conclusion.
75
Cf. LE 27:490 (p. 259): ‘if we wish, with Schiller, to assume a worth arising therefrom
[i.e. from obeying the moral law], it is nothing more than man’s respect for the moral
law, and that provides no ground for supposing a charm that attracts us to fulfilling
it. That is contradicted by the authority of the laws, which enjoins absolute obedience,
and awakens resistance and struggle, which we perceive in fulfilling them’.
76
Cf. MM 6:457 (pp. 575–6): ‘But while it is not in itself a duty to share the sufferings (as
well as the joys) of others, it is a duty to sympathize actively in their fate; and to this
end it is therefore an indirect duty to cultivate the compassionate natural (aesthetic)
feelings in us, and to make use of them as so many means to sympathy based on moral
principles and the feeling appropriate to them. It is therefore a duty not to avoid the
places where the poor who lack the most basic necessities are to be found but rather to
seek them out, and not to shun sickrooms or debtors’ prisons and so forth in order to
avoid sharing painful feelings one may not be able to resist. For this is still one of the
 Hegel’s critique of Kant (via Schiller) 133

confusion between the holy and the human will, treating the latter as
if it were capable of being the former, and thus of avoiding the ‘neces-
sitation’ of duty.77 The interest of Schiller’s work, however, is the way
in which it seems to question the truth of these Kantian assumptions,
by adopting a more Aristotelian model of the virtuous person. For, on
this model, the virtuous agent can be viewed as one in whom (on some
occasions, at least) base inclinations are simply lacking (as a result of
their correct upbringing and moral education), so we do not need to
assume (as Kant does above) that they have been ‘trumped’ by a con-
trary inclination that leads the agent to do what is right rather than
what is wrong, or that in the virtuous individual, reason is required
to marshal other inclinations to fight against them. On this model,
then, we can claim with Schiller that reason and inclination are in
harmony, without claiming that this harmony is achieved because the
moral individual finds themselves with an inclination to do what is
right that overcomes contrary inclinations; rather, this can also be
achieved by someone who, as a result of the previous effects of proper
habituation, now has no inclination drawing them away from what rea-
son tells them is the right thing to do.78

impulses that nature has implanted in us to do what the representation of duty alone
might not accomplish’.
77
Cf. the discussion of Schiller in the Vorarbeiten and the Vigilantius lecture notes, espe-
cially 23:100 and LE 27: 489–91 (pp. 258–60). For a passage that may seem more
Schillerian, see Relig 6:58 (p. 102), where Kant says that we should not seek to ‘extir-
pate’ our natural inclinations in a Stoic manner, as there need not necessarily be any-
thing ‘reprehensible’ about them. However, in fact, Kant’s main point in this passage
is not to claim that inclinations present no moral obstacle to us, but that the Stoics
are wrong to make this the primary or central difficulty, which is with our will and its
tendency to incorporate non-moral maxims, in a way that is the source of our radi-
cal evil: ‘That is, the first really good thing that a human being can do is to extricate
himself from an evil which is to be found not in his inclinations [as the Stoics wrongly
thought] but in his perverted maxims and hence in freedom itself. Those inclinations
only make more difficult the execution of the good maxims opposing them; whereas
genuine evil consists in our will not to resist the inclinations when they invite trans-
gression, and this disposition is the true enemy’ (ibid., note).
78
GD 20:297 (p. 160): ‘In grace … the mind governs with liberality, because it is the mind
that sets nature in action here, and finds no opposition to quell’. Cf. also Schiller’s
account of ‘the late Duke Leopold of Brunswick, standing upon the banks of the
raging waters of the Oder, [who] asked himself if at the peril of his life he ought to ven-
ture into the impetuous flood in order to save some unfortunates who without his aid
were sure to perish’, where Schiller writes that while the Duke’s ‘natural instinct’ for
self-preservation counselled otherwise, nonetheless once he had determined how he
should behave using his reason, his ‘susceptibility for the beautiful’ meant that ‘when
reason gave the order, the feelings would place themselves on the same side, and he
134 Understanding Mor al Obligation

And yet, given the complexities in his engagement with Kant, and
his insistence that we treat this sort of agent as perhaps no more
than an ideal anyway, Kant can scarcely be criticised for not seeing
in Schiller the resources with which this sort of challenge to his posi-
tion might be mounted, particularly when (as we have seen) Schiller
himself did so much to emphasise his own desire to be accepted
within the Kantian fold.79 Nonetheless, whatever the final truth con-
cerning where in the end Kant and Schiller stand in relation to one
another, the fact remains that to Schiller’s successors, and to Hegel in
particular, this dialectic between them served to highlight the kind
of dualism that Kant’s solution to the problem of moral obligation
seemed to require, and raised questions over whether this dualism
was an appropriate way of conceiving of the human condition and
our relation to the moral life. What Schiller made clear, therefore,
was that Kant’s account of the moral ‘must’ relies on taking this ten-
sion between reason and inclination seriously, as it is this tension
that forms the basis for Kant’s account of the kind of necessitation
that duty involves; and like Kant himself, Schiller showed how a form
of ethical engagement might be possible without this tension and
thus without this necessitation – but where Kant had firmly put this
possibility beyond us, Schiller had seemed to make it perhaps more
realisable in part, even while following Kantian orthodoxy in also
making its full realisation (as ‘the beautiful soul’) no more than an
ideal. Because of the uncertainty surrounding his differences from
Kant, however, Schiller therefore remains no more than a transi-

would do willingly that which without the inclination for the beautiful he would have
had to do contrary to inclination’ (MUAM 21:33–34 (pp. 97–8)). For further discus-
sion of Schiller’s position here, see Ives 1970.
79
This goal can be seen clearly in his response to Kant in a letter of 13th June 1794, in
which Schiller writes: ‘Only the vitality of my desire to make the results of the moral
doctrine established by you agreeable to a part of the public which seems to be fleeing
from them even now, and the eager wish to reconcile a not unworthy part of humanity
with the strictness of your system, could make me appear for a moment to be your adver-
sary, a part for which I am in fact very little suited, and which I am still less inclined to
play’ (27:13). Cf. also AE 1st Letter, §3, 20:309–10 (pp. 3,5), where Schiller addresses
Prince von Augustenburg in the first letter by telling him that ‘it is for the most part
Kantian principles on which the following theses are based’, and that ‘[c]oncerning the
ideas which prevail in the Practical part of the Kantian system only the philosophers
are at variance; the rest of mankind, I believe I can show, have always been agreed’ – all
that makes these ideas seem problematic is their ‘technical form’. However, as Beiser
notes (Beiser 2005: 183), in letters to others Schiller places himself at a greater distance
from Kant.
 Hegel’s critique of Kant (via Schiller) 135

tional figure,80 taking us to the considerably less equivocal position

to be found in Hegel.

From Schiller to Hegel

Although Schiller is not accorded by Hegel the kind of detailed discus-
sion that he offers other philosophers of the period in his Lectures on the
History of Philosophy and elsewhere, there is nonetheless little doubting
his influence on Hegel’s thought. Schiller (in this respect like Goethe)
is frequently mentioned, if not given any lengthy treatment, and invari-
ably referred to in a very positive manner. Hegel’s attitude is perhaps
best summarised in his Introduction to the Lectures on Aesthetics, where he
testifies forcefully to the power and impact of Schiller’s attempt to resolve
the dualisms of the modern world: ‘This unity of universal and particu-
lar, freedom and necessity, spirit and nature, which Schiller grasped
scientifically as the principle and essence of art and which he laboured
unremittingly to call into actual life by art and aesthetic education, has
now, as the Idea itself, been made the principle of knowledge and exist-
ence, and the Idea has become recognized as that which alone is true
and actual’.81 Immediately preceding this statement, Hegel has summa-
rised his understanding of the Letters on Aesthetic Education as involving
the attempt to reconcile reason and nature in a way that leads to beauty;
and this (Hegel notes) also reflects Schiller’s position in the earlier essay
On Grace and Dignity, where he remarks in this context of Schiller’s praise
there of women that occurs towards the end of the essay:
Now in the conflict of these opposite sides [viz. reason and nature],
aesthetic education is precisely to actualize the demand for their medi-
ation and reconciliation, since, according to Schiller, it proceeds by
so developing inclination, sensuousness, impulse, and heart that they
become rational in themselves; and in this way reason too, freedom,
and spirituality emerge from their abstraction and, united with the nat-
ural element, now rationalized, acquire flesh and blood in it. The beau-
tiful is thus pronounced to be the mutual formation of the rational and
the sensuous, and this formation to be the genuinely actual.82

80
For a discussion of Schiller that emphasises how his commitment to an essential
Kantian framework made it hard for him to develop his insights in a consistent man-
ner, see Henrich 1982, esp. pp. 250–4.
81
LA 13:91 (i, pp. 62–3).
82
LA 13:91 (i, p. 62). Cf. also LA 13:17 (i, p. 4): ‘[Art] has frequently been recommended
as a mediator between reason and sense, between inclination and duty, as a reconciler
136 Understanding Mor al Obligation

This, we have seen, provides an accurate enough summary of Schiller’s

position on what the ideal of beauty might amount to, albeit trans-
posed somewhat into more Hegelian language. It is notable in view of
our subsequent discussion, however, that there is no mention here of
dignity or of the sublime.
When it comes to Hegel’s critical account of Kant’s treatment of
obligation, and how this rests on a dualism of reason and inclination,
it is therefore no surprise that several aspects of that response relate
back to Schiller. Five of these aspects can be highlighted: the con-
trast between the ethics of virtue and an ethics of duty; the contrast
between ancient and modern ethics; the significance of love; the rela-
tion between this dualism and the social and political order in which
the individual lives; and the suggestion that because of this dualism,
Kant’s notion of autonomy is limited and inadequate. I will discuss
each of these aspects in turn.

From an ethics of duty to an ethics of virtue. As we have seen, an important

element in Schiller’s attempt to ‘temper’ the impression of harshness
and dualism in Kant’s ethics was to turn from considerations of
duty to considerations of virtue, in a way that related back to a more
Aristotelian way of viewing ethical life. Schiller thus ushered in a
debate over whether Kant’s picture is to be preferred over Aristotle’s,
and indeed over whether in the end there is such a fundamental
difference between them, where both debates continue to this day.83
Hegel follows Schiller in introducing a contrast between ‘binding
duty’ and virtue, where the former ‘appear[s] as a limitation … to the
drives of the natural will’84 in a way that the latter does not, for virtue
is ‘[t]he ethical, in so far as it is reflected in the naturally determined
character of the individual as such’.85 Much like Schiller, therefore,

of these colliding elements in their grim strife and opposition’. Hegel notes that in
response, ‘it may be maintained that in the case of these aims of art … nothing is
gained for reason and duty by this attempt at mediation, because by their very nature
reason and duty permit no mixture with anything else; they could not enter into such
a transaction, and they demand the same purity which they have in themselves’. Hegel
finally deals with this objection prior to his discussion of Schiller: see LA 13:80–82 (i,
pp. 53–5).
83
See, for example, Sherman 1997.
84
PR §149, 7:297 (p. 192). For a helpful discussion of Hegel’s account of virtue, see
Buchwalter 1992.
85
PR §150, 7:298 (p. 193).
 Hegel’s critique of Kant (via Schiller) 137

Hegel turns to Aristotle to find an alternative to Kant’s dualism of rea-

son and inclination:
So here we shall introduce the principal definitions pertinent to the
will and to the concept of virtue [in Aristotle]. In defining the con-
cept of virtue from a practical standpoint Aristotle distinguishes the
soul’s rational side from its irrational side. To the rational side belong
judiciousness, prudence, knowledge, and wisdom in general. The other
or irrational side encompasses sensation, inclination, and passion. And
virtue consists in the unity of the rational side with the irrational side.
When the inclinations, passions, and the like are so disposed towards
λóγος or reason as to do what it commands, then we have virtue.86

This alignment of reason and inclination in the virtuous person does

not mean, Hegel points out, that such a person should just follow their
inclinations without their reason playing a role; if this happens, and
the individual does a good thing, the best they can be called is good
hearted, not virtuous, because ‘there is no virtue if insight is wanting
[or] the λóγος is inferior, because logos is requisite for virtue’.87 On the
other hand, Hegel argues, while ‘Socrates locates virtue simply in know-
ledge’, for Aristotle ‘goodness requires the presence of an irrational
drive even though reason judges and determines the drive’.88
Hegel also accords with Schiller in accepting an Aristotelian notion
of habituation and second nature to go alongside this account of vir-
tue. This allows for individuals to overcome their original, non-moral,
inclinations and dispositions, and to replace them with ones that are
moralised and so in accord with reason, in a way that is required on
this model of the virtuous individual. Thus, Hegel declares, ‘Education
[Pädagogik] is the art of making human beings ethical: it considers

86
LHPii Hegel 1989: 92 (p. 255). Cf. also SC i: 326 (p. 214): ‘This expanded content
[which Jesus identifies with righteousness] we may call an inclination so to act as the
laws may command, i.e., a unification of inclination with the law whereby the latter loses
its form as law [my emphasis] … Similarly, the inclination [to act as the laws may com-
mand], a virtue, is a synthesis in which the law (which, because it is universal, Kant
always calls something “objective”) loses its universality and the subject its particu-
larity; both lose their opposition, while in the Kantian conception this opposition
remains, and the universal becomes the master and the particular the mastered. The
correspondence of inclination with law is such that law and inclination are no longer
different; and the expression “correspondence of inclination with the law” is therefore
wholly unsatisfactory because it implies that law and inclination are still particulars,
still opposites’.
87
LHPii Hegel 1989: 92 (p. 255). 88 LHPii Hegel 1989: 92 (p. 255–6).
138 Understanding Mor al Obligation

them as natural beings and shows them how they can be reborn, and
how their original nature can be transformed into a second, spiritual
nature so that this spirituality becomes habitual to them’.89 Along with
Schiller, therefore, Hegel rejects the idea that we are good from birth
and so from nature in that sense;90 on the other hand, he thinks that
our natural selves can be developed to become good, at first under the
tutelage of reason and then in conformity with it.91 Hegel thus adopts
a strongly anti-Rousseauian view of the value and methods of educa-
tion, claiming that ‘[o]ne of the chief moments in a child’s upbringing
is discipline, the purpose of which is to bring the child’s self-will in
order to eradicate the merely sensuous and natural’;92 only in this way,
Hegel thinks, can the individual arrive at the second nature needed for
virtuous behaviour.
Hegel thus follows Schiller in accepting that Kant’s emphasis on
duty is an important corrective to those who hold that we are some-
how born with good dispositions are so are moral by nature in this
sense. On the other hand, he does not see Aristotle’s conception of
virtue as making this mistake, while it in its turn acts as a corrective to

89
PR §151Z, 7:302 (p. 195). Cf. AE 4th Letter, §3, 20:317 (p. 19): ‘Hence it will always
argue a still defective education if the moral character is able to assert itself only by
sacrificing the natural’.
90
For Schiller, cf. AE 3rd Letter, §5, 20:315 (p. 15), where Schiller characterises man’s
‘natural character’ as ‘selfish and violent’; and AE 24th Letter, §§2–7, 20:389–93
(pp. 173–9), where Schiller characterises man’s natural state in amoral terms:
‘Unacquainted as he is with his own human dignity, he is far from respecting it in
others; and, conscious of his own savage greed, he fears it in every creature which
resembles him’ (AE 24th Letter, §2, 20:389 (p. 173)). For Hegel, cf. PR §151, 7:301
(p. 195): ‘But if it is simply identical with the actuality of individuals, the ethical [das
Sittliche], as their general mode of behaviour, appears as a custom [Sitte]; and the habit
of the ethical appears as a second nature which takes the place of the original and
purely natural will and is the all-pervading soul, significance, and actuality of individ-
ual existence [Dasein]’; and PR §18Z 7:69 (p. 51): ‘The Christian doctrine that man is
by nature evil is superior to the other according to which he is good. Interpreted philo-
sophically, this doctrine should be understood as follows. As spirit, man is a free being
[Wesen] who is in a position not to let himself be determined by natural drives. When
he exists in an immediate and uncivilized [ungebildeten] condition, he is therefore in a
situation in which he ought not to be, and from which he must liberate himself. This is
the meaning of the doctrine of original sin, without which Christianity would not be
the religion of freedom’.
91
Cf. LHP 19:223 (ii, p. 205): ‘[W]hen a beginning from virtue has been made, it does
not necessarily follow that the passions are in accordance, since often enough they are
quite the reverse … But then the subject must, in this separation of his activity, bring
likewise his passions under the subjection of the universal, and this unity, in which the
rational is pre-eminent, is virtue’.
92
PR §174Z, 7:327 (p. 211). Cf. also PR §20, 7:71 (p. 52) and §187, 7:343–5 (pp. 224–6).
 Hegel’s critique of Kant (via Schiller) 139

the Kantian morality of duty, which always sees a tension between our
rational and natural capacities, even in the well brought-up and hence
virtuous individual. As a result, he takes Aristotle to have arrived at
the best account:
This is the correct definition. On the one side it involves suppression
of the passions and on the other side it is directed against the ideal of
virtue and also against the view that the inclinations, drives, and so
forth are intrinsically good. Both these extreme views have [neverthe-
less] been put forward often enough in recent times. So, what is simply
natural is no virtue, [even though] there is a saying that ‘a human being
who is fine and noble by nature is nobler than duty and far above it’.
On the other side we find the view that duty should be done ‘as duty’,
without taking other factors into account and without defining the par-
ticular as a moment of the whole.93

Hegel thus refuses to set our ‘drives’ and our ‘determinations of rea-
son’ against one another; rather, the goal is to bring them together
within a whole, an aim that is stated clearly in the Introduction to his
Philosophy of Right.94

Ancient vs modern ethics. In Schiller, as we have discussed, this turn to

an ethics of virtue is linked with a broader contrast between Greek
and modern thought, where the dualism characteristic of the latter
was seen as allowing an ethics of duty to emerge and to eclipse the
Hellenic focus on virtue. Schiller, of course, was by no means the only
one during this period to contrast the Greeks to ‘us Moderns’ in this
way; but again, in leading him to adopt a similar view, Schiller was
clearly an influence on Hegel, as were others (such as Wincklemann,
Schelling and Hölderlin).95 From Schiller in particular, Hegel learnt
to see Kant as a representative of a distinctively modern turn, to which
classical Greek thought represented a significant alternative, thanks
to its more harmonious conception of the individual and their place

93
LHPii Hegel 1989: 93 (p. 256).
94
Cf. PR §19 7:70 (p. 51): ‘Underlying the demand for the purification of the drives is the
general idea [Vorstellung] that they should be freed from the form of their immediate
natural determinacy and from the subjectivity and contingency of their content, and
restored to their substantial essence. The truth behind this indeterminate demand is
that the drives should become the rational system of the will’s determination; to grasp
them thus in terms of the concept is the content of the science of right’.
95
Cf. Pinkard 2000: 32–3. For more general discussion, see also White 2002: 10–16.
140 Understanding Mor al Obligation

in the world. However, because of Schiller’s remaining admiration for

Kant, Hegel also learnt from him to see the Greeks as possessing this
harmony in an ‘immediate’ way, as a result of a kind of intellectual
and social innocence and immaturity, so that the idea of returning
to this lost golden age could only ever be a superficial solution to the
problems we face in overcoming the dichotomies of modernity.
Schiller’s attitude is nicely reflected in a passage from his marginal
notes to an essay by Wilhelm von Humboldt, an attitude that is equally
recognisable in Hegel:
Are not those three stages which we can distinguish in all empirical
knowledge likely to hold approximately for the general development of
human culture?
1. The object stands before us as a whole, but confused and fluid.
2. We separate particular characteristics and distinguish; our know-
ledge is now distinct, but isolated and limited.
3. We unite what we have separated, and the whole stands before us
again, no longer confused, however, but illuminated from all sides.
The Greeks found themselves in the first of these three phases. We find
ourselves in the second. The third, therefore, we may still hope for, and
when it comes we shall no longer yearn for the Greeks to return.96

This may be directly compared to Hegel’s own account of the Greeks,

according to which sources of the kind of disharmony we experience
in the modern world had not yet emerged, but having done so, they
can no longer go unrecognised and unacknowledged. Thus, as much
as we might nostalgically admire it, there is no going back to this lost
innocence, so that instead harmony must be achieved in a different
way, in which the pressures that lead to its breakdown are properly
appreciated and accommodated within a unity that also incorporates
difference within it.97

Schiller (1943–) 21:63; cited in AE, p. 234.

96

Cf. DFS 2:20–22 (pp. 89–91), where Hegel claims on the one hand that ‘[d]ichotomy is
97

the source of the need of philosophy’, and that the ‘sole interest of Reason’ is to suspend
such rigid antitheses’ as those between ‘Reason and sensibility, intelligence and nature
and … absolute subjectivity and absolute objectivity’; on the other hand, he insists that
this does not mean that ‘Reason is altogether opposed to opposition and limitation’,
‘[f]or the necessary dichotomy is One factor in life’ – ‘[w]hat Reason opposes, rather,
is just the absolute fixity [my emphasis] which the intellect gives to the dichotomy’. This
characteristic Hegelian search for ‘unity-in-difference’ and its associated attitude to
the Greeks may therefore be traced back to Schiller, therefore (although again, not
Schiller alone).
 Hegel’s critique of Kant (via Schiller) 141

As we shall see in the next chapter, the way in which the ancient
Greeks might be used to provide an alternative model for ethics,
from which a different conception of duty and obligation might be
recovered, plays a significant role in Hegel’s thinking, just as it did in
Schiller’s – as also did the idea that this recovery could not be direct or
immediate. We shall also later see whether the fact that their picture of
the Greeks was somewhat mythical means that the non-Kantian alter-
native that Hegel bases upon it is fatally flawed as a result.

Love vs law. Another important theme in Hegel that has its roots in
his reading of Schiller (amongst others) is the focus on love as an
important feature of human relations, in a way that again casts doubt
on and contrasts with the more Kantian outlook.
As has been discussed, Schiller gives love an important role in
On Grace and Dignity, while it had also figured in his earlier writings.
Schiller was fully aware of the complexities and dangers of love in the
human context, and how it might easily become corrupted into ego-
ism and mere sensuousness; hence his insistence that it be balanced
with respect. Nonetheless, by providing us with a model of grace, love
rightly understood shows how an individual can relate to the beloved
in a way that properly respects their needs and interests, without the
lover being required to suppress their contrary inclinations.98 In this
sense of duty and obligation, then, it is inappropriate to introduce
such notions into the relation of love.
While Schiller retains his allegiance to Kant in nonetheless treat-
ing this model with caution and warning of its dangers,99 in his early
writings and particularly in ‘The Spirit of Christianity’, Hegel wanted
to take it further, and used love to offer an alternative to the law

98
Cf. NLBF 21:24 (p. 177): ‘Of all the inclinations that are decided from the feeling
for the beautiful and are special to refined minds, none commends itself so much
to the moral sense as the ennobled instinct of love; none is so fruitful in impressions
which correspond to the true dignity of man. To what an elevation does it raise human
nature! and often what divine sparks does it kindle in the common soul! It is a sacred fire
that consumes every egoistical inclination [my emphasis], and the very principles of moral-
ity are scarcely a greater safeguard of the soul’s chastity than love is for the nobility of
the heart’.
99
Cf. NLBF 20:24–5 (pp. 177–8): ‘But do not follow this guide [of love] until you have
found a better … [T]he passion of love sets [up sophisms] against conscience (whose
voice thwarts its interests), making its utterances despicable as suggestions of selfish-
ness, and representing our moral dignity as one of the components of our happiness
that we are free to alienate’.
142 Understanding Mor al Obligation

conception of ethics, treating Jesus as its spokesman in the Sermon on

the Mount.100 Hegel thus treats love as a way of reconciling the oppos-
ition between morality and inclination, such that the former loses its
commanding force:
The command ‘Thou shalt not kill’ is a maxim which is recognized as
valid for the will of every rational being and which can be valid as a
principle of a universal legislation. Against such a command Jesus sets
the higher genius of reconcilability [Versöhnlichkeit] (a modification
of love) which not only does not act counter to this law but makes it
wholly superfluous; it has in itself a so much richer, more living, fullness
that so poor a thing as a law is nothing for it at all. In reconcilability the
law loses its form, the concept is displaced by life.101

For Hegel, therefore, it is possible to find in the love of people for one
another that is preached by Jesus ‘this extinction of law and duty’.102
Hegel’s concern with love as a feature of our moral lives continues
into his mature writings, including the Philosophy of Right; however,
in this work he is more wary of its potentially problematic nature,
and so is closer to Schiller in this respect. This can be seen clearly
in his treatment of the family in the first section of his discussion of
Ethical Life. While according it some value, here Hegel notes ‘the
transient, capricious, and purely subjective aspects of love’, which fol-
lows from the fact that ‘love’, as a feeling [Empfindung] is open in all
respects to contingency’.103 Thus, Hegel emphasises, while marriage
should not be thought of as a purely contractual relation in a Kantian

100
Cf. Beiser 2005: 84: ‘But we must beware of romantic intoxication. Schiller’s concept
of love has been misinterpreted as much as the beautiful soul. It has been taken as evi-
dence for Schiller’s adherence to an evangelical or Pauline ethic, an ethic that replaces
the moral law with love. This was indeed how the young Hegel interpreted it in his
Geist der Christenthums. But the mature Schiller never had any such programme’.
101
SC 1:327 (p. 215).
102
SC 1:336 (p. 223). Cf. also SC 1:325 (p. 213), where Hegel agrees with Kant that ‘in love
all thought of duties vanishes’, but not because (as Kant thinks) love is a feeling and
so cannot be commanded (cf. MM 6:401 (p. 530)), but because (as Kant also thinks),
‘duties require an opposition, and an action that we like to do requires none’, where
this opposition is the case of an action done out of love. Cf. a cancelled passage from
SC: ‘A command can express no more than an ought or shall, because it is a univer-
sal, but it does not express an “is”; and this at once makes plain its deficiency. Against
such a command Jesus set virtue, i.e., a loving disposition, which makes the content
of the command superfluous and destroys its form as a command, because that form
implies an opposition between a commander and something resisting the command’
(SC 1:327 note (p. 215)).
103
PR §161Z, 7:310 (p. 201).
 Hegel’s critique of Kant (via Schiller) 143

manner, it also cannot be equated just with love, if it is going to have

the kind of stability and objectivity required. Here, clearly, Hegel ech-
oes Schiller’s concerns about erotic love and desire. Nonetheless, a
more ethical notion of love also emerges within marriage, which is
less about feeling and closer to ‘trust, and the sharing of the whole of
individual existence [Existenz]’.104 Thus, just as Schiller saw the need
to balance love with respect, so too Hegel sees the need to balance
love as desire with the recognition that one’s relation to the other
goes beyond this feeling, whilst at the same time this prevents the
relation being more akin to continence than virtue; if it becomes
so, ‘the moment of feeling’ has been lost, and the marriage is over,
as each takes themselves to be acting against their inclinations and
hence bound to the other simply by duty. Hegel thus introduces a
complexity into his understanding of love that is not present in his
early writings, akin to Schiller’s own insistence on the need to bal-
ance love with respect. Nonetheless, taken in this ‘higher’ form,105
love should enable those who have this relation to feel that they are
not constrained in their behaviour towards the beloved, in just the
way that Hegel had argued in his earlier work.

The aesthetic state and the rational state. A further important aspect
in which Schiller may be taken to have influenced Hegel is in the
former’s conception of the ‘aesthetic state’, which, as we have seen, was
developed in the Aesthetic Letters. Here, Schiller presented an attempt
to overcome the Kantian dualism between ‘reason and sensuousness,
duty and inclination’ in social terms, based on his vision of how society
might in the end operate, in such a way as to bring this dualism to an
end. Moreover, he draws a contrast between the aesthetic state and ‘the
ethical state of duties’, in which ‘Man sets himself over against man with
all the majesty of the law, and puts a curb upon his desires’.106 By making
this as much a political and social issue as a personal one, Schiller
added a new dimension to the debate. Whilst in the end Schiller’s
claims about the achievability of such a state remained modest at best,

104
PR §163, 7:313 (p. 202).
105
Cf. PR §164, 7:316 (p. 205), where Hegel speaks of a love that has an ‘ethical charac-
ter’, as involving ‘that higher suppression and subordination of mere natural drive
which is already naturally present in shame and which the more determinate spiritual
consciousness raises to chastity and purity [Zucht]’.
106
AE 27th Letter, §9, 20:410 (p. 215).
144 Understanding Mor al Obligation

by taking the matter in this direction at all, Schiller arguably opened

up a way of thinking about it that diverged from Kant’s.
At the same time, because in the end Schiller’s hopes for the
aesthetic state are limited, this perhaps contributed to Hegel’s
unwillingness to follow this social theory in its aesthetic dimension,
according to which art and aesthetic education might in themselves
enable individuals to get beyond the dualisms that Schiller had iden-
tified in the modern world. To this extent, Hegel’s appeal to the
rational state instead serves as a departure from Schiller’s position.
Nonetheless, in so far as Hegel’s hopes for the rational state closely
resemble Schiller’s hopes for the aesthetic one, the latter’s influence
is clear and significant.107

The limits of Kantian autonomy. A final and perhaps most important

element in Schiller’s reception of Kant, was his suggestion that Kant’s
dualistic picture did not itself allow for full autonomy, even though
the intention of his ethics was to avoid the heteronomy of other
moral theories. As Schiller put this concern in one of the passages
from his Kallias Letters to Körner that we have cited already: ‘When
the form of the non-reasonable [nicht-vernunftig] is determined by
reason (theoretical or practical, both are the same here), its natural
determination is constrained, and beauty cannot arise’.108 He puts the
point equally clearly in a later letter I have also cited: ‘We never want
to see coercion, even if it is reason itself which exercises it; we want
even nature’s freedom to be respected because “we regard every being
in aesthetic judgement as an end in itself”, and it disgusts (outrages)
us, for whom freedom is the highest thing, that something should be
sacrificed for something else, and used as a means’.109
For Hegel, too, the concern was that by retaining the dualism
between reason and inclination, Kant did not really allow for the real-
isation of true autonomy, despite his championing of freedom. Hegel
puts the point forcefully in ‘The Spirit of Christianity’, turning against
him the case Kant himself had made against various forms of authori-
tarian religious life:

107
For a discussion of the history of the idea of the aesthetic state and its ‘Aufhebung’ in
Hegel, see Chytry 1989, especially pp. 72–105 and 178–218.
108
KL 26:195 (p. 156). 109
KL 26:198 (p. 159).
 Hegel’s critique of Kant (via Schiller) 145

[B]etween the Shaman of the Tungus, the European who rules the
church and state, the Voguls, and the Puritan on the one hand,110 and
the man who listens to his own command of duty, on the other, the
difference is not that the former make themselves slaves, while the lat-
ter is free, but that the former have their lord outside themselves, while
the latter carries his lord in himself, yet at the same time is his own
slave. For the particular – impulses, inclinations, pathological love, sen-
suous experience, or whatever else it is called – the universal is neces-
sarily and always something alien and objective.111

He then goes on to contrast the Kantian picture with what he takes to

be the paradigm offered by Jesus in the Sermon on the Mount:
To complete subjection under the law of an alien Lord, Jesus opposed
not a partial subjection under a law of one’s own, the self-coercion of
Kantian virtue, but virtues without lordship and without submission,
i.e., virtues as manifestations of love.112

It is clear that for Hegel in this early period, Kant’s account of the obli-
gations of morality stood in the way of his providing a fully adequate
picture of autonomy, where obedience had just been transferred from
without to within.
In the Philosophy of Right, this theme continues in the opening sec-
tion of ‘Ethical Life’, where Hegel makes the striking claim that ‘[t]he
individual … finds his liberation in duty’.113 At first, it may appear that
Hegel intends this claim in a Kantian manner, where the freedom
in question consists in the individual escaping from his phenomenal
self, so that he is ‘liberated from his dependence on mere natural
drives’.114 However, Hegel also says that on his conception of duty,
the individual is also freed from ‘the burden he labours under as
a particular subject in his moral reflections on what one should do
and what one would like to do’,115 suggesting that the tension between
these latter two is overcome in a way that is also necessary for free-
dom. This is confirmed by the way in which Hegel then moves on to
a discussion of virtue, in the way that we have discussed, whereby the
tension between sensuousness and reason can be resolved, and true
autonomy achieved.

110
Kant brought up these as examples in Relig 6:176 (p. 195).
111
SC 1:323 (p. 211). 112
SC 1:359–60 (p. 244).
113
PR §149, 7:297–8 (p. 192). 114
PR §149, 7:298 (p. 192).
115
PR §149, 7:298 (p. 192), translation modified.
146 Understanding Mor al Obligation

Going beyond Schiller. We have seen, then, how there is a considerable

overlap between the outlooks of Schiller and Hegel, both in the areas
where they acknowledge Kant’s achievements, and where they identify
his limitations. In many respects, therefore, Hegel’s critique of Kant
may be thought of in Schillerian terms.
However, Schiller also remains a transitional figure in this
debate, because he holds back where Hegel wants to go further.116
Fundamentally, I believe, this is because Schiller retains an attach-
ment to the Platonistic sublime, as something lying beyond and outside
nature conceived of as an appearance, which corresponds to Kant’s
conception of the transcendental self that sets duty above desire.
Hegel, by contrast, works to overcome such transcendence, and looks
for a more unified and immanent picture, in a way that is influential
not only in his ethics, but in his metaphysics more broadly.117
As a result, as we have seen, Schiller continues to agree with Kant
that a full harmony between our reason and our sensible natures is an
unattainable ideal, and thus that Kant’s account of duty and obligation
as arising from this tension between them remains a viable picture for
Schiller, while duty conceived of in these terms cannot be removed
from our ethical life as he sees it. Having thus offered us a model of
virtue, grace, and beauty as an alternative, Schiller in the end agrees
with Kant that this is not an alternative that we can ever fully realise,
so that his allegiance to Kant remains intact.
As a consequence, therefore, there is for Schiller no pressure to
arrive at an account of notions like duty and obligation that do away
with the Kantian model, because Schiller retains the sort of dualism
which can still make sense of these notions in Kantian terms. Schiller
therefore never faces the need to offer a different solution to the prob-
lem of moral obligation. However, because of his more radical position,

116
Cf. White 2002: 30: ‘Hegel’s reaction to Kant was far more radical than Schiller’s.
Whereas Schiller acknowledged Kant’s dualism of reason and inclination but tried to
paper it over, Hegel attempted to demonstrate that it did not really exist at all. Schiller
had maintained that a human being would be best off it the recommendations of
reason and the urging of inclination were consistent, and that such a condition was
rationally to be sought. Hegel went further, by trying to argue that the dualism pre-
supposed by both Kant and Schiller was entirely misconceived’.
117
Cf. PR §§133–6, 7:250–54 (pp. 161–4), where Hegel both praises the ‘loftiness’ [der
hohe Standpukte] or ‘sublimity’ [erhaben] of Kant’s outlook, and gently mocks it – so
that while such talk ‘glorifies the human being and fills his heart with pride’, it can
also become ‘wearisome’, particularly when (as Hegel thinks) it gives us little help in
dealing with the realities of the world.
 Hegel’s critique of Kant (via Schiller) 147

in which Schiller’s Aristotelian picture of the virtuous individual is

treated as no mere ideal, Hegel is led to come up with an account of
obligation and duty that is different from Kant’s, as we shall now see.
In the next chapter, therefore, I shall examine how Hegel tried to over-
come the problem of moral obligation, having been convinced by the
implications of Schiller’s position that Kant’s hybrid solution involved
paying too great a price.
 5

H ege l’s s olu t ion to t h e probl e m

of mor a l obl ig at ion

In discussions of the transition from Kant to Hegel, it is common-

place to characterise it in terms of the move from the individual to the
social. In this chapter, I too will be following this pattern in consider-
ing Hegel’s solution to the problem of moral obligation.
What I earlier called the standard story concerning the history of
Kantian and post-Kantian ethics takes a similar trajectory, in offering a
social solution to the Kantian paradox, of the sort proposed by Pippin
and Pinkard: because in Hegel the legislating subject is a ‘we’ not an
‘I’, this is supposed to help resolve the problem of emptiness faced
by the constructivist account. But it is not clear how much this can
help. As one critic has put it: ‘there is no value in a system of mutual
constraint which harmonizes the various rational natures’ choosing
unless that choosing is itself valuable. For one valueless rational nature
to constrain itself out of deference to other valueless rational natures
is just one more version of arbitrary self-launching’.1 The problem thus
seems to be no different at the Hegelian social level than it is at the
Kantian individual one, so it is hard to see why, if the latter is deemed
problematic, the former should be deemed any less so: both seem
to be equally empty unless we recognise an antecedent background
of value which determines the way in which we take ourselves to be
constrained.
In fact, however, just as it can be claimed that the standard story
does not do justice to Kant’s value realism (as we saw in Chapter 1), so
the same can be claimed of Hegel, where again for Hegel as well as for
Kant, that value can be said to reside in our freedom, as the highest

1
Regan 2002: 287. I have discussed the difficulties with the ‘social solution’ to the prob-
lem of self-legislation further in Stern 2004b.

148
 Hegel’s social command account of obligation 149

form of self-actualisation. Thus, as well as defending a realist view of

Kant, Allen Wood likewise defends a comparable view of Hegel: ‘Like
Kantian ethics, [Hegel’s ethical theory] is based on the value of free-
dom or rational selfhood’.2 Hegel acknowledges that value at the out-
set of the Philosophy of Right, and so makes it the basis for his reflections
on social and political philosophy as well as ethics: ‘Right [Recht] is
something utterly sacred, for the simple reason that it is the putting into
being [Dasein] of the absolute concept, of self-conscious freedom’;3
and he comments later that ‘Right is concerned with freedom, the
worthiest and most sacred possession of man’.4
Now, of course, constructivist readers of Hegel have also noted the
significance that Hegel attaches to freedom; and, because they also
accept the argument from autonomy, they have gone on to claim that
this means that Hegel cannot be conceived of as a realist about value.5
However, we have given reasons previously to think that this argument
is less than cogent, and that questions about autonomy only become
pressing when considerations about obligation and duty arise, not value
as such. This, too, seems to be how Hegel conceives of the issue as it
figures in Kant’s ethics. For, he suggests, it was because Kant was one
of the first to see that ‘human freedom is the ultimate pivot upon

2
Wood 1990: 31. Hegel’s commitment to realism is also noted by Dudley Knowles:
see Knowles 2002: 215, where he refers to the long Remark to PR §140, 7:265–80
(pp. 170–83).
3
PR §30, 7:83 (p. 59; translation modified).
4
PR §215Z, 7:368 (p. 247). Cf. Kant LE 27:344 (p. 125): ‘Freedom is thus the inner worth
of the world’, and GMM 4:436 (p. 85): ‘Autonomy is therefore the ground of the dignity
of rational nature and of every rational nature’.
5
In this connection, Pippin is fond of quoting from EL §31Z, 8:98 (p. 69): ‘This pure
being at home with ourselves [Beisichsein] belongs to free thought, that voyages in the
open, where nothing is below us and nothing is above us, and we stand in solitude
with ourselves alone’, taking this to be Hegel’s view of modernity, as finding ourselves
in the position of needing to create our own values around us, where we cannot rely
on anything being there already – this constitutes our solitude, and also our freedom
(see e.g. Pippin 1991: 67). However, although it can be read in this way, the meaning of
this text is not altogether clear. For, while Hegel is certainly contrasting our position
to something like Scholastic metaphysics, which was not free because it ‘adopted its
content as something given, and indeed given by the Church’, it is less clear that he
is rejecting realism, in so far as he claims that the Greeks also ‘thought freely’, where
it seems that all that is necessary for this is to avoid any such appeal to authorities in
one’s speculations: ‘We must imagine the ancient philosophers as men who stand right
in the middle of sensory intuition, and presuppose nothing except the heavens above
and the earth beneath, since mythological representations have been thrown aside. In
this simple factual environment, thought is free and withdrawn into itself, free of all
[given] material, purely at home with itself’.
150 Understanding Mor al Obligation

which humanity turns’,6 that the notion of duty became problematic

for him, for it is qua duty that the free will apparently comes to be
bound.7 And we have already seen how Hegel, following Schiller, is dis-
satisfied with the account of duty and obligation that Kant then offers,
as arising from the dualistic nature of the human will.8
Thus, while in this chapter I will follow the standard story inasmuch
as I will treat the Hegelian move from the individual to the social as
crucial, I will depart from that story in not seeing this as an attempt
to resolve the Kantian paradox of self-legislation; rather, I will see it as
an attempt to offer an account of duty and obligation that avoids the
dualism of Kant’s hybrid view. For reasons that will become clear as we
proceed, we might therefore call Hegel’s position a ‘social account’.
Before turning to a discussion of Hegel himself, however, I would like
to give some context to that discussion by seeing how other theorists
have used social accounts to provide an understanding of duty and
obligation.

Social command accounts of moral obligation

In his paper ‘Radical Virtue Ethics’, Kurt Baier attempts to tackle the
claim made by some of the more extreme proponents of virtue the-
ory that there is a clear disjunction between an ethics of virtue and
a morality of duty and obligation, and that we should aim to replace
the latter with the former (rather than having it complement it, as sug-
gested by more moderate voices). One target Baier chooses is Elizabeth

6
LHPiii Hegel 1986: 168 (p. 190).
7
Cf. LHPiii Hegel 1986: 169 (p. 191).
8
Another aspect of Hegel’s critique of Kant’s dualism, which we cannot consider fur-
ther in detail here, is his objection to Kant’s treatment of immortality as one of the
‘postulates of pure practical reason’ (cf. CPrR 5:122 (p. 238)). Hegel interprets this as
an attempt by Kant to ‘soften’ this dualism, but in a way that is ultimately inconsistent:
‘Perfected morality must remain a Beyond [for Kant]; for morality [for him] presup-
poses the difference of the particular and universal will. It is a struggle, the deter-
mination of the sensuous by the universal; the struggle can only take place when the
sensuous will is not yet in conformity with the universal. The result is, therefore, that
the aim of the moral will is to be attained in infinite progress only; on this Kant founds
… the postulate of the immortality of the soul, as the endless progress of the subject
in his morality, because morality itself is incomplete, and must advance to infinity. The
particular will is certainly something other than the universal will; but it is not final or
really unchangeable’ (LHP 20: 269–70 (iii, pp. 461–2; translation modified)). Cf. also
Hegel, PS 3:446–7 and 457–60 (pp. 368–9 and pp. 377–80), and Stern 2002: 169–78
for further discussion.
 Hegel’s social command account of obligation 151

Anscombe and her 1958 article ‘Modern Moral Philosophy’. In that

article, Anscombe famously argues that as we can no longer take ser-
iously the idea that there might be a divine legislator, so we must give
up talking about moral duties and obligations, as such notions don’t
really make sense in the absence of the divine command model.9 As
we have already seen, Anscombe is scornful of attempts to replace this
model with the Kantian idea of self-legislation, while she is equally dis-
missive of other approaches that have been tried, such as appeals to
conscience, custom and contract.10 Instead, she argues, we should now
set out to do ethics without the notions of duty and obligation that are
central to our distinctively moral thinking, and instead return to some-
thing more like the virtue ethics of the ancient Greeks, which did not
rely on any such notions.
Now, many critics of Anscombe have questioned whether she is
right to contrast the outlook of the ancients and the moderns here,
and have argued instead that the Greeks also operated with the dis-
tinctively moral ought.11 Baier, however, takes a different line of attack,
and argues that where Anscombe goes wrong is in thinking that once
the idea of a divine legislator has been rejected, it then inevitably fol-
lows that the normative notions of duty and obligation must be given
up. For, he claims, these notions do not necessarily imply this idea;
instead, he writes:
The fact that there are these wrongs, duties, and obligations implies
merely that (i) there is adequate moral reason for me to do and no
adequate moral reason for me not to do, that is, ‘compelling’ moral rea-
son for me to do these things, and (ii) that there is an adequate reason
for society ‘to insist’, that is, set up suitable social pressures to ensure
that its members follow these compelling moral reasons (for, obviously,
the mere existence of such compelling moral reasons does not actually
compel people).12

It is point (ii) that I want to say is characteristic of social command

accounts of moral obligation and duty. But let me first say something
about point (i), which is also an important part of such an account.

9
See Anscombe 1958, especially pp. 5–6/1981, esp. p. 30–1.
10
Anscombe 1958: 13–15/1981: 37–8.
11
Cf. Louden 1992: 34–39; White 2002: especially Chapter 3; and Rosler 2005: espe-
cially Chapter 4, where all three give extensive references to other texts on the issue
(and where Brochard 1901 foreshadows Anscombe in a notable way).
12
Baier 1988: 128.
152 Understanding Mor al Obligation

What is notable in relation to (i) is that Baier here relies on a dis-

tinction between what is morally good or bad on the one hand, and
what is obligatory or required on the other, where he is appealing only
to the former in (i). We have seen a similar distinction at work in what
I called ‘intermediate’ positions in the natural law debate (such as
Suarez’s), and also in Kant, according to whom acts can be morally
good for the divine will without putting that will under any obligation.
Moreover, once moral goodness and moral obligatoriness are distin-
guished in this manner, the former is treated as a necessary condition
for the latter, so that in order for an act to be obligatory, what one is
required to do must be good, giving one an antecedent moral reason
to act in this way. This therefore means that on this account, I can-
not be under a moral obligation to count grains of sand or commit
genocide,13 as the necessary condition for obligatoriness stated in (i)
would not be met.
However, again in common with the ‘intermediate’ positions and
with Kant, Baier does not treat moral goodness as a sufficient condi-
tion for obligatoriness and duty: something further must be added to
what is morally good if it is to count as an obligation. As we have seen,
on divine command theories, what is said to be added is a directive to
act in this way from God, while on Kant’s hybrid theory, what is said
to be added is the constraint reason exercises here over desire. Baier’s
approach amounts to what I am calling a social command account, by
contrast, because he claims that what is added is the demand by soci-
ety to act in accordance with what one has moral reason to do.
Now, in a note to the passage we are discussing, Baier refers back
to John Stuart Mill, where Mill’s discussion of duty and obligation in
Utilitarianism has frequently been seen as an inspiration for this sort
of approach.14 There, Mill writes, ‘[w]e do not call anything wrong,
unless we mean to imply that a person ought to be punished in some
way or other for doing it; if not by law, by the opinion of his fellow-
creatures; if not by opinion, by the reproaches of his own conscience’.15

13
This seems to be Anscombe’s worry concerning views of this sort, where she considers
the idea that the ‘norms’ of society might be used to retain ‘a law conception without a
divine legislator’: ‘[J]ust as one cannot be impressed by Butler when one reflects what
conscience can tell people to do, so, I think, one cannot be impressed by this idea if
one reflects what the “norms” of a society can be like’ (Anscombe 1958: 13/1981: 37).
14
Others who have found Mill’s position valuable include also Gibbard 1990: especially
pp. 40–5; Adams 1999: esp. pp. 233–48; Skorupski 1999: 137–59; and Darwall 2006:
p. 27 and pp. 92–3. Cf. also the talk of ‘social pressure’ in Hart 1994: 86.
15
Mill 1969: Chapter v, p. 246.
 Hegel’s social command account of obligation 153

We may hold that there are strong non-moral or even moral grounds
for someone to do something, and so ‘dislike or despise him for not
doing’ it – so we might think there are strong aesthetic reasons to
appreciate the music of Bob Dylan, or strong moral reasons to avoid
eating meat, and so ‘dislike or despise’ those who do not do either
or both of these things; but we might also think that as things stand,
the kind of ‘social pressure’ to act in these ways does not exist, and
with good reason, so that individuals cannot be said to be under any
obligation or duty here. It is an interesting question, of course, why we
might think there is good reason not to have any such social pressure
in certain cases of moral goodness; but that need not detain us now. It
is also an interesting question whether it is actual ‘social pressure’ that
is needed to make something obligatory, or just sufficient reasons for
such pressure to obtain, whether or not it actually does;16 but again, I
do not want to dwell on this issue here. The crucial issue for our pur-
poses now is that both Mill and Baier take it that being morally good
is not sufficient for duty and obligation; it is also required that society
holds us responsible for so acting, and will apply some form of pun-
ishment or sanction to us if we do not, in a way that puts us under an
obligation to act in this manner.
The social command account may therefore be seen to share some-
thing with the ‘intermediate’ divine command account, which also
takes it that God must command that a morally good act be performed
before it is rendered obligatory; but these latter accounts take God to
be the source of the demand, whereas on social command accounts, the
claim is that society can act as this source, as a legitimate authority in its
own right. This is why Baier thinks that the social command account
can be employed when faith in a divine legislator has been lost.17 On
this view, therefore, what makes an act obligatory is that the command
to so act obtains, where this command involves sanctions that attach to
not so acting, and where society is taken to have the rightful authority
to impose such sanctions and thus to issue commands.

16
For arguments to the effect that the demand needs to be actual, see Adams 1999:
245–6.
17
Cf. also Darwall 2006 especially pp. 100–15, where he suggests that there is a concep-
tual pressure to move from the latter view to the former, from ‘morality as accountabil-
ity to God’ to ‘morality as equal accountability’ to fellow rational agents, thus turning
Anscombe’s position ‘on its head’. See also Darwall 2004. For some doubts about
Darwall’s argument here, and Darwall’s response, see Watson 2007: 40–6 and Darwall
2007: 65–9.
154 Understanding Mor al Obligation

It is worth emphasising two features of this position, in order to

forestall two important potential objections to it. The first concerns
how this account deals with the motivations of the moral agent: for,
it might be said, in so far as obligatoriness is tied here to sanctions
applied by society to certain actions, then the motivation to act mor-
ally is rendered problematic, as it would then appear to be based on
the fear of such sanctions being applied to the agent. However, in
response to this worry, the social command theorist can argue that
while the possibility of punishment is what renders an act obligatory,
this need not figure in the motivations of the agent who acknowledges
the obligations on them that then obtain – rather, such an agent may
still be motivated by simply recognising this obligatoriness and thus
that an action is her duty (just as someone can see that she would not
be legally obliged to pay her taxes unless some sanction was attached
to not doing so, while at the same time paying them out of a recogni-
tion that this is what the law requires, rather than a concern to avoid
such sanctions). A second issue relates to whether this social command
account can accommodate the autonomy of the agent; for, it might be
felt, on this account the community seems to exercise authority over
the individual in a way that renders them heteronomous. To this con-
cern, however, there are several possible replies, including an appeal
to the way in which authority is here exercised by a legitimate source,
and also by a community of which individuals themselves form a part.
More radically, as Stephen Darwall has suggested recently, it can be
argued that this moral community itself must be ‘understood as a
cooperative of mutually accountable free and rational beings’,18 and
thus not as a threat to that freedom.
However, a proponent of Anscombe’s position in ‘Modern Moral
Philosophy’ might reply that even if the social command account can
be made to work in this way, it doesn’t really touch a deeper point that
Anscombe’s article brings out, which is that in fact no such replace-
ment is really needed: to many, what Anscombe’s article shows in the

18
Darwall 2006: 244. Cf. also ibid., p. 35: ‘But isn’t acting on demands that others can
make of one heteronomy rather than autonomy, being governed by them rather than
by oneself? … The response to this objection is that when one decides [to act on] …
what the moral communities authoritatively demands, the second-personal perspec-
tive of a member of the moral community is as much one’s own as anyone else’s. One
demands the conduct of oneself from a point of view one shares as a free and rational
person’.
 Hegel’s social command account of obligation 155

case of the virtues is that ethics can be and has been conducted with-
out the notions of moral duty and obligation playing any fundamen-
tal role, in a way that arguably makes virtue ethics much superior to
the ‘peculiar institution’ of morality in which such notions have their
home.19 On this view, then, the question is why duty and obligation
need form part of our ethical thinking at all, even if Baier and the
social command account can succeed in showing how we might make
them intelligible in secular terms.
In her article ‘Moral Obligations and Social Commands’, Susan
Wolf provides an interesting answer to this question. Against the
Anscombian suggestion that we could ‘do without the concept of obli-
gation’, she argues that ‘[t]here are good reasons … for wanting our
moral and ethical framework to contain a distinction between the
obligatory and morally desirable’.20 She identifies two such reasons.
First, by treating some acts as obligatory, we are able to operate with
a higher level of expectation that people will do them, and so can
coordinate our social lives more effectively. Second, this distinction
also prevents society and individuals from becoming morally over-
­demanding, where we can insist on those types of moral action that we
treat as duties, but leave others to the discretion of individuals.21 When
it comes to obligation, therefore, Wolf concludes that ‘[w]e have rea-
son … to find some alternative to Anscombe’s proposal that we simply
do ethics without that concept’22 – where the alternative she proposes
is a position that retains the concept by offering an account of it that
treats it as a social command.
We have thus seen in a preliminary way how a social command
account of obligation might be made to work. We will now see how
this can provide a model for thinking about the Hegelian approach
to these issues.

19
Cf. Taylor 1988. 20
Wolf 2009: 347.
21
Although I will not consider the dialectic any further here, a virtue ethicist might
reply, perhaps, that any decent account of the virtues will have these elements already
built in – so that, for example, it is part of the notion of being a virtuous individual that
no extreme self-sacrifice is involved. I take it that Wolf might then respond, however,
by suggesting that either this constraint is based on some dubious notion of eudaimo-
nia, or that it relies on the very distinction between the moral and the obligatory that
she is trying to articulate.
22
Wolf 2009: 348.
156 Understanding Mor al Obligation

Sittlichkeit and duty

As these issues are often presented, it is assumed that if one moves (as
we have seen that Hegel does) from something like a Kantian view of
morality, to a position that claims to be more like that of the ancient
Greeks, then the result will be that notions such as duty and obliga-
tion will drop out of the picture. Thus, for example, Henry Sidgwick
famously draws a contrast between the Kantian ‘imperatival’ view of
morality and the Greek ‘attractive’ view as the two major options here,
where the former is marked by the dualism of reason and inclination,
and the latter is not: as a result, in a way that resembles Anscombe’s
position, Sidgwick holds that when it comes to ‘Greek thinkers’, then
‘[t]heir speculations can scarcely be understood by us unless with a
certain effort we throw the quasi-jural notions of modern ethics aside’,
and so no longer see them as asking questions like ‘What is Duty and
what are its grounds?’23 And when it comes to Hegel, therefore, it may
equally seem that his intentions in recovering the outlook of Greek
ethics must have also been to move from the ‘imperatival’ to the
‘attractive’, and thus to end up with an ethics in which notions like
duty and obligation have no real place.
This view of Hegel is simplistic, however, for although he rejects
Kant’s account of duty and obligation, these are not then notions that
he tries to abandon or do without: on the contrary, they are central to
the Philosophy of Right. He makes their significance clear immediately
prior to his discussion of Kant in the sub-section on ‘The Good and
Conscience’: ‘The essential character of the will for me is duty … In
doing my duty, I am with myself [bei mir selbst] and free. The merit and
exalted viewpoint of Kant’s moral philosophy are that it has empha-
sized this significance of duty’.24 Thus, instead, Hegel tries to retain
these notions, but to offer an account of them that goes against both
divine command theory and the hybrid theory put forward by Kant,
and to offer what I would claim is a social command theory instead, as
we shall now see.
Having criticised the Kantian theory of duty and obligation in the
‘Morality’ section of the Philosophy of Right,25 Hegel provides this social
23
Sidgwick 1981: 106. For a criticism of Sidgwick’s view of the Greeks here, see White
1992 and 2002.
24
PR §133Z, 7:250–1 (p. 161).
25
Cf. also SL 5:147–8 (p. 136): ‘Duty is an ought directed against the particular will,
against self-seeking desire and capricious interest and it is held up as an ought to the
will in so far as this has the capacity to isolate itself from the true … The philosophy of
 Hegel’s social command account of obligation 157

command account in the concluding ‘Ethical Life’ section. The latter

takes into consideration not only the individual will, but also the ‘laws
and institutions which have being in and for themselves’.26 As a result, the
individual can be seen to be part of an ‘ethical substance’ that con-
sists of ‘laws and powers’,27 where ‘these substantial determinations are
duties which are binding on the will of the individual’.28 Because of the
authority of these duties over the lives of individuals, and of the rela-
tive unimportance of individuals within the social order, it can appear
to them that the moral law has a divine origin, as it did in pre-modern
societies.29 But this is to neglect the social basis of these obligations,
and that while the social order is a substance to which individuals
relate as ‘accidents’, nonetheless these accidents are required by the
substance in order to be actual.30 Hegel makes clear, therefore, that he
sees divine command accounts of obligation as based on a picture of
our relation to the world that has been surpassed, where these obliga-
tions are now better accounted for as an aspect of our existence within
the social environment of ethical life.
Hegel also makes clear that these obligations arise not only through
positive law, but also through the unwritten laws of customs [Sitten].31
Between them, in an ethical community, these will lay down ‘what
someone must do and what the duties are which he has to fulfil in
order to be virtuous’.32 In fact, Hegel claims, to the extent that the indi-
vidual lives within a well-developed ethical community, virtue will be
less important than what he calls ‘rectitude’ [Rechtschaffenheit]: for, in
most cases, it should be apparent to the individual how they are called
upon to behave from within the established norms of that community,

Kant and Fichte sets up the ought as the highest point of the resolution of the contra-
dictions of Reason; but the truth is that the ought is only the standpoint which clings
to finitude and thus to contradiction’.
26
PR §144, 7:294 (p. 189). 27
PR §146, 7:295 (p. 190).
28
PR §148, 7:296–7 (p. 191).
29
Hegel makes two references to divine command theories in this respect at the start of
‘Ethical Life’. Cf. PR §144Z, 7:294 (p. 189): ‘If we consider ethical life from the objec-
tive point of view, we may say that ethical man is unconscious of himself. In this sense,
Antigone proclaims that no one knows where the laws come from: they are eternal’;
and PR §145Z, 7:294 (p. 190): ‘Whether the individual exists or not is a matter of
indifference to objective ethical life, which alone has permanence and is the power by
which the lives of individuals are governed. Ethical life has therefore been represented
to nations as eternal justice, or as gods who have being in and for themselves, and in
relation to whom the vain pursuits of individuals are merely a play of the waves’. Cf.
also PS 3:321–2 (pp. 261–2). 30 EL §151, 8:294–7 (pp. 225–6).
31
PR §151, 7:301 (p. 195). 32
PR §150, 7:298 (p. 193).
158 Understanding Mor al Obligation

without any need for the kind of practical deliberation required in

the exercise of virtue.33 Hegel is not saying here that such deliberation
is never required, or that the individual ought not to ‘think for them-
selves’ in ethical matters; he is just saying that a properly structured
system of ethical duties should make such ‘hard cases’ less common,
while warning that the appeal to individual virtue can become a hypo-
critical smokescreen for avoiding one’s genuine obligations in moral
matters.34
As a result of the ‘laws and powers’ of the community, therefore,
the individual will find duties that are ‘prescribed, expressly stated,
and known to him within his situation’.35 These ethical laws may then
appear to have ‘an absolute authority and power, infinitely more firmly
based than the being of nature’.36 At the same time, however, Hegel
argues that in so far as they stem from the ethical community, such
laws are ‘not something alien to the subject’ but something to which
‘the subject bears spiritual witness … as to its own essence’.37 He sug-
gests, therefore, that while the motivations for obeying the duties that
the ethical substance lays down may start with ‘certain particular ends,
interests, and considerations, with hope or fear, or with historical
presuppositions’,38 nonetheless this would put the individual in a too
‘external’ relation to that ethical substance, which must be less instru-
mental, and more like a ‘relationless identity – in which the ethical is
the actual principle [Lebendigkeit] of self-consciousness’.39 We should
not think, therefore, that just because something is an obligation
because it is required by the social group, that the motivating reason

33
Cf. PR §150, 7:299 (pp. 193–4): ‘Within a given ethical order whose relations are fully
developed and actualized, virtue in the proper sense has its place and actuality only in
extraordinary circumstances, or where the above relations come into collision …
This is why the form of virtue as such appears more frequently in uncivilized societies
and communities, for in such cases, the ethical and its actualization depend more
on individual discretion and on the distinctive natural genius of individuals. In this
way, the ancients ascribed virtue to Hercules in particular. And since, in the states of
antiquity, ethical life had not yet evolved into this free system of self-sufficient devel-
opment and objectivity, this deficiency had to be made good by the distinctive genius
of individuals’. In his discussion of Hegel’s concept of virtue, Buchwalter points to
some passages outside the Philosophy of Right, mainly in Hegel’s early writings, where
he expresses reservations about rectitude (see Buchwalter 1992: 558–61); but I find it
hard to see any such reservations in the text of the Philosophy of Right itself.
34
Cf. Hegel’s discussion of ‘Virtue and the Way of the World’ in PS 3:283–91
(pp. 228–35).
35
PR §150, 7:298 (p. 193). 36 PR §146, 7:295 (p. 190).
37
PR §147, 7:295 (p. 191). 38
PR §147, 7:296 (p. 191).
39
PR §147, 7:295 (p. 191).
 Hegel’s social command account of obligation 159

the individual has for complying with it comes from these external
ends: rather, it can be based on the recognition of it as an obligation,
grounded in the acknowledged authority of the ethical community
over the individual, where at the same time the individual is part of
this group, and so not subordinated to it as by an alien will.
Now, of course, as we have seen, if a social command account of
this kind is going to be plausible, it can only treat what is required by
society as a necessary condition for creating a moral obligation; for, if
it were to also treat it as a sufficient condition, then the worry would
arise that on this account anything required by society would amount
to an obligation. It is therefore important that Hegel considers these
requirements as laid down by the rational state, which is seeking to
uphold the freedom of its individual citizens: without this constraint,
it is clear that it would not have the legitimacy to create genuine duties
for people to follow.
That this is so is clear from the way Hegel treats the contrast as he
sees it between duties as they are conceived in modern societies, and
those that were accepted within the states of classical antiquity. In the
latter case, Hegel holds, obligations were imposed on citizens prior
to a properly developed conception of freedom, so that the liberty of
individuals to pursue their own particular concerns and interests was
not acknowledged, and demands were thereby made by society on indi-
viduals that failed to take this into account. Hegel makes clear that he
believes the view of freedom held in this period was limited and one-
sided,40 and that accordingly this pre-modern conception of duties
within the state was flawed, despite the fact that they corresponded to
what the ‘ethical substance’ of the time required of individuals.41
There are therefore constraints on what our duties can be, for while
there are moral reasons for society to insist on certain actions and
to forbid others in order for society to function, there are also moral
reasons, based on our freedom, not to want to take from us the ability

40
Cf. PR 260Z, 7:407 (p. 283): ‘In the states of classical antiquity, universality was indeed
already present, but particularity [Partikularität] had not yet been released and set at
liberty’.
41
Cf. PR §261, 7:409 (p. 284): ‘The abstract aspect of duty consists simply in disregard-
ing and excluding particular interests as an inessential and even unworthy moment.
But if we consider the concrete aspect, i.e. the Idea, we can see that the moment of
particularity is also essential, and that its satisfaction is also necessary; in the process
of fulfilling his duty, the individual must somehow attain his own interest and satisfac-
tion or settle his own account’.
160 Understanding Mor al Obligation

to pursue the different concerns and interests that belong to us as indi-

viduals – our commitments to specific causes, to different pursuits,
our allegiances to family and friends, and so on. The goal, then, is to
find a balance between these two apparently conflicting desiderata,
so that society is structured in such a way as to allow the latter to feed
into the former. As Hegel puts it: ‘Particular interests should certainly
not be set aside, let alone suppressed; on the contrary, they should
be harmonized with the universal, so that both they themselves and
the universal are preserved … In the state everything depends on the
unity of the universal and the particular’.42 A good deal of the point
of Hegel’s account of the structures of the rational state is to show how
this balance can be achieved.
The result is, therefore, that Hegel can claim that the duties which
are enforced within the rational state in this way are not threats to the
freedom of the individual. They may appear to be a threat, but only
if one has what Hegel calls an ‘abstract’ conception of freedom, as
the ability to do whatever one has chosen to do, simply because it has
been so chosen. This view is something that Hegel rejects, however,
on the grounds that freedom in this form is self-contradictory: on the
one hand, the individual cannot allow anything it has so chosen to
govern its actions, as this would be to limit its capacity to choose and
hence its freedom as it conceives it; on the other hand, if it does not
limit its capacity to some extent, it can never act, as to act is always
to will something in particular – so it must therefore in the end just
accept something as what it will do, which it then cannot claim to have
chosen, but rather to have been given to it.43 Once one accepts a less
abstract notion of freedom, therefore, Hegel argues that it can be
accommodated within the account of duties that he presents, as these
arise within the rational state.44
We have seen, then, that Hegel’s account of duty as this arises for
the individual within ethical life can plausibly be considered to be
a form of social command account, where what renders something
a duty or obligation for an individual is the ‘absolute authority and
42
PR §261, 7:409 (p. 285) and §261Z, 7:410 (p. 285; translation modified).
43
Cf. PR §15, 7:65–68 (pp. 48–9).
44
Cf. LA 13:136 (i, p. 98): ‘In a state which is really articulated rationally all the laws
and organizations are nothing but a realisation of freedom in its essential character-
istics. When this is the case, the individual’s reason finds in these institutions only the
actuality of his own essence, and if he obeys these laws, he coincides, not with some-
thing alien to himself, but simply with what is his own. Caprice [Willkür], of course,
is often equally called “freedom”; but caprice is only non-rational freedom, choice
 Hegel’s social command account of obligation 161

power’ of the ethical community. And we have also seen how Hegel
came to develop this account as an alternative to both a divine com-
mand theory (which is seen as a kind of primitive forerunner of the
social command account), and to Kant’s hybrid theory (with its dualis-
tic conception of the will).

Social roles as sources of obligation

Despite my claims above, however, some may feel unpersuaded that
Hegel’s position does fit the profile of a social command account,
where ‘[a]ccording to social theories of the nature of obligation, hav-
ing an obligation to do something consists in being required (in a
certain way, under certain circumstances or conditions), by another
person or a group of persons, to do it’.45 Instead, it may be argued that
what Hegel offers is a social role account of duty and obligation, which
treats these issues in a different way. Indeed, at least since Bradley’s
famous discussion of ‘My Station and Its Duties’ in Ethical Studies, it
is just this approach that has been most commonly identified as a dis-
tinctively ‘Hegelian’ contribution to such matters, and it may therefore
seem wrong-headed of me to suggest otherwise.
I want to argue now, however, that it is a mistake to see Hegel offer-
ing any such ‘social role’ account of duty and obligation in any way
that meaningfully contrasts with the social command account offered
above – and that the same is true of Bradley. I will begin, therefore, by
saying more about the social role theory, and how it differs from the
social command account as characterised above.
In his paper ‘Role Obligations’, Michael Hardimon writes about
such obligations as follows:
As I shall understand the term, a ‘role obligation’ is a moral require-
ment, which attaches to an institutional role, whose content is fixed
by the function of the role, and whose normative force flows from the
role.46

On this view, then, what makes it the case that someone has an obli-
gation to do something is that this obligation relates to the role they
occupy, as father, teacher, citizen, or whatever. The role the person

and self-determination issuing not from the rationality of the will but from fortuitous
impulses and their dependence on sense and the external world’.
45
Adams 1999: 242. 46 Hardimon 1994: 334.
162 Understanding Mor al Obligation

occupies, therefore, is understood to be the basis on which certain

actions come to be required of them, in a way that renders these
actions obligatory.
The question that arises, then, is how roles give rise to this obliga-
toriness, and so make certain actions required. What is it about roles
that can bring it about that the person in one has an obligation or set
of obligations?
Now, some have argued that there is really nothing special about
roles as such that accounts for this, and that any obligatoriness they
may appear to involve can be explained by a further normative feature
in a reductive way – where the most common candidate for this feature
is some sort of contractual story. Thus, it is argued, individuals in social
roles have obligations, because they have in various ways contracted or
agreed to do certain things, and that because it is obligatory to keep
one’s contracts and agreements, so it is obligatory to act in these ways.
Of course, we then need some account of the obligatoriness of keep-
ing contracts and agreements: but whatever account we offer of that, it
would seem to prevent us appealing to roles as a distinctive explanation
of how obligatoriness arises.
However, role theorists have objected that these contractual stories
are in fact implausible as accounts of how roles give rise to duties and
obligations, on the grounds that few if any roles are contractual in this
way. Thus, it is argued, there is no contract underlying the relation
between parents and children, but we nonetheless think individuals
have obligations here by virtue of occupying the respective roles; and
even in cases where the contractual idea is more plausible, perhaps,
the kinds of obligations we attribute to people in those roles are again
often too concrete and specific to have any clear contractual source, of
an actual or even hypothetical kind. It seems, then, that the obligatori-
ness that comes from roles must be understood in a different way.47
One commonly canvassed alternative is what I will call an identifica-
tory account. Here, the idea is, what explains the obligatoriness that
comes with occupying a role is that the individual’s own identity as
a person is tied up with that role, such that not performing it is no
longer an option for them, but becomes a requirement on them; and
if the role involves doing some moral good, then it constitutes a moral
requirement. Thus, the claim is, it is through identifying with the role

For a flavour of these debates, which I cannot go into fully here, see Hardimon’s art-
47

icle, and the response in Simmons 1996b. See also Sciaraffa 2011.
 Hegel’s social command account of obligation 163

that the actions associated with it acquire their obligatory force. As

John Horton puts this view:
[B]oth the family and the political community figure prominently in
our sense of who we are; our self-identity and our understanding of our
place in the world … It should not be surprising, therefore, that some
institutional obligations, through their deep-rooted connections with
our sense of who we are and our place in the world, have a particularly
fundamental role in our moral being. That these kind of institutional
involvement generate moral obligations, and these obligations rather
than standing in need of justification may themselves be justificatory, is
only to be expected.48

Now, in a way that is ironic given her close association with Kant,
the person who has most developed this sort of account of obligation
within contemporary ethics is Christine Korsgaard. This is reflected
in her conception of practical identity, which is ‘a description under
which you find your life to be worth living and your actions to be worth
undertaking’.49 Some of these identities can be, and for most will be,
tied in with an individual’s social roles, whilst others (such as ‘being a
human being’) may not:
Practical identity is a complex matter and for the average person there
will be a jumble of such conceptions. You are a human being, a woman
or a man, an adherent of a certain religion, a member of an ethnic
group, a member of a certain profession, someone’s lover or friend, and
so on. And all of these identities give rise to reasons and obligations.
Your reasons express your identity, your nature; your obligations spring
from what that identity forbids.50

Korsgaard’s claim, then, is that ‘[a]n obligation always takes the form
of a reaction against a threat of a loss of identity’,51 where for most of
us, this will mean our social roles will bring in obligations, as for most
of us our identity is bound up with such roles to at least some degree.
And, where those roles involve moral goods, therefore the obligations
48
Horton 1992: 150 and 157. 49 Korsgaard 1996c: 101.
50
Korsgaard 1996c: 101.
51
Korsgaard 1996c: 102. Cf. also p. 18: ‘[Moral claims on us] must issue in a deep way
from our sense of who we are’, and pp. 239–40: ‘You may be tempted to do something
but find that it is inconsistent with your identity as a teacher or a mother or a friend,
and the thought that it is inconsistent may give rise to a new incentive, an incentive
not to do this thing. As Luther’s “here I stand, I cannot do otherwise” reminds us, the
human heart, being human, discovers itself not only in spontaneous desire, but in
imperatives’.
164 Understanding Mor al Obligation

they give rise to are in turn themselves moral; while if the role does not
involve such goods (as in the case of the member of the Mafia brought
up against Korsgaard by Cohen), they are not.52
We therefore seem to have here a social account of obligation that
is nonetheless distinct from and a competitor to the social command
theory: for here, it is the identification of the individual with their role
that explains obligatoriness, not the ‘social pressure’ exerted on the
individual, as on the social command account.
Now, one important historical source of the role view is generally
taken to be Bradley, where it is then also widely held that Bradley is fol-
lowing Hegel. If this is right, and if the social role and the social com-
mand views are in competition, then it would follow that my account
of Hegel as adopting the latter and not the former must be mistaken.
Thus, when it comes to Bradley, A. John Simmons cites the following
remarks, as evidence of his commitment to the identification thesis,
and thus to the social role theory:
We have found ourselves when we have found our station and its duties,
our function as an organ in the social organism … If we suppose the
world of relations, in which [an Englishman] was born and bred, never
to have been, then we suppose the very essence of him not to be; if we
take that away, we have taken him away … The state … gives him the life
that he does and ought to live.53

And, as the chapter on ‘My Station and Its Duties’ in Ethical Studies,
from which these remarks are taken, is clearly signposted by Bradley as
Hegelian in inspiration, it may seem that Hegel’s position too should
be interpreted in these terms, as a social role theory.
To see why this view is mistaken, however, we can begin by seeing
how it is mistaken even of Bradley himself, and that in fact he is also
much more plausibly read as a social command theorist.
That this is so can be made plain once one recalls the structure
of the dialectic in Ethical Studies, and the place of the chapter (or
‘essay’) on ‘My Station and Its Duties’ within it. Up to this point,
Bradley has considered two contrasting approaches, both of which are
said to have some merit, but neither of which is wholly satisfactory as
things stand. The first is ‘pleasure for pleasure’s sake’, which has the

52
Cf. Korsgaard 1996c: 256–8.
53
Bradley 1927: 163, 166, 174. Simmons cites these remarks (from a different edition) in
Simmons 1996a: 262 note 38/2001: 80–1 note 38.
 Hegel’s social command account of obligation 165

advantage of thinking about how morality might relate to the individ-

ual’s ­‘self-realisation’, but does so in way that has a narrow and mis-
taken view of what this amounts to, namely pleasure. The second is
‘duty for duty’s sake’, which rightly scorns the latter idea as simplistic,
and instead conceives of the self to be realised as the pure will, and so
understands morality in terms that are merely formal. Again, accord-
ing to Bradley, there is some value to viewing morality in this way
(where he is of course thinking of Kant),54 but for familiar Hegelian
reasons (including the dualism we have discussed above),55 he rejects
it as ultimately unsatisfactory.
What is needed, therefore, is some sort of synthesis or ‘Aufhebung’
of these views, which Bradley tries to offer in ‘My Station and Its
Duties’.56 What we are looking for, then, is a position that has a con-
ception of duty that overcomes the problems with the Kantian position
and which also relates it to a notion of self-realisation that is less crude
than the one offered by the outlook of ‘pleasure for pleasure’s sake’.

54
In his discussion of ‘duty for duty’s sake’, Bradley does not claim to be doing full just-
ice to Kant’s position – though he gives this a characteristically Bradleyean sting in
the tail: ‘As I said before, this is not a statement of the Kantian view; that view is far
wider, and at the same time more confused’ (Bradley 1927: 148, note 1; cf. also p. 142,
note 2).
55
Cf. Bradley 1927: 146–7: ‘Morality is the activity of the formal self forcing the ­sensuous
self, and here first can we attach a meaning to the words “ought” and “duty” … [I]f
our self were a pure, unalloyed will, realizing itself apart from a sensuous element,
the word “ought” would … be meaningless. It is the antagonism of the two elements
in one subject which is the essence of the ought … The ought is the command of the
formal will, and duty is the obedience, or, more properly, the compulsion of the lower
self by that will, or the realisation of the form in and against the recalcitrant matter
of the desires’. Bradley also follows Hegel in objecting to the ‘empty formalism’ of the
Kantian approach, which he sees as arising out of this dualism. Cf. also Bradley 1927:
181: ‘In “duty for duty’s sake” we were always unsatisfied, no nearer our goal at the end
than at the beginning. There we had the fixed antithesis of the sensuous self on one
side and a non-sensuous moral ideal on the other – a standing contradiction which
brought with it a perpetual self-deceit, or the depressing perpetual confession that I
am not what I ought to be in my inner heart, and that I never can be so’; and p. 215,
where Bradley criticises the view that ‘morality is a life harassed and persecuted every-
where by “imperatives” and disagreeable duties, and that without these you have not
got morality’, commenting that ‘it is sufficient to remark that [this view] stands or falls
with the identification of morality with unwilling obedience to law, and that, accord-
ing to the common view, a man does not cease to be good so far as goodness becomes
natural and pleasant to him’.
56
In fact, as is sometimes missed, the dialectic continues even after ‘My Station and Its
Duties’, which itself proves problematic in various ways (see Bradley 1927: 202–6). But
that need not concern us here, as the reading of Bradley as a social role theorist con-
cerning obligation is generally supported by this chapter alone.
166 Understanding Mor al Obligation

Bradley makes this clear when he summarises how things stand at the
start of the chapter:
We have learnt that the self to be realized is not the self as this or that
feeling, or as any series of the particular feelings of our own or others’
streams or trains of consciousness. It is, in short, not the self to be
pleased … And, passing then to the opposite pole, to the universal as
the negative of the particulars, to the supposed pure will of duty for
duty’s sake, we found that too was an unreal conception …
But let us view this [negative] result, which seems so unsatisfactory,
from the positive side; let us see after all with what we are left. We have
self-realization left as the end, the self so far being defined as neither a
collection of particular feelings nor an abstract universal. The self is to
be realized … as will, will not being merely the natural will … but the
will as the good will, i.e. the will that realizes an end which is above this
or that man, superior to them, and capable of confronting them in the
shape of a law or an ought. This superior something, further, which is
a possible law or ought to the individual man, does not depend for its
existence on his choice or opinion.57

Bradley’s position, then, is that what his previous discussion has shown
is that we need a view that allows for self-realisation on the one hand,
and duty on the other, without treating the former as mere pleasure
or hedonistic well-being, and the latter as something empty, formal
and dualistic – where it is precisely in a view that tries to achieve both,
that these respective limitations will be overcome. Bradley’s positive
suggestion, therefore, is that if we think of the individual as following
duties that relate to a good that is more than his individual good, then
at the same time self-realisation will be achieved, and these duties will
be given a content and context as well as a non-dualistic force over us,
in a way that will enable a satisfactory ‘middle way’ to be found.
And then, Bradley claims, this is just what one will get within a state,
in which the individual is both part of the general good of the commu-
nity, and also able to find itself fully realised by participating in that
community as a result. Thus, Bradley declares, in a passage of consid-
erable rhetorical force, by living within a ‘social organism’ of this sort,
where one has a ‘station and its duties’ through which one contributes
to this goal, and therefore also has contentful and objective require-
ments laid upon one, by a society in which one also flourishes, then a
notable advance towards dialectical stability will have been achieved:

57
 Bradley 1927: 160–2.
 Hegel’s social command account of obligation 167

Here, and here first, are the contradictions which have beset us solved –
here is a universal which can confront our wandering desires with a
fixed and stern imperative, but which yet is no unreal form of the mind,
but a living soul that penetrates and stands fast in the detail of actual
existence. It is real, and real for me. It is in its affirmation that I affirm
myself, for I am but as a ‘heart-beat in its system’. And I am real in it;
for, when I give myself to it, it gives me the fruition of my own personal
activity, the accomplished ideal of my life which is happiness. In the
realized idea which, superior to me, and yet here and now in and by me,
affirms itself in a continuous process, we have found the end, we have
found self-realization, duty, and happiness in one – yes, we have found
ourselves, when we have found our station and its duties, our function
as an organ in the social organism.58

My claim is, then, that up to this point, Bradley is offering a social

command account, whereby on the one hand the state59 is such as to
‘confront our wandering desires with a strict and firm imperative’
because of its authority over us, but where on the other hand ‘when I
give myself up to it’, the state ‘gives me the fruition of my own personal
activity, the accomplished ideal of my life which is happiness’. Bradley
makes the nature of his position fully clear when he writes: ‘[The state]
speaks the word of command and gives the field of accomplishment,
and in the activity of obedience it has and bestows individual life and
satisfaction and happiness’.60
Likewise, I would argue, from what we saw before in the previous
section, Bradley is in effect paraphrasing Hegel here (as he would
no doubt happily grant), and Hegel’s claim that ‘[i]n the state every-
thing depends on the unity of universal and particular’.61 It is precisely

58
Bradley 1927: 163.
59
As a note in the second edition of Ethical Studies to the passage just cited makes clear,
Bradley does not think that this can be achieved only within the state as a political com-
munity, but also within other kinds of community; but as the note also makes clear, in
writing this passage, it was really the state he had in mind.
60
Bradley 1927: 184–5. Cf. also pp. 203–4, where in considering the limitations on the
position he has outlined in ‘My Station and Its Duties’, Bradley observes that: ‘It is
necessary to remark that the community in which he is a member may be in a confused
or rotten condition, so that in it right and might do not always go together’ – that is,
the commanding authority of the state, which is the basis of duty, may go astray.
61
Cf. Bradley 1927: 202: ‘The theory which we have just exhibited [of my station and
its duties] … satisfies us, because in it our will attain their realization; the content of
the will is a whole, is systematic; and it is the same whole on both sides. On the out-
side and the inside alike we have the same universal will in union with the particular
personality’.
168 Understanding Mor al Obligation

this, as we have seen, that allows Hegel to also strike the balance that
Bradley is after, between duty as imposed by the state on the one hand
and the interests of the individual on the other, so that by having the
source of those duties in the command of the rational state, the indi-
vidual has obligations, has their ‘particularity’ taken into account, and
is lifted above the narrow and egoistic concerns of the pre-social indi-
vidual. By thinking of duty in these terms, as imposed by society on the
individual who has a place and role within it, the dialectical harmony
that both Hegel and Bradley are looking for can be achieved, but only
because obligations are seen to arise from the social community of
which they are part, and which has the self-realisation or freedom of
its citizens (which for Bradley and Hegel are in effect the same thing)
at its heart.
However, if this shows him to be a social command theorist, what
of the passages in which Bradley seems to make so much of the way in
which an individual’s identity is bound up with their role, and which
have led so many to interpret him as a social role theorist concerning
obligation? And similarly, what of the comparable passages in Hegel?
When it comes to Bradley, I think the simple answer is as follows.
These ‘identificatory’ passages are there not to support an identifica-
tory account of social roles, but to answer three very significant objec-
tions to any social command theory, namely:
(a) that the state which Bradley claims has the authority to give indi-
viduals their duties does not really exist and is a myth, because it
can always be reduced to a mere collection of individuals, with
nothing but the authority of individuals over one another
(b) that self-realisation does not require social membership, so that
there is no essential connection (as Bradley claims there is)
between a morality of social duties and self-realisation
(c) that individuals must always see the authority of the state as taking
away their freedom.
All three objections can be urged by the ‘individualist’, who does not
think Bradley’s vision of the ‘social organism’ is at all plausible, where
it is the position of this individualist that Bradley outlines immedi-
ately after the passage that we just cited, with its high-flown talk of the
‘social organism’:
‘Mere rhetoric’, we shall be told, ‘a bad metaphysical dream, a stale old
story once more warmed up, which can not hold its own against the
 Hegel’s social command account of obligation 169

logic of facts. That the state was prior to the individual, that the whole
was sometimes more than the sum of its parts, was an illusion which
preyed on the thinkers of Greece. But that illusion has been traced to
its source and dispelled, and is in plain words exploded. The family,
society, the state, and generally every community of men, consists of
individuals, and there is nothing in them real except individuals’.62

Now, it is also clear that it is in order to refute just this view that Bradley
turns to his claim about the dependence of individuals for their iden-
tity on society and their place within it:
[W]e meet the metaphysical assertion of the ‘individualist’ with a mere
denial; and, turning to facts, we will try to show that they lead us in
another direction. To the assertion, then, that selves are ‘individual’ in
the sense of exclusive of other selves, we oppose the (equally justified)
assertion, that this is a mere fancy. We say that, out of theory, no such
individual men exist; and we will try to show from fact that, in fact, what
we call an individual man is what he is because of and by virtue of com-
munity, and that communities are thus not mere names but something
real, and can be regarded (if we mean to keep to facts) only as the one
in the many.63

After a long disquisition in support of this view, which hinges on how

much an individual’s identity depends on his place within a social
framework, Bradley concludes:
In short, man is a social being; he is real only because he is social, and
can realize himself only because it is as social that he realizes himself.
The mere individual is a delusion of theory; and the attempt to realize it
in practice is the starvation and mutilation of human nature, with total
sterility or the production of monstrosities.64

Bradley’s response to the reductionist objection that there cannot be

any social commands, because ‘in fact’ there is no social organism, is
that the reduction cannot work, as without the social organism there

62
Bradley 1927: 163. Bradley does not identify precisely whom he was thinking of as hold-
ing this individualist position, but Peter Nicholson plausibly suggests that he ‘perhaps
had in mind such writers as Bentham, J. S. Mill, and Herbert Spencer’ (Nicholson 1990:
24). For the problem this position raises for the social command theorist, see Wolf
2009: 345, where she notes that for ‘the question of society’s existence is … a legitimate
and serious question … To be sure, we live among other people – in a neighborhood, a
state, a world. But is any collection of them sufficiently organized and unified to consti-
tute a group that can be seen to issue commands in the requisite sense?’.
63
Bradley 1927: 166. 64
 Bradley 1927: 174.
170 Understanding Mor al Obligation

is ‘in fact’ no individual. What we see in this talk of identity and one’s
place in society, therefore, is not a defence of an identificatory theory
of obligation, but a defence of the idea of society that is needed by the
kind of social command theory that Bradley has put forward earlier in
the chapter. It is also needed to substantiate his crucial link between
duty and self-realisation, which on the individualist position does not
require the person to have any place within a social whole, while it also
shows that this social will is not alien to the agent’s own will.
Bradley thus uses his ‘identificatory’ claims as a way of supporting
his anti-individualism and his account of the social organism, which
he needs in order to defend his social command theory:
(a) there is such a thing as the state or society, that can issue
commands.
(b) the individual can realise themselves by following these social
duties.
(c) the individual need not feel ‘alienated’ from these duties as exter-
nal impositions, in so far as they are essentially bound up with this
social whole.
It can be argued, then, that Bradley’s focus on the social identity of
the individual does not show that his account of duty based on roles
is intended to be identificatory, but rather forms part of an approach
that fits better with a social command model.
And something similar is true, I would argue, of Hegel – but where
for Hegel, the identificatory talk is there not to show what we can
defend a non-reductionist view of the state, but rather that (as we have
seen) social duties are not inimical to the freedom of the individual,
in so far as this ‘ethical substance’ is not ‘alien to the subject’ – where
‘[o]n the contrary, the subject bears spiritual witness to them as to its
own essence’.65 In Hegel, then, claims of this sort also do not support
a social role reading, but the one offered above, that treats Hegel as a
social command theorist when it comes to understanding how moral
obligations are possible.

65
PR §147, 7:295 (p. 191).
 Pa rt i i i

K i e r k eg a a r d
 6

K i e r k eg a a r d’s c r i t iqu e of H ege l

In an earlier chapter, it was argued that Kant turned away from a div-
ine command account of obligation, to offer instead a hybrid account,
while we have just seen that Hegel then turned away from this to offer
his social command account instead. In this chapter, the wheel turns
again, as Kierkegaard’s critique of the latter takes us back to a divine
command account.
However, whilst it is scarcely surprising to say that Kierkegaard was
a critic of Hegel in some broad sense,1 and also possibly to say that he
was a divine command theorist in some broad sense,2 it is less easy to
narrow down these aspects of his position, so to say exactly what these
criticisms amount to, and exactly what form of divine command theory
Kierkegaard was proposing. When it comes to the former, we need to
substantiate that it was Hegel’s social command account of obligation
that formed the focus of Kierkegaard’s objections, and not just other
issues that have no impact on this question; and when it comes to the
latter, we need to substantiate that Kierkegaard was offering a divine
command account of obligation, and not an ethic of a different sort, or
a divine command theory of a more radically voluntaristic kind, which

1
Those who read Kierkegaard as a critic of Hegel include: Thulstrup 1980; Crites 1972;
Collins 1983; Westphal 1998; Taylor 2000. For a more revisionist account, which argues
that Kierkegaard’s real target was the members of the school of Danish Hegelians,
rather than Hegel himself, see Stewart 2003. However, whilst Stewart succeeds in add-
ing a lot of fascinating detail to the story of Kierkegaard’s encounter with Hegel and
Hegelianism, I am not in the end convinced that he succeeds in overturning the more
standard view. For responses to Stewart along these lines, see Westphal 2004; Pattison
2005: 28–33; and James 2007.
2
For characterisations of Kierkegaard in relation to divine command theory, see Quinn
1996 and 1998; and Evans 2004. For the suggestion that Kierkegaard is not a divine
command theorist, see Green 1992: 202; Ferreira 2001: 40–2, 242, and also Ferreira
2002, esp. pp. 149–52; and Manis 2009.

173
174 Understanding Mor al Obligation

treats the good and right as altogether dependent on God’s command,

not merely for their obligatory force.
My discussion of Kierkegaard will be structured as follows. In
this chapter, which focuses on Kierkegaard’s critique of Hegel, I will
look mainly at the two pseudonymous works that contain objections
to Hegel’s ethical outlook, namely Fear and Trembling, and Either/Or.
However, in part because they are pseudonymous, and in part also
because their intentions are mainly negative, it is unwise to see in these
writings the full extent of Kierkegaard’s positive position as a whole –
where for this, in the next chapter, we will turn to Kierkegaard’s Works
of Love.

Kierkegaard’s critique of Hegel in Fear and Trembling

Published by Kierkegaard in 1843, Fear and Trembling was written
under the pseudonym of Johannes de Silentio, and is centred around
the story of the binding of Isaac by Abraham (Genesis 22:1–19). It can
be seen as a ‘singularly problematic text’,3 for a variety of reasons –
not only because of its pseudonymous authorship and curious literary
form; for its hints at a hidden meaning; and for its tantalising relation
to Kierkegaard’s biography at this period; but also because of the need
to do justice to its ethical and religious radicalism on the one hand,
without on the other tipping it into a position that is so extreme that
it becomes difficult to take seriously. In particular, to do justice to its
radicalism, it can seem necessary to interpret it as offering a strongly
voluntaristic divine command theory, as many have done;4 but then
it has appeared easy to dismiss it, given the rebarbative implications
of such a theory.5 Faced with this difficulty, defenders of Kierkegaard
then point to the pseudonymous nature of the work, its clear polem-
ical intent, as well as its possibly secret messages, and argue that this
is not really his own position and that it is misleading to think of
Kierkegaard himself as a divine command theorist at all; but then,
as a result, Fear and Trembling becomes rather marginalised.6 Or it is

3
Pattison 2005: 115.
4
For a recent reading of Kierkegaard along these lines, see Irwin 2009: 313–5 and
322–4. See also Olafson 1967: 28–31.
5
Cf. Green 1993: 198: ‘A more serious problem is that if Fear and Trembling defends a div-
ine command ethic, it is a forbidding and frightening ethic’.
6
Cf. Ferreira 2001: 40.
 Kierkegaard’s critique of Hegel 175

argued that Fear and Trembling has no real relation to divine command
approaches, and is given a different focus: but this is to go against what
appears to be a natural reading of the work.
In what follows, I hope to avoid this oscillation, by treating the text
as putting forward a divine command account of obligation in contrast
to Hegel’s social command account, but where (as we have seen) this
sort of ‘intermediate’ divine command theory is distinct from any
strong voluntarism. However, as we shall go on to see, it still does have
a radical potential of a different kind that the social command theory
does not, which Kierkegaard exploits in his dialectic with Hegel, so
that none of the text’s tendency to disturb need be lost. The root of
Kierkegaard’s concern here, I will argue, has to do with the relation
between ethics and faith: on Hegel’s social theory of obligation, there
is a huge cost in religious terms, as such a theory cannot treat the good
and the right as transcendent and thus beyond our full comprehen-
sion, while for Kierkegaard it is precisely this transcendence which it
is necessary to acknowledge if we are to stand in the proper relation
to the divine. This transcendent conception of moral value will still
mean that Fear and Trembling represents a radical challenge to secular
ethicists, who will characteristically take it that moral value is broadly
graspable within the human perspective; but it is a challenge distinct
from that posed by a voluntaristic divine command theory, and it is
one that Kierkegaard thinks it is necessary to preserve in order to
make sense of religious faith.
That this is the key focus of Fear and Trembling is made clear by the
way in which the text is ‘framed’, by its Preface at the beginning and
its Epilogue at the end. In both, it is religious faith and in particular
the devaluing of faith that is the primary concern – where Kierkegaard
treats his contemporary Hegelians as symptomatic in this respect,
who, as a result of following Hegel, have been led to believe that faith
is easily come by and surpassed. In the Preface, de Silentio draws a
parallel in this respect between faith and doubt, which contemporary
Hegelians also take in their stride and effortlessly ‘go beyond’, and he
contrasts the modern outlook with that of previous eras, where faith
and doubt were taken more seriously, both in terms of how long it took
to properly come to terms with them, and of how hard they were to
transcend.
De Silentio sees that what might make this possible is the Hegelian
‘System’, in which religion is ‘sublated’ by philosophy, so that ‘the
whole content of faith [is converted] into conceptual form’ and so
176 Understanding Mor al Obligation

made intelligible.7 But he confesses that he himself cannot grasp this

‘System’, as he is ‘not at all a philosopher’,8 so that faith for him is
much harder to deal with and ‘get beyond’. In the Epilogue, de Silentio
makes clear that in fact, it cannot be genuine faith that has been ‘sub-
lated’ in this way: ‘There are perhaps many in every generation who
do not even come to [faith]; but nobody goes further’:
Whether there are also many in our age who do not discover it, I do not
decide; I dare only refer to myself, who does not conceal that it may not
happen for a long time to come for him, yet without his therefore wish-
ing to deceive himself or the great by making it into a trifling matter,
into a childhood malady one must wish to get over as soon as possible.9

De Silentio then compares himself not to a merchant who dumps

spices in the sea to raise their price artificially in a sluggish market, but
to someone who, in the foregoing discussion of Abraham, has shown
what faith really amounts to, in order to combat the complacency with
which it is treated by the Hegelians, by bringing out its truly difficult
and challenging nature.10
Now, within the main body of the text, the primary term with
which de Silentio tries to bring out the difficult nature of faith is ‘the
absurd’ – for this, he argues, is how Abraham must appear to those
without faith (of whom he is one),11 but where if this absurdity were
lacking, Abraham would not be a ‘knight of faith’, but another kind of
figure entirely. De Silentio presents two dimensions to this absurdity
in connection with the binding of Isaac,12 relating to two ways in which

7
FT 3:59 (p. 5). Cf. Hegel LHP 18:100 (i, p. 79; translation modified): ‘Thus Religion
has a content in common with Philosophy the forms alone being different; and the
only essential point is that the form of the Concept should be so far perfected as to be
able to grasp the content of Religion’.
8
FT 3:59 (p. 5). 9 FT 3:167 (p. 108).
10
FT 3:166 (p. 107): ‘is what [the present generation needs] not rather an honest earnest-
ness that fearlessly and incorruptibly calls attention to the tasks, an honest earnestness
that lovingly preserves the tasks, that does not anxiously want to rush precipitously to
the highest but keeps the tasks young, beautiful, delightful to look upon, and inviting
to all, yet also difficult and inspiring for the noble-minded (for the noble nature is
inspired only by the difficult)?’.
11
Cf. FT 3:84 (p. 26): ‘By no means do I have faith’; FT 3:85 (p. 28); ‘I can well endure
living in my own fashion, I am happy and content, but my joy is not that of faith and in
comparison with that is really unhappy’.
12
De Silentio also mentions other aspects of absurdity in relation to the Abraham story
more generally, such as his emigration and his belief that Sarah would have a child:
see FT 3:69–70 (p. 14).
 Kierkegaard’s critique of Hegel 177

Abraham might be thought of as less than a knight of faith, where

each dimension relates roughly to the two halves of the main part of
the book – namely, the half covering the ‘Attunement’, the ‘Tribute
to Abraham’ and the ‘Preliminary Outpouring of the Heart’, and the
half containing the three ‘Problems’.
The first dimension in which Abraham’s position can be viewed as
absurd, and yet in a way which also qualifies him as a knight of faith,
is not as immediately ethical as the second, though it does relate to
it nonetheless. This first dimension of absurdity is brought out by de
Silentio by contrasting Abraham’s way of responding to God’s com-
mand to sacrifice Isaac, and what his own response would have been
in the same situation. That is, rather than carrying it out with dread,
foreboding, or resignation, Abraham set about carrying it out with joy.
De Silentio attributes this difference to a belief that Abraham had,
which could not be had by someone like himself who lacked faith,
and who therefore could not share in Abraham’s joyous demeanour –
namely, the belief that God demanded that he sacrifice Isaac, but also
that the demand would be waived, where Abraham takes both equally
seriously in a way that defies ordinary comprehension:
But what did Abraham do? He arrived neither too early nor too late. He
mounted the ass and rode slowly along the way. During all this time he
believed; he believed that God would not demand Isaac of him, while
he still was willing to sacrifice him if it was demanded. He believed by
virtue of the absurd, for human calculation was out of the question, and
it was indeed absurd that God, who demanded it of him, in the next
instant would revoke the demand. He climbed the mountain, and even
at the moment when the knife gleamed he believed – that God would
not demand Isaac.13

Perhaps feeling, however, that this does not quite capture the full
nature of the relation between Abraham’s faith and the absurd, de
Silentio adds a further level to the account, which is not only that God
would allow Abraham to keep Isaac despite requiring that he be sac-
rificed, but that God would allow him to kill Isaac, but somehow give
him back:
Let us go further. We let Isaac actually be sacrificed. Abraham believed.
He did not believe that he would be blessed one day in the hereafter but
that he would become blissfully happy here in the world. God could give

13
FT 3:86–7 (p. 29).
178 Understanding Mor al Obligation

him a new Isaac, call the sacrificed one back to life. He believed by vir-
tue of the absurd, for all human calculation had long since ceased.14

As de Silentio observes in the preceding ‘Tribute to Abraham’, not

only here but throughout his life, Abraham ‘left one thing behind,
took one thing with him. He left his worldly understanding behind
and took faith with him; otherwise he undoubtedly would not have
emigrated but surely would have thought it preposterous’.15
In the ‘Preliminary Outpouring of the Heart’, this aspect of
Abraham’s position is then used to draw a contrast between faith and
what de Silentio calls ‘infinite resignation’, where the latter involves
abandoning the joys, passions, and pleasures of ordinary existence, for
the sake of some higher cause, while the former somehow manages to
retain a commitment to the finite despite all that is being asked of it.
De Silentio’s suggestion is that it is the absurdity of Abraham’s belief
that makes this commitment possible for him, as the more reasonable
position would seem to be resignation, as the reasonable view is that
everything has been lost in sacrificing Isaac – not just Isaac himself, but
all Abraham’s hopes for his legacy and for his people that have been
founded on Isaac’s continuance of his line. De Silentio thus pictures
the knight of faith dwelling contentedly within the finite, mundane
world – perhaps as an ordinary tax-collector, rather than as any sort
of other-worldly ascetic. In order to achieve this, like Abraham, this
ordinary believer must have made the movement of faith, of renunci-
ation followed by return:
And yet, yet – yes, I could fly into a rage over it, if for no other reason
than out of envy – yet this person has made and at every moment is
making the movement of infinity. He empties the deep sadness of exist-
ence in infinite resignation, he knows the blessedness of infinity, he
has felt the pain of renouncing everything, the dearest thing he has in
the world, and yet the finite tastes every bit as good to him as someone
who never knew anything higher, for his remaining in finitude has no
trace of a dispirited, anxious training, and yet he has this confidence
to delight in it as if it were the most certain thing of all. And yet, yet
the whole earthly figure he presents is a new creation by virtue of the
absurd. He resigned everything infinitely and then grasped everything
again by virtue of the absurd.16

14
FT 3:87 (pp. 29–30).
15
FT 3:69 (p. 14).
16
FT 3:91 (p. 34).
 Kierkegaard’s critique of Hegel 179

De Silentio goes on to develop the contrast, with his tale of a young lad
who forms a doomed attachment to a princess, who can never consum-
mate his love: the ‘knight of resignation’ finds his love transfigured by
his abandonment of his early hopes, while the ‘knight of faith’ retains
his place within the finite, by retaining the belief that somehow the
princess will be his in the end, even while he suffers through the pain
of knowing that she will not.17
In this way, therefore, de Silentio takes himself to have shown how
difficult it is to make sense of faith, for faith can only take on its char-
acteristic feature by virtue of its connection with the absurd; once a
more reasonable attitude is adopted, it becomes something more like
infinite resignation. So, the first criticism of the Hegelian is that he
has underestimated this absurdity. The second, and related, criticism,
is that he has therefore mischaracterised faith and confused it with
infinite resignation: for, although faith involves the transcendent, for
de Silentio it also brings with it precisely the kind of ‘being at home in
the world’ that the Hegelian claims to provide, but which the Hegelian
thinks requires immanence and not transcendence.18 Far from los-
ing touch with the finite through the transcendent, de Silentio sug-
gests, it is only through the latter that the finite is genuinely retained,
which is otherwise in danger of being lost in the attitude of infinite

17
FT 3:96–7 (p. 39): ‘We shall now let the knight of faith appear in the incident previ-
ously mentioned. He does exactly the same as the other knight, he infinitely renounces
the love that is the content of his life and is reconciled in pain. But then the miracle
occurs. He makes yet another movement more wonderful than anything, for he says:
“I nevertheless believe that I shall get her, namely by virtue of the absurd, by virtue of
the fact that for God everything is possible.” The absurd does not belong to the distinc-
tions that lie within the proper compass of the understanding’. Cf. 3:99 (p. 42): ‘But by
my own strength I cannot get the least bit of what belongs to finitude, for I continually
use my strength to resign everything. By my own strength I can give up the princess,
and I shall not become a sulker but find joy and peace and rest in my pain. But by my
own strength I cannot get her back again, for I use all my strength just for the act of
resigning. But by faith, says that miraculous knight, by faith you will get her by virtue
of the absurd’. Cf. also 3:97 (p. 40), where de Silentio contrasts the faith of the young
boy with a girl whose optimism is just based on ‘childlike naiveté and innocence’, and
who is not therefore aware of all that stands in the way of her hopes being fulfilled, and
so experiences no difficulty in her optimism.
18
Cf. EL §38Z 8:109 (p. 78), where Hegel expresses his admiration for this side of empiri-
cism, though of course he is critical of it in other respects: ‘From Empiricism the call
went out: “Stop chasing about among empty abstractions, look at what is there for the
taking, grasp the here and now, human and natural, as it is here before us, and enjoy it!”
And there is no denying that this contains an essentially justified moment. This world,
the here and now, the present, was to be substituted for the empty Beyond, for the spi-
derwebs and cloudy shapes of the abstract understanding’.
180 Understanding Mor al Obligation

resignation – an attitude which the Hegelian confuses with genuine

faith.
We have seen, then, how in the first half of the book, de Silentio
draws out the constitutive connection between faith and the absurd –
where up to this point, the absurdity in question has primarily been
metaphysical in a broad sense (how can Isaac be sacrificed and yet
live? how can God intend him to be sacrificed yet equally intend to
stop the sacrifice? how can the princess come to love the young lad,
given all the obstacles that stand in the way?). If this absurdity did
not form part of religious life, de Silentio is clearly arguing, such life
would not be truly possible. My claim now will be that the discussion at
this point moves on to a different kind of absurdity – an ethical absurd-
ity – that is equally said to form a crucial part of the religious life in a
way that the Abraham story also brings out, where it is this that would
be lost if the Hegelian account of obligation were accepted, and with
it the possibility of religious faith.
However, it is important to also recognise that the absurdity encoun-
tered so far can be thought of as having an ethical dimension. For,
behind Abraham’s belief that God is working in these mysterious ways
is also his belief that somehow through all this, God’s earlier promises
to Abraham regarding his own happiness and that of his people will
be kept – though this is also equally mysterious for him and seemingly
absurd, given that if Isaac dies as God seems to require, Abraham will
lose all that matters to him and his people will have lost their next
leader. Nonetheless, as C. Stephen Evans notes, ‘Abraham simply rests
unwaveringly in his trust in God’s goodness’19 without doubting that
goodness, even though he has no real idea how it is being displayed
in what he is being asked to do. It is this transcendent aspect to what
is good, I will now argue, that is explored more explicitly in the subse-
quent discussion of the ethical in the second part of the book.
De Silentio’s turn to the ethical is signalled at the end of the
‘Preliminary Outpouring’,20 which moves to the discussions of three
‘Problems’ raised by the Akedah, beginning with ‘Is there a teleological
suspension of the ethical?’, and continuing with ‘Is there an absolute

Evans 2006: xix.

19

20
FT 3:103 (p. 46; my emphasis): ‘It is now my intention to draw out in the form of prob-
lems the dialectical factors implicit in the story of Abraham in order to see what a pro-
digious paradox faith is – a paradox that is capable of making a murder into a holy act well
pleasing to God, a paradox that gives Isaac back again to Abraham, which no thought
can lay hold of because faith begins precisely where thinking leaves off’.
 Kierkegaard’s critique of Hegel 181

duty to God?’, and ‘Was it ethically defensible of Abraham to conceal

his undertaking from Sarah, from Eliezer, from Isaac?’. My suggestion
will be that in each of these discussions, a key theme is that for faith
to retain its character, it must retain its connection to the ethically
absurd, where this is only possible if a social command theory of obli-
gation is rejected in favour of a divine command theory of obligation,
for only then can what is right or good be seen as possibly outstrip-
ping our understanding in a way that this conception of the absurd
requires.21
In the first Problem, de Silentio begins with a characterisation of
Hegelian Sittlichkeit, which involves not merely a Kantian conception of
the universality of ethical principles,22 but also the universality of the
ethical community of which the individual is part, and within which
he encounters no ultimate antithesis between his merely particular
interests and the universal good of the social whole, as that distinc-
tion becomes blurred. Within this conception, as we have seen, duties
are enforced through the community, with this social end or telos in
view, of creating a harmonic social order, one in which the individual
is not crushed or subordinated, but finds their higher realisation. As a
result, therefore, the good underlying the imposition on the individ-
ual of any duty is transparent, and what the individual is obliged to do
can be justified to all by appealing to it directly.
This does not mean that on this account no conflict can arise, of
course, as there will be circumstances in which the individual may
be called upon to undergo great personal sacrifice for the sake of the
well-being of the community. In this case, de Silentio argues, the indi-
vidual is put in the position of what he calls the tragic hero, of which

21
This aspect of Kierkegaard’s account has been emphasised in Outka 1973: 236 p. 236:
‘Abraham’s own antecedent criteria of right and wrong are not antecedently authori-
tative. For in a fashion akin to Job, he must finally defer to a wisdom superior to his
own. His obedience may presuppose a general confidence in the wisdom of God’s
commands, but it does not require in the situation a perfect understanding in accord-
ance with his own autonomous moral lights. In this life, at least, he must be prepared
to change his mind. So he sets out, knowing that it is God who tries him, but not fully
understanding the point of the command’. Cf. also pp. 240–4, and also Outka 1993:
213: ‘Fear and Trembling … focuses on the danger ethics presents insofar as it sets ante-
cedent terms for the individual’s personal relation to God. We cannot fully anticipate
what God may command us to do’.
22
Whilst it was once customary to take Kant to be Kierkegaard’s focus here, it has increas-
ingly been recognised that it is Hegel that forms his primary target. See, for example,
Westphal 1981: 73–4/1991: 76–7. See also Evans 2009: 103–4. For a more complete set
of references to different views on this issue, see Lee 1992: 102–3, note 3.
182 Understanding Mor al Obligation

he gives three exemplars: Agamemnon, Jephthah, and Lucius Junius

Brutus. All three are fathers who sacrifice their offspring for the sake
of the good of society, where their duties are laid down as a result of
their place within it, qua king or civic leader. The end for which these
men acted is therefore clear, as are the requirements upon them; and
while we may sympathise with them greatly, the ethical import of their
actions is made transparent by the good realised by what they choose
to do, where in each case their duties to the family are outweighed by
their obligations to the state:
The tragic hero still remains within the ethical. He lets an expression
of the ethical have its telos in a higher expression of the ethical; he
reduces the ethical relation between father and son or daughter and
father to a sentiment that has its dialectic in its relation to the idea of
the ethical life. Here, then, there can be no question of the teleological
suspension of the ethical itself.23

However, while these men might make for tragic heroes, they can-
not (in the manner of Abraham) make for knights of faith, not only
because (unlike Abraham) they do not believe their children will be
returned to them,24 but also because (again unlike Abraham) the
duties relating to their actions are of a civic kind, as is the moral value
that belongs to them. Abraham, by contrast, acted because God com-
manded him to, where the good to be realised by sacrificing Isaac is
opaque and unknown, and where that good is given priority over the
moral value of acting to preserve the social order:
The case is different with Abraham [from that of the tragic hero]. By his
act he transcended the whole of the ethical and had a higher telos out-
side, in relation to which he suspended it. For I would certainly like to
know how Abraham’s act can be brought into relation to the universal,
whether any connection can be discovered between what Abraham did

23
FT 3:109 (pp. 51–2).
24
Cf. FT 3:108–9 (p. 51): ‘When at the decisive moment Agamemnon, Jephthah, and
Brutus heroically have overcome the pain, heroically have lost the beloved and merely
must complete the deed externally, there never will be a noble soul in the world with-
out tears of sympathy for their pain and tears of admiration for their deed. However, if
at the decisive moment these three men were to add to their heroic courage with which
they bore their pain the little phrase, “but it will not happen,” who then would under-
stand them? If as an explanation they added, “we believe it by virtue of the absurd,”
who then would understand them better? For who would not easily understand that it
was absurd, but who could understand that one could then believe it?’.
 Kierkegaard’s critique of Hegel 183

and the universal other than that Abraham overstepped it. It is not to
save a people, not to uphold the idea of the state, not to appease angry
gods that Abraham does it.25

The fact that Abraham did not act to achieve any social good, and
thus did not act in accordance with the duties that might legitimately
be imposed upon him within the framework of Sittlichkeit, means that
he can appeal to no such conception of moral value to legitimate his
actions. All he can do is appeal to the fact that God requires these
actions of him and that they are therefore his duty, but where the link
between that duty and moral value can no longer be discerned in the
way that it can in the case of the tragic hero:
Why does Abraham do it then? For God’s sake, and what is altogether
identical with this, for his own sake. He does it for God’s sake because
God demands this proof of his faith; he does it for his own sake so that
he can prove it. Hence the unity is quite rightly expressed in a word
always used to denote this relation: it is a trial, a temptation. A tempta-
tion; but what does that mean? That which ordinarily tempts a person,
to be sure, is whatever would keep him from doing his duty, but here the
temptation is the ethical itself, which would keep him from doing God’s
will. But then what is the duty? Well, the duty is precisely the expression
for God’s will.26

The latter claim, therefore, shows that Abraham is much more than
a tragic hero, and why a ‘new category’ is needed ‘for understanding
Abraham’.27
Now, what makes this ‘new category’ that of the knight of faith is
(as we saw previously) the link to the absurd, where here the link is
based on the epistemic and moral uncertainty that Abraham is under
and embraces, in a way that the tragic hero is not. For the tragic hero
knows how his duty connects to the good, where Abraham does not,
and so takes an enormous risk in acting as he does and taking it to be
his duty to sacrifice Isaac, since God would not make it obligatory for
him to kill Isaac by commanding it unless it were good in some way –
but he has no idea how this might be so. In view of this, he then could
use this as reasonable grounds on which to reject the command, by
taking it to show that it is not God who is commanding him, or that

25
FT 3:109 (p. 52).
26
FT 3:109 (p. 52).
27
FT 3:110 (p. 52).
184 Understanding Mor al Obligation

he is misunderstanding what is being commanded.28 Abraham does

not do this, however, because he has the humility to simply trust in
God, a humility which he could not exercise if the Hegelian pos-
ition were right, and the connection here between moral value and
duty could always be made clear, as it can on the social command
account.
It is thanks to this opacity, however, that Abraham’s position is
fraught with epistemic and moral risk, because he cannot ever be cer-
tain he is not deluded and that his trust is not entirely misplaced:
[T]he tragic hero gives up the certain for the even more certain, and
the eye of the beholder rests confidently upon him. But the one who
gives up the universal in order to grasp something higher that is not
the universal, what does he do? Is it possible that this can be anything
other than a temptation? And if it is possible but the single individual
then made a mistake, what salvation is there for him? He suffers all the
pain of the tragic hero, he destroys his joy in the world, he renounces
everything and perhaps at the same moment blocks himself from the
sublime joy which was so precious to him that he would buy it at any
price. The observer cannot understand him at all, nor confidently rest
his eyes upon him.29

It is by contemplating this possibility of radical error in Abraham’s

actions, a possibility that does not exist for the tragic hero, that
Abraham becomes a figure of ‘fear and trembling’,30 as one looks on
at his action with dread. We can miss this, de Silentio argues, because
we just think of the result of what actually happened, and how in
the end things worked out well, as the moral value in what God was
intending here is made clear again (God was trying to test Abraham,
to work out the extent of his faith, to demonstrate his disapproval of
child sacrifice, and so on). But for Abraham at the time, of course, this
was not clear, and hence the awesome nature of his decision, and its
apparent absurdity – where, de Silentio notes, a similar uncertainty

28
Cf. Kant, Relig 6:99 note (p. 134): ‘[I]f an alleged divine statutory law is opposed to a
positive civil law not in itself immoral, then is there cause to consider the alleged div-
ine law as spurious, for it contradicts a clear duty, whereas that it is itself a divine com-
mand can never be certified sufficiently on empirical evidence to warrant violating on
its account an otherwise established duty’; and also Relig 6:186–7 (pp. 203–4) and CF
7:63 note (p. 283).
29
FT 3:110 (p. 53). 30 Cf. FT 3:111 (p. 53).
 Kierkegaard’s critique of Hegel 185

related to those who first had faith in Christ, and to the actions of the
Apostles.31
In the second Problem (‘Is there an absolute duty to God?’), de
Silentio is again concerned to show how a Hegelian ethics makes faith
impossible. In this ethics, he allows, one might say that in some sense
‘every duty, after all, is duty to God’; but the content for these duties
really comes from the moral values inherent within ethical life, so that
the appeal to the divine in fact acts nothing: ‘God becomes an invis-
ible vanishing point, an impotent thought, his power being only in the
ethical which fills all existence’.32 There is thus no inner moment of
fateful decision in the light of the divine, and thus there is no faith:
‘Faith, on the contrary, is this paradox, that inwardness is higher than
outwardness’.33 De Silentio makes clear that by the ‘inwardness’ of faith
here he does not mean anything like mystical feeling, which he agrees
that philosophy would be right to ‘get beyond’; rather, he means this
vital connection with the absurd: ‘Faith is preceded by a movement of
infinity; only then does faith commence, unexpectedly, by virtue of
the absurd’.34
As a knight of faith, therefore, Abraham recognises a duty to God
here, which cannot be given any grounding in the duties of ethical
life. But this means it cannot be related to any social ends – where
normally we would take it that if an action is not so related, then it
is grounded in self-interest instead. However, that is not the basis of
Abraham’s action, which comes from his acknowledgement of a duty
to God. Here, then, we have a duty that is nonetheless not mediated by
the ‘universal’ of Sittlichkeit, by an appeal to the ‘common good’. The
paradox involved in thinking of Abraham’s action as a duty to God,
therefore, is that he is not acting for the general good (and so is acting
egoistically?), but he is acting to sacrifice all he holds dear (and so is

31
FT 3:115 (p. 58; translation modified): ‘One is moved, one returns to those beautiful
times when sweet, tender longings lead one to the goal of one’s desires, to see Christ
walking about in the promised land. One forgets the anxiety, the distress, the para-
dox. Was it so easy a matter not to make a mistake? Was it not appalling that this person
who walked among others was God? Was it not terrifying to sit down to eat with him?
Was it so easy a matter to become an apostle? But the outcome, the eighteen centur-
ies, it helps; it lends a hand to that paltry deception whereby one deceives oneself and
others. I do not feel brave enough to wish to be contemporary with such events, but
for that reason I do not judge harshly of those who made a mistake [and doubted that
Jesus was God] nor slightly of those who saw things rightly’.
32
FT 3:117 (p. 59; translation modified).
33
FT 3:117 (p. 60). 34
FT 3:118 (p. 61).
186 Understanding Mor al Obligation

not acting egoistically?) – a paradox that can be resolved by recognis-

ing a good beyond the general good that forms the basis of a duty to
God alone, who cannot then be reduced to a ‘vanishing point’.35
De Silentio underlines how such duties to God can go beyond our
civic duties by quoting from Luke 14:26, and the ‘hard saying’: ‘If any
man cometh unto me and hateth not his own father and mother and
wife and children and brothers and sisters, yea, and his own life also,
he cannot be my disciple’.36 De Silentio resists all attempts to ‘soften’
these words; on the other hand, he does not take them to mean lit-
eral hatred either, as Abraham would not be Abraham unless he loved
Isaac. What he is doing, however, is something that from any social
conception of ethics must appear to be one of hatred and not of love,
of murder and not of cherishing – just as the disciple may be required
to renounce family and friends if asked to devote his life to God. De
Silentio therefore also resists the thought that perhaps we can find
some social good here, such as the good of the Church,37 as in this
case we would have a tragic hero again, and not a knight of faith. De
Silentio concludes the second Problem with further reflections on the
difference between the two, and how the certainties of the former
make their position relatively easy compared to the latter, notwith-
standing the undoubted sacrifices required of both.
In the third and final Problem (‘Was it ethically defensible of
Abraham to conceal his undertaking from Sarah, from Eliezer, from
Isaac?), de Silentio treats at some length a feature of Abraham’s pos-
ition that he has touched on in earlier discussions: namely, how that
position necessarily isolates him from other people, and cuts him off
from normal communication with them. This, again, relates to the

35
FT 3:120 (p. 62): ‘Thus if one sees a person do something that does not conform to the
universal, one says that he hardly did it for God’s sake, meaning thereby that he did it
for his own sake. The paradox of faith has lost the intermediate factor, i.e. the univer-
sal. On the one hand, it is the expression for the highest egoism (doing the frightful
deed for one’s own sake); on the other hand, it is the expression for the most absolute
devotion (doing it for God’s sake). Faith itself cannot be mediated into the universal,
for it is thereby annulled’.
36
Cf. FT 3:120–1 (p. 63).
37
Cf. FT 3:123 (p. 65): ‘Furthermore, the passage in Luke must be understood in such a
way that one perceives that the knight of faith has no higher expression of the univer-
sal (as the ethical) at all in which he can save himself. If we thus let the church require
this sacrifice from one its members, then we have only a tragic hero. For the idea of
the church is not qualitatively different from that of the state, inasmuch as the single
individual can enter into it by a simple mediation’.
 Kierkegaard’s critique of Hegel 187

deep uncertainty concerning the value of what he is about to do, an

uncertainty that does not attach to the actions of the tragic hero. The
latter can point to a generally recognised good that would be realised
in his sacrifice, whereas the former cannot, and is therefore conscious
that in the eyes of others it cannot be explained or justified. Given
this opacity concerning the moral value of his action, he does not and
cannot expect to be able to convince others of the worthiness of what
he is doing: he is alone, and cannot expect anyone to follow him, as
in this situation, each must judge how things stand for themselves. As
de Silentio has put the point in the second Problem: ‘Whether the
single individual is now actually situated in a state of temptation or is a
knight of faith, only the individual himself can decide’.38
In the third Problem, therefore, the contrast with the Hegelian pos-
ition is that nothing in Hegel’s conception of moral value can prevent
the grounds of an ethical action from being transparent, so that here
Abraham’s inability to communicate his purposes to Sarah, Eliezer and
Isaac would be a sign that he is in the wrong. But in fact the paradox
of Abraham’s position is that he is in the right, but he cannot say what
its rightness consists in or offer any ethical grounding for it, because
God has only commanded him to sacrifice Isaac, without vouchsafing to
him the reason why, where from the human perspective those reasons
are utterly opaque. Thus, if Abraham were asked to justify his actions,
he could not – just like the person who does what is wrong, but where
that incommunicability is not a sign that the act actually is wrong. The
opacity of the moral value in question, therefore, renders the action
something that must be concealed, while the Hegelian, who has no such
opacity within his account, cannot make sense of this, and so must mis-
takenly take Abraham’s silence to show that he is acting in moral error.
De Silentio again contrasts Abraham’s position to that of the tra-
gic hero, who can speak and explain his actions, and also to other
cases where silence might be permitted on grounds that do not apply
to Abraham – for example, where that silence would save somebody
else. Abraham’s silence has an altogether deeper source, based on the
transcendent grounding of what he is called upon to do, where that
grounding is something he cannot articulate to others by explaining
to them what makes his actions right. The tragic hero, by contrast,
‘ought not to be afraid of having overlooked anything’,39 and so can

38
FT 3:127 (p. 69; translation modified).
39
FT 3:159 (p. 100).
188 Understanding Mor al Obligation

explain what he takes to be the moral value in what he is doing, even if

others might then disagree with him about that. By contrast, Abraham
understands that if he did try to explain himself to Sarah, Eliezer,
and Isaac, his action would inevitably start to seem unwarranted in
his eyes, in a way that would constitute a temptation to do wrong. It is
this threat that compels him into silence: where normally it is the pos-
sibility of silence that allows us to contemplate wrongdoing, here it is
the possibility of communication, as the attempt at public justification
makes clear exactly how difficult it is to supply.
I have argued, therefore, that Fear and Trembling involves a critique
of Hegelian ethics from the perspective of faith, and the claim that this
requires us to accept a divine command account of obligation instead.
Within Sittlichkeit, the most that one can be is a tragic hero, because
one’s duties are transparently grounded in the moral values recog-
nised within the social order. If we are to understand how Abraham
could be a knight of faith, however, he must find himself in a situation
where this sort of transparency breaks down, which can only occur if
God is seen as the source of obligation for him, and not society. For it
is possible to imagine God grasping the good in a way that we cannot,
and so putting us under obligations in a way that are opaque from
the human perspective, and which can therefore suspend the nor-
mal sense of where our duties lie. Only in this manner, Kierkegaard is
arguing, can we put Abraham in the right relation to the absurd, and
so see him not just as a tragic hero, but as a knight of faith.
In this way, therefore, we can do justice to the radical and disturb-
ing implications that Kierkegaard clearly wanted to draw out from the
Abraham story, without needing to commit Kierkegaard to a strongly
voluntarist version of the divine command account in order to do
so. As the contrast with Hegel has, I hope, made clear, it is sufficient
to contrast their positions as accounts of obligation, and so to treat
Kierkegaard’s position as an ‘intermediate’ divine command theory,
which views obligatoriness as constituted by God’s will, rather than
the good and the right as such. For, in the account I have offered,
the radical nature of Kierkegaard’s account comes from allowing the
good and the right to be beyond our cognitive grasp, not from the fact
that they depend on the arbitrary determination of the divine will.40

Along similar lines, Philip Quinn has argued that even if one abandons radical vol-
40

untarism in favour of some sort of ‘modified’ divine command theory of the sort pro-
posed by Robert Adams, nonetheless ‘[t]o preclude the possibility of a credible divine
 Kierkegaard’s critique of Hegel 189

There is, however, a way in which Kierkegaard’s critique of Hegel

in Fear and Trembling is dialectically limited (though it would scarcely
have struck Kierkegaard as such): namely, that it rests on the strat-
egy of showing that Hegel’s ethical position is in the end inadequate,
because it cannot do justice to the nature of faith. Now, against Hegel
himself, this is arguably an effective strategy, as Hegel did claim to be
able to accommodate the latter – so if Kierkegaard has succeeded in
showing that in fact his position in ethics makes this impossible, then
this might well be considered an important consideration against it.
But for those who are not committed to this part of the Hegelian pro-
gramme, and who also take matters of faith and religion rather lightly,
it may appear that the Kierkegaardian strategy can easily be shrugged
off: for they will feel that even if Kierkegaard is right that endorsing
the social command account of obligation comes at a cost in religious
terms, this is of little significance to them, as it is a price they are happy
enough to pay. In other words, it may be felt that in Fear and Trembling
Kierkegaard has not yet given us a critique of the Hegelian position
that will have much purchase on those who lack Kierkegaard’s com-
mitment to the religious life. For this, there is perhaps more reason to
turn to Either/Or.

Kierkegaard’s critique of Hegel in Either/Or

Either/Or is also a pseudonymous work, and is also published by
Kierkegaard in 1843, a few months before Fear and Trembling. It consists
of two volumes, put together by one ‘Victor Eremita’, where the papers
of an aesthete known as ‘A’ make up the first volume, while those of
a person known as ‘B’ make up the second one, where ‘B’ is revealed
to be a Judge William. His papers mainly consist of two long letters
that he has sent to A, together with a concluding sermon that he has
received from a country priest. Broadly speaking, the structure of the
work can be taken to reflect Kierkegaard’s three ‘spheres of existence’,
namely the aesthetic (volume 1), the ethical (William’s papers in vol-
ume 2), and the religious (the sermon).41 As with Fear and Trembling, the

command to practice human sacrifice would be to attempt to domesticate the tran-

scendent, which is at odds with its fearful and dangerous character’ (Quinn 2002:
465). Cf. also Wainwright 2005: 205–8.
41
I say ‘broadly speaking’, because, as we shall see, there are significant religious elem-
ents in Judge William’s ethical position; the question is, however, whether those elem-
ents take the religious dimension sufficiently seriously.
190 Understanding Mor al Obligation

question is again whether the ethical can be developed in a way that is

prior to the religious, but where this is not here posed by making the
matter of faith quite so central, but rather the question of the ethical
itself – what are the limitations of an ethics conceived of in this way?
Judge William presents himself as critical of some aspects of what
he takes to be Hegelian doctrine, most especially of its treatment of
contradiction and of its attendant inability (as he sees it) to take indi-
vidual choice seriously.42 Nonetheless, perhaps in one of those ironies
of failed self-knowledge that Kierkegaard delighted in as an author,
William’s own ethical outlook is shown to be that of a Hegelian.43 This
can be seen primarily in the way in which Judge William acts as a
spokesman for the family, to the value of which he seeks to win round
the A of volume 1. In doing so, the Judge was not only following Hegel
in treating the family as a central institution of ethical life, but he also
defends it as such in clearly Hegelian terms, through the working of
‘love’s dialectic’,44 whereby the romantic love that A champions is ‘sub-
lated’ within the familial form (where it may not be coincidental to all
this that William was also Hegel’s second name).
As the first stage in this dialectic, A’s romantic love is characterised
in terms of immediacy:
But first I shall indicate the characteristics of romantic love. One could
say in a single word: it is immediate. To see her and to love her would
be one and the same, or even though she saw him but one single time
through a crack in the shuttered window of a virgin’s bower, she neverthe-
less would love him from that moment, him alone in the whole world.45

Romantic love involves no rational reflection, but only the imme-

diacy of feeling or impulse, where it is then related to the sensuous
experience of beauty. At the same time, however, it takes itself to be
distinguished from mere lust or carnal desire, so that it ‘has an analogy

42
This mostly occurs in the opening parts of the Judge’s second letter, on ‘The Balance
Between the Aesthetic and the Ethical in the Development of the Personality’. This
critical element is reflected also in the comments that the pseudonymous author of the
Concluding Unscientific Postscript makes about ‘the ethicist in Either/Or’: see CUP 7:438
note (p. 503).
43
Jon Stewart has also emphasised the connections with Hegel here, criticising attempts
by Niels Thulstrup to minimise these: see Stewart 2003: 225–9. For Thulstrup’s treat-
ment, see Thulstrup 1980: 324–8.
44
EO 2:17 (p. 18). For a comparative study, see Perkins 1967.
45
EO 2:18–19 (pp. 19–20).
 Kierkegaard’s critique of Hegel 191

to the moral in the presumed eternity, which ennobles it and saves it

from the merely sensuous’.46 The question Judge William presses, how-
ever, is whether it can maintain this distinction in a stable way, unless
in incorporates more of morality: for in fact, through being based on
feeling, it is no more eternal and enduring than lust itself. It may then
be said, however, that the eternal consists in ‘living in the present’,
whereby the moment of union is treated as if it lasts forever, where
marriage is then rejected as a prolonging of this precious instant into
the tedium of days and years together – an attitude encapsulated in
the Byronic sentiment that ‘love is heaven, marriage is hell’.47 While
a naïve romantic love is therefore prepared to accept the marriage
ceremony as a joyful ‘festivity’, this more sophisticated form of roman-
tic love dreads the thought of love going cold, and so insists that if
marriage occurs, it can always be terminated if this happens; the eter-
nal is therefore lost again, as divorce is possible on this basis at any
moment. Judge William diagnoses a deep melancholy here, a kind of
paranoid fear of abandonment and loss of affection, together with a
morbid sense that perhaps one could do better elsewhere: ‘What can
one depend upon; everything may change; perhaps even this being I
now almost worship can change; perhaps later fates will bring me in
contact with another being who for the first time will truly be the ideal
of which I have dreamed’.48 Likewise, the lover may fear that they may
themselves change and also leave the beloved in a hopeless relation-
ship, and so again shun marriage. In a typically Hegelian reversal of
the dialectic therefore, romantic love that started with the eternal and
with joy has ended with the transitory and with despair.
In response, the reaction may then be to turn from the romantic
position to its opposite, which moves from immediacy and feeling to
mediation and reflection, as the individual contemplates the advan-
tages to be gained from a marriage of convenience, with its basis in ‘the
understanding’ and ‘common sense’. This may appear to be a more
ethically satisfactory position than romantic love, but in fact Judge
William questions this: ‘Insofar as it has neutralized the sensuous in
marriage, it seems to be moral, but a question still remains whether
this neutralization is not just as immoral as it is unesthetic’.49 It has also

46
EO 2:20 (p. 21).
47
EO 2:21 (p. 22).
48
EO 2:24–5 (p. 25).
49
EO 2:25–6 (p. 27).
192 Understanding Mor al Obligation

not solved the problem of the relation to the eternal, for a marriage of
convenience may be dissolved at any time, when it outlives its useful-
ness. Judge William brings out the dialectical tension between the two
positions so far discussed by quoting from ‘a commonsensical little
seamstress’ in a ‘recent play’, who ‘makes the shrewd comment about
fine gentlemen’s love: They love us, but they do not marry us; they do
not love the fine ladies, but they marry them’.50
Judge William then sets out, in a Hegelian manner, to find some
way of achieving a ‘mediated immediacy’ here, and thus of ‘sublating’
both positions in a higher one that retains what is worthwhile in each
while overcoming their ‘one-sidedness’51 – for otherwise, we will be left
with a lasting antithesis between inclination and duty, feeling and rea-
son, and the temporal and the eternal. As Judge William puts it:
The question remains whether the immediate, the first love, by being
caught up into a higher, concentric immediacy, would not be secure
against this scepticism so that the married love would not need to
plough under the first love’s beautiful hopes, but the marital love would
itself be the first love with the addition of qualifications that would not
detract from it but would ennoble it. It is a difficult problem to pose,
and it is of the utmost importance, lest we have the same cleavage in the
ethical as in the intellectual between faith and knowledge.52

Judge William thus takes as his ‘task’53 the attempt to show how this
mediation between romantic love and marriage can be achieved, and
the opposition that A sees between them can be overcome. It is this
task that he carries out in the rest of his long first letter.
Clearly, Judge William’s position here echoes Hegel’s own, where, as
we have seen,54 he too argues that the immediacy of love as mere feel-
ing must be transmuted, and given a more ‘ethical character’ in mar-
riage, in which love can take on a higher and more satisfactory form.
Likewise, Judge William’s strategy is to argue that while first love can
be a unity of sensuousness and spirituality, freedom and necessity, and
eternity and temporality, it is so only in ‘immediate’ form, where this is
more properly realised when love is contained within marriage:

50
EO 2:26 (p. 28).
51
Cf. EL §§79–82, 8:168–79 (pp. 125–33).
52
EO 2:28 (p. 29).
53
Cf. EO 2:29 (p. 31): ‘So you see the nature of the task I have set for myself: to show that
romantic love can be united with and exist in marriage – indeed, that marriage is its
true transfiguration’.
54
See above, Chapter 4, §Love vs Law.
 Kierkegaard’s critique of Hegel 193

[First love] is implicit in the unity of contrasts that love is: it is sensuous
and yet spiritual; it is freedom and yet necessity; it is in the instant, is
to a high degree present tense, and yet it has in it an eternity. All this
marriage also has; it is sensuous and yet spiritual, but it is more, for the
word ‘spiritual’ applied to the first love is closest to meaning that it is
psychical, that it is the sensuous permeated by spirit. It is freedom and
necessity, but also more, for freedom applied to the first love is never-
theless actually rather the psychical freedom in which the individuality
has not yet purified itself of natural necessity … Even more than the
first love, it is an interior infinitude, for marriage’s interior infinitude
is an eternal life.55

Judge William then considers a series of objections to this view. The

first set of objections centres on the ends of marriage (that it is no
more than a ‘school for character’, a means of producing children,
or of acquiring a home), where the concern is that these all deprive it
of any relation to love.56 He then focuses on the marriage ceremony,
and its apparently empty rituals, where the congregation and the pub-
lic nature of marriage more generally seem to intrude on the privacy
and inwardness of love. He also considers the threat to love within
marriage posed by the kind of external trials that life can throw at the
married couple, as well as the more internal threat of the dread hand
of custom and familiarity, and of duty.
Judge William in all three cases attempts to defend marriage against
the objections that he imagines A as raising, where it is the final criti-
cism concerning duty that is of particular concern to us here. For,
again, the Judge adopts a Hegelian line on this question, arguing that
it is a mistake to see any antithesis between duty and inclination, and
thus to assume that because it involves duties, marriage must in the
end suppress love: ‘So, then, you regard duty as the enemy of love, and
I regard it as its friend … For me, duty is not one climate, love another,
but for me duty makes love the true temperate climate, and this unity
is perfection’.57 The mistake A is making, William argues, is to think
(in a Kantian manner) that because love is an emotional state and

55
EO 2:55–6 (pp. 60–1). Cf. also 2:42 (p. 45), where the argument for this conclusion
starts: ‘So we turn back to the first love. It is the unity of freedom and necessity. The
individual feels drawn by an irresistible power to another individual but precisely
therein feels his freedom. It is a unity of the universal and particular; it has the univer-
sal as the particular even to the verge of the accidental. But all this it has not by virtue
of reflection; it has this immediately’.
56
Hegel deals with these aspects of marriage in PR §§161–80, 7:308–38 (pp. 200–18).
57
EO 2:133 and 134 (pp. 146 and 147).
194 Understanding Mor al Obligation

something one feels willingly, it cannot be commanded;58 but in fact,

in marriage one’s duty is to love in this way, so that ‘[d]uty is always
consonant with love’.59
Nonetheless, Judge William recognises, A may reply that while duty
makes sense at a civic level, it makes no sense within a purely personal
relationship, where ‘if I form a close friendship with another person,
love is everything here. I acknowledge no duty’.60 William responds
to this objection by arguing that even when it comes to civic duties,
an individual may find in them his highest realisation, and yet it still
makes sense to call them his duties – not because he does not want to
carry them out, but because they are what are expected of him within
society.61 Likewise, when it comes to love, an individual can expect
certain things of himself and blame himself for failing in them, so
that it also makes sense to think of love as a duty,62 without assuming
that it becomes so because it goes against his inclinations; this would
indeed make a duty to love into a contradiction, in the manner that A
mistakenly takes it to be. Judge William thus concludes this part of the
discussion, on which the philosophical reflections contained in this
first letter end, with a final statement of his position on this issue:
But I have not been afraid of duty; it has not appeared to me as an enemy
that would disturb the fragment of joy and happiness I had hoped to
rescue in life, but it has appeared to me as a friend, the first and only
confidant in our love.63

58
Cf. MM 6:401 (p. 530): ‘Love is a matter of feeling, not of willing, and I cannot love
because I will to, still less because I ought to (I cannot be constrained to love); so a duty
to love is an absurdity … What is done from constraint … is not done from love’.
59
EO 2:135 (p. 149). 60
EO 2:136 (p. 150).
61
EO 2:137 (pp. 151–2): ‘[Y]ou think that all the rest of life can be construed within
the category of duty or its opposite and that it has never occurred to anyone to apply
another criterion; marriage alone has made itself guilty of this self-contradiction. You
cite as an example the duty to one’s occupation and think that this is a very appropri-
ate example of a pure duty-relationship. This is by no means the case. If a person were
to view his occupation merely as the sum total of assignments he carries out at specific
times and places, he would demean himself, his occupation, and his duty. Or do you
believe that such a view would make for a good public official? Where, then, is there
room for the enthusiasm with which a person devotes himself to his occupation, where
is there room for the love with which he loves it?’.
62
EO 2:138 (p. 152): ‘In marriage, however, the internal is primary, something that can-
not be displayed or pointed to, but its expression is precisely love. Therefore, I see no
contradiction in its being required as duty, for the circumstance that there is no one
to supervise is irrelevant, since he can indeed supervise himself’.
63
EO 2:138 (p. 153).
 Kierkegaard’s critique of Hegel 195

Likewise, in the second letter to A, where Judge William moves beyond

his discussion of marriage and considers the contrast between the aes-
thetic and the ethical more broadly, this is a theme that he comes
back to frequently. Once again, William expresses himself in Hegelian
terms, contrasting his position with a more Kantian one:
The ethical is the universal and thus the abstract. That is why in its per-
fect abstraction it is always interdictory. Thus the ethical takes the form
of law … When the ethical becomes more concrete, it crosses over into
the category of morals … But the ethical is still abstract and cannot be
fully actualized because it lies outside the individual. Not until the indi-
vidual himself is the universal, not until then can the ethical be actual-
ized … The person who views life ethically sees the universal, and the
person who lives ethically expresses the universal in his life. He makes
himself the universal human being, not by taking off his concretion, for
then he becomes a complete non-entity, but by putting it on and inter-
penetrating it with the universal … The ethical individual, then, does
not have duty outside himself but within himself.64

Judge William therefore contends in a Hegelian manner that there is

no fundamental antithesis between the individual and the ethical, no
element of Kantian struggle: ‘So the personality does not have the eth-
ical outside itself but within itself and it bursts forth from this depth’.65
Judge William also follows Hegel in stressing how life within Sittlichkeit
will bring self-realisation to the individual, as he is able to exist as a
concrete particular within the social whole.66
Judge William then considers three possible objections to the pic-
ture that he has presented. The first is a worry about relativism: for,
‘[w]hen living for the fulfillment of duty is made a person’s task in
life’, then ‘what is often pointed out is the scepticism that duty itself is
unstable, that laws can change’, where this issue is said to apply par-
ticularly to ‘civic virtues’. Judge William admits that this may be true
up to a degree, but only within limits, for ‘[t]his scepticism, however,

64
EO 2:229–30 (pp. 255–6) 65 EO 2:230 (p. 257).
66
EO 2:235 (p. 262): ‘The person who has ethically chosen and found himself possesses
himself defined in his entire concretion. He then possesses himself as an individual
who has these capacities, these passions, these inclinations, these habits, who is subject
to these external influences, who is influenced in one direction thus and in another
thus. Here he then possesses himself as a task in such a way that it is chiefly to order,
shape, temper, inflame, control – in short, to produce an evenness in the soul, a har-
mony, which is the fruit of the personal virtues’.
196 Understanding Mor al Obligation

does not apply to the negative aspect of morality’ (that is, certain fun-
damental ethical prohibitions), ‘for that continues unchanged’.67
The second objection turns out to be more fundamental for the
critique of Hegelianism that is to follow. This is that duty is something
beyond us, something that we may be utterly unable to do, because it
demands too much of us: ‘The duty is the universal. What is required
of me is the universal; what I am able to do is the particular’.68 Judge
William argues, however, that from a Hegelian perspective, this worry
is misplaced. For while doing one’s duty in a general sense is aimed
at the good of all, and while none of us can achieve this as individ-
uals, I can still be capable of doing my particular duties, as the things
required of me by virtue of my place within the wider whole, where it
is at this wider level that the good of all is realised:
I never say of a man: he is doing his duty or duties; but I say: He is doing
his duty; I say: I am doing my duty, do your duty. This shows that the indi-
vidual is simultaneously the universal and the particular. Duty is the
universal; it is required of me. Consequently, if I am not the universal,
I cannot discharge the duty either. On the other hand, my duty is the
particular, something for me alone, and yet it is duty and consequently
the universal. Here personality appears in its highest validity. It is not
lawless; neither does it itself establish the law, for the category of duty
continues, but the personality takes the form of the unity of universal
and particular. That this is so is clear; it can be made understandable to
a child – for I can discharge the duty and yet not do my duty, and I can
do my duty and yet not discharge the duty.69

Judge William therefore presents it as a fundamental feature of his

Hegelian position, and also one of its great virtues, that it treats our
ethical obligations as attainable by the individual; for otherwise, A
would be right in viewing the ethical standpoint as an alienated form
of life, in which we could no longer feel ‘at home in the world’.
A third objection William considers is a historicist worry that if we
endorse this Hegelian picture, we will have to assess people’s ethical
standing by the moral practices of their own society; but these can vary
greatly, so that we might have to say that those who do unspeakable
things are nonetheless to be considered good, because that is what
their practices encouraged and sanctioned. Judge William’s response

67
EO 2:236 (p. 263).
68
EO 2:236 (p. 263).
69
EO 2:236 (pp. 263–4).
 Kierkegaard’s critique of Hegel 197

is twofold. On the one hand, he argues, we should be careful not

to exaggerate the historical relativity that can be found, so that for
example, ‘there has never been a nation that believed that children
should hate their parents’.70 On the other hand, he allows that prac-
tices within societies may still vary greatly, so that ‘savages [have] prac-
tised the custom of putting their aged parents to death’.71 However,
he claims, we should be careful about assuming that the individuals
involved in these practices were therefore evil: for what they were
intending to do thereby was often good (such as ensuring the parents’
souls reached heaven, for example), so that the divergence is about the
facts (concerning how the afterlife works and whether there is one),
not a deep ethical variation which cannot be sufficiently marked on
the Hegelian account.72 At the same time, Judge William argues, if one
does not accept the moderate historicism of his position, then one will
be searching for an ‘Archimedean point from which one can lift the
world’, a perspective outside all our historical ethical practices from
which to judge them – but this, the Judge urges, is as hopeless as look-
ing for the source of the Niger, where it must be admitted that ‘no one
knows where it is’.73
After prematurely promising A that ‘[h]ere I shall bring my theor-
izing to an end’, by then recounting his own childhood experience
of the inculcation in him of a sense of duty, Judge William finally
takes up the theme of the relation between obligation and beauty.
Where A sees a clear antithesis here, the Judge again follows Hegel’s
more Schillerian outlook, and insists that the latter can be found in
the former, even in the duty to work. Once more, William insists that
this is so because within the social whole within which the individual
works, he also contributes to his self-realisation, so that work need not
be seen as alien to him or his nature.74 Having accepted this view of

70
EO 2:237 (p. 265). Kierkegaard is doubtless expecting us to have in mind the passages
from Luke that were quoted in Fear and Trembling which we have cited earlier, concern-
ing the ‘remarkable teaching’ that duty to God may require the hatred of family.
71
EO 2:237–8 (p. 265).
72
‘The ethical always resides in this consciousness [of intending to do good or evil],
whereas it is another question whether or not insufficient knowledge is responsible’
(EO 2:238 (p. 265; translation modified)).
73
EO 2:238 (p. 265).
74
Cf. EO 2:262 (p. 292): ‘The ethical thesis that every human being has a calling
expresses, then, that there is a rational order of things, in which every human being, if
he so wills, fills his place in such a way that he simultaneously expresses the universally
human and the individual’. For Hegel on work, cf. PR §§196–8, 7:351–3 (pp. 231–3).
198 Understanding Mor al Obligation

work, and also the earlier one of marriage, Judge William’s ‘hero’ is
ready to settle equably into the kind of contented bourgeois life that
Hegel too treats as an ideal within Sittlichkeit:
Hence our hero lives by his work; his work is also his calling; there-
fore he works with a will. Since it is his calling, it places him in touch
with other people, and in carrying out his task he accomplishes what
he would wish to accomplish in the world. He is married, content in his
home, and time runs smoothly for him.75

And, as the Judge later in effect confesses, ‘our hero’ is Judge William
himself, who concludes by putting forward his own life to A as a para-
digm of ethical existence and the kind of satisfactions it can bring.76
Before turning finally to the critique that Kierkegaard sets up in
the last part of Either/Or, it is important to the force of that critique to
emphasise that Judge William has also presented himself as includ-
ing a religious dimension within his perspective on the ethical, where
this is comfortably accommodated within his conception of civic life.77
Judge William thus presents himself as a Christian, who sees marriage
and the having of children in religious terms.78 He therefore accepts
A’s challenge to show that, like the ethical, Christianity is also not
inimical to the aesthetic, properly conceived, and so is not opposed to
all notions of sensuousness and beauty.79 He also accepts A’s challenge
concerning the Christian view of sin, pointing out that the Church
does allows marriage, while the Judge clearly finds any idea of ori-
ginal sin hard to accept, particularly as this relates to the individual
rather than just human kind in general.80 He also rejects mysticism as
a model for the Christian life,81 in large part because this removes an
individual from their civic responsibilities: ‘It is especially as a married
man and as a father, that I am an enemy of mysticism’.82 While it can-
not be gone into fully here, in his attitude to the relation between the

75
EO 2:293–4 (p. 305). 76
 See EO 2:290–1 (pp. 323–4).
77
Cf. Rudd 1993: 141–3, where Rudd comments that for the Judge, ‘his religion is a sort
of metaphysical epiphenomenon of his ethics – a halo on his head, but no part of his
body’ (p. 141). Norman Lillegard has protested against views of William that treat him
as ‘an insipid “Hegelian” bourgeoisie who essentially lives without God while hiding
in a civil righteousness’; but even he admits that it is ‘nonetheless true that he sees
the relation to God as always being mediated by and thus limited by the ethical’: see
Lillegard 1995: 108. On this issue, see also Watkins 1995.
78
EO 2:27–9 (pp. 28–29); 2:51–4 (pp. 55–8); 2:82–6 (pp. 89–94); 2:291 (p. 324).
79
EO 2:44–6 (pp. 48–50). 80
EO 2:83–4 (pp. 91–2).
81
EO 2:216–25 (pp. 241–50). 82
EO 2:219 (p. 244).
 Kierkegaard’s critique of Hegel 199

religious and the ethical, Judge William’s approach once again mir-
rors Hegel’s to a large degree, and shares many of the latter’s assump-
tions concerning the place of religion within social life.83
Now, as other commentators have noted, it is against this back-
ground that we need to understand the concluding part of Either/Or,
and its role as a critique of the foregoing.84 This section takes the form
of a sermon sent to Judge William by a pastor in Jutland who was a
student friend; William now passes it on to A some time after he has
received the Judge’s letters. The juxtaposition here is startling, and
tells us a good deal about what will follow in the Works of Love.
The title of the sermon is ‘The Edification that Lies in the Thought
That in Relation to God We Are Always in the Wrong’.85 This is an
immediate challenge to Judge William’s Hegelian outlook, not only in
violating Hegel’s dictum that ‘philosophy must beware of the wish to
be edifying’,86 but also in suggesting that our duty may be beyond us.
As we have seen, this was a concern about the ethical that was raised by
A, who feared that it might involve a constant process of Kantian self-
overcoming, in which the individual must set aside their own desires
and interests in such a way to act as morality requires. In response,
Judge William had argued, because within civic life the duties of each
particular individual can be limited and yet still realise the good of
the whole, the ethical is therefore not so strenuous as to make its real-
isation difficult or impossible for the individual, in a way that would
result in a clash between our moral and non-moral selves. Likewise,
for Hegel, ‘[a]n ethical order provides individuals with a generally
satisfying mode of life, so that they are seldom called upon to make
great personal sacrifices for others’.87 This does not mean, of course,
that they live merely within the range of their narrow self-interest, as
those interests will include a concern with others and their well-being,
built on a variety of social ties through the family, civil society, and
the state; but on the other hand, it does mean that there are limits to

83
Cf. PR §270, 7:415–31 (pp. 291–303).
84
Cf. Westphal 1998: 106–7, where he notes that ‘Either/Or is Janus faced by virtue of this
ending, which in the fewest of words puts the long, Hegelian exposition of the ethical
in question’. See also Perkins 1995 and Law 1995. Curiously, in his very full discussion
of the Hegel/Kierkegaard relation, Jon Stewart does not mention this part of Either/
Or, and thus presents the latter as a ‘pro-Hegelian’ text.
85
I have preferred the translation of ‘Edification’ to ‘Upbuilding’, and will continue to
use this throughout.
86
PS 3:17 (p. 6). 87
Wood 1999: 210.
200 Understanding Mor al Obligation

the demands placed upon them as individuals, as those demands are

mediated through the workings of the community as a whole, thus
reducing the requirements on each to a manageable level, one that is
consistent with the concerns they can be expected to have developed
as virtuous individuals living within a well-ordered society that cares
for its citizens. This is why Judge William thinks he can satisfactorily
answer the aesthete’s concerns about duty, because he thinks he can
rightly claim that the moral demand on us is not so great as to outstrip
our human capacity to meet it.
By characterising Judge William for us, however, and then juxta-
posing his position with that of his pastor friend as well as with that
of the aesthete, Kierkegaard seems to have wanted to highlight what
he sees as the limitations of the Judge’s Hegelian outlook: for it is pre-
cisely the ethical undemandingness of William’s life that this juxtapos-
ition shows up, and the extent of his moral complacency. Again, this is
not because William is any sort of straightforward egoist or amoralist,
who denies that morality has any call on him at all; it is just that he is
sure that this call is a moderate one, that can be met relatively easily
within the terms of his bourgeois life. He thus feels every confidence
that if he satisfies his place in society, his ethical requirements will
have been honoured, which is precisely why he thinks he can calm A’s
fears concerning the ethical. At the same time, he is uncomfortable
with the suggestion that the Christian outlook somehow goes beyond
this, with its claims about sin, repentance, and grace;88 properly con-
ceived, he clearly hopes, nothing in the Christian outlook threatens
his ethical stance.
What the pastor’s sermon offers, however, is a challenge to the Judge’s
comfortable conception of the moral demand, taking as its theme that
in fact ethically considered, ‘we are always in the wrong’. The text that
the pastor chooses for his sermon is Luke 19:41–8, in which Jesus pre-
dicts the downfall of Jerusalem, and then ejects the merchants and
money-changers from the temple – where the mistake is to confuse
civic virtue with what is required of us in order to ‘live in the right’.89

88
Robert Perkins has emphasised the limited view that Judge William has of these
notions: see Perkins 1995: 215–7.
89
Cf. Hannay 1991: 64: ‘[Jesus’s] point is not, of course, by association to accuse incum-
bents of positions of civic and political responsibility of thievery; it is rather … to bring
out the incompatibility between civically defined virtue … and the notion of a tran-
scendent God as the source and guarantor of personal value or fulfilment’.
 Kierkegaard’s critique of Hegel 201

But, it might be said, isn’t this to ask more of us than can be reason-
ably expected? The pastor is scornful of this sort of response:
If I should speak in a different way, I would remind you of a wisdom you
certainly have frequently heard, a wisdom that knows how to explain
everything easily enough without doing an injustice either to God or
to human beings. A human being is a frail creature, it says; it would be
unreasonable of God to require the impossible of him. One does what
one can, and if one is ever somewhat negligent, God will never forget
that we are weak and imperfect creatures. Shall I admire more the sub-
lime concepts of the nature of the Godhead that this ingenuity makes
manifest or the profound insight into the human heart, the probing
consciousness that scrutinizes itself and now comes to the easy, cozy
conclusion: One does what one can?90

While someone with Judge William’s ethical sensibilities may feel he

has ‘done what he can’, the pastor makes clear that the right attitude is
the ‘dread’ that one has fallen short, that one could have done more,
that one has fooled oneself into thinking that this was all that was
required when in fact more is really demanded. ‘So every more earnest
doubt, every deeper care is not calmed by the words: One does what
one can’.91
The pastor then moves on to consider whether anything edifying
can be found in the thought that ‘as against God we are always in the
wrong’, and if so, how? Being in the wrong pains us, though it may be
edifying to think that this sense of wrongness may then induce us to
be better; but here we are considering the claim that we are always in
the wrong, so this sort of edification will not apply in this case. The
pastor then considers a situation in which you love someone who does
wrong to you; here, he suggests, you might find some edification in
the thought that you would rather it were you who had done wrong.
But still, you don’t believe here that you actually have done wrong,
so that again this does not really fit the position the pastor is consid-
ering. Nonetheless, he suggests, thinking of a relationship in which
love plays an important role is a significant step forward, where what
is in question is love of an infinite being like God. But again, there
is no edification in believing that one is in the wrong by comparing
oneself to God’s infinite goodness and then realising that one always

90
EO 2:310 (pp. 344–5). 91
EO 2:311 (p. 346).
202 Understanding Mor al Obligation

falls short.92 If it is to be edifying, the sense of one’s wrongness must

not come about through reflection in this way: rather, it must come
through the love one feels for God:
Therefore this, that in relation to God you are always in the wrong, is
not a truth you must acknowledge, not a consolation that alleviates your
pain, not a compensation for something better, but it is a joy in which
you win a victory over yourself and over the world, your delight, your
song of praise, your adoration, a demonstration that your love is happy,
as only that love can be with which one loves God.
Therefore this thought, that in relation to God we are always in the
wrong, is an edifying thought; it is edifying that we are in the wrong,
edifying that we are always in the wrong.93

It is this relation to God that Judge William, who thinks he is mostly in

the right,94 fails to attain, and from which he is therefore cut off.
The pastor recognises, however, that this thought that we are always
in the wrong might be taken to be paralysing: for if this is so, why should
we bother trying to act rightly at all, and not simply despair of ethical
action altogether?95 The pastor concedes that if we come to the con-
clusion that we are always wrong by comparing ourselves to the way in
which God is always right, we might indeed become despairing in this
way. However, he has already argued that this is not how the thought of
our wrongness should be seen as coming about; rather, we have come
to this belief out of our love for God, so that it brings with it joy and a
positive sense of potentiality, not gloomy inadequacy and hopelessness:
‘In relation to God we are always in the wrong – this thought puts an
end to doubt and calms the cares; it animates and inspires to action’.96

92
Cf. EO 2:314–5 (p. 350): ‘You acknowledge, then, that God is always in the right, and as a
consequence of that you are always in the wrong, but this acknowledgement did not edify
you. There is nothing edifying in acknowledging that God is always in the right, and con-
sequently there is nothing edifying in any thought that necessarily follows from it. When
you acknowledge that God is always in the right, you stand outside God, and likewise
when, as a conclusion from that, you acknowledge that you are always in the wrong’.
93
EO 2:315–6 (p. 351).
94
Cf. EO 2:213 (p. 237), where Judge William affirms that it is ‘a sign of a high-minded
person and a deep soul if he is inclined to repent’, but on the other hand he thinks it
can be done ‘without too much pondering’.
95
EO 2:317 (p. 353). Hegel was himself always keen to stress the dangers of paralysis in a
certain kind of overwrought moral sensibility: see, for example, his discussion of the
beautiful soul as well as his critique of Kant’s postulates in PS 3:453–94 (pp. 374–409).
96
EO 2:317 (p. 353). There are obvious parallels between what the pastor says here about
the possibility of joy in the face of God’s demand, and what Johannes de Silentio said
about Abraham’s joy in Fear and Trembling.
 Kierkegaard’s critique of Hegel 203

Kierkegaard may therefore be read as setting the pastor’s theologic-

ally informed ethics alongside Judge William’s more Hegelian and
secular one, according to which one ‘does what one can’ within the
duties prescribed by the state – a state that seeks to balance the ‘par-
ticularity’ of our interests with the ‘universality’ of the general good,
and thus does not ask too much of us as individuals. According to the
pastor’s theological ethics, by contrast, one is always ‘in the wrong’,
always falling short of what is required, where nonetheless this remains
an edifying and not dispiriting perspective, in so far as these require-
ments come to us from a loving God.
This contrast is not fully developed or defended by the pastor –
whose brevity marks another contrast with Judge William and his
loquaciousness. Instead, the pastor concludes with what sounds like
a warning of what might happen if a secularised Hegelianism were to
triumph, and the properly theological tradition in ethics were lost:
Could you wish that you might be in the right; could you wish that the
beautiful law which for thousands of years has carried the generation
through life and every member of the generation, that beautiful law,
more glorious than the law which carries the stars on their paths across
the arch of heaven, could you wish that that law would break, an even
more terrible catastrophe than if the law of nature lost its power and
everything disintegrated into dreadful chaos? Could you wish that?97

Here, it is clear, Kierkegaard intends us to see in the pastor’s words a

fundamental challenge to the Hegelian position represented by Judge
William, who had tried to present ethics as going with the grain of
our human capacities, while the pastor sets it at odds with them, in a
way that puts us always in moral error. The challenge here is not only
a religious one (though of course it is also that), but is also based on a
concern with how ethics itself is to be conceived: namely, as a radical
demand that we cannot fulfil on our own, and which therefore only
makes sense when understood in theological terms. It is this challenge,
I will now suggest, that is more fully explored in the Works of Love, to
which we turn next.

97
EO 2:318 (p. 354).
 7

K i e r k eg a a r d’s s olu t ion to t h e

probl e m of mor a l obl ig at ion

Works of Love was published in 1847, and is one of a series of ‘Christian

discourses’ that appeared under Kierkegaard’s own name around this
time. Its central focus is love and the various forms that it can take,
and in particular how love figures in what Jesus gives as the second
commandment, or the so-called ‘royal law’: ‘You shall love your neigh-
bour as yourself’.1 I will suggest in this chapter that one central pur-
pose of Kierkegaard’s extended meditation on this commandment
is to show how the limitations he has identified with the Hegelian
position require us to take up a divine command account of obliga-
tion. Kierkegaard’s strategy, I will claim, is to emphasise how on the
one hand morality asks much more of us than can be accommodated
within the Hegelian picture (which, as we have seen, was already the
theme of the pastor in Either/Or), while on the other hand we can only
make sense of this demandingness if we think we are being asked to
act in that way by God, whose capacity to judge, assist and forgive us
differs from that of any other kind of obligating source. Thus, for
Kierkegaard, only a divine command theory can make intelligible a
morality that expects more of us than the limited requirements to be
found within Judge William’s vision of ethical life, and its expectation
that one merely ‘does what one can’.

The love commandment and the ‘moral gap’

Near the beginning of Works of Love, Kierkegaard writes:

1
Cf. Matthew 19:16–19, Matthew 22:35–40, Mark 12:28–34, John, 15:17, Luke 10:25–28,
James 2:8–9 (it is in James that the commandment is referred to as the ‘royal law’). Cf.
also Leviticus 19:18.

204
 Kierkegaard’s account of obligation 205

Take a pagan who is not spoiled by having learned thoughtlessly to pat-

ter Christianity by rote or who has not been spoiled by the delusion of
being a Christian – and this commandment, ‘You shall love’, will not
only surprise him but will disturb him, will be an offense to him.2

The pagan thinks of love as ‘the play of feelings, drives, inclinations,

and passions, in short that play of the powers of immediacy, that cel-
ebrated glory of poetry in smiles or in tears, in desire or in want’,3 and
thus as something we willingly feel for others on a natural basis. He
will therefore find it curious indeed to speak of a command to love, in
part because he will not see how love, as a feeling, can be commanded
at all,4 but also because he will see no need for the command, in so
far as on his conception of love, it is something we are spontaneously
inclined to feel in any case, and so does not have to be required of us in
this manner. For the pagan, therefore, ‘You shall love … contains this
apparent contradiction: to love is a duty’.5
The perspective of the true Christian on love, however, is not like
that of the pagan: ‘what a difference between this and the earnestness
of eternity, the earnestness of the commandment in spirit and truth,
in honesty and self-denial!’.6 To the Christian, then, it makes sense to
think of love as being demanded of us, because love as he conceives it
goes beyond our natural disposition. However, his outlook will then
seem shocking to the pagan in another way, as it now appears that the
commandment is asking us to ‘overcome ourselves’, raising the ques-
tion of how we can manage to obey it. Looked at from a pagan per-
spective, therefore, the love commandment causes ‘offense’, because it
seems to be requiring of us what cannot be achieved, and so to mock
us and our human capacities.7

2
WL 9:28 (p. 25). 3 WL 9:29 (p. 25).
4
As Hannay notes, this was Kant’s view of ‘pathological love’: see Hannay 1991: 254–62.
Cf. Kant GMM 4:400 (p. 55), MM 6:401 (p. 530), MM 6:410 (p. 537).
5
WL 9:27 (p. 24). 6 WL 9:29 (p. 25).
7
As Anthony Rudd has noted, J. L. Mackie may be taken to be responding to the love
commandment in this ‘pagan’ manner when he writes: ‘The biblical commandment
“Thou shalt love thy neighbour as thyself”, though it has its roots in a mistranslation
of a much more realistic rule, is often taken as prescribing a universal and equal con-
cern for all men. So interpreted, it is, as Mill says, effectively equivalent to the utilitar-
ian principle. And it is similarly impracticable. People simply are not going to put the
interests of their ‘neighbours’ on an equal footing with their own interests and specific
purposes and with the interests of those who are literally near to them. Such moral
concern will not be the actual motives of their choices, nor will they act as if they were’
(Mackie 1977: 130–1; cited in Rudd 1993: 164). Rudd also cites Steiner 1971: 43.
206 Understanding Mor al Obligation

What makes this sort of pagan response ‘unspoilt’ and there-

fore honest in Kierkegaard’s eyes is that at least it involves the proper
acknowledgement of how radical and strenuous the love command-
ment really is. By contrast, someone who we might called the ‘dishon-
est Christian’ (the Christian of the ‘Christendom’ that Kierkegaard
so much despised) takes herself not to reject the commandment, but
to abide by it – and yet this conformity is achieved only by watering it
down so that it can be met with little difficulty. As a result, Kierkegaard
thinks, the love commandment has been ‘tamed’ and been made to
look innocuous, so that we can easily forget what it really involves,8
while at least the ‘honest pagan’ takes it seriously, even if he then
rejects it as a consequence. Thus (as we have seen), just as Kierkegaard
presents the story of Abraham, or the ‘hard saying’ in Luke, in such a
way as to disturb and challenge us, so too he thinks that we are right
to find the love commandment problematic, given all that he takes it
to involve.
In this way, therefore, the love commandment highlights for
Kierkegaard what John Hare has called the ‘moral gap’. Such a gap
arises, Hare argues, for any ethical position which holds that a require-
ment of morality outstrips our capabilities for meeting it. When such
a gap is proposed, Hare suggests, there are four main ways one might
respond, two of which deny that the gap in fact exists, and two of which
attempt to bridge it. Thus, one might claim that in fact there is no gap,
as our capacities are greater than is being supposed; or alternatively,
one might claim that the moral demand is being set too high, so that
in fact our ordinary capacities are adequate to meet it. On the other
hand, one could accept that such a gap exists, but argue that there are
nonetheless ways we can be helped to deal with it, either by giving God
a role in the account, or in some more naturalistic manner.9
In these terms, therefore, Kierkegaard may be seen as using the
love commandment to insist against the first two strategies that there

8
Cf. WL 9:29–32 (pp. 25–29), esp. 9:30 (p. 27): ‘A tool that a craftsman uses becomes
dull over the years, and a spring loses its tension and becomes weak, but that which has
the tension of eternity retains it totally unchanged through all ages. When a strength-
tester has been used for a long time, eventually even the weak can meet the test, but
eternity’s standard of strength by which every person is to be tested – whether he will
have faith or not – remains totally unchanged through all ages’.
9
See Hare 2002: 25–6 and 36–7 for an overview of these positions, which are then
treated at length in subsequent chapters. Hare himself discusses Kierkegaard primar-
ily in Chapter 8, while other writers who employ Hare’s notion of a ‘moral gap’ in ana-
lysing Kierkegaard include Quinn 1998 and Evans 2004: 49 and 82.
 Kierkegaard’s account of obligation 207

really is a moral gap; and the sermon from the pastor in Either/Or may
be viewed as attempting to overturn the Hegelian strategy adopted by
Judge William, who, as we have discussed, appeals to the assistance of
the state in realising the full moral demand in a way that then reduces
the burden of obligation on each particular individual to a manageable
level.10 The latter strategy is apparent in William’s distinction between
duty in general and my duty, where the former is realised by the com-
munity as a whole, rendering the latter more attainable for each citi-
zen. For Kierkegaard, however, the upshot is not unlike the position of
the ‘dishonest Christian’, whereby the Judge may think he is abiding by
a Christian ethic,11 but in fact he is failing to recognise the full extent
of what is required of him as an individual. Thus, using the love com-
mandment to eliminate the other alternatives, Kierkegaard shows that
we must both accept the moral gap, and also treat God as the only
viable means of dealing with it – where from this claim, a divine com-
mand account of obligation is seen to follow.
We will consider this second step in the third section; in the next
section, I want to consider further the way in which Kierkegaard uses
the love commandment to highlight the moral gap. What is it about
the commandment that makes it so difficult for us to achieve, and why
must it be understood in this strenuous way?

What is the love commandment?

Kierkegaard’s case for the strenuousness of the love commandment
turns on a close consideration of what is meant here by love, and by
neighbour.

10
At the start of Chapter 7, Hare himself briefly indicates that he would see Hegel as
offering to bridge the moral gap in a naturalistic, non-theological manner, but does
not say any more: see Hare 2002: 170.
11
As we discussed in the previous chapter, William undoubtedly uses religious language
in presenting his ethical outlook, and is prepared to talk in terms of sin and repent-
ance; but the question is whether he is taking these issues with the ‘seriousness of the
true Christian’, and thus genuinely facing up to the ‘moral gap’ that this implies. This
issue is discussed in Perkins 1995. Cf. CUP 7:227 (p. 268): ‘Sin was not brought up
in any of the pseudonymous books. The ethicist in Either/Or [i.e. Judge William] did
indeed give a religious touch to the ethical category of choosing oneself by accom-
panying the act of despair with repenting oneself out of continuity with the race, but
this was a vitiation that no doubt had its basis in the aim of keeping the work ethical –
quite as if in accordance with my wishes – namely, in order that every factor should
become clear separately’.
208 Understanding Mor al Obligation

Kierkegaard begins by considering self-love, the kind of love for

oneself that one is then also meant to show the neighbour. Now,
Kierkegaard notes, basing the commandment round this sort of self-
love might seem to make the commandment too easy: for, it could be
claimed, wouldn’t it be more high-minded if the demand were that
we love the neighbour more than ourselves?12 Kierkegaard thinks that
this high-mindedness is vacuous, however, while also being morally
problematic, as it is only through the comparison with self-love that we
come to recognise the need to prevent others harming their own inter-
ests.13 On the other hand, the relation to self-love might seem to make
the commandment too hard in a rather straightforward way: for one
might think that our love for ourselves is so self-centred that it could
not possibly be shown to others. Kierkegaard argues, however, that it is
questionable whether individuals who display this kind of self-centred-
ness really love themselves at all or know where their well-being truly
lies, while a genuine self-love is rich enough to allow the individual to
also feel for others and not merely focus on their own concerns.14

12
Cf. WL 9:22–3 (pp. 18–19): ‘But would this really be the highest; would it not be pos-
sible to love a person more than oneself ? Indeed, this kind of poetic effusion is heard
in the world. Would it perhaps then be so that it is Christianity that is unable to soar
that high and therefore (probably also because it addresses itself to simple, everyday
people) is left miserably holding to the requirement to love the neighbour as oneself
… – would this perhaps be so? Or would we, since we do make a concession to cel-
ebrated love in comparison with commanded love, meagerly praise Christianity’s level-
headedness and understanding of life because it more soberly and more firmly holds
itself down to earth, perhaps in the same sense as the saying “Love me little, love me
long?” ’.
13
Cf. WL 9:24 (p. 20): ‘a human being you shall love as yourself. If you can perceive what
is best for him better than he can, you will not be excused because the harmful thing
was his own desire, was what he himself asked for. If this were not the case, it would be
quite proper to speak of loving another person more than oneself, because this would
mean, despite one’s insight that this would be harmful to him, doing it in obedience
because he demanded it, or in adoration because he desired it. But you expressly have
no right to do this; you have the responsibility if you do it, just as the other has the
responsibility if he wants to misuse his relation to you in such a way’.
14
Cf. WL 9:26–7 (p. 23): ‘Whoever has any knowledge of people will certainly admit that
just as he has often wished to be able to move them to relinquish self-love, he has also
had to wish that it were possible to teach them to [properly] love themselves. When the
bustler wastes his time and powers in the service of futile, inconsequential pursuits,
is this not because he has not learned rightly to love himself? When the light-minded
person throws himself almost like a nonentity into the folly of the moment and makes
nothing of it, is this not because he does not know how to love himself rightly? When
the depressed person desires to be rid of life, indeed, of himself, is this not because he
is unwilling to learn earnestly and rigorously to love himself?’.
 Kierkegaard’s account of obligation 209

In fact, Kierkegaard holds, when we think about the kind of erotic

love praised by ‘the poet’, we can see how it can be easy to get beyond
this narrow-minded focus merely on oneself: for in erotic love, we
turn from a preoccupation just with ourselves and what we want,
to a devoted concern for the happiness and well-being of the other.
Likewise, the same is true in the case of love between friends, where
this is also a more stable form of affection than we find in erotic love.
One might think, therefore, that these are the kinds of love that the
commandment is enjoining us to show to one another.
However, despite these features of erotic love and friendship,
Kierkegaard insists that neither are the real focus of the love com-
mandment: for, the love commandment enjoins us to love the neigh-
bour, who is not a particular individual for whom one feels some sort
of emotional attachment; it is simply the one who happens to present
herself to you as in need of assistance (which for Kierkegaard may be
as much spiritual as physical, to help them love God), where it could
be any person that calls upon you in this way, making them genuinely
‘other’ and requiring you to set aside all your preferences and thus to
act in a properly self-denying manner:
The same holds true of friendship as of erotic love, inasmuch as this,
too, is based on preference: to love this one person above all others, to
love him in contrast to all others. Therefore the object of both erotic
love and of friendship has preference’s name, ‘the beloved’, ‘the friend’,
who is loved in contrast to the whole world. The Christian doctrine, on
the contrary, is to love the neighbour, to love the whole human race, all
people, even the enemy, and not to make exceptions, neither of prefer-
ence nor of aversion…
Who, then, is one’s neighbour [Næste]? The word is obviously derived
from ‘nearest [Nærmeste]’; thus the neighbour is the person who is
nearer to you than anyone else, yet not in the sense of preferential love,
since to love someone who in the sense of preferential love is nearer
than anyone else is self-love – ‘do not the pagans also do the same?’ …
The concept ‘neighbour’ is actually the doubling of your own self; ‘the
neighbour’ is what thinkers call ‘the other’, that by which the selfish-
ness in self-love is to be tested.15

Thus, Kierkegaard argues, when it comes to the love commandment,

pagans do not ‘do the same’ as Christians: for, while pagans had dis-
tinguished between erotic love and the love involved in friendship,

15
WL 9:23 (p. 19) and 9:25 (p. 21).
210 Understanding Mor al Obligation

‘[p]aganism had never dreamed of this’, namely ‘self-denial’s love for

the neighbour’: ‘no one in paganism loved the neighbour; no one sus-
pected that he existed. Therefore what paganism called love, as dis-
tinguished from self-love, was preference. But if passionate preference
is essentially another form of self-love, then one sees here again the
truth in the saying of the venerable fathers: “that the virtues of pagan-
ism are glittering vices” ’.16
Kierkegaard considers a possible response, however, which is that
surely erotic love and love in friendship also involve self-denial, as one
gives oneself over to the other in a way that can entail enormous self-
sacrifice? Nonetheless, Kierkegaard insists, while both do indeed get
beyond crude egoism, they still remain self-regarding, as in both love
is given to the other out of admiration one feels for him or her, and
thus the hope of gaining love from them in return:
[T]here must be admiration in the relationship of erotic love, and the
stronger, more intense the admiration is, the better, says the poet. Well,
now, to admire another person is certainly not self-love; but to be loved
by the one and only admired one, would not this relation turn back
in a selfish way into the I who loves – his other I ? And so it is also with
friendship. To admire another person is certainly not self-love, but to be
the one and only friend of this one and only admired person – would
not this relation turn back in an alarming way into the I from which
we proceeded? Is not this plainly self-love’s danger – to have one single
object for its admiration when this one and only admired person in turn
makes one oneself the sole object of his erotic love or his friendship?17

To see the love commandment as about one’s beloved or friend, there-

fore, is still to fail to see the kind of ‘self-denial’ it involves: for while it
may indeed be true that one can be asked to do a great deal by both,
and more than one has any immediate desire or interest in doing, it is
nonetheless the case (Kierkegaard thinks) that this care for the other
in the end comes back to the care one has for oneself.
As a consequence of this self-denial, therefore, Kierkegaard thinks
the ‘honest pagan’ is indeed correct to recognise the kind of difficul-
ties that the love commandment poses for us, as we are required to
go beyond all our natural feelings and inclinations. One might think,
perhaps, that it will be easy to follow it, for its very exaltedness may
seem to make it attractive to us: ‘Indeed, is there anything for which

16
WL 9:55–6 (p. 53).
17
WL 9:57 (pp. 54–5).
 Kierkegaard’s account of obligation 211

the human being as such, anything for which the natural man is more
desirous than for the highest!’.18 If we think this way, Kierkegaard
thinks, we are deceiving ourselves, where once again the honest pagan
is nearer to the mark in facing up to the difficulties involved:
No, the essentially Christian is certainly the highest and the supremely
highest, but, mark well, in such a way that to the natural man it is an
offense … So it is also with the commandment to love the neighbour.
Only confess it, or if it is disturbing to you to have it put this way, well,
I myself will confess that many times this has thrust me back and that I
am yet very far from the delusion that I fulfil this commandment, which
to flesh and blood is an offense and to wisdom foolishness.19

Kierkegaard therefore insists that the love commandment, properly

understood, should seem shocking to people in what it asks of them;
and if it does not cause ‘offense’ in this manner, then this is probably
because they have watered it down.
One particularly problematic aspect of the commandment, in the
way that it asks us to go against the grain of our natural dispositions
and how our capacity for love generally operates, is the implication
that in being required to love one’s neighbour, one is also required
to love one’s enemy. Again, Kierkegaard accepts that asking this of
us may well seem an ‘offense’ and ‘foolishness’ to ‘flesh and blood’;
for, perfectly understandably, ‘[p]eople think that it is impossible for
a human being to love his enemy, because, alas, enemies are hardly
able to stand the sight of one another’.20 If we can see the enemy as our
neighbour, Kierkegaard suggests, we might then find it in ourselves to
love him; but at the same time, he allows, this underscores yet further
how hard seeing others as our neighbours can be.
The commandment is also problematic in the way that it relates
neighbour love to erotic love and the love of friendship.21 For, although
it does not abolish the latter kinds of love, in so far as neighbour love
excludes no one, and so encompasses within it love of the beloved or
friend,22 it nonetheless puts this on a different kind of footing. What
happens, Kierkegaard thinks, is that neighbour love is given a priority,

18
WL 9:60 (p. 58). 19 WL 9:60–1 (pp. 58–9). 20
WL 9:69 (p. 68).
21
This aspect of Kierkegaard’s position has been much discussed, and has been the
focus of criticism. For a critical response, see Løgstrup 1997. For further discussion,
see Ferreira 2001: 53–64 and Evans 2004: 203–15.
22
Cf. WL 9:64 (p. 63): ‘Do not delude yourself into thinking that you could bargain, that
by loving some people, relatives and friends, you would be loving the neighbour … No,
212 Understanding Mor al Obligation

so that (for example) ‘your wife must first and foremost be to you your
neighbour; that she is your wife is then a more precise specification of
your particular relationship to each other’.23 As Kierkegaard notes, ‘[t]o
drives and inclination this is no doubt a strange, chilling inversion; yet it
is Christianity and no more chilling than the spirit is in relation to the
sensate or the sensate-psychical’.24
Now, Kierkegaard recognises that by underlining the strenuousness
of the love command in these ways, we may come to feel disheartened
or fearful, since it may seem that we are being asked to do so much
that we can only ever fall short.25 It is just this, however, that makes the
commandment a religious matter for Kierkegaard, a question of faith:
Do not cease to believe because the commandment almost offends you,
because the discourse does not sound flattering like that of the poet,
who with his songs insinuates himself into your happiness, but sounds
repelling and terrifying, as if it would frighten you out of the beloved
haunts of preferential love – do not for that reason cease to believe it.
Bear in mind that just because the commandment and the discourse
are like this, for that very reason the object can be the object of faith!26

We can now turn to considering this last claim: for, it will be argued, it
is precisely the difficulty of meeting the love commandment that gives
it its religious significance, where the moral gap that Kierkegaard has
opened up then makes room for God, and a divine command account
of obligation.

Bridging the gap: the love commandment and God

In the chapter of Works of Love entitled ‘Love is the Fulfilling of the
Law’, Kierkegaard declares that human beings have staged a ‘mutiny’

love the beloved faithfully and tenderly, but let love for the neighbour be the sanctify-
ing element in your union’s covenant with God’.
23
WL 9:135 (p. 141). 24 WL 9:135 (p. 141).
25
Cf. WL 9:128–9 (p. 134): ‘Is there any more accurate expression of how infinitely far a
person is from fulfilling the requirement than this, that the distance is so great that he
actually cannot begin to calculate it, cannot total up the account! Not only is so much
neglected every day, to say nothing of what guilt is incurred, but when some time has
passed, one is not even able to state accurately the guilt as it once appeared to oneself,
because time changes and mitigates one’s judgment of the past – but, alas, no amount
of time changes the requirement, eternity’s requirement – that love is the fulfilling of
the Law’.
26
WL 9:64 (p. 62).
 Kierkegaard’s account of obligation 213

against God, by attempting to make ourselves into the source of the

moral law and its obligations.27 This mutiny is based on a determin-
ation to assert our freedom: ‘The abominable era of bond service is
past, and so there is the aim of going further – by means of the abom-
ination of abolishing the person’s bond service in relation to God’:28
Yet this bond service is found to be a burdensome encumbrance and
therefore there is a more or less open intent to depose God in order
to install human beings – in the rights of humanity? No, that is not
needed; God has already done that – in the rights of God. If God is
dismissed, the place will indeed be vacant.29

However, like Anscombe in our own time,30 Kierkegaard holds that

this attempt to replace God as the source of obligation with a more
humanistic model can only lead to incoherence and confusion, so that
‘[a]s a reward for such presumption, all existence will in that way prob-
ably come closer and closer to being transformed into doubt or into a
vortex’.31 As with Anscombe, a central difficulty raised by Kierkegaard
concerns how any one of us, either individually or collectively, can have
the kind of authority to impose law on ourselves or each other.
Against a Kantian model of self-legislation, Kierkegaard argues that
it would then be ‘left up to my arbitrary decision to assume one thing
as the Law’s requirement today and something else tomorrow’,32 so
that we end up with ‘pure arbitrariness’.33 Kierkegaard expands on
this point further in a journal entry from around 1850:
Kant was of the opinion that man is his own law (autonomy) – that is,
he binds himself under the law that he himself gave himself. Actually,
in a profounder sense, this is how lawlessness or imaginary constructing
is posited. This is not being rigorously earnest any more than Sancho
Panza’s self-administered blows to his own bottom were vigorous. It is
impossible for me to be really any more rigorous in A than I am or wish
to be in B. Constraint there must be if it is going to be in earnest. If I am
bound by nothing higher than myself and I am to bind myself, where
would I get the rigorousness as A, the binder, which I do not have as B,
who is supposed to be bound, when A and B are the same self?…

27
WL 9:113 (p. 117). Cf. also PC 12:84 (p. 88): ‘Strangely enough, this deification of the
established order is the perpetual revolt, the continual mutiny against God’. For help-
ful discussion of Kierkegaard’s position here, see Evans 2004: 127–32.
28
WL 9:112 (p. 115). 29 WL 9:112 (p. 115).
30
For further comparative discussion of Anscombe and Kierkegaard, see Cantrell
2009.
31
WL 9:112 (p. 115). 32 WL 9:112 (p. 115). 33 WL 9:112 (p. 116).
214 Understanding Mor al Obligation

The maxim which I give myself is not only not a law, but there is a law
that is given me by one higher than myself, and not only that, but this
lawgiver takes the liberty of taking a hand in the capacity of tutor and
bringing constraint to bear.34

Kierkegaard thus foreshadows Anscombe’s later critique of the idea

of self-legislation, that without ‘superior power in the legislator’, this
does not amount to a genuine constraint.
At the same time, Kierkegaard is equally critical of social attempts
to account for obligation, where again the problem is one of author-
ity. Within a social order, there is no individual or group of individ-
uals obviously fitted to take on the role of legislator, which is why the
Kantian ends up legislating over himself.35 An alternative, perhaps,
might be a collective account, whereby the people as a whole impose
the requirement on the individual: but how is any such group to be
assembled, in order for its will to be determined and for this binding
of the individual to take place? And how are disagreements within the
group to be handled over what is to be required and what is not? If this
is to be decided by some sort of vote, how large should it be?36
As well as being unstable in these respects, Kierkegaard also thinks
this kind of social model has problematic ethical implications. For,
firstly, there is a sense in which each individual now relies on the rest
to tell him where his moral requirements lie, instead of taking respon-
sibility of determining this for himself: ‘[i]n this way all human life
transforms itself into one big excuse’.37 Secondly, the individual may
feel on this account that they can never arrive at the point of being sure
that the requirements being imposed upon them are the definitive set,
as the determination of these requirements is caught up in a constant
process of negotiation within society. At any particular moment in

34
JP 102 A 396 n.d., 1850 (i, §188, p. 77).
35
Cf. WL 9:112 (p. 115): ‘Since one person does not stand essentially higher than another,
it is left entirely up to my arbitrary decision with whom I will affiliate in the determin-
ation of the highest unless I myself – even more arbitrarily, if possible – could be in a
position to hit upon a new determination and as a recruiter win alliance for it’.
36
Cf. WL 9:112 (p. 115): ‘Or should the determination of what is the Law’s requirement
perhaps be an agreement among, a common decision by, all people, to which the indi-
vidual then has to submit? Splendid – that is, if it is possible to find the place and fix
a date for this assembling of all people (all the living, all of them? – but what about
the dead?), and if it is possible, something that is equally impossible, for all of them
to agree on one thing! Or is perhaps the agreement of a number of people, a certain
number of votes, sufficient for the decision? How large a number is necessary?’.
37
WL 9:112 (p. 116).
 Kierkegaard’s account of obligation 215

time, therefore, the individual may well regard these requirements as

arbitrary, compared with those that might be laid down at ‘the end of
history’ – a clear dig at the Hegelian position.38
These problems can be resolved, Kierkegaard argues, if we turn
from the Kantian and Hegelian accounts to a divine command
account, so that ‘all of us, each one separately, receives our orders at
one place, if I may put it this way, and then each one unconditionally
obeys the same orders’:39
When this is the case, there is durability in existence, because God has
a firm hold on it. There is no vortex, because each individual begins,
not with ‘the others’ and therefore not with evasions or excuses, but
begins with the God-relationship and therefore stands firm and thereby
also stops, as far as he reaches, the dizziness that is the beginning of
mutiny.40

Concerned with our own autonomy, therefore, we may seek to offer

something other than a divine command account of obligation; but,
Kierkegaard argues, none of these alternatives can be considered sta-
ble, because in none of them is the figure of authority who is able to
issue any such requirement.
Now, Kierkegaard maintains, as a consequence of the ‘mutiny’ that
has increasingly taken hold within the modern world, the require-
ments of the moral law have been scaled back, with the result that
the love commandment itself has been watered down accordingly. For,
once a divine command ethic is abandoned, we lose notions such as
divine assistance and forgiveness that go with it, which are needed to
bridge the ‘moral gap’ and to make sense of that law in a strenuous
form: for, it is only in so far as we think of that law as divinely com-
manded that we can then also apply these concepts of assistance and
forgiveness to it, in a way that is needed if we are to see the love com-
mandment in the right light.

38
WL 9:113 (p. 116): ‘If, now, this inhumanly protracted labour on the common agree-
ment among all people is not finished in one evening but drags on from generation to
generation, how the individual comes to begin will then become purely accidental: it
will depend on when the individual joined the game, so to speak. Some would begin at
the beginning but would die before they had come halfway; others would begin mid-
way but would die without seeing the end, which no one actually would see, inasmuch
as it would not come until all was finished, world history had ended, because only then
would it be fully known what the Law’s requirement is’.
39
WL 9:114 (p. 117). 40
WL 9:114 (p. 118).
216 Understanding Mor al Obligation

Kierkegaard makes his position here clear in a journal entry from

1851, entitled ‘The Christian Emphasis’:
Christianly the emphasis does not fall so much upon to what extent or
how far a person succeeds in meeting or fulfilling the requirement, if
he actually is striving, as upon his getting an impression of the require-
ment in all its infinitude so that he rightly learns to be humbled and to
rely upon grace.
To scale down the requirement in order to be able to fulfill it better
(as if this were earnestness, that now it can all the more easily appear that
one is earnest about fulfilling the requirement) – to this Christianity in
its deepest essence is opposed.
No, infinite humiliation and grace, and then a striving born of grati-
tude – this is Christianity.41

Kierkegaard thus proposes a dilemma: We can either take the love

commandment seriously, with the conception of neighbour love that
it involves; but then we must see it as divinely commanded, as only this
will enable us to bridge the moral gap with the conceptions of forgive-
ness and grace that make sense of the fact that it is demanded of us,
even though we cannot really expect to abide by it. Or we can abandon
a divine command theory, but then accept merely a watered-down ver-
sion of the love command, as without this theological context, we can
no longer appeal to the notions of forgiveness and grace that render
the command intelligible in its strong form.
It might be said in response, that from a Kantian perspective this
dilemma is not exhaustive, for Kant also allows a role for God as a
judge within his ethics, but without endorsing a divine command the-
ory in a Kierkegaardian fashion. However, I think that precisely for
this reason, Kierkegaard would reject Kant’s position as unstable, to
the extent that (as we have seen) the ethical is made prior to the reli-
gious on Kant’s account, so that precisely in doing so the ethical is
already scaled back. The law that God is then called upon to uphold
(as it were) is one that is first conceived without him, and thus cannot
be a law that contains the love commandment within it, but is fitted to
a pre-religious sensibility. Far from escaping the dilemma, therefore,

41
JP 103 A 734 n.d. 20, 1851 (i, §993, pp. 433–4). Cf. also SLW 6:443 (p. 476): ‘The aes-
thetic sphere is the sphere of immediacy, the ethical the sphere of requirement (and
this requirement is so infinite that the individual always goes bankrupt), the religious
the sphere of fulfilment’.
 Kierkegaard’s account of obligation 217

from a Kierkegaardian perspective, Kantian ethics may instead appear

simply to confirm its force.
Thus, for Kierkegaard, one of the fundamental effects of the ‘mutiny’
he has described is that now ‘[t]here actually is a conflict between what
the world understands and what God understands by love’,42 and thus
between what the world understands by the love commandment and
what genuine Christians understand by it. For the ‘worldly’, who trace
obligations back to the human level, the command encompasses noth-
ing much more than the kind of love based on our natural inclina-
tions towards our nearest and dearest; but for the latter, who trace
obligations back to God, it involves the much more demanding love
of the ‘other’, for which we require divine support: ‘And when eternity
says, “You shall love”, it is responsible for making sure that this can be
done’.43
We can therefore understand the Works of Love by seeing this as the
issue which determines the architectonic of the book as a whole, as
Bruce Kirmmse has rightly noted:
Works of Love is divided by SK into two books or ‘sequences’ [Følger] of
discourses. The first book posits the notion of the Neighbor and of the
Law’s demand, and we are compelled to confront the radical absolute-
ness of Christian ethics and our inability to live accordingly. We are led
to the threshold of grace. The second book deals with one who has been
granted the grace of God and who is active as ‘the loving person’ [den
Kjerlige]. The two halves of Works of Love thus relate to one another as
‘theory’ (the Law) and ‘practice’ (the Gospel). The spark that connects
them – and that carries one from the helplessness of legal obligation to
the ‘sheer activity’ of loving – is grace.44

Kierkegaard’s concern with the ‘moral gap’, and how to bridge it, is
thus the hinge on which his ‘Christian discourse’ turns.
Finally, once Kierkegaard has completed this double-movement, he
is willing in the conclusion to introduce a perspective from which (in
a Kantian manner) the love commandment is no longer a command-
ment, as it can be followed in its most demanding form, and yet with-
out any sense of struggle – where this is the perspective of ‘the Apostle
John’, who is able to say not that we shall love one another, but rather

42
WL 9:115 (p. 118).
43
WL 9:44 (p. 41).
44
Kirmmse 1990: 312.
218 Understanding Mor al Obligation

‘Beloved, let us love one another’,45 thereby dropping the strongly

imperatival form. Kierkegaard comments on this as follows:
These words, which have apostolic authority, also have, if you consider
them, an intermediate tone or an intermediate mood in connection
with the contrasts in love itself. The basis of this is that they are by one
who was perfected in love. You do not hear the rigorousness of duty in
these words; the apostle does not say, ‘You shall love one another’; but
neither do you hear the vehemence of poet-passion and of inclination.
There is something transfigured and beatific in these words, but there
is also a sadness that is agitated over life and mitigated by the eternal. It
is as if the apostle said: ‘Dear me, what is all this that would hinder you
in loving, what is all this that you can win by self-love! The command-
ment is that you shall love, but ah, if you will understand yourself and
life, then it seems that it should not need to be commanded, because
to love people is the only thing worth living for, and without this love
you are not really living. Moreover, to love people is the only blessed
comfort both here and in the next world; and to love people is the only
true sign that you are a Christian’ – truly, a profession of faith is not
enough either.46

If we could only see things aright, therefore, and properly understand

ourselves, God and love itself, God’s injunction to us would no longer
need to be uttered as a command. This is not because John is telling us
that love of the neighbour would then somehow matter less and so no
longer be required – in that sense, Kierkegaard insists, the command
is eternal and ‘is not changed in the slightest way, least of all by an
apostle’.47 Rather, just as Kant’s conception of the holy will opens up
a perspective from which the moral demand no longer appears to the
agent as a demand, so John’s words indicate a similar relation to God,
who would no longer have to require love of us; but this is a position we
can hardly hope to reach, though it is the one from which the apostle
himself speaks:
The change then can only be that the person who loves becomes more
and more intimate with the commandment, becomes as one with the
commandment, which he loves. This is why he [i.e. John] can speak
so gently, so sadly, almost as if it had been forgotten that love is the
commandment. If, however, you forget that it is the apostle of love who

45
WL 9:355 (p. 375). See I John 4: 7.
46
WL 9:355 (p. 375).
47
WL 9:356 (p. 375).
 Kierkegaard’s account of obligation 219

is speaking, you misunderstand him, because such words are not the
beginning of the discourse about love but are the completion. Therefore
we do not dare to speak this way. That which is truth on the lips of the
veteran and perfected apostle could in the mouth of a beginner very
easily be a philandering by which he would leave the school of the com-
mandment much too soon and escape the ‘school-yoke’. We introduce
the apostle speaking; we do not make his words our own but make our-
selves into listeners: ‘Beloved, let us love one another!’48

48
WL 9:356 (pp. 375–6). Cf. also WL 9:97 (p. 99), where Kierkegaard speaks of Christ
as ‘the downfall of the Law; because he was what it required. Its downfall, its end –
for when the requirement is fulfilled, the requirement exists only in the fulfillment,
but consequently it does not exist as requirement’; however, ‘[e]ven though the law is
abolished [in Christ], it still stands here with its power and fixes an everlasting chas-
mic abyss between the God-man and every other person’ (WL 9:99 (p. 101)). Jamie
Ferreira has rightly stressed the Lutheran background to Kierkegaard’s position here,
and Luther’s claim that ‘if every man had faith we would need no more laws [for]
everyone would of himself do good works all the time, as his faith shows him’ (Ferreira
2001: 241–2).
 C onc lusion: F rom K a n t to
K i e r k eg a a r d – a n d b ack ag a i n ?

In the foregoing discussion, I have traced the issue of moral obligation

as it has run through the work of Kant, Hegel, and Kierkegaard, shap-
ing their relation to each other. In doing so, I have argued, I have put
forward a different picture of the development of moral philosophy
during this period from the one offered by what in the Introduction
I called the standard story, which begins from what turns out to be a
misconceived version of Kant’s argument from autonomy. I therefore
hope to have offered a more convincing account of that argument,
and also of the trajectory of moral philosophy from Kant onwards.
However, in following this journey, we have in fact been led round
in a circle. For, as I presented it, Kant’s argument from autonomy was
not directed against moral realism as such, but rather against divine
command accounts of obligation; but in arriving at Kierkegaard, I
have also claimed, it is with a divine command account that we have
ended up. The obvious question this raises, then, is whether one of
the three positions that make up this circle is to be preferred over the
others, or whether their respective merits and demerits put them on a
par, locked in a perpetual dialectical struggle with one another with-
out hope of resolution.
In order to ascertain whether this is indeed so, in conclusion I will
look at the critical moves that took us from one theory to the next, and
thus consider the following questions: how far should considerations
of autonomy drive one from a divine command theory to a Kantian
hybrid theory? How far should considerations of dualism drive one
from a hybrid theory to a Hegelian social command theory? And how
far should considerations of complacency drive one from a social com-
mand theory to a Kierkegaardian divine command theory? If we can
form some assessment of the relative weight of these arguments of the

220
 Conclusion 221

one theory against the others, we can ascertain where (if anywhere) we
might be able to bring the dialectic to a conclusion.
As we shall now see, the issues raised by these debates are large and
complex, and I cannot pretend to deal with them all here. Nonetheless,
by thinking them through, we can see how these positions are finely
balanced against each other; thus in the end, it may appear, all that is
achieved is a kind of equipollence, making it hard to choose between
the theories in a definitive manner, where none can claim to have
resolved the problem of moral obligation without a significant cost.

The Kantian argument from autonomy and the

Kierkegaardian divine command theory
While the argument most commonly used against divine command
theories is the Euthyphro dilemma, the argument from autonomy
has a claim to run a close second.1 Moreover, as recent discussions
of the Euthyphro dilemma have shown, the effectiveness of the lat-
ter is limited, as it only really bites against radical, fully voluntaristic
versions of divine command theories, not classical ‘intermediate’ div-
ine command theories, or more recent so-called ‘modified’ versions,2
while the argument from autonomy applies to all such theories, and
so is apparently of greater potency in this respect. For, the dilemma,
as standardly posed, is: does God command an action because it is
right, or is it right because God commands it? On the first horn, it
looks as if this allows that the rightness of an action is independent of
God’s command, which is what the radical voluntarist denies, so that
they seem forced to opt for the second horn. However, on the second
horn, it looks like what is right becomes arbitrary, as arising out of the
will of God that is itself then unconstrained by any prior values in a
way that makes it hard to see how a will of this sort could constitute
what is right for us to do. But once we move to an ‘intermediate’ or
‘modified’ divine command theory, this dilemma loses its force: such
a theory can allow that what is right or good is independent of God’s
command, but holds instead that this only becomes obligatory as result

1
There are, of course, a number of other significant objections to be made to divine com-
mand theories as well. For a helpful survey, see Wainwright 2005: 106–17. Wainwright
discusses the Euthyphro Problem on pp. 73–83 and the argument from autonomy on
pp. 117–22.
2
‘Modified’ divine command theory is particularly associated with the work of Robert
Merrihew Adams: see Adams 1973 and 1979b.
222 Understanding Mor al Obligation

of the command. A divine command theorist of this more moderate

sort can therefore hold that God commands actions because they are
right or because he is good, while still claiming to be a divine com-
mand theorist about what makes these right actions obligatory for us,
as actions we now must do. The Euthyphro dilemma turns out to be
limited in its application, in a way that the argument from autonomy
is not.
Now, when it comes to Kierkegaard, I have suggested above that he
was a divine command theorist of this moderate, ‘intermediate’, sort,
whose God puts us under obligations but without operating outside
any prior order of values or norms, even though that order may not be
wholly within our grasp as finite beings. As a result, if this interpret-
ation is correct, then (as we have just seen) the Euthyphro objection
to Kierkegaard’s position drops away; this then leaves the argument
from autonomy as perhaps the primary difficulty it faces. We there-
fore need to consider how effective this argument is, and how theists
have responded to it.3 In general, their response has been that divine
command theory is compatible with autonomy in certain fundamental
respects, while where it is not, their critics have exaggerated the sig-
nificance and value of such autonomy.
Firstly, then, autonomy can be viewed as the capacity for decid-
ing or thinking for oneself and not abdicating this responsibility to
others: does a divine command theory take this away?4 When it comes
to Kierkegaard, his position on this issue is complex. On the one
hand, he clearly holds that individuals must determine for themselves
whether to accept God’s authority, and must take full responsibility for
the outcome of that decision: this is partly why, of course, Abraham’s
action is a matter of ‘fear and trembling’. Thus, as one commenta-
tor has put it, for Kierkegaard ‘[a]utonomy is positively insisted on
in the sense that an agent must freely yield to God’s sovereignty and
assent to his command … If anything, Kierkegaard emphasizes more

3
Perhaps the most influential recent response has been by Robert Merrihew Adams, who
has then been followed by John Hare, Stephen Evans, and others. Adams deals with this
issue in Adams 1979a; he develops the response further in Adams 1999: 270–6. For
Hare, see especially Hare 2009: 266–9, while Evans briefly discusses the issue in rela-
tion to Kierkegaard in Evans 2004: 303–4. For a complementary but somewhat differ-
ent approach to the problem, see Quinn 1978: 1–22.
4
Cf. Løgstrup 1997: 2: ‘If a religious proclamation is not understandable in the sense
that it answers to decisive features of our existence, then to accept it is tantamount
to letting ourselves be coerced – whether by others or by ourselves – for faith without
understanding is not faith but coercion’.
 Conclusion 223

strongly than many defenders of moral autonomy the importance of

independent acceptance as an implicate of faith’.5 Likewise, and again
in a way that is illustrated by Abraham’s ‘trial’, we are not expected to
abandon all our moral criteria and conceptions in considering God’s
commands: indeed, it is the fact that such criteria and conceptions
remain in play that makes Abraham’s action such a remarkable one.
It is a mistake to think, therefore, that in speaking of a teleological
suspension of the ethical, Kierkegaard is suggesting that our ordinary
moral thinking should simply be given up when faced by a divine com-
mand, and that the former counts for nothing when confronted by
the latter; rather ‘suspension’ is here being used in a Hegelian sense
of ‘sublated’ or ‘aufgehoben’, not simply abandoned, as if we then hand
all guidance in moral matters over to God in an arbitrary manner,
any more than a soldier who follows the orders of a superior whom he
trusts and respects. In these respects, therefore, autonomy may be said
to be preserved within Kierkegaard’s account.
On the other hand, on the analysis of Kierkegaard’s position offered
previously, there will be acknowledged limits to our capacity for crit-
ical reflection, in so far as the moral good is treated as transcend-
ent and thus beyond our abilities to fathom, with the result that God
can require us to do things that violate our criteria and mean that in
such circumstances we must be willing to set them aside.6 It is debat-
able, however, how much this violates principles of autonomy.7 It seems
that this certainly does not preclude us from taking a critical stance
altogether, for we are still required, as autonomous agents, to weigh
very carefully when admitting to this kind of ignorance, for fear that it
becomes an excuse for the abdication of all responsibility for our deci-
sions – where this then resembles cases where we accept the guidance
of experts on some matter, whose knowledge goes beyond ours, but
where we retain our autonomy by scrutinising carefully the credentials
of those who we follow in this manner. Thus, while it may be true that
according to Kierkegaard, Abraham ‘must defer to a wisdom superior

5
Outka 1973: 234–5.
6
Cf. WL 9:24 (p. 20): ‘But you shall love God in unconditional obedience, even if what he
requires of you might seem to you to be your own harm, indeed, harmful to his cause;
for God’s wisdom is beyond all comparison with yours, and God’s governance has no
obligation of responsibility to your sagacity’.
7
Cf. again Løgstrup 1997: 2, where the passage just cited continues: ‘However … the
understandability of a proclamation does not mean that it must concur with our own
formulations or accommodate itself to the solutions we have ourselves imagined; it
does not mean that we cannot take offense at it’.
224 Understanding Mor al Obligation

to his own’,8 it could be argued that nothing prevents this deference

from being of a thinking and reflective kind, in a manner that is suffi-
cient to preserve our autonomy. We have seen, then, how on this first
conception of autonomy, a way can be found to largely accommodate
it within a divine command account.
A second conception of autonomy is perhaps more problematic for
the divine command account, however, and it is this conception that
seems to play the central role in the actual Kantian argument that we
considered in Chapter 2. The issue here has less to do with the auton-
omy we exercise in thinking for ourselves, and more to do with our
motivations, and what leads us to act in moral matters. The worry is, as
Robert Adams has put it, that ‘perhaps the introduction of divine com-
mands in ethics threatens to debase our motivation in well-doing’,9
so that in Kant’s terms, that ‘well-doing’ now becomes heteronomous
rather than autonomous. To illustrate the problem, Adams introduces
two examples: the Conscientiously Obedient Relief Worker, and the
Autonomous Relief Worker. The former does good in relieving fam-
ine, but ‘only because he has been instructed by his employers to do
it, and believes unquestioningly that that morally obliges him to do
it’.10 The latter, on the other hand, ‘directs the food to the poor, not
just because he has been instructed to do so, but primarily because
he cares about the needs of the poor’,11 and therefore thinks this is
the right thing for him to do. It may appear that this worker is more
autonomous than the first, as he is acting in a way that is directed by
what he takes to be the right course of action, rather than out of con-
sideration for an authority which renders that action obligatory for
him.
How does that consideration render the first worker’s action heter-
onomous? One way, made central by Kant, is if this means that his
action is based on fear of punishment or the hope of reward, or some
other non-moral concern of this sort. And, of course, this could indeed
be the sort of thing that is motivating the Conscientiously Obedient
Relief Worker, where Kant might then plausibly say that his behav-
iour is no longer fully autonomous (even though, on a more Humean
approach, this could also be denied, providing the desires on which

8
Outka 1973: 236.
9
 Adams 1979a: 194/1987: 126. Cf. Adams 1999: 274–6.
10
Adams 1979a: 193/1987: 125.
11
Adams 1979a: 193/1987: 125.
 Conclusion 225

he is acting are ones that have been suitably ‘endorsed’ by the agent
themselves). But, even allowing the Kantian to make this move, the
divine command theorist could still reply that this does not have to
be what is going on in the case of this worker: rather, his motivation
might be that he feels a duty or obligation to help the hungry, where
he thinks this duty or obligation is grounded in the fact his employers
have instructed him to do it as legitimate authorities in this matter,
but where he is no less acting from and motivated by duty than the
second relief worker, for whom the obligation arises simply because he
takes it to be right for him to act in this way.12 It is not clear, therefore,
that by introducing the command of God as the source of our obliga-
tions here, the divine command theorist has necessarily ‘debased’ the
motivations of the agent involved, and so rendered them heteronom-
ous in this Kantian sense. For many such theorists, it is the normative
authority of God over us that makes his command obligatory, not his
coercive power as such, where it is in acknowledging the former and
not the latter that the motivations to obey his commands can arise
(and where that normative authority may or may not be said to itself
involve such coercive powers). It therefore appears mistaken to suggest
that in obeying this authority, the motivations of the agent must be
related to considerations of fear and favour, rather than respect for
the authority and the obligations it lays down as such.13

12
This sort of position is also suggested by an example discussed by Korsgaard, involv-
ing a student who takes a course because of the reasons that make it worthwhile, and
another student who takes the course because they are required to do so by the aca-
demic department to which they belong. She argues that if the latter student is not
acting out of fear or favour, but because they recognise the department as a legitimate
authority in this case who can make the course obligatory, then in following the course
they are no less autonomous than the first student. See Korsgaard 1996c: 25–7 and
105–7. Hare has used Korsgaard’s position here to defend the compatibility of auton-
omy with divine command accounts: see e.g. Hare 2009: 268–9.
13
Cf. Korsgaard 1996c: 24–5, which emphasises this point in relation to Pufendorf and
Hobbes: ‘[B]oth Pufendorf and Hobbes believed that no one could be a legislator
without the power to impose sanctions to enforce his law. And it is frequently inferred
that the point of those sanctions is to provide the subjects of the law with the motives
to obey it. Actually, however, both of these philosophers though that morally good
action is action which proceeds from what we would now call the motive of duty. One
does the right thing because it is the right thing, because it is the law, and for no other
reason … Why then are the sanctions needed? The answer is that they are necessary to
establish the authority of the legislator’. Cf. also Darwall 2006: 108. Of course, though
they defend the divine command theory in this respect, Korsgaard and Darwall are
critical of it in other ways.
226 Understanding Mor al Obligation

However, the Kantian might reply, it still remains an essential part

of a theological ethics such as Kierkegaard’s that one’s concern is to
do more than merely what one takes to be one’s duty; one is also, qua
religious believer, motivated by the thought that in following what is
required, one is bringing one’s behaviour into line with God’s will,
where this is taken to be a matter of supreme importance14 – indeed,
if one didn’t think this were the case in so acting, one would then lose
much of one’s motivation for doing so. And in this respect, it could
also be said, the divine command theory differs problematically from a
social command theory like Hegel’s: for while the latter treats a socially
imposed requirement as the source of obligatoriness, one’s willing-
ness to comply with the obligation will not ultimately rest on a desire
to subordinate oneself to the ‘social will’ in this matter, and if it did,
this would seem to compromise one’s autonomy. Rather, one acknow-
ledges that the social command makes the action into a duty, but one
does it because it is recognised as such, not because in so doing one is
conforming to the will of another. There is thus a difference between
acknowledging that this will renders an action obligatory, and hence
seeing this will as authoritative in this regard, and doing the action
because one thereby does what that authority wills, where it is in obedi-
ence to that will that the fundamental value in so acting is seen to lie.
The issue is, therefore, whether the theological case involves the com-
promising of autonomy in this manner.
In response to this worry, the strategy that some contemporary div-
ine command theorists have adopted is to admit that while the agent
here is not just acting autonomously, what we have instead is a case of
theonomy, where the latter can nonetheless involve both autonomous
action and a conformity to the will of another, so that autonomy is
not really given up. The term theonomy is particularly associated with
Paul Tillich,15 while the idea is developed in the following manner by
Robert Adams:

14
Cf. Kant, Relig 6:103–5 (pp. 137–8). And cf. Swinburne 1974: 213: ‘For a theist, a man’s
duty is to conform to the announced will of God’.
15
Cf. Tillich 1946: 80: ‘The words “autonomy,” “heteronomy,” and “theonomy’ answer
the question as to where the “nomos” or law of life is rooted in three different ways:
Autonomy asserts that man as the bearer of universal reason is the source and measure
of culture and religion – that he is his own law. Heteronomy asserts that man, being
unable to act according to universal reason, must be subjected to a law, strange and
superior to him. Theonomy asserts that the superior law is, at the same time, the inner-
most law of man himself, rooted in the divine ground which is man’s own ground: the
law of life transcends man, although it is at the same time his own’.
 Conclusion 227

Let us say that a person is theonomous to the extent that the following is
true of him: He regards his moral principles as given him by God, and
adheres to them partly out of love or loyalty to God, but he also prizes
them for their own sakes, so that they are the principles that he would
give himself if he were giving himself a moral law. The theonomous
agent, in so far as he is right, acts morally because he loves God, but also
because he loves what God loves. He has the motivational goods both of
obedience and of autonomy.16

The claim is, then, that theonomy involves two aspects: acting out of
concern for what is right or what one takes to be one’s duty, but also
out of concern for how this puts one in relation to God. Thus, if one
holds that only the former characterises the autonomous moral agent,
it could be said that the divine command model of moral agency does
not do full justice to autonomy; but the claim is that it is quite possible
to combine this with the latter, without any significant loss. In this
case, the individual is indeed seeking to do what God asks of him, out
of love, loyalty, respect, gratitude, or some such reason (rather than
fear or hope of reward), and this is his motivation to act in accord-
ance with the obligation that God has laid down; but at the same time,
what God has ordered him to do is what he would himself choose to
do were he to determine his action solely by himself, in such a way
that his autonomy is therefore also preserved.17 Thus, the suggestion
is that while one is here setting out to conform to God’s will, there is
still autonomy, because that will is directing you to do what is right and
what you would will for yourself.
Now, of course, the Kantian might still argue that this combination
is unstable, and that autonomy cannot be combined with obedience
in this manner, in so far as obedience is guided by the desire to con-
form to the will of another, and so takes away a crucial element of

16
Adams 1979a: 194/1987: 126. Cf. also Hare 2001: 114–5. As Hare notes (p. 115,
note 61), one might also think that ‘[a]utonomous submission to political authority
(“fredonomy”?) has the same structure’. Cf. also Connell 1992. Not all writers on this
issue treat theonomy and autonomy as compatible, however: cf. Westphal 1988: 2:
‘[Kierkegaard’s] ethics is theonomous, and thus heteronomous’.
17
Cf. also Hare 2001: 115, where Hare associates this sort of picture with Scotus: ‘[T]here is
nothing heteronomous about willing to obey a superior’s prescription because the super-
ior has prescribed it, in a discretionary way, as long as the final end is shared between
us, and we have trust also about the route. The dichotomy which the usual version of the
argument [from autonomy] relies upon is false. The dichotomy is: either our own wills
entirely or entirely the will of another. What human moral life is actually like on the
Scotist picture is a complex and rich mixture’.
228 Understanding Mor al Obligation

self-direction that is required in order for the agent to be autonomous.

Alternatively, however, the Kantian might say that any such combin-
ation is unstable as on a third conception of autonomy, this is incom-
patible with obedience of any kind, as it must involve self rule or self
government, comprising only ‘submission to laws that one has made
oneself’, so that ‘[t]he autonomous man, insofar as he is autonomous,
is not subject to the will of another’.18 It is a concern with autonomy
in this sense, as we have seen, that Kierkegaard credits with causing
the ‘mutiny’ that has put man and not God into the position of moral
legislator, of treating ourselves (either individually or collectively) as
the source of our obligations and duties. We cannot think of ourselves
as free, in this way, unless God has been displaced from his position of
authority, and we have come to occupy his place.
Now, as theists will often allow, divine command theory is indeed
incompatible with autonomy taken in this way; but they may also see that
admission as causing no difficulty for them, because any such notion
of autonomy as self-legislation can be rejected in a Kierkegaardian or
Anscombian manner, as ultimately empty, wrong-headed, or incoher-
ent.19 However, as we have seen, some of these criticisms may depend
on the assumption that this sort of autonomy commits one to a self-
legislation picture that goes ‘all the way down’, and so ends up in the
same sort of vacuousness and arbitrariness that has also been associ-
ated with the equivalent theistic picture of a divine legislator operat-
ing in a valueless universe. I have argued above, however, that while
some constructivists take Kant’s talk of self-legislation in this extreme
way, this is not in fact the correct approach; and I have also argued that
once it is seen to be compatible with elements of moral realism, then

18
Wolff 1970: 14. Cf. Feinberg 1973: 161: ‘I am autonomous if I rule me, and no one else
rules I’.
19
Cf. Adams 1999: 271, where he characterises the Kantian sense of autonomy as being
one with which divine command theory is not compatible: ‘There are some moral
theories – Kant’s being the most famous example – in which the autonomy of the
moral agent is treated as constituting the nature of the moral law. According to Kant,
the moral law is essentially a law that we give ourselves, because it is constituted and
shaped by the nature of practical reason, which is each agent’s own practical reason
although it is the same in all of us. This Kantian theory is an alternative to divine
command theories of the nature of obligation, and is clearly inconsistent with them’.
Adams himself refrains from criticising the Kantian position, but Hare and Evans
do – with the complication that they do not think this position was Kant’s because
Hare reads Kant as a divine command theorist, as we have seen, and Hare has influ-
enced Evans to give up the standard constructivist account. See Hare 2009: 176–83
and 264–6, and Evans 2004: 127–9.
 Conclusion 229

the worry about the possible emptiness of this form of autonomy as

self-legislation can perhaps be resolved – where what is morally good
is said to be a duty or obligation for us as a result of the structure of
our own wills, rather than through the willing of another being, and
so is self-legislated in this sense. We also saw that by conceiving Kant’s
position in this ‘hybrid’ manner, other concerns raised by Anscombe
or Kierkegaard can claim to be satisfactorily settled.20 If the Kantian
conception of autonomy as self-legislation can indeed be shown to be
coherent in these ways, therefore, then the argument from autonomy
does seem to tell against the latter, by showing (as the theist will gen-
erally admit) that his divine command theory treats another will as the
source of obligatoriness, where it therefore cannot accommodate such
autonomy at least to this degree.
However, of course, at this point, the theist could turn to a differ-
ent critique of the Kantian position, namely the one offered by Hegel,
to try to show that there are indeed problems with explaining in a
Kantian manner how obligation can arise from the structure of the
will, in a self-legislated and hence autonomous way; for, as we have
seen, if the Kantian accounts for this using the hybrid theory with its
associated contrast between the holy and the human will, then he may
avoid the objections mentioned above, but instead face the difficulties
raised against that theory by Hegel. In following through the debate
between the divine command theorist and the Kantian, therefore, it is
to the Hegelian critique of Kant that we must now turn.

The Hegelian argument from dualism and the

Kantian hybrid theory
Just as divine command theorists have attempted to resist the force
of Kant’s argument from autonomy, so Kantians have challenged
Hegelian claims that Kant’s position must involve an insupportable
dualism, between reason and desire, duty and inclination. For, as we
have seen, Kant’s account of obligation turns neither on a natural law
theory nor a divine command theory, but instead on his account of
the difference between the holy will and the human will: because we
are subject to non-moral inclinations, what is good or right for us to
do is presented to us in an imperatival and hence obligatory form as
involving a moral ‘must’, whereas for the holy will who lacks any such

 See Chapter 4, pp. 103–4.

20
230 Understanding Mor al Obligation

inclinations, this is not the case. Hegel’s concern (inspired in part by

Schiller) is that this account relies on a picture of us as ‘fallen’ and lim-
ited moral beings, who must always face inclinations that pull against
the moral course of action, rather than what we desire being in line
with what reason tells us is the right thing to do. Hegel therefore sets
out to overcome the ‘tension between reason and desire’ that Kant
takes to be ‘central to our moral experience’.21
In response, however, it can be argued that while Kant does indeed
rely on his distinction between the divine and human will to explain
why obligatoriness is a feature of morality for us, nonetheless Hegel
exaggerates the role that a tension of this sort is meant to play in that
distinction, so that Kant’s ‘dualism’ is in fact less extreme than the
Hegelian critique implies. Instead, it could be claimed, all Kant needs
and all he is appealing to is a weaker notion, namely the idea that non-
moral action is always possible for us in a way that it is not for the holy
will; this means that morality serves to constrain us to the extent that
it prevents us from taking this option, but where this need not involve
any great inner battle between the forces of duty and inclination, rea-
son and desire. That this is the best way to understand Kant’s position
has been argued by Philip Stratton-Lake:
The moral law appears to us as an imperative because we do not neces-
sarily will in accordance with it. It appears, therefore, as a constraint
for a finite rational will. It is easy to take the notion of a constraint
as implying that our natural inclinations are in some way essentially
opposed to the requirements of the moral law. But Kant thinks that the
moral law appears to us as necessitating, or constraining, not because
our natural inclinations are intrinsically immoral, but because of the
contingency of the connection between what we are inclined to do and
what we ought to will. The notion of a constraint should not, therefore,
be understood as presupposing a conception of inclination as essen-
tially opposed to morality (a view that Hegel and Hegelians are fond of
ascribing to Kant), but as expressing the fact that the moral law places a
rational limit on the practical possibilities open to us in certain circum-
stances, and is recognised as such. The moral law does not appear to a
perfectly rational being as a rational constraint because it does not limit
the possibilities open to such a will. This is because such a being does
not have possibilities open to it which can conflict with, and hence can
be limited by, the moral law.22

21
Schneewind 1992: 317/2010: 258.
22
 Stratton-Lake 2000: 37–8. Cf. also Stratton-Lake 1996: 50.
 Conclusion 231

On this sort of account, therefore, we can explain why Kant uses the
language of constraint and necessitation, in so far as morality limits
our options in a way that it does not for the holy will, but not in a way
that introduces the sort of dualistic picture that so concerned Schiller,
Hegel, and many others.
However, while this reading of Kant might make his position more
palatable to some tastes, the textual evidence is against it. Stratton-
Lake cites in his support the following sentence from The Metaphysics
of Morals, which may indeed seem to make nothing more than the
point he attributes to Kant, that morality constrains us merely to the
extent of limiting the various options proposed to us by our inclina-
tions: ‘The very concept of duty is already the concept of a necessitation
(constraint) of free choice through the law’. However, Kant immedi-
ately continues:
This constraint may be an external constraint or a self-constraint. The
moral imperative makes this constraint known through the categorical
nature of its pronouncement (the unconditional ought). Such con-
straint, therefore, does not apply to rational beings as such (there could
also be holy ones) but rather to human beings, rational natural beings,
who are unholy enough that pleasure can induce them to break the
moral law, even though they recognize its authority; and even when
they do obey the law, they do it reluctantly [ungern] (in the face of oppos-
ition from their inclinations), and it is in this that such constraint prop-
erly consists.23

Kant seems unequivocal here in emphasising the element of struggle

and consequent reluctance involved in obligation, and so exactly the
element in his view that Schiller and Hegel found so problematic. And
this is by no means an isolated passage, so that while he may have
thought we can develop inclinations that lead us to act well and not
just badly,24 Kant also seems to have held that where we are conscious
of a duty, we must invariably experience reluctance, resistance and a
sense of hindrance – not just an awareness of the existence of possible

23
MM 6:379 (p. 512). Cf. Allison 1990: 156: ‘insofar as we act from duty we act on the
basis of the recognition of a moral constraint and, therefore, not gladly’; and Allison
2001: 609: ‘[H]appiness must always insist on its “rights”, which entails that there will
always be “resistance” to the demands of morality when they infringe on them’.
24
Cf. Relig 6:58 (p. 102): ‘Considered in themselves natural inclinations are good, i.e. not
reprehensible, and to want to extirpate them would not only be futile but harmful and
blameworthy as well; we must rather only curb them’.
232 Understanding Mor al Obligation

non-moral options, but also the thwarted desire to take them.25 For
Kant, it seems, the concept of duty only makes sense in this context.
Moreover, if we downplay these aspects of Kant’s view, we also argu-
ably lose much of the explanatory power of Kant’s position, which is to
account for the ‘necessitating’ character of morality, the way it appears
to us to have a compelling or imperatival nature; a quality that has led
many (as we have seen) to adopt divine command theories, but where
Kant seems to rely instead on the ‘clash’ between the forces of reason
and desire, duty and inclination, to account for this experience.26 If,
however, morality is just seen as depriving us of certain possibilities of
non-moral action which are open to us but not to the holy will, then it
is hard to see why this should appear to really constrain us in the way
that Kant intends, unless the inclination to act non-morally is actually
there, in which case the opposition between morality and our non-
moral inclinations is brought back into the picture. After all, there
are many courses of action that are possible to me, but where because
I have no desire to do them, I will experience no sense of compulsion
if those possibilities are then closed off. If we go too far in depriving
Kant of his ‘dualism’, therefore, we run the risk of losing this aspect of
his account, which is required if he is to do justice to what he is setting
out to explain.27

25
See the passages referred to above, Chapter 3 note 49.
26
Cf. Sidgwick 1981: 77 (my emphasis): ‘Such cognitions [of the rightness and wrong-
ness of conduct], again, I have called “dictates,” or “imperatives”; because, in so far
as they relate to conduct on which anyone is deliberating, they are accompanied by
a certain impulse to do the acts recognised as right, which is liable to conflict with other
impulses’. As Sidgwick makes clear on pp. 34–5 (to which he is referring back here), it
is this conflict which he makes crucial to the commandingness of morality, and which
he sees as distinctive to us as moral agents, much like Kant – where also much like
Kant (I have argued) he treats this picture as compatible with a realism about what is
right: ‘In fact, this possible conflict of motives seems to be connoted by the term “dic-
tate” or “imperative,” which describes the relation of Reason to mere inclinations or
non-rational impulses by comparing it to the relation between the will of a superior
and the wills of his subordinates. This conflict seems also to be implied in the terms
“ought,” “duty,” “moral obligation,” as used in ordinary moral discourse: and hence
these terms cannot be applied to the actions of rational beings to whom we cannot
attribute impulses conflicting with reason. We may, however, say of such beings that
their actions are “reasonable,” or (in an absolute sense) “right.”’
27
Cf. Kant’s response to Schiller in Religion within the Boundaries of Mere Reason, where
Kant makes clear that he is unwilling ‘to associate gracefulness with the concept of duty
… For the concept of duty includes unconditional necessitation, to which gracefulness
stands in direct contradiction’ (Relig 6:23 note (p. 72)). Baxley chooses to downplay
the phenomenological aspect of Kant’s position, though admits it is there: see e.g.
Baxley 2010: 112. By connecting it to Kant’s idealistic account of how the moral law
 Conclusion 233

Nonetheless, it might be argued that Kant himself tones down this

distinction, in so far as he points out to Schiller that, on his view, it
is a mistake to see the virtuous person as acting joylessly or without
cheer when she acts dutifully,28 where it could then be said that in this
case no actual conflict need be occurring in the dutiful moral agent,
and that therefore Schiller and Hegel are mistaken if they think that
this is so.29 However, the fact that Kant expects and encourages us to
get joy from acting out of duty, and not just feel fearful or miserable
about it, does not show that he thought no non-moral inclinations are
involved, for the joy we are expected to feel is precisely the pleasure we
take in overcoming such temptations in a Herculean manner, where
in cultivating the taste for such pleasures we are likely to be better at
defeating them in the future. Thus, even where satisfaction is to be
found in acting dutifully (as Kant undoubtedly thought it is), this is
a satisfaction that involves the defeat of our non-moral inclinations,
not the absence of them, where it is still their presence that explains
the imperatival nature of that duty in the first place, in conforming to
which one can then feel happiness and not despair.
However, even if it is accepted that it is important for an account of
obligation to explain the imperatival force that morality has for us as a
feature of our moral experience, it might be said that alternative and
better ways of doing this can be found using other aspects of Kant’s
position, so that this way of drawing a distinction between the holy will
and our own can be abandoned, and with it any reliance on the sup-
posed dualism of the latter.
An alternative strategy of this sort has been put forward recently by
Christine Korsgaard. On the one hand, she has little sympathy with
attempts to downplay the role of necessitation as a psychological force
in our moral lives30 and to replace it with what she calls the ‘Good
Dog’ picture of the virtuous agent, as someone ‘whose desires and
inclinations have been so perfectly trained that he always does what

appears to us as finite creatures, I hope to have shown how the phenomenological

element is crucial to his approach.
28
Cf. Relig 6:23 note (p. 72).
29
See Baxley 2010: 103, 133, and 117, where she criticises Schiller for thinking that for
Kant, ‘the moral law takes the form of an imperative only insofar as a person actually
has, or is tempted by, inclinations that run contrary to moral concerns’.
30
Cf. Korsgaard 2009: 3: ‘[T]he normativity of obligation is, among other things, a psy-
chological force. Let me give this phenomenon a name, borrowed from Immanuel
Kant. Since normativity is a form of necessity, Kant calls its operation within us – its
manifestation as a psychological force – necessitation’.
234 Understanding Mor al Obligation

he ought to do spontaneously and with tail-wagging cheerfulness and

enthusiasm’.31 On the other hand, she does not think necessitation can
best be accounted for by adopting the opposite model of the ‘Miserable
Sinner’, which treats human beings as ‘in a state of eternal reform,
who must repress [their] unruly desires in order to conform to the
demands of duty’.32 Korsgaard claims that ‘[t]he opposition between
the two pictures is shallow, for they share the basic intuition that the
experience of necessitation is a sign that there is something wrong
with the person who undergoes it’, and thus they both ‘denigrate the
experience of necessitation’.33 She also argues that both pictures fail
to give an adequate account of how we are necessitated, or how it is we
come to be bound to what is right or good.34
Now, Korsgaard sees Kant’s attempt to offer an account of necessita-
tion which employs the contrast between us and the holy will as fitting
into the flawed ‘Miserable Sinner’ model; and though she accepts that
therefore Kant in part adopted this model, she thinks he was wrong to do
so, for the reasons outlined above. As she puts her position: ‘Kant him-
self seems to fall into this error in the Groundwork, when he suggests that
necessitation is only experienced by an imperfectly rational will’.35 But
she thinks we can use different Kantian materials to develop a better and
less shallow way of understanding necessitation, which traces the issue
back to the struggle for self-constitution, and what is involved in that:
I believe that these theories [associated with the ‘Good Dog’ and
‘Miserable Sinner’ models] both underestimate and misplace the role
of necessitation in our psychic lives. There is work and effort – a kind
of struggle – involved in the moral life, and those who struggle success-
fully are the ones whom we call ‘rational’ or ‘good’. But it is not the
struggle to be rational or to be good. It is, instead, the ongoing struggle for
integrity, the struggle for psychic unity, the struggle to be, in the face of
psychic complexity, a single unified agent. Normative standards – as I
am about to argue – are the principles by which we achieve the psychic
unity that makes agency possible. The work of achieving psychic unity,
the work that we experience as necessitation, is what I am going to call
self-constitution.36

31
Korsgaard 2009: 3 32 Korsgaard 2009: 3
33
Korsgaard 2009: 3–4. 34 Cf. Korsgaard 2009: 4–7.
35
Korsgaard 2009: 4 note 5; and cf. Korsgaard 1997: 240 note 52/2008: 52 note 39,
where she states that ‘the view that the will’s imperfection is what makes us subject to
an ought … is a red herring here’.
36
Korsgaard 2009: 7.
 Conclusion 235

Korsgaard thus develops what she still thinks of as a Kantian account

of necessitation, but one that makes no appeal to his conception of the
holy will, and the contrast that conception allows Kant to draw with a
finite will such as ours.
Now, it is not possible here to work through Korsgaard’s position in
any detail, as she offers a complex view that requires extensive consider-
ation to do it justice. However, I think there is an obvious potential dif-
ficulty for it, which does not seem to afflict the Kantian position which
she rejects as based around the ‘struggle to be good’. For, that position
locates duty and obligation as arising from the way in which the good-
ness and rightness present themselves to us, and thus as residing in the
moral situation, so that it is what is right and good that obligate and
necessitate; but, by contrast, the worry about Korsgaard’s account is
that it puts this necessitation in the wrong place (so to speak), where it
is the struggle of the subject to unify itself that is responsible for this
as a ‘psychic force’, where this would seem to lack any underlying con-
nection to anything normative. Someone might well think, therefore,
that if there is any degree of struggle in morality, and if this is to be
used to explain its obligatory force, then this struggle should revolve
around the good and the right, not on the difficulties faced by the self-
constituting subject – and this is more readily available on the picture
of Kant’s position that Korsgaard rejects than on the one with which
she replaces it.
It would seem, then, that responses to Hegel’s concerns that
attempt to ‘soften’ the dualism that he identifies, or to make it
less central to the Kantian story, do not do justice either to Kant’s
own outlook, or to the explanatory force of his account. The ques-
tion still remains, however, whether the Hegelian is right to find
anything objectionable in this account. We must therefore consider
whether the Kantian can respond adequately to what I take to be the
Hegelian’s two most fundamental concerns: firstly, that the account
of duty which Kant offers results in a picture of our moral capacities
that is both overly negative and demeaning; and secondly, that this
account ends up compromising the very autonomy that Kant set out
to champion and to make fundamental to his position. As we have
seen, both challenges may be said to arise out of Schiller’s response
to Kant, and both are well put by him in a central passage from On
Grace and Dignity, where he asks: ‘Does mankind have to be accused
and humiliated simply by the imperative form of the moral law, and
does the most sublime document of its greatness also have to be a
236 Understanding Mor al Obligation

certification of its frailty?’;37 and, ‘Just because the moral weakling

would like to introduce a certain laxity into the law of reason, to
make it a toy for his own convenience, does this mean that a rigidity
has to set in, transforming the most powerful expression of moral
freedom into merely an honourable kind of servitude?’38
As we have discussed, when it comes to the first issue, a contrast is
often drawn between Kant’s position and Aristotle’s, who is said to
have had a more exalted view of our moral capacities, according to
which virtue is both attainable by us and involves a harmony between
inclination and correct ethical behaviour. This does not mean that the
virtuous agent can give herself over to inclination, and abandon reason
in determining what she will do: but (it is suggested) she can be in a
position where her inclinations stand in no tension with reason, pull-
ing her away from the moral course of action. This way of understand-
ing Aristotle’s position has been presented as follows:
According to this account, Aristotle does not believe that if a person
is acting out of virtue, there is any opposing motivational force that
interferes or generates conflict. This idea has been put by saying that
in the relevant sorts of cases there is no ‘internal friction’, as there is in
someone who, out of ‘strength of will’ resists a desire that runs counter
to what he has decided is best to do. If there is no such friction, however,
then it seems that there should be no occasion for a person who acts out
of virtue to command some part of himself to do what he has decided is
best; and therefore there should be no occasion for imperatives.39

It may seem, then, that we can straightforwardly turn to Aristotle if we

no longer wish to find ourselves ‘accused and humiliated’ in a Kantian
manner, by finding in his model of the virtuous person just the kind
of harmonious and ‘frictionless’ moral life that Kant is only prepared
to attribute to the holy will, and not us; and as we have seen, it is clear

37
GD 20:286 (p. 151). 38
GD 20:285 (p. 151).
39
White 2002: 119. Cf. also pp. 224–5, where White provides a nice illustration of what
he takes Aristotelian virtue to be like on the standard reading: ‘We could compare the
good man to someone who, instead of being either gluttonous or completely uninter-
ested in food, finds his appetite adjusted perfectly to his own correct judgements about
what he rationally ought to eat. Thus, as he walks down the buffet table, he judges cer-
tain foods good for him and others not, and finds his appetite waxing and waning in
glad unison with his judgements about nutrition’ (p. 225). White puts forward Urmson
and McDowell as proponents of the ‘standard reading’, citing Urmson 1988, especially
pp. 66–7, and McDowell 1979 and 1980.
 Conclusion 237

that both Schiller and Hegel took themselves to be following Aristotle

in making a move of just this sort.
There are, however, several ways in which the Kantian might respond
to any such use of Aristotle’s conception of virtue against them, some
of which deny the contrast that is here drawn between their position
and the one actually adopted by Aristotle,40 and some of which accept
the contrast but argue nonetheless that this tells in their favour.
A first way to deny the contrast might be to argue that even on
Aristotle’s conception of virtue, conflict of a Kantian kind is recog-
nised to be present. Thus, as we have seen previously,41 when Aristotle
comes to speak of courage, for example, he allows that the courageous
person will still feel emotions and inclinations (such as fear or a desire
to avoid injury) that in some sense are out of line with the course of
action she sees to be right; Aristotle’s position here may therefore be
said to be little different from the Kantian picture, and also to be more
plausible as an account of the virtue (for if a person did not feel such
things and have these desires, would they not be showing the indif-
ference to life more typical of the psychopath than the courageous
individual?).42 So, when it comes to this kind of case, it could be said,
the Aristotelian and Kantian accounts turn out to be broadly similar,

40
Within the contemporary Kant literature, there has been an increasing desire to find
a rapprochement between the two thinkers: for notable examples of this trend, see
Korsgaard 1996b and Sherman 1997.
41
Chapter 4 pp. 122–3
42
Cf. Broadie 1994: 234: ‘[T]he virtuous person [would be] a repugnant person if being
virtuous entails never desiring to do or have what it would be currently better not to.
For whenever the right course of action requires one to give up some otherwise attract-
ive option, the virtuous person must care not at all about giving it up. He must regard
it as no loss; he must feel no tremor of regret at finding himself placed so as to be
obliged to forgo that thing. Yet what if the case is one where, on moral grounds, he has
to say No to some honour – and the honour is both significant and in this case richly
deserved? Or if he has to incur a serious financial loss, one that perhaps affects others
for whom he is responsible? Or he must leave the place where he has made his life, and
say goodbye for ever to valued friends? On the picture suggested (and often suggested
as Aristotle’s), virtue is not only the opposite of admirable, but – especially for the
good person – psychologically impossible! Not to mind or care about the things which
are morally precluded indicates an agent who fails to value such goods as they should
be valued. But how can it be consistent with virtue not to be alive to the value of what
should be valued?’. Cf. NE, Book iii, §7, 1115b24–28, p. 1761, where Aristotle argues
that the person who feels no fear when faced by danger is not brave but mad: ‘Of those
who go to excess he who exceeds in fearlessness has no name (we have said previously
that many states have no names), but he would be a sort of madman or insensible per-
son if he feared nothing, neither earthquakes or waves, as they say the Celts do not’.
238 Understanding Mor al Obligation

where the latter is no more ‘disparaging’ of our capacities than the

former, but where both are simply realistic portraits of how things
stand for us.
However, it can be replied, a contrast between Aristotle and Kant
still remains. For, in the first place, while this is perhaps not always
completely clear, it appears that for Kant the contrary emotions and
inclinations represent a greater temptation to us than they do for
Aristotle’s virtuous individual,43 who may indeed feel those emotions
and desires, but who still arguably gives them no real weight in her
deliberations, so that her action is still ‘frictionless’ in this sense; and,
as we discussed previously, while the attempt has been made to ‘soften’
the Kantian picture in this respect, there appear to be good exegetical
and philosophical reasons not to do so. In the second place, while when
it comes to a virtue like courage, the Aristotelian might be prepared to
allow that such countervailing emotions and inclinations can co-exist
with it, this seems less plausible in the case of other virtues, such as
benevolence or charity. For, as Philippa Foot points out: ‘charity is …
a virtue of attachment as well as action, and the sympathy that makes
it easier to act with charity is part of the virtue’.44 In this case, there-
fore, if one did not have a kindly disposition, but instead felt emotions
and desires of a misanthropic sort, then one could scarcely be said to
possess the virtue, where here there is a divergence from the case of
courage. Likewise, it might be said, it is plausible to think of temper-
ance as a case in which the individual feels no attraction towards those
things that in the circumstances would constitute excess. Thus, even if
the contrast between Kant and Aristotle may perhaps not apply to all
the virtues, it arguably at least applies to some of them.
Nonetheless, it could be said, there is a second way of denying that
any such contrast exists, which is to claim that just as Kant allows for
the case of a holy will, but treats it as an ideal, so too Aristotle’s virtu-
ous individual is equally an ideal, where no less than Kant he thought

43
Cf. MM 6:383 (p. 515), where Kant characterises finite holy wills as beings who ‘could
never be tempted to violate duty’. Cf. also 6:396–7 (pp. 526–7). Given that in Relig,
Kant portrays Jesus as having been subject to temptation (see 6:62 (p. 105) and 6:64
(p. 106)), this suggests that he was not a finite holy will – which may make it hard to
know who Kant would have put into this category (angels, perhaps – or maybe certain
saintly characters?). Alternatively, it could be said that Jesus still counts as a finite holy
will, because although he was subject to temptations, he nonetheless had no propen-
sity to succumb to them thanks to his innate purity of will (cf. Baxley 2003: 564 note 9
or 2010: 55 note 9) – where it is this innate goodness that distinguishes him from us.
44
Foot 1978: 14. Cf. also pp. 11–12.
 Conclusion 239

of us as falling short of it.45 However, while it may be true that for

Aristotle, most of us may be no more than continent most of the time,
there is no reason to think he believed that it was unattainable in the
manner of Kant’s holy will, which belongs to a different order of being
altogether.46 In fact, had he held this, it would seem he would not have
bothered to discuss virtue at all, as he puts aside discussion of ‘super-
human’ virtue for just that reason, as an inappropriate topic for a book
like the Ethics, which is supposed to be relevant to creatures like us.
It would seem, then, that a contrast between Aristotle and Kant still
remains.
Another way of dealing with the situation, however, might be to
accept the contrast, but to claim that Kant’s position is to be preferred.
A first reason to prefer it might be that it simply provides a more accur-
ate and plausible account of what moral life is like for us as the crea-
tures we are. Or, secondly, it could be argued that if one looks deeply at
why Kant sees a tension or ‘friction’ here, but Aristotle does not, then
it will be seen that the grounds underlying Aristotle’s more optimistic
outlook are themselves questionable: namely, that he is committed to
eudaimonism, or that he operated with a comparatively undemand-
ing conception of what could be asked of a moral agent. We therefore
need to consider each of these claims in turn.
The first Kantian response, then, might be that Kant’s conception
of how we are as moral agents is not in fact ‘demeaning’, but simply
level-headed and insightful. Thus, much as we might like to think of
ourselves as conforming to a more Aristotelian model, this is in fact
not the case. That does not mean, on the one hand, that Aristotle has
a completely rosy view of virtue as completely ‘frictionless’, for as we

45
White raises this possibility, noting that ‘It might even be the case that “the good man”
is, in Aristotle’s view, a sheer idealization which is never realized in any actual human
being’ (White 2002: 225). Cf. also McDowell 1978: 28/1998: 92: ‘[Aristotle’s] view of
virtue obviously involves a high degree of idealization; the best we usually encounter
is to some degree tainted with continence. But in a view of what genuine virtue is,
idealization is not something to be avoided or apologized for’. And cf. Grenberg 2005:
52–3.
46
As Sarah Broadie has pointed out to me, the Nicomachean Ethics (especially) is addressed
to the politicos or statesman running the city, and in the last chapter deals with the
question of how to set up political institutions conducive to good moral education.
This means that the concern with virtue is concern for what moral educators might
reasonably hope to foster city-wide, and thus for what is broadly attainable. As some
evidence that this was Aristotle’s view, see NE Book i, §8, 1099b18–20, p. 1737, where
Aristotle says that happiness on his account is available to many because virtue, the
main component, is acquirable by our efforts.
240 Understanding Mor al Obligation

have seen in the case of courage and other examples, this is not the
case; but on the other hand, while Kant’s outlook still differs from
Aristotle’s (as we have seen), this still does not mean that Kant in his
turn is committed to an excessively gloomy picture. For, while we have
resisted readings that try to do away with the tension in our moral
experience altogether, it could be said that this tension should still
be understood in a relatively moderate way, that is arguably realistic
rather than degrading.
Thus, for example, in a recent account of Kant’s position,47 Anne
Margaret Baxley has claimed that it is a mistake to think that Kant’s
virtuous individual, whenever she comes to act, does so in the face of a
genuine inclination to do otherwise. On the other hand, this does not
mean that such inclinations are not ‘in the background’, but in such a
way as to have been repressed so that at the time of acting they do not
get in the way or represent any real temptation – where for the merely
continent person they do. Baxley thus compares the situation to that
of the recovered and the recovering alcoholic or food addict. For the
recovered alcoholic, who has already gone through a maintenance
programme to suppress her desires and to build up her resistance to
them, while it is the case that she can be tempted, she is not actually
tempted by a drink – though there is still enough ‘friction’ to explain
the sense of necessitation involved when she turns down a whisky. On
the other hand, she is different from the recovering alcoholic, who
also turns down the offer of a dram, but who has to struggle against a
real temptation to grab the drink when she does so. Similarly, there-
fore, one might think of the virtuous agent as undergoing a constant
process of maintenance and control with respect to her inclinations,
so that in that sense virtue for her is not entirely ‘effortless’ in the man-
ner of a holy will, while still she differs from the continent person in
not feeling genuinely tempted to act non-morally when the occasion
arises, in part because her ongoing programme of maintenance has
been successful. Thus, it could be said, while from an Aristotelian (or
Schillerian, or Hegelian) perspective, Kant’s virtuous person is still no
more than continent, in fact Kant nonetheless has a virtue/continence
distinction of his own that he can draw; and, it is one that both accounts
for why even the virtuous person will feel the force of the moral ‘must’
and thus have an experience of duty, and is also more realistic as an

47
 Baxley 2010: 134.
 Conclusion 241

account of what virtue in human beings actually involves, where the

less strenuous Aristotelian model is still implausibly rosy.
Now, of course, it could still well be that the Hegelian will claim
that this Kantian picture remains ‘demeaning’: for to compare the
virtuous individual to the recovered alcoholic is still seemingly to put
her in a rather bad light, and still to endorse the kind of ‘autocratic’
picture that Schiller deplored as an account of virtue. But Kant could
claim that as a model for what virtue is for human beings, this is the
best that can be hoped for, if we are to take seriously the fact of our
‘fallen’ natures, and the real possibility that our reason will be over-
come – just as the recovered alcoholic must recognise the constant
threat of a return to her addiction and the desires that go with it. Were
this danger to be eradicated altogether, Kant holds, we would indeed
have attained the status of holy wills, and so be beyond both duty
and virtue. Thus, as Baxley puts it: ‘Virtue, on this Kantian picture,
should therefore not be mistaken for a fixed, settled moral disposition
acquired by habituated feelings and desires (or any other feature of
sensibility) towards right action [as it is on the Aristotelian concep-
tion]. It is rather a lifelong project requiring our sustained efforts in
maintaining a moral way of thinking’.48 In the end, therefore, while
the Kantian outlook may show less faith in our nature as moral beings
than the Aristotelian is inclined to feel, in a way that then underpins
Kant’s correspondingly more ‘strenuous’ picture of virtue, the Kantian
can nonetheless respond by claiming that this lack of faith is (sadly)
fully justified, and his conception of virtue is thus better informed
than that of his critics.
Such claims about human nature are scarcely to be adjudicated
here, however, and no doubt the suspicion concerning Kant will still
remain, that he exaggerates the difficulties that stand in our way as
moral beings – to use Baxley’s analogy, it is more attractive to think of
ourselves as well-adjusted drinkers who can handle alcohol with little
further thought, rather than recovered alcoholics who must constantly
be on guard unless we slip back into the abyss. For an Aristotelian
process of moral education to have got merely to this point, therefore,
is arguably for that process to have failed. And, the Hegelian will no
doubt feel, Kant’s picture is still closer to a model of continence rather

48
Baxley 2010: 131–2; cf. also pp. 80–1. For another treatment that defends Kant’s
account as offering a realistic view of human nature and our capacities, see Grenberg
2005: 49–55.
242 Understanding Mor al Obligation

than of genuine virtue, and still too closely tied to a Christian concep-
tion of the ‘fall’ and its associated notion of ‘radical evil’.49
It might be said, however, that this is also too bleak a view of the
Kantian position, which is not (as in the case of the recovered addict)
that misdirected desires and inclinations are always something that
can happen to us, in the sense that they will happen unless we remain
vigilant in controlling them; rather, that these desires and inclinations
are merely logically or metaphysically possible for us, just because we
are human and not divine – this makes us capable of having non-moral
inclinations in a way that a holy will cannot, but not in a way that
implies this is likely or a real threat to our behaviour.50 It could then be
argued that Kant’s critics must concede this point, for surely even the
very best of us, no matter how good we actually are and how unlikely
it is that we will be tempted to act otherwise, may still possibly be so
tempted, in this minimal sense. To think otherwise, it may be said,
really is to confuse us with gods in an absurdly self-aggrandising and
self-deluding manner.
Now, I think the response to this sort of view can have two aspects.
At an interpretative level, it seems implausible to claim that Kant’s pos-
ition was as attenuated as this. For, in so far as Kant treats the self as
a member not just of the ‘intelligible world’, but also the ‘world of
sense’, he takes it that we are commonly and easily prone to ‘impulses
of sensibility’ that lead us astray,51 and not merely that such impulses
are logically or metaphysically possible for us – even if they can be

49
Cf. Fackenheim 1954: 340, who notes that Schiller wrote to Körner characterising Kant’s
position on ‘the propensity of the human heart to evil’ as ‘scandalous’ [empörend] to
his feelings (28th February 1793, 26:219; cf. also the letter to Bartholomäus Fischenich
20th March 1793, 26:235, and GD 20:286 note (p. 151)). As far as I know, Hegel does
not explicitly discuss Kant’s doctrine, though of course he does discuss the issue of evil,
original sin, and related notions extensively, both in PR §139 7:260–65 (pp. 167–70),
and in his various lectures cycles on religion. For a nuanced discussion of Hegel’s pos-
ition, see Dews 2008: 81–117.
50
Cf. Baxley 2010: 112, my emphases: ‘But, for Kant, the imperatival force of the moral
law presupposes only the possibility of transgressing the moral law, not the existence
of actual contrary-to-duty inclinations. Kant would therefore insist that moral norms
take the form of imperatives even for Schiller’s beautiful soul, simply in virtue of the
fact that it is possible that she can violate duty (even if she never does and even if she is
never tempted to act contrary to duty)’.
51
Cf. GMM 4:454–5 (p. 100–1). Cf. also NF 19:222–3 (p. 456): ‘One must coerce oneself
to prudent and morally good actions. Hence imperativi. The reason is that one’s power
of choice [Willkür] is also sensible, and the first movement stems from the sensible. The
more one can coerce oneself through pragmatic coercion, the freer one is. This coer-
cion nevertheless occurs per stimulus, but indirecte, namely one proceeds in accordance
 Conclusion 243

controlled in the manner Baxley suggests. So, it looks as if Baxley’s

recovered alcoholic model is the best interpretation of Kant’s position,
which the Aristotelian and Hegelian can then reasonably challenge on
the lines we have considered. And secondly, at a philosophical level,
the issue is still whether the Kantian needs the clash between reason
and desire or sensibility to be more genuine than this picture suggests:
for why, it if is merely logically or metaphysically possible that I could
have desires that go against the moral law, should that explain the
sense of imperatival or necessitating force that belongs to it? Again,
this contrasts with Baxley’s picture of the recovered alcoholic, where
real control and self-restraint is still going on in a way that can account
for the necessitation involved; but this would be lost when the position
is that desire is opposed to reason only in this very weak sense, as some-
thing that could (as a logical or metaphysical possibility) depart from
it. So again, it is the position discussed previously that seems to be the
one to be taken seriously, but which may then intelligibly be disputed
by the Aristotelian or Hegelian critic.
However, we can now turn to a second type of Kantian response
to his opponent, which is not just to question the applicability of the
Aristotelian model to creatures like us, but to cast suspicion on the
thinking that underpins it, where the first objection of this sort cen-
tres on Aristotle’s eudaimonism, and the claim that he therefore sees
little need for effort in being virtuous, as ultimately such virtue is in
the interests of the agent herself. Now, Aristotle’s eudaimonism is a
complex and highly controversial issue, as is eudaimonism as such,
and Kant’s relation to it, where none of this can be dealt with fully
here.52 In general, of course, it may certainly be plausible to claim that
one powerful way in which to explain how our inclinations might fall
into line with what is right to do, is if doing the latter will lead to our
greater happiness or well-being. However, while Aristotle may indeed

with reflection. Moral coercion is external through the power of choice of another;
and if we are free from this, inner coercion still remains’; and CPrR 5:84 (pp. 207–8):
‘For, being a creature and thus always dependent with regard to what he requires for
complete satisfaction with his condition, he can never be altogether free from desires
and inclinations which, because they rest on physical causes, do not of themselves
accord with the moral law, which has quite different sources; and consequently, with
reference to those desires, it is always necessary for him to base the disposition of his
maxims on moral necessitation, not on ready fidelity but as respect, which demands
compliance with the law even though this is done reluctantly’.
52
For some further discussion, see Irwin 1996 and 2009: 124–46.
244 Understanding Mor al Obligation

have thought there is some such connection between acting morally

and living well, he does not seem to have been so crudely eudaimonis-
tic as to think that it is only through seeing this connection that an
individual will come to be virtuous. On the contrary, if the individual
is brought to feel and desire rightly on these grounds, he is surely lack-
ing in virtue.53 Thus, it could be said, Aristotle believed that there was
no inevitable tension between our desires and our moral actions, not
because he took it that the moral agent would see the latter as satisfy-
ing the former in some egocentric or prudential form, but because he
believed the kind of socialisation and habituation involved in moral
education could take us beyond such egocentric or prudential con-
cerns, thus alleviating the tension accordingly.
Finally, we can conclude this discussion of Aristotle’s relation to
Kant by considering a second critical diagnosis of Aristotle’s position,
which is to suggest that his outlook is more ‘frictionless’ than Kant’s
here, not because of his eudaimonism, but because his conception of
our ethical requirements was rather modest.54 Indeed, it is arguable
that Aristotle is precisely the kind of ‘honest pagan’ that Kierkegaard
had in mind, who he thought would be understandably outraged by
the kind of extreme demand on us made by the love commandment
(as Kierkegaard conceives it). However, how far this kind of critique of
Aristotle (even if true) is effective in showing that Kant’s position is to
be preferred will be considered further in the next section, when we
come to consider the comparable challenge to Hegel, so discussion of
it can be postponed until then.
Before we come to that, however, I will close this section by con-
sidering the second objection to Kant’s position that concerns the
Hegelian, which is not just that it is ‘humiliating’, but also that it ‘trans-
forms the most profound expression of moral freedom into merely an
53
John McDowell, in particular, has protested against seeing Aristotle’s position as
eudaimonist in this crude form: ‘[The] proper manifestation [of virtue] is renoun-
cing, without struggle, something that in the abstract one would value highly (physical
pleasure, security of life and limb). The lack of struggle is ensured by keeping the
attention firmly fixed on what Aristotle calls “the noble”; not by a weighing of attrac-
tions that leads to the conclusion that on balance the virtuous course is more desir-
able’ (McDowell 1978: 27/1998: 91–2).
54
Cf. LE 29:604 (p. 229), where Kant notes that the ethical ideal of holiness is a specific-
ally Christian innovation, unknown to the Greeks: ‘In the Gospel we also find an ideal,
namely that of holiness. It is the state of mind from which an evil desire never arises.
God alone is holy and man can never become so, but the ideal is good … The ancient
philosophers never got to that point, though it has been said that they have enunciated
all that is moral in the Bible’.
 Conclusion 245

honourable kind of servitude’, as Schiller put it. The concern here is

that, on the Kantian model, while the individual is no longer bound by
an external will, her reason nonetheless still plays a constraining role,
thereby taking away her autonomy in a different but no less significant
manner. Richard Rorty, for one, clearly shares this suspicion; he there-
fore treats Kant as a ‘transitional figure’, who ‘substituted “Reason”
for “God” as the name of a law giver’, giving up divine commands but
setting one part of ourselves over another instead – thus, Rorty thinks,
Nietzsche was right to say that ‘a bad smell of blood and the lash hangs
over Kant’s categorical imperative’.55
An obvious response for the Kantian to make to this sort of objec-
tion is to argue that the conception of autonomy being employed here
by their critic is misguided: whilst autonomy might require freedom
from external constraint, there is no loss of autonomy when that con-
straint is internal, and thus when reason rules over the agent’s inclina-
tions. Indeed, it could be argued, it is precisely when the individual
exercises authority over these inclinations that autonomy is realised, as
then the individual shows herself to be in reflective control over her-
self in a way that fits perfectly with the etymological idea of autonomy
as ‘self-rule’.
Now, it is unlikely that the concept of autonomy is clear and
unequivocal enough in its meaning to allow any appeal to concep-
tual analysis to settle this dispute.56 For someone whose sensibility
has been shaped by Kant, there may indeed appear to be no lack of
autonomy in an individual whose reason is said to be in command
in this manner. Or, it could be said, even if we do not have auton-
omy here, we nonetheless have what Kant called ‘autocracy’, which
involves ‘mastery over oneself’,57 and the ‘capacity to master one’s

55
Rorty 2004: 198/2007: 187. Rorty presumably has in mind the following passage from
Nietzsche’s Genealogy of Morals: ‘It was in this sphere then, the sphere of legal obliga-
tions, that the conceptual world of “guilt”, “conscience”, “duty”, “sacredness of duty”
had its origin: its beginnings were, like the beginnings of everything great on earth,
soaked in blood thoroughly and for a long time. And might one not add that, funda-
mentally, this world has never since lost a certain odor of blood and torture? (Not
even in good old Kant: the categorical imperative smells of cruelty.)’ (Nietzsche 1980
5:300/1969: 65, Essay 2, §6). Cf. also Schopenhauer 1962: §4, pp. 648–9/1965: 54–5:
‘Now as theological ethics is essentially dictatorial, the philosophical has also appeared
in the form of precept and moral obligation … A commanding voice, whether coming
from within or from without, cannot possibly be imagined except as threatening or
promising’.
56
For a discussion of its myriad forms, see Dworkin 1988 and O’Neill 2000.
57
LE 27:362 (p. 139).
246 Understanding Mor al Obligation

inclinations when they rebel against the law’.58 However, from a more
Hegelian perspective, it will appear that freedom cannot be properly
realised until some greater degree of reconciliation has been recog-
nised within the individual, and harmony achieved – only then am I
‘with myself’ and therefore free.59 And, of course, Kant himself allows
that it is more admirable if no such exercise of authority is needed,
which is why the holy will is set above us as an ideal.60 Whether or not
we qualify as heteronomous in so far as we do not achieve this ideal,
the more fundamental question in dispute, therefore, is whether Kant
is right to treat such harmony as unattainable by us as human beings,
or whether the Hegelian can justifiably claim that it is possible for
us to be wills of this ‘higher’ sort, beyond the kind of dualism Kant
seemed to think was inevitable for us.
Answering this question, however, will take us back again to
Kierkegaard’s critique of Hegel: for Kierkegaard’s objection to the
Hegelian is that there is a price to be paid for holding that there is
no such dualism, namely that the moral demand is correspondingly
reduced, to ensure that it does not go too much against the grain of
our natural dispositions. And it should come as no surprise that the
Kantian and Kierkegaardian can make common cause against the
Hegelian on this issue. For, while I have argued above that Kant is not
a divine command theorist, and so unlike Kierkegaard does not use
the strenuousness of the moral law as an argument in favour of the
latter position, commentators like Hare are nonetheless right that like
Kierkegaard, there is a significant role for a ‘moral gap’ in his the-
ory, which then allows him to introduce a belief in God on a practical
basis, as making sense of how the problem of ‘radical evil’ might then
be resolved through divine assistance.61 It is therefore no surprise that
these Kierkegaardian considerations should enter into the dialectic
58
MM 6:383 (p. 515). For further discussion of this notion of autocracy in Kant, see
Baxley 2010 esp. pp. 50–84.
59
Cf. PR §133Z, 7:251 (p. 161).
60
In the Metaphysics of Morals, Kant quotes from Albrecht Haller’s poem Über den Ursprung
des Übels that ‘Man with all his faults / Is better than a host of angels without’, where
he then notes that because of the kind of self-constraint that it involves ‘[v]irtue so
shines as an ideal that it seems, by human standards, to eclipse holiness itself, which is
never tempted to break the law’. However, Kant makes clear that he thinks it would be
a mistake to value virtue over holiness in this way. See MM 6:396–7 (pp. 526–7); see
also Relig 6:65 (p. 107) and LR 28:1077 (p. 411).
61
Cf. Hare 2002: Chapters 1–3. And cf. Relig 6:44–5 (p. 90). Hare argues that, when it
comes to the problem of radical evil, Kant’s own attempt to use divine assistance to
bridge the gap fails, as it is in internal contradiction with his commitment to the ‘Stoic
 Conclusion 247

between the Kantian and Hegelian at this point, over whether it can be
right to assume that no such gap exists, or whether to hold that it does
not is to either artificially ‘puff up’ our moral capacities, or to compla-
cently lower the moral demand. The debate between the Kantian and
the Hegelian therefore pushes us on to the next position in the circle,
which is the Kierkegaardian critique of Hegel’s approach.

The Kierkegaardian argument from complacency and

the Hegelian social command theory
The suspicion that Hegel’s social and moral philosophy is somehow
conservative and limited has of course hung over his Rezeptionsgeschichte
since its very beginning. However, in recent years that suspicion has
come to be increasingly challenged, as the progressive and critical
aspects of Hegel’s work have been rightly underlined.62 One familiar
way in which to proceed with this kind of re-evaluation is to note that
while Hegel was indeed concerned to avoid any empty utopianism,
nonetheless his account of the social structures that he upholds does
not pertain to just the existing state of his time, but to one that is fully
‘rational’ and developed, in a way that takes his work beyond an appeal
merely to how things happen to be63 – in Hegel’s terminology, there-
fore, his focus is not simply on existence, but on actuality, which is the
condition of things when they have properly realised their natures.64
Now, in response to the Kierkegaardian concerns about Hegelian
Sittlichkeit amounting to no more than an ethic of fulfilling one’s
socially determined duties, it seems perfectly possible to adopt a simi-
lar strategy, and to argue that it is only with a rational state that Hegel
would expect Judge William’s vision of ethical life to be realised, in
which the duties falling to the individual are not expected to be too
strenuous, as a good deal is taken care of through the state. Thus, it
could be argued, Hegel only envisages this happening within the state
when it is fully developed, where problems of poverty, alienation, lack
of recognition, large-scale inequality, and so on have been resolved, so

maxim’, ‘that a person herself must make or have made herself into whatever, in a
moral sense, whether good or evil, she is to become’ (p. 60).
62
For a representative product of this reappraisal of Hegel’s work, see Stewart 1996.
63
Cf. the papers by Yovel 1975; Fackenheim 1969; Jackson 1987, all reprinted in Stewart
1996 – where this trope is now commonplace in the Hegel literature. For some further
discussion, see Stern 2006.
64
Cf. EL §6, 8:47–8 (pp. 28–30).
248 Understanding Mor al Obligation

that we can indeed lead lives in which our individual duties no longer
put a weighty burden on us – just as, in a system of well-organised
taxation and social care, the need for us to come to the aid of others
would in general not arise, as the individual is simply called upon to
play their part within the whole. In a sense, therefore, no less than
Kant or Kierkegaard, Hegel does not expect us to be able to live mor-
ally without assistance; but here, the assistance comes in the form of a
social system which takes away the kinds of circumstances that place
a heavy expectation on individuals, thereby ensuring that the actual
demands made of us can in fact be met with relatively little difficulty,
in ways that should fit comfortably enough with the dispositions of a
properly brought-up moral individual – much like the demands of a
well-regulated marriage, such as Judge William’s.
However, the convincingness of this reply may then depend on how
realistic one takes Hegel’s conception of the rational state to be, which
may in turn lead to a dilemma for the Hegelian position. For, if the
Hegelian makes the state something that lies within our grasp rela-
tively easily, then (from where we now stand), it seems unlikely that
strenuous requirements for aid and assistance to others will go away.
‘The poor’, it might be said, ‘are always with us’, as are other forms of
social injustice, need, and hardship, where the moral demands they
set up look sufficiently difficult as to generate a tension between what
moral action is called for and what our inclinations may be. On the
other hand, the Hegelian could respond to this objection by insisting
that within the rational state, all these difficulties can indeed be over-
come. But then the danger is that this position will seem implausibly
utopian, and open to just the sort of objections that Hegel himself
directed against fanciful positions of this kind.
In order to avoid this dilemma, the Hegelian might choose to take
a different approach: this is to challenge whether the Kierkegaardian
is justified in claiming that anything as demanding as the love com-
mandment really is a moral requirement on us. Kierkegaard, after
all, expected many to be ‘outraged’ by it, and that the ‘honest pagan’
would reject it as absurd: who is to say that this is not the right reac-
tion? As Simon Critchley notes, and as Kierkegaard fully recognised,
one response to the love commandment is simply to dismiss it as over-
inflated and fanciful, and not really part of morality at all:
This is a ridiculous demand! Just consider for a moment what Christ is
saying to his audience: you might have heard the wisdom of Leviticus
 Conclusion 249

that you should love your neighbour as yourself, but that is not enough,
you should also love your enemies, you should love those who curse, des-
pise, hate and willfully persecute you … Christ’s argument here is that
if you love only those who love you in return, then you are not open to
the more radical demand of the stranger, the foreigner, the adversary.
If you love only your own brethren, the people of your tribe, nation or
community, then you are no better than the publicans, the publicani,
those docile servants of Gentile oppressors who dutifully do whatever
the Romans ask of them. What Christ demands of his audience, which
is – unless it might be forgotten – a Jewish audience, is that if they wish
to be truly the children of their Father, God, then they must subject
themselves to this exorbitant demand. That is, Christ is asking his audi-
ence to be perfect, god-like, ‘even as your Father which is in heaven’.65

Of course, Kierkegaard still might be right in saying that with the advent
of Christianity, God was then able to play a role in our moral thinking
that made such moral demands thinkable; but once Christianity has
been rejected, is there not then reason to go back to the more ‘pagan’
perspective, and so to accept a correspondingly reduced conception
of what morality requires? While, of course, there have always been
and will continue to be those who stress the radical call that morality
makes on us, there are also those who would reject this as mere rhet-
oric and bad-faith, insisting instead that, as James Griffin has put it,
‘[m]oral norms must be tailored to fit the human moral torso’, where
‘a norm that ignores the limited nature of human agents is not an
“ideal” norm but no norm at all’.66
However, this sort of response may just seem to reinforce the kind
of complacency charge that Kierkegaard makes against Hegel: for,
on this basis, might we not be forced to abandon such central moral
notions as the love commandment, and the associated idea that we
should love our enemy, on the grounds that these cannot reasonably
be said to fit our capacities, as such ‘tailoring’ requires? In mitigation,
perhaps, the Hegelian could remind his opponent that we are talking
here about what is obligatory, not of what is good or right: it is still com-
patible with the Hegelian view that great moral goodness lies in these
actions, where all that is being questioned (on this line of response) is

65
Critchley 2007: 52. For further discussion of the love commandment in a biblical and
theological context, see Goodman 2008.
66
Griffin 1992: 131. Cf. also Griffin 1996: 105. As part of his argument, Griffin appeals
to the principle of ‘ought implies can’. I discuss whether he is right to do so in Stern
2004a.
250 Understanding Mor al Obligation

whether it is intelligible to say that we are obliged to undertake them,

or whether they are in fact supererogatory for creatures like us.
Moreover, it could be said, precisely because Hegel has a more opti-
mistic and Aristotelian conception of how our dispositions can be
moralised, and taken beyond merely narrowly prudential and egocen-
tric concerns, there is no reason to think that this process of ‘tailor-
ing’ will leave us with no norms of recognisable moral substance at
all: even while wanting to go with the grain of our natures more than
Kant, Hegel believes we nonetheless have moral capacities enough to
make it the case that the normally virtuous individual will display con-
cern for others in numerous ways, even if this might not satisfy the
more exalted requirements of a Kierkegaardian. To the Kantian and
the Kierkegaardian, perhaps, this may look suspiciously like ‘puffing
up’ our moral capabilities in a dubious way, in order to avoid redu-
cing the moral demand too far; but Hegel’s counter-claim is precisely
that the Kantian and Kierkegaardian conception of those capacities is
excessively pessimistic, and perhaps too easily influenced by Christian
preoccupations with our ‘fall’.
In the end, though, the Hegelian will still probably agree that even
if his greater optimism here is justified, and no mere empty ‘puffing
up’, it remains the case that the love command, at least as interpreted
by Kierkegaard, is still extremely demanding for us. So the dialectic
will come back to the question whether Kierkegaard is right to accept
such demands as part of morality, or whether in scaling them back,
Hegel is either responding to the radical nature of ethics in a compla-
cent and bourgeois manner which says it is all right to ‘do what one
can’, or relying on a utopian ideal of the state that merely papers over
the challenge that these commandments represent, and which in the
end need to be made sense of in theistic terms if we are to do proper
justice to them.67

67
Cf. LPRiii: 53–4 (p. 118), where Hegel discusses the love commandment. On the one
hand, taken as prescribing ‘love of humankind in general’, Hegel dismisses it as a
‘lame abstraction’, on the grounds that ‘[t]he human beings whom one can love are
a few particular individuals. The heart that seeks to embrace the whole of humanity
within itself indulges in a vain attempt to spread out its love until it becomes a mere
pretense [Vorstellung], the very opposite of what love is’. On the other hand, Hegel
takes Jesus to have been telling the apostles that they should work for the good of
the community and thus for others in this sense: ‘[They are] to make only this unity,
this community in and for itself their goal – not the liberation of humanity [as] a pol-
itical goal – and [they are] to love one another for its sake’. To the Kierkegaardian,
this may precisely look like scaling back the demand on the individual on the one
 Conclusion 251

As well as this issue of complacency, moreover, there are other

grounds on which the Kierkegaardian can question the sort of social
command view with which Hegel replaces the Kantian hybrid theory.
For, as we have seen, Kierkegaard objects that there can be morally
objectionable results on a view of this sort; and worries of this sort
have been amplified recently by other proponents of a divine com-
mand approach, such as Robert Adams and Stephen Evans. Adams,
in particular, has argued that, on the one hand, societies have
demanded behaviour of their citizens which we would not intuitively
want to accept as morally obligatory (such as informing on fellow
citizens in Nazi Germany, or going in for various ritualistic rites of
passage), while on the other hand they have failed to demand behav-
iour that we nonetheless think is obligatory (such a treating others
with fairness, or respecting their property). If we adopt the social
command theory, therefore, the worry is that we will end up with
counter-intuitive consequences of this kind – where no such prob-
lems are said to afflict a divine command account, on which God
only commands what is morally good, and all such moral goods are
commanded by him.68
Now, Susan Wolf has responded to objections of this sort by
Adams in ways that the Hegelian could also follow in replying to
Kierkegaard. In response to the first worry, she has emphasised that
even if social command is a necessary condition for obligation, it
need not be a sufficient condition, where moral goodness may also
be said to be required, thereby ruling out the worry that a morally
bad action could be rendered into a duty. She admits, however, that
the second worry is harder to resolve, and in the end she bites the
bullet, allowing that on the social command theory, there may be
societies (however rare, and however distant from ours), in which no
such command exists and likewise no such obligation – though one

hand, while on the other relying on the community of which one is part to take on
what needs to be done, in a way that is over-optimistic. Cf. also LPRiii: 149 (p. 218),
where Hegel also understands the commandment in terms of ‘mutual love of the
community’, and ‘not an impotent love of humanity in general’. See also LPRiii: 284
(p. 367).
68
Cf. Adams 1999: 247–8, and Evans 2004: 1–3, 16–17, and 129–32. However, even on
the divine command account, there may still seem to be a related worry, which is that if
God did not exist, and so did not make these commands, certain actions which we take
to be morally required would no longer be obligatory. For some further discussion of
this complex issue, see Wainwright 2005: 110–15.
252 Understanding Mor al Obligation

can still say that what they did was morally bad, even if they had no
duty not to do it:
If acceptance of a Social Command Theory of moral obligation pre-
vents one from being entitled to say that the members of an extremely
historically distant society violated their obligations when they acted in
morally horrific ways, it does not prevent one from saying all sorts of
other things in criticism of that society’s practices. It does not prevent
one, for example, from being entitled to say that the practices were
morally abhorrent; that the victims of these practices were abominably
treated; or, that to live in such a society would have been in certain
respects very bad. Nor does it prevent one from being entitled to say
that there is decisive moral reason to resist such practices – at least for
us, but also possibly also for them.69

As Wolf’s reply grants, therefore, the social command theory does not
have a direct response to the Kierkegaardian worry, but rather makes
an attempt to defuse it. As a result, therefore, we can scarcely expect
the Kierkegaardian to accept this reply, where he will again feel justi-
fied, as Adams and Evans do, in turning instead to a divine command
approach, in an attempt to offer a conception of obligation that is less
historicist and more universal.

Conclusion
We have seen, then, how Kierkegaard’s critique of Hegel’s position
can be developed, and the difficulty the Hegelian will have in dispos-
ing of it entirely. However, we have also seen that the same is true of
Kant’s critique of the sort of divine command theory that Kierkegaard
puts in its place, as well as Hegel’s critique of Kant’s own alternative,
namely the hybrid theory. It therefore appears that we are faced with
a dialectical circle of positions, each of which has advantages over
the one it tries to supersede, but each of which has its drawbacks too.
We have found, then, that it is hard for the deadlock between these
three approaches to be broken. At the same time, however, we have
seen how these three approaches can be used to criticise other alter-
natives, such as natural law theories. Equally, as we have also seen,
there are grounds to reject the claim made by those who suggest that
the concept of moral obligation should be simply abandoned, and

69
Wolf 2009: 364.
 Conclusion 253

who therefore attempt to turn their back on the issue. Thus, it would
appear, as no solution can claim victory over its rivals and critics, the
significant challenge of properly theorising and accounting for moral
obligation remains as a problem that is still to be resolved – not only
by the tradition of thought that we have been considering, but by us
as well, as its inheritors.
 Bi bl io gr a ph y

Acton, H. B. (1970), Kant’s Moral Philosophy (London: Macmillan).

Adams, Robert Merrihew (1973), ‘A Modified Divine Command Theory of
Ethical Wrongness’, in Gene Outka and John P. Redder, Jr (eds.), Religion
and Morality (Garden City: Anchor), 318–47; reprinted in Adams 1987:
128–43.
(1979a), ‘Autonomy and Theological Ethics’, Religious Studies 15: 191–4;
reprinted in Adams 1987: 123–7.
(1979b), ‘Divine Command Metaethics Modified Again’, Journal of Religious
Studies 7: 66–79; reprinted in Adams 1987: 128–43.
(1987), The Virtue of Faith and Other Essays in Philosophical Theology (Oxford
University Press).
(1999), Finite and Infinite Goods: A Framework for Ethics (Oxford University
Press).
Allison, Henry E. (1990), Kant’s Theory of Freedom (Cambridge University Press).
(2001), ‘Ethics, Evil and Anthropology in Kant: Remarks on Allen Wood’s
Kant’s Ethical Theory’, Ethics 111: 594–613.
Alston, William P. (1989a), ‘Some Suggestions for Divine Command Theorists’,
in Michael Beaty (ed.), Christian Theism and the Problems of Philosophy
(University of Notre Dame Press), 303–26; reprinted in Alston 1989b:
253–73.
(1989b), Divine Nature and Human Language: Essays in Philosophical Theology
(Ithaca: Cornell University Press).
Ameriks, Karl (2003), ‘On Two Non-Realist Interpretations of Kant’s Ethics’, in
his Interpreting Kant’s ‘Critiques’ (Oxford University Press), 263–82.
Annas, Julia (1995), The Morality of Happiness, new edn (Oxford University
Press).
Anscombe, G. E. M. (1958), ‘Modern Moral Philosophy’, Philosophy 33: 1–19;
reprinted in Anscombe 1981: 26–42.
(1981), The Collected Philosophical Papers of G. E. M. Anscombe, volume iii: Ethics,
Religion and Politics (Oxford: Blackwell).
Aristotle (1984), The Complete Works of Aristotle: The Revised Oxford Translation,
edited by Jonathan Barnes, 2 vols. (Princeton University Press).
Baier, Kurt (1988), ‘Radical Virtue Ethics’, in Peter A. French, Theodore E.
Uehling, and Howard K. Wettstein (eds.), Ethical Theory: Character and

255
256 Bibliogr aphy

Virtue, Midwest Studies in Philosophy Volume xiii (Notre Dame University

Press), 126–35.
Barbeyrac, Jean (1735), ‘Judgement d’un anonyme sur l’original de cet abrégé’,
in Pufendorf 1735: ii, 379–435; translated as Barbeyrac 2003.
(2003), ‘The Judgment of an Anonymous Writer on the Original of this
Abridgement’, translated by David Saunders in Pufendorf 2003: 267–305.
Baumgarten, Alexander Gottlieb (1760), Initia philosophiae practicae primae
(Halle: Hemmerde); reprinted in Kant (1900–), vol. xix: 7–91.
Baxley, Anne Margaret (2003), ‘The Beautiful Soul and the Autocratic Agent:
Schiller’s and Kant’s “Children of the House”’, Journal of the History of
Philosophy 41: 493–514.
(2008), ‘Does Kantian Virtue Amount to More Than Continence?’, Review of
Metaphysics 56: 559–86.
(2010), Kant’s Theory of Virtue: The Value of Autocracy (Cambridge University
Press).
Beck, Lewis White (1960), A Commentary on Kant’s ‘Critique of Practical Reason’
(Chicago University Press).
Beiser, Frederick (1992), Enlightenment, Revolution and Romanticism: The Genesis
of Modern German Political Thought, 1790–1800 (Cambridge, MA: Harvard
University Press).
(2005), Schiller as Philosopher: A Re-examination (Oxford University Press).
Bradley, F. H. (1927), Ethical Studies, 2nd edn (Oxford University Press).
Brandt, R. B. (1964), ‘The Concepts of Moral Obligation and Duty’, Mind 73:
374–93.
Brink, David O. (1989), Moral Realism and the Foundations of Ethics (Cambridge
University Press).
Broadie, Sarah (1994), ‘Another Problem of Akrasia’, International Journal of
Philosophical Studies 2: 229–42.
Brochard, Victor (1901), ‘La morale ancienne et la morale moderne’, Revue
Philosophique 51: 1–12.
Buchwalter, Andrew (1992), ‘Hegel’s Concept of Virtue’, Political Theory 20:
548–83.
Cantrell, Michael A. (2009), Kierkegaard and Modern Moral Philosophy: Conceptual
Unintelligibility, Moral Obligations and Divine Commands, unpublished Ph.D.
dissertation, Baylor University.
Chalier, Catherine (2002), What Ought I to Do? Morality in Kant and Levinas,
translated by Jane Marie Todd (Ithaca: Cornell University Press).
Chytry, Josef (1989), The Aesthetic State: A Quest in Modern German Thought
(Berkeley: University of California Press).
Clarke, Samuel (1711), A Discourse Concerning the Unchangeable Obligations of Natural
Religion, and the Truth and Certainty of the Christian Revelation, 3rd edn, cor-
rected (London: James Knapton); partially reprinted in Schneewind (ed.)
2003: 295–311.
Cohen, G. A. (1996), ‘Reason, Humanity, and the Moral Law’, in Korsgaard
1996b: 167–88.
Collins, James D. (1983), ‘The Attack Upon Hegelianism’, in his The Mind of
Kierkegaard (Princeton University Press), 98–136.
 Bibliogr aphy 257

Connell, George B. (1992), ‘Judge William’s Theonomous Ethic’, in George

B. Connell and C. Stephen Evans (eds.), Foundations of Kierkegaard’s Vision
of Community: Religion, Ethics, and Politics in Kierkegaard (New Jersey:
Humanities Press), 56–70.
Crisp, Roger (2004), ‘Does Modern Moral Philosophy Rest on a Mistake?’, in
Anthony O’Hear (ed.), Modern Moral Philosophy (Cambridge University
Press), 75–94.
(2006), Reason and the Good (Oxford University Press).
Critchley, Simon (2007), Infinitely Demanding: Ethics of Commitment, Politics of
Resistance (London: Verso).
Crites, Stephen (1972), In the Twilight of Christendom: Hegel vs. Kierkegaard on Faith
and Religion (Chambersburg: American Academy of Religion).
Crosby, John F. (2004), ‘The Personal Encounter with God in Moral Obligation’,
in his Personalist Papers (Washington: Catholic University of America Press),
64–92.
Crusius, Christian August (1744), Anweisung vernünftig zu leben, 3rd edn (Leipzig:
Johann Friedrich Gleditsch; reprinted Hildesheim: Georg Olms, 1969);
partially translated in Crusius 2003.
(1766), Entwurf der nothwendigen Vernunftwahrheiten (Leipzig: Johann Friedrich
Gleditsch).
(2003), Guide to Rational Living, translated by J. B. Schneewind in Schneewind
(ed.) 2003: 569–84.
Cudworth, Ralph (1996), A Treatise Concerning Eternal and Immutable Morality
[1731], edited by Sarah Hutton (Cambridge University Press).
Cullity, Gareth and Berys Gaut (1997), ‘Introduction’, in Gareth Cullity and Berys
Gaut (eds.), Ethics and Practical Reason (Oxford University Press), 1–28.
Cumberland, Richard (2005), ‘A Treatise Concerning the Obligation,
Promulgation, and Observance of the Law of Nature’, printed as Appendix
ii to his A Treatise of the Laws of Nature [1727], translated by John Maxwell,
edited by Jon Parkin (Indianapolis: Liberty Fund).
Darwall, Stephen (1992), ‘Internalism and Agency’, Philosophical Perspectives 6:
155–74.
(1995), The British Moralists and the Internal “Ought” 1640–1740 (Cambridge
University Press).
(1997), ‘Reasons, Motives, and the Demands of Morality: An Introduction’, in
Stephen Darwall, Allan Gibbard, and Peter Railton (eds.), Moral Discourse
and Practice (Oxford University Press), 305–12.
(2004), ‘Autonomy in Modern Natural Law’, in Natalie Brender and Larry
Krasnoff (eds.), New Essays in the History of Autonomy (Cambridge University
Press), 110–29.
(2006), The Second-Person Standpoint: Morality, Respect, and Accountability.
(Cambridge, MA: Harvard University Press).
(2007), ‘Reply to Korsgaard, Wallace, and Watson’, Ethics 118: 52–69.
Dean, Richard (2006), The Value of Humanity in Kant’s Moral Theory (Oxford
University Press).
Deligiorgi, Katerina (2006), ‘Grace as Guide to Morals? Schiller’s Aesthetic
Turn in Ethics’, History of Philosophy Quarterly 23: 1–20.
258 Bibliogr aphy

Descartes, René (1984), The Philosophical Writings of Descartes, translated by John

Cottingham, Robert Stoothoff, and Dugald Murdoch, 2 vols. (Cambridge
University Press).
Dews, Peter (2008), The Idea of Evil (Oxford: Blackwell).
Dworkin, Gerald (1988), The Theory and Practice of Autonomy (Cambridge
University Press).
Engstrom, Stephen (1996), ‘Happiness and the Highest Good in Aristotle and
Kant’, in Stephen Engstrom and Jennifer Whiting (eds.), Aristotle, Kant,
and the Stoics: Rethinking Happiness and Duty (Cambridge University Press),
102–38.
Evans, C. Stephen (2004), Kierkegaard’s Ethic of Love: Divine Commands and Moral
Obligations (Oxford University Press).
(2006), ‘Introduction’ in Kierkegaard 2006: vi–xxx.
(2009), Kierkegaard: An Introduction (Cambridge University Press).
Fackenheim, Emil (1954), ‘Kant and Radical Evil’, University of Toronto Quarterly
23: 339–53.
(1969), ‘On the Actuality of the Rational and the Rationality of the Actual’,
Review of Metaphysics 89: 690–98; reprinted in Stewart (ed.) (1996): 42–52.
Falk, W. D. (1944), ‘Morals Without Faith’, Philosophy 19: 3–18; reprinted in Falk
1986: 163–79.
(1945), ‘Obligation and Rightness’, Philosophy 20: 129–47.
(1947–48), ‘“Ought” and Motivation’, Proceedings of the Aristotelian Society 48:
492–510; reprinted in Falk 1986: 21–41.
(1950), ‘Morality and Nature’, Australasian Journal of Philosophy 28: 69–92;
reprinted in Falk 1986: 180–97.
(1986), Ought, Reasons, and Morality: The Collected Papers of W. D. Falk (Ithaca
and London: Cornell University Press).
Feinberg, Joel (1973), ‘The Idea of a Free Man’, in James F. Doyle (ed.),
Educational Judgments (London: Routledge & Kegan Paul), 143–69.
Ferreira, M. Jamie (2001), Love’s Grateful Striving: A Commentary on Kierkegaard’s
‘Works of Love’ (Oxford University Press).
(2002), ‘The Glory of a Long Desire: Need and Commandment in Works
of Love’, in Inglof U. Dalferth (ed.), Ethik der Liebe: Studien zu Kierkegaards
‘Taten der Liebe’ (Tübingen: Mohr Siebeck), 139–53.
Finnis, John (1980), Natural Law and Natural Rights (Oxford University
Press).
(2002), ‘Natural Law: The Classical Tradition’, in Jules Coleman and Scott
Shapiro (eds.), The Oxford Handbook of Jurisprudence and Philosophy of Law
(Oxford University Press), 1–60.
Foot, Philippa (1978), Virtues and Vice and Other Essays in Moral Philosophy
(Oxford: Blackwell).
Formosa, Paul (forthcoming), ‘Is Kant a Moral Constructivist or a Moral
Realist?’, European Journal of Philosophy.
Galvin, Richard (forthcoming), ‘Rounding up the Usual Suspects: Varieties of
Kantian Constructivism in Ethics’, The Philosophical Quarterly.
Garner, Richard T. (1990), ‘On the Genuine Queerness of Moral Properties
and Facts’, Australasian Journal of Philosophy 68: 137–46.
 Bibliogr aphy 259

Gaut, Berys (1997), ‘The Structure of Practical Reason’, in Garrett Cullity and
Berys Gaut (eds.), Ethics and Practical Reason (Oxford University Press),
161–88.
Gauthier, Jeffrey A. (1997), ‘Schiller’s Critique of Kant’s Moral Psychology:
Reconciling Practical Reason and an Ethics of Virtue’, Canadian Journal of
Philosophy 27: 513–44.
Gibbard, Allan (1990), Wise Choices, Apt Feelings: A Theory of Normative Judgment
(Oxford University Press, 1990).
Goethe, J. W. von (1963), Goethe’s Faust [1828–9], translated by Walter Kaufmann
(New York: Anchor Books).
Goodman, Lenn E. (2008), Love Thy Neighbor as Thyself (Oxford University
Press).
Gordon, Peter E. (2005), ‘Self-Authorizing Modernity: Problems of Inter­
pretation in the History of German Idealism’, History and Theory 44:
121–37.
Green, Ronald M. (1992), Kierkegaard and Kant: The Hidden Debt (Albany: SUNY
Press).
(1993), ‘Enough is Enough! Fear and Trembling is Not About Ethics!’, Journal of
Religious Studies 21: 191–209.
Grenberg, Jeanine (2005), Kant and the Ethics of Humility: A Story of Dependence,
Corruption, and Virtue (Cambridge University Press).
Griffin, James (1992), ‘The Human Good and the Ambitions of Consequen­
tialism’, in Ellen Frankel Paul, Fred D. Miller, and Jeffrey Paul (eds.), The
Good Life and the Human Good (Cambridge University Press), 118–32.
(1996), Value Judgement: Improving Our Ethical Beliefs (Oxford University
Press).
Guyer, Paul (1996), Kant and the Experience of Freedom (Cambridge University
Press).
(2000), Kant on Freedom, Law, and Happiness (Cambridge University Press).
(2007), Kant’s ‘Groundwork for the Metaphysics of Morals’ (London: Continuum).
Haakonssen, Knud (1996), Natural Law and Moral Philosophy: From Grotius to the
Scottish Enlightenment (Cambridge University Press).
Haldane, John (1991), ‘Medieval and Renaissance Ethics’, in Peter Singer (ed.),
A Companion to Ethics (Oxford: Blackwell), 133–46.
Hannay, Alasdair (1991), Kierkegaard, revised edn (London: Routledge).
Hardimon, Michael O. (1994), ‘Role Obligations’, The Journal of Philosophy 91:
333–63.
Hare, John E. (2000a), ‘Kant’s Divine Command Theory and its Reception
within Analytic Philosophy’, in D. Z. Phillips and Timothy Tessin (eds.),
Kant and Kierkegaard on Religion (New York: Palgrave Macmillan),
263–77.
(2000b), ‘Kant on Recognizing our Duties as God’s Commands’, Faith and
Philosophy 17: 459–78.
(2001), God’s Call: Moral Realism, God’s Commands, and Human Autonomy
(Grand Rapids: William B. Eerdmans).
(2002), The Moral Gap: Kantian Ethics, Human Limits, and God’s Assistance,
paperback edn (Oxford University Press).
260 Bibliogr aphy

(2009), God and Morality: A Philosophical History (Oxford: Wiley-Blackwell).

Hart, H. L. A. (1994), The Concept of Law, 2nd edn (Oxford University Press).
Hegel, G. W. F. (1892–96), Lectures on the History of Philosophy, translated by
E. S. Haldane and Frances H. Simson, 3 vols. (London: K. Paul, Trench,
Trübner); reprint edn (Lincoln: University of Nebraska Press, 1995).
(1948), Early Theological Writings, translated by T. M. Knox (Philadelphia:
University of Pennsylvania Press).
(1969), Science of Logic [1st edn 1812, 3rd edn 1816], translated by A. V. Miller
(London: George Allen & Unwin).
(1969–71), Theorie Werkausgabe: Werke in zwanzig Bänden, edited by Eva
Moldenhauer and Karl Markus Michel (Frankfurt: Suhrkamp).
(1970), Philosophy of Nature: Part ii of the Encyclopaedia of the Philosophical Sciences
[1st edn 1817, 3rd edn 1830], translated by M. J. Petry, 3 vols. (London:
George Allen & Unwin).
(1975), Hegel’s Aesthetics: Lectures on Fine Art, translated by T. M. Knox, 2 vols.
(Oxford University Press).
(1977a), The Difference Between Fichte’s and Schelling’s System of Philosophy [1801],
translated by H. S. Harris and Walter Cerf (Albany: SUNY Press).
(1977b), Phenomenology of Spirit [1807], translated by A. V. Miller (Oxford
University Press).
(1984), Vorlesungen über die Philosophie der Religion, Teil 3, Die vollendete Religion,
vol. v of G. W. F. Hegel: Vorlesungen: Ausgewählte Nachschriften und Manuskripte,
edited by Walter Jaeschke (Hamburg: Felix Meiner).
(1986), Vorlesungen über die Geschichte der Philosophie, Teil 4, Philosophie der mit-
telalters und der neueren Zeit, vol. ix of G. W. F. Hegel: Vorlesungen: Ausgewählte
Nachschriften und Manuskripte, edited by Pierre Garniron and Walter
Jaeschke (Hamburg: Felix Meiner).
(1989), Vorlesungen über die Geschichte der Philosophie, Teil 3, Griechische Philosophie
ii. Plato bis Proklos, vol. viii of G. W. F. Hegel: Vorlesungen: Ausgewählte
Nachschriften und Manuskripte, edited by Pierre Garniron and Walter
Jaeschke (Hamburg: Felix Meiner).
(1991a), Elements of the Philosophy of Right [1821], edited by Allen W. Wood and
translated by H. B. Nisbet (Cambridge University Press).
(1991b), Logic: Part I of the Encyclopaedia of the Philosophical Sciences [1st edn
1817, 3rd edn 1830], translated by T. F. Geraets, W. A. Suchting, and H. S.
Harris (Indianapolis: Hackett).
(2006), Lectures on the History of Philosophy 1825–6, volume ii: Greek Philosophy,
edited by Robert F. Brown, translated by R. F. Brown and J. M. Stewart
(Oxford University Press).
(2007), Lectures on the Philosophy of Religion, volume iii: The Consummate Religion,
edited by Peter C. Hodgson, translated by R. F. Brown, P. C. Hodgson, and
J. M. Stewart (Oxford University Press).
(2009), Lectures on the History of Philosophy 1825–6, volume iii: Medieval and
Modern Philosophy, edited by Robert F. Brown, translated by R. F. Brown
and J. M. Stewart (Oxford University Press).
Henrich, Dieter (1965), ‘Das Prinzip der kantischen Ethik’, Philosophische
Rundschau 2: 29–34.
 Bibliogr aphy 261

(1982), ‘Beauty and Freedom: Schiller’s Struggle with Kant’s Aesthetics’, in

Ted Cohen and Paul Guyer (eds.), Essays in Kant’s Aesthetics (University of
Chicago Press, 1982), 237–60.
Herman, Barbara (1993), The Practice of Moral Judgment (Cambridge, MA:
Harvard University Press).
Hills, Alison (2008), ‘Kantian Value Realism’, Ratio 21: 182–200.
Hobbes, Thomas (1998), On the Citizen [1651], edited and translated by Richard
Tuck and Michael Silverthorne (Cambridge University Press).
Horton, John (1992), Political Obligation (London: Methuen).
Hunter, Ian (2003), ‘Arguments Over Obligation: Teaching Time and Place in
Moral Philosophy’, in J. B. Schneewind (ed.), Teaching the New Histories of
Philosophy (Princeton University Center for Human Values), 131–68.
Hursthouse, Rosalind (1999), On Virtue Ethics (Oxford University Press).
Hutcheson, Francis (2002), An Essay on the Nature and Conduct of the Passions and
Affections, with Illustrations on the Moral Sense [1st edn 1728, 3rd edn 1742],
edited by Aaron Garrett (Indianapolis: Liberty Fund).
Idziak, Janine Marie (1979), ‘Divine Command Morality: A Guide to the
Literature’, in Janine Marie Idziak (ed.), Divine Command Morality: Historical
and Contemporary Readings (New York and Toronto: Edwin Mellen Press),
1–38.
Irwin, Terence (1996), ‘Kant’s Criticisms of Eudaemonism’, in Stephen
Engstrom and Jennifer Whiting (eds.), Aristotle, Kant, and the Stoics:
Rethinking Happiness and Duty (Cambridge University Press, 1996),
63–101.
(2004), ‘Kantian Autonomy’, in John Hyman and Helen Steward (eds.), Agency
and Action (Cambridge University Press), 137–64.
(2006), ‘Aquinas, Natural Law, and Aristotelian Eudaimonism’, in Richard
Kraut (ed.), The Blackwell Guide to Aristotle’s ‘Nicomachean Ethics’ (Oxford:
Blackwell), 323–35.
(2007), The Development of Ethics: A Historical and Critical Study, volume i: From
Socrates to the Reformation (Oxford University Press).
(2008), The Development of Ethics: A Historical and Critical Study, volume ii: From
Suarez to Rousseau (Oxford University Press).
(2009), The Development of Ethics, A Historical and Critical Study, volume iii: From
Kant to Rawls (Oxford University Press).
Ives, Margaret C. (1970), The Analogue of Harmony: Some Reflections on Schiller’s
Philosophical Essays (Pittsburgh: Duquesne University Press).
Jackson, M. W. (1987), ‘Hegel, the Real and the Rational’, International Studies in
Philosophy 19: 11–19; reprinted in Stewart (ed.) (1996): 19–25.
James, David (2007), ‘The Absolute Paradox: Kierkegaard’s Argument Against
Hegel’s Account of the Relation of Faith to Philosophy’, Kierkegaardiana 24:
102–20.
Johnson, Robert N. (2007), ‘Value and Autonomy in Kantian Ethics’, in Russ
Shafer-Landau (ed.), Oxford Studies in Metaethics, volume ii (Oxford
University Press), 133–48.
Kain, Patrick (1999), Self-Legislation and Prudence in Kant’s Moral Philosophy, Ph.D.
dissertation, Notre Dame University.
262 Bibliogr aphy

(2004), ‘Self-Legislation in Kant’s Moral Philosophy’, Archiv für Geschichte der

Philosophie 86: 257–306.
(2005), ‘Interpreting Kant’s Theory of Divine Commands’, Kantian Review 9:
128–49.
(2006), ‘Realism and Anti-Realism in Kant’s Second Critique’, Philosophy
Compass 1: 449–65.
Kant, Immanuel (1900–), Kants gesammelte Schriften, edited by the Royal Prussian
(later German, then Berlin-Brandenberg) Academy of Sciences (Berlin:
Georg Reimer, later Walter de Gruyter).
(1956), Critique of Practical Reason [1788], translated by Lewis White Beck
(Indianapolis: Bobbs-Merrill).
(1996a), Practical Philosophy, translated and edited by Mary J. Gregor
(Cambridge University Press).
(1996b), Religion and Rational Theology, translated and edited by Allen W.
Wood and George di Giovanni (Cambridge University Press).
(1997), Lectures on Ethics, edited by Peter Heath and J. B. Schneewind, trans-
lated by Peter Heath (Cambridge University Press).
(1998a), Critique of Pure Reason [1st edn 1781, 2nd edn 1787], translated and
edited by Paul Guyer and Allen W. Wood (Cambridge University Press).
(1998b), Groundwork of the Metaphysics of Morals [1785], translated and edited
by Mary J. Gregor (Cambridge University Press).
(2000), Critique of the Power of Judgment [1790], edited by Paul Guyer, trans-
lated by Paul Guyer and Eric Matthews (Cambridge University Press).
(2002a), Theoretical Philosophy 1755–1770, translated and edited by David
Walford, in collaboration with Ralf Meerbote (Cambridge University
Press).
(2002b), Theoretical Philosophy after 1781, edited by Henry Allison and Peter
Heath, translated by Gary Hatfield, Michael Friedman, Henry Allison, and
Peter Heath (Cambridge University Press).
(2004), Vorlesung zur Moralphilosophie, edited by Werner Stark (Berlin: Walter
de Gruyter).
(2005), Notes and Fragments, edited by Paul Guyer, translated by Paul Guyer,
Curtis Bowman, and Frederick Rauscher (Cambridge University Press).
Kerstein, Samuel J. (2002), Kant’s Search for the Supreme Principle of Morality
(Cambridge University Press).
Kierkegaard, Søren (1901–06), Søren Kierkegaard’s Samlede Vœrker, edited by
A. B. Drachman, J. L. Heiberg, and H. O. Lange, 14 vols. (Copenhagen:
Gyldendal).
(1909–78), Søren Kierkegaard’s Papirer, edited by P. A. Heiberg, V. Kuhr,
and E. Torsting, supplementary vols. edited by Niels Thulstrup, 22 vols.
(Copenhagen: Gyldendal).
(1987), Either/Or [1843], Part ii, edited and translated by Howard V. Hong
and Edna H. Hong (Princeton University Press).
(1988), Stages on Life’s Way [1845], edited and translated by Howard V. Hong
and Edna V. Hong (Princeton University Press).
(1991), Practice in Christianity [1850], edited and translated by Howard V.
Hong and Edna H. Hong (Princeton University Press).
 Bibliogr aphy 263

(1992), Concluding Unscientific Postscript to ‘Philosophical Fragments’: volume 1:

The text [1846], edited and translated by Howard V. Hong and Edna H.
Hong (Princeton University Press).
(1995), Works of Love [1847], edited and translated by Howard V. Hong and
Edna H. Hong (Princeton University Press).
(1999), Søren Kierkegaard’s Journals and Papers, 2nd edn, edited and translated
by Howard V. Hong and Edna H. Hong, assisted by Gregor Malantschuk, 7
vols. (Bloomington: Indiana University Press).
(2006), Fear and Trembling: A Dialectical Lyric [1843], translated by Sylvia Walsh,
edited by C. Stephen Evans and Sylvia Walsh (Cambridge University Press).
Kirmmse, Bruce H. (1990), Kierkegaard in Golden Age Denmark (Bloomington:
Indiana University Press).
Knowles, Dudley (2002), Hegel and the ‘Philosophy of Right’ (London: Routledge).
Korkman, Petter (2001), Barbeyrac and Natural Law, Ph.D. dissertation (Helsinki:
Åkademi University).
(2003), ‘Voluntarism and Moral Obligation: Barbeyrac’s Defence of Pufendorf
Revisited’, in T. J. Hochstrasser and P. Schröder (eds.), Early Modern Natural
Law Theories: Context and Strategies in the Early Enlightenment (Dordrecht:
Kluwer), 195–225.
Korsgaard, Christine M. (1983), ‘Two Distinctions in Goodness’, The Philosophical
Review 92: 169–95; reprinted in Korsgaard 1996: 249–74.
(1989), ‘Kant’, in Robert Cavalier, James Gouinlock, and James Sterba (eds.),
Ethics in the History of Western Philosophy (New York: St Martin’s Press),
pp. 201–43; reprinted as ‘An Introduction to the Ethical, Political, and
Religious Thought of Kant’ in Korsgaard 1996a: 3–42.
(1996a), Creating the Kingdom of Ends (Cambridge University Press).
(1996b), ‘From Duty and for the Sake of the Noble: Kant and Aristotle on
Morally Good Actions’, in Stephen Engstrom and Jennifer Whiting (eds.),
Aristotle, Kant, and the Stoics: Rethinking Happiness and Duty (Cambridge
University Press), 203–36; reprinted in Korsgaard 2008: 174–206.
(1996c), The Sources of Normativity, with G. A. Cohen, Raymond Geuss, Thomas
Nagel, and Bernard Williams, edited by Onora O’Neill (Cambridge
University Press).
(1997), ‘The Normativity of Instrumental Reason’, in Garrett Cullity and
Berys Gaut (eds.), Ethics and Practical Reason (Oxford University Press),
213–54; reprinted in Korsgaard 2008: 27–68.
(1998), ‘Introduction’, in Kant 1998b: vii–xxx.
(2007), ‘Autonomy and the Second Person Within: A Commentary on Stephen
Darwall’s The Second Person Standpoint’, Ethics 118: 8–23.
(2008), The Constitution of Agency: Essays on Practical Reason and Moral Psychology
(Oxford University Press).
(2009), Self-Constitution: Agency, Identity, and Integrity (Oxford University Press).
Krasnoff, Larry (2008), Hegel’s ‘Phenomenology of Spirit’: An Introduction (Cambridge
University Press).
Lafont, Cristina (2004), ‘Moral Objectivity and Reasonable Agreement: Can
Realism Be Reconciled with Kantian Constructivism?’, Ratio Juris 17:
27–51.
264 Bibliogr aphy

Langton, Rae (2007), ‘Objective and Unconditioned Value’, Philosophical Review

116: 157–85.
Larmore, Charles (2008), The Autonomy of Morality (Cambridge University
Press).
Law, David R. (1995), ‘The Place, Role, and Function of the “Ultimatum” of
Either/Or, Part Two, in Kierkegaard’s Pseudonymous Authorship’, in Robert
L. Perkins (ed.), International Kierkegaard Commentary: Either/Or (Macon:
Mercer University Press), 233–57.
LeBar, Mark (2008), ‘Aristotelian Constructivism’, Social Philosophy and Policy
25: 182–213.
Lee, Seung-Goo (1992), ‘The Antithesis between the Religious View of Ethics
and the Rationalistic View of Ethics in Fear and Trembling’, in Robert L.
Perkins (ed.), International Kierkegaard Commentary: ‘Fear and Trembling’ and
‘Repetition’ (Macon: Mercer University Press), 101–26.
Leibniz, Gottfried Wilhelm (c 1702), ‘Méditation sur la notion commune de
la justice’, in G. W. Leibniz, Rechtsphilosophisches aus Leibnizens ungedruckten
Schriften, edited by Georg Mollat (Leipzig: Robolsky, 1885), 56–82; trans-
lated as Leibniz 1988a.
(1706), ‘Monita quaedam ad Samuelis Pufendorfii principia, Gerh. Wolth.
Molano directa’, in G. W. Leibniz, Opera omnia, edited by Ludwig Dutens,
vol. iv, 3 (Geneva: De Tournes, 1768; reprinted Hildesheim: Olms, 1989),
275–83; translated as Leibniz 1988b.
(1710), Essais de Théodicée, in G. W. Leibniz, Die philosophischen Schriften, edited
by C. J. Gerhardt, 7 vols. (Berlin: Weidmannsche Buchhandlung, 1875–90;
reprinted Hildesheim: Georg Olms, 1978), vi: 1–487; translated as Leibniz
1998.
(1988a), ‘Meditation on the Common Concept of Justice’ in his Political
Writings, 2nd edn, edited and translated by Patrick Riley (Cambridge
University Press), 45–64.
(1988b), ‘Opinion on the Principles of Pufendorf’, in his Political Writings,
2nd edn, edited and translated by Patrick Riley (Cambridge University
Press), 64–75.
(1998), Theodicy: Essays on the Goodness of God, edited by Austin Marsden Farrer,
translated by E. M. Huggard (La Salle: Open Court).
Lillegard, Norman (1995), ‘Judge William in the Dock: MacIntyre on
Kierkegaardian Ethics’, in Robert L. Perkins (ed.), International Kierkegaard
Commentary: Either/Or (Macon: Mercer University Press, 1995), 83–112.
Løgstrup, Knud Ejler (1997), The Ethical Demand, translated by Theodore I.
Jensen and Gary Puckering, revised by Hans Fink and Alasdair MacIntyre
(University of Notre Dame Press).
Louden, Robert B. (1992), Morality and Moral Theory: A Reappraisal and
Reaffirmation (Oxford University Press).
McDowell, John (1978), ‘Are Moral Requirements Hypothetical Imperatives?’,
Proceedings of the Aristotelian Society, supplementary volume 52: 13–29;
reprinted in McDowell 1998: 77–94.
(1979), ‘Virtue and Reason’, The Monist 62: 331–50; reprinted in McDowell
1998: 50–76.
 Bibliogr aphy 265

(1980), ‘The Role of Eudaimonia in Aristotle’s Ethics’, Proceedings of the African

Classical Associations 15: 1–14; reprinted in McDowell 1998: 3–22.
(1998), Mind, Value, and Reality (Cambridge, MA: Harvard University Press).
MacIntyre, Alasdair (1967), A Short History of Ethics (London: Routledge &
Kegan Paul).
(1985), After Virtue, 2nd edn (London: Duckworth).
Mackie, J. L. (1977), Ethics: Inventing Right and Wrong (Harmondsworth:
Penguin).
Manis, R. Zachary (2009), ‘Kierkegaard and Divine-Command Theory: Replies
to Quinn and Evans’, Religious Studies 45: 289–307.
Mill, J. S. (1969), Utilitarianism, in Collected Works of John Stuart Mill, vol. x, edited
by J. M. Robson (University of Toronto Press), 203–60.
Mouw, Richard J. (1990), The God Who Commands (University of Notre Dame
Press).
Murdoch, Iris (1970), The Sovereignty of Good (London: Routledge).
Neiman, Susan (1994), The Unity of Reason (Oxford University Press).
Nicholson, Peter P. (1990), The Political Philosophy of the British Idealists: Selected
Studies (Cambridge University Press).
Nietzsche, Friedrich (1969), On the Genealogy of Morals [1887], translated by
Walter Kaufmann and R. J. Hollingdale, in ‘On the Genealogy of Morals’ and
‘Ecce Homo’ (New York: Vintage Books).
(1980), Sämtliche Werke: Kritische Studienausgabe in 15 Bänden, edited by G. Colli
and M. Montinari (Berlin: De Gruyter).
Norton, Robert E. (1995), The Beautiful Soul: Aesthetic Morality in the Eighteenth
Century (Ithaca and London: Cornell University Press).
Nowell-Smith, P. H. (1961), ‘Morality: Religious and Secular’, Rationalist Annual:
5–22.
Nussbaum, Martha (2001), Upheavals of Thought: The Intelligence of Emotions
(Cambridge University Press).
Olafson, Frederick A. (1967), Principles and Persons: An Ethical Interpretation of
Existentialism (Baltimore: John Hopkins Press).
O’Neill, Onora (2000), ‘Agency and Autonomy’, in her Bounds of Justice
(Cambridge University Press), 29–49.
Outka, Gene (1973), ‘Religious and Moral Duty: Notes on Fear and Trembling’, in
Gene Outka and John P. Reeder, Jr. (eds.), Religion and Morality: A Collection
of Essays (Garden City: Anchor Books), 204–54.
(1993), ‘God as the Subject of Unique Veneration: A Response to Ronald M.
Green’, Journal of Religious Ethics 21: 211–16.
Parkin, Jon (1999), Science, Religion and Politics in Restoration England: Richard
Cumberland’s ‘De Legibus Naturae’ (Woodbridge: Boydell Press).
Paton, H. J. (1967), The Categorical Imperative, 6th edn (London: Hutchinson).
Pattison, George (2005), The Philosophy of Kierkegaard (Chesham: Acumen).
Perkins, Robert L. (1967), ‘The Family: Hegel and Kierkegaard’s Judge William’,
Hegel Jahrbuch: 89–100.
(1995), ‘Either/Or/Or: Giving the Parson His Due’, in Robert L. Perkins (ed.),
International Kierkegaard Commentary: Either/Or (Macon: Mercer University
Press), 207–31.
266 Bibliogr aphy

Pidgen, Charles (1988), ‘Anscombe on “Ought”’, Philosophical Quarterly 38:

20–41.
Pink, Thomas (2004a), ‘Action, Will and Law in Late Scholasticism’, in Jill Kraye
and Risto Saarinen (eds.), Moral Philosophy on the Threshold of Modernity
(Dordrecht: Springer), 31–50.
(2004b), ‘Moral Obligation’, in Anthony O’Hear (ed.), Modern Moral Philosophy
(Cambridge University Press), 159–85.
Pinkard, Terry (2000), Hegel: A Biography (Cambridge University Press).
(2002), German Philosophy 1760–1860: The Legacy of Idealism (Cambridge
University Press).
(2003), ‘Subjects, Objects, and Normativity: What is it like to be an Agent?’,
International Yearbook of German Idealism 1: 201–18.
(2007), ‘Was Pragmatism the Successor to Idealism?’, in Cheryl Misak (ed.),
New Pragmatists (Oxford University Press), 142–68.
Pippin, Robert B. (1991), Modernism as a Philosophical Problem: On the Dissatisfactions
of European High Culture (Oxford: Blackwell).
(2000a), ‘Hegel’s Practical Philosophy: The Realization of Freedom’, in Karl
Ameriks (ed.), The Cambridge Companion to German Idealism (Cambridge
University Press), 180–99.
(2000b), ‘Kant’s Theory of Value’, Inquiry 43: 239–66.
(2008), Hegel’s Practical Philosophy: Rational Agency as Ethical Life (Cambridge
University Press).
Pistorius, Hermann Andreas (1975), ‘Rezension der Grundlegung zur Metaphysik
der Sitten’ [1786], reprinted in Rüdiger Bittner and Konrad Cramer (eds.),
Materialen zu Kants ‘Kritik der praktischen Vernunft’ (Frankfurt: Suhrkamp),
144–60.
Prauss, Gerold (1983), Kant über Freiheit als Autonomie (Frankfurt:
Klostermann).
Price, Richard (1948), A Review of the Principal Questions in Morals [1st edn 1758,
3rd edn 1787], edited by D. D. Raphael (Oxford University Press).
Prichard, H. A. (2002), ‘Kant’s Fundamental Principles of the Metaphysic of Morals’,
in Jim MacAdam (ed.), Moral Writings (Oxford University Press), 50–76.
Pufendorf, Samuel (1672), Elementorum jurisprudentiae universalis libri duo
(Cambridge: John Hayes; reprinted Oxford University Press, 1931); trans-
lated as Pufendorf 1931.
(1682), De officio hominis et civis juxta legem naturalem libri duo (Cambridge: John
Hayes; reprinted Oxford University Press, 1927); translated as Pufendorf
1991.
(1688), De jure naturae et gentium (Amsterdam: Andreas van Hoogenhuysen;
reprinted Oxford University Press, 1934); translated as Pufendorf 1934.
(1735), Les Devoirs de l’Homme et du Citoien, translated by Jean Barbeyrac,
2 vols. (Amsterdam: De Coup and Kuyper).
(1931), The Two Books of the Elements of Universal Jurisprudence, translated by
William Abbott Oldfather (Oxford University Press).
(1934), Of the Law of Nature and Nations, translated by C. H. Oldfather and
W. A. Oldfather (Oxford University Press).
 Bibliogr aphy 267

(1991), On the Duty of Man and Citizen, edited by James Tully and translated by
Michael Silverthorne (Cambridge University Press).
(2003), The Whole Duty of Man, According to the Law of Nature, translated by
Andrew Tooke, edited by Ian Hunter and David Saunders (Indianapolis:
Liberty Fund).
Pugh, David (1996), The Dialectic of Love: Platonism in Schiller’s Aesthetics (Montreal
and Kingston: McGill-Queen’s University Press).
Quinn, Philip L. (1978), Divine Commands and Moral Requirements (Oxford
University Press).
(1996), ‘The Divine Command Ethics in Kierkegaard’s Works of Love’, in Jeff
Jordan and Daniel Howard-Snyder (eds.), Faith, Freedom and Rationality
(Lanham: Rowman and Littlefield), 29–46.
(1998), ‘Kierkegaard’s Christian Ethics’, in Alastair Hannay and Gordon
D. Marino (eds.), The Cambridge Companion to Kierkegaard (Cambridge
University Press), 349–75.
(2002), ‘Obligation, Divine Commands and Abraham’s Dilemma’, Philosophy
and Phenomenological Research 64: 459–66.
Rauscher, Frederick (2002), ‘Kant’s Moral Anti-Realism’, Journal of the History of
Philosophy 40: 477–99.
Rawls, John (1971), A Theory of Justice (Cambridge, MA: Harvard University
Press).
(1980), ‘Kantian Constructivism in Moral Theory’, The Journal of Philosophy
77: 515–72; reprinted in Rawls 1999: 303–58.
(1989), ‘Themes in Kant’s Moral Philosophy’, in Eckart Förster (ed.), Kant’s
Transcendental Deductions (Stanford University Press, 1989), pp. 81–113;
reprinted in Rawls 1999: 497–528.
(1993), Political Liberalism (New York: Columbia University Press).
(1999), Collected Papers, edited by Samuel Freeman (Cambridge, MA: Harvard
University Press).
(2000), Lectures on the History of Moral Philosophy, edited by Barbara Herman
(Cambridge, MA: Harvard University Press).
Reath, Andrews (1994), ‘Legislating the Moral Law’, Nous 28: 436–64; reprinted
in Reath 2006: 92–120.
(2003), ‘Value and Law in Kant’s Moral Theory’, Ethics 114: 127–55.
(2006), Agency and Autonomy in Kant’s Moral Theory: Selected Essays (Oxford
University Press).
Regan, D. H. (2002), ‘The Value of Rational Nature’, Ethics 112: 267–91.
Reiner, Hans (1951), Pflicht und Neigung (Meisenheim am Glan: Hain); trans-
lated as Reiner 1983.
(1983), Duty and Inclination, translated by Mark Santos (The Hague:
Nijhoff).
Roehr, Sabine (2003), ‘Freedom and Autonomy in Schiller’, Journal of the History
of Ideas 64: 119–34.
Rorty, Richard (2004), ‘Trapped Between Kant and Dewey: The Current
Situation of Moral Philosophy’, in Natalie Brender and Larry Krasnoff
(eds.), New Essays on the History of Autonomy (Cambridge University Press),
268 Bibliogr aphy

195–214; reprinted as ‘Kant vs. Dewy: The Current Situation of Moral

Philosophy’ in Rorty 2007: 183–202.
(2007), Philosophy as Cultural Politics: Philosophical Papers, Volume 4 (Cambridge
University Press).
Rosler, Andrés (2005), Political Authority and Obligation in Aristotle (Oxford
University Press).
Rousseau, Jean-Jacques (1964), Contrat social [1762], edited by Robert Derathé
et al., in his Oeuvres completes, edited by Bernard Gagnebin, Marcel Raymon
et al., 5 vols. (Paris: Gallimard, 1959–95), volume iii, 347–470; translated
in Rousseau 1997: 39–152.
(1997), ‘The Social Contract’ and Other Later Political Writings, edited and trans-
lated by Victor Gourevitch (Cambridge University Press).
Rudd, Anthony (1993), Kierkegaard and the Limits of the Ethical (Oxford University
Press).
Schiller, Friedrich (1904), Philosophische Schriften, Schillers Sämtliche Werke:
Säkular-Ausgabe, vol. xxii, edited by Oskar Walzel (Stuttgart: J. G. Cotta).
(1943), Schillers Werke: Nationalausgabe, edited by Julius Petersen et al. (Weimar:
Böhlaus).
(1967), On the Aesthetic Education of Man in a Series of Letters [1794], translated by
Elizabeth M. Wilkinson and L. A. Willoughby (Oxford University Press).
(2003), ‘Kallias or Concerning Beauty: Letters to Gottfried Körner’ [1793],
translated by Stefan Bird-Pollan, in J. M. Bernstein (ed.), Classical and
Romantic German Aesthetics (Cambridge University Press), 145–84.
(2005), On Grace and Dignity [1793], translated Jane V. Curran, in Jane V.
Curran and Christophe Fricker (eds.), Schiller’s ‘On Grace and Dignity’ in its
Cultural Context: Essays and a New Translation (Rochester: Camden House),
123–70.
(2006), Aesthetical and Philosophical Essays, translated anonymously (Middlesex:
Echo Library).
Schmucker, Josef (1961), Die Ursprünge der Ethik Kants in seinen vorkritischen
Schriften (Meisenheim: Hain).
Schneewind, J. B. (1986), ‘The Use of Autonomy in Ethical Theory’, in Thomas
C. Heller, Morton Sosna, and David E. Wellbery (eds.), Reconstructing
Individualism: Autonomy, Individuality, and the Self in Western Thought
(Stanford University Press), 64–75.
(1990), ‘The Misfortunes of Virtue’, Ethics 101: 42–63; reprinted in Schneewind
2010: 176–201.
(1992), ‘Autonomy, Obligation, and Virtue: An Overview of Kant’s Moral
Philosophy’, in Paul Guyer (ed.), The Cambridge Companion to Kant
(Cambridge University Press), 309–41; reprinted in Schneewind 2010:
248–76.
(1996a), ‘Kant and Stoic Ethics’, in Stephen Engstrom and Jennifer Whiting
(eds.), Aristotle, Kant, and the Stoics: Rethinking Happiness and Duty (Cambridge
University Press), 285–302; reprinted in Schneewind 2010: 277–95.
(1996b), ‘Voluntarism and the Foundations of Ethics’, Proceedings and Addresses
of the American Philosophical Association 70: 25–41; reprinted in Schneewind
2010: 202–21.
 Bibliogr aphy 269

(1998), The Invention of Autonomy: A History of Modern Moral Philosophy

(Cambridge University Press).
(2000), ‘Hume and the Religious Significance of Moral Rationalism’, Hume
Studies, 26: 211–23; reprinted in Schneewind 2010: 222–36.
(2002), ‘Why Study Kant’s Ethics?’, in Allen W. Wood (ed. and trans.), Immanuel
Kant, Groundwork for the Metaphysics of Morals (New Haven and London: Yale
University Press), 83–91; reprinted as ‘Why Study Kant’s Groundwork?’ in
Schneewind 2010: 241–7.
(ed.) (2003), Moral Philosophy from Montaigne to Kant (Cambridge University
Press).
(2010), Essays on the History of Moral Philosophy (Oxford University Press).
Schönecker, Dieter (1999), Kant: Grundlegung iii. Die Deduktion des kategorischen
Imperativs (Freiburg: Alber).
Schopenhauer, Arthur (1962), ‘Über die Grundlage der Moral [1841]’, in
Wolfgang Löhneysen (ed.), Sämtliche Werke (Stuttgart: Cotta/Insel), iii:
629–815; translated as Schopenhauer 1965.
(1965), On the Basis of Morality, translated by E. F. J. Payne (Indianapolis:
Bobbs-Merrill).
Schwaiger, Clemens (1999), Kategorische und andere Imperative: Zur Entwicklung
von Kants praktischer Philosophie bis 1785 (Stuttgart: Frommann-
Holzboog).
(2009), ‘The Theory of Obligation in Wolff, Baumgarten, and the Early Kant’,
translated by Nicholas Walker, in Karl Ameriks and Otfried Höffe (eds.),
Kant’s Moral and Legal Philosophy (Cambridge University Press), 58–76.
Sciaraffa, Stefan (2011), ‘Identification, Meaning, and the Normativity of Social
Roles’, European Journal of Philosophy 19: 107–28.
Selden, John (1892), The Table Talk of John Selden [1689], edited by Samuel Harvey
Reynolds (Oxford: Clarendon Press).
Sensen, Oliver (2009), ‘Dignity and the Formula of Humanity’, in Jens
Timmermann (ed.), Kant’s ‘Groundwork of the Metaphysics of Morals’: A Critical
Guide (Cambridge University Press), 102–18.
Sensen, Oliver(2011), ‘Kant’s Conception of Inner Value’, European Journal of
Philosophy 19: 262–80.
Sharpe, Lesley (1995), Schiller’s Aesthetic Essays: Two Centuries of Criticism
(Columbia: Camden House).
Sherman, Nancy (1997), Making a Necessity of Virtue: Aristotle and Kant on Virtue
(Cambridge University Press).
Sidgwick, Henry (1981), The Methods of Ethics, 7th edn, republished (Indianapolis:
Hackett).
Silber, John R. (1959), ‘The Copernican Revolution in Ethics: The Good
Reexamined’, Kant-Studien 51: 85–101; reprinted in Robert Paul Wolff
(ed.), Kant: A Collection of Critical Essays (London: Macmillan, 1968),
266–90.
Simmons, A. John (1996a), ‘Associative Political Obligations’, Ethics 106: 247–73;
reprinted in Simmons 2001: 65–92.
(1996b), ‘External Justifications and Institutional Roles’, The Journal of
Philosophy 93: 28–36; reprinted in Simmons 2001: 93–101.
270 Bibliogr aphy

(2001), Justification and Legitimacy: Essays on Rights and Obligations (Cambridge

University Press).
Skorupski, John (1999), Ethical Explorations (Oxford University Press).
Smith, John (1859), Select Discourses [1600], 4th edn, edited by Henry Griffin
Williams (Cambridge University Press).
Smith, Michael (1994), The Moral Problem (Oxford: Blackwell).
Steiner, George (1971), In Bluebeard’s Castle: Some Notes Towards the Redefinition of
Culture (New Haven: Yale University Press).
Stern, Robert (2002), Hegel and the ‘Phenomenology of Spirit’ (London:
Routledge).
(2004a), ‘Does “Ought” Imply “Can”? And Did Kant Think it Does?’, Utilitas
16: 42–61.
(2004b), ‘Pinkard on German Idealism’, Bulletin of the Hegel Society of Great
Britain, 49/50: 1–17.
(2006), ‘Hegel’s Doppelsatz: A Neutral Reading’, Journal of the History of
Philosophy 44: 235–66.
(2007), ‘Freedom, Self-Legislation and Morality in Kant and Hegel:
Constructivist vs Realist Accounts’, in Espen Hammer (ed.), German Idealism:
Contemporary Perspectives (London: Routledge), 245–66.
(2010), ‘Moral Scepticism and Agency: Kant and Korsgaard’, Ratio 23:
453–74.
Stewart, Jon (ed.) (1996), The Hegel Myths and Legends (Evanston: Northwestern
University Press).
(2003), Kierkegaard’s Relation to Hegel Reconsidered (Cambridge University
Press).
Stratton-Lake, Philip (1996), ‘In Defence of the Abstract’, Bulletin of the Hegel
Society of Great Britain 33: 42–53.
(2000), Kant, Duty and Moral Worth (London: Routledge).
Street, Sharon (2008), ‘Constructivism about Reasons’, in Russ Shafer-Landau
(ed.), Oxford Studies in Metaethics, volume iii (Oxford University Press),
207–46.
(2010), ‘What Is Constructivism in Ethics and Metaethics?’, Philosophy Compass
5: 363–84.
Suarez, Francisco (1612), De Legibus, ac Deo Legislatore, partially reprinted in
Selections from Three Works of Francisco Suarez, vol. i (Oxford University Press,
1944); translated as Suarez 1944.
(1944), On Laws and God the Lawgiver, in Selections from Three Works of Francisco
Suarez, vol. ii, translated by Gwladys L. Williams, Ammi Brown and
John Waldron (Oxford University Press), 3–646; partially reprinted in
Schneewind (ed.) 2003: 68–86.
Swinburne, R. G. (1974), ‘Duty and the Will of God’, Canadian Journal of
Philosophy 4: 213–27.
Superson, Anita M. (2009), The Moral Skeptic (Oxford University Press).
Taylor, Mark C. (2000), Journeys to Selfhood: Hegel and Kierkegaard, reprint edn
(New York: Fordham University Press).
Taylor, Richard (1988), ‘Ancient Wisdom and Modern Folly’, in Peter A. French,
Theodore E. Uehling, and Howard K. Wettstein (eds.), Ethical Theory:
 Bibliogr aphy 271

Character and Virtue, Midwest Studies in Philosophy Volume xiii (Notre Dame
University Press), 54–63.
Thulstrup, Niels (1980), Kierkegaard’s Relation to Hegel, translated by George L.
Stengren (Princeton University Press).
Tillich, Paul (1946), ‘Religion and Secular Culture’, The Journal of Religion 26:
79–86.
Timmermann, Jens (2006), ‘Value without Regress: Kant’s “Formula of
Humanity” Revisted’, European Journal of Philosophy 68: 69–93.
(2007), Kant’s ‘Groundwork of the Metaphysics of Morals’: A Commentary
(Cambridge University Press).
Urmson, J. O. (1988), Aristotle’s Ethics (Oxford: Blackwell).
Vasquez, Gabriel (1606), Commentaria ac Disputationes in Primam Secundae Sancti
Thomae, 2 vols. (Ingolstadt).
Wainwright, William J. (2005), Religion and Morality (Aldershot: Ashgate).
Watkins, Julia (1995), ‘Judge William – A Christian?’, in Robert L. Perkins (ed.),
International Kierkegaard Commentary: Either/Or (Macon: Mercer University
Press), 113–24.
Watson, Gary (2007), ‘Morality as Equal Accountability: Comments on Stephen
Darwall’s The Second-Person Standpoint’, Ethics 118: 37–51.
Westphal, Merold (1981), ‘Abraham and Hegel’, in Robert L. Perkins (ed.),
Kierkegaard’s ‘Fear and Trembling’: Critical Appraisals (The University of
Alabama Press), 62–80; reprinted in Westphal 1991: 61–84.
(1988), ‘Commanded Love and Moral Autonomy: The Kierkegaard-Habermas
Debate’, Kierkegaard Studies Yearbook: 1–22.
(1991), Kierkegaard’s Critique of Reason and Society (Pennsylvania State University
Press).
(1998), ‘Kierkegaard and Hegel’, in Alastair Hannay and Gordon D. Marino
(eds.), The Cambridge Companion to Kierkegaard (Cambridge University
Press), 101–24.
(2004), ‘Review of Stewart’, Søren Kierkegaard Newsletter 48: 10–15.
White, Nicholas (1992), ‘The Attractive and the Imperative: Sidgwick’s View of
Greek Ethics’, in Bart Schultz (ed.), Essays on Henry Sidgwick (Cambridge
University Press), 311–32.
(2002), Individual and Conflict in Greek Ethics (Oxford University Press).
Willaschek, Marcus (2006), ‘Practical Reason: A Commentary on Kant’s
Groundwork of the Metaphysics of Morals (GMS ii, 412–17)’, in Christoph
Horn and Dieter Schönecker (eds.), Groundwork for the Metaphysics of Morals
(Berlin: Walter de Gruyter), 121–38.
Wolf, Susan (2009), ‘Moral Obligations and Social Commands’, in Samuel
Newlands and Larry M. Jorgensen (eds.), Metaphysics and the Good: Themes
from the Philosophy of Robert Merrihew Adams (Oxford University Press),
343–67.
Wolff, Christian (1733), Vernünfftige Gedancken von der Menschen Thun und
Lassen, zu Berförderung ihrer Glückseeligkeit, 4th edn (Frankfurt and
Leipzig), facsimile edition in Gesammelte Werke, edited by Jean École et al.
(Hildesheim: Georg Olms, 1976), vol. i.4; partially translated in Wolff
2003.
272 Bibliogr aphy

(2003), Reasonable Thoughts About the Actions of Men, for the Promotion of Their
Happiness, translated by J. B. Schneewind in Schneewind (ed.) 2003:
333–50.
Wolff, Robert Paul (1970), In Defense of Anarchism (New York: Harper
Torchbooks).
Wood, Allen W. (1990), Hegel’s Ethical Thought (Cambridge University Press).
(1999), Kant’s Ethical Thought (Cambridge University Press).
(2008), Kantian Ethics (Cambridge University Press).
Yovel, Yirmiahu (1975), ‘Hegel’s Dictum that the Rational is Actual and the
Actual is Rational: Its Ontological Content and its Function in Discourse’,
in Werner Becker and Wilhelm K. Essler (eds.), Konzepte der Dialektik
(Frankfurt: Klostermann), 111–27; reprinted in Stewart (ed.) (1996):
26–41.
 I n de x

Acton, H.B., 34n.70 Buchwalter, Andrew, 136n.84, 158n.33

Adams, Robert Merrihew, 42n.2, 152n.14,
153n.16, 221n.2, 224, 226, 228n.19, Cantrell, Michael A., 213n.30
251, 252 Chalier, Catherine, 11n.11
Allison, Henry E., 105n.2, 118n.37, Chytry, Josef, 144n.107
231n.23 Clarke, Samuel, 48, 50, 53, 69, 70n.5
Alston, William P., 76n.23 Cohen, G.A., 13n.17, 15n.22, 72n.14, 164
Ameriks, Karl, 34n.70 Collins, James D., 173n.1
Annas, Julia, 132n.74 Connell, George B., 227n.16
Anscombe, G.E.M., 14, 42n.1, 69n.1, constructivism, 8–10, 13–14, 89,
150–1, 154–5, 156, 213, 214, 228, 229 see also Kant, Immanuel: and
anti-realism, 7, 9, 13, 32, 41, 89–91, 97, 99 constructivism
Aquinas, Thomas, 13n.17, 44n.5, 69 continence, 110, 111, 112, 122, 132, 143,
argument from autonomy, 1, 3, 7, 8, 8n.3, 239, 240, 241
10, 14, 15, 18, 22, 24, 26, 41, 42, 49, Crisp, Roger, 4n.2, 69n.1
53, 57, 57, 65, 99, 220, 221–9 Critchley, Simon, 248
Aristotle, 110–11, 122–3, 133, 136, 236–44 Crites, Stephen, 173n.1
autonomy, 1, 2, 4, 23, 154, Crosby, John F., 47n.9
see also argument from autonomy; Crusius, Christian August, 50, 58, 63–4,
Kant, Immanuel: and autonomy 65, 73–4
Cudworth, Ralph, 69, 70n.5
Baier, Kurt, 150, 151–5 Cullity, Gareth, 9n.4
Barbeyrac, Jean, 49, 70, 71–3 Culverwell, Nathaniel, 98
Baumgarten, Alexander Gottlieb, 50, 51, Cumberland, Richard, 47n.8, 73n.16
55n.43, 77n.26
Baxley, Anne Margaret, 105n.2, 112n.22, Darwall, Stephen, 3n.1, 10n.9, 12, 12n.12,
114n.26, 130n.71, 130n.69, 131n.73, 22n.37, 24n.39, 91n.65, 152n.14,
132n.74, 232n.27, 233n.29, 238n.43, 153n.17, 154, 225n.13
240–1, 246n.58 Dean, Richard, 32n.63, 32n.62, 34n.71,
beautiful soul, 112, 117, 118, 119, 120, 121, 35n.74, 36n.75
125, 126, 130 Deligiorgi, Katerina, 105n.2
Beck, Lewis White, 76n.24, 105n.2 Descartes, René, 70n.5
Beiser, Frederick, 12n.12, 105n.2, 106n.5, Dews, Peter, 242n.49
124n.54, 134n.79, 142n.100 divine command theory, 2, 3, 7, 22, 41,
Bradley, F.H., 161, 164–70 42, 42n.2, 45, 46, 47, 70, 71–4, 98–9,
Brandt, R.B., 4n.2 152, 156, 215, 220, see also modified
Brink, David O., 93n.69 divine command theories; obligation:
Broadie, Sarah, 122n.45, 237n.42, intermediate accounts of
239n.46 divine grace, 216, 217
Brochard, Victor, 151n.11 Dworkin, Gerald, 245n.56

273
274 Index

Engstrom, Stephen, 39n.82 Henrich, Dieter, 130n.69, 135n.80

Euthyphro dilemma, 8, 22, 23, 35, 103, Herman, Barbara, 9n.6, 27n.47
221, 222 heteronomy, 3, 42n.2, 89, see also Kant,
Evans, C. Stephen, 173n.2, 180, 181n.22, Immanuel: and heteronomy
206n.9, 211n.21, 213n.27, 222n.3, Hills, Alison, 9n.4, 27n.47, 31
228n.19, 251, 252 Hobbes, Thomas, 45n.5, 47n.8, 50, 71n.9,
externalism, existence, 80–97 225n.13
Hölderlin, Friedrich, 139
Fackenheim, Emil, 242n.49, 247n.63 holy will, 2, 66, 68, 75–89, 103, 104, 110,
Falk, W.D., 95–7 119, 125, 131, 133, 218, 229, 230–1,
Feinberg, Joel, 228n.18 235, 239, 240
Ferreira, M. Jamie, 173n.2, 174n.6, Horton, John, 163
211n.21, 219n.48 Humboldt, Wilhelm von, 140
Finnis, John, 44n.3 Hume, David, 26, 52
Foot, Phillipa, 122n.45, 238 Humean constructivism, 15n.21
Formosa, Paul, 14n.19, 32n.62, 33n.65 Hunter, Ian, 71n.7
Hursthouse, Rosalind, 120n.42, 122n.45
Galvin, Richard, 10n.7 Hutcheson, Francis, 74n.19
Garner, Richard, 52n.33
Gaut, Berys, 9n.4, 36n.75 Idziak, Janine Marie, 48n.10
Gauthier, Jeffrey A., 105n.2 imperatives, categorical, 25, 78–80,
Gibbard, Alan, 152n.14 see also necessitation
Goethe, J.W. von, 117n.33, 135 imperatives, hypothetical, 25, 78–9, 98,
Goodman, Lenn E., 249n.65 see also necessitation
Gordon, Peter E., 16n.26 internalism, existence, 91–7
Green, Ronald M., 173n.2, 174n.5 internalism, motivational, 10
Grenberg, Jeanine, 239n.45, 241n.48 Irwin, Terence, 25n.41, 34n.70, 44n.4,
Griffin, James, 249 46n.7, 69n.3, 69n.1, 71n.7, 76n.23,
Guyer, Paul, 27n.47, 31n.60, 32n.61, 90n.63, 174n.4, 243n.52
33n.66, 35, 79n.32, 130n.70 Ives, Margaret C., 134n.78

Haakonssen, Knud, 43n.3 Jackson, M.W., 247n.63

Haldane, John, 43n.3 James, David, 173n.1
Hannay, Alasdair, 200n.89, 205n.4 Johnson, Robert N., 22–4, 34n.71
Hardimon, Michael O., 161
Hare, John E., 34n.70, 42n.2, 58–67, 206, Kain, Patrick, 34n.70, 58n.48
222n.3, 225n.12, 227n.16, 228n.19, Kant, Immanuel, 1–3
246, 246n.61 and autonomy, 1, 2, 7–8, 10, 11–12, 15,
Hart, H.L.A., 152n.14 18–20, 23, 24, 31, 34, 41, 42, 53–5,
Hegel, G.W.F., 1, 2 57, 65, 68, 99, 221–9, 228n.19, 245
account of obligation, 2–3, 156–61, 168, and constructivism, 2, 7, 8n.3, 10n.7,
173, 220–1, 226 10–13, 15, 20, 24–5, 33, 43, 89–91,
and Aristotle, 137–9 99, 228
and constructivism, 148, 149 and divine command theory, 53–67, 68,
and Kantian paradox, 1, 2, 15–16 74–5, 80, 98–9, 173, 216
and realism, 148–9 and externalism, 91–7
and social roles, 161, 164 and heteronomy, 18–20, 24, 25, 34,
and the Greeks, 139–41 53–7, 64, 99
and virtue, 137–9 and internalism, 91–7
as critic of Kant, 104, 136–47, 156–7, and natural law, 68, 74–5, 98–9
229–47 and realism, 2, 8, 26–40, 68, 89–91, 95,
influence of Schiller, 134, 135–47 99, 148
on freedom, 144–5, 148, 159, 160 and self-legislation, 99, 151, 213
on love, 141–3 and the Formula of Humanity, 28–9,
on the love commandment, 250n.67 32–3, 34, 39
on the state, 143–4, 159, 160, 247–8 hybrid theory of, 2, 68, 76–99, 103–4,
see also Kierkegaard, Søren: critique of 147, 150, 152, 156, 173, 220, 229
Hegel; Sittlichkeit on autocracy, 245
 Index 275

on duty and inclination, 2, 3, 76, 77, 78, Leibniz, Gottfried Wilhelm, 21, 49–50,
81, 86, 87, 88, 90, 94, 98, 104, 105, 51, 56n.44, 70–1, 74, 98
114, 132, 137, 229, 230, 231, 232, 238, Levinas, Emmanuel, 11n.11
243 Lillegard, Norman, 198n.77
on possibility and actuality, 91n.64 Løgstrup, Knud Ejler, 211n.21, 222n.4,
on Schiller, 130–5, 232n.27, 233 223n.7
on virtue, 130–1, 233, 236 Louden, Robert B., 151n.11
see also Hegel, G.W.F.: as critic of
Kant; holy will; Kantian paradox; MacIntyre, Alasdair, 17
Kierkegaard, Søren: and Kant; Mackie, J.L., 53n.33, 205n.7
maxims; Schiller, Friedrich: as Manis, R. Zachary, 173n.2
critic of Kant; transcendental maxims, 81–3
idealism McDowell, John, 236n.39, 239n.45,
Kantian paradox, 1, 2, 13, 13n.14, 14, 15, 244n.53
103, 148, 150 Mill, J.S., 152–3
Kerstein, Samuel J., 54n.36, 83, 84 modified divine command theories, 221
Kierkegaard, Søren, 1, 2–3, 16–18 moral demand, 3, 200, 206, 207, 218, 246,
and divine command theory, 173–5, 247, 248, 249, 250
204, 215–19, 220–9 moral realism, see realism
and faith, 175–6, 177–80, 185, Mouw, Richard J., 42n.2
188, 212 Murdoch, Iris, 17
and intermediate divine command
theory, 175, 188 naturalism, 10
and Kant, 181n.22, 217, 218, 221–9 necessitation, 2, 52, 53, 77, 78, 79, 80, 81,
and knight of faith, 176, 177, 178, 182, 83, 127, 130, 131, 133, 134, 230, 231,
183, 185 232, 233n.30, 233–5
and the absurd, 176–80, 183, 185 Neiman, Susan, 11n.11
and the honest pagan, 206, 209, 210, Nicholson, Peter P., 169n.62
211, 244, 248, 249 Nietzsche, Friedrich, 245, 245n.55
and the love commandment, 204–19, Norton, Robert E., 106n.4, 121n.43
250 Nowell-Smith, P.H., 42n.2
and the teleological suspension of the Nussbaum, Martha, 132n.74
ethical, 180, 223
and the tragic hero, 181, 187 O’Neill, Onora, 245n.56
critique of Hegel, 173–203, 204, 207, obligation, 2–3, 7
214–15, 252–3 and autonomy, 42
Either/Or, 189–203, 207 and law, 42
Fear and Trembling, 174–89 identificatory accounts of, 162–4
Works of Love, 204–19 intermediate accounts of, 44, 46, 98,
King, William, 50 152, 153, 222
Kirmmse, Bruce, 217 natural law accounts of, 43, 45, 46, 47,
Knowles, Dudley, 149n.2 51, 68–74, 98–9
Korkman, Petter, 71n.7 nature of, 45–8
Körner, Christian Gottfried, 105, 109, problem of, 2, 40, 43–52, 76, 104, 134,
111, 112, 144, 242n.49 146
Korsgaard, Christine M., 9, 12, 12n.13, social command accounts of, 2, 150–5,
13n.15, 14, 20n.34, 20n.33, 22n.37, 167, 168, 170, 184, 214, 226, 252
25n.43, 33n.68, 33n.64, 42, 69n.2, social role accounts of, 161–70
90n.63, 163–4, 225n.13, 225n.12, voluntarist accounts of, 43, 47, 50, 174
233n.30, 233–5, 237n.40 see also argument from autonomy;
Krasnoff, Larry, 16n.26, 17n.28 divine command theory; Hegel,
G.W.F.: account of obligation;
Lafont, Christine, 11n.11, 20 Kant, Immanuel: hybrid theory of;
Langton, Rae, 27n.47, 35 Kierkegaard, Søren: and divine
Larmore, Charles, 13 command theory; modified divine
Law, David R., 199n.84 command theory
LeBar, Mark, 12n.12 Olafson, Frederick A., 8n.3, 174n.4
Lee, Seung-Goo, 181n.22 Outka, Gene, 181n.21, 223n.5
276 Index

Parkin, Jon, 47n.8 on beauty, 105, 107, 108, 109, 112,

Paton, H.J., 79n.32, 83n.44, 105n.2, 114n.26 118–19, 127–8, 135
Pattison, George, 173n.1, 174n.3 on dignity, 112, 118–26, 127–8
perfectionism, 18, 19, 19n.32, 21, 25, 51, on duty, 105, 107, 131
55, 57, 63, 64n.66 on grace, 112, 117, 118–26, 127–8
Perkins, Robert L., 190n.44, 199n.84, on love, 124, 141–3
200n.88, 207n.11 on the aesthetic state, 128–9, 143
Pidgen, Charles, 69n.1 on the sublime, 112, 118, 127–8, 146
Pink, Thomas, 46n.7, 69n.1 on virtue, 114–15, 119, 121, 125
Pinkard, Terry, 13n.14, 16n.26, 17n.28, see also beautiful soul; Hegel, G.W.F.:
139n.95, 148 influence of Schiller
Pippin, Robert, 13n.14, 15, 16n.25, Schmucker, Josef, 50n.21
16n.24, 27n.47, 148, 149n.5 Schneewind, J.B., 7, 12n.13, 12n.12, 12,
Pistorius, Herman Andreas, 36–40 20n.33, 32n.62, 36n.75, 50, 50n.18,
Plato, 8, 113 71n.7, 99, 132n.74
Prauss, Gerold, 105n.2 Schönecker, Dieter, 27n.47, 29n.54
Price, Richard, 69 Schopenhauer, Arthur, 14n.18, 245n.55
Prichard, H.A., 86–8, 96 Schwaiger, Clemens, 50n.21, 77n.26
Pufendorf, Samuel, 4, 45n.5, 49–50, 51, Sciaraffa, Stefan, 162n.47
70–1, 225n.13 Scotus, Duns, 45n.5
Pugh, David, 112n.22, 128n.63 second nature, 111, 137
Selden, John, 47n.8, 73n.16, 73n.15
Quinn, Philip L., 173n.2, 188n.40, self-legislation, 1, 2, 11, 12, 13, 15, 18,
206n.9, 222n.3 228, see also Anscombe, G. E. M.;
Kant, Immanuel: and self-legislation;
Rauscher, Frederick, 12n.12 Kantian paradox
Rawls, John, 8, 9, 11, 12, 18–22, 20n.33, Sensen, Oliver, 26n.46, 32n.63, 32n.62,
24, 24n.39, 25–6, 36, 41 33n.67, 33n.65, 36n.75
realism, 1, 7, 8–10, 11, 13, 16, 18, 20, 23, Sharpe, Lesley, 124n.54
25, 41, 57, 89–91, 148, 220, 232n.26, Sherman, Nancy, 136n.83, 237n.40
see also Kant, Immanuel: and realism; Sidgwick, Henry, 156, 232n.26
value Silber, John R., 8n.3, 36n.75
Reath, Andrews, 12, 12n.12, 20n.33, Simmons, A. John, 164
21n.36, 24n.39, 27n.47 Sittlichkeit, 157, 160, 181, 183, 185, 188,
Regan, D.H., 21n.35, 148n.1 195, 198, 247
Reiner, Hans, 105n.2, 130n.69 Skorupski, John, 152n.14
Reinhold, Karl Leonhard, 126 Smith, John, 50, 50n.20
Roehr, Sabine, 105n.2 Smith, Michael, 92n.67
Rorty, Richard, 245 ‘the standard story’, 1, 2, 7, 8, 18, 26, 40,
Rosler, Andrés, 132n.74, 151n.11 41, 57, 99, 103, 148, 150, 220
Ross, W.D., 96 Steiner, George, 205n.7
Rousseau, Jean-Jacques, 1, 20n.33, 138 Stern, Robert, 10n.8, 29n.56, 148n.1,
Rudd, Anthony, 198n.77, 205n.7 150n.8, 247n.63, 249n.66
Stewart, Jon, 173n.1, 190n.43, 199n.84
scepticism, moral, 10, 10n.8, 29, 52, 93 Stratton-Lake, Philip, 230, 231
Schelling, F.W.J., 139 Street, Sharon, 8, 15n.21
Schiller, Friedrich Suarez, Francisco, 44n.4, 46, 46n.6,
Aesthetic Letters, 126–9, 135 47n.9, 49, 98, 152
and the Greeks, 128, 139–41 Superson, Anita, 93n.68
as critic of Kant, 104–5, 109–10, Swinburne, Richard, 226n.14
113–21, 125–6, 129–35, 136–47, 230,
231, 235, 237, 241, 245 Taylor, Jeremy, 73n.16
Grace and Dignity, 111–40, 141, 235 Taylor, Mark C., 173n.1
Kallias Letters, 105–11, 115, 120, 122, Taylor, Richard, 155n.19
125, 126, 144 theonomy, 226–7
on autonomy, 106, 108–9, 144 Thulstrup, Niels, 173n.1, 190n.43
 Index 277

Tillich, Paul, 226 Wainwright, William J., 189n.40, 221n.1,

Timmermann, Jens, 10n.8, 29n.56, 251n.68
33n.64, 34n.70, 35n.72, 77n.27, Watkins, Julia, 198n.77
79n.32, 105n.2, 112n.22 Watson, Gary, 153n.17
transcendental idealism, 10, 11, 31, 79, Westphal, Merold, 173n.1, 173n.1,
80, 91, 146 181n.22, 199n.84, 227n.16
White, Nicholas, 139n.95, 146n.116,
Urmson, J.O., 236n.39 151n.11, 156n.23, 236n.39, 239n.45
Willaschek, Marcus, 88n.57
value, 1, 8, 11, 23, 26–40, 32n.62, 149, 175 Wincklemann, Johann Joachim, 139
Vasquez, Gabriel, 46, 46n.7, 69 Wolf, Susan, 155, 169n.62, 252
virtue, 110, 111, 112, 114, 119, 122, 128, Wolff, Christian, 50–1, 73
133, 136, 139, 145, see also Hegel, Wolff, Robert Paul, 228n.18
G.W.F.: and virtue; Kant, Immanuel, Wood, Allen W., 27, 34n.70, 87n.54,
on virtue; Schiller, Friedrich: on 105n.2, 149
virtue
virtue ethics, 150, 151, 155 Yovel, Yirmiahu, 247n.63