MODERN MORAL PHILOSOPHY

In this magisterial study, one of our leading moral philosophers refutes

the charge (originally made by Elizabeth Anscombe) that modern ethics
is incoherent because it essentially depends on theological and religious
assumptions that it cannot acknowledge. Stephen Darwall’s panoramic
picture starts with the seventeenth-century thinker Grotius and tells the
story continuously down to the time of Kant, exploring what was in fact
a completely new way of doing ethics based on secular ideas of human
psychology and universal accountability. He shows that thinkers from
Grotius to Kant are profoundly united by this modern approach, and that
it helped them to create a theory of natural human rights that remains
of great political relevance today. He further shows that this new way of
thinking provides conceptual resources that are far from exhausted, and
that moral philosophy in this idiom still has a vibrant future.

Stephen Darwall teaches philosophy at Yale University. He is the

author of The British Moralists and the Internal ‘Ought’, 1640–1740 (Cam-
bridge University Press, 1995), and of many other publications in moral
philosophy and its history, including The Second-Person Standpoint:
Morality, Respect, and Accountability (2006).

Published online by Cambridge University Press

Published online by Cambridge University Press
 MODERN MORAL PHILOSOPHY
From Grotius to Kant

STEPHEN DARWALL
Yale University

Published online by Cambridge University Press

 Shaftesbury Road, Cambridge CB2 8EA, United Kingdom
One Liberty Plaza, 20th Floor, New York, NY 10006, USA
477 Williamstown Road, Port Melbourne, VIC 3207, Australia
314–321, 3rd Floor, Plot 3, Splendor Forum, Jasola District Centre,
New Delhi – 110025, India
103 Penang Road, #05–06/07, Visioncrest Commercial, Singapore 238467

Cambridge University Press is part of Cambridge University Press & Assessment,

a department of the University of Cambridge.
We share the University’s mission to contribute to society through the pursuit of
education, learning and research at the highest international levels of excellence.

www.cambridge.org
Information on this title: www.cambridge.org/9780521860475
DOI: 10.1017/9781139025065
© Stephen Darwall 2023
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 & Assessment.
First published 2023
Printed in the United Kingdom by TJ Books Limited, Padstow, Cornwall
A catalogue record for this publication is available from the British Library.
A Cataloging-in-Publication data record for this book is available from the Library
of Congress
ISBN 978-0-521-86047-5 Hardback
Cambridge University Press & Assessment 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.

Published online by Cambridge University Press

 For Jerry, David, and Bill

Published online by Cambridge University Press

Published online by Cambridge University Press
 CONTENTS

Preface xi
Acknowledgments xiv

Introduction  1
Anscombe’s Challenge  1
Sidgwick’s Contrast  4
Modern Moral Philosophy’s Shadow I   6
Modern Metaethics  7
Modern Moral Philosophy’s Shadow II   8
The Case of Kant   10
(Relatively) Modest Aspirations   10

1 Grotius  15
Classical Natural Law: Aquinas and Suárez   16
Grotius on “ius,” Morality, and Obligation   20
Morality as a Distinctive Source of Reasons   25
Freedom, Self-Rule, and the Right to Punish   31
Publicly Articulable General Principles: A
“Science” of Morality   34
Sociability as the “Fountain of Right”   36

2 Hobbes and Pufendorf   39

Hobbes  43
Desire and the Good   44
Obligation and the Right   48
Grounding the Right in the Good?   52
Irreducibly Deontic Laws of Nature?   54
A Metaethics of Right?   59
Pufendorf  61
Moral Powers I   65
A “Sociable” Attitude   70
Equal Human Dignity   73
Holding to Obligations in the Natural State   76

vii

Published online by Cambridge University Press

 viii contents
Moral Powers II   79
Moral Agency  82

3 Locke and Cumberland   87

Locke: Natural Law, Motive, and Sanction   90
Morality as Solution to the Human Natural Collective
Action Problem  93
Right and Good   95
Law, Accountability, and Autonomy   98
Cumberland  100
Practical Propositions  102
Obligation  103
Empirical Grounds for the Law of Nature, Understood
Egocentrically  104
Practical Dictates of Reason, Agreement, and the Best End   105
Obligation, Reconsidered  108

4 Spinoza, Cudworth, Shaftesbury, and Leibniz   111

Spinoza  116
Against Deontic Morality: Religious or Secular   116
The Good: Knowledge, Power, and Perfection   119
Revised Deontic Moral Concepts and Democratic Politics   122
“Bear[ing] Calmly” and “Living Blessedly”: Spinoza on Freedom,
Happiness, and Power   125
Cudworth  128
Against Positivism  128
Metaphysical Idealism?  130
Ethical Idealism?  131
Love  134
Self-Determination  135
Free Will and Obligation, Animal and Moral   137
Shaftesbury  139
An Ethics of Virtue   140
Virtue and Moral Sense   142
Virtue and Self-Reflective Moral Agency   145
The “Obligation” to Virtue   146
Authorship and Self-Determination   147
Ancient or Modern?   151
Leibniz  152
Perfectionism: Identifying Goods (Personal and Impersonal)
and Justice  154
Freedom, Agency, and Necessity   157
Right and Claim   158

Published online by Cambridge University Press

 contents ix

5 Hutcheson and Butler   165

Hutcheson  172
Empiricist Naturalism and Moral Sense   173
Moral Virtue Ethics: Constructing and Deriving Other Moral
Notions and Standards   176
Calm Desires, Reason, and Moral Sense   178
Obligation  181
Butler  183
The Human Moral Psychological Economy   185
Normative Practical Judgment: Wide and Narrow   188
The Authority of Conscience   192
The Right and the General Good   197
Against Egoism  199

6 Hume and Smith   201

Hume  203
Against Practical Reason   206
Against Moral Rationalism and for Moral Sentimentalism   207
Justice, the Artificial Virtue   213
Smith  222
Sympathy: Attributing Mental States and Judging
Their Propriety  224
The Metaethics of Propriety   228
Justice, Mutual Respect, and Accountability   231
Dignity and Self-Command, the “Respectable” Virtue   234

7 The British Rationalists and Reid   237

Fordyce  239
Clarke  243
Balguy  248
Price  250
Reid  258

8 Rousseau and Kant   269

Rousseau  274
Kant: Groundwork I  278
Groundwork II: FUL  287
Groundwork II: FH, FA, and FKE   299
The Dignity of Persons and Autonomy of the Will   306
Autonomy in Groundwork III  309
The Critique of Practical Reason: Material vs. Formal
Practical Principles  314
The Fact of Reason   322

Published online by Cambridge University Press

 x contents
Necessitation and Respect   325
Kant on the Right and the Good   330
Freedom, Constraint, and Right in The Metaphysics of Morals  334
Kant’s Racism and Sexism, and a Final Note of Humility   340

Works Cited   343

Index  363

Published online by Cambridge University Press

 PREFACE

This is the first of a projected two-volume work on the history of moral philoso-
phy in the modern period, beginning with Grotius and extending through the
end of the twentieth century, more or less. This first volume will end with Kant,
and the next will begin with Fichte and Hegel. As the Introduction to follow
makes clear, even at two completed volumes the work will be far from com-
prehensive. It will be framed by the challenge Elizabeth Anscombe famously
issued in “Modern Moral Philosophy” that Western ethical philosophers’ focus
on deontic morality during this period has been substantially misplaced (Ans-
combe 1958).1
My attention will be devoted mainly to modern philosophers’ attempts to
respond to Anscombe’s challenge (often avant la lettre), that is, to theorize,
defend, and ground deontic morality. However, I will also be concerned
with attempts to criticize or limit morality, for example, in Hegel, Marx, and
Kierkegaard, as well as to undermine it completely, as in Nietzsche (and, argu-
ably, Spinoza). As we shall see, these philosophers pay homage to the power
and influence of the modern idea of morality as well, even as they seek to
oppose or restrict it. And there will be discussions too of aspects of modern
ethical thought that do not focus on deontic morality at all, for example, ethics
of virtue and feminist critiques of orthodox moral theory in the late twentieth
century.
I began to work on this project over twenty years ago.2 Painfully aware of my
ignorance of various periods and important figures, especially post-Kantian
Continental thinkers, I undertook to fill in some of the gaps. Supported by a
grant from the National Endowment for the Humanities in 1998–99, I began to
study Fichte and Hegel and came upon Allen Wood’s Hegel’s Ethical Thought.
There I found Wood’s chapter on recognition and its section “Fichte’s Theory
of Recognition.” I read this with something approaching ecstatic endorsement,

1
I will thus be neglecting a number of figures who are less significant from this perspective,
such as Descartes, Malebranche, Voltaire, Diderot, Montesquieu, and d’Holbach, among
others.
2
When Paul Guyer and Gary Hatfield invited me to contribute to a series of histories of
modern philosophy they were editing for Cambridge University Press.

xi

https://doi.org/10.1017/9781139025065.001 Published online by Cambridge University Press

 xii preface

with a resounding “yes!” as Nietzsche might put it (Wood 1990: 77–93;

Nietzsche 2007: 377). Encountering for the first time Fichte’s idea that it takes
a “summons” (Aufforderung) from another person to gain practical conscious-
ness of our own freedom, and that this reciprocal awareness of summoned and
summoner alike can ground fundamental principles of a theory of equal right,
it felt as if the scales had fallen from my eyes (Fichte 2000: 35–83).
I came face-to-face with an idea that, as I allowed it to penetrate my thought
fully over the weeks and months to come, seemed to me to provide the missing
piece necessary to adequately ground morality, something I had been search-
ing for perhaps my entire philosophical life. When I combined “Fichte’s Point,”
as I came to call it, with Pufendorf and Grotius on sociability, Reid on the
“social operations of the mind,” Smith on the role of empathy (“sympathy”) in
judgments of justice, and, most importantly, Strawson on the role of the “par-
ticipant” stance in moral responsibility, these insights appeared to me to point
toward a fundamental reorientation of the foundations of morality and moral
theory, providing a grounding for both in the “second-person standpoint.”
I pretty much put my historical project on hold that year. The history of eth-
ics could wait for a while; after all, the texts weren’t going anywhere. I began to
work on what became The Second-Person Standpoint almost immediately and for
the last twenty years have pursued what I have taken to be the fruits of Fichte’s
insight (and the complementary insights of other historical figures) through many
articles and books (Darwall 2006, 2013a, 2013b).
Some of those essays were explicitly historical – on Grotius, Pufendorf,
Smith, Kant, and Nietzsche – as I found central elements of the second-
personal framework in these writers, sometimes under attack, as in Nietzsche
(Darwall 2013a, 2013b). Increasingly, I came to realize that even when modern
philosophers did not make explicit use of Fichte’s insight, it was implicit in
their thought, nonetheless.
A good example is the distinction between law and counsel Suárez drew at
the beginning of the early modern period, which influenced so much of the
moral philosophy that followed.3 Suárez notes a fundamental conceptual con-
trast between a normative area that is tied intrinsically to accountability and
free moral agency (law) and one that is not (counsel). There can be beings who
are agents having normative reasons but who lack the kind of freedom (moral
agency) that Fichte relates distinctively to rights. (This, I came to see, was
essentially the same as Cudworth’s distinction between “animal free will” and
“moral free will,” which I discuss in Chapter 4.) Suárez’s distinction was taken

3
Suárez was not the first to draw this distinction, although his version was especially salient
to modern moral philosophers who followed. Schneewind describes Grotius’s invocation
of it as following “the Scholastic tradition” (Schneewind 1998: 74). I am indebted to an
anonymous referee to the Press for pressing this point.

https://doi.org/10.1017/9781139025065.001 Published online by Cambridge University Press

 preface xiii

up explicitly by Grotius, Hobbes, and Kant and implicitly by many others. It

is in the background of almost all modern moral philosophy, as will begin to
become clear in the Introduction.
If the texts stayed pretty much the same during the last twenty years, his-
torical scholarship has not. The most prominent intervention has been Terence
Irwin’s magisterial The Development of Ethics, almost two volumes of which are
devoted to the modern period (Irwin 2008, 2009). Many excellent smaller stud-
ies have appeared as well. Just within Fichte scholarship, both Michelle Kosch
and Allen Wood have published superb works (Kosch 2018; Wood 2016). And
Kosch has also brought out an unsurpassed treatment of Kierkegaard’s ethics
(Kosch 2006). These are only a few examples. The last twenty plus years have
been a period of extraordinary flourishing in the history of modern philosophi-
cal ethics.4
I return to this project now with some trepidation, therefore, but also
with real excitement. There really does seem to me to be an untold story of
this period that emphasizes Anscombe’s challenge and modern philosophers’
attempts to deal with it. This volume, and the one to follow, are my attempt to
tell that story.
There is, moreover, even greater reason now to consider whether such a
story can be told, since Irwin’s Development of Ethics expresses skepticism
on this score. On Irwin’s telling, the history of modern philosophical ethics
is largely continuous with what he calls “Aristotelian Naturalism.” In this
respect, I will be writing against Irwin’s account, echoing Anscombe and other
historians who mark a significant contrast between modern ethical thought
and the ancient and medieval philosophy that preceded it.5
Since I have “a horse in the race,” readers will want to keep my philosophical
biases in mind. I can only hope that my having something at stake as a phi-
losopher makes for a more philosophically engaged (and engaging) histori-
cal account. My biases may make me sensitive to things others might not see,
though perhaps I am just looking through “second-personal glasses” that dis-
tort my reading of texts and obscure important historical context. I hope my
fellow historians of ethics, and you, dear reader, will keep me honest. Recalling
the old Russian proverb that Reagan loved to quote to Gorbachev – “Trust, but
verify” – I hope you will ignore the first and pursue the second.

4
Another striking example is Hurka’s study of British ethical thought from Sidgwick
through Ewing (Hurka 2014).
5
Here I have in mind Sidgwick, Schneewind, and Rawls, as I make clearer in the Introduction.

https://doi.org/10.1017/9781139025065.001 Published online by Cambridge University Press

 ACKNOWLEDGMENTS

I have been fortunate throughout my career in having mentors, colleagues, and

students who have helped me both philosophically and as an historian. Two of
the most prominent historians of ethics of the twentieth century were among my
mentors. As a graduate student at the University of Pittsburgh, I was in the first
seminar that J. B. Schneewind taught on Kant, along with Hobbes and Hume. At
that point, Schneewind was mostly absorbed in his landmark work on Sidgwick,
but one could already see in that seminar the seeds of what would emerge almost
thirty years later as The Invention of Autonomy, one of the most important works
on modern moral philosophy ever written (Schneewind 1977, 1998). Schneewind
has been both a model and source of encouragement ever since.
The other great twentieth-century historian of modern ethics I was fortunate
to learn from was William Frankena. Frankena published relatively little of his
historical scholarship, but his knowledge of this period was vast and deep. I got
to know him first in 1979–80 when he was a fellow at the National Humanities
Center in North Carolina. I was then an assistant professor at the University of
North Carolina at Chapel Hill, where I had also come under the tutelage of W.
D. (David) Falk, from whom I got an almost spiritual sense of the importance of
the early modern British moralists.6 Falk was at the National Humanities Center
that year as well, as were my former dissertation director Kurt Baier – with Falk,
an important figure in the “good reasons” tradition in mid-twentieth century
metaethics – and Annette Baier, who was in the process of becoming a leading
figure on Hume. To say that this was an ideal environment in which to have nur-
tured an interest in the history of modern moral philosophy is understatement.7

6
As I acknowledge in Darwall (1995: ix).
7
I didn’t actively pursue this interest right away, since I was still at work on Impartial Reason
(1983). But it did not take long. In 1986–87, two years after I moved to Michigan, I spent
a year in England doing research for The British Moralists and the Internal ‘Ought’ (1995).
Much of this was in the British Library, where David Falk had first read the British moral-
ists when he fled the Nazis. Falk did some of the first work on internalism in ethics. Indeed,
he invented the term (Falk 1948). And he was a powerful proselytizer for it, both philo-
sophically and in eighteenth-century British moral philosophy.

xiv

Published online by Cambridge University Press

 acknowledgments xv

I spent the fall semester of 1982 as a visitor at the University of Michigan,

talking often with Frankena. When I took a permanent position there in
1984, it was, in effect, to complete a new “Michigan Three,” with Allan Gib-
bard and Peter Railton – the original three having been Frankena, Stevenson,
and Brandt. (Ultimately, it became a “Michigan Five,” with Elizabeth Ander-
son and David Velleman.) Gibbard jokes that his becoming a noncognitivist
must have been determined by occupying Stevenson’s old budget line. I’m not
sure how it worked budgetarily, but Railton’s ethical naturalism was certainly
close to Brandt’s, and my interests were uncannily akin to Frankena’s. It was at
Michigan that I first attempted to do serious history of ethics, in early modern
British moral philosophy, certainly under Frankena’s influence, and with the
inspiration of Falk and Schneewind. It is with great gratitude that I dedicate
this volume to Jerry, David, and Bill.
I should also express my deep gratitude to John Rawls. In addition to
being the most important political philosopher of the twentieth century
and having a deep appreciation of metaethical issues that he normally kept
in the background, Rawls was a very careful reader of modern moral and
political philosophy. His posthumously published Lectures on both sub-
jects are both important works from which I here draw (Rawls 2000, 2007).
It was mostly as a philosophical and human model that Rawls most influ-
enced me, however. A Theory of Justice has shaped much of my own philos-
ophy from the beginning, and I have always been very grateful for Rawls’s
interest in a young philosopher’s work.
And then there is Yale. One of the attractions of the Yale Department,
which I have certainly enjoyed since joining it, is that it provides the best mix
of history of philosophy and philosophy proper of any department I know.
All the historians have serious philosophical interests that inform their his-
torical work, and almost all the non-historians are literate in and apprecia-
tive of philosophy’s history. And that includes the graduate students. There
is not a whiff of “Those who can, do philosophy, and those who cannot, do
history of philosophy.” It has been a pleasure and a privilege to be able to do
this work in such a supportive atmosphere. I am especially grateful to my fel-
low historians of modern philosophy, Michael Della Rocca, Paul Franks, and
Kenneth Winkler.
I owe a special debt to the students with whom I have done courses and
worked on dissertations in the history of modern moral philosophy, both at
Michigan and at Yale. Some of these seminars were taught with other faculty –
Scott Shapiro, Matthew Smith, and Gideon Yaffe – from whom I have learned

On reflection now, it is clear to me that my pursuing the history of internalism in Darwall

(1995) after having engaged with it as a metaethical position in Darwall (1983) is something
like the pattern instantiated in the present historical project in relation to my own second-
personal philosophizing (most obviously, in Darwall [2006]).

Published online by Cambridge University Press

 xvi acknowledgments

much. At Michigan, I was privileged to be able to sit in on courses with Edwin

Curley, Louis Loeb, and Michelle Kosch. Also, I have discussed aspects of this
book and been helped by countless scholars at these and other institutions,
most prominently among those I remember: Kate Abramson, Julia Borcherd-
ing, Rachel Cohon, Remy Debes, Richard Dees, Aaron Garrett, Michael Gill,
Charles Griswold, Paul Guyer, Thomas Hill Jr., T. H. Irwin, Christine Kors-
gaard, Matthew Leisinger, Sharon Lloyd, Tito Magri, Geoffrey Sayre-McCord,
Tamar Schapiro, Kelley Schiffman, Susanne Sreedhar, Nicholas Sturgeon, and
Allen Wood.
I am grateful more than I can express to those to whom I am closest.
Julian has listened to me talk about figures in this volume since he was
six and suggested as a title for my British moralists book Philosophy in
the Minds of Ten. I always learn from our conversations. I have discussed
philosophy and its history with Will for almost as long as we have con-
versed. No one holds my feet to the fire more (especially when it comes to
Marx and Nietzsche, who will appear in the next volume). I owe the deci-
sion to include Spinoza to Will. Long conversations with Frank Marotta
fed my soul towards the end of writing this volume, and Sidney Phillips
helped keep me on an even keel throughout the writing. Finally, Laura has
sustained me throughout, sometimes challenging but always supporting –
invariably, just what I needed.
I owe many debts in connection with this book’s publication, most espe-
cially to Paul Guyer and Gary Hatfield who first convinced me to write it over
twenty years ago. Terence Moore was then my editor at Cambridge Univer-
sity Press. Both he and Hilary Gaskin since have provided very helpful advice,
as did also an anonymous reader for the Press. For inestimable help in copy-
editing and other editorial assistance at late stages, I am very grateful to Molly
Montgomery.
Finally, material from earlier publications is included here, and I am grateful
to the publishers listed below:

“Grotius at the Creation of Modern Moral Philosophy,” Archiv für Geschichte

der Philosophie 94 (2012): 94–125. Reprinted with the permission of Walter
de Gruyter.
“Pufendorf on Morality, Sociability, and Moral Powers,” The Journal of the His-
tory of Philosophy 50 (2012): 213–38. Reprinted with the permission of The
Journal of the History of Philosophy.
The British Moralists and the Internal ‘Ought’: 1640–1740. Reprinted with the
permission of Cambridge University Press.

Published online by Cambridge University Press

 u

Introduction

Anscombe’s Challenge
“Modern Moral Philosophy” is the title Elizabeth Anscombe gave to her famous
broadside against the ethical philosophy she found around her in the middle of
the twentieth century (Anscombe 1958).1 Anscombe’s chief complaint was that
her contemporaries remained under the influence of a fundamentally mistaken
turn that Western ethical thought had taken in the modern period, beginning
roughly, I will suggest, with Hugo Grotius in the seventeenth century.2
Anscombe’s most basic problem with modern ethical philosophy was that it
is, as she put it, “moral philosophy.” It concerns, even if not exclusively, what it
calls morality: a set of putatively universal deontic or “juridical” norms of right
and wrong that purport to obligate all normal human adults, indeed, all moral
agents, as it terms beings who have the capacity to respond to obligation (5).3
Anscombe called this a “law conception” of ethics (5), since it posits a moral
law of right and wrong, what Grotius and his followers called “natural law.”
All societies are structured by laws or social norms (mores) in some way, of
course. But what modern ethical philosophers called and still call “morality”
transcends socially constituted norms. Of any social or legal obligation, we can
always ask whether it obligates us morally, that is, whether it would be morally
wrong, and not just against social convention or illegal, or even our society’s
moral beliefs, to violate it. Moral right and wrong can never be settled by look-
ing simply to a society’s laws or mores. They concern morality’s norms.

1
Thomas Nagel also describes Anscombe’s article as a “broadside” (Nagel 2022).
2
Anscombe had other concerns also. Partly, her target was the “consequentialism” that char-
acterized, she said, “every single English academic moral philosopher since Sidgwick” (1958:
10). By “consequentialism,” she meant views according to which sufficiently extreme con-
sequences might justify otherwise immoral acts. By this standard, even deontologists like
W. D. Ross count as consequentialists. “Consequentialism” has come to have a narrower use,
referring to theories of the right that are ultimately based on the goodness of consequences
(either directly, of actions, or indirectly, of rules, practices, or motives that dictate them).
3
Except when context requires it, I shall use these terms more or less synonymously, although
strictly speaking, “deontic” terms express concepts that do not require actual socially con-
stituted accountability structures to be instantiated, although “juridical” terms do.

https://doi.org/10.1017/9781139025065.002 Published online by Cambridge University Press

 2 introduction

This is not always apparent, since we use “morality” both as a count noun
and as a non-count noun. In the count-noun sense there are as many morali-
ties as there are societies with different mores (and perhaps different individual
moral codes). Moralities, in this sense, are things we can count in the actual
world. “Morality” in the sense that modern moral philosophers are concerned
with, however, is not a countable aspect of actuality. It is an essentially norma-
tive deontic structure.4 When modern moral philosophers use “morality” to
refer to this normative structure, they are using it in a non-count sense that
contrasts with the count-noun sense of “morality.”
Moreover, what modern philosophers call “moral agents,” those who are
subject to morality, is determined not by membership in any actual society,
but by having certain capacities of thought and will that philosophers dub
“moral agency.” These include the capacity to guide themselves by the moral
law, which binds each simply as one moral agent among others. “Moral agent,”
or “Person,” as Locke puts it, “is a Forensick Term” that “belongs only to intel-
ligent Agents capable of a Law” (Locke 1975: 346).
Anscombe thought it obvious, however, that no practical law of any kind can
exist without a lawgiver. So there can be a moral law only if it has a legislator
whose jurisdiction transcends any posited earthly realm, even, indeed, that of
the international law (jus gentium) that Grotius himself helped to originate in
the early modern period.5 Anscombe concluded that there can be such a thing
as morality only if it is legislated by God.
The idea that morality binds only because it is divinely legislated has certainly
been represented in modern moral philosophy, for example, in seventeenth-
century natural lawyers like Pufendorf and Locke. But it has been a decidedly
minority view. Most modern philosophers have been content to employ the
essentially juridical concept of morality without any such grounding in divine
legislation.6 Grotius is an excellent example. And many deny that morality is
even the kind of thing that could be legislated, even by God. It is not some actual
thing that was made actual by creation or legislation. It is an essentially norma-
tive deontic structure.

4
This also can also be confusing, since “normative” is often used, mostly outside of philoso-
phy, to refer to actual social norms or to what is “normal” by their lights. For a helpful
discussion, see Parfit’s distinction between “rule-involving” and “reason-involving” con-
ceptions of normativity (2011: I:144–148).
5
Grotius’s De Jure Belli Ac Pacis (The Rights of War and Peace) was published in 1625
(Grotius 2005).
6
And even when philosophers like Pufendorf take this position, they often feel philosophi-
cal pressure to argue that the reasons that are distinctively tied to moral obligations are not
eudaimonistic or egoistic. On this aspect of the modern view, see the next section. This is a
crucial distinction between medieval theological voluntarists, like Ockham and Scotus, and
modern ones. I am indebted to an anonymous reviewer for the Press for asking me to clarify
this.

https://doi.org/10.1017/9781139025065.002 Published online by Cambridge University Press

 anscombe’s challenge 3

Anscombe thought that this left modern deontic concepts with no “discern-
ible content except a certain compelling [‘psychological’] force” (Anscombe
1958: 18). Deontic moral concepts certainly do not present as psychological
concepts, however. How could they present as such and have the distinctive
normative purport that has been thought to be morality’s hallmark? We take
morality to purport to be essentially normative, to provide us with normative
reasons for acting.7 Indeed, we take morality to purport to obligate us in an
essentially normative way, to make violations not just unwise or imprudent but
morally wrong: things we are accountable for not doing and for which we incur
culpability and guilt when we do them without excuse. How could anything
like that follow from psychological facts alone?
So powerful is this idea and so insidious is the moderns’ error, Anscombe
thought, that the concept of moral obligation is often invoked as a necessary
ground even for divine law itself. In order for God’s laws to obligate, it is said –
for example, by the seventeenth-century Cambridge Platonist Ralph Cud-
worth – we must be obligated to follow God’s legislation independently of his
legislative act.8 Any moral obligation to comply with God’s commands cannot,
Cudworth argued, itself be created by his commands. If Anscombe is right,
however, the very idea of such a ­legislation-independent obligation is funda-
mentally confused. Philosophers who employ it end up deploying a contentless
concept having only psychological force.
“Anscombe’s Challenge,” as we can call it, indicts much of the thought that
ethical philosophers have produced in the West since the early seventeenth
century. As we shall see, Anscombe is unquestionably right that a central fea-
ture of “modern moral philosophy” has been its fundamentally deontic or
juridical character. Modern moral philosophers have indeed been concerned
to articulate, understand, defend, and attempt to ground morality with its dis-
tinctively obligating normativity.
Even modern critics of the idea of morality – most prominently, Nietzsche,
but also more qualified critics like Bernard Williams – have focused on the
deontic features that Anscombe identifies. In seeking to overturn or at least
rein in morality, they have implicitly accepted her point that deontic morality
has been a, if not the, central focus of modern ethical thought (Nietzsche 2007;
Williams 1985: 1–4, 174–196).

7
Even those who deny that morality’s imperatives are “categorical” and reason-giving for
every person do not contest that they purport to be (see, e.g., Foot 1972).
8
“It was never heard of, that any one founded all his Authority of Commanding others, and
others’ Obligation or Duty to Obey his Commands, in a Law of his own making, that men
should be Required, Obliged, or Bound to Obey him” (Cudworth 1996: I.ii.3). For a dis-
cussion of Cudworth’s claim, which, in effect, argues that it begs the question against the
Anscombean view, see Schroeder 2005a.

https://doi.org/10.1017/9781139025065.002 Published online by Cambridge University Press

 4 introduction

To appreciate the distinctiveness of this conception of deontic morality,

compare it to the view of ethics one finds in Plato and Aristotle. Notably, the
term “morality” does not even appear in standard English translations of Pla-
to’s Republic or Aristotle’s Nicomachean Ethics, though it sometimes does in
commentaries on these texts.9 For Plato and Aristotle, the central ethical con-
cepts all concern species of the good: virtue, intrinsically good or noble (kalon)
action, and the good or benefit (eudaimonia) of human beings. Much of what
modern ethical philosophers consider under the heading of morality, Plato and
Aristotle discuss under the virtue of justice or intrinsically good just action.
When Socrates is challenged by Glaucon and Adeimantus at the beginning
of Plato’s Republic to say why we should be just, his reply is not that justice
is morally obligatory or that others’ rights provide, in themselves, reasons to
respect them. Rather, Plato has Socrates argue that justice is both instrumen-
tally and intrinsically good for the just person.10
But consider how a modern like H. A. Prichard responded to arguments like
this near the beginning of the twentieth century. Prichard maintains that such
arguments “rest” “moral philosophy … on a mistake” (Prichard 1912: 21–37).
That complying with the moral law can benefit us, even intrinsically, is an
important fact; but it is not one, Prichard argues, that can either explain why
morality obligates us or establish any reason for being moral that might flow
directly from its doing so. For these tasks, Socrates’s argument provides a rea-
son of the wrong kind.

Sidgwick’s Contrast
The great nineteenth-century moral philosopher Henry Sidgwick made a simi-
lar point when he wrote that according to “the Greek schools” of ethics,
Right action is commonly regarded as only a species of the Good. … Their
speculations can scarcely be understood by us unless with a certain effort
we throw the quasi-jural notions of modern ethics aside, and ask (as they
did) not “What is Duty and what is its ground?” but “Which of the objects
that men think good is truly Good or the Highest Good?” (Sidgwick 1967:
105–106)11

9
This can be confirmed by a search of electronically available translations, such as W. D.
Ross’s and Terence Irwin’s translations of Aristotle’s Nicomachean Ethics, or Benjamin
Jowett’s or Paul Shorey’s translations of Plato’s Republic. “Moral” appears in Aristotle, of
course, for example, in his contrast between theoretical and moral virtue. But there it sim-
ply means virtues that are connected to character and the will.
10
On this point and Prichard’s response, discussed presently, see Brown 2007.
11
John Rawls’s Lectures on the History of Moral Philosophy, which aims to cover “Modern
Moral Philosophy: 1600–1800,” begins with an invocation and brief discussion of
Sidgwick’s contrast (Rawls 2000: 1–2). In my discussion of Sidgwick in the companion

https://doi.org/10.1017/9781139025065.002 Published online by Cambridge University Press

 sidgwick’s contrast 5

As Anscombe would later, Sidgwick argued that modern moral philosophy dif-
fers from the ethical thought of the ancient Greeks in viewing the “quasi-jural”
or deontic notion of moral duty or right as distinct from any “species of the
Good.”
Sidgwick drew a further, related contrast between ancient and modern eth-
ics concerning what philosophers these days call “normativity” or the force of
normative reasons or oughts:
[I]n Greek moral philosophy generally, but one regulative and governing
faculty is recognised under the name of Reason …; in the modern ethical
view, when it has worked itself clear, there are found to be two, — Universal
Reason and Egoistic Reason, or Conscience and Self-love (1964: 198).12

Plato, Aristotle, and other ancient thinkers tend to be eudaimonists, holding

that all normative reasons for action must derive from the agent’s own good
or happiness (eudaimonia) broadly conceived, even when it comes to virtue
and justice.13 In Socrates’s exchange with Glaucon and Adeimantus, it is sim-
ply assumed by all parties that if Socrates cannot establish that it is intrinsi-
cally or extrinsically beneficial to the just person to be just, he will not have
shown any reason for them to act justly rather than, say, simply to appear to
be doing so.
According to Sidgwick, however, “the modern ethical view” allows for the
position that Prichard clearly assumes, namely, that the fact that an action
would unjustly wrong someone and therefore be morally wrong is or entails
in itself sufficient reason not to do it. “Conscience,” the mental power through
which we make moral judgments, can be a “regulative and governing faculty”
in its own right, an aspect of practical reason itself.14 According to Sidgwick’s
moderns, deontic morality can provide an independent source of normative
reasons and oughts that are additional to those provided by any species of the
good. As we shall see, even modern eudaimonists or rational egoists who hold
that all reasons for acting must come from (the agent’s) good, like Locke, none-
theless often maintain that moral right and obligation are independent ethical
ideas that cannot be reduced to any “species of the good.”

volume to this one, I will discuss how Sidgwick himself uses deontic terms in a broader
sense than the quasi-jural, accountability-entailing sense. In this, he is followed by Broad
and Ross. Ewing notes the distinction between these in Ewing 2012a. For discussion, see
Hurka, 2014. I am indebted to an anonymous reviewer for the Press for requesting clarifi-
cation here.
12
For an excellent discussion of this passage, see Frankena 1992. For a defense of a “reason-
implying” conception of normativity, see Parfit 2011: I:144–148.
13
The Stoics complicate this picture, though Irwin characterizes them as eudaimonists in
Irwin 2003.
14
For an excellent history of the idea of conscience, see Sorabji 2014.

https://doi.org/10.1017/9781139025065.002 Published online by Cambridge University Press

 6 introduction

There are two separate but related aspects to what we can call “Sidgwick’s
Contrast.” First, many moderns claim that the deontic concept of moral right
or obligation is irreducible to any species of the good. And second, an impor-
tant strain of modern ethical philosophy, at least, holds that the right has a
normative practical force that is additional to that of the good.

Modern Moral Philosophy’s Shadow I

Obviously, I have chosen my title with Anscombe’s essay in mind. I agree with
Anscombe, Sidgwick, and others (most notably, John Rawls and J. B. Schnee-
wind) that the “quasi-jural” or “law conception” of morality has been at the
very heart of ethical philosophy in the West, roughly from Grotius on (Rawls
2000; Schneewind 1998). Of course, not all modern ethical philosophy can be
categorized as moral philosophy in Anscombe’s sense. Just to take one example,
G. E. Moore famously argued in Principia Ethica (Moore 1993) that the concept
of intrinsic good (though not that of a person’s good) is the single fundamental
ethical concept, an ineliminable kernel of any ethical concept.
About the concept of moral right, Moore there says that “to assert that a
certain line of conduct is, at a given time, absolutely right or obligatory is obvi-
ously to assert that more good or less evil will exist in the world, if it be adopted
than if anything else be done instead” (Moore 1993: 77). Clearly, there is noth-
ing irreducibly juridical or deontic about the notions expressed by “right” and
“obligatory” as Moore defines them in Principia. If what it means to say that
an act is morally right or obligatory is just that it produces the most good, then
the claim that it is morally right or obligatory to produce the most good is not
a deontic normative claim. It is what Parfit calls a “concealed tautology” (Parfit
2011: II:276).
In using deontic moral terms like “right” and “obligatory” to express his
views, however, Moore is arguing against and arguably trying to co-opt more
orthodox deontologically minded moral philosophers of his time like Prichard.
Thus, even though Moore’s philosophical project is not itself moral philosophy
in Anscombe’s sense, his use of deontic moral terms is testimony to modern
moral philosophy’s powerful influence.
Even more obviously, if anyone counts as a modern ethical philosopher,
Nietzsche surely does, though he is hardly a moral philosopher in Anscombe’s
sense. Even so, the idea of morality unquestionably looms large in Nietzsche’s
thought. On the Genealogy of Morality is devoted to understanding what
Nietzsche argues to be the concept’s questionable origins and to a critique of
“the value of these [i.e., morality’s] values” (Nietzsche 2007: 7).
Thus, even when ethical philosophers in the West from the seventeenth cen-
tury on have not been doing moral philosophy, strictly so called, their thought
has often been shaped by it. Either their terms bear the influence of modern
moral philosophy, as with Moore; their projects are defined in opposition to it,

https://doi.org/10.1017/9781139025065.002 Published online by Cambridge University Press

 modern metaethics 7

as with Nietzsche; or their thought is formulated and received against modern

moral philosophy’s background and in its shadow.

Modern Metaethics
My aim in this book and the one to follow is to investigate central aspects of
Western ethical philosophy from the publication of Hugo Grotius’s The Rights
of War and Peace in 1625 through the end of the twentieth century, using Ans-
combe’s and Sidgwick’s characterizations as reference points. Mostly I will be
concerned with philosophers’ attempts to understand, articulate, defend, and
ground morality, as well as to grapple with the distinctive metaphysical and
epistemological questions that arise in relation to it, metaethical questions, as
we now call them. Are there moral truths, and if so, what makes them true?
What is the relation between any metaphysical basis morality might require
and that needed by the natural sciences that began to take a recognizably
modern shape in the seventeenth century? Has modern moral philosophy the
resources to respond adequately to Anscombe’s Challenge and to Nietzsche’s
and his followers’ critiques? On the epistemological side: can we acquire moral
knowledge? And if so, how might that be related to moral motivation?
Although philosophers since the ancient Greeks have concerned themselves
with fundamental metaphysical and epistemological questions about ethics,
metaethics as a subfield of ethical philosophy did not exist before the modern
era, arguably, not before the twentieth century.15 It was only in the aftermath of
Moore’s Principia Ethica that philosophers began to distinguish and focus on
so-called second-order issues about ethical language and concepts and their
relation to questions in the philosophy of language and mind, metaphysics, and
epistemology as a distinct area of inquiry, separable from “first-order” “norma-
tive ethics” (Mackie 1977).
Surely, one aspect of the modern period that stimulated metaethical thought
was the decline of Aristotelian teleological metaphysics alongside the rise of
modern science. If nature is not itself teleological, this forces the question of
what place value can have “in a world of fact” (Köhler 1938). Another factor
is that metaethical questions sometimes turn on technical issues in logic and
the philosophies of language and mind, whose tools required the increasing
specialization of the twentieth century to be developed sufficiently to pursue
metaethics in a focused, fruitful way.
I speculate that a further important factor that led to metaethical reflection
in the modern period, however, concerns Sidgwick’s Contrast. So long as there
is believed to be only a single fundamental ethical concept, the good, the ques-
tion need not arise of what makes something an ethical or normative concept

15
A search of JSTOR reveals “metaethics” first being used in its contemporary sense in
Wisdom 1948.

https://doi.org/10.1017/9781139025065.002 Published online by Cambridge University Press

 8 introduction

in general, or of what the “sources of normativity” are (Korsgaard 1996a). No

distinction between evaluative and normative concepts and questions seems
possible. Once, however, there are thought to be two fundamental ethical
notions, the good and the right, these more general and fundamental metaethi-
cal questions begin to seem unavoidable.16
Moreover, once it is believed, for example, by the early modern natural law-
yer Francisco Suárez, that the deontic moral concept of right has a distinctive
normative profile that differs from that of the good – that the morally right
obligates whereas the good recommends – these philosophical questions can
seem even more urgent. As we shall see, Suárez makes a fundamental distinc-
tion between law and counsel, the influence of which extends through Gro-
tius, Hobbes, Kant, and the moral philosophy that follows them.17 This focuses
Prichard’s question and also opens up philosophical space for thinkers like
Nietzsche to argue that even if the good can be adequately grounded, the puta-
tively obligatory normativity of morality (the right) cannot be.
In addition to these metaethical issues, modern moral philosophers have
also grappled with substantive normative questions both at the level of nor-
mative moral theory and in thinking about specific cases. The familiar dispute
between deontology and consequentialism is itself a creature of the modern
period, requiring morality’s hallmark deontic categories even to be formulated.
It is worth bearing in mind that though it begins with a view about the good,
consequentialism is itself a view about the good’s relation to the right. Mill
begins Utilitarianism, indeed, by saying that nothing is “more significant of the
backward state” of ethical knowledge than the “little progress” that has been
made in determining “the criterion of right and wrong” (Mill 2002: I:1). It is to
this matter of deontic morality that utilitarianism and consequentialism more
generally have historically been addressed.

Modern Moral Philosophy’s Shadow II

Modern philosophy about morality will not, however, be my sole focus. I seek
also to understand and situate aspects of modern ethical philosophy that are
not moral philosophy in Anscombe’s sense. Some of these concern traditional
questions about the good – both, what kind of life is best for human beings,

16
It was only toward the end of the twentieth century when Gibbard introduced the idea
that there is a single normative concept, which can be expressed equivalently by “ought”
or by “normative reason” (in Gibbard’s terms, what “makes sense”), that metaethics came
to be regarded as a species of the larger genus of metanormative theory, which concerns
normativity more generally, including in, for example, normative epistemology (1990). As
we shall see, Gibbard follows a strain of thought that arguably begins with Sidgwick’s idea
that ought is the fundamental ethical concept (Sidgwick 1967: 23–38).
17
Suárez was not the first to draw this distinction. On this see Preface, n.3.

https://doi.org/10.1017/9781139025065.002 Published online by Cambridge University Press

 modern moral philosophy’s shadow ii 9

and, what is intrinsically choiceworthy – questions that have been a staple of

ethical philosophizing since the ancient Greeks. But much such philosophy in
the modern period, I shall be arguing, must nonetheless be understood in rela-
tion to modern moral philosophy, even when its aims are profoundly different.
Sometimes this is because, as with Nietzsche and Williams, ethical conceptions
are put forward in opposition to morality, as a replacement for or at least as a
counterbalance to it. And sometimes, as with Moore, a philosopher’s project
may not itself concern deontic morality, but still be formulated in moral phi-
losophy’s distinctive deontic terms.
Different varieties of modern virtue ethics provide a particularly interest-
ing example. Some virtue ethicists, like Francis Hutcheson in the seventeenth
century and Rosalind Hursthouse in the late twentieth, put forward their the-
ories in a moral philosophical idiom. Hutcheson is concerned with what he
calls “moral goodness,” which he distinguishes from “natural goodness.” And
Hursthouse draws on an account of moral virtue to ground a theory of what
makes actions morally right or wrong (Hutcheson 2004; Hursthouse 1999). But
modern virtue ethicists also often advance their virtue theories in opposition to
deontic moral theories. Hume is a leading example. And a major aim of Ans-
combe’s “Modern Moral Philosophy” in 1958, after all, was to wean her readers
from that subject and return them to a virtue-centered Aristotelian approach.
As it happened, the 1960s and 1970s intervened, with movements for social
and political equality that made issues of justice and rights inescapable, form-
ing a backdrop for Rawls’s monumental A Theory of Justice, which commanded
the moral philosophical scene through the 1970s (Rawls 1971). This required
deontic moral philosophy, since there can be no moral rights without moral
duties.
It was only in the 1980s, in a very different sociopolitical climate, that the
revival of virtue ethics for which Anscombe called began to take place, fre-
quently formulated in non-deontic terms, as by writers like Alasdair MacIntyre
and Annette Baier, and sometimes put forward in opposition to morality, as
in Michael Slote’s From Morality to Virtue (MacIntyre 1981; Baier 1985; Slote
1992).18

18
Another precursor was Philippa Foot, whose landmark papers, “Moral Arguments” and
“Moral Beliefs,” appeared roughly contemporaneously with Anscombe’s (1958 and 1959,
respectively), and whose Virtues and Vices (Foot 1978), appeared just before the heyday
of virtue of ethics in the 1980s. In effect, Foot worked out both the metaethics and norma-
tive ethics of the kind of neo-Aristotelian virtue ethics for which Anscombe called. And
she expressed her skepticism of orthodox moral philosophy in “Morality as a System of
Hypothetical Imperatives” (1972).
Baier’s work was less concerned with the systematic development of virtue ethics than
with defending a broadly Humean virtue approach against more orthodox normative
moral theory (see, e.g., “Doing Without Moral Theory?” in Baier 1985).

https://doi.org/10.1017/9781139025065.002 Published online by Cambridge University Press

 10 introduction

The Case of Kant

Perhaps the most fascinating example of a philosopher whose thinking is
shaped, and in some ways arguably distorted, by the modern moral philosophi-
cal frame, is Kant. On the face of it, Kant’s ethics can seem a paradigm of moral
philosophy in the modern idiom, a textbook example of deontological ethics
and an archetypical theory of the “moral law.” I shall argue, however, that on
closer inspection the deontic categories of duty, obligation, and the moral law
are, for Kant, only the shadow for finite rational beings of how a fully ratio-
nal being would choose and act. Fully rational deliberation, according to Kant,
always only involves questions of the good (albeit the intrinsically choicewor-
thy rather than agents’ good – das Gute vs das Wohl) and not deontic moral
questions.
This means that duty, obligation, and the moral law have no independent
normative force for Kant. Kant’s response to Anscombe’s Challenge is that it
is reason that commands us finite rational beings to do what it is best (what
there is most reason) for us to do, hence what a fully rational being would do in
our place. But two questions then arise. What can give reason this distinctively
deontic power, if it is itself concerned exclusively with the good? And how can
reason create any further reason to do something by commanding what it has
already determined there is independently reason to do? Any such deontic
aspect would seem to be normatively epiphenomenal. And if that is so, moral-
ity itself may end up being epiphenomenal on Kant’s view also.
Despite this, Kant’s emphasis on freedom and what he calls “autonomy” is
unquestionably modern and unlike anything to be found among the ancient
Greeks or in ethical philosophy of the medieval period that derives from them.
Moreover, I shall argue that the emphasis on the distinctive freedom of a delib-
erating rational agent in modern philosophers like Cudworth, Locke, and
Samuel Clarke even before Kant, as well as in Kant himself, derives from their
taking it to be necessary for the very possibility of morality and moral obliga-
tion. And this surely is an essential feature of philosophers who follow in Kant’s
wake, most obviously, Fichte and Hegel, but also Kierkegaard and, arguably,
even Nietzsche.

(Relatively) Modest Aspirations

I do not aspire to anything like a comprehensive treatment of modern ethics in
this volume. Even were I capable of writing such a book, we already have in T.
H. Irwin’s magisterial, three-volume The Development of Ethics a more widely
ranging and detailed treatment, even of just modern ethics, than perhaps any-
one else can or will be able to provide (Irwin 2007, 2008, 2009). Irwin devotes
almost all of his second and third volumes to the modern period, nearly nine-
teen hundred pages. I seek something significantly more modest.

https://doi.org/10.1017/9781139025065.002 Published online by Cambridge University Press

 (relatively) modest aspirations 11

There is a further difference between Irwin’s treatment and what I aim for
here. Irwin tends to downplay the significance of Anscombe’s and Sidgwick’s
contrasts and argues that much of the modern period carries forward a tra-
dition of ethical thought that is continuous with that of the ancient Greeks.
What Irwin calls Aristotelian naturalism is a combination of views he finds first
in Aristotle, but which he argues carries through the ancient Stoics, notably
Cicero, Aquinas’s classical theory of natural law, and, in Irwin’s view, many
modern figures from Grotius on, including Grotius himself.
Aristotelian naturalists, according to Irwin, hold a teleological view of practi-
cal reason; they maintain that rational action aims always and only at the good.
They are also eudaimonists, holding that the final rational end is the agent’s
own good or eudaimonia. They thus deny that deontic standards of moral right
or obligation can provide reasons for acting that are independent of the agent’s
good. As Irwin emphasizes, this does not mean that Aristotelian naturalists
cannot recognize distinctively moral virtues or intrinsically good moral acts
that differ from other nonmoral virtues or noble action. Like Cicero, they can
recognize a category of duty (officium) or right action (honestum) and hold,
moreover, that there is reason to conform to moral duty or right even when it is
not advantageous in the sense of being instrumentally beneficial (commodium
or utile), because it is intrinsically beneficial.
Nevertheless, for Aristotelian naturalists like Cicero and Aquinas, duty and
right are not identified independently of the good; they rather concern a spe-
cies of good, the common good. Also, any reason that agents have to comply
with their moral duty depends on their own good or eudaimonia, albeit by vir-
tue of compliance’s being intrinsically, rather than just instrumentally, good
for them. There is, as Cicero puts it, a bonum honestum, an intrinsic benefit in
being moral and just, as indeed Socrates argues in The Republic and as Aristotle
implies in the Nicomachean Ethics when he maintains that eudaimonia consists
in virtuous (including just) activity (Irwin 2007: 620, 2008: 31–32).
Finally, Irwin calls this position Aristotelian naturalism since it maintains
that an intrinsically beneficial, virtuous life is one that best realizes our rational
human nature. This means that when Irwin classifies a modern moral theorist
like Grotius as an Aristotelian naturalist, he is interpreting him as holding that
morality’s normative force derives from the fact that moral conduct realizes the
agent’s good by virtue of realizing his rational human nature. On an Aristote-
lian naturalist view, morality can have no normative force that is independent
of the agent’s good; nothing can. It can provide, in Suárez’s terms, only counsel
and not genuinely obligating law.
Irwin writes at the beginning of The Development of Ethics that were he
to give his volumes an “ampler title” on the model of some seventeenth- or
eighteenth-century works, he might have chosen The Development of Ethics:
being a selective historical study of moral philosophy in the Socratic tradition
with special attention to Aristotelian naturalism (2007: 1).

https://doi.org/10.1017/9781139025065.002 Published online by Cambridge University Press

 12 introduction

I agree that viewing the history of ethics, including that of the modern
period, through the lens of Aristotelian naturalism provides an interesting
and insightful interpretative prospect. But it is far from the only one. The his-
tory of modern ethics presented in this book will be guided by a very different
viewpoint, one that, in my view, Irwin’s Development insufficiently appreciates.
The thread that will run throughout these volumes is that ethics in the mod-
ern period is best appreciated in relation to moral philosophy conceived in the
deontic terms Anscombe identifies, through the lens of Sidgwick’s claim that
much modern ethical thought recognizes the morally right as having an inde-
pendent normative force that is irreducible to that of the good.
I would like to say something in conclusion here, however inadequate, about
race, sex, and gender as these enter into our history. No one can reasonably
doubt that the larger culture that produced and supported philosophy and
its publication during the modern period was, and remains, sexist and racist,
patriarchal and white supremacist. It is not at all surprising, therefore, not just
that so little published work was produced by women or people of color, but
also that so many philosophers, including many here discussed, were them-
selves racist, sexist, or both. At least one philosopher who will be especially
central to our story, Kant, himself produced a theory of races (Kant 2013b).
There is really no way to deal adequately with issues of race, sex, and gender,
and other forms of epistemic and, indeed, philosophical oppression in a his-
tory such as this. I agree with Lucy Allais and Dilek Huseyinzadegan that, for
example, Kant’s racism is not a sufficient reason not to study his works (Allais
2016; Huseyinzadegan 2018). Kant’s ideas have not simply had great impor-
tance for philosophy; his doctrine of the equal dignity of persons has also had
special significance in movements to resist racist and sexual oppression.19 Still,

19
It is difficult to believe, for example, that it was simply a coincidence that in the years just
following the civil rights movement, three important works of moral and political phi-
losophy appeared in quick succession – Rawls’s A Theory of Justice, Wolff’s In Defense of
Anarchism, and Nozick’s Anarchy, State, and Utopia – all of which appealed to Kant’s doc-
trine of equal dignity in their foundations (Rawls 1971; Wolff 1970; Nozick 1974). Hegel
says, no doubt rightly, that philosophy’s “owl of Minerva begins its flight only at the onset
of dusk” only after progressive social and political changes have taken place. But even if
Rawls’s, Wolff’s, and Nozick’s Kantian philosophizing played no causal role in the civil
rights movement, it seems pretty likely that Kant’s notion of the equal dignity of persons
played an important background role in helping to create the intellectual political culture
that made the civil rights movement possible. That the idea had become undeniable in the
1960s was surely part of Kant’s legacy.
Another important reason to study figures like Kant, including their most objectionable
writings, is, as Allais argues, to help us better understand how someone like Kant could
come to write what he did, so that we, who are no less subject to racist and sexist biases our-
selves, might avoid similar errors of our own (Allais 2016). And the same arguably holds for
other important figures we will be considering. As important as it is to pursue the genea-
logical and self-critical work Allais calls us to, however, this is not work we shall do here.

https://doi.org/10.1017/9781139025065.002 Published online by Cambridge University Press

 (relatively) modest aspirations 13

some of what Kant says certainly bears the imprint of his racial, sexual, class,
and Eurocentric position, and, as Allais and Huseyinzadegan point out, study-
ing his works in context provides an opportunity for us to think about how our
own moral and philosophical thought and practice bear those marks as well
(Allais 2016; Huseyinzadegan 2018). We shall have occasion to return to this
theme at the end of this volume’s final chapter, which is on Rousseau and Kant.
There will be many more men discussed in the following pages than women,
as well as more white philosophers than philosophers of color.20 Although
women philosophers of the early modern period have begun to be more closely
studied, many, like Mary Astell, Margaret Cavendish, Catharine Trotter Cock-
burn, Anne Conway, Mary Shepherd, and Princess Elisabeth of Bohemia, are
more notable for their metaphysical and epistemological views than for their
ethical philosophy (Atherton 1994).21 It is only in the second half of twentieth
century, beginning in the 1950s and 1960s, that women came to be widely rec-
ognized as important ethical philosophers. Anscombe’s Challenge in “Modern
Moral Philosophy” (in 1957) frames our study, albeit retrospectively. And the
post–World War II period also features Foot, Beauvoir, and Murdoch, all of
whom produced ethical philosophy of enduring value.22 Their works and those
of numerous later prominent women philosophers, like Hursthouse, (Annette)
Baier, Wolf, and Korsgaard will be central to the closing chapters of the volume
to follow.23
Finally, that volume will conclude with a chapter titled “Theorizing Oppres-
sion and Nonideal Theory,” which will discuss a steadily growing movement
by academic philosophers in the latter third of the twentieth century and early
twenty-first attempting to come to terms with systematic sexual and racial
oppression. This begins with philosophical responses to the civil rights move-
ment and continues with discussions of affirmative action and other attempts

20
The latter will begin to be discussed in the projected volume to follow the present one.
21
Damaris Masham, who did have interesting views on love and friendship, will enter into
our story because of the link she provides between Cudworth and Locke on free will. Some
of her most interesting thoughts occur in her correspondence with Leibniz on metaphysi-
cal topics.
22
See Benjamin Lipscomb’s The Women Are Up to Something for a fascinating account
of Anscombe, Foot, Murdoch, and Mary Midgley’s philosophical lives, which began
when they found themselves together as students at Oxford just before World War II
(Lipscombe 2021).
23
This may be an appropriate place to explain my pronoun use in this work. Following the
well-argued proposal of Dembroff and Wodak, I shall use “they” for both singular and
plural neutral pronouns (except within quotations) (Dembroff and Wodak 2018). Happily,
this practice is becoming increasingly widespread. I shall not, however, pursue this practice
inside quotations, except infrequently when context requires it. On reflection, it seems wis-
est to allow the philosophers I am quoting to speak for themselves. I am indebted here to
counsel from Michael Della Rocca, Molly Montgomery, Laura Radwell, and Ken Winkler.

https://doi.org/10.1017/9781139025065.002 Published online by Cambridge University Press

 14 introduction

to address past racial and gender injustice and to eliminate it in the present
(see, e.g., Boxill 1984, 2003). As formal and informal barriers to entry to aca-
demic philosophy began to be removed for women and philosophers of color,
the topics that are considered to be moral philosophical topics of significance
have changed accordingly.
The most notable development has been the pursuit of what has come to be
called, following Charles Mills, “nonideal” moral and political theory, which
attempts to theorize justice, not under ideal conditions, but in circumstances of
historical and ongoing oppression, most especially the kind of white suprem-
acy that has characterized the United States (Mills 1997, 2005). Although
the terms “ideal” and “nonideal theory” come from Rawls, Mills has been
the founding and central figure of nonideal theory, showing philosophers the
need for moral- and political-philosophical accounts of urgent injustice that
Rawls’s ideal theory leaves out. Also important have been Elizabeth Anderson
and Tommie Shelby (Anderson 2010, 2014; Shelby 2005, 2016). Nonideal moral
and political theories have been pursued as well in fields like epistemology, as
in Miranda Fricker’s pathbreaking work on epistemic injustice (Fricker 2007).
All of this will be discussed at the end of the volume to follow the current one,
which will take us from Fichte and Hegel through the end of a (long) twentieth
century. The present volume, however, ends with Kant and begins with a moral
philosopher who is justifiably regarded as the founding figure of modern moral
philosophy, Hugo Grotius. To him we now turn.

https://doi.org/10.1017/9781139025065.002 Published online by Cambridge University Press

 1

Grotius

Courses in the history of modern philosophy generally focus on ­epistemology

and metaphysics and begin with Descartes’s engagement with skepticism in the
Meditations. But where should a history of modern ethical thought begin? It
has been widely believed, not least by early moderns themselves, that in the sev-
enteenth century, ethical philosophy came to assume a self-consciously “mod-
ern” form, opposing itself to “ancient” Greek ideas and to medieval scholastic
approaches that derived, at least in part, from Aristotle.
Many have thought that Hugo Grotius (1583–1645) helped initiate this mod-
ern form. In the mid-eighteenth century, Jean Barbeyrac’s Historical and Criti-
cal Account of the Science of Morality praised Grotius as “the first who broke
the ice” of “the Scholastic Philosophy; which [had] spread itself all over Europe”
(Barbeyrac 1749: 66, 67). More recently, Knud Haakonssen has written that
ethical philosophers of the time thought “that something decisively new hap-
pened with Grotius” (Haakonssen 1996: 15). To many thinkers of the early
modern period, Grotius’s The Rights of War and Peace (RWP), published in
1625, was not simply an original treatise on international law; it seemed to set a
new agenda in ethical and political philosophy across the board.1 A strong case
can be made, indeed, that the very concepts to which “morality” and “science”
in Barbeyrac’s title referred were themselves substantially shaped by Grotius
and his influence.2
So we begin our history of modern moral philosophy with Grotius, and in
this chapter I shall attempt to say why.3 This is a pressing question, since Irwin
maintains that Barbeyrac “misunderstands Grotius in representing him as a
pioneer” (Irwin 2008: 99). According to Irwin, Grotius follows the main lines of

1
De Jure Belli Ac Pacis is also often translated as On the Law of War and Peace.
2
For Barbeyrac, “morality” consists of a body of “rules” (“laws of morality”) we are “obligat[ed]”
to comply with just because we are “corporeal rational creature[s]” (Barbeyrac 1749: 2–3, 5). The
modern project Barbeyrac terms the “science of morality” is the task of articulating and defend-
ing these “rules,” including by providing some philosophical account of their distinctively oblig-
atory character. This, I shall argue, is the problematic that Grotius bequeathed to early modern
ethical philosophy, or at least, that many early moderns plausibly took to be his bequest.
3
I draw heavily on Darwall (2012a).

15

https://doi.org/10.1017/9781139025065.003 Published online by Cambridge University Press

 16 grotius

an “Aristotelian naturalism,” which also characterized classical Thomist natu-

ral law, with which Grotius’s early modern natural law is frequently contrasted.
If Irwin is right, however, there was nothing fundamentally original in Grotius.
What was importantly new in Grotius that helped produce a distinctively
“modern” ethical philosophy? Schneewind has argued that Grotius’s original-
ity consisted in his “confrontation” with skepticism, which “was simply not an
issue for the classical natural law theorists” who preceded him (Schneewind
1998: 71). If so, Grotius might stand to modern ethical philosophy in something
like the relation Descartes stands to modern epistemology. But although RWP
virtually begins with the skeptical challenge that justice is a mere “Chimera”
since “Nature prompts all men … to seek their own particular advantage,”
Irwin correctly notes that skeptical objections of this kind were well known
to the ancients, as Grotius’s putting it in the mouth of Carneades effectively
acknowledges (Grotius 2005: I, 79, Irwin 2008: 94–96).4 It is largely the same
challenge that Glaucon and Adeimantus pose to Socrates in the Republic: Why
should we be just when it is disadvantageous to be so? If, however, serious
engagement with skepticism was not new with Grotius, what was?
In this chapter, I shall argue that what was novel in Grotius is a combination
of three doctrines that will loom large throughout the modern period: (i) moral
right (natural law) has a distinctively obligating normativity that cannot be
reduced to normative reasons concerning the good that advise or “counsel”
action rather than demand it; (ii) morality includes basic universal claim rights,
what Grotius dubs “perfect rights” (thereby initiating the perfect/imperfect
right and correlative perfect/imperfect duty distinctions); and (iii) morality
would exist and obligate us “even if we were to suppose … that there is no
God” (Grotius 2005: I, xxiv).5 By affirming that morality has obligating author-
ity but not deriving it from God’s command, Grotius made himself liable to
Anscombe’s Challenge.

Classical Natural Law: Aquinas and Suárez

For Grotius, morality is embodied in what he calls “natural law.” But there is
nothing exclusively modern in the idea of natural law itself. It goes back to
the Stoics and was, in Aquinas’s classical natural law theory, part of scholastic
orthodoxy. What, if anything, was different about Grotius’s conception?

4
Because it is more generally available, I will be using the 2005 LibertyClassics edition
(Grotius 2005), edited by Richard Tuck. It was translated by John Morrice (Grotius 2005:
I: xxxv). In at least some places, which I will note below, Morrice appears to have worked
from Barbeyrac’s French translation, which is not entirely faithful to Grotius’s Latin text.
(Grotius 1724). When appropriate, I will supplement with Francis Kelsey’s translation
(Grotius 1925).
5
This passage occurred in the Prolegomena to the first edition of De Jure Belli.

https://doi.org/10.1017/9781139025065.003 Published online by Cambridge University Press

 classical natural law: aquinas and suárez 17

For Aquinas, laws of nature are teleological standards inherent in our ratio-
nal nature that concern our common good (Aquinas 1997: Q91–93). This is
familiar Aristotelian naturalist doctrine. But what makes these standards laws?
There is, of course, a general sense in which “law” can refer to any standard or
norm, but that is insufficient to distinguish Anscombe’s juridical notion. Not
just any standard involves obligation in the sense of something for which we
are answerable, where issues of culpability, guilt, and innocence are automati-
cally involved. If I believe the opposite of what is entailed by things I know, I
violate a standard of reason (so a standard inherent in my rational nature), but,
so far anyway, nothing juridical need be involved. No issue of culpability, guilt,
or innocence necessarily arises. From the fact that someone manifests some
mistake or fault of reasoning or errs in failing to respond to normative reasons,
it does not follow that they are to blame or that this is their fault in that distinc-
tive sense (Pink 2007). But this is precisely what is involved in the modern “law
conception” of morality Anscombe identifies. So the fact that an ethical con-
ception like Aquinas’s is advanced in legal terms is not enough to make it a “law
conception” in Anscombe’s sense.
Even before Grotius, Francisco Suárez (1548–1617) criticized Aquinas’s the-
ory precisely on the grounds that it could not explain why we would have any
obligation to follow natural law. Suárez agreed with Aquinas that, for example,
telling falsehoods is intrinsically “repugnant” to rational nature, but he pointed
out that this is insufficient to create any obligation to tell the truth (Suárez
1944: 181–183). To understand natural law as genuinely obligating law, Suárez
believed, it is necessary to see its dictates as authoritatively addressed demands
with which we are accountable for compliance. And this, he thought, requires
seeing natural law as issuing from God’s authoritative command.6 In effect,
Suárez held that Aquinas’s view was not subject to Anscombe’s Challenge
because it was not even a theory of obligating law.
Suárez crystallized his idea in a fundamental distinction he made between
law and “counsel,” the influence of which would be felt throughout the early
modern period. According to Suárez, law has a conceptual connection to obli-
gating demands and accountability that normative reasons recommending an
action as good (advisable) do not.
[C]ounsel is excluded from law. … The word promulgation implies an
order for the purpose of creating an obligation and it is in this respect most
of all that counsel differs from law [i.e., in not being “promulgated” and
obligation creating] (Quoted in Schneewind 1990: I.74–75).7

6
Here I draw on Darwall (2003).
7
From De Legibus, first published in 1612. Jean Bodin made a related distinction between “com-
mand” and “counsel” in The Six Books of the Republic, published in 1577 (2004: 119). For dis-
cussion of Bodin’s distinction and its role in early modern British thought, see Paul (2020).

https://doi.org/10.1017/9781139025065.003 Published online by Cambridge University Press

 18 grotius

It is difficult to overestimate the importance of Suárez’s insight for the moral phi-
losophy that followed him. The distinction between law (or command) and coun-
sel is explicit in Grotius, Hobbes, and Kant. All these philosophers distinguish
between normative reasons favoring or counseling an action, on the one hand,
and the idea that the action is obligatory or demanded (commanded) through the
moral law, on the other. And I shall be arguing that it is widely implicit elsewhere.
Several ideas are packed into Suárez’s notion of obligating law. First, he says,
“ordering pertains to the will,” so obligating moral norms or laws must aim to
direct a will; only thus can they have “binding force” (Suárez 1944: 66, 67). So,
second, moral norms are God’s will as addressed to us, hence to our rational
wills second personally (Darwall 2006). But Suárez’s idea is not that God seeks
to determine our wills directly. If that were so, we could not fail to comply (“all
these precepts would be executed”), since God is omnipotent (Suárez 1944: 55).
Rather, God wills “to bind” his subjects by addressing legitimate demands to them
through commands that they can then rationally choose to follow for what they
can regard to be good reasons (55). Thus, third, the commands that create natural
law must be addressed to human beings as free and rational. Laws of nature can
exist “only in view of some rational creature; for law is imposed only upon a nature
that is free, and has for its subject-matter free acts alone” (37). Finally, fourth, we
are accountable for complying with moral obligations. If we do not “voluntarily
observe the law,” we are culpable (“legal culprits in the sight of God”) (132).
In holding that the law of nature obligates because it is God’s command,
Suárez is implicitly accepting the premise of Anscombe’s Challenge: any law
such as morality purports to be can bind us only if it is divinely legislated. How-
ever, Suárez’s theological voluntarism is not really essential to his conceptual
distinction between law and counsel. Grotius, after all, invokes the distinction,
as we shall see, but denies that moral obligation is grounded in divine com-
mand. The law, he says, would hold “even if we were to suppose … that there is
no God” (Grotius 2005: I, xxiv). This is what makes him subject to Anscombe’s
Challenge (as a modern moral philosopher).
The crucial conceptual distinction is between (deontic) concepts like law,
authority, command, and obligation, which are conceptually tied to account-
ability and culpability, on the one hand, and the idea of normative reasons
weighing in favor of or counseling action, which is not, on the other. Con-
sistently with the conceptual connection between law and accountability, we
might be accountable for complying with the moral law even if the latter did
not depend on divine command. We might be accountable to one another, and
ourselves, as representative persons or members of the moral community.8

8
This is the view I argue for in Darwall (2006). Indeed, I argue there and in the next chapter
on Pufendorf, that features internal to the logic of accountability make theological volun-
tarism inherently unstable as an account of the moral law.

https://doi.org/10.1017/9781139025065.003 Published online by Cambridge University Press

 classical natural law: aquinas and suárez 19

Again, Grotius invokes Suárez’s distinction between law and counsel, but
denies his theological voluntarism.9 But if Suárez’s objection has force against
Aquinas, then why would it not also against Grotius? If Grotius rejects theo-
logical voluntarism as an explanation of the obligatory force of natural law, then
what separates his view from Aquinas’s? In the next section, I shall show how
despite his rejection of theological voluntarism, Grotius’s conception of natural
law is nonetheless juridical in Anscombe’s sense. And in the section following, I
shall argue that, as against both Aquinas and Suárez central aspects of Grotius’s
view commit him to accepting Sidgwick’s mark of the modern, that morality
(natural law) provides a source of normative reasons that are additional to those
provided by the agent’s good, however broadly construed. Finally, we will con-
sider Grotius’s response to Anscombe’s Challenge that morality requires a foun-
dation in divine command. Grotius claims that our rational and sociable nature
is sufficient to ground morality. This, he says, is the “fountain of right.”
But why, first, would Suárez not count as a modern by Anscombe’s stan-
dards? The reason is that although Suárez holds obligation to be essential to
law, he does not think it is necessary for standards of moral right and wrong,
nor for there to be normative reasons to comply with these standards. As odd
as it may sound to modern ears, Suárez holds that an act’s being morally wrong
does not in itself entail any obligation not to do it.
Like Aquinas, Suárez is in this respect an Aristotelian naturalist (Irwin 2008:
38–41). Although Suárez holds that it takes God’s command to create laws that
place us under obligation, this adds obligation to what it is already intrinsi-
cally right, and wrong not, to do.10 As Irwin puts it, Suárez believes that “if we

9
Tuck points out that in the Prolegomena to the first edition, Grotius says that the law of
nature “necessarily derives from intrinsic principles of a human being” and that the law
would hold “even if we were to suppose … that there is no God, or that human affairs are
of no concern to him” (Grotius 2005: I, xxiv; III, 1748–1749). Tuck notes that Grotius is
less direct on this point in later editions. There Grotius says that “Natural Right [the Law
of Nature] is the Rule and Dictate of Right Reason, shewing the Moral Deformity or Moral
Necessity there is in any Act, according to its Suitableness or Unsuitableness to a reason-
able Nature, and consequently, that such an Act is either forbid or commanded by GOD,
the Author of Nature” (I, 150–151, emphasis added). This might encourage the Suárezian
thought that genuinely obligating natural laws require an authoritative divine direction
that is consequent upon any intrinsic reasonableness or unreasonableness and that the
latter is impotent to provide this all by itself. However, Grotius then adds that actions that
are thus suitable or unsuitable to a reasonable nature are “in themselves either Obligatory
or Unlawful, and must, consequently, be understood to be either commanded or forbid
by God himself” (I, 151–152, emphasis added). This means that the obligatory character of
natural law depends not on divine legislation, but vice versa.
10
“This will of God, prohibition or prescription, is not the whole character of the goodness
and badness that is present in the observance or transgression of natural law, but it assumes
in the actions themselves some necessary rightness or wrongness, and joins to them a spe-
cial obligation of divine law” (Suárez 1944: ii.6.11, quoted in Irwin 2008: 38n).

https://doi.org/10.1017/9781139025065.003 Published online by Cambridge University Press

 20 grotius

abstract divine commands from the natural law, what is left is morality (hon-
estas), not just natural goodness” (31). God’s commands do not make what
is wrong wrong; they add an obligation to avoid the wrong, so that wrongful
actions end up being, as it were, doubly wrong: wrong in and of themselves, but
also because they disobey and so wrong God.
Therefore it is necessary that it add some obligation of avoiding the evil
that is evil from itself and by its own nature. Further, there is no contradic-
tion if a thing that is right from itself has added to it an obligation to do
it, or if a thing that is wrong from itself has added an obligation to avoid
it (Suárez 1944: ii.6.12, Irwin 2008: 29b).

As Irwin emphasizes, Suárez follows Aquinas in holding (i) that standards of

right and wrong are inherent in the nature of rational human beings; (ii) that
moral rightness, so understood, is an intrinsically beneficial good, the “bonum
honestum,” which is distinct from the pleasant and the useful; and (iii) that this
good provides the ultimate normative reasons agents have to do what is morally
right for its own sake (Irwin 2008: 31, 2007: 606). As eudaimonists, Aquinas and
Suárez both hold that any reasons agents have to do what is morally right must
be rooted in their own good, where this is the intrinsic good of bonum honestum.
This ends up making Suárez pre-modern by Anscombe’s and by Sidgwick’s
standards. Both Aquinas and Suárez have a way of marking out “moral” goods
within the class of intrinsic goods, since only these goods and virtues distinc-
tively concern the common good (Irwin 2007: 615–619, 2008: 66–67). But nei-
ther holds that moral right (honestas) is essentially juridical nor that it involves
a non-eudaimonist source of normative reasons.
Since there is nothing essentially juridical about honestas, “morality” as
Aquinas and Suárez conceive it is in this respect unlike morality in Anscombe’s
(allegedly) modern sense. And a similar point holds for Sidgwick’s contrast. A
mark of “ancient” ethics, for Sidgwick, again, is that “Virtue or Right action is
commonly regarded as only a species of the Good” (Sidgwick 1967: 105–106).
Since Suárez and Aquinas both hold that honestas provides no non-eudaimonist
ground for action, but is instead a distinctive kind of intrinsic benefit (honestum
bonum) whose normative weight can be fully captured within a eudaimonist
framework, both count as pre-moderns by Sidgwick’s standard also. Neither
brings an intrinsic connection to obligation into their conception of “morality,”
and neither has any reason to think that morality requires a kind of reason for
action that differs from eudaimonist considerations of the agent’s own good.

Grotius on “ius,” Morality, and Obligation

Grotius disagrees with Aquinas and Suárez on both counts. To begin to under-
stand Grotius’s position on these issues, however, we need to examine a three-
way distinction he draws at the beginning of RWP between things correctly

https://doi.org/10.1017/9781139025065.003 Published online by Cambridge University Press

 grotius on “ius,” morality, and obligation 21

called ius.11 First, Grotius says, “ius” can “signif[y] merely that which is just” or
at least not unjust, where “that is unjust which is repugnant to the Nature of a
Society of Reasonable Creatures” (Grotius 2005: I, 136). So far, this seems thor-
oughly within the classical natural law and Aristotelian naturalist framework;
no essentially juridical notions need be involved.
Grotius then quotes Florentinus’s remark that “Nature has founded a kind
of Relation between us,” and Seneca’s saying that human beings “are born for
Society, which cannot subsist but by a mutual Love and Defence of the Parts”
(I, 136). He next distinguishes two different kinds of “relations” that are found
in societies. Some are relations of “unequals,” such as “Parents and Children,
Masters and Servants, King and Subject,” where one individual has authority
over and the standing to govern another. Other relations are of “equal[s],” such
as “Brothers, Citizens, Friends and Allies,” where each party is conceived to be
self-governing, at least so far as their relations to one another are concerned (I,
136). In the former instance, superiors have a “Right of Superiority”; in the lat-
ter, each has a “Right of Equality.”12 Grotius summarizes: “So that which is just
takes place either among Equals, or amongst People whereof some are Gover-
nors and others governed, considered as such” (I, 137).
Grotius thus defines the human relations he is concerned with in terms
of reciprocal recognition of relative authority. So understood, a relationship
essentially includes the authority or standing those within it have to make
claims on and demands of others to whom they are related by it; the relations
are already conceived as inherently involving rights (ius) of this distinctive
kind. Since relations of these sorts are essentially conceived in terms of ius, it
follows straightway that anything that is contrary to the nature of people who
are related in these defined ways will also be contrary to ius in a legal/juridical
sense.
It is often noted that Grotius seeks to ground natural law in a form of “socia-
bility” that is distinctive of human beings (I, 79–87). It is easy to mistake the
force of his idea, however, since it can seem an attempt to derive moral obliga-
tions from some form of benevolent concern or affinity with others or some
desire to be with or in agreement with them. But no such attempt can succeed
in grounding juridical moral notions, especially since, as we shall see, Grotius
himself distinguishes between what we owe to others and what we might prop-
erly be moved by love or by some sense of merit or desert to provide for them.
Only in the former case, he holds, are genuine obligations (with associated
rights) involved.
Moreover, we have seen that the distinctive kind of social relations Grotius
has in mind are themselves already conceived in terms of relative authority,
that is, in terms of the standing that the related parties have to make claims on

11
Grotius’s translators alternately translate “ius” as “law” or “right.”
12
Kelsey translates these as “rectorial law” and “equatorial law,” respectively (Grotius 1925: 37).

https://doi.org/10.1017/9781139025065.003 Published online by Cambridge University Press

 22 grotius

and demands of one another.13 This gives a distinctive color to Florentinus’s

remark that “Nature has founded a kind of Relation between us”; it suggests
that this fundamental social relation must be conceived in terms of authority
also. And that affects how Grotius’s concept of “sociability” might most chari-
tably be understood.
I shall argue that when Grotius says that “sociability” is “the Fountain of
Right” (I, 85–86), the most promising philosophical option is to interpret him
as holding that it is our standing to make reasoned claims and demands of one
another at all that underlies the more specific rights and obligations that are
contained in the law and right of nature. In other words, human beings share
an equal authority over one other. Only such an interpretation of sociability
can bring it into the conceptual space of law rather than “counsel.”
To be clear, this interpretation goes well beyond what Grotius explicitly says.
My point is that the idea of a standing to make claims on and demands of one
another is implicit in aspects of his thought to which he is committed. And I
maintain that it has a fundamental and distinctive importance for claims of
moral obligation and rights that he, and after him so many philosophers of the
modern period, wished to defend.
The second meaning of ius that Grotius mentions brings out the idea of
authoritative standing even more clearly. Grotius says that this second sense
is different from the first, “yet aris[es] from it.” In the second sense, “Right
is a moral Quality annexed to the Person, enabling him to have, or do, some-
thing justly” (I, 138). Here Grotius introduces his famous distinction, which
will reverberate throughout the modern period, between perfect and imper-
fect rights (generating the modern distinction between perfect and imperfect
duties or obligations).14 A perfect right is a “Faculty” of the person, a norma-
tive “power” over oneself or, in special cases, over others, as when a relation of
“superiority” like parent/child, master/servant, or king/subject obtains. Perfect
rights are also implicated in property relations and contracts.
It is clear that Grotius identifies perfect rights with what we now call claim
rights, which generate correlative obligations. Barbeyrac’s French translation
(from which Morrice’s English translation draws) makes this explicit by adding

13
It is worth noting in this connection that Pufendorf understands sociability (socialitas) in
similar terms: “By a sociable attitude, we mean an attitude of each man towards every other
person, by which each is understood to be bound to the other by … a mutual obligation”
(1934: 208). Pufendorf of course differs from Grotius in seeking to derive natural law from
divine command (including God’s “fundamental” command to have a “sociable attitude”)
rather than from sociability. Even so, he and Grotius apparently agree in understanding
sociability itself as involving relations of mutual right, obligation, and hence, authority,
albeit understood formally.
14
First, in Pufendorf (e.g., 1934: 90). On the significance of Grotius’s distinction, see
Schneewind (1998, esp. pp. 78–79, 133–134).

https://doi.org/10.1017/9781139025065.003 Published online by Cambridge University Press

 grotius on “ius,” morality, and obligation 23

that perfect rights involve (in Morrice’s English) “The Faculty of demanding
what is due, and to this answers the Obligation of rendering what is owing”
(I, 139).15 It follows that there can be perfect rights only if natural law includes
genuine obligations to respect them.
An imperfect right, on the other hand, is not a “Faculty” but an “Aptitude.”
Under this heading, Grotius includes considerations of “Worth” and “Merit” that
can recommend actions as more or less worthy or meritorious, but that no one
has standing to claim or demand (I, 141). “Prudent management in the gratuitous
Distribution of Things” to which no individual or society has a valid claim may
nonetheless recommend giving preference to “one of greater before one of less
Merit, a Relation before a Stranger, a poor Man before one that is rich” (I, 88).
However, Grotius insists that “Ancients” like Aristotle, and even “Moderns” who
follow him are mistaken if they think this a matter of right, though they may take
considerations of these kinds to be included within what they call “justice” (it is
what Aristotle and his followers include under “distributive justice”).16 “Right,
properly speaking, has a quite different Nature,” namely, “doing for [others]
what in Strictness they may demand” (I, 88–89).17
In its final sense, “ius” or “Right” “signifies the same Thing as Law when
taken in its largest Extent, as being a Rule of Moral Actions, obliging us. … I
say obliging: for Counsels, and such other Precepts, which however honest and
reasonable they be, lay us under no Obligation, come not under this Notion of
Law, or Right” (I, 147–148).
Grotius here explicitly invokes Suárez’s distinction between law and coun-
sel. Even in “its largest Extent,” ius is to be distinguished from counsel. It is
of the nature of law and right to obligate; ius makes legitimate demands of us
to which we are accountable for conforming. Normative reasons that simply
weigh in favor of or “counsel” action lack this conceptual connection to the
deontic; they do not obligate of themselves. This does not mean that there is not
good reason to do what we are obligated to do. Far from it. It just means that
being obligated to do something cannot itself be captured through the weight
of reasons.

15
In his notes, Barbeyrac defends this addition and argues that it should be read to apply
not just to the case of contracts, as the Latin text might suggest, but to all perfect rights.
Whether or not this is defensible as a translation, it seems clear that Barbeyrac is correct in
thinking that if perfect rights are claim rights, then Grotius is committed to it.
16
On the curious difference between this traditional notion and our contermporary notion
of distributive justice, see Fleischacker (2004: 17–28).
17
Kelsey translates this last passage from section 10 of the Prolegomena as “leaving to
another that which belongs to him, or in fulfilling our obligations to him” (Grotius 1925:
13). Although it is clear from the text later that Grotius has Aristotle’s distributive justice in
mind, the language of “ancients” and “moderns” is Barbeyrac’s. Kelsey has: “Long ago the
view came to be held by many, that this discriminating allotment [distributive justice] is
part of law” (1925: 13).

https://doi.org/10.1017/9781139025065.003 Published online by Cambridge University Press

 24 grotius

Grotius follows this with his definition of the “law” or “right” of nature:
Natural Right is the Rule and Dictate of Right Reason, shewing the Moral
Deformity or Moral Necessity there is in any Act, according to its Suitableness
to a reasonable Nature (I, 150–151).

Barbeyrac notes that other editions interpolated “and Sociable” between “rea-
sonable” and “Nature” and says there is some reason to believe that these were
simply left out by a printer or transcriber (I, 151, note 2). As he points out, when
Grotius distinguishes between a priori and a posteriori proofs of laws of nature,
he brings in sociability explicitly. A posteriori proofs appeal to a consentium
gentium, that is, to something being “generally believed to be” natural law “by
all, or at least, the most civilized Nations” (I, 159). An a priori proof, by contrast,
proceeds by “shewing the necessary Fitness or Unfitness of any Thing, with a
reasonable and sociable Nature” (I, 159).
We shall return to how Grotius might hope to ground laws of nature in rea-
sonable sociability in the final section. Since Grotius rejects Suárez’s theological
voluntarism, if he is to hold, like Suárez, that ius differs from counsel because of
its distinctively obligating normativity, he needs some other way of accounting
for this. Here, again, he faces Anscombe’s Challenge.
Already, however, we can glimpse why Grotius might think that norms that
can be justified in this way would have to be genuinely obligating. If, as we
noted above, Grotius conceived of sociability as itself involving a fundamen-
tal standing to make claims and demands of one another, then being contrary
to our reasonable and sociable character would mean being at odds with our
standing to make reasoned or reasonable claims and demands of each other,
not simply being contrary to standards of rational thought and action in gen-
eral, even standards rooted in our rational nature however social that nature
might be.18 Only if it can be justified in some such way, indeed, would a “law
of nature” be a genuine ius. Only if it can be grounded in an authority to issue
demands would it be able to “lay us under” a genuine obligation in a way that a
mere counsel cannot.
We can also now see that Grotius, unlike Suárez, is in no position to hold that
obligation, though necessary for law, is inessential to moral right and wrong.
Grotius ties his third definition of right – which he follows with his definition of
natural right as “Suitableness to a reasonable Nature” – to obligation. Although he
agrees with Aristotelian naturalists like Aquinas and Suárez that natural law and
right are grounded in rational (and sociable) nature, he nonetheless holds that
they entail obligation and that the latter does not come into existence through
God imposing it on us through command. Moral right and wrong as Grotius

18
For the difference between “sociability” in Grotius’s sense and “social” in our ordinary
sense and its philosophical significance, see Darwall (2014c).

https://doi.org/10.1017/9781139025065.003 Published online by Cambridge University Press

 morality as a distinctive source of reasons 25

understands them are already essentially juridical in Anscombe’s (and Sidg-

wick’s) sense. Something can be morally wrong only if we are under a legitimate
demand not to do it. “Counsel,” however reasonable, is an insufficient ground.
For Grotius, therefore, there is no notion of moral right and wrong that is
independent of moral obligation. Considerations of common good, taken by
themselves, are simply insufficient to show that anything is morally right or
wrong in the sense with which Grotius is distinctively concerned. The most
they can provide is counsel, not authoritative demand.

Morality as a Distinctive Source of Reasons

We turn now to Grotius’s relation to Sidgwick’s claim that modern ethical phi-
losophers came to recognize morality (natural law) as a source of normative
reasons for acting that is independent of the agent’s good, therefore, eudai-
monist. The point is somewhat delicate since, as Irwin has shown, a good bit
of what Grotius says can be accommodated within an Aristotelian naturalist,
eudaimonist framework according to which virtue – justice in particular – is
intrinsically beneficial (part of the bonum honestum), whether or not it benefits
agents instrumentally or promotes their partial interests, say, by making it like-
lier that others will act justly towards them in the future (Irwin 2003: 351–352).
A crucial passage occurs right at the outset of RWP when Grotius poses a
fundamental skeptical challenge to his ideas, one that will be echoed later by
Hobbes’s “foole” (Hobbes 1994a: XV.§4), Hume’s “sensible knave” (Hume
1975b: 256), and Kant’s worry that morality might be a “chimerical idea with-
out any truth” (Kant 1996a: 4:445). Who better to pose this challenge, Grotius
says, than the ancient skeptic Carneades, who held that “Laws … were insti-
tuted by Men for the sake of Interest” (Grotius 2005: I, 79).
As to that which is called Natural Right, it is a mere Chimera. Nature
prompts all Men … to seek their own particular Advantage: So that either
there is no Justice at all, or if there is any it is extreme Folly, because it engages
us to procure the Good of others, to our own Prejudice (I, 79).19

19
Cf. Hobbes:
The fool hath said in his heart, there is no such thing as justice; and sometimes also
with his tongue; seriously alleging that: “every man’s conservation, and content-
ment, being committed to his own care, there could be no reason, why every man
might not do what he thought conduced thereunto: and therefore also to make, or
not make; keep, or not keep covenants, was not against reason, when it conduced to
one’s benefit” (Hobbes 1994a: XV.§4).
And Hume:
[T]hough it is allowed, that, without a regard to property, no society could sub-
sist; yet, according to the imperfect way in which human affairs are conducted, a

https://doi.org/10.1017/9781139025065.003 Published online by Cambridge University Press

 26 grotius

According to Carneades, there is only one source of reasons for acting, the
agent’s own interest or good; therefore, there can be no reason to follow any
law that might conflict with that.
But what precisely is the challenge? If we were talking about law ordinarily
so called – the laws of actual societies or even international law – egoism of this
sort might pose no fundamental obstacle, since it seems no part of the concept
of the law of any actual state or even of international law that such a law exists
only if those subject to it have reason to respect it just because it applies to
them. Laws of these kinds can exist even if the only motives for following them
are self-interested desires to avoid sanctions.
This is not, however, the case with natural law, as that idea operated within
both the classical theory deriving from Aquinas and the modern version I am
claiming comes most clearly from Grotius. Something can be a natural law or
right only if it entails normative reasons for agents to act. And if so, Carneades’s
challenge is that “natural right” so understood is a “mere Chimera.” The only
source of normative reasons is the agent’s own interest.
Now it is important to see that although Aquinas and Suárez would cer-
tainly deny Carneades’s claim that no law is such that there is necessarily rea-
son to follow it, they nonetheless accept the eudaimonist assumption behind
his claim, as Aristotelian naturalists do more generally. Aquinas and Suárez
think there is necessarily reason to follow natural law because they believe
that the agent’s good necessarily coincides with it. In fact, for Aquinas, natu-
ral law and the agent’s good turn out to provide the very same normative
standard.
According to Aquinas, natural law is simply a formulation of “eternal law,”
God’s ideal or archetype for all of nature – “the exemplar of divine wisdom …
moving all things to their due end” – as it applies to rational human beings
(Aquinas 1997: Q93.1). This is Thomas’s synthesis of Aristotelian teleology with
the idea of divine rule. Eternal law specifies the perfection or ideal state of every
natural being, and so “rule[s] and measure[s]” them. Rational beings, however,
are subject to eternal law in a special way since, having “a share of the eternal
reason,” they can act in the light of their awareness of eternal law, choosing to
follow or flout it (Q91.3). What Aquinas calls “natural law” is simply the eternal
law made accessible to and applicable by rational creatures in practice (Q91.2).
It follows that the content of natural law cannot differ from that of the eter-
nal law for rational human beings and that, since what eternal law “requires”

sensible knave, in particular incidents, may think, that an act of iniquity or infidelity
will make a considerable addition to his fortune, without causing any considerable
breach in the social union and confederacy. That honesty is the best policy, may be
a good general rule; but is liable to many exceptions: And he, it may, perhaps, be
thought, conducts himself with most wisdom, who observes the general rule, and
takes advantage of all the exceptions (Hume 1975b: 256).

https://doi.org/10.1017/9781139025065.003 Published online by Cambridge University Press

 morality as a distinctive source of reasons 27

for any being is simply its good and perfection, natural law and the agent’s own
good are the very same normative standard. To follow natural law just is to
pursue one’s own good properly understood, and vice versa.20
It is also worth noting that on Aquinas’s picture, since individual human
beings realize their respective goods only within the overall teleological
scheme specified by eternal law, any fundamental conflict between individu-
als’ interests, properly understood, is ruled out – harmony is guaranteed by
perfectionist-teleological metaphysics. The moderns who follow Grotius,
however, assume that there can be genuine conflicts of interest; or at any rate,
they believe that it cannot be assumed that there cannot be. Individuals will
rationally believe that some rational conflicts are likely, even if that is only
because no one can rationally believe that everyone else will believe that there
won’t be.
For most of Grotius’s followers, the natural laws that comprise Barbeyrac’s
“morality” purport to provide a source of reasons that is distinct from and
potentially in conflict with self-interest.21 That is what makes Carneades’s chal-
lenge genuine for them and explains why it compels the attention of thinkers as
different as Hobbes, Hume, and Kant. For modern natural lawyers like Hobbes
and Locke, moreover, it is central to their conception of the natural (moral)
law that morality provides reasons for acting that trump considerations of self-
interest in precisely those instances where the collective result of each indi-
vidual’s pursuing their own good, or their own conception of a good life for
them would be worse for everyone.22 As these Grotius-influenced moderns see
it, the problem of social order is a collective action problem to which morality
provides the solution.23 Everyone does better if everyone follows the moral law
than each would do were they to pursue their own good, or, at any rate, to pur-
sue their own conception of their good.
From the perspective of his followers, then, the problem that Grotius appears
to pose right at the outset of the modern period is how to show why individuals
should respect such laws even when they would be better off individually (or,
at least sensibly believe they would) by departing from them. In other words,
since there cannot even be such a law unless it provides a source of reasons
independent of self-interest, why suppose that such a law exists? The problem
that Grotius thereby appears to set for the modern period is thus to show why
collectively advantageous, putatively obligating laws are genuinely obligating

20
At least, this equivalence holds with matters of common good.
21
On this in relation to what Schneewind calls the “Grotian Problematic” (in Schneewind
1998: 119–129) see Darwall (1999a: 340–341).
22
Although both of them believe that this is ultimately only because of sanctions (secular, for
Hobbes, and divine, for Locke). See my discussions of Hobbes and Locke on this point in
Darwall (1995: 36–44, 74–79).
23
For the classic discussion of collective action problems, see Olson (1971).

https://doi.org/10.1017/9781139025065.003 Published online by Cambridge University Press

 28 grotius

and reason-giving rather than chimerical, as Carneades claims, even though

following them is individually disadvantageous.
Even so, it is possible also to interpret Grotius, as Irwin does, as not himself
posing this modern problematic, however central it might become to later writ-
ers (Irwin 2003: 351–352). It is certainly true, moreover, that Carneades’s origi-
nal challenge, as well as the way it was understood by the Aristotelian naturalist
tradition before Grotius, does not require the modern Sidgwickean assumption
that morality provides a source of reasons that is additional to the agent’s own
interest properly understood. Carneades himself challenged law and justice on
roughly the same grounds that Glaucon and Adeimantus do in the Republic
(359), namely, that they are no more than artificial conventions instituted for
mutual instrumental advantage and that there is no intrinsic reason to follow
them when it is contrary to one’s (instrumental) interest to do so. And Grotius
by and large just quotes Carneades’s own challenge.24
Irwin notes that Aristotelian naturalists had a clear reply to Carneades’s chal-
lenge that is fully consistent with eudaimonism. Moreover, some of what Gro-
tius says certainly suggests that his own response to the skeptical challenge is
simply that of Aristotelian (or as Irwin calls it elsewhere, “Stoic”) naturalism,
and hence that he accepts eudaimonism and so is not a modern by Sidgwick’s
lights (Irwin 2003). Grotius says that where Carneades goes wrong is in failing
to appreciate that a “desire for society,” or the “Disposition the Stoicks termed
‘oikeiôsis’,’ [sociability],” is an essential aspect of human nature. The Stoics held
that actions can be, as Irwin puts it, “morally right (honestum) because they are
appropriate for human nature,” given our natural sociability, and therefore that
there is “a natural basis for justice, apart from the usefulness of justice in main-
taining society” (2003: 352). And they held also that justice’s being expressive
of our nature and thus “natural” makes just action intrinsically beneficial to the
just agent, whether it is useful, that is, instrumentally beneficial to them or not.
So understood, Grotius’s reply to Carneadean skepticism is a response of the
same general kind as Socrates’s reply to Glaucon and Adeimantus in Plato’s
Republic. Plato and the Stoics might disagree about why justice is intrinsically
beneficial to the just agent, and perhaps also about what justice is. But they
agree that the fundamental reason to be just derives from the agent’s own good.
And so, Irwin argues, does Grotius.
However, this ancient reply to Carneadean skepticism will not do as a
response to a skeptical challenge to the existence of natural law as Grotius
understands it, and therefore as he must seek to defend it. The reason is, as Gro-
tius himself explicitly points out, that “ancient” conceptions of right and law

24
As it is reported in Lactantius’s Divine Institutes: “Nature prompts all Men … to seek their
own particular Advantage: So that either there is no Justice at all, or if there is any it is
extreme Folly, because it engages us to procure the Good of others, to our own Prejudice”
(I, 79; see also Long 1986: 104).

https://doi.org/10.1017/9781139025065.003 Published online by Cambridge University Press

 morality as a distinctive source of reasons 29

lack any conceptual connection to obligation and legitimate demands. Grotius

contrasts the Aristotelian conception of distributive justice that “ancients” and
even some “moderns” include under the concept of right with right “properly
speaking,” since the latter includes doing for others “what in Strictness they
may demand” (88–89). And he ties the broadest sense of right he is concerned
with (the third sense of ius distinguished above) to obligation in the juridical
Anscombean sense. Unlike “Counsels” and other “reasonable” “Precepts,” law
and right “lay us under … Obligation” (148).
The problem with the classical Stoic and Aristotelian naturalist reply is that it
can provide no more than reasonable counsel, however weighty the reasons of
the agent’s good supporting the counsel might be. Therefore, even if it responds
adequately to Carneades’s challenge in the terms in which Carneades himself
raised it, it is impotent to respond to a challenge to natural right and law, as
Grotius understands these. However good it might be for us to comply with
a standard or norm, that would not yet show that we lie under any obligation
to comply with the standard or that that is something that can legitimately be
demanded of us. The most the Aristotelian naturalist response can support is
“reasonable counsel.” It therefore cannot yet show that what we call natural law
or right actually is a law or a right as Grotius understands these, since it would
not yet have established its obligatory character.
Once we have the distinction between a mere “counsel,” however well sup-
ported by reasons of extrinsic or of intrinsic value, on the one hand, and an
obligating demand, on the other, we are committed to the idea that there must
a source of reasons for acting other than the agent’s own good. The most the lat-
ter can provide is counsel, so if Grotius’s natural law and right are to exist, there
must be another source of reasons for following them. And if there is not, then
the ideas of natural law and right are only chimerical and “there is no justice at
all,” just as Grotius has Carneades say.
It might be replied, however, that even if the arguments of the last two para-
graphs show that eudaimonist reasons cannot establish natural law and right as
genuinely obligating, they do not show that eudaimonism cannot provide an
adequate account of the reasons for complying with the law. After all, that was
Suárez’s position. Why could it not have been Grotius’s also?
Part of the problem is that, unlike Suárez, Grotius has no way of separating
the grounds of natural law from the reasons for compliance. Suárez was an
externalist in the sense that he held that what grounds something as obligating
law, God’s command, is distinct from what provides reasons for compliance.
Only considerations of the agent’s own good can supply the latter, even if some
are related to God’s commands, for example, through the good of obedience
and the avoidance of sanctions.
Unlike Suárez however, Grotius has no independent way of grounding nat-
ural law. To the contrary, he appeals to human beings’ rational and sociable
nature both to reply to Carneades’s challenge and as the “Fountain of Right”

https://doi.org/10.1017/9781139025065.003 Published online by Cambridge University Press

 30 grotius

(79–81, 86). In this respect, Grotius is an internalist. What grounds natural law
and right as genuinely obligating must simultaneously supply normative rea-
sons for compliance.
Although this ups the ante, it nonetheless gives Grotius a significantly more
satisfying position philosophically. The externalist must hold that the fact that
an action is legitimately demanded of an agent, and that they would be blame-
worthy for failing to comply were they to lack some further justification or
excuse, does not itself entail any normative reason for them to comply. Any rea-
sons for compliance must come from elsewhere – from the agent’s own good
for a eudaimonist like Suárez. But however broadly we construe the agent’s wel-
fare, as including, for example, the bonum honestum or obedience to God as
intrinsic goods, it seems nonetheless to supply a “reason of the wrong kind”
for their compliance with moral obligations and legitimate demands (Prichard
2002, Strawson 1968, Darwall 2006).
When we hold someone to account for doing something through what P. F.
Strawson calls “reactive attitudes” like moral blame, we thereby imply that
there were conclusive reasons for the agent not to have done what they did
(Williams 1995, Gibbard 1990: 42). We presuppose that there are reasons for
acting consisting in or entailed by the legitimate demands or obligations them-
selves, whether or not it will benefit the agent to comply. If, for example, we
owe something to someone as a matter of right, it seems beside the point to ask
whether discharging our obligation to them will benefit us.
We can agree with Irwin that nothing just said entails that Grotius himself
saw that his conception of natural law requires a defense going beyond Aristo-
telian naturalism or that he rejects its eudaimonism. But there is no doubt that
many of his followers saw this, as we can see from the way in which Grotius’s
Carneadean trope is repeated in Hobbes’s reply to “the foole” (Hobbes 1994a:
XV.§4), Hume’s reply to the “sensible knave” (Hume 1975b: 256), and Kant’s
response to the charge that morality might be a “chimerical idea without any
truth” (Kant 1996a: 4:445). These all assume that acting against morality might
sometimes coincide with the agent’s own interest in fact.25 Consequently, they
all assume that morality can exist only if eudaimonism is false, only if, that
is, there can be reasons for agents to act that do not derive from their own
good. Whether or not Grotius considered himself a modern in Sidgwick’s
sense, therefore, his account of moral right as entailing obligation and legiti-
mate demand, and thus as being distinct from counsel, might reasonably have

25
It is consistent with this that philosophers like Hobbes can hold that self-interested eudai-
monist considerations still come in at another level. Hobbes is frequently seen as some kind
of “rule” or “indirect” egoist. The point is that they do not deny the Carneadean claim that
the wrong or unjust act may be beneficial. Obviously, for Kant, eudaimonist considerations
don’t enter at any level.

https://doi.org/10.1017/9781139025065.003 Published online by Cambridge University Press

 freedom, self-rule, and the right to punish 31

been seen by his followers as committing himself to the “modern” outlook as

Sidgwick conceived it.

Freedom, Self-Rule, and the Right to Punish

An important theme of Richard Tuck’s writings is that the moral/political con-
ceptual framework we have inherited from the early moderns, one that bases
political authority on the rights and dignity of individuals, begins with Gro-
tius.26 Part of Grotius’s purpose in writing both RWP and the earlier On the
Law of Prize and Booty27 was to argue that organized groups of individuals,
like the Dutch trading companies with which he and his family were involved,
were like states in having a right to punish wrongs that did not violate their
own rights but where the perpetrators might not otherwise be held responsible.
Implicit throughout is a conception of individual persons as having the author-
ity to rule themselves as well as the standing to hold one another responsible for
respecting natural rights and law.
Consider, first, the following passage, which brings out Tuck’s analogy
between the personal and the political:
But as there are several Ways of Living … and every one may chuse
which he pleases of all those Sorts; so a People may chuse what Form of
Government they please: Neither is the Right which the Sovereign has over
his Subjects to be measured … but by the Extent of the Will of those who
conferred it upon Him (Grotius 2005: I, 262).

Or the following, often cited by Tuck, from On the Law of Prize and Booty:
God created man autexousion, “free and sui iuris,” so that the actions of
each individual and the use of his possessions were made subject not to
another’s will but to his own. Moreover, this view is sanctioned by the com-
mon consent of all nations. For what is that well-known concept, “natural
liberty,” other than the power of the individual to act in accordance with
his own will? And liberty in regard to actions is equivalent to ownership
in regard to property. Hence the saying: “every man is the governor and
arbiter of affairs relative to his own property” (Grotius 1950: 18).

26
“We take for granted that the language in which we still describe this autonomous,
right-bearing individual is in fact a language to describe states or rulers. When Hart
in his famous 1955 essay ‘Are There Any Natural Rights?’ said about promising that
‘the promisee has a temporary authority or sovereignty in relation to some specific
­matter over another’s will,’ he was drawing on precisely this tradition which we find
articulated for the first time in [Grotius’s] De Indis [Grotius’s favored title for On the
Law of Prize and Booty]’ (Tuck 2001: 84–85). See also Tuck (1981, 1993); and,
especially, Tuck (2001: 1–9, 83–89).
27
Grotius wrote On the Law of Prize and Booty between 1604 and 1606.

https://doi.org/10.1017/9781139025065.003 Published online by Cambridge University Press

 32 grotius

The “well-known concept” of “natural liberty,” is a reference to Fernando

Vázquez de Menchaca; but however well established the idea of self-rule was
in some form or other, it seems clear that Grotius took the idea of a natural
right to govern oneself significantly further. In Vázquez de Menchaca’s hands,
for example, “natural liberty” seems to refer alternately to a psychic faculty for
free choice shared by rational agents, on the one hand, and to a Hohfeldian
liberty, that is, to a range of permitted choices that violate no law or obliga-
tion, on the other.28 With Grotius, however, the right to rule oneself evidently
includes a Hohfeldian claim right and therefore entails a consequent obliga-
tion of others to allow one to do so.29
“Right properly and strictly so called,” Grotius says, “consists in leaving oth-
ers in quiet Possession of what is already their own [including the ‘Power …
over ourselves, which is term’d Liberty’ (2005: I, 138)], or in doing for them
what in Strictness they may demand” (I, 88–89). When we fail to abstain from
what belongs to others (including, again, by interfering with their liberty), their
right gives them standing to demand “Restitution” of what we have taken, inso-
far as this is possible, and “Reparation” of any “Damage done through our own
Default” (I, 86). These reparative responses are all “due” to others not just in
the sense that it is fitting that they have them or even that they deserve them,
but also that these others have “the Faculty of demanding what is due,” and that
we consequently have “the Obligation of rendering what is owing” (I, 139). As
Barbeyrac remarks in a footnote to the passage quoted at the beginning of the
last paragraph, “When we Repair the Damage he has sustained in his Person,
Goods, or Reputation, whether designedly or through Inadvertency, we restore
what we had taken from him, and what was his own, which he had a strict Right
to demand” (I, 88–89n).30
It follows that natural right, as Grotius understands it, includes
­obligation-entailing claim rights of individuals to demand that others conduct
themselves toward them in various ways: “the Abstaining from that which
is another’s, the Restitution of what we have of another’s, or of the Profit we
have made by it, the Obligation of fulfilling Promises, [and] the Reparation

28
For an excellent discussion of Vázquez de Menchaca on natural liberty, see Brett (1997:
165–204). Brett argues that Hobbes uses “natural liberty” to refer to these two things also
(Brett 1997: 205–235). For Hohfeld’s classification of rights, see Hohfeld (1923).
29
Or, at least, it was taken to include such a claim right, as is shown by Barbeyrac’s remarks
quoted in the next paragraph.
30
Similarly, when Grotius considers an argument on behalf of the Maccabees that “they acted
by Vertue of the Right which their Nation had to demand Liberty, or the Power of govern-
ing themselves,” which right we know from the first passage quoted above that Grotius
must hold derives from individuals’ right to choose, he implicitly accepts the premise that
that right would include a right to “demand Liberty.” (However, he also claims that the
Maccabees had lost their right to liberty by earlier conquest (I, 359).)

https://doi.org/10.1017/9781139025065.003 Published online by Cambridge University Press

 freedom, self-rule, and the right to punish 33

of a Damage done through our own Default” (I, 86). And this shapes how the
law of nature must itself be understood; it must include respect for these claim
rights.31
Grotius’s doctrine of natural right thus shapes his theory of natural law in the
direction of a conception of morality as requiring protections of certain basic
interests of individuals. But there is also a second way in which Grotius points
toward what will become a signal modern idea: that the dignity of individuals
is at the center of the moral law. In discussing Grotius’s views on natural rights,
Tuck aptly compares them to H. L. A. Hart’s characterization of rights in his
landmark paper, “Are There Any Natural Rights?” In Tuck’s words, rights, for
Grotius and Hart, “constitut[e] a kind of sovereignty for the individual over
parts of his life” (Hart 1955, Tuck 2001: 9, see also 84–85).32
In addition, Grotius holds a view about the standing of individuals to punish
violations of natural law that connects the law of nature to the dignity of indi-
viduals in yet a further way.
Is the power to punish essentially a power that pertains to the state
[respublica]? Not at all. On the contrary, just as every right of the mag-
istrate comes to him from the state, so has the right come to the state
from private individuals; and similarly, the power of the state is the
result of collective agreement. … Therefore, since no one is able to
transfer a thing that he never possessed, it is evident that the right of
chastisement was held by private persons before it was held by the state
(Grotius 1950: 91).

31
This is quite different from a conception of natural liberty as a Hohfeldian liberty (arguably,
as in Vázquez de Menchaca and later in Hobbes), which simply entails permissions without
any accompanying obligations. Hobbes’s “right of nature,” for example, of each “to use his
own power, as he will himself, for the preservation of his own nature” (Hobbes 1994a: xiv,
§4), entails no obligations of others to allow him to do so. It simply means that such an exer-
cise of liberty by the agent themselves is “blameless.” The term Hobbes uses in The Elements
of Law is “blameless liberty.”: “And that which is not against reason, men call RIGHT, or jus,
or blameless liberty of using our own natural power and ability” (Hobbes 1994b: 79).
32
An important additional aspect of Grotius’s idea is best brought out by reference to
another classic paper on rights, namely, Joel Feinberg’s “The Nature and Value of Rights”
(Feinberg 1980). This is the thought that claim rights involve the authority of the right
holder to claim or demand certain treatment. This goes beyond it simply being the case
that others should treat us in certain ways. It is the additional idea that right holders have
the authority to demand that they be so treated and to demand restitution or reparation
should they not be. As Feinberg puts it, “it is claiming that gives rights their special moral
significance” (Feinberg 1980: 155). Without rights, although others might accept norms
that require them to treat us in certain ways, we could not claim this is as our right. We
would have, as Feinberg says, no place to “stand, … look others in the eye” and make claims
on one another (Feinberg 1980: 151). This gives us a dignity; as Feinberg stresses, it makes
others accountable or answerable to us (Darwall 2006, 2013a). Better, our sharing these
rights makes us mutually accountable.

https://doi.org/10.1017/9781139025065.003 Published online by Cambridge University Press

 34 grotius

As further evidence, Grotius adds an argument that will later be picked up by

Locke in The Second Treatise of Government in support of his view that individ-
uals in the state of nature have a right to punish that is additional to their right
to seek reparation for violation of their own rights (Locke 1988: 272 (II.9)) –
namely, that states normally claim the right to punish wrongs against not only
their own citizens, but also against foreigners, “yet it derives no power over
the latter from civil law, which is binding upon citizens only because they have
given their own consent” (Grotius 1950: 91–92).33
According to Grotius, then, individuals have not only the authority to
demand compliance with their own rights, and to demand restitution and rep-
aration when these are violated, they also have the standing to demand that
the moral law be complied with in respect of others’ rights also. In effect, they
have an authority of membership in an assumed moral community of mutually
accountable equals.
This adds a distinctive element to the moral law. Since the law requires
respect for each person’s rights, it involves obligations that are, in the first
instance at least, to patients, that is, those whom we affect by our actions. But
patients are not the only ones who have the standing to demand compliance
with these rights, and hence the law; all others do as well, since they have the
standing to punish and not just to secure restitution or reparation on behalf
of the victim. And since that is so, moral obligations involve, in the second
instance, responsibilities to all persons and not just to patients. Here we have
at least a strong suggestion, perhaps the first, of a conception of moral com-
munity as mutual accountability. Morality imposes genuine obligations that we
are accountable for complying with, where this accountability involves being
answerable to one another.

Publicly Articulable General Principles: A “Science” of Morality

It should now be evident that the central elements of the conception of natural
law that in the mid-eighteenth century Barbeyrac referred to as “morality” were
present already in Grotius (Barbeyrac 1749). Grotius holds that all individuals

33
Grotius is not so explicit about individuals’ right to punish in RWP. There he says that
the “Person to whom the Right of Punishing belongs, is not determined by the Law of
Nature” (II, 955). All “natural Reason” tells us for sure is that “a Malefactor may be pun-
ished, but not who ought to punish him” (II, 955). Though it “suggests … it is fittest to be
done by a Superior,” it does not show that “to be absolutely necessary” (II, 955). Even if this
seems weaker, there is no reason to think that Grotius has gone back on his earlier view.
Indeed, in the passage where he proclaims sociability the “Fountain of Right,” he adds that
that this Right includes not just “Abstaining from that which is another’s,” “Restitution,”
“Reparation,” and “the Obligation of fulfilling promises,” as I mentioned before, but also
“the Merit of Punishment among Men” (I, 86).

https://doi.org/10.1017/9781139025065.003 Published online by Cambridge University Press

 publicly articulable general principles: a “science” of morality 35

are subject to universal laws just by virtue of being rational sociable persons,
irrespective of their local, national, or religious differences; that these laws
impose genuine obligations; and, therefore, that agents are responsible for com-
plying with them (and so subject to punishment if they do not). Moreover,
these obligations include respecting natural rights that any rational moral agent
has, including a right of autonomy or self-rule. The classic problem Grotius
bequeathed to the modern period was how to establish that such universal gen-
uinely obligating laws actually exist and are not merely chimerical, especially
if their existence does not derive from divine command. Grotius initiated the
form of modern moral philosophy that provoked Anscombe’s Challenge.
I said at the beginning that Grotius was also an important source for what
Barbeyrac called the “science of morality.” This terminology is not Grotius’s,
nor is it familiar today, so we should ask what Barbeyrac intended by it. By
a “science of morality,” Barbeyrac means a publicly accessible formulation of
basic “Principles and Rules of Morality” together, perhaps, with some account
of their power to obligate (2).34 He says that discovering moral “principles and
rules” requires no “inquiry into the impenetrable Secrets of Nature,”35 and that
it is available to everyone, including “Persons, of the lowest Rank” (3).
It can be argued that a “science” of morality in this sense is precisely what
Grotius was attempting himself to provide. It is worth quoting at length an
excellent observation Tuck makes on this point in his introduction to The
Rights of War and Peace:
The Indian Ocean and the China Sea were an arena in which actors had to
deal with one another without the overarching frameworks of common
laws, customs, or religions; it was a proving ground for modern politics
in general, as the states of Western Europe themselves came to terms with
religious and cultural diversity. The principles that were to govern ­dealings
of this kind had to be appropriately stripped down: there was no point
in asserting to a king in Sumatra that Aristotelian moral philosophy was
­universally true (Grotius 2005: I, xviii).

Grotius’s project was to formulate “minimalist” principles of obligation. The

point was to specify principles that people can readily agree they should hold
one another and themselves accountable for complying with regardless of
national, cultural, and religious differences.
It is, indeed, because laws of nature (moral laws) purport to provide a
basis for holding one another answerable, regardless of these more specific

34
In his view, again, theological voluntarism provides the latter.
35
“When we say, That Man is subject to Law; we mean nothing by Man, but a corporeal rational
Creature: What the real Essence, or other Qualities of that Creature are, in this Case, is no
Way consider’d” (1749: 4). Compare this to Locke’s claim that “Person” is a “forensic term”
(i.e., competent to be held accountable) that does not refer to a real essence (Locke 1975: 346).

https://doi.org/10.1017/9781139025065.003 Published online by Cambridge University Press

 36 grotius

differences, that they must be presumed to be publicly formulable and gen-

erally accessible without assuming any specific cultural or religious doctrine.
Aristotle can intelligibly hold that someone of perfect virtue has a kind of prac-
tical wisdom that requires years of cultivation and experience and that may not
even be possible for many people. Why think that everyone should be able to
attain excellence of any kind or be capable of noble conduct?
Once, however, we have a conception of a standard that people can be held
accountable for complying with, it seems that we must assume that anyone
subject to it has what it takes to comply, that compliance requires no esoteric
knowledge or special talents or wisdom that ordinary people cannot be pre-
sumed to have. The very idea of a standard of mutual accountability seems to
require that it be capable of some public “minimalist” formulation, in some-
thing like the golden rule, which Hobbes says is “intelligible, even to the mean-
est capacity” (Hobbes 1994a: xv, 35), or the Categorical Imperative, or Grotius’s
injunction that we not deprive others of what is theirs. Of course, these for-
mulations are not self-interpreting; they require judgment, and there can be
reasonable disagreements about what they require. But the point remains that
they require no special skill or controverted religious or cultural tradition that
anyone subject to them cannot reasonably be supposed to share.
Seen in this light, Grotius’s a posteriori arguments for putative laws of nature
take on a new significance. A showing that a putative law, say, the obligation to
keep promises, has “the Consent of all Nations” (Grotius 2005: I, 160–161), that
it is part of a consentium gentium, is not just evidence that there are good rea-
sons to keep promises. It is an argument that this is something we can reason-
ably demand of one another because we can reasonably expect (epistemically
and morally) that others will reasonably accept it also.
So viewed, a “science” of morality is a distinctively modern project; it is an
essential element of a mutually accountable social order having the same pub-
licity requirements as the rule of law ordinarily so called.36

Sociability as the “Fountain of Right”

How are we to understand Grotius’s claims that “Sociability” is “the Fountain
of Right” (I, 85–86) and that what makes something in accord with or contrary
to natural law is its “Fitness or Unfitness … with a reasonable and sociable
Nature” (I, 159)? The first statement of these ideas comes just after Carneades’s
challenge, to which Grotius replies by saying that human beings have a “Desire
of Society” (I, 79). It is, however, important to understand just what Grotius
takes our “sociable” nature to be, such that it might be able to ground natural
law, as he understands it, that is, as a genuinely obligating law.

36
For further elaboration of this point see Chapter 12 of Darwall (2006), especially the discus-
sion of Bentham.

https://doi.org/10.1017/9781139025065.003 Published online by Cambridge University Press

 sociability as the “fountain of right” 37

It seems clear, first, that sociability cannot be anything like benevolence, or

the desire for the good of others or even of all, for at least two different reasons.
First, even if such a desire were universal in the human species, it would be
unable, by itself, to ground the idea that we are under any obligation to promote
human welfare, that is, that we are responsible or accountable for doing so.
In being benevolent, we see the flourishing of human beings as a good thing,
but that is, by itself, insufficient to warrant the thought that we are obligated
to bring this about, that it is something that can be legitimately demanded of
us. Benevolence as such is focused on the good, not on the right, so it can only
ground apparent counsel, not law.
Second, the very same form of collective action problem that arises with
respect to individual good also arises with respect to the overall good. There
are cases, notoriously those involving justice or fairness, where an agent could
produce more overall good by violating a norm of justice as, for example, in
Hume’s famous example of restoring stolen property to a “seditious bigot.”37
Neither can a desire to live among or in agreement with other human
beings, or the Stoic idea that affiliation with other human beings is part of our
common good, help to ground obligation and right. Like any consideration
of good, the most such considerations can justify is some form of counsel, not
obligating law.
Grotius is actually quite specific in defining the precise kind of “Desire of
Society” that he has in mind.
[A] certain Inclination to live with others of his own Kind, not in any
Manner whatsoever, but peaceably, and in a Community regulated
­according to the best of His Understanding (I, 80–81).

To this he adds that mature human beings develop a “peculiar Instrument”

that is necessary for such a “Community,” namely, “the Use of Speech” (I,
85). And humans have also the related “Faculty of knowing and acting,
according to some General Principles” (I, 85). Other animals live together,
and many seem capable of acting out of something like affection or concern
for at least some others of their kind. What is distinctive about human beings
in this regard is their capacity for and disposition towards a particular kind
of social order, namely, one mediated by the common acceptance of formu-
lable, public “General Principles.” Human beings thus have a capacity for and
drive toward a distinctive kind of society, namely, “A Society of reasonable
Creatures” (I, 136).
Recall the passages we considered earlier about the distinctive relations
that define such a society, for example, Florentinus’s remark that “Nature has

37
“When a man of merit, of a beneficent disposition, restores a great fortune to a miser, or
a seditious bigot, he has acted justly and laudably; but the public is a real sufferer” (Hume
1978: 497).

https://doi.org/10.1017/9781139025065.003 Published online by Cambridge University Press

 38 grotius

founded a kind of Relation between us” (I, 136). The examples that Grotius gives
of social relations are all, again, relations of relative authority, involving stand-
ings to make claims and demands and to hold accountable, whether the relations
are reciprocal (“Right of Equality”) or not (“Right of Superiority”) (I, 136–137). If
we put these passages together with those quoted in the last paragraph concern-
ing the “Social Faculty” (I, 87), what we get is a conception of sociability as the
human capacity for and disposition toward a distinctive kind of social order,
namely, the very kind of order that Grotius is himself trying to found in RWP.
So understood, sociability is the fountain of an order involving all persons that is
mediated by universal “General Principles” enshrining rights with which indi-
vidual members take themselves to have standing to demand compliance.
This suggests that we might best understand Grotius as holding that the law
and right of nature are grounded in the capacity of rational sociable persons to
recognize their common competence and authority to make reasoned claims
and demands against one another and to live with one another on terms that
respect this common standing. Of course, this does not tell us much about how
such a rationale might work in any detail. But it does point us toward a way of
thinking about the modern conception of natural law (morality) that appre-
ciates its legal/juridical aspect and, consequently, the distinguishing form the
Carneadean challenge must take once there is a commitment, as Sidgwick says,
to a source of normative reasons that is independent of self-interest.38
To sum up this chapter, Grotius put forward a moral conception according
to which morality or natural law: (i) has a distinctively obligating normativity
that cannot be reduced to normative reasons concerning the good that advise
or “counsel” action rather than demanding it; (ii) includes basic universal
claim rights, what Grotius dubs “perfect rights” (thereby initiating the perfect/
imperfect right and perfect/imperfect duty distinctions); and (iii) would exist
and obligate us “even if we were to suppose … that there is no God” (I, xxiv).
This conjunction of claims played a significant role in shaping “modern moral
philosophy.” (i) and (iii) together lead to Anscombe’s Challenge.
Moral philosophers who followed were unquestionably influenced by Gro-
tius’s approach to the subject. Although Grotius’s answer to Anscombe’s Chal-
lenge – that sociability is the “Fountain of Right” – was not explicitly taken up
by many who directly followed him, we shall nonetheless see in the next chap-
ter how a similar conception of sociability entered into Pufendorf’s thought,
although not as a fundamental ground. Whether or not modern moral philoso-
phers have accepted Grotius’s answer, many have clearly accepted the combi-
nation of Grotian views that make them subject to Anscombe’s Challenge. It is
a challenge we have been grappling with ever since Grotius.

38
This idea has clear affinities with Scanlon’s contractualist ideal of the moral community as
one of mutual justification (Scanlon 1998; see also Darwall 2006).

https://doi.org/10.1017/9781139025065.003 Published online by Cambridge University Press

 2

Hobbes and Pufendorf

If it is widely believed that modern ethical thought began with Hugo Grotius,
there is a virtual consensus that modern political thought began with Thomas
Hobbes (1588–1679). The Hobbesian project of grounding political authority
on the sparest assumptions consistent with the emerging modern sciences and
without the “Summum Bonum … of the old moral philosophers” seemed to
mark a new departure in political philosophy that was, if anything, even more
recognizable and radical than the Grotian turn in ethical philosophy overall
(Hobbes 1994a: XI.1).
In this chapter, we shall be concerned with Hobbes’s similarly spare approach
to ethics. Hobbes defines “moral philosophy” variously as “the science of the
laws of nature,” “the science of … good and evil,” and “the science of virtue and
vice” (XV.40). What makes moral philosophy a “science,” in Hobbes’s view, is
its concern with “knowledge of consequences” (V.17), where “consequence”
means conceptual or logical entailment rather than causal effect. “Ethics,”
Hobbes says, draws “[c]onsequences from the passions of men” (IX.table).
“Science” is the product of “reason,” which is “nothing but reckoning …
of the consequences of the general names agreed upon for the marking and
signifying of our thoughts” (V.2). All thoughts originate in sense, whose
cause is invariably some “external body, or object which presseth the organ
proper to each sense” (I.2, 4). We use names to signify our experiences and
“cogitate[e]” concerning their causes and effects (I.2; IV.3). The form of rea-
soning involved in Hobbesian science is not, however, inductive cogitation;
it is drawing out deductive consequences of agreed “general names” signify-
ing thoughts that originate in sense experience.
To this concept empiricism, Hobbes adds a materialist metaphysics. To have
meaning and reference, names must refer either to some “matter or body,” to
“some accident or quality which we conceive to be in [bodies],” to “proper-
ties of our own bodies” such as sensings or imaginings (“fancies”), or to other
names (IV.15–18). Where, however, can ethical thought and language fit into
such a materialist, nominalist picture? It would seem that the most that a
Hobbesian science could tell us is what is the case, not what ought to be or what
we ought to do.
39

https://doi.org/10.1017/9781139025065.004 Published online by Cambridge University Press

 40 hobbes and pufendorf

Hobbes makes it clear that he rejects the central concept of Aristotelian tele-
ological metaphysics. “A final cause has no place but in such things that have
sense and will” (Hobbes 1839: I.131), and here final causes are to be assimilated
to efficient causes provided by desires and aversions (1994a: VI.2). It appar-
ently follows that the language of Aristotelian naturalism can be no more than
“insignificant words of the Schools” (IV.1).
If Hobbes was like Grotius in aiming to overthrow Scholastic orthodoxy,
he was also like him in wanting to secure ethical and political thought on new
foundations. Moreover, Hobbes shared the Grotian project of grounding a dis-
tinctively juridical ethical conception – a “science of the laws of nature” – a
“modern moral philosophy” or “science of morality” (Barbeyrac 1749). Finally,
Hobbes and Grotius were united in their concern to defend “minimalist” foun-
dations for moral and political thought that could survive strong evaluative
disagreements, whether international, as with Grotius, or intra-national, as
with the religious differences of the English Civil War that Hobbes hoped to
overcome.
But here the similarities end. Hobbes’s empiricism and materialism impose
philosophical constraints on his thought that go well beyond anything limit-
ing Grotius. Grotius was a philosophical lawyer who began with agreed legal
doctrines, a consentium gentium, and sought deeper rationales. Hobbes, by
contrast, came to moral and political philosophy with prior commitments in
metaphysics, epistemology, and the philosophies of language and mind. These
constrained not just his method, but also his views about how ethical and polit-
ical thought and their respective objects might be so much as possible.
This can make Hobbes’s thought especially relevant to us today. Hobbes’s
materialism and empiricism force him to confront metaethical questions about
the metaphysics of ethics, and related issues in the philosophies of language
and mind, in something like the form they take in contemporary discussions.
Moreover, Hobbes’s own views bear an interpretation that is not unlike so-
called anti-realist or expressivist positions in contemporary metaethics.
One obvious difference between Grotius and Hobbes concerns sociability.
Both Hobbes and Grotius are “methodological individualists” in political phi-
losophy in the sense that both seek to ground political authority in the per-
spective of independent individuals. But Hobbes’s individualism runs deeper.
Whereas Grotius arguably takes human sociability to entail a basic interdepen-
dent normative authority that can ground reciprocal claim rights (the “Foun-
tain of Right”), Hobbes’s fundamental commitments rule out any such mutual
normative standing. Individual human beings no more instantiate natural nor-
mative relationships than do “mushrooms” “sprung out of the earth” (Hobbes
1983: 8.1). Obligations come into existence only as the result of individuals’ vol-
untary acts “when a man hath … abandoned or granted away his right” (1994a:
XIV.7). “There [is] no obligation on any man which ariseth not from some act

https://doi.org/10.1017/9781139025065.004 Published online by Cambridge University Press

 hobbes and pufendorf 41

of his own” (XXI.10). Absent such undertakings, there is nothing in a state of

nature to ground reciprocal obligations and claims.
The natural human condition, Hobbes says, is a basic right of liberty, not of
claim.

The Right of Nature … is the liberty each man hath to use his own
power, as he will himself, for the preservation of … his own life, and conse-
quently of doing anything which, in his own judgment and reason, he shall
conceive to be the aptest means thereunto (XIV.1).

Hobbes does say that “laws of nature” bind individuals even in the state of
nature. But though he also says that these natural laws and the right of nature
differ “as much as obligation and liberty,” this juridical language substantially
softens under analysis. First, in the state of nature when one cannot be assured
that others will conform to natural laws, such as the first, to “seek peace,” or the
third, to “perform their covenants made,” the laws of nature obligate only “in
foro interno, that is to say, they bind to a desire they should take place” (XIV.4;
XV.1, 36). And second, Hobbes says that the laws of nature are called “law, but
improperly; for they are but the conclusions or theorems concerning what con-
duceth to the conservation and defence of themselves” (XV.41).
Taken literally, this last statement reduces laws of nature to empirical gener-
alizations that lack any explicit ethical or normative force. Such generalizations
are philosophically respectable by Hobbes’s empiricist, materialist lights. But
the cost seems to be that anything distinctively ethical or normative has been
reduced away. Again the problem arises: If language and thought can only gain
content by referring to material things and properties we are aware of through
sense experience, how can there be meaningful ethical language and thought?
As a modern, indeed, Hobbes actually faces two distinct problems here. How
can there be contentful claims and thoughts about the good? And how can
there be such claims and thoughts about obligation and the right?
Samuel Pufendorf (1632–1694) will give us an especially apt comparative
perspective in this chapter. Like Hobbes, Pufendorf is concerned with plac-
ing morality (“moral entities”) in relation to an empirically graspable realm
of “physical entities” (Pufendorf 1934: 5). But Hobbes and Pufendorf disagree
radically about what brings juridical morality into play. Pufendorf is a theo-
logical voluntarist who holds that obligations all derive ultimately from “impo-
sition” by God’s authoritative command. For Hobbes, by contrast, it is only by
voluntarily laying down natural liberties that individuals become obligated to
one another.
There is actually an issue for Hobbes here that Pufendorf more actively con-
fronts. How exactly is it possible to “lay down” a right and thereby create an
obligation? As we shall see, Hobbes seems simply to assume the existence of
what Pufendorf calls “moral powers” (“normative powers,” in contemporary

https://doi.org/10.1017/9781139025065.004 Published online by Cambridge University Press

 42 hobbes and pufendorf

terms), the voluntary exercise of which can put us under obligation (Raz 1972),
like the power to promise or contract. It turns out that Pufendorf has thought
through what such powers must consist in and imply in ways that Hobbes
apparently has not.
It is important to see that Hobbes and Pufendorf agree about the fundamen-
tal role of normative “moral powers” in underwriting moral obligations. For
Hobbes, genuine obligations can arise only from exercising a power to under-
take them voluntarily by laying down rights (liberties). And for Pufendorf,
all moral obligations derive from God’s “imposition” through exercising his
moral power as a superior. As we shall see, both ultimately face the problem
Cudworth raised for all such voluntarist views, namely, that the very existence
of moral powers may presuppose moral obligations already in the background.
Pufendorf also provides an interesting counterpoint to Hobbes on socia-
bility. Pufendorf agrees with Hobbes, against Grotius, that no natural basis
for reciprocal claim rights can exist without the exercise of moral powers.
According to Hobbes, we become obligated to one another by laying down
rights through covenant in establishing the sovereign’s authority. According
to Pufendorf, all obligations ultimately derive from God’s exercising his moral
power of authoritative command. But despite agreeing on this, Hobbes and
Pufendorf fundamentally disagree about the natural moral state human beings
are in in relation to one another before any human beings have exercised their
moral powers on each other.
According to Pufendorf, the fundamental law of nature derives from God’s
commanding us to “cultivate and preserve towards others a sociable attitude,”
where this involves viewing one another with mutual respect as having a dig-
nity that grounds equal basic reciprocal obligations and rights (Pufendorf
1934: 208). In effect, Pufendorf’s position is that our fundamental obligation
to God is to view one another as sociable in Grotius’s sense, that is, as sources
of mutual obligations and claim rights, what Rawls will call toward the end of
the twentieth century “self-originating source[s] of valid claims” (Rawls 1980:
546). According to Hobbes, however, human beings can have no obligations to
or rights against one another without voluntarily laying down or transferring
natural liberty rights. Pufendorf is a sharp critic of this Hobbesian position.
These interrelations between Grotius, Hobbes, and Pufendorf crystallize in
their respective views of war and peace. Grotius’s founding work is titled The
Rights of War and Peace. And Hobbes’s first law of nature is “to endeavour
peace” so long as there is hope of “obtaining it,” but otherwise to “seek and use
all helps and advantages of war” (Hobbes 1994a: XIV.4). Pufendorf, for his part,
defines peace as the state human beings are in when they bear one another a
reciprocally respectful, sociable attitude:
Peace is that state in which men dwell together in quiet and … render their
mutual dues, as of obligation … (Pufendorf 1934).

https://doi.org/10.1017/9781139025065.004 Published online by Cambridge University Press

 Hobbes 43

However, Grotius, Hobbes, and Pufendorf exemplify three fundamentally

different positions on this topic of common vital concern. Grotius holds that
human beings naturally instantiate a basic normative sociable, hence peace-
ful, relation independently of any voluntary undertaking or the exercise of
any moral power. Pufendorf agrees with Grotius that such a basis for mutu-
ally respectful peaceful relations exists independently of any voluntary human
undertaking, but nonetheless agrees with Hobbes, and against Grotius, that it
takes a voluntary action – God’s exercise of his moral power of command – to
ground this natural sociable state. Finally, Hobbes holds that the state of nature
is not one of reciprocal obligation and right (so, is not a Pufendorfian state
of peace) in any sense, thus disagreeing with both Grotius and Pufendorf. As
Hobbes sees it, the only way peace can be achieved is through human covenants
in which people lay down their (liberty) right to all things and authorize and
agree to obey a sovereign who can provide the background institutional condi-
tions necessary to make peace possible.

Hobbes
We turn now to Hobbes’s attempts to work out an ethical philosophy whose
tenets can be both “significant” and philosophically defensible.1 Recall that
Hobbes treats “ethics” and “moral philosophy” as sciences, therefore, as ratio-
nal inquiries into the “consequences” of the “general names” with which they
are concerned. Ethics seeks “consequences from the passions of men.” And the
part of moral philosophy that concerns obligations and claim rights, “the Sci-
ence of Just and Unjust,” seeks “Consequences from speech … [i]n contract-
ing” (Hobbes 1994a: IX.table).2
We can put the two questions about the good and the right we raised above
in Hobbes’s scientific idiom. It would seem, first, that any consequences that
might follow from “the passions of men” can only concern the nature and cir-
cumstances of human passions in fact. But claims about the good, for example,
are not claims about what people actually desire; they are about what is worthy
of desire, what is desirable or there is reason to desire. So if ethics concerns
consequences of the passions, and these include only what follows from our
passions in fact, how can genuinely ethical claims about the good be part of a
philosophically respectable ethics, as Hobbes understands it?
Similarly, second, Hobbes holds that strictly juridical matters of obligation,
claim right, and justice arise only through contract. Since the very idea of a
promise or contract is to undertake some kind of obligation, Hobbes might

1
“Significant” in Hobbes’s sense of meaningful and securing reference to the material world.
2
See Pettit 2009 for an account of Hobbes that stresses this aspect. Cf. the discussion of the
role of language in Grotian sociability in Chapter 1.

https://doi.org/10.1017/9781139025065.004 Published online by Cambridge University Press

 44 hobbes and pufendorf

seem to be on firmer ground in holding that the “science of just and unjust”
consists of “consequences from speech in contracting.” Even so, a dilemma
looms: either whether a promise or contract has been executed is itself a norma-
tive matter or it is not. If it is, that will assure normative “consequences” from
contracting, but then it will be hard to see how contracting can itself consist in
the simple holding of material facts concerning speech and social convention.
On the other horn, the more empirically and materialistically respectable we
make contracting, the harder it will be to see how any genuinely obligating nor-
mativity can follow from it. And there will remain as well the further problem I
alluded to briefly above concerning how to understand the normative power to
contract within the normative materials that Hobbes’s state of nature provides.
We shall investigate these matters in turn presently. When we get to Hobbes
on the right, it will be important to bear in mind that although Hobbes’s official
view of obligation is that it can arise only through a voluntary laying down of
right, as in contract, Hobbes also sometimes says that laws of nature obligate,
and he defines natural laws in explicitly deontic terms. A law of nature, he says,
“is a precept or general rule, found out by reason, by which a man is forbid-
den” to act in certain ways (as it happens, “to do that which is destructive of his
life or taketh away the means of preserving the same”) (XIV.3). “Forbidden”
is a deontic term. So there will be two separate problems of right. How can
any precept that can forbid action follow from the passions of men? And how
can obligation, more strictly so called, follow as a consequence of “speech in
contracting”?

Desire and the Good

We turn now to Hobbes on the good. The dominant view has been that Hobbes
has a very simple way of answering the question of how claims about the good
can be consequences of human passions, namely, by holding that these claims
are about the desires of those who make them. A common view is that Hobbes
is a value subjectivist (Gauthier 1969: 7; 1979: 548; Hampton 1986: 29; Kavka
1986: 47). When Hobbes writes that “[g]ood and evil are names that signify our
appetites” (Hobbes 1994a: XV.40), and that whatever a person desires they
“calleth good” (XV.7), he is frequently read as saying that “good” means some-
thing like “desired by me.”
But although this makes good sense of Hobbes’s claim that ethical proposi-
tions of the good concern consequences of human desires, it gives him the bur-
den of explaining how facts about an agent’s psychology can have normative
practical force. Why does the fact that I actually desire something make it the
case that there is a reason for me to desire and seek it? That I do desire some-
thing seems to be one thing; that it is desirable or good, something I should
desire or seek, another. Reductionism seems to change the subject from ethics
to psychology.

https://doi.org/10.1017/9781139025065.004 Published online by Cambridge University Press

 desire and the good 45

There is a further problem with subjectivism that must also infect a subjec-
tivist interpretation of Hobbes. When two people, A and B, say of something,
X, that it is good and that it is not good, respectively, they seem to disagree
about some matter concerning which both cannot be correct. If subjectivism is
true, however, what A really says is that she, A, desires X, and what B says is that
he, B, does not. But if this is so, then there is no issue on which A and B differ.
A will not be affirming what B denies, nor vice versa. Moreover, subjectivism
is not simply problematic in its own right. It is especially dubious as an inter-
pretation of Hobbes, since Hobbes says that disagreements about good and evil
can be grave enough to lead to war (XV.40).
A major reason given for attributing a subjectivist view to Hobbes, again, is
his saying that what we desire we call good, and that what we are averse to or
“hate” we call evil (VI.7). But this no more supports subjectivism than would
saying that whatever we believe we call true support subjectivism about truth.
No one would claim that when we say that something is true, we are saying that
we believe it. That would run over a distinction between expressing and self-
attributing a belief, just as the subjectivist interpretation of Hobbes’s text blurs
the analogous distinction between expressing and self-attributing a desire.
It is more charitable, both philosophically and interpretatively, to read
Hobbes as holding something closer to what is nowadays called expressivism
or projectivism than it is to interpret him as a metaethical subjectivist (see, e.g.,
Blackburn 1984; Gibbard 1990).3 When people say that something is good, they
are not saying that they desire it; they are expressing their desire for it and, per-
haps, projecting a property (goodness) onto the object of their desire in some-
thing like the way that Hobbes holds, following Galileo, that our color sense
projects colors onto objects: “[T]he real and very object seems invested with the
[color] fancy it begets in us” (Hobbes 1994a: I.4).
Hobbes explicitly analogizes the case of value to that of color. Though there
is nothing more to color metaphysically, Hobbes says, than “motion caused
by the action of external objects,” these motions are nonetheless “to the sight”
as “light and colour” (VI.9). Similarly, Hobbes holds that although desires are
similarly motions in our bodies, these also have appearances, in this case as of
an object’s having the property of being good.
Chapter VI of Leviathan, “Of the … Passions, and the Speeches by Which
They Are Expressed,” contains a careful analysis of “the Interiour Beginnings
of Voluntary Motion” in appetites and aversions. Desires always have some
“fancied” object toward which they move us (VI.1, 2). The fancy (imagining of
an object) that accompanies desire is “delight” or “pleasure” (VI.10).
As in sense that which is really within us is … only motion caused by
the action of external objects (but in appearance, to the sight, light and

3
In the following paragraphs, I draw from Darwall 2000.

https://doi.org/10.1017/9781139025065.004 Published online by Cambridge University Press

 46 hobbes and pufendorf

colour …) so when the action is continued … to the heart, the real effect
there is nothing but motion or endeavour [desire or aversion] … But to
the appearance … is that we either call Delight, or Trouble of Mind
(VI.9).

This delight or pleasure, Hobbes says, is “the appearence, or sense, of good;

and molestation or displeasure, the appearence, or sense, of evil” (VI.11). Just
as color sensation projects apparently real, but non-existent, color proper-
ties onto objects, so too do desire and aversion project the ethical properties
of goodness and evil onto their respective objects. When I see a banana, the
material thing before me causes me to project the property of yellow onto the
banana, as though it had this property intrinsically. Similarly, when the sight of
a banana stimulates a desire in me to eat it, that causes me to see the banana as
though it, or at least eating it, has the property of goodness intrinsically.
This interpretation explains how Hobbes can hold that disagreements about
the good are capable of leading to “controversies, and at last war” (XV.40).
According to a subjectivist account, people who utter apparently conflicting
sentences about the good are not really expressing disagreement; to the con-
trary, they agree about their disagreeing attitudes. But on an expressivist or
projectivist interpretation, this is not true. People uttering such sentences are
expressing their conflicting desires, hence their disagreement. Such passionate
disagreements can indeed lead to war.
But what about Hobbes’s claim that “[g]ood and evil are names that signify
our appetites and aversions” (XV.40)? Here again, Hobbes treats claims about
good and evil as being on par with those involving color. With color, Hobbes
says, “we bring into account the properties of our own bodies whereby we make
such distinction” (IV.17). Similarly, all that can “enter into account” for the
good metaphysically are material motions, in this case, desires and aversions
(IV.15). Nevertheless, when we judge objects to be colored or good, we do not
judge or see them as causing these material motions; we see them as though
they have an intrinsic property that Hobbes believes they do not actually pos-
sess, in one case a phenomenal color property, in the other, an evaluative or
normative property.4

4
Hobbes also says that “good” and “evil” are “ever used with relation to the person that useth
them,” and that although there is no rule of good inherent in “the nature of objects them-
selves,” a rule can be given by “the person of the man” (VI.7). Moreover, Hobbes frequently
talks about good and evil in an explicitly relativized way. “The voluntary acts of every man,”
he says, invariably aim at “some good to himself” (XIV.8). And he defines benevolence
as “desire of good to another” (VI.22). Don’t all these remarks point in the direction of
subjectivism rather than projectivism? Not necessarily. If speakers use “good” and “evil”
to express their desires, then they use them “with relation to” themselves even if they do
not assert some relation to themselves (as subjectivism requires). When Hobbes speaks of
things being good to people, this can easily be interpreted as referring to what people think

https://doi.org/10.1017/9781139025065.004 Published online by Cambridge University Press

 desire and the good 47

Would not, however, knowledge of the projective character of ethical judg-

ments undermine ethical thought and practice? Should not Hobbes have
believed that in putting forward a projective theory of value judgment he was
placing a significant obstacle in his readers’ way in accepting the normative
propositions he wished to convince them of in Leviathan?
Hobbes does not directly confront this question, but he might answer it by
arguing that since evaluative thought is always an expression of desires, it is
simply unavoidable for us so long as we have desires. “[L]ife itself is but motion,
and can never be without desire” (VI.58). Once we are in the state of desire, we
are per force deliberating and subject to deliberative ethical thought.5
All deliberation begins with desire (VI.49). And “[f]rom desire, ariseth the
thought of some means we have seen produce the like of that which we aim at
… till we come to some beginning within our own power” (III.4). “[I]n delib-
eration,” Hobbes adds, “the appetites and aversions are raised by the foresight
of the good and evil consequences and sequels of the action whereof we delib-
erate” (VI.57). The relevant evaluative “foresight” must itself be provided by
desires that are upstream in the deliberative process. Under the influence of the
desires that begin deliberation, we are disposed to see some consequences of
alternative actions as good and others as evil, and these thoughts then give rise
to further (instrumental) desires leading ultimately to the “last appetite” (will
or volition) and finally to action.
In effect, Hobbes is working with a distinction between theoretical and
practical standpoints. Thought and discourse about good and evil encode
an agent’s view from the deliberative perspective their desires provide. Thus
although Hobbes calls “moral philosophy … the science of what is good and

good or what appears good to them. And we can understand Hobbes’s claim that voluntary
action invariably aims at some good to the agent as saying that voluntary action invariably
aims at something the agent thinks good or that seems good to her. Since Hobbes holds
that all voluntary motions result from desires and aversions, it will be true that all voluntary
motion aims at something that seems good to the agent. As opposed to “vital motion,” “ani-
mal” or “voluntary motion” invariably involves a “fancy” or “imagination,” a “precedent
thought of whither, which way, and what” (VI.1), which is “the appearence, or sense, of
good” (VI.1,11).
5
Even if we cannot stop it appearing to us as though some things are good and ought to be
brought about, might we nonetheless become alienated from these appearances? Compare
the case of color. A projectivist about color can hold that although, strictly speaking, all
color judgments are false, our practical purposes are nonetheless better served by speaking
and thinking as though they are not, normalizing our color judgments to the experiences
of the normally sighted under normal conditions. We can easily imagine a philosophically
sophisticated, projectivist interior decorator whose “first-order” thought and speech about
color are regimented in this way. Such a philosophical position would apparently pose no
obstacle to facility, even expertise, in color judgments for practical purposes as a decorator.
Similarly, Hobbes might reply that the point and function of ethical terms and concepts is
practical – even more so, indeed, than those of color.

https://doi.org/10.1017/9781139025065.004 Published online by Cambridge University Press

 48 hobbes and pufendorf

evil” (XV.40), he must have held that it is a science we engage in from a practi-
cal point of view, as we discover how the consequences to be drawn from “the
passions of men” bear on the ends that drive our deliberative thought.6
Since the good is the currency of desires, and deliberation is driven by desires,
the good must be the currency of practical reason. Where, then, can Hobbes
locate the right or judgments of right? Considerations of right can provide nor-
mative reasons for action only under “the guise of the good.” This would not
rule out considerations of right from being or entailing normative practical
reasons. It would just mean that it is impossible for someone to take deontic
considerations as reasons and act on them without a desire whose appearance
and expression is that such action would be good.

Obligation and the Right

It is common to interpret Hobbes as holding that the normativity of the “laws of
nature” ultimately derives from how conformance (or perhaps the disposition
to conform) advances agents’ self-preservation or good (Gauthier 1967, 1979;
Hampton 1986; Kavka 1986, 1995; Darwall 1995). This interpretation is encour-
aged by Hobbes’s saying that laws of nature are really “theorems concerning
what conduceth to the conservation and defense of themselves” (Hobbes 1994a:
XV.41). But while such good-based, instrumentalist approaches can plausibly
claim to ground a kind of normativity for the laws of nature, it is ultimately
one of the good. Such approaches cannot explain how Hobbes might account
for any deontic sense in which someone might be “forbidden” to act against
self-preservation.
We shall return to how Hobbes might reasonably hope to provide a more
satisfyingly independent metaethics of right, one that is similar to his metaeth-
ics of good, presently. First, however, we should consider what Hobbes says
about the distinctive deontic claims of right to which he is explicitly committed,

6
This does not mean that Hobbes thought his readers’ ethics would have been entirely unaf-
fected by accepting his analogy between color and value. To the contrary, Hobbes must
have thought they would be affected and for the good. Hobbes was a notorious critic of reli-
gious “superstition” and its intellectual expression in the metaphysics of the schools. When
superstitious believers expressed their ethical convictions, he thought, they were simply
giving voice to desires, albeit ones highly disciplined by religious ritual and practice. The
faithful themselves, however, likely saw both their desires and their ethical convictions as
responding to real objective values or “final causes.” Hobbes of course thought that this
was nothing but “insignificant speech” and that nothing stood behind their ethics but the
desires of which their thought was an appearance and their discourse an expression. Even
so, he might also have believed that their ethics were, at least partly, “ideological” in the
sense that some of the desires their ethics expressed would not have existed but for their
superstitious metaphysical beliefs. Convincing his readers of an expressivist metaethics of
the good might help unmask and overturn their ideological ethics.

https://doi.org/10.1017/9781139025065.004 Published online by Cambridge University Press

 obligation and the right 49

namely, the obligations that are created by voluntarily laying down or transfer-
ring rights.7
It is the right of nature that individuals are said to lay down in creating the
sovereign’s authority by contract. Although Hobbes holds that it is only by
laying down this right that genuine obligations are created, the right of nature
is itself a putative deontic moral fact. It is a moral liberty. This is brought out
by Hobbes’s De Cive formulation that it is not “reprehensible … for a man
to use all his endeavours to preserve and defend his body” (1983: I.7), and by
his calling the right of nature a “blameless liberty” (1994b: 14.6). Even if the
laws of nature are ultimately to be interpreted in non-deontic terms, the right
of nature itself purports to be a truth about how it is morally permissible for
agents to act.
Hobbes is clear, however, that when the right of nature is abandoned or laid
down, this gives rise to genuine obligations. Hobbes defines obligation, indeed,
as the upshot of such voluntary undertakings: “When a man hath in either
manner abandoned or granted away his right, then he is said to be Obliged
or Bound not to hinder those” from benefiting from the right (1994a: XIV.7).
The question we must now investigate is: How can Hobbes account for such
voluntarily created obligations within his metaethical framework?
Hobbes gives different answers to this question in his earlier works, De
Cive and The Elements of Law, and in Leviathan, respectively.8 The key to
his earlier answer has to do with conceptual connections Hobbes makes
between obligation, liberty, and deliberation, and with the idea, which he
abandons in Leviathan, that when people contract to do something, they
thereby conclude deliberation about whether to do it. Hobbes’s later answer
in Leviathan is what scholars have devoted the greatest attention to, inter-
preting it either in instrumentalist terms (Gauthier 1969, 1979; Hampton
1986; Kavka 1986; Darwall 1995) or as grounded in a fundamental “reciproc-
ity theorem” (Lloyd 2009).
A central move in Hobbes’s earlier works is the idea that when deliberation
has concluded in an intention not to perform some act, we are no longer free to
take it, and “where Liberty ceaseth, there beginneth Obligation” (Hobbes 1983:
II.10). We can begin to see how this is supposed to work with contracting by
considering what Hobbes says in De Cive about “natural obligation,” a species
of obligation he does not recognize in his later works:
[T]here are two Species of natural obligation, one when liberty is taken
away by corporall impediments. … The other when it is taken away by
hope, or fear, according to which the weaker despairing of his own power
to resist, cannot but yield to the stronger (1983: XV.7).

7
For an excellent account, see Abizadeh 2018.
8
The paragraphs to follow draw from Darwall 1995.

https://doi.org/10.1017/9781139025065.004 Published online by Cambridge University Press

 50 hobbes and pufendorf

The first sense does not even purport to be deontic. It is roughly the sense in
which we say we were “obliged” not to do something when circumstances pre-
vented us from doing it, which Hart famously contrasted with obligation (Hart
1961: 6–8). The second sense, however, is one that Hobbes intends for deontic
purposes, since he holds it to be the sense in virtue of which God has supe-
rior authority over us. What this authority consists in, according to Hobbes,
is God’s irresistible power over us, which we cannot but despair of opposing.
This gives Hobbes’s Leviathan claim, also in De Cive, that no human
being’s power is irresistible, since “the weakest man [can] kill the stron-
gest,” normative relevance (Hobbes 1983: I.3). Because of this natural fact, no
human being can be under natural obligation to any other. But this is not the
case with our relation to God: “there is no kicking against the pricks” (XV.7).
Although attempting to oppose human power is always an open deliberative
option, it can never be one with respect to God, since God controls the very
power with which we would try to resist. We therefore despair of opposing
God’s power, and this closes deliberation about whether to try to do so. And
when our deliberation with respect to an alternative is removed by despair,
we are no longer free to do that alternative. “He that deliberates is so farre
forth free” (II.8). The “last act of deliberating” is that “whereby the liberty
of non-performance is abolisht” (II.10). And “where Liberty ceaseth, there
beginneth Obligation” (II.10).
Hobbes’s problem, again, is to account for obligation, or at least obligation
judgments (similarly to judgments of good), within a materialistic framework.
We can at least see how this is supposed to work with the natural obligations
owed to God, though we may wonder how anything genuinely normatively
deontic has been accounted for. In his earlier works, Hobbes attempts to account
for obligations that contracts create along similar lines. In De Cive, Hobbes
holds that once a contract is executed, that closes the parties’ deliberations about
whether to perform as contracted, thereby limiting their liberty to do so.
The promises which are therefore made for some benefit received (which
are also Covenants) are tokens of the Will; that is, … of the last act of
deliberating, whereby the liberty of non-performance is abolisht, and by
consequence are obligatory; for where Liberty ceaseth, there beginneth
Obligation (II.11).

If an agent’s liberty not to perform is to be restricted by concluding a delibera-

tion, it cannot be by forming an intention to contract, but only by intending
to perform the contracted action. But Hobbes identifies these normally distin-
guished intentions: “For by contracting for some future action, he wills it done”
(III.3; see also II.14).9
9
Hobbes distinguishes the deliberation-concluding, hence obligating, will in contract from
a promise of future free gift, which fails to obligate. When one person promises to benefit

https://doi.org/10.1017/9781139025065.004 Published online by Cambridge University Press

 obligation and the right 51

Hobbes underscores this point with an analogy between injury, the break-
ing of a contract, and absurdity, the maintaining of contradictory proposi-
tions in disputation. “For by contracting for some future action, he wills it
done; by not doing it, he wills it not done, which is to will a thing done and
not done at the same time, which is a contradiction” (III.3; see also 1994b:
I.xvi.2).
De Cive’s account of the obligation created by contract fits well within
Hobbes’s materialistic framework, even if it is implausible on other grounds.
But Leviathan’s approach is quite different. Hobbes retains the thesis that con-
cluding deliberation restricts liberty, but he explicitly rejects De Cive’s analysis
of why that is. “Every deliberation is then said to end, when that whereof they
deliberate, is either done, or thought impossible, because till then we retain the
liberty of doing or omitting” (1994a: VI.52).10 Of course, Hobbes is not com-
mitted to thinking in De Cive that it is literally impossible to violate contract.
He treats the situation of a contracting party as that of an individual forming
an intention to do something in the future. Obviously, such an individual may
fail to follow through when the time comes. It follows that the only kind of
contract that can possibly limit the liberty to perform must be one to perform
immediately.
Gone also is any talk of natural obligation. Nor do we find Hobbes saying in
Leviathan that contract ends deliberation, thereby limiting liberty and, for that
reason, obligating performance. The contrast between non-obligating prom-
ises of free gift and obligating contract is described without any mention of
their respective relations to deliberation and freedom.11
Importantly, Leviathan gives a non-reductive definition of obligation as
the condition that results when a person has “abandoned or granted away his
right” (XIV.7). Since Hobbes defines “contract” as a “mutual transferring of
right,” that people are obligated to keep their covenants will simply follow as a
theorem of the “science of just and unjust” (XIV.11, IX.table).12

another in the future, receiving nothing now in return, not even a reciprocal promise, “a
promiser of this kind must be understood to have time to deliberate, and power to change
that affection as well as he to whom he made that promise, may alter his desert” (II.8). Since
“he that deliberates is so farre forth free,” he cannot be said to be obligated. But Hobbes
thinks this is not true with contracts. Contracts obligate precisely because they conclude
deliberation about performance.
10
But see LeBuffe 2003.
11
Hobbes does retain the analogy between injury and contradiction, but in more cautious
terms. He no longer says that a violated contract involves an absurd willing and unwilling
of the same action. Rather, the injury is “to undo that which, from the beginning, he had
voluntarily done,” which might plausibly be taken to mean that violating a contract is like
trying to undo it (1994a: XIV.7).
12
Hobbes reserves “covenant” for a contract to perform in the future in return for the other’s
earlier performance (XIV.11).

https://doi.org/10.1017/9781139025065.004 Published online by Cambridge University Press

 52 hobbes and pufendorf

Grounding the Right in the Good?

But although this is assured as a conceptual truth, Hobbes seems to appreci-
ate that more needs to be done to establish that covenanters are obligated in
any genuinely normative sense. No substantive normative claim can follow
from a definition. It is notable that in addition to this definition, Hobbes both
introduces as a third law of nature “that men perform their covenants made”
and then argues for this law across several of the most important paragraphs
in Leviathan (XV.1–6). These include the famous challenge of the fool that
seems, as we noted in Chapter 1, so clearly to echo Carneades’s skeptical chal-
lenge, which Grotius discusses. The fool argues that whether or not breaking
covenants “may be called injustice,” such actions are not against reason when
they are not against the agent’s good (XV.4). And if they are not against reason,
it would seem to follow that such actions are not violations of obligation in any
genuinely normative sense.
To focus the issue, Hobbes considers cases in which there is no reasonable
suspicion that the other covenanter will not perform their part as, for example,
when they have “performed already” (XV.5). Commentators generally agree
that Hobbes’s reply to the fool emphasizes the potential costs to the agent of
violating covenant, or of being someone who is disposed to do so, including
being excluded from valuable agreements as an unreliable partner, which
agreements Hobbes thinks offer the only exit from a violent state of nature.
There is also consensus that Hobbes grants the fool’s premise that, despite
these costs, there can be cases in which violating covenant actually can pay. But
Hobbes urges that even though unforeseeable circumstances might “turn” vio-
lation to the violator’s “benefit, such events do not make it reasonably or wisely
done” (XV.5). Being able to maintain one’s reputation while violating covenant
is something no one can “reckon upon” (XV.5).
One way to take Hobbes is as arguing that benefiting by violating contract
can never “reasonably be expected” (XV.7).13 To use contemporary jargon, it
can never maximize expected utility, or maximize the utility we should reason-
ably expect given our evidence, to violate contract, and because this is so, doing
so is always against reason. It is hard to believe, however, that this would always
be true. But even if there are exceptions when we would most reasonably expect
violation to be optimific, Hobbes might still argue that even in these cases, vio-
lation would nonetheless be imprudent.
The problem with doing what seems best by the light of available evidence
is that poor evidence sheds poor light. Hobbes seems willing to grant the fool
that if we could know that violating contract would be for our good, we should

13
Hobbes actually uses this phrase in a slightly different context later in the paragraph –
gaining sovereignty by rebellion – but to a similar purpose.

https://doi.org/10.1017/9781139025065.004 Published online by Cambridge University Press

 grounding the right in the good? 53

violate; it would be contrary to reason not to. The problem, he appears to be

saying, is that we are never able to know this, and the stakes are so high that
no matter how close to knowledge we may seem, we are never close enough to
warrant taking a chance (Kavka 1986, 1995).
So far, this is an argument for the third law of nature, “that men perform
their covenants made.” But how does this relate to any obligation that contract
might create? The right that covenanting parties in the state of nature are said to
“lay down” is the right of nature: the agent’s liberty “of doing any thing, which
in his own judgment, and reason, he shall conceive to be the aptest means” to
“the preservation of his own nature” and good (Hobbes 1994a: XIV.1). We have
just seen that because of the central role covenant plays in extricating human
beings from the state of nature and keeping them in civil society, the third law
of nature dictates keeping covenant even when contracting parties believe,
however reasonably, their interests would be better served by violating it.
It follows that the right of nature must be qualified for the case of contract.
When contract is involved, right reasoning in the service of the agent’s own
good is not then the same as what the agent “in his own judgment, and rea-
son … shall conceive to be the aptest means thereunto.” In this case, right rea-
soning recommends keeping contract regardless of what the agent believes,
even reasonably, the relation to be between doing so and their own preser-
vation and good. It follows, therefore, that when an agent contracts, it then
becomes true that the full and unqualified right of nature no longer applies to
the act they have contracted to perform. By covenanting a person does some-
thing that makes it no longer advisable for them to base their determination of
whether to keep covenant on their own preservation and good.
Hobbes might thus argue that there is a sense in which in the case of cov-
enant, the right of nature does not apply and is, therefore, “laid down.” And
because it is, the person ends up obligated in Hobbes’s official sense (as a theo-
rem of the science of just and unjust) with normative force being provided by
the third law of nature, backed by Hobbes’s reasoning in his reply to the fool.
While this line of reasoning goes some way to addressing the problem
Hobbes faces, there are various reasons why it cannot be fully satisfying. The
problem, recall, is how to account for the normativity of right in Hobbes’s
framework. The most significant difficulty with Hobbes’s reply as we have so far
interpreted it is that it remains grounded entirely in the agent’s good. Even on
the most sophisticated version, which argues that agents will do better if they
treat covenants as binding, and as giving them overriding reasons for acting,
even when the obligatory action goes against the agent’s good, normativity is
ultimately coming only from the good. It is being argued, in effect, that agents’
goods are best promoted when they reason as though they were bound by inde-
pendent normative constraints of right. But, as Prichard would later argue, this
seems an argument of the wrong kind to show that deontic moral requirements

https://doi.org/10.1017/9781139025065.004 Published online by Cambridge University Press

 54 hobbes and pufendorf

of right are genuinely valid and binding on their own. It just shows that we each
do better if we suppose that.

Irreducibly Deontic Laws of Nature?

But perhaps there is a way of interpreting Hobbes that avoids attributing to
him an implausible, arguably impossible, attempt to account for the normativ-
ity of the right (or judgments of the right) solely in terms of that of the good
(or judgments of the good). Sharon Lloyd argues that there is substantial tex-
tual evidence that Hobbes was committed to a fundamental ideal of reciproc-
ity or reasonableness that is like the idea of “the reasonable” as it operates in
the moral and political thought of John Rawls and T. M. Scanlon in the late
twentieth century (Lloyd 2009; Rawls 1971, 1980; Scanlon 1998). What Lloyd
calls Hobbes’s “Reciprocity Theorem” is a formulation of the Golden Rule that
Hobbes says summarizes the laws of nature: “Do not that to another, which
thou wouldst not have done to thyself” (Hobbes 1994a: XV.35). An even better
formulation for Lloyd’s purposes is: “Do not that to another, which thou think-
est unreasonable to be done by another to oneself” (XXVI.13).
So interpreted, the laws of nature embody constraints on reasonable con-
duct, where “the reasonable” is conceived not in terms of advancing the agent’s
good, even indirectly – what Rawls calls “the rational” – but within the irreduc-
ible conceptual space of the right. Conduct is reasonable or morally permis-
sible, on this view, when there is no reasonable objection to it. And conduct
is unreasonable, or morally wrong, when there is such reasonable objection,
when people could reasonably demand that one not do it – as Hobbes puts it in
terms of “the law of the Gospel,” “whatsoever you require that others should do
to you, that do ye to them” (XIV.5).
An interpretation of laws of nature in terms of reasonable requirement
or demand would explain why they are formulated in deontic terms as
“precept[s] … by which a man is forbidden to do that which is destructive of his
life” (XIV.3). It also puts them on the “law” side of Suárez’s law/­counsel ­contrast.
Hobbes himself draws the distinction between law and counsel ­somewhat dif-
ferently, since he prefers to reserve “law,” properly so called, for commands of
the sovereign (or God) whom we are “formerly obliged” to obey (XXVI.2).14

14
As Curley’s glossary makes clear, by “formerly obliged,” Hobbes means obliged by a for-
mer command. Cf. also: “They who less seriously consider the force of words, do some-
times confound law with counsel. … We must fetch the distinction between counsel and
law, from the difference between counsel and command. Now COUNSEL is a precept,
in which the reason of my obeying it is taken from the thing itself which is advised; but
COMMAND is a precept, in which the cause of my obedience depends on the will of the
commander. For it is not properly said, thus I will and thus I command, except the will
stand for a reason. Now when obedience is yielded to the laws, not for the thing itself, but

https://doi.org/10.1017/9781139025065.004 Published online by Cambridge University Press

 irreducibly deontic laws of nature? 55

His official use of “law” ties it conceptually to “obligation” in his technical sense
of the bond that derives from contracting. But that does not stop him from
using deontic language in defining laws of nature. Nor does it stop him from
holding that we are accountable for conforming to the laws of nature and that
violations are culpable. Indeed, he says that the reason the laws of nature must
be able to be summed into a readily usable formula like the Golden Rule is that
otherwise we could not be justifiably blamed for violations.
[T]o leave all men inexcusable, they [the laws of nature] have been con-
tracted into one easy sum, intelligible, even to the meanest capacity; and
that is, Do not that to another, which thou wouldest not have done to thyself
(XV.35).

To this he adds:
[H]e has no more to do in learning the laws of nature, but (when weighing
the actions of other men with his own, they seem too heavy) to put them
into the other part of the balance, and his own into their place, that his own
passions, and self-love may add nothing to the weight; and then there is
none of these laws of nature that will not appear unto him very reasonable
(XV.35).

To work out whether demands of others are reasonable or not, we must con-
sider whether we can accept others’ making the very same demands of us in
similar situations. Similarly, we should permit ourselves to do things to others
only if we are prepared to permit others to do the very same thing to us in simi-
lar conditions.
This also provides a plausible reading of the right of nature, the “blame-
less liberty” of action in the state of nature. As noted above, Hobbes writes
in De Cive that it is not “reprehensible … for a man to use all his endeavours
to preserve and defend his body” (1983: I.7). What the right of nature is say-
ing, on this interpretation, is that there is no reasonable objection – no one
can be justifiably blamed – for doing what they believe necessary to preserve
themselves in the state of nature (Lloyd 2009: 18–19). If others could be relied
on to treat each other in peaceful ways, then one would be bound to treat
them peacefully also. But given the insecurities of the state of nature, no one
can reasonably rely on this. And because they cannot, no one can reason-
ably demand that others forego doing what they think necessary to preserve
themselves.

by reason of the adviser’s will, the law is not a counsel, but a command, and is defined thus:
LAW is the command of that person (whether man or court) whose precept contains in it
the reason of obedience: as the precepts of God in regard of men, of magistrates in respect
of their subjects, and universally of all the powerful in respect of them who cannot resist,
may be termed their laws” (Hobbes 1983: XIV.1; see also Hobbes 1994a: XXV).

https://doi.org/10.1017/9781139025065.004 Published online by Cambridge University Press

 56 hobbes and pufendorf

So far, this is all evidence that Hobbes understands the laws of nature as
independent norms of right that people are appropriately held to and that
“oblige conscience” (Hobbes 1983: III.28). But what about evidence that seems
to count against this reading and in favor of a more orthodox instrumentalist
good-based interpretation: Hobbes’s reply to the fool and his doctrine that the
laws of nature do not bind in foro externo and are not genuine laws without the
backing of the sovereign?15
The first thing to notice is that even if Hobbes’s reply to the fool is that fools
would do better were they to pursue their interests indirectly and reason by
the third law of nature, thereby keeping contracts even when this is expect-
ably against their interest, this would not commit Hobbes to thinking that the
authority of natural laws derives from this fact. So far as the dialectic goes, this
reply could simply be one that proceeds from premises that Hobbes accepts
arguendo for dialectical purposes. On reflection, it is somewhat puzzling that
interpreters have given so much weight to this passage as evidence for Hobbes’s
most fundamental normative justificatory views. So far as the text goes, the
situation could be much the same as in Bishop Butler’s famous “cool hour”
passage, in which Butler assumes for the sake of argument something he actu-
ally denies, namely, that the authority of conscience depends on support from
self-interest.16 Moreover, there are ways of interpreting Hobbes’s reply within
the kind of interpretation we have just been considering.
Hobbes says that the fool “declares he thinks it reason to deceive those
that help him”; and again: “he declareth that he thinks he may with reason
do so” (1994a: XV.5, emphasis added). Why does Hobbes represent the fool
as “declaring” these things?17 If what were in question were simply how one
might most sensibly promote one’s interest, whether directly or indirectly, why
would Hobbes be talking about the fool’s declarations? Surely the fool would
try to cover his tracks and be more or less successful in doing so. If, however,
what is in question is a principle of right, of what one may legitimately demand
of people and what one may not, of what people may and may not do, then
this is indeed properly determined by what the fool is prepared to declare and
endorse publicly. If they necessarily concern mutual accountability, principles
of right are subject to a test of public endorsability by their very nature.18

15
Lloyd argues against the orthodox interpretation of Hobbes’s reply to fool in Lloyd 2009:
302–317.
16
Butler argues that even if reasons for acting were ultimately grounded in self-interest, there
would still be reason to be virtuous, since a virtuous life will give the virtuous agent the
best life possible (Butler 2017: XI.20). For discussion, see Darwall 1995: 265n–266n, and
Chapter 5 below.
17
For a different explanation see Hoekstra 1997. See also Lloyd 2009: 310–312 for a criticism
of Hoekstra that is closer to the view presented here.
18
On this point, see Darwall 2006: 313–315.

https://doi.org/10.1017/9781139025065.004 Published online by Cambridge University Press

 irreducibly deontic laws of nature? 57

Note next that what Hobbes then says is that a fool who declares the view that
covenants may be breached or falsely entered into when that turns out to be
useful can be accepted into society through covenant “only by the error of them
that receive him” (XV.5). The orthodox interpretation is that someone who
reasons egoistically rather than by the third law of nature will be received into
society only by others’ mistaking his real intentions. An alternative interpreta-
tion that is closer to the text, however, is that someone who puts forward what
the fool “declares” as a proposition of right cannot be so received, except by way
of a mistake not about the fool’s intentions, but about the obligation to keep
covenants and so about the right itself. So understood, the mistake that both
the fool and his potential covenanters make is not a strategic error in advancing
their respective interests; both make a mistake of right. A covenant, by its very
nature, is something that transfers a right; but no such transfer is possible if
both parties know that one of them has no intention of following through. Even
“reasonable suspicion” of non-performance is enough, Hobbes says, to render
a covenant “void” (XIV.18). In other words, the view that covenants may be
broken or entered into falsely for reasons of self-interest is itself mistaken as a
proposition of right. And anyone who attempts to covenant with someone who
declares such a view must mistake the nature of covenants, since covenants are
void on any “reasonable suspicion” of non-performance.
What then, about Hobbes’s view that the laws of nature bind to external
conduct but rarely in the state of nature and require the sovereign’s sanctions
to provide genuinely obligating law? Here again, it might seem that Hobbes
is saying that the normative force of any duty or obligation to follow natu-
ral law must derive from reasons of self-interest coming from the sovereign’s
sanctions.
Although Hobbes clearly does say that it takes the sovereign’s commands
to create genuine law, even to make the “improperly called” laws of nature
laws properly so-called, it is nonetheless of great importance to him that sub-
jects see themselves as obligated to follow the sovereign’s commands and not
just as having reasons of self-interest for doing so (1983: II.10; 1994a: XIII.13,
XXVI.passim). Sovereigns are established, as it were, either from below, by sub-
jects covenanting with one another to authorize and follow a sovereign (1994a:
XVII), or from above, by conquest. An obligation to follow the sovereign’s
commands follows directly from the covenant authorizing him. Even in con-
quest, Hobbes is clear that “it is not … victory that giveth the right of dominion
over the vanquished, but his own covenant,” which the vanquished enter into
when they submit to the conqueror and promise to obey as the price of survival
(XX.11). Hobbes distinguishes between the condition of a slave who is kept in
thrall by chains and threats of reprisals, on the one hand, and a “servant” who
“upon promise not to run away, nor to do violence to his master, is trusted by
him” (XX.10, 12), on the other. And he explicitly analogizes the case of servants
to that of submitting subjects.

https://doi.org/10.1017/9781139025065.004 Published online by Cambridge University Press

 58 hobbes and pufendorf

If masters and sovereigns genuinely trust covenanting servants and subjects,

respectively, they must think that believing oneself to be obligated can make a
difference in conduct that cannot fully be explained by motives of self-interest.
And this is far from the only place in which Hobbes relies on this difference
between motivation by self-interest and motivation by a sense of right. The
most central, of course, is in the making of the covenant itself. Since a covenant
involves a transfer of rights by its very nature, it is only possible if the covenant-
ing parties see themselves as giving themselves and one another a reason to
comply that is not reducible to self-interest. Or to put the point the other way
around, individuals who recognize and know one another to recognize only
reasons of self-interest cannot covenant.
There are many other places also where Hobbes recognizes the possibility of
conscientious conduct. Some particularly important examples for his political
philosophy are instances in which sovereigns’ commands conflict with sub-
jects’ consciences or, contrariwise, when a proposed act of sedition would vio-
late conscience. Thus, Hobbes says that “subjects will hardly be drawn into the
field and fight with courage against their consciences” (2014: 130).19
We should also note that even if Hobbes’s official view is that it takes the
sovereign’s command to make genuine law, Hobbes cannot possibly think that
this is also necessary for obligation, since the sovereign can himself exist only
if a covenant that establishes him and his authority obligates independently of
his command. Moreover, there are reasons of right why, at least in Hobbes’s
view, it takes the sovereign’s command to create genuine law. These reasons fit,
briefly, under two main headings: publicity and assurance.
Genuine law requires genuine publicity – public practices and institutions –
and those are unavailable in the state of nature by definition. Second, like any
covenant, the covenant establishing the sovereign requires joint performance
and is void under reasonable suspicion of non-performance. Covenanters will
need assurance, therefore, not just that there will be reasons of right to com-
ply, but that less conscientious individuals will have reasons to comply beyond
their being obligated to.
The crucial point is not that the covenanters might have a strategic, self-
interested reason for this assurance; they have reasons for assurance that are
grounded in the authority of right itself. Perfectly conscientious beings who
know one another to be so might not need the assurance provided by the
sovereign’s sanctions, but ordinary human beings, knowing one another to
be guided both by conscience and by self-interest, do. Since reasonable sus-
picion of non-performance voids contract, even conscientious subjects have
reasons of right for wanting there to be reasons of self-interest for compliance

19
And that “though a man be discontent, yet if in his own opinion there be no just cause of
stirring against, or resisting the government established, nor any pretence to justify his
resistance, and to procure aid, he will never show it” (1994b: II.8).

https://doi.org/10.1017/9781139025065.004 Published online by Cambridge University Press

 a metaethics of right? 59

also. But this does not mean that the reasons of right derive their author-
ity from self-interest in any way. It just means that they would not bind to
performance in foro externo unless sanction-creating reasons of self-interest
were in play also.
Suppose, therefore, that we understand Hobbes to hold that the laws and
right of nature are genuinely deontic propositions of right that cannot be
reduced to propositions of good (despite Hobbes’s saying that they are really
“theorems” about what conduces to the good of self-preservation (1994a:
XV.41)). A substantial metaethical problem remains, namely, that although
Hobbes provides a metaethical account of claims about the good, he provides
no independent account of claims of right. Moreover, it would seem to follow
from his claims about deliberation and desires that the only way consider-
ations of right can enter practical reasoning is by being translated into terms
of the good.

A Metaethics of Right?
It is worth considering, therefore, whether Hobbes might nonetheless be able
to provide such an independent metaethics of right that could satisfy the same
empiricist, materialist constraints that help motivate his metaethics of good.
I believe he could, by tying deontic moral claims to the expression, not of
desires, but of the distinctive attitudes through which we hold ourselves and
one another accountable, “reactive attitudes,” as Strawson famously calls them,
attitudes like guilt, moral blame, and what Strawson calls the “sense of obliga-
tion” (Strawson 1968: 86).
We have already seen how Hobbes holds both laws and the right of nature
to a standard of justified blame. The right of nature is a “blameless liberty,” and
the laws of nature must be formulable in something like the Golden Rule in
order to leave “men inexcusable” (Hobbes 1983: I.7; 1994a: XV.35).
In The Elements of Law, Hobbes contrasts human beings with other “living
creatures” who “have no conception of right and wrong” (1994b: I.19.5). Such
creatures, Hobbes says, experience “pleasure and pain” and so have desires and
aversions. What they lack is any attitude of “censure of one another” (I.19.5).
If Hobbes were to accept a projectivist metaethics of right according to which
judgments of right and their conceptual correlates express reactive attitudes
like indignation, blame, and censure, he could then explain why in lacking
these attitudes, other creatures lack a “conception of right and wrong,” though
they are capable of desire, deliberation, and will. Such beings would lack a
conscience.
We have seen that Hobbes regards the concepts of moral right and wrong as
intrinsically related to moral responsibility, blame, and excuse. And Strawson-
ian reactive attitudes find their way into his moral psychology also. For exam-
ple, Hobbes says that “indignation” is “[a]nger for great hurt done to another,

https://doi.org/10.1017/9781139025065.004 Published online by Cambridge University Press

 60 hobbes and pufendorf

when we conceive the same to be done injury” (1994a: VI:21).20 Of course,

if Hobbes were to think that indignation is a reaction to a prior belief about
injury and wrong, he could not then say that these beliefs about right project
the reactive attitude. But there is no philosophical reason for Hobbes to be a
“judgmentalist” about reactive attitudes that would not apply also to being a
judgmentalist about desire. Neither view, moreover, seems to fit with Hobbes’s
fundamental philosophical orientations.
The idea would not be that deontic claims of right express attitudes like
blame and guilt directly. The normative claims that apparently directly
express these attitudes are claims of blameworthiness and culpability. A
metaethics of right that is analogous to, but nonetheless independent of,
Hobbes’s metaethics of good could then exploit a conceptual connection
that seems to exist between wrongness and blameworthiness. It is an appar-
ent conceptual truth that an act is morally wrong if, and only if, it is an act
of a kind that it would be blameworthy to perform were the agent to do so
without excuse (Mill 2002: V.14; Brandt 1979: 163–176; Gibbard 1990: 42;
Skorupski 1999: 142; Darwall 2006). Hobbes could thus say that claims of
right express attitudes of censure and blame indirectly via this conceptual
connection.
How, then, would a projectivism about the right fit with a projectivism about
the good? Hobbes believes that deliberation is but the succession of appetites
and aversions and that the will is simply the last appetite before action. There
is no doubt that such a stripped-down view of deliberation and the will fits less
comfortably with a doctrine of the authority of right than a moral psychol-
ogy and philosophy of action that connect these more tightly to autonomous
agency. Only the latter, it would seem, can best fit with a moral and political
philosophy that give authority and consent fundamental roles.
Nonetheless, there is nothing flatly inconsistent between Hobbesian delib-
eration and a projectivist metaethics of right. Just as a Kantian ethics and moral
psychology can fit with a sufficiently formal Humean theory of motivation
(e.g., by making Rawls’s distinction between object-dependent and principle-
dependent desires), so also might a projectivism about the right fit with a
Hobbesian projectivism about the good (Rawls 2000). Thus we might regard
one of the “subjects” that Hobbes discusses in the passage quoted above – the
subject who forbears self-interested rebellion because they lack “just cause” –
as having their reactive attitudes affect their desires in ways that make their
judgments of good influenced by their judgments of right. Realizing that such
rebellion would be blameworthy by imaginatively placing themselves in an
impartial position and blaming rebellion from that perspective, they desire not

20
See also Hobbes on “revengefulness”: “desire, by doing hurt to another, to make him con-
demn some [act] of his own” (1994: VI.34). I am indebted here to Moya Mapps.

https://doi.org/10.1017/9781139025065.004 Published online by Cambridge University Press

 pufendorf 61

to rebel, which they express with the judgment that that would be bad to do
(and good to forbear doing) because it would be wrong.
I know of no evidence that Hobbes considered a projectivist ethics of right
along these lines. The point is that it is a metaethical position that can fit with
his normative claims of right that resists reduction to claims about the good.
And it would fit into his empiricist materialist framework no less well than does
his projectivist metaethics of good.
In conclusion, I want to note one last problem that Hobbes’s views face even
if he can adequately underwrite the independent normative claims of right that
are embodied in the laws and right of nature. When Hobbes defines obligation
as the state resulting from the transfer or laying down of a right in contract, this
definition rings true because of its association with our ordinary conceptions of
how contractors can give one another rights. But is this a plausibility to which
Hobbes is really entitled? Our ordinary notion of contract is like that of prom-
ise in being something that comes about through the exercise of normative or
moral powers that the contracting parties (or promisor and promisee) recipro-
cally recognize each other as having and exercising. It is impossible to contract
or to promise without the normative power to do so and without a partner who
recognizes both themselves and one as having that power.
Even if, consequently, Hobbes can ground both the laws and the right of
nature as genuine deontic facts, the question will remain whether these norma-
tive materials are sufficient to explain the existence of normative powers that
individuals must possess in order to contract or lay down their rights. We turn
next to examine the views of a philosopher, Samuel Pufendorf, who thought
extensively about what such normative powers must consist in and presuppose
and who criticized Hobbes partly on the grounds that the normative materials
provided by a Hobbesian state of nature are insufficient to underwrite these
powers.

Pufendorf
It is only relatively recently that scholars have begun to appreciate Samuel
Pufendorf’s importance to the history of ethics.21 The signal element of Pufen-
dorf’s ethics for recent commentators is his idea that morality arises “by impo-
sition,” by God imposing his superior will on a world that, if not completely
value free, nonetheless can contain no moral value of or on its own. But how,
exactly, is imposition accomplished? According to Pufendorf, human beings
do not simply defer to God in the way elephant seals do to a dominant male.
Rather, imposition is realized through human recognition of God’s authority

21
Thanks largely to Schneewind 1987 and 1998, and Haakonssen 1996. See also Korsgaard
1996 and Irwin 2008. What follows draws heavily from Darwall 2012.

https://doi.org/10.1017/9781139025065.004 Published online by Cambridge University Press

 62 hobbes and pufendorf

to direct and hold us answerable. This brings into play a whole battery of con-
cepts – recognition, accountability, imputation, and authority – along with the
capacities to operate with them in practical thought.
What is brilliantly original in Pufendorf, and less well noted by commenta-
tors, is his appreciation of these conceptual connections along with his aware-
ness of their implications for moral psychology. Authority is a normative
“moral power,” as Pufendorf calls it, which agents can exercise only within a
mutually recognized social, moral space that is constituted by their respec-
tive obligations to and rights against one another, and whose exercise directly
affects those rights and obligations. As Pufendorf sees it, only “sociable” beings
with the capacity for mutual recognition are capable of moral obligation.
Pufendorf was far from the first thinker to hold some version of a divine
command theory of morality. But he may have been the first to try to work out
what such a view must look like if it is to take seriously the conceptual links
between authority, recognition, and accountability, as well as the psychology
necessary for these to be realized in moral life. This may, however, introduce
instability into Pufendorf’s overall view. Whereas Pufendorf seeks to derive
human moral powers, equal dignity, and sociability from God’s superior moral
power, the very idea of moral powers, including God’s, may already presuppose
a more basic moral power or dignity that is shared by any being who is capable
of sociable relations. This is a view that, as we saw, is much closer to Grotius’s.
If, as Sidgwick maintained, modern ethical philosophy is characterized by its
focus on the “quasi-jural” concepts of morality and moral obligation, arguably
no early modern thinker had a more juridical theory of these concepts, or more
legalistic normative moral theories deploying them, than Pufendorf (Sidgwick
1967: 6). Pufendorf’s works, Elements of Universal Jurisprudence (1660), The
Law of Nature and Nations (1672), and On the Duty of Man and Citizen Accord-
ing to Natural Law (1673), were among the most significant moral and political
texts of the seventeenth century (Pufendorf 1934, 1991, and 2009). If less phil-
osophically powerful than Hobbes, Pufendorf stands as the most systematic
expositor of the modern natural law tradition, more philosophically insightful
than Grotius and Cumberland, and arguably no less so than Locke.
The most prominent aspect of Pufendorf’s thought is his theological volun-
tarism. Like Suárez before him, Pufendorf holds that any genuine law must
obligate and that it can do so only if it consists of legitimate demands from
a superior authority.22 It follows that there can be moral obligations, hence a
natural law, only if we are subject to God’s commands.
At the same time, Pufendorf was definitely a post-Grotian. He follows Gro-
tius’s distinction between perfect and imperfect human rights, which like Gro-
tius, he makes central to his account of morality, and which he carries forward

22
For Suárez, see Schneewind 1998: 58–66, Darwall 2003: 116-118, and Irwin 2008: 1–69.

https://doi.org/10.1017/9781139025065.004 Published online by Cambridge University Press

 pufendorf 63

with a correlative distinction between perfect and imperfect obligations.23

Pufendorf’s characterization of the former distinction broadly follows Grotius.
Grotius says that a perfect right involves a “power of demanding” that to which
one has the right (Grotius 2005: I, 139). Pufendorf also speaks of an “active
qualit[y] … by virtue of [which] something can be demanded of another”
(Pufendorf 1934: 19) and adds to this a “power” “to compel” others not to vio-
late the right, whether by “directing action against [them] before a judge, or,
where there is no place for that, by force” (1934: 19; 2009: 92).
Imperfect rights, by contrast, involve for Grotius and Pufendorf only an
“aptitude” rather than a “faculty” or “power.” An imperfect right to benefi-
cence, for example, consists in another person’s having a moral reason, per-
haps a weighty one, to benefit the rightholder, where beneficence is nothing
that “can be extorted from him against his will” (Pufendorf 1934: 20) or that
“in Strictness [one] may demand” (Grotius 2005: I, 89). Pufendorf and Gro-
tius differ in their language somewhat, since Grotius reserves “obligation” for
Pufendorf’s “perfect obligations,” which entail perfect rights. Where Pufendorf
speaks of “imperfect obligations,” Grotius seems happier saying that anything
but rights “strictly so called” “lay us under no obligation” (Grotius 2005: I, 88,
148). But there is no real difference here. To bring their usages into alignment,
we need only interpret Pufendorf’s “imperfect obligation” to mean something
that Grotius would say gives moral counsel but does not obligate.
A second Grotian element in Pufendorf’s thought is his emphasis on human
“sociality” or “sociability” (socialitas). The ways in which Pufendorf and Gro-
tius employ the concept seem very different, however, at least at first sight.
Whereas Grotius says that “Sociability … is the Fountain of Right, properly so
called” (I, 85–86), socialitas figures in Pufendorf’s thought not as the founda-
tion of natural law or right, but as the central constituent of the “fundamental
law of nature[’s]” content: “Every man, so far as in him lies, should cultivate and
preserve toward others a sociable attitude” (Pufendorf 1934: 208; cf. 1991: 35).
There are two differences here. One is that, according to Grotius, sociability
grounds natural law, whereas for Pufendorf, natural law can only be grounded
in God’s superior (foundational) authority. Moreover, Pufendorf does not
simply hold, as Suárez did, that God’s authority is necessary only for moral

23
Mautner 1999 understands Tuck as holding that Pufendorf denies natural perfect rights,
since according to Tuck, Pufendorf maintains that perfect rights require a “network of
social relations” through which they entail correlative obligations to right holders. (See
Tuck 1981: 161). I shall argue presently that although Pufendorf does believe that perfect
rights and obligations are social – indeed, they partly constitute the condition Pufendorf
calls sociability (socialitas) – that does not mean they are not natural. It is worth remem-
bering that Grotius himself held that perfect rights entail correlative obligations, since he
held that a perfect right includes a “[f]aculty of demanding what is due,” to which “answers
the Obligation of rendering what is owing” (Grotius 2005: I, 139).

https://doi.org/10.1017/9781139025065.004 Published online by Cambridge University Press

 64 hobbes and pufendorf

obligation, with our rational (and social) nature being sufficient to ground
natural moral rightness and wrongness, as on the Thomist classical natural
law view.24 Pufendorf thinks that the whole moral realm, moral properties
and moral “entities” of every kind, depends on the “imposition” of natural
law through God’s supremely authoritative will (1934: 5–8). Not even “good
repute and turpitude can be conceived to exist before law, and without the
imposition of a superior” (27).25
The second difference is that Pufendorf holds that perfect rights and obliga-
tions between human beings derive from a fundamental natural law, obedience
to which is owed to God. In other words, it is only because God commands
a “sociable attitude” between us and because this attitude itself involves an
acknowledgment of perfect, “mutual” obligations that these reciprocal obliga-
tions and rights between human beings exist. Contrary to Grotius, therefore,
Pufendorf believes that without God’s authoritative will neither natural law in
general nor rights and (“directed” or “bipolar”) obligations between human
beings could possibly exist.26
Despite these differences, there is an important similarity between Grotius’s
and Pufendorf’s conceptions of sociability. Both understand socialitas as essen-
tially involving relationships that are already conceived in terms of mutual rec-
ognition of individuals’ respective authorities or normative “powers,” that is, in
terms of their capacities to have obligations to and rights against one another.
“By a sociable attitude,” Pufendorf says, “we mean an attitude … by which each
is understood to be bound to one another … by a mutual obligation” (208).
As we noted in Chapter 1, Grotius’s notion of sociability can be interpreted
similarly. Grotius quotes Seneca’s saying that “[n]ature has founded a kind of
Relation between us” (Grotius 2005: 136), which Grotius interprets in terms
of mutual recognition of relative authority, involving either a “Right of Equal-
ity” or some justifiable “Right of Superiority” (136–137). Moreover, an essential
aspect of Pufendorf’s theory, as we shall see, is that mutual perfect obligations
are tied to a view of the equal dignity of any being capable of sociability and,
accordingly, to the idea that persons share a basic authority over themselves
and their own lives. This also echoes Grotius.
Of even more fundamental importance is the role that authorities and “moral
powers” play, according to Pufendorf, in constituting the whole of morality,
both making it up and bringing it about. It is no exaggeration to say that Pufen-
dorf’s expressed view is that every obligation, whether owed to God or to other
human beings, can exist only through the exercise of moral powers, either
God’s power to obligate human beings through his authoritative direction or

24
A point that Irwin emphasizes in Irwin 2008.
25
Likewise, “we call an action good morally, or in moral estimation, which agrees with law”
(1934: 114).
26
For a discussion of “bipolar obligation” see Thompson 2004 and Darwall 2013a.

https://doi.org/10.1017/9781139025065.004 Published online by Cambridge University Press

 moral powers i 65

our power to obligate ourselves to one another through agreements and “pacts.”
However, Pufendorf cannot, in the end, sustain this position. And seeing why
this is so makes it possible to appreciate something profoundly significant in
the very idea of a normative power to bind oneself or others as Pufendorf con-
ceives of it and as it came to be conceived in a tradition of moral philosophical
thought that derived from him and flourishes still.

Moral Powers I
The concept of “moral power” lies at the core of Pufendorf’s moral theory.27
God creates the law of nature, and so morality itself, by exercising a moral
power, his authority over us. And Pufendorf holds that the fundamental natu-
ral law he thereby creates demands a sociability that itself includes recognizing
various moral powers human beings have regarding one another, including the
power to obligate ourselves through “pacts,” from which, like Hobbes, Pufen-
dorf believes that civil society and the authority of the state derives. If Hobbes
and Grotius can be read, as they commonly are, as trying to base morality and
natural law on rational self-interest (or perhaps reciprocity) and sociability,
respectively, then Pufendorf is best interpreted as making it derive from moral
powers. According to Pufendorf, all moral obligations, indeed, any “moral
entity” or “mode” at all, exist only through the exercise of moral powers.
The idea of “normative” or moral power has become familiar in contempo-
rary philosophy, first, through Hohfeld’s famous distinctions between species
of rights and correlative obligations – claims, liberties, privileges, and powers –
and, more recently, by Joseph Raz’s writings on “normative powers” (Hohfeld
1923; Raz 1972 and 2002a: 98–104). The basic idea can be gotten across with
either of two examples that play an important role in Raz’s work – promises
and authoritative directives (or orders) (Raz 1977 and 2002b: 3–27).
When someone promises to do something, they voluntarily take on an obli-
gation to a promisee by addressing the intended promisee in a distinctive way.
As Pufendorf puts it, when I make a “perfect promise,” “I not only intend in fact
to be obligated, but also confer a right upon another to demand what I promise
as quite simply owed to him” (Pufendorf 1991: 70). Promising presupposes the
moral or normative power that is exercised in making a promise, namely, the
power to bind oneself to another with the promise.28 This is a power to change
the moral relations obtaining between promiser and promisee – their respec-
tive rights against and (directed or bipolar) obligations to one another – by
a voluntary address, a promise to the promisee. Similarly, when someone in
authority orders another to do something, they thereby make the person who

27
One source is Suárez. See Schneewind 1990: 70.
28
For accounts of promising that stress this aspect see Raz 1977, Watson 2009, and Darwall 2013i.

https://doi.org/10.1017/9781139025065.004 Published online by Cambridge University Press

 66 hobbes and pufendorf

is subject to their authority obligated to them to comply. As with the power to

bind oneself by a promise, someone can obligate another through an order only
if they have the relevant moral or normative power, in this case, the authority to
issue the order and thereby bind someone subject to their authority.
As Pufendorf brings out with beautiful clarity, powers of the kind involved in
pacts and authoritative demands are fully normative or “moral.” By exercising
the power, someone directly changes their moral relations to the person they
address. These changes are what Pufendorf calls the power’s “moral effects”
(1934: 18). Exercising moral powers creates new rights and obligations, not
just by creating some change in the natural or physical world that, when com-
bined with existing moral obligations, has moral implications for the changed
physical or natural circumstances, as, for example, when one incurs an obliga-
tion to make amends to someone by having said something hurtful. In such a
case, one is already under an obligation to make amends to anyone one hurts,
whether one has already hurt someone or not. A promise, by contrast, creates
a new bipolar obligation to the promisee. We can of course say that a general
obligation to keep promises or comply with authoritative orders already exists,
but any such general obligation will itself depend on the normative powers to
voluntarily create the more specific bipolar obligations of which promise and
authority themselves respectively consist.
Pufendorf defines a moral power as “that by which a man is able to do some-
thing legally and with a moral effect” (18, emphasis added). “Moral effects” are
simply the newly created rights and obligations together with the account-
ability and legitimate sanctions they entail.29 They are part of a moral order or
“space,” as Pufendorf calls it, which he explicitly analogizes to natural or physi-
cal space (6–7). To grasp the precise character and significance of his idea, we
should note its relation to a fundamental distinction Pufendorf makes between
moral entities, on the one hand, and natural or physical entities, on the other.30
Physical things, their attributes, and modes occupy space and time and are
related to one another through causal laws. The physical causal order, brought
into existence by God’s creative act, is “physically complete” (5–6). Moral enti-
ties and their various modes are “superadded,” Pufendorf says, “by intelligent
agency to physical things and modes” (5). Whereas “the original way of pro-
ducing physical entities is creation, the way in which moral entities are pro-
duced” is through “imposition” (5). Pufendorf does hold that imposition occurs
when God creates human beings with their distinctive nature, including the

29
“Power is that by which a man is able to do something legally and with a moral effect. This
effect is that an obligation is laid upon another to perform some task, or to admit as valid
some of his actions, or not to hinder them, or that he shall be able to confer upon another a
power of action or possession, which the other did not formerly possess” (1934: 18).
30
Compare Kant’s later distinction between a realm governed by causal laws of nature and a
moral realm governed by laws of freedom.

https://doi.org/10.1017/9781139025065.004 Published online by Cambridge University Press

 moral powers i 67

capacity to understand his will for them, but it is only by God’s imposing his
will on human beings through making manifest his will to them through his
creative act that moral entities and modes are produced. As we shall see, genu-
ine imposition can occur only within a reciprocally recognitional social space.
Moral entities and aspects are not themselves part of the “physically com-
plete” natural world God creates. Unlike any physical “active force” consisting
in an “ability to produce any physical motion or change in any thing,” the dis-
tinctive “active force” of moral entities “consist[s] only in this, that it is made
clear to men along what line they should govern their liberty of action” (6,
emphasis added).31 Moral entities are inherently directive; their laws are inher-
ently normative in a way that the causal laws of physical nature are not. And
they entail accountability and legitimate sanctions: “[I]n a special way men are
made capable of receiving some good or evil and of directing certain actions
towards others with a particular effect” (6). The “particular effect” is not physi-
cal or natural, but “moral.”
Moral entities’ “efficacy” “flows from the fact that, as man’s creator,” God
“has the right to set certain limits to the liberty of will” and, therefore, to
impose legitimate sanctions should these be violated (6).32 It is God’s author-
ity or moral power that enables him to obligate his creatures through his will.
Since the rights and obligations that are created by God’s commands are perfect
rights and obligations, it follows on Pufendorf’s view that the power to create
such “moral effects” includes the authority or power to justifiably use sanctions
and coercion to secure compliance with the obligations (“by the threat of some
evil” (6)). This has important implications for Pufendorf’s account of the will
and moral motivation, as we shall consider below.
The only obligations that flow directly from God’s commands are obliga-
tions human subjects owe to him. Thus far, therefore, the only moral power
in play is God’s. But God commands us to be sociable to one another and, as
will be evident presently, Pufendorf thinks that sociability includes the mutual

31
Compare Basil Kennett’s translation: Moral entities’ “force” consists in “in shewing men
how they ought to govern their Freedom of Action” (Pufendorf 1710: 3).
32
Compare Kennett: “[H]e [God] by his Right of Creation hath the Power of circumscrib-
ing, within proper Limits, that Liberty of Will with which he indulg’d Mankind, and
when it grows refractory, of turning it which way soever he pleaseth, by the Force of some
threatned Evil” (Pufendorf 1710: 3). (Compare the following passage from De Officio: “An
obligation is introduced into a man’s mind by a superior, by one who has not only the
strength to inflict some injury on the recalcitrant, but also just cause to require us to curtail
the liberty of our will at his discretion” (1991: 28)). Knud Haakonssen has impressed upon
me that the Latin text mentions no “right of creation,” but only that as creator, God is
free to circumscribe his creatures’ will. It is clear, however, that Pufendorf distinguishes
between a superior’s power to threaten natural evils and his “moral power” to create
“moral effects,” which includes inferiors’ being held accountable and therefore subject to
deserved sanctions.

https://doi.org/10.1017/9781139025065.004 Published online by Cambridge University Press

 68 hobbes and pufendorf

recognition of an equal human dignity that grounds basic human moral pow-
ers, for example, to bind ourselves to one another through pacts of the sort that
are necessary to legitimate civic association and the state.
Here again, Pufendorf distinguishes between human natural and moral
powers along with the respective natural and moral effects of their exercise:
In man the power to act is twofold. One is natural power to act, through
which he is able by his natural strength to perform an action, or to neglect
it. … But moral power in man is that whereby he is able to perform a vol-
untary action legitimately and with a moral effect, that is to say, so that
this action shall … be able to produce moral effects in others (2009: 229).

A “moral effect,” again, “is that an obligation is laid upon” someone (1934:
18). And since the obligation is a perfect one, it includes the standing to
“deman[d]” and, within proper procedures, to “compel” compliance (1934:
19; 2009: 92). It follows that the exercise of moral powers, whether by God
or derivatively by human beings, brings about obligations that make men
“capable of receiving some good or evil” in a distinctively legal way, that is,
as justly deserved benefits or sanctions (1934: 6). The obligations that are cre-
ated by exercising a moral power entail a standing to hold those obligated to
one accountable.
Pufendorf’s theory is thus designed to capture the conceptual intuition
about the right that we mentioned earlier in connection with Suárez and
Hobbes, namely, that something can be genuinely obligatory only if those who
are obligated can justifiably be held responsible or accountable for compliance.
By addressing his will to human beings, God makes them “moral causes,” that
is, agents to whom actions and their effects can be imputed and for which they
are thereby accountable.
The formal nature of a moral action, Pufendorf says, “consists of its ‘imputa-
tivity’,” “whereby the effect of a voluntary action can be imputed to an agent.”
Whether the effects be “good or evil,” “he must be responsible for both” (68).
Pufendorf’s Latin makes clearer that he means “accountable.” What can be
imputed to us as a “moral cause” is what we must answer for or provide some
account of (“ratione rederre”) (1672: 61).33 The “primary axiom in morals,”
Pufendorf says, is that “a man can be asked for a reckoning” for anything in his
power.34 Or, equivalently, “any action controllable according to a moral law,
the accomplishment or avoidance of which is within the power of a man, may
be imputed to him” (1934: 70).

33
I am indebted here to Knud Haakonssen.
34
Compare a parallel passage from De Officio: “The primary axiom in moral disciplines
which look at the subject from the point of view of the human court is held to be: a man
may be held accountable for those actions which it is in his power whether they are done or
not” (Pufendorf 1991: 23).

https://doi.org/10.1017/9781139025065.004 Published online by Cambridge University Press

 moral powers i 69

Pufendorf holds, again, that all moral “entities” and “modes,” all things
properly considered “moral,” depend upon God’s having exercised his moral
power to impose law. Pufendorf recognizes a sense of “natural goodness” that
is independent of imposition (“the native power to produce an effect good and
useful to mankind”), but this is insufficient “to constitute an action in the field
of morals” (28). Any moral “entity,” quality, or “mode” whatsoever, he thinks,
exists only because of authoritatively imposed law. “That reason should be able
to discover any morality in the actions of a man without reference to a law, is as
impossible as for a man born blind to judge between colours” (28).
This is an illuminating analogy. Hobbes was but the first of a number of modern
philosophers who would analogize the metaphysics of the ethical to that of color.
In the eighteenth century, Hutcheson and Hume would draw a similar analogy
between color sense and “moral sense” or sentiment.35 For Pufendorf, however,
the perspective from which we attribute moral qualities is neither that of desire
nor of any sentiment or attitude like moral approbation or admiration, but rather
the point of view of a will under authoritative direction, that is, of one will being
addressed demands by another who has the authority or “moral power” to make
demands of, and so to direct, them legitimately, and who has, by virtue of that,
the standing to hold them accountable through legitimate sanctions.
The way Pufendorf connects moral obligation, accountability, and legiti-
mate sanctions to his theory of the will and moral motivation will occupy us
further below. So also will the question of how precisely “mutual” (bipolar)
perfect human obligations and rights, including human moral powers, along
with an equal basic human dignity underlying them, can all be grounded in
God’s moral power. Before we turn to these questions, however, we should
briefly note a fundamental problem, which we shall consider in more detail
below, that afflicts any attempt to construct the whole of morality from the
exercise of moral powers. Pufendorf seems dimly aware of this problem, but I
think he has no good solution to it.
Pufendorf makes a crucial distinction between imposing oneself on
another(’s will) by force and doing so legitimately, with authority. And he con-
nects this to an equally fundamental, and reciprocal, distinction within the
obligated agent’s motivation and practical reasoning between being moved by
fear of a threatened sanction (“sense of an impending evil”), on the one hand,
and being moved by respect for the legitimacy of the sanction (seeing that the
evil “falls upon him justly”), on the other (91). In the latter case, Pufendorf says,
the obligated will must be able “to judge itself worthy of some censure, unless it
conforms to a prescribed rule” (91). Moral agency, he holds, requires the capac-
ity for conscientious self-judgment.
35
Thus Hume: “Vice and virtue, therefore, may be compared to sounds, colours, heat, and
cold, which, according to modern philosophy, are not qualities in objects, but perceptions
in the mind” (Hume 1978: 486).

https://doi.org/10.1017/9781139025065.004 Published online by Cambridge University Press

 70 hobbes and pufendorf

We have seen already that Pufendorf holds that someone can be put under
obligation only through the exercise of a moral power.36 And we know too that
he thinks any human moral power must derive ultimately from God, specifi-
cally, from the obligation to follow the fundamental law of nature, which God
legislates by exercising his moral power. But where does God’s moral power
come from? In imposing his will, God makes natural law. But Pufendorf also
says that “we are obligated by a law, because we owed beforehand obedience to
its author” (89). And this now seems to close a circle: God imposes obligations
on us by exercising his moral power, which he apparently has only if we are
already obligated to act as he directs “beforehand.”37
This need pose no insoluble metaphysical problem, since Pufendorf can con-
sistently hold that God’s creative act simultaneously gives him moral power to
direct us – by making it the case that we are indebted to him (through grati-
tude) for the great benefit of creation – and expresses his manifest will for (and
to) us, and so exercises this moral power over us, thereby creating obligations
to act as he directs.38 Granted, this isn’t quite “beforehand,” but that doesn’t
seem required anyway. What matters is only that God have moral power at the
time he exercises it. However, a normative or moral issue will remain. God’s
authority to obligate by imposing his will must be normatively independent of
its exercise. That is, the moral fact that, as Pufendorf holds, God’s gift indebts
his creatures to him to follow his will cannot itself depend on God’s will, since
God’s will can obligate only if that fact independently holds.
Still, Pufendorf can consistently maintain that no actual obligations (and
in this sense no “moral entities”) exist before God’s creative, will-imposing
act, and not just trivially, that is, because before that there exist no subjects to
be obligated. Pufendorf can hold that it takes God’s creative act to bring into
existence any actual moral powers, including God’s, and therefore any moral
effects or obligations. Nevertheless, the problem remains that for that to be the
case it must be independently true – true independently of God’s will – that any
such creative benefaction would make the created rational beneficiaries their
creator’s legitimate subjects (in a fully normative sense) and, consequently, that
their creator would have moral power over them.

A “Sociable” Attitude
We turn now from the obligation to comply with laws of nature in general,
which derives from God’s moral power to obligate us through his superior will,
to “mutual” (directed or bipolar) human obligations and the moral powers
that human beings have to directly affect their moral relations and create new
36
For this claim and its defense, see Mautner 1999: 175.
37
Again, this is Cudworth’s objection to all forms of metaethical voluntarism.
38
Below I shall show that Pufendorf pursues this line.

https://doi.org/10.1017/9781139025065.004 Published online by Cambridge University Press

 a “sociable” attitude 71

bipolar obligations and rights. All obligations to follow the laws of nature are,
in the first instance, owed to God. Even when God directs us to benefit and not
harm other human beings, any obligations that are thereby created are owed
only to God. No obligation to others to act in these ways directly follows. As
far as laws of nature in general are concerned, the only being with grounds for
complaint about non-compliance is God.
So far, therefore, we lack any account of human perfect rights. Like Gro-
tius, however, Pufendorf holds that human beings have natural perfect rights
(and, Pufendorf adds, perfect obligations) and that these entail the standing to
demand compliance and even to compel it when directive judicial procedures
run out. What brings human moral powers and rights into view is the law of
nature’s demand to take a sociable attitude to one another.
“It is,” Pufendorf says,

[A] fundamental law of nature, that “Every man, so far as in him lies,
should cultivate a sociable attitude, which is peaceful and agreeable at all
times to the nature and end of the human race” (208).

But how does this make us obligated to one another? Assume, for example,
that we are required by the fundamental natural law to cultivate benevolence
and act beneficently toward each other. This would not entail that others have
any right to our beneficence or that we are obligated to them to act benefi-
cently. Of course, Pufendorf does not believe that there is any perfect right
to beneficence. But he does think we have perfect rights not to be harmed
in various ways and that these imply correlative obligations to others not to
harm them (128). Nonetheless, from the simple fact that God commands
that we not harm one another, no reciprocal rights and obligations follow. If
“mutual” human obligations and rights are to follow from the fundamental
law of nature, therefore, a “sociable attitude” must be understood in some
other way.
And Pufendorf does seem to understand sociability differently, indeed, in
such a way that mutual obligations are already built into it. “By a sociable atti-
tude,” he says,

We mean an attitude of each man towards every other man, by which each
is understood to be bound to the other by kindness, peace, and love, and
therefore by a mutual obligation (208).

Pufendorf evidently understands sociability as itself including the recognition

of mutual obligations. Regarding others from the perspective of this attitude
must be already to see them as having rights against one and, consequently, to
regard oneself as obligated to them. It must be to see them as related to others
by the same equal “relationship” of mutual accountability that Grotius holds
sociability to include.

https://doi.org/10.1017/9781139025065.004 Published online by Cambridge University Press

 72 hobbes and pufendorf

As Pufendorf understands inherently “social” attitudes of kindness, peace,

and love, these already involve a kind of mutual recognition and respect.
Indeed, Pufendorf explicitly defines peace, and its contrary war, in terms of
mutual obligations whose reciprocal recognition the state of peace includes.
[P]eace is that state in which men dwell together in quiet and without
violent injuries, and render their mutual dues, as of obligation (9).

The part of sociability consisting in peacefulness, therefore, must already

include recognition of reciprocal human rights and duties inherent in it.
Similarly, war is “the state of men who are naturally inflicting or repelling
injuries [violations of rights] or are striving to extort by force what is due to
them” (9). It follows that human beings cannot see themselves as at peace or
at war without taking themselves to have reciprocal rights and obligations,
indeed, perfect ones. To view other human beings sociably in Pufendorf’s
sense, whether in war or in peace, is to view them as having rights and obliga-
tions that justify directive and sometimes coercive action that would other-
wise amount to brute force. This, if you like, is what it is to view someone as a
fellow human being as opposed to as a brute.
We could put the same point by saying that, for Pufendorf, the natural
state of human beings is an inherently social one, not just in the sense of
involving affiliating attitudes and relations that lead human beings natu-
rally to want to live together without conflict, but also as including social
attitudes in which recognition of reciprocal human rights and obligations is
already intrinsic.
Recall Pufendorf’s fundamental distinction between the physical order,
with its distinctive physical entities, powers, and (temporal and spatial)
states, and the “super-added” “moral” order with its own entities, powers,
and states created by “imposition.” Concerning the latter, Pufendorf says,
“every state presupposes in the man who is said to be in that state a cer-
tain respect and attitude toward others, since, indeed, every state is accom-
panied by a right or obligation which cannot be understood without an
object for its force” (9). Because of God’s imposition at the creation, human
beings find themselves in a moral state in relation to him by virtue of his
moral power. In addition, God’s command that we cultivate sociability
puts us also into a moral relation to one another, that is, into moral “states”
that, Pufendorf says, “more expressly denote a relation toward other men,”
“since they signify distinctly the mode in which men mutually transact
their business” (9).39

39
Note that the idea is not that mutual recognition is the natural human state, but that it
is inherent to the moral state that God commands his creatures to be in regarding one
another. I am indebted to Knud Haakonssen for pressing me on this.

https://doi.org/10.1017/9781139025065.004 Published online by Cambridge University Press

 equal human dignity 73

Equal Human Dignity

Human beings thus enjoy “an equality of right, which has its origin in the fact
that an obligation to cultivate a social life is equally binding on all men, since it
is an integral part of human nature as such” (333).40 Because, Pufendorf writes,
“[H]uman nature belongs equally to all men, and no one can live a social life
with a person by whom he is not rated as a fellow man” it follows, as a precept
of natural law, that “[e]very man should esteem and treat another man as his
equal by nature, or as much a man as he is himself” (330).41 This is a corollary of
the fundamental natural law. A sociable attitude, as Pufendorf understands it,
includes mutual respect for one another’s equal dignity as human beings. It is a
“precept of natural law,” therefore, that everyone should “permit others to enjoy
the same right that he himself does” (336).42
To regard others socially, consequently, is to see them as sharing a com-
mon basic standing with oneself that grounds or includes an ensemble of basic
human rights to make demands of one another and hold each other to those
demands. People “who, from a special feeling of their own superiority, would
have every liberty reserved for themselves alone … and claim honor before all
other men” are “plainly unsocial” (336). It is not too much to say that Pufendorf
holds that complying with the fundamental law of nature requires viewing oth-
ers as having a sociability that can be, as Grotius holds, “the Fountain of Right.”
Sociability thus includes the recognition of all men as equally entitled to fun-
damental respect or, as Pufendorf usually puts it, “esteem.” “Esteem of per-
sons in communal life … outside of states consists in this, that [the other] is
regarded as the kind of person with whom … it may be possible to have inter-
course” (2009: 94). “Intercourse,” like “sociable,” is a normatively loaded term
for Pufendorf. Just as taking a “sociable attitude” toward someone is already to
view them as entitled to certain forms of treatment and to hold others to these,
so likewise, to relate to someone as apt for “intercourse” is to recognize them
as having this same fundamental dignity and thereby to respect (in Pufendorf’s
term, “esteem”) them as an equal to whom one is accountable for complying
with “mutual” obligation and correlative rights.
“Esteem” refers to a form of valuation Pufendorf uniquely applies to “per-
sons” and their distinctive value: “the value of persons in common life” (1934:
1229). He contrasts esteem with the way “things” are valued. “In common life,”
Pufendorf says, persons are “valued by” esteem and things are valued by their

40
“Suitable” may be a better translation of Pufendorf’s “comitetur” than the Oldfathers’
“integral.” I am indebted to Knud Haakonssen here.
41
Knud Haakonssen suggests that a better translation than “treat another man as his equal by
nature” is “treat another man as if his equal by nature.”
42
Cf. Hobbes on “the law of the Gospel,” “whatsoever you require that others should do to
you, that do ye to them” (1994a: XIV.5).

https://doi.org/10.1017/9781139025065.004 Published online by Cambridge University Press

 74 hobbes and pufendorf

“price,” anticipating Kant’s later distinction between dignity and price (Kant
1996a: 4: 435), but in a way that understands, perhaps more explicitly than Kant
will, esteem or respect for human dignity to be a form of “sociability” or inter-
personal relating (Darwall 2008). To esteem or respect others in this sense is
to regard them as having the standing for “intercourse” as equals, along with
whatever moral powers that standing involves (including the power to promise
and enter into pacts).
In connecting rights and dignity in this way, Pufendorf begins a line of
thought that will prove powerful and influential. Human beings carry with
them “a most sensitive self-esteem” and are “rarely less and often more dis-
turbed” by disrespect for their dignity “than if an injury were being offered
[to their] person and property” (Pufendorf 1934: 330). When one’s person or
property is violated, the injury is generally twofold, indeed, since some insult
is at least implicitly added, and so a second injury of disrespect, to the primary
injury. The “most telling reply with which the rude insults of other men is met
is, ‘I am not a dog or a beast, but as much a man as you are’” (330).
Because of these connections between sociability, respect for equal dignity,
and rights, arrogance “giv[es] occasion to a breach of peace” (335–336). Peace
just is the condition in which people acknowledge one another’s dignity and
rights, so anyone who, in word or deed, presumes a superiority over others that
cannot itself be justified from a fundamental standpoint of equality – like, for
example, that of a political official that is properly grounded in an agreement or
“pact” – undermines peace and threatens war. To be peaceful, one must treat
others not as “inferiors,” but as equal “men.”43
Pufendorf contrasts his account of natural human equality with Hobbes’s
in two respects. Concerning Hobbes’s idea that human powers are sufficiently
equal to ground mutual “diffidence” or distrust, Pufendorf replies that a great

43
Pufendorf holds that husbands have unequal rights over their wives because these are
grounded in the “matrimonial pact” (Pufendorf 2009: 370–371).

Human obligations based in equal dignity are reciprocal or “mutual” in senses going
beyond their involving mutual treatment. People can expect the compliance of others, in
the non-epistemic sense of being entitled to expect compliance of them, only if they com-
ply themselves. “[I]f a man wishes to avail himself of the assistance of other men …, he
must in turn lend his own talents to their accommodation” (1934: 335). The point is not one
of prudence, that others will only cooperate if one reciprocates oneself. It concerns rather
the reciprocity of right: Others will no longer be obligated to one to cooperate if one does
not cooperate with them. “In this respect,” Pufendorf says, “the obligation to exercise the
duties of natural law toward others, although enjoined by the supreme will of God, agrees
with an obligation arising from a convention [or ‘mutual agreement’].” Just as someone
who makes an agreement but fails to do their share loses any claim to others’ compliance
and is rightly subject to their complaint and legal action, so someone who fails to discharge
their obligations to others “cannot demand any longer those duties from the other, and the
other has the further right to use force in making him render satisfaction” (1934: 333; see
also 264).

https://doi.org/10.1017/9781139025065.004 Published online by Cambridge University Press

 equal human dignity 75

“variety” of natural powers can actually exist in a “just harmony” precisely

because of an equality of right that human beings can jointly recognize (Hobbes
1994a: XIII.4; Pufendorf 1934: 332). Second, Pufendorf rejects Hobbes’s natural
“right to all things” as not a “real right,” since it entails no correlative “moral
effect” (obligation) “in the rest of [humankind],” “those who have the same
nature as I” (Pufendorf 1934: 391). “[I]t is absurd to try to designate as a right …
[something] which all other men have an equal right to prevent one from exer-
cising” (391). For Hobbes, the “right of nature” is a “blameless liberty” that
does not purport to give one a claim or right to demand anything from others
(Hobbes 1983: I.14.6). By Pufendorf’s lights, however, this effectively defines the
state of nature as a state of war of all against all.
There are fundamental issues that confront Pufendorf’s attempt to ground
morality in moral powers. First, Pufendorf’s basing human moral powers,
equal dignity, and mutual obligations and rights on God’s authoritative will
faces the problem that taking a sociable attitude toward others is arguably not
the kind of thing that can be done in obedience to anyone else’s legitimate
demands, not even God’s. We regard others sociably if we regard them as hav-
ing the standing to make claims on us because of our shared rational and social
nature. There seems to be no way that this common human dignity could arise
by God’s commands, for example, as by a delegation of his authority. We will
return to this problem in the next section.
More generally, as Pufendorf is himself dimly aware, it is possible for any
being, God or human beings, to create obligations and rights through the exer-
cise of their moral powers only if the parties who come thereby to have the new
obligations and rights (i.e., through the power’s exercise) are already related
to one another morally in ways that entail mutual obligations. Whether the
moral powers in question are human or divine, Pufendorf seems committed
to thinking that the very existence of a moral power presupposes that both the
person having it and those with respect to whom it may legitimately be exer-
cised are already obligated to one another not to use violence or force, that is, to
be “sociable” to one another.
This distinction between force and respectful sociable intercourse is reflected
further in a distinction Pufendorf makes in his account of practical reasoning
and moral motivation between the will’s being moved by “compulsion,” on
the one hand, and by “obligation,” on the other (Pufendorf 1934: 386). In the
former, “the mind is forced to something by mere external violence,” whereas
“whatever we do from an obligation is understood to come from an intrinsic
impulse of the mind” (386). Obligation thus requires the possibility of con-
scientious motivation that differs from fear of sanctions. At the same time,
however, sanctions, most especially God’s, play a virtually ineliminable role in
Pufendorf’s overall picture. For Pufendorf, the moral motive is a combination
of “fear mingled with reverence” (1934: 95; 1991: 28).

https://doi.org/10.1017/9781139025065.004 Published online by Cambridge University Press

 76 hobbes and pufendorf

Holding to Obligations in the Natural State

We should now consider two fundamentally different ways, defined recipro-
cally in terms of two different kinds of standing or authority, in which human
beings are responsible for complying with moral obligations, according to
Pufendorf. Obligations to comply with laws of nature in general are owed to
God. Even when natural laws concern our actions toward other human beings,
even indeed, when they concern the very actions to which others may have a
perfect right, the fact that these acts are required by natural law entails only
that God has the authority to demand that we perform them and to hold us
responsible for doing so. At the same time, God commands us to cultivate a
“sociable attitude,” and since this attitude itself involves regarding one another
(and ourselves) as having equal dignity and perfect rights, we can comply with
this fundamental natural law only if we “esteem” each other as equals and hold
ourselves responsible to one another for according each other our perfect rights.
Complying with our obligations to God consequently requires us to see our-
selves as obligated to one another. And we can do that only by recognizing our
shared authority to demand certain treatment and to hold each other respon-
sible for this treatment. Human obligations to one another are “mutual” in
the sense that they imply correlative rights and that rights to others’ peaceable
actions toward us are conditional on peaceable treatment of them. Nature does
not “allow,” Pufendorf says, that one can violate their obligations to another,
“while the other remains bound [to him] to keep the peace with him” (1934:
264–265).44
So long as we give others no cause for complaint of our actions toward them,
we retain the right that they act peacefully and sociably toward us. Threats to
our person can therefore legitimate self-defense and uses of force that would
otherwise count as violence (i.e., forceful violations of others’ rights), and viola-
tions of our personal rights can warrant compensation and forceful attempts to
secure it (1934: 264–265; 2009: 92).45

44
If someone “undertakes against me such things as tend to my destruction, it would be a
most impudent thing for him to demand of me that I should thereupon hold his person
inviolate” (265).
45
Someone’s violent taking (i.e., without right) of that to which one has a perfect right consti-
tutes injury or “damage” (damnum) to which one thereby acquires a right to compensation
(1934: 314). Natural law “requires that men voluntarily perform and do for each other what
they, for any reason, owe others, and freely offer to make good any damage which has been
done to others” (825).
The authority to demand respect for perfect rights, recognized in common law, thus
includes the standing to direct others forcefully in ways that would otherwise violently
injure or “damage” them and so give them cause for similar forceful “complaint,” as well as
to demand compensation when others violate their rights.
Powers of these kinds are like those that victims are generally taken to have under the
law of torts; both involve a distinctive form of accountability to the victim. Only the right

https://doi.org/10.1017/9781139025065.004 Published online by Cambridge University Press

 holding to obligations in the natural state 77

Although individuals may lose their right to demand respect for their per-
fect rights when they violate others’, this does not mean that others are thereby
relieved from the obligation of natural law to act peaceably toward them. Viola-
tion of a reciprocal or “mutual” obligation cancels individuals’ reciprocal rights,
but the natural law may still require actions to which these individuals are no
longer entitled since “that wherein the other party fell short of justice, can be
made up by the author of the obligation” (1934: 333). God creates natural law by
imposing his authoritative will on human beings and simultaneously laying on
sanctions for violators as well as benefits that can compensate victims.46
Since someone who actually violates your rights shows themselves willing to
do so, Pufendorf holds that in addition to demanding compensation, victims
may also demand that the wrongdoer “repent” and “give his word that he will
not offend in the future” (825). But Pufendorf is clear that individuals have no
standing to punish the wrong or “defect” itself nor to try to collect anything like
punitive damages. “[W]hen reparation of damages has been offered,” the vic-
tim’s “loss” made good, and the wrongdoer has given the victim “guarantees …
for the future,” then the victim incurs an obligation to “freely pardon the other
when he requests it” and “live with him thereafter on peaceful terms” (825,
1160). Further punitive measures make a victim “responsible for the breach of
peace and the altercation that follows” (825).

holder has the standing to decide whether or not to seek compensation for violations of
their rights, only they can forgive the violator, only they can consent to what would other-
wise constitute violation of their rights, and so on. But now note an important difference
between the responsibilities and powers that are involved in reciprocal human obligations
to and rights against one another, on the one hand, and those involved in the obligation
to comply with natural law in general, which mirrors the familiar distinction between tort
law and criminal law.
Whereas tort law is concerned with compensation, criminal law is concerned with pun-
ishment. And whereas it is up to the victim whether to bring a tort action, whether to pros-
ecute and punish violations of criminal law is not appropriately the victim’s decision; it is up
to the people and their representatives. Now, as we have seen, Pufendorf holds that human
beings have natural perfect rights that legitimate actions defending them and seeking com-
pensation for their violation. But he holds that no person in a state of nature has the author-
ity to punish violations of natural right or law; only God has that authority or power. This
is because genuine punishment can be done only with superior authority or “sovereignty,”
and the only natural sovereignty is God’s. Human beings are natural equals, and human
sovereignty can arise only through conventions and pacts (1934: 1161; 2009: 296).
46
Individuals do have the standing to demand and defend their perfect rights, and to hold
others accountable for violating them by seeking compensation, but only God has the
standing to punish violations of natural law. These two standings, corresponding to those
implicit, respectively, in tort and criminal law, are reflected in a distinction Pufendorf
makes between two different “features” or “factors” of wrongful treatment of human
beings, namely, between the “loss” that wrongs “directly or indirectly bring upon another”
(the tortious injury or “damage”) and the amount or degree of “the deviation … from the
law” that the wrong itself involves (the criminal wrong and its culpability) (1160).

https://doi.org/10.1017/9781139025065.004 Published online by Cambridge University Press

 78 hobbes and pufendorf

Unlike Grotius, Pufendorf takes the view that there is no natural human
authority to punish. Outside the state, in the condition of “natural liberty,”
there is no such thing as human punishment. “In the proper sense of the word,”
human punishment “follows upon human sovereignty” and right, and the
natural human condition is an equality of dignity (“esteem”) and right in which
no one has any standing or power in the state of nature to punish any other
(1161). Forcible defenses of human rights, including “by means of war,” are not
“punishments in the proper sense of the term” (1161).47
We return now to the problem I noted briefly in the last section, namely,
how human dignity, mutual human powers, and rights can be derived or
grounded in God’s authoritative command. We might think solutions to both
problems can be found in the idea of delegated powers and authorities. Surely,
God’s authority might encompass the power to delegate his power to punish,
so perhaps there is some way for states and magistrates to acquire their puni-
tive powers in this way (2009: 212). Similarly, might we not regard individu-
als’ authorities to demand certain treatment of themselves, seek compensation,
defend themselves, and so on, as delegated powers? So considered, human
beings would ultimately be acting on God’s behalf when they make and defend
claims. They would have certain “legal” powers so to act in defense of them-
selves, but such self-defense would be duly authorized ultimately by God –
ultimately, they would be defending, as it were, God’s right that his orders be
complied with and that his creatures not be damaged against his expressed will.
There is no obvious problem in thinking of human beings as having cer-
tain derived legal powers and rights in this way. The problem is that what God
demands of human beings, according to Pufendorf, is that they take a socia-
ble attitude and that this attitude involves a form of regard for human beings,
respect or “esteem,” as having an equal basic dignity just because they are ratio-
nal and sociable, that is, as being worthy of esteem just because they are beings
“with whom … it may be possible to have intercourse” (2009: 94). To have
sociable respect or “esteem” for someone in this sense is not just to count them
as fitting some legal category, as an official with delegated authority might rec-
ognize someone as qualifying for certain legal treatment. It is to view or regard
them as having an intrinsic dignity or value in themselves and so being intrinsi-
cally worthy of respect or “esteem” for no further reason. Despite the fact that
Pufendorf holds that the mutual obligations entailed within sociability them-
selves derive from God’s command (the “fundamental law of nature”), it sim-
ply seems impossible to come to have a sociable attitude of esteem for someone

47
However, civil magistrates do have the right to punish. Human sovereignty and unequal
authority arise by convention or pact (949–966). But even assuming that human beings
have the moral power to bind themselves to collectively constituted authority, it is not
clear, as Grotius had argued before, how any such authority could thereby acquire a right
to do a kind of thing, punish, that no individual had the authority to do in the first place.

https://doi.org/10.1017/9781139025065.004 Published online by Cambridge University Press

 moral powers ii 79

for the reason that God commands it. One could, of course, desire to acquire the
attitude for this reason, or undertake steps to try to inculcate it for this reason.
But trying to see someone as intrinsically worthy of esteem or respect for this
reason would be like trying to form an intrinsic desire for a saucer of mud for
some external reason having nothing to do with any features of mud that one
might be able to see as intrinsically desirable.
We can put the same point by saying that even if God’s commands give us
a reason of some kind for being in the mental state of thinking that human
beings have dignity and equal perfect rights, they give us no reason whatso-
ever for thinking that the propositions we would thereby affirm would actu-
ally be true.48 This would mean that a theory like Pufendorf’s might have to
be an “error theory” of human mutual rights and obligations. It might explain
why we have reason, indeed are morally obligated, to think in terms of mutual
human obligations and rights and their associated human moral powers, but
would nonetheless hold that, unlike thoughts about God’s authority and our
obligations to him, these propositions are not actually true.

Moral Powers II
We turn now to whether it is possible to derive all of morality from God’s
authoritative will, that is, through his exercising his moral powers, and the way
the issue of its possibility is reflected in Pufendorf’s theory of moral agency
and the will. I believe this is not possible, that Pufendorf is aware of this prob-
lem, and that this awareness shows itself in some deeply significant things he
says about obligation and moral motivation (though these are in tension with
other central aspects of his view). Whether or not Pufendorf’s views fully hang
together, analyzing them can help us see things of fundamental importance
about the nature of moral obligation and moral or normative powers that will
reverberate through modern moral philosophy.
Pufendorf’s awareness of these issues comes in his discussion of certain dif-
ferences between obligations created by law through an exercise of the moral
power of sovereignty or authority, and obligations created by agreements, that
is, exercises of the power to promise and create pacts. A central difference, he
says, is that “[s]ince agreements depend in their origin upon our choice, it is
first determined what is to be done before we are obligated in action” (1934: 89).
With laws and authoritative commands, however, the order is reversed: “[W]e
are first obligated in action, and then what is to be done is determined” (89). It
is only when a promise or pact is consummated that the obligation we thereby
undertake exists, but “we are obligated by a law, because we owed beforehand

48
The reasons would be what Parfit calls “state-given” rather than “object-given” reasons
(Parfit 2011). For further discussion of this problem, see Darwall 2006: for example, 15–17.

https://doi.org/10.1017/9781139025065.004 Published online by Cambridge University Press

 80 hobbes and pufendorf

obedience to its author” (89). Laws thus presuppose that we are already obli-
gated before their promulgation in ways that agreements do not. But if we can
be obligated by God’s commands only if we are obligated “beforehand” to obey
him, then not all obligations can come through God’s exercising his moral
power of command. There must be at least one obligation existing already.
Now if this were as far as things had to go, the problem might be easily
enough contained. Even if Pufendorf must admit that God’s moral power
or authority to command cannot itself derive from the exercise of that very
authority, he might nonetheless hold that no actual obligations “in action” exist
until God exercises this power. It must be true independently of God’s exercis-
ing his authority, of course, that anyone would be obligated by his commands
were God to issue any. Even so, any actual obligations to act or forbear action
would await God’s command.
But we can press the problem further. In virtue of what does God have his
superior authority? If Pufendorf took God’s authority simply to be a brute nor-
mative fact, he might deflect this question. He does not, however, and for good
philosophical reasons that are reflected in his theories of moral motivation and
the will. As we will examine in more detail presently, Pufendorf makes a crucial
distinction between the practical reasoning of an agent under obligation and
one who is moved by the desire to avoid harmful consequences or negative
sanctions. The idea seems to be that, owing to the conceptual connections that
Pufendorf insists on between obligation and accountability, someone can be
under a moral obligation to do something only if they can hold themselves to
the relevant demand through recognizing its legitimacy. “An obligation affects
the will morally,” he says; “it is forced of itself to weigh its own actions, and to
judge itself worthy of some censure, unless it conforms to a prescribed rule”
(91). Someone can be accountable only by holding themselves accountable. If,
consequently, moral obligations are that for which we are appropriately held
answerable by God, it follows that any such obligations must be things God can
expect us to accept as legitimate demands and judge ourselves censurable for
failing to obey.49
It would follow that God’s authority, indeed any authority over us whatso-
ever, must be something we are able to accept just by virtue of our being in a
position to be obligated to him. Either, therefore, this authority must be self-
evident to any rational human being or it must be something any such being
can appreciate good reasons for accepting. And Pufendorf holds that there are
such reasons. Mere power is insufficient to “lay an obligation,” though Pufen-
dorf does hold that the ability to create negative consequences or sanctions is
necessary. What is needed in addition to this, he says, is either that the putative

49
Note also Pufendorf’s remark that in order for law to exist it must be “communicated to the
subject in such a way that he recognizes he must bend himself to it” (1934: 89).

https://doi.org/10.1017/9781139025065.004 Published online by Cambridge University Press

 moral powers ii 81

authority “have done me some special service” or “that I should of my own

accord consent to his direction” (101). Whatever force the former condition has
in general, “this is all the more true if I am indebted to him for my very being”
(101). God’s authority is thus grounded in the greatest debt of gratitude that a
person could possibly have.
Whatever we think about such an idea in its own terms, it creates an obvious
problem for the position that all moral obligations derive from God’s exercis-
ing his moral power over us. For now it seems, as we saw before, that God
can have this moral power only if there already exists a freestanding obligation
of gratitude.50 Even so, however, Pufendorf might still hold that, in a way, all
actual moral obligations depend on God’s exercising his moral power, since
God’s benefaction and legal imposition might be conceived to be a single action
accomplished in the creation that simultaneously creates his authority and
exercises it.
In any case, there is also a different issue that Pufendorf faces in maintaining
that all of morality derives from God’s exercising his authority over us, which
afflicts any attempt to understand moral powers as prior to obligation. The
problem is that, properly understood – and as Pufendorf seems to understand
it – the idea of a moral power presupposes that those who exercise it and those
with and for whom it is exercised (for whom the “moral effects” of correlative
obligations and rights are created) must already have a moral relationship that
entails obligations to and rights against one another.
To see this, we can begin with the human power to obligate through pacts. By
virtue of agreements or “conventions,” individuals obligate themselves to one
another to obey and give each other the right to expect their obedience. Thus is
political sovereignty created by “consent.” But note how sovereignty, consent,
and agreement function in this line of thought. The natural human condition
is an “equality of right” where it takes individuals’ consent to justify directing
and using force against them as a putative authority would (333). It is part of
the very idea of a moral power of consent that consenting legitimates direction
and force that without consent would have been wrong, indeed that would have
wronged the consenter had they not consented. The same is true of the ideas of
agreement, convention, or pact. It is part of the very idea that one can create
an obligation, say, to give someone something as a part of an agreement, that it
would be wrong for the other simply to take the thing from one if one had not
agreed, and, indeed, that one would be wronged by such an unjustified taking.
The general point is that the exercise of a moral power can legitimate direc-
tion and the use of force if, and only if, it is assumed by the reciprocally recog-
nizing parties to its exercise that the thereby legitimated coercive action would

50
A further problem is that Pufendorf holds the obligation of gratitude to be imperfect
(Schneewind 1998: 136).

https://doi.org/10.1017/9781139025065.004 Published online by Cambridge University Press

 82 hobbes and pufendorf

have wronged its recipient were the directing agent to have lacked the relevant
power or were somehow to have exercised it improperly. The very existence of
moral powers to create new obligations and rights therefore apparently presup-
poses already existing moral relationships, with already existing mutual rights
and obligations, between those who have the power and those with respect to
whom it can legitimately be exercised. Finally, since this is a point about moral
powers in general, moreover, the same would seem to apply to God’s authority
over his human subjects.
The arguments of the last three paragraphs have been entirely conceptual,
drawing out consequences of the concepts of consent, agreement, and the exer-
cise of moral powers in general. I am not saying that Pufendorf was himself
aware of these entailments or even that he would have accepted them had they
been pointed out to him. But if they hold, as I submit they do, then they pose
a fundamental challenge to any attempt to derive all moral relations from the
exercise of moral power, God’s or any other beings.
There are places where Pufendorf seems implicitly to accept the general
point.
So that a man may not be able to complain that wrong has been done
him when he is compelled to adapt himself to the free choice of a second
person, it is necessary that the authority in question also be legitimate
(2009: 206).

If this is implicit in the moral power of authority, it will obviously apply as much
to God’s as to any human authority. Although Pufendorf believes that God has
an authority over human beings that may legitimate whatever directive treat-
ment of them he pleases, it seems that he is committed by an aspect of the idea
of moral power that he himself accepts to thinking that the created beings with
respect to whom God exercises his moral power must have moral standing
independently of his superior authority. (Of course, they would not have had
that power unless he created them, but that is another matter.) Indeed, it seems
to follow, oddly enough, that God could not have the superior authority that
he distinctively has unless it were already the case that human beings had an
authority they share with him, that is, a standing that makes unauthorized force-
ful treatment of them wrong, indeed, that would make it wrong them.

Moral Agency
I have been arguing that Pufendorf is committed to a fundamental distinction
between authorized directive or forceful action, which is legitimated by moral
powers and mutual obligations and rights and which therefore does no wrong
or “damage” to its recipient, on the one hand, and illegitimate force, which
does wrong and injury, on the other. In other words, the very framework of
moral powers seems to presuppose that brute force and coercion – force that

https://doi.org/10.1017/9781139025065.004 Published online by Cambridge University Press

 moral agency 83

is not so legitimated – is both wrong and wrongs its recipients (violates their
rights and the dignity that gives them standing to complain). If that is so, then
whatever authorities or powers exist, it is guaranteed to be the case that anyone
with whom one can have sociable “intercourse” in the sense of a reciprocally
recognizing exercise of moral power must already be assumed to be able to be
wronged by brute coercion and, therefore, that they likewise have the power to
complain about unauthorized use of force against them. It apparently follows
that the very idea of moral powers would commit Pufendorf to holding not
only that God has moral powers that cannot be created by their exercise, but
that so also must anyone who can be affected “morally” by the exercise of moral
powers – that is, any agent who can acquire new obligations and rights through
the exercise of moral powers in social intercourse.
In this last section, I want to bring out a way in which this distinction in
the space of interpersonal relations is reflected in a distinction that Pufendorf
makes in the internal space of moral phenomenology in his account of moral
agency and conscience. Both distinctions, it is useful to see, amount to a con-
trast with Hobbes’s philosophy, at least, as Pufendorf understands Hobbes.51
Just as Pufendorf is concerned to argue against Hobbes that the state of nature
involves a robust equality of right that cannot be reduced to equal natural
power (or a right that is merely a “blameless liberty”), so also is he concerned
to argue that conscience and the sense of obligation cannot be reduced to the
fear of sanctions.
It is useful to begin with an insightful distinction Pufendorf makes between
shame and conscience. This arises in connection with an objection he consid-
ers that “because the very blood seems to have a kind of natural sense of base
deeds, since it brings blushes to our cheeks … as we … feel ashamed,” our sense
of the moral quality of actions must also therefore be able to be caused naturally
just by considering the nature of the actions themselves, independently of any
“law” or “imposition” (1934: 31). Pufendorf grants that we can feel shame in
response to moral wrongs and defects of moral character among other things,
but he denies that shame is especially attuned to the moral. “[S]hame arises not
merely because of some base action, but also from anything, even though not
morally base, which is thought to diminish our reputation,” like “shortness of
stature, lameness, baldness” and so on (31–32). Since shame doesn’t distinguish
between the moral and the non-moral, it can give us no distinctive sense of the
former. So the fact that it arises naturally without imposition is no evidence
that any distinctively moral emotion might be independent of imposition.

51
Pufendorf’s picture of Hobbes is something of a caricature. Hobbes also distinguishes, as
we saw, between slaves who are moved only by fear of the “natural” consequences of dis-
obedience and servants who, having promised to obey, are bound “in conscience” (1994a:
xx.10).

https://doi.org/10.1017/9781139025065.004 Published online by Cambridge University Press

 84 hobbes and pufendorf

Pufendorf contrasts shame with conscience. Conscience, he says, is a “judg-

ment passed on moral actions by the understanding, in so far as it can take
cognizance of laws, and so is responsible to the lawgiver for their execution and
observance” (41). But though it involves the “understanding,” Pufendorf con-
ceives of conscience also in affective or emotional terms. Conscience’s judg-
ment when contemplating action already taken (“consequent conscience”)
involves the agent’s “approving” or “condemning,” “attended by peace of mind
or agitation” (1934: 41; see also 2009: 266). Similarly, when it comes to con-
science’s role in deliberation (“antecedent conscience”), Pufendorf draws a
contrast intrapersonally, as we have seen, between “compulsion” and action
from obligation, in a way that echoes the interpersonal distinction between
brute coercion and legitimate direction we noted above.
This forms the main difference between obligation and compulsion, since
in the latter the mind is forced to something by merely external violence
contrary to its intrinsic inclination, while whatever we do from obligation
is understood to come from an intrinsic impulse of the mind, and with the
full approbation of its own judgment (1934: 386).

And again:
An obligation differs in a special way from coercion … the latter only
shakes the will with an external force, and impels it to choose … only by the
sense of an impending evil (91).

An obligation, by contrast, “affects the will” not “by some natural weight,” but
“morally.” It “fills [the agent or will’s] very being with such a particular sense,
that it is forced of itself to weigh its own actions, and to judge itself worthy of
some censure, unless it conforms to a prescribed rule” (91).
Nothing could seem clearer than the contrast these passages draw between
the will’s autonomous motivation to comply with obligation and external,
“violent,” “coercive” avoidance of negative “natural” consequences to which
someone displeased with the agent’s conduct, even a displeased God, might
have the power to subject the agent. Things are not so neat, however. When
we view what is left out of the passage quoted toward the beginning of the last
paragraph, we see that Pufendorf says that the “peace of mind and agitation” of
consequent conscience come from an expectation “of the blessing or the wrath
of the lawgiver, as well as the goodwill or anger of other men” (41), not just from
the agent’s own approval or censure, as reading without the elided passages
might lead us to expect. Moreover, in the context from which the passage in
the last paragraph comes, we find Pufendorf saying that “nothing can constrain
the human mind, as it deliberates on the future to do or avoid anything, except
reflections on the good and evil which will befall others and ourselves from
what we do” (91). This suggests that, contrary to what we might have expected
when Pufendorf says that action from obligation involves an “intrinsic impulse

https://doi.org/10.1017/9781139025065.004 Published online by Cambridge University Press

 moral agency 85

of the mind,” the only thing that can move an agent is desire to have or to
avoid (naturally) good and evil consequences. And if this is right, then what
can the real difference be between the will’s complying with obligation “of its
own accord” and its doing so to avoid evil consequences?
It seems likely that Pufendorf’s thought is in some tension here. On the one
hand, the elements of his thought that I have been emphasizing push in the
direction of a firm distinction between motivation by sanctions and consci-
entious motivation. On the other, Pufendorf apparently lacks a theory of the
will that is fully adequate to capturing the distinction: “it belongs to the nature
of the will always to seek what is inherently good, and to avoid what is inher-
ently evil” (56). All motivation is apparently good regarding rather than right
regarding. Pufendorf tries to maintain the distinction, however, but in a way
that can be accommodated within his theory of the will. He says that while both
avoidance of natural evils and obligation involve fear of an evil or “some object
of terror,” in the case of obligation a man is forced “to acknowledge of himself
that the evil … falls upon him justly” (91)52
Pufendorf consistently maintains that there must be actual sanctions in
order for a person genuinely to be obligated. “An obligation is properly laid
on the mind of man by a superior, that is, by one who has both the strength to
threaten some evil against those who resist him, and just reasons why he can
demand that the liberty of our will be limited at his pleasure” (1934: 95; cf. 1991:
28). The threat of actual sanctions is thus necessary for obligation. But Pufen-
dorf emphasizes that his view differs from Hobbes’s, at least as he understands
Hobbes, since he holds that sanctions are not sufficient to obligate. It is also
necessary that there be “just reasons” for the sanction, hence that it falls on one
justly. The moral motive, again, is “fear mingled with reverence” (1934: 95).
It is a sign of the tension in Pufendorf’s thought that it is not clear exactly
how to understand the desire to avoid justified sanctions when, as Pufendorf
seems to be supposing, it involves no desire to avoid what would justify sanc-
tions. To hold that the threat of actual sanctions is necessary for obligation and
moral motivation, Pufendorf must be thinking that the agent’s own judgment
that his action is worthy of censure and sanction can do no independent moti-
vational work. It is as if conscience involves being “forced of itself to weigh its
own actions,” but also that judging that a specific action would make one “wor-
thy of some censure” leaves one unaffected until one judges that some sanction
(or censure) other than one’s own censure would actually take place. But what
motivational work is conscience then doing? It is as if the judgment that one is
worthy of censure is utterly external and without affect; in other words, that it
involves no implicit censure or blame of oneself.

52
More precisely, the evil is “non immerito” (not unmerited). I am indebted here to Knud
Haakonssen.

https://doi.org/10.1017/9781139025065.004 Published online by Cambridge University Press

 86 hobbes and pufendorf

Motivationally, it is hard to see the difference between a desire to avoid justi-

fied sanctions that involves no desire to avoid what would justify sanction and
the desire to avoid sanctions period. But Pufendorf is at pains to distinguish
these. Of course, if someone were to lack a general desire to avoid sanctions
and desired to avoid them only on the condition they were justified, then the
desire to avoid justified sanctions would clearly be motivationally distinct from
the desire to avoid sanctions generally. But Pufendorf of course thinks that we
do have a general desire to avoid sanctions, whether they are justified or not.
Someone who wants to avoid sanctions on the condition they are justified is
most plausibly seen as wanting to avoid what would justify sanctions, whether
or not the sanction will actually be applied.53
Pufendorf’s attempt to account for morality as a realm of moral “entities” and
properties that arise through God’s exercising his moral power to impose obli-
gations on human beings, including mutual obligations and rights they acquire
through the demand to be sociable, thus ends up a fascinating, but unstable,
balance of elements from Grotius and Hobbes. Like Grotius, Pufendorf empha-
sizes sociability, which he ties explicitly to the recognition of equal dignity and
mutual rights, something that is only implicit in Grotius. But whereas sociabil-
ity is foundational for Grotius, it is derivative for Pufendorf, and in a way that I
have argued makes Pufendorf’s view unstable. On the other hand, a connection
between morality and the “natural” power to impose sanctions, which is so
important for Hobbes in creating a robust realm of obligations in foro externo,
is no less essential for Pufendorf. But here again, this introduces instability into
Pufendorf’s theory, since he is at pains to distinguish his moral psychology and
view of the natural moral condition from Hobbes’s.
In the end, however, what seems most interesting in Pufendorf is the central-
ity of moral powers and its connection to reciprocal recognition and account-
ability. But here again, we encounter a fundamental instability, since the idea
of a “moral power” as Pufendorf understands it seems already to presuppose
a fundamentally equal sociability, dignity, and right, even though Pufendorf
seeks to derive these equal moral powers from God’s superior moral power.

53
Of course, the sanctions might be justified simply by the fact that they are legitimated by the
authority of the agent or body who sets them, say, God or some civil authority. I am not sug-
gesting that anything in Pufendorf’s thought requires justification on a case-by-case basis.

https://doi.org/10.1017/9781139025065.004 Published online by Cambridge University Press

 3

Locke and Cumberland

All three of the philosophers we have considered so far – Grotius, Hobbes,

and Pufendorf – are canonical representatives of the early modern natural
law tradition. In this chapter, we consider the tradition’s other two main fig-
ures: John Locke (1632–1704) and Richard Cumberland (1631–1718). Locke is
justly famous, of course, though more for his political philosophy and the more
familiar topics in epistemology and metaphysics treated in the Essay Concern-
ing Human Understanding than for his ethics. Cumberland is much less well
known. However, Cumberland’s contemporaries saw him, along with Grotius
and Pufendorf, as one of a “triumvirate of seventeenth-century founders of the
‘modern’ school of natural law” (Haakonssen 2001: 29). Moreover, Cumber-
land has special relevance to moral philosophy of the present day, since he was
among the first to attempt to “locate” moral truths and facts in relation to, and
perhaps reduce them to, those that are confirmable by the empirical sciences, a
naturalist program that remains vital today.1
The early modern natural lawyers put the juridical idea of morality (natural
law) at the center of ethical thought. But the word “morality” rarely appears
in translations of Grotius and Pufendorf (nor, indeed, in Hobbes), though it
features frequently in Barbeyrac’s notes to his early eighteenth-century trans-
lations and, of course, in his Historical and Critical Account of the Science of
Morality, which preceded his 1729 translation of Pufendorf’s De Jure Naturae
et Gentium.
By the time Locke published his Essay in 1690, however, the use of “moral-
ity” in our contemporary deontic sense to refer to universal deontic norms that
obligate all “moral agents” was evidently sufficiently well established in Britain
that Locke could refer to the golden rule, called by Hobbes the “sum” of the
laws of nature (Hobbes 1994b: XV.35), as “that most unshaken Rule of Moral-
ity” (Locke 1975: 68).2 Indeed, according to James Tyrell, the very project of

1
See Jackson 1998 on the “location problem.”
2
Locke says in the Essay that “Person is a forensick term, appropriating actions and their
merit; and so belongs only to intelligent Agents capable of a law” (1975: 346). About this,
Edmund Law writes, “Now the word person, as is well observed by Mr. Locke … is …
used in the strict forensic sense, denoting some such quality or modification in man as

87

https://doi.org/10.1017/9781139025065.005 Published online by Cambridge University Press

 88 locke and cumberland

the Essay had its roots in discussions Locke had in or before 1671 “about the
principles of morality and revealed religion” (Woolhouse 2007: 98).3 A central
Lockean claim in the Essay, in fact, is that “Morality [is] amongst the Sciences
capable of Demonstration” (Locke 1975: 549).
That “morality” had its modern deontic sense in England by this time is
illustrated also by Cumberland. Cumberland’s De legibus naturae disquisitio
philosophica originally appeared in 1672, although it was not translated into
English until 1727 as A Treatise of the Laws of Nature. Like Locke, Cumberland
is concerned to argue that the “rules” or “principles of morality” are not innate,
but that a “science of morality” can nonetheless be firmly established through
philosophical reflection on materials that are given through observation and
experience (Cumberland 2005: 264, 297, 489).
Although he is much less well known than Locke now, Cumberland actu-
ally has greater significance for the history of ethics, although not, of course,
for the history of political thought. Important present-day positions in moral
philosophy – for example, reductionist forms of ethical naturalism and empiri-
cal naturalist strains of utilitarianism – arguably find their first formulation in
Cumberland.
Cumberland is a reductionist both in normative ethics and in metaethics.
At the normative level, he argues that “all the Laws of Nature” or “true Rules
of Morality” can be “reduc’d to that one, of Benevolence towards all Ratio-
nals,” where benevolence is understood to aim at the “common Happiness of
all rational Beings” (237, 413).4 Metaethically, Cumberland’s project is to show
that “the Whole of moral Philosophy, and of the Laws of Nature, is ultimately
resolv’d into natural observations known by the Experience of all Men, or into
Conclusions of Natural Philosophy” (291). Unlike contemporary naturalists,
however, and certainly unlike Hobbes, Cumberland hastens to add that he
intends “natural philosophy” to be understood in “a large Sense” that includes
empirical inquiry into “the Nature of Souls” no less than “natural Bodies” (291).
Both Cumberland and Locke conceive of morality in terms of natural law,
following Grotius, Hobbes, and Pufendorf. But whereas the latter writers, I
have been stressing, maintain essentially juridical elements in their concep-
tions of natural law, right, and obligation, Cumberland and, in some ways,

denominates him a moral agent, or an accountable creature; renders him the proper subject
of laws, and a true object of rewards or punishments” (Law 1824: III, 179).
3
Draft B of the Essay contains a Table of Contents that includes within Locke’s projected
critique of innate ideas: “The goodnesse of god never imprinted those Ideas of him that
are to be found in mens minds. Noe god, noe idea of morality, which yet may be approved,
moral rules & conscience about them prove them not innate” (Locke 1990: 87). See also note
8 regarding this use of “morality” in Locke’s manuscripts on the law of nature, composed in
the 1660s.
4
“True Rules of Morality” is Maxwell’s translation of “veris Morum regulis” (Cumberland
1683: 118–119).

https://doi.org/10.1017/9781139025065.005 Published online by Cambridge University Press

 locke and cumberland 89

Locke, attempt to account for and ground a juridical conception of morality in

non-juridical terms, more specifically in good-based considerations that neces-
sarily weigh with a rational agent from a practical or deliberative point of view.
Accounting for morality’s normative force motivationally arguably operates
at some level in Hobbes’s thought also, but Cumberland’s reductionism goes
farther than anything in Hobbes. And Cumberland attempts to derive conclu-
sions that are explicitly unHobbesian. He advertises that in his Treatise, “Mr.
Hobbes’s Philosophy, as well Moral as Civil, are consider’d and refuted” (236).
More specifically, Cumberland aims to demonstrate that an empirically based
ethical philosophy leads not toward a Hobbesian “Epicurean[ism],” but in the
direction of what will later be called utilitarianism.5
What practical reason recommends to each agent, according to Cumber-
land, is not the agent’s own interest considered independently of others, but
rather the “happiness of all rational Beings.” Hobbes never attempts to reduce
moral or ethical thought or discourse to natural philosophy, holding that, like
thought and talk about color, it expresses rather than describes users’ mental,
and ultimately bodily, states. Cumberland holds, by contrast, that “practical
Propositions” and “Dictate[s] of Reason” can be formulated equivalently as
oughts, as obligating laws, or as propositions of natural philosophy concern-
ing what best conduces to the common happiness. “In my Opinion,” he writes,
“these several Forms of Speech, relating to the Law of Nature, mean the same
thing” (482–483).
In this way, Cumberland anticipates yet another important strand of modern
ethical thought. In eighteenth-century British moral philosophy, “obligation”
comes frequently to operate as a synonym for a rational motive or normative rea-
son for acting. Shaftesbury’s Inquiry Concerning Virtue or Merit (1711) provides
an excellent example. In Part I, Shaftesbury considers what virtue is and then
asks, in Part II, “[W]hat obligation is there to virtue, or what reason to embrace
it?” (Shaftesbury 2001: 192). Similarly, Hutcheson and Hume distinguish between
“natural” and “moral obligations” (Hume 1978: 3.2.2.23), holding the first to con-
sist, in Hutcheson’s words, in “a Motive from Self-interest, sufficient to deter-
mine all those who duly consider it” (Hutcheson 2004: 177). Moreover, Hume
and Hutcheson both understand moral obligation not as an irreducibly deontic
or juridical concept, but in terms of moral sense or sentiment.
In this chapter, we shall consider “naturalizing” moves that Cumberland,
and also Locke, make in these directions in attempting to understand moral-
ity and its normativity. Cumberland, especially, is an important influence on
Hutcheson, and through him, on Hume. Both Cumberland and Locke hold

5
252–253, where Cumberland explains his dialectical reasons for not basing a defense of
“natural Religion and Morality” on Platonism. These include that such a basis “can never be
prov’d against the Epicureans, with whom is our chief Controversy.”

https://doi.org/10.1017/9781139025065.005 Published online by Cambridge University Press

 90 locke and cumberland

that in being morally obligated to do something, the obligated agent must have
normative reasons to do it (what I call “morality/reasons existence internal-
ism”).6 And both hold that such normative reasons can exist only if the agent
could, under appropriate conditions, be motivated so to act (“reasons/motives
existence internalism”).7 That said, there are also significant elements in the
thought of both that tie both to the juridical moral philosophizing of Grotius
and, ­especially, Pufendorf.
Cumberland presents himself throughout as faithful to the theological vol-
untarist idea that morality and natural law cannot exist without a lawgiver,
although, by the time he has finished, Cumberland’s idea of the divine will
seems almost indistinguishable from his notion of practical reason. And an
important line of thought leading Locke to revise his views on freedom of the
will in the second edition of the Essay runs right through a connection Locke
assumes between moral obligation and justified punishment or accountabil-
ity. In these ways, Cumberland and Locke are important transitional figures
between more explicitly juridical theories of morality represented by Grotius,
Pufendorf, and, to a lesser extent, Hobbes, and less legalistic moral philoso-
phies to follow, including many that criticized the modern natural law tradi-
tion, like those of Shaftesbury, Leibniz, Hutcheson, and Hume.

Locke: Natural Law, Motive, and Sanction

We begin with Locke. Although Locke’s Essay (published in 1690) was the main
repository of his moral philosophy published during his lifetime, Locke’s first
theorizing about morality and natural law were contained in Latin manuscripts
composed in the 1660s that are now referred to as Essays on the Law of Nature
(Locke 2002).8 The main lines of Locke’s views in the Essays are not dissimi-
lar to Pufendorf’s Elements of Universal Jurisprudence, which Locke evidently
acquired soon after its publication in 1660 (Locke 2002: 38). Like Pufendorf,
Locke holds that morality and natural law depend upon divine command. And
though Locke has nothing like Pufendorf’s sophisticated, systematic theory of
“moral powers,” something akin to Pufendorf’s distinction between moral and
physical entities, along with the “imposed” and essentially “directive” character

6
Although Locke, as we shall see, does not think that moral facts or thoughts – for example,
that an action would be morally required – are intrinsically motivating themselves. In this
respect, he is an externalist. For a discussion of different species of internalism and external-
ism in moral philosophy, see Darwall 1997.
7
In Darwall 1995, I argue that both Cumberland and Locke can be placed within an empirical
naturalist tradition of British internalism about morality.
8
Note that the “modern” use of “morality” noted above appears also here, for example, “rules
of morality” (“morum regulas”) (Locke 2002: 166–167).

https://doi.org/10.1017/9781139025065.005 Published online by Cambridge University Press

 locke: natural law, motive, and sanction 91

of the former, seems to lie behind Locke’s Essay distinction between “ectypal”
and “archetypal” concepts, his corresponding distinction between “real” and
“nominal essences,” and his associated claim that “Morality [is] amongst the
Sciences capable of Demonstration” (because moral concepts are archetypal
and thus concern nominal and not real essences, which can only be discovered
empirically) (1975: 549).
Where Locke departs decisively from Pufendorf, and also from Grotius, is
in the latter’s doctrine that human moral agency essentially involves a social
nature. By the time of the Essay, Locke is explicitly a rational egoistic hedonist,
holding that normative reasons for acting can come only from considerations
of the agent’s own pleasure. Already in the Essays, Locke holds that the very
raison d’être of morality derives from rationally irresolvable conflicts of self-
interest to which the only solution is divine, sanction-backed legislation. More-
over, if it wasn’t already clear in the Essays, it is in Locke’s later writing, that
since he holds the agent’s pleasure to be the only source of normative reasons,
he cannot accept the fundamental distinction Pufendorf draws between being
moved by the legitimacy of sanctions and being moved by fear of sanctions
pure and simple.
This is a deep difference between Locke and Pufendorf. I have been suggest-
ing that much more is packed into sociable nature, as Pufendorf and Grotius
understand it, than is normally appreciated. To play the role it does in their
thought, sociability cannot just be an affiliative nature, like Stoic oikeiôsis, that
gives human beings a genuinely common good. It must involve a reciprocally
recognizable common standing or authority that can justify legitimate claims
and demands. So Locke’s disagreement with Pufendorf and Grotius concerns
not simply human psychology, or even the nature of human good. It concerns
whether moral agency essentially includes the capacity to recognize and be
guided by normative reasons of a kind that differ from those of good conceptu-
ally, viz., legitimate claims and demands of right that presuppose the capacity
for accountability and reciprocal recognition of authority.
Locke is clear, however, that morality and moral obligation are themselves to
be understood not in terms of agents’ goods, but in irreducibly juridical terms.
Natural law contains “all the requisites of a law” (2002: 111). “It is the decree
of a superior will”; “it lays down what is and what is not to be done”; and “it
binds men, for it contains in itself all that is requisite to create an obligation”
(111–113). Like Pufendorf, Locke holds that natural law comes into existence
through divine imposition. Because God has superior authority over us and so
the “right” to our obedience, we become obligated through his decreed will.9

9
Somewhat like Pufendorf, but also in ways that anticipate his labor theory of property in
the Second Treatise, Locke holds that God acquires this right through the creation (Locke
2002: 185).

https://doi.org/10.1017/9781139025065.005 Published online by Cambridge University Press

 92 locke and cumberland

Now so far, Locke could simply be taking Suárez’s position, as did his Brit-
ish contemporary, Nathaniel Culverwell.10 Suárez rejected Aquinas’s view that
a teleological order, Aquinas’s “eternal law,” with a robust common human
good, is sufficient to constitute natural law. Thomist eternal law is sufficient,
Suárez held, to provide natural law’s content, but not yet to give it the form
of law. For this, something explicitly juridical, God’s authoritative command,
is required. But though authoritative imposition is necessary for a law of any
kind, it is not necessary, Suárez held, for what Irwin calls “natural rightness”
(Irwin 2008: 28–61). The standards of conduct that give natural law’s content
are provided by the teleological order of nature, which also grounds natural
law’s reason-giving force. Independently of divine command, there would be
reason enough to comply with natural law owing to the common goods it real-
izes that are essential to human agents’ telos. God’s imposition simply adds a
further reason (of obedience).
This is not, however, Locke’s position in the Essays. Locke declares at the out-
set that he assumes “there will be no one to deny the existence of God, provided
he recognizes … that there is a thing that deserves to be called virtue or vice”
(2002: 109). Locke assumes, that is, that people will recognize that there cannot
be right or wrong, virtue or vice, without God’s command. Like Pufendorf,
Locke holds that all moral properties and “entities,” obligations and “moral
good or virtue,” depend on a moral law that is authoritatively imposed by God
(2009: 109). Moreover, Locke holds that the content of natural law is not pro-
vided by what is good for agents to do independently of authoritative demand.
Utility is not the basis of the law or the ground of obligation, but the con-
sequence of obedience to it. Surely, it is one thing for an action of itself to
yield some profit, another for it to be useful because it is in accordance with
the law, so that if the law were abolished it would have no utility whatever:
for example, to stand by one’s promise, though it were to one’s own hin-
drance (2002: 215).

It is true, as Irwin frequently notes, that Suárez and Aristotelian naturalists

more generally distinguish between what is useful or expedient (utile) and nat-
urally right actions that are intrinsically good (the “right as good” or bonum
honestum) (Irwin 2008: 31, 34). However, Locke can hardly be saying that natu-
ral law consists of standards that dictate acts that are inexpedient, but nonethe-
less beneficial to agents since they realize intrinsic common goods. If Locke
thought that, he would not say that natural law can require one “to stand by
one’s promise, though it were to one’s own hindrance.”
Locke’s view is that goods and evils that come to an agent from following or
violating natural law derive not, or at least not always, from the acts themselves

10
For a discussion of Culverwell, see Darwall 1995, esp. pp. 27–33.

https://doi.org/10.1017/9781139025065.005 Published online by Cambridge University Press

 solution to the human natural 93

considered independently of their being morally required or forbidden, but

from consequences of their respecting or violating the law, that is, from their
being obedient or disobedient.
In fact we must distinguish between an action as such and obedient action,
for an action itself can be inexpedient … whereas an obedient action is use-
ful in so far as it averts a penalty due to a crime … And thus the rightness
of an action does not depend on its utility; on the contrary, its utility is a
consequence of its rightness (Locke 2002: 215).

This rejection of Suárezian natural rightness is literally the last word of Locke’s
Essays. And it makes clear the central role of divine sanctions in his view.
Locke holds that both God and “the immortality of souls … must be neces-
sarily presupposed if natural law is to exist” (173). Without God, there will be no
law-maker and so no source for natural law. But why the immortality of souls?
Just as, Locke says, “[T]here is no law without a law-maker,” “law is to no purpose
without punishment” (173). Natural law must presuppose immortality of the soul
because (eternal) divine sanctions are necessary for law, and these require that
human souls be immortal so that the requisite punishment can be meted out.

Morality as Solution to the Human Natural Collective

Action Problem
Locke must be thinking that the reason divine supernatural sanctions are nec-
essary for natural law is that without them human beings would lack adequate
reason to comply with natural law. We know he thinks that adequate reasons
cannot be independent of God’s authoritative command. But if the soul’s
immortality is necessary for the kind of punishment that natural law requires,
it must be because natural law requires sanctions of a kind that only God could
mete out supernaturally, through eternal goods and evils.
Locke’s egoistic hedonism is clearer in the Essay:
If willing be but the being better pleased, as has been shewn, it is easie to
know what ’tis determines the Will, what ’tis pleases best: every one knows
’tis Happiness, or that which makes any part of Happiness, or contributes
to it; and that is it we call Good (1975: 249n).

Some form of egoism is already evident in the Essays, however, in Locke’s holding
that divine sanctions are necessary to establish natural law and give it a “purpose.”
A second reason Locke gives for thinking that the natural law cannot be grounded
simply in agents’ interests is the fact that human interests can conflict. If natural
law were grounded simply in agents’ interests, conflicts of interest would make
it impossible for people jointly to obey natural law. But “it is impossible that the
primary law of nature is such that its violation is unavoidable” (2002: 211).

https://doi.org/10.1017/9781139025065.005 Published online by Cambridge University Press

 94 locke and cumberland

Human goods conflict because of natural scarcity. Nature’s goods “have

been bestowed in a definite way and in a predetermined quantity; they have
not been fortuitously produced nor are they increasing in proportion to what
men need or covet. Clothes are not born with us, nor do men like tortoises, pos-
sess and carry about shelters” (211). And when one person takes away from the
common stock of natural goods, “he takes away from another man’s heap the
amount he adds to his own” (211).
These themes are familiar from the Second Treatise, but already in the Essays
Locke is concerned about the role of property in human good. It is the human
need and desire to appropriate that makes for conflicts of interest that human
beings cannot resolve without divine intervention in the form of supernatural-
sanction-backed natural law. “The strongest protection of each man’s private
property,” Locke writes in the Essays, “is the law of nature,” which enables a
man to “be master of his property … and pursue his own advantage” without
undermining others’ capacity to pursue theirs (207).
Left to our own devices, we human beings face a classic collective action
problem, something like an n-person prisoners’ dilemma, which our created
egoistic nature bequeaths to us. We humans see that this is so, moreover, and
the very empirical evidence of divine design that persuades us that a supremely
wise, benevolent, and powerful creator exists, can convince us also that God
must command us to act on mutually advantageous laws to solve our collective
action problem. Natural law or morality is the solution to the collective action
problem that human beings face that would otherwise defeat them.
Since we know that God is perfectly wise, powerful, and benevolent, we
have sufficient evidence that he wills us to conform to principles that it is in the
interest of each that all follow. But since, as Locke supposes, only agents’ own
goods give them reasons for acting, however broadly their good may be con-
ceived, it follows also that God’s will would be vain if he did not provide further
incentives, in the form of eternal sanctions, to comply with the law. And for
these to exist, the soul must be immortal. In other words, from the evidence
that supports the existence of an omnipotent, omniscient, and omnibenevo-
lent creator, Locke believes we can draw the following further conclusions: the
existence of natural law and eternal supernatural sanctions, and the immortal-
ity of the soul. All, he must be thinking, are necessary conditions for the very
possibility of morality and reasonable social order, given the circumstances of
human life as we find it.
The role of divine sanctions in giving human beings incentives without
which they would lack good reason to comply with natural law is even clearer
in later writings and the Essay.11 Before we consider these, however, we should
11
See, however, Sheridan 2007, which notes that there are places where Locke allows that the
pleasure of doing our duty or love can motivate moral action, for example: “Happiness …
is annexed to loving others or our doing our duty” (Locke 1997: 319).

https://doi.org/10.1017/9781139025065.005 Published online by Cambridge University Press

 right and good 95

notice a deep tension already apparent in the Essays between Locke’s egoism
and his juridical view of moral obligation. We have seen that Locke is clear that
there is a conceptual difference between natural law and (agents’) good. Show-
ing that something will benefit an agent, intrinsically or extrinsically, does not
establish that it is legitimately demanded of them. No claim of right follows
from these facts of good. For deontic claims to hold, authority and accountabil-
ity relations must be in place that can legitimate demands and make the agent
answerable for complying with them. Neither can the content of legitimate
demands be given by the agent’s good. Rather, natural laws consist of norms
directing agents to perform acts, like keeping their promises or respecting the
property of others, even to their own “hindrance.” Everyone’s conforming to
such norms benefits everyone, however.

Right and Good

Natural law is thus juridical in concept and form. Like Suárez and Pufendorf,
Locke holds that something does not count as law and cannot create genuine
obligations unless it exists within a framework of authority and accountability.
This is the conceptual or formal difference between the deontic concepts of
law, right, and moral obligation, on the one hand, and the concept of good,
whether extrinsic (utile) or intrinsic and common goods (bonum honestum),
on the other. Further, Locke disagrees with Suárez, and agrees with Pufendorf,
in thinking that natural law differs from the good also as a matter of content.
There is no natural moral rightness that is a species of intrinsic (common)
good. The right cannot be understood as a bonum honestum.
But this does not mean that human agents lack adequate reasons of self-
interest for complying with natural laws created by God’s commands. To
the contrary, since God punishes disobedience, there is always adequate self-
interested reason to comply. Still, though God’s sanctions are what give agents
reasons to comply with his legitimate demands, they are not what obligate them
to do so.
All obligation binds conscience and lays a bond on the mind itself, so that
not fear of punishment, but a rational apprehension of what is right, puts
us under an obligation, and conscience passes judgment on morals, and, if
we are guilty of a crime, declares that we deserve punishment (185).

This means, as we shall see more clearly presently, that there is, for Locke, a gap
between conscientious judgments of right, legitimate demand, obligation, and
deserved punishment, on the one hand, and rational motivation on the other.
Though it is guaranteed that there will always be sufficient reasons to comply
with moral obligations, that is not because moral obligations entail or create
reasons for acting themselves.

https://doi.org/10.1017/9781139025065.005 Published online by Cambridge University Press

 96 locke and cumberland

The coincidence between obligation and practical reasons is guaranteed

extrinsically through divine sanctions. In this sense, Locke is a moral exter-
nalist. He is a judgment externalist, since he thinks that the judgment that
something is morally obligatory is not inherently motivating. And he is, in one
important sense, a (morality/reasons) existence externalist, since he thinks that
something’s being morally obligatory does not in itself give him any reason,
much less overriding reason, to do it.12
We should compare Locke’s view with Pufendorf’s distinction between being
moved by fear of sanctions and acting “from obligation” through an “intrinsic
impulse of the mind” consisting in the agent’s own “censure” of violating moral
obligations and their sense that they would be subject to sanctions that would
fall upon them “justly” (Pufendorf 1934: 91, 386). As we noted, Pufendorf also
says things that suggest a view of motivation that is more like Locke’s. But in the
quoted passages, at least, Pufendorf clearly distinguishes himself from Locke
in holding that genuine obligation is impossible unless moral agents have the
capacity to be moved by moral obligations intrinsically through their own
acceptance of authority that legitimates demands, and through their capacity
to hold themselves responsible to authority through their own censure.
That this is not Locke’s view is even clearer in his later works. Thus we find in
a 1693 entry in his Commonplace Book:
The pleasure that a man takes in any action or expects as a consequence of
it is indeed a Good in it self able & proper to move the will. But the Moral
Rectitude of it considered barely in it self is not good or evill nor in any way
moves the will but as pleasure & pain accompanies the action it self or is
looked on to be a consequence of it. Wch is evident from the punishments
& rewards wch god has annexed to moral rectitude or pravity as proper
motives by ye will wch would be needless if moral rectitude were it self good
and moral pravity evil (quoted in Colman 1983: 48–49).

In the Essay, Locke holds that “Good and Evil … are nothing but Pleasure and
Pain” and, that since
it would be utterly in vain, to suppose a Rule set to the free Actions of Man,
without annexing to it some Enforcement of Good and Evil, to deter-
mine his Will, we must, where-ever we suppose a Law, suppose also some
Reward or Punishment annexed to that Law (Locke 1975: 351).

Locke distinguishes three different kinds of laws – divine law, civil law, and the
law of “Opinion or Reputation” by their respective “Enforcements, or Rewards
and Punishments” (351).

12
In another sense, he is a morality/reasons existence internalist, since morality can only
exist through God’s commands and sanctions, which guarantee that there is always suf-
ficient self-interested reason for complying with morality.

https://doi.org/10.1017/9781139025065.005 Published online by Cambridge University Press

 right and good 97

In the Essay’s final chapter, Locke advances a three-fold division of knowl-

edge: logic and semantics, theoretical knowledge or natural philosophy, and
practical knowledge (1720).13 Practical knowledge concerns “[t]hat which Man
himself ought to do, as a rational and voluntary Agent, for the attainment of
any End, especially Happiness.” And Locke defines “Ethicks,” as “the seeking
out those Rules, and Measures of humane Actions, which lead to Happiness,
and the Means to practise them” (1975: 720).
It is the connection to agents’ pleasure or happiness, their only rational
motive, that gives ethical thought its practical point for Locke. The writings of
ethical writers like “Aristotle” or “Confucius,” which attempt to pick out vir-
tues and vices, “evaporat[e] only into words, disputes, and niceties.” “Without
showing a law that commands or forbids them,” and that includes sanctions
for disobedience, “moral goodness will be but an empty sound” (King 1830:
129–130). What makes the conceptions figuring in natural law and moral good
and evil more than mere names is their practical force. It is because God’s sanc-
tions connect natural law to agents’ goods that they become practical. Only in
this way can human beings acquire motives to make the world fit archetypal
moral notions and make them genuinely practical.
Locke contrasts the intrinsically practical knowledge at which ethics aims
with knowledge of “[t]he Nature of Things, as they are in themselves, their
Relations, and their manner of Operation” (1975: 720). Knowledge of our
moral obligations does not require knowledge of real essences or substances,
most importantly, of human beings themselves. “For as to Substances, when
concerned in moral Discourses, their divers Nature are not so much enquir’d
into as supposed, v.g. when we say that Man is subject to Law: We mean noth-
ing by Man but a corporeal rational Creature: What the real Essence or other
Qualities of that Creature are in this Case, is no way considered” (516). The
basis for the law of nature thus involves nothing about the real essence of
human beings.
This marks a real difference between Locke and Suárez – and a similarity
between Locke and Pufendorf – that points in the direction of the difference
between views called “constructivist” in contemporary discussion and “natu-
ralist” views, whether Aristotelian naturalist in Irwin’s sense, or in our contem-
porary sense of non-teleological ethical naturalism. There is an important way
in which Locke “naturalizes” natural law, since he holds that the juridical, obli-
gating character of natural law is, from a practical point of view, a fifth wheel.
It has no intrinsic deliberative role, and without being connected to human
motivation (through sanctions that are, albeit, supernatural), moral, irreduc-
ibly juridical concepts appear to be mere names lacking practical force, which
are no part of “Ethicks,” by Locke’s criterion.

13
Here, I follow Darwall 1995: 45–46.

https://doi.org/10.1017/9781139025065.005 Published online by Cambridge University Press

 98 locke and cumberland

Even so, once we have the concept of a person, that is, a rational corporeal
creature subject to law (346, 516), know that rational creatures can act only on
their own good, and know enough about the human situation to draw conclu-
sions about God’s will and sanctions, we know enough to draw conclusions
about natural law. No further inquiry into the real essence of human beings is
required.14

Law, Accountability, and Autonomy

Before turning to Cumberland, we should note one further feature of Locke’s
view that manifests a tension he copes with between holding that moral con-
cepts are essentially deontic or juridical, and so irreducible to the good, on the
one hand, and holding, on the other, that only the agent’s good gives them
normative reasons for acting and so is intrinsically “practical” and therefore
“ethical.” Since morality is tied to accountability through the nature of juridical
concepts and, since, Locke came to believe, somewhat like Pufendorf, someone
can intelligibly be held answerable for complying with law only if they have the
agential capacities sufficient to determine themselves to comply with it, Locke
required a theory of free will sufficient to explain how human agents could
intelligibly be thought subject to natural law.
In the first edition of the Essay, Locke already declares his agreement with
“those who cannot conceive, how any thing can be capable of a Law, that
is not a free Agent” (76). By “capable of law,” again, Locke means capable
of being subject to law and obligated by it. But Locke’s view of freedom
there, that freedom consists in the absence of obstacles to agents’ doing
what they will, was largely Hobbesian (276). And he follows Hobbes also in
holding that the question about whether the will itself can be free involves
a category error (241, 247). In the first edition, Locke has no conception of
a capacity for free determination of the will itself, that is, of autonomy or
self-determination.

14
There are places in the Essays where Locke’s view can seem much like Suárez’s, for exam-
ple, when Locke uses Suárez’s term “debitum naturale,” saying that there is a “natural obli-
gation … to fulfil the duty which it lies upon one to perform by reason of one’s nature”
(Locke 2002: 181). But Locke there adds: “or else submit to the penalty due to the perpe-
trated crime” (181). For Locke, sanctions for violating divinely imposed law are necessary
to give natural law normative or practical force, whereas Suárez’s idea of a natural duty
consists of a standard agents already have sufficient eudaimonistic reason to comply with
(bonum honestum) independently of God’s commands and sanctions. What Locke obvi-
ously means by “duty which it lies upon one to perform by reasons of one’s nature” is that
the content of natural law derives from incontestable features of the human condition that
require no inquiry into human essences: corporeal rational beings placed in circumstances
of relative scarcity with some means for mutually advantageous collective action, but who
are unable, without God’s sanctioned law, rationally to undertake it.

https://doi.org/10.1017/9781139025065.005 Published online by Cambridge University Press

 law, accountability, and autonomy 99

But why would such a conception be necessary? Why after all, are only “free
Agent[s]” “capable of a Law”? The answer has to do with conceptual connec-
tions Locke sees between the concepts of person, imputation, and accountabil-
ity. “Person,” Locke says, “is a Forensick Term appropriating Actions and their
Merit; and so belongs only to intelligent Agents capable of a Law” (346). The
identity of a person over time, “whereby it becomes … accountable,” presup-
poses an agent who consciously “owns and imputes to it self past Actions” on
the same grounds “that it does the present” (346).
Still, why does such agency require anything beyond the human capacity
to grasp natural law and that God punishes disobedience? If people were able
to do whatever they think will be best for them in the long run, nothing more
might be necessary, since once they are aware of divine sanctions, they would
naturally prefer not to violate natural law. By the time of the second edition,
however, Locke came to believe that this is not so and significantly revised his
account of free agency.15
What is it that determines the Will in regard to our Actions? And that upon
second thoughts I am apt to imagine is not, as is generally supposed, the
greater good in view: But some (and for the most part the most pressing)
uneasiness a Man is at present under (250–251).

The problem is that even if people know that they will be punished for violat-
ing natural law, and even if they believe complying with the law is important
for their greater long-term good, they will not comply with it if the threat of
punishment does not make them sufficiently uneasy or if they do not desire to
avoid the punishment first and foremost.16
If, consequently, human agents are to have the capacity to guide themselves
by natural law, they will require some capacity to bring the strength of desire
(“present uneasiness”) into alignment with their assessments of long-term
benefit. In the Essay’s second edition, Locke describes precisely such a power of
self-determination.
For the mind having in most cases, as evident in Experience, a power to sus-
pend the execution and satisfaction of any of its desires, and so all, one after
another, is at liberty to consider the objects of them; examine them on all sides,
and weight them with others. In this lies the liberty Man has; and from the not
using of it right comes all that variety of mistakes, errors, and faults which we
run into, in the conduct of our lives, and our endeavours after happiness (263).

15
Mostly, in Chapter 21, “Of Power.” These are analyzed much more fully in Darwall 1995:
149–175, from which I here draw, and Yaffe 2000.
16
One example Locke gives in the second edition is humorous when viewed against our
contemporary background (say, after doing a Google search on “akrasia” and “smoking”):
“Bread or Tobacco may be neglected, where they are shewn to be useful to health, because
of an indifferency or disrelish to them” (1975: 280).

https://doi.org/10.1017/9781139025065.005 Published online by Cambridge University Press

 100 locke and cumberland

However strong our current desires, we have the power to step back from them
and “examine” their objects “on all sides.” “Thus, by a due consideration and
examining any good proposed, it is in our power to raise our desires, in a due
proportion to the value of that good, whereby in its turn, and place, it may
come to work upon the will, and be pursued” (262). “[H]e that has a power to
act, or not to act according as such determination directs,” Locke concludes, “is
a free Agent” (266).
Despite the fact that human beings always act on their strongest desire, then,
they can nonetheless be held “answerable” and “justly incur punishment”
when they violate natural law. “If the neglect or abuse of the Liberty he had, to
examine what would really and truly make for his Happiness, misleads him,
the miscarriages that follow on it, must be imputed to his own election” (271).
Locke’s doctrine of self-determination, though prompted by the very same
consideration of natural law’s conceptual connection to accountability as
Pufendorf’s account of moral agency, nonetheless proceeds entirely within
an egoistic hedonist framework. Contra Pufendorf, though we must be self-
determining agents to be answerable and so “capable of a Law,” this does not
include any “intrinsic impulse of the mind” to recognize and hold ourselves
accountable to legitimate authority. For Locke deontic moral thoughts are,
in themselves, motivationally inert. That the capacity for free agency is made
necessary by the moral law does not mean that morality must have any intrin-
sic practical “force” for the agent itself. God’s sanctions bring a coincidence
between natural law and our greatest long-run pleasure, and self-determining
moral agency is a hedonistic capacity to bring the strengths of our current
desires into alignment with that long-run pleasure. For Locke, any normative
force the right has can come only from the good.
The connection Locke draws between autonomy or self-­determination and
being subject to the moral law is one whose power and influence reverberated
throughout the history of modern moral thought. It is possible that Locke him-
self first encountered the idea in the philosophy of the Cambridge Platonist
Ralph Cudworth, to be discussed in the next chapter.17 We will find other ver-
sions of this notion in the thought of Shaftesbury, Butler, Rousseau, and Kant, as
well as in forms of neo-​Kantian moral philosophy of the late twentieth century.

Cumberland
We turn now to Cumberland. Cumberland is best known for being, as Ernest
Albee put it, “the true founder of English utilitarianism” (Albee 1957: 1).
Although historians of ethics duly note this, they tend to take less account of
17
In Darwall 1995, I suggest that Locke may have had access to Cudworth’s manuscripts on
freedom of the will (see Darwall 1995: 172–175). However, see Broad 2006, which argues
that this is unlikely, and that it is more likely that Locke learned Cudworth’s ideas through

https://doi.org/10.1017/9781139025065.005 Published online by Cambridge University Press

 cumberland 101

Cumberland’s views concerning the nature of practical reason and its relation
to morality.18 It may be, however, that these foundational aspects of Cumber-
land’s thought have even more historical significance.19
We have been stressing the Sidgwick/Anscombe thesis that much of mod-
ern moral philosophy departs from an Aristotelian naturalist framework by
insisting on the juridical or deontic character of the right as a normative notion
that cannot adequately be conceived as a “species of the Good” (Sidgwick 1967:
105–106). But there is a second way in which some modern thinkers diverged
from Aristotelian naturalism, even about the good, namely, by rejecting the
eudaimonist thesis that practical reason necessarily aims fundamentally at
eudaimonia or the agent’s own good.
Cumberland makes an important move here. Arguing against Hobbesian
egoism, he maintains that others’ goods are no less good than the agent’s own
and therefore are no less choiceworthy. He holds, first, that as a matter of ratio-
nal will and action, it is no less possible for agents to aim directly at the good of
others than it is for them to aim at their own. And second, he holds they have
no less reason to do so. Thus begins a tradition according to which practical
reason includes rational benevolence no less than self-regard that runs through
Hutcheson and Sidgwick.
Cumberland also arguably stands at the beginning of another, empiri-
cal naturalist tradition in metaethics that seeks to “resolv[e]” “the Whole of
moral Philosophy … into natural Observations known by the Experience of
all Men, or into Conclusions of true Natural Philosophy” (Cumberland 2005:
291).20 Cumberland advances a naturalist account of the good: the “best Effect”
accomplishable by human action is the “greatest Happiness of all rational
Agents” (258, 487). And although he is formally a theological voluntarist, and
recognizes that without its distinctive deontic features, morality will not take
the form of a law, he ultimately seeks to reduce these naturalistically as well.
A Treatise of the Law of Nature, Cumberland says, is “a Confutation of
the Elements of Mr. Hobbes’s Philosophy” (287). Against Hobbes, Cum-
berland seeks to establish that there are “certain Propositions of unchange-
able Truth, which direct our voluntary Actions … and impose an Obligation

conversations with Damaris Cudworth Masham, in whose house Locke spent his last years
(Broad 2006). Carter 2010 notes contrary evidence.
18
See Albee 1957: 1–51, Sidgwick 1964: 173–175. See also Sharp 1912, Haakonssen 1996: 50–51
and 2001, and Crisp 2019: 33–48.
19
I follow closely, and sometimes draw text from, Darwall 1995: 80–108.
20
But note: “Natural Philosophy, in the large Sense I now use it, does not only comprehend all
those Appearances of natural Bodies, which we know from Experiment, but also inquires
into the Nature of our Souls, from Observations made upon their Actions and distinguishing
Perfections, and at length leads Men, by the Chain of natural Causes, to the Knowledge of the
first Mover, and acknowledges him to be the Cause of all necessary Effects” (2005: 291).

https://doi.org/10.1017/9781139025065.005 Published online by Cambridge University Press

 102 locke and cumberland

to external Actions, even without Civil Laws, and laying aside all Consid-
eration of those Compacts, which constitute Civil Government” (289). We
will consider his accounts of the directive and of the obligatory aspects of
natural law in turn.

Practical Propositions
To “direct” human action, natural philosophy must establish what Cumberland
calls “practical propositions.” Cumberland accepts the traditional Aristotelian/
Thomist account of action as invariably aiming at the good, though he seeks
also to reinterpret it. As we’ve just noted, he holds that agents can aim at others’
goods directly and not just as a species of their own. And Cumberland’s version
of the traditional account is empirical naturalist. An end is simply an “Effect”
the idea of which, “preconceiv’d in the Mind, first moves a Rational Agent to
intend the producing it” (506). Practical propositions therefore “pronounc[e]
concerning the Consequences of human Action” (258).
For laws of nature to be practical propositions, then, they must concern
“effects.” But this can hardly be sufficient for them to be directive. Any empiri-
cally confirmable proposition concerning what will happen if one does some-
thing counts as practical by this standard. Laws of nature aim, however, to
direct action more specifically, to say what one should do when doing one thing
would have one effect and doing something else would have another. More is
required for directing action than just pointing out the consequences of vari-
ous alternatives before the agent. The practical question is which alternative to
choose.
The law of nature directs us to benevolence or concern for the common
good. Unlike other practical propositions, whose guidance depends entirely
upon an optional end – for example, those that concern what will happen if one
performs some geometrical construction – “the Effects of a care of the common
Good … nearly concern all” (263). We are not free not to have the end the law
of nature directs us to; we have it necessarily.
It is not entirely clear exactly what Cumberland supposes this necessary end
to be. Here is his statement of the “fountain” of all natural law:
The greatest Benevolence of every rational Agent towards all, forms the hap-
piest State of every, and of all the Benevolent, as far as it is in their Power,
and is necessarily requisite to the happiest State, which they can attain, and
therefore the common Good is the supreme Law (292).

It is natural to assume that “Effects of a care of the common Good” at least

include the agent’s own happiness. Given some other things Cumberland says,
we might suppose that he is saying that the law of nature directs every agent to
be benevolent because that is necessary to the agent’s own happiness.

https://doi.org/10.1017/9781139025065.005 Published online by Cambridge University Press

 obligation 103

[A]n Action is … necessary to a rational Agent, when it is certainly of the

Causes necessarily requisite to that Happiness, which he naturally, and
consequently necessarily, desires (554).

As we will see, however, Cumberland clearly thinks we can at least no less ratio-
nally aim at others’ happiness as we can aim at our own. And his considered
view seems to be that the happiness of all is a better end to aim at, indeed that it
is the “best effect” or end.

Obligation
So far, then, we have the idea that the law of nature directs us to benevolent
concern for all because the “Effects of a care of the common Good” inescapably
concern any rational agent. We shall return to the further issue we have just
broached – whether the inescapable end is to be understood egocentrically or
impartially – presently. First, however, let us begin to understand the sense in
which Cumberland thinks the law of nature obligates.
This is essentially the same issue as whether morality provides genuine
law. “Laws,” Cumberland holds, are “nothing but practical Propositions, with
Rewards and Punishments annex’d, promulg’d by competent Authority” (253;
see also 543). So far, this is basically the same view as Locke’s and Pufendorf’s.
But there are important differences in Cumberland. First, it is important to
Locke’s view that morality’s sanctions are supernatural. Our natural condition,
Locke thinks, is like a prisoner’s dilemma, which can be solved only by God’s
creating supernatural, eternal rewards and punishments that can offset natural
individual advantages of collectively disadvantageous action. And second, it is
crucial to Locke’s view that these be genuine sanctions, that is, that the rewards
and punishments be consequences, respectively, of obedience and disobedience
of God’s authoritative will.
Both are differences from what Cumberland holds. Although Cumberland
accepts eternal sanctions in Locke’s sense (600), they play no role in estab-
lishing natural law. The “Rewards and Punishments” with which he is mostly
concerned are those that are “naturally connected” to the actions natural law
directs. The natural connection between benevolence and “the happiest state of
any and all the Benevolent” is sufficient to make the latter a reward. If this natu-
ral connection can be established and “reduc’d into the Form of [a] Practical
Propositio[n],” then that is “all that is essential to a Law” (499).
Cumberland believes that human beings can solve their collective problems
with motives they already have at hand in knowing the natural consequences of
their motivated actions. The “Rewards and Punishments” through which God
enacts practical propositions that constitute the law of nature are none other
than those referred to in the law of nature itself. Thus, unlike both Locke and

https://doi.org/10.1017/9781139025065.005 Published online by Cambridge University Press

 104 locke and cumberland

Pufendorf, Cumberland holds that the fact that the law of nature proceeds from
God plays no role in giving human agents motives to follow it. If the natural
realm were to exist without God, human agents would have the same reasons to
be mutually benevolent. Unlike Locke and Pufendorf, therefore, Cumberland
does not hold that agents’ motives for compliance with natural law depend
upon its being law, formally so called. The thought that in contravening natural
law one is disobeying God or violating his authoritative will has no motivational
force.
Formally, again, Cumberland holds that law must be “promulg’d by compe-
tent authority” (253). But God’s authority turns out not to be independent of a
proper understanding of natural law itself. As Cumberland puts it,
[I]t amounts to the same thing, when we say, “That the Obligation is an Act
of the Legislator”, or the First Cause; as if in this place we had call’d it, “An
Act of the Law of Nature.” For the Legislator obliges by the Law sufficiently
promulg’d, and he sufficiently promulges it, when he discovers to our
Minds, “That the prosecution of the Common Good is the Cause necessar-
ily requisite to that Happiness, which every one necessarily desires” (555).
It turns out, therefore, that the existence of normative reason for following
natural law’s “direction” and the obligatory normative force that is distinc-
tive of moral law are both reduced to the empirical naturalistic generalization
that is the natural law’s content. Thus are “moral Philosophy” and “the Laws of
Nature … ultimately resolv’d into natural Observations known by the Experi-
ence of all Men, or into Conclusions of true Natural Philosophy” (291).

Empirical Grounds for the Law of Nature,

Understood Egocentrically
So far, we have said nothing about why Cumberland thinks the natural law
holds as a matter of empirical fact. We have provisionally understood the law
to assert that all rational human agents are likeliest to have the happiest lives
within their power if they act with the greatest benevolence “to all rationals” –
to fellow human beings and to God. But why does Cumberland think this is so?
All his foundational theses would count for nothing against Hobbes unless he
can establish this apparently unlikely empirical claim in the face of Hobbes’s
powerful arguments to the contrary.
Unlike Hobbes and Locke’s Epicureanism, Cumberland holds a more Aris-
totelian theory of well-being as the enjoyable exercise of human faculties (523).
As we shall better appreciate in the next section, Cumberland believes he can
reasonably claim that acting with universal benevolence more satisfyingly
exercises our powers than does acting with purely self-regarding motives (266).
An extensive benevolence is also intrinsically satisfying because it maximally
actualizes the joy of love and the tranquility of a mind without any conflict

https://doi.org/10.1017/9781139025065.005 Published online by Cambridge University Press

 practical dictates of reason, agreement 105

between what it wills for itself and what it wills for others. Benevolence is use-
ful as well, moreover, since it makes agents less vulnerable to various maladies
(both psychological and physiological) that afflict the self-absorbed. “Hatred
and Envy, which fill the Mind of him who regards his own Good only, are neces-
sarily accompanied with Trouble and Sadness, Fear and a Solitary State, which
are evidently inconsistent with a Happy Life” (528). Finally, many of the most
important human goods are common goods, through which human beings
enjoy in common their joint exercise of natural powers.21
Part of Cumberland’s disagreement with Hobbes thus concerns the nature
of human good. Hobbes sees the state of nature as one of conflict owing to
natural human desires for power, glory, and scarce resources. Cumberland, by
contrast, holds that the peace that derives from mutual non-aggression and,
even more, from mutual benevolence, is not only the best outcome agents can
rationally hope for given the power behind others’ interests. Rather, mutual
benevolence and its fruits fulfill human beings’ most important needs and
desires.
We need not worry about the details of Cumberland’s empirical arguments.
Whether the law of nature provides every agent with adequate reasons, based
in their own good, for being benevolent may matter less to his overall argument
than it initially seems. We shall see why in the next section.

Practical Dictates of Reason, Agreement, and the Best End

We begin with Cumberland’s account of the will. Cumberland holds the
traditional view that all action aims at some good, but he modifies it in two
important ways. First, as we have seen, he takes action to be fundamentally
instrumental, and therefore, to be capable of being justified only by the good-
ness of natural effects it realizes. And second, he takes these justifying effects
to include others’ goods no less than the agent’s, since they are no less good. It
is possible, Cumberland argues, for human agents “to exercise Love, not only
towards themselves, but also towards God, and Men, partaking of the same
rational Nature with themselves” (312). “‘[T]he Understanding is capable of
judging, what promotes the Good of others, as well as what promotes our own;’
nor is there any Reason, ‘Why we cannot will those same things, which we have
judg’d to be good’” (473).22

21
We might say that this is an Aristotelian Naturalist strain of his thought in Irwin’s sense.
22
Sharp points out that “the principle … that egoism and altruism have the same psycho-
logical root” “lies at the very foundation of [Cumberland’s] system,” since Cumberland
believes that “the good as such, independently of its relationship to this or that possible
possessor, tends to arouse the desire to realize it” (Sharp 1912: 380). As we will see in
the next point, all desirable goods are goods of some being or other. This also would be
Hutcheson’s view, arguably under Cumberland’s influence.

https://doi.org/10.1017/9781139025065.005 Published online by Cambridge University Press

 106 locke and cumberland

Cumberland’s account of action and practical reason involves three signifi-

cant moves. The first, again, is that rational will is instrumental and exclusively
consequence directed. The agent’s end is always some (believed) good effect,
that is, some state of the world that is naturally ­producible by some action in
the agent’s power.
Cumberland’s second move is to hold that the value of a consequential state
is itself resolvable into the natural goods (well-being) of affected persons. This
is a move that becomes characteristic of reductive naturalist utilitarianism, so
long as it is supposed, as Cumberland does, that the good of person can be
understood naturalistically.
It is a consequence of these first two claims that when an effect is aimed at
as good, the will’s object is invariably the good of someone. Thus, Cumberland
agrees with Locke that it is impossible for anyone to desire or be motivated
directly by the right or by moral good; all desires are for natural goods, that is,
for the natural states that constitute well-being. Where Cumberland departs
from Locke is in holding that people can directly desire the goods of others no
less than their own.
This is Cumberland’s third move: that the good of others is no less good, and
so no less choiceworthy, than is the agent’s own good. It follows, Cumberland
argues, that since an agent has reason to will an end just insofar as it is a good
effect, and since the good of others is no less good than the agent’s own, the agent
has the same reason to aim at the good of others that they do to aim at their own.
Because we can equally well act for others’ sake as we can for our own, the
moral question arises when people’s goods conflict of whose good it is morally
right to realize.
There is no doubt, but that Man has a natural Power, or Will, which he
himself may determine to act which way he pleases. But when we are
inquiring into the Right of Acting, the Question is, “Which among those
Actions which are in our power, are lawful?” … Any one can either hang,
or throw down a Precipice, either himself, or any other innocent person
(359–360).

If the goodness of effects or ends is the only source of normative reasons, the
fact that an act will be good for the agent cannot be a reason to act for his sake as
opposed to someone else’s, if an alternative would be no less good for someone
else. Cumberland will conclude that the “best end” for practical reason to aim
at must be the greatest good of all.
This line of thought lies behind remarks Cumberland makes about consis-
tency and agreement. We noted earlier Cumberland’s claim that someone who
wills the same for themselves as they do for others has a kind of “inward Peace”
(296). But Cumberland goes farther to say that the costs of willing different
effects for ourselves than for others is not just psychic turmoil but a kind of
deep unreasonableness.

https://doi.org/10.1017/9781139025065.005 Published online by Cambridge University Press

 practical dictates of reason, agreement 107

It is evident, “That no Action relating to others can be consistent with those

necessary and right Actions conducing to our own Good, unless the Practical
Dictates of Reason, by which we are determin’d to that Action, … enjoin us
to desire such things to them as to our-selves” (525).

Cumberland objects on these grounds to the thinking underlying Hobbes’s

right of nature:

The reason of Hobbes’s making so gross a Blunder in this Argument was

because he did not observe, “That there was the same Standard to all, by
which the Reason of every one is to be tried, whether it be right or no” (381).

Of course, in one obvious sense, the right of nature does give all the same stan-
dard. The problem, from Cumberland’s point of view, given his instrumental-
ist conception of rational action, is that any standard of practical reason can be
derived only from the goodness of ends or effects. The requisite rational standards
can only come from an “End necessary to all rational Beings” (381). Cumberland
is clearly ruling out any agent-relative standard that gives each agent a different
target effect or state, and so a different end. What is needed for there to be a com-
mon standard of practical reason is for there to be a common end.
“Reason will not suffer,” Cumberland says, “that the greatest Private Good
should be propos’d as the ultimate End” (529). Were that so, “Actions truly
Good will be in mutual opposition to one another, which is impossible” (529).
When people’s goods conflict, moreover, “there is no Cause, why the Hap-
piness of any one of these [say, the agent themselves] should be his ultimate
End, rather than the Happiness of another should likewise be his ultimate End”
(529). Cumberland therefore concludes that

Reason dictates to neither, that he should propose to himself his own

Happiness only, as his greatest End, but to every one, rather his own in
conjunction with the Happiness of others; and this is that Common Good,
which we contend is to be sought after (529).

This is a line of reasoning that is similar to Sidgwick’s leading to his princi-

ple of Rational Benevolence, which comes from the thought that “the good of
any one individual is of no more importance from … the point of view of the
Universe.” And it is similar also to Moore’s famous “refutation” of egoism in
Principia Ethica (Sidgwick 1967: 381; Moore, 1993: 99–102). Of course, neither
Sidgwick nor Moore are ethical naturalists, but otherwise Cumberland’s idea
is very similar.
For Cumberland, there is one fundamental ethical standard, the good, more
precisely, the goodness of states of the world or “effects.” Whether conduct is
justified, dictated by practical reason, depends on the goodness of its effects,
and others’ goods are no less good than is the agent’s own. It follows that reason
must dictate to every agent that they pursue their own good “in conjunction

https://doi.org/10.1017/9781139025065.005 Published online by Cambridge University Press

 108 locke and cumberland

with the Happiness of others.” “Let Hobbes tell me,” Cumberland writes, “what
the addition of a proper name does, toward making the former Proposition
[that each act only for their own self-preservation] a more evident Dictate of
Reason,” than that each should promote the happiness of all (Cumberland
2005: 532). What Cumberland is objecting to, of course, is not really Hobbes’s
use of proper names; that might be ruled out as inappropriately particularistic
in a universal norm. It is, rather, Hobbes’s use of the reflexive first-person pro-
noun. The ruling out of agent-relatively is a deep feature of a view of practical
reason that takes the goodness (simpliciter) of an effect as what most funda-
mentally underlies there being reason to aim at it.
Cumberland concludes that the “common Good is the best and greatest End,
which [rational beings] can propose to themselves” (269). There are, again, three
steps that lead Cumberland to this conclusion. The first is that the best end is the
best effect, the best state of affairs producible by any act in the agent’s power. The
second is the reductive naturalist move that the goodness of a state is constituted
by the natural goods of individuals. Thus, once he has established the two theses,
Cumberland thinks he has a very short argument (his third step) for the conclu-
sion that practical reason dictates promoting the greatest aggregate good:
“The Happiness of All is greater than the like Happiness of any smaller
Number.” But “that Happiness is greatest, which is greater than any other
assignable.” Nor is it a different Judgment, that by which we affirm, “The
greatest Happiness of all Rational Beings is the greatest or chief End, which
any Rational Agent can pursue” (537).

“The greatest overall happiness,” is, if you like, the naturalistic content of
“best end.” Cumberland says that his “Method” is to show “[h]ow to reduce
whatsoever the Moralists have said concerning the Means of obtaining the
best End into Theorems concerning the Power of human Actions in produc-
ing the Effects propos’d” (259). Once “best end” is interpreted naturalistically
as “greatest overall happiness,” natural philosophy suffices to establish truths
concerning what will best promote it. Importantly, Cumberland notes that the
“happiness of all” provides a common public standard of action to which every
moral agent has access. He explicitly contrasts this standard with Aristotelian
practical wisdom, which only the virtuous can achieve (275).23

Obligation, Reconsidered
If any ideas are likely to prove resistant to Cumberland’s program of natural-
ist reduction, they are the deontic ideas of authority, right, and obligation.
How are we to understand the law of nature as genuinely obligating law?

23
On this point, see Heydt 2014: 19.

https://doi.org/10.1017/9781139025065.005 Published online by Cambridge University Press

 obligation, reconsidered 109

Cumberland is formally a theological voluntarist, holding that anything with

the form of law requires both competent promulgation and authoritative sanc-
tions. These aspects of his view, however, appear ­ultimately to be removable
formal scaffolding.
This is revealed in his view of the source of God’s authority or right of
rule. Cumberland says that he used, “as most others do, to deduce the Divine
Dominion intirely from his being the Creator” (672). This, of course, is Locke’s
view, as it is in essentials, Pufendorf’s also. But Cumberland tells us that he
came to realize that “all Right is a Power granted or permitted by some Law …
But Law there is none prior to the Natural Law, of that Dictate of the Divine
Wisdom, concerning the Best End, and the Means thereto necessary” (672). The
ground for God’s right that we obey the law of nature is simply that it is the
law of nature, that it indicates the “Best End, and the Means thereto necessary.”
Indeed, Cumberland holds that we can judge the truth and bindingness of the
law of nature for ourselves: “God will determine the same End and Means to be
best, which the Reason of any Man truly judges to be so” (537). Once we have
determined that the best end is the good of all, and that benevolence is the best
means to our necessary end,
there is no room to doubt, but that we shall here also have God’s concur-
rence. For, since He himself is Rational, and it cannot be conceiv’d, how
he can act rationally, without proposing an End to himself, nor can there
be a greater End than the aforesaid Aggregate of all Good Things; we cannot
but think, he judges this to be the best End that he can propose to himself
(537–538).
Cumberland goes so far, indeed, as to say that we “believe the sacred Scriptures
to be the Word of God, the Author of Nature, because they every where illus-
trated, confirm, and promote the Law of Nature” (281).
In the end, Cumberland’s position has the effect of making not just God’s
authority epiphenomenal, but the very ideas of right, obligation, and authority
epiphenomenal also. In his view, there simply is no genuine normative stan-
dard or “dictate of practical reason” that is independent of the good. “There is
nothing that can superinduce a Necessity of doing or forbearing anything, upon
a Human Mind deliberating upon a thing future, except Thoughts or Proposi-
tions promising Good or Evil, to ­ourselves or others, consequent upon what we
are about to do” (554).
Ultimately, therefore, there are three reductionist programs represented
in Cumberland. We have mentioned two before: reductive naturalism and
the reduction of moral norms or laws of nature to Cumberland’s one utilitar-
ian formulation. But there is also a third, namely, Cumberland’s project of
cloaking his thought in the deontic terms of the right that are characteristic
of modern moral philosophy while nonetheless maintaining that these terms
can ultimately be explained away so that ultimately there is only one final

https://doi.org/10.1017/9781139025065.005 Published online by Cambridge University Press

 110 locke and cumberland

normative standard: the good, which can be given the naturalistic interpreta-
tion he proposes. Later moral philosophers took up all three of these projects.
In the next chapter, we turn to the thought of thinkers who, whether they were
prepared to think deontic thoughts or not, nonetheless were all profoundly
critical of the early modern natural law tradition we have been investigating
to this point.

https://doi.org/10.1017/9781139025065.005 Published online by Cambridge University Press

 4

Spinoza, Cudworth, Shaftesbury, and Leibniz

All the philosophers we have considered to this point have been representa-
tives of the early modern natural law tradition. The seventeenth-­century natu-
ral lawyers conceived of morality in fundamentally deontic or juridical terms,
and they attempted to explain morality’s power to obligate in ways that might
link satisfactorily with complementary theories of accountable moral agency
and practical reason. Although there were non-voluntarist versions – Grotius’s
and, in some ways, Cumberland’s – the most influential forms were voluntarist,
mostly theological voluntarisms of the sort defended by Locke and Pufendorf.
In this chapter, we consider a group of philosophers who formed their
views in opposition to voluntarist natural law. Although there are important
differences between Baruch (Bento or Benedict) Spinoza (1632–1677), Ralph
Cudworth (1617–1688), Anthony Ashley Cooper (the 3rd Earl of Shaftesbury
(1671–1713)), and Gottfried Wilhelm Leibniz (1646–1714), all four were united
by their opposition to voluntarism.1 For voluntarists, even natural law is a kind
of positive law since it is created by God’s discretionary imposition. Against
this, Cudworth complained that voluntarism makes morality “positive, arbi-
trary and factitious only” and unable to explain morality’s “eternal and immu-
table” character (Cudworth 1996: I.i.1). Fundamental moral truths such as
the wrongness of gratuitously harming an innocent do not simply happen to
obtain; they hold independently of the contingencies of any possible world.
Neither can morality depend on discretionary legislation, even God’s. What
makes such an act wrong is the kind of act it is; an act cannot be made wrong
“by will,” but only by “its own nature” (I.ii.3).
Cudworth also complained about the way natural law theories tend to focus
on external conduct rather than on internal motive and character. He contrasts
a “dead Law of outward Works” with the “inward Soul and Principle of Divine

1
All but Spinoza were united also by their personal and intellectual connections to Ralph
Cudworth’s daughter, Damaris Cudworth Masham (1659–1708). As noted in the last chap-
ter, it was likely because of Masham that Locke learned of Cudworth’s views on autonomy
and freedom of the will. Locke was also a close mentor of Shaftesbury and introduced him
to Masham. Finally, Masham also had a philosophical correspondence with Leibniz. See
Broad 2006 and Hutton 1993.

111

https://doi.org/10.1017/9781139025065.006 Published online by Cambridge University Press

 112 spinoza, cudworth, shaftesbury, and leibniz

Life … that enliveneth and quickeneth … all our outward Performances” (1969:
123). Although Cudworth is anxious also to account for morality’s power to
obligate, and to provide an account of self-determining moral agency he argues
to be necessary for moral accountability and obligation, he nonetheless under-
stands morality as most fundamentally involving the virtuous human capac-
ity to manifest and share God’s love (the “Principle of Divine Life”). Indeed,
Cudworth ties morality and love together. The moral law is ultimately a “law of
love” (124). And the “moral free will” that Cudworth holds to be necessary for
accountable moral agency essentially includes the capacity for love.
Recall Locke’s view of self-determination as the capacity to step back from
desires, form a judgment of what will bring the agent the greatest long-run
pleasure, and then act on that considered judgment. Like Locke, Cudworth
holds that free will involves self-conscious practical judgment.2 But the Lock-
ean form that judges “private utility” Cudworth calls “animal free will,” and
Cudworth holds that this is insufficient for accountable moral agency (~1670:
4980.171, 4982.89). “Moral free will” requires, “besides the spring of animal
life,” that we also have a “spring or principle of another superior intellectual
life,” namely, the “love or charity” that is the “life and soul of all morality”
(~1670: 4982.20; 1678: I.iv.9).
In turning from natural law’s focus on conduct to motive, character, and
virtue, Cudworth provided an alternative to the law conception that resonated
with Shaftesbury and Leibniz. Like Cudworth, Shaftesbury and Leibniz argue
that theological voluntarism is powerless to explain God’s own goodness or
justice as deriving from God’s commands (Shaftesbury 2001: I.iii.2.49; Leib-
niz 1988: 64–75). And Leibniz echoes Cudworth’s complaint that voluntarism
could not explain morality’s “eternal truths” (Leibniz 1988: 71).
For Shaftesbury and Leibniz, Cudworth’s most important move, though,
was his change in focus from act to motive, and specifically, to love. Shaftesbury
writes in his Preface to the Sermons of Cudworth’s fellow Cambridge Platonist
Benjamin Whichcote that “it is unaccountable that men who profess a reli-
gion where love is chiefly enjoined, where the heart is expressly called for, and
outward action without that is disregarded … should combine to degrade the
principle of good nature, and refer all to reward” (Shaftesbury 1698).
Whereas Leibniz’s anti-positivism was provoked by Pufendorf, Shaftesbury’s
was prompted by Locke. Shaftesbury objects against Locke that by arguing that
human moral conduct is possible only through divine sanctions, Locke “struck
at all fundamentals, threw all order and virtue out of the world” (1900: 403).
Similarly, Leibniz complains against Pufendorf that a morally good person is

2
Indeed, as I suggested in the last chapter, there is reason to think that Locke was influenced
to adopt this view by his exposure to Cudworth’s ideas, either directly by Cudworth’s man-
uscripts on freedom of the will or by conversation with Damaris Cudworth Masham.

https://doi.org/10.1017/9781139025065.006 Published online by Cambridge University Press

 spinoza, cudworth, shaftesbury, and leibniz 113

not one who simply performs certain “external acts.” He does what is good and
right, “not out of hope or fear” of sanctions, but from a core “inclination of his
soul,” “love for God or … neighbor” (Leibniz 1988: 72, 73).
Even more than Cudworth, Shaftesbury puts forward an ethics of virtue in
opposition to an ethics of law, one that provides a model for the later virtue eth-
ics of Hutcheson and Hume. And Shaftesbury’s thought inspired Hume’s and
Hutcheson’s in yet a further way. Hume is famous for his doctrine that “moral
distinctions [are] not derived from reason,” but “from a moral sense” (Hume
1978: 455, 470). The phrase “moral sense” comes into Hume’s thought most
directly from Hutcheson, who made his theory of the moral sense the foun-
dation of his moral philosophy. However, Hutcheson took the phrase “moral
sense” from Shaftesbury (Shaftesbury 2001: II.46).
In Hutcheson and Hume’s hands, the idea of a moral sense or sentiment
is deployed in opposition to eighteenth-century forms of ethical rationalism,
represented in the English context by contemporaries like Samuel Clarke, John
Balguy, William Wollaston, and, later, Richard Price. But the debates between
rationalism and sentimentalism that were a staple of eighteenth-century moral
philosophy were much more sharply focused than anything we can find in the
earlier philosophers investigated in this chapter. Cudworth, Shaftesbury, and
Leibniz can fruitfully be seen as important bridge figures between early mod-
ern natural law and the ­eighteenth-century rationalists and sentimentalists.
The latter’s sharply drawn distinctions were much more muted in the earlier
writers.
Cudworth was an important influence on Richard Price. But Price and the
other British rationalists were more clearly rational intuitionists, and their eth-
ical theorizing took a more deontological form. Where Cudworth identified
moral reason with an essentially practical form of love, the eighteenth-century
rationalists tended to see it as an intuitive perception or cognition of deontic
moral truths or facts. Their sentimentalist opponents found Cudworth’s virtue-
centered approach much more to their liking than the deontological theories of
the eighteenth-century British rationalists.
Similarly, when Shaftesbury talked of a “moral sense,” he did not mean to
distinguish it from reason, as his followers Hutcheson and Hume would. To
the contrary, Shaftesbury’s thought was much closer to that of Cambridge Pla-
tonists like Cudworth and Whichcote, who tended to blur lines between the
rational and the affective. Like the Cambridge Platonists, Shaftesbury argues
that moral goodness is ultimately grounded in creative and practical powers
that are intrinsic to the mind. Here again, we find thinkers we investigate in
this chapter functioning as bridge figures to later writers who develop one or
another aspect of their thought in a more single-minded way.
For his part, Spinoza is also a bridge figure, but of an importantly differ-
ent kind. In many ways, Spinoza stands apart from the main trends of moral

https://doi.org/10.1017/9781139025065.006 Published online by Cambridge University Press

 114 spinoza, cudworth, shaftesbury, and leibniz

philosophy in the seventeenth and eighteenth centuries. He does not so much

bridge developments within early modern ethics as form a bridge from modern
ethical thought back to premodern Aristotelian naturalist forms and forward
to later ethical philosophers like Nietzsche. Not that Spinoza is an Aristote-
lian. Like Hobbes, Spinoza rejects final causes and a human telos. Also, freedom
plays a central role in his thought that it did not for the ancients. Spinoza is a
eudaimonist, however, indeed a perfectionist eudaimonist, like Aristotle. His is
an ethics of virtue that ultimately identifies all of the following: knowledge, vir-
tue, right, and well-being. Also like Aristotelian naturalists, Spinoza claims that
virtue promotes the common good, thereby unifying personal and impartial
good, and rejecting the modern division between them (Spinoza 1985a: 128).3
Spinoza is a unifier par excellence. He is, first and foremost, a systematic
metaphysician who presents his views in more geometrico. And his metaphys-
ics does not simply ground his ethics. Since, for Spinoza, all ethical concepts are
ultimately metaphysical, every ethical claim he makes is really a metaphysical
claim. Spinoza’s “ethics” consists of metaphysical propositions expressed with
ethical language. And his metaphysics is obviously unified also; it is a form
of metaphysical monism. Spinoza believes that there is only one substance –
Nature with its necessary laws – and that all other metaphysical categories – for
example, properties and what we take to be the (Aristotelian) substances that
instantiate them – are either attributes of Nature or modes of those attributes.
The most obvious contrast is with Descartes’s dualism. For Spinoza, con-
sciousness and extension are not properties of distinct substances, mind and
matter, but attributes of one and the same substance (Nature). And individual
thinking beings and extended bodies are modes of these different attributes,
respectively. Natural necessities and causal relations obtain within attributes
but not across them. Thus, causal relations are impossible between mental and
bodily phenomena. Bodies and minds are modes of fundamentally distinct
attributes of the same substance.
Spinoza titled his great metaphysical work Ethics, but this is something of a
misnomer, since relatively little of it is focused on traditional ethical questions,
not to mention issues of deontic morality. Spinoza recognizes, as we shall see,
no concept of moral obligation or right that is distinct from that of the good.
And his theorizing about the good life (alternatively, “virtue” or “perfection”)
is, again, reducible to propositions of metaphysics.
This is perhaps clearest in Spinoza’s account of perfection. Typically, perfec-
tionists like Aristotle or Aquinas advance their views in irreducibly teleological
terms that set ideals of perfection for different natural (teleological) kinds. But
Spinoza thinks this involves a fundamental metaphysical error, since “Nature

3
For the role of the common good in Aristotelian naturalists, see Irwin 2007: 615–619; 2008:
66–67.

https://doi.org/10.1017/9781139025065.006 Published online by Cambridge University Press

 spinoza, cudworth, shaftesbury, and leibniz 115

does nothing on account of an end” (1985: 544). Spinoza does think that every
natural thing has a directed nature, what he calls a conatus or “striving.” But
this just means that everything has a nature that is expressible in natural laws of
efficient causation, rather than involving anything like final causation in Aris-
totle’s sense.
For Spinoza, there are (naturally necessary) truths about how every natural
thing will tend to behave if it is not perturbed by external causes acting on it.
What it is for something to be perfect in the sense in which Spinoza is con-
cerned in his ethical writings, then, is simply for the process of its striving in
this efficient causal sense to be unimpeded. This makes perfection, as Curley
notes, “a nonevaluative, metaphysical” notion (Spinoza 1985b: 650). “By per-
fection,” Spinoza says, “I understand only reality,” hence nothing about how
reality is not and ought to be (252).
Thus not only, as we shall see, does Spinoza not recognize any notion of right
that is distinct from that of good, but so also are all ethical (or other normative)
notions ultimately metaphysical for him. Far from facing the problem of how
to account for distinct normativities of right and good, therefore, Spinoza side-
steps the whole problem. Nothing in his thought corresponds to normativity as
ethical philosophers normally conceive it. There are no oughts, only is’s.
Spinoza is especially skeptical of morality as it is conceived by modern moral
philosophers. He rejects deontic moral notions, the psychological attitudes he
takes to be connected to them (like indignation and blame), and the presuppo-
sitions of free will they require. Like theological voluntarists, he takes these all
to be most at home when tied to a doctrine of eternal sanctions. But he rejects
the personal God that doctrine requires. Moreover, differently from Pufendorf,
he takes the requisite sanctions not to be tied to a sense of their justifiability that
can motivate, but only to motivation through fear. Far from realizing moral
virtue, conduct that is motivated in this way is, by definition, imperfect, unfree,
and, strictly speaking, as we shall see, a matter of “passion” rather than our own
“action” (493, 497).
Like Nietzsche, who acknowledges Spinoza’s influence in this respect, Spi-
noza both decisively rejects the doctrine of free will necessary for deontic
morality and diagnoses the latter as unhealthy, what Nietzsche will call an
“illness” (Nietzsche 2007: 7). We will see resonances of this latter idea also in
Shaftesbury and Hume. Arguably, Spinoza begins a kind of naturalistic critique
of deontic morality that runs through Hume, Nietzsche, and into the twenti-
eth century with Foot, Williams, and Annette Baier (Foot 2002; Williams 1985;
Baier 1993). Although in many ways he is a bridge to premodern ethics, in oth-
ers, he can seem the most modern thinker discussed in this book.
Perhaps Spinoza’s greatest influence outside his metaphysics was his mas-
sive effect on eighteenth-century political philosophy. As Jonathan Israel has
shown, Spinoza provided much of the inspiration for a democratic, egalitarian,

https://doi.org/10.1017/9781139025065.006 Published online by Cambridge University Press

 116 spinoza, cudworth, shaftesbury, and leibniz

libertarian movement of political thought that Israel calls the “Radical Enlight-
enment” (Israel 2001). The central work here is Spinoza’s Theological-Political
Treatise, which sets out Spinoza’s democratic views. As we are interested in
Spinoza’s ethics, we will consider his democratic politics less for their own sake
than for their grounding in Spinoza’s ethical and metaphysical views.

Spinoza
What seemed most radical to Spinoza’s contemporaries was his atheism. Spi-
noza rejected this characterization himself. Indeed, on Spinoza’s definition – “a
substance consisting of an infinity of attributes” – God is foundational for his
metaphysics (Spinoza 1985b: 409). “God” is Spinoza’s term for the one neces-
sarily existing substance, which Spinoza also calls “Nature,” and whose nec-
essary essence determines everything, from the motion of the tiniest physical
particle to the most brilliant cogitations and courageous actions of human
beings (549).
Spinoza thus denied the orthodoxy of Christian Europe and of his own Jew-
ish immigrant community in Amsterdam. That this made Spinoza an outsider
is an understatement, as Spinoza was ultimately excluded even from his own
community.4 Spinoza thus became an outsider even to a group that were peren-
nially excluded themselves. And the situation was more complicated still. The
Portuguese Jews of Amsterdam had suffered through the Inquisition in Portu-
gal, surviving largely as conversos, or converts to Christianity, at least publicly.
Those, like Spinoza’s family, who wished to continue to live as Jews, could do
so only secretly until they were forced to emigrate from Portugal. This required
them to manage two identities simultaneously – a public Christian persona and
a private Jewish one. Yovel plausibly claims that this cultural experience carried
seeds of the liberalism we find in Spinoza’s thought (Yovel 1992). Liberal sepa-
ration of church and state became a central tenet of the Radical Enlightenment
and the political philosophy that followed in its wake, culminating, perhaps, in
Rawls’s Political Liberalism at the end of the ­twentieth century (Rawls 1993).

Against Deontic Morality: Religious or Secular

Despite these liberal aspects, there were elements of Spinoza’s thought that cut
against the main lines of the modern conception of morality that was so impor-
tant for the Radical Enlightenment. There can be natural rights, after all, only if
there are natural duties, and for most philosophers of the early modern period,
duties are conceptually connected to accountability and, for many, to divine
punishment. On the orthodox view, God punishes wrongdoers who (freely)

4
For an excellent account of this and Spinoza’s life more generally, see Nadler 2018.

https://doi.org/10.1017/9781139025065.006 Published online by Cambridge University Press

 against deontic morality: religious or secular 117

violate his law. For theological voluntarists like Pufendorf and Locke, the very
existence of morality depends on God’s superior authority to hold human
beings accountable. Spinoza denies that any such personal God can possibly
exist, however. Where orthodoxy defines God as “omniscient, omnipotent, …
[and] infinite[ly] compassion[ate]” (Spinoza 1985a: 88), Spinoza claims that
none of these properties can hold of a necessarily existing substance:
[T]hese things [properties] are only certain modes of the thinking thing …
[and] can neither be nor be understood without that substance of which
they are modes. That is why they cannot be attributed to him, who is a
being existing of himself, without anything else (89).
Spinoza’s “him” and “himself” retain the cover of his claim to theism, but his
argument ensures that anyone convinced by it can no longer be an orthodox
Christian or Jew.
Thought and extension, for Spinoza, are attributes that necessarily pre-
suppose a substance, Nature, to which they apply. Individual thinking and
extended things are nothing but modes of these attributes, as are all mental and
bodily phenomena. It follows that any particular action or willing that ortho-
doxy supposes to be instantiated by God and human beings, respectively, can-
not have the properties orthodoxy attributes to them, including the authority
relations that voluntarists like Locke and Pufendorf hold to be necessary for
the existence of deontic morality. Even non-voluntarists like Cudworth and
Leibniz must therefore be mistaken in thinking that God could be a perfectly
loving mind who wishes his creatures well and, therefore, that they act in ways
that will promote their good.
A second blow against orthodox doctrine was what Spinoza’s critics called
his “fatalism.” All natural events are necessarily determined by earlier ones in
accordance with natural law, including human actions that modern moralists
take to result from free will. Against this, Spinoza’s correspondent, Henry Old-
enburg, complained that it makes deontic morality’s connection to account-
ability impossible to justify.
You appear to postulate a fatalistic necessity in all things and actions. If
this is conceded and affirmed, they say, the sinews of all laws, all virtue
and religion are severed, and all rewards and punishments are pointless. …
[N]o one will thus be without excuse in the sight of God (quoted in Nadler
2006: 84).

There is yet a third respect in which Spinoza’s Ethics threatened modern

orthodoxy, both religious and secular versions. It is a central plank of mod-
ern thought that moral conduct expresses our nature as free moral agents.
On Spinoza’s metaphysics and moral psychology, however, the only motive
that sanctions can create is fear, whether the sanctions are divine or secular.
And any conduct that is motivated by fear, according to Spinoza, cannot be

https://doi.org/10.1017/9781139025065.006 Published online by Cambridge University Press

 118 spinoza, cudworth, shaftesbury, and leibniz

self-determined or even an instance of willed “action,” properly speaking:

something we really do. As we shall see, Spinoza holds it to be an instance of
“passion” in which we are “acted upon” by things external to us. Action takes
place only “when something happens, in us or outside us, of which we are the
adequate cause” (Spinoza 1985b: 493).
On the mainstream view, when we avoid wrongdoing because it would be
wrong, we exercise our capacity for autonomous moral agency. Spinoza turns
this view on its head. He thinks we are most unfree, and do not strictly speaking
even act at all, when we are motivated by accountability concerns, since these
are realized in our psychic economy only through fear, a passion that puts us
in “bondage” rather than enabling free action (505, 548). Unlike Pufendorf and
Cudworth, Spinoza lacks anything like conscience in his moral psychology.
Other passions that Spinoza takes to be involved in deontic morality are even
more toxic: blame, indignation, and – he holds – hate, anticipating Nietzsche.
“Indignation is Hate toward someone who has done evil to another,” and
“[h]ate can never be good” (535, 571). Indignation, blame, hate, and the like are
all negative emotions and can have no intrinsic value.
We feel these negative emotions, moreover, only through our own igno-
rance. When we fully understand someone’s behavior – someone else’s or
our own – we cannot really be angry with them, since their behavior is sim-
ply the necessary result of the one necessarily existing substance; indeed, their
behavior is the causal consequence of factors outside of them rather than their
own actions. True knowledge yields “true love of one’s fellow man, which …
disposes us so that we never hate him, or are angry with him, but are instead
inclined to help him and bring him into a better condition” (1985a: 128).
“Remorse and Repentance,” Spinoza says, “occur only by surprise” (114). It
is impossible for us to grasp that something is good and not do it. If it turns
out not to be good, its lack of goodness must take us “by surprise.” Moreover,
there is nothing good in these feelings themselves. “[O]n the contrary,” they
are “injurious [in themselves], and consequently evil … and to be shunned
and fled” (115). Here again, Spinoza sounds themes that will be developed by
Nietzsche. Indeed, many elements of Spinoza’s ethics – including his doctrine
of conatus and power, his critique of deontic morality and freedom of the will,
and his emphasis on value as positive and bad as lack or privation (“perfection”
as “reality”) – find later expression in Nietzsche. When Nietzsche first discov-
ered Spinoza’s writings, he wrote to a friend, “I have a forerunner. And what a
forerunner” (quoted in Young 2010: 319).
There are, however, significant differences between Spinoza and Nietzsche.
One of the most striking is what we might call the “flatness” of Spinoza’s moral
psychology in contrast with the depth of Nietzsche’s. There is nothing in Spi-
noza like ressentiment or Nietzsche’s “dark workshop” of the unconscious. And
complex and turbulent emotions like hatred and jealousy, which Nietzsche

https://doi.org/10.1017/9781139025065.006 Published online by Cambridge University Press

 the good: knowledge, power, and perfection 119

reveals to be tortuously and torturously targeted in the human psyche, are sig-
nificantly simplified and compressed in Spinoza’s psychology. “Blame,” for
example, is “the Sadness with which we are averse to [someone’s] action” (Spi-
noza 1985b: 510). “Hate” is “Sadness, accompanied also by the idea of an exter-
nal cause” (511), and “Jealousy … is … nothing but a vacillation of mind born
of Love and Hatred together, accompanied by the idea of another who is envied”
(514). “Sadness,” moreover, is simply Spinoza’s term for the most general nega-
tive response to the transition from a more to a less perfect state of oneself (500).
All things considered, one feels it likelier that Nietzsche does justice to the
subtlety, complexity, and depths of the human psyche than that Spinoza does.

The Good: Knowledge, Power, and Perfection

We turn now to Spinoza on the good. In the end, something’s being good and
its being determined by the laws of its nature amount to the same thing for
Spinoza. “[N]othing happens in nature which can be attributed to any defect in
it” (492). Every occurrence is determined by Substance’s necessary nature, and
every natural thing’s conatus determines it to remain in existence as the natural
thing it is. When something is realized in this natural-law-determined sense,
it is thereby “perfected,” independently of any telos or end. This, however, can
be impeded by other things and their contrary strivings; and when that occurs,
there is necessarily imperfection or bad, which is simply the lack of perfection
or good. There is no such thing as intrinsic badness, for Spinoza, just the lack of
intrinsic goodness.
When bad occurs for some natural thing, that means that some other thing
or things impede its perfection but thereby perfect themselves. “[G]ood and
evil are nothing but relations” (1985a: 92).
[G]ood and evil, say Peter’s goodness and Judas’s evil, have no defini-
tions apart from the [particular] essence[s] of Judas and Peter, for these
[essences] alone [a]re in Nature, and without them [the goodness of Peter
and the evil of Judas] cannot be defined (93).5

When bad for one thing occurs, good must occur for others. And the good of
the system of Nature as a whole is always necessarily continuously realized,
since that just consists in everything’s being determined by its necessary nature.
Spinoza realizes, like Hobbes, however, that even if “good and evil are not
things … which are in Nature,” we speak and think as though they are. And
he has, also like Hobbes, an error theory that explains why we do. Spinoza and
Hobbes agree that the language of good and evil (bad), and thoughts we for-
mulate with it, are expressions of practical states of mind. Hobbes holds, recall,

5
See Curley’s note at 1985a: 93n.

https://doi.org/10.1017/9781139025065.006 Published online by Cambridge University Press

 120 spinoza, cudworth, shaftesbury, and leibniz

that whatever we desire, we call good (Hobbes 1994a: VI.7). Similarly, Spinoza
says that “each one, from his own affect, judges, or evaluates, what is good and
what is bad” (Spinoza 1985b: 516). And even more emphatically:
[W]e neither strive for, nor will, neither want, nor desire anything because
we judge it to be good; on the contrary, we judge something to be good
because we strive for it, will it, want it, or desire it (500).

Hobbes’s projectivist account of value judgment is more detailed (and explic-

itly modeled on his Galilean account of color) than Spinoza’s is; but in essen-
tials, they are the same. We speak and think with the concepts of good and evil
when we have desires and aversions. These practical states of mind include see-
ing their objects as good and bad themselves.
Something that is distinctive in Spinoza, and absent from Hobbes, however,
is an account of objective good that is projected by will rather than by desire.
We will consider this distinction in more detail below when we examine Spi-
noza’s views on the kind of freedom the will realizes in contrast with the bond-
age of desire and appetite, and will’s relation to knowledge and ignorance. The
relevant point here is that although for Hobbes, will is simply the “last appe-
tite” before action, Spinoza distinguishes between appetites and desires, and
the behavior they cause, on the one hand, and will and action, in his distinc-
tive senses, on the other. And he makes a related distinction between appar-
ent good and objective good.6 Spinoza’s distinction between will and desire is
based on a distinction between practical states that are grounded in knowl-
edge and understanding and those that are not. Spinoza defines “Will” as a
thing’s “striving” when it is “related only to the Mind” or “Intellect” and is
based on “adequate” rather than “inadequate” or “confused” ideas of its object
(498–500).
Guided only by the Intellect, the Will is informed by “clear and distinct
ideas,” whereas desires and appetites are based on “confused ideas” (499). Will
and genuine action are informed by the Intellect and its clear view of Nature.
Unlike Descartes, from whom the language of “clear and distinct” obviously
derives, and who distinguishes between will and intellect, Spinoza holds that
no such distinction exists (Della Rocca 2008: 23, 134). The Mind or Intellect
has its own striving, and that is what Spinoza means by “Will.” There is no real
distinction between Will and Intellect for Spinoza.
This makes possible a companion account of real or objective good in
contrast with the apparent good projected by desires and appetites. Objec-
tive good is what we see the objects of will as having when we have adequate
ideas of their real (non-ethical) nature, and (merely) apparent good is what
we see the objects of desire as having when we have confused ideas of their

6
These are my terms, not Spinoza’s.

https://doi.org/10.1017/9781139025065.006 Published online by Cambridge University Press

 the good: knowledge, power, and perfection 121

nature.7 This account is still projectivist; it denies that there are any value prop-
erties (that are not reducible to natural non-­evaluative properties) in nature.
And it remains relational: objective goodness is always in relation to individual
beings. “[G]ood and evil” are thus “nothing but relations” (Spinoza 1985a: 92).
Objective goodness is the property that the will projects onto its objects, where
will is the practical attitude that is based on an adequate intellectual grasp of its
objects’ real (non-ethical) nature. Obviously, this latter grasp cannot involve
projection itself.
But although objective good is relational, Spinoza holds that objective good
is universal with respect to the beings who share a nature. We humans all indi-
vidually will the completion or perfection of human nature in ourselves.
I shall understand by good what we know certainly is a means by which we
may approach nearer and nearer to the model of human nature that we set
before ourselves. By evil, what we certainly know prevents us from becom-
ing like that model (1985b).

By “model of human nature,” Spinoza cannot mean some ideal at which we

ought to aim in an irreducibly normative sense. There are, again, no oughts that
are not reducible to is’s. A “model of human nature” is simply a fact about what
one’s intellectual striving would lead to were it not impeded by the strivings of
others, where that includes not just other human beings, but anything exter-
nal to us whatsoever. Spinoza identifies human perfection, so understood, with
“virtue” and “power,” which he defines as perfection; “insofar as we attribute
something to them … that involves … lack of power,” “we call them imper-
fect” (545). As Nadler notes, “we can set up the following equation for Spinoza:
virtue=knowledge=activity=freedom=power=perfection” (Nadler 2006: 256).
Here again are ideas from which Nietzsche will obviously draw. Every-
thing, Spinoza thinks, has a conatus for power that Nietzsche will call the “will
to power.” For Spinoza, this only counts as “will,” properly so called, when
it is the conatus of Mind or Intellect, which necessarily realizes freedom and
overcomes the bondage of appetites and non-­intellectual affections. But this
also is a doctrine with which Nietzsche will largely agree; it is only when we
grasp the way nature actually works, and the genealogy of our own ideas and
desires, Nietzsche holds, that we can throw off the superstitious, shackling, and
unhealthy modern notion of morality and become genuinely free and autono-
mous. All of this – together with the idea that good is essentially positive and
that bad is merely its lack – Nietzsche gets from Spinoza.
Even though objective good involves a “model of human nature,” it remains
an agent-relative, eudaimonist concept. Spinoza has no notion of agent-neutral

7
Cf. the discussion of Hutcheson’s distinction between calm desires and passions and his
related accounts of the good in Chapter 5.

https://doi.org/10.1017/9781139025065.006 Published online by Cambridge University Press

 122 spinoza, cudworth, shaftesbury, and leibniz

or of impartial good – what Hutcheson will term “public good,” what

Sidgwick will call “good from the point of view of the universe,” or what
Moore will dub, “intrinsic value” (Hutcheson 2004: 102; Sidgwick 1967: 381;
Moore 1993). It is consistent with objective good in Spinoza’s eudaimonist
sense that our objective goods might be irremediably at odds. If human
nature were essentially competitive, then it would simply be impossible for
human virtue, power, and perfection to be realized jointly. If any were to
realize the perfection and power toward which their conatus tends, it would
be at the cost of others not realizing their objective goods.
Spinoza’s solution to this problem is like that of Aristotelian naturalists.
Human nature is essentially non-competitive, and much of what is good for
human beings is good for us in common. Spinoza accepts Aristotle’s character-
ization of human beings as social animals (Spinoza 1985b: 564). No one seems
more excellent to us, Spinoza says, “than those that agree entirely with our
nature.” When “two individuals” work together, they “compose an individual
twice as powerful as each one.” Therefore, “to man, then, there is nothing more
useful than man” (556).
Man … can wish for nothing more helpful to the preservation of his being
than that all should so agree in all things that the Minds and Bodies of all
would compose, as it were, one Mind and one Body; that all should strive
together, as far as they can, to preserve their being; and that all, together,
should seek for themselves the common advantage of all (556).

Given our common intellectual nature, each of us will realize our own virtue,
perfection, good, and power only if we can do so together.

Revised Deontic Moral Concepts and Democratic Politics

Presently, we will consider how Spinoza’s idea of the common good underlies
his democratic political philosophy. First, however, we should see how Spinoza
provides a revisionary account of deontic moral concepts, one that is not unlike
similarly subversive accounts in Shaftesbury, Leibniz, Hutcheson, and Hume.
Again, as Israel emphasizes, Spinoza’s democratic politics were especially
important for the Radical Enlightenment, and, as we noted, deontic morality
was an essential companion to Enlightenment ideas of natural rights. For Spi-
noza’s views to be maximally helpful for the Radical Enlightenment, therefore,
it was essential that Spinoza have non-deontic replacements that could do the
conceptual work of natural duties and rights within his non-deontic ethical
framework.
Because Spinoza is a thoroughgoing monist, he cannot accept that good and
right are fundamentally distinct ethical notions, a distinction Sidgwick claimed
as a mark of modern moral thought. For Spinoza, there can be only one ethical

https://doi.org/10.1017/9781139025065.006 Published online by Cambridge University Press

 revised deontic moral concepts and democratic politics 123

notion, the good, understood eudaimonistically. And even that must be identi-
fied with the metaphysical fact of a thing realizing its natural tendencies, unim-
peded by others’ natural strivings. What Spinoza calls “ethics” is ultimately a
non-normative subject, part of metaphysics.
What is at issue here, however, is the connection between the good, as Spi-
noza theorizes it, and the concept of right or moral duty, both as it functions
within moral and political philosophy and as Spinoza can understand it within
his framework. Here is a helpful passage for seeing how Spinoza attempts to do
this:
[H]uman power is very limited and infinitely surpassed by the power of
external causes. So we do not have an absolute power to adapt things out-
side us to our use. Nevertheless, we shall bear calmly those things which
happen to us contrary to what the principle of our advantage demands,
if we are conscious that we have done our duty, that the power we have
could not have extended itself to the point where we could have avoided
those things, and that we are a part of the whole of nature, whose order we
follow (594).

Note, first, Spinoza’s language of “demands” and “duty,” and his saying that the
demand in question comes from “our advantage,” as is required if moral duty is
ultimately to be understood eudaimonistically. Next note a distinction that Spi-
noza implies that is similar to a familiar distinction within moral philosophy
between subjective and objective duty or rightness.
What is objectively right is whatever would realize our greatest good if our
actions were not impeded. Since, however, we are only “a part of the whole of
nature,” others may frustrate our best efforts (as might also imperfect informa-
tion and perhaps yet other things). So we should “bear calmly” when things
don’t proceed as planned. When we do our best, given the obstacles we could
reckon on, Spinoza says, “we have done our duty.” Of course, we will not nec-
essarily have done what was objectively right in the sense of what would have
brought about our greatest good absent these obstacles. So Spinoza must be
here intending “duty” in a different, more subjective sense.
However this might be worked out in detail, several things emerge from
this passage. Most obviously, Spinoza understands (objective) duty in terms of
actions that are necessary to bring about the perfection of the agent’s nature,
where this is simply what the agent themselves would bring about if their
efforts were not impeded by “external causes.” This is “what the principle of our
advantage demands” or, as he says elsewhere, “what reason prescribes to us”
(555, 594). Moreover, such action is naturally (hence, for Spinoza, rationally)
necessary – the only non-normative sense of “necessity” that Spinoza recog-
nizes – in two distinct ways. It is, first, causally necessary in order to achieve the
agent’s perfection; it is a necessary means. And second, it is naturally necessary
that a perfect agent would take the requisite action and that it would achieve the

https://doi.org/10.1017/9781139025065.006 Published online by Cambridge University Press

 124 spinoza, cudworth, shaftesbury, and leibniz

agent’s ­perfection, were there no impediments to its doing so. That is a natu-
rally ­necessary truth.
In this way, Spinoza achieves three conceptual reductions that reform the
orthodox modern concept of moral duty or obligation. First, he reduces the
idea of instrumental rationality to that of causally necessary means and natu-
rally necessary action. Second, he reduces the idea of moral duty and obligation
to that of an act’s being instrumentally required to achieve the agent’s good.
And third, he reduces the (identical) notions of moral duty, virtue, perfection,
and the agent’s good to the aforementioned natural necessities. The right is
reduced to the good and the good is reduced to the naturally necessary. There
are no unreduced oughts, including the moral ought.
These identities are reflected in the following passage:
For a perfect man is moved to help his fellow man only by necessity with-
out any other cause. And therefore he finds himself all the more obliged to
help … [those who] have greater need … (1985a: 117).

Significantly, this is a case of acting to benefit another in need. By definition,

as we have seen, perfection concerns the realization of the agent’s own good.
This example thus illustrates Spinoza’s view that since we are social, intel-
lectual beings, our goods are common in ways they might otherwise not be.
They are jointly realizable; so a perfect human realizes their own good in cases
like this by responding to the greater needs of others. Moreover, the virtu-
ous do what they do by natural “necessity.” They simply do what they would
do (if unimpeded) in accordance with the laws of their nature. They are thus
“obliged.” This can be seen as a version of Augustine’s “Love and do what you
want” (Augustine 1990: XIV.110). Spinoza’s might read: “Understand (your
nature) and do what you will.”
This clearly reshapes the deontic concept of moral duty in a subversive way.
It stipulates meanings for deontic vocabulary so that the concepts they express
can fit within a eudaimonistic virtue ethics. We will see an identical move in
two of Spinoza’s rough contemporaries who are also critics of early modern
natural law: Shaftesbury and Leibniz. Shaftesbury repurposes “obligation” to
mean a reason provided for the agent’s own good when he argues that agents
have an “obligation to virtue” in the fact that they will live happier lives if they
are virtuous (Shaftesbury 2001: II.i.2.173). Hutcheson and Hume, who fol-
low Shaftesbury’s moral sentimentalism, also follow him in this usage, which
Hume dubs “interested obligation” (Hume 1975b: 278).
For his part, Leibniz understands moral obligation or “moral necessity”
in a way that reduces the deontic concept to one involving (non-normative)
natural necessity and eudaimonist virtue. Differently from Spinoza, and
looking forward more to Hutcheson, Leibniz understands virtue in terms
of benevolence or “charity.” “Moral necessity,” he says, “is equivalent to

https://doi.org/10.1017/9781139025065.006 Published online by Cambridge University Press

 “bear[ing] calmly” and “living blessedly” 125

‘natural’ for a good man” (Leibniz 1988: 171).8 Unexpectedly, perhaps, we will
find a similar usage in Kant.9
If we revise the concept of moral duty in the way we have outlined, we can see
how Spinoza can ground his (suitably revised) democratic political philosophy
in his eudaimonist ethical monism. The crucial element will be the conception
of common good on display in the two passages last quoted. Individuals real-
ize their own good in concert with others, and since Intellect is essential to the
nature of each, the concert must be intellectual. Collective intellectual work in
the political sphere is what democracy is: the people working out what to do
in their common life by reasoning together. Thus Spinoza writes: “A man who
is guided by reason is more free in a state, where he lives according to common
decision, than in solitude, where he obeys only himself” (Spinoza 1985b: 587). We
will see this same theme in Rousseau in Chapter 9.
It was Spinoza’s Theological-Political Treatise, published anonymously in
1670, that set the democratic political agenda for the Radical Enlightenment
(2016).10 It is written in deontic, juridical terms without explicit instructions
about how these are to be understood in Spinoza’s revisionist lexicon. Spino-
za’s readers thus read its claims about natural right and its justification of the
state in comparison with Hobbes’s and likely interpreted their terms similarly.
Two signal differences between their systems were Spinoza’s championing of
democracy and his claim that individuals need not give up their natural rights
to establish the state, as against Hobbes’s view that it takes a contract through
which all lay down their natural rights to establish an absolute sovereign. What
grounds both of Spinoza’s positions is his monistic metaphysics along with his
view of our intellectual collaborative nature.

“Bear[ing] Calmly” and “Living Blessedly”: Spinoza

on Freedom, Happiness, and Power
Spinoza tells us that “the Desire … to live blessedly, or well, to act, etc. is the very
essence of man” (1985b: 557–558). Spinoza’s “etc.” covers a lot. In addition to
the items in Nadler’s “identity – virtue=knowledge=activity=freedom=power=
perfection, we could add happiness, and living and “bearing calmly” (Nadler
2006: 256). All are roughly identical for Spinoza. “[T]he foundation of virtue
is this very striving to preserve one’s own being, and … happiness consists in
man’s being able to preserve his being” (Spinoza 1985b: 555–556). “Activity” is
conduct that realizes our (intellectual) nature as we necessarily strive to do as

8
Similarly, Leibniz defines justice as the “charity of the wise,” thereby reducing justice to
benevolence in a way that is like, but different from, Hutcheson (Leibniz 1988: 171). For
Hutcheson, see the discussion in Chapter 5.
9
See Chapter 8.
10
See Jonathan Israel’s introduction to Spinoza 2007 and Israel 2001.

https://doi.org/10.1017/9781139025065.006 Published online by Cambridge University Press

 126 spinoza, cudworth, shaftesbury, and leibniz

the kind of beings we are. And this activity is, simultaneously, virtuous, perfect-
ing, power exercising and power realizing, free, and that in which our happi-
ness consists.
Spinoza’s Hebrew name was “Baruch,” or “blessed.” Perhaps one of Spi-
noza’s greatest achievements was to show how it is possible, both in theory
and in practice, to “live blessedly” outside a religious context. In his own life,
Spinoza faced extraordinary challenges. An excluded Jew who was officially
separated from his own community in Amsterdam, he made his living as a lens
grinder and died at the age of forty-four, probably from pulmonary tubercu-
losis brought on by the silica in glass dust (Nadler 2006: 407). By all accounts,
however, Spinoza lived largely without complaint and unaffected by the nega-
tive passions and emotions that make it difficult to “bear” life “calmly.” To put
it in the reductive terms of our own age, Spinoza lived “positively,” and didn’t
give in to “negative stuff.”
If these contemporary terms seem overly simple or superficial, they are not
far from Spinoza’s own formulations. Somewhat like Hobbes, Spinoza has a
taxonomy of the emotions that works with two valences, positive and negative
“affect,” with every passion or emotion, however complex, being composed of
a combination of beliefs or imaginings and positive or negative affects toward
the believed or imagined state. To take an example, Joy and Sadness, which are
basic affects for Spinoza, are felt when “the Mind passes to a greater perfection”
and when “it passes to a lesser perfection,” respectively (1985b: 501). This is only
one instance of how Spinoza’s psychology flattens our internal life. Sadness in,
say, the loss of a child hardly seems well captured by this formula. “Grief” does
not appear in Spinoza’s lexicon.
Similarly, “Hate” is simply “an inclination to avoid something which has
caused some evil,” which is only the privation of good, and “Hatred toward him
who has done evil to another” is “Indignation” (1985a: 107; 1985b: 507). This
significantly deflates both hatred and indignation, taking, as it were, the “fire”
out of both.
Similarly with Love, which for Spinoza is simply a species of Joy “accom-
panied by the idea of an external cause” (1985b: 511). As Spinoza analyzes
it, love is not really an attitude toward any being for their sake. Neither is it
connected to deeply felt attitudes like joy, sadness, and grief on the beloved’s
behalf. Spinoza’s love seems to lack the emotional depth of heartfelt human
connection.
Throughout, it is important to bear in mind that Spinoza holds that nega-
tive emotions always involve the error of failing to understand the world as
it really is. “Hate, then, the direct opposite of Love, arises from that error
which comes from opinion” (1985a: 101). The way to “bear” things “calmly,”
therefore, is simply to understand, to perfect our intellectual nature, as we
naturally strive to do. We are subject to passions or affects only when, not

https://doi.org/10.1017/9781139025065.006 Published online by Cambridge University Press

 “bear[ing] calmly” and “living blessedly” 127

having exercised our intellectual powers, we have inadequate or confused

ideas. Acting on such confusion is not really action, properly so called, nor
does it result from will, again, properly so called. As such, “[t]he Mind strives
to imagine only those things that posit its power of acting” (1985b: 525). Only
good can come from our intellectual striving: perfection, action, virtue, hap-
piness, and blessedness. And only bad from emotions and passions, includ-
ing from love as we ordinarily understand it, namely, as involving emotional
vulnerability to another.
Spinoza’s view of the relation of reason, passion, and action is quite close to
Hutcheson and Hume on the “calm” versus the “violent” passions, and Hume’s
famous claim that “reason is the slave of the passions” (Hutcheson 2002; Hume
1978: 415). Spinoza does hold that it is only through the Intellect that we are able
to break the bondage of non-rational affects and realize free action, through
our intellectual nature’s being an “adequate cause” (Spinoza 1985b: 493). This
may make it seem as if reason is the master, rather than the slave, of the pas-
sions, as Hume says. But there is no real difference with Hume. The doctrine
that Hume opposes is the one we find in Plato and, among the moderns, in
Butler and Kant, among many others, that reason is the faculty of normative
judgment and that it can motivate, that there is such a thing as what Kant calls
“practical reason.”
Spinoza denies this doctrine also. What the Intellect grasps, according to
Spinoza, is nothing normative; it grasps “laws of nature” rather than “laws of
freedom,” as Kant will put it (Kant 1996b: 5: 65). And although it is impor-
tant to Spinoza that passions’ strengths, indeed, that their very existence,
depend on the absence of Intellect, he decidedly does not hold, as Plato and
Kant do, that we can do what is good and right, regardless of the strengths of
the passions. And he also denies the mainstream modern moral philosophi-
cal view, again, that anyone subject to morality can comply with it simply by
exercising the capacity for free moral agency that makes them thus subject.
On all these points, he agrees with Hutcheson and Hume (and, of course,
with Nietzsche).
Spinoza’s distinction between passion and will thus anticipates
Hutcheson’s and Hume’s distinctions between the “calm” and “violent pas-
sions.” Like Spinoza, Hume and Hutcheson draw the line based on whether a
motivating state is informed by reason or intellect. Violent passions are those
that are not, in the current terms, “reasons responsive,” those that do not
stem from reason or intellect. Hutcheson also holds a version of Spinoza’s
thesis that we are moved to bring about good when we maximize “calmness”
through theoretical reason or intellect. The difference is that Spinoza’s view
is steadfastly eudaimonist, whereas Hutcheson holds the “modern” view that
“calm desires” can either be for the agent’s own good or for good impartially
considered.

https://doi.org/10.1017/9781139025065.006 Published online by Cambridge University Press

 128 spinoza, cudworth, shaftesbury, and leibniz

Cudworth
A particularly interesting comparison and contrast with Spinoza is the Cam-
bridge Platonist Ralph Cudworth.11 Like Spinoza, Cudworth is a rationalist and,
moreover, one who, unlike the eighteenth-century British rationalists who draw
from him, emphasizes non-deontic over deontic ethical aspects. But Cudworth
is quite unlike Spinoza in holding that ethics cannot be reduced to (non-ethical)
metaphysics and that reason is the faculty of irreducibly ethical cognition.
Cudworth was best known during his lifetime for his metaphysical thought
and for his masterful command of classical philosophical and theological texts.
Both were on display in his massive work, The True Intellectual System of the
Universe, the only work he ever published, in 1678. But just as Locke’s Essay was
written with an eye to issues “about the principles of morality and revealed reli-
gion” (Woolhouse 2007: 98), so also had Cudworth’s System a moral agenda.
Cudworth tells us in the Preface that he originally projected three volumes, all
to be concerned with divine government. The first was to focus on refuting argu-
ments for atheism and “democritick fatalism,” the second, to argue for an essen-
tially good God, and the third to argue that there is “a Liberty, or Sui-Potestas, in
Rational Creatures, as may render them Accountable,” “that we are so far forth …
Masters of our own Actions, as to be Accountable to Justice for them, or to make
us Guilty and Blame-worthy for what we doe Amiss” (Cudworth 1678: Preface).
In the event, Cudworth published only the first volume. The projected
second and third books of the System remained unpublished, although the
posthumously published Treatise Concerning Eternal and Immutable Moral-
ity (1731) probably included material from the second. Cudworth left lengthy
manuscripts, many on freedom of the will, which must represent attempts at
the third.12 Thus even in the System Cudworth is attempting to come to grips
with metaphysical issues that are especially relevant to morality conceived as
standards with which we are distinctively accountable for compliance.

Against Positivism
We shall return to Cudworth’s writings on moral accountability and “moral
free will” presently. They represent some of the most philosophically sophisti-
cated reflections of the period on autonomy and the will. Moreover, they show

11
For a more detailed version, see Darwall 1995: 109–148, from which I here draw. Also, the
Cambridge Platonism Sourcebook is a treasure trove of Cudworth’s writings and Cambridge
Platonism more generally. Most important for our purposes, it has digital transcriptions
of Cudworth’s manuscripts with which we shall be concerned. www.cambridge-platonism
.divinity.cam.ac.uk/about-the-cambridge-platonists/circle-network-constellation
12
Cudworth ~1670. Only one of these (Mss. 4978) has ever been published: A Treatise of
Freewill, which is included with Cudworth 1996. (It was originally published in 1838.) For
a useful overall discussion of the provenance, history, and contents of the manuscripts, see
Passmore 1951: 107–113.

https://doi.org/10.1017/9781139025065.006 Published online by Cambridge University Press

 against positivism 129

that Cudworth was concerned, as Pufendorf and Locke had been in their own
ways, with coming to terms with the conceptual demands placed on our ethical
thought, and on our moral psychology, by morality’s fundamentally deontic
character. First, however, we should consider Cudworth’s criticisms of volun-
tarist attempts to ground moral deontology.
This is a primary focus of the Treatise, which seeks to argue against any view
of morality as “positive, arbitrary, and factitious only,” whether this view takes
a Hobbesian form or is advanced by “modern theologers” who seek to derive
morality from the will of God (1996a: I.i.1,5). Against these views, Cudworth
argues that if fundamental moral terms like “moral good and evil, just and
unjust” are to have meaning, and if they are not “names for nothing else, but
willed and commanded,” then what they refer to “cannot possibly be arbitrary
things, made by will without nature” (I.ii.1). Cudworth thinks he can exclude
the second alternative, that moral terms include reference to will or command
in their content. If, for example, “morally wrong” simply meant being contrary
to the sovereign’s or God’s command, then “it is wrong to act contrary the
sovereign’s or God’s command” would become a mere tautology without nor-
mative force. For moral propositions to be essentially normative, then, some
further claim of the legitimacy of the sovereign’s or God’s commands must
either be asserted or assumed. Moreover, Cudworth continues,
it was never heard of, that anyone founded all his authority of command-
ing others, and others’ obligation or duty to obey his commands, in a law
of his own making, that men should be required, obliged, or bound to obey
him. … For if they were obliged before, then this law would be in vain and
to no purpose. And if they were not before obliged, then they could not
be obliged by any positive law, because they were not previously bound to
obey such a person’s commands (I.ii.3).

Any authority necessary for God’s commands to obligate can therefore not
itself be created by God’s command.
As we noted in earlier chapters, this is a point that Locke and Pufendorf
effectively grant, since they hold that God has authority by virtue of our com-
plete dependency on him for all that is good in our lives. So although they
think that all (other) facts about our moral obligations depend upon God’s
legislative will, they nonetheless must implicitly accept a background moral
fact, namely, that God has an authority that makes his commands binding
obligations for his creatures, an authority that is grounded in our dependency
on him, not in his commands. They can still hold, as we also noted, that any
actual obligations depend on God’s positive legislative acts. But they are com-
mitted to the background moral fact necessary for God’s legislation actually to
bind nonetheless.
The thrust of Cudworth’s anti-voluntarist arguments in the Treatise is to
show that moral properties must depend on the nature of things that have
them and that natures are “eternal and immutable.” If stealing is wrong, it is

https://doi.org/10.1017/9781139025065.006 Published online by Cambridge University Press

 130 spinoza, cudworth, shaftesbury, and leibniz

because it has a nature that makes it wrong. Not even God can make stealing
not be wrong by fiat, any more than he can make the angles of a triangle sum to
something other than two right angles. To change an action’s moral properties,
therefore, it would have to be possible to change its nature without, per impos-
sible, changing what it is.

Metaphysical Idealism?
Now this might seem to suggest a kind of Platonic realism that holds that
moral properties and facts are completely mind independent. But although
­Cudworth’s thought clearly owes much to Plato, it may owe more to a
­neo-­Platonist like Plotinus.13 Cudworth considers the question, “where these
immutable entities do exist,” and answers, “first that as they are considered
­formally, they do not properly exist in the individuals without us” (IV.iv.4).
Natural philosophy has shown that no substantial forms exist in material
­bodies. But, he continues, “[n]either do they exist somewhere else apart for [sic]
the individual sensibles, and without the mind” (IV.iv.4). “These ­intelligible
ideas or essences of things … exist, nowhere but in the mind it self.”
Cudworth remarks that he actually agrees with Hobbesian nominalism that,
quoting Hobbes, “‘There is nothing in the world universal but names,’” if this
be taken, Cudworth adds, to include all things “existing without the mind” (IV.
iii.15). What Cudworth denies is that universal natures and essences, includ-
ing “eternal and immutable morality,” though mind dependent, are not real.
Writing about relations like cause and effect and whole and part, he says that
though these are “notions of the mind and modes of conceiving in us that only
signify what things are relatively to intellect,” this does not make them less real.
To the contrary, mind or “intellect … hath more of entity in it than matter or
body” (IV.ii.5).
Cudworth’s metaphysics are, therefore, some form of idealism, as we shall
see is borne out also by what he says about moral properties and facts more
specifically. As is often the case with idealisms, Cudworth’s is driven largely
by epistemological considerations. Knowledge of objects is impossible without
universal notions. Since these cannot be extracted from experience, they must
be prior to it. “Knowledge,” he says, “is a comprehension of a thing prolep-
tically, and as it were a priori” (III.iii.5). And this means that the categories
that make knowledge possible must be present in the very activity of mind.
He therefore holds all knowledge to be ultimately reflexive: “Knowledge and
intellection doth … comprehend its object within itself, and is the same with
it.” “The intellect and the thing known are really one and the same” (III.iii.4). It

13
Passmore notes that Coleridge thought that Cudworth should be described as a Plotinist
rather than a Platonist (Passmore 1951: 14).

https://doi.org/10.1017/9781139025065.006 Published online by Cambridge University Press

 ethical idealism? 131

follows that the “intelligible forms by which things are understood or known,
are not stamps or impressions passively printed upon the soul from without,
but ideas vitally protended or actively exerted from within itself” (IV.i.1).14
All universal natures and essences, including moral ones, therefore, are
modifications of mind. They have reality only so long as mind, specifically,
God’s archetypal mind of which all created minds are ectypes, exists. Although
he is no theological voluntarist, Cudworth actually agrees with Pufendorf and
Locke, that morality could not exist without God. No universal natures could.
“If … there be Eternal Intelligibles or Ideas, and Eternal Truths; … then must
there be an Eternal Mind Necessarily Existing, since these Truths and Intelli-
gible Essences cannot possibly be any where but in a Mind” (1678: 736).

Ethical Idealism?
Since Cudworth holds fundamental moral truths to be “eternal and immuta-
ble,” it would seem that these too would have to have the same mind depen-
dence. This is borne out in the Treatise’s anti-voluntarist arguments. Cudworth
distinguishes between “natural” and “positive” goodness and justice in the pro-
cess of arguing against the voluntarists that unless at least something is natu-
rally just or good (e.g., complying with a sovereign’s commands in general),
then nothing can be positively just or good. Notice, however, how Cudworth
makes the distinction.
[T]here are some things which the intellectual nature obligeth to of itself
(per se) and directly. … Other things there are which the same intellectual
nature obligeth to by accident only, and hypothetically, upon condition of
some voluntary action of our own or some other person’s … (1996a: I.ii.4).

Since Cudworth’s doctrine is that it is the eternal and immutable character of

the act that makes it right or wrong, required or prohibited, we might expect
him to say that the obligation derives directly from that. But Cudworth empha-
sizes that the obligation consists in a “relation to the intellectual nature” and
also that in cases when someone is obligated by a positive command, that “it is
not the mere will and pleasure him that commandeth that obligeth to do posi-
tive things commanded, but the intellectual nature of him that is commanded”
(I.ii.4).15

14
A. O. Lovejoy remarked that Cudworth’s epistemology involved a “Copernican Revolu­
tion” not unlike the more familiar Kantian version: “Knowledge depends upon the ‘con-
formity of objects to our mode of cognition’ rather than upon the conformity of our mode
of cognition to objects” (Lovejoy 1908). Universal natures exist formally by way of cogni-
tive modes – thus does Cudworth’s epistemology drive his metaphysics.
15
Gill writes that “the intellectual nature that obliges to do what is naturally good does not
seem to be our intellectual nature, but rather the intellectual nature, that is the nature of

https://doi.org/10.1017/9781139025065.006 Published online by Cambridge University Press

 132 spinoza, cudworth, shaftesbury, and leibniz

Once we recall that Cudworth believes that all “intelligible essences” and
“eternal truths” are “modifications of mind and intellect” (IV.iv.8), these
claims are no longer puzzling. Anything having universal reality must involve
some relation to intellectual nature.
What is especially notable is that Cudworth holds that the “modifications of
mind” in which moral natures and facts consist are essentially practical, moti-
vating states of mind. This comes out in two remarkable passages. One occurs
early in the Treatise, closely following the passages about obligation and intel-
lectual nature we just noted:
[I]f the rational or intellectual nature in itself were indetermined and
unobliged to any thing, and so destitute of all morality, it were not possible
that any things should be made morally good or evil, obligatory or unlaw-
ful (debitum or illicitum) or that any moral obligation should be begotten
by any will or command whatsoever (I.ii.6).

Note the connection Cudworth draws here between being obligated to do

something and being “determined” to do it. If it is not clear already in this pas-
sage that Cudworth has in mind a form of practical determination or motiva-
tion that is not necessarily in play with other intellectual essences, it becomes
clear in a following remarkable passage in which Cudworth claims agreement
with, of all philosophers, Hobbes. Here is what Cudworth says:
[I]t is truly affirmed by the author of the Leviathan, “That there is no
common rule of good and evil to be taken from the nature of the objects
themselves,” that is either considered absolutely in themselves, or rela-
tively to external sense only, but according to some other interior anal-
ogy which things have to a certain inward determination in the soul itself
(IV.iv.4).

So far, this just asserts Cudworth’s agreement with Hobbesian nominalism so

far as mind-independent aspects of “the world” are concerned. But Cudworth
then goes on to distinguish the distinctively practical modifications of mind in
which moral natures and facts consist:
Not that the anticipations of morality spring merely from intellectual
forms and notional ideas of the mind … but from some other more inward
and vital principle, in intellectual beings as such, whereby they have a natu-
ral determination to do some things and to avoid others (IV.iv.4, emphasis
added).

reality as it exists eternally and immutably, independent of our particular minds” (Gill
277). It will become clear below, however, that Cudworth holds that we are capable of being
obligated only because we ourselves can share (God’s perfect) intellectual nature.

https://doi.org/10.1017/9781139025065.006 Published online by Cambridge University Press

 ethical idealism? 133

Cudworth concludes that if the soul “hath no innate active principle of its own, …
upon this hypothesis there could be no such thing as morality” (IV.iv.4).16
As we’ve seen, Cudworth holds that all universal essences exist by way of
mind, ultimately God’s archetypal mind. It follows that the existence of moral-
ity, like any universal essence or fact, depends upon the existence of God. But
notice how Cudworth puts the point:
[I]t is not possible that there should be any such thing as morality, unless
there be a God, that is, an infinite eternal mind that is the first original and
source of all things, whose nature is the first rule and exemplar of morality
(IV.vi.13).

God’s idea of a triangle must serve as the archetype of triangles, but even if this
is a modification of God’s mind, this does not mean that God’s nature is “the
first rule and exemplar” of triangularity.
The reason that God’s nature is the rule and exemplar of morality is that God
is, for Cudworth, an ideal of moral agency. Perfect mind is essentially practical;
it is in its nature “determined to do some things and avoid others.” Cudworth is

16
In Darwall 1995 I argue that this distinguishes Cudworth’s view from that of the
­eighteenth-century British rational intuitionists like Clark and Price. And I argue there
that Cudworth’s view amounts to a kind of autonomist internalism, according to which
obligation consists in motivation that arises through the agent’s exercise of their power
for self-determination. Michael Gill criticizes my arguments for these claims and argues
that Cudworth is best seen as an intuitionist for whom moral truths are akin to truths
of geometry (Gill 2006: 275–277). Gill acknowledges that Cudworth holds morality to be
mind dependent, but argues that morality “cannot be constituted by the motives of human
agents” (275). That is of course true, but it does not mean that moral obligation cannot
depend on human motives. In my view, Cudworth holds, like Kant, that the possibility of
moral obligation only arises for imperfect rational beings who have in addition to what
Kant calls “pure practical reason” other, not purely rational, sources of motivation. So any
motivation that is part of intellectual nature that can be necessary for or part of moral obli-
gation must be shared by human moral agents.
Gill claims that Cudworth’s arguments against voluntarism “imply that morality con-
sists of principles that are independent of anyone’s motives,” including God’s. Although
it seems clear that Cudworth is assuming the goodness of God’s love as a background fact
that is itself independent of any fact about what God necessarily (lovingly) desires, it does
not follow from this that facts about moral conduct are also thus independent. Hutcheson,
for example, holds that facts about the moral choiceworthiness of acts derive from the
moral goodness of benevolence. What moral agents should do, he thinks, is whatever
we would do if perfectly benevolent. Of course, Hutcheson holds the moral goodness of
benevolence to be contingent, and Cudworth holds it to be necessary, but their views might
otherwise be similar in this respect. So long as it is a necessary truth that God is loving, and
a necessary truth also what a perfectly loving being would want for the circumstances of
human life, moral truths will be “eternal and immutable.” For a criticism of Gill’s claim
that for Cudworth the truths of morality are like those of geometry, see Leisinger 2019.

https://doi.org/10.1017/9781139025065.006 Published online by Cambridge University Press

 134 spinoza, cudworth, shaftesbury, and leibniz

of course careful to distinguish his view from theological voluntarists. Morality

does not depend upon God’s arbitrary will. If God’s will were utterly arbitrary,
there would be no such thing as morality. According to the theological volun-
tarists, “there is no pattern or archetypal exemplar of morality in God” (~1670:
4979.87).17 It is only because God’s perfect mind does provide an exemplar of
will and motivation – that is, because perfect mind is essentially practical – that
eternal and immutable morality exists.

Love
However, what motivation is essential to perfect mind? Cudworth does not
say in the Treatise, but the general shape of his answer is clear from his other
writings. In the process of discussing Plato’s Form of the Good in the System,
Cudworth complains that “Plato sometimes talks too Metaphysically and
Clowdily about it.” “Nevertheless,” he continues, “he plainly intimates … that
this Nature of Good which is also the Nature of God, includes Benignity in it”
(1678: 205). Then he writes,
But the Holy Scripture without any Metaphysical Pomp and Obscurity, tells
us plainly Both what is that Highest Perfection of Intellectual Beings, which
is … Better than Reason and Knowledge, and which is also the Source, Life
and Soul of all Morality, namely that it is Love or Charity (205).

From there Cudworth proceeds to quote at length the famous passage on love
from Paul’s First Epistle to the Corinthians, adding that without love, “I have
no Inward Satisfaction, Peace or True Happiness” and an “destitute of all True
Morality, Vertue, and Grace.”
The fundamental ethical motive, essential to perfect mind, thus, is love.18
Cudworth tells us little about love’s precise character. He says love “dispenses it
self Uninvidiously,” with an equal regard of some sort, but he does not try to for-
mulate any principle like Cumberland’s “common good of all rationals.” Since
Cudworth says that “justice is an essential branch of this Divine Goodness,”
he must hold that love and justice cannot conflict. But it is impossible to be
sure whether he holds that love is benevolence tempered by justice, or whether
he thinks that justice can somehow be resolved into benevolence (I.iv.9).19 It
should not be surprising that Cudworth is unclear on this point since the fact
that universal benevolence might conflict with other moral concerns such as

17
This passage comes from Cudworth’s manuscripts on freedom of the will.
18
Indeed, at one point, Cudworth, like Cumberland, identifies God with love: “in a Rectified
and Qualified sence, this may pass for true Theology; That Love is the Supreme Deity and
original of all things: namely, if by it be meant, Eternal, Self-originated, Intellectual Love”
(1678: 123, misnumbered 117).
19
I am indebted to Roberts unpublished for this passage.

https://doi.org/10.1017/9781139025065.006 Published online by Cambridge University Press

 self-determination 135

justice or honesty was not well appreciated until Butler’s Dissertation upon the
Nature of Virtue some fifty or so years later.
As mentioned earlier, Cudworth opposes an ethics of love to a “dead Law
of outward Works” (123). God’s love makes morality possible; without it there
would be no distinction between moral good and evil. Because we can partici-
pate in God’s (practical) intellectual nature, we can be moral agents. “Love is at
once a Freedome from all Law, a State of purest Liberty, and yet a Law too, of
the most constraining and indispensable Necessity” (1969: 124).
Unlike the natural lawyers, Cudworth’s ethics do not start from the prob-
lem of obligation. Even so, as the passages quoted earlier from the System
show, the legitimation of moral obligation is a primary purpose behind Cud-
worth’s metaphysical speculations as well as his writings on freedom of the will.
Although his ethics is at bottom an ethics of love, Cudworth also aims to argue
that love grounds moral obligation. We are obligated, most fundamentally,
to give expression to the source of morality we have within, thereby enliven-
ing what would otherwise be the “dead carkasse” of outward acts. The agent’s
“intellectual nature” is what makes them obligated, and love is the practical
form of the intellect.
This motivational source is, however, only one necessary condition of moral
obligation. The other is what Cudworth calls “free will,” the agent’s capacity to
determine conduct by freely forming their own practical judgment.

Self-Determination
This is the subject of Cudworth’s surviving manuscripts, which are galvanized
by the problem of accountability: conditions for the very possibility of moral
“guilt” and “blame” (1678: Preface). If obligation is distinctive of morality, it is
distinctive of obligation that we are accountable for acting as we are obligated.
When we take up an attitude toward, make a judgment of, or act in response to
persons as responsible for their acts, we impute their acts to them. We regard
persons as the causes of their own acts, imputing to the person the determina-
tion of the act.
Although there is a weak sense in which someone might “blame” a defective
alarm clock as the cause of their oversleeping, Cudworth notes that they do not
“imput[e] to [the clock] its being the cause of its own moving well or ill” (1996:
155). By contrast, when “we blame a man for any wicked actions, … we blame
him not only as doing otherwise than ought to have been done, but also than he
might have done, …, so that he was himself the cause of the evil thereof” (155).
Our blame takes a self-­determining person as object. Commendation or blame
not “only signif[ies] our approbation or dislike of the [acts] themselves … but
doth also reflect upon the person as the cause of either to himself when it was
possible he might not have been so” (~1670: 4980.31).

https://doi.org/10.1017/9781139025065.006 Published online by Cambridge University Press

 136 spinoza, cudworth, shaftesbury, and leibniz

Cudworth holds that morality can genuinely obligate only agents who can
determine themselves to act morally. He uses a variety of terms for the requi-
site power, sui potestas, as well as some of his own coinage: “autexousy” and
“autexousious power”.20 Cudworth contrasts his account of self-determining
imputable agency with Hobbes’s account of the will as the “last appetite.” The
will, Cudworth says, “is no particular appetite.” It involves “the soul compre-
hending its whole self.” Volition is the agent’s “last practical judgment,” he
says, using this phrase in his own technical sense of a more (or less) considered
judgment made actively by the agent themselves in comprehension, more or
less, of their various motives (4982.80).
Hobbes’s account of will as “nothing but alternate passions where the victory
at last falls to that passion whose necessary force is most prepondering” might
aptly describe the behavior of “brutes.” But they lack the self-awareness neces-
sary for accountable agency. “[T]here is no … thing in them [brutes] that takes
notice of all, no reduplicate self-active principle that … exerting itself more or
less can oversway things one way or the other” (4980.7).21
In addition to Hobbes’s theory, Cudworth rejects two other views he thinks
do not adequately capture the role of “self-comprehensive” practical thinking
and judgment in forming the will. One is the indifferency view, which holds
that agents simply plump for some option or other after the “last dictate of the
understanding” (1996: 168). This alternative makes free choice blind and “mere
irrationality” (169). A second view agrees with this criticism of the first, but
holds that rational action is the result of a necessary process in which agents
always do what they believe best. Cudworth rejects this account because it
reduces moral agency to processes that occur in an agent rather than anything
the agent themselves directs. A belief of “the understanding” is something the
agent has or a state they are in. By contrast, a considered practical judgment
is something the agent themselves makes (178–179). An agent’s power of self-
determination consists in its capacity to act, not just from itself, but also “on
itself,” to be “at once both agent and patient,” as it directs its own practical
all-things-considered deliberation and act on the basis of that (~1670: 4982.4).
As Locke would later, Cudworth makes an important distinction between
merely believing that an action is the best thing to do and actively judging it to
be so as the result of the agent’s own deliberative thought.22 Were the under-
standing directly to determine the will, then there would “nee[d] nothing else
to be done in order to virtue [sic] but only the informing of a man’s speculative

20
“Autonomy” only had a political use before English translations of Kant’s “autonomie.”
21
As we shall see, Cudworth maintains there is a kind of free will whose scope covers actions
motivated by desires for our overall happiness. But though he calls this “animal free will,”
he does not mean that (other) animals actually have it.
22
As mentioned in the last chapter, it is arguably under Cudworth’s influence that Locke
makes this distinction.

https://doi.org/10.1017/9781139025065.006 Published online by Cambridge University Press

 free will and obligation, animal and moral 137

understanding concerning the nature of good” (4982.10).23 Like Locke in the

revised Essay, Cudworth holds that we do not always do what we think best,
but that we do always do what we ourselves judge best on the basis of our own
(more or less) considered self-reflection. The will just is the “last practical judg-
ment,” which is the conclusion to this motivation-infused process (4982.80).
“[W]e are not,” Cudworth writes, “merely passive to our own practical judg-
ments and to the appearances of good, but contribute something of our own to
them” (1996: 179). Human beings can do nothing to affect their fundamental
sources of motivation. These are, as Cudworth puts it, motives of the “animal”
and “divine life,” respectively, that together constitute our “necessary nature”
(4980.56). What we can do (in addition to acquiring virtuous habits (4980.57,
61)) is to undertake to get a more or less comprehensive grasp of our own prac-
tical situation in self-­comprehensive deliberation.
An agent can thereby affect the “light” in which they view options, or even
what options they have in view, as “these may be very different accordingly as
we more or less intensely consider or deliberate which is a thing … in our own
power” (1996: 179). Thus, “we ourselves finish or determine the last conclusive
visum or phantasy of good, which is called the last or ruling practical judg-
ment” (~1670: 4980.239). The requisite practical judgment can thus only be
made by agents with free will. “[It] is made up partly from nature, some natural
congruity or other, higher or lower, partly from we ourselves, the soul self-
comprehensive or self-active” (4980.239).
Cudworth distinguishes between successful guidance by correct percep-
tions of value (“liberty” or “sui potens”), which God has, from “free will” or
“sui potestas,” the power we imperfect rational beings have to make and act
on considered, self-reflective practical judgments to the best of our ability
(4980.31, 44). In addition to these, there is also the voluntariness of exercising
or failing to exercise this power. Sin or fault “is not to be imputed to free will
as if it were actively exercised by freewill.” Rather “it is nothing but a voluntary
deficiency of a good power” (4982: 22).24 What makes wrongdoing voluntary
is that the agent “might have made a better judgment than now he did, had he
more intensely considered, and more maturely deliberated” (1996: 179). The
agent “is conscious also that it has a self-active autexousious power to excite
and command a full consultation, deliberation, and search” (~1670: 4980.66).

Free Will and Obligation, Animal and Moral

The capacity for self-comprehensive choice is not, however, sufficient for moral
accountability and, therefore, for moral obligation. Cudworth distinguishes

23
On this, see Leisinger 2021.
24
Roberts unpublished emphasizes this “asymmetrical” character of Cudworthian liberty.

https://doi.org/10.1017/9781139025065.006 Published online by Cambridge University Press

 138 spinoza, cudworth, shaftesbury, and leibniz

between a kind of agency and free will, “animal free will,” which though it
involves self-reflective self-determination is not yet sufficient for “moral free
will,” and therefore, for moral agency and obligation.
According to Cudworth, human beings participate in two “stories of life” –
“animal life” and “divine life.” The former is composed of desires, feelings, and
experiences arising from our bodily, “animal” nature; the latter, of the instincts
of love and “honesty” that are essential to perfect mind. A being might have only
animal ends and still have a kind of rationality, one Cudworth calls “inferior
reason,” that involves a “larger comprehension of [its] own utility” (4979.5).
Cudworth stresses that this is different from particular animal passions and
appetites, even when they are combined with instrumental rationality individ-
ually. Inferior reason’s “larger comprehension of our own utility” is tethered
neither to specific desires nor to any particular point in time (4980.6).25
A being restricted to animal desires and inferior reason can still have a kind
of free will, which Cudworth calls “animal free will.” And Cudworth holds that
this is sufficient for a kind of “animal” obligation.
[I]t is possible that the soul may be here defective in the exertions of this
autexousious faculty … in … succumbing passively to the urging impor-
tunity of the present appetite … and then it is conscious to itself that it is
faulty and blameworthy, not as guilty of any moral evil properly called sin,
[but] because it was wanting to itself within the sphere of the animal life as
to private utility, and hereupon follows … regret of mind, self-displeasure
(4982.17).

Animal free will, including inferior reason, is necessary for animal obligation.
And the proper exercise of this power is sufficient to discharge the obligation.
Animal free will and inferior reason are, however, insufficient for moral obli-
gation. Cudworth pointedly notes (with an eye to Hobbes) that were agents
only so far equipped, “there is no doubt but they might have societies, polities,
and laws … enforced with punishments and rewards to good purpose, in order
to the advantage of private persons and the safety of the whole, which is the
very constitution of the Leviathan” (4980.9). But this would not be an “obliga-
tion truly moral.” Against Hobbes, Cudworth echoes Pufendorf: “Laws could
no otherwise operate or seize upon them than by taking hold of their animal
selfish passions … and that will allow of no other moral obligation than this
utterly destroys morality” (4980.9).26
In order to have moral obligations, an agent must have “moral free will,”
which involves the same power of self-comprehensive self-command, but com-
bined with the distinctive motives of the divine life and, therefore, “superior

25
“Utility” here is restricted to animals’ desires only.
26
Compare this with Shaftesbury’s similar remark that by making morality depend on sanc-
tions, Locke “threw … virtue out of the world.”

https://doi.org/10.1017/9781139025065.006 Published online by Cambridge University Press

 shaftesbury 139

reason,” alongside animal desires and inferior reason. “For the better expla-
nation” of moral obligation, Cudworth writes, “it must first be supposed that
besides the spring of animal life, there is in the soul, a spring or principle of
another superior intellectual life” (4982.20). What Cudworth calls alternately
“superior reason,” “love,” “the principle of honesty,” and “conscience” exists in
us not as God’s overflowing love is in him. God is love essentially and necessar-
ily. Though we participate in the divine life, we have “a principle of animal life
in us besides” (4982.20). So, unlike God, we imperfect rational beings require a
power of free will to enable us to self-reflectively guide our lives by this superior
motive. We require moral free will. In this way, Cudworth combines an eth-
ics of love with the characteristics of moral agency necessary for morality to
obligate us.

Shaftesbury
Anthony Ashley Cooper, the 3rd Earl of Shaftesbury, was the grandson of the
1st Earl, the famous Whit politician. When Shaftesbury was three, his grandfa-
ther assumed his formal guardianship and turned over the direction of much
of his education to the 1st Earl’s medical attendant and sometime advisor
and secretary, John Locke. Thus began a complex relationship.27 Shaftesbury
remained indebted to Locke, but his philosophical instincts ran directly oppo-
site to Locke’s.28
Shaftesbury was more attracted to the Cambridge Platonists’ affective ratio-
nalism, represented by Cudworth. He especially reviled Locke’s voluntarism,
since it made morality external to the moral agent and reduced moral moti-
vation to prods and pulls rather than anything expressive of the agent’s own
virtue or moral character. As Shaftesbury saw it, Locke’s voluntarism was even
more pernicious than Hobbes’s, since the latter’s “slavish” political views made
his form of voluntarism less likely to get a foothold. “It was Locke,” Shaftesbury
wrote, “that struck the home blow … [and] threw all order and virtue out of the
world” (Shaftesbury 1900: 403).
Shaftesbury also opposed Locke’s methods. Anglophone philosophy of the
modern period was generally marked by a desire for analytical rigor and conso-
nance with developing scientific understanding. Even Cudworth was a member
of the Royal Society.29 In his first Enquiry (1748), Hume famously distinguished
between philosophy that seeks to edify and a more scientific sort that seeks
“those original principles, by which, in every science, all human curiosity must
be bounded” (Hume 1975a: 5). Shaftesbury’s philosophy is more of the edifying

27
For details, see Voitle (1984).
28
In what follows, I draw from Darwall 1995: 176–206.
29
Formally called The Royal Society of London for Improving Natural Knowledge.

https://doi.org/10.1017/9781139025065.006 Published online by Cambridge University Press

 140 spinoza, cudworth, shaftesbury, and leibniz

sort. He frequently scorns the metaphysical and epistemological writings of

contemporaries who, like Locke, are stimulated by advances in natural philoso-
phy: “What specious exercise is found in those which are called ‘Philosophi-
cal Speculations’? The formation of ideas, their compositions, comparisons,
agreement and disagreement” (Shaftesbury 1900: 267). Such inquiries, he com-
plains, make a person “neither better, nor happier, nor wiser” (269). “The most
ingenious way of becoming foolish,” Shaftesbury famously remarked, “is by a
System” (2001: I.290). For Shaftesbury, philosophy is the edifying study of the
good – an attempt to find, or more accurately, to inspire in the agent a fixed and
steady aim necessary for them to project an enduring, integral self.30
It is ironic, therefore, that Shaftesbury is best remembered among anglophone
philosophers for his idea of the “moral sense,” which inspired Hutcheson’s and
Hume’s empiricist moral sentimentalism. Sidgwick, for example, treats Shaft-
esbury as the first moralist to make the empirical study of mental phenomena
the basis of ethics, thereby leading to Hutcheson and Hume’s sentimentalism
(Sidgwick 1964: 184–191).31 But Shaftesbury’s moral sense has greater affinities
to the affective or practical rationalist themes we found in Cudworth. Cassirer
wrote, indeed, that it was “principally Shaftesbury” who saved “the Cambridge
School from the fate of a learned curiosity” and made it a “philosophic force in
the centuries to come” (Cassirer 1953: 160; see also Cassirer 1951).

An Ethics of Virtue
Shaftesbury’s ethics mark a radical departure from early modern natural law.
Deontic notions are almost entirely absent from Shaftesbury’s writings, and
when they enter, Shaftesbury tends to recast them in subversive ways. For
example, after having provided an account of “what virtue is and to whom the
character belongs” in Book I of the Inquiry, Shaftesbury begins Book II by say-
ing that “[i]t remains to inquire, What Obligation there is to Virtue” (Shaftes-
bury 2001: II.78). Any impression that Shaftesbury is attempting to account for
deontic aspects of morality within an ethics of virtue is immediately dispelled,
however, when he continues: “or what Reason to embrace it” (II.78). This

30
Diderot wrote in his Encyclopédie, “There are very few errors in Locke and two few truths
in milord Shaftesbury: the former is only a man of vast intellect, penetrating and exact,
while the latter is a genius of the first order.” Whereas “Locke has seen, Shaftesbury has
created, constructed, and edified. To Locke we owe some great truths coldly perceived,
methodically developed, and dryly presented; and to Shaftesbury, some brilliant schemes
often poorly grounded, though full of sublime truths” (translated and quoted in Grean
1967: x–xi).
31
Martineau said that Hutcheson worked out Shaftesbury’s “fruitful hints and construct[ed]
from them a systematical psychology” (Martineau 1901 v. 2, 514). And Albee wrote, “it
is customary to regard Hutcheson’s system as the logical development of Shaftesbury’s”
(Albee 1957: 58).

https://doi.org/10.1017/9781139025065.006 Published online by Cambridge University Press

 an ethics of virtue 141

already is a move from a notion on the law or command side of Suárez’s dis-
tinction, obligation, to a counsel notion. Moreover, Shaftesbury’s eudaimonism
leads him to interpret this question of practical reason in self-interested terms.
The question of whether there is an obligation to virtue, in Shaftesbury’s hands,
becomes that of whether a virtuous life best realizes the virtuous agent’s good.
Hutcheson and Hume will follow Shaftesbury’s revisionist lead. What Shaftes-
bury calls obligation will be what Hume calls “natural” or “interested” obliga-
tion (Hume 1975b: 268; 1978: 498).
So viewed, what makes something “obligating” has nothing to do with
legitimate demand or accountability, not even with a reformed notion of natu-
ral law such as we find in Cumberland, nor even with Cudworth’s idea of a
“law of love.” For Shaftesbury, morality primarily concerns good motives or
“affections” rather than conduct. “[I]t is by affection merely that a Creature
is esteem’d good or ill” (II.22). And when Shaftesbury talks about wrongful
actions, he tends to have in mind actions that result from bad motives (II.32),
turning a law-based ethical conception on its head.
Shaftesbury’s ethics are steadfastly an ethics of virtue. What makes an affec-
tion a good one? Shaftesbury’s answer derives from his theory of nature: a good
affection is a natural one. Indeed, “natural” is often synonymous with “good”
for Shaftesbury. Thus, the passage quoted in the last paragraph ends: “good or
ill, natural or unnatural” (II.22). In the background is a teleological picture
of nature in which subsystems function together to realize a well-functioning
whole. All species and individual members have natural constitutions that are
fitted, indeed designed, for specific functional roles in the “System of all Things”
or “Universal Nature” (II.20, 105). Whether an individual or species is “really”
good or ill depends on whether it enhances or detracts from the good function-
ing of the whole (II.20).
Though ultimately good and ill must be reckoned in relation to the whole
system of nature, Shaftesbury generally writes as though whether an affection is
good or bad primarily concerns how it relates to the flourishing of the species,
which he generally calls the “publick Interest” (II.31). An animal whose violent
instincts must be kept in check by “fear of his Keeper” is not good themselves,
just as Hobbes and Locke’s human agents, kept in check by fear of God, cannot
be good on Shaftesbury’s picture (II.26).32

32
Shaftesbury sometimes writes as if natural affections have the public interest as object, but
his considered view seems to be that natural affections are those functionally suited to pro-
mote the public interest, regardless of their object (Trianosky 1978). “A good Creature is
such a one as by the natural Temper or Bent of his Affections is carry’d primarily and
immediately, and not secondarily and accidentally, to Good, and against Ill” (Shaftesbury
2001: II.26). Good affections are those whose primary function is to realize species good.
“When in general, all the Affections or Passions are suted to the publick Good, or good of
the Species, … then is the natural Temper intirely good” (II.27).

https://doi.org/10.1017/9781139025065.006 Published online by Cambridge University Press

 142 spinoza, cudworth, shaftesbury, and leibniz

There can also be natural and unnatural degrees of affections like self-love
and benevolence. Self-love tends to be “injurious to social Life” only when
immoderate, so only an immoderate degree of self-love is unnatural (II.140).
Similarly, “even as to Kindness and Love of the most natural sort, (such as that
of any Creature for its Offspring) if it be immoderate and beyond a certain
degree,” it can be unnatural also (II.27).

Virtue and Moral Sense

Although having affections adapted to species’ goods is sufficient for the “mere
Goodness” of any sensible creature, Shaftesbury maintains that “that which is
call’d Virtue or Merit” can be realized only by a self-­reflective rational agent
(II.28). Only an agent who is guided by reflective assessments of their own
character and motive (through “moral sense”) or moral conscience can direct
their moral life in a way that is genuinely virtuous. Shaftesbury here embarks
on a path that will prove influential, leading to Butler’s idea of the authority of
conscience and arguably even to Kant on moral worth. Interestingly, however,
it is not a path that his main sentimentalist followers, Hutcheson and Hume,
will follow. To the contrary, the idea that moral goodness must involve the
agent’s moral self-direction is more characteristic of writers who stress morali-
ty’s deontic character and connection to accountability. Rationalists like Clarke
and Price take this position, as does Butler. Shaftesbury’s reasons for ­taking the
position are different, as we shall see.
For rational beings “capable of forming general Notions of Things, not only
the outward Beings which offer themselves to the Sense,” but also actions and
affections can be “fram[ed]” in thought as “rational Objects” and contemplated,
along with their relation to the rest of nature (II.28, 36). When natural affec-
tions, “being brought into the Mind by Reflection, become Objects, … there
arises another kind of Affection towards those very Affections themselves,
which have been already felt, and are now become the Subject of a new Liking
or Dislike” (II.28). This “reflex approbation,” as Butler puts it, or “natural moral
Sense,” as Shaftesbury does, is a second-order approval of natural affections
(Butler 2017: Pr.26; Shaftesbury 2001: II.28, 46).
But what guarantees that moral agents will approve of natural affections,
affections that tend toward the public interest, on reflection? The answer has
to do with Shaftesbury’s teleology of nature, his “aestheticism,” and his view of
the nature of beauty.

The Mind, which is Spectator or Auditor of other Minds, cannot be with-

out its Eye and Ear, so as to discern Proportion, distinguish Sound, and
scan each Sentiment or Thought which comes before it. It can let noth-
ing escape its Censure. It feels the Soft and the Harsh, the Agreeable and
Disagreeable, in the Affections; and finds a Foul and Fair, a Harmonious

https://doi.org/10.1017/9781139025065.006 Published online by Cambridge University Press

 virtue and moral sense 143

and a Dissonant, as really and truly here, as in any musical Numbers, or in

the outward Forms or Representations of sensible Things. Nor can it with-
hold its Admiration and Exstasy, its Aversion and Scorn, any more in what
relates to one than to the other of these subjects (Shaftesbury 2001: II.29).

Moral goodness, for Shaftesbury, is a species of beauty: a harmony, ­proportion,

or beauty of the mind. Since beauty is naturally agreeable – something we con-
template with pleasure – we cannot but approve it.
Functional adaptation and order are part of what beauty is. “Beauty and
Truth are plainly join’d with the Notion of Utility and Convenience, even in
the Apprehension of every ingenious Artist” (III.181). Similarly, the “inward
Beauty of the Body” is “the Harmony and just Measures of the rising Pulses,
the circulating Humours, and the moving Airs or Spirits” (III.181). Thus, the
adaptation of natural affections to their end, the public interest, is beautiful and
agreeable also. Virtue is “moral beauty,” as Hume will hold also (Shaftesbury II.
29; Hume 1978: 527).
Shaftesbury says that beauty and deformity are “immediately perceiv’d by
a plain internal Sensation” (Shaftesbury 2001: II.285). Nevertheless, “moral
sense” is, for him, plainly a cultivated taste, one that develops through free pub-
lic critical discourse about the public good and “liberal Education” (II.293).
“Taste or Judgment … can hardly come ready form’d with us into the World;”
“antecedent Labour and Pains of Criticism” are necessary (III.164,165). In free
conversation we “polish one another, and rub off our Corners and rough Sides
by a sort of amicable Collision” (I. 64). “There is no real Love of Virtue, without
the knowledge of Publick Good. And where absolute Power is, there is no Pub-
lick” (I.107). The sense of moral beauty, even if immediate when properly culti-
vated, nonetheless develops through public liberal critical discourse informed
by reflection on the public good.
But what makes the relevant perception a feeling? How do we “feel” moral
beauty and respond to it in agreeable way? What makes virtue “amiable”?
Shaftesbury holds that though there may be “no real Amiableness or Defor-
mity in moral Acts, there is at least an imaginary one of full force. Tho perhaps
the Thing itself shou’d not be allow’d in Nature, the Imagination or Fancy of
it must be allow’d to be from Nature alone” (II.43). Natural affections’ suit-
ability to public good is itself part of nature, but their agreeableness is not.
It is response dependent. Here we see seeds of the moral sentimentalism of
Hutcheson and Hume.
But quite unlike the later empiricist sentimentalists, Shaftesbury does not
hold that the agreeable and disagreeable responses of “moral sense” are con-
tingent features of human nature. When Shaftesbury says that the imaginative
reflective response to natural affections is “from Nature alone,” he does not
mean just that it has aspects of nature as object, or even that it is the product of
natural causation. Those are true of Hutcheson’s and Hume’s versions of the

https://doi.org/10.1017/9781139025065.006 Published online by Cambridge University Press

 144 spinoza, cudworth, shaftesbury, and leibniz

moral sense or sentiment also. He means that it is a product of the same orga-
nizing rational principle or mind that is immanent in nature, in which we as
contemplating rational minds sympathetically participate. Shaftesbury’s theol-
ogy, philosophy of nature, and notions of “noble Enthusiasm” (III.33), love,
creative inspiration, and mind come together to create a version of the doctrine
of moral sense that is far from the empiricist sentimentalism of Hutcheson and
Hume and closer to Cambridge Platonists like Cudworth. Although moral
sense clearly involves feeling for Shaftesbury, it is more accurate to think of him
as a kind of rationalist moral sense theorist. Moral sense involves the creative,
framing power of reason, not the passive reception of sensations.
The form and structure manifested by the natural affections, and by all natu-
ral beauty, evidence the design of a unifying creative mind. Shaftesbury’s deity
is not the God of orthodox religion.33 The creative mind evident in nature is not
a transcendent being but an organizing principle immanent in nature. A “The-
ist” believes that “every thing is govern’d, order’d, or regulated for the best, by
a designing Principle or Mind, necessarily good and permanent” (II.11). Since
atheism, and worse, “daemonism,” conflict with the contemplation of natural
affections as aspects of a benevolently designed natural order, these outlooks
tend to undermine the reflective love or admiration on which virtue super-
venes.34 “[H]ow little dispos’d must a Person be, to love or admire any thing as
orderly in the Universe, who thinks the Universe it-self is a Pattern of Disorder”
(II.70). Therefore, “the Perfection and Height of Virtue must be owing to the
Belief of a God” (II.77)).
It is important to Shaftesbury’s view that the contemplation of beauty,
including moral beauty, involves a kind of feeling that itself depends on ratio-
nal reflection or “framing” (II.36). Although Shaftesbury is a sharp critic of
the sectarian emotionalism generally called “enthusiasm” in early modern
England, he vigorously defends what he calls “noble Enthusiasm” (III.33). The
former is arational or irrational; the latter, the result of creative imaginative
activity involved in rationally grasping order and design. This is the “Inspira-
tion” he calls “a real feeling of the Divine Presence” (I.53). Inspiration by the
divine is but a heightened version of our capacity to be moved by beauty and
order more generally: “Is there a rational and admired enthusiasm that belongs
to architecture, painting, music, and not to this [the order of the universe]?”
(1900: 33).
When we admire and are inspired by beauty, Shaftesbury holds, the ulti-
mate object of our attitude is the ordering mind beauty manifests. Though the

33
See Grean 1967, chs. 4, 5, and 7, for a helpful discussion of Shaftesbury’s religious and theo-
logical views.
34
Shaftesbury’s categories of atheism and daemonism correspond to Cudworth’s “atheisti-
cal” fatalism and “divine fate immoral,” respectively.

https://doi.org/10.1017/9781139025065.006 Published online by Cambridge University Press

 virtue and self-reflective moral agency 145

occasion for our agreeable experience will be the beautiful object we contem-
plate, what we really admire is the designing mind that formed the object.
“The art then is the Beauty.” “Right.” “And the Art is that which beauti-
fies.” “The same.” So that the Beautifying, not the Beautify’d, is the really
Beautiful’ … “[T]he Beautiful … [is] in the… forming Power.” … What is it
you admire, but Mind, or the Effect of Mind? ’Tis Mind alone which forms
(2002: II.404, 405).

Similarly, the contemplation of beauty involves creative, forming mental pow-

ers. “[N]either can Man by the same Sense or brutish Part, conceive or enjoy
Beauty: But all the Beauty and Good he enjoys is in a nobler way, and by the
help of what is noblest, his Mind and Reason” (II.425). When we contemplate
beauty, we therefore have sympathetic union with the same creative intelli-
gence that formed the beauty we contemplate. We are inspired and moved by
the creative mind we thereby share.

Virtue and Self-Reflective Moral Agency

Recall Shaftesbury’s thesis that although all affections that tend to the public
good are natural or good, only agents who can self-reflectively approve their
own good affections are genuinely virtuous. Why is this? One answer is implicit
in what we have just noticed about the ultimate object of our contemplation of
beauty being beautifying mind. Natural affections are beautiful, but if they are
in no way regulated by the agent’s own moral sense, the mind to which their
beauty must be referred is not the agent’s but the creative mind that is implicit
in all natural order. If that were so, beauty would be in the agent’s mind no dif-
ferently than it would be in the workings of a healthy body. The agent’s mind
would only be the “beautified,” not also “beautifying.” The beauty displayed by
their mind would be in them, but not really theirs. But if they have the capacity
to shape their conduct and affections by their own moral sense, they become
not just a place where beauty is displayed, but a creator of moral beauty them-
selves. The “first Order of Beauty,” Shaftesbury says, has “a power of making
other Forms,” of “forming Forms” (II.405). Virtue, Shaftesbury may be think-
ing, is moral beauty of the “first order.”
This rationale for emphasizing moral autonomy or authorship is rather
different from the one we found in Cudworth and Locke, which is related to
accountability. But although Shaftesbury focus is elsewhere, there is at least one
place where he sounds Locke and Cudworth’s theme.
Neither can any Weakness or Imperfection in the Senses be the occasion
of Iniquity or Wrong; if the Object of the Mind it-self be not at any time
absurdly fram’d, nor any way improper, but sutable, just, and worthy of
the Opinion and Affection apply’d to it. For if we will suppose a Man who,

https://doi.org/10.1017/9781139025065.006 Published online by Cambridge University Press

 146 spinoza, cudworth, shaftesbury, and leibniz

being sound and intire both in his Reason and Affection, has nevertheless so
deprav’d a Constitution or Frame of Body, that the natural Objects are, thro’
his Organs of Sense, as thro’ ill Glasses, falsely convey’d and misrepresented;
’twill be soon observ’d, in such a Person’s case, that since his Failure is not in
his principal or leading Part; he cannot in himself be esteem’d iniquitous, or
unjust (II.32–33).

In the background of this remark, as we shall appreciate better presently, is a

view of self-critical self-determination that is very similar to those we found
in Cudworth and Locke. Whatever direct influence there is of Cudworth and
Locke here, the intellectual source Shaftesbury claims is Epictetus.35 Rational
agents have a “ruling,” “leading,” or “governing principle,” a hegemonikon,
that gives them critical leverage over their opinions and affections (1900: 112,
113, 130).
This is the point of Shaftesbury’s contrast between an agent whose injur-
ing behavior results from defective “Organs of Sense,” and one whose actions
reflect “absurdly fram’d” opinions and affections. In the former case, “since his
Failure is not in his principal or leading Part; he cannot in himself be esteem’d
iniquitous, or unjust” (2001: II.32–33). We cannot attribute the defect properly
to them rather than, say, to their eyes.

The “Obligation” to Virtue

Having determined in Book I of the Inquiry Concerning Virtue or Merit what
virtue is, Shaftesbury writes at the beginning of Book II: “It remains to inquire,
What Obligation there is to Virtue; or what Reason to embrace it” (II.78). As we
noted earlier, in posing the question in this way, Shaftesbury runs right over
Suárez’s distinction between law and obligation, on the one hand, and norma-
tive reasons for acting, or “counsel,” on the other. Hobbes and Cumberland,
who did much to naturalize natural law, still felt the need to work within it.
And even Cudworth, who rejected natural law’s focus on external conduct, was
nonetheless concerned to show that we are accountable for conforming to a
“law of love.” By the time we reach Shaftesbury’s Inquiry (1711), however, the
natural law framework had been so revised from within, and recast from with-
out, that it was possible for Shaftesbury to pose his question in superficially
deontic terms without apparently feeling he had to provide any justification
for doing so. After arguing in Book II that a virtuous life best realizes the virtu-
ous agent’s good, Shaftesbury concludes that there is indeed an “obligation to
virtue,” even, indeed, on the Cartesian supposition, that our life is but a dream,
thereby using “obligation” for only the second time in the book.

35
For references and further discussion of Shaftesbury’s relation to Ralph Cudworth,
Damaris Cudworth Masham, and Locke, see Darwall 1995: 191n.

https://doi.org/10.1017/9781139025065.006 Published online by Cambridge University Press

 authorship and self-determination 147

Although Shaftesbury’s account of virtue stresses other-regarding natural

affections, he is nonetheless a eudaimonist. To argue that there is an obligation
to virtue, in Shaftesbury’s deflationary sense, it is necessary to show that the vir-
tuous life promotes the virtuous agent’s good: “[T]he Judgment we are to make
of Interest, and the Opinion we shou’d have of Advantage and Good … is what
must necessarily determine us in our Conduct, and prove the leading Principle
of our Lives” (2001: I.307). Hutcheson and Butler later criticize Shaftesbury for
his rational egoism, but his eudaimonism was of the sort Irwin calls Aristotelian
naturalism, which held that natural affections and virtue are intrinsically ben-
eficial for the virtuous person. Hutcheson would later hold that others’ goods
are no less capable of motivating rational action than is the agent’s. And Butler
would hold that conscience, a close relative of Shaftesbury’s moral sense, is a
more authoritative source of reasons than either the agent’s good or the good
of others.
The human psyche is a system, and the harmony that moral sense disinter-
estedly approves of is no less essential to the agent’s flourishing. Shaftesbury
argues that a virtuous character is essential for psychic balance, both because
it is the only balanced way of having diverse satisfactions and because it is also
balance enhancing in itself. Love, in some ways a master virtue for Shaftesbury,
is also the “Master-Pleasure”; “out of … Community or Participation in the
Pleasures of others, and Belief of meriting well from others… wou’d arise more
than nineTenths of whatever is enjoy’d in Life” (II.104, 109). We want not only
esteem from others, but “merited Esteem,” Shaftesbury says, sounding a theme
that will loom large also in Adam Smith’s moral philosophy (II.112).
Noting that his argument is cast almost entirely in terms of the internal, psy-
chological benefits of virtues, Shaftesbury concludes by remarking that it sur-
vives even Cartesian skepticism:
Nor is it of any concern to our Argument, how these exterior Objects stand;
whether they are Realitys or mere Illusions; whether we wake or dream. …
In this Dream of Life, therefore, our Demonstrations have the same force;
our Balance and OEconomy hold good, and our Obligation to Virtue is in
every respect the same. (II.173)

Authorship and Self-Determination

When we earlier considered Shaftesbury’s reasons for holding that genuine
virtue requires self-direction through moral sense, I mentioned that he relies
on a Lockean/Cudworthian account of self-determination that he attributes
to Epictetus. Both in Advice to an Author and in his Stoic-inspired notebooks,
Shaftesbury elaborates on this process at some length (2001: I.154–364; 1900).
Though the former is formally addressed to literary authors, Shaftesbury
remarks elsewhere that this is “Pretence”; his real aim is “to correct Manners,

https://doi.org/10.1017/9781139025065.006 Published online by Cambridge University Press

 148 spinoza, cudworth, shaftesbury, and leibniz

and regulate Lives” (2001: III.187). He argues that through the self-reflective
process he calls “soliloquy” (also “self-­converse”), agents can genuinely author
their own conduct and lives.
Cudworth and Locke hold the power of self-critical self-determination to
be necessary for moral agency because they think we could not otherwise be
accountable for acting morally. Shaftesbury also argues that exercising this
power is necessary to be morally good, but his reasons have less to do with
accountability than with the nature of aesthetic creation and authorship, as
well as with the integrity of the self. Even so, Shaftesbury describes a form of
self-determination that involves agents taking something like a juridical or
deontic attitude toward themselves, that is, as a robust form of self-government.
Shaftesburean soliloquy is a distinctive form of self-therapy that involves not
just a reflective grasp of motivations and vivid imagination of their objects, or
even bringing these impressions together to make all-things-considered judg-
ments. In addition, Shaftesbury describes a form of self-analytic reality therapy
in which an agent seeks an objective picture of the sources of their desires and
the desires’ place in the natural order. He agrees with Locke that there is no
direct correlation between the press of present desire and the real contribu-
tion its object would make to the agent’s good. Locke’s strategy is imaginative
examination of the real nature of desires’ objects. This may work so long as the
lack of congruence is the result of misapprehension or distorted salience. But
the existence and strength of desire may bear no simple relation to the agent’s
conception of its object. The desire may have a different motivational source.
One kind of case that concerns Shaftesbury involves motivated ignorance,
where a relatively superficial desire is partly explained by another, deeper
desire that motivates the first, explained also by an ignorance of that moti-
vation. Getting an adequate conception of the superficial desire’s object may
leave the motivational force of the secretly motivating desire unaffected. To
gain leverage on this, an agent must somehow bring the relation to the deeper
desire to consciousness.
The very capacity of rational agents to be aware of their own desires may
require that deeper motivating desires be “sly” and “insinuating” (I.173). If an
agent would be less likely to act on a desire were they to be aware of having it,
then the desire may motivate a more surreptitious pursuit of its object, moti-
vating superficial desires the satisfaction of which will also tend to satisfy the
deeper desire and mask its operation at the same time.
“One would think,” Shaftesbury writes, “there was nothing easier for us,
than to know our own Minds, and understand … what we plainly drove at,
and what we propos’d to our-selves, as our End, in every Occurrence of our
Lives” (I.171). But this is not so. Ignorance of some of our desires is motivated
by those desires. “[T]he chief Interest of Ambition, Avarice, Corruption, and
every sly insinuating Vice, is to prevent this Interview” of ourselves and the

https://doi.org/10.1017/9781139025065.006 Published online by Cambridge University Press

 authorship and self-determination 149

consequent discovery of the vices (I.173). Vicious desires are likely to express
themselves “tacitly and murmuringly,” “by a whisper and indirect insinuation,
imperfectly, indistinctly, and confusedly” (1900: 166). Vanity, for instance,
may motivate a desire for political office out of proportion to the attraction
this would otherwise have and in such a way that the desire’s real motivational
source remains hidden.
A different kind of case is the sort of false enthusiasm or emotional con-
tagion that Shaftesbury sarcastically calls the “inspiring Disease” (2001: I.45).
Here one catches a feeling or passion neither by responding to its object, nor
by a desire to emulate, but by “insensible Transpiration” (I.45). Here again, a
desire’s strength will be relatively independent of one’s conception of its object
and may be relatively immune to having a more accurate conception.
Cases of these kinds require a form of self-analysis through which agents
bring their desires and their mutual relations to light and realistically exam-
ine them. So long as an agent views things only from the perspectives of their
desires and emotions, the agent is in the grip of those desires and emotions.
Avowing a desire is already a step away from being possessed by it. One now
occupies a different standpoint (not from the desires, but on the desire). The
desiring view of things can be seen for what it is, a view of things one has by
virtue of having the desire.
Shaftesbury also calls soliloquy “self-converse” (I.166), because it involves
both giving desires voice and addressing oneself and them from a more realistic
perspective. Listening to one’s own unconstrained soliloquizing can enable one
to identify desires that would otherwise have remained hidden. Moreover, pri-
vacy disarms the desire to please, enabling one to be more honest with oneself.
“There is nothing in the management of the visa,” Shaftesbury writes, “than
to have a sort of custom of putting them into words, making them speak out
and explain themselves as it were viva-voce, and not tacitly and murmuringly”
(1900: 166).36 As “our own Subjects of Practice” we can “divide our-selves into
two Partys” and make our thoughts and feelings the objects of our own aware-
ness and consideration (2001: I.169).
The agent can then interrogate the source of their desires:
Thus I contend with Fancy and Opinion; and search the Mint and Foundery
of Imagination. For here the Appetites and Desires are fabricated. Hence
they derive their Privilege and Currency. If I can stop the Mischief here,
and prevent false coinage; I am safe (I.320).

Desires aim to take over the agent’s practical perspective and, in effect, to
impersonate the agent. Shaftesbury’s “advice to authors,” therefore, is that
they directly address their desires in a way that disarms their motivational

36
Cf. Cudworth on determining the last “visum” or practical judgment before acting.

https://doi.org/10.1017/9781139025065.006 Published online by Cambridge University Press

 150 spinoza, cudworth, shaftesbury, and leibniz

leverage. “Idea! wait a-while till I have examin’d thee, whence thou art, and
to whom thou retain’st. Art thou of Ambition’s Train? Or dost thou promise
only Pleasure” (I.320). When a desire’s strength depends partly on ignorance
of its source, appreciating its etiology can help an agent affect its strength in the
direction of their overall good.
After this scrutinizing of desires and their causes, the agent must still
“recollect” themselves and place present desires in the framework of their most
important concerns over time. “Thus at last a Mind, by knowing it-self, and
its own proper Powers and Virtues, becomes free, and independent. It sees its
Hindrances and Obstructions, and finds they are wholly from it-self, and from
Opinions wrong-conceived. The more it conquers in this respect … the more it
is its own Master” (III.204).
Although Shaftesbury shares Locke’s rational egoism, he rejects his hedo-
nism. To the contrary, the more reflective half of Shaftesbury’s divided self is
a better or nobler self who represents an ideal of character (given by their own
moral sense) to which they are committed and on which the goodness of their
life is staked.
And thus as long as I find Men either angry or revengeful, proud or asham'd,
I am safe: For they conceive an Honourable and Dishonourable, a Foul and
Fair, as well as I (II.421).

Shaftesburean self-converse involves responsibility to a personification of an

ideal of character to which the agent is committed. The agent views themselves
from the perspective of someone who realizes, or respectably pursues, their
character ideal and, identifying with their better self, judges what course they
should take, taking account of the higher values on which their self-esteem is
staked. Consciousness of being a scoundrel makes life miserable, but that is
because one thinks villainy base, not vice versa. “[W]e can never really blush
for anything beside what we think truly Shameful, and what we shou’d still
blush for, were we ever so secure as to our Interest” (II.419).
More clearly than Locke, and even more than Cudworth, Shaftesbury con-
ceives of self-critical self-determination in juridical terms as self-­government.
This is really the only place where deontic notions play an important role in
his thought. “How do I govern myself?” “This,” he writes in his notebooks, “is
a matter, and the only matter” (1900: 102). For Shaftesbury, soliloquy involves
recognizing and subjecting ourselves to the authoritative prescriptions of our
better selves, our “ruling principle.” In dividing ourselves into two, we give
one perspective authority as “our Counsellor and Governor” (2001: I. 169).
“As cruel a Court as the Inquisition appears; there must, it seems, be full as
formidable a one, erected in our-selves” (I.186). There must be a “superior
part which disciplines, instructs, and manages,” one with “authority and com-
mand” (1900: 169).

https://doi.org/10.1017/9781139025065.006 Published online by Cambridge University Press

 ancient or modern? 151

Shaftesbury connects the capacity for self-government to the ability to unify

and give integrity to a life and to agency over time. Through self-governing
soliloquy, we “gain … a Will” (2001: I.187). Only in this way can an agent “be
sure of his own Meaning and Design; and as to all his Desires, Opinions, and
Inclinations, be warranted one and the same Person to day as yesterday, and to
morrow and to day” (I.187). Identifying soliloquy with philosophy itself, Shaft-
esbury writes, “[T]is the known Province of Philosophy to teach us our-selves,
keep us the self-same Persons, and so regulate our governing Fancys, Passions,
and Humours, as to make us comprehensible to our selves” (I.283).

Ancient or Modern?
Shaftesbury does not fit easily into Sidgwick’s categories. On the one hand,
his fundamental philosophical commitments conform to Irwin’s Aristotelian
naturalist model: (i) eudaimonism about practical reason, (ii) the claim that
virtue is an intrinsic good (for the virtuous agent), and (iii) the thesis that this
good realizes the agent’s nature. Furthermore, Shaftesbury steadfastly rejects
law-based ethical conceptions, especially in prominent voluntarist formula-
tions, and defends an ethics of virtue. On the other hand, there are elements
of Shaftesbury’s thought that lend themselves to development in more distinc-
tively modern ways by philosophers who follow him. The clearest example is
Shaftesbury’s “moral sense,” which Hutcheson and Hume will work out in an
empiricist framework that is consonant with the emerging empirical sciences.
In Hutcheson’s hands, a Shaftesburean ethics of virtue takes a distinctively
modern form with moral goodness being determined entirely by disinter-
ested approval from an impartial perspective completely independently of the
agent’s flourishing. And Shaftesbury’s focus on other-regarding virtues of nat-
ural benevolence and love will help inspire Hutcheson’s repudiation of ancient
eudaimonism; Hutcheson will argue that benevolence is no less an ultimate
source of motivation than self-love.37
Finally, Shaftesbury’s focus on authorship and autonomy and his doctrine
that genuine virtue must involve the moral agent’s determining themselves
by their own moral judgment will also loom large in the modern period, for
example, in Butler’s doctrine of the authority of conscience and Kant’s idea that
genuine moral worth must be self-reflectively moral. It is notable in this con-
nection that the one place where deontic notions figure prominently in Shaft-
esbury’s thought is in his view of autonomy as self-government through which
an agent holds themselves to be obligated by the judgments of their better self.
Here again we see ideas that will be taken in a more explicitly modern direction
by Butler and Kant. In these ways Shaftesbury is, like Cudworth, a bridge figure.

37
Another influence on Hutcheson is Cumberland. On this, see Chapter 4.

https://doi.org/10.1017/9781139025065.006 Published online by Cambridge University Press

 152 spinoza, cudworth, shaftesbury, and leibniz

Leibniz
Leibniz is a bridge figure also. Like Shaftesbury, Leibniz sounds classical (Aris-
totelian) naturalist themes, though with a more systematic metaphysical under-
pinning. Both are philosophical optimists, holding that reality is a harmonious
whole. But Leibniz’s theology is more orthodox and philosophically elaborated
than Shaftesbury’s. Leibniz’s God is a transcendent being who creates “the best
of all possible worlds” (Leibniz 1985: 228). And Leibniz’s theodicy is rooted
in explicit doctrines concerning the natures of being and perfection, the prin-
ciple of sufficient reason, the identity of indiscernibles, and the analyticity of all
truth. Leibniz is concerned less with edification, and more with working out a
comprehensive rationalist metaphysics that can fit with modern science.38
Leibniz was also a harmonizer by philosophical temperament. He sought
to bridge differences between Protestants and Catholics, and although he was
definitely sensitive to conceptual differences between the good and the right,
he attempted to demonstrate fundamental coincidences between them even so.
Schneewind writes that Leibniz’s conception of morality is “centered on love,”
with “law and obligation hav[ing] a vanishingly small part to play” (Schneewind
1998: 246). This is correct in the main, but deontic ideas nonetheless enter into
Leibniz’s thought in surprising and interesting ways, as we shall see.
Like Cudworth and Shaftesbury, Leibniz is a sharp critic of all forms of vol-
untarism, both secular and theological. But whereas Cudworth and Shaftes-
bury focus their criticisms on Hobbes and Locke, respectively, Leibniz’s main
opponent is Pufendorf. Also, like Cudworth, Leibniz holds that God’s loving
nature is a perfectly virtuous exemplar we are to emulate. More, Leibniz’s
official view is that moral obligation or “moral necessity” derives from this.
“Moral” in “moral necessity,” he says, means “that which is equivalent to ‘natu-
ral’ for a good man” (Leibniz 1988: 171). In other words, what a perfectly virtu-
ous being would do is what we imperfect beings morally should do. To say that
something is obligatory on this definition is not to say anything intrinsically
deontic or juridical. It is rather to say that it is what a perfectly virtuous exem-
plar would do as a matter of natural necessity. Like Shaftesbury’s, this appears
to be a revisionary view of obligation.39
Leibniz seeks also to understand justice in terms of the universal benevo-
lence of the ideally virtuous, apparently reducing justice to benevolence, and so
the right to the good. Justice is, he says, “the charity of the wise” (171). Because
“the good man is benevolent toward all,” “the highest rule of right is that every-
thing is directed to the greatest general good or common felicity” (quoted in

38
It should be remembered not only that Leibniz helped invent the calculus, but also that he
thought his dynamic teleological doctrines of monads provided a necessary foundation for
the emerging natural philosophy (science).
39
In Chapter 8 I shall argue that, remarkably, Kant also holds a version of this same doctrine.

https://doi.org/10.1017/9781139025065.006 Published online by Cambridge University Press

 leibniz 153

Johns 2013: 83). At one point, indeed, Leibniz formulates this in terms that will
be taken up by Hutcheson and later utilitarians as the greatest happiness prin-
ciple: “to act in accordance with supreme reason is to act in such a manner that
the greatest quantity of good available is obtained for the greatest multitude
possible and that as much felicity is diffused as the reason of things can bear”
(Hruschka 1991: 166).40
But this apparent reduction of the right to the good is deceiving. Some of
Leibniz’s earliest writings concern jurisprudence and the “science of right”
(Johns 2013).41 And various arguments and observations Leibniz makes con-
cerning his theory of justice as rationally regulated benevolence also suggest
that he takes conceptual connections between the right and legitimate demands
to underlie and affect the interpretation of his theory. As an example, Leibniz
holds that justice (conceptually) concerns acting in ways to which no one has
a legitimate complaint. “Everyone will perhaps agree, perhaps, to this nominal
definition, that justice is a constant will to act in such a way that no one has
a reason to complain of us” (Leibniz 1988: 53).42 Leibniz parlays this concep-
tual claim into a fundamental principle of equity or reciprocity that it is always
unjust not to provide others what one would ask or demand from them. Any
case for anything approaching utilitarianism on this basis is less on the grounds
of equal love than on the grounds of equal respect for all persons as having a
common authority to make claims and demands of one another.
Another aspect of Leibniz’s importance for the moral philosophy that will
follow him is his significance for Kant. Much of Kant’s critical or transcen-
dental turn in moral philosophy, similar to that in the Critique of Pure Reason,
came from his rejection of Wolffian rationalism, which was close to Leibniz’s.
Kant was especially critical of Leibniz’s compatibilist view of freedom, which
he disdainfully called “the freedom of a turnspit” (Kant 1996b: 5:97). Despite
this, the details of Kant’s account of moral obligation will mirror Leibniz’s idea
of moral necessity. For Kant, as for Leibniz, obligation derives from laws of the
nature of a perfect agent (in Kant’s terms, a “holy will”). The major difference
concerns their respective views of what such laws consist in. For Leibniz, they
are perfectionist standards that at once unify the agent’s good, justice, and the
good of all. For Kant, they are “laws of freedom.” This is the signature Kantian
doctrine of “autonomy”: “the property of the will by which it is a law to itself
(independently of any property of the objects of volition)” (Kant 1996a: 440).
This difference notwithstanding, however, Kant’s moral philosophy, often

40
According to Hruschka 1991, Hutcheson gives his own translation of this passage of
Leibniz’s as stating his own view in Hutcheson 2004: 125.
41
New Method for Learning and Teaching Jurisprudence was written in 1667. Johns 2013 con-
tains a translation and analysis of this work from which I shall quote below.
42
From Meditation on the Common Concept of Justice, written in 1702–1703.

https://doi.org/10.1017/9781139025065.006 Published online by Cambridge University Press

 154 spinoza, cudworth, shaftesbury, and leibniz

taken to be a paradigm of deontological ethics, turns out to retain Leibnizian

elements that require significant alteration in this common view.
As we noted above, we must complicate the kind of characterization of Leib-
niz’s ethics that Schneewind gives. To be sure, the most prominent elements of
Leibniz’s thought are a good-based ethics of virtue with justice and obligation
being ultimately either reduced to or identified with God’s perfectly virtuous
universal benevolence. However, as I have indicated, there are also irreducibly
deontic elements that resist such an analysis that significantly complicate the
picture.

Perfectionism: Identifying Goods (Personal and

Impersonal) and Justice
Let us begin, though, with the better-known elements of Leibniz’s thought.
Most fundamentally, Leibniz’s ethics are, like Shaftesbury’s, an ethics of vir-
tue. Against the natural lawyers’ focus on conduct or “external acts,” Leibniz
objects that moral goodness concerns the agent’s motives (Leibniz 1988: 73).
Moreover, Leibniz treats justice primarily as a virtue, holding even that the jus-
tice of actions depends on motivation.
[H]e who acts well, not out of hope or fear, but by an inclination of his soul,
is so far from not behaving justly that, on the contrary, he acts more justly
than all others, imitating, in a certain way, as a man, divine justice (72).

Against voluntarists like Pufendorf, Leibniz argues that their theory makes it
impossible to say that God himself is good and just, since goodness and jus-
tice could not possibly themselves derive simply from God’s commands. It is
rather, Leibniz argues, God’s benevolent goodness that underwrites his author-
ity and therefore the legitimacy of his commands. Of course, neither Pufendorf
nor Locke denies that God’s goodness is independent of his command. Their
position is that without his actual commands there is no deontic morality,
nothing that genuinely obligates moral agents.
Leibniz might argue, however, that they effectively credit God’s goodness as
the source of his authority of command in arguing, as at least Pufendorf seems
to, that God’s authority derives from something like a debt of gratitude. If it is
ultimately because of God’s goodness to us that we owe him obedience, then his
authority is ultimately rooted in his benevolence.
Not only goodness, however, but justice itself is “an essential attribute of
God” (71). So the standard of justice cannot derive from God’s “free will.”
When God acts or wills justly, there must be reasons – reasons of justice – that
weigh with him and make his resulting will and actions just. Leibniz concludes
that for us to act justly, we should act on these reasons also and thereby “imi-
tate divine justice” (72). Indeed, it appears to follow that we would have these

https://doi.org/10.1017/9781139025065.006 Published online by Cambridge University Press

 perfectionism 155

reasons even if God did not exist, though there are complications here that we
must consider below.
When he comes to discuss obligation, Leibniz takes over Grotius and Pufen-
dorf’s term “moral necessity,” but gives it a subversive redefinition that seems
to rob it of any distinctively deontic normativity (Grotius 2005: I, 150–151;
Pufendorf 1934: 20). Whereas moral necessity for Grotius and Pufendorf is tied
to accountability for compliance with legitimate demands, Leibniz identifies
moral necessity with a necessity of the virtuous person’s nature. “Right is a kind
of moral possibility, and obligation is a moral necessity. By moral I mean that
which is equivalent to ‘natural’ for a good man” (Leibniz 1988: 171). Leibniz uses
“right” and “morally possible,” as we shall see presently, to mean actions that
do no wrong (are morally permissible) and that wrong no one (do not violate
anyone’s rights). And clearly he intends moral necessity to entail wrongness,
that is, being morally impermissible, or wrong not to do. Indeed, Leibniz works
out a deontic logic, which he calls “modes of right,” in his Elements of Natural
Right (1669–1671), which makes these entailments explicit (Johns 2013: 48–56).
What I want to focus on now, however, is the way Leibniz reduces moral
possibility (also “moral power” in his earlier text on the “science of right”) and
moral necessity to what is compossible with, and what is necessary to, God’s
wise and benevolent nature, respectively. Plainly there is nothing essentially
deontic or juridical in the ideas of perfect wisdom and perfect benevolence.
“A good man,” Leibniz says, “is one who loves everybody, in so far as reason
permits” (Leibniz 1988: 171). Leibniz follows this with his famous definition of
justice as “the charity of the wise man” (1988: 171). If there is any doubt about
Leibniz’s meaning, it is cleared up in the following passage: “moral necessity …
is always followed by its effect in the wise. This kind of necessity is happy and
desirable” (1985: 395). Moral necessity is natural necessity, that is, nomological
necessity, of the wise person’s nature. It is what such a person invariably does
or would do. And what they would do is what we imperfect beings should do,
what we are, Leibniz says, obligated to do.
Similarly, “right,” for Leibniz, has the meaning of “not wrong.” And some-
thing is not wrong, hence right, if it is not contrary to the wise person’s nature,
if it is something the wise person might do, consistently with their nature. If an
action is something a wise person might do, if it is compossible with their nature,
then it is something imperfect beings may do. Like Shaftesbury’s implicit defi-
nition of obligation in terms of normative reasons, these are revisionary defini-
tions of obligation and right, respectively. There is nothing essentially deontic
or juridical in them, nothing that puts them on the law or demand, as opposed
to on the counsel, side of the Suárezian ledger.
Moreover, like Aquinas’s classical natural law view, Leibniz’s position
requires a metaphysical guarantee of harmonies and coincidences, both within
morality and between morality and self-interest. Within morality, Leibniz’s

https://doi.org/10.1017/9781139025065.006 Published online by Cambridge University Press

 156 spinoza, cudworth, shaftesbury, and leibniz

perfectionism leads him ultimately to identify all of the following: justice,

enlightened benevolence, wisdom, virtue, and happiness. (We’re beginning to
get into Spinoza territory here. What we would need to get all the way there
would be to be identify all of these with reality itself.)
First, again, justice is “the charity of the wise,” so whatever an enlightened
love of all leads to is just. Subject to some caveats we shall note presently, this
guarantees a connection between justice and overall good. Indeed, even in
his early jurisprudential writings, Leibniz identifies justice and injustice with
“what is useful or harmful to the public,” respectively (Johns 2013: 153).
But neither are wisdom and benevolence really distinct, for Leibniz. Every
being’s good is its perfection, and though the good life is pleasurable, that is
not because pleasure is intrinsically good. Rather, “[p]leasure is a knowledge
or feeling of perfection, not only in ourselves, but also in others” (Leibniz 1988:
83). This assures that knowledge of (the prospect of) good (perfection) in any-
one will lead to a desire for it for its own sake. Indeed, it apparently follows that
this knowledge itself constitutes either pleasure in or a desire for the good of
others, depending on whether the latter is actual or envisioned as a possibility
in prospect. Since wisdom includes knowledge of perfection, virtue or justice
can also be described as “the habit of acting according to wisdom” (83).
Leibniz’s identification of moral goodness and justice with universal benev-
olence led him to assert a version of the greatest happiness principle, for per-
haps the first time in the history of ethical thought: “To act in accordance with
supreme reason is to act in such a manner that the greatest quantity of good
available is obtained for the greatest multitude possible and that as much felic-
ity is diffused as the reason of things can bear.”43 The utilitarians would reject
Leibniz’s perfectionism, but formally draw the same maximizing consequences
from equal concern for the good of all.
Despite Leibniz’s stress on benevolence, however, he is ultimately a
eudaimonist for whom genuine normative reasons must be grounded in the
agent’s own good. But his perfectionism and views about the relation between
knowledge of perfection and pleasure and desire assure that there cannot be any
deep conflict between the agent’s goodness and the good of all. This is confirmed
in a criticism Leibniz makes of Grotius’s response to Carneades’s challenge that
justice is foolish since it is “folly to care for another at loss to oneself” (quoted
in Johns 2013: 28). Grotius, recall, argues that justice is grounded in sociability
and not in the agent’s own good, however broadly construed. To this Leibniz
replies that acting against one’s own good is indeed foolish, and that the proper

43
This comes from a review Leibniz published anonymously, “Observationes de Principio
Juris,” published in 1700 in Monathlicher Auszug/Aus Allerhand Neu-herausgegebenen/
Nützlichen und Artigen Büchern (Hanover: 1700), p. 378. It is quoted (and translated) in
Hruschka 1991: 166. Hruschka argues that this is the first formulation of the greatest happi-
ness principle.

https://doi.org/10.1017/9781139025065.006 Published online by Cambridge University Press

 freedom, agency, and necessity 157

response to the Carneadean challenge is the Ciceronian one that justice realizes
an essential intrinsic benefit (what Cicero calls bonum honestum) even when it
is disadvantageous (not instrumentally beneficial or utile).44 Knowledge that
an action will bring about others’ good is pleasurable in itself.
Despite these coincidences between justice, overall good, and the agent’s
good, Leibniz nonetheless argues that not everyone is sufficiently wise to appre-
ciate them. “If God did not exist” and did not reward and punish virtue and
vice, respectively, Leibniz writes, even “wise men would have no more cause to
be benevolent than such as would be required for their own welfare.”45 But no
problem arises for the wise, since they appreciate the connections between jus-
tice, public good, and the agent’s good that Leibniz draws. Eternal sanctions are
necessary, however, for the rest of us who truly believe eudaimonism (that our
own good is the only source of normative practical reasons) but fail to appreci-
ate Leibnizian perfectionist connections. God thereby “ensures that every good
act will be beneficial, and every bad one harmful, to the agent. So that not even
a man who endures torture and death for the public good can be regarded as
an idiot” (Hostler 1975: 59). God assures a harmony between even the appar-
ent eternal interests of all members of the “Universal Republic of Spirits” or
“Kingdom of Grace,” guaranteeing for each a coincidence between virtue and
good (17).

Freedom, Agency, and Necessity

If only because they were to constitute such an important point of departure
for Kant, it is important to say something about Leibniz’s views on free agency
and necessity and how these interact with his ethical claims. Leibniz famously
distinguishes between metaphysical necessity and moral necessity. Something
is metaphysically necessary if it is something that must take place “because the
opposite implies a contradiction” (Johns 2013: 129). A moral necessity, by con-
trast, is one “whereby a wise being chooses the best, and every mind follows the
strongest inclination.”46
God creates the best of all possible worlds and does so by moral necessity.
God, like any agent, does what seems best to him, and God, in his wisdom, sees,
and is thereby moved to create, the best world. But there were other possible
worlds in the sense that their existence would in themselves imply no contra-
diction. A contradiction only arises when we add in their having to have been
created by God.

44
See Johns 2013: 28. For a discussion of Cicero on this point, see Irwin 2007: 600.
45
Translated and quoted in Hostler 1975: 59.
46
These passages come from Leibniz’s correspondence with Clarke. They are quoted in Johns
2013: 56. Leibniz accuses Clarke of failing to make this distinction.

https://doi.org/10.1017/9781139025065.006 Published online by Cambridge University Press

 158 spinoza, cudworth, shaftesbury, and leibniz

If this is the best of all possible worlds, the question might arise, what rea-
son can we have to do one thing rather than another? It is a central aspect
of Leibniz’s metaphysics, however, that all substances act teleologically with
some form of perception and appetite. There is a sense, therefore, in which all
substances act for reasons coming from their apprehension, often confused,
unconscious, and subrational, of the good. All activity, whether God’s, our
imperfect rational activity, or the activity of subrational substances aims at
the good. The best of all possible worlds is thus composed of substances that
aim at their respective goods. It is therefore essential to the world as we find it
that rational substances face the question of what is the greatest good they can
achieve for themselves.
Leibniz is a compatibilist. All substances aim at the good, and their choices
are necessitated by their nature through their perception, however confused, of
their goods. But Leibniz holds that this necessitation in no way impedes free-
dom. Liberty or freedom, for Leibniz, is a power of action. And action is, in its
nature, aimed at the good. So genuinely free action is action where this aim suc-
ceeds, where we are determined by perceptions and appetites that successfully
lead us to genuine good. In this way, moral necessity is thoroughly compatible
with genuine freedom or liberty.
Leibniz denies not just the liberty of indifference, that also that there exists
or need be anything like a faculty of will that can choose arbitrarily. Locke and
Cudworth deny these, but they hold that in addition to beliefs and desires, gen-
uinely free agents also have a self-reflective capacity to direct their attention –
as Locke puts it, “exerting that dominion [the mind] takes itself to have over
any part of a man” (Locke 1975: II.21.15). According to Leibniz, all beings act for
the good; rational beings differ from subrational ones only in being determined
by less confused and more accurate perceptions of it.

Right and Claim

To this point, I have been emphasizing the non-deontic character of Leibniz’s
ethics, both his eudaimonism and his apparent reduction of justice to universal
benevolence. But this is only part of the story, albeit its dominant theme. There
are also irreducibly deontic elements in Leibniz’s thought. These are most nota-
ble in his early writings on jurisprudence, both in his “science of right” (New
Method of Learning and Teaching Jurisprudence) and in his Elements of Natural
Right. Just to give a sense of the deontic theme, consider the following passage
from the Elements:
The just is not accurately enough defined as the public useful, since it may
be permitted to prefer the ruin of many to my own ruin. … The just is the
balanced relationship between the love of myself and the love of neighbor
(quoted in Johns 2013: 32).

https://doi.org/10.1017/9781139025065.006 Published online by Cambridge University Press

 right and claim 159

From the perspective of universal benevolence, the ruin of many is clearly worse
than an individual’s ruin, as Leibniz himself here acknowledges. Nevertheless,
it is not unjust, Leibniz is saying, for an agent to act to protect themselves even
at greater universal cost. Justice is the “balanced relationship” between self-
love and the love of others. And Leibniz seems to be ruling out the possibility
that this balance is to be determined in terms of universal good, that is, in terms
of what maximizes benefit overall.
If this is indeed Leibniz’s view, it follows that he cannot consistently hold
that the deontic ideas of right, obligation, and the like – those on the law side of
the Suárezian divide – can be reduced to those of good – those on the norma-
tive reasons or counsel side. We get some more indication of how Leibniz must
be thinking in his “science of right.”
We can begin with his definition of right and obligation. As Christopher
Johns points out, Leibniz writes in later notes, “Right and obligation are not
to be treated separately. They constitute one relation composed from both.”47
This suggests something like the kind of correlativity claim associated with
Hohfeld and Michael Thompson (Hohfeld 1923; Thompson 2004). Obligations
that are owed to someone entail the latter’s claim right to the obligation’s being
fulfilled. There is really just a single deontic/juridical relation viewed from two
different “poles,” as Thompson puts it (Thompson 2004: 335).
When, however, we look at Leibniz’s initial definitions of right and obliga-
tion in this early work from the perspective of his later definitions, this sugges-
tion may seem dispelled. “The morality,” Leibniz writes, “that is, the justice or
injustice of the act,” “springs from” a “moral quality” of the person (Johns 2013:
154). This quality is “twofold, namely, the power to act and the necessity to act;
therefore, moral power is called right, and moral necessity is called obligation”
(154). Recall now from Leibniz’s deontic logic that “right” must have the force
of “all right” or not wrong, as opposed to right in the sense of obligatory. After
all, Leibniz is here distinguishing right from obligation.
But when we take this together with Leibniz’s later accounts of justice and
moral necessity as “the charity of the wise,” we apparently get accounts that
lack distinctively deontic/juridical entailments. Being right, or not unjust, sim-
ply means being compossible with God’s wise benevolence, something that
deliberation from the perspective of equal concern for all would not rule out.
Similarly, being obligatory or morally necessary means being an act that these
(good-based) grounds of action converge on.
Indeed, even if we read “right” as expressing an irreducibly deontic relation,
being not unjust is only a Hohfeldian liberty right (that one does no wrong or
does no one a wrong), not a claim right that would entail a correlative obli-
gation. This, recall, was the difference between Hobbes’s right of nature and

47
From “Notae in Tabulam Jurisprudentiae” (1696). Quoted in Johns 2013: 9.

https://doi.org/10.1017/9781139025065.006 Published online by Cambridge University Press

 160 spinoza, cudworth, shaftesbury, and leibniz

the natural rights that Pufendorf recognizes (and consequently, the source of
Pufendorf’s objections to Hobbes). A liberty right consists just in the absence
of obligation to act otherwise; it entails no obligations to others like the correla-
tives of claim rights.
Two sections later, however, Leibniz writes:
The causes of right in one person are a kind of loss of right in another; that
is, the second person has an obligation to the first. Conversely, acquiring
an obligation from another is the cause of recuperation of right (156).

This is, of course, precisely the correlativity point about directed (bipolar)
obligations and claim rights. Moreover, Leibniz sketches out on this basis a
jurisprudence that bears a remarkable resemblance to Kant’s Doctrine of Right.
Each person has a fundamental right of freedom, which includes a right to the
body. Everyone’s right of freedom imposes correlative obligations on others
and is constrained by their similar rights. Moreover, this “right in persons is
called right to coercion,” justifying the use of force in its defense (154).
Obviously, this line of thought requires an irreducibly deontic/juridi-
cal framework. There is simply no way to get the relevant entailments to go
through on Leibniz’s official good-based definitions of justice, right, and obli-
gation. Moreover, Leibniz derives from his fundamental moral qualities three
deontic “precepts” of decreasing stringency: (i) “strict right”: do no harm, (ii)
“equity”: give everyone their due, and (iii) “piety”: live honorably and respect
authority.
We find similarly irreducibly deontic thoughts in Leibniz’s Elements of Nat-
ural Right where Leibniz elaborates the precept of equity in terms of reciprocal
claims or demands: “The equitable is to grant to the other as much as one would
demand from another” (32). Since it is formulated in terms of demands, this is a
deontic formulation of the Golden Rule. The root idea is that it is a fundamen-
tal violation of justice as reciprocity to deny others not what one might hope
or be pleased by from them, but what one would claim or demand from them.
We find corollary ideas in Leibniz’s Meditation on the Common Concept of
Justice.

Everyone will agree, perhaps, to this nominal definition, that justice is a

constant will to act in such a way that no one has a reason to complain of
us (Leibniz 1988: 53).

There is an obvious affinity between this formulation and that concerning

reciprocal demands from the Elements. Whatever can justify complaint can
justify a demand, and vice versa.
In an especially interesting series of steps, beginning with the obligation
of strict right – to do no harm – Leibniz then justifies increasingly expansive
duties in terms of the Golden Rule, expressed, as above, in terms of reciprocal

https://doi.org/10.1017/9781139025065.006 Published online by Cambridge University Press

 right and claim 161

claims and demands. The first concerns helping someone avoid an evil when
one can do so at little cost.

Let him ask himself what he would say and hope for if he should find him-
self actually on the point of falling into an evil, which another could make
him avoid by a turn of his hand (54).

Leibniz assumes his readers will easily agree that refusing such help would jus-
tify complaint, which one can check by applying the Golden Rule formulated in
terms of demand or complaint to oneself.
Next Leibniz considers a case where someone cannot prevent evil, but can
relieve it at little cost. Suppose one is drowning and another will not throw one
a rope, as they could easily do. Would not one, would not anyone, complain?
And next, a case where another is in a position to help one acquire a good.

A great good is going to come to you; an impediment appears; I can remove

that impediment without pain: would you not believe yourself to have a
right to ask it of me, and to remind me that I would ask it of you, if I were
in a similar position? (55)

Leibniz’s final step is to pose a case in which he asks, “[H]ow will you refuse the
only remaining request, that is to secure a great good for me, when you can do
it without inconveniencing yourself in any way?” and can only give as a reason
for refusal “a simple ‘I don’t want to’.” (55). In all these cases, others would have
“reasons of complaint.”
Leibniz concludes with a “principle of equity,” which holds “that one
should grant [to others] whatever one would wish in a similar situation, with-
out claiming to be privileged … or to be able to allege one’s will as a reason”
(56). It is clear from the context, that by “wish,” Leibniz means something
that is naturally addressed to its object, like a request, claim, or demand,
that is, not an idle wish, or even an ardent desire that one would not impose
on others. The basic idea is that equity is rooted in the idea that any claims
we make on others must be grounded in a common basic authority to make
claims of one another. So understood, equity is a fundamentally deontic or
juridical idea, squarely on the law or command side of the Suárezian idea.
Indeed, it seems closely related to Grotius’s claim that sociability is the “foun-
tain of right.”
How should we relate the deontic themes we have just been discussing to
Leibniz’s identification of justice with the charity of the wise? And to his related
support for something like the principle of utility or the “greatest happiness”?
If Leibniz aims to reduce justice to universal good as a metaethical thesis,
holding that the good is the only fundamental notion of which justice can be
understood as a “species,” to use Sidgwick’s terms, then these deontological
claims apparently conflict with that. On the other hand, if Leibniz is making the

https://doi.org/10.1017/9781139025065.006 Published online by Cambridge University Press

 162 spinoza, cudworth, shaftesbury, and leibniz

substantive normative claim that promoting overall good is the fundamental

just-making feature, there may be no deep conflict.
This is brought out in Leibniz’s response to an objection that a judge could
never justly sentence a criminal because they would not want to be sentenced
were they in the criminal’s shoes.
The reply is easy: the judge must put himself not only in the place of the
criminal, but also in that of others who are interested that the crime be
punished. And the balance of good (in which the lesser evil is included)
must determine it (1988: 46).

As Irwin points out, reasoning of this kind would seem to be able to justify
the kinds of trade-offs for which utilitarianism is famous (or infamous) (Irwin
2008: 319). Indeed, it appears to be an argument of the very same kind that R.
M. Hare will advance in the late twentieth century for act utilitarianism as an
ideal morality (Hare 1982). The important thing to realize, however, is that if an
argument like this one of Leibniz’s ends up supporting utilitarianism, it does
so on very different grounds than that this is what is recommended from the
standpoint of universal benevolence. It derives it from a principle of univer-
sal respect for individuals’ authority or standing to make claims and demands.
Indeed, Leibniz says, “there is in justice a certain respect for the good of others”
(Leibniz 1969: 136).
It is important to be clear that the ethical position that, following Bentham
and Mill, we call utilitarianism is a normative ethical position, and, more par-
ticularly, that it is a normative thesis about the right. It is the view that the
determinant of whether an action is morally right or not is its relation (direct
or indirect) to promoting universal good or happiness. It is not simply a claim
about the good.
Historically, there have been three different ways of supporting this norma-
tive deontic view. One has been to anchor it in universal benevolence or care,
as Hutcheson does. A second is to argue that from an impartial or impersonal
point of view, everyone’s good or happiness matters equally. Sidgwick explic-
itly does this with respect to good, and early chapters of Mill’s Utilitarianism
might be read as implicitly doing it with respect to happiness. And a third is
to derive utilitarianism from universal respect. As I see it, economic forms of
utilitarianism that give equal weight to preferences (“consumer sovereignty”)
implicitly invoke universal respect in this way.
Respect and benevolence or care are different attitudes (Darwall 2006: 126–
130). Benevolence and care involve desires for someone’s good or happiness,
however that might come about. (Recognition) respect, on the other hand, is
concerned fundamentally with how we conduct ourselves toward someone.
Whereas benevolence is targeted on the good, respect is focused on the right –
on doing right by, or justice to, its object.

https://doi.org/10.1017/9781139025065.006 Published online by Cambridge University Press

 right and claim 163

Leibniz’s ideas seem to hover between these. On the one hand, he lays out an
ethics of virtue with his leading virtues being forms of love and benevolence.
On the other, he discusses fundamental relations of right, and some of his argu-
ments for universal beneficence (if not benevolence) are rooted in a principle
of equity or equal respect. Finally, there are places, as we noted above, where
Leibniz seems to pull away from the kind of impartial weighing of interests that
universal beneficence requires. If I am “permitted to prefer the ruin of many to
my own ruin” (quoted in Johns 2013), then anyone is. So perhaps justice and
morality does not demand universal beneficence.
Thinking along these lines might lead us to favor as a theory of right not act
utilitarianism but a form of rule utilitarianism. We start with the idea that every-
one has the same claim-making authority but want to know what demands we
can justifiably make of one another. We then restate this question as “Given
that we all have the same claim-making authority, what demand-warranting
principles or rules should there be?” Since our equal authority warrants weigh-
ing everyone’s preferences equally, we then reinterpret this as “What demand-
warranting principles or rules are there such that their acceptance would
maximize preference-­satisfaction?” It seems clear, as rule utilitarians of all stripes
have historically argued, that such rules would not require universal beneficence.
In thinking about Leibniz’s view in relation to the good and the right, it is
worth noting some criticisms Barbeyrac makes of Leibniz and of Leibniz’s cri-
tique of Pufendorf, together with some comments about these by Schneewind
and Irwin. These center around Barbeyrac’s complaint that, unlike Pufendorf,
Leibniz fails to appreciate the fundamentally deontic character of morality and
that moral obligations purport to give us reasons that cannot be reduced to our
own good. Barbeyrac’s most important criticism, according to Schneewind, is
that whereas Leibniz holds “that all reasons for action, and hence all motives,
arise from considerations of just one kind,” “moral law brings into our deliber-
ations reasons for action that are different in kind” from considerations of our
own good (Schneewind 1998: 256). To this we might add that deontic consid-
erations of moral obligation differ also from those of universal or impersonal
good, even if the moral law requires universal beneficence.
Irwin defends Leibniz, and implicitly, Aristotelian naturalism, against this
critique (Irwin 2008: 327–330). He notes that Barbeyrac oversimplifies Leib-
niz’s eudaimonism. As can be seen in Leibniz’s comments on Grotius’s response
to Carneades’s challenge, Leibniz (and Aristotelian naturalists generally) can
distinguish between the agent’s advantage or narrow or instrumental interests
and the intrinsic benefit that is realized by benevolent actions for the common
good (the bonum honestum). Secondly, Barbeyrac’s critique that Leibniz fails
to appreciate morality’s deontic character, which Pufendorf’s account explains
better, rests, Irwin argues, on the implausible voluntarist view that obligation
must be voluntarily imposed.

https://doi.org/10.1017/9781139025065.006 Published online by Cambridge University Press

 164 spinoza, cudworth, shaftesbury, and leibniz

Both of these defenses can be resisted. As we noted in connection with Gro-

tius’s response to the Carneadean challenge, the bonum honestum provides a
reason of a fundamentally different kind from (deontic) morality as Grotius
understands it, and hence, a reason of the wrong kind to support or be entailed
by it. This is essentially Barbeyrac’s “deeper criticism,” as Schneewind pres-
ents it. And second, Irwin is assuming that the criterion for being on the law,
command, or obligation side of the Suárezian divide is imposition by superior
authority. It is certainly true that Suárez and Pufendorf, as theological volun-
tarists, would agree with this second point. But there is a more fundamental
conceptual connection between these deontic notions and accountability. The
voluntarists accept this connection, but hold that one can be accountable only
for what has been imposed by superior authority. However, this can be denied,
as it was, indeed, by Grotius and has been by later philosophers like Butler, the
British rationalists, and many philosophers today.
One of the things that is so interesting about Leibniz is that he actually insists
on this connection, as his “nominal definition” of justice in terms of justified
complaint and his formulation of the principle of equity make clear. At the
same time, however, the dominant elements of his thought are a perfectionist
theory of virtue and the good that recall classical naturalist themes.

https://doi.org/10.1017/9781139025065.006 Published online by Cambridge University Press

 5

Hutcheson and Butler

Francis Hutcheson (1694–1746) and Bishop Joseph Butler (1692–1752) were

true contemporaries whose lives overlapped for fifty years, and whose best-
known works appeared within a year of one another. Although they also had
similar intellectual temperaments and styles, their philosophical approaches
differed in important ways that reverberated through the moral philosophy
that followed.
Hutcheson’s An Inquiry into the Original of Our Ideas of Beauty and Vir-
tue was first published in 1725 (2004), and Butler’s Sermons appeared in 1726
(2017). Both were sharply analytical thinkers, and both shared a keen insight
into aspects of human psychology centrally involved in moral life, what we now
call moral psychology. Moreover, both laid great stress on a feature of moral
psychology they held to be implicated in moral judgment. Hutcheson called
this, following Shaftesbury, “moral sense,” and Butler referred to it alternately
as the “principle of reflection” or “conscience.”
For both Hutcheson and Butler, as for Shaftesbury, moral judgment involves
the human capacity to reflect upon and respond affectively to motives, char-
acters, and actions in a way that abstracts (more or less) from self-interest and
other forms of partiality as well as from the effects of mood and passion. How-
ever, the role of disinterested, dispassionate moral reflection for these two phi-
losophers is profoundly different. According to Hutcheson, it is an observer’s
moral sense that enables a person to make moral evaluations of motivation
and actions, but these evaluations are not self-reflexive – the agent is not evalu-
ating their own, or even others’, moral sense. For Butler, on the other hand,
conscience is a moral faculty by which agents crucially make judgments of
themselves, judgments they employ to shape their own actions. Conscience is
action-guiding for Butler in a way that moral sense is not for Hutcheson.
Butler agrees, moreover, with the Shaftesburean doctrine that genuine virtue
requires self-reflective moral agency. Indeed, Butler holds that conscience is a
necessary condition for moral agency itself. For Hutcheson, by contrast, moral
sense is inessential in virtuous conduct, at best, and, at worst, it can undermine
virtue by giving an otherwise virtuous agent motive to indulge complacently in
reflecting on their own virtue. It can give them, in Bernard Williams’s words,
“one thought too many” (Williams 1981a: 18).
165

https://doi.org/10.1017/9781139025065.007 Published online by Cambridge University Press

 166 hutcheson and butler

Considering Hutcheson and Butler’s respective relations to Leibniz can

bring out a related difference. Like Leibniz, Hutcheson takes the most virtuous
(morally best) motive to be a universal form of benevolence or love. Though
morally best, Hutchesonian love is not itself moralized. It is a desire for what
Hutcheson calls the “natural good” (well-being or benefit) of all beings. This
entails the point we just noted. Unlike Shaftesbury and Butler, and perhaps
like Leibniz, virtue for Hutcheson does not itself require moral judgment or
moral sense.1 Indeed, since Hutcheson holds that all desires (other than ara-
tional appetites and passions) have someone’s welfare or natural good as their
object, Hutcheson holds that it is actually impossible for moral sense to moti-
vate action directly. Butler, again, holds that virtue consists in conscientious
self-guidance.
A second difference springs from this. The object of universal benevolence
is, Hutcheson and Leibniz believe, the greatest happiness of all. In fact, as we
noted in the last chapter, Hutcheson may have taken the formulation of the
greatest happiness principle in his Inquiry directly from Leibniz.2 Butler, how-
ever, argues against Hutcheson that virtue cannot be reduced to benevolence.
Partly he has in mind the Shaftesburean (and later Kantian) doctrine we have
just mentioned: genuine virtue involves moral autonomy, that is, the agent’s
directing themselves by their own moral judgment. But Butler also rejects the
proto-utilitarian thesis that morality always requires, or even recommends, the
action that would bring the greatest happiness overall. There are other morally
relevant considerations, like justice and honesty, that, Butler argues, cannot be
reduced to promoting the general welfare.
Hutcheson is perhaps most significant for his influence on Hume. Several
familiar Humean doctrines – for example, that morality is derived from senti-
ment rather than reason, that reason is incapable of motivating, and Hume’s
account of obligation – come directly from Hutcheson. Hume is drawn to
Hutcheson because Hutcheson provides a promising way of accounting for
key features of ethics and ethical judgment within a fundamentally empiricist
framework that is consonant with the emerging experimental sciences. At the
same time, there are important differences between Hutcheson and Hume that
make Hutcheson a particularly interesting figure from the perspective of the
history of modern moral philosophy.
As may have been evident in what we have just noted about Hutcheson’s eth-
ics of virtue as universal love, explicitly deontic morality is not at the center of
his theorizing. When, indeed, he embraces the greatest happiness principle, it

1
I say, “perhaps like Leibniz,” since we noted some passages where Leibniz seems to hold that
virtue involves respect for others’ goods, and respect (recognition of others’ moral standing)
is moralized in its nature.
2
See Chapter 5, footnote 39.

https://doi.org/10.1017/9781139025065.007 Published online by Cambridge University Press

 hutcheson and butler 167

is not as an account of moral obligation or right, but more precisely of the mor-
ally choiceworthy (of what moral sense recommends concerning “election”)
(Hutcheson 2004: 125). In his Illustrations on the Moral Sense, Hutcheson goes
so far, indeed, as to term “ought” an “unlucky Word in Morals” (2002: 156).
These Hutchesonian ideas have clear resonances in Hume. Like Hutcheson’s
moral sense, what Hume calls moral sentiment is a disinterested response to
motive and character. As Hume memorably puts it: “[T]he external perfor-
mance has no merit” (Hume 1978: 477). And like Hutcheson’s, Hume’s ethics
are decidedly virtue theoretic. The only place that genuinely deontic notions
enter for Hume, like the ancients, is the virtue of justice, which Hume holds
to be an “artificial” rather than a natural virtue. Quite unlike Hume, however,
Hutcheson holds that there is a bright line between moral goodness and other
estimable or praiseworthy natural abilities and talents. This, indeed, is a direct
consequence of Hutcheson’s doctrine of the moral sense, which Hume also
rejects, though both are moral sentimentalists.
The Inquiry’s signature thesis is that the ideas of moral goodness and moral
evil cannot be reduced to those of natural goods and evils. Hutcheson accepts
Locke’s concept empiricism according to which simple, no-further-analyzable
ideas can be received only through some sense. It follows that if, as Hutcheson
believes, moral goodness and evil are to be understood in terms of the dis-
tinctive responses of “approbation” and “condemnation,” respectively, and if
“Approbation and Condemnation are probably simple ideas, which cannot
be farther explained” (Hutcheson 2004: 217), then there must be a dedicated
sense, the moral sense, through which we have these ideas. Hutcheson there-
fore recognizes a fundamental distinction between moral approval and disap-
proval (“condemnation”), on the one hand, and other, non-moral forms of
esteem and disesteem, however disinterested and dispassionate these may be.
Hume will reject both doctrines. He will reject Hutcheson’s theory of moral
sense in favor of his thesis that moral sentiment is produced by the same gen-
eral psychological mechanisms (sympathy and the association ideas) that are
in play in non-moral esteem and disesteem. And whereas Hutcheson claims
that “involuntary Disadvantage[s]” and “Fortune” cannot “exclude any Mortal
from the most heroick Virtue,” Hume will hold that the difference between
“moral virtues” and “natural abilities” is merely “verbal” and not worry much
about the relation of virtue to luck (Hutcheson 2004: 134; Hume 1975a: 279).
Unlike ancient versions, Hutcheson’s virtue ethics is a theory of distinc-
tively moral virtue, not in Aristotle’s sense of concerning passions and actions
rather than intellect, but as eliciting responses that are constitutive of a dis-
tinctive kind of non-Aristotelian value: moral goodness. As Hutcheson under-
stands them, virtues realize moral goodness by eliciting, or being such as to
elicit, the irreducibly moral idea he calls “approbation,” which non-morally
estimable traits and characters do not elicit. Canvassing cardinal Aristotelian

https://doi.org/10.1017/9781139025065.007 Published online by Cambridge University Press

 168 hutcheson and butler

virtues such as temperance and courage, Hutcheson says that these are genu-
ine virtues only to the extent that they are “dispositions universally neces-
sary to promote publick Good, and denote Affections toward rational Agents”
(Hutcheson 2004: 102).
In this way, then, Hutcheson counts as a modern moral philosopher. His
theories of moral goodness and the morally choiceworthy are intended to have
narrower scope than the kinds of goodness that are in play in ancient ethics.
They implicate responses, approbation and condemnation, that Hutcheson
takes to be distinctive of morality. When Hutcheson puts forward a version
of the greatest happiness principle, it is “[i]n comparing the moral Qualitys of
Actions” (with moral sense) “in order to regulate our Election among various
Actions propos’d” (125).
Despite this modern moral focus, deontic language and notions of obligation
and right enter into Hutcheson’s thinking only as derivative of moral goodness.
Hutcheson derives both his principle of moral choiceworthiness and his theses
about justice and rights from his fundamental theory of moral good.
There is another way in which Hutcheson rejects the Aristotelian naturalism
that is typical of ancient ethics; Hutcheson denies any form of eudaimonism.
He holds that considerations of other people’s welfare or “natural good” are no
less rationally choiceworthy than those concerning the agent’s own. Though he
believes that reason by itself cannot motivate an agent to pursue others’ good
or, for that matter, the agent’s own, he nonetheless thinks that calm rational
reflection can produce both self-regarding and other-regarding desires. As
we shall see, Hutcheson believes there are “two grand determinations” of our
nature: “one toward our own greatest happiness, the other toward the greatest
general good” (1755: 50). In this he follows Cumberland. Although Sidgwick
cites Butler as the source of his own “dualism of practical reason,” it is more
accurate to say that Sidgwick’s predecessors are Cumberland and Hutcheson
(Sidgwick 1967: x–xi).
This brings us to Butler. The Butlerian dualism to which Sidgwick pointed
was not between egoism (or self-love) and universal benevolence, but between
self-love and conscience. Butler was famous for the doctrine that conscience
has a unique “authority,” but he also says that conscience and self-love are “the
chief or superior principles in the nature of man: because an action may be suit-
able to this nature, though all other principles be violated; but becomes unsuit-
able, if either of those are” (Butler 2017: III.9). Being natural or suitable to our
nature are normative properties for Butler. Butler’s claim, in a more contempo-
rary idiom, is, roughly, that being contrary to the agent’s good and being ruled
out by conscience, that is, being morally wrong in the agent’s own view, are
both sufficient reasons not to do something.
Despite this passage’s appearances, however, Butler’s position is not that
self-love and conscience are equals in what Butler calls “authority.” Butler is

https://doi.org/10.1017/9781139025065.007 Published online by Cambridge University Press

 hutcheson and butler 169

careful to use “authority” only in connection with conscience and never with
self-love.

It is by this faculty [conscience], natural to man, that he is a moral agent,

that he is a law to himself: by this faculty, I say, not to be merely considered
as a principle in his heart, which is to have some influence as well as others;
but considered as a faculty in kind and in nature supreme over all others,
and which bears its own authority of being so (II.8).

Like Shaftesbury and Hutcheson, Butler sees our psychology as a kind of system.
Though he uses a variety of metaphors to describe it – that of an economy, for
example – his favored way of putting his idea is that the human psyche has a con-
stitution, where this is to be understood in the same juridical terms that apply to
a political body or state. It is only by treating conscience as authoritative, Butler
argues, by respecting its authority, that we are able to constitute ourselves as moral
agents.
Butler’s doctrine of the authority of conscience puts his ethics into a fun-
damentally different normative space than Hutcheson’s. It puts moral auton-
omy – the agent’s capacity to be a “law to himself” – at the heart, not just of
Butler’s view of moral virtue, but of what it is to be a moral agent. And it under-
stands the autonomy that grounds moral agency in essentially deontic or jurid-
ical terms.
Conscience, both for Butler and in our contemporary sense, is not a capac-
ity to make just any assessments of ourselves. It is a faculty with which we
make a distinctive kind of deontic judgment through which we hold our-
selves accountable (Darwall 2014b; Sorabji 2014). “[T]here is no man,” Butler
writes, “but would choose, after having had the pleasure or advantage of a
vicious action, to be free of the guilt of it, to be in the state of an innocent
man” (Butler 2017: VII.16). “[V]ice,” he concludes, cannot be the happi-
ness, but must upon the whole be the misery, of such a creature as man …,
an accountable agent” (VII.16). If, consequently, an action cannot be suit-
able to our nature if it is contrary to either self-love or conscience, it is not
because these have equal authority. Butler believes that the vicious implicitly
acknowledge the superior authority of their own consciences (over self-love)
through feeling guilt.
It is important to appreciate how Butler’s non-eudaimonist identification
of autonomy with the authority of conscience differs from the way auton-
omy features in the thought of Cudworth, Locke, and Shaftesbury. Although
Locke and Cudworth are led to accounts of autonomy because they think it
necessary for moral agency, and though Cudworth distinguishes between the
autonomy (“moral free will”) that is necessary for moral agency and an “animal
free will” that is sufficient for a self-reflective rational agency that falls short
of full accountability, neither Cudworth nor Locke questions a fundamentally

https://doi.org/10.1017/9781139025065.007 Published online by Cambridge University Press

 170 hutcheson and butler

eudaimonist framework. Most obviously, Locke’s egoistic hedonism forces him

to the view both that human agents have adequate reason to avoid moral wrong
only by virtue of divine sanctions and that autonomy consists in the capacity
to step back from the press of current desires and think about the long-term
consequences of wrongdoing for the agent so that they can bring their current
desires into line with their judgments about the latter. But even Cudworth, for
whom Lockean autonomy would amount only to animal free will, understands
moral free will not in terms of responsiveness to non-eudaimonistic consider-
ations of right but to higher goods. Even when deontic morality is in play, it
is not, for Cudworth, as a source of reasons that are irreducible to the agent’s
good. And the same is true of Shaftesbury.
Butler’s doctrine of the authority of conscience marks an important depar-
ture from this view. Indeed, Butler criticizes Shaftesbury on precisely these
grounds: “The not taking into consideration this authority, which is implied in
the idea of reflex approbation or disapprobation, seems a material deficiency
or omission in Lord Shaftesbury’s Inquiry concerning Virtue” (Pr.26). Though
Shaftesbury has shown “beyond all contradiction, that virtue is naturally the
interest or happiness, and vice the misery, of such a creature as man,” he has
not shown that moral sense or conscience has authority itself, an authority,
indeed, that can override considerations of the agent’s good, whether it is con-
ceived in terms of higher goods or not (Pr.26).
Recall Sidgwick’s claim that modern moral philosophy differs from ancient
ethics in recognizing two ultimate sources of normative reasons:

[I]n Greek moral philosophy generally, but one regulative and govern-
ing faculty is recognised under the name of Reason …; in the modern
ethical view, when it has worked itself clear, there are found to be two,—
Universal Reason and Egoistic Reason, or Conscience and Self-love
(Sidgwick 1964: 198).

In these terms, we might say that it is with Butler’s view of the authority of con-
science that “the modern ethical view … work[s] itself clear.”
Butler is not the first philosopher to hold that agents are capable of act-
ing rationally for ends other than their own good. Both Cumberland and
Hutcheson hold that an agent can aim rationally at others’ goods no less than
at their own. However, in holding that both their own good (considered from
a first-person perspective) and universal good (considered third personally)
are the two “grand determinations” of a rational human agent, Hutcheson also
maintains that there is no deliberative perspective that agents can occupy that
can enable them to decide between these should they conflict. It falls to God
to ensure that the correct results from these two perspectives always coincide.
Butler rejects this view. He denies that virtue can be reduced to benevolence,
holds that morality can require us to act in ways that are neither recommended

https://doi.org/10.1017/9781139025065.007 Published online by Cambridge University Press

 hutcheson and butler 171

by the agent’s or by others’ goods, and holds that accountable moral agents
must be able to treat these requirements as authoritative reasons.
Butler’s view differs also from Cumberland’s. Unlike Hutcheson, Cumber-
land maintains that agents can ask whether they should pursue their own or
universal good, in the case that these conflict. Indeed, that they can do this
plays a significant role in his argument for the rationality of universal benevo-
lence and the greatest good of all being the “best end.” But although this is so,
Cumberland holds a fundamentally teleological conception of practical reason,
according to which all normative reasons for action are reducible to goods.
Butler denies this. He rejects Hutcheson and Cumberland’s view that virtue
can be reduced to benevolence, as noted above, and holds that conduct can be
ruled out by conscience on grounds other than the good. Conscience may dis-
approve of injustice, for example, even if it produces an “overbalance of happi-
ness” (Butler 2017: D.8).
Significantly, Butler reserves the deontic/juridical term “authority” for con-
science. In so doing, however, he slides over a distinction between a different
sense of “authority” that is also not well marked among contemporary philoso-
phers these days. Sometimes in present-day philosophical discussion, “author-
ity” can express a broadly normative notion. When philosophers discuss, for
example, “first-person authority” in epistemology and the philosophy of mind,
they are not concerned with anything distinctively deontic or juridical. They
are simply considering whether, say, the fact that someone believes they are
in pain provides normative reason to believe they are that is not created by the
beliefs of other observers. Similarly, philosophers querying the “authority of
desire” are asking whether the fact that an agent desires something creates any
normative reason for them to bring it about. In neither case is anything essen-
tially deontic in question.
Used more narrowly, however, “authority,” has an essentially deontic/jurid-
ical use that is conceptually connected to obligation and accountability in ways
that put it on the law or demand side of Suárez’s distinction between law and
counsel. Clearly, this is the use that is in play when we consider issues of politi-
cal legitimacy or of one person’s authority over another. But it is also the sense
that is distinctively concerned with deontic morality and moral obligation. To
be sure, there are also uses of “obligation” in early modern writers like Shaftes-
bury, Leibniz, and, as we shall see, Hutcheson and Hume, that are non-deontic.
But these are reformative and subversive. The sense in which conscience in “the
modern ethical view” is held to be authoritative is not just that it responds to
especially weighty normative reasons, but also that these can obligate in the
sense that we are accountable for complying with them.
As we shall see, Butler tends to include both of these claims when he says
that conscience has supreme authority. Conscience is the faculty necessary for
accountable moral agency, through which we recognize morality’s distinctive

https://doi.org/10.1017/9781139025065.007 Published online by Cambridge University Press

 172 hutcheson and butler

capacity to obligate. But Butler holds as well that conscience, which he also
calls the “principle of reflection,” is necessary for genuinely rational agency in
general. In this latter case, conscience is a broader capacity not just to recognize
deontic obligatory norms, but also to make and guide ourselves by judgments
of normative reasons more generally. It will be important to keep these claims
separate and to try to understand the relation between them in his thought.
Generally, I will use “conscience” to refer to the former and “principle of reflec-
tion” to refer to the latter.
Butler identifies practical normative judgment in deontic terms. In doing so,
he turns eudaimonism on its head. Whereas eudaimonism grounds all norma-
tive practical reasons in the agent’s good, Butler’s view can appear to ground
them all in the right, as if there would be no normative reason to pursue self-
interest unless there were an obligation to do so. A sympathetic reading dis-
tinguishes, however, between narrower and broader notions of conscience in
Butler’s thought. So understood, Butler’s “broader” claim of the authority of
the “principle of reflection” is that rational autonomy involves agents’ guiding
themselves by their own judgments of normative reasons, and that doing so
requires distinguishing whether an action will benefit the agent and whether
there is therefore reason for them to perform it as different questions. “Every-
thing is what it is and not another thing” (Pr.39).
A central question of metaethics as it has been pursued in the twentieth and
twenty-first centuries has concerned the nature or “sources of normativity,”
where this latter is understood in terms of the grounds of normative reasons
understood most generally (e.g., Korsgaard 1996a). On a eudaimonist or Aris-
totelian naturalist view, there is only one fundamental normative concept: the
good. So long as the good is held to be the only source of normative reasons,
there can seem no important conceptual distinction between the good and
normative reasons. Once, however, a second normative concept, the right, is
distinguished, the separate, more fundamental concept of normative reason
or ought comes more clearly into view. In addition to issues concerning the
nature and grounds of the good and the right, there is the more fundamental
question of the nature of normative reasons and oughts, hence of normativity
itself. Arguably, Butler’s doctrine of the authority of conscience put modern
moral philosophy on a track to distinguish and ultimately focus on this latter
question.

Hutcheson
We noted in the last chapter that it has become customary to view Shaftesbury’s
thought retrospectively, through the lens of the moral sentimentalism his writ-
ings inspired in Hutcheson and Hume. Shaftesbury’s thought was unques-
tionably salient for Hutcheson. To the title of Inquiry’s first edition (originally

https://doi.org/10.1017/9781139025065.007 Published online by Cambridge University Press

 empiricist naturalism and moral sense 173

published in 1725) Hutcheson appended: “In which the principles of the late
Earl of Shaftesbury are Explain’d and Defended, against the Author of The
Fable of the Bees” (Hutcheson 2004: 199).3 What was primarily at issue between
Shaftesbury and Mandeville was whether all human motivation is ultimately
egoistic or whether there is such a thing as disinterested benevolence.
But although Hutcheson clearly sided with Shaftesbury on the question of
psychological egoism, there are many ways in which his thought departed from
Shaftesbury’s and in which he understood himself to be doing so. Whereas
Shaftesbury’s moral sense must be understood within a broadly rationalist
epistemology that owes much to the Cambridge Platonists, Hutcheson’s epis-
temology is more thoroughly Lockean. That moral concepts and judgments
require a special sense follows from what Hutcheson argues is the simplicity
or irreducibility of fundamental moral ideas together with the Lockean prem-
ise that simple ideas can only be received through senses. Unlike Shaftesbury,
Hutcheson was thoroughly in the empirical naturalist tradition of Hobbes,
Cumberland, and Locke.
Hutcheson follows Shaftesbury’s virtue ethics, but, as we noted, denies Shaft-
esbury’s claim that genuine virtue must involve self-reflective guidance through
moral sense.4 And Hutcheson also denies Shaftesbury’s claim that the “obliga-
tion to virtue” is eudaimonistic, derived from the agent’s own good. Hutcheson
agrees with Shaftesbury’s attempt to reform the use of “obligation” away from
expressing a distinctively deontic/juridical notion and toward the more general
idea of normative reasons and oughts. But he denies that the normative reasons
that ground virtue must ultimately be eudaimonistic. He praised Shaftesbury’s
project of demonstrating the congruence of virtue and interest as “the principal
Business of the moral Philosopher” (179), and asserted throughout his writ-
ings that self-interest constitutes an obligation to virtue. But he also insisted
against Shaftesbury that the virtuous person has normative reason to act virtu-
ously quite independently of any good to self, intrinsic or extrinsic, just in the
fact that the action would benefit others or, more accurately, in the fact that it
would benefit someone. And Hutcheson introduced a second reformative use
of “obligation” – what Hume would later call “moral obligation” – that is tied
directly to the approval of moral sense (Hume 1978: 498).

Empiricist Naturalism and Moral Sense

It is illuminating to view Hutcheson’s Inquiry as an extended argument that
morality can be understood, in all its complexity, within the framework of
empiricist naturalism. Broadly accepting Lockean doctrines about the powers of

3
Bernard Mandeville’s Fable of the Bees was published in 1714 (Mandeville 1988). Hereafter, I
will omit the original dates of publication of Hutcheson’s, Butler’s, and Hume’s writings.
4
In what follows, I draw from Darwall 1995: 207–243.

https://doi.org/10.1017/9781139025065.007 Published online by Cambridge University Press

 174 hutcheson and butler

reason and the empirical sources of ideas and knowledge, Hutcheson attempts
to show, against Hobbes and Locke, that morality has its own natural place.
Most significant, he argues, is a fundamental distinction between moral and
natural good that can be located naturalistically. Natural good is benefit or wel-
fare, which Hutcheson follows Locke in identifying with pleasure.5 Moral good,
by contrast, is realized only in the motives of a moral agent and is recognized
when contemplating those motives disinterestedly. Moral good is irreducible
to natural good, but that does not make it any less a natural feature or aspect
of nature. Whether a motive is morally good is determined entirely by facts of
nature and is just as open to empirical investigation as natural good is.
As we will see presently, Hutcheson holds that moral good and evil are
definable, respectively, in terms of the simple ideas of “approbation” and “con-
demnation.” Roughly, moral goods are motives whose contemplation elicits
approbation, and moral evils are motives that elicit condemnation. Hutcheson
holds that all other moral qualities – the morally choiceworthy, rights (along
with the distinction between perfect and imperfect rights), and moral obliga-
tion – are definable in terms of moral good and evil, hence approbation and
condemnation. Placing moral value properly in nature therefore provides a
naturalistic foundation for all of morality.
Later editions of the Inquiry begin with the following definition:
The word Moral Goodness, in this Treatise, denotes our Idea of some
Quality apprehended in Actions, which procures Approbation, attended
with Desire of the Agent’s Happiness. Moral Evil, denotes our Idea of
a contrary Quality, which excites Condemnation or Dislike (Hutcheson
2004: 85, 217).

To this, Hutcheson adds: “Approbation and Condemnation are prob-

ably simple Ideas, which cannot be farther explained” (217). Any impression
of a hedge with “probably” is removed by Hutcheson’s definition of moral
sense:

A Determination of our Minds to receive the simple Ideas of Approbation

or Condemnation, from Actions observ’d, antecedent to any Opinions of
Advantage or Loss to redound to ourselves from them (100, 221).

Since Hutcheson defines a sense, in general, as “a Determination of the Mind,

to receive any Idea from the Presence of an Object, which occurs to us, inde-
pendent on our Will” (90), the simple, irreducible character of approbation
and condemnation entails Hutcheson’s hypothesis of the moral sense.
When Hutcheson says that moral goodness is a quality of actions “that
procures Approbation,” he evidently means that it is the apprehension or
5
“The Pleasure in our sensible Perceptions of any kind, gives us our first Idea of natural
Good, or Happiness” (2004: 86).

https://doi.org/10.1017/9781139025065.007 Published online by Cambridge University Press

 empiricist naturalism and moral sense 175

contemplation of the quality that does so, not the quality directly itself. Moral
goodness elicits approbation by way of its contemplation. And Hutcheson’s
view is that the qualities of actions that do elicit approbation are all and always
motives of action, most directly, at least. Moral qualities, for Hutcheson, are
always of actions as motivated. As he sometimes puts it, the simple moral ideas
of approbation and condemnation arise only from contemplating “Affections
of rational Agents” (90).
Moral good contrasts with natural good in two ways. Natural good is simply
whatever causes pleasure, mediately or immediately, but moral good is that
the contemplation of which causes approbation. What is morally good is the
contemplated quality itself. Hutcheson does hold that approbation is a kind of
pleasure. So, to whatever extent moral good tends to cause pleasant contem-
plation of itself, this will make moral good (mediately) naturally good for the
person contemplating it. But this does not make moral goodness a species of
natural goodness. A quality might be no less morally good even if its existence
did not occasion its own pleasurable contemplation, if, say, it tended to distract
attention from itself.
The second thing that distinguishes moral good from natural good in general
is that Hutcheson takes approbation to be a specific kind of pleasure – a simple
idea, in Hutcheson’s empiricist theory of concepts – just as he takes condemna-
tion to be a distinctive kind of pain. This point will prove especially important
in a disagreement Hutcheson has with Hume over whether the most plausible
form of moral sentimentalism is a moral sense theory, or whether, as Hume
believes, approbation can be explained as resulting from more general psycho-
logical mechanisms: sympathy and the association of ideas and impressions.
Finally, it is worth noting that Hutcheson says not that moral goodness is the
property of causing approbation and love, but that it is that quality (motive),
if any, that causes these (when contemplated). He appends to his definitions
of moral good and evil: “We must be contented with these imperfect Descrip-
tions, until we discover … what general Foundation there is in Nature for this
Difference of Actions, as morally Good or Evil” (85). Empirical inquiry is nec-
essary to see whether some natural quality actually fits this description.
This suggests that Hutcheson intends his definition as a reference fixer,
pointing to a place in the natural causal order and identifying moral good with
whatever happens to occupy it. His hypothesis is that every quality that fills the
bill is some form of benevolence. If this hypothesis proves to be correct, then
moral good turns out to be identical with benevolence.
However, there are other places where Hutcheson seems to identify moral
goodness with something like the disposition to cause approbation (on
reflection).6 He sometimes treats beauty as depending on an “internal sense”

6
I am indebted here to Nicholas Sturgeon.

https://doi.org/10.1017/9781139025065.007 Published online by Cambridge University Press

 176 hutcheson and butler

in this way.7 And he gives his readers every reason to think he regards beauty
and moral goodness as having the same ontological status. In the Illustra-
tions, moreover, he says that the ideas received by both moral and internal
senses are, like those of “Colours, Sounds, Tastes, Smells, Pleasure, Pain,”
“only Perceptions in our own Minds, and not Images of any like external
Quality” (2002: 177).8
However we read Hutcheson on this issue, it is clear that he holds moral
good to be a feature of the natural order and that he thinks that every qual-
ity the contemplation of which causes approbation is some form of benevo-
lence.9 Section I of the Inquiry is devoted to showing that we all do experience
distinctive pleasurable and painful states of mind when we contemplate cer-
tain motives disinterestedly, that is, that we have a moral sense. And Section
II argues that every motive whose contemplation causes approbation either
is a form of benevolence (desire for someone’s natural good) or is approved
because of its relation to benevolence. Hutcheson argues also that our approval
of benevolence increases depending upon the extent of the benevolence we are
contemplating, with the greatest approbation being of universal benevolence.10
In every case, Hutcheson’s method of argument is the same: appeal to intro-
spection guided by judiciously selected cases.

Moral Virtue Ethics: Constructing and Deriving

Other Moral Notions and Standards
Moral good and evil are constructed in a straightforward way from the simple
ideas of approbation and condemnation, as qualities (motives) of action that,
when contemplated, give rise to approbation and condemnation, respectively.
So understood, moral good and evil are features that are assessed from an
observer’s rather than an agent’s perspective. We apprehend these qualities not
in deliberating about what to do, hence looking for reasons (motives) to guide
action. Rather we encounter them as features of motivated actions, once an

7
“[W]ere there no Mind with a Sense of Beauty to contemplate Objects, I see not how they
could be call’d Beautiful” (2004: 27).
8
Hutcheson contrasts these with the ideas of “Extension, Motion, Rest.” He also consider
the possibility that God could have constituted us with a moral sense that approved of
“Barbarity, Cruelty, and Fraud,” rather than benevolence, and implies, but does not say,
that this would make these traits morally good (2004: 197).
9
For discussion of the issue of whether Hutcheson deserves to be called a “moral realist,” see
Norton 1982, Winkler 1985, and Kail 2001. See also Frankena 1955.
10
“Benevolence is a Word fit enough in general, to denote the internal Spring of Virtue. …
But to understand this more distinctly, ’tis highly necessary to observe, that under this
Name are included very different Dispositions of the Soul. Sometimes it denotes a calm,
extensive Affection, or Good-will toward all Beings capable of Happiness or Misery. …
[This] sort is above all amiable and excellent” (2004: 237).

https://doi.org/10.1017/9781139025065.007 Published online by Cambridge University Press

 moral virtue ethics 177

agent has already settled the practical question for themselves of how to act or
has acted on some reason or motive.
Moral good and evil are identical with vice and virtue, for Hutcheson, but it
is important to see that Hutcheson’s is a narrower, distinctively moral, concep-
tion of virtue and vice than that of ancient ethicists like Aristotle. Hutcheson’s
is a moral virtue ethics. What makes something a virtue, for Hutcheson, is not
that it is a human excellence through which we flourish, or that contemplating
it gives rise to esteem of a broader kind, but that it gives rise to a distinctively
moral kind of approbation. And vices are features that elicit not contempt,
shame, or disgust, but rather condemnation. Although deontic ideas are not, as
we shall see, at the center of Hutcheson’s thought, condemnation suggests the
deontic/juridical nexus of accountability, culpability, and guilt. This grounds
a profound disparity between Hutcheson and Hume, who will deny that any
real difference between moral virtue and non-moral forms of human excel-
lence exists.
Approbation and condemnation arise, again, from an observer’s, rather than
from an agent’s, perspective. Knowing which motives are morally good does
not directly answer an agent’s question of what, or of what there is most moral
reason, to do. However, it is clear that Hutcheson thinks it can provide an
answer indirectly if we assume, as he evidently does, that we should approach
the deliberative question from the perspective provided by the motive that is
morally best.
Moral sense can “regulate our Election among various Actions propos’d”
(2004: 125) indirectly by, first, determining that universal benevolence is the
morally best motive and then working out which act is most highly recom-
mended from the perspective of that motive. Since the object of universal
benevolence is the greatest natural good or happiness of all, Hutcheson con-
cludes that “that Action is best, which procures the greatest Happiness for the
greatest Numbers; and that worst, which, in like Manner, occasions Misery”
(125). In this way, the most morally choiceworthy act is derived from the mor-
ally best motive.
Hutcheson’s formula is a proto-utilitarian account of the morally choice-
worthy that is arguably closer to Bentham and Mill than to Cumberland’s
“common good of all rationals.” But it is important to see that Hutcheson gets
to the greatest happiness principle in a profoundly different way, from within
an ethics of virtue, than do the later utilitarian writers. For Hutcheson, it is the
moral goodness of universal benevolence, the fact that benevolence elicits the
simple moral idea of approbation, that grounds everything.
It is significant that Hutcheson uses the term “best” (most morally choice-
worthy) rather than morally “right,” as later utilitarians will. Like Shaftesbury
before him, and Hume after, Hutcheson is a virtue ethicist who is, at best,
ambivalent about deontic moral conceptions. Although his “condemnation”

https://doi.org/10.1017/9781139025065.007 Published online by Cambridge University Press

 178 hutcheson and butler

has a definite deontic resonance, he subversively redefines moral obligation in

aretaic terms.
Hutcheson is most explicit about his project of constructing all moral
notions out of the two simple, distinctively moral ideas in the Inquiry’s Section
VII: “A Deduction of some Complex moral Ideas, viz. of Obligation, and Right,
Perfect, Imperfect, and External, Alienable, and Unalienable, from this moral
Sense” (177). Just as Hutcheson’s virtue ethics of universal benevolence lead to
a proto-utilitarian theory of moral choiceworthiness, so also does he derive a
proto-utilitarian theory of rights, perhaps the first in the history of ethics, on
the same basis.
“The true Original of moral Ideas,” Hutcheson reiterates, is “[t]his moral
Sense of Excellence in every Appearance, or Evidence of Benevolence” (177).
And “[f]rom this Sense too we derive our Ideas of Rights” (183).
Whenever it appears to us, that a Faculty of doing, demanding, or pos-
sessing any thing, universally allow’d in certain Circumstances, would
in the Whole tend to the general Good, we say, that any Person in such
Circumstances, has a Right to do, possess, or demand that Thing. And
according as this Tendency to the publick Good is greater or less, the Right
is greater or less (183).

Hutcheson then accounts for distinctions among rights within this proto-
utilitarian framework. Rights are “perfect” (coercively enforceable) if what
they protect is so important to the public good that their universal violation
“would make human Life intolerable” and if “allow[ing] a violent Defence, or
Prosecution of such Rights,” even outside of civil government, would have bet-
ter consequences than “the Violation of them with Impunity” (183). Imperfect
rights, by contrast, are those that are not appropriately coercively enforced,
where “a violent Prosecution of such Rights, would generally occasion greater
Evil than the Violation of them” (184). Similarly, the distinction between alien-
able and inalienable rights depends on whether “a Power … to transfer such
Rights may serve some valuable Purpose” (186).11

Calm Desires, Reason, and Moral Sense

In the next section, we will consider Hutcheson’s view of obligation, specifically
of what Hume will call later “moral obligation” (Hume 1978: 498). First, how-
ever, we should look more carefully at Hutcheson’s rejection of eudaimonism
and his view of practical reason, from which Hume will also borrow.
In An Essay Concerning the Passions and the Affections, Hutcheson lays out a
systematic theory of action and practical reason. His main object is to argue that
disinterested desires for the good of others are no less possible, and indeed, no less

11
For more detail, see Darwall 2022.

https://doi.org/10.1017/9781139025065.007 Published online by Cambridge University Press

 Calm Desires, Reason, and Moral Sense 179

rational in the only philosophically respectable sense, than self-regarding desires.

As Hutcheson sees it, the main obstacle to “conceiving disinterested Desires”
has been that philosophers, like Locke, have misconceived desires as “uneasy
Sensation[s] in the absence of good,” rather than as simple ideas that can only be
identified by their functional role in motivating action (Hutcheson 2002: 28).
“Desire is as distinct from any Sensation, as the Will is from the Understanding or
Senses. This everyone must acknowledge, who speaks of desiring to remove Uneas-
iness or Pain” (28–29). Once the clear distinction between desire and (unpleasant)
sensation is made, Hutcheson thinks, it should appear obvious that disinterested
desires are no less possible (or, indeed, “rational”) than interested ones.
Although Hutcheson argues famously in his Illustrations on the Moral Sense,
from which Hume will borrow liberally, that reason is motivationally inert in
itself, he nonetheless thinks that action and motivation can be more or less
based on rational apprehension. Indeed, he makes a fundamental distinc-
tion between “desires” or “affections,” on the one hand, and “passions” and
“appetites” that operate not through rational apprehension of their objects but
through sensation, on the other. “Desire” and “pure Affection … arise necessar-
ily from a rational Apprehension of Good or Evil,” whereas “there are in our
Nature violent confused Sensations, connected with bodily Motions,” which are
“denominated Passions” (51).
The goods or evils the rational apprehension of which gives rise to desires
and affections are natural goods and evils (pleasure or welfare and pain
or illfare). As Hutcheson argues in Illustrations, desire does not result from
rational apprehension alone; some motivational susceptibility or “Instinct” is
required (51). But it is important to see that the requisite instinct embeds a
conception of rationality. It is a disposition to desire natural goods (as it will
turn out, both for self and for others) when these are rationally apprehended – as
Hutcheson says, “calmly.” In this way calm desires and affections are, unlike
passions and appetites, “rational Desire[s]” (52). On this basis, he defines
“[t]he Will, or Appetitus Rationalis” as “the disposition of Soul to pursue what
is presented as good, and to shun Evil” (214).
The Essay lays out some twenty “Natural Laws of pure Affection” (13) that
describe our tendency to desire good, “as it appears to our Reason or Reflec-
tion,” without the attendant sensual confusion and violence of the passions (31).
Taken together, these imply that rational reflection on our own good leads us to
desire our own greatest good or pleasure, and that rational reflection on natural
goods, conceived disinterestedly, leads us to desire the greatest happiness of
the greatest number. “The Strength either of the private or publick Desire of any
Event, is proportioned to the imagined Quantity of Good” (37).
Hutcheson is not claiming that we all tend to desire the greatest good of all or
even our own greatest good. Rather, he is claiming that it is a law of our psycho-
logical nature that we tend to desire in proportion to quantity of good when we
contemplate the latter calmly and rationally.

https://doi.org/10.1017/9781139025065.007 Published online by Cambridge University Press

 180 hutcheson and butler

When the Soul is calm and attentive to the Constitution and Powers of
other Beings, their natural Actions and Capacities of Happiness and
Misery, and when the selfish Appetites and Compassions and Desires
are asleep, … there is a calm impulse of the Soul to desire the greatest
Happiness and Perfection of the largest System within the Compass of its
Knowledge (1755: 10).

Similarly, in the intrapersonal case, “there is found in the human Mind, when it
recollects itself, a calm determination toward personal happiness of the highest
kind it has any notice of” (50).
Hutcheson’s laws of calm desire thus embed an ideal of rational delibera-
tion. They are empirical generalizations, but about an ideal human agent whose
deliberations are free of the perturbing influence of passion and appetite, and
who is perfectly informed about, and attentively considering, all natural goods
in prospect. When human beings instantiate this ideal, they raise the “two
grand determinations” of our nature: “one toward our own greatest happiness,
the other toward the greatest general good” (50).
Since all desires must have natural goods as objects, it is easy to see why
Hutcheson was committed to holding that moral sense cannot motivate
directly. Hutcheson did hold that the motive we mostly highly approve through
moral sense, calm “extensive” (universal) benevolence, is one that rational
deliberation on natural goods will give us. But he believed that not only was an
apprehension of the moral goodness or virtue of benevolence not itself neces-
sary for moral virtue – it could actually interfere with it. The only way he had
of understanding Shaftesbury’s (and later, Butler’s) doctrine that self-reflective
approval is necessary for genuine virtue was that such self-approbation offers
the agent a natural good: the “secret Sense of Pleasure” of contemplating his
own virtue (2004: 87).
Hutcheson rightly believed he had sketched in the Essay a substantial alter-
native to Shaftesbury. Shaftesbury included benevolent motivation in human
nature but thought it could form no basis for a reflective endorsement of the
good life. Shaftesbury’s “obligation to virtue” remained steadfastly eudaimonist,
only the agent’s interest on the whole can provide, he thought, a framework
necessary to order pursuits in a rational way. The Essay’s central thesis, by con-
trast, is that “calm extensive benevolence” gives rationally reflective human
agents a second organizing motive. Just as agents can order the goods they
can realize for themselves within their good on the whole, so can they likewise
order goods they can realize impersonally. Particular benevolence stands to
calm extensive benevolence as agents’ desires for specific personal goods stands
to calm self-love or the desire for their own greatest good.
Ultimately, there are three reasons why Hutcheson was committed to the
view, denied by Butler, that moral sense cannot directly motivate the actions of
which it approves. First, moral sense receives sensations, and Hutcheson explic-
itly distinguishes these from desires, which can exclusively motivate. Second,

https://doi.org/10.1017/9781139025065.007 Published online by Cambridge University Press

 obligation 181

all calm desires are for natural goods, for oneself or others, and never for moral
good itself. And third, Hutchesonian moral sense approves of motives and not
acts considered independently of these and so cannot possibly motivate action.

Obligation
This brings us to Hutcheson’s view of obligation. Here again, Hutcheson
departs from Shaftesbury. Hutcheson rejected both Shaftesbury’s view that
agential self-direction through moral sense is necessary for genuine virtue and
that the (only) “obligation to virtue” is provided by motives of self-interest.
Hutcheson agreed with Shaftesbury that virtue ultimately coincides with self-
interest. Otherwise, the two “grand determinations” of practical reasoning,
universal benevolence and self-love, would conflict with no possible delib-
erative perspective to resolve them. But Hutcheson rejected Shaftesbury’s
eudaimonism and so his view that normative reasons to be virtuous are all
grounded in the agent’s own good. Universal benevolence is no less a source of
normative reasons than is self-love.
Shaftesbury’s reduction of the obligation to virtue to “what Reason [there
is] to embrace it” (Shaftesbury 2001: II.78), and then, to eudaimonist rea-
sons of the agent’s good, was, as we have noted, a subversive redefinition that
reduces away any Suárezian deontic/juridical content. Hutcheson shared
Shaftesbury’s desire to center his ethics on virtue rather than on deontic
morality, and so shared also Shaftesbury’s rationale for reforming definition
of deontic notions. “[O]ught,” Hutcheson said, is an “unlucky Word in Mor-
als” (Hutcheson 2002: 156).
Hutcheson might naturally have followed Shaftesbury’s reduction of obliga-
tion to normative reasons. From there he could have departed and said that
the obligation to virtue was not (simply) self-interest, but also the very con-
siderations of others’ (or impersonal) natural good, anyone’s weal and woe, to
which the morally best motive, universal benevolence, responds. He did not,
however, take that route.
Like Shaftesbury, Hutcheson accepts a species of obligation that Hume will
later call “natural obligation” or, more helpfully, “interested obligation” (Hume
1978: 498; 1985b: 278):

A Motive from Self-Interest, sufficient to determine all those who duly

consider it, and pursue their own Advantage wisely, to a certain Course of
Actions (Hutcheson 2004: 177).

But to this he adds a second species, what Hume will call “moral obligation,”
which involves moral sense. There are subtle differences between his account in
the Inquiry and his later Illustrations account. The latter is cleaner and gives his
mature contrast between the obligation of self-interest and that of moral sense:

https://doi.org/10.1017/9781139025065.007 Published online by Cambridge University Press

 182 hutcheson and butler

When we say one is obliged to an Action we either mean, 1. That the

Action is necessary to obtain Happiness to the Agent, or to avoid Misery:
Or, 2. That every Spectator, or he himself upon Reflection, must approve
his Action, and disapprove his omitting it, if he considers fully all its
Circumstances (2002: 146).

In defining moral obligation in relation to moral sense in this way, it might be

thought that Hutcheson had no real disagreement with Shaftesbury. Although
Shaftesbury did not call the approval of moral sense obligation, this might be
thought more of a semantic than a substantive disagreement.
There were, however, two substantial disagreements. First, as we have noted,
Hutcheson holds that moral sense cannot directly motivate, since all motives
are for natural goods. Second, later editions of the Inquiry make clear that what
Hutcheson has in mind by moral sense is not just any second-order “Liking
or Dislike,” as Shaftesbury put it (Shaftesbury 2001: II.28), but a disposition
to have distinctive moral responses of approbation and condemnation, the
“simple ideas” that define moral sense and from which, Hutcheson believes, all
other (complex) moral notions can be deduced (Hutcheson 2004: 100, 217, 221).
Once Hutcheson had accepted Shaftesbury’s reduction of one species of
obligation to normative reasons of self-interest, it might seem natural to under-
stand a moral species of obligation as consisting in the consideration of disin-
terested natural good to which the morally best motive, benevolence, responds.
In tying moral obligation to the approbation and condemnation of moral
sense, however, Hutcheson ties it to sensations or perceptions that he believes
are incapable of motivation. We might reasonably ask Hutcheson, then, what
is the genus of which his different senses of “obligation” are both species? What
can unify them?12

12
Hutcheson faces a similar unification problem in his account of practical reasoning. Like
Sidgwick’s dualism of practical reason, Hutcheson holds that practical reasoning can be
grounded either in calm-self love (the agent’s overall interest) or calm extensive (universal)
benevolence (the greatest good of all). By the time he wrote A System of Moral Philosophy
he no longer found it satisfactory to think that rational deliberation could be bifurcated in
this way.
But here arises a new perplexity in this complex structure where these two principles
seem to draw different ways. Must the generous determination, and all its particu-
lar affections, yield to the selfish one. … Or must the selfish yield to the generous?
Or can we suppose that in this complex system there are two ultimate principles
which may often oppose each other, without any umpire to reconcile their differ-
ences (1975: 50–51).
The Essay doctrine was that there can be no practical umpire between calm benevo-
lence and calm self-love. Through moral sense we approve of benevolence and not self-
love (except when we consider ourselves disinterestedly), but this approbation is, in itself,
deliberatively inert. It can give considerations of general happiness no greater weight as
practical reasons. But if no human deliberative adjudication is possible, neither is any

https://doi.org/10.1017/9781139025065.007 Published online by Cambridge University Press

 butler 183

Butler
We turn now to Butler. In light of his influence on modern moral philoso-
phy, it is remarkable how little of Butler’s life was actually devoted to writing
on the subject. When Hume listed Butler as one of “some late philosophers

needed. Since God has designed a pre-established harmony between self-love and benevo-
lence, the consistency of rational human agency is preserved.
In the System, however, Hutcheson writes that this “must appear a complex confused
fabric, without any order or regular consistent design” (74). Moreover, he adds that its
solution requires “a distinct consideration of [the] moral faculty. By means of it, all is capa-
ble of harmony” (74). But how can this be, given that moral sense is practically inert?
The System bears the clear imprint of Butler’s influence. Given his moral psychology,
and his reduction of virtue to benevolence, Hutcheson can hardly take Butler’s authority
of conscience into his system in Butler’s own terms. It is fascinating, however, to see how
Butler’s idea can be recast in Hutchesonian idiom.
At his most Butlerian, Hutcheson writes that the “moral sense from its very nature
appears to be designed for regulating and controlling all our powers” (61). But note
Hutcheson’s twist:
This moral faculty plainly shews that we are also capable of calm settled universal
benevolence, and that his is destined, as the supreme determination of the generous
kind, to govern and control our particular generous as well as selfish affections; as
the heart must entirely approve its doing thus in its calmest reflections (74).
Butler of course holds that it is conscience that ultimately governs. Hutcheson, however,
holds, against Butler, both that the will can aim only at natural goods (rather than directly
at moral good or at the right), and that moral sense approves of universal benevolence
overriding all other desires for natural goods, including calm self-love.
But this really gives us no more than we had before the System. Although moral sense
approves of benevolence over self-love, nothing whatsoever follows from this about the
relative weight of the agent’s good and the good of all as normative reasons for acting.
Thus Hutcheson faces a dilemma in the System. On the one hand, his theory of practi-
cal reasoning and the will, carried forward from the Essay, restricts rational motivation
to calm self-love and benevolence and seems to provide no place for genuinely practi-
cal judgment between these should they conflict. On the other, he wants to hold some
version of Butler’s position that the moral faculty provides the agent with just such a
perspective.
His solution is grounded in a doctrine of superior and inferior pleasures that closely
resembles Mill’s more familiar later version together with the observation that apparently
self-sacrificing benevolent actions provide higher pleasures than those that seem less self-
sacrificing. These higher pleasures always erase the conflict between benevolence and self-
love and always make the sacrifice only apparent and never real.
The doctrine of higher and lower pleasures is already in the Essay. Like Mill, Hutcheson
takes the testimony of experienced judges to determine superiority and inferiority of plea-
sures. “It is obvious that ‘those alone are capable of judging, who have experienced all the
several kinds of Pleasure, and have their Senses acute and fully exercised in them all’” (2002:
89). Moreover, Hutcheson also there says that the pleasures of virtue are superior to all
others. “[A]ll virtuous Men have given Virtue this testimony, that its Pleasures are supe-
rior to any other, nay to all others jointly” (89). That this is the testimony of the virtuous
themselves should not be dismissed as self-serving bias. To the contrary, since they alone

https://doi.org/10.1017/9781139025065.007 Published online by Cambridge University Press

 184 hutcheson and butler

in England, who have begun to put the science of man on a new footing,” he
was judging Butler on the strength of several sermons preached at the Roll
Chapel in London and published as part of a collection in 1726 (Hume 1978:
xvi; Butler 2017). Although others of Butler’s Sermons have had continuing
interest – for example, those on resentment and self-deceit – Butler’s cen-
tral ethical ideas are mainly laid out in only five sermons (and a preface),
three on human nature and the superior authority of conscience, and two
on “the love of our neighbor.” Besides his “Dissertation of the Nature of
Virtue,” Butler’s only other philosophical work was his massive Analogy of
Religion (1736), with which the “Dissertation” was published, which argued
that revealed religion is fully consistent with then current experimental sci-
ence. Mostly, Butler was an Anglican clergyman, then bishop, who devoted
himself to church affairs.
Butler’s ethics have had a profound and continuing influence on the sub-
ject. His impact on Hume, Hutcheson, and Adam Smith was substantial. And
the Sermons were reprinted more during the nineteenth century than any
other British ethical work (Schneewind 1977: 7). Sidgwick’s Methods of Ethics
is full of references to Butler. Also, as I mentioned earlier, the terms Sidgwick
uses to refer to the two “regulative and governing facult[ies]” that he claims
modern moral philosophy comes to recognize are Butler’s: “conscience” and
“self-love” (Sidgwick 1964: 198). Even if Cumberland and Hutcheson stand
more clearly behind Sidgwick’s “dualism of practical reason,” it was Butler’s
“authority of … conscience” that more evidently marked modern moral phi-
losophy’s expression of the independent normativity of the right (Sidgwick
1967: x–xi; Butler 2017: Pr.14). Cumberland and Hutcheson’s views were dual-
isms of the good.13

will have had their “Senses acute and fully exercised” in their benevolence, their testimony
is dispositive evidence.
Hutcheson makes clear that the superior pleasures of benevolence consist not just in the
joyful sense that others are benefited, but also in the awareness that one has benefited them
oneself and also, he seems to add, that this is morally good.
These joys are different from the sympathetick, which may arise from that happi-
ness of others to which our affections and actions contributed nothing. … When we
find our own whole soul kind and benign, we must have a joyful approbation: and a
further and higher joy arises from exercising these affections in wise and beneficent
offices (1975: 1331).
Even if, therefore, moral sense cannot regulate deliberation and so directly govern in the
way required by Butler’s authority of conscience, it is nonetheless integral to a harmony of
practical reason that God affords us. Whereas Butler makes moral autonomy the center of
moral virtue and unity something the agent can themselves achieve in their own practical
reasoning, albeit with God-given materials, both Hutchesonian virtue and unified practi-
cal reasoning depend more directly on God’s grace.
13
As, actually, was Sidgwick’s.

https://doi.org/10.1017/9781139025065.007 Published online by Cambridge University Press

 the human moral psychological economy 185

Analytical moral philosophy of more recent times is also greatly indebted

to Butler. Butler was a master of subtle, important distinctions. His marvelous
remark, “[e]verything is what it is, and not another thing” (Butler 2017: Pr.39),
provided the epigraph for G. E. Moore’s agenda-setting work of twentieth-
century analytical ethics, Principia Ethica.14 And Butler’s distinction between
“mere power and authority” (II.14) anticipated Sidgwick’s contrast between
non-normative judgments and normative or “ethical judgments,” as well as
Moore’s juxtaposition of naturalistic properties with the irreducible property
of goodness (Sidgwick 1967: 23–38; Moore, 1993: 53–88). Finally, W. D. Ross’s
famous “deontological” objections to Moore’s “ideal utilitarianism” – that con-
siderations other than the goodness of consequences “make right acts right” –
bear a clear relation to Butler’s objections, in his Dissertation, to Hutcheson’s
claim that “virtue is resolvable into benevolence” (Ross 2003). There Butler
argues, anticipating Ross, that we “condemn falsehood, unprovoked violence,
[and] injustice” and “approve of benevolence to some preferably to others,
abstracted from all consideration, which conduct is likeliest to produce an
overbalance of happiness or misery” (Butler 2017: D.8).

The Human Moral Psychological Economy

What is most notable in Butler is his sophisticated and systematic moral psy-
chology, along with his penetrating analysis of different motivational sources
within the human psychic “economy” or “constitution,” and his powerful, sug-
gestive ideas of how these fit together within moral agents – or, better, how
moral agents fit them together themselves in autonomously determining their
own moral conduct.
Following Shaftesbury, Butler distinguishes between “private” and “public”
“passions and affections” based on their function and tendency to promote
individual and overall good, respectively, whether or not they have these as
intentional objects. Thus, hunger is a private passion, since it leads people even
without “any conviction from reason of the desirableness of life” to preserve
themselves and their own good (I.7). Butler does not intend these categories
to be mutually exclusive. A passion can be both public and private, as are, for
example, “desires of esteem from others” (I.7).
“Everything is what it is, and not another thing” (Pr.39). And what distin-
guishes different passions is that they are for one thing rather than for another.
Butler insists that desires invariably have specific intentional objects. It is
essential to his famous argument against psychological egoistic hedonism that
“all particular appetites and passions are towards external things themselves,
distinct from the pleasure arising from them” (XI.6).

14
Saul Kripke invokes it also in Kripke 1981: 94.

https://doi.org/10.1017/9781139025065.007 Published online by Cambridge University Press

 186 hutcheson and butler

There could be no enjoyment or delight for one thing more than another,
from eating food more than from swallowing a stone, if there were not an
affection or appetite to one thing more than another (XI.6).

A person’s good or “happiness,” Butler holds, “consists only in the enjoyment

of those objects, which are by nature suited to our several particular appetites,
passions, and affections” (XI.9).
Self-love, it follows, is a second-order desire for the satisfaction of one’s first-
order desires. Benevolence is likewise second-order; it is the desire for the sat-
isfaction of others’ first-order desires. We will consider some implications of
these claims that Butler makes use of in arguing against psychological egoism
below. Here, however, we should note that benevolence and self-love are alike
in standing “above” particular public and private desires, respectively, if not yet
in an order of justification, in the sense that an agent must be able to view first-
order desires with reflective distance even to have these second-order sources
of motivation. They are desires of which only mutually aware, self-reflective
beings are capable.
Butler uses the language of normative hierarchy, “superiority” and “infe-
riority,” to express the claim that there is normative reason to act on one
desire rather than another. In this way, Butler calls self-love a “superior
principle” to specific private desires. Moreover, as we have already noted,
Butler classes self-love with conscience as being “the chief or superior prin-
ciples in the nature of man: because an action may be suitable to this nature,
though all other principles be violated; but becomes unsuitable, if either of
those are” (III.9).
This might suggest a Sidgwickian “dualism of practical reason,” like
Hutcheson’s, with conscience playing the dual to self-love that universal (calm
extensive) benevolence plays in Hutcheson’s view. For Hutcheson, recall,
self-love and universal benevolence are independent, final, incommensurable
sources of normative practical reasons. Butler’s statement might suggest, as
Sidgwick took it, that this is what Butler held about self-love and conscience
(Sidgwick 1967: xi).
There is, however, a fundamental difference between self-love and con-
science that is central to Butler’s claim that only conscience has what he calls
“authority.” As Butler understands self-love, namely, as a second-order desire
for the satisfaction of one’s first-order desires, it has no normative content. But
that is what precisely is not true of the “principle of reflection” or conscience. As
Butler conceives it, the principle of reflection uniquely has authority, because it
uniquely makes normative judgments.
But there is a superior principle of reflection or conscience in every man,
which distinguishes between the internal principles of his heart, as well
as his external actions: which passes judgment upon himself and them
(Butler 2017: II.8).

https://doi.org/10.1017/9781139025065.007 Published online by Cambridge University Press

 the human moral psychological economy 187

The terms of judgment that Butler then mentions are distinctively moral and
deontic. Conscience “pronounces determinately some actions to be just, right,
good; others to be in themselves evil, wrong, unjust” (II.8). This suggests that But-
ler takes the judgments of conscience always to be judgments of deontic morality.
However, Butler clearly thinks we make normative practical judgments that are
not distinctively moral in the deontic sense. Someone’s “strongest desire” might
be for some “present gratification” even at the cost of “certain ruin.” In such a
situation, Butler says, self-love is “superior” to the former desire, however strong
the former might be (II.10). In such a case, “without particular consideration of
conscience, we may have a clear conception of the superior nature of one inward
principle to another” (II.11). Acting on the stronger desire is “unnatural,” Butler
asserts, using “unnatural” in the same normative sense as “inferior” to express the
claim that the desire for present gratification, even if stronger than self-love, gives
rise to less weighty normative reasons for action than does self-love.
Butler clearly intends his claim that self-love is a superior principle to
the stronger desire for lesser gratification, or that acting on the latter desire
is unnatural or contrary to our nature, as a normative judgment. It is, again,
a judgment we might express in a contemporary idiom by saying that there
is, or that the agent has, weightier normative reason to do what is for their
greater good in the long run rather than to act for lesser but closer goods. But
the only faculty of normative practical judgment in Butler’s moral psychology
is the principle of reflection or conscience, conceived in the broad sense. It is
a signal difference between Hutcheson’s and Butler’s pictures that Butler takes
the capacity for self-regulating normative judgment via the principle of reflec-
tion to be an essential feature of autonomous agency. Whereas Hutcheson
holds that there is no deliberative perspective from which agents can decide
between self-love and benevolence, Butler conceives the principle of reflection
to be unlike Hutchesonian moral sense in being a faculty of practical normative
judgment by which agents determine themselves in their own deliberations. It
is, he says, what makes autonomous agency possible.
In the case in question, however, the kind of normative judgment that is
called for, between self-love and the desire for a lesser, closer pleasure, is not
one of deontic morality. That is why Butler says that we can judge the norma-
tive superiority of self-love “without particular consideration of conscience”
(II.11). Since what is at issue is not a deontic moral question, no issue of con-
science in this narrower sense is involved. But there is a broader sense in which
the “principle of reflection,” which phrase I shall use to express this sense, is
the seat of normative practical judgment in Butler’s moral psychology.15 In this

15
Henceforth, I will reserve “conscience” for the faculty through which we make judgments
of deontic morality, as well as judgments, for example, of moral goodness that derive from
deontic morality.

https://doi.org/10.1017/9781139025065.007 Published online by Cambridge University Press

 188 hutcheson and butler

wider sense, we make judgments of superiority/inferiority and naturalness/

unnaturalness through our principle of reflection.
Thus Butler says in the Preface to the Sermons that his aim is to show that vir-
tue accords with our nature, using “nature” in the same wider normative sense
we have just been using. And he notes there what he describes as an “inward
feeling” or “inward conviction” of the “ancient moralists,” “which they chose
to express in this manner, that … [virtue] consists in following nature” (Pr.13).
To this Butler adds, that such “inward perceptions … deman[d] in all cases to
govern such a creature as man” (Pr.16). This makes it clear that Butler’s view
is that the “inward” judgments of naturalness and superiority – in our terms,
judgments of normative practical reasons – employ the principle of reflection:
“you cannot form a notion of this faculty … without taking in judgment, direc-
tion, superintendency” (II.14). Autonomous agents can be a law to themselves
only because they have the capacity to make, and act on, their own normative
practical judgments.

Normative Practical Judgment: Wide and Narrow

The theme of autonomy and making one’s own judgments is evident through-
out the Sermons from the outset. The Preface begins with these words:

Though it is scarce possible to avoid judging, in some way or other, of

almost everything which offers itself to one’s thoughts; yet it is certain, that
many persons, from different causes, never exercise their judgment, upon
what comes before them, in the way of determining whether it be conclu-
sive, and holds (Pr.1).

“Arguments are often wanted for some accidental purpose,” Butler continues,
“but proof as such is what they never want for themselves, for their own satis-
faction of mind, or conduct in life” (Pr.1; see also Pr.2–7).
Although judgment of some kind is at least implicit in virtually all thought,
Butler decries the self-indulgent rationalization, self-deception (IV, VI, X),
inattentive thought, and complacent reluctance to make careful judgments that
he finds around him. “People habituate themselves to let things pass through
their mind … rather than to think of them.” “Review and attention, and even
forming a judgment, becomes fatigue, and to lay anything before them that
requires it, is putting them quite out of their way” (Pr.4).
The theme here is of people failing to make use of their faculty of judgment,
both generally and in “conduct in life” (Pr.1). The latter faculty is the “principle
of reflection.” To make the best sense of Butler’s idea, again, we need to dis-
tinguish between a narrower reflective faculty that makes deontic moral judg-
ments, “conscience,” and a broader faculty of normative practical judgment in
general, for which we are reserving “the principle of reflection.” When Butler

https://doi.org/10.1017/9781139025065.007 Published online by Cambridge University Press

 normative practical judgment: wide and narrow 189

argues, for example, in Sermon II, that we have reason to live a morally virtuous
life (in his terms, that such a life accords with our nature), we should regard this
argument as addressed to our principle of reflection and not to our conscience
in the narrow deontically moral sense.
It is a tautology that it is wrong to do what is wrong, which we can only judge
with narrow conscience. Butler obviously thinks he is making an argument in
Sermon II and not just pointing to a tautology. He must think that in arguing
that, in our terms, there is sufficient reason not to do wrong, he is not just argu-
ing that wrongdoing violates moral obligation or goes against the balance of
moral reasons. He is aiming to support the further claim that wrongdoing is
against what one has reason to do all things considered, or against the balance
of normative practical reasons sans phrase.
It is also clear that when Butler argues, throughout the Sermons, that virtue
accords with our nature, or that vice and morally wrong conduct are unnatural,
he is not simply asserting what Irwin calls the Aristotelian naturalist claim that
virtue, because it realizes human nature, realizes eudaimonia or human good
or well-being. Aristotelian naturalism is eudaimonist, so were we to interpret
Butler’s claims in this way, he would be grounding normative reasons to be
virtuous or to act rightly in the agent’s good. But that would commit Butler to
the view that self-love is the only ultimate source of normative reasons. Even if
Butler were to hold, as Sidgwick sometimes interprets him, that self-love and
conscience are independent sources of normative reasons that are on a par, he
would have to reject this eudaimonist view.
Butler explicitly considers Shaftesbury’s view that the “obligation to virtue”
consists in reasons for acting grounded in the agent’s interest. Like Hutcheson,
Butler grants that Shaftesbury “has shewn beyond all contradiction, that vir-
tue is naturally the interest or happiness, and vice the misery, of such a crea-
ture as man” (Pr.26). But Butler explicitly denies Shaftesbury’s eudaimonism.
“[W]­hatever their opinion be concerning the happiness of virtue,” he writes,
people are still “under the strictest moral obligations” to avoid wrongdoing and
vice (Pr.27).16
It is an important fact that the object of self-love, for Butler – the satisfaction
of the agent’s first-order desires, their “happiness” – is itself non-normative. Of
course, Butler believes that all agents have reason to further their happiness.
But that this is so is a further normative claim or judgment that is not entailed

16
Butler’s famous “cool hour” passage might be thought to be some evidence against this:
“Let it be allowed, though virtue or moral rectitude does indeed consist in affection to and
pursue of what is right and good, as such; yet, that when we sit down in a cool hour, we can
neither justify to ourselves this or any other pursuit, till we are convinced that it will be
for our happiness, or at least not contrary to it” (XI.20). It is clear enough, however, that
Butler’s “Let it be allowed” signals a dialectical supposition rather than any assertion of
eudaimonism. I discuss this more carefully at Darwall 1995: 265n–266n.

https://doi.org/10.1017/9781139025065.007 Published online by Cambridge University Press

 190 hutcheson and butler

by the natures of happiness and self-love themselves. The judgment that some-
thing will promote the agent’s happiness is “one thing,” we might say, echoing
Butler’s famous phrase, and the judgment that there is therefore reason to do it
is “another.” The second is a normative judgment, while the first is not.
Though self-love responds to normative practical reasons, it does not itself
make or involve normative judgments. The only sources of normative judg-
ment are the principle of reflection and conscience. As we are using Butler’s
terms, as we noted above, conscience is the faculty we employ when we judge
actions in moral deontic terms. And the principle of reflection is the faculty
through which we judge the existence of normative practical reasons and
oughts in general.
Consider in this light the evidence Butler cites that human beings actually
have a principle of reflection or conscience. Butler asks us to imagine “a man”
who has helped “an innocent person in great distress” but who also “in the
fury of anger” does “great mischief to a person who had given no just cause of
offence” (I.8).

Let the man … coolly reflect upon them afterwards, without regard to their
consequences to himself: to assert that any common man would be affected
in the same way towards these different actions, that he would make no
distinction between them, but approve or disapprove of them equally, is
too glaring a falsity to need being confuted (I.8).

Several things about this example are worthy of comment. First, although we
are told that the man acts in anger in the second case, we are not told what the
motive of action actually is in either case. The object of reflective judgment is
an act in a circumstance (we can consider the agent’s anger as part of the cir-
cumstances). In this way, both wide and narrow Butlerian reflective judgment
(the principle of reflection and conscience) differ from both Hutcheson’s moral
sense and Hume’s moral sentiment. The direct object of the latter is always, at
least in the first instance, an agent’s motive, whereas the object of Butlerian
reflective judgment is standardly an action in a circumstance.17
A related, second point is that the standard case of conscience or the prin-
ciple of reflection is that of an agent reflecting on their own actions and situ-
ations, whether from the deliberative perspective or post hoc, considering
what they did given the circumstances of choice that faced them. Again, this
is no part of the Hutcheson/Hume view of moral sense/moral sentiment. Of
course, Hutcheson and Hume believe that, in principle, observers can judge
the moral goodness and badness of their own motives no less than they can
those of others (supposing that they are capable of disinterestedly bracketing

17
Direct, because Hutcheson holds that moral sense can indirectly (and derivatively) approve
of acts as choiceworthy.

https://doi.org/10.1017/9781139025065.007 Published online by Cambridge University Press

 normative practical judgment: wide and narrow 191

their own interests). And, also of course, Butler holds that reflective judgments
they make of themselves commit agents to making similar judgments of oth-
ers in the same situation. Butler’s point is not that in the agent’s distinguishing
their judgments of their action toward the innocent in distress in the first case
from their judgment of angry injury of someone against whom they had no just
claim, the agent would be likely to make a different judgment of themselves than
they would of someone else in the very same situation. To the contrary, Butler
is clearly assuming that such judgments would have to be identical on pain of
inconsistency.
Rather, Butler’s point is that an agent’s reflective judgment of themselves
plays a fundamentally different role in their moral psychological economy than
their judgments of others’ actions possibly can because, first, it occurs within
the agent’s own “economy,” and, second, because it purports to play an author-
itative, directive role within that economy. In judging that they really should
not have angrily injured someone who had done nothing to deserve their ill
usage, they do not merely note a fact about themselves. They come to a norma-
tive judgment that can play a role in holding themselves to the norm whose
acceptance the normative judgment expresses.
Third, and this is shared with Hutchesonian moral sense, the deliverances of
conscience purport to be disinterested and dispassionate. Taking it for granted
that, in the heat of anger, the “man” may well have construed his situation as
one in which his victim “had it coming,” Butler asks him to “coolly reflect” on
what he has done when anger’s heat has subsided. Moreover, he is to reflect back
on his actions “without regard to their consequences for himself.” This does not
mean that consequences for, as it were, “the agent,” are to be abstracted. What
he is to ignore is that the agent was he himself. Thus even though it is impor-
tant, as we noted in the last point, that judgments of the principle of reflection
or conscience are reflective, that is, that agents make these judgments about
themselves, it is also important that agents make these judgments about them-
selves as agents facing specific kinds of alternatives in a specific kind of situation.
This commits agents to making identical judgments of others in the very same
kind of circumstance facing the very same kinds of alternatives.
Notice now that these conditions are no less relevant to practical norma-
tive judgment than they are to deontic moral judgments. Thus consider But-
ler’s example of someone who does not defer a lesser gratification at the cost
of losing a much greater one later. (In Butler’s actual example, the cost is “cer-
tain ruin” (II.10).) Even if, as we may suppose, no deontic question of right or
wrong is at issue, we can well imagine Butler saying that it would be “too glar-
ing a falsity to need being confuted” (I.8) to assert that someone could decide
they had equally good reason to act prudently and to act imprudently for no
creditable purpose after they had considered both options coolly and disinter-
estedly. Both deontic moral judgments and normative practical judgments in

https://doi.org/10.1017/9781139025065.007 Published online by Cambridge University Press

 192 hutcheson and butler

general purport, equally, to be dispassionate, disinterested, and to be based on

an informed and adequate grasp of their objects.
It is no argument against this that, as Butler is clearly assuming, there are
practical normative reasons for agents to act for their own interests and to pur-
sue other forms of self-partiality. This means that agents will correctly take the
fact that an action is in their interest as a normative reason to do it. But though
this fact is “interested” (i.e., self-interested, a fact about their interests), the
fact that this (self-interested) fact is a normative reason for them is not self-
interested. It is not essentially about their interests. The fact that an action is in
my interest is a reason for me to take it if, and only if, the fact that an act is in the
agent’s interest is a reason for that agent to take it. To assess these facts, I have
to abstract from their relation to my interests as such.

The Authority of Conscience

In the Preface and Sermon I, Butler argues that our moral psychological econ-
omy contains various specific public and private desires, self-love, benevolence,
and the principle of reflection. In Sermon II he turns to arguing that there are
normative differences of superiority and inferiority between these, and that it
is uniquely true of conscience or the principle of reflection that it has authority:
that it alone is “in nature supreme” over all other principles, that it “ought to
preside over and govern all the rest” (II.1).18
It is Sermon II where Butler attempts to demonstrate the existence of dis-
tinctive “moral obligations” (II.1) as against Shaftesbury’s view that the obliga-
tion to virtue consists (eudaimonistically) only in self-interest (Pr.26). Butler is
also implicitly arguing against Hutcheson’s view that moral obligation can be
grounded in a third-personal, observer’s mental state that can get no direct pur-
chase in deliberation. His aim is to show that moral obligations are grounded in
conscience’s supreme deliberative authority.
There are three distinguishable strains of argument in Sermon II for con-
science’s deliberative authority and thus for the claim that we should govern
deliberation by its deliverances. One is a teleological/functional argument that
appeals to conscience’s manifestly superintendent function. A second places
the first in a theological framework, arguing that it is manifestly God’s purpose
that we function in this way. There is clear textual support for both of these
arguments, but both face apparently insuperable objections to which But-
ler himself seems committed. A third argument seems both more promising

18
Note that though we are using “principle of reflection” and “conscience” to refer, respec-
tively, to faculties of all-things-considered practical judgment and deontic moral judg-
ment, Butler does not himself distinguish these. His claim here applies most obviously to
overall practical judgment.

https://doi.org/10.1017/9781139025065.007 Published online by Cambridge University Press

 the authority of conscience 193

philosophically, and truer to Butler’s claim that it is by virtue of conscience

that human beings “are a law to themselves” and that they are “moral agent[s]”
(II.4, 8, 9). Because this argument proceeds from necessary conditions for the
very possibility of moral agency, we can call it Butler’s “transcendental argu-
ment.” It holds that the validity of moral obligation, and normative reasons to
act as obligated, are necessary conditions of the possibility of moral agency.19
Let us consider Butler’s teleological/functional argument first. Butler first
discusses conscience’s functional role in Sermon I. It is evident, he argues, that
conscience restrains us from doing harm and leads us to do good to others. In
this way, its function is the same as any “public” principle. But conscience, by
its nature, is less vulnerable to the effects of mood and passion, so it can move
us when natural affections are challenged. Thus, Butler points out, parents who
may find their love and affections for their children taxed by “labour and dif-
ficulties” can be “carrie[d]” to care for them by their conscientious sense that
doing so is their “proper business” as parents (I.8).
In Sermon I, though, Butler is content to argue that conscience is simply a
principle in our nature, that it has some independent role in our moral psy-
chological economy. In Sermon II, Butler argues not just that conscience is
designed to lead us to do good when it “happen[s] to be stronger than other
principles,” but that we are so designed that “the whole character be formed
upon thought and reflection; that every action be directed by some determinate
rule, some rule other than the strength and prevalency of any principle or pas-
sion” (II.3). So Butler now asks, “What sign is there in our nature … that this
was intended by its Author?” (II.3). What evidence does our nature provide
that it has a design in which conscience’s function is to superintend?
The most obvious consideration, Butler thinks, is that conscience uniquely
represents itself as superintendent. Butler also offers a second, coordinate con-
sideration. Benevolence, self-love, and other public and private affections are
all “in a degree equally natural,” and are all equally part of our natural design.
However, “we have no method of seeing the particular degrees in which one
or the other is placed in us by nature” (II.8). So we cannot successfully govern
ourselves by any of these more specific principles or by a judgment of their
naturalness, understood teleologically. When they conflict, we have no way of
telling, by looking to their place in our design, which we should follow. None of
these can be a “law to us” (II.8).
Conscience is God’s solution to this design problem. Butler continues:
But there is a superior principle of reflection or conscience in every man,
which distinguishes between the internal principles of his heart, as well
as his external actions: which passes judgment upon himself and them;

19
What follows draws on Darwall 1995: 262–283.

https://doi.org/10.1017/9781139025065.007 Published online by Cambridge University Press

 194 hutcheson and butler

pronounces determinately some actions to be in themselves just, right,

good; others to be in themselves evil, wrong, unjust (II.8).

By giving us intuitive judgments of which principles we should follow and

which actions we should perform, God has given us a way of regulating our-
selves that does not require detailed knowledge of the precise place of specific
principles in our design.
Assume, then, that Butler has shown that, in the terms of his famous meta-
phor, we function properly when we virtuously follow conscience much as a
watch functions properly when it accurately gives the time (Pr.14). Add that
this is God’s evident intention. The problem is that these premises can hardly
be dispositive by themselves, without additional normative premises, for
beings that have a principle of reflection.
Butler himself makes the relevant point in the Preface just after the watch
analogy. Machines, like watches, are “inanimate and passive: but we are agents.
Our constitution is put in our own power. We are charged with it: and there-
fore are accountable for any disorder or violation of it” (Pr.14). The principle
of reflection makes possible our moral agency, our existence as accountable
moral beings. Of course, Butler is supposing here that it makes us accountable
for “any disorder or violation” in our functioning. But that can be true only if it
is independently true that we should function properly and/or align ourselves
with God’s intentions, and that we can judge our alignment through our prin-
ciple of reflection. Otherwise, we can hardly intelligibly be held accountable for
our proper functioning.
Of course, there is a notion of correct or proper functioning on which, if we
are designed to be governed by conscience, it simply follows that we function
properly only if we are so governed. But it is hard to see how any normative
practical reasons can follow from this. Any normative reason we could have to
realize proper functioning so understood must be judged independently of any
such function-relative norm. And Butler is committed to the view that we can
make normative judgments only by employing our principle of reflection itself.
Also of course, Butler holds that what determines our proper functioning is
not just any design, but that of an omniscient, omnibenevolent designer. That
there is normative reason to realize our proper functioning seems much more
plausible in light of that fact. But here again, any plausibility this normative
hypothesis has, Butler has to think, is something we can judge only by employ-
ing our principle of reflection. Ultimately, therefore, it is hard to see how any
argument from design – wise and benevolent, or otherwise – can work without
already assuming that we should govern ourselves by our principle of reflec-
tion. The reason this is so, again, is Butler’s claim that we are accountable for
following conscience. This engages the Pufendorfian point that we can be held
so accountable only if we can judge our accountability through conscience
itself.

https://doi.org/10.1017/9781139025065.007 Published online by Cambridge University Press

 the authority of conscience 195

Let us turn now to the transcendental strain in Butler’s thinking. The meta-
phor that Butler uses most frequently for our moral psychology is that of a
constitution. He clearly has in mind “constitution” not just in the generic
sense of what something is made of, but in the juridical/political sense of a
constitutional order. Viewed in this light, the thesis of conscience’s authority
has a distinctive juridical/deontic cast. It ties Butler’s claim to his themes of
self-government, being a “law to [one]self,” and to his contrast between “mere
power and authority” (II.8, 14) in a special way. A constitutional order deter-
mines the identity and form of government of a political state. It establishes
the possibility of the rule of law and what will count as the state’s judgment on
matters of law and governance.
A constitutional order requires something from which authority flows, and
for Butler, the seat of authority is the principle of reflection or conscience. It
is what makes possible a person’s making their own normative judgments
and guiding themselves by them, thereby being a “law to” themselves. Follow-
ing our distinction between narrow conscience, which makes deontic moral
judgment, and a broader principle of reflection, which makes judgments of
normative practical reasons in general, we may distinguish similarly between
Butlerian claims about necessary conditions for the possibility of autonomous
rational agency in general and for moral agency in particular.
Taking autonomous rational agency first, this just is an agent’s guiding
themselves by their own normative practical-reason judgments, which is only
possible if they have the capacity to make such (putatively dispassionate, disin-
terested, reflective) judgments through a principle of reflection. Such an agent
can hardly put in question whether they should follow their principle of reflec-
tion as their ultimate court of appeal without putting in question their very
capacity for reasoned action. They can of course doubt their judgment on par-
ticular matters in the sense that they can doubt that they are judging correctly.
But they can hardly ask, in general, what reason they have to do what they
themselves judge they have reason to do without fundamentally undermining
the rational self-trust necessary for reasoned agency. In this sense, following
the direction of their principle of reflection is just what reasoned agency is.
Once, however, we have the distinction between the broader principle of
reflection and narrow conscience, it would seem to be possible for a rational
agent to sensibly ask whether, and if so, why, they should follow their (narrow)
conscience. Butler claims that moral agency is impossible without conscience,
and this seems importantly correct. His point is not just that autonomous moral
agency is impossible without it, but that anything worth calling moral agency at
all is impossible without it. This is the claim he shares with Shaftesbury against
Hutcheson. Although there is undeniably something good about ingenuous
benevolence, a benevolent being who was not even capable of morally assessing
their own conduct or guiding themselves by that assessment would not really
be morally virtuous or even a moral agent yet.

https://doi.org/10.1017/9781139025065.007 Published online by Cambridge University Press

 196 hutcheson and butler

Essential to the capacity for moral agency, Butler holds, is accountability.

Unlike a “passive” mechanism like a watch, moral agents are accountable for
their own operations (Pr.14). Accountability for complying with deontic moral
norms, however, requires not just the ability to accord with the norms, but to
follow them in the sense of holding oneself to them. And this requires the capac-
ity to make and be moved by deontic normative judgments.
There is a minimal sense of rational agency, namely responsiveness to nor-
mative reasons, that does not require reflective normative judgment. And it
is of course true that rational agents, indeed moral agents, may better accord
with normative reasons, or even moral obligations, if they respond correctly to
the facts that are reasons (or their moral obligations), irrespective of their own
normative reasons (or obligation) (see, e.g., Arpaly 2004). But as Pufendorf
pointed out, someone can intelligibly be held accountable for discharging their
obligations only if they can hold themselves to them, and this requires, in But-
ler’s terminology, conscience.
But what is the relation between conscience and the principle of reflection?
Might not a moral agent judge themselves obligated to do something through
conscience, but still ask, using their principle of reflection, whether they have suf-
ficient normative reason to discharge their obligation? Although this is plainly a
more coherent question than whether they have reason to do what (they judge)
they have reason to do, there may nonetheless be features of accountable moral
agency that effectively close it in the way Butler clearly thinks it closed.
Recall Butler’s remark that “vice cannot be the happiness, but must upon the
whole be the misery, of such a creature as man …, an accountable agent” (Butler
2017: VII.16). The reason this is so, to the extent that it is, is due to the fact that
conscience is not merely the faculty of judgment in the sense of judging whether
something is the case. It is also the faculty through which we judge ourselves and
our actions in a deontic sense. And not just in the sense of estimating our posi-
tive and negative features, but in coming to what P. F. Strawson calls “reactive
attitudes” that express verdicts toward our actions, like guilt and blame. I think
it is fair to say that Butler would recognize his own doctrine of the authority of
conscience in the following passage from Samuel Clarke’s 1705 Boyle Lectures.

For the Judgment and Conscience of a Man’s own Mind … that his Actions
should be conformed to such or such a Rule or Law; is the truest or formall-
est Obligation. … For whoever acts contrary to this sense and conscience
of his own mind is necessarily self-condemned (Selby-Bigge: 1964: I.165).

The reason that acknowledged, unexcused wrongdoing can make moral agents
miserable is that they hold themselves accountable through the attitude of self-
blame or guilt, that is, by having a guilty conscience.
Narrow conscience is thus a faculty through which we make not just any
normative assessments of ourselves. It is one whose currency are distinctive

https://doi.org/10.1017/9781139025065.007 Published online by Cambridge University Press

 the right and the general good 197

(reactive) attitudes through which we hold ourselves accountable (Darwall

2006 and 2014b). But this means that the deliverances of (narrow) conscience
have implications for normative reasons, hence judgments of the (broad) prin-
ciple of reflection.
It is arguably a presumption of blame as a holding-accountable attitude that
there cannot be sufficient reason to perform culpable wrongs. It is a kind of
self-contradiction to blame someone but simultaneously to acknowledge that
they had sufficient reason to do what one is blaming them for (Darwall 2006:
98; Williams 1995: 40–44; Gibbard 1990: 299–300; Shafer-Landau 2003: 181–
183; Skorupski 1999: 42–43). But if that is so, then blame cannot be warranted,
an act cannot be culpable, unless there is, by virtue of its culpability, sufficient
reason not to perform it.
Even if such a line of thought can establish normative practical reasons to act
morally, or, in Butler’s terms, that vice is unnatural, it would not yet establish
that there actually is such a thing as morality and its distinctive obligating nor-
mativity. It would not yet, that is, fully respond to the Carneadean challenge that
morality might be chimerical. It would just mean that if there is such a thing as
moral right and wrong, then there is never sufficient reason to do what is wrong.
But where can skepticism about the right get a foothold that would not apply
equally to the good? Once we have the general concept of normativity that seems
implicit in both the good and the right, then any skepticism about normativity
in general will apply equally to both. Butler’s distinction between self-love and
the principle of reflection opens up the space to distinguish the general notion
of normative reasons. The dialectic should then naturally turn to examining the
way in which normativity enters into the concepts of good and right. An attrac-
tive view is Sidgwick’s, that the concept of the good involves that of (reason-
supported) justified desire. If the concept of right is tied to accountability, as
Butler seems to be supposing, then a complementary thought to Sidgwick’s is
that the right is therefore also tied to normative reasons, in this case, reasons
for the distinctive attitudes through which we hold one another accountable,
reactive attitudes such as guilt and blame (Gibbard 1990: 42). As Mill puts the
point, “We do not call anything wrong, unless we mean to imply that a person
ought to be punished for doing it … [for example], by the reproaches of his
own conscience” (Mill 2002: 5.14). Unless a skeptic can give some argument
for doubting the existence of normative reasons for the attitudes of conscience
that would not also apply to other attitudes, such as desire, then it would seem
that conscience and therefore morality’s authority is vindicated (Darwall 2016).

The Right and the General Good

Conscience, as Butler understands it, is unlike Hutcheson’s moral sense in hav-
ing an essentially practical role. The capacity to govern deliberation by consci-
entious normative judgment makes human beings “moral agents” and a “law

https://doi.org/10.1017/9781139025065.007 Published online by Cambridge University Press

 198 hutcheson and butler

to themselves” (Butler 2017: II.4, 8). For Hutcheson, moral sense always has
motives and motivated actions in view, seen from an observer’s perspective
(which an agent might of course adopt themselves). Butler’s conscience, on the
other hand, is essentially agential. Not only motives, but also “external actions”
themselves are objects of conscientious judgment (II.8). It is essential to moral
agency, as Butler conceives it, that moral agents are capable of deciding not to
do something on the grounds that, as they judge through conscience, it would
be wrong.
So far, we have two contrasts between Hutcheson and Butler that go together.
For Hutcheson, moral sense always has a motive as object, whereas for Butler,
we can also approve of actions (in circumstances) directly. And second, the per-
spective of moral sense for Hutcheson is that of an observer, whereas Butlerian
conscience is agential. Yet a third contrast flows from these two. Hutcheson
holds that all approved motives are desires for states of affairs in which natural
goods are realized. Butler, however, holds that among the “external actions”
that conscience approves are deeds considered not just for their naturally good
or evil consequences, but also on the basis of other right- and wrong-making
features of the acts themselves.
Thus, Butler rejects Hutcheson’s view that “virtue is resolvable into benev-
olence” (Butler 2017: D.8), not just on the grounds that genuine virtue must
involve moral self-government through conscience, but also because con-
science approves and disapproves of conduct on other, non-consequentialist
grounds.
The fact then appears to be, that we are constituted so as to condemn false-
hood, unprovoked violence, injustice, and to approve of benevolence to
some preferably to others, abstracted from all consideration, which con-
duct is likeliest to produce an overbalance of happiness or misery (D.8).20

Butler makes, moreover, a point that will loom large in defense of indirect utili-
tarian theories of right to follow, such as rule-utilitarianism, namely, that there
is no reason to think that it would maximize the general happiness for human
beings to be governed by a desire to maximize it. It may be that God is univer-
sally benevolent and “that He foresaw that this [non-utilitarian] constitute of
our nature would produce more happiness, than forming us with a temper of
mere general benevolence” (D.8). From this, Butler draws the same conclusion
as later indirect or rule-utilitarians: what we are obligated to do is not to pro-
mote overall net happiness but to follow dictates, the general internalization of
which has these good effects.

20
The similarity between this list and Ross’s examples of nonconsequentialist prima facie
duties is uncanny (Ross 2003).

https://doi.org/10.1017/9781139025065.007 Published online by Cambridge University Press

 against egoism 199

Against Egoism
In addition to his claims about the authority of conscience and his critique of
Hutcheson’s proto-utilitarian focus on benevolence, Butler is best known for
his “refutation” of psychological egoism. This is somewhat ironic since many of
the considerations Butler draws on are common to Shaftesbury and Hutcheson
also. However, Butler’s arguments are crisp and convincing and have proven
the most memorable.
Butler notes that it is tautologous, in one sense, that people always do what
they desire, but that this is impotent to show anything about the objects of their
desires. It is the latter, however, that is really in dispute between those who
hold that all human motivation is egoistic and those who deny this. Then But-
ler points out that whenever something makes someone happy or gives them
pleasure, this invariably involves the person’s having some positive attitude
to some particular thing that makes them happy or that gives them pleasure.
“There could be no enjoyment or delight from one thing more than another,
from eating food more than from swallowing a stone, if there were not an affec-
tion or appetite to one thing more than another” (XI.6). If I get pleasure from
eating pad Thai but not from swallowing stones or iPads, that must be because
I like the former experience and do not like the latter. It simply cannot be the
case that the only thing I like or desire is pleasure, conceived generically, since I
could not get pleasure from eating pad Thai unless I liked features of the experi-
ence that are intrinsic to it.
Like Hutcheson and Shaftesbury, Butler was concerned not just to show that
we have desires that are independent of self-love or a desire for our own good
or happiness. He also wanted to establish that we can desire things beyond our
own experiences, including the good of others and moral action. However, nei-
ther of the arguments mentioned so far can establish that. It is consistent with
them that all desires are for experiences of various kinds or, more generally,
for states of oneself. But Butler believed that psychological egoism can only
seem to be plausible when we think, mistakenly, as he has shown, that it is sup-
ported by some such confusion as the arguments we have already mentioned
unmask. And once we see that what is at issue is an empirical question, the
evidence will be overwhelming that people are capable of taking pleasure in, or
being made happy by, things other than their own experiences, for example, by
the happiness of others. Whether benevolence exists “is … to be judged of and
determined in the same way other facts or matters of natural history are.” By
that standard, “that there is some degree of benevolence amongst men, may be
as strongly and plainly proved in all these ways, as it could possibly be proved,
supposing there was this affection in our nature” (I.6n).
If the object of self-love can be accomplished only when we satisfy our
desires for specific objects, among which may be the good of others, we already

https://doi.org/10.1017/9781139025065.007 Published online by Cambridge University Press

 200 hutcheson and butler

have the result that a person’s interest may not be advanced to the degree that
they concern themselves with it. It will be realized only insofar as they realize
the objects of their various desires. Moreover, Butler argues, a person may have
too great a concern for their own good for their own good. Like the person at a
party who constantly asks, “Are we having fun yet?” they may be too focused on
being happy to enjoy things that would make them happy. “Disengagement is
absolutely necessary to enjoyment; and a person may have so steady and fixed
an eye upon his own interest, whatever he places it in, as may hinder him from
attending to gratifications within his reach, which others have their minds free
and open to” (XI.9). Butler refers to this as a kind of paradox (Sidgwick called
Butler’s point the “paradox of egoistic hedonism” (Sidgwick 1967: 136)). But it
is fundamentally the same kind of point Butler makes against universal benev-
olence. Just as universal benevolence, universally realized, is unlikely actually
to bring about the greatest overall happiness that is its object, so also is self-
love, realized in an individual agent, unlikely to bring about its object.21
Despite the relative brevity of his philosophical writings, Butler ended up
making several significant contributions to the history of modern moral phi-
losophy. In normative moral theory, Butler articulated what would become
standard objections to consequentialism, but he also made points, those noted
in the last paragraph, that would point the way to “indirect” consequentialist
theories. By his sharp conceptual distinctions, for example, between power and
authority, and his insistence on the independent normative authority of right
and conscience, Butler’s thought served to “work clear,” in Sidgwick’s terms,
the modern idea of the independence of right and good in a way that could
clarify the distinctively metaethical question of the sources of normativity in
general, as well as that of moral right in particular. Finally, Butler’s subtle and
careful treatment of human moral psychology and of the role of conscience
within it took important steps in the direction of articulating a conception of
accountable moral agency necessary for a modern moral philosophy.

21
This is what Parfit calls being “indirectly self-defeating” in Parfit 1986.

https://doi.org/10.1017/9781139025065.007 Published online by Cambridge University Press

 6

Hume and Smith

Like Butler and Hutcheson, David Hume (1711–1776) and Adam Smith (1723–
1790) were contemporaries. More, they were interlocutors, compatriots, and
friends, who, along with Hutcheson, helped shape the remarkable period
of intellectual activity in eighteenth-century Scotland known as the Scot-
tish Enlightenment. Hume and Smith inherited Hutcheson’s sentimentalist
approach: a form of moral empiricism that is opposed to ethical rationalism
and that continues to find resonance today. Hume’s version has had the great-
est influence; he is, of course, one of the greatest philosophers of all time.
Though the main lines of Hume’s critique of rationalism come straight from
Hutcheson, Hume pushed them farther and more memorably. It was Hume,
moreover, who placed moral philosophy most compellingly within an empiri-
cist naturalist framework.
We also owe to Hume a formulation of perhaps the most fundamental prob-
lem of metaethics, what Michael Smith calls the “moral problem” (Smith 1994).
Moral judgments, Hume says, “go beyond the calm and indolent judgments
of the understanding” in seeming to be intrinsically motivating (Hume 1978:
457). When someone’s expressed moral or ethical judgments leave them utterly
unmoved, we doubt their judgments’ sincerity. But Hume famously held that
only desires can motivate. According to the “Humean theory of motivation,”
belief and reason are motivationally inert (Smith 1994). Nonetheless, ethical
judgments also seem belief-like in being subject to genuine disagreement con-
cerning their correctness or truth. The metaethical challenge is how to recon-
cile these three things.
Hume raises a related issue in his famous “is”/“ought” passage, which ques-
tions how reasoning from non-normative premises formulated with “is”’s
can arrive at conclusions expressed with “ought”’s. This “ought, or ought not,”
Hume writes, “expresses some new relation or affirmation” (Hume 1978:
469). This leads to the question: What is the “source” of “ought”’s “normativ-
ity?” as Christine Korsgaard puts it (Korsgaard 1996a). Following Hume, it
is often thought that normative judgments necessarily play a role in guiding
the attitudes they normatively assess in a way that non-normative judgments
do not.
201

https://doi.org/10.1017/9781139025065.008 Published online by Cambridge University Press

 202 hume and smith

For his part, Adam Smith is, of course, best known for his writings on eco-
nomics. But Smith’s Theory of Moral Sentiments is arguably one of the greatest
works on moral psychology ever written, whose moral philosophical signifi-
cance may only now be becoming clear (Fleischacker 1991, 2013; Griswold 1999;
Darwall 2004, 2006). Smith shows the ubiquity of imaginative perspective-
taking in our mental moral lives, both in attributing mental states to others
(getting “inside their heads”) and in everyday normative judgments of the fit-
tingness, or “propriety” as Smith calls it, of attitudes to the objects they have in
view.
For example, we often grasp that another person is angry and what they are
angry at by imaginatively placing ourselves in their perspective and imagining
what is in view from that standpoint and how to respond to it. Determining
what their state of mind is is not itself a matter of ethical or normative judg-
ment. But we do also often judge whether someone’s response to some object
of thought is a justified or “proper” one. According to Smith, this is something
we do by an imaginative projection into the other’s “shoes” – though the pro-
jection is regulated impartially (with Smith’s “impartial spectator”), so that we
enter into their shoes not as them, or even as ourselves, but as someone in their
situation. If we share their response after such an impartially regulated pro-
jection – in Smith’s terms, “sympathize” or have “fellow-feeling” with it – we
judge their response proper or justified in the sense of being fitting to its object.
Smith makes an important connection between fellow-feeling and mutual
respect and accountability. Our capacity to take on others’ perspectives and
regulate our conduct toward them from an informed and impartial second-
personal point of view figures centrally both in Smith’s account of justice and
in his connection between the human capacity for unforced agreement (in par-
ticular, free economic exchange) and mutual respect.
This turns out to be central to solving what has been called the “Adam Smith
Problem”: How can Smith both praise the free market’s “invisible hand” for
turning self-interestedly motivated conduct into public benefits, and also be as
concerned as he is with moral sentiments? How could the same mind have pro-
duced both The Wealth of Nations and The Theory of Moral Sentiments?
The answer is that Smith thinks that these are inextricably connected. In the
very section of The Wealth of Nations in which Smith attributes the division of
labor on which national wealth depends to the human capacity for “truck, bar-
ter, and exchange,” he remarks that this capacity makes possible a relation of
mutual respect (Grotius and Pufendorf’s “sociability”) between human beings
that other species cannot have (Smith 1976: 25).
As we will see, Smith echoes Pufendorf’s connection between sociability,
mutual respect, equal dignity, and mutual accountability. “A moral being is an
accountable being,” Smith writes (1982: 111). When another person injures us,
Smith says, what “chiefly enrages us … is the little account he seems to make

https://doi.org/10.1017/9781139025065.008 Published online by Cambridge University Press

 hume 203

of us, the unreasonable preference which he gives to himself over us” (96). We
are outraged by the violation of our dignity and what Kant will call the other’s
“self-conceit” (Kant 1996b: 5: 73–77).1 As with Kant, this focus on individual
dignity gives Smith a principled way of opposing the imposition of costs on
some for the greater benefit of others. Here, Smith writes, our concern is rightly
not for overall utility, but for the “very individual who [would be] injured”
(Smith 1982: 90).
Hume is more equivocal on these matters. Like Smith, and unlike Hutcheson,
he stresses the contrast between justice and public utility in individual instances,
but argues that the moral foundation of justice is ultimately utility nonetheless.
Hume lacks Pufendorf’s or Smith’s focus on equal dignity or mutual respect
and any related emphasis on autonomy and accountability. Indeed, he also
rejects Hutcheson’s sharp separation of moral evaluation from other forms of
esteem. What leads Hume to these positions is his “experimental” approach,
which he pursues from the perspective of an “anatomist” concerned to discover
the mind’s “most secret Springs and Principles” (Hume 1932: I.32–33; 1978: xi,
620).2 Frequently, when Hume uses “moral philosophy,” it is to refer to empiri-
cal moral psychology, analogously to “natural philosophy”’s early modern ref-
erence to the emerging natural sciences.
At the same time, Hume is also a moral philosopher in our contemporary,
normative sense. But even here, his approach is decidedly tied to an observer’s
third-person perspective rather than the first-person standpoint of a deliberat-
ing agent or the second-person point of view of mutually accountable moral
agents.3

Hume
Hume’s empiricist naturalism is on display from outset of A Treatise of Human
Nature, the first two volumes of which Hume published in 1739 at the age of
twenty-eight. Its subtitle announces the Treatise as “an attempt to introduce
the experimental method of reasoning into moral subjects” (Hume 1978: xi).
Hume’s ambition, it is often said, was to be the “Newton of the moral sciences”
(Passmore 1968: 131).4 Although Hutcheson also broadly shared this approach,
Hume’s Newtonianism led him to deny two central tenets of Hutcheson’s

1
On this element of Smith, see Debes 2012.
2
Although Hume’s contrast between the approach of the painter and the anatomist is famil-
iar from the Treatise, this characterization comes from a letter to Hutcheson. On this, see
Immerwahr 1991; see also Abramson 2007.
3
For an argument that Hume can provide a first-personal account of moral agency, see
Abramson 2002.
4
See, however, McIntyre 1994 for an argument that Samuel Clarke was actually an earlier
contender for this title.

https://doi.org/10.1017/9781139025065.008 Published online by Cambridge University Press

 204 hume and smith

moral sentimentalism. These were, first, Hutcheson’s hypothesis of a dedi-

cated moral sense, what Hume termed an “original instinct,” and second,
Hutcheson’s claim that the object of moral sense, moral goodness or virtue, is
fundamentally distinct from other estimable qualities to which we respond in
non-moral ways. Hume argued on Newtonian grounds that a simpler expla-
nation of approbation and disapprobation is provided by all-purpose mecha-
nisms of sympathy and the association of ideas that are in play throughout our
psychology. No independent moral “sense” is required. Further, Hume argued
that any distinction between “moral virtues” and estimable “natural abilities” is
purely verbal (Hume 1978: 606).
Hume’s ethics are, like Hutcheson’s, an ethics of virtue, but Hume denied
the tenet that made Hutcheson’s a modern moral virtue ethics in contrast to
Aristotle’s. Like Shaftesbury, Hutcheson rejected the juridical model of the
modern natural lawyers. But he nonetheless insisted against ancient ethi-
cal theorists that moral good and evil differ fundamentally from other things
the ancients and their followers called virtues and vices, like temperance and
gluttony. Only the former were objects of the distinctively moral approval
and disapproval that he called “approbation” and condemnation, respec-
tively. Moreover, Hutcheson held that all moral goodness consists in forms of
benevolent concern for human (non-moral) good. And he argued that because
universal benevolence is the morally best motive, and because universal benev-
olence aims at the “greatest happiness of the greatest number,” this goal pro-
vides the standard of moral choiceworthiness. These ideas led him to embrace
one of Sidgwick’s marks of “modern moral philosophy,” namely, that universal
benevolence is no less a source of normative reasons than is egoistic concern
for the agent’s own good. And this led naturally, in Sidgwick and later writers,
to utilitarian and consequentialist theories of moral conduct more generally.
According to Hutcheson, moral good and evil are qualities the contempla-
tion of which gives rise to irreducible moral feelings or “ideas” of approbation
and condemnation, respectively. Hume denied that any such distinctive feel-
ings requiring an “original instinct” or “sense” exist, however. Sympathy and
the association of ideas can explain our human contemplative responses quite
generally, and what Hutcheson called moral goodness is only one of a wide
variety of things that evoke favorable disinterested contemplative response.
Any distinction between “moral virtues” and “natural abilities” like wit, Hume
said, is only “verbal.” Finally, Hume also denied that benevolence enjoys any-
thing like the priority even among moral virtues that Hutcheson claimed.
Hume agreed with Butler that virtues like justice and fidelity can conflict with
benevolence and sometimes trump it.
Hume’s appreciation of the differences between benevolence and justice, and
of the latter’s special connection to problems of collective action that can be
solved only through mutually advantageous practices and institutions in which

https://doi.org/10.1017/9781139025065.008 Published online by Cambridge University Press

 hume 205

participants do their fair share, marked an important contribution to the his-

tory of ethical thought. Hume showed that just practices are vulnerable not just
to free-riding “sensible knave[s]” who wish to advance their self-interest, but
also to unjust benevolence, as when someone seeks to advance the public good
rather than repay a loan to a “seditious bigot” (1975a: 282–283; 1978: 497).
It is essential to Hume’s view that justice is established by conventions or
rules that are not simply beneficial overall but mutually advantageous in the
sense that their benefits extend to pretty much everyone who is bound by
them.5 Even so, Hume’s account of what he calls the “moral obligation” of jus-
tice is in terms of public utility, which sympathy and the association of ideas
turn into observers’ approbation. This means that for Hume it is the ben-
eficial consequences of justice overall, though not in every specific case, that
grounds its moral value. This gives Hume an especially interesting relation to
utilitarianism.
Hume is not a utilitarian in the sense of Bentham and Mill. To begin with,
Hume is a virtue theorist and uninterested in articulating what Mill will call a
“criterion of right and wrong” (Mill 2002: I.1). This he shares with Hutcheson.
However, Hutcheson, unlike Hume, does put forward a standard of moral
choiceworthiness, and that standard is, moreover, a utilitarian one: Moral sense,
Hutcheson says, “regulate[s] our Election” by preferring acts to the degree that
they “procur[e] the greatest Happiness for the greatest Numbers” (Hutcheson
2004: 125). Even so, Hutcheson does not accept a consequentialist, much less
a utilitarian, theory of virtue. The moral goodness of benevolence is intrinsic
and fundamental, irrespective of its consequences, as is evident to moral sense.
Broadly speaking, however, Hume is a consequentialist in these respects. It
is our sympathy with the good effects of virtuous traits and the practice of jus-
tice that leads us to approve of them. This was not lost on Bentham. Bentham
wrote that when he read the Treatise, he “felt as if the scales had fallen from
[his] eyes”; he “learned to see that utility was the test and measure of all virtue”
(Bentham 1838: 268n).6
As we noted in the last chapter, Butler’s objections to Hutcheson’s claim
that “virtue is resolvable into benevolence,” that justice, fidelity, and truth tell-
ing are moral obligation-making independently of their consequences, would
become stock counterexamples to utilitarianism for later critics. There was,
though, a further issue regarding whether moral goodness can be reduced to
benevolence that concerned Butler’s doctrine of the authority of conscience.
Here Butler held, as against Hutcheson, that genuine moral virtue involves

5
Hume restricts, however, just practices to those having a roughly equal antecedent stand-
ing. I will discuss a passage in which this becomes starkly clear below (Hume 1985: 190). For
a discussion of Hume’s racism, see Garrett 2004.
6
For more discussion, see Darwall 1994.

https://doi.org/10.1017/9781139025065.008 Published online by Cambridge University Press

 206 hume and smith

moral autonomy: self-guidance by the moral agent’s own conscientious moral

judgments. And Butler held, again, that ordinary moral agents’ judgments are
not utilitarian.
This illustrates what would become a deep difference between utilitarians
and their opponents. Like Butler, deontological critics of utilitarianism gener-
ally give a central role to moral autonomy. They frequently see self-direction as
essential both to a morally good life – as in Kant’s doctrine that genuine moral
worth involves conduct resulting from agents’ own conception of their moral
duty – and as a source of non-utilitarian duties and rights – for example, rights
to lead one’s own life and to be treated, as Kant says, never simply as means but
always as an end – and the correlative moral duties these rights entail.
It is significant, in this connection, that Hume gives so little weight to
moral autonomy. Indeed, when Hutcheson, impressed by Butler’s defense of
the authority of conscience, attempted to give it some role in his later works
(e.g., in System of Moral Philosophy (Hutcheson 1755)), albeit still holding that
moral sense or conscience is motivationally inert, Hume criticized him.7 Hume
wrote to Hutcheson: “You seem here to embrace Dr. Butler’s Opinion” that
conscience “has an Authority distinct from its Force and Durableness, and that
because we think it ought to prevail” (Hume 1932: I.47). Against this, Hume
urged that conscience or moral sense is simply like any “Instinct or Principle”
in “approv[ing] of itself upon reflection” (I.47). If anything, then, Hume down-
plays moral autonomy even more than Hutcheson.

Against Practical Reason

Hume mounts his major metaethical arguments, first, against rationalism, and
second, for moral sentimentalism, at the beginning of Book III of the Treatise.
However, a central premise in these arguments, that neither actions nor pas-
sions can be “contrary to truth or reason” (1978: 415), is argued for in II.ii.3
(“Of the Influencing Motives of the Will”). “Nothing is more usual in philoso-
phy,” Hume says, than “talk of the combat of passion and reason,” including
by philosophers, like Butler, who say that every motive is to be brought into
“conformity with that superior principle” (413). Against this, Hume announces
that he will argue for two claims: (a) “reason alone cannot be a motive to any
action of the will,” and (b) reason “can never oppose passion in the direction of
the will” (413).
The first is what Michael Smith calls the “Humean theory of motivation,”
though Hume generally prefers to say that reason rather than belief is impo-
tent to motivate. (On Hume’s theory of reason, these come to the same thing.)

7
Hutcheson’s System was published posthumously in 1755. Hume read an earlier copy that
circulated among friends. For discussion, see Darwall 1995: 237–242.

https://doi.org/10.1017/9781139025065.008 Published online by Cambridge University Press

 against moral rationalism and for moral sentimentalism 207

Hume recognizes two kinds of reasoning: demonstrative reasoning concern-

ing the “relations of our ideas” and probabilistic reasoning about “causes and
effects” (413–414). Reasoning of either kind can directly affect only our beliefs.
It is only when beliefs are combined with desires that action is motivated.
And when this happens, it is passion or desire that provides the motivational
impetus.
It appears to follow immediately that in one sense reason and belief “can
never oppose passion”: They cannot counter passion’s motivational impetus.
But Hume also argues that reason cannot oppose passions and actions in a dif-
ferent way; passions and actions can neither be “contrary or conformable to
reason” since they cannot themselves be reasonable nor unreasonable.
This follows rather directly from Hume’s conception of reason, which he
tells us in Book III is “the discovery of truth and falsehood” (485). It is a conse-
quence that only what can be true or false can be reasonable or unreasonable.
Passions and actions have no “representative quality”; unlike ideas, including
those with the “vivacity” to be beliefs, passions do not purport to be “copies” of
“objects which they represent” (97, 415). A belief is “contradictory to truth and
reason” if it is in “disagreement” “with those objects which [it] represent[s]”
(415). Actions and passions cannot be unreasonable in this way.
Hume allows that we sometimes call passions and actions “unreasonable”
when they are accompanied by or based on false beliefs, but it is really only
the beliefs that are unreasonable “strictly and philosophically” (415–416). “’Tis
not contrary to reason,” he concludes, “for me to chuse my total ruin, to pre-
vent the least uneasiness of an Indian or person wholly unknown to me. ’Tis as
little contrary to reason to prefer even my own acknowledg’d lesser good to my
greater, and have a more ardent affection for the former than the latter” (416).

Against Moral Rationalism and for Moral Sentimentalism

Having made these arguments about reason and the will in Book II, Hume
begins Book III by arguing on their basis that “Moral Distinctions” are “not
deriv’d from Reason” (455). Although he gives a battery of other considerations,
the central arguments against metaethical rationalism depend on these two.
If it is impossible for passions and actions to be in accord with or contrary to
reason, then a passion or action can hardly be against morality because it is
against reason. If reason just is “the discovery of truth and falshood” (458), then
reason cannot dictate passions and actions. Therefore, morality cannot dictate
an action or passion on the grounds that reason does.
Obviously, this Humean argument depends entirely on Hume’s definition of
reason, which even many modern-day Humeans would reject as overly narrow,
since they often want to hold that there is such a thing as instrumental irratio-
nality: failing to take necessary means to ends. Hume himself held that even

https://doi.org/10.1017/9781139025065.008 Published online by Cambridge University Press

 208 hume and smith

such irrationality is not against reason, “strictly and philosophically” (415).

At best, it typically involves irrational belief. Kant will later argue that reason
imposes formal constraints on practical reasoning: “hypothetical impera-
tives” of instrumental rationality, but also “categorical imperatives” that are
grounded in the conditions of free rational agency. For this dispute between
Hume and his critics to be more than definitional, it must be set in the context
of a whole network of substantive philosophical issues concerning the nature of
practical reason, deliberation, and the will.
By far Hume’s most influential argument against metaethical rationalism
depends on his claim that ethical judgments “go beyond the calm and indolent
judgments of the understanding” (457) together with the Humean theory of
motivation advanced in Book II. The direct argument from his definition of
reason attempts to show that morality cannot be derived from reason meta-
physically, that moral dictates are neither grounded nor consist in dictates of
reason. Reason can only dictate belief and not action. The motivational argu-
ment attempts to show that neither can moral distinctions derive from rea-
son epistemically. Moral knowledge and judgment go “beyond” anything with
which reason or belief provides us, since moral judgments can move us directly.
For this argument to establish its conclusion, it must hold not just that there
is some connection between moral judgment and motivation, but that the lat-
ter is intrinsic to the former. Otherwise, moral judgments might be made by
reason with motivation being provided by some passion that is connected to it,
perhaps even necessarily. Kant himself actually agrees that motivation always
requires desire; he just holds that some desires derive from reason (“pure prac-
tical reason”).8
Ethical rationalists like Samuel Clarke held that through reason we can per-
ceive relations of “fitness” between actions and circumstances (e.g., that grati-
tude is a fitting response to benevolence). Hume argued that this rationalist
position faced the challenge of proving “a priori, that these relations, if they
really existed and were perceiv’d, wou’d be universally forcible and obliga-
tory,” by which he meant motivating (466). Not all rationalists have accepted
this internalist challenge – prominently, Ross does not – but the eighteenth-
century rationalists usually did, as we shall later see.
It is not clear, indeed, exactly what connection between ethical judgment
and motivation Hume himself accepts (Brown 1988). His challenge to the ratio-
nalists that they must show an a priori connection to motivation is rooted in
the fact that they hold morality to be “eternal and immutable,” in Cudworth’s
phrase. Since Hume does not hold this view, he need not face this challenge.
Hume holds that it is contingent psychological human processes of sympathy

8
“Desire is the faculty to be, by means of one’s representations, the cause of the objects of
these representations” (Kant 1996c: 6:211; see also Kant 1996b: 5:9).

https://doi.org/10.1017/9781139025065.008 Published online by Cambridge University Press

 against moral rationalism and for moral sentimentalism 209

and the association of ideas that make ethical judgments motivating. Whether
this motivation is intrinsic to ethical judgment depends on exactly what Hume
takes such judgments to be, and this is, at best, a complicated matter. Whatever
the details of Hume’s own view, there is no question that, as Michael Smith’s
formulation of “the moral problem” makes clear, the claim that motivation
is intrinsic to moral judgment has had a powerful influence in contempo-
rary metaethics, leading many philosophers to forms of anti-realism, whether
expressivist or error-theoretic (e.g., Stevenson 1937; Mackie 1977).
Again, Hume holds that reasoning can only establish analytic propositions
about “relations of our ideas” or “matters of fact” through probabilistic causal
reasoning (Hume 1978: 413, 468). No substantive ethical proposition is ana-
lytic, so that means reasoning can establish ethical propositions only if they are
“matters of fact.” And here, Hume invites us to consider how ethical judgment
actually works. Of course, it must begin with “matters of fact” concerning the
object we are judging. But that does not mean the judgment with which we
conclude itself concerns a matter of fact.
Take any action allow’d to be vicious; Wilful murder, for instance. Examine
it in all lights, and see if you can find that matter of fact, or real existence,
which you call vice. In which-ever way you take it, you find only certain
passions, motives, volitions, and thoughts. There is no other matter of fact
in the case. The vice entirely escapes you, as long as you consider the object.
You never can find it, till you turn your reflexion into your own breast,
and find a sentiment of disapprobation, which arises in you, towards this
action (468).

When we make ethical judgments, it is on the basis of what we take to be “mat-

ters of fact” concerning the thing we are judging. Our concluding judgment
cannot then be just another such fact; it is a further judgment that “go[e]s
beyond” the facts.
So far, this should be uncontroversial. Indeed, the irreducibility of nor-
mative ethical judgments to non-normative “matters of fact” on whose basis
ethical judgments are made is something on which ethical rationalists them-
selves generally insist. A conceptual distinction between the normative and
the non-normative is implicit in Butler’s distinction between power and
authority and, later, in Richard Price’s claim that “right” and “wrong” express
“simple ideas” (Price 1974: 41). It is even clearer, of course, in Sidgwick on the
irreducibility of “ought” and Moore on the “naturalistic fallacy” (Sidgwick
1967: 25; Moore 1993).
When, therefore, Hume famously proclaims a conceptual gap between
claims expressed with “is” and those expressed with “ought,” this is com-
mon ground with his rationalist opponents. Where Hume departs in a way
that he believes rationalists cannot follow is in holding that moral judgments
“go beyond the calm and indolent judgments of the understanding” because

https://doi.org/10.1017/9781139025065.008 Published online by Cambridge University Press

 210 hume and smith

the former have, and the latter lack, motivational impetus. It is because ethical
judgments express or somehow otherwise involve sentiment that there is no
rational path to them from beliefs about matters of fact.
Hume’s sentimentalism has been a source of inspiration to recent philoso-
phers who have developed a variety of anti-rationalist metaethical views, from
naturalist versions of subjectivism and constructivism, to forms of expressiv-
ism and error theory (e.g., Brandt 1979; Railton, 1986; Dreier 1990; Street 2008;
Stevenson 1937; Blackburn 1984; Gibbard 1990; Mackie 1977). The differences
between these later views depend on subtle distinctions – for example, between
expressing and attributing sentiments – that Hume himself does not make or,
at least, consistently observe. It is probably ahistorical folly to attempt to deter-
mine exactly what metaethical position Hume held in any fine-grained way.
For instance, right after the passage quoted above, Hume writes, “So that
when you pronounce any action or character to be vicious, you mean noth-
ing, but that from the constitution of your nature you have a feeling or senti-
ment of blame from the contemplation of it” (Hume 1978: 469). Taken literally,
this is a version of subjectivism: ethical judgments self-attribute sentiments
(Dreier 1990). However, Hume also says that “[w]e do not infer a character to
be virtuous, because it pleases: But in feeling that it pleases after such a particu-
lar manner, we in effect feel that it is virtuous” (1978: 471). This runs against
subjectivism. It suggests something closer to the projectivist or expressivist
account of value judgment that we found in Hobbes. Similarly, in the second
Enquiry, Hume contrasts reason-based judgments of the understanding and
ethical judgments in the following terms:
The former conveys the knowledge of truth and falsehood: the latter gives
the sentiment of beauty and deformity, vice and virtue. The one discovers
objects as they really stand in nature, without addition or diminution: the
other has a productive faculty, and gilding or staining all natural objects
with the colours, borrowed from internal sentiment, raises in a manner a
new creation (1975a: 294).
This also suggests a projectivist or expressivist view (Blackburn 1984: 210–212).
However its details are developed, Hume clearly espouses some version
of metaethical sentimentalism in opposition to any form of rationalist real-
ism. As we noted earlier, this was common ground with Hutcheson, even if
Hutcheson tended more single-mindedly to metaethical subjectivist natural-
ism than Hume with his expressivist leanings. Hume titles the main sentimen-
talist section of the Treatise (III.i.2) in explicitly Hutchesonian terms: “Moral
Distinctions Deriv’d From a Moral Sense.” However, just after delimiting
moral sentiment as of the “peculiar kind which makes us praise or condemn,”
suggesting Hutcheson’s doctrine of the moral sense and its “simple,” irreduc-
ibly moral “ideas” of approbation and condemnation, Hume makes a turn to
explicitly reject Hutcheson’s view (1978: 472).

https://doi.org/10.1017/9781139025065.008 Published online by Cambridge University Press

 against moral rationalism and for moral sentimentalism 211

“From what principles,” he then asks, does moral sentiment derive (473)?
What is the underlying psychological mechanism? Hume’s Newtonianism leads
him to seek the simplest explanation. Hutcheson thought he had a sufficiently
simple explanation since he thought the object of approbation is always some
form of benevolence; a dedicated sense that approves of it is simple enough.
But, like Butler, Hume rejected Hutcheson’s normative theory of moral good-
ness, if only because justice can conflict with benevolence no less than it can
with self-interest. There are many different kinds of things we approve, and
“’tis absurd to imagine, that in every particular instance, these sentiments are
produced by an original quality and primary constitution” (473). Any hypoth-
esis of a dedicated “moral sense” must thus be suspect on Newtonian grounds.
Moreover, Hume rejected Hutcheson’s bright line between moral good-
ness and estimable “natural abilities” such as wit. He thought he had a simpler
explanation that could encompass all our approvals and disapprovals without
distinguishing between moral and non-moral approbation. The materials of
this explanation were quite general phenomena of “sympathy” (in Hume’s spe-
cial sense) and the association of ideas that are involved throughout our mental
lives.
What Hume calls “sympathy” is a psychological process through which an
idea of a feeling or passion is “converted into an impression, and acquires such
a degree of force and vivacity, as to become the very passion itself” (317). Sym-
pathy, for Hume, is much like what psychologists these days call “emotional
contagion” in which we catch feelings from others, but not by taking their
perspective or considering the object that others’ feelings have in view. “The
human countenance, says Horace, borrows smiles or tears from the human
countenance” (1975a: 220). This example just is emotional contagion: “catch-
ing” others’ feelings in their presence. The mechanism of sympathy at work in
Hume’s account of the moral sentiment is, however, somewhat different, since
it can be employed in solitary thought from the armchair.
Hume’s mechanism, again, transforms an idea of a feeling into the feeling
“itself.” For example, if I have the idea of someone feeling pleasure, sympathy
then transmutes that idea into a pleasant feeling (impression) in me. The senti-
ments implicated in ethical judgment are felt impressions, and it is Humean
sympathy that brings these feelings into our mental economy in making ethical
judgments. The rest of the work is done by the association of ideas, a general
mechanism that takes us from idea to idea, when these are related by “resem-
blance, contiguity, and causation” (1978: 60). Putting these two mechanisms
together provides a significantly simpler explanation of our moral sentiments,
Hume argues, than any that can be provided by the hypothesis of a moral sense.
Hume’s explanatory theory also ends up valorizing a broadly consequential-
ist, almost utilitarian, normative theory of virtue and vice. As the chapter titles
of the second Enquiry make clear, Hume believes that all virtues are qualities

https://doi.org/10.1017/9781139025065.008 Published online by Cambridge University Press

 212 hume and smith

that are either “immediately agreeable” or “useful” to ourselves or to others

(1975a: 233, 250, 261). To see how his psychology of moral sentiment works,
consider benevolence. When we contemplate the motive or trait of benevo-
lence, our thoughts are carried by an association of ideas to the normal and
usual consequences of benevolence, that is, the pleasurable states of mind
of those who benefit from actions motivated by it or by directly pleasurably
engaging with the benevolent person. Sympathy then transforms ideas of these
pleasurable states into pleasant feelings in us, which we associate back with the
benevolent motives or traits we are contemplating. We express these pleasant
contemplative feelings when we judge the motive or trait of benevolence itself
to be good or virtuous.
There are some important wrinkles in Hume’s view. First, Hume holds
that our judgments involve less the actual consequences of some motive or
trait on particular occasions than their normal consequences. “Virtue in rags
is still virtue” (1978: 584). Second and more importantly, because we tend to
sympathize more with people who are psychologically closer to us, we have
to “correc[t] our sentiments, or at least … our language” by fixing on some
“steady and general points of view” and thinking and speaking as though we
occupied them (581–582). “[N]otwithstanding [the] variation of our sympa-
thy, we give the same moral approbation to the same moral qualities in China
as in England” (581).
We “correct” the sentiments that sympathy and the association of ideas
themselves provide by “plac[ing] ourselves” in a common “general” point of
view. And we correct our language by “apply[ing] the terms expressive of our
liking or dislike … as if we remain’d in one point of view. Experience soon
teaches us this method of correcting our sentiments, or at least, of correcting
our language” (582).
In the Treatise, Hume’s perhaps idiosyncratic view of how we do this is by
fastening on the perspective of someone in “that narrow circle, in which [the
agent who is the object of our judgment] moves” (602). In the Enquiry, he
offers a more social-constructionist proposal. There he speaks of “render[ing]
our sentiments more public and social” (1975a: 229). “The intercourse of sen-
timents, therefore, in society and conversation, makes us form some general
unalterable standard, by which we may approve or disapprove of characters
and manners” (229).
In public moral discourse, sympathy helps supply a general standard that
can solve the problem that “variation … of sympathy” in private moral judg-
ment itself causes. Moral discussion expresses sentiments, and interlocutors
sympathetically reverberate with one another’s feelings. We can think of moral
discussion of morals, then, as a kind of tuning of a large social orchestra, where
what brings the various instruments into tune is their reverberation with one
another.

https://doi.org/10.1017/9781139025065.008 Published online by Cambridge University Press

 justice, the artificial virtue 213

Justice, the Artificial Virtue

Hume’s ethics are, like Hutcheson’s, an ethics of virtue. Both are opposed to the
modern natural lawyers’ focus on conduct and obligation. And both carry for-
ward Shaftesbury’s subversive reconceptualizing of obligation in non-deontic
terms. When Shaftesbury asked, “What Obligation there is to Virtue” (Shaftes-
bury 2001: II.78), he was inquiring what normative reason we have to be virtu-
ous (and assuming that any such reasons would have to be self-interested). To
this, Hutcheson added the non-eudaimonist premise that any person’s good
can ground normative reasons, not just the agent’s. This, again, was his “mod-
ern turn.” And he added a sense of “obligation” consisting in the deliverances
of moral sense. Hume follows Hutcheson, holding that, in addition to the “nat-
ural obligation” of self-interest, moral sentiment provides a “moral obligation”
(Hume 1978: 497, 498). This becomes clearest in Hume’s discussion of justice,
to which we now turn.
In one way, Hume goes farther in the direction of a virtue ethics than
Hutcheson does. Though Hutcheson downplays questions of morally right
and wrong conduct, and though virtue or moral goodness is for him, in the
first instance, anyway, a matter of motive and not of conduct, Hutcheson does
recognize and theorize a kind of morally good conduct, namely, the morally
choiceworthy, considered independently of its motive. As we saw, this is where
he formulates his version of the greatest happiness principle. Hume does not
follow him in this. At the beginning of Treatise III.2, on justice, Hume writes,
“[W]hen we praise any actions, we regard only the motives that produced
them, and consider the actions as signs or indications of certain principles in
the mind and temper. The external performance has no merit” (477).
Hume’s analysis of justice in terms of mutually advantageous rule-­governed
cooperative practices is path breaking and brilliant. However, it is important
to bear in mind, as the passage just quoted makes clear, that his primary focus
is on the virtue of justice, and this, for Hume, must consist in some kind of
motive. Nevertheless, justice differs fundamentally from other motives, like
benevolence, in that its object is a kind of conduct considered independently
of its consequences, rather than some result, like someone’s natural good,
whether the agent’s, like self-interest, or someone else’s, like benevolence.
Hume distinguishes between virtues he calls “natural” and those he calls
“artificial.” The distinction is roughly this: artificial virtues require a back-
ground of conventional rule-governed practices in order to be realized; the nat-
ural virtues do not. In holding that there are both natural and artificial virtues,
Hume places himself between Hobbes and Hutcheson. As against Hobbes, and
with Hutcheson, Hume holds that there can be virtue or moral goodness in a
state of nature. Benevolence is to be approved of whether or not there exists
law or custom. But as against Hutcheson, and with Hobbes, Hume holds that

https://doi.org/10.1017/9781139025065.008 Published online by Cambridge University Press

 214 hume and smith

there are moral traits that do require human artifice, and that do not reduce to
benevolence in any of its forms – namely, justice. This latter claim is an echo of
Butler, with a distinctly Hobbesian twist.
Since virtues for Hume, as for Hutcheson, consist in virtuous motives, Hume
argues for the artificiality of the virtue of justice by arguing that no reliable
motive to justice “proceeds immediately from original principles,” like benev-
olence or self-love, “without the intervention of thought or reflexion” (484).
Neither self-love nor “regard to publick interest” leads us to be just directly.
As far as the former is concerned, “’tis certain that self-love, when it acts at its
liberty, instead of engaging us to honest actions, is the source of all injustice and
violence” (480). Regarding the latter, Hume gives several reasons why regard
for the public interest cannot be the motive to just acts. One is that, like self-
interest, the public interest can conflict with justice in particular cases. Thus a
loan “to a seditious bigot” might be secret and its being repaid be contrary to
the public interest, however much justice requires its repayment (497). Second,
we have but limited generosity, and “experience sufficiently proves that men, in
the ordinary conduct of life, look not so far as the public interest, when they pay
their creditors, perform their promises, and abstain from theft, and robbery,
and injustice of every kind” (481).
The virtue of justice thus cannot be identical either with self-love or with
benevolence, since neither reliably motivates just action, even, indeed, when
rules of justice have been conventionally established. But what about the moral
sentiment? Hume thinks that does occur naturally. And he certainly thinks
we approve of justice – otherwise we could not judge justice a virtue. Even so,
Hume holds that there is a conceptual bar to the moral sentiment being, as he
puts it, the “first motive” to just acts. The reason, again, is that the “external per-
formance” of a just act can have no “merit” in itself, since it derives any merit
it has from its motive (477). Therefore, “[t]o suppose, that the mere regard to
the virtue of the action, may be the first motive, which produc’d the action, and
render’d it virtuous, is to reason in a circle” (478).
Like Hutcheson, Hume holds that since the moral sentiment approves of
motives, it is impotent to motivate any act itself, any choice of a thing to do,
considered independently of motivation. Hutcheson didn’t think, however,
that justice posed any special problem in this regard, since he thought that jus-
tice and the public interest never conflict, so the virtuous motive of benevo-
lence is sufficient to motivate just acts. No separate virtuous motive of justice is
required. Hume, however, denied this.
A sense in which the virtue of justice, unlike benevolence, is artificial,
according to Hume, is that we only require such a virtue because of contingent
features of human life that make necessary the conventional establishment of
rule-governed practices of property, promise, and contract. If human generos-
ity were unlimited, or human goods superabundant, we might have no need
for such practices and there would be no virtue of justice. By contrast, natural

https://doi.org/10.1017/9781139025065.008 Published online by Cambridge University Press

 justice, the artificial virtue 215

virtues like benevolence are not conditional on the instituting of human prac-
tices to deal with such human limitations.
What Rawls would call, after Hume’s account, the “circumstances of
­justice” – limited generosity and moderate scarcity – create the need and the
opportunity for individuals to establish mutually advantageous rule-­governed
practices of property, promise, and contract (Rawls 1971: 126–129). To see how
Hume thinks this works, consider the case of property.
Without an institution of property, because of moderate scarcity and lim-
ited generosity, our access to many goods will be relatively unstable. We will
not, absent some special remedy, be able to rely on strangers not to take the
benefits of our own industry. But, though neither self-love nor benevolence
can be relied upon to motivate others to leave us in possession of our goods,
“nature provides a remedy in the judgment and understanding, for what is
irregular and incommodious in the affections” (Hume 1978: 489). We can
come to see together that it is mutually advantageous to jointly accept and
follow some set of rules concerning the legitimate acquisition and transfer of
property, rules that forbid appropriating without consent what others have
legitimately acquired.
Thus “members of [a] society” gain a “general sense of common interest”
in everyone’s abiding by a rule that, say, requires people to restrain from tak-
ing the possessions of friends and strangers alike, even when it is in their own
interest to do so (490). Each “observe[s] that it will be for his own interest to
leave another in the possession of his goods, provided he will act in the same
manner with regard to me” (490). This sense of common interest does not how-
ever remain, as it were, in members’ individual private consciousnesses. It is
rather made “public and social” by, Hume says, “all the members of the society
express[ing it] to one another” (1975a: 229; 1978: 490). “When this common
sense of interest is mutually express’d, and is known to both,” Hume writes, “it
produces a suitable resolution and behaviour. And this may properly enough
be called a convention or agreement betwixt us” (490).
Although we often use the term “convention” to refer to customs or rules,
what Hume means by it is more like a convergence or convening of society
members’ minds, the creation of a kind of “common mind” (Pettit 1993) that
somehow establishes rules. The latter, like those that structure property, are
both the object and the upshot of the convention. There are some puzzles, how-
ever, about exactly how they are. It is clear that Hume thinks that the conven-
tion produces a “suitable resolution” of members “to regulate their conduct by
certain rules,” for example, those of property (Hume 1978: 490). And it is also
clear that Hume thinks, as against Hobbes, that they do not, indeed, that they
cannot, do this by anything like a mutual contract or promise. Hobbes faced
the problem, recall, of explaining how the normative power to enter into any
contract to establish the sovereign can be explained from the meager normative
materials he provides. For Hume, any normative power of promise or contract

https://doi.org/10.1017/9781139025065.008 Published online by Cambridge University Press

 216 hume and smith

would itself depend on social rules that it would take a convention to establish.
So the convention cannot be a mutual promise or contract itself.
It seems closest to Hume’s meaning to take the convention to consist simply
in the mutually expressed “general sense of common interest” in everyone’s
regulating their conduct by the rules. In his famous example, “[t]wo men who
pull the oars of a boat, do it by an agreement or convention, tho’ they have
never given promises to each other” (490). If you and I each simply want to get
across a lake, not caring whether the other does, but realize that we each can
only if we row together, then our “general sense of common interest” will be
sufficient to motivate us to do what we need to do to get across. No joint resolu-
tion is required.
If we take our cue from the rowing example, then Hume would be saying, at
least initially, that a mutually expressed “general sense of the common inter-
est” in everyone regulating their conduct by rules of justice, such as those
structuring an institution of property, can be sufficient to produce a “suitable
resolution” of members “to regulate their conduct” by these rules. So although
self-love is often a source of “injustice and violence” (480), reflection on the
mutual benefits of restraining ourselves from taking the property of others can
be a sufficient motive to make us do so. “There is no passion,” Hume writes,
that can control “the interested affection, but the very affection itself, by an
alteration of its direction” (492).
However, there are special features of the rowing example that are not gen-
eralizable to all attempts to produce a sufficiently widespread resolution to
comply with rules of justice. Rowers are easily able to surveil their partners
and withhold their contribution when their partners do. And they enjoy the
benefits of their partners’ rowing just in case their partners enjoy their recipro-
cal benefits, indeed, simultaneously. None of this is true with all attempts to
establish rules of justice. Finally, and perhaps most importantly for Hume’s
account of the virtue of justice, a well-established rule-governed practice of
property will exist only if a sufficient number of members of society develop a
new motive. A rule-governed practice of justice requires that a sufficient num-
ber take what Hart called an “internal point of view” on the rules and accept
them as obligating norms (Hart 1961: 55–57). Nothing like this is involved in
the rowing example, although of course, the rowers might have a natural sense
of justice or reciprocity requiring them to do their fair share in a cooperative
enterprise. But this is something Hume explicitly denies. Justice, he thinks, is
an artificial virtue.
This means that Hume cannot really escape a version of the problem that
beset Hobbes’s attempt to derive obligations of justice from a covenant to estab-
lish the sovereign without sufficient normative materials to explain the norma-
tive power to make and be obligated by such a covenant in the first place. Hume
rejects the Hobbesian project of grounding justice in contract or promise, since

https://doi.org/10.1017/9781139025065.008 Published online by Cambridge University Press

 justice, the artificial virtue 217

whatever is required to explain the obligating power of promises would have

to be able to ground justice itself already. Hume holds, of course, that promises
gain their power to obligate only because a social convention to regulate con-
duct by a rule of promise-keeping exists. But if the convention just consists in
a “general sense of common interest,” even when this is mutually expressed, it
will only give members self-interested reasons to comply and then only when
this is necessary to produce others’ compliance. A convention, understood in
these interested terms, is impotent to establish rules that people can genuinely
respect or violate and to create the acceptance of such rules upon which a rule-
governed practice depends.
Hobbes and Hume both see clearly that there are strong self-interested rea-
sons to want justice practices in place (whether these require an absolute sov-
ereign, as Hobbes believes, or not). This insight is not, of course, new to the
modern period. When Glaucon challenges Socrates to justify justice in Plato’s
Republic, he states what he takes to be the common view that people are ben-
efited by others’ just conduct; he just wants Socrates to tell him why people
should act justly themselves in cases where it is not necessary to procure others’
justice (as it would not be if one had the ring of Gyges). He wants to know what
non-instrumental motive people have to be just. I mentioned this at the outset
of this book because the only such non-instrumental motives Socrates consid-
ers are intrinsic benefits of self-interest, not any motive of right that, Sidgwick
claims, the moderns begin to recognize when they “work” their “idea[s] clear”
(Sidgwick 1964: 198).
Hume calls self-interest the “natural obligation” to justice (in the Enquiry:
“interested obligation”) (Hume 1978: 498; 1975a: 278). He does think that the
natural obligation of self-interest is initially sufficient to get justice practices
started. “To the imposition, then, and observance of these rules, both in gen-
eral, and in every particular instance, they are at first mov’d only by a regard to
interest; and this motive, on the first formation of society, is sufficiently strong
and forcible” (1978: 499). But only initially. When society becomes sufficiently
“numerous,” and the effects of injustice are sufficiently dispersed and diffuse,
self-interest turns from being an ally to an enemy of justice. When that hap-
pens, any natural obligation to justice is insufficient. There must then also be a
moral obligation to be just if rule-governed practices of justice are to flourish or
even survive (499).
The basic problem, of course, is that of the “free rider,” of which Hume’s
“sensible knave” provides a textbook example (1975a: 282). “[H]onesty” or jus-
tice is often “the best policy,” but while good as a “general rule,” it is nonetheless
“liable to many exceptions.” Someone “conducts himself with the most wis-
dom,” says the knave, “who observes the general rule, and takes advantage of
all the exceptions” (1975a: 282–283). To use Rawls’s famous contrast between
a “summary” and a “practice” conception of rules, the knave treats rules of

https://doi.org/10.1017/9781139025065.008 Published online by Cambridge University Press

 218 hume and smith

justice as summary “rules of thumb” that have no binding force except as

guides to their own interest (Rawls 1955). Rules of justice cannot, however, be
so conceived and survive. It is only when conceived as practice rules that can be
respected or violated that rules can structure practices of justice like property,
promise, and contract.
Hume suggests at various points that the non-self-interested motive that
practices of justice require is a distinctively moral one. He says that their “very
equity and merit” is the only “universal motive for observing the laws of equity”
(Hume 1978: 483). Also, “we have no motive leading us to the performance of
promises, distinct from a sense of duty” (518). But now a problem emerges.
Like Hutcheson, Hume understands moral obligation in terms of the
moral sentiment. To say we are morally obligated to something is just to say
that moral sentiment or sense approves of it. But also like Hutcheson, Hume
holds that moral sentiment only directly approves of motives and never acts
considered independently of them. Indeed, Hume goes farther in this than
Hutcheson, since Hutcheson defends a derived sense in which moral sense can
approve of acts as choiceworthy – that is, as options considered from a delib-
erative perspective. Unlike Hutcheson, Hume gives no account of such delib-
erative judgments.
This commits Hume to the view that the moral sentiment is impotent to
motivate action from the deliberative point of view. Approbation is some-
thing we feel from an observer’s (third-personal) standpoint when we con-
sider motive and character and the motivated actions that flow from these.
Therefore, despite the fact that Hume seems to be saying that just acts can
be – indeed that they often are – motivated by a sense of moral obligation and
duty, such a statement is actually inconsistent with his considered view of the
moral sentiment.
For Hume, the virtue of justice, like any virtue, must be expressed in a
motive. But what is the motive of justice? On what motive does a just person
act? If we rule out a sense of moral duty or obligation on the grounds just men-
tioned, then what motives are left? We have seen that both self-interest and
benevolence are also ruled out. The “natural obligation” of self-interest may be
sufficient to help constitute a convention to establish rules of justice initially,
but when society becomes “numerous” it will not reliably motivate compliance.
Benevolence is ruled out, first, because its limitation is one of the “circum-
stances of justice” that make just practices necessary, and, second, because, as
in the case of a loan to a “seditious bigot,” benevolence conflicts with justice.
We can see our way to a possible solution if we note a further puzzle in
Hume’s text. Hume writes that only after the “convention, concerning absti-
nence from the possessions of others, is enter’d into … there immediately arise
the ideas of justice and injustice; as also those of property, right, and obligation”
(490). This suggests that all of these ideas are “artificial” in Hume’s sense that

https://doi.org/10.1017/9781139025065.008 Published online by Cambridge University Press

 justice, the artificial virtue 219

they have no application outside the artifice of convention.9 Moreover, “obliga-

tion” in this sense would, like “property,” be defined within the rules of justice.
We have already noted two senses in which Hume uses “obligation” that are
defined independently of rules of justice. Although Hume thinks that there is a
natural obligation and a moral obligation to be just, these species of obligation
are not themselves artificial in the way Hume thinks justice is. Neither depends
upon conventionally established rule-governed practices of justice.
The kinds of rules Hume is concerned with are directive in their nature.
A rule of property, for example, directs agents to abstain from others’ pos-
sessions.10 It requires that people so abstain; if they do not, they violate the
rule. We might say, meaning nothing more, that it obligates them to abstain.
We can call this third Humean sense rule obligation. Recall Hart’s distinc-
tion between the “internal” and “external” aspects of rules (Hart 1961: 55–57).
From an external third-personal point of view, we can say, at a given social
stage, that rules of justice have been established when a sufficient number of
people respect and regulate their conduct by the rules and forbear violating
them. Here we speak, as it were, not from the “internal” perspective of mem-
bers who so regulate themselves, but from an “external” observer’s standpoint.
From this perspective, any talk of obligation and justice is de facto and non-
normative. When, however, we speak from the perspective of those respecting
and regulating their conduct by the rules, we take up an “internal” standpoint,
and when we use “obligation” from this perspective, it now has a de jure nor-
mative sense.
Above I mentioned that a fundamental problem faces any philosopher, like
Hobbes or Hume, who wants to move from a self-interested justification for
individuals to desire the existence of a social order in which they are bound by
normative bonds of the right, obligations, that differ from normative reasons
grounded in their own good (or, for that matter, in the good of others). It will
be impossible to establish any such order on those grounds, since it provides a
reason of the wrong kind to support claims of right and obligation in their own
terms. (We also noted that Pufendorf faces a version of this problem, albeit a
different version, when he attempts to ground mutual accountability and the
dignity of persons in divine command.)
Any such attempt must ultimately “rest” its “moral philosophy on a mis-
take,” in Prichard’s memorable phrase (Prichard 1912). Rules of justice are only
established when sufficient numbers accept the rules and treat them as obligat-
ing norms, that is, as giving them adequate reason to be just simply because

9
As we saw in Chapter 3, Hobbes says similarly that the ideas of justice and obligation apply
only after the covenant, though, like Hume, he needs these ideas elsewhere, for example, to
ground the covenant itself.
10
Hereinafter, I follow Darwall 1995: 297–318.

https://doi.org/10.1017/9781139025065.008 Published online by Cambridge University Press

 220 hume and smith

they are so obligated, quite independently of whether the just conduct is for
their good, or, for that matter, whether it promotes the public good.
In many places, Hume describes the motive of justice in something approach-
ing these terms. Just people “lay themselves under the restraint of such rules”;
they observe them “strict[ly],” and treat them as “sacred and inviolable” (Hume
1978: 499, 533–534). One way of solving the problem his views about moral
sentiment, virtuous motive, and the lack of merit of “external performance[s]”
create would be for him to hold that the virtue of justice consists in the distinc-
tive form of motivation involved in accepting rules of justice as binding norms.
So understood, just acts, considered independently of their motivation, would
still lack moral goodness or “merit” in themselves. They would nonetheless be
obligatory in the sense of instantiating rule obligation. And when just actions
are motivated by the just person’s acceptance of the rules as binding norms, we
can then say that such actions are morally good or have merit.
All virtues, for Hume, recall, must either be useful or beneficial, either to
the agent or to others. And Hume specifically says that “sympathy with pub-
lic interest is the source of the moral approbation, which attends” the virtue of
justice (499–500). As with all the virtues, the idea is that when we contemplate
the just person’s motivating sense of justice, we are carried by an association of
ideas to thoughts of its usual consequences, generally, pleasures constituting
the “public interest,” and sympathy then turns thoughts of these pleasures into
a suitably “general” or disinterested pleasing sentiment in the observer, which
constitutes their moral approbation.
To summarize this way of interpreting Hume’s theory of justice, then: there
is, first, the natural obligation to justice, which is the self-interested motive
members of society have to bring about conventions establishing rules of jus-
tice. There is the rule obligation that is internal to the established rules. And
there is the moral obligation to justice consisting in approbation of the just
person’s characteristic motivation, a sense of justice consisting in acceptance of
the rules, by sympathy with this motive’s usual publicly beneficial effects.
To be sure, for Hume to hold this view of the motive of justice, he would have
to revise the theory of motivation he takes from Hutcheson, according to which
all actions seek to promote goods. “The will,” Hume says, “exerts itself, when
either the good or the absence of the evil can be attain’d by any action of the
mind or body” (439; see also 399).11 But no other major changes in view would
seem to be required.
To complete this discussion of Hume on justice, we should note a remark
that Hume makes about the extent of justice in the Enquiry that illustrates jus-
tice’s artificial character from Hume’s perspective and that will mark an impor-
tant contrast with Smith.

11
We noted in Chapter 2 that a similar point holds for Hobbes and Pufendorf.

https://doi.org/10.1017/9781139025065.008 Published online by Cambridge University Press

 justice, the artificial virtue 221

Were there a species of creatures intermingled with men, which, though

rational, were possessed of such inferior strength, both of body and mind,
that they were incapable of all resistance, and could never, upon the high-
est provocation, make us feel the effects of their resentment; the necessary
consequence, I think, is that we should be bound, by the laws of humanity
to give gentle usage to these creatures, but should not, properly speaking,
lie under any restraint of justice with regard to them, nor could they pos-
sess any right or property (1975a: 190).

The reason Hume thinks we would owe no obligations of justice to any crea-
tures, however rational, who “could never … make us feel the effects of their
resentment,” is that no convention for mutual restraint could possibly be
advantageous to us. Failing to give them “gentle usage” would still be every bit
as vicious as it would otherwise be. But such creatures could have no right to
this and failing to give it to them would not be a violation of a just claim they
could justifiably resent.
This brings out an important aspect of Hume’s artificiality claim that will
mark a significant contrast with Smith. For Hume, resentment functions to
create costs for the resented that have to be figured in to determine whether
accepting rules that give victims specific convention-based authorities and
right would be mutually advantageous or not. If others’ power is sufficiently
inferior, then restraining our treatment of them by such rules will not be
advantageous to us, and we will consequently owe them no obligations of jus-
tice. For Smith, by contrast, whether we owe someone an obligation of justice
depends on whether their resentment would be justified, or, in Smith’s term,
“proper.” To be sure, the “effects” of resentment that Hume most likely has in
mind include sympathetic ones. And his point might partly be that we would
be unable to have effective moral discourse with such creatures, since such dis-
course is carried by a sympathy-infused intercourse of sentiments, and sympa-
thy varies with psychic distance. But here again, a deep difference with Smith
emerges.
Humean sympathy operates from an observer’s third-person perspective,
whereas, for Smith, we judge whether a sentiment like resentment is proper by
putting ourselves in the shoes of the person we are judging. For Hume, moral
judgments are akin to aesthetic assessment; they concern what he often calls
“moral beauty” (e.g., 1978: 465). For Smith, the locus of moral judgment is
interpersonal and, sometimes, as with justice, second personal. We judge pro-
priety by placing ourselves in one another’s perspective; and we assess justice
by considering whether holding-­accountable emotions (what Strawson called
“reactive attitudes”) like resentment would be justified from that point of view
(Strawson 1968; Darwall 2006). It is thus not surprising that Smith, like Butler
and Kant, and unlike Hutcheson and Hume, places great weight on the moral
autonomy we must grant others if we are to insist on their answerability for
their own choices.

https://doi.org/10.1017/9781139025065.008 Published online by Cambridge University Press

 222 hume and smith

Smith
In the last chapter, we noted that an important difference between Butler and
Hutcheson, and we can now add Hume, is Butler’s emphasis on the “authority
of conscience.” As we saw, Butler takes the faculty of “conscience” or the “prin-
ciple of reflection” to be a necessary condition of moral agency, and he takes
virtue to consist in the excellent exercise of this faculty. In the latter, Butler
follows Shaftesbury, who distinguishes between “natural” (estimable) motives
of which conscience or “reflex approbation” approves, and genuine “virtue or
merit,” which consists in a moral agent’s directing themselves by their own
reflective moral judgments (Butler 2017: Pr.26; Shaftesbury 2001: II.28).
For Hutcheson and Hume, virtue or merit consists not in any motivation
provided by moral sense or sentiment but in motives of which it approves.
Indeed, since Hutcheson and Hume both hold that the object of approbation is
always in the first instance motives and never acts considered from a delibera-
tive perspective independently of motives, it is impossible for moral sentiment
to provide what Hume calls the “first motive” to any virtuous conduct (Hume
1978: 478). Strictly speaking, approbation cannot motivate from a (first-person)
deliberative standpoint at all; it is an observer’s third-person response to con-
templated motive and motivated action.
In this dispute, Adam Smith is decidedly with Butler (and Shaftesbury)
against Hutcheson and Hume. So last chapter’s debate between Hutcheson and
Butler is carried forward in this chapter between Hume and Smith, with Smith
taking Butler’s part. Smith holds both that the capacity for moral autonomy or
“self-command” – to direct our lives by our own moral judgments – is essential
for moral agency and that its excellent exercise constitutes “perfec[t] virt[ue]”
(Smith 1982: 237, 241, 266).12
Hutcheson and Hume hold, again, that moral judgments fundamentally
concern motives and motivated actions, viewed from an observer’s third-
personal perspective. For Hutcheson, moral sense is a disinterested response
to motives considered in themselves, quite independently of their effect and of
their response to the agent’s deliberative context. For Hume, we form moral
judgments by sympathy with the target motive’s usual effects. For Smith, by
contrast, basic moral judgments, indeed basic normative judgments, concern
what Smith calls a mental state’s “propriety,” which is what contemporary phi-
losophers call its “fit” or “fittingness” to its objects, viewed from the agent’s per-
spective (D’Arms and Jacobson 2000a and 2000b; Chappell 2012).
Attitudes and emotions, including intentions and choices, all have inten-
tional objects, which they standardly represent or “construe” in evaluative or
normative terms (Roberts 2003). Desires, for example, represent their objects as

12
“Self-command is not only itself a great virtue, but from it all the other virtues seem to
derive their principal lustre” (1982: 241).

https://doi.org/10.1017/9781139025065.008 Published online by Cambridge University Press

 smith 223

desirable, as something there is reason to desire. Resentment construes its object

as resentable in the sense of involving injury or unjust treatment to the resenter.
Intentions represent an intended action as something to be done. And so on.
Something is said to be a fitting object of desire, then, just in case it is desir-
able; it is a fitting object of resentment just in case it is resentable (warrants
resentment); or it is a fitting object of intention just in case it is an act that is
to be done. In general, an object X is a fitting object of attitude A just in case
X is “A-able” or “A-worthy” (on the model of “desirable” or “blameworthy”),
that is, worthy of being A-ed in virtue of features of it (i.e., of X together with
what Smith calls the “situation” in which X is presented in deliberation to the
agent). These are features that would correctly be taken by the agent as norma-
tive reasons to have attitude A and on the basis of which the agent could form
the attitude. Attitudes are mental states that agents have for reasons (Anscombe
1957). And they are fitting to their objects just in case the object in the presented
situation has features that give the agent “reasons of the right kind” to form the
relevant attitude (Hieronymi 2005, 2013). When this is true, then X is “A-able”
or “A-worthy,” and so it is fitting to have A toward X.
Parfit gets at roughly the same idea when he distinguishes between “object-
based” and “state-based” reasons for having an attitude. Reasons are of the
“wrong,” state-based kind when they are given, not by features of the attitude’s
object in the presented “situation,” but, for example, by pragmatic reasons why
it would be a good thing for the agent to have that attitude (to be in that “state”)
(Parfit 2011). Suppose someone offers you a prize to form a desire for something
that is utterly undesirable, like a saucer of mud. Any reason that might give
you to have that desire could not, by hypothesis, bear on the mud’s desirability;
neither is it a consideration on the basis of which you could form a desire for
the mud from a deliberative point of view (though you could so form a desire
to have that desire). An attitude, A, then, is fitting to its object, just in case the
object is “A-able” or “A-worthy,” that is, an object that warrants having atti-
tude A toward it in virtue of object-based reasons, which are A-able-making
features, where these are reasons to have A toward X and on the basis of which
the agent could form the attitude.
Now for Smith, again, basic normative judgments concern what he calls
“propriety.” But this turns out to be same thing as fittingness, as we will see
presently. For now, however, note that if it is, this means that Smith brings the
agent’s point of view into moral judgments in a way that Hutcheson and Hume
do not. The question of whether benevolence is proper in the sense of fitting
to its object in the situation as it is presented to a deliberating agent is simply
a different matter than whether it is the object of moral sense, in Hutcheson’s
sense, or whether it would tend to have pleasurable consequences that Humean
sympathy would transmute into contemplative pleasure from a third-person
standpoint (Hume’s “moral sentiment”).

https://doi.org/10.1017/9781139025065.008 Published online by Cambridge University Press

 224 hume and smith

Sympathy: Attributing Mental States and Judging Their Propriety

Hume, recall, uses “sympathy” to refer to a process that takes an observer from
an idea of an attitude or feeling to a version of the feeling itself. Like emotional
contagion, it involves catching others’ feelings, but does not require being in
others’ presence. Humean sympathy can work from the armchair. Emotional
contagion and Humean sympathy share a fundamental feature, however,
that contrasts with what Smith calls sympathy, namely, that they need not be
directed to the same object.
Take emotional contagion first. I may “catch” another’s anger by seeing their
angry face without putting myself in their shoes to attempt to grasp what they
are angry at. It might even make me angry at them. From the point of view of
emotional contagion, the object does not matter; we catch the other’s feeling
but supply an object ourselves. Similarly, when in the armchair I come to have a
version of others’ pleasure through Humean sympathy by first having the idea
of their pleasure, there is nothing in the process that takes me imaginatively
into their perspective so that our pleasures will have the same object.
For Smith, however, “[s]ympathy … does not arise so much from the view
of [a] passion,” as with Hume, “as from that of the situation which excites it”
(Smith 1982: 12). We sympathize by putting ourselves in others’ “situations”
in their shoes, that is, by viewing the objects of their mental states as they are
presented to others from their view of their situation.
Now, we are concerned mostly with the role that sympathy plays, according
to Smith, in judgments of mental states’ propriety, that is, as fitting or “suitable
to their objects.” “To approve of the passions of another … as suitable to their
objects,” Smith writes, “is the same thing as to observe that we entirely sympa-
thize with them” (16). And again, the point we should continue to consider is
that Smith holds that we sympathize with others in the sense he has in mind by
placing ourselves in their perspectives and viewing the objects of their mental
states from that standpoint.
First, however, we should note that Smith holds that imaginative projec-
tion into others’ points of view is also central even just to attributing mental
states to them and to grasping what their situation is like for them, regardless
of their mental states’ normative status. Here Smith anticipates the notion of
“simulation” as it figures in contemporary philosophy of mind (Goldman 1992;
Gordon 1992). “As we have no immediate experience of what other men feel,”
Smith writes, “we can form no idea of the manner in which they are affected,
but by conceiving what we ourselves should feel in the like situation” (Smith
1982: 9). This we do by “plac[ing] ourselves” in others’ “situation[s]” imagina-
tively (9).13

13
There is empirical evidence for Smith’s simulationist account. For example, Amos Tversky
presented subjects with the following story:

https://doi.org/10.1017/9781139025065.008 Published online by Cambridge University Press

 attributing mental states and judging their propriety 225

When the function of imaginative projection is epistemic, as it is in mental

state attribution or in trying to grasp what someone’s experience is like, as when
“our brother is upon the rack” and we want to know “what he suffers,” then the
imaginative exercise is to imaginatively take on as many features of the person
into whose shoes we are projecting as possible, to “become in some measure the
same person with him” (9). Here the knowledge we gain is not normative. It is
of how things actually are with the person we are trying to know.
But Smith holds that we also make normative judgments of propriety by
imaginative projection. And here the exercise must be somewhat different,
since there has to be critical distance between what mental state someone actu-
ally is in and whether that state is proper in the sense of fitting or “suitable to
its object.” It cannot be the case that determining what someone is feeling also
establishes that their feeling or attitude is proper.
Note also that what Smith calls “sympathy” is not the imaginative projection
itself, but a sharing of feeling as a result of it. “Sympathy” “denote[s] our fellow-
feeling with any passion whatever,” and that can be determined only after we
know what passion another person actually has, Smith thinks, by imaginative
projection (10).
The question of whether a mental state is proper is not particularistic in the
way that the question about what some person’s mental state actually is is. We
imaginatively become the “very person” to determine the latter, but we cannot
do that and attain the critical distance necessary to judge the latter. Rather,
the question of whether a mental state is proper or suitable to its object is the
question of what would be a suitable or proper response for someone facing the
deliberative “situation” that the particular person confronts.
“To approve of another man’s opinions is to adopt those opinions, and to
adopt them is to approve of them” (17). The “objects” to which beliefs or opin-
ions are proper or suitable are the believed propositions together with the epis-
temic “situation” confronting those who hold them, that is, what they take as

Mr. Crane and Mr. Tees were scheduled to leave the airport on different flights, at
the same time. They traveled from town in the same limousine, were caught in a
traffic jam, and arrived at the airport 30 minutes after the scheduled departure time
of their flights. Mr. Crane is told that his flight left on time. Mr. Tees is told that his
was delayed, and just left five minutes ago. Who is more upset? (Kahneman and
Tversky 1982).
If you are like 96 percent of Tversky’s subjects, you will say more or less instantly that
Mr. Tees is more upset than Mr. Crane. How did you decide? Alvin Goldman argues that it
is unlikely that we all possess common-sense theories that overlap on a generalization that
we apply to the instant case or that we all think consciously about how we ourselves have
felt in the past and infer by analogy (Goldman 1992). It seems far likelier that you simply
unconsciously projected into Tees’s and Crane’s respective situations and imaginatively
simulated feelings “in their shoes.”

https://doi.org/10.1017/9781139025065.008 Published online by Cambridge University Press

 226 hume and smith

evidence or reasons to hold their opinions. We determine the propriety of oth-

ers’ opinions by imaginatively confronting their epistemic situation and delib-
erating about what seems true (hence, what to believe) from that point of view.
This is not the same thing, again, as determining what they actually do think.
When we are considering whether someone’s beliefs are proper, we assume
that we already know what they think and consider the further normative ques-
tion of what to think (or what there is reason for them to think) in light of the
epistemic situation confronting them. This is not a particularistic question; it
is the question of what one should think given the evidence. And because it is,
then, unless there are relevant differences between the other’s epistemic situa-
tion and our own, there is no difference between judging the other’s opinions
proper and adopting them ourselves (unless, of course, we are epistemically
weak-willed).
And what holds for opinions holds also for other mental states, at least
for those one can adopt for reasons, like emotions and attitudes, including
intention and choice. To judge that a mental state is suitable to its object and
situation, as these present themselves to the agent in deliberation, just is,
according to Smith, to attempt to sympathize with, that is, to “adopt” the
attitude oneself imaginatively feels when one projects into the agent’s
deliberative situation. Suppose, for example, that someone is angry with
another person for having stepped on their foot. To work out whether their
anger is proper, one puts oneself imaginatively in the position of someone
who has had their foot stepped on, along with any other relevant features of
their situation. To sympathize or have “fellow-feeling” with their anger is to
share it from that perspective. If one does, one judges their anger proper.
Suppose one judges their anger improper, that it is out of proportion to what
has provoked it. There is a sense in which one might nonetheless still sym-
pathize with this particular individual being thus angry. One might attempt
imaginatively to become “the same person” (9), that is, that very individual
with their own foibles, personal history, and so on, in order to appreciate what
it is like for that particular person to have their feet stepped on. You might still
think their response is not suitable to its object, though you can see how this
individual person would have it. There are reasons that explain the person’s
response without, however, justifying it.
The general lesson is that it is always necessary to factor the agent’s cir-
cumstances, in the broadest sense, into two parts. What Smith generally
means by the agent’s “situation” is the set of situational features to which the
agent responds – the situation that the agent faces or confronts. The ques-
tion of propriety is whether a mental state is suitable or proper to the situ-
ation confronting the agent and to which the state is a response. From this
perspective, the agent drops out of the picture as the particular agent they
are. They confront a situation that, in principle, anyway, anyone might face.

https://doi.org/10.1017/9781139025065.008 Published online by Cambridge University Press

 attributing mental states and judging their propriety 227

In deliberating about what to feel or do in that situation, they do not face a

question that is essentially about them as the particular individual they are.
In addition to this first set of features, however, there will be further particu-
larizing features that together comprise the second set and that explain why
they, as particular individuals, respond as they do to the situation character-
ized by the first set.
Suppose, for example, that I have an irrational fear of flying, that my fear is
improper because it is not suitable to the danger to which it responds. My fear
may have many different causes, and if someone wants to comprehend my fear
and what it is like, and not assess its propriety, they will have to imaginatively
take on my particular idiosyncrasies from this second set that, because they are
no part of the situation I am confronting (but rather characterize me as I actu-
ally confront it), are irrelevant to my fear’s propriety. “When I condole with
you for the loss of your only son,” Smith writes, “I consider what I should suffer
if I was really you, and I not only change circumstances with you, but I change
persons and characters” (317). Condolence seeks not normative assessment of
others’ responses to misfortune, but to compassionate their responses as they
actually are.
Normative assessment of states of mind as proper or improper is different,
however. Here it is a mistake to take on features from the second set that are
not present in the first. The issue is how to respond to a presented situation
(features in the first set), what response would be proper or suitable to it, not
what explains an actual response or what that response is like. This is where
Smith’s idea of the “impartial spectator” enters. Because the question of propri-
ety is always about how an agent should response to a presented situation, what
matters is not how the other actually responds, or even how one would actually
respond oneself. To grasp these (perhaps potential) actualities, we must imag-
ine ourselves in the other’s situation not as them, as we do in condolence, or,
in the latter case, as ourselves. Knowing that I have an irrational fear of flying,
for example, I can’t imagine being in an airplane passenger’s position as myself
without importing my irrational fear. But that is obviously irrelevant to what
level of fear is proper.
Smith’s phrase “impartial spectator” can misleadingly suggest an observer’s
third-personal view. But when he says we should “consider what will be the
sentiments of the cool and impartial spectator” (38), he is not recommending
our considering the kind of third-personal response involved in Hutchesonian
or Humean moral sense or sentiment. Neither of those involves sympathy in
Smith’s distinctive sense, that is, sharing the agent’s response to the presented
“situation” from their point of view. Rather, we think as “cool and impartial
spectators” when we imagine ourselves into the other’s situation as some-
one, rather than as any particular person, whether ourselves or the person
themselves.

https://doi.org/10.1017/9781139025065.008 Published online by Cambridge University Press

 228 hume and smith

So the difference between the imaginative exercise involved in understand-

ing the other’s mental state and assessing its propriety is between taking on all
of the features of the both sets, on the one hand, and taking on only the features
of the first in order to imagine what to feel or do as someone facing the agent’s
deliberative situation and, Smith therefore thinks, what they should feel or do.
There is no reason, of course, to suppose that anyone can flawlessly pursue the
latter imaginative exercise by themselves or that people’s attempts will auto-
matically converge. So Smithian moral epistemology will inevitably involve
taking other people’s sympathetic propriety judgments into account. This is
a further aspect of the “impartial spectator” idea. Attempting to put myself in
someone’s situation as someone, I will inevitably be influenced by factors in
my second set that are irrelevant to what a proper response to a situation com-
prised by the first set would be. Each of us attempts to make such judgments
impartially (as someone), but we are all subject to errors we can hope to cor-
rect only by reasoning together in ways that reveal previously unacknowledged
partialities.
We should note also that we can ask more fine-grained normative questions
by bringing in features from the second set. For example, even though the fea-
tures to which an improper (irrational) fear respond – the seeming danger – do
not warrant that fear or acting as if the dangers were real, we can still ask what
someone should do given their irrational fear. Even though it is thinking about
a particular person’s situation that generates it, this question is still not particu-
laristic in the way questions about what a particular mental state is or is like for
the person having it are. In the latter case we attempt imaginatively to take on
all features from both sets, as noted above, while in the first we take only those
features from the second set that are part of the (more fine-grained) delibera-
tive situation the particular agent faces, not further features from the second set
that might explain how they would in fact respond to it.

The Metaethics of Propriety

At the outset of Part VII of TMS, “Of Systems of Moral Philosophy,” Smith dis-
tinguishes two questions that a systematic treatment of moral philosophy must
answer. One is a fundamental normative question: “[W]herein does virtue
consist?” The second is “[B]y what power or faculty in the mind is it, that this
character, whatever it be, is recommended to us” (265). As Smith understands
it, this second question is not the purely epistemic issue of how we know virtue
and what it consists in. Rather, he is concerned with what “recommends” vir-
tue, with the power or faculty through which we recognize virtue’s normativity,
such that we ought or have normative reason to be virtuous so understood.
Smith canvasses four answers to this metaethical question. The first is the
one implicit in Shaftesbury’s argument for the “obligation to virtue”: that virtue

https://doi.org/10.1017/9781139025065.008 Published online by Cambridge University Press

 the metaethics of propriety 229

is “recommended to us by self-love” (266).14 The second is offered by ratio-

nalists, like Clarke, Balguy, and Price, that virtue is recommended “by reason,
which points out to us the difference between one character and another, in
the same manner as it does that between truth and falsehood” (266). The third
is Hutcheson’s hypothesized “moral sense.” And the fourth is “some other
principle in human nature, such as a modification of sympathy” (266). Smith
includes Hume in this fourth category with himself, but explicitly distinguishes
between their views.
Smith regards Hutcheson as having decisively refuted both egoistic and
rationalist responses (315–321). He allows that reason has a role to play in deter-
mining “general maxims of morality,” but only “from experience and induc-
tion” (319). In terms of Mill’s contrast in Utilitarianism between the “intuitive”
and “inductive” schools of ethical philosophy, Smith counts as an inductivist.
Reason cannot give us our “first perceptions of right and wrong,” Smith writes;
these are “founded… [upon] immediate sense and feeling,” as Hutcheson had
shown (320–321).
Smith agrees with Hume, against Hutcheson, that the theory of the moral
sense is to be rejected in favor of the idea that we judge moral qualities through
sympathy. But he disagrees sharply with Hume about how sympathy enters
into moral judgment and, relatedly, what the object of moral judgment most
fundamentally is, that is, the answer to the first question: “Wherein does vir-
tue consist?” For Hume and Hutcheson, the moral sentiment assesses motives,
considered either in themselves (Hutcheson) or in relation to their usual effects
(Hume). For Smith, by contrast, moral judgments, and normative judgments
more generally, most fundamentally concern the relation between agents’
mental states and the “situations,” faced from the agent’s perspective, to which
the mental states respond. The fundamental normative property for Smith is
fittingness or propriety.
So Smith’s own answer to the first question is propriety, and he thinks there
are really only two alternatives to propriety: benevolence and prudence (267).
I mentioned above that Smith sides with Butler against Hutcheson and Hume
on the role of moral self-direction or autonomy in moral virtue. For Smith,
this follows from the fact that virtue is propriety. If that is so, then virtue con-
sists not “in any one species of affections,” whether considered in themselves
or through their usual effects, “but in the proper government and direction
of all our affections … according to the objects they pursue, and the degree of
‘vehemence’ with which they pursue them” (286). Smith evidently thinks that
“proper government and direction” of the affections and maintaining proper
affections amount to the same thing.

14
See the discussion of this point in Shaftesbury in Chapter 4.

https://doi.org/10.1017/9781139025065.008 Published online by Cambridge University Press

 230 hume and smith

We will return to the centrality of “self-government” to Smith’s ethics, in

contrast, again, to Hutcheson and Hume below. To conclude this section,
however, we should return to metaethics. Smith clearly considers himself a
sentimentalist, but of a different kind than Hutcheson and Hume. Like Hume,
he rejects a Hutchesonian moral sense, but the way sympathy enters into his
metaethics is profoundly different than the way it enters Hume’s. Humean
sympathy works from a third-person perspective on an evaluated agent,
whereas sympathy of the kind that Smith believes is at work in judgments of
propriety can arise only after an imaginative projection into the agent’s point
of view.
For Hume, sympathy’s role in moral judgment is identical to its role in aes-
thetic judgments of beauty; indeed, Hume frequently refers to virtue as moral
beauty. Both involve contemplation from an observer’s third-­personal point of
view. Smith explicitly rejects this Humean view. As it happens, Smith accepts
Hume’s theory of aesthetic judgment. Hume is right, he thinks, that our judg-
ments of beauty, for example, of a “convenient and well-contrived building”
are ultimately grounded in utility, which lead through Humean sympathy to
pleasure in a contemplating observer. Smith notes that Hume takes the same
aesthetic view of virtue.
The same ingenious and agreeable author who first explained why utility
pleases, has been so struck with this view of things, as to resolve our whole
approbation of virtue into a perception of this species of beauty which
results from the appearance of utility (188).

However, Smith rejects Hume’s aestheticizing of moral judgment and vir-

tue. “The sentiment of approbation always involves in it a sense of propriety
quite distinct from the perception of utility” (188). Whether a given motive
or intention is proper is a matter of whether it responds fittingly to the delib-
erative situation that an agent confronts. And this is something we can judge
not by remaining in an observer’s third-person position, but by projecting into
the agent’s perspective and judging the situation as it confronted them.
Smith makes another argument against identifying moral judgments with
aesthetic judgments. Since the latter do not implicate an essentially interper-
sonal point of view, a person might “grow up” “without any communication
with society” and still be as disposed to note aesthetic features, including of
their own actions and affections. They might view their “own temper and char-
acter with that sort of satisfaction with which we consider a well-contrived
machine … or with that sort of distaste of dissatisfaction with which we regard
a very awkward and clumsy contrivance” (192). But these responses would dif-
fer fundamentally from moral praise or blame. “He would not exult from the
notion of deserving reward in the one case, nor tremble from the suspicion of
meriting punishment in the other” (193). These latter sentiments “suppose the
idea of some other being” who can hold the person accountable (193).

https://doi.org/10.1017/9781139025065.008 Published online by Cambridge University Press

 justice, mutual respect, and accountability 231

Smith’s answer to his second, “normativity” question is thus, like Hume’s, a

version of the fourth, “sympathy” approach, but a very different version than
Hume’s. As with Hume, there are various more finely delimited metaethical views
that might be consistent with Smith’s approach. One can imagine a naturalist
empiricist view on which propriety consists in the convergence of fellow-feeling
with a given attitude when imaginative projection meets some naturalistically
specifiable impartiality conditions. Or there might be an expressivist version
according to which propriety judgments express sympathy with a given attitude,
perhaps under conditions meeting impartiality standards that one accepts. Or
there might be a projectivist, error-theoretical view that takes a similar shape. All
seem consistent with what Smith actually says; as with Hume, it is ahistorical folly
to attempt to place Smith determinately into one or the other of these.

Justice, Mutual Respect, and Accountability

Smith may be somewhat overstating his contrast between judgments of propri-
ety and aesthetic judgments. It is certainly true that the former differs from the
latter in involving a projection into the agent’s point of view in a way that aes-
thetic judgments do not. Smith’s considered view is that it is only moral judg-
ments of merit and demerit, especially as these implicate judgments of justice,
that involve the distinctive responses that P. F. Strawson will call “reactive atti-
tudes” (two centuries later) and that are held from an essentially “participant,”
that is, “interpersonal” or second-personal, perspective of mutually account-
able equals (Strawson 1968; Darwall 2006).
Justice, for Smith, is a kind of demerit, where merit and demerit are distin-
guished among moral qualities by their respective relations to patients’ war-
ranted reactive feelings, specifically, to proper gratitude and resentment, and,
consequently, to reward and punishment (Smith 1982: 67–108). “The sentiment
which most immediately and directly prompts us to reward, is gratitude; that
which most immediately and directly prompts us to punish, is resentment” (68).15
It is a necessary condition of merit and demerit that these attach to proper
and improper actions and states of mind, respectively. But that is insufficient,
since, for example, not all impropriety warrants resentment. Injustice consists
in improper conduct (motivated by improper motives) to which the proper
response is a reactive feeling like resentment and motives to challenge, resist, or
punish the conduct forcefully.16

15
Strawson cites gratitude and resentment as paradigmatic “reactive attitudes” that are felt
from a second-personal perspective of implicit relating to someone (Strawson 1968).
16
What follows draws from Darwall 1999b. Smith holds, of course, that these feelings can
also be held improperly and, moreover, that “we ought always to punish with reluctance,
and more from a sense of the propriety of punishing, than from any savage disposition to
revenge.” (1982: 172; see also 160–161).

https://doi.org/10.1017/9781139025065.008 Published online by Cambridge University Press

 232 hume and smith

This already sheds light on one important way in which Smith’s treatment
of justice differs from the tradition of Hutcheson and Hume. For Hutcheson,
the primary moral fact from which all others derive is that universal benevo-
lence is the morally best motive, this being established by moral esteem from an
observer’s point of view. Since universal benevolence aims at everyone’s hap-
piness equally, Hutcheson draws the conclusion that “that action is best, which
procures the greatest happiness for the greatest numbers” (Hutcheson 2004:
125). A similar line of reasoning leads to Hutcheson’s theory of rights. Some-
one has a right “to do, possess, or demand” something just in case “a faculty
of doing, demanding, or possessing [it], universally allow’d in certain circum-
stances, would in the whole tend to the general good” (183). Hume’s theory of
justice differs from Hutcheson’s in (at least) the respect that he holds that rules
of justice cannot be established unless they are, and are seen to be, in everyone’s
interest, and not just for the general good. Nonetheless, Hume’s account of
the “moral obligation” to justice is roughly the same as Hutcheson’s since it
depends on justice’s tendency to advance the “public interest.”17
On Smith’s view, however, we judge injustice, not from an observer’s per-
spective, but by projecting ourselves impartially into the agent’s and, crucially,
into the patient’s point of view. In particular, something is unjust only if it is
proper to feel resentment toward it: “the violation of justice … is the proper
object of resentment” (Smith 1982: 79). And we can judge whether resentment
is proper only by imagining what it would be like to be in the shoes of the par-
ties who are affected and considering (impartially) what to feel – what any one
of us would or should feel – in their situation. Since injustice is a vice, it is also
essential that its motives be improper. But this it shares with every other vice.
What is distinctive about injustice is that it is properly resented and resisted,
and this is something we can assess only from the perspective of individual
patients.
Earlier we noted in connection with Hume’s theory of justice his remark
that we would owe no obligations of justice to a “species of [rational] crea-
tures” of such inferior strength that “they were incapable of all resistance, and
could never, upon the highest provocation, makes us feel the effects of their
resentment” (Hume 1975a: 190). For Hume, justice is an artificial virtue estab-
lished, if at all, by a mutually advantageous convention. In such an instance,
however, no such mutually advantageous convention would exist. Smith’s
view is importantly different in these respects. Justice is not conventional, for
Smith. To determine whether such creatures might be unjustly treated by our
“provocations,” we look not to their relative power, but to whether putting
ourselves (impartially) in their place would lead us to experience resentment

17
“Sympathy with public interest is the source of the moral approbation which attends that
virtue [justice]” (Hume 1978: 499–500; see also 498).

https://doi.org/10.1017/9781139025065.008 Published online by Cambridge University Press

 justice, mutual respect, and accountability 233

from that point of view. Whether the provokers were able to “feel the effects of
[our] resentment” would obviously be irrelevant to whether to feel it (from the
weaker species’ point of view), and therefore whether their resentment would
be proper, hence, whether the provocation would be unjust.
Although like all vicious action injustice proceeds from “motives which are
naturally disapproved of,” injustice is distinguished by causing “injury,” “positive
hurt to some particular persons” that is “the proper object of resentment, and of
punishment, which is the natural consequence of resentment” (Smith 1982: 79).
It is no surprise, therefore, to find Smith concluding that it is wrong for some-
one to injure another in order to benefit themselves “though the benefit to the
one should be much greater than the hurt or injury to the other” (138). From an
agent-neutral observer’s standpoint, in which benefits and harms are accounted
as off-setting gains or losses, such an action might seem justified. Nonetheless,
Smith believes, we cannot justify it when we view things from the perspective of
the injured parties themselves, if we enter impartially into their standpoint.
Smith thus endorses the “remarkable distinction between justice and all
the other social virtues” that we first noted with Grotius (80). “[W]e feel our-
selves to be under a stricter obligation to act according to justice, than agree-
ably to friendship, charity, or generosity” (80). Justice consists in “rules” we
must strictly observe, the latter in recommended “precepts” we are not strictly
obligated to follow (80). In his chapter on the “systems of moral philosophy,”
Smith distinguishes between what he calls “ethics” (roughly the “other social
virtues”) and natural jurisprudence or justice. Although the ancients have
excellent accounts of the former, he thinks, it is only with the moderns that
we begin to get adequate accounts of justice. “Grotius,” he says, “seems to have
been the first who attempted to give the world any thing like a system of those
principles” (341–342).
Smith links this individual-relative metaethics of moral judgment (and
patient-relative metaethics of justice) to a doctrine of the dignity of individuals
and their mutual accountability. Because we judge that anyone would properly
resent various kinds of injury and violation, we see these as forms of injustice
that are appropriately resisted and reacted to with force. This gives us a way
of grounding rules of justice that impose “strict,” enforceable obligations and
define rights of individuals that we are mutually accountable for respecting. “A
moral being,” Smith says, “is an accountable being,” who “must give an account
of its actions to some other, and that consequently must regulate them accord-
ing to the good-liking of this other” (111n.3). Although Smith says that we are
“principally accountable to God,” he quickly adds that each “must necessarily
conceive himself as accountable to his fellow-creatures, before he can form any
idea of the Deity …” (111n.3).18
18
Smith did not retain this passage in the sixth edition of TMS. I am indebted here to Charles
Griswold and Vivienne Brown.

https://doi.org/10.1017/9781139025065.008 Published online by Cambridge University Press

 234 hume and smith

There is an element of Smith’s thought here that echoes the theme of the
equal dignity of persons that we saw in Pufendorf and will see perhaps most
forcefully in Kant. We resent attempted injuries, not just to our possessions
or bodies, but also to our persons. What most “enrages us against the man who
injures or insults us, is the little account which he seems to make of us” – “that
absurd self-love, by which he seems to imagine, that other people may be sac-
rificed at any time, to his conveniency” (96, emphases added). In other words,
we resent disrespect of our dignity, our status as someone who may not (not to
say, cannot) be treated in certain ways. What our resentment is “chiefly intent
upon, is not so much to make our enemy feel pain in his turn, as … to make
him sensible, that the person whom he injured did not deserve to be treated in
that manner” (95–96).19 Punishment aims to make the other “sensible” of our
dignity, to feel respect for us.
Smith’s view of the equal dignity of individuals provides some of the moral
underpinning for his praise of free markets in The Wealth of Nations. Market
exchanges occur between independent equals who pursue their respective inter-
ests, not by “servile and fawning attention to obtain” each other’s goodwill, but
through mutually advantageous, respectful free exchange (1976: 26). Of course,
Smith’s main topic is the relation between markets and the creation of wealth,
but even here the theme of equal dignity is in the background: “Masters of all
sorts … make better [though humiliating] bargains with their servants” during
tough times when their servants are “more humble and dependent” (101, 412).20

Dignity and Self-Command, the “Respectable” Virtue

Earlier we noted that Smith follows Butler’s (and Shaftesbury’s) doctrine that
virtue involves not simply estimable motives, but “self-government” or “self-
command”: agents governing themselves by their own consciences. Again, this
is a significant contrast with Hutcheson and Hume.
This difference with Hume arises, indeed, within a distinction that Smith
makes between what he calls the “amiable” and “respectable” virtues, which
distinction he actually takes from Hume! In the Treatise, Hume distinguishes
between “amiable” and “awful” virtues; contemplation of the former produces
“love,” whereas the latter inspires “esteem.” “We could wish to meet with the
one character in a friend; the other character we wou’d be ambitious of in our-
selves” (1978: 607).

19
In the introduction to their edition of The Theory of Moral Sentiments, Raphael and
Macfie point out that Smith could, in his “Letter to the Editors of the Edinburgh Review”
of July, 1775, “describe, from his own reading, … Rousseau’s Discourse on Inequality”
(1982: 10).
20
See Fleischacker 2005 for an account of Smith’s concerns about justice for the poor.

https://doi.org/10.1017/9781139025065.008 Published online by Cambridge University Press

 dignity and self-command, the “respectable” virtue 235

Self-government is Smith’s paradigm of a respectable virtue: “of that com-

mand of the passions which subjects all the movements of our nature to what
our own dignity … and the propriety of our own conduct require” (Smith
1982: 23). This connection between dignity, respect, and self-command
shows up most powerfully in some remarks Smith makes about slavery.
Smith regarded it as a symptom of the failure of all earlier social ideals that
they depended on subservience and slavery. The classical republicans’ focus
on collective liberty and the virtues of civic participation was fine as far as
it went, but it was necessarily confined to a narrow elite. And ancient ideals
were decisively marred by their reliance on slavery (Smith 1976: 377–378; see
Muller 1995). Smith’s criticism of the chattel slavery of his own time is espe-
cially moving.
There is not a negro from the coast of Africa who does not, in this respect,
possess a degree of magnanimity which the soul of his sordid master is
too often scarce capable of conceiving. Fortune never exerted more cruelty
over her empire over mankind, than when she subjected those [African]
nations of heroes to the refuse of the jails of Europe (1982: 206).

Magnanimity, for Smith, is a form of self-command, the ability to bear up

with dignity in the face of violent emotions, to regulate grief, for example, by
what an “impartial spectator” in the situation could sympathize with. As an
aspect of self-command, magnanimity is a “respectable” rather than an “amia-
ble” virtue. This is significant, since it is the link to self-command that evidently
underlies slavery’s special horror for Smith. The subjection of such dignified,
self-commanding “heroes” to men so unable or unwilling to control their own
arrogance and self-conceit that they appear powerless even to conceive the vir-
tue of self-command is especially appalling to him.
Here we see, again, the Rousseauean theme in Smith that leads toward Kant.
From our own egocentric standpoints, there is a tendency to judge value and
importance by their importance to us, from our own points of view. To insist on
this self-preference, however, is to manifest the “arrogance of … self-love” (83).
When, however, we view ourselves and our situation from a perspective any-
one could take on us, we see that we are “but one of the multitude, in no respect
better than any other in it” (137). Only from this impartial standpoint can a
person assess the propriety of their own motives and achieve self-command,
thereby “humbl[ing] the arrogance of his self-love” (83). From this perspective,
blind self-­preference seems “shamefu[l]” (137). Smith and Kant both stress that
we are humbled also when we recognize this authority through the eyes of oth-
ers, as when we appreciate their self-command. That is why self-command is an
“awful” or “respectable,” rather than amiable, virtue for Smith. In Kant’s terms,
the other’s “example holds before me a law that strikes down my self-conceit”
(Kant 1996b: 5:77).

https://doi.org/10.1017/9781139025065.008 Published online by Cambridge University Press

 236 hume and smith

There is an irony in the fact that Smith’s emphasis on self-command occurs

in the context of a distinction between the respectable and the amiable virtues
that he takes from Hume. Whereas Smith links his paradigmatic respectable
virtue, self-command, to themes of equal dignity and mutual accountability
that are marks of a distinctively modern conception of morality, Hume intro-
duces the category of virtues that are “awful” rather than “amiable” as part of an
argument that “natural abilities” like wit that occasion esteem rather than love
have no less merit, and that the modern distinction between “moral virtues”
and “natural abilities” is merely a verbal one (Hume 1978: 606).

https://doi.org/10.1017/9781139025065.008 Published online by Cambridge University Press

 7

The British Rationalists and Reid

Hume’s claim that “moral distinctions derive” from sentiment rather than rea-
son, echoed in different ways by Hutcheson and Smith, was advanced in oppo-
sition to a group of philosophers writing in the early eighteenth century now
known as the “British rationalists,” which included Samuel Clarke (1675–1729),
John Balguy (1686–1748), and William Wollaston (1659–1724). Richard Price’s
Review of the Principles of Morals (1758) was a later entrant in this debate on the
rationalist side as, in some ways, was Thomas Reid’s Essays on the Active Pow-
ers of the Human Mind (originally published in 1788), though Reid, like Butler,
was hesitant to characterize himself as a rationalist. (Price lived from 1723 to
1791 and Reid from 1710 to 1796.) In this chapter, we shall focus on the central
elements of eighteenth-century British rationalism, beginning with Clarke and
Balguy, but concentrating mostly on the more philosophically developed and
interesting versions in Price and Reid.
The philosophical, metaethical claims that divided rationalists and senti-
mentalists were primarily epistemological and metaphysical. Concerning epis-
temology: How and through what faculties do we judge, and perhaps know
the truth of, moral and ethical propositions? And metaphysically: In what do
moral and ethical properties consist (if they exist)? Are there moral and ethi-
cal facts, and, if so, how do they relate to facts of other kinds? The most obvi-
ous disagreement was epistemological: whether the faculty of moral judgment
and knowledge is some version of the understanding, intellect, or reason, or
whether it consists in sentimental or affective sensibility of some other kind.
But these epistemological differences were often grounded in metaphysical
disagreements.
The rationalists followed Cudworth in arguing that morality is “eternal and
immutable” and that it could not, therefore, be response-dependent in the ways
moral sentimentalism affirms (Clarke 1978a, 1978b: 596; Price 1974: 50). Moral
obligations hold at the most fundamental level not just for moral agents, con-
stituted as they are in, say, the human species, but for any possible moral agent.
For Hume, however, “[v]ice and virtue … may be compared to sounds [and]
colours,” which “according to modern philosophers,” are secondary quali-
ties that depend essentially on contingent human sensibilities like our sense
237

https://doi.org/10.1017/9781139025065.009 Published online by Cambridge University Press

 238 the british rationalists and reid

of color (Hume 1978: 468). Because they thought morality to be “eternal and
immutable,” the rationalists held that moral propositions could not be judged
by any such sensibility or sense that a possible moral agent might lack.
There was a related issue concerning the capacities that are necessary for
moral agency and action. For Hutcheson and Hume, what Hume called the
“first motive” to virtuous conduct is always some motive, like benevolence,
that, as the rationalists saw it, a moral agent might conceivably lack. Rational-
ists like Balguy, for their part, tended to hold that moral virtue must involve
motivation provided by a faculty that is necessary to be subject to morality at all
as they saw it, the faculty of reason. “[S]upposing us void of natural Compas-
sion, as well as Benevolence; might we not possibly be induced to attempt the
Relief of a Person in Distress, merely from the Reason of the Thing, and the
Rectitude of the Action?” (Balguy 1976: I. 12). Balguy decidedly answered “yes,”
as did Clarke and Price. Moral virtue, for them, involves the proper exercise of
the faculty (reason) that is an essential feature of moral agency.
Partly what was at issue was the dispute that, as we have seen, divided Shaft-
esbury, Butler, and Smith, on the one side, and Hume and Hutcheson on the
other – namely, whether virtue necessarily involves self-direction through an
agent’s own moral judgment or conscience. The rationalists followed Butler
in this, holding that virtue is conscientious conduct and that conscience oper-
ates through the agent’s own reason. Butler’s doctrine does not strictly require
rationalism, as the example of Smith, and perhaps Butler himself, proves. Some
version of rationalism is difficult to avoid, however, if some form of reason is
thought to be necessary and sufficient for moral agency.
Here it is significant that the British rationalists were all attempting to the-
orize a deontic conception of morality alongside a matching conception of
moral agency, in a way that Hume and Hutcheson were not. Theirs was defi-
nitely a “law conception,” which, like Suárez and his natural law followers, they
took to be essentially related to accountability. Moral agents are, Clarke says,
“Accountable Beings” whose wills can and morally must (it is “(morally speak-
ing) necessary” that they) “be determined in every Action by the Reason of the
thing, and the Right of the Case” (Clarke 1978b: 613). Price echoes this theme in
a Butlerian idiom: “Every being endowed with reason, and conscious of right
and wrong, is, as such, necessarily a law to himself” (Price 1974: 119). Because
they have reason and conscience, moral agents are “accountable, guilty, and
punishable” for wrongful conduct (119).
As we shall see, there is arguably a deep connection between the British
rationalists’ focus on deontic morality and their rationalism.1 But this is not a
necessary connection. Nothing rules out a sentimentalist metaethics of deontic
morality in principle. Owing to the conceptual connection between deontic

1
Here I follow Schiffman 2015.

https://doi.org/10.1017/9781139025065.009 Published online by Cambridge University Press

 fordyce 239

moral concepts like obligation and accountability, there will indeed be dialecti-
cal pressure to hold, with Pufendorf, that an agent can be a subject of obliga-
tion, and hence a moral agent, only if they have whatever faculties are necessary
to hold themselves accountable (in Butler’s terms, to be a “law to themselves,”
or in Locke’s, to be “capable of a law”). But these might involve sentiment, as
Smith arguably held. I suggested earlier that we might develop a metaethics of
judgments of right that ties them to reactive attitudes, considering Hobbes’s
claim that the “sense of right and wrong” that distinguishes human beings from
other species consists in our susceptibility to (reactive) attitudes through which
we “censure” one another.
As we shall see, a central plank of British rationalism in its most developed
form, especially in Price, is the thesis that the deontic idea of moral obligation
is a fundamental moral concept that resists definition in non-deontic-moral
terms. They were not, however, the first to notice this or make this claim. Argu-
ably, the first was David Fordyce (1711–1751), who was not a rationalist at all, but
arguably a kind of sentimentalist.2 It is remarkable, indeed, to see how closely
Fordyce cleaves to the views we have already noted as characteristic of the Brit-
ish rationalists without embracing their rationalism. Before we begin to con-
sider the rationalists’ views in more detail, it might be useful to frame them
with a short discussion of Fordyce.

Fordyce
What is most interesting about Fordyce from our perspective is that he insists
on the indefinability of moral obligation, a thesis that looms large in the case
for British ethical rationalism in its most developed form in Price. Unlike
Price, however, Fordyce denies that the faculty of moral judgment is reason.
Fordyce’s view of reason seems thoroughly within an empiricist framework
deriving from Locke by way of Hutcheson. Fordyce holds, like Locke and
Hutcheson, that all ideas come from sense experience, and that different simple
ideas come from different “senses” (Fordyce 2003: 34). That, recall, was what
lay behind Hutcheson’s doctrine of the moral sense, which, he held, receives
the simple ideas of approbation and condemnation through which we attribute
moral good and evil.
Fordyce agrees with Hutcheson’s general epistemology, but the simple idea
on which Fordyce focuses is not that through which an observer judges the
moral goodness of motives and motivated action, but the distinctively deontic
idea of a morally required act. It is, “the Idea of Moral Obligation,” he says,
that we must “trace” “to our inmost Sense and Experience” (22). Fordyce uses
Hutcheson’s term, “moral sense.” But he is referring to the conscientious sense

2
William Frankena used to make this claim to me in conversation.

https://doi.org/10.1017/9781139025065.009 Published online by Cambridge University Press

 240 the british rationalists and reid

an agent has that they ought or must perform some act, not an observer’s sense
of the moral goodness of a motive (what Strawson calls the “sense of obliga-
tion” (Strawson 1968: 86)).
Although it is set within a Hutchesonian empiricist epistemology, Fordyce’s
view of the moral sense is actually much closer to Butler’s conscience. Like But-
ler, Fordyce emphasizes conscience’s “authority”: “[T]here is another directing
or controuling Principle, which we call Conscience, or the Moral Sense, which
by a native kind of Authority, judges of Affections and Actions, pronouncing
some just and good, and others unjust and ill” (19). Fordyce’s including affec-
tions (motives) here might seem to diminish any difference from Hutcheson,
but his examples make clear that, as with Butler, conscience has a major agential
role in directing us to perform morally obligatory acts. The instances he gives of
the sense of moral obligation have as their objects acts rather than motives. “Let
us suppose, for instance, a Parent, a Friend, a Benefactor, reduced to a Condi-
tion of the utmost Indigence and Distress, and that it is in our Power to give
them immediate Relief. To what Conduct are we obliged?” (24). Or “Again, let a
Friend, a Neighbour, or even a Stranger, have lodged a Deposit in our Hands, and
after some time reclaim it, no sooner do these Ideas of the Confidence reposed
in us, and of Property not transferred, but deposited, occur, than we immediately
and unavoidably feel, and recognize the Obligation to restore it” (24).
And the undefinable or simple idea that Fordyce insists on is the deontic idea
of moral obligation. “[W]hen we use these Terms, Obligation, Duty, Ought,
and the like, they stand for a simple Idea, an original uncompounded Feeling
or Perception of the human Mind, as much as any Idea whatsoever, and can no
more be defined than any other simple Idea” (28). This, of course, is the idea
that Shaftesbury, Hume, and Hutcheson subversively redefine in non-deontic
terms of self-interest (Hume’s “natural” or “interested obligation”) or esteem
and disesteem of motivation (Hume’s “moral obligation”).
Nonetheless, Fordyce insists, against what will be the main tendency of ethi-
cal rationalism, that the source of this fundamental deontic moral idea is not
reason, but sentiment or sense. Like Hume and Hutcheson, he restricts rea-
son to demonstrative and probabilistic reasoning, including of the aptness of
means to accomplishing ends. And like both, he denies reason’s capacity to
motivate on its own: “Reason may perceive a Fitness, or Aptitude to a certain
End, but without some Sense or Affection we cannot propose, or indeed have
any Idea of an End, and without an End we cannot conceive any Inducement
to Action” (35). “We might … perceive all the possible Reasons, Relations, and
Differences of Things, and yet be totally indifferent to this or that Conduct,” he
says, likely with Clarke in his sights (35).
In a passage that recalls Hume’s challenge that one can establish through rea-
son every “matter of fact in the case,” and nonetheless find that “[t]he vice entirely
escapes you” until sentiment is engaged (Hume 1978: 468), Fordyce writes:

https://doi.org/10.1017/9781139025065.009 Published online by Cambridge University Press

 fordyce 241

We employ Reason, and worthily employ it in examining the Condition,

Relations, and other Circumstances of the Agent … or, in other words, the
State of the Case (Fordyce 2003: 27).
But this judgment of reason does not yet give us the idea of moral obligation.
To make the further judgment that an agent is morally obligated to act in the
circumstances that reason has determined to be the case, we require moral
sense. Reason can tell us that property has been deposited with and not trans-
ferred to us, but it takes moral sense to conclude that we are morally obligated
to return it.
Fordyce explicitly considers Samuel Clarke’s rationalist view, which we shall
examine below, that reason can judge relations and “Finesses” that constitute
obligations in themselves (Fordyce 2003: 33). To Clarke’s claim that it is self-
evident to reason that gratitude is a fitting, obligatory response to benevolence,
Fordyce acknowledges that through reason we can “compar[e]” “the Nature
or Character of a Benefactor, and of a Beneficiary” and “the Relation between”
them through reason (34). But he asks,
Are any of these Ideas that which we understand by the Moral Duty or
Obligation, the Idea of Gratitude due to the Benefactor from the Beneficiary?
This is evidently a distinct Perception, obvious to some Sense, Organ, or
Power of Perception, but not the Result of Reasoning (34).
Even as he insists on the irreducibility of moral obligation that is a hallmark
of British rationalism, therefore, Fordyce remains steadfastly empiricist. “The
Nature, the Reasons, and the Relations of things would never have suggested
to us this simple Idea of Moral Obligation without a proper Sense susceptible
of it” (40).
Fordyce’s decidedly deontic focus does, however, run against the main lines
of empiricist sentimentalism in Hutcheson and Hume, as does his focus on
the authority of conscience. Connected to these is a doctrine he shares with
Butler about the connection between moral goodness and conscientiousness,
the agent’s self-regulation through conscientious deontic moral judgment. The
doctrine is a staple of British rationalism and opposes Hume’s and Hutcheson’s
skepticism about the moral goodness, and even, indeed, the very possibility,
of conscientious motivation. Fordyce defines “morally good Action” as action
“to fulfil a Moral Obligation knowingly and willingly.” “And a morally bad
Action,” or an immoral Action, is action “to violate a Moral Obligation know-
ingly and willingly” (25).
It is crucial to Fordyce, as it was to Butler, that Conscience makes us a “Law
to Ourselves” (40). Both take conscience to be essential to moral agency, to be
the very capacity that makes us subject to morality, and both hold that moral
goodness or virtue consists in the excellent exercise of conscience. It is only
because conscience or the moral sense “is interwoven with the very Frame and

https://doi.org/10.1017/9781139025065.009 Published online by Cambridge University Press

 242 the british rationalists and reid

Constitution of our Nature” that “We are in the strictest Sense a Law to Our-
selves” (40).
It is a notable feature of the use of “obligation” in the moral philosophy of
the seventeenth and eighteenth centuries that it sometimes refers to a moti-
vational state and sometimes expresses a deontic normative relation. We see
this in Fordyce (as we had in early British writers), and we will see it too in
the British rationalists Clarke and Balguy and, remarkably, also in Kant. Price
will make a special point of marking this ambiguity so that he can focus on the
deontic normative relation.
At one point, Fordyce defines what he calls a “State of Moral Obligation”
as “That State in which we feel ourselves moved … and prompted to a cor-
respondent Conduct” by our sense (through conscience) that we “ought” to
act (24). Conscience speaks to us “with a Voice irresistibly audible and com-
manding …, with an Authority which no person can silence without being
self-condemned” (24). A state “in which we feel ourselves moved” is a moti-
vational state. But what conscience tells us, of course, is that we ought to act,
not just that we feel moved to, and this is a normative relation between us and
the relevant act. Conscience’s content or object is not a motivational state, but
the normative relation or state of which it makes us aware: that we ought or
are obligated to act.
Fordyce then describes “the State of Moral Obligation” in the following
terms:

that State in which a Creature, endued with such Senses, Powers, and
Affections as Man, would condemn himself, and think he deserved the
Condemnation of all others, should he refuse to fulfill it (25).

The “it” to be fulfilled is the obligation of which Conscience gives us the sense.
Fordyce’s way of describing the State of Moral Obligation here is thus notably
not as the mental state of being moved. It is rather things being such that the
agent would be in the mental state of condemning themselves were they not
to have fulfilled the obligation of which their conscience gives them the sense.
To be sure, this is still a motivational claim, only a counterfactual one. If, how-
ever, Fordyce had said not that the agent “would condemn himself,” but that he
should, that would make the relevant state normative rather than motivational
or psychological.
In any case, Fordyce clearly sees obligation as something one can either suc-
ceed or fail at fulfilling, where success and failure are tied to accountability. This
makes obligation something normative (or at least, believed to be normative) in
a distinctive deontic sense. Falling short of an obligation is, moreover, failure to
meet not just any normative standard, but one that elicits self-condemnation.
Moral obligations are, in their nature, the sort of thing for which we hold our-
selves accountable in our own conscience.

https://doi.org/10.1017/9781139025065.009 Published online by Cambridge University Press

 clarke 243

Clarke
We find this same ambiguity in Samuel Clarke, who is generally taken to be
the originator of British Ethical Rationalism, though Price draws inspiration
from Cudworth as much as from Clarke. Clarke’s rationalism was doctrinally
cleaner than Cudworth’s, since Cudworth identifies “superior reason” with
love. It was Clarke’s rationalism, laid out in his Boyle Lectures of 1704 and 1705,
to which Hutcheson, Hume, and, as we have just seen, Fordyce opposed their
moral sentimentalism (Clarke 1978a, 1978b; Hutcheson 2002: 155–160).3 And
later British rationalists all recognized Clarke as their predecessor.
Clarke was most famous for the doctrine that there are eternal relations of
“fitness” and “unfitness” between moral agents, deliberative circumstances,
and actions that are self-evident to reason. There is a “Fitness or Suitableness
of certain Circumstances to certain Persons, and an Unsuitableness of others,
founded in the nature of Things and the Qualifications of Persons, antecedent
to all positive appointment whatsoever,” Clarke writes, in opposition to vol-
untarism of all forms (Clarke 1978b: 608). “’Tis undeniably more Fit … that all
Men should endeavour to promote the universal good and welfare of all” (609).
This “Branch” of the “Rule of Righteousness” Clarke calls that of “Love” (619).
But it is not the only one. The fundamental rule of right has another branch,
“Equity,” which Clarke formulates as reciprocity or the Golden Rule: “so to
deal always with another, as he would reasonably expect that Others should in
like Circumstances deal with him” (619). This also is self-evidently “Fit” (609).
Unlike Butler, Clarke does not consider the possibility that Equity and Love
might conflict.
Fitness, as Clarke understands it, is plainly a normative relation. He tends to
use “reasonable,” “fit” or “fitting,” and “right” as rough synonyms. Thus, “’tis
without dispute Fit and reasonable in itself” that we should “preserve the Life
of an innocent” person who is in our power, or “deliver him from imminent
danger” (609). Being “fit,” for Clarke, is much the same as being proper for
Smith. Recall that Smith takes propriety to be a relation that holds between
some “case” or “situation” (what Clarke tends to call a “circumstance”) that
an agent confronts and some act or response they might have in the light of it.

3
William Wollaston (1659–1724) is also a target of Hutcheson and Hume, but what is distinc-
tive about his The Religion of Nature Delimited (originally published in 1728) is an attempt
to reduce normative judgments and properties of moral good and right to the property
of truth (Wollaston 1978). Wollaston argued that morally wrong actions express false-
hoods, for example, in Wollaston’s best case, that lying “says” that something false is true.
Sentimentalists and rationalists were agreed that it is reason that judges truth and false-
hood, but they generally denied that moral judgments and properties can be thus reduced.
That was a centerpiece of Hume’s argument that “moral distinctions” cannot “derive” from
reason, after all (Hume 1978: 455, 470). Later rationalists like Price agreed (Price 1974:
125–126).

https://doi.org/10.1017/9781139025065.009 Published online by Cambridge University Press

 244 the british rationalists and reid

Similarly, an act is fitting for Clarke when it is a fitting response to the “circum-
stance” in which it is taken. Fittingness and propriety are normative relations
for both Clarke and Smith. Clarke uses “fitting” in a way that is substantially the
same as the way the term functions in metaethical debates in the twentieth and
twenty-first centuries (Broad 1962; Ewing 2012a, 2012b; D’Arms and Jacobson
2002a, 2002b; Howard 2019). What is distinctive in Clarke is the claim that nor-
mative fittingness is something that can be self-evident to reason.
Clarke famously analogizes morality to mathematics: “that from the differ-
ent relations of different Persons one to another, there necessarily arises a fit-
ness or unfitness of certain manners of Behaviour of some persons towards
others, is as manifest, as that the Properties which flow from the Essences of
different mathematical Figures, have different congruities or incongruities
between themselves” (Clarke 1978b: 608). But here an issue arises in under-
standing Clarke’s view.
Clarke takes it to be undeniable by reason “[t]hat there are Differences of
things; and different Relations, Respects or Proportions, of some things toward
others” (608). These “Differences” and “Relations” derive from “the nature of
Things”; they are “natural relations” (608, 615). In one sense even sentimental-
ists like Hume can agree that the analogy with mathematics carries through.
That mathematical objects are what they are and have whatever properties they
do in virtue of their natures is available to reason; for Hume, this concerns
“relations between ideas” that are discoverable by reason.
Similarly, facts about “Circumstances” (Smith’s “situations” and “cases”)
are what Hume calls “matters of fact” that probabilistic causal reasoning can
reveal. As we recently saw, Fordyce also agrees with this: “We employ Reason,
and worthily employ it in examining the Condition, Relations, and other Cir-
cumstances of the Agent … or, in other words, the State of the Case” (Fordyce
2003: 27). Since in terms of Hume’s “is”/“ought” distinction, these are all “is”’s,
sentimentalists like Hume, Hutcheson, and Fordyce can happily agree that rea-
son can determine them.
What they disagree about and what is consequently at issue is whether rea-
son can determine the kind of normative claims we express with “ought.” And
here it is important to appreciate that Clarke himself distinguishes between
natural “Differences” and “Relations” that constitute the “Circumstances” of
action, on the one hand, and the normative judgment of an action, that it is “fit”
or “reasonable,” on the other. “Fitness or Suitableness” is, he says, “founded in
the nature of Things” (Clarke 1978b: 608). The normative relation is grounded
or “founded” in natural relations that make up circumstances. It is not itself
constituted by them.

These 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 upon them so to do (596).

https://doi.org/10.1017/9781139025065.009 Published online by Cambridge University Press

 clarke 245

Clarke importantly distinguishes, therefore, between relations that themselves

constitute the circumstances about which a normative judgment is made and
normative judgments of fitness and reasonableness of action in those circum-
stances that are grounded in these. From the natural relations and differences
that constitute the former, “there necessarily arises … a fitness or unfitness”
they do not themselves constitute (608).
This is all common ground with the sentimentalists. All can agree with a
basic “supervenience” claim that came to be a staple of metaethical debates in
the twentieth century from G. E. Moore on: Things can differ in their norma-
tive properties only if they differ in the properties that constitute their natures
(Moore 1993). As it came to be expressed, evaluative or normative properties
are grounded in or supervene on properties constituting the nature of what is
said to have the evaluative or normative property. Just as this would be shared
by cognitivists like Moore and non-cognitivists like Hare in the twentieth cen-
tury, so also was it common to Clarke, Fordyce, and Hume in the eighteenth.
What was at issue was what faculties we make the normative judgments
with, the judgments Clarke expressed with terms like “fit” and “reasonable.”
Clarke held the view that these also were judgments of reason. And here, Clarke
faced Fordyce’s criticism. Fordyce agreed with Clarke that reason can deter-
mine, for example, the relations in which a benefactor stands to a beneficiary,
but he denied that these necessarily give us “that which we understand by the
Moral Duty or Obligation, the Idea of Gratitude due to the Benefactor from the
Beneficiary.” He held that “this is evidently a distinct Perception, obvious to
some Sense, Organ, or Power of Perception, but not the Result of Reasoning”
(Fordyce 2003: 33, 34). “The Nature, the Reasons, and the Relations of things,”
Fordyce argued, “would never have suggested to us this simple Idea of Moral
Obligation without a proper Sense susceptible of it” (40).
Thus, Clarke distinguished between reason-made judgments of circum-
stances and the nature of things, on the one hand, and normative judgments
that are based on them, on the other, just as the sentimentalists did. But he
asserted against them that the latter are made by reason also. Clarke could not
respond directly to the sentimentalist critique, since its first focused form in
Hutcheson’s Illustrations Upon the Moral Sense appeared only a year before
Clarke died in 1729. To do so, however, he might have availed himself of the
theme, central to later rationalists like Balguy and Price, that we mentioned
earlier – namely, that normative judgments concerning “eternal and immu-
table” morality could not be made through a contingent human sensibility that
possible moral agents might lack. The judgment, for example, that it is wrong to
cause gratuitous harm, or that we should help “deliver” others “from imminent
danger,” evidently purports to concern any possible moral agent, and not just
human beings (Clarke 1978a: 194). How could some contingent sensibility akin
to the human color sense be sufficient to judge that?

https://doi.org/10.1017/9781139025065.009 Published online by Cambridge University Press

 246 the british rationalists and reid

This brings us to the tendency in eighteenth-century moral philosophy we

remarked on earlier to use “obligation” to refer both to a normative relation and
to a motivational state. Clarke also exemplifies this ambiguous usage. Generally,
as in the passage quoted above, he understands “obligation” in the former way, as
expressing, synonymously with “duty,” a normative relation of fitness and unfit-
ness that is grounded in “the eternal and necessary differences of Things” (596).
The nature of the circumstances, he says, “lay an Obligation” on the agent (596).
There is, however, a notable exception.
For the Judgment and Conscience of Man’s own Mind, concerning
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 whoever acts contrary to this sense and conscience of his
own mind, is necessarily self-condemned (614).
Here is the same ambiguity we saw in Fordyce. And in both philosophers it is
tied, significantly, to a Butlerian view of the authority of conscience along with
a focus on deontic morality tied to accountability.
Clarke acknowledges that “the ambiguous use of [‘obligation’] as a Term of
Art, has caused some perplexity and confusion” (614). But he holds that there is
an underlying unity in the ambiguity. “The original Obligation … is the eternal
Reason of Things,” at least when this concerns reasons of right (614). When it
does, judgments of fitness and reasonableness are expressible with deontic terms
like “right,” “duty,” and “obligation.” Clarke follows Butler (and Pufendorf) in
holding that “the force” of moral obligations is “particularly discovered and
applied to every Man” through each person’s conscience with which they can
hold themselves accountable by self-­condemnation (614). It follows that obliga-
tion as a normative relation can be in force only if it is possible for agents who are
subject to it to instantiate the relevant motivational state in themselves through
conscience. Clarke’s two uses of “obligation” thus go together in a way that the
ambiguous, subversive uses by Shaftesbury, Hutcheson, and Hume do not.
Clarke holds that there are two fundamentally different aspects or “branches”
of deontic morality or the “Rule of Righteousness” as regards our “Fellow-
Creatures,” which he calls “Equity” and “Love,” respectively:
That in particular we so deal with every Man, as in like Circumstances we
could reasonably expect he should deal with Us; and that in general we
endeavour, by a universal Benevolence, to promote the welfare and happi-
ness of all Men (619).

He does not say, as Butler does, that what is morally right can sometimes con-
flict with the greatest overall good that universal benevolence would recom-
mend. But unlike Hutcheson, universal benevolence does not function for
Clarke as a source of moral direction that is independent of conscience and the
rule of right. As with Butler, the latter is fundamental. Moral agency consists

https://doi.org/10.1017/9781139025065.009 Published online by Cambridge University Press

 clarke 247

in conscientiously holding oneself accountable for compliance with the rule of

right, and this requires both equitably asking no more of others than we would
have them ask of us and promoting human good (though they may come to the
same thing).
As we saw in the last chapter, a major prong of Hume’s critique of rational-
ism is that the rationalists cannot explain the connection between moral judg-
ment and motivation. One version Hume gives seems especially targeted on
Clarke:
In order, therefore, to prove, that the measures of right and wrong are eter-
nal laws, obligatory on every rational mind, ’tis not sufficient to shew the
relations upon which they are founded: We must also point out the con-
nexion betwixt the relation and the will; and must prove that this connex-
ion is so necessary, that in every well-disposed mind, it must take place and
have its influence (Hume 1978: 465).

One might think that rationalists could have avoided Hume’s critique by
simply denying the motivational internalism that underlies it, as some
twentieth-century intuitionist externalists, like Ross, did. Importantly, how-
ever, Clarke does not take this tack, and neither do Balguy and Price, who
follow his lead.
There are, moreover, good philosophical reasons for them not to do so,
owing to the connections between deontic morality, conscience, and account-
able moral agency that, as we have just noted, they were all concerned to draw. If
deontic morality is necessarily connected to (any possible) moral agent’s capac-
ity to hold themselves accountable through their own conscience, then moral
agents must be able to be moved by deontic moral judgments of themselves.
Clarke is the first of the British rationalists to defend a motivational inter-
nalism by invoking a parallel between the way a rational mind is necessarily
moved to believe what it judges to be true and the way the will is necessarily
moved to what a rational agent judges to be fitting and right.
’[T]is as natural and (morally speaking) necessary, that the Will should
be determined in every Action by the Reason of the thing, and the Right
of the Case; as ’tis natural and (absolutely speaking) necessary, that the
Understanding should submit to a demonstrated Truth (Clarke 1978b: 613).

But note the distinction Clark makes between the theoretical and practical
cases. Whereas believing a demonstrated truth is “absolutely” necessary, doing
what one judges right is “morally” necessary.
[A]ssent to a plain speculative Truth, is not in a Man’s Power to withhold;
but to Act according to the Plain Right and Reason of things, this he may,
by the natural Liberty of his Will, forbear (613).

Balguy and Price will argue that the connection is tighter still.

https://doi.org/10.1017/9781139025065.009 Published online by Cambridge University Press

 248 the british rationalists and reid

Balguy
Like Cudworth and Leibniz, Clarke argued against all forms of voluntarism
that they made moral right and wrong “positive” and “arbitrary” (Cudworth
1996: I.i.1; Clarke 1978b: 608). Sentimentalists also were opposed to the volun-
tarist’s moral positivism. Balguy, however, argued against sentimentalism that
it also made morality “positive”. About Hutcheson’s version, he wrote, “Virtue
appears in it to be of an arbitrary and positive Nature; as entirely depending
upon Instincts, that might originally have been otherwise” (Balguy 1976: 8–9).
Balguy argued that this was true in several ways. It is the case, first, that, for
Hutcheson, the property of moral goodness or virtue is a contingent, response-
dependent property, like color. Second, the response on which virtue depends,
the moral sense, is not itself intrinsic to moral agency, according to Hutcheson.
Possible moral agents might lack it. Hutcheson did not concede this point; he
proclaimed it. Indeed, Hutcheson thought that virtuous moral conduct does
not involve the agent’s moral sense at all; it consists entirely in benevolence.
Third, Balguy argued that benevolence is not intrinsic to or necessary for moral
agency either. Had we “been formed destitute of Natural Affection” and “void
of natural Compassions, as well as Benevolence,” we would still be agents who
are fully capable of moral virtue (12).
Like Butler and Clarke, Balguy holds that virtue is realized by exercising the
faculty that makes us moral agents and subject to moral obligation. And Balguy
follows Clarke in holding that the requisite faculty is reason. On Hutcheson’s
view, by contrast, non-rational beings can be virtuous no less than we can be.
“[I]f Virtue be only kind Instincts, or Affections,” then how can we “deny the
Virtue of Brutes,” Balguy asks (14). He holds that virtue consists entirely in
agents’ determining themselves by their own rational judgment. Nothing can
“be Virtue, but what consists in a Rational Determination of the Mind” (41). “I
cannot have any other Idea of Moral Merit, than conforming, or endeavouring
to conform our Actions to the Reasons of Things” (60).
Reason-determined action is not only necessary for moral goodness, it is
also sufficient. Therefore, as against Hutcheson, benevolence is not only insuf-
ficient, it is also not necessary. Even were we “void of natural Compassion, as
well as Benevolence; might we not possibly be induced to attempt the Relief of
a Person in Distress, merely from the Reason of the Thing, and the Rectitude of
the Action?” (12). “Might not some such Maxim as that of doing as we would be
done unto, offer it self to our Minds, and prevail with us to stretch out a helping
hand upon such an occasion[?]” (12–13). Not only can reason effectively direct
us to what is right, it also enables us to hold ourselves accountable if we do not
follow its direction: It “would prompt us to undertake it, and condemn us if we
omitted it” (13).
For Clarke and Balguy, rational agency and moral agency are coterminous.
Unlike Butler, who seems implicitly to recognize a distinction between wider

https://doi.org/10.1017/9781139025065.009 Published online by Cambridge University Press

 balguy 249

judgments of practical reason (the “principle of reflection”) and narrower, fun-

damentally deontic judgments of conscience, Balguy speaks indifferently of
conforming to “the Reason of the Thing” and to “the Rectitude of the Action.”
“The End of Rational Actions, and Rational Agents, consider’d as such,” he
says, “is Reason or Moral Good” (48).
This is most evident in the analogy Balguy takes over from Clarke between
the way the rational will aims at doing what is right and reasonable, on the
one hand, and the way theoretical reason aims at truth, on the other: “The
same necessity which compels men to assent to what is true, forces them
to approve what is right and fit” (45). By approval, it is important to see,
Balguy does not mean the kind of observer’s esteem for the motive of a
completed or intended action that Hutcheson and Hume use “approbation”
to express. He means a motivating state an agent is in when they view an act
they consider from a deliberative perspective. “What is the Reason exciting
a Person to the Choice of a Virtuous Action?” he asks, using Hutcheson’s
terminology for what contemporary philosophers call a “motivating rea-
son” (Hutcheson 2002: 138). Balguy’s answer: the agent’s “very Approbation
of it is itself a sufficient [‘exciting’] Reason, wherever it is not over-ruled by
another more powerful” (Balguy 1976: 45). When an agent judges an act
they are considering to be “right and fit,” they are then necessarily already in
a state moving them to do it. Balguy allows that they might akratically fail to
do it, but only through a more powerful motive. They require no additional
desire to be motivated beyond what their judgment of practical reason itself
provides.
Balguy goes farther in pursuing the analogy with belief and truth than
Clarke. Clarke held that while it is “absolutely speaking” necessary that belief
be determined by demonstrated truth, it is only “morally speaking” necessary
that a rational agent do what they judge reasonable and fit. For Clarke the anal-
ogy breaks down owing to the “natural liberty of [the] will,” which does not
hold with assent or belief; neither can be voluntarily adopted (Clarke 1978b:
613). Balguy, however, does not make this distinction.
[W]hy then do we approve? … I answer in one word, Necessity. The
same Necessity which compels Men to assent to what is true, forces them
approve what is right and fit (Balguy 1976: 45).

Motivation is connected as necessarily to judging an act to be right and fit as

belief is to seeing a demonstration that something must be true.
Like Clarke, Balguy calls the motivational state involved in judging the rea-
sonableness of an act a kind of obligation.
Internal Obligation is a State of the Mind into which it is brought by the
Perception of a plain reason for acting, or forbearing to act, arising from the
Nature, Circumstances, or Relations of Persons or Things (31).

https://doi.org/10.1017/9781139025065.009 Published online by Cambridge University Press

 250 the british rationalists and reid

As with Clarke, three things must be distinguished: (a) the circumstances and
relations that make an act reasonable or fit (the facts that are “plain reason[s]
for acting,” like that an act would avoid harm), (b) the act’s being reasonable or
fit (owing to the facts in (a)), and (c) the mental (motivational) state of taking
an act to be reasonable or fit (believing (b) in the light of (a)). (b) is a normative
relation between an act and the non-­normative facts in (a), and (c) is the non-
normative state of accepting (b), which is a normative proposition or fact. Like
Clarke, Balguy calls (c) a kind of obligation, but (c) consists in taking oneself to
be under an obligation in the normative deontic sense (12).4

Price
Richard Price makes this very point in the course of arguing that the concept of
obligation is indefinable. Quoting Balguy’s definition of “internal obligation,”
Price notes that it conflates “the effect of obligation perceived” with “obligation
itself” (Price 1974: 114). Noting the general tendency to use “obligation” to refer
to a distinctive motivational state, including by Clarke, Price argues that this
usage fails to distinguish sufficiently between “motive” and “obligation” (114,
118). As Price also puts it, it conflates “an act and object of the mind,” “a percep-
tion and the truth perceived” (117).5
Price argues that “fit,” “reasonable,” “right,” “morally good,” “obligatory,”
and “ought” all express the same fundamental indefinable normative idea
when applied to acts (41, 104–130). “Fitness and unfitness … when applied to
actions, … generally signify the same with right and wrong” (104). Similarly,
“[m]orally good and evil, reasonable and unreasonable” applied to acts “evi-
dently mea[n] the same” as “right and wrong, fit and unfit” (104). Crucially,
for Price, “right” and “wrong,” are indefinable “simple ideas” in Locke’s sense.
“Our ideas of right and wrong are simple ideas, and must therefore be ascribed
to some power of immediate perception in the human mind” (41). So also,
therefore, are the ideas referred to by the synonymous terms “reasonable” and
“unreasonable,” “fit” and “unfit,” and so on. They all express the same funda-
mental indefinable normative ideas.
Like Balguy and Clarke, Price does not mark Suárez’s distinction between
law and counsel, between being obligatory and being recommended by nor-
mative reasons for acting. The critical element of Suárez’s contrast, again, is
that obligations are necessarily connected to accountability; issues of culpa-
bility necessarily arise with obligations in a way they do not with choosing

4
“[B]y the help of Reason and Reflection, [we can] discover ourselves to be under Obligations,
and that we ought to return good Offices, or Thanks, according to our Abilities” (12).
5
Price does note, however, that “in common language, the term obligation often stands for
the sense and judgment of the mind concerning what is fit or unfit to be done” (117). Dafydd
Daniels emphasizes this point in Daniels 2020.

https://doi.org/10.1017/9781139025065.009 Published online by Cambridge University Press

 price 251

less choiceworthy options. Price does hold that an intrinsic connection exists
between obligation and accountability. “Would a person who either believes
there is no God, or that he does not concern himself with human affairs, feel no
moral obligations, and therefore not be at all accountable?” he asks, implying
that accountability is entailed by obligation (106–107). But Price does not mark
any contrast between the narrower deontic normative accountability-entailing
relation expressed by “obligation” and the more general normative notion
expressed by “normative reason” and “fit” and “unfit” when these are used in
more general normative senses.
Also like Balguy, Price’s foil in much of his moral philosophy is Hutcheson.
Hutcheson too stressed that all moral thought involves simple fundamen-
tal moral ideas, but for Hutcheson, these were approbation and condemna-
tion, ideas that we have from an observer’s perspective when we contemplate
motives and the intended or completed actions they motivate. For Price, how-
ever, the fundamental moral ideas of fitness and unfitness, right and wrong,
are ideas we have in connection with alternatives for choice: acts, considered
mostly independently of motive, as we might consider them from an agential
perspective in deliberating about what to do. It is only when we form an inten-
tion or act that actual motivation is revealed in our reasons for so intending
or acting. Here again, we see a contrast between deontic morality’s focus on
accountability and moral agency, on the one hand, and Hume and Hutcheson’s
virtue-ethical focus on motive and character, on the other.6
The difference with Hutcheson that Price is most concerned to stress, how-
ever, concerns the source of our fundamental moral ideas. Where Hutcheson is
led by Lockean empiricism to infer the existence of a moral sense, Price defends
the ethical rationalist thesis that fundamental moral ideas come from reason.
Here he differs as well from Fordyce, who also holds that the idea of moral
obligation is indefinable, but who argued against Clarke that the source of this
idea is not reason but sense. Against Hutcheson (and Fordyce), Price declares:
What is the power within us that perceives the distinctions of right and
wrong?
My answer is. The understanding (17).

Price complains that moral sense theories like Hutcheson’s make moral
judgments matters of taste and treat morality as response-dependent in the
same way color is. So conceived, morality becomes “the effect of a positive con-
stitution of our minds,” resulting from “an implanted and arbitrary principle
by which a relish is given us for certain moral objects and forms and aversion
to others” (14). Moral sense theorists often opposed their ideas to early modern

6
This is reflected in Gary Watson’s distinction between “responsibility as accountability”
and “responsibility as attributability” (Watson 1996).

https://doi.org/10.1017/9781139025065.009 Published online by Cambridge University Press

 252 the british rationalists and reid

forms of voluntarism that modeled morality on positive law. But Balguy and
especially Price argued that the moral sense theory ultimately made morality
depend no less on arbitrary posits, in this case contingent aspects of the human
psyche rather than voluntary dictates of the will. Price argued, following Cud-
worth, that morality is “eternal and immutable”: objective rather than subjec-
tive, inherently rational rather than arbitrary, necessary rather than contingent,
and universally binding on any possible moral agent (52, 85, 89).
Against the sentimentalists’ empiricist theory of ideas, Price contended that
it could give no account, not just of moral ideas, but of a whole host of concepts
that are of fundamental importance for thought in general, whether theoretical
or practical. Drawing on Plato’s Theaetetus and, especially, Cudworth’s True
Intellectual System of the Universe, Price argued that the concepts of universal-
ity, necessity, substance, power, causation, and the like, necessary for active
thought at all, could not possibly derive from phenomenal sensations alone.
Sense is essentially particularistic and passive – it receives particular images.
(This, of course, is Berkeley’s reason for rejecting “abstract ideas.”) Sense “lies
prostrate under its object,” Price says, using a phrase of Cudworth’s (19).7 This
is a familiar theme among rationalists (think of Descartes’s piece of wax in the
Meditations) (Descartes 2017). And the empiricists themselves acknowledged,
sometimes happily, that many such thought-structuring ideas cannot derive
from sense. Sometimes they outright rejected the rationalists’ concepts (Berke-
ley on “abstract ideas”) or attempted alternatively to provide “skeptical” ways
of explaining them (like Hume on causation and, arguably, moral properties
such as virtue).
In arguing so clearly that “the power … that understands … and that com-
pares all the objects of thought, and judges of them, is a spring of new ideas” like
universality and necessity, as well as the fundamental moral ideas involving
these, Price’s thought looks forward to Kant (18). “Sensation is only a mode
of feeling in the mind; but knowledge implies an active and vital energy of the
mind” (20). Thought and knowledge require something like the spontaneity of
reason, no less in the practical than in the theoretical sphere. When we make
moral judgments, taking, for example, someone to be under a moral obliga-
tion, we think with concepts that cannot come from sense, since they essen-
tially involve the ideas of universality and necessity that cannot possibly derive
from it.
Price and the intuitionists, however, give rational “intuition” or “immedi-
ate perception” of the understanding a fundamental role in their moral episte-
mology that it does not have in Kant’s. “Some truths there must be, which can
appear only by their own light, and which are incapable of proof; otherwise

7
“Sense, that suffers from external objects, lies as it were prostrate under them” (Cudworth
1996: 54).

https://doi.org/10.1017/9781139025065.009 Published online by Cambridge University Press

 price 253

nothing could be proved, or known; in the same manner as … if there were no

simple or undefinable ideas, there could be no complex ideas” (98). Intuition is
the source both of fundamental moral ideas and of our knowledge of the truth
of propositions we are able to formulate with them. “It is on this power of intu-
ition, essential, in some degree or other, to all rational minds, that the whole
possibility of reasoning is founded” (98).
Price thus agrees with the sentimentalists that knowledge is grounded in
immediate perception and expanded from that foundational basis by reason-
ing. And they agree also that all simple ideas must be given in “immediate per-
ception.”8 But they disagree about what the requisite faculty of perception is.
For empiricists, it is given entirely by sense experience, whereas for rational
intuitionists like Price it is a form of intellectual perception that, like Carte-
sian “clear and distinct perception,” is possible only with the rational mind or
understanding.
Sense experience, Price argues, is irremediably imagistic, particularistic, and
subjective, whereas knowledge requires, quite generally, formulation in uni-
versal concepts and objectivity. And fundamental moral knowledge, moreover,
is necessary rather than contingent. “[M]orality is a branch of necessary truth”
(85). Eternal and immutable morality is constituted by deontic standards that
obligate any possible “moral agent.” They are laws for moral agents as such.
Price stresses, again, that unlike sense experience, intuition or perception by
the understanding is active rather than passive. It involves “an active and vital
energy of the mind” (20). This contrast carries through also to Price’s account
of self-determining agency which plays a central role in his account of deontic
morality.
As with Butler, it is important to Price that virtue or moral goodness consists
not in any form of motivation to which a moral agent might be (passively) sub-
ject or simply have. It involves the agent determining themselves by their own
judgment of how they morally should act, as Butler would put it, through their
own conscience. “[A]n agent cannot justly be denominated virtuous,” Price
writes, “except he acts from a consciousness of rectitude” (184). “It is the actual
conformity of the wills of moral agents to what they see or believe to be the fit-
nesses of things, that is the object of our praise and esteem” (184).
“Moral capacit[y],” the power or competence required to be subject to deon-
tic moral standards of “rectitude” and to hold oneself accountable for com-
pliance to them, requires “the power of acting and determining” (181). True
agency, Price holds, cannot be understood through any form of causation
or determination that one passively undergoes. Rather agency is something
agents actively exercise and themselves determine, so that they cause their own
actions. “As far as it is true of a being that he acts, so far he must himself be the

8
“Given” in the sense of the “myth of the given” (Sellars 1956).

https://doi.org/10.1017/9781139025065.009 Published online by Cambridge University Press

 254 the british rationalists and reid

cause of the action” (181). Price calls this power “moral capacity” since he holds
it to be essential to the accountability that deontic standards of “rectitude” nec-
essarily involve.
[I]t is hard to say what virtue and vice, commendation and blame, mean,
if they do not suppose agency, free choice, and an absolute dominion over
our resolutions.—It has always been the general, and it is evidently the nat-
ural sense of [humankind], that they cannot be accountable for what they
have no power to avoid (182).

Here is another aspect of the contrast between Price’s focus on deontic moral-
ity and the version of virtue ethics we find in Hutcheson and Hume. The latter
feel no dialectical pressure in the direction of a theory and ideal of autonomous
moral agency that Price so evidently does.
Price does not draw incompatibilist conclusions from his doctrine of “lib-
erty” as agent causation. His focus is squarely on ethics rather than metaphysics.
If … advocates of the doctrine of necessity should find, that what they
mean by necessity is not inconsistent with the ideas of agency and self-­
determination, there will be little room for farther disputes (183).

Price is happy to allow that doctrines of universal causation might be true as a

metaphysical matter. All that matters is that the idea of self-­determination be in
good working order for the purposes of deontic morality.
Price holds that agents determine themselves by making their own judg-
ments of what is fitting and right, and by acting on those.
The intellectual nature is its own law. It has, within itself, a spring and
guide of action which it cannot suppress or reject. Rectitude is itself an end,
an ultimate end, an end superior to all other ends (187).

We determine ourselves by guiding ourselves by the law of our intellectual

nature, which makes us active beings.
In a pointed contrast with Hutcheson, Price distinguishes between “instinc-
tive” and “rational” benevolence. Whereas Hutcheson holds that virtue is
benevolence pure and simple, with guidance by the moral sense being not
only unnecessary, but not actually directly possible, Price claims that any such
“instinctive benevolence is no principle of virtue” (191). “Rational benevolence” is
virtuous, Price allows, but that is because “[r]ational benevolence entirely coin-
cides with rectitude, and the actions proceeding from it, with the actions pro-
ceeding from a regard to rectitude” (191). Virtue always consists in the agent’s
doing what they think fitting and right because they themselves think so. That is
obviously consistent with beneficence, and it can be consistent also with benev-
olence, so long as the agent’s governing motivation is a “regard to rectitude.”
Again, Price does not distinguish between rectitude in a narrower, distinc-
tively deontic sense and a broader notion of fittingness or support by normative

https://doi.org/10.1017/9781139025065.009 Published online by Cambridge University Press

 price 255

practical reasons. This gives him a very tight circle of concepts when it comes
to the question of why be moral. We have seen that Butler can be interpreted
in a way that makes this question at least superficially open by distinguishing
between narrowly moral deontic judgments of right, made through conscience,
on the one hand, and broader normative judgments of reason (what Butler calls
“superiority” and “inferiority”) that are made with the “principle of reflection.”
But Price makes no such distinction. He says that intellectual nature or rational
agency aims at doing what is “fitting” and “right” indifferently.
This turns the question of “why be moral (do what is morally obligatory or
right)?” into the question of “why do what is fitting (do what there is good
reason to do)?” If that question asks for anything like justification, it is self-
answering. But what explains why the agent’s answering the normative ques-
tion motivates them? This puts us back in the territory of Hume’s challenge:
How can we “prove a priori that these relations, if they really existed and were
perceived, would be universally forcible and obligatory” (Hume 1978: 466)?
What is the “connexion betwixt the relation and the will” (465)?
Price, like Clarke and Balguy, takes up this internalist challenge, refusing
to take refuge in an externalist intuitionism as some later intuitionists, like
Ross, do. “It seems extremely evident,” Price says, “that excitement belongs
to the very ideas of moral right and wrong” (Price 1974: 186). By this, Price
clearly does not mean that an act’s being unfitting or wrong is the same thing
as the agent’s being motivated not to do it, or even that the latter is entailed by
the former. That would run over the distinction he is so concerned to make
against Clarke and Balguy between the normative claim that an act is fitting,
right, or obligatory, and the psychological proposition that someone perceives
or accepts that and is moved by it. It would blur the distinction between nor-
mative and motivating reasons. Price clarifies his meaning, saying that “excite-
ment” “is essentially inseparable from the apprehension of” the “ideas of moral
right and wrong” (186). “When we are conscious that an act is fit to be done, or
that it ought to be done,” he continues, “it is not conceivable that we can remain
uninfluenced, or want a motive to action” (186).
But Price thinks the explanation of this is so straightforward almost to
be trivial. “Why a reasonable being … has a disposition to follow reason…;
why he chuses to do what he knows he should do…, are questions which
need not, and deserve not to be answered” (187). This follows directly from
Price’s account of rational agency, agent causation, and self-determination.
­Something does not so much count as the action of a self-determining rational
agent unless it is chosen under the guise of the reasonable, fitting, and right. It
is important, again, that these latter amount to the same concept for Price. He
works with no distinction between law and counsel; if it so much as counsels,
“reason commands” (188).
Although it may seem to, this need not rule out the possibility of akrasia,
action contrary to the agent’s considered beliefs about what they have most

https://doi.org/10.1017/9781139025065.009 Published online by Cambridge University Press

 256 the british rationalists and reid

reason to do. Price could hold that there is a difference between the latter and
what seems to the agent most fitting in the moment because of framing effects
and the pressure of nonrational motivation. Price does not, however, make this
move.
Price thus assembles in its most mature form among the British rational-
ists an account of deontic morality that puts together mutually supporting ele-
ments that appear also in Clarke and Balguy, as well as Cudworth and Butler.
1. Morality comprises a set of deontic norms, laws of “rectitude” that
2. universally bind all (any possible) moral agents,
3. making them accountable for compliance.
4. Moral agency involves the capacity for self-determination, which is
5. the capacity to do, and hold oneself accountable for doing, what one thinks
right.
6. Morally good (virtuous) action consists in the excellent exercise of this
capacity (action from “consciousness of rectitude”).
This is not a potpourri of views. All six are at least mutually supporting and
perhaps even mutually entailing. If there is a fundamental organizing idea, it is
the notion that deontic norms of right involve accountability in their nature.
As writers as diverse as Butler and Pufendorf illustrate, once we have that idea,
the capacities necessary for accountable agency follow in its train, and good
moral agency as their excellent exercise follows accordingly.
This package of views does not entail metaethical rationalism, even less ratio-
nal intuitionism, as the example of Fordyce proves. But if, like Price, Clarke,
and Balguy, one treats rational agency as coterminous with moral agency, and
makes no distinction between what there is normative reason (what it is fitting)
to do, and what is morally demanded, required, or obligatory, then some form
of rationalism seems virtually unavoidable. If rational agency is sufficient to
make one subject to moral obligation, then practical reason must be sufficient
also for accountability. And if that is so, then reason itself must equip an agent
with all that is necessary for right and wrong to be evident to them and for them
to hold themselves to these standards.9
The central importance for Price of conscientious conduct shows itself also
in a distinction he makes between “abstract” and “practical” forms of virtue.
Abstract virtue is the same as fittingness and right. It is what “it is right such
an agent, in such circumstances, should do,” “considered independently of the
sense [belief] of the agent” (177). Practical virtue, by contrast, is what an agent
should do, given their beliefs. Since there are two ways this might be taken, it is
important to see how Price understands it. By “practical virtue,” Price does not
mean what an agent should do, conditional on what they believe, that is, if their

9
See here Schiffman 2015.

https://doi.org/10.1017/9781139025065.009 Published online by Cambridge University Press

 price 257

beliefs were to be true. Practical virtue rather means what an agent should do
conditional on their believing as they do.
This goes farther than simply saying that moral goodness or virtue consists
in the agent’s following their conscience. It says, in addition, that the content of
the agent’s conscience also affects what they should do. Price no doubt thinks,
like Butler, that most people’s consciences are reliable much of the time. But he
acknowledges that we can be mistaken. So he says that “[o]ur rule is to follow
our consciences steadily and faithfully, after we have taken care to inform them
in the best manner we can” (179). He counsels caution, however. “[W]­here we
doubt,” we are “to take the safest side, and not venture to do any thing concern-
ing which we have doubts, when we know there can be nothing amiss in omit-
ting it” (179).
But what determines abstract virtue or moral right? Price formulates this as
the question “[W]hat are the chief heads of virtue[?]” (meaning abstract virtue
or right) (131). Here Price is an intuitionist also, but in a yet further normative-
theoretic sense that became popular in the nineteenth and twentieth centuries.
Like early twentieth-century intuitionists, such as Ross and Prichard, Price
held that there are a multiplicity of principles of right (what Ross called “prima
facie duties”) that are self-evident to rational intuition and that can neither be
derived from nor explained by more fundamental principles (Ross 2003). Price
was what Sidgwick called a “dogmatic intuitionist,” what Mill called an “intui-
tive” moralist, and what Rawls called simply an “intuitionist” (Sidgwick 1967:
102; Mill 2002: 2; Rawls 1971, 3).
Price’s intuitionism in this pluralist sense closely follows Butler’s critique of
Hutcheson’s doctrine that virtue can be “resolved into” benevolence, laid out
in Butler’s Dissertation, which Price quotes virtually in its entirety (Price 1974:
131–132). For his part, Price argues that the “heads” of virtue include duties
to God, duties of self-regard, of beneficence, of gratitude, of veracity (under
which he, implausibly, considers promissory obligations), and duties of justice
(131–164). Although he evidently holds these “heads” or “principles” to be irre-
ducibly normatively plural, he nonetheless says that “they all run up to one gen-
eral idea, and should be considered as only different modifications and views of
one original, all-governing law” (165). What he evidently means by this is not
that there is a single principle from whose content the content of all the rest can
be derived (in the way he thinks that the duty to keep promises can be derived
from the duty to tell the truth), but just that they all “are alike our indispensable
duty.” “It is the same authority enjoins, the same truth and right that oblige,
the same eternal reason that commands in them all” (165). All are equally part
of the same body of law of “rectitude”; all derive from “intellectual nature” or
reason.
Price recognizes that these different “principles” or “heads” can conflict.
When they do, it can be difficult determinately to weigh their conflicting

https://doi.org/10.1017/9781139025065.009 Published online by Cambridge University Press

 258 the british rationalists and reid

claims. “[S]o variously may different obligations combine or oppose each other
in particular cases, and so imperfect are our discerning faculties, that it can-
not but happen, that we should be frequently in the dark …” (167). As with
Ross, who will hold that it is only prima facie duties that are self-evident, and
that how to balance them in particular cases may not be, so Price says that it is
“[t]he principles themselves” that “are self-evident,” even if “obscurity attend[s]
several cases wherein a competition arises between the several principles of
morality” (168).

Reid
We prefaced our examination of the canonical British ethical rationalists –
Clarke, Balguy, and Price – with a short discussion of Fordyce. Fordyce accepts
all the elements of the deontic moral conception that find their most mature
formulation in Price, but he denies metaethical rationalism. He agrees that
morality comprises a set of norms of right that obligate all moral agents, making
them accountable for compliance, with moral agency including the capacities
for self-determination and holding oneself accountable through conscience.
And he accepts that virtuous or morally good action consists in the excellent
exercise of this capacity; it is conscientious conduct. But whereas the rational-
ists identify conscience with reason, Fordyce claims it requires a “Moral Sense”
(Fordyce 2003: 12).
What drives Fordyce to this conclusion is his Lockean theory of ideas. Like
Hutcheson, he accepts that moral concepts involve a fundamental simple idea
that can be received only through a distinct, moral sense. But Fordyce departs
from Hutcheson on what the fundamental moral idea is. For Hutcheson, it is
moral goodness of motives, the essential kernel of which is approbation, an
observer’s response upon contemplating the motive of benevolence. Against
this, Fordyce agrees with the rationalists that the fundamental concept of
morality is the deontic idea of moral duty or obligation (28).
As we have recently seen, Price makes this point a central part of his argu-
ment that the source of our moral ideas is reason or “The Understanding”
rather than sense (Price 1974: 17, 41). No sense, as the empiricists understand
them, could possibly give rise to the idea of moral duty with its entailments of
necessity, universality, and objectivity.
To end this chapter, we turn to an important philosopher, Thomas Reid,
who agreed with the rationalists on almost all these points, but who is nonethe-
less not a canonical ethical rationalist. Like Fordyce, Reid is prepared to call the
“original power of the mind” through which we “have the notions of right and
wrong” and also “perceive certain things to be right, and others to be wrong,”
“moral sense” (Reid 1969: 231). And Reid is skeptical about canonical rationalist
formulations like “acting contrary to the relations of things, … to the fitness of

https://doi.org/10.1017/9781139025065.009 Published online by Cambridge University Press

 reid 259

things” (472). Clarke and Balguy’s “phrases have not the authority of common
use” and so are no part of moral commonsense (472).
Like Butler, from whom he draws many of his arguments and terminology,
Reid carefully avoids taking a definite side in the debate between rationalists
and sentimentalists. In the Dissertation, Butler says that it matters little whether
conscience is called “moral reason” or “moral sense” (Butler 2017: D.1) and
then uses phrases that seem purposely designed to blur the lines: “sentiment of
the understanding” and “perception of the heart” (D.1). Reid is similarly cagey,
wishing not to enter into “subtile disputes among modern philosophers” about
whether the moral “power of the mind” that is the source of moral ideas and
perceptions “ought to be called reason, or be not rather some internal sense
or taste” (Reid 1969: 71, 231). Reid also follows Butler in saying that the moral
faculty is “cool,” “dispassionate,” and authoritative. It operates “in a manner
so like to judgment or reason, that even those who do not allow it to be called
by that name, endeavour to account for its having always had that name” (71).
Reid actually agrees with Price that no moral sense, as empiricists like Hume
and Hutcheson understand it, could possibly be a source of moral ideas and
knowledge. Reid’s critique of the empiricist theory of ideas and sense experi-
ence is a centerpiece of his epistemology quite generally. As Reid sees it, the
central problem with early modern empiricism is that it takes the objects of
sense experience to be “sensations in our own minds” (458) rather than ordi-
nary objects and their properties. Empiricists end up restricting the scope of
justified empirical judgments to mental phenomena: sensations and feelings.
This makes empirical judgment irremediably subjective.
Reid thinks, by contrast, that we use our senses to make justified objective
empirical judgments directly about ordinary objects and their properties.
Moreover, as we shall see presently, he advances a careful analysis of justi-
fied intuitive moral knowledge through moral sense. Thus, although Reid is
happy to use Hutcheson’s term “moral sense,” he rejects Hutcheson’s empiri-
cist understanding of that term. “I find no fault with the name moral sense,”
Reid says (398). But he adds that “[m]odern philosophers have conceived of
the external senses as having no other office but to give us certain sensations,
or simple conceptions, which we could not have without them” (398–399).
“[T]his notion has been applied to the moral sense. But it seems to me a mis-
taken notion in both” (399).
Reid’s Essays on the Active Powers of the Mind (1788) virtually begins with
the signature rationalist (and Butlerian) claim that “[e]very thing virtuous and
praiseworthy must lie in the right use of our power; every thing vicious and
blameable in the abuse of it” (1). The relevant “active power” is the “capacity
of self-government,” which Reid identifies with the capacity to govern one-
self by a “conception of a law” and “its obligation” (2). Virtue is conscientious
conduct, an agent’s determining and holding themselves to the moral law. It

https://doi.org/10.1017/9781139025065.009 Published online by Cambridge University Press

 260 the british rationalists and reid

is the excellent exercise of the capacity that makes them moral agents, subject
to deontic morality, and accountable for compliance with it. “Two things are
implied in the notion of a moral and accountable being; understanding and
active power” (315). A moral agent “must understand the law to which he is
bound, and his obligation to obey it” (315). And a moral agent must have the
“power to do what he is accountable for” (316).
Reid defines “active power” as “an attribute in a being by which he can do
certain things if he wills” (38). And he similarly defines “will” as a “power to
determine, in things which he conceives to depend upon his determination”
(57). With these, he advances a subtle theory of agency as agent causation that
is significantly more sophisticated than Price’s. Reid distinguishes the (active)
power of will from the exercise of that power: “the act of determining” or “voli-
tion” (57). And he distinguishes the latter from all “motives” or “incitements”
to the will, which other philosophers mistakenly identify as “modifications”
of the will (58). The exercise of will is always (as Cudworth and Locke both
emphasized) a self-­conscious self-determining activity, rather than anything
that happens in or to the agent.10 Reid carefully botanizes the wide variety of
motives and “incitements” to which human beings are subject, distinguishing
between “mechanical” instincts and habits, “animal” appetites and passions,
and “rational” motives of prudence and moral duty. Throughout, he is clear
that it is always the agent that wills, and so causes, their own actions. Motives
“influence” but do not cause actions. Only agents can do that.
An especially important part of Reid’s action theory for his ethics is a mental
state he calls “a fixed purpose or resolution with regard to our future conduct”
(83). This is the kind of prospective intention on which philosophers only really
came to focus in the wake of Michael Bratman’s work on intention in the late
twentieth century (Bratman 1987). It is close, however, to what Kant consid-
ers under the heading of adopting or “setting” an end (e.g., Kant 1996c 6:239).
We form a fixed purpose or resolution when we form a firm intention to do
something in the future. Resolutions thus involve the will for Reid in a way that
motives like desire can never, however strong they might be.
Since Reid takes virtue necessarily to involve the will, it can never be consti-
tuted by motivating states, like desires, to which an agent is passively subject.
Virtuous agency thus involves self-determination, and virtuous agents adopt
fixed purposes and resolutions. The virtue of justice, Reid writes, is “nothing
but a fixed purpose, or determination, to act according to the rules of justice”
(Reid 1969: 85). So indeed, Reid says, are all virtues; “[t]hey are all fixed pur-
poses of acting according to a certain rule” (85). (By “rule” here Reid means a
rule of right.)

10
Reid quotes approvingly Locke’s definition of will as “an act of the mind, knowingly exert-
ing that dominion it takes itself to have over any part of the man, by employing it in, or
withholding it from any particular action” (58).

https://doi.org/10.1017/9781139025065.009 Published online by Cambridge University Press

 reid 261

“By this,” Reid says, “the virtues may be easily distinguished … from natu-
ral affections that bear the same name” (85). Hutcheson, by contrast, identi-
fies benevolence with a desire for the good of others (or for one’s own good,
impartially considered). Reid holds that, so understood, benevolence cannot
be a virtue. According to Reid, “[t]he virtue of benevolence is a fixed purpose
or resolution to do good when we have opportunity, from a conviction that it is
right, and is our duty” (86).
The boundaries of virtue and right are coterminous with those of voluntary
moral agency. “Hard would it be indeed to think, that a man should be born
under a degree of reprobation, because he has the misfortune of a bad natural
temper” (86). Reid analyzes active power into three “voluntary” “operations
of mind” (76–89). Two of these, attention and deliberation, are familiar from
Cudworth on the “autexousious power” and from Locke’s account of “free
agency” in the second edition of the Essay. Moral agents can voluntarily direct
their attention to consider alternative choices before them in deliberating
about what to do (though “[w]e ought not to deliberate in cases that are per-
fectly clear” (80)). To these, Reid adds the voluntary forming of fixed purposes
or resolutions (83).
Together these constitute our job as moral agents and all we can be held
accountable for:
If any man could be found who, in the whole course of his life, had given
due attention to the things that concern him, had deliberated duly and
impartially about his conduct, had formed his resolutions, and executed
them according to his best judgment and capacity, surely such a man
might hold up his face before God and man, and plead innocence (92).

Like Butler, Reid treats both what Butler calls “reasonable self-love” and
conscience as rational principles of action. “To act with a view of some distant
interest, or to act from a sense of duty, seems to be proper to man as a reason-
able being” (83). Both a regard to “what is good for us upon the whole” and the
concern to do “what appears to us to be our duty” are “rational principles” (202–
203). Both are “leading and governing principle[s],” in Butler’s sense (202).
This raises a number of issues, however, as it did with Butler. The “rational
principle of a regard to our good upon the whole” “is so similar to the moral
principle, or conscience, and so interwoven with it,” Reid says, “that both are
commonly comprehended under the name of reason” (210). We are “self-
condemned” when we go against our duty and when we pursue a nearer lesser
good over a greater one in the further future (209). A regard to our own greatest
good, like conscience, Reid writes, “gives us the conception of a right and a wrong
in human conduct,” but he quickly adds, “at least of a wise and a foolish” (210).
That is the problem. Reid’s considered view is that the idea of right or moral
obligation is a different notion from that of the agent’s good. It is a distinc-
tively deontic “relation of its own kind … perhaps too simple to admit of logical

https://doi.org/10.1017/9781139025065.009 Published online by Cambridge University Press

 262 the british rationalists and reid

definition” (229; see also 223). The idea of right contains a conceptual connec-
tion to accountable agency that the concept of the agent’s own good does not
have. To constitute the normative relation of obligation, an action must be
“voluntary” and “within the sphere of [the agent’s] natural power,” and the
agent must have the “understanding and will” enabling them to know their
obligations and to hold themselves to them (229). Nothing like this is true of
the concept of the agent’s greatest a good or a rational concern for it.
Prudence and conscience are, therefore, “distinct principles” (210). It is con-
science or the “moral faculty” that is the “original power of the mind” through
which we “have the notions of right and wrong” and also “perceive certain
things to be right, and others to be wrong” (231). Like Butler, Reid reserves
“authority” for conscience rather than self-regard (254). It would seem to fol-
low, as it does for Butler, therefore, that the only way acting against rational
self-regard can make us self-condemned in the deontic accountability sense is
if imprudence also violates conscience. Only conscience or the moral faculty
provides us with the deontic conceptual materials to blame or condemn our-
selves. These are, Reid emphasizes, “acts of judgment” that involve the “exer-
cise [of the] moral faculty” (464). Moreover, Reid says, the idea of right that is
essential to conscience’s judgment cannot be “resolved into that” of our “inter-
est, or what is most for our happiness” (223).
“Prudence is a virtue,” like “benevolence,” “fortitude,” justice, and others.
But the “essence and formal nature” of every virtue, including prudence, “must
lie in something that is common to all these, and to every other virtue” (399).
And this, Reid clearly says, “can be nothing else but the rectitude of such con-
duct …, which is discerned by a good man” through the exercise of the moral
faculty (399).11
A deeper point is that although the “regard to duty” and the “regard to our
good on the whole” are both “motives” or “rational principles of action,” Reid
speaks only of conscience or the “moral faculty” as a “power,” one involving
both the power to understand the right and the active power to determine our-
selves to act in light of it.12 Butler does not make this distinction, as he has a less
sophisticated theory of action than Reid does.
We noted in our discussion of Butler that it is possibly profitable to read
him through the perspective of a distinction between a broader “principle of

11
Cf., however, “We ought to prefer a greater good, though more distant, to a less. … A
regard to our own good, though we had no conscience, dictates this principle; and we can-
not help disapproving the man that acts contrary to it” (362). This is like the passage in
Butler’s sermons where he says that, in a similar case, “we may have a clear conception of
the superior nature of one inward principle to another” (self-love over a particular present
appetite) “without particular consideration of conscience” (Butler 2017: II.11). See p. 179
above for discussion of this same issue in Butler.
12
“Of Regard to our Good on the Whole,” is the title of Essay III, Chapter II.

https://doi.org/10.1017/9781139025065.009 Published online by Cambridge University Press

 reid 263

reflection” that makes normative judgments of practical reason in general, and

conscience, which makes narrower judgments of deontic morality. It is more
difficult to read such a distinction into Reid, however. The reason is that Reid
explicitly says that the deliberative question is “whether the action under …
deliberation ought to be done, or ought not,” and for Reid, like the British
rationalists, the only ought is the narrow deontic moral ought of right and obli-
gation. “[T]he notion or conception of duty,” Reid says, can be defined “only
by synonymous words or phrases, … as when we say that it is what we ought
to do” (223).
There seems to be no place in Reid’s thought for the more general practical
normative notion of what a person should do in the sense of what there is most
normative reason for them to do. Had Reid theorized with this more general
notion, he might have made a distinction between the moral active power and a
more general active power of practical reason that could have had the power of
rational prudence as a species.
Although Reid recognizes the difference between self-regard and conscience
(regard to duty) as rational motives or “principles of action,” he holds that
Divine providence guarantees that there can be no genuine conflict between
them. Any “opposition between the two governing principles … is merely
imaginary” (256). So long as “the world is under a wise and benevolent admin-
istration, it is impossible, that any man should … be a loser by doing his duty”
(256–257).
Reid’s normative moral theory is laid out in his lectures on practical ethics
(Reid 2007). He holds with Price that moral knowledge and reasoning must
“rest upon one or more first principles of morals, whose truth is immediately
perceived without reasoning” (1969: 235). And like many of the British ratio-
nalists, he makes the golden rule a “first principl[e] in morals”: “that we ought
not to do to another, what we should think wrong to be done to us in like cir-
cumstances” (234).
Of particular interest in Reid’s metaethics is an argument he makes against
empiricist sentimentalism, represented most clearly by Hume, that crystallizes,
arguably for the first time, a contrast between cognitivism and non-cognitivism
that will loom large in metaethical debates of the twentieth century. Reid crisply
states the issue in terms that refer explicitly to the sentiment that figures cen-
trally in the moral sentimentalism of Hutcheson and Hume.
[F]or half a century, it has been a serious dispute among philosophers,
what this approbation and disapprobation is, whether there be a real judg-
ment included in it, which, like all other judgments, must be true or false;
or, whether it includes no more but some agreeable or uneasy feeling, in
the person who approves or disapproves (457).
For Hutcheson, approbation is the simple idea that moral sense receives and
through which moral goodness is defined. And although Hume rejects moral

https://doi.org/10.1017/9781139025065.009 Published online by Cambridge University Press

 264 the british rationalists and reid

sense as part of our “primary constitution,” approbation figures prominently in

his sympathy-based sentimentalism (Hume 1978: 473).
When we discussed Hume, we noted that he formulates his view in different
ways, some of which suggest a subjectivist form of cognitivism, some of which
suggest a non-cognitivist or expressivist reading. Hume himself doesn’t seem
to notice the differences, so it seems anachronistic to attribute one or the other
of the views clearly to him. But Reid, perhaps for the first time, shows precisely
what the difference and issue is between these alternatives.
The key question, Reid says, is whether a given sentiment, like approbation,
necessarily involves judgment. As Reid understands it, a judgment necessar-
ily has an object, which is always some “proposition” that is capable of being
true or false. “In judgment, we can distinguish the object about which we
judge, from the act of the mind in judging of that object” (Reid 1969: 460). And
“[e]­very judgment must necessarily be true or false” (460). Feelings, by con-
trast, are distinguished by their felt phenomenal features, which may be either
“agreeable” or “uneasy” (457).
Reid allows that in “many operations of the mind,” feeling and judgment
are “inseparably conjoined under one name.” The crucial question is whether
the sentiment expressed by what a moral sentimentalist calls moral “judg-
ments” (as even sentimentalists are disposed to call them) necessarily involves
a judgment in the sense that Reid distinguishes, namely, an attitude toward a
proposition that is held to be true. Were Hume to confront the issue framed
in this way, it would force him to make a choice. Since he holds that “reason
is the discovery of truth and falsehood,” then it will follow from the claim that
moral “judgments” are made through sentiment rather than reason that they
are not really judgments in Reid’s sense. It would mean that Hume’s holding
that moral judgments express, rather than self-attribute, feelings would com-
mit him to metaethical non-cognitivism, the view that moral judgments can-
not be true or false.
In these terms, Hutcheson, however, would seem clearly to count as a cogni-
tivist, since he holds that moral goodness is the property of (motivated) actions
that “procures” approbation, in the sense that disinterestedly contemplating
the action causes it. That is a proposition that can be straightforwardly true
or false. The point is that, unlike Hutcheson, Hume seems tempted by non-
cognitivism when he says such things as “morality is more properly felt than
judged of” and that “we do not infer a character to be virtuous, because it
pleases,” but rather our “judgment” of virtue is itself an expression of our plea-
sure (Hume 1978: 470, 471).
Reid also offers two further arguments against understanding moral judg-
ments in a subjectivist way. “Suppose … my friend says, Such a man did well
and worthily; man’s conduct is highly approvable,” and then “in relation to the
same case … says, The man’s conduct gave me a very agreeable feeling.” That
these two statements mean different things is shown, first, by the fact that the

https://doi.org/10.1017/9781139025065.009 Published online by Cambridge University Press

 reid 265

friend’s first statement is plainly not about themselves, whereas the second one
is. A second “reason these two speeches cannot mean the same thing,” Reid
writes, is that one can deny the first, but not the second, “without an affront”
(Reid 1969: 464–465). It is presumptuous and potentially insulting to deny that
a person feels what they say they feel, though it is not to express disagreement
with their moral opinions. These arguments work against only analytical or
definition forms of subjectivism, and only crude ones at that. More sophisti-
cated versions, like Hutcheson’s can avoid them.
To conclude this chapter, we should note some particularly interesting and
philosophically important aspects of Reid’s claims about justice, especially in
the context of his critique of Hume’s claim that justice is an artificial virtue.
These recall somewhat similar points in Smith and, in some ways, also look
backward to Grotius and Pufendorf on sociability, and forward to later devel-
opments in Fichte and Hegel, as well as in recent moral philosophy.13
Hume’s claim that justice is artificial, recall, is that justice requires a rule-
structured practice that is established by a mutually advantageous “convention”
(Hume 1978: 490). Reid concedes that a “conception of the virtue of justice” is
impossible until people “have lived some time in society.” But he denies that
this shows that justice is conventional, since reason itself depends on develop-
ment and maturation within a social context (Reid 1969: 405). The sense of
justice arises simultaneously, he claims, with the capacity for moral judgment
in general (406). Moreover, this sense already involves the apprehension of “an
obligation to justice, abstracting from the consideration of its utility” (406).
As evidence of these claims, Reid cites the fact that the concepts of favor
and injury come with normal human (social) development, independently of
Humean conventions, and that these concepts presuppose those of justice and
injustice. The very idea of a favor is of an intentional benefit that goes beyond
anything its beneficiary has the standing to expect or demand as their just due.
And the notion of an injury is of a harm that violates a norm of justice.14 It is
one thing to see ourselves as benefited or harmed by others, even intentionally,
and another to regard another’s act as a favor or injury. The latter only makes
sense in relation to warranted expectations that Reid collects under the concept
of justice. What justice requires and allows, as he puts it, “fills up the middle
between these two” (410).
Reid explicitly connects these notions, moreover, to reactive attitudes in the
same way Smith does: “A favour naturally produces gratitude. An injury done
to ourselves produces resentment; and even when done to another, it produces
indignation” (410). Before we have the concepts of favor and injury, we can, of

13
For example, in Darwall 2006, 2013b, 2013c. In what follows I draw from Darwall 2006:
192–200.
14
Obviously, what Reid means by “injury” is nothing physical or medical, but something that
is appropriately resented, as in, “Sir, you do me an injury.”

https://doi.org/10.1017/9781139025065.009 Published online by Cambridge University Press

 266 the british rationalists and reid

course, be pleased when we are benefited, especially when it is intentional, and

displeased when we are harmed. But to have the concept of a favor or injury, we
must have some notion of what we can justifiably expect or demand from oth-
ers. Favor and injury thus presuppose a standing to have expectations and make
demands of one another and hold each other accountable. “He perceives that
injury is done to himself, and that he has a right to redress. … Even the injurious
person is conscious of his doing injury; he dreads a just retaliation; and if it be
in the power of the injured person, he expects it as due and deserved” (416–417).
Among the most interesting of Reid’s criticisms are those he makes of Hume
on the obligation to keep promises. Hume’s account of promising is notori-
ously byzantine,15 and it would take us too far afield to attempt to master all its
intricacies. Suffice it to say that although Hume sensibly understands promis-
ing to involve the voluntary undertaking of an obligation (“the willing of that
obligation, which arises from the promise”), his moral psychology and virtue
ethics require that the obligation be understood via the moral sentiment’s role
in an individual’s psychic economy (Hume 1978: 516–517). Against this, Reid
points out that the ideas of promise and its resulting obligation are essentially
social, they are to others, and, therefore, involve what he calls “social operations
of the human mind.” Social operations contrast with “solitary” ones in that the
former, unlike the latter, “cannot exist without being expressed by words or
signs” to another person (Reid 1969: 438).
Neither, Reid argues, can the obligation to keep promises depend on the use-
fulness of a conventional social practice of promising, as Hume supposes is true
of the moral obligation to justice generally. Reid anticipates the recent psychol-
ogy of moral development in arguing that children are capable of understand-
ing, and naturally accept, the basic fairness and reciprocity involved in keeping
faith long before they have any conception of its social utility (444).16
Reid stresses throughout the necessity of a pre-conventional form of recip-
rocally recognized mutual obligation and accountability that individuals must
already implicitly recognize in order for them to come to have genuine con-
ventional obligations at all. For this point, it does not much matter whether we
reserve the words “promise” and “contract” for undertakings that require spe-
cific conventional contexts. If we do, then the point will remain that we could
not come to have the conventionally established obligations of promise and
contract unless we were capable of the pre-conventional second-personal obli-
gations that make conventional obligations generally possible in the first place.

15
Or, as Hume might have put it, “Romish,” since he says that promising is “one of the most
mysterious and incomprehensible operations that can possibly be imagin’d, and may even
be compar’d to transubstantiation, or holy orders, where a certain form of words, along
with a certain intention, changes entirely the nature of an external object, and even of a
human creature” (Hume 1978: 524).
16
See, for example, Engelmann and Tomasello 2019.

https://doi.org/10.1017/9781139025065.009 Published online by Cambridge University Press

 reid 267

Suppose we reserve “promise” for a conventionally defined obligation that is

undertaken with tokens of “promise” and its synonyms, and we reserve “con-
tract” for those that require as context the law of contract. So defined, promise
and contract will be essentially conventional, and whatever distinctive obli-
gations they carry will depend in some way on their respective conventional
contexts. They will depend, as Hume puts it, on a certain “form of words” and
therefore on the employment of “symbols or signs instituted” by “the conven-
tions of men” (522). But if Reid is right, these will themselves depend on the
possibility of voluntarily undertaking reciprocal obligations that are not con-
ventional. It must be possible for individuals to create binding conventions by
a kind of agreement that is itself neither conventional (in the usual sense) nor
simply a coordination of self-interested conditional intentions, as in Hume’s
boat-rowing example or in an iterated prisoner’s dilemma.
Reid faults Hume for failing to grasp promising’s essentially “social,” recip-
rocally recognizing character. Commanding, testifying, requesting, bargain-
ing, and promising are all, Reid argues, irreducibly “social”: “They cannot exist
without being expressed by words or signs, and known to the other party”
(438). It is impossible, moreover, “to resolve” social operations “into any mod-
ification or composition of the solitary” (438). It is consequently impossible
to understand promising in terms of such “solitary” Humean mental acts as
“willing [an] obligation” (understanding “obligation” as Hume does) or as the
expression of “a resolution to perform the promised act” (Hume 1978: 516–518).
Additionally, even if the distinctive forms that contract and promise take
in established practices depend on “human invention,” these nonetheless pre-
suppose a second-personal sociality that makes conventional practices pos-
sible in the first place. Reid makes a similar claim about language. No doubt
much of language is conventional also, but without a background of natural,
non-conventional dispositions to respond reciprocally, conventional linguistic
discourse would be impossible: “[T]his intercourse, in its beginning at least,
must be carried on by natural signs, whose meaning is understood by both par-
ties, previous to all compact or agreement” (Reid 1969: 439). “The power which
[a human being] has of holding social intercourse with his kind, by asking and
refusing, threatening and supplicating, commanding and obeying, testifying
and promising,” cannot be human invention, but must rather be a distinctive
“part of our constitution, like the powers of seeing and hearing” (439).
Contemporary epistemological discussion has drawn from related Reidian
claims about testimony (Coady 1992; Burge 1993; Moran 2005). Reid argues that
our practices of reason-giving cannot get off the ground unless we are disposed
to give others authority in theoretical reasoning. We must treat their testimony
as providing reasons for belief in addition to any that others themselves offer
in support of what they say. As Pettit and Smith argue, when we conduct seri-
ous “conversation of an intellectual kind,” we must impute to one another the
authority to give us reasons to believe things (Pettit and Smith 1996: 430–431).

https://doi.org/10.1017/9781139025065.009 Published online by Cambridge University Press

 268 the british rationalists and reid

Reid notes that “as soon as they are capable of understanding declarations,”
children “are led by their constitution to rely upon them.” And “[t]hey are no
less led by constitution to veracity and candour, on their own part” (Reid 1969:
444). Were children not naturally disposed to trust what they are told and to
avow candidly their own beliefs, it is difficult to see how they could possibly be
inducted into practices of theoretical reason-giving or even learn a language.
Reid argues that the analogous points must hold also with practical trust-
worthiness. Fidelity must be the natural default in the practical realm no less
than in the theoretical. Children naturally trust others’ assurances about their
future actions no less than what they say about other matters (444). It would
be impossible, moreover, for us to make agreements, promises, and contracts,
unless we already took ourselves to be accountable to one another and accorded
each other the authority to undertake these distinctive obligations. Violations
of agreements or promises do not simply give the lie to declarations of future
actions, of course. They are distinct forms of injustice with their own charac-
teristic remedies. If I agree or promise to do something, then I am obligated to
make that happen in a way I am not normally obligated to make what I assert
true. I am obligated only to assert what is true, or what I believe to be true.
Again, it doesn’t matter if we reserve “promise” and “contract” for distinc-
tive conventionally defined methods of voluntarily undertaken reciprocal rec-
ognized obligations. Reid’s point is that such conventional obligations will be
possible only if we can presuppose an authority to make claims of one another
and undertake voluntary obligations at all that does not itself depend upon ante-
cedent conventions. We can hear in Reid’s words echoes of Smith on exchange:
One boy has a top, another a scourge; says the first to the other, if you will
lend me your scourge as long as I can keep up my top with it, you shall next
have the top as long as you can keep it up. Agreed, says the other. This is a
contract perfectly understood by both parties, though they never heard of
the definition given by Ulpian or by Titius. And each of them knows, that
he is injured if the other breaks the bargain, and that he does wrong if he
breaks it himself (437).

Like Smith, Reid thus brings mutual accountability, and reactive attitudes as
their medium, directly into his account of justice. And, even more than Smith,
he stresses the idea of a distinctive “social” authority that justice presupposes,
most obviously in the case of mutually undertaken obligations through prom-
ise and contract. This gives what Grotius and Pufendorf called “sociability” an
essential role in an adequate philosophical theory of justice. And it suggests
the possibility that deontic morality itself, with its conceptual connection to
accountability, cannot be adequately accounted for except in some such terms.17

17
As I argue in Darwall 2006, 2013b, 2013c.

https://doi.org/10.1017/9781139025065.009 Published online by Cambridge University Press

 8

Rousseau and Kant

No philosopher is more strongly associated with deontological ethics or is a

more canonical modern moral philosopher than Immanuel Kant (1724–1804).1
In this chapter, we focus on Kant, though we begin, after a brief introduction
of Kant and his historico-philosophical significance, with some discussion of
Rousseau. Jean Jacques Rousseau (1712–1778) is best known as a political phi-
losopher, but there are elements of his thought that have great importance for
moral philosophy and its history and, especially, for Kant. Kant credits Rous-
seau as the source of his signature claim of the equal dignity of rational per-
sons. And Rousseau’s conception of political society as an “association” that
“defend[s] and protect[s] the person and goods of each associate with the full
common force,” but where each “nevertheless obey[s] only himself,” has obvi-
ous resonances with Kant’s “kingdom of ends” in which all are governed by
self-legislated law (Rousseau 2003: 49–50; Kant 1996a: 4:433). At the same
time, Rousseau offers important points of contrast to Kant. Whereas Rous-
seau’s emphasis is essentially social and political, Kant will attempt to argue
for morality as a common law binding all agents that is grounded in practical
reason alone. Rousseau, by contrast, points toward an alternative grounding
in sociability that is reminiscent more of Grotius, Pufendorf, Smith, and Reid.
We will consider this in more detail presently. First, however, let’s set the stage
for Kant.
Bernard Williams calls Kant “the philosopher who has given the purest,
deepest, and most thorough representation of morality” (Williams 1985: 174).
Williams’s famous critique of the “morality system,” whose “special signifi-
cance in modern Western culture” he decries, largely targets Kant (1985: 6).
The aspects on which Williams focuses are mainly the deontic features empha-
sized also by the British rationalists, namely, that morality comprises a set of
deontic norms that universally bind all (and any possible) moral agents; that
moral agency involves the rational capacity for free self-determination by these
norms; and that moral virtue consists in the excellent exercise of this capacity
in conscientious conduct.

1
“If any philosopher is regarded as central to deontological moral theories, it is surely
Immanuel Kant” (Alexander and Moore 2016).

269

https://doi.org/10.1017/9781139025065.010 Published online by Cambridge University Press

 270 rousseau and kant

As we saw, what underlay the British rationalists’ account of free rational

agency was their acceptance of the Suárezian conceptual connection between
deontic morality and accountability. Moral agents can intelligibly be held
accountable for compliance with deontic moral norms only if they have the
capacity to hold themselves accountable in their own practical reasoning. The
rationalists identified this capacity with reason and held it to include con-
science, in Butler’s narrower, distinctively moral sense.2
Notoriously, the rationalists provided no account of how reason includes
conscience, but they made no apologies for this, since they thought any attempt
to provide such an account would risk reducing ethics to psychology. This
brings us to Kant.
Kant is a profoundly systematic thinker, so it is possible fully to grasp his
moral thought only if we appreciate it within the context of his “Copernican
revolution” and “critical philosophy” more generally. It is here that we find the
deepest motivations for his Copernican turn in practical and moral philosophy,
as against more “dogmatic” forms of rationalism represented by the British
rationalists and, for Kant himself, by Leibniz and Christian Wolff (1679–1754).
Kant held that just as theoretical philosophy, including natural philosophy and
metaphysics of nature, is impossible without a prior “critique of pure [theoreti-
cal] reason,” so also does moral philosophy require a critique of practical reason.
This is what prior rationalist moral philosophers failed to provide, according
to Kant. Even when, like Cudworth and Price, they appreciated the incapac-
ity of empiricist and sentimentalist accounts to explain fundamental concepts,
like universality and necessity, and saw, as well, that theoretical and practi-
cal reason must be “active” faculties unlike any “passive” reception of sense
impressions, they provided no systematic account or “critique” of reason’s
powers in which a proper theoretical or practical philosophy might adequately
be situated.
From Kant’s perspective, this was a failure in both theoretical and practical
philosophy. Rationalist theoretical philosophers, like Leibniz and Wolff, failed
to see that the powers of theoretical reason are limited by possible experience
and the transcendental conditions of its possibility. Rationalist attempts to
extend theoretical knowledge beyond these bounds and maintain, for example,
a doctrine of universal causation, can always be countered by equally good argu-
ments for the contrary position. Uncritical theoretical rationalisms unavoid-
ably encounter “paralogisms” and “antinomies” of “pure reason” (Kant 1998).
Kant thought that this problem was compounded in practical or moral phi-
losophy. Theoretical cognition is, by its nature, of objects. Although “[g]eneral
logic abstracts … from any relation of [cognition] to the object, and considers

2
See the distinction made in Chapter 6 between Butlerian conscience in a narrow sense tied to
deontic morality and the “principle of reflection” as a broader capacity of practical reason.

https://doi.org/10.1017/9781139025065.010 Published online by Cambridge University Press

 rousseau and kant 271

only … the form of thinking in general,” the “transcendental logic” from which
the “categories” of empirical cognition are drawn necessarily concerns “rules
of the pure thinking of an object” (1998: A55/B79–80, 102, emphasis added).
Moreover, any object of human cognition must be experienced as “given” in
space and time.
But whereas Kant famously holds that pure theoretical thought involv-
ing only “the form of thinking,” without some content or putative object, is
“empty,” he maintains that the fundamental concepts of practical thought
“have as their basis the form of a pure will as given within reason and therefore
within the thinking faculty itself,” irrespective of the will or practical thought’s
objects (1998: A51/B75; 1996b: 5:66). Kant calls the fundamental concepts of
practical thought “categories of freedom.” And he defines the “positive concept
of freedom” that is realized in pure practical reason, which he calls “autonomy,”
as “the property of the will by which it is a law to itself (independently of any
property of the objects of volition)” (1996a: 4:440, 446; 1996b: 5:65).
It follows that what Kant calls “practical law” and (under human conditions,
as we shall see) “moral law,” must be able to be grounded in a “formal principle”
of the “pure will,” or, as he also calls it, “pure practical reason” (1996b: 5:21–28).
“If a rational being is to think of his maxims as practical universal laws, he can
think of them only as principles that contain the determining ground of the will
not by their matter,” but through their form (5:27).
In the Critique of Practical Reason, Kant’s argument is carried by a distinc-
tion between “material” and “formal” principles, where material principles are
grounded empirically in inclination and formal principles are not. Kant’s ratio-
nalist opponents agree that nothing that is empirically given can ground funda-
mental practical or moral norms. Rationalist principles would count as formal
by Kant’s definition in the second Critique. It seems clear in the Groundwork,
however, that Kant intends his fundamental argument to cut more deeply. The
difference between autonomy and heteronomy, which ultimately coincides
with Kant’s distinction between formal and material principles as well as that
between categorical and hypothetical imperatives, is between principles of
action that can be grounded entirely in the form of the will and those that can-
not, that is, whether any content or object is provided by inclination or by “rep-
resentations of reason” (1996a: 4:441). It is worth quoting what he says there at
length.
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.
This relation, whether it rests upon inclination or upon representations of
reason, lets only hypothetical imperatives become possible (4: 441).

https://doi.org/10.1017/9781139025065.010 Published online by Cambridge University Press

 272 rousseau and kant

This is only the briefest presentation of Kant’s argument, which we will need
to analyze at greater length. I present it briefly now to give some indication of
how Kant’s ethical rationalism differs from that of the British rationalists. By
Kant’s lights, all prior rationalists are “dogmatic,” in the sense he outlines in
the Critique of Pure Reason of attempting to “ge[t] on solely with pure cogni-
tion from (philosophical) concepts according to principles, which reason has
been using for a long time without first inquiring in what way and by what right
it has obtained them” (1998: Bxxxv). Prior rationalists proceeded uncritically.
Had they pursued a proper critique of practical reason, Kant thinks, they would
have seen that moral laws can be grounded only in the form of the will or auton-
omy, again, “the property of the will by which it is a law to itself (independently
of any property of the objects of volition)” (1996a: 4:440).
Autonomy in this defined sense is only one of the things Kant means by
“freedom.” It is this autonomy he has in mind when he refers to practical laws
as “laws of freedom,” in contrast to the “laws of nature” that structure the
objects of theoretical cognition, and when he calls the fundamental concepts of
practical thought “categories of freedom” (1996a: 4:387; 1996b: 5:65). But this
is far from the only sense of “freedom” featuring prominently in his thought.
Another is “freedom of choice” (Willkür) – the “faculty to do or refrain from
doing as one pleases” – which contrasts with pure practical reason or autonomy
(Wille) and is exercised when agents determine themselves by self-given law
(1996c: 6:213).
A third is an essentially interpersonal sense, that in which one person can
restrict or violate another person’s freedom. This is the sense primarily in play
in The Doctrine of Right, for example, in the Universal Principle of Right: “Any
action is right if it can coexist with everyone’s freedom in accordance with a
universal law” (6:230). The freedom to which all persons are entitled is “exter-
nal” freedom or “freedom in the external use of choice” (6:214). Freedom of
choice is internal to the agent, whereas external freedom concerns our capacity
actually to bring about what we choose, or would choose, to accomplish in the
world; it concerns matters external to the agent.
In the Preface to the Critique of Practical Reason, Kant says that “the con-
cept of freedom … constitutes the keystone of the whole structure of a system
of pure reason, even of speculative reason” (1996b: 5:3–4). “Freedom” here is
akin to autonomy, the will’s capacity to be a law to itself and, equivalently, Kant
maintains, to be bound and to bind itself by the moral law. His argument in the
second Critique is that we cannot avoid concluding that we are free in this sense
because we are inescapably aware of being bound by the moral law (“the fact of
freedom”), and that we could be so bound only if we are free: “freedom is real,
for this idea reveals itself through the moral law” (5:4). Freedom in this sense is
the linchpin of Kant’s whole critical philosophy.
Our focus is Kant’s moral philosophy, understood broadly to include his
philosophy of right. So, in addition to appreciating autonomy’s fundamental

https://doi.org/10.1017/9781139025065.010 Published online by Cambridge University Press

 rousseau and kant 273

role in Kant’s critical practical philosophy, we shall need to understand how

this sense of “freedom” relates to the other senses – “freedom of choice” and
“external freedom” – in Kant’s systematic moral theories, including his theo-
ries of virtue and right in the Metaphysics of Morals.3
We should bear in mind throughout that the connection between deontic
morality and accountability is an important source of the accounts of free moral
agency in philosophers like Clarke, Price, and Reid. Only self-determining
moral agents can, they thought, be intelligibly held accountable. This turns
out to be much less clear with Kant. I shall be arguing that although some of
Kant’s arguments borrow plausibility from this idea, it does not, indeed it can-
not, play any fundamental role in his thought. The reason is that for Kant, the
distinctively deontic aspects of morality – in his terms, morality itself strictly
so called – turn out not to be intrinsic to practical reason at all. They are nor-
matively epiphenomenal features of the finite human perspective that can play
a deliberative role only from that point of view and not from the standpoint of
practical reason itself. In contemporary parlance, these deontic aspects neither
create nor respond to any additional normative reasons. Even so, I shall argue,
Kant’s argument for freedom in the second Critique trades on accountability-
related features of deontic morality. Ultimately, Kant’s philosophical case for
autonomy must rest on overall systematic considerations of transcendental
idealism and the critical philosophy more generally.
That deontic moral concepts have no place in pure practical reason itself is
Kant’s view throughout his moral philosophical writings, although it becomes
clearest in The Critique of Practical Reason: “Pure reason is practical of itself
alone and gives (to the human being) a universal law which we call the moral
law” (1996b: 5:31, emphasis added). There what Kant called in the Groundwork
the Categorical Imperative and termed the “supreme principle of morality,”
he calls the “Fundamental Law of Pure Practical Reason” (1996a: 4:440; 1996b:
5:30). “The moral law is, in other words, for the will of a perfect being a law of
holiness, but for the will of every finite rational being a law of duty, of moral
necessitation and of the determination of his actions through respect for this
law” (1996b: 5:82).
Morality and, in particular, moral obligation, is unique to finite beings who
must restrain their inclinations through self-presented oughts and imperatives
that are internal to the (phenomenal) feeling of respect, thereby “necessitat-
ing” their compliance not normatively, but by a natural law of psychology. Kant
understands obligation to be what he calls “necessitation,” by which he means a
psychological process, mediated by the feeling of respect, through which finite
agents overcome inclinations to diverge from practical law – as they see it,
moral law – by representing reason’s directives as imperatival commands. “For

3
Guyer 2006 is especially good on this point.

https://doi.org/10.1017/9781139025065.010 Published online by Cambridge University Press

 274 rousseau and kant

human beings and all created rational beings moral necessity is necessitation,
that is, obligation” (5:81).
It is only from a finite, self-restraining perspective – through the phenom-
enal feeling of respect – that the deontic moral concepts of ought, obligation,
demands, and duty get purchase. They have no deliberative role from the per-
spective of practical reason itself. This makes moral obligation ultimately a psy-
chological rather than a normative relation in a way that is not unlike what
Price objected to in Balguy, Clarke, and sentimentalists like Hutcheson and
Hume. We shall return to this aspect of Kant’s view later.

Rousseau
It will provide useful context for our investigation of Kant’s moral philoso-
phy if we begin with a brief discussion of Rousseau. Central elements of Kant’s
view seem clearly to come from Rousseau, although Rousseau’s theories rely
on a grounding of morality that differs profoundly from Kant’s in pure practi-
cal reason (from a first-person agential perspective). What is fundamental for
Rousseau is something more like “sociability” as it operates in the thought of
Grotius and Pufendorf. This grounding involves an essentially interpersonal,
or second-personal, perspective.
Where Kant and Rousseau agree is on the equal dignity of persons and on a
kind of freedom or autonomy that persons can realize when they comply with
laws each legislates from a common (first-person plural) point of view. Kant
credits Rousseau as having “set [him] straight” about the former. What gives
human life the greatest value is not, as Kant had previously thought, admi-
rable pursuits, like the search for “knowledge,” that create a hierarchy of value
depending on how excellently people pursue them, but something all human
beings have in common. From Rousseau, Kant learned not to value the admi-
rable as “constitut[ing] the honor of humanity,” but rather to “honor men”
themselves, each and every one, as having intrinsic value – Kant’s “dignity” –
on which basis it is possible to “establish the rights of humanity” (Kant 1904:
20:44). Rousseau thus revealed for Kant the equal dignity of persons, which
stands in contrast to what Nietzsche will call the “rank-defining” values recog-
nized in hierarchical orders by honor respect, esteem, and contempt (Nietzsche
2007: 11; Darwall 2013c).
In Of the Social Contract, Rousseau says that the fundamental problem of
political philosophy is to “find a form of association that will defend and pro-
tect the person and goods of each associate with the full common force, and
by means of which, each, uniting with all, nevertheless obey only himself and
remain as free as before” (Rousseau 2003: 49–50). Rousseau goes on to say
that “the social contract provides the solution” to this problem (50). Central to
Roussseau’s solution is his idea of the “general will”: “[e]ach of us puts his person

https://doi.org/10.1017/9781139025065.010 Published online by Cambridge University Press

 rousseau 275

and his full power in common under the supreme direction of the general will”
(50). In contrast with the “will of all” – the aggregate of “private interest[s]” –
the general will is essentially public, general, and common in two distinct
senses (60). Both the will’s form and its content or object are general.
The will is formed from a shared, first-person plural perspective – it is what
we will – and its object is never something particular, but always something
universal or general (67). Only if these two conditions are met can what is
willed be validated as law. Since the perspective of the general will is one that
anyone who is subject to it can share (as a member of the association whose will
it is), those who are required to comply with it end up “obey[ing] [themselves]”
and are not simply subjected to external force.
Kant’s treatment of the moral law and its relation to the kingdom of ends has
the same structure, as we shall see. This is a second thing Kant gets from Rous-
seau. The perspective from which moral agents must be able to will the prin-
ciples (maxims) of their wills, Kant says, is one in which they “abstract from …
all the content of their private ends,” therefore arriving at “common laws” in a
“kingdom of ends” (Kant 1996a: 4:433). This means that moral persons are sub-
ject to laws that can issue from a “general will” in both of Rousseau’s senses –
from a shared general perspective and concerning matters of general principle
rather than issues involving any individuals in particular.
Kant also gets from Rousseau the idea that in governing themselves by laws
that can be grounded in this way, persons realize a kind of freedom or auton-
omy that differs from natural liberty or independence. In a state of nature, out-
side of a social contract, all persons equally share a “natural independence,”
which “common freedom is a consequence of man’s nature” (Rousseau 2003:
42, 63). “[A]s soon as he has reached the age of reason, he is the sole judge of
the means proper to preserve himself, he becomes his own master” (42). Every
rational agent, consequently, has a claim right to autonomy to make their own
decisions. But the kind of autonomy to which one can claim a right differs fun-
damentally from the kind that is realized by the social contract, what Rousseau
calls “moral freedom.” It is this moral freedom that is conceptually related to
yet a third kind of freedom that Kant calls “autonomy.”4
We can see the relations between these three different things that “auton-
omy” can mean by considering what Rousseau says about mastery and slavery.
The Social Contract begins with this famous passage:
Man is born free, and everywhere he is in chains. One believes himself the
others’ master, and yet is more a slave than he. How did this change come
about? I do not know. What can make it legitimate? I believe I can solve
this question (41).

4
For the relations between these, see Darwall 2013g.

https://doi.org/10.1017/9781139025065.010 Published online by Cambridge University Press

 276 rousseau and kant

The freedom with which we are born is natural liberty or independence, the
“common freedom” consisting in the claim right to be our “own master[s]”
(42). Any attempt to usurp this freedom is therefore illegitimate. The only
way this freedom can legitimately be constrained is by an individual’s consent
(through their private will) or by the general will (as it involves the individual’s
will as an equal member). Those who attempt to master others in slavery vio-
late this “common freedom” without either legitimate consent or authorization
from the general will.5
But what does Rousseau mean by saying that anyone who believes them-
selves an “others’ master … is more a slave than they” (41)? Even masters have
the same right to autonomy, that is, natural independence or “common free-
dom,” as those they seek to enslave. Rousseau does not appear here to be saying
that they thereby forfeit that right. In what sense, then, do masters make them-
selves “more a slave”? The obvious answer is that they fail to obey themselves
in the sense that those who govern themselves by the general will do. They
therefore fail to realize autonomy as something that can be achieved. This is the
same general sense of “autonomy” that philosophers have in mind when they
talk about someone succeeding in governing themselves by their own values or
moral principles. In Kant’s terms, it consists in a moral agent’s succeeding in
following self-legislated law.
For Kant, this will bring yet a third kind of autonomy into play that Rous-
seau does not discuss: “autonomy of the will,” that is, “the property of the will
by which it is a law to itself (independently of any property of the objects of
volition)” (Kant 1996a: 4:440). It is no understatement to say that autonomy in
this sense is the linchpin of Kant’s whole moral philosophy. The very possibility
of morality, he thinks, depends upon autonomy of the will, which Kant ties to
pure practical reason. The will instantiates autonomy if, and only if, it is practi-
cal reason. There is, therefore, practical reason only if the will has autonomy.
In the Groundwork, Kant attempts to show that autonomy in this sense is
an inescapable presupposition of practical reasoning from the agent’s (first-
person) standpoint. As we shall see, however, it is doubtful that any such argu-
ment can succeed, as Kant himself seems to have appreciated when he wrote the
Critique of Practical Reason.6 Rousseau has no systematic moral philosophy, so
he does not himself take a stand on this issue. He is, however, an acute observer
and theorist of moral psychology. There are in his writings both an insightful
appreciation of the complexity of our social nature and an implicit distinction
between what Kant, following him, calls “unsociable sociability” and sociable
sociability. It is a distinction that points to the possibility of a grounding of

5
To justify this claim, Rousseau gives a classic argument against the possibility of individuals
consenting to their own slavery on the grounds that any such attempt would be nullified by
the fact that persons cannot renounce or alienate their own “quality as man” (45).
6
See the discussion of the Critique of Practical Reason below.

https://doi.org/10.1017/9781139025065.010 Published online by Cambridge University Press

 rousseau 277

morality (and, indeed, of autonomy of the sociable will) that is like the one we
found in Grotius (Kant 2009: 13).
The central Rousseauean theme here is the distinction between what Rous-
seau calls amour propre, on the one hand, and sociability of the kind that makes
morality and egalitarian politics possible, on the other. Rousseau introduces
amour propre in The Second Discourse on the Origin of Inequality, published in
1754, eight years before The Social Contract (Rousseau 2019). There Rousseau
notes the difference between two kinds of self-love: amour de soi and amour
propre.7 Amour de soi is the drive for self-preservation and other goods for the
self that we have in common with the rest of the animal kingdom indepen-
dently of society. Amour propre, on the other hand, is essentially social. It is
only in societies that human beings begin to look at each other and take an
interest in how they are looked at or regarded. This interest in others’ (and
one’s own) esteem is what Rousseau means by “amour propre” (218). “Vanity,”
as it is sometimes translated, shows itself in a whole host of competitive and
invidious syndromes that Rousseau contrasts with sociability among equals,
as in morality and the social contract. To form a general will, members of a
society must relate to one another as members and see one another as having a
standing to participate independently of how they regard one another in terms
of esteem or contempt.
In a famous passage from Idea for a Universal History with a Cosmopoli-
tan Aim, Kant takes special note of the distinction between what he calls the
“unsociable sociability” created by amour propre and the sociable sociability
necessary for egalitarian society (Kant 2009: 13–14). Human beings naturally
have a “propensity to enter into society, … combined with a thoroughgoing
resistance that constantly threatens to break up this society.” The “unsociable”
aspect involves amour propre, including “spiteful competitive vanity” (14).
Kant’s point in the passage, however, is not to decry the threat that unsociable
sociability poses to sociable sociability. It is rather to analyze the former’s his-
torical role in moving humanity from “an arcadian pastoral life of perfect con-
cord, contentment and mutual love” that, however desirable in itself, would
ultimately leave humanity impotent to achieve history’s “cosmopolitan aim.”
Without competitive, rivalrous instincts, though human beings would be “as
good natured as the sheep they ten[d],” human life would have no “greater
worth than that of their domesticated beasts …, [and] humanity would eter-
nally slumber undeveloped” (14).
All these themes are in Rousseau. In the second Discourse, Rousseau says that
it is amour propre that allows “all our faculties [to be] developed,” and “reason
[to] become active” (Rousseau 2019: 170). Rousseau also notes that mutual love
or “pity” characterizes humanity’s natural pre-social state as it does those of

7
For an excellent analysis, see Neuhouser 2008.

https://doi.org/10.1017/9781139025065.010 Published online by Cambridge University Press

 278 rousseau and kant

other animals (152). And Rousseau draws a contrast, as does Kant, between
pity-motivated, loving conduct, on the one hand, and action that realizes what
Rousseau calls “moral freedom” and Kant calls “moral worth,” on the other
(Kant 1996a: 4:398). Finally, Rousseau maintains, as Kant will also, that moral
freedom consists in guiding oneself by a law that one wills for oneself from a
shared perspective as an equal person, which requires equal respect for all.
The crucial difference between Rousseau and Kant resides in the transcen-
dental ambitions of Kant’s critical philosophy. Kant bids to ground morality
in inescapable presuppositions of pure practical reason. Rousseau has no such
ambitions; his aim is to solve the basic problem of political philosophy. His
solution – grounding political legitimacy and obligation in a shared human
capacity for sociable “moral freedom” – points the way toward a similar
grounding for morality of the kind we encountered first in Grotius.

Kant: Groundwork I
Now to Kant. We begin with Kant’s signature work, The Groundwork of the Meta-
physics of Morals, published in 1785. Deontic morality’s conceptual features are
implicated in Kant’s thought right from the outset. After distinguishing between the
“laws of nature” and “laws of freedom” and their respective “sciences” – “physics”
and “ethics” – Kant turns to a brief statement of the need for “a pure moral philoso-
phy, completely cleansed of everything that may be only empirical” (4:388–389).
For, that there must be such a philosophy is clear of itself from the com-
mon idea of duty and of moral laws. Everyone must grant that a law, if it
is to hold morally, that is, as a ground of an obligation, must carry with it
absolute necessity (4: 389).

“[T]he command ‘thou shalt not lie’,” he adds, “does not hold only for human
beings, as if other rational beings did not have to heed it” (4:389).8 This exam-
ple may seem to presuppose Kant’s notorious absolutism about lying, but all
Kant’s readers need grant is that in a case where it is agreed that it would be
wrong for human beings to lie, it would be wrong also for any rational agent
to, absent some distinguishing feature that would affect the wrongness of lying
were that feature present in the human case.
All Kant needs for his illustration is that moral laws purport to hold in gen-
eral for any being who has the capacity to be bound by them – for any rational
agent (assuming for the moment that rational agency is necessary and suf-
ficient for moral agency) – and that what moral laws dictate, they prescribe
for any rational being who happens to be in the relevant situation. It will then

8
Though, again, any such duty or command can apply only to finite rational agents.

https://doi.org/10.1017/9781139025065.010 Published online by Cambridge University Press

 Kant: Groundwork I 279

follow “from the common idea of duty” that deontic moral truths, if they hold
at all, necessarily hold universally for any possible rational agent (4:389).9 And
that will show that the foundations of morality must be a priori. (At this point,
Kant is evidently assuming, like the rationalists, that rational agency is both
necessary and sufficient for moral agency, though he will later revise that posi-
tion, distinguishing between practical and moral laws, with moral laws holding
only for finite rational agents like human beings and practical laws holding for
all possible rational beings.)
The Groundwork’s project, Kant tells us, “is … nothing more than the search
for and establishment of the supreme principle of morality” (4:392). This, of
course, is the Categorical Imperative (CI), which Kant formulates in different
but, he argues, equivalent ways. Kant ends the Preface with an outline of the
Groundwork’s method, which brings out the centrality of deontic morality to
the book’s structure.
Sections I and II are to proceed analytically by drawing out the consequences
of what Kant hopes will seem like uncontroversial truisms about morality
(4:392). At this point, Kant will simply be “explicating the generally received
concept of morality” (4:445). By analyzing the concepts of morality – moral
duty and moral goodness – and relying only on what is truistic concerning
them, Kant believes he can show that the CI in its various formulations follows.
Most importantly, by the end of Section II, Kant takes himself to have shown
that if these truistic assumptions hold, then “[a]utonomy of the will,” the will’s
being a law to itself “independently of any property of the objects of volition,”
follows as a consequence, and vice versa (4:440).
Kant is careful to point out, however, that no such analytical argument can
establish that any of these assumptions and, hence, their consequences, actually
holds. It is consistent with the soundness of his analytical arguments in Sections
I and II that all of the ideas that he has argued stand together also all fall together
and that deontic moral concepts turn out to be “chimerical idea[s] without any
truth” (4:445). This is the functional equivalent in Kant’s moral works of the
challenges to morality posed by Grotius’s Carneades, Hobbes’s “foole,” and
Hume’s “sensible knave,” although it is not put in explicitly egoistic terms.10 No
analytical argument can establish morality’s fundamental normative claims or
its necessary presuppositions. That, Kant says, “requires a possible synthetic use
of pure practical reason,” and the only way of showing that is possible is by “a
critique of [the practical] rational faculty itself” – a critique of practical reason,
which Kant attempts in a first version in Section III of the Groundwork and pur-
sues more systematically in the second Critique (4:445; 5:3–163).

9
Kant’s position seems to be that The Fundamental Law of Pure Practical Reason and the
Categorical Imperative (the moral law) have the very same contents and so call for the
same acts, but their prescriptions create duties or obligations only for finite rational agents.
10
Discussed in Chapters 1, 2, and 6, respectively.

https://doi.org/10.1017/9781139025065.010 Published online by Cambridge University Press

 280 rousseau and kant

We can put Kant’s point and plan in terms of what Allison calls the “rec-
iprocity thesis,” namely, that morality and the CI hold if and only if auton-
omy of the will does (Allison 1990: 201–213). This is what Kant’s arguments in
Groundwork I and II are supposed to establish. But these analytical arguments
are insufficient to show that any of the key deontic moral ideas is not “chi-
merical.” Establishing that requires a critique of practical reason. The critical
strategy Kant pursues in Groundwork III differs substantially from his argu-
ment in the second Critique. In Groundwork III, Kant attempts to show that
autonomy is an inescapable presupposition of deliberative practical rationality
quite generally. From there, he can deduce the equivalent moral concepts and
platitudes as consequences. In presupposing autonomy, deliberating agents are
committed to morality. In the second Critique, however, Kant pursues some-
thing like the reverse strategy. He argues that awareness of being bound by the
moral law is, if not a presupposition of any rational action (we will see why he
cannot think that), nonetheless an undeniable feature of human practical con-
sciousness. “Consciousness of [the] fundamental law” is, Kant says, “a fact of
reason.” Implicit in that consciousness is the realization that we can determine
ourselves by the moral law, that our will has the property of autonomy (1996b:
5:30–31).
Kant’s view that deontic moral concepts can have no place from the perspec-
tive of pure practical reason vastly complicates this second argument, since its
plausibility trades, I shall argue, on accountability-related features of deontic
morality. It is hard to see, therefore, how someone who accepts Kant’s view
of the normative epiphenomenality of deontic morality should be convinced
that the fact that we must assume we are free when we are aware of being mor-
ally bound from the human perspective can show that we really are free, that
autonomy of the will really holds. If Kant’s argument in Groundwork III works,
and autonomy really is an inescapable presupposition of practical reason, this
would solve his problem. But Kant probably ended up thinking it does not,
and, as we shall see, he was ultimately right to think it doesn’t.
We turn now to a more careful laying out of Kant’s analytical arguments, first
in Groundwork I. Section I begins with Kant’s famous claims about the unique
goodness “without limitation” of a “good will” and his related claims about the
moral worth of actions that express goodness of will – as he argues, actions that
do not simply accord with duty but that are also done “from duty.” These con-
clude with a short argument for morality’s “supreme principle,” CI, in its “uni-
versal law” formulation (FUL): “I ought never to act except in such a way that I
could also will that my maxim should become a universal law” (1996a: 4:402).11

11
I take the following abbreviations from Wood 1999: FUL (Formula of Universal Law), FLN
(Formula of the Law of Nature), FH (Formula of Humanity as End-in-Itself), FA (Formula
of Autonomy), FKE (Formula of the Kingdom of Ends).

https://doi.org/10.1017/9781139025065.010 Published online by Cambridge University Press

 Kant: Groundwork I 281

It is worth analyzing Kant’s claims and arguments in some detail in order to

bring out both the ways in which Kant relies on familiar features of deontic
morality to carry his arguments and how his understanding of moral obliga-
tion as a psychological process (“necessitation”) – one through which finite
human agents represent to themselves morality’s distinctive deontic aspects
(oughts, imperatives, and commands) that are normatively epiphenomenal
from the perspective of pure practical reason – is essential to comprehending
otherwise puzzling aspects of his claims in Groundwork I.
We should first get clear about the kind of evaluative claim Kant is making
when he says that only a good will is “good without limitation” (4:393). Several
things make it obvious that the evaluations with which Kant is concerned in
Groundwork I are internal to (deontic) morality, that Kant is saying that the
only thing that is morally good without limitation is a good will.
First, Kant tells us that his arguments in the first two sections aim to
“explicat[e] the generally received concept of morality” (4:445). Second, it is
evident that Kant intends his claims about the unique “moral worth” of actions
done “from duty” to be understood as claims about actions that manifest good
will, and he puts this also by saying that only such actions are “morally good”
(4:390, 397–398). Third, Kant defines moral duty in terms of good will: “the
concept of duty … contains that of a good will though under certain subjective
limitations” (4:397). This definition can seem puzzling, since if a good will is a
will that acts from duty, how can duty itself be understood in terms of the good
will without circularity? The key to unlocking this puzzle, as we shall see, is
Kant’s doctrine of necessitation, according to which duty is the presentational
aspect under which a finite rational agent must represent the practical law in
order to determine themselves by it.
Understood as a claim within morality about moral goodness, Kant’s claim
that the good will is uniquely “good without limitation” is truistic. If morality
essentially concerns self-determined compliance with moral laws, then moral
goodness can only be excellent moral self-governance, and the moral value of
anything else must be reckoned in relation to that. Considered in themselves,
there is nothing morally good about the talents of mind, temperament, and
gifts of fortune with which Kant contrasts good will, however valuable, even
intrinsically, these might be in other ways (4:393).
Kant’s claim that the moral worth of actions depends on their being done
from duty, and not just in accord with it, appears to follow fairly directly from
the unique moral goodness of the good will. We have seen before in Butler and
the British rationalists the claim that genuine moral virtue consists uniquely in
self-consciously moral self-direction, and all these philosophers agree that that
is realized by moral agents exercising the capacity that makes them subject to
(deontic) morality in the first place.
Now ethical philosophers of pretty much any stripe can agree with Kant’s
claim that the prudent shopkeeper of his first example, who charges the same

https://doi.org/10.1017/9781139025065.010 Published online by Cambridge University Press

 282 rousseau and kant

price to everyone, passing up opportunities to take advantage even of “inexpe-

rienced customer[s],” but only because honesty is the most expedient policy,
does not display virtuous or morally worthy conduct (4:397). One does not
have to accept a “modern” conception of deontic morality to agree with that.
It becomes clearest that Kant is relying on a shared deontic moral common
sense in his third example, in which one person helps another not out of self-
interest but from benevolence or sympathy (4:398). Kant actually describes
the example in terms that suggest that the person’s motive is ultimately self-
serving, since he depicts the agent as “tak[ing] delight in the satisfaction of oth-
ers so far as it is their own work” (4:398). His example would be cleaner if the
person were motivated simply by care or benevolent concern focused entirely
on another person and their welfare, being concerned to help them only if it is
good for them. We do better to think about such a case.
Why, we might then ask, is Hutcheson not right that benevolent motivation
is morally good irrespective of whether the agent is moved by duty? Kant says
that the good will is (morally) good in itself, not in virtue either of “what it
effects” or even “its fitness to attain some proposed end, but only because of its
volition,” because of its “maxim” or the “principle” that the agent actually acts
on (4:394, 397–398). But Hutcheson agrees with that. The disposition to aid
people simply because it will benefit them, he thinks, is morally good in itself,
regardless of what benevolent action actually accomplishes. This, again, is a
key difference between him and Hume. What can Kant say to a Hutchesonian
reader of his (revised) third example?
It is instructive to compare Kant’s praise of the good will at the beginning
of Groundwork I to the famous passage about charity or love from Paul’s First
Letter to the Corinthians. About the good will, Kant says that “like a jewel, it
would shine by itself,” even if “by a special disfavor of fortune” it had no power
to bring about good effects (4:394). Paul says something similar about love:
“though I bestow all my goods to feed the poor, … and have not charity [love],
it profiteth me nothing” (1 Corinthians 13:3). Kant, Hutcheson, and Paul can
all agree that the goodness with which they are all concerned is intrinsic to the
agent’s motivation or will.
Their disagreement concerns a fundamental difference in ethical concep-
tions. In the background of Kant’s claims is obviously a deontic moral concep-
tion, which he evidently assumes his readers will share when they fully think
things through. From the perspective of deontic morality, the problem with
benevolent concern or love is that what it moves us to do can conflict with what
we are morally obligated to do. This is a point on which Butler and the British
rationalists insist. It is most obviously true when love is focused on the good of
individuals or specific groups, but it is also true of Hutcheson’s most favored
moral motive, universal or “calm extensive benevolence,” which aims at the
greatest overall or “public good.” Morality may sometimes require agents to

https://doi.org/10.1017/9781139025065.010 Published online by Cambridge University Press

 Kant: Groundwork I 283

forego promoting that also, if, for example, doing so is incompatible with jus-
tice. Indeed, even if act utilitarianism is correct, the concept of moral right is
clearly different from that of optimally beneficent conduct.
The modern deontic conception of morality holds, moreover, that moral
agents are accountable for complying with moral obligations and, therefore,
that they should hold themselves to doing what is right and avoiding wrongdo-
ing. Even if, consequently, moral agents’ best attempts to hold themselves to
deontic moral standards and do what is right were to lead them astray in a par-
ticular case – indeed, even if it were likelier to do so than their being motivated
by benevolence – that would be irrelevant to their moral goodness. According
to deontic morality, the question of an agent’s goodness entirely concerns how
excellently they exercise the capacity that makes them subject to morality and
capable of being morally obligated in the first place.12
To see the point from a different perspective, recall Balguy’s query that were
we “void of natural Compassion, as well as Benevolence; might we not possibly
be induced to attempt the Relief of a Person in Distress, merely from the Rea-
son of the Thing, and the Rectitude of the Action?” (Balguy 1976: 12). If con-
science or rational agency is sufficient to make one an accountable moral agent,
there could be moral agents completely lacking benevolence. If, consequently,
benevolence were itself a source of moral worth, there would be possible moral
agents who completely lack the capacity for a form of moral goodness. From
the perspective of deontic morality that is an absurd consequence.
It should be clear by this point that Kant’s claims in the first half of Ground-
work I borrow plausibility from an assumed deontic conception of morality.
What I want now to show is that implicit also in this beginning part of the
Groundwork is Kant’s view of obligation as “necessitation,” which ultimately
makes deontic moral concepts and standards normatively epiphenomenal
from the perspective of pure practical reason. Obviously, the British rational-
ists shared no part of this view of morality.
The point arises in two things Kant says in Groundwork I about the concept
of duty. I have already mentioned the first, namely, Kant’s remark that the idea
of duty “contains that of a good will though under certain subjective limita-
tions and hindrances” (Kant 1996a: 4:397). This already seems puzzling, since
if a good will is moved by the thought that an act is a duty, it would seem that
the idea of duty would have to be defined independently of the will. A similarly
puzzling passage is what Kant calls “the third proposition,” which occurs after
his claims about the good will and the moral worth of actions from duty, and
which he says follows from them: “duty is the necessity of an action from respect
for law” (4:400). This also is puzzling. We might have thought that being a duty

12
Irwin seems implicitly to be making this point in Irwin 2009: 26–27.

https://doi.org/10.1017/9781139025065.010 Published online by Cambridge University Press

 284 rousseau and kant

was a distinctive normative status – being required or called for by law – being,
that is, necessary in a normative sense. What could Kant mean by the neces-
sity of an action from respect for law? Respect and motivation by respect are
psychological phenomena. What can they have to do with an action’s being
necessary in any normative sense?
The answer is that they do not, on Kant’s view. We shall have to await a full
discussion of the role of respect in Kant’s moral psychology in the second Cri-
tique to see how this is supposed to work in detail. But we can already see that
what Kant is implicitly referring to is what he calls “necessitation,” a process
that takes place in finite rational agents (in the “phenomenal world,” as he
calls in in the first Critique (Kant 1998: A273)). Through it, from the feeling of
respect, they restrain inclinations contrary to practical reason’s dictates and
determine themselves to accord with those dictates by representing them as
deontic laws. Those laws obligate them through what Kant calls a “command
(of reason)” (4:413). Respect is a feeling that is the phenomenal cause (event
cause in the phenomenal world) of the morally worthy actions that on the
noumenal level moral agents cause themselves through pure practical reason.
Respect works by representing practical reason’s dictates in distinctively deon-
tic imperatival terms.
The point here is that Kant takes deontic moral ideas, like duty, to be con-
cepts whose function is to mediate this process. As he puts it in the Metaphysics
of Morals, “[t]he very concept of duty is already the concept of a necessita-
tion (constraint) of free choice through the law” (1996c: 6:379).13 Similarly, in
Groundwork II:
The representation of an objective principle, insofar as it is necessitating
for a will, is called a command (of reason), and the formula of the com-
mand is called an imperative.
All imperatives are expressed by an ought and indicate by this the rela-
tion 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 (1996a: 4:413).

The deontic notions of ought, obligation, duty, demands (commands), and

imperatives are all defined, for Kant, in relation to “necessitation,” thus always
only in relation to finite, imperfect rational agents. Whereas perfect rational

13
“Such a 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 it reluctantly (in the face of opposition from
their inclinations), and it is in this that such constraint properly consists” (6:379).

https://doi.org/10.1017/9781139025065.010 Published online by Cambridge University Press

 Kant: Groundwork I 285

agents deliberate in terms of the (practically) good – what would be choice-

worthy or good to do – the practical thought of finite rational agents must be
mediated by deontic moral notions. The latter have no place in pure practical
rational thought.
For this reason, deontic moral notions end up being normatively epiphe-
nomenal. The deontic claim that I ought or must (am under an imperative)
to do something is the same normative claim as that it would be good or best
that I do it (“they say that to do or to omit something would be good”), but as
addressed to someone who does not automatically act for those reasons because
they can have inclinations to do otherwise. The obligation gives me no further
normative reason to act; all normative reasons are already figured into what
makes the act good or best to do, and no additional reasons flow from duty
or the ought. We shall return to this matter both in our discussion of Kant on
imperatives and the CI and in seeking to understand Kant’s view of the role of
the feeling of respect.
Having seen now how Kant’s claim about the good will and morally worthy
(morally good) action in Groundwork I are most plausible within an assumed
deontic moral framework, and also how what Kant says about the concept of
duty has to be understood in terms of his view that deontic moral claims are
ultimately normatively epiphenomenal, let us turn to the argument Kant there
gives for the universal law formulation of the CI (FUL). The argument, which
takes place in a breathtakingly swift two sentences, can be convincing only on
the same deontic moral assumptions we needed to see how Kant’s claim about
the good will could be defended against Hutchesonian or Pauline objections.
Here is Kant’s argument:
But what kind of law can that be, the representation of which must deter-
mine the will, even without regard for the effect expected from it, in order
for the will to be called good absolutely and without limitation? Since I
have deprived the will of every impulse that could arise for it from obey-
ing some law, nothing is left but the conformity of actions as such with
universal law, which alone is to serve the will as its principle, that is, I ought
never to act except in such a way that I could also will that my maxim should
become a universal law (4:402).

We begin with the idea of a good will and then ask what must characterize
any law by which such a will could be governed. Assume, then, that a good
will is (morally) good in virtue of its principle or maxim. Again, someone who
thinks that love or benevolence is intrinsically morally good can agree with
that. To get the further Kantian claim that moral goodness consists in acting
from duty, we must make the further deontic moral assumption that moral
agents are accountably subject to morality by virtue of rational capacities that
require nothing like love or benevolence, and that moral evaluation concerns
how they exercise those capacities.

https://doi.org/10.1017/9781139025065.010 Published online by Cambridge University Press

 286 rousseau and kant

But is even that sufficient to yield FUL? Is the idea that the good will acts
from and not only in accord with duty sufficient to show that duty requires
acting only on principles (maxims) that one could will to be a universal law?
That depends on how we understand duty and what it means to be moved by it.
Begin with Kant’s idea of practical law or an “objective law of reason” (4:413). If
a law of reason says, “that to do or to omit something would be good,” or even
that it would be the best or most choiceworthy act, it is hard to see how treating
it as a practical law in that sense and being governed by it would commit one
to FUL. To see the problem, imagine someone who holds rational egoism as a
theory of what it is most rationally choiceworthy to do, and who acts in their
own interest only because they think that that is what practical law prescribes.
Holding that will commit them to egoism as a universal law, and to thinking
that everyone should follow it. But would it commit them to willing, or to being
able rationally to will, that everyone follow it?14
To be clear, we are to imagine not just someone who desires most funda-
mentally that their interests be promoted and who happens (idly perhaps)
to believe that promoting their interests themselves is rational. We need to
consider someone who is disposed to promote their interests only when and
because that is what it is best or most choiceworthy for them, or for anyone in
their position, to do, that is, because it is dictated by a practical law in Kant’s
sense. We should imagine someone, then, who is reluctant to promote their
own interests, so that doing so runs against their temperament, and who must
restrain their self-denying inclinations to make themselves promote their own
interests on principle. Someone can evidently fit this profile without willing or
being prepared to will that others act on this principle also. They would have to
believe, again, that others should act on this principle. But that would not itself
give them any reason to will that others do so. Indeed, to the extent that agents’
interests conflict, the principle they would be treating as practical law might
dictate willing that others not follow it, or at least not willing that they do so.
If the concept of duty is identified with practical law, or even with the moral
law as Kant officially understands that (practical law from a finite human per-
spective), it seems entirely possible to act from duty in this sense without being
committed to FUL. But notice now how much more plausible Kant’s argu-
ment for FUL is when we bring in the accountability-related aspects of duty as
a deontic moral concept. If moral duty is the idea, not just of what agents have
most reason to do, or of what it would be best or most choiceworthy for them
to do, but of what they are accountable for doing through exercising the moral
powers that make them thus accountable, then there is an obvious movement
of ideas in the direction of FUL. What we are accountable for is only what can
justifiably be demanded of us. And whatever else a demand involves, it is an

14
On this point, see Aune 1979: 29–30 and Allison 1991.

https://doi.org/10.1017/9781139025065.010 Published online by Cambridge University Press

 groundwork ii: ful 287

expression of will. As Strawson points out, the states of mind through which we
hold one another and ourselves accountable – blame and self-condemnation or
guilt – implicitly make demands (Strawson 1968: 85). If moral duty and obliga-
tion are conceptually tied to accountability, then moral wrong is conceptually
tied to justified demands and blameworthiness (lacking excuse) (Darwall 2006,
2013a, 2013b).
Imagine now that someone holds and is governed by the belief not just that
it is always best for them (as for anyone) to do what is in their interest, but that
acting in their own interest is a moral duty for which they and others are to be
held accountable. Willing that they act in their own interest on these grounds
will effectively be to will that the rational agent they happen to be do so. It will
be part of holding themselves accountable for their conduct as one moral agent
among others. They cannot consistently will on these grounds that they pro-
mote their own interest, therefore, unless they are simultaneously prepared to
will that others do so as well. So understood, FUL seems something like a moral
axiom or principle that is presupposed by the very idea of deontic morality.
It is no surprise, therefore, that principles of reciprocity or the golden rule
that are close cousins of FUL are ubiquitous among deontic moral philoso-
phers, including even Hobbes, but more obviously, British rationalists like
Clarke, Balguy, and Price. When Balguy argues against Hutcheson that through
the rational moral faculty we “might … be induced to attempt the Relief of
a Person in Distress, merely from the Reason of the Thing, and the Rectitude
of the Action,” he adds: “Might not some such Maxim as that of doing as we
would be done unto, offer it self to our Minds, and prevail with us to stretch
out a helping hand upon such an occasion” (Balguy 1976: 12–13). Not only does
reason effectively direct us to what is right, he thinks. It also enables us to hold
ourselves accountable if we do not follow its direction: It “would prompt us to
undertake it, and condemn us if we omitted it” (13).
If this is right, some principle of reciprocity like FUL is itself internal to the
rational moral capacity that makes moral agents subject to deontic morality.
The very idea of deontic morality presupposes that moral agents have what is
necessary and sufficient to be subject to morality in the rational capacity to
hold themselves and others accountable for compliance with deontic moral
norms. This notion underlies both the most plausible version of Kant’s claims
that moral goodness of will and worth of actions involve the motive of duty and
his argument for FUL on their basis.

Groundwork II: FUL

Section II of the Groundwork continues Kant’s analysis of the “generally
received concept of morality” (Kant 1996a: 4:445). Its main function is to elabo-
rate and defend the different formulations Kant gives of the CI, which Kant
argues are all equivalent to each other and to autonomy of the will. Section II

https://doi.org/10.1017/9781139025065.010 Published online by Cambridge University Press

 288 rousseau and kant

thus establishes the “reciprocity thesis”: The will is subject to the moral law,
in each of its formulations, if, and only if, the will is autonomous. It is, Kant
says, a “[t]ransition” to the synthetic “critique of pure practical reason” – which
is necessary to establish that morality is “not a chimerical idea without any
truth” – in Groundwork III (4:445).
The first formulation of the CI is FUL: “[A]ct only in accordance with that
maxim through which you can at the same time will that it become a univer-
sal law” (4:421). As Guyer points out, Kant’s argument for FUL in Section II
reprises his argument in Section I, but this time he argues “from the analysis of
the concept of a rational being” rather than “from the common-sense notions
of good will and duty” (Guyer 2006: 184). In Section II, the argument goes
through a connection between rational agency and action on practical norms
or laws, Kant’s distinction between hypothetical and categorical imperatives,
and an analysis of the concept of a categorical imperative. Kant attempts in Sec-
tion II “to derive [moral laws] from the universal concept of a rational being as
such” (Kant 1996a: 4:412). I will claim, however, that Kant’s argument can plau-
sibly establish its conclusion only if we read into it the same accountability-
related aspects of deontic moral concepts that we saw were necessary to shore
up the parallel argument in Section I.
We can begin with a famous passage:
Everything in nature works in accordance with laws. Only a rational being
has the capacity to act in accordance with the representation of laws, that
is, in accordance with principles, or has a will. Since reason is required for
the derivation of actions from laws, the will is nothing other than practical
reason (4:412).
There is an implied contrast here between two kinds of laws: “laws of nature,”
in accordance with which everything in nature works, and “laws of freedom,”
normative action-guiding principles on which rational agents act. The intuitive
point can be put in terms of a distinction between explanatory and norma-
tive reasons and a distinctive kind of explanation of action in terms of agents’
reasons.
If I pick up a rock from the ground to examine it and then drop it to return
it to its previous place, there will be explanations of both my and the rock’s
behavior. My behavior might be explained in many ways, and some of these
take the same form that reasons for the rock’s behavior do – efficient causal
explanations proceeding from prior states of the world. But there is a distinc-
tive kind of explanation of my conduct that is unavailable in the case of the
rock, namely, in terms of the reasons on or for which I acted – my reasons for
having dropped the rock. Maybe I thought it would be better, for some reason
or other, for the rock to be returned to the earth rather than remaining with me.
Or maybe I did it simply because I wanted to, not in the sense that my mental
state of wanting caused my action (though perhaps that was also true), but in

https://doi.org/10.1017/9781139025065.010 Published online by Cambridge University Press

 groundwork ii: ful 289

the sense that I took my desire to be a reason for me to drop the rock. Whatever
it was, my reason for acting will have been something I took to be a (normative)
reason to drop the rock. Nothing was or could have been the rock’s reason in
this sense for falling. Only rational agents have their reasons (agents’ reasons)
for what they do, where those are considerations they themselves take to be
normative reasons for them to act as they do (Darwall 1983: 30–31).
An agent’s reasons in this sense are near enough to what Kant calls the “sub-
jective principle” or “maxim” of an agent’s will (Kant 1996a: 4:401n). The cru-
cial point is the connection Kant draws between acting “in accordance with
principles” and acting “in accordance with the representation of laws,” where
“laws” are not natural explanatory laws but the normative practical “laws of
freedom” with which Kant contrasts explanatory laws. Maxims or principles
in this sense are not themselves states of an agent’s psychology, motivating
“springs of action” in the early modern sense; they are rather considerations
that can be formulated and acted upon, though of course an agent’s doing so
itself involves their psychology and motivational states.
Kant’s thought is that to act on a principle in this sense is to act on something
one represents as a universal norm that holds, at the most fundamental level,
for any rational agent: a “practical law.” In letting go of the rock for the reason
that I do, I am committed to thinking that any rational agent would have a
similar reason in similar circumstances. If my reason is that I simply want to let
go of the rock, then I must grant that any rational agent would similarly have
such a reason to do what they want. Or if I thought that the reason was that
a disturbed part of nature should be returned to its undisturbed state, then I
should think that all rational agents have reason to act according to this same
principle. Whenever rational agents act, Kant is saying, they are committed to
accepting some universal practical law by their action.
Having made this point, Kant turns to a brief discussion of necessitation in
anticipation of distinguishing between hypothetical and categorical impera-
tives and, ultimately, giving his parallel derivation of FUL from the very idea of
a categorical imperative and the “concept of a rational being as such” (4:412). If
a being is perfectly rational in the sense that “reason infallibly determines the
will,” Kant says, then its actions “that are cognized as objectively necessary,”
that is, as fulfilling practical law, “are also subjectively necessary,” in the sense
that the action happens by what Kant calls an “inner necessity” of the agent’s
perfect rational nature (1996a: 4:412; 1996c: 6:222).15 Perfectly rational beings
invariably do (by a kind of nomological or natural necessity) whatever “reason
independently of inclination cognizes as practically necessary, that is as good”

15
“[A] practical law … 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)”
(1996c: 6:222).

https://doi.org/10.1017/9781139025065.010 Published online by Cambridge University Press

 290 rousseau and kant

(1996a: 4:412). This is an important point: The normative concept of pure prac-
tical reason is that of good or, as Kant also puts it, “practical good” (4:413). An
“objective law of reason say[s] that to do or omit something would be good”
(4:413).
When, however, actions that are “objectively necessary are subjectively
contingent” – that is, when agents have potentially conflicting non-rational
sources of motivation, what Kant calls “inclinations” – then “the determination
of such a will in conformity with objective laws is necessitation” (4:405, 413).
Finite rational agents do not accord with objective practical law through an
“inner necessity” of their rational nature; they must somehow “restrain” incli-
nations that might lead them astray (1996c: 6:222, 6:481). Kant holds that they
can do this only by representing practically necessary (i.e., good) conduct as
commanded by reason and therefore what they ought or must do. And we do
that, he claims, through the deontic, juridical lens provided by the feeling of
respect (about which, more later).
The representation of an objective principle, insofar as it is necessitating
for a will, is called a command (of reason), and the formula of the com-
mand is called an imperative (1996a: 4:413).

“All imperatives are expressed by an ought,” Kant says. They “indicate by this
the relation of an objective law of reason to a will that by its subjective constitu-
tion is not necessarily determined by it (a necessitation)” (4:413).16
Perfect and imperfect rational wills both stand under objective or practical
laws, but only imperfect ones are subject to oughts, imperatives, duties, and
obligations, that is, to what Kant calls morality. Kant’s reason for holding this
is that the deontic moral concepts only have place and function within neces-
sitation. “A perfectly good will would, therefore, equally stand under objective
laws (of the good), but it could not on this account be represented as necessi-
tated to actions in conformity with the law” (4:414). Necessitation can only take
place in finite rational agents. “Hence no imperatives hold for the divine will
and in general for a holy will: the ‘ought’ is out of place here, because volition is
of itself necessary in accord with the law” (4:414).
The imperatival forms that moral laws take are “categorical imperatives.”
These represent actions as objectively necessary (and therefore, rationally com-
manded) in and of themselves, “without reference to another end” (4:414).

16
This does not mean that the will of a perfectly rational agent is determined by a necessita-
tion. To the contrary, as we shall soon see, necessitation can only take place in imperfect,
finite agents. Kant’s point is that the only way a finite agent can purposely act from practi-
cal reason is through a necessitation, but that they do not invariably do so; sometimes they
give in to conflicting inclinations. Thus, 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” (4:413).

https://doi.org/10.1017/9781139025065.010 Published online by Cambridge University Press

 groundwork ii: ful 291

“[H]ypothetical” imperatives, by contrast, represent actions as “good merely as

a means to something else” (4:414).
To be imperatives at all, even hypothetical imperatives must go beyond
claims about what it would be good to do, whether categorically or relative to
some further assumed end. They must be expressible by oughts in the impera-
tival mood. But there are puzzles about how exactly this is supposed to work
within Kant’s picture of practical reason, puzzles that are reflected in recent
debates about the nature of instrumental rationality. This famous passage is
central to these debates: “Whoever wills the end also wills (insofar as reason has
decisive influence on his actions) the indispensably necessary means to it that
are within his power” (4:417). On this basis, Thomas Hill formulates what he
calls “The Hypothetical Imperative,” which stands in relation to specific hypo-
thetical imperatives as the CI stands to specific categorical imperatives: “If a
person wills an end and certain means are necessary to achieve that end and are
within his power, then he ought to will those means” (Hill 1973: 429).
This might be taken as an axiom of instrumental rationality. But if it is, how
is the “ought” claim to be understood? Recent discussions of this issue dis-
tinguish between “wide scope” and “narrow scope” readings. Read with wide
scope, the ought does not occur within the conditional, but governs it. It says
that agents ought to make it the case that: if they will an end and believe a means
indispensably necessary, then they will the means. This they can do equally by
taking the means or by renouncing the end, and the wide scope norm does
not distinguish between these. On a narrow scope reading, “ought” occurs
within the conditional, making it a genuine conditional “ought” that can be
discharged by modus ponens if the conditions of the agent’s having the relevant
end (and belief) are met.
Considered as an axiom of instrumental reasoning, a wide scope reading
would seem to have obvious advantages. Contemporary philosophers who
defend this reading, like Michael Bratman, point out that a narrow scope read-
ing has the unwelcome consequence that an agent can “bootstrap” themselves
into having a reason for promoting any arbitrary end, whatever reasons there
might be not to adopt it, just by adopting it (Bratman 1981). The right way to
think of a norm of instrumental rationality, these philosophers argue, is as a
norm of practical coherence. So understood, the norm says nothing whatso-
ever about what agents have normative reason to do, although it does dictate
the transfer of the support of normative reasons from end to means (Darwall
1983: 14–16, 44–49). Similarly, it transfers lack of support from means back to
ends.
It is very difficult to see how a narrow scope reading can plausibly be con-
strued as a truistic axiom of practical reasoning. But how should we interpret
Kant’s dictum? Hill interprets it with wide scope, which makes it more plausible
as a structuring principle of practical reason. Lending even more plausibility to

https://doi.org/10.1017/9781139025065.010 Published online by Cambridge University Press

 292 rousseau and kant

this reading, Kant says that his claim is an analytic proposition (Kant 1996a:
4:417). How could that possibly be true if it is read with narrow scope? It would
mean that philosophers, like Bratman, who deny the narrow scope reading are
trapped in contradiction or conceptual incoherence themselves.
Despite this, there are strong reasons for thinking that a narrow scope read-
ing is actually truer to Kant’s text (Schroeder 2005b). For one thing, Kant makes
a distinction between different kinds of hypothetical imperatives: “problem-
atic” imperatives that concern possible ends and “assertoric” imperatives con-
cerning ends an agent actually has that would be unmotivated on a wide scope
reading. As interesting as these issues are in their own right, however, I pro-
pose mostly to ignore them. We should note, though, one puzzling thing Kant
says about assertoric hypothetical imperatives in comparison with categorical
imperatives that invokes Suárez’s distinction between counsel and command
or law, a distinction that has loomed so large in our reading of modern moral
philosophy to this point.
Kant includes “imperatives of prudence” among assertoric hypothetical
imperatives, since he thinks that human finite rational beings have their own
happiness as an end “by a natural necessity” (Kant 1996a: 4:415, 4:417). Now
on a narrow scope reading, hypothetical imperatives are genuine conditional
oughts, no different in their logical form from statements like, “If you have
made a promise to do something, you should do it,” except that the condition
concerns an agent’s optional ends. But there is an important sense in which Kant
holds that human agents having their own happiness as end is not optional,
since they have it by “natural necessity.” From the human perspective, there-
fore, it is hard to see what the analytical difference is between an ought that is
conditional on an unavoidably satisfied condition and an unconditional ought.
However, Kant holds that although it is determinately the case that human
beings have their own happiness as an end, this end is itself too indeterminate
and vague to give definite rational guidance. In other words, there may be no
actions that are “indispensably necessary” for this end or, more plausibly, even
if some action or policy were to be indispensably necessary, no human agent
would ever be able to know it. A human agent “is not capable of any principle
by which to determine with complete certainty what would make him truly
happy, because for this omniscience would be required” (4:418).
Kant says on this basis that “imperatives of prudence cannot, to speak pre-
cisely, command at all, that is, present actions objectively as practically neces-
sary; that they are to be taken as counsels (consilia) rather than as commands
(praecepta) of reason” (4:418). This obviously invokes the Suárezian vocabu-
lary, but to make a very different contrast than the one Suárez made and that
influenced early modern philosophers like Grotius and Hobbes. Even if some
means is known to be indispensably necessary for an unavoidable end, that
would show nothing about whether taking it was called for by obligating law
in Suárez’s sense of something one is accountable for complying with, where

https://doi.org/10.1017/9781139025065.010 Published online by Cambridge University Press

 groundwork ii: ful 293

non-compliance is culpable lacking excuse. This is yet another instance where

Kant does not so much account for deontic moral normativity as attempt to
reinterpret it within his own theory.
We turn now to Kant’s derivation of the first formulation of the CI (FUL)
from “the concept of a rational being as such,” which parallels Groundwork
I’s argument for FUL based on the concepts of a good will and morally worthy
action.
[W]hen I think of a categorical imperative I know at once what it contains.
For, since the imperative contains, beyond the law, only the necessity that
the maxim be in conformity with this law … nothing is left with which the
maxim of action is to conform but the universality of a law as such; and this
conformity alone is what the imperative properly represents as necessary.
There is, therefore, only a single categorical imperative and it is this: act
only in accordance with that maxim through which you can at the same time
will that it become a universal law (4:420–421).

We know by an analysis of the concept of a rational agent that rational agents

act on “representation[s] of laws,” where these are practical laws that apply uni-
versally to all rational agents (4:412). The ideas of rational agent and (univer-
sal) practical law are matched concepts; each requires the other. There can be
practical laws only if they can be represented as such and acted on by rational
beings, considered as such. But that can be so, Kant here argues, only if FUL
holds, only if rational agents are, as such, required to act only on maxims that
they can will to become universal law.
This argument has the same logical structure as the one we examined before
in Groundwork I, and it faces the same logical gap. It seems entirely possible
for an agent both to treat a principle as a universal law applying equally to any
rational agent and to be motivated to act on the principle entirely on those
grounds without that committing them in any way to willing or being ratio-
nally able to will that every rational agent do so as well. Here again, they will
be committed to thinking that any agent should act on the principle, but that
is a different matter, which we can show with the same example we employed
before. An agent might have a self-denying temperament and inclinations,
hold a rational egoist theory of practical law, and (albeit reluctantly) guide their
conduct by their (putative) representation of practical law, without that com-
mitting them to willing or being rationally able to will that all agents act on the
same principle. It would, again, commit them to thinking that others should act
on the egoist principle as well, but not to willing that they do so.
With the argument in Groundwork I we saw that this gap can be closed when
the idea of moral duty is interpreted in the distinctively obligating terms that
entail accountability. If a moral agent determines themselves to comply with
moral obligation as part of holding themselves accountable for compliance
then that does commit them to being able to will that others do so as well. But

https://doi.org/10.1017/9781139025065.010 Published online by Cambridge University Press

 294 rousseau and kant

nothing in the bare concept of a rational agent commits us to anything like that.
We will return to these issues in connection with the Critique of Practical Rea-
son, where Kant argues that principles like that of rational egoism are material
rather than formal, and that genuine practical laws must be formal. The point
here is that nothing like this can fall out of any analysis of the “concept of a
rational being as such” (4:412).
FUL is followed directly by FLN: “[A]ct as if the maxim of your action were
to become by your will a universal law of nature” (4:421). Following Rawls and
others, we can treat FUL and FLN together as involving the same CI procedure,
as Rawls calls it (Rawls 2000).17 Roughly, we test whether everyone’s acting on
the agent’s maxim as if by a law of nature is something the agent could ratio-
nally will, consistently with their acting on the maxim themselves.
Since the CI invariably tests agents’ maxims or subjective principles of act-
ing, it cannot be applied until an agent’s maxim exists or is at least assumed
to exist. To apply it in practical reasoning, therefore, an agent must be tenta-
tively decided to do something for some reason, or at least actively considering
whether to act for that reason. The CI is then applied to whatever the agent’s
reason for acting would then be. Whether the agent will – or is actively deciding
to – do something for a particular reason is a matter of psychological fact. If I
am considering doing something for one set of reasons, and cannot rationally
will a world (natural realm) in which as a matter of natural law everyone acts
similarly for the same set of reasons, but there exists a second set of reasons,
which are not my reasons for wanting to perform the relevant action, but which
I could rationally will everyone to act on, that does not help. What matters is
my actual maxim or reasons. Applying FUL in deliberation obviously requires
self-awareness – if not, in many cases, brutal honesty with oneself. What CI
dictates depends on the agent’s actual maxim, and dishonest agents are only
fooling themselves if they think they can game the system and, like tax lawyers,
cook up more fine-grained universalizable maxims.
What Rawls calls the “Four-Step CI-Procedure” then goes roughly as fol-
lows: (a) formulate the agent’s maxim (say, “Whenever in circumstances C,
do A”), (b) formulate a universalized version of the maxim as a candidate for
practical law (say, “Everyone should, when in circumstances C, do A”), (c) for-
mulate a possible world in which it is a psychological law that everyone when
in C does A (on the maxim) and always has, and (d) conjoin that world with
the actual world to determine what a (c) world that is otherwise as close to the
actual world would be like. We then apply FUL by determining whether the (d)
world is something the agent could rationally will, consistent with acting on
their tentative actual maxim as given in (a).

17
Unless context clearly signals otherwise, I will use FUL to refer indeterminately to FUL
and FLN.

https://doi.org/10.1017/9781139025065.010 Published online by Cambridge University Press

 groundwork ii: ful 295

Kant illustrates FUL with four examples, the same four he will use in illus-
trating the Formula of Humanity (FH). We need not consider all four. It will
do to restrict ourselves to the two most commonly discussed. These will be
sufficient to illustrate a basic distinction Kant holds FUL to reveal between
cases where a possible world modeled on the agent’s maxim (a (d) world) can-
not “even be thought without contradiction,” on the one hand, and, on the
other, cases where such a world is a rationally thinkable possibility, but none-
theless not one that can be rationally willed (4:424). This distinction tracks
and explains a related Kantian distinction in normative moral theory between
“strict” or “narrow” duties and “wide” or “meritorious” duties, respectively.
Roughly, strict duties narrowly dictate specific actions, whereas wide duties
require more general ends, giving agents some discretion in how they promote
them. Like most commentators, I will assume that this distinction is roughly
the same as the one Kant marks out by speaking of “perfect” and “imperfect
duties,” as he does more frequently.18 Kant holds that strict duties are violated
by actions the universalization of whose maxims cannot rationally be con-
ceived – that involve a contradiction in thought – and that wide duties are vio-
lated by actions on maxims whose universalization can be rationally thought
but not rationally willed.
Kant’s first two examples are meant to illustrate two kinds of perfect duties,
the first, toward oneself (against suicide) and the second, toward others
(against insincere promises), both of which are characterized by the fact that,
as Kant argues, a world in which agents acts on their conflicting maxims can-
not even be rationally thought. And his second two are intended to exemplify
imperfect duties – one toward oneself (to develop one’s talents) and the other
toward others (to come to their aid) – both of which are characterized by the
fact that though their relevant (d) worlds can be thought, they cannot be willed
rationally.
Consider first the second of Kant’s four examples, which is supposed to
illustrate a strict duty to others. An agent intends to borrow money with no
intention of repaying the loan and plans to disguise this fact to the lender. To
keep it simple, suppose the person does this just because they believe it to be in
their interest to do so, abstracting from any features of urgent need that might
complicate things. Their maxim is something like, “Whenever I am in circum-
stances in which it is in my interest to make an insincere promise, I will do so.”
Without going through all four steps of Rawls’s CI procedure, let’s consider
the possibility of a world in which everyone acts on this maxim, whenever it
applies to their situation and always has. The imagined world is otherwise like
the actual world. Kant argues that such a possible world is not rationally con-
ceivable, since if everyone were to act on the maxim, it would make the promise

18
See, for example, Johnson 2016.

https://doi.org/10.1017/9781139025065.010 Published online by Cambridge University Press

 296 rousseau and kant

and the end one might have in it impossible, as all potential lenders “would
laugh at all such expressions as vain pretenses” (4:422).
We should grant that credit would quickly dry up in a world of insincere
promises and that this is certainly relevant to the moral impermissibility of
self-serving insincere promises. But Kant’s claim that such a world is not
even thinkable is probably too strong, as reflection on our actual world might
quickly show. Credit would no doubt be reduced, but that would not mean that
people could not still sometimes find themselves facing a willing lender foolish
enough to think they could sniff out insincerity. We need not worry about that
problem though, since there are larger ones.
What seems to be doing the work in Kant’s example, is the idea of a practice,
in this case promising, that has certain background conditions, where depar-
tures prohibited by the practice tend to undermine the conditions on which the
practice itself depends.19 In such a case, one cannot imagine everyone making
the practice-prohibited departures, because this removes the condition neces-
sary for the practice, and hence for the practice-prohibited departures, to exist
in the first place. But what moral force does this have? If the practice is, like
promising, one that is valuable or mutually advantageous (or which an agent
is otherwise committed to willing), then the fact that everyone’s acting on a
maxim will undermine the practice seems to count in favor of its being wrong
for one to do so. But what if the practice has no particular value? Or better, what
if the practice is thoroughly evil?
John Rawls describes a hypothetical practice he calls “telishment,” which is
like punishment except that innocent people are capriciously selected for sanc-
tions, say, as part of a reign of terror (Rawls 1955). Suppose that this practice
depends on official underlings who are called upon to administer telishment.
The regime tells them that if they refuse to go along with the practice and telish
innocents, other officials can be found who will. The regime has a name for
disobeying a command to telish: nullishing. To forestall a fussy objection, sup-
pose that it is intrinsic to the idea of nullishing that it can occur only when
there is a going practice of telishment. Only an official who has been com-
manded to telish can nullish. Suppose finally that if a critical mass of officials
were to nullish when commanded to telish, this would undermine the practice
of telishment.
Now since one can nullish only if one has been commanded to telish, if
nullishing tends to undermine the practice of telishing, it also tends to under-
mine the possibility of nullishing. In this respect nullishing is like promise-
breaking. (There are still differences, of course. If someone were commanded
to telish under the circumstances brought about by universal nullishing, they
could still nullish, although one could not break a promise in circumstances

19
Here I follow my discussion in Darwall 1998: 160–161.

https://doi.org/10.1017/9781139025065.010 Published online by Cambridge University Press

 groundwork ii: ful 297

brought about by universal false promising, since the promise could not have
been made. In both cases, however, universal action on the maxim removes
the opportunity to act on the maxim.) But imagine a public-spirited official
who has been commanded to telish and who is considering nullishing in order
to help destroy the practice of telishment. If we cannot rationally conceive a
world in which everyone makes self-serving, false promises, so also would we
be unable rationally to conceive a world in which everyone nullishes in such a
situation. But how can this show that it is morally unacceptable to nullish, espe-
cially if one’s purpose is to undermine the practice of telishing and, even more,
if the practice is thoroughly evil?
Clearly something has gone wrong here. What has gone wrong, I think, is
that Kant has been misled by certain features of his example and generalized
too quickly. He would have done better to rely exclusively on the test of whether
it is possible to will a world in which everyone acts on one’s maxim. This gives
us the “morally expected” answer in both the promising and telishing/nullish-
ing cases. As Kant himself says about a self-serving false promise, if everyone
were to act on the maxim that gives rise to it, “it would make the promise and
the end one might have in it itself impossible” (Kant 1996a: 4:422). That matters
because in making such a promise oneself, one wills this purpose (that others
trust one so that one can benefit from their trust), but in willing everyone act
on the maxim, one wills in contradiction with this purpose. This is not so in the
nullishing case. The official who is determined to nullish in order to help end
telishing does not will anything that is in conflict with what they will in willing
that all act on their maxim. Quite the contrary.
Or just consider a person who, for whatever reason, is opposed to the prac-
tice of promising and who makes false promises not to advantage themselves,
but as a step in bringing down the practice. Suppose they favor everyone’s liv-
ing more spontaneously without having to rely on formal guarantees. It is hard
to see how the fact that everyone’s acting on their maxim would tend to make it
impossible to promise falsely should speak to them as a reason for them not to
do so. “That is exactly why I intend to do it,” they might reasonably say.
In sum, it is hard to see how Kant’s contradiction-in-thought test creates
anything but mischief for his view. It seems better to rely on the contradiction-
in-will test, which would pretty readily rule out self-serving insincere promises
in the kind of cases Kant has in mind, since they would tend to make it much
less likely that the agent would be in a position to make the kind of promise
they are tentatively considering.
Let us move to the fourth example where Kant explicitly applies the test. Here
we are to consider a person who is in a position to help another, but decides
not to do so, let us suppose, simply because it is not in their interest to do so.
We can imagine them to be a rational egoist who will do anything, including
aiding others, if – and only if – it advances their own interests. Their universal-
ized maxim is: “Everyone should help others only if, and to the extent, that it

https://doi.org/10.1017/9781139025065.010 Published online by Cambridge University Press

 298 rousseau and kant

advances their self-interest.” Kant agrees that we can easily imagine an associ-
ated (d) world that is like our world except that everyone acts on this maxim
as though it is a law of nature. The question is whether an agent can rationally
will this world. Kant argues that they cannot, because any such attempt to will
the requisite (d) world while willing also what one wills in acting, that one act
on the maxim oneself, would “conflict with itself” (4:423). The agent would be
willing that their own interests be promoted (in acting on the maxim oneself),
but also willing that they not be promoted, assuming that the (d) world would
not be in their interests, since it would rob themselves “of all hope of [others’]
assistance” (4:423).
This certainly seems a more promising line of moral analysis than the
contradiction-in-thought test, and it seems to be getting at something genu-
inely morally significant. But there still appear to be obvious problems. Sup-
pose someone is in a sufficiently powerful position such that in order to get the
assistance they need they do not have to rely (much anyway) on other people
being prepared to override their self-interested actions. When they need oth-
ers’ help, suppose, they can purchase it. Of course, they may want things from
others that cannot be purchased, but Kant clearly does not intend his analysis
to depend on anything like that. Suppose, then, that when a sufficiently power-
ful agent considers the (d) world associated with everyone’s (including their)
invariably self-serving action and compares it with a world in which everyone
constrains their pursuit of self-interest when they can help others at little cost
to themselves, they vastly prefer the former. What rational incoherence would
there then be in their will?
If FUL would not rule out purely self-interested refusal to aid others, or
perhaps even self-interested harming, that is problematic. It seems obvious
that the basic idea that Kant is trying to get at with the Categorical Imperative
should rule out conduct whose universalization is rationally acceptable only
owing to knowledge of relatively advantageous circumstances. That the formu-
lation does not already suggests that an improved formulation should rule out
such self-serving construals.
There is a further issue. FUL seems addressed to the question of whether it
is morally permissible for someone to perform some action on some specific
maxim or principle. Take the second example, again. Even the contradiction-
in-will test, it seems, can only show that it is wrong to make insincere promises
if one does so in order to advantage oneself. The conflict in the will comes from
willing that one promote one’s own interests by insincerely promising, as willing
that everyone do so is to will a (d) world that is against one’s interest, since prom-
issory credit will be too tight to be able to promise self-advantageously oneself.
Suppose even that the problem we were just considering, that the agent might be
prepared to accept the (d) world on balance because they have a highly advanta-
geous bargaining position, can be solved. The further problem is that even if it

https://doi.org/10.1017/9781139025065.010 Published online by Cambridge University Press

 groundwork ii: fh, fa, and fke 299

can be shown that it would be morally impermissible to make insincere prom-

ises self-servingly, that would not show that making insincere promises would be
wrong if done from other motives. Imagine, again, someone who is opposed to
the practice of promising quite generally, and who promises insincerely to try to
hasten the practice’s demise. Here what the person wills in insincerely promising
themselves is not in conflict with what they would will in willing its universaliza-
tion in the associated (d) world. The two would be entirely in sync.

Groundwork II: FH, FA, and FKE

We might put the problem we just noted by saying that although FUL plausibly
provides a principle of moral consistency, concerning what moral agents can
permissibly do consistently with acting on their intended maxim, it seems pow-
erless to generate claims about the wrongness of acts considered independently
of motive. Or, at least, FUL can generate such results only if the argument can
rely on some facts about ends to which agents are rationally committed indepen-
dently of the universalization test. There needs to be some content in the agent’s
will to generate a conflict with a willed (d) world. In Kant’s fourth example, the
content comes from the agent’s actual maxim, but if we don’t have that, there
seems to be nothing left to generate conflict, unless we can assume some content
or end to which a rational agent is necessarily committed independently of FUL.
Kant seems to have something like this problem in mind when he consid-
ers rational action from the point of view of its end. Kant takes the capacity of
“setting … ends” to be intrinsic to “humanity” (1996c: 6:387). “[S]ubjective”
ends “that a rational being proposes at his discretion as effects of his actions
(material ends) are all only relative” and “can therefore furnish … no practical
laws” (1996a: 4:428). This problem would be solved, Kant says, if “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” (4:428). He
then famously declares:
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 by this or that will at its
discretion (4:428).

Kant follows this with the Formula of Humanity (FH):

So act 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 merely as a means
(4:429).

It is difficult to overstate the philosophical, indeed historical, significance of

this idea (or, more properly, set of ideas). Kant distinguishes between an end in
the sense of a possible “effect” or state of the world that one might desire and

https://doi.org/10.1017/9781139025065.010 Published online by Cambridge University Press

 300 rousseau and kant

therefore aim to bring about, on the one hand, and, on the other, a very differ-
ent way of acting for someone or something’s sake, namely, not to promote
or bring it about, but as a distinctive way of valuing it – as a form of respect.
Respect is like desire in that it can provide content to the will, but it does so in
a fundamentally different way, by regulating one’s conduct toward its object
rather than motivating one to produce or acquire the object.
Kant’s radical claim then is that irrespective of their specific talents and
accomplishments all rational human beings have a common dignity as “per-
sons” that “marks them out as an end it itself” (4:428). Kant credits Rousseau
for this landmark idea:
I am an investigator by inclination. I feel a great thirst for knowledge and
an impatient eagerness to advance [it]. … There was a time when I thought
that all this could constitute the honor of humanity, and I despised the
mob, which knows nothing about it. Rousseau set me straight. This daz-
zling excellence vanishes; I learn to honor men, and would consider myself
much less useful than common laborers if I did not believe that this consid-
eration could give all the others a value, to establish the rights of humanity
(Kant 1904: 20:44).

One can hardly think of a positive normative thesis has been more central to
mainstream modern moral and political thought as it developed after Kant than
the claim that all rational human persons have a common dignity that grounds
equal basic human rights. We saw anticipations of this thought in Pufendorf
and Adam Smith, but Kant’s formulation and defense of it has certainly proven
the most influential.
What is perhaps most remarkable is Kant’s ambition of establishing this
powerful normative thesis as part of a theory of practical reason inescapable
from commitments of any rational agent in deliberation. Korsgaard and Wood
have argued on Kant’s behalf that agents can rationally value the ends they set
themselves only if they value themselves (Korsgaard 1996a: 122; Wood 1999:
129–130). I have criticized these arguments elsewhere (Darwall 2006: 229–235;
see also Cohon 2000 and Bukoski 2018). I agree, as a matter of normative
moral theory, that the most plausible justification for respecting someone’s
will and choices, whether someone else’s or, indeed, one’s own, is respect for
them and their dignity as equal moral persons, as being, in Rawls’s terms, “self-
originating source[s] of valid claims” (Rawls 1980: 546; Darwall 2013b: 93–114).
What I question is how to ground this basic normative thesis in a theory of
practical reason without assuming the philosophical baggage of Kant’s tran-
scendental idealism.20 This issue will return when we consider Kant’s defense
of autonomy of the will, since the defense is ultimately the linchpin in Kant’s

20
In Darwall 2006, I argue that the most philosophically plausible grounding of the dignity of
persons is within the second-person standpoint.

https://doi.org/10.1017/9781139025065.010 Published online by Cambridge University Press

 groundwork ii: fh, fa, and fke 301

case that the whole package of mutually involving deontic moral ideas are not
chimerical. I shall argue, again, that Kant’s attempts to situate morality within
pure practical reason cannot succeed, that crucial moves in his arguments bor-
row plausibility from deontic moral claims that can play no justifying role in
his official framework, and that these are better grounded in a way that retains
their distinctive normativity.
No matter how FH is justified or grounded, however, there are questions
about how it is to be understood, applied, and related to the other formulations
of the CI, especially if we seek interpretations on which these are all equivalent.
If what it is to be a rational agent is to be able to act on a conception of law, that
is, for reasons – importantly, one’s own reasons – then to treat a rational agent
as an end in the sense that Kant has in mind must involve respecting them as an
independent will. As with FUL, we get a more determinate idea of what Kant
thinks this requires in his discussion of his same four examples. Especially illu-
minating, again, are the second and fourth.
The second example, of a self-serving insincere promise, seems tailor made
for Kant’s characterization of “mak[ing] use of another human being merely as
a means” (Kant 1996a: 4:429). Kant helpfully adds: “For, he whom I want to use
for my purposes by such a promise cannot possibly agree to my way of behaving
toward him and so himself contain the end of this action” (4:430). This naturally
suggests that FH requires something like actual or at least hypothetical consent
(Korsgaard 1996b: e.g., 110–114). Consent is a kind of authority – the power to
permit or prohibit. Regulating conduct by it implicitly recognizes a kind of
authority persons have just by having certain rational capacities. It treats their
rational humanity as an end in itself, not as something to be promoted, but as
something to be respected in the recognition sense (Darwall 1977, 2006, 2013d).
About the second example, Kant adds:
This conflict with the principle of other human beings is seen more dis-
tinctly if examples of assaults on the freedom and property of others are
brought forward. For then it is obvious that he who transgresses the rights
of human beings intends to make use of the person of others merely as
means, without taking into consideration that, as rational beings, they are
always to be valued at the same time as end (Kant 1996a: 4:430).

This is an important linking text between the Groundwork and the Doctrine of
Right. Although Kant’s derivation of the Universal Principle of Right (UPR)
in the latter is carried out primarily in terms that are internal to the theory of
right, it is important to bear in mind that the UPR must also be anchored in
Kant’s moral philosophy. Rights require correlative duties, and what underlies
these duties is FH and the equal dignity of persons.21
21
Guyer emphasizes the centrality of freedom to this linking of FH to the Universal Principle
of Right. To comply with FH, he argues, we must “act only on maxims that could be

https://doi.org/10.1017/9781139025065.010 Published online by Cambridge University Press

 302 rousseau and kant

Kant’s analysis of his fourth example proceeds along somewhat different

lines. Even if we must have others’ actual or hypothetical consent to make use of
them and their rational powers, we do not always need their consent to decline
to advance their ends and projects. The duty to come to others’ aid is a “wide,”
imperfect, or “meritorious duty” (4:424, 4:430). Barring emergencies, we do no
wrong in privileging our own ends and projects over others’. The duty to aid
requires, rather, that we make the happiness of others our end in some more
general way. Others’ happiness is “an end that is also a duty” (1996c: 6:382). In
this sense, Kant says, the ends of others “must as far as possible be also my ends,
if that representation is to have its full effect in me” (1996a: 4:430).
In what sense, however, must those with whom we interact be able to “con-
tain the end” of our actions? Suppose, for example, that someone is convicted
of an egregious tax fraud in accordance with due process in, we can add, a per-
fectly just republic. In what sense must the convicted person be able to “contain
the end” of the sentencing judge’s action? Must the sentenced be able to share
the end of their own detention and, perhaps, reform?
It helps here to bring in what Kant counts as “the third practical principle
of the will, as supreme condition of its harmony with universal practical rea-
son” – what Wood calls the Formula of Autonomy (FA): “the idea of the will of
every rational being as a will giving universal law” – and to read FH in light of it.
(4:431; Wood 1999). Ultimately, we will want to have a way of understanding all
the formulations of the CI as equivalent at the most fundamental level.
We give ourselves universal law, and so treat rational nature in ourselves as
an end in itself, when we comply with FUL (as so far interpreted), namely, by
acting only on principles whose universalization we ourselves can will. Simi-
larly, it might be argued, we treat rational nature in others as an end in itself
when we restrict our acts toward them to those on principles whose univer-
salization they can will also. Even if someone convicted of tax fraud may not
share the end of their own punishment, they are far likelier to be able to will the
universalized form of the judge’s maxim. They are unlikely to be anarchists. So
long as they are not, they will be capable of sharing the judge’s end in this law-
giving sense.
This may give us a way of linking FUL and FH through FA. We give our-
selves the law, and treat ourselves as beings who do, Kant must think, when we
regulate our practical reasoning by not acting on maxims that we cannot will
as universal law (thus complying with FUL). So far, however, this only requires
treating others as beings who can give themselves the law by regulating their
conduct by FUL as well.
This would involve seeing ourselves and others as “lawgiver[s]” in the sense
that Kant distinguishes in The Metaphysics of Morals, of being “author[s]” “of

accepted by everyone else as preserving their capacity for free choice as well” (Guyer 2006:
194–195).

https://doi.org/10.1017/9781139025065.010 Published online by Cambridge University Press

 groundwork ii: fh, fa, and fke 303

the obligation in accordance with the law, but not always the author of the law”
(1996c: 6:227). We “author” obligations when we impose the law on ourselves
by regulating our own conduct by FUL through necessitation and respect. So
far, this requires no particular treatment of others, except insofar as that is
required by FUL already. We do not need the idea that rational persons are
ends in themselves, and so must be treated as such, to fully regard them as law-
givers in this “author of obligation” sense. For FA to require FH and not just
FUL, FA must require that rational moral agents be able to give themselves the
law in some further sense.
We noted above that to the extent that FUL only prohibits acting on agents’
actual maxims, it seems powerless to prohibit or require the performance of
certain acts, or even, indeed, the adoption of certain ends, considered inde-
pendently of the agent’s actual maxims. To solve this problem a revision of
FUL would have to permit action when a given maxim or principle applies (and
not just when it is actually acted on) based on whether its associated (d) world
can rationally be willed. It would thus have to seek to rule out or rule in action
through conflicts between willing the associated (d) world and something the
agent can be assumed to will as a rational (moral) person independently of
FUL. It could not rely on conflicts with the agent’s actual maxims.
This, again, seems exactly what the idea of rational nature as an end in itself is
supposed to provide. And this suggests a further sense in which rational beings
must be assumed to be “will[s] giving universal laws” that is additional to the
sense of authoring obligation through the self-application of FUL (1996a: 4:431).
The content of a law that can require more than moral consistency and prohibit
or require certain actions or ends independently of agents’ actual motives must
somehow be grounded in “objective ends” that are provided, Kant holds, by
rational beings as ends in themselves (4:427–428). And this means that rational
beings must be able not just to author the obligation of the law whatever its
content, but that rational nature as an end in itself must somehow ground the
content of moral law.
Practical laws are, by definition, normative for rational beings. And moral
laws are, by definition, requirements and prohibitions for human rational
beings. But what is the content of practical and moral law? Whatever it is, it
must be the same or common for all human rational agents. I can give myself
the moral law in a “content-affecting” sense, therefore, only if any and every
human moral agent can give themselves that same law. One and the same moral
law must have its source in any and every rational agent’s will. I alone can give
myself my obligation to comply with the law (through “necessitation”), but I
cannot by myself be the source of obligating law’s content. That can have its
source in my will only if its source is rational nature generally. And that can be
true only if the candidate (d) world can be rationally willed also by those with
whom I interact. And this gives us something like FH: Act toward others only
on, and according to, principles that they can will to be universal laws also.

https://doi.org/10.1017/9781139025065.010 Published online by Cambridge University Press

 304 rousseau and kant

Failure to appreciate this more robust sense in which moral law must have
its source in rational will is, Kant thinks, the explanation of why “all previous
efforts that have ever been made to discover the principle of morality … had to
fail” (4:432). “[I]t never occurred” to earlier philosophers that a moral agent “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” (4:432). It is hard not to hear in these
remarks an echo of Rousseau’s famous formulation of his aim in The Social
Contract:
To find a form of association that will defend and protect the person and
goods of each associate with the full common force, and by means of which
each, uniting with all, nevertheless obey only himself and remain free as
before (Rousseau 2003: 49–50).

It is not surprising, therefore, that Kant proceeds directly to his idea of a “realm”
or “kingdom” of ends.
For, all rational beings stand under the law that each of them is to treat
himself and all others never merely as means but always at the same time as
ends in themselves. But from this there arises a systematic union of rational
beings through common objective laws, that is, a kingdom, which can be
called a kingdom of ends (Kant 1996a: 4:433).

The moral law is essentially a law for rational human agents in common.
Consequently, the only way it can issue from the will of any particular agent is if
it likewise issues from the will of every agent. And this can be so only if it can be
grounded not in what some particular agent wills from their own perspective,
but in what any and all agents rationally will from a perspective they can share
and occupy in common.
Recall now the problem we faced with Kant’s treatment of the fourth exam-
ple of FUL, where we considered the possibility of an advantageously situated
agent who was prepared to accept others’ acting only in their own interest if the
agent could thereby get the benefit of acting in their own interest themselves.
Clearly, the fact that they themselves are thus advantageously situated has no
bearing on whether what they propose to do or accept can be sanctioned by
practical or moral law. If, consequently, their being prepared to will the rel-
evant (d) world is conditional on this contingent circumstance, which is arbi-
trary with respect to the bindingness of practical law, then they can hardly take
this to show that they are acting in accord with practical law.
To solve this problem within the CI procedure what we would need to know
is not what (d) world the agent can rationally will from their own, perhaps idio-
syncratic, perspective, but what they, and indeed any agent, could rationally
will from a standpoint they can share in common as rational moral agents.
Kant gives the following clue.

https://doi.org/10.1017/9781139025065.010 Published online by Cambridge University Press

 groundwork ii: fh, fa, and fke 305

Now since laws determine ends in terms of their universal validity, if we

abstract from the personal differences of rational beings as well as from all
the content of their private ends we shall be able to think of a whole of all
ends in systematic connection … that is, a kingdom of ends, which is pos-
sible in accordance with the above principles (4:433).

As Rawls points out, this suggests a “limit” on information that should be

available in any perspective from which the question of whether a (d) world
associated with some candidate principle can be rationally willed should be
considered. Rawls’s own device from A Theory of Justice of a veil of ignorance
seems a necessary feature of any adequate CI procedure (Rawls 1971, 2000:
139).22 It should ask not just what (d) worlds the agent, or even the particular
persons they are interacting with, can rationally will, but what (d) worlds a
person can rationally will from behind a veil of ignorance with respect to “per-
sonal differences,” from the perspective, that is, of a rational human person as
such.
A veil of ignorance will solve the idiosyncrasy problem, but, if anything, it
exacerbates the difficulty of getting moral laws with content that require spe-
cific acts (and ends) in specific circumstances. We still need something to give
content to rational willing from behind the veil. We need, in another words,
something to play an equivalent role to that played by “primary goods” (and
what Rawls later called “highest” and “higher order interests”) in A Theory of
Justice and his later writings (Rawls 1971, 1980). In discussing this aspect of
Kant’s views in Lectures on the History of Moral Philosophy, Rawls suggests that
we import into the CI procedure some similar theory of “true human needs”
(Rawls 2000: 232–234). If this seems overly contingent for Kant’s purposes, we
might rather employ an account of interests or needs for rational agents con-
sidered as such, which for the case in which rational agents are human, would
approximate “true human needs.”23
We might then model the perspective of rational willing relevant to common
laws in a kingdom of ends quite closely on Rawls’s hypothetical choice in his

22
Rawls himself argues that such limits on information are already implicit in FUL and cites
Kant’s remark in the Typic in the second Critique: “[A]sk yourself whether, if the action
you propose were to take place by a law of the nature of which you were yourself a part,
you could indeed regard it as possible through your will. Everyone does, in fact, appraise
actions as morally good or evil by this rule. Thus one says: if everyone permitted himself
to deceive when he believed it to be to his advantage … or looked with complete indiffer­
ence on the need of others, and if you belonged to such an order of things, would you be
in it with the assent of your will?” (1996b: 5:69). About this passage, Rawls writes that he
“find[s] it hard to read” it “without feeling that such an idea [limits on information] is
implicit” (Rawls 2000: 176).
23
Though on some conceptions of human needs, these could differ from needs of rational
persons who happen be human.

https://doi.org/10.1017/9781139025065.010 Published online by Cambridge University Press

 306 rousseau and kant

original position (1980). The veil of ignorance would “abstract from the per-
sonal differences of rational beings as well as from all the content of their pri-
vate ends” (Kant 1996a: 4:433) and some theory of “primary goods” for rational
persons considered as such could provide grounds for rationally willing (d)
worlds from that perspective.
Guyer rightly emphasizes the role of freedom in connecting Kant’s moral
and political philosophy; in particular, the connection between FH and UPR:
“Any action is right if it can coexist with everyone’s freedom in accordance
with a universal law” (1996c: 6:230). Thus, Guyer interprets FH as, “act only on
maxims that could be accepted by everyone else as preserving their capacity
for free choice as well” (Guyer 2006: 194). There is no doubt that Kant thinks
that freedom in this sense is an interest, arguably the highest-order interest,
of rational agents, considered as such. So interpreted, FH would be very close
to a Formula of the Kingdom of Ends (FKE) interpreted along the Rawlsian
lines I have just been suggesting: “Act only on (or in accord with) principles that
anyone could, from a perspective reflecting only their knowledge and interests as
rational agents, will all rational agents to act on (for the case of being human).”
We might then regard FKE as the best formulation of the “very same law” that
FUL, FH, and FA attempt to formulate, with increasing success, as they move
in the direction of FKE. Where Rawls speaks of a “Kantian Interpretation” of
his theory of justice as fairness, we might speak of a “Rawlsian Interpretation”
of the CI.

The Dignity of Persons and Autonomy of the Will

“[M]orality,” Kant says, “is the condition under which alone a rational being
can be an end in itself, since only through this is it possible to be a lawgiv-
ing member in the kingdom of ends” (Kant 1996a: 4:435). “Hence morality,
and humanity insofar as it is capable of morality, is that which alone has dig-
nity” (4:435). Famously, Kant distinguishes between a kind of value that can
be “priced” in something like the market sense of exchange for equivalents,
which shows itself through strength of preference or desire, and a different sort
of value, dignity, to which the evaluative response is a fundamentally different
kind of attitude: respect. “[R]espect alone provides a becoming expression” for
the attitude that responds to what has dignity (4:436). Dignity is an “incompa-
rable worth” “raised above all price” (4:436, 4:434).
Kant’s deep philosophical point concerns a distinction between different
kinds of value that is manifested in different evaluative responses: desire and
respect. This difference is reflected also in the different ways these evaluating
states of mind respectively relate to action, to the different kinds of ends associ-
ated with each. The objects of desire are possible states of affairs that actions can
promote or bring about as “effects” (even when the effect is the trivial one of an

https://doi.org/10.1017/9781139025065.010 Published online by Cambridge University Press

 the dignity of persons and autonomy of the will 307

action’s having been performed). Respect shows itself not in motivating action
for the sake of some end in the “effect” sense – a possible state of the world –
but rather through self-regulating deontic moral action, as when one acts “for
the sake of duty” or the moral “law alone,” or “from the idea of the dignity of a
rational being” (Kant 1996b: 5:85, 5:81; 1996a: 4:434).
There is actually some unclarity concerning what Kant thinks the precise
relations are between respect, moral law, morally good action and good will,
the capacity for morally good (deontically self-regulating) conduct, and respect
for persons exactly are. And a complicating factor is that there are also different
kinds of respect and that Kant himself distinguishes between a kind of respect
that shows itself in action (“observantia” or “respect in the practical sense”) and
the feeling of respect (“reverentia”) (1996c: 6:449, 6:402).
One important distinction between kinds of respect is that between “recog-
nition respect” and “appraisal respect” (Darwall 1977). To see the difference,
notice that we can think both that there is a sense in which all persons are enti-
tled to respect and a sense in which how much respect a person deserves or
has earned depends on how they conduct themselves. The former, recognition
respect, consists in how we treat someone or something. We respect persons in
the recognition sense when we constrain our conduct toward them, recogniz-
ing that it would be wrong (and wrong them) not to do so. We thereby regard
and treat them as beings we are required to treat always as ends in themselves
(in the relevant sense) and never as means only. Respect of the latter kind,
appraisal respect, shows itself not directly in deliberate action or treatment, but
in an appraising attitude (moral esteem) that responds to how the objects of
our respect conduct themselves, their actions and character.
Kant’s own distinction between observantia and reverentia does not map
cleanly onto the difference between recognition and appraisal respect. Obser-
vantia, “respect in the practical sense,” seems to be a form of recognition
respect. But reverentia, or the “feeling” of respect, can arise both when we
contemplate morally good actions and agents, and also when we impose the
moral law on ourselves (within morally good action from the agent’s point of
view). These matters will prove crucial to understanding respect’s role in Kant’s
theory of morally good action and the process of “necessitation” in the second
Critique.
On the most common interpretation, Kant holds that persons, that is, beings
with the capacity for moral action, have dignity as ends in themselves and can
never (rightfully) be treated as mere means. But in the Groundwork, Kant also
says that “a morally good disposition,” “virtue,” and the “cast of mind” “that
practices” self-regulation by the moral law have dignity (1996a: 4:435–436).24

24
For some of the complications here, see Dean 2006, Sensen 2011, and Darwall 2013e.

https://doi.org/10.1017/9781139025065.010 Published online by Cambridge University Press

 308 rousseau and kant

If we distinguish two distinct, but related, forms of dignity that can be objects,
respectively, of recognition respect for persons (as ends in themselves) and of
appraisal respect (moral esteem) for action and agents motivated by the former,
then we can sort Kant’s use of “dignity” out along these lines. Just as “respect”
is ambiguous as between recognition and appraisal senses, so also might we
distinguish between associated senses of “dignity.”
Recall Kant’s remark that Rousseau taught him that only by “honor[ing]
men” in general could he adequately appreciate a kind of human value that
could “establish the rights of humanity” (1904: 20:44). The idea that human
rights are grounded in human dignity has proven an historically powerful idea.
The beginning articles of Germany’s Grundgesetz assert it with notable clarity:
(Article 1) Human dignity shall be inviolable. To respect and protect it shall
be the duty of all state authority.
(Article 2) The German people therefore acknowledge inviolable and
inalienable human rights as the basis of every community, of peace and of
justice in the world (Tschentscher 2012).

Respect for rights is not, like moral esteem, something that can be earned. It
can rather be demanded or claimed. This is the sense of “respect” and “dignity”
that Kant must have mind in the following passage.
But a human being regarded as a person, that is, as the subject of a mor-
ally practical reason, is exalted above any price … he possesses a dignity
(absolute inner worth) by which he exacts respect for himself from all other
rational beings in the world. He can measure himself with every other
being of this kind and value himself on a footing of equality with them.
Humanity in his person is the object of the respect which he can demand
from every other human being, but which he must also not forfeit (Kant
1996c: 6:434–435).

The only kind of respect we can “exact” or “demand” is recognition respect, in

this case for our dignity as persons and for the rights this dignity grounds. Any
connection between dignity and rights, therefore, will have to concern respect
and dignity in the recognition rather than in the appraisal or esteem sense.
We have, then, a host of fundamental deontic moral claims that Kant regards
as entailing one another: the unique moral goodness of the good will and self-
governing moral conduct, the various formulations of CI (FUL, FH, FA, and
FKE), and the claim that moral agents have a dignity that is “beyond price.”
Linking these ideas is Kant’s thesis that “the dignity of humanity consists just in
this capacity to give universal law, though with the condition of also being itself
subject to this very lawgiving” (1996a: 4:440).
Kant then introduces his definition of autonomy of the will: “the prop-
erty of the will by which it is a law to itself (independently of any property of
the objects of volition),” which will play the central role in Kant’s critique of

https://doi.org/10.1017/9781139025065.010 Published online by Cambridge University Press

 autonomy in groundwork iii 309

practical reason, first, in Groundwork III, and later in Kant’s second Critique.
So defined, autonomy is, Kant claims, within the circle of mutually entailing
fundamental moral claims just listed. If it can be validated, it assures the pos-
sibility of categorical imperatives grounded in the “capacity to give universal
law, though with the condition of also being itself subject to this very lawgiv-
ing” (4:440). There can be a moral law only if those subject to it are assured the
capacity to govern themselves by it, without having to assume any desire for
any object or “effect,” like benevolence. This is assured if, and only if, the will is
autonomous in Kant’s sense.
We noted above that connections like these can plausibly be drawn if we take
adequate note of deontic morality’s conceptual tie to accountability. It seems to
go with the very idea of a moral law one is accountable for complying with that
an agent can intelligibly be subject to it if, and only if, they have the capacity to
hold themselves accountable for compliance in their own practical reasoning.
This, again, was a central point of Pufendorf’s. I have been arguing that even
though Kant’s arguments in Groundwork I and II borrow plausibility from
morality’s accountability-involving character, his view of the epiphenomenal
nature of deontic concepts and claims from the perspective of pure practical
reason does not permit him to take advantage of these conceptual connections.
Kant seeks to situate autonomy within a critique of practical reason in
Groundwork III as an inescapable presupposition of pure practical reason, irre-
spective of deontic moral ideas. At the end of Section II, Kant says that until
they can be validated by a critique of practical reason, it cannot be ruled out
that all of the central (deontic moral) ideas and claims in play in Sections I and
II, autonomy of the will included, are “chimerical idea[s] without any truth”
(4:445). However, even if Kant’s critical validation of autonomy in Ground-
work III succeeds, I shall argue, his view of necessitation and the epiphenom-
enal character of the deontic may end up showing that even if autonomy of
the will is not chimerical, there is a sense in which deontic moral concepts and
claims are, for pure practical reason at least.

Autonomy in Groundwork III

Kant begins his argument with a “negative” definition of freedom as agent
causation through the agent’s own practical reasoning, “independently of
alien causes determining it” (4:446). Kant will add to this a “positive concept
of freedom,” but first we should have before us the principle that will do the
major work in Kant’s argument: “[E]very being that cannot act otherwise than
under the idea of freedom is just because of that really free in a practical respect”
(4:448). We need not concern ourselves with the relation between freedom “in
a practical respect” and “transcendental freedom” (1996a: 4:448, 1996b: 5:3).
However central that question is to metaphysics, it would seem that if Kant

https://doi.org/10.1017/9781139025065.010 Published online by Cambridge University Press

 310 rousseau and kant

can show that freedom is an inescapable deliberative presupposition of rational

agency, that would be sufficient to establish its bona fides for moral philosophy,
at least, insofar as it attempts to ground normative convictions from a practical
point of view.
In one sense, it should be uncontroversial that a deliberating agent must
presuppose that they can choose to act on the basis of their deliberation, for
what they take be reasons for acting. We need not assume we are infallible, that
everything we take to be reasons really are or that they have the weight we take
them to have. Nor need we assume that we are not subject to biases of different
kinds, though the fact that we are should make us epistemically humble and
cautious and lead us, when necessary, to check our reasoning both intra- and
interpersonally. We just have to assume we are not hopeless and proceed on
the assumption that there are normative reasons for and against alternative
actions, and that we can attempt (and are attempting) to determine what they
are with an eye toward acting on our determination.
Suppose you believe, however, that your deliberative thoughts are being con-
trolled by an evil demon, and that the demon will cause you to do the opposite
of whatever you conclude you should do. Can you now deliberate? Simply hav-
ing that belief about what will happen does not prevent you from turning your
mind to the practical question of what to do and trying to answer it for yourself
(Korsgaard 1996c). In your deliberation, you might simply try to ignore the
belief. Can you deliberate with the belief in mind? Thinking does not so much
as count as deliberation unless it is aimed at making up your mind between
what you take to be practical alternatives of what you take to be normative
reasons. The only way you can take account of the belief, consequently, is if you
can see its practical relevance – as affecting what you have normative reason
to do. That practical issue is simply a different question from the theoretical
one we are supposing you have already made up your mind on, namely, your
belief concerning how your reasoning and conduct will be determined in fact
by the demon. To think about the practical question you will have to ignore, at
some level of reflection anyway, the theoretical question of how your practical
thinking is in fact proceeding. You will reason practically as if you were doing
so “independently of alien causes.”
That’s not quite the same thing as assuming that your practical thinking
actually is independent of causes that are “alien” in the sense of affecting it
in untoward ways. But it may be close enough for Kant’s purposes. It is the
assumption of what Kant terms the “positive concept of freedom” that really
matters for his argument, however, since Kant will be arguing that that is the
concept of autonomy of the will. He will be claiming that since we cannot delib-
erate except under the assumption of autonomy, we are entitled to conclude all
of the deontic moral claims that he claims follow from it. Having established in
Section II that “a free will” in this positive sense “and a will under moral laws

https://doi.org/10.1017/9781139025065.010 Published online by Cambridge University Press

 autonomy in groundwork iii 311

are one in the same,” Kant can conclude the latter from the former inescapable
assumption (Kant 1996a: 4:447).
We should grant that positive freedom of some form is a necessary delibera-
tive assumption. There might be nothing “alien” causing one’s practical think-
ing to go awry even as it misfires persistently; it might be a completely random
process. We should grant, therefore, that deliberation requires the assumption
that it is not, that one’s practical thought can somehow be guided by norma-
tive reasons and whatever norms or practical laws these involve. Just as ordi-
nary theoretical thought about the world must proceed, Kant argues, under the
guise of “laws of nature,” so also must practical deliberative thought be under
the guise of “laws of freedom” or practical laws (4:387). (Kant tends not to use
the currently familiar vocabulary of normative reasons, preferring to speak of
actions that are dictated by practical law as being “good” to do. Imperatives
drawn from practical law, he says, “say that to do or to omit something would
be good” (4:413). However, we can translate between these two vocabularies if
we simply understand normative reasons as facts that tend to make an action
good to do in Kant’s sense.)
Now so far, it is important to see that the relevant presuppositions of nega-
tive and positive freedom are in no way unique to or distinctive about practical
reasoning.25 Whether we deliberate about what to believe or what to do, we
must assume that we can think freely of alien causes and in accordance with
reasons and rational norms. And Kant says as much: “[O]ne cannot possibly
think of a reason that would consciously receive direction from any other quar-
ter with respect to its judgments, since the subject would then attribute the
determination of his judgment not to his reason but to an impulse” (4:448).
However, Kant identifies the requisite positive practical freedom with
autonomy of the will. “[W]hat, then, can freedom of the will be other than
autonomy, that is, the will’s property of being a law to itself? [‘independently
of any property of the objects of volition’]” (4:447, 4:440). But why should we
suppose that? Based on what we have said so far, the only sense in which Kant
has vindicated a presupposition of positive freedom is one that is common to
both practical and to theoretical reason. As we saw earlier, however, autonomy
of the will has no structural analogue in theoretical reason in Kant’s thought.
General logic concerns the pure forms of theoretical thought, but it cannot
generate contentful theoretical thought without bringing in pure concepts of
the understanding (the categories) that have to do with the objects of thought
and their possible properties. For Kant, the idea that theoretical reason or the
understanding could be a law to itself independently of any properties of its
possible objects is a non-starter. Pure theoretical thought involving only “the

25
What follows draws from Darwall 2006: 224–229.

https://doi.org/10.1017/9781139025065.010 Published online by Cambridge University Press

 312 rousseau and kant

form of thinking,” without some content or putative object, is, he says, “empty”
(1998: A54/B78, A51/B75; 1996b: 5:66).
Perhaps, however, there is something about the practical standpoint in par-
ticular that requires a deliberating agent to presuppose autonomy. Consider,
then, how deliberation ordinarily proceeds from the practical perspective. The
primitive deliberative phenomenon is the act of taking some fact about the
world as a reason to do something, a consideration that tends to make some
action a good thing to do. Wondering how to spend a free evening, I scan the
newspaper to find an attractive possibility, say, going to a film. What I take as
reasons for wanting to go, and for going, are aspects of the possible state of see-
ing the film, on the basis of which I might want to see it. It would be enjoyable,
say, or, if I cannot articulate any such reason, there is the fact that, as it might
seem to me, seeing the film would be good, a possible state or outcome that
there is some reason to bring about.26 My reasons for going will derive from
properties of the object of my desire (and so my volition); whether I go will
depend on good-making features, as these seem from my perspective, of the
possible state of the world that would be the outcome of my action. Nothing yet
seems to require the thought that what I take to be normative reasons for acting
must be grounded somehow in the pure form of the will.
But do we yet have genuine agency and a will? For Kant, rational action
involves not just beliefs and desires, but also some norm or principle the agent
accepts and implicitly makes their own in acting on it (1996a: 4: 412). To involve
my will in an intelligent pursuit of an outcome I desire, I must deliberate on the
basis of some rational norm, one I take to apply validly to any possible rational
agent. And I must presuppose that I am bound by such norms as a condition
of the intelligibility of my own deliberation. But does this require us to presup-
pose autonomy? For this assumption to amount to autonomy, I would have to
be required by the logic of my deliberative situation to presuppose that (at least
some) reason-involving norms are valid independently of any properties of the
objects that the norms counsel me to bring about. What in the deliberative con-
text forces us to assume this?
Consider what norm I might accept in deciding to go to the film because I
expect the state of seeing it to be good (for whatever good-making reasons my
consideration of the state presents me with)? The state of my enjoyably view-
ing the film will seem to me to be an intrinsically good thing, a state that, in
Moore’s words, “ought to exist for its own sake” (Moore, 1993: 34). A natural
answer to the question of what norm I might accept is Moore’s answer. I might

26
Of course, these might include facts about myself, my expected mood, how I would expect
to enjoy the film, and so on, that I take as reasons for going, and for wanting to go. But these
are still facts about the objects of my desire most properly understood, not the fact that I
have the desire itself.

https://doi.org/10.1017/9781139025065.010 Published online by Cambridge University Press

 autonomy in groundwork iii 313

accept the act-consequentialist norm of always doing whatever promotes good

or desirable states or outcomes, and I might think, in accepting that, that any
rational agent should do likewise.
Nothing in the deliberative context, I should make clear, forces the accep-
tance of this norm. The point is that nothing seems to preclude it and that
accepting it would seem quite natural from an ordinary practical standpoint.
Just as belief aims at accurate representation of the world as it actually is, so
from the practical standpoint of an agent with desires might action seem to aim
at bringing about intrinsically desirable states or outcomes, as these seem from
the perspective of the agent’s desires. Compare Kant: “[T]heoretical cognition
[is] that through which I cognize what exists, and practical cognition as that
through which I represent what ought to exist” (Kant 1998: A633/B661).
But do we even have an agent yet? Obviously, a being who deliberates about
what to do simply from the perspective of their current desires and beliefs, and
who is incapable of stepping back and critically revising these, is not an agent
in any sense we should be interested in here. A deliberating agent must, as
Korsgaard says, both be, and be able to see themselves as, “something over and
above” their desires who “chooses which desire to act on” (Korsgaard 1996a:
100). But it is quite possible for an agent to do that without assuming autonomy
of the will, although they will have to assume autonomy in the familiar sense of
being able to act on critically revised desires and so to make their actions “their
own.” As we have been thinking of it, the ordinary agent’s desires are identified
with their current evaluations of the possible states of the world that are their
objects. They can certainly step back from these and reevaluate, getting a bet-
ter conception of the features of these states on which they take their value to
supervene and, hence, of desired states’ value (desirability).
Just as a theoretical reasoner can bring experience and reflection to bear on
the dispositions to belief involved in their current experiences – for example,
overriding or defeating any tendency to believe that an apparently bent stick
in water before them really is bent – so also can a practical reasoner analo-
gously critically revise their desires. Plainly, this picture of deliberation is still
much too naïve. But my point in setting it out is not to defend it, but to suggest
that nothing in the presuppositions to which a deliberating agent is committed
seems to rule it out. It follows that Kant’s argument for autonomy in Ground-
work III fails. Naïve practical reasoning is surely intelligible, even if it is mis-
taken.27 If it is to be ruled out, and autonomy of the will ruled in, it would have

27
I have only recently been able to study Tamar Schapiro’s Feeling Like It in which she gives
a compelling systematic account of inclinations in Kant’s sense as incompatible with this
picture, which, as she points out, is that of the “standard” Humean desire/belief and more
rationalist Scanlonian “practical thinking” accounts of motivation. Schapiro’s account fits
perfectly with Kant’s and is intended to show how autonomy is implicated in “the moment

https://doi.org/10.1017/9781139025065.010 Published online by Cambridge University Press

 314 rousseau and kant

to be on systematic metaphysical, metaethical grounds provided by the critical

philosophy and transcendental idealism more generally.
In setting out this flatfooted alternative to autonomy-grounded normative
reasons and practical laws, I have helped myself to various notions – like that
of an intrinsically desirable outcome – that are obviously problematic from
Kant’s point of view. My point has been that nothing in inescapable assump-
tions to which rational deliberation is committed seems to rule such notions
out. It will become clearer in our consideration of the Critique of Practical Rea-
son that Kant takes the sources of motives and normative reasons to include
only (“material”) “inclinations,” on the one hand, and purely “formal” practical
reason, on the other. We shall want to know what grounds this picture, which
rules out the very possibility of motivating states of mind, like benevolence and
love, for example, that apparently respond to sources of value that neither do
nor could derive from autonomy of the will.

The Critique of Practical Reason: Material vs. Formal

Practical Principles
However satisfied Kant was with his vindication of autonomy, and of morality
on its basis, when he published the Groundwork in 1785, he had evidently grown
dissatisfied by the time he published the Critique of Practical Reason (Kritik der
praktischen Vernunft (KpV)) in 1788. Rather than arguing that we can establish
the validity of morality through an independent argument for autonomy as an
inescapable presupposition of practical reason, Kant here argues that although
“freedom is indeed the ratio essendi of the moral law, the moral law is the ratio
cognoscendi of freedom” (Kant 1996b: 5:4n). Through our awareness of moral
obligation, of being subject to moral imperatives and oughts, we see that we
must be capable of regulating our conduct by these. Here, I shall suggest, Kant
implicitly relies on the same accountability-related considerations that under-
lie the British rationalists’ arguments for freedom. If this is so, however, it will
raise a fundamental problem for Kant, since he believed that deontic morality
has no normative significance for pure practical reason. That this is his con-
sidered view is made evident again in the second Critique. Here he lays out
the psychological details of necessitation (obligating self-constraint) in his
account of the moral psychology of respect. This will mean, ultimately, that no
argument for autonomy as an essential aspect of pure practical reason can be
mounted based on what Kant calls the “fact of reason” – the human awareness

of drama” when an agent steps back from the perspective of inclination and deliberates
rationally. Though Schapiro’s account is doubtlessly closer to the human (“finite”) case, it
is still not clear how it must presuppose autonomy in Kant’s sense, viz., that the form of the
will alone is sufficient to provide normative grounds for action (Schapiro 2021).

https://doi.org/10.1017/9781139025065.010 Published online by Cambridge University Press

 the critique of practical reason 315

of being morally obligated – since, on his view, it is not really a fact of pure
practical reason at all (5:31).
Section I of Chapter I (“On the Principles of Pure Practical Reason”) of
KpV begins with definitions on the basis of which Kant proves three “theo-
rems,” ultimately deriving the Critique’s version of the CI (the Groundwork’s
“supreme principle of morality”), which Kant here calls the “Fundamental Law
of Pure Practical Reason” (1996a: 440, 1996b: 5:30). Unlike the Groundwork,
whose first two sections proceed by analyzing the “generally received concept
of morality” (1996a: 4:445), the arguments of KpV are meant not to rely on
(deontic) moral premises at all. KpV’s arguments are to be based on necessary
features of practical reason alone.
Kant begins by defining “[p]ractical principles” as “propositions that con-
tain a general determination of the will, having under it several practical rules”
(1996b: 5:19). Practical principles are “subjective, or maxims, when the condi-
tion is regarded by the subject as holding only for his will” (5:19). And they are
“objective, or practical laws, when the condition is cognized as objective, that is
as holding for the will of every rational being” (5:19).
Already, we face some complexities of interpretation. In our discussion of
the Groundwork, we were understanding maxims (“subjective principles”) in
terms of agents’ actual reasons for acting. Maxims, so interpreted, are neces-
sarily connected to psychological states of an agent in a particular deliberative
context, although it is a distinctive mental state the content of which can be
formulated, from the agent’s point of view, as a (putative) normative principle
of action (in our contemporary parlance, in terms of putative normative rea-
sons or practical oughts that provide the agent’s reason for acting). If we were
to carry that usage through here, the relevant “holding” would be psychological
and not normative. A practical principle would hold, in this sense, if, and only
if, it actually motivates the agent. It is doubtful, however, that this is how Kant
regards subjective principles’ “holding” in KpV, since he speaks of the agent
regarding subjective principles as “holding” (only for their own wills). The rel-
evant “holding” here must be normative, though “regard” obviously refers to a
psychological state.
But who then is “the subject” whose “regard” determines whether a candi-
date principle is subjective or objective? Presumably, the subject to whom the
normative principle is supposed to apply. This is not, however, some particu-
lar actual agent, as in the case of an actually motivated action. It is anyone to
whom the principle applies as formulated. But then why would whether the
agent regards the principle as holding (normatively) for their own will matter
rather than whether a given practical principle validly holds normatively only
for their will?
Similarly, Kant defines principles as “objective” practical principles or “prac-
tical laws, when the condition is cognized as objective, that is, as holding for the

https://doi.org/10.1017/9781139025065.010 Published online by Cambridge University Press

 316 rousseau and kant

will of every rational being” (5:19). “Cognized,” unlike “regarded,” is factive,

so it entails that the principle universally holds normatively, but even so, why
is not this latter sufficient? Why does the principle also have to be cognized as
universally normatively valid?
Nonetheless, Kant’s practice is to speak of practical principles not just as
normative formulae, but as essentially tied to agents in particular deliberative
contexts. This is clear in his reasoning for Theorem I: “All practical principles
that presuppose an object (matter) of the faculty of desire as the determining
ground of the will are, without exception, empirical and can furnish no practi-
cal laws” (5:21). The “matter” of a desire is its object (“an object whose reality
is desired”). Kant then says that “when the desire for this object precedes the
practical rule and is the condition of its becoming a principle, then I say …
that this principle is in that case always empirical” (5:21). This last definition
makes being “empirical” (later, Kant says “material”) a matter of how the prin-
ciple functions in a particular deliberative “case.” When an agent begins with a
desire for some object and then, simply out of that desire, adopts some practical
rule or principle, then the resulting principle is empirical.
In such a particular “case,” desire or inclination is doing all the motivational
work and principles are, in Humean fashion, slaves to it. The “determining
ground of choice” in that case is then the “representation of an object” from
the perspective of the preceding desire “by which the faculty of desire is deter-
mined to realize the object” (5:21). “[I]n such a case the determining ground
of choice must always be empirical,” and as the principle is “based only” on
this empirical condition, it cannot be regarded (or cognized) as a practical law,
since no such empirical condition can be assumed to hold (either normatively
or psychologically) for all rational beings (5:21–22).
It follows from this reasoning that if an agent adopts a practical principle or
rule motivated only by their desires for specific (empirical) objects, in order to
realize their desires’ objects as their action’s effects, then they cannot be cogniz-
ing the principle as a practical law. Suppose the principle is something like, “Do
whatever will maximize the satisfaction of your desires,” in universalized form:
“Agents should maximize the satisfaction of their desires.” If the only thing
leading one to adopt such a principle for one’s own conduct is desire, that can
hardly justify acceptance of the principle as one on which all rational agents
should act.
As important as that is, however, it would not rule out adopting the very
same principle, with very same content, on other grounds, even, indeed,
because the agent took the universalized version of it to be a practical law. A
principle might be empirical or material in the sense that satisfying empirical
desires is part of what the principle recommends, without its being true that in
some particular deliberative “case” the agent in question adopts and acts on it
only because they are motivated to do so by their empirical desires. Versions
of eudaimonism or rational egoism have been advanced as theories of practical

https://doi.org/10.1017/9781139025065.010 Published online by Cambridge University Press

 the critique of practical reason 317

reason since the ancients. If anything, it has been the default view. Imagine
someone who accepts rational egoism and who, like Butler, identifies agents’
good or happiness with desire-satisfaction. Arguably, Kant accepts some such
theory of happiness himself.
Imagine someone now who adopts the principle of maximizing their own
desire satisfaction not because they are motivated to do so by their own desires,
but because they accept that that is what any rational agent should do. Again,
their acceptance of the latter is also not motivated by their own desires. Sup-
pose they independently come to accept desire-satisfaction egoism as a theory
of practical reason. If they do, then though their adopted principle (maxim)
will have the very same content as “empirical” or “material” principles in Kant’s
sense, their principle will not count as “empirical” or “material” in the sense
that Kant seems to be relying on. The agent’s desires for specific objects will
indeed “preced[e] the practical rule,” but the desires will not be “the condition
of its becoming a principle” in sense of the agent’s adopting of the principle’s
being “based only” on such a “subjective condition of receptivity to a pleasure
or displeasure” (5:21).
If “empirical” (or “material”) principles are defined as those that are based
only on empirical subjective conditions in this particular-agent-guiding sense,
then Theorem I follows fairly readily. But so understood, both this definition
and Theorem I concern not principles as individuated by their content but con-
sidered as particular agents’ principles. What Theorem I then says is that if what
motivates an agent’s adopting and following a principle of whatever content
is their empirical desires (what Kant generally calls “inclinations”), then the
principle cannot be functioning as a practical law in the agent’s reasoning. The
agent cannot be treating the principle as an objective principle that applies to
all rational agents.
So far, however, this says nothing about the content of empirical desires. It is
consistent with the definition of empirical principles that the principle “Maxi-
mize desire-satisfaction” is not itself an empirical desire.
Ultimately, Kant wants to derive the Fundamental Law of Pure Practical
Reason, KpV’s equivalent of the CI: “So act that the maxim of your will could
always holds at the same time as a principle in a giving of universal law” (5:30).
Theorem III is a key linking premise: “If a rational being is to think of his
maxims as practical universal laws, he can think of them only as principles
that contain the determining ground of the will not by their matter but only
by their form” (5:27). If Kant can show that practical laws have to be grounded
in the form of the will rather than anything that could independently give it
content, then it would seem that that would establish autonomy of the will
(“the property of the will by which it is a law to itself (independently of any
property of the objects of volition)”) (1996a: 4:440). And all the formulations
of the CI would then follow from that by virtue of the analytical arguments of
Groundwork II.

https://doi.org/10.1017/9781139025065.010 Published online by Cambridge University Press

 318 rousseau and kant

Unlike the Groundwork, however, KpV’s arguments are meant to be free of

any premises about (deontic) morality. Kant’s proofs of Theorems I through
III, and of the Fundamental Law of Pure Practical Reason following from them,
are, like those of Groundwork I and II, analytical, but they attempt to analyze
the concepts of rational agency and norms of practical reason (practical laws).
Even if Kant can show that the Fundamental Law follows from these concepts,
however, that will not dispatch the worry he raises about the possibly “chimeri-
cal” character of morality. It would show that autonomy (and therefore, deon-
tic moral concepts) are not only mutually entailing, but that they also entail, are
entailed by, the concepts of rational agency and practical law. But it would fol-
low from that entailment that deontic moral concepts are not chimerical only if
it could be shown that the concepts of rational agency and practical law are not
chimerical themselves. Such a claim could not be analytic; it would have to be
synthetic a priori. And the only way to establish such a synthetic a priori truth,
Kant thinks, is through a critique of practical reason (of the kind that Kant
unsuccessfully attempted in Groundwork III). In KpV, Kant tries to establish
the claim through his “fact of reason” argument, which we shall consider in the
next section.
The question we face now is how Kant can get from Theorem I, considered
in the particular-agent-guiding sense in which it is most plausible, to Theorem
III, considered in the sense necessary to establish autonomy. Theorem III says,
again, that “[i]f a rational being is to think of his maxims as practical laws, he
can think of them only as principles that contain the determining ground of
the will not by their matter but only by their form” (1996b: 5:27). For Theo-
rem III to entail that the will is a “law to itself” independently of any possible
contents (“objects of volition”) (1996a: 4:440), it must be read in a way that
rules out interpretations consistent with the particular-agent-guiding sense
of Theorem I.
The issue is how to interpret Theorem III’s requirement that to think of max-
ims as practical laws, the agent must see them as “contain[ing] the determining
ground of the will not by their matter” (1996b: 5:27). What seems to follow from
Theorem I as we have been interpreting it is that insofar as an agent is moved
to adopt and act on a maxim solely because of their empirical desires, then they
cannot be viewing their principle as a practical law. The “matter” is the “deter-
mining ground” of their will here, as their will is being formed entirely under
the motivational influence of their empirical desires. As we saw, however, if it
is interpreted in this way, Theorem I is consistent with the matter of an agent’s
desires being the “determining ground” of their will in a different sense that
does not rule out their viewing the principle of their will as a practical law. To
take the most obvious case, again, they might hold it to be a practical law that
agents should maximize the satisfaction of their desires. They might not adopt
this maxim simply owing to the motivational pressures of their desires.

https://doi.org/10.1017/9781139025065.010 Published online by Cambridge University Press

 the critique of practical reason 319

How should Theorem III be interpreted if it is to follow from Theorem I?

(Theorem II, that all material practical principles “come under the general
principle of self-love,” does no independent work in proving Theorem III
(5:22).) The sense in which an agent must think of their maxims as “containing
the determining ground of the will not by their matter” would have to be the
same particular-agent-guiding sense. Interpreted thus far, indeed, Theorem III
just is Theorem I. Theorem I states that agents must not regard their own adop-
tion of and action on principles as being entirely motivationally driven by their
(empirical) desires if they are to regard them as practical laws. But Theorem
III adds a further positive requirement. To view principles as practical laws an
agent must regard them as “containing the determining ground of the will …
only by their form” (5:27).
This, then, is the question we must therefore confront: how to interpret the
phrase “only by their form” in order to make it true both that (i) Theorem III
follows from the most plausible version of Theorem I and that (ii) Theorem III
entails autonomy of the will, that the will can be a “law to itself” independently
of any fact about its possible contents (“any property of the objects of voli-
tion”)? And there seems to be no way to do this. Suppose we interpret “only by
their form” in such a way that the agent we considered – the one who believes
that adopting the principle of maximizing their desire satisfaction is dictated
by a rational norm of action (practical law) – satisfies the “only by their form”
condition. Theorem III will then follow from Theorem I all right, but it will not
entail autonomy of the will. It will not follow from Theorem III that any, much
less every, practical law is grounded only in the form of the will, independently
of any properties of the will’s possible content.
Or take a different kind of case, where an agent regards certain possible
states of the world as intrinsically good or desirable in G. E. Moore’s sense
(Moore 1993). Imagine, for example, that benevolence or love involves regard-
ing the realization of the good or happiness of the beloved as good in itself.
Suppose then that a loving agent adopts the principle of acting to bring about
the beloved person’s good whenever that is possible. Their maxim need not be
“empirical” or “material” in Theorem I’s most plausible sense. Even if, indeed,
they regard their appreciation of the intrinsic value of their beloved’s welfare as
being connected to their love or benevolence epistemically (which we can grant
for these purposes involves an empirical desire), it does not follow that they
would have to see their principle of action as itself grounded in their desire. To
the contrary, they would see their desire as itself a response to their beloved’s
value and the intrinsic desirability of their well-being. They would regard the
beloved’s well-being as providing a(n agent-neutral) normative reason for
them, but also, in principle, for anyone, to promote it (Darwall 2002: 69–72). In
that sense, they would see their principle as a practical law. If, again, this were
to count as their adopting the principle “only by [its] form,” it would obviously

https://doi.org/10.1017/9781139025065.010 Published online by Cambridge University Press

 320 rousseau and kant

not involve their seeing their principle as deriving simply from the pure form
of the will.
There is also the case we considered in our discussion of Groundwork I and
II that revealed a gap in Kant’s arguments for the CI, both from the goodness
of the good will and from the idea of rational action. An agent might adopt
the principle of prudence, of doing whatever is good for them (whether that
consists in desire-satisfaction or not), not out of their desire for their own good
but because they think it is normative for rational agents (or rational human
agents) as such. Again, that adoption would be grounded only in the princi-
ple’s form in the sense that follows plausibly from Theorem I, but the principle
would not necessarily be grounded in the pure form of the will.
Alternatively, we might try to interpret Theorem I in some different way that
does entail an interpretation of Theorem III required for it to entail autonomy
of the will. The problem will now be the reverse, namely, that the requisite defi-
nitions will no longer yield a Theorem I that is plausible and uncontroversial.
The various possibilities we have just been considering will no longer be consis-
tent with Theorem I so interpreted. But that will be because these alternatives to
the idea that normative reasons and rational norms of action can be grounded
only in the pure form of the will have been ruled out by definition in the first
place.
At this point, we might step back from the letter of KpV and reflect on what
we could think of as the spirit of Kant’s critical philosophy in relation to it. The
general thrust of Kant’s Copernican Revolution is to seek the grounds of theo-
retical and practical cognition, not as “dogmatic” philosophers (like the Brit-
ish rationalists) supposed in what is somehow “given” to us, but in structures
a mind must itself impose in order for there to be such a thing as theoretical
or practical cognition at all. So viewed, autonomy is what would have to play
the relevant structuring role on the practical side. As against the intuitionists,
Kant does not grant the validity of practical norms independently of practi-
cal reason’s exercise. Practical reason is not a faculty of intellectual perception
of independent normative truths. Rather normative practical truths are made
true by facts about the exercise of practical reason.
Constructivist or “procedural realist” interpretations of Kant, like those
of Rawls and Korsgaard, emphasize this aspect of Kant’s thought (Korsgaard
1996a, 1996b; Rawls 2000). What Rawls calls the CI procedure can be inter-
preted as a “procedure of construction” that is not simply an epistemic device
for discovering deontic moral truth but a procedure the correct application
of which explains deontic moral truth. Understood at this level, however, the
CI procedure might be seen as a normative deontic moral theory rather than a
metaethical theory that might compete with realist and irrealist theories (or be
a version of either). It could be seen as operating at the same level as contrac-
tualism, as, for example, Derek Parfit understands what he calls the “Kantian
Contractualist Formula” (Scanlon 1998; Parfit 2011).

https://doi.org/10.1017/9781139025065.010 Published online by Cambridge University Press

 the critique of practical reason 321

Korsgaard’s constructivist, “procedural realist” interpretation of Kant aims

to cut more deeply to the metaphysical, metaethical level. It is of the nature
of practical normative truth that it must be able to be grounded in an agent’s
deliberative context, from the practical point of view. Kant “start[s] from the
fact that when we make a choice, we must regard its object as good” (Korsgaard
1996a: 122). We cannot, however, just take the goodness of the objects of our
desires for granted. “Being human we must endorse our impulses before we
can act on them” (122). We must ask whether what we take to be good really is,
and what makes it good. “Rejecting one form of realism,” Korsgaard continues,
Kant “decided that the goodness was not in the objects themselves” (122). We
would not find them good, after all, unless we desired them. “Kant saw that
we take things to be important because they are important to us – and he con-
cluded that we must therefore take ourselves to be important” (122).
This is the Kantian argument for FH that I mentioned in passing above (see
also Wood 1999: 129–130). As I said there, although we can agree, as a matter of
normative moral theory, that the most plausible justification for taking some-
one’s will and choices, either one’s own or someone else’s, as reason giving is
respect for them and their dignity as equal moral persons, there are insuperable
objections to the argument at the level of the metaphysics of value (Darwall
2006: 229–235, 2013b: 93–114). The idea that the value of persons grounds a
constructive procedure might be most plausible within deontic moral theory,
in something like the way the Rawlsian interpretation of CI discussed above
treats it.
Similarly, Korsgaard discusses an “appeal to autonomy” as a source of moral
normativity according to “Kant and Kantian constructivists” like Rawls: “Kan-
tians believe that the source of the normativity of moral claims must be found
in the agent’s own will” (Korsgaard 1996a: 19). This recalls the Groundwork’s
FA: “the idea of the will of very rational being as a will giving universal law”
(Kant 1996a: 4:432). As we discussed before, however, FA functions within
deontic moral theory as a way of linking FUL to FH and FKE. It results from
unpacking the “generally received concept of morality” (4:445). Nothing that
underlies the attractiveness of FA and of autonomy as the source of moral nor-
mativity in that context, however, can support autonomy in a critique of practi-
cal reason of the kind Kant thinks necessary to establish that deontic morality
is not chimerical.
Perhaps we do better, however, to regard Kant not as giving specific argu-
ments for a constructivist metaethics but as putting forward a systematic meta-
ethical alternative to forms of realism. Whether it should be accepted would
ultimately require assessing the overall theoretical costs and benefits as judged
in comparison with other metaethical theories. And to be sure, there is some-
thing undeniably attractive about the thought of being able to derive substantive
normative practical conclusions from procedural constraints of practical rea-
son. There can be widespread agreement that some rational requirements – for

https://doi.org/10.1017/9781139025065.010 Published online by Cambridge University Press

 322 rousseau and kant

example, those of instrumental rationality and similar constraints of rational

coherence that John Broome has studied – are built into the very possibility of
rational agency (Broome 2013). But it can reasonably be doubted whether all
normative reasons for acting or considerations that can make an action good
(even indeed for any rational agent) must be able to be derived from procedural
rationality and its requirements (Scanlon 2014: 90–11; Parfit 2011). To take per-
haps the most powerful counterexample, benevolence and love seem to present
us with access to the agent-neutral value of their objects as reason-giving in
themselves, and it is difficult to see how any such value could be grounded in
the form of the will or requirements of rational agency.
We are left, then, with the problem that we do not yet have any grounding
for autonomy, as Kant seems to believe we need. KpV’s analytical arguments
fail to establish autonomy and the Fundamental Law of Pure Practical Reason.
Even if these analytical arguments were successful, however, that would sim-
ply widen the circle of reciprocal concepts. The (moral) analytical arguments
of Groundwork I and II established the reciprocity thesis – that morality and
autonomy imply one another. But that still left open the question of whether
either concept was more than “chimerical.” Similarly, were KpV’s (rational)
analytical arguments to succeed, they would establish a wider reciprocity
thesis – that morality, autonomy, and practical reason imply one another. But
strictly speaking, it would be compatible with this wider reciprocity thesis that
none of these concepts are instantiated, that all are chimerical.

The Fact of Reason

Groundwork III sought to fill this gap by arguing that autonomy is an inescap-
able presupposition of practical reason. If that strategy had succeeded, it would
have shown, together with the reciprocity thesis, that any agent is rationally
committed in their deliberations to morality, the CI, and the dignity of persons
as well. And that would arguably be all the vindication that deontic morality
would need or could possibly hope for.28

28
But would it not then be true that if Kant could establish a wider reciprocity thesis between
rational agency, practical law, autonomy, and morality, that a strategy similar to that of
Groundwork III, but beginning from the premise that deliberation must assume some
concepts of rational agency and practical law, could then work? The problem would be
exactly how to understand “rational agency” and “practical law” in such a presuppositional
argument. As we noted when we considered Kant’s argument in Groundwork III, even if a
deliberating agent has to assume some concept of rational agency and practical law in order
to deliberate intelligibly at all, it seems not to be true that they need to assume concepts that
are plausibly equivalent to autonomy and those of deontic morality. Even if, for example,
they must presuppose rational requirements of procedural rationality, these will be insuf-
ficient in themselves to generate contentful normative practical reasons (practical-good-
making, or choiceworthy, considerations). And even if a deliberating agent must assume

https://doi.org/10.1017/9781139025065.010 Published online by Cambridge University Press

 the fact of reason 323

In KpV, Kant abandons this strategy, arguing that although “freedom is

indeed the ratio essendi of the moral law, the moral law is the ratio cognoscendi
of freedom” (1996b: 5:4n). What does the work is Kant’s discussion of two vivid
examples, the second of which exemplifies the way in which consciousness of
the moral law is the ratio cognoscendi of freedom, what Kant calls “a fact of rea-
son” (5:30–31). Based on this discussion, Kant asserts: “Consciousness of this
fundamental law may be called a fact of reason because one cannot reason it out
from antecedent data of reason, for example, from consciousness of freedom
(since this is not antecedently given to us)” (5:31).
In the first example, Kant imagines “someone [who] asserts of his lustful
inclination that, when the desired object and the opportunity are present, it is
quite irresistible to him” (5:30). About this case, Kant dryly observes:
[A]sk him whether, if a gallows were erected in front of the house where
he finds this opportunity and he would be hanged on it immediately after
gratifying his lust, he would not then control his inclination. One need not
conjecture very long what he would reply (5:30).

This example illustrates a kind of freedom, but one that is entirely compatible
with actions always being motivated by inclinations whose objects are invari-
ably some “effect” beyond the will’s exercising itself in a law-guided way. It
just shows that if an agent were placed in a situation in which they were more
strongly inclined toward something other than the object of their lust, they
would be moved by that stronger inclination.
The first case serves as a foil to set up Kant’s discussion of the second:
But ask him whether, if his prince demanded, on the pain of the same
immediate execution, that he give false testimony against an honorable
man whom the prince would like to destroy under a plausible pretext, he
would consider it possible to overcome his love of life, however great it
may be (5:30).

By contrast with the first case, Kant hazards no speculation about what the per-
son would do in the second instance, nor does he speculate about what the per-
son themselves thinks they would do. He confidently declares that the person
“must admit without hesitation that it would be possible for him” not to falsely
testify about the “honorable man” (5:30). Kant then concludes: “He judges,
therefore, that he can do something because he is aware that he ought to do it
and cognizes freedom within him, which, without the moral law, would have
remained unknown to him” (5:30).

that contentful normative practical reasons exist (and therefore that there are valid practi-
cal laws), that will not show that all or even any such reasons must be able to grounded in
the form of the will or be accessible by the exercise of practical rationality (understood in
some formal or procedural way) alone.

https://doi.org/10.1017/9781139025065.010 Published online by Cambridge University Press

 324 rousseau and kant

Several aspects of the second example in contrast with the first bear com-
ment. The most obvious aspect of it, first, is Kant’s contrast between what
might be plausibly thought to be the strongest possible inclination and what
the agent must assume to be a source of motivation other than inclina-
tion, a source that is capable of motivating contrary to even the strongest
inclination.
Second, the kind of freedom the agent must unhesitatingly admit is not
just that they are capable of performing an act of some given kind, considered
independently of motive, for example, that they are capable of performing the
physical actions that constitute forbearing false testimony against an honor-
able person. The agent must admit that they can successfully be moved not to
do this because they know it would be wrong. The deep point is that in having
this bit of deontic moral knowledge, agents also thereby implicitly know that
they can comply with their moral obligation simply by exercising the capacity
that makes them subject to the moral law.
To bring out the distinctive importance of this point, suppose that the per-
son in the second example is threatened not with their own death, but with
serious injury to someone they care for. And stipulate that the presence of this
threat, like the death threat, does not change the wrongness of the false testi-
mony. Love and benevolent concern are different from (mere) inclinations in
that they involve seeing their objects and their well-being as having an agent-
neutral value that is independent of any inclination or desire-like mental state
through which the value is felt or appreciated (Darwall 2002: 69–72). But here
again, Kant’s diagnosis applies. In judging that it would be wrong to testify
falsely against an innocent person, even at the cost of injury to a loved one,
one seems committed to judging that one can not do so, indeed, that one can
just because one knows it is wrong. Or suppose the threat is to destroy some-
thing else one thinks to have significant intrinsic value, say a library of precious
books. The same reasoning seems to apply.
Most importantly, third, an essential aspect of Kant’s thought experiment is
that it involves consciousness of the moral law. “He judges, therefore, that he
can do something because he is aware that he ought to do it” (Kant 1996b: 5:30).
“[W]ithout the moral law,” “freedom … would have remained unknown to
him” (5:30). Recall that for Kant, “ought,” imperatives, and obligation are the
mark of deontic morality, which are all distinctive of finite imperfect rational
agential thought and which have no place in the reasoning of perfectly rational
agents.
As we noted in our discussion of the British rationalists, there is a plau-
sible line of thought explaining the fact of reason that has to do with deontic
morality’s conceptual connection to accountability. Since one can intel-
ligibly hold accountable only agents who have the capacity to hold them-
selves accountable in their own practical thought, it seems a conceptually

https://doi.org/10.1017/9781139025065.010 Published online by Cambridge University Press

 necessitation and respect 325

necessary truth that an agent’s actions can be morally wrong, a violation of

their moral obligation, if, and only if, they have the capacity to know that
and be appropriately moved by this knowledge. It must be the case, in other
words, that exercise of the very capacities that make them subject to the
moral law and obligated by it suffices to enable them to comply. I submit that
the plausibility of Kant’s claims about the second example and, therefore,
of the fact of reason, itself depends on deontic morality’s accountability-
involving character.
Again, Kant calls the capacity of rational agents to act on a law that is
grounded only in the form of the will, independently of any facts about the
“objects of volition,” “autonomy” (1996a: 4:440). Autonomy thus involves
the capacity to act for reasons and norms that are not based in the goodness
of any states the will might bring about as effects, but in formal features of
will. This is where the versions of Kant’s second example we considered, in
which agents take themselves to be morally obligated to forgo the promo-
tion of agent-neutral goods, are especially apt. It might well be that many
normative reasons for acting concern such agent-neutral goods, which we
are nonetheless required by deontic morality, on occasion, to forgo. Espe-
cially in such cases, we see that in taking ourselves to be so obligated, we
must “admit without hesitation” that we can do so just because we ought (in
the deontic moral sense). The moral law might indeed be the only substan-
tive norm of practical reason that is grounded in the form of the will, with
autonomy being guaranteed by deontic morality’s conceptual connection to
accountability.

Necessitation and Respect

I have been emphasizing that, for Kant, deontic moral concepts play no role
in pure practical reasoning; they function only in the “necessitating” practical
thought of finite rational agents through which the imperfect agents “restrain”
themselves to do what they already have, independently, good reason to do.
Considerations of moral right and wrong themselves provide no further nor-
mative reasons. Kant develops an empirical account of how necessitation
works through the feeling of respect in Chapter III of KpV, “On the incentives
of pure practical reason,” to which we now turn (1996b: 5:71).
An action can have “moral worth” only if “the moral law determine[s]
the will immediately” (5:71). But how does this happen empirically? How
is “immediately” free action on a law of pure practical reason possible in
the empirical (“phenomenal”) world? For Kant, all phenomenal events
must have phenomenal causes. This is guaranteed transcendentally, since
anything empirical, any possible object of experience, must be conceived
through the category of causation in space and time. This is no less true of

https://doi.org/10.1017/9781139025065.010 Published online by Cambridge University Press

 326 rousseau and kant

actions than it is of anything else that can be experienced in time. It follows

for Kant that however much a moral action must be a product of free prac-
tical reason, its empirical manifestation must nonetheless be empirically
caused.
Kant’s candidate for this cause is the feeling of respect. When we freely
choose to do something because morality requires it, we are moved, as a mat-
ter of empirical psychology, by the feeling of respect. Kant calls this an a priori
feeling, since its existence can be established on a priori grounds (5:79). That
there must be some such feeling follows from the fact that if moral actions,
determined “immediately” by a law of pure practical reason, are to manifest
themselves empirically, they must have empirical causes, and this, Kant thinks,
must be a feeling of some kind.
Kant calls an “incentive” “the subjective determining ground of the will
of a being whose reason does not by its nature necessarily conform with the
objective law” (5:72). Only finite beings with an empirical nature are sub-
ject to incentives; a “divine will” is not (5:72). But although incentives oper-
ate through empirical (felt) causes by definition, the “determining ground” of
the agent’s will, the reason for which the agent acts, is not the empirical feeling
itself. “[R]espect for the law is not the incentive to morality; instead it [respect]
is morality itself subjectively considered as an incentive” (5:76). Moral action
done “from duty,” for the reason that the moral law requires it, expresses itself
in the empirical world as action done from respect (5:81). It is through respect
that the moral law manifests itself empirically and so can give rise “immedi-
ately” to moral action.
It is helpful to begin with the following remarkable passage:

[B]efore a humble common man in whom I perceive uprightness of

character in a higher degree than I am aware of in myself my spirit bows,
whether I want it or whether I do not and hold my head ever so high, that
he may not overlook my superior position. Why is this? His example holds
before me a law that strikes down my self-conceit when I compare it with
my conduct, and I see observance of that law and hence its practicability
proved before me in fact (5:77).

“Humble,” “eminent,” and “superior position” all signal that Kant is here con-
trasting moral respect with the form that respect takes in recognizing social
status. The passage comes directly after Kant’s quotation of a saying of Fon-
tenelle’s: “I bow before an eminent man, but my spirit does not bow” (5:76).
Bowing is an essentially social form of respect. It is a form of deference that
supports or honors, and thereby helps constitute, a (high) social status. It is a
species of recognition respect, honor respect, as I have elsewhere called it, that
is the very medium through which hierarchies of honor and status are socially
constructed (Darwall 1977, 2013d). To be “eminent” or to have “superior

https://doi.org/10.1017/9781139025065.010 Published online by Cambridge University Press

 necessitation and respect 327

position” just is to be recognized as such through others’ (publicly) treating

one as having that status, for example, by bowing.
Fontenelle says that though he bows before an eminent man, his “spirit does
not bow.” This makes an important point. Eminent status need not be jeopar-
dized by insincere expressions of honor. So long as his bow is performatively
received as sincere by the characters in the the social drama in which he takes
part, whether or not it is sincere or is even believed to be, that is enough to
make it a genuine case of honor respect and to confirm or further enhance the
eminent social status of the actual individual to whom the bow is made (Dar-
wall 2013d). Of course, if an honoring gesture wears its insincerity on its sleeve,
that can undermine a bow’s social significance. All of the socially significant
“action” goes on in public.
Kant tells us, first, that unlike Fontenelle who bows before an eminent
man but whose “spirit does not bow,” his spirit does bow before a “humble
common man.” And Kant adds that his spirit bows, “whether I want it or
whether I do not and hold my head ever so high, that he may not overlook my
superior position” (1996b: 5:77, emphasis added). The part I have italicized
exemplifies performative contempt within the social drama in which Kant
partakes (Darwall 2018). Holding one’s head high so that a social inferior
“may not overlook” one’s superior position “puts” an “inferior” individual
“in their place.”
It seems most natural to take the kind of moral respect Kant here describes to
be appraisal respect or moral esteem rather than recognition respect (Darwall
1977). Any reaction to someone’s “uprightness of character” is from an observ-
er’s standpoint in appraising that person’s (motivated) conduct or character; it
is not from a deliberative perspective in deciding how to treat someone as a way
of recognizing them.
The feeling of respect that manifests itself unbidden when we contemplate
good will or “uprightness of character” is moral appraisal respect (1996b: 5:77).
To play the role it is slated for in Kant’s moral psychology, however, respect
must be felt, and be able to motivate action, from the deliberative standpoint.
It must be for the moral law, and for the dignity of persons as mandated by
it, as making claims on our conduct. Only then is respect for “morality itself
subjectively considered as an incentive” (5:76). It must be the form of respect
that is realized in an upright person themselves when they deliberate about
what to do.
What is front and center in Kant’s example, however, is appraisal respect
for an ordinary person of low status. Disarmed by their humility in the face
of Kant’s public pride, Kant involuntarily feels moral esteem for their char-
acter. It is important that this unbidden respect is for uprightness, not for
anything especially ethically distinguished. It is for someone who conscien-
tiously follows the very same deontic standards that obligate us all as one

https://doi.org/10.1017/9781139025065.010 Published online by Cambridge University Press

 328 rousseau and kant

person among others, for an ordinary moral agent who simply does their job
as a moral agent. The man’s example, Kant says, “holds before me a law that
strikes down my self-conceit when I compare it with my conduct, and I see
observance of that law and hence its practicability proved before me in fact”
(5:77).
Kant describes the experience of respect as a kind of humiliation. We do not
simply find ourselves with unbidden moral esteem for a humble upright per-
son, we also feel the force of a normative standard, the moral law, that “strikes
down” or “humiliates” “self-conceit” in us. In feeling unbidden moral appraisal
respect for them, one is brought face to face with a standard of conduct (the
moral law) that applies equally to one as well. Their “example holds before me
a law that strikes down my self-conceit” (5:77). And this, Kant must be think-
ing, leads unavoidably from moral appraisal respect for the upright person to
moral recognition respect of the law itself, to emulating the moral recognition
respect that the upright person themselves feels in being moved to comply with
the moral law “immediately” or “by morality itself subjectively considered as
an incentive” (5:76).
But what is self-conceit? And what is the “humiliation” it receives that Kant
thinks is part of felt respect for the moral law? Kant distinguishes two forms
of “self-regard”: “self-love” and “self-conceit” (5:73). Self-love is a “predomi-
nant benevolence toward oneself,” roughly caring more for oneself and more
about one’s well-being than about others’. Self-conceit, by contrast, consists
in “satisfaction with oneself” or “esteem for oneself” that is independent of and
unconstrained by the moral law.
Self-love need only by “restrict[ed]” “to the condition of agreement with”
the moral law (5:73). Unlike self-conceit, self-love does not purport to provide
any standard for esteem or satisfaction with ourselves at all, much less one that
is unconstrained by the moral law. Concern for oneself or desire for one’s own
well-being, however strong, need not pronounce on estimability.
By self-conceit, however, Kant means forms of self-satisfaction or self-
esteem on grounds that purport to be independent of and unconstrained by
the moral law. At its most extreme, this can include forms of ego-obsession
and involvement that entail seeing oneself as inherently better than or supe-
rior to others. But it also encompasses any standard of esteem that is inde-
pendent of and unconstrained by the common moral standard of the moral
law.
Internalized conceptions of social, racial, and gender privilege are paradigm
examples. In our passage, Kant’s gentleman is able to maintain his high social
standing while feeling unbidden moral esteem for a “humble common man.”
So long as his continuing to enact his social superiority (“holding his head so
high”) is just an act, there may not yet be any self-conceit to be struck down. But
suppose his “character” does not simply so act in the social drama, but that he,

https://doi.org/10.1017/9781139025065.010 Published online by Cambridge University Press

 necessitation and respect 329

the individual playing this role, indulges in feeling justified in so acting. Now
we have self-conceit. He, the actual human individual, is feeling self-satisfied
esteem for himself as an honorable person to whom people should defer inde-
pendently of the moral law.
It is this latter, self-esteem’s form in self-conceit, that cannot coexist simulta-
neously with his moral esteem for the common humble upright person. Esteem
for uprightness just is esteem for being governed “immediately” by the moral
law, a normative standard that is common to all persons, irrespective of their
rank, status, or personae, any such considerations to the contrary notwithstand-
ing. From the perspective of moral appraisal respect for uprightness, we see
that “all claims to esteem for oneself that precede accord with the moral law
are null and quite unwarranted” (5:73). From this standpoint, any standard for
esteem deriving from social status hierarchies are unmasked as mere preten-
sion. And even standards of genuine human excellence are seen to be limited
and regulated by the moral law.
Thus is self-conceit struck down or humiliated. Like humiliations of more
familiar kinds, a “superior” person feeling moral esteem for a “lower” upright
person cannot escape a felt lowering that implicitly recognizes that any stan-
dard that might rationalize their superiority is bogus (self-conceit’s “illusion”
(5:76)), since they cannot possibly be more estimable than an upright person
who is governed by a law that binds both in common as persons (5:77). And
this humiliation has a painful edge that Kant holds to be essential to respect
(5:73).
In feeling respect for a common upright person’s good will, one esteems
them for their exercise of a capacity that is shared between respecter and
respected. The upright person provides an “example” of “immediately” acting
on the moral law (“from duty”), “and hence its practicability [is] proved before
me in fact” (5:71, 5:77, 5:81). “Practicability” here cannot just mean that every-
one can act in accordance with the moral law, that it never demands anything it
is impossible to do. We hardly need the example of an upright person to prove
that it is possible to keep our promises, tell the truth, and act in conformity with
the moral law (as Kant says, “legally,” in our external conduct (5:151)). What the
common upright person shows by example is that, just like the respected, the
respecter, whoever they may be, can also act “immediately” on the law, embody-
ing its “spirit” and not just its “letter” (5:72). It demonstrates the possibility of
acting not just in accord with duty, but from duty in a way that gives action
“genuine moral worth” (1996a: 4:398).
This brings us to respect’s relation to necessitation and, through it, to the
deontic moral representations – duty, oughts, and imperatives – that mediate
respect for the moral law. Kant lays this out in the following passage.
The consciousness of a free submission of the will to the law, yet as com-
bined with an unavoidable constraint put on all inclinations though

https://doi.org/10.1017/9781139025065.010 Published online by Cambridge University Press

 330 rousseau and kant

only by one’s own reason, is respect for the law. … An action that is
objectively practical in accordance with this law, with the exclusion of
every determining ground of inclination, is called duty, which, because
of that exclusion, contains in its concept practical necessitation, that is,
determination to actions however reluctantly they may be done (1996b:
5:80).

From the beginning of the Groundwork, Kant observes a distinction between

practical law, which applies to all rational beings, perfect and imperfect alike,
and the moral law and its associated duties, obligations, oughts, and impera-
tives, that apply only to finite, imperfect rational agents who, in addition to
having practical reason, are subject also to inclinations. The latter must restrain
potentially conflicting inclinations through their reason, and the process
through which they do it is necessitation, mediated by the feeling of respect,
which itself involves deontic representations of what one must do – what, Kant
says, reason commands one to do. “The representation of an objective princi-
ple, insofar as it is necessitating for a will, is called a command (of reason), and
the formula of the command is called an imperative” (1996a: 4:413). “[D]uty
is,” Kant says, “the necessity of an action from respect for law” (4:400).
The moral law is, in other words, for the will of a perfect being a law of holi-
ness, but for the will of every finite rational being a law of duty, of moral
necessitation and of the determination of his actions through respect for
this law (1996b: 5:83).

This means, again, that deontic notions of duty, wrong, requirement, prohibi-
tion, and so on, are no part of pure practical reason per se. They enter from
the perspective of finite rational agents in whom pure reason can be practical
only through necessitating representations of the practical law as something
they must constrain themselves to comply with through respect for reason’s
commands.

Kant on the Right and the Good

Recall Sidgwick’s claim that whereas the ancients thought of virtue and choice-
worthy conduct as different species of the good, the modern period has been
marked by philosophers who recognize duty as an independent ethical concept
and who believe that moral agents have a “regulative or governing faculty,”
“conscience,” that responds to it (Sidgwick 1964: 198). These moderns think,
that is, that conscience is a part of practical reason in the sense that deontic
morality is an independent source of normative reasons.
It follows from our analysis in the last section, however, that if conscience
operates via moral necessitation, it cannot be part of pure practical reason, as
Kant understands it, and therefore that Kant actually rejects this “modern”

https://doi.org/10.1017/9781139025065.010 Published online by Cambridge University Press

 kant on the right and the good 331

view. Perfectly rational beings have neither duties nor conscience, since the
“moral necessitation” that is essential to these concepts is foreign to their
nature. Deontic moral concepts like duty arise only from the perspective of an
imperfect rational being who must conceive of practical laws as authoritative
commands addressed to them and which they must hold themselves account-
able for complying with in their conscience. Moral duty is impotent, for Kant,
to provide any normative reason for acting itself. All normative reasons are
registered already in the practical law, which is not itself deontic. If reason
could create further reasons by commanding what there is most reason to do,
it would be capable of indefinitely magnifying even small difference in the rela-
tive weight of reasons by infinite reiteration.29
What Kant calls “the wonderful capacity within us which we call conscience”
arises only from the perspective of a finite rational agent requiring necessita-
tion (1996b: 5:98). In The Metaphysics of Morals, Kant describes the process
through which human beings can hold themselves accountable in their own
conscience. “A human being who accuses himself in conscience must think of a
dual personality in himself, a doubled self” (1996c: 6:438n). We erect “an ideal
person (the authorized judge of conscience)” within ourselves who occupies
the perspective of pure practical reason (6:439). This personification of reason
“impose[s] all obligation,” and “must be, or be thought as, a person in relation to
whom all duties whatsoever are to be regarded as also his commands” (6:439).
We thus see ourselves as subject to the authoritative commands of practical
reason and hold ourselves accountable for complying with them.
The right in Sidgwick’s deontic sense is thus not a concept of pure practical
reason; it enters neither into practical laws nor into the practical reasoning of
a perfectly rational being. “The only objects of a practical reason are therefore
those of the good and the evil,” Kant says, where these are respectively under-
stood as the “necessary object[s] of the faculty of desire … [and] aversion … in
accordance with a principle of reason,” respectively (1996b: 5:58).
Rawls is famous for the claim that “the priority of right” over the good “is a
central feature of Kant’s ethics,” as it is of Rawls’s own conception of justice as
fairness (Rawls 1971: 31). In support, Rawls cites a section of KpV (Rawls 2000:
230–232).30 The best passage for Rawls’s purposes is the following:
[T]he concept 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 … after it and by means of it (Kant 1996b: 5:63).

29
Thus suppose that there is slightly more reason to do A than B. Reason might then com-
mand doing A, creating even more reason to do A. But because there is now more reason,
reason might command doing A because of this greater reason, creating yet more reason to
do A. And so on.
30
Rawls refers to 5:62–65.

https://doi.org/10.1017/9781139025065.010 Published online by Cambridge University Press

 332 rousseau and kant

Kant certainly does hold that the concepts of good (das Gute) and evil (das
Böse) must be distinguished from those of well-being (das Wohl) and ill-being
(das Übel), respectively, and that the “objects of a practical reason” cannot be
derived from these latter empirical concepts (5:58–60). Since Sidgwick and
Rawls understand the good as well-being (das Wohl), it follows that Kant does
hold that “the objects of a practical reason” cannot be derived from the good as
Sidgwick and Rawls understand it.
When, however, Kant says that “the concept of good and evil must not be
determined before the moral law,” he means nothing beyond the KpV’s signa-
ture thesis that the object of pure practical reason must be conceived formally
through freedom, and, therefore, in accordance with the Fundamental Law
of Pure Practical Reason (5:63). For Kant, this is a claim about the practical
good, not about deontic morality and right as he understands it. He does not
mean that the concept of good (das Gute) must be determined subsequent to
the moral law, conceived in the distinctively deontic terms through which it
is given to imperfect rational beings. It remains true for Kant that “[t]he only
objects” of pure practical reason are “those of the good and the evil” (5:58).31
“Imperatives,” for Kant, “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 rep-
resented to it that it would be good to do that thing” (1996a: 4:413).
The question before any rational agent is what it would be good or best to do
in the sense of what there is most reason to do. When Kant says that “the con-
cept of good is … derived from an antecedent practical law,” what he means is
that the concept of good that is the object of pure practical reason is simply that
of “practical good” or the choiceworthy: what it would be best to do, not well-
being or what would be good to happen (either from the agent’s point of view
or from the “point of view of the universe” (agent-neutral value)) (Kant 1996b:
5:58, 5:114; Sidgwick 1967: 381). “Good or evil always signifies a reference to the
will insofar as it is determined by the law of reason” (1996b: 5:60).
If “right” is taken in its distinctive deontic sense, therefore, Kant does not
hold that the right is prior to the good as he, Kant, understands the good (das
Gute). The question of pure practical reason is what is good or best to do. That
must be determined independently of any issue of moral duty or obligation. The
latter are simply the form that reason’s deliverances of what it is good or best
to do take from the perspective of a will that can have motives to act otherwise.
Rawls used the slogan “the right is prior to the good” to contrast his and
Kant’s view with teleological or consequentialist theories in which “the good

31
Cf. “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 con-
formity with law since of itself, by its subjective constitution, it can be determined only
through the representation of the good” (Kant 1996a: 4:414). For an insightful analysis of
this passage and its significance, see Kosch, 2006: 47–50.

https://doi.org/10.1017/9781139025065.010 Published online by Cambridge University Press

 kant on the right and the good 333

is defined independently of the right” and the right is determined by the good,
usually as what maximizes it, either directly or indirectly (Rawls 1971: 25). Con-
sequentialists frequently identify the good with well-being (das Wohl), and
always with what is good to happen (with valuable states of affairs). Rawls is
right to insist that Kant rejects any such view. Kant’s point in the second Cri-
tique is that what it is good to do can only be determined by the free use of prac-
tical reason, hence not (simply anyway) by what would promote well-being
or would be good to happen. This is identical with the Groundwork’s thesis of
“autonomy of the will” (1996a: 4:440).
Nevertheless, if we understand the good (das Gute), as Kant does himself, to
fundamentally concern “the will insofar as it is determined by the law of rea-
son,” then Kant does not hold that the right, conceived deontically, can provide
a source of reasons that do not derive from the good (1996b: 5:60). Neither is it
the case that considerations of moral duty can figure as good-making features
of action, as providing reasons that bear on what it is good or best to do. The
common thought that we might have had sufficient reason to do something if
it had not been wrong or violated an obligation cannot get a foothold. Being
moral obligatory or a duty is simply the shadow that good or reason-supported
actions cast for imperfect rational agents.
We are left, finally, with the question of Kant’s success in vindicating “the
concept of freedom,” which “constitutes the keystone of the whole structure of
a system of pure reason” (5:3–4). Departing from Groundwork III’s strategy of
arguing that autonomy is an inescapable presupposition of practical reason,
Kant argues in KpV that we can be assured of our freedom through our aware-
ness of being bound by the moral law (the “fact of reason”) (5:30–31). I have
argued, however, that the plausibility of Kant’s analysis of his key example of
the “fact of reason” relies on the same conceptual connection between deontic
morality and accountability that underlay the British rationalists’ arguments
for free moral agency. It turns out, however, that Kant rejects Butler’s and the
British rationalists’ view that conscience is a form of practical reason. So he
cannot accept that when conscientious agents act from duty – having as their
reason that their action is morally required and taking it that that (deontic
moral) fact is itself conclusive normative reason for them so to act – that they
are right. On Kant’s considered view, deontic moral facts are no part of pure
practical reason and can provide no normative reasons for acting at all. It fol-
lows that even if ordinary moral agents necessarily take the fact that they are
morally obligated to act as guaranteeing that they have (as accountable agents)
the capacity to comply for that very reason, Kant himself is committed, as a
theorist, to thinking that they are acting on a kind of illusion. Any normative
reason to act is figured already into what would have made their action a good
thing to do independently of its being their moral duty.
I should emphasize, however, that in stressing Kant’s treatment of the moral
deontic as normatively epiphenomenal, I am not questioning his commitment

https://doi.org/10.1017/9781139025065.010 Published online by Cambridge University Press

 334 rousseau and kant

to the claim that the content of the moral law – the CI in its various formula-
tions – is grounded in pure practical reason. Indeed, that is part of my point. All
the normative work, for Kant, is done by practical law: what practical reason
establishes as good or best for agents to do. That an action is a moral require-
ment or duty normatively adds nothing. Neither am I questioning that what
practical reason dictates, for Kant, is respect and support for the freedom of
rational agents. Above we noted with approval Guyer’s interpretation of FH as
“act only on maxims that could be accepted by everyone else as preserving their
capacity for free choice as well” (Guyer 2006: 194). It is undeniably Kant’s view
that this formulation of the CI, like FUL and FA, is a dictate of pure practical
reason. I have, to be sure, disputed Kant’s arguments for these claims in the
Groundwork and in KpV. But that is unquestionably his view. I have suggested
that the most promising defense of it lies in systematic considerations of the
critical philosophy overall and what Rawls calls Kant’s “moral constructivism”
(Rawls 2000: 237). Even here, however, constructivism may prove to be a more
promising account of the metaethics of morality than of practical reason more
generally.

Freedom, Constraint, and Right in The Metaphysics of Morals

Necessitation is the process through which finite agents constrain their incli-
nations in order to comply with the law of their own will or practical reason,
thereby realizing the freedom Kant calls “autonomy.” This is, we might say,
the “internal” account of the relation between freedom and constraint for
Kant. Autonomy is realized in finite agents by “an internal but intellectual con-
straint,” which we impose on ourselves through the feeling of respect (Kant
1996b: 5:32). When inclinations tempt us away from what it would be best to
do, they require “a resistance of practical reason” (5:32).
Kant also has an “external” account of the relation between freedom and
constraint, one that parallels his internal account of moral, “internal lawgiv-
ing,” and opens up an entirely new philosophical area, which Kant pursues in
the Doctrine of Right (included in his Metaphysics of Morals (originally pub-
lished in 1797)).32 Just as morality, for Kant, is fundamentally concerned with
internal lawgiving, constraint, and freedom (autonomy), so, in parallel, is his
philosophy of right, which examines “the external … relation of one person
to another,” concerned with “external lawgiving,” constraint, and freedom
(1996c: 6:229–230).
Hegel famously remarked in his Elements of the Philosophy of Right, published
almost twenty-five years later in 1821, that “the right of subjective freedom is the
32
We should note that Johann Gottlieb Fichte’s Foundations of Natural Right, which
observes a similar distinction between ethics and theory of right, was published just before,
in 1796–1797 (Fichte 2000).

https://doi.org/10.1017/9781139025065.010 Published online by Cambridge University Press

 freedom, constraint, right in the metaphysics of morals 335

pivotal and focal point in the difference between antiquity and the modern age”
(Hegel 1991: 151). Kant’s Doctrine of Right is in this sense a thoroughly modern
text, providing, along with Fichte’s Foundations of Natural Right (1796–1797),
the model for Hegel’s philosophy of right and establishing freedom as a, if not
the, central concept of political and legal philosophy ever since.
The Metaphysics of Morals begins with a discussion of “preliminary con-
cepts,” organized around the relation of different forms of freedom to “inter-
nal” and “external lawgiving” (Kant 1996c: 6:221–228). “Moral laws” involve
“internal lawgiving” through necessitation (6:221–222). The moral deontic
notions of “duty,” the morally “permitted,” and “obligation” are all defined in
relation to self-constraint (6:222). “[O]bligation involves not merely practical
necessity (such as a law in general asserts) but also necessitation,” that is, the
psychological process we have been analyzing (6:223). “Deed[s]” are actions
only “insofar as [they] come under obligatory laws” and can be “imputed” to
the “person” who is obligated to perform them (6:223). Moral laws thus admit
only of “internal lawgiving” (6:220–221).
By contrast, Kant uses “external law” to refer to “[o]bligatory laws for which
there can be an external lawgiving,” where that can operate through external
constraints in a way the “internal lawgiving” of the moral law cannot (6:220,
6:224). “All duties,” Kant says, “involve a concept of constraint through a law”
(6:394). “Ethical duties” can involve only internal lawgiving and constraint
(through necessitation), and “duties of right involve a constraint for which
external lawgiving is also possible” (6:394).
Ethical duties and duties of right mark out the two parts of The Metaphysics
of Morals: Doctrine of Right and Doctrine of Virtue, respectively. It is impor-
tant to keep in mind, however, that even though duties of right cannot be ethi-
cal duties in Kant’s technical sense, he clearly thinks we are morally obligated
not to do what duties of right proscribe.33 The point of Kant’s terminology is
that what ethical duties require is not (external) actions but the adoption and
maintenance of ends. They are “ends that are also duties,” duties of “wide obli-
gation,” as he also puts it, such that we have some discretion (Spielraum) in
determining which specific actions to perform in their pursuit (6:385). Duties
of right are “of narrow obligation,” requiring specific actions that are capable of
being exacted by external constraint and legislation.
The Doctrine of Virtue systematically explores the two “ends that are also
duties”: “one’s own perfection” and “the happiness of others” (6:385). The for-
mer, Kant considers to be duties to oneself under various headings: “as an
Animal Being,” “as a Moral Being,” and so on. And the latter are all “duties to
others,” including the “Duty of Love” and “the Duty of Respect” (6:492–493).34

33
On this point, see Guyer 2006: 240.
34
I discuss the duty of respect in relation to Kant’s views on respect in Darwall 2013e.

https://doi.org/10.1017/9781139025065.010 Published online by Cambridge University Press

 336 rousseau and kant

What makes them all ethical duties, again, is that it is impossible for them to
be exacted by external constraint. Though “I can indeed be constrained by oth-
ers to perform actions …, I can never be constrained by others to have an end:
only I myself can make something my end” (6:381). I do so through free self-
constraining necessitation; external constraint is of no avail.
Duties of right, by contrast, do admit of external constraint. Indeed, Kant
holds, they can be rightfully coerced, and the Doctrine of Right lays out how
this can be. What is especially interesting from the point of view of the themes
we have been discussing is that Kant holds that what makes ethical duties
and duties of right both duties is that both essentially involve constraint. “All
duties,” again, “involve a concept of constraint through law” (6:394). Indeed,
though the Gregor translation usually reserves “necessitation” for the self-
constraint involved in inner lawgiving, it sometimes uses “necessitation” to
refer to external constraint.
The very concept of duty is already the concept of a necessitation (con-
straint) of free choice through the law. This constraint may be an external
constraint or a self-constraint (6:379).

Kant uses both Zwang (“force”) and Nöthigung (“compulsion” or “constraint”)

to refer to self-constraint and to external constraint. The crucial point is that
even when it is backed by reason, a constraint is itself the application of force
rather than reason. Rousseau is famous (or infamous) for claiming in The Social
Contract that citizens who go against the general will must be “forced to be
free” (Rousseau 2003: 53). It seems no exaggeration to use this same language
to describe Kant’s position concerning both internal and external legislation.
We have seen how Kant thinks self-constraint works through the (painful)
feeling of respect in response to what we represent to ourselves as reason’s
commands (from, as it were, a second-person standpoint of accountability for
compliance), thereby forcing ourselves to realize autonomy (following the law
of our own will). Even if this necessarily involves the appearance of further nor-
mative reason beyond those practical-good-making considerations that back
reason’s command (more on this presently), Kant’s position is that any such
appearance is illusory and therefore that we are only able to get ourselves to
do what there is independently good reason to do through self-applied force
(Zwang).
A major aim of the Doctrine of Right is to carry the idea of freedom-preserving
constraint through law that is part of the concept of duty in general, and to
demonstrate how practical reason also dictates external constraint to preserve
external freedom. Kant defines a “doctrine of right” as “[t]he sum of those
laws for which an external lawgiving is possible” (6:229). His method,
­therefore, is not, as Ripstein has stressed, to “apply general moral principles to the
factual circumstances that make political society necessary” (Ripstein 2009: 1).

https://doi.org/10.1017/9781139025065.010 Published online by Cambridge University Press

 freedom, constraint, right in the metaphysics of morals 337

Rather Kant works from within what he takes to be the different conceptual space
of the right.
The concept of right, insofar as it is related to an obligation correspond-
ing to it (i.e., the moral concept of right), has to do … only with the exter-
nal and indeed practical relation of one person to another (Kant 1996c:
6:230).

“All that is in question,” Kant says, “is the form in the relation of choice on the
part of both, insofar as choice is regarded as free, and whether the action of one
can be united with the freedom of the other in accordance with a universal law”
(6:230).
On this basis, Kant lays out the “Universal Principle of Right”: “Any action
is right if it can coexist with everyone’s freedom in accordance with a univer-
sal law” (6:230). This idea drives Kant’s argument. Actions that are rightful
by the Universal Principle are thereby within a sphere of protected external
freedom. If, consequently, someone “hinders” such an action, thereby doing
something that cannot itself be rightful (because it cannot coexist with every-
one’s freedom in accordance with universal law), they not only do wrong,
they wrong the hindered agent, thereby violating their right (6:230–231). More
specifically, they violate what Kant calls the “one innate right”: “[f]reedom
(independence from being constrained by another’s choice), insofar as it can
coexist with the freedom of every other in accordance with a universal law”
(6:237). It is “the only original right belonging to every man by virtue of his
humanity” (6:237).
The crucial point for our purposes is that Kant takes the right to freedom to
justify external constraint. When someone hinders another’s external freedom,
the hindered person is justified in compelling the hinderer to desist, thereby
“hindering … [the] hindrance” (6:231). That this is rightful and no wronging
of the hinderer follows analytically, Kant thinks, from the Universal Principle
of Right. “Hence there is connected with right by the principle of contradiction
an authorization to coerce someone who infringes upon it” (6:231). “Right and
authorization to use coercion,” Kant says, “therefore mean one and the same
thing” (6:232).35
For Kant, consequently, external restraint preserves external freedom in a
way that is entirely analogous to the way internal self-constraint realizes free-
dom in the form of autonomy. Through the latter we force ourselves to comply
with the moral law, while through the former we force others to keep their

35
In 6:232–233, Kant lays out an analogy between a “construction” of the concept of right
that exhibits the rightfulness of a “fully reciprocal and equal coercion” and analogous geo-
metrical constructions. For a masterly discussion of this and the Doctrine of Right overall,
see Ripstein 2009.

https://doi.org/10.1017/9781139025065.010 Published online by Cambridge University Press

 338 rousseau and kant

actions within the sphere of rightful external freedom under universal law.
There is, however, a further, deeper way in which Kant’s account of the role
of constraint in duty – both in our duties of right to others and in our duties
(period) under the moral law – substitutes force for any recognition of duties as
entailing distinctive deontic reasons for acting.36
To see how this is so, notice how Kant’s understanding of a right as a jus-
tification for coercion compares to the different notion (which we found, for
example, in Grotius) of claim rights as a justification for claiming or demand-
ing certain treatment. Grotius and Kant agree that interfering with another
person’s freedom in a way that violates the Universal Principle of Right wrongs
them (6:230–231). What this means for Grotius is that it violates their claim
right; it is something they have the authority to demand we not do. But Kant’s
innate right to freedom is not really a claim right, though he does use the lan-
guage of wronging, which might be thought to entail, as a conceptual mat-
ter, the violation of a claim right (Kant 1999c: 6:256; Hohfeld 1923; Thompson
2004; Darwall 2013a). Rather, Kant’s view seems to be that when one’s free-
dom is hindered, one does no wrong (and, in Kant’s lexicon, does not wrong
the hinderer) if, in response, one hinders the hinderer by coercively restrain-
ing them.
This is a deep conceptual difference. When one makes a justified demand of
another person, one presumes an authority to direct their conduct through a
distinctive kind of reason for acting (what we might call an authority-based or,
as I might put it, a “second-personal reason”) that is grounded in one’s authori-
tative demand. And this demand is grounded in one’s assumed authority or
claim right (Feinberg 1980; Darwall 2013a). The claim right against others not
to be hindered by them is conceptually equivalent to their obligations to one
not to hinder one (Hohfeld 1923). And just as accountability is conceptually
implicated in moral obligations (period), so also is accountability to the obligee
conceptually implicated in moral obligations owed to them and, correlatively,
in the obligee’s entailed claim rights (Darwall 2013a).
The contrast between this picture and Kant’s is stark. It is the difference
between an essentially interpersonal realm of joint reasoning – in which agents
justify themselves to one another through reasons that depend upon their com-
mon authority to demand such justification – and a multi-personal realm in
which agents have the authority to exert interpersonal force. On the first, the
“realm of rights” is one in which agents are presumed to be able to recognize
and act on essentially interpersonal deontic reasons. On the second, no such
reasons are acknowledged, and force is substituted for authoritative direction
through mutually acknowledged reasons of right.

36
See Darwall 2013a for the distinction between “relational,” “directed,” or “bipolar obliga-
tions” owed to someone and moral obligations “period” (or “pure and simple”).

https://doi.org/10.1017/9781139025065.010 Published online by Cambridge University Press

 freedom, constraint, right in the metaphysics of morals 339

This point can be brought out by considering Joseph Raz’s theory of author-
ity as involving the capacity to create “preemptive” and “exclusionary” reasons
(Raz 1986). When someone with the authority to direct someone’s contact
issues a legitimate directive, Raz argues, they thereby create a new “preemp-
tive” reason for that person to act as directed and exclude from their legitimate
deliberative consideration any reasons there might otherwise have been for
them to act differently (Raz 1986: 42). If this influential idea is correct, it follows
that if someone holding a claim right has the authority to demand compliance
with the right, then the right holder has the authority to create a reason for oth-
ers to respect the right that would not have existed but for the right’s and the
authority’s existence.37
Demanding respect for a claim right, on this picture, necessarily implicates
a distinctively interpersonal form of reason-giving. On Kant’s picture, by con-
trast, the innate right not to have one’s freedom hindered creates no reason
against hindering that one could authoritatively give to potential and actual
hinderers in interpersonal practical reasoning. The authority that goes with
the right is the authorization to coerce, to hinder others’ hinderings. The right
holder has the authority to say, “Stop!” as a way of causing or forcing the hin-
derer to desist, but not with the illocutionary force of a justified demand that
creates preemptive or exclusionary reasons.
Consider now the case of internal self-constraint and legislation. External
hinderings hinder rightful external freedom. Is there anything comparable to
this in the internal case? Both internal and external constraints are hinderings of
threats to freedom. External constraints hinder threats to external freedom, and
internal self-constraint hinders threats to autonomy. But are both hinderings of
hindrance? At one point, Kant calls inclinations that threaten compliance with
the moral law “hindering impulses” (Kant 1996c: 6:405). So the hindered agent’s
hindering of these impulses is indeed a hindering of hindrances.
Recall that according to Kant’s moral psychology, self-constraint always
operates through the feeling of respect, which brings deontic notions and prop-
ositions into view. Perfect rational beings have no need of deontic morality’s
notions and stand only under practical law, not under moral law, strictly so
called. Their deliberations proceed entirely in terms of what it would be good
or best to do, what, as we might say, there is most reason for them to do. Finite
imperfect rational beings, however, stand under moral imperatives that are

37
Most usually, of course, the claim right is thought to entail a reason for compliance,
whether or not that is actually demanded by the right holder. If you have a claim right that
I not hinder your freedom, then you do not have to demand that I not do so in order for me
to have a reason not to hinder you deriving from your right. In such cases, the authority to
demand can be understood as an authority to “second” or “insist on” a justified demand
that is assumed to already be “in force” by virtue of the right, which entails a reason for
compliance that would not have existed but for the right.

https://doi.org/10.1017/9781139025065.010 Published online by Cambridge University Press

 340 rousseau and kant

“expressed by an ought” (1996a: 4:413). But this gives them no new reasons for
acting. 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 rep-
resented to it that it would be good to do that thing” (4:413). In other words,
they express no new or different normative proposition – they express the same
normative propositions the person would accept if perfectly rational – but they
do so in a distinctively peremptory way.
Kant’s view is that from the perspective of (the self-restraining feeling of)
respect, imperatives have the apparent content of reason’s commands. If Raz’s
influential picture is correct, however, authoritative directives like commands
create preemptive and exclusionary reasons for acting. The reason-commanded
agent would not only have good and sufficient reason to do something. They
would also have a further preemptive reason to do it, one that would exclude
any reasons to the contrary and with which they would be accountable for
compliance (Darwall 2013f). On Kant’s view, however, self-regulation through
respect does not give the agent a further preemptive reason for complying with
reason’s demand. As Kant sees it, “the view from respect” is illusory; it is the
misleading appearance of a non-rational goad that finite rational agents need
to overcome their inclinations against doing what they have sufficient reason to
do already. It is, like the interpersonal case, an application of force rather than
reason, albeit one that, also like the interpersonal case, has reason’s backing.

Kant’s Racism and Sexism, and a Final Note of Humility

It is appropriate at this point in this chapter, and in the book overall, to say
something about Kant’s views on race, sex, and colonial position; how these
affected his moral philosophy; and what we can learn from that about the his-
tory of modern moral philosophy more generally and about our own moral the-
orizing today. As I mentioned in the Preface, no one can doubt that the culture
that produced and supported modern moral philosophy and its history was, and
remains, sexist and racist, patriarchal and white supremacist. Testimony to this
is the virtual absence from these pages of women and thinkers of color.
In recent years, however, scholars have become significantly more interested
in the sociopolitical context of modern philosophy, in identifying excluded
thinkers, and in critically assessing modern philosophy, and modern philoso-
phers, from this perspective. Because of his importance, but also because he
provides such an egregious example, it is not surprising that much of this new
scholarship has focused on Kant (Allais 2016; Bernasconi 2001, 2002, 2003; Eze
1997; Hay 2013; Herman 1993; Hill and Boxill 2001; Langton 1992; Lu-Adler
2022a, 2022b; Mills 1997, 2005a, 2005b, 2017; Spivak 1999; Varden 2020).38

38
I have been helped by thinking through these issues with Moya Mapps.

https://doi.org/10.1017/9781139025065.010 Published online by Cambridge University Press

 kant’s racism and sexism, and a final note of humility 341

Some of this work provides detailed analysis of Kant’s racism and sexism.
And Kant provides much grist for the mill of racial and gender critique.
As Huseyinzadegan notes, “Kant’s problematic claims … range from the
unnaturalness of a female philosopher, ‘who might as well have a beard,’ the
stupid things that a black carpenter said ‘because he was black from head
to foot,’ the poor women ‘living in the greatest slavery in the Orient,’ to the
‘sheep-like existence of the inhabitants of Tahiti’” (Huseyinzadegan 2018: 1).
Nor are these throw-away remarks. Kant had a theory of race that gave con-
temporary racism a philosophical pedigree (Kant 2013b). Moreover, Kant’s
racism and sexism directly influenced his moral and political philosophy, at
least in applying it to actual circumstances. Thus, he held in The Doctrine of
Right that women were justifiably denied voting rights and could at best be
“passive citizens” owing to their dependency on their husbands (Kant 1996c:
6:314).39
Of course, this does not necessarily impugn Kant’s fundamental moral and
political ideas; perhaps he just misapplied them because of his various privi-
leged biases. Some Kant scholars argue, indeed, that Kant’s moral and political
philosophy can be put to use to resist racist and sexist oppression. This is most
obvious in the case of doctrines of equal dignity and rights that fueled so much
of the civil rights and liberation movements of the 1960s and beyond. But it is
also arguably the case in a more fine-grained way, as Helga Varden argues is
true of some of Kant’s moral and political doctrines in the Doctrine of Right
(Varden 2020).
Perhaps the most interesting perspective that close study of Kant, and other
modern moral philosophers, along these lines can reveal is its capacity to help
us gain critical purchase on the role of bias in our own moral theory and prac-
tice. This is a point made by both Allais and Huseyinzadegan (Allais 2016;
Huseyinzadegan 2018). Though we are the beneficiaries of an improved moral
understanding that is partly the result centuries of resistance by oppressed the-
orists and activists, readers of this book, as well as its author, are likely subject
to distortions by virtue of our own privileged sociopolitical positions. Taking
account of these may help us to pursue moral theory and practice from a more
critically informed perspective, as, indeed, Kant’s critical philosophy itself
encourages us to do.
But sociopolitical distortions run deeper. Perhaps it is the case, as Evgeny
Pashukanis argued, following Marx, that the very concepts of modern moral
philosophy – those of the moral subject, deontic morality, accountability, and
so on – are the superstructural ideology of an underlying capitalist mode of
production (Pashukanis 2003). Nietzsche, of course, makes a structurally simi-
lar argument in The Genealogy of Morality (Nietzsche 2007).

39
He also says the same of wage laborers because of their dependency on their bosses.

https://doi.org/10.1017/9781139025065.010 Published online by Cambridge University Press

 342 rousseau and kant

But here again, perhaps these and similar genealogies, including evolution-
ary ones, that seek to debunk morality can be resisted on the grounds that mod-
ern moral philosophy and its history enable us to appreciate. Even if a given
genealogy is correct, it does not follow that the ideas it explains are mistaken.
Indeed, it may be possible to combine modern moral philosophy, sociopolitical
history, and evolutionary theory to come to an overall understanding both of
why core modern moral philosophical conceptions are valid philosophically
and of how we have evolved, both biologically and socioculturally, to think they
are (Tomasello, 2018; Buchanan and Powell 2018).

https://doi.org/10.1017/9781139025065.010 Published online by Cambridge University Press

 WORKS CITED

Abizadeh, Arash (2018). Hobbes and the Two Faces of Ethics. Cambridge:
Cambridge University Press.
Abramson, Kate (2002). “Two Portraits of the Humean Moral Agent,” Pacific
Philosophical Quarterly 83: 301–334.
Abramson, Kate (2007). “Hume’s Distinction between Philosophical Anatomy
and Painting,” Philosophy Compass 2: 680–698.
Albee, Ernest (1957). A History of English Utilitarianism. London: Allen & Unwin.
Alexander, Larry and Michael Moore (2016). “Deontological Ethics,” Stanford
Encyclopedia. https://plato.stanford.edu/entries/ethics-deontological/
Allais, Lucy (2016). “Kant’s Racism,” Philosophical Papers 45: 1–36.
Allison, Henry (1990). Kant’s Theory of Freedom. Cambridge: Cambridge
University Press.
Allison, Henry (1991). “On a Presumed Gap in Kant’s Derivation of the Categorical
Imperative,” Philosophical Topics 19: 1–15.
Anderson, Elizabeth (1999). “What Is the Point of Equality?” Ethics 109: 187–237.
Anderson, Elizabeth (2010). The Imperative of Integration. Princeton: Princeton
University Press.
Anderson, Elizabeth (2012). “The Fundamental Disagreement between Luck
Egalitarians and Relational Egalitarians,” Canadian Journal of Philosophy 36
(Supp.): 1–23.
Anderson, Elizabeth (2014). “Social Movements, Experiments in Living, and
Moral Progress: Case Studies from Britain’s Abolition of Slavery.” The Lindley
Lecture, The University of Kansas.
Anderson, Elizabeth (forthcoming). The Great Reversal: How Neoliberalism
Turned the Work Ethic against Workers. Cambridge: Cambridge University
Press.
Annas, Julia (1993). The Morality of Happiness. Oxford: Oxford University Press.
Anscombe, G. E. M. (1957). Intention. Oxford: Basil Blackwell.
Anscombe, G. E. M. (1958). “Modern Moral Philosophy,” Philosophy 33: 1–19.
Aquinas, Thomas (1997). “Summa Theologica,” in Basic Writings of Thomas
Aquinas, v. II, ed. Anton C. Pegis. Indianapolis, IN: Hackett Publishing
Company. Originally published in 1265–1273.
Aristotle (1999). Nicomachean Ethics, ed. and trans. Terence Irwin. Indianapolis,
IN: Hackett Publishing Company.
343

Published online by Cambridge University Press

 344 works cited

Aristotle (2000). Nicomachean Ethics, ed. and trans. Roger Crisp. Cambridge:
Cambridge University Press.
Arpaly, Nomy (2004). Unprincipled Virtue: An Inquiry into Moral Agency. Oxford:
Oxford University Press.
Atherton, Margaret (1994). Women Philosophers of the Early Modern Period.
Indianapolis, IN: Hackett Publishing Company.
Augustine (1990). Homilies on the First Epistle of John, in The Works of Saint
Augustine: A Translation for the 21st Century, intro., trans., and notes Boniface
Ramsay. New Hyde Park, NY: New City Press.
Aune, Bruce (1979). Kant’s Theory of Morals. Princeton: Princeton University
Press.
Austin, J. L. (1957). “A Plea for Excuses,” Proceedings of the Aristotelian Society 57:
1–30.
Baier, Annette (1985). Postures of the Mind: Essays on Mind and Morals.
Minneapolis: University of Minnesota Press.
Baier, Annette (1993). “Moralism and Cruelty: Reflections on Hume and Kant,”
Ethics 103: 436–457.
Balguy, John (1976). The Foundation of Moral Goodness, 2 vol. in one. New York:
Garland Publishing Co., Inc. Originally published in 1728, 1729.
Barbeyrac, Jean (1718). Jugement d’un anonyme … avec dex reflexions du tra-
ducteur … in Pufendorf (1718).
Barbeyrac, Jean (1749). “An Historical and Critical Account of the Science of
Morality,” in The Law of Nature and Nations, eds. Samuel Pufendorf, trans.
Basil Kennett, ed. Jean Barbeyrac, 5th ed. London, pp. 1–73.
Batson, C. Daniel and Laura L. Shaw (1991). “Evidence for Altruism: Toward a
Pluralism of Prosocial Motives,” Psychological Inquiry 2: 107–122.
Bentham, Jeremy (1838). The Works of Jeremy Bentham, ed. John Bowring.
Edinburgh: William Tait.
Bernasconi, Robert. (2001). “Who Invented the Concept of Race? Kant’s Role in
the Enlightenment Construction of Race,” in Race: Readings in Continental
Philosophy, ed. Robert Bernasconi. Oxford: Blackwell, pp. 11–36.
Bernasconi, Robert (2002). “Kant as an Unfamiliar Source of Racism,” in
Philosophers on Race: Critical Essays, eds. Julia K. Ward and Tommy L. Lott.
Oxford: Blackwell.
Bernasconi, Robert (2003). “Will the Real Kant Please Stand Up: The Challenge
of Enlightenment Racism to the Study of the History of Philosophy,” Radical
Philosophy 117: 13–22.
Blackburn, Simon (1984). Spreading the Word: Groundings in the Philosophy of
Language. Oxford: Oxford University Press.
Bodin, Jean (2004). On Sovereignty, ed. and trans. John H. Franklin. Cambridge:
Cambridge University Press.
Boxill, Bernard (1984). Blacks and Social Justice. Lanham, MD: Rowman &
Littlefield.

Published online by Cambridge University Press

 works cited 345

Boxill, Bernard (2003). “A Lockean Argument for Black Reparations,” The Journal
of Ethics 7: 63–91.
Brandt, Richard (1979). A Theory of the Good and the Right. Oxford: Oxford
University Press.
Bratman, Michael (1981). “Intention and Means-End Reasoning,” The Philosophical
Review 90: 252–265.
Bratman, Michael (1987). Intentions, Plans, and Practical Reason. Cambridge,
MA: Harvard University Press.
Brett, Annabel S. (1997). Liberty, Right and Nature: Individual Rights in Later
Scholastic Thought. Cambridge: Cambridge University Press.
Broad, C. D. (1962). Five Types of Ethical Theory. London: Routledge & Kegan Paul.
Broad, Jacqueline (2006). “A Woman’s Influence? John Locke and Damaris
Masham on Moral Accountability,” Journal of the History of Ideas 67: 489–510.
Broome, John (2013). Rationality through Reasoning. Oxford: Wiley/Blackwell.
Brown, Charlotte (1988). “Is Hume an Internalist,” Journal of the History of
Philosophy 26: 69–87.
Brown, Lesley (2007). “Glaucon’s Challenge, Rational Egoism, and Ordinary
Morality,” in Pursuing the Good: Ethics and Metaphysics in Plato’s Republic,
eds. Douglas Cairns and Fritz-Gregor Herrmann. Edinburgh: University of
Edinburgh Press.
Brunero, John and Niko Kolodny (2018). “Instrumental Rationality,” Stanford
Encyclopedia of Philosophy.
Buchanan, Allen and Russell Powell (2018). The Evolution of Moral Progress: A
Biocultural Theory. New York: Oxford University Press.
Bukoski, Michael (2018). “Korsgaard’s Argument for the Value of Humanity,” The
Philosophical Review 127: 197–224.
Burge, Tyler (1993). “Content Preservation,” Philosophical Review 102: 457–488.
Butler, Joseph (2017). Fifteen Sermons and Other Writings on Ethics, ed. David
McNaughton. Oxford: Oxford University Press. Originally published in 1726.
Reference to sermon and paragraph numbers.
Cambridge Platonists Project, www.cambridge-platonism.divinity.cam.ac.uk/
view/texts/normalised/index
Carter, Benjamin (2010). “The Standing of Ralph Cudworth as a Philosophy,” in
Insiders and Outsiders in Seventeenth-Century Philosophy, eds. G. A. J. Rogers,
Tom Sorrell, and Jill Kraye. New York: Routledge.
Cassirer, Ernst (1951). The Philosophy of the Enlightenment, trans. Fritz C. A.
Koelln and James P. Pettegrove. Princeton: Princeton University Press.
Cassirer, Ernst (1953). The Platonic Renaissance in England, trans. James
Pettegrove. Austin: University of Texas Press.
Chappell, Richard Yetter (2012). “Fittingness: The Sole Normative Primitive,”
Philosophical Quarterly 62: 684–704.
Clarke, Samuel (1978a). A Discourse Concerning the Being and Attributes of God, in
Works, v. ii. New York: Garland Publishing, Inc. Originally published in 1704.

Published online by Cambridge University Press

 346 works cited

Clarke, Samuel (1978b). A Discourse Concerning the Unchangeable Obligations and

the Truth and Certainty of the Christian Revelation, in Works, v. ii. New York:
Garland Publishing, Inc. Originally published in 1705.
Coady, C. A. J. (1992). Testimony: A Philosophical Study. Oxford: Oxford University
Press.
Cohon, Rachel (2000). “The Roots of Reasons,” The Philosophical Review 109:
63–85.
Colman, John (1983). John Locke’s Moral Philosophy. Edinburgh: Edinburgh
University Press.
Constant, Benjamin (2003). Principles of Politics Applicable to All Governments.
Indianapolis, IN: Liberty Fund. Originally published in 1810.
Cooper, John M. (1996). “Justice and Rights in Aristotle’s Politics,” Review of
Metaphysics 49: 859–872.
Crisp, Roger, Sacrifice Regained. Oxford: Oxford University Press.
Cudworth, Ralph (1969). “A Sermon Preached before the House of Commons,”
in The Cambridge Platonists, ed. C. A. Patrides. Cambridge: Cambridge
University Press, pp. 90–127.
Cudworth, Ralph (~1670). Manuscripts on freedom of the will. British Library,
Additional Manuscripts, nos. 4978–4982.
Cudworth, Ralph (1678). The True Intellectual System of the Universe. London:
Richard Royston.
Cudworth, Ralph (1996a). A Treatise Concerning Eternal and Immutable Morality,
ed. Sarah Hutton. Cambridge: Cambridge University Press.
Cudworth, Ralph (1996b). “A Treatise of Freewill,” in Cudworth 1996.
Cumberland, Richard (1683). De legibus naturae disquisitio philosophica, 2nd ed.
Lubeck and Frankfurt: S. Otto & J. Wiedemeyer. (First edition published in
1672.)
Cumberland, Richard (2005). A Treatise of the Laws of Nature, trans., with an intro.
and appendix, John Maxwell, ed., with a foreword, John Parkin. Indianapolis,
IN: Liberty Fund. (Based on 1727 edition.)
Daniel, Dafydd Mills (2020). Ethical Rationalism and Secularisation in the British
Enlightenment: Conscience and the Age of Reason. Oxford: Palgrave Macmillan.
D’Arms, Justin and Daniel Jacobson (2000a). “The Moralistic Fallacy: On the
‘Appropriateness’ of Emotions,” Philosophy and Phenomenological Research 61:
65–90.
D’Arms, Justin and Daniel Jacobson (2000b). “Sentiment and Value,” Ethics 110:
722–748.
Darwall, Stephen (1977). “Two Kinds of Respect,” Ethics 88: 36–49.
Darwall, Stephen (1980). “Is There a Kantian Foundation for Rawlsian Justice?”
in John Rawls’ Theory of Social Justice, ed. H. G. Blocker and E. Smith. Athens:
Ohio University Press.
Darwall, Stephen (1983). Impartial Reason. Ithaca, NY: Cornell University Press.
Darwall, Stephen (1994). “Hume and the Invention of Utilitarianism,” in Hume
and Hume’s Connexions, eds. M. A. Stewart and J. Wright. Edinburgh:
Edinburgh University Press.

Published online by Cambridge University Press

 works cited 347

Darwall, Stephen (1995). The British Moralists and the Internal “Ought”: 1640–
1740. Cambridge: Cambridge University Press.
Darwall, Stephen (1997). “Reasons, Motives, and the Demands of Morality: An
Introduction,” in Moral Discourse and Practice, eds. S. Darwall, A. Gibbard,
and P. Railton. New York: Oxford University Press.
Darwall, Stephen (1998). Philosophical Ethics. Boulder, CO: Westview Press.
Darwall, Stephen (1999a). “The Inventions of Autonomy,” European Journal of
Philosophy 7: 339–350.
Darwall, Stephen (1999b). “Sympathetic Liberalism: Recent Work on Adam
Smith,” Philosophy & Public Affairs 28: 139–164.
Darwall, Stephen (2000). “Normativity and Projection in Hobbes’s Leviathan,”
The Philosophical Review 109: 313–347.
Darwall, Stephen (2002). Welfare and Rational Care. Princeton: Princeton
University Press.
Darwall, Stephen (2003). “Autonomy in Modern Natural Law,” in New Essays on
the History of Autonomy, eds. Larry Krasnoff and Natalie Brender. Cambridge:
Cambridge University Press.
Darwall, Stephen (2004). “Equal Dignity in Adam Smith,” Adam Smith Review 1:
xxx.
Darwall, Stephen (2006). The Second-Person Standpoint: Morality, Respect, and
Accountability. Cambridge, MA: Harvard University Press.
Darwall, Stephen (2008). “Kant on Respect, Dignity, and the Duty of Respect,”
in Kant’s Virtue, ed. Monika Betzler. Berlin: Walter de Gruyter, pp. 175–200.
Reprinted in Darwall 2013c.
Darwall, Stephen (2012a). “Grotius at the Creation of Modern Moral Philosophy,”
Archiv für Geschichte der Philosophie 94: 296–325. Reprinted in Darwall
2013c.
Darwall, Stephen (2012b). “Pufendorf on Morality, Sociability, and Moral Powers,”
The Journal of the History of Philosophy 50: 213–238. Reprinted in Darwall
2013c.
Darwall, Stephen (2013a). “Bipolar Obligation,” in Morality, Authority, and Law:
Essays in Second-Personal Ethics I, ed. Stephen Darwall. Oxford: Oxford
University Press, pp. 20–39.
Darwall, Stephen (2013b). Morality, Authority, and Law: Essays in Second-Personal
Ethics I. Oxford: Oxford University Press.
Darwall, Stephen (2013c). Honor, History, and Relationship: Essays in Second-
Personal Ethics II. Oxford: Oxford University Press.
Darwall, Stephen (2013d). “Respect as Honor and as Accountability,” in Darwall
2013c.
Darwall, Stephen (2013e). “Kant on Respect, Dignity, and the Duty of Respect,” in
Darwall 2013c.
Darwall, Stephen (2013f). “Authority and Second-Personal Reasons for Acting,”
in Darwall 2013b.
Darwall, Stephen (2013g). “The Value of Autonomy and Autonomy of the Will,”
in Darwall 2013a.

Published online by Cambridge University Press

 348 works cited

Darwall, Stephen (2013h). “Forcing Freedom” (critical review of Arthur Ripstein’s

Force and Freedom), Legal Theory 19: 89–99.
Darwall, Stephen (2013i). “Demystifying Promises,” in Darwall 2013c.
Darwall, Stephen (2014a). “Agreement Matters” (critical Notice of Derek Parfit’s
On What Matters), The Philosophical Review 123: 79–105.
Darwall, Stephen (2014b). “Review of Richard Sorabji’s Moral Conscience,
Its History through the Ages: 5th Century BCE to the Present.” Notre Dame
Philosophical Reviews. https://ndpr.nd.edu/news/56313-moral-conscience-
through-the-ages-fifth-century-bce-to-the-present/
Darwall, Stephen (2014c). “The Social and the Sociable,” Philosophical Topics, spe-
cial issue, eds. James Conant and Sebastian Rödl 42: 201–217.
Darwall, Stephen (2016). “Making the Hard Problem of Moral Normativity
Easier,” in Weighing Reasons, eds. Errol Lord and Barry Maguire. Oxford:
Oxford University Press.
Darwall, Stephen (2018). “Contempt as an Other-Characterizing, ‘Hierarchizing’
Attitude,” in The Moral Psychology of Contempt, ed. Michelle Mason. Lanham,
MD: Rowman & Littlefield.
Darwall, Stephen (2022). “Hutcheson in the History of Rights,” Journal of Scottish
Philosophy 20: 85–101.
Darwall, Stephen, Allan Gibbard, and Peter Railton (1997). “Toward Fin de Siècle
Ethics: Some Trends,” in Moral Discourse and Practice, eds. S. Darwall, A.
Gibbard, and P. Railton. New York: Oxford University Press.
Dean, Richard (2006). The Value of Humanity in Kant’s Moral Theory. Oxford:
Oxford University Press.
Debes, Remy (2012). “Adam Smith on Dignity and Equality,” British Journal for the
History of Philosophy 20: 109–140.
De Grouchy, Sophie (2019). Sophie de Grochie’s Letters on Sympathy, trans Sandrine
Bergès, introduction, glossary, and commentary, Eric Schliesser and Sandrine
Bergès. Oxford: Oxford University Press.
De Kenessey, Brendan (2017). Joint Practical Deliberation. Diss., Massachusetts
Institute of Technology.
Della Rocca, Michael (2008). Spinoza. London and New York: Routledge.
Descartes, René (2017). Meditations on First Philosophy: With Selections from the
Objections and Replies. Cambridge: Cambridge University Press.
Dorsey, Dale (2010). “Hutcheson’s Deceptive Hedonism,” The Journal of the
History of Philosophy 48: 445–467.
Dreier, James (1990). “Internalism and Speaker Relativism,” Ethics 101: 6–26.
Elster, Jon (1986). An Introduction to Karl Marx. Cambridge: Cambridge
University Press.
Engelmann, Jan and Michael Tomasello (2019). “Children’s Sense of Fairness as
Equal Respect,” Trends in Cognitive Science 23: 454–463.
Engels, Friedrich (1975). “Socialism: Utopian and Scientific,” in Collected Works
of Karl Marx and Friedrich Engels, v. 24. New York: International Publishers.
Ewing, A. C. (2012a). The Definition of Good. London: Routledge & Kegan Paul.

Published online by Cambridge University Press

 works cited 349

Ewing, A. C. (2012b). Second Thoughts in Moral Philosophy. Abingdon, Oxford:

Routledge. Reprints. Originally published in 1959.
Eze, Emanuel Chukwudi 1997. “The Color of Reason: The Idea of ‘Race’ in Kant’s
Anthropology,” in Postcolonial African Philosophy: A Critical Reader, ed.
Emanuel Chukwudi Eze. Oxford: Blackwell.
Falk, W. D. (1948). “‘Ought’ and Motivation,” Proceedings of the Aristotelian
Society 48: 113–138.
Feinberg, Joel (1980). “The Nature and Value of Rights,” in Rights, Justice, and the
Bounds of Liberty. Princeton: Princeton University Press.
Fichte, Johann Gottlieb (2000). Foundations of Natural Right, ed. Frederick
Neuhouser, trans. Michael Bauer. Cambridge: Cambridge University Press.
Fleischacker, Samuel (1991). “Philosophy in Moral Practice: Kant and Adam
Smith,” Kant-Studien 82: 249–269.
Fleischacker, Samuel (2004). A Short History of Distributive Justice. Cambridge,
MA: Harvard University Press.
Fleischacker, Samuel (2005). A Short History of Distributive Justice. Cambridge,
MA: Harvard University Press.
Fleischacker, Samuel (2013). “Adam Smith,” Stanford Encyclopedia of Philosophy,
https://plato.stanford.edu/entries/smith-moral-political
Foot, Philippa (1958). “Moral Arguments,” Mind 67: 502–513.
Foot, Philippa (1959). “Moral Beliefs,” Proceedings of the Aristotelian Society 59:
83–104.
Foot, Philippa (1972). “Morality as a System of Hypothetical Imperatives,” The
Philosophical Review 81: 305–316.
Foot, Philippa (1978). Virtues and Vices, and Other Essays in Moral Philosophy.
Oxford: Basil Blackwell.
Foot, Philippa (2002). “Nietzsche’s Immoralism,” in Moral Dilemmas, and Other
Topics in Moral Philosophy. Oxford: Oxford University Press.
Fordyce, David (2003). The Elements of Moral Philosophy, in Three Books with
a Brief Account of the Nature, Progress, and Origin of Philosophy, ed. with an
intro. by Thomas Kennedy. Indianapolis, IN: Liberty Fund. Originally pub-
lished in 1754.
Forsyth, Murray (1982). “The Place of Richard Cumberland in the History of
Natural Law Doctrine,” Journal of the History of Philosophy 20: 23–42.
Frankena, William (1942). “Obligation and Value in the Ethics of G. E. Moore,” in
The Philosophy of G. E. Moore, ed. Paul A. Schilpp. La Salle, IL: Open Court.
Frankena, William (1955). “Hutcheson’s Moral Sense Theory,” Journal of the
History of Philosophy 16: 356–375.
Frankena, William (1963). Ethics. Englewood Cliffs, NJ: Prentice Hall, Inc.
Frankena, William (1992). “Sidgwick and the History of Ethical Dualism,” in Essays
on Henry Sidgwick, ed. Bart Schultz. Cambridge: Cambridge University Press.
Fricker, Miranda (2007). Epistemic Injustice. Oxford: Oxford University Press.
Garrett, Aaron (2004). “Hume’s ‘Original Difference’: Race, National Character
and the Human Sciences,” Eighteenth-Century Thought 2: 127–152.

Published online by Cambridge University Press

 350 works cited

Gauthier, David P. (1967). “Morality and Advantage,” The Philosophical Review

76: 460–474.
Gauthier, David P. (1969). The Logic of Leviathan: The Moral and Political Theory
of Thomas Hobbes. Oxford: Clarendon Press.
Gauthier, David P. (1979). “Thomas Hobbes: Moral Theorist,” Journal of Philosophy
76: 547–559.
Gibbard, Allan (1990). Wise Choices, Apt Feelings. Cambridge, MA: Harvard
University Press.
Gill, Michael (2006). The British Moralists on Human Nature and the Birth of
Secular Ethics. Cambridge: Cambridge University Press.
Goldman, Alvin I. (1992). “In Defense of the Simulation Theory,” Mind and
Language 7: 104–119.
Gordon, Robert M. (1992). “The Simulation Theory: Objections and
Misconceptions,” Mind and Language 7: 11–34.
Grean, Stanley (1967). Shaftesbury’s Philosophy of Religion and Ethics: A Study in
Enthusiasm. Athens: Ohio University Press.
Green, Leslie (2003). “Legal Positivism,” Stanford Encyclopedia of Philosophy.
https://plato.stanford.edu/entries/legal-positivism/
Griswold, Charles (1999). Adam Smith and the Virtues of Enlightenment.
Cambridge: Cambridge University Press.
Grotius, Hugo (1724). Le Droit de La Guerre et de la Paix, trans. Jean Barbeyrac.
Amsterdam: Pierre de Coup.
Grotius, Hugo (1925). On the Law of War and Peace, trans. Francis W. Kelsey, intro
James Brown Scott. Oxford: Clarendon Press.
Grotius, Hugo (1950). De Iure Praedae Commentarius; Commentary on the Law of
Prize and Booty, trans. Gwladys L. Williams with the collaboration of Walter
H. Zeydel. Oxford: Clarendon Press. Original manuscript of 1604.
Grotius, Hugo (2005). The Rights of War and Peace, 3 vols., ed. Richard Tuck. From
the 1738 English translation by John Morrice of Jean Barbeyrac’s French trans-
lation, with Barbeyrac’s notes. Indianapolis, IN: Liberty Fund. Originally pub-
lished in 1625.
Guyer, Paul (2006). Kant. New York: Routledge.
Haakonssen, Knud (1996). Natural Law and Moral Philosophy: From Grotius to the
Scottish Enlightenment. Cambridge: Cambridge University Press.
Haakonssen, Knud (1998). “Divine/Natural Law Theories of Ethics,” in Cambridge
History of Seventeenth-Century Philosophy, vol. 2, 2 vols., eds. Daniel Garber
and Michael Ayers. Cambridge: Cambridge University Press, pp. 1317–1357.
Haakonssen, Knud (2001). “The Character and Obligation of Natural Law
According to Richard Cumberland,” in English Philosophy in the Age of Locke,
ed. M. A. Stewart. Oxford: Oxford University Press, pp. 29–47.
Habermas, Jürgen (1990a). Moral Consciousness and Communicative Action, trans.
Christian Lenhardt and Shierry Weber Nicholsen. Cambridge, MA: MIT Press.
Hampton, Jean (1986). Hobbes and the Social Contract Tradition. Cambridge:
Cambridge University Press.

Published online by Cambridge University Press

 works cited 351

Hannay, Alastair (2021). “Kierkegaard against the Herd,” iai News. https://iai.tv/
articles/kierkegaard-against-the-herd-auid-1826?_auid=2020
Hare, R. M. (1982). Moral Thinking: Its Levels, Methods, and Point. Oxford: Oxford
University Press.
Hart, H. L. A. (1955). “Are There Any Natural Rights?” The Philosophical Review
64: 175–191.
Hart, H. L. A. (1961). The Concept of Law. Oxford: Clarendon Press.
Hart, H. L. A. (2012). The Concept of Law. 3rd ed., ed. Leslie Green. Oxford: Oxford
University Press.
Hay, Carol (2013). Kantianism, Liberalism, and Feminism: Resisting Oppression.
New York: Palgrave-McMillan.
Hegel, Georg Wilhelm Friedrich (1991). Elements of the Philosophy of Right, eds.
Allen W. Wood and Hugh B. Nisbett, eds. Cambridge: Cambridge University
Press. Originally published in 1821.
Herman, Barbara (1993). “Could It Be Worth Thinking about Kant on Sex
and Marriage?” In A Mind of One’s Own: Feminist Essays on Reason and
Objectivity, eds. Louise M. Anthony and Charlotte Witt. Boulder, CO:
Westview Press.
Heydt, Colin (2014). “Utilitarianism before Bentham,” in The Cambridge
Companion to Utilitarianism, eds. Ben Eggleston and Dale Miller. Cambridge:
Cambridge University Press, pp. 16–37.
Hieronymi, Pamela (2005). “The Wrong Kind of Reason,” The Journal of
Philosophy 102: 437–457.
Hieronymi, Pamela (2013). “The Use of Reasons in Thought (and the Use of
Earmarks in Argument),” Ethics 124: 114–127.
Hill, Thomas E., Jr. (1973). “The Hypothetical Imperative,” The Philosophical
Review 82: 429–450.
Hill, Thomas E. and Bernard Boxill. (2001). “Kant on Race,” in Race and Racism,
ed. Bernard Boxill. New York: Oxford University Press.
Hobbes, Thomas (1839). The English Works of Thomas Hobbes, 11 vols., ed. Sir
William Molesworth. London: J. Bohn.
Hobbes, Thomas (1983). De Cive, the English Version, Entitled in the First Edition,
Philosophical Rudiments Concerning Government and Society, ed. and trans.
Howard Warrender. Oxford: Clarendon Press. References are to chapter and
paragraph number.
Hobbes, Thomas (1994a). Leviathan, ed. Edwin Curley. Indianapolis, IN: Hackett
Publishing Co., Inc. Originally published in 1651. References are to chapter and
paragraph number.
Hobbes, Thomas (1994b). The Elements of Law, Natural and Politic, ed. J. C. A.
Gaskin. Oxford: Oxford University Press.
Hobbes, Thomas (2014). Behemoth, v. x of Clarendon Edition of the Works of
Thomas Hobbes. Oxford: Oxford University Press.
Hochschild, Adam (2005). Bury the Chains: Prophets and Rebels in the Fight to
Free an Empire’s Slaves. Boston: Houghton Mifflin Company.

Published online by Cambridge University Press

 352 works cited

Hockett, Robert and Aaron James (2020). Money from Nothing: Or, Why We Should
Stop Worrying about Debt and Learn to Love the Federal Reserve. Brooklyn, NY:
Melville House Publishing.
Hoekstra, Kinch (1997). “Hobbes and the Foole,” Political Theory 25: 620–654.
Hohfeld, Wesley Newcomb (1923). Fundamental Legal Conceptions as Applied in
Judicial Reasoning and Other Legal Essays. New Haven, CT: Yale University
Press.
Hostler, John (1975). Leibniz’s Moral Philosophy. London: Duckworth.
Howard, Christopher (2019). “The Fundamentality of Fit,” Oxford Studies in
Metaethics 14: 216–236.
Hruschka, Joachim (1991). “The Greatest Happiness Principle and Other Early
German Anticipations of Utilitarian Theory,” Utilitas 3: 165–177.
Hudson, W. D. (1970). Modern Moral Philosophy. Garden City, NY: Doubleday.
Hume, David (1932). The Letters of David Hume, ed. J. Y. T. Grieg. Oxford:
Clarendon Press.
Hume, David (1978). A Treatise of Human Nature, ed. L. A. Selby-Bigge, 2nd ed.,
with rev. P. H. Nidditch. Oxford: Oxford University Press. Originally pub-
lished in 1739, 1740.
Hume, David (1975a). An Enquiry Concerning Human Understanding, in Enquiries
Concerning Human Understanding and Concerning the Principles of Morals,
ed. L. A. Selby-Bigge, 3rd. ed., rev. P. H. Nidditch. Oxford: Clarendon Press.
Originally published in 1748.
Hume, David (1975b). “An Enquiry Concerning the Principles of Morals,” in
Enquiries Concerning Human Understanding and Concerning the Principles of
Morals, ed. L. A. Selby-Bigge, 3rd. ed., rev. P. H. Nidditch. Oxford: Clarendon
Press. Originally published in 1751.
Hurka, Thomas (2014). British Ethical Theorists from Sidgwick to Ewing. Oxford:
Oxford University Press.
Hursthouse, Rosalind (1999). On Virtue Ethics. Oxford: Oxford University Press.
Huseyinzadegan, Dilek (2018). “For What Can the Kantian Feminist Hope?
Constructive Complicity in Appropriations of the Canon,” Feminist Philosophy
Quarterly 4.1: 1–25.
Hutcheson, Francis (1755). A System of Moral Philosophy, 2 vols. Glasgow: R. A.
Foulis. Facsimile edition in Collected Works, vols. 5 and 6. Hildesheim: Olms,
1969–1971.
Hutcheson, Francis (2002). An Essay on the Nature and Conduct of the Passions
and Affections, with Illustrations on the Moral Sense, ed. with an intro., Aaron
Garrett. Indianapolis, IN: Liberty Fund. Originally published in 1728, 2nd ed.,
1730, 3rd ed., 1742. This edition is based on the first edition but also includes
passages from later editions.
Hutcheson, Francis (2004). An Inquiry into the Original of Our Ideas of Beauty and
Virtue in Two Treatises, ed. with an intro., Wolfgang Leidhold. Indianapolis,
IN: Liberty Fund. (Based on the 2nd edition, 1726; 1st edition published in
1725.) This edition also includes passages from the revised 5th edition 1753.

Published online by Cambridge University Press

 works cited 353

Hutcheson, Francis (2007). A Short Introduction to Moral Philosophy, ed. with an

intro., Luigi Turco. Indianapolis, IN: Liberty Fund. Originally published in
1747.
Hutton, Sarah (1993). “Damaris Cudworth, Lady Masham: Between Platonism
and Enlightenment,” British Journal for the History of Philosophy 1: 29–54.
Immerwahr, John (1991). “The Anatomist and the Painter: The Continuity of
Hume’s Treatise and Essays,” Hume Studies 1: 1–17.
Irwin, T. H. (2003). “Stoic Naturalism and Its Critics,” in The Cambridge
Companion to the Stoics, ed. Brad Inwood. Cambridge: Cambridge University
Press, pp. 345–364.
Irwin, Terence (2007). The Development of Ethics: A Historical and Critical
Study, Volume I: From Socrates to the Reformation. Oxford: Oxford
University Press.
Irwin, Terence (2008). The Development of Ethics: A Historical and Critical Study,
Volume II: From Suarez to Rousseau. Oxford: Oxford University Press.
Irwin, Terence (2009). The Development of Ethics: A Historical and Critical Study,
Volume III: From Kant to Rawls. Oxford: Oxford University Press.
Israel, Jonathan (2001). Radical Enlightenment: Philosophy and the Making of
Modernity. Oxford: Oxford University Press.
Jackson, Frank (1998). From Metaphysics to Ethics: A Defence of Conceptual
Analysis. Oxford: Oxford University Press.
Johns, Christopher (2013). The Science of Right in Leibniz’s Moral and Political
Philosophy. London: Bloomsbury.
Johnson, Robert (2016). “Kant’s Moral Philosophy,” Stanford Encyclopedia of
Philosophy. https://plato.stanford.edu/entries/kant-moral/
Kahneman, Daniel, and Amos Tversky (1982). “The Simulation Heuristic,” in
Judgment Under Uncertainty, eds. D. Kahneman, P. Slovic, and A. Tversky.
Cambridge: Cambridge University Press.
Kail, P. J. E. (2001). “Hutcheson’s Moral Sense: Skepticism, Realism, and Secondary
Qualities,” History of Philosophy Quarterly 18: 57–77.
Kames, Lord (Henry Home) (2005). Essays on Morality and Principles of Natural
Religion, 3rd ed. Indianapolis, IN: Liberty Fund. Originally published in 1776.
Kant, Immanuel (1904). Educational Theory of Immanuel Kant, trans. and ed.
Edward Franklin Buchner. Philadelphia: J. B. Lippincott Company.
Kant, Immanuel (1996a). “Groundwork of the Metaphysics of Morals,” in Practical
Philosophy, trans. and ed. Mary J. Gregor. Cambridge: Cambridge University
Press. References are to page numbers of the Preussische Akademie edition.
Originally published in 1785.
Kant, Immanuel (1996b). “Critique of Practical Reason,” in Practical Philosophy,
trans. and ed. Mary J. Gregor. Cambridge: Cambridge University Press. Page
references are to page numbers of the Preussische Akademie edition.
Kant, Immanuel (1996c). “Metaphysics of Morals,” in Practical Philosophy, trans.
and ed. Mary J. Gregor. Cambridge: Cambridge University Press. Page refer-
ences are to page numbers of the Preussische Akademie edition.

Published online by Cambridge University Press

 354 works cited

Kant, Immanuel (1998). Critique of Pure Reason, trans. and ed. Paul Guyer and
Allen Wood. Cambridge: Cambridge University Press. Originally published
in 1781. Page references are to page numbers of the Preussische Akademie
edition.
Kant, Immanuel (2007). Anthropology, History, and Education, ed. Robert Louden
and Günther Zöller. Cambridge: Cambridge University Press. Page references
are to page numbers of the Preussische Akademie edition.
Kant, Immanuel (2009). Idea for a University History with a Cosmopolitan
Aim (1784), trans. Allen Wood, eds. Amelie O. Rorty and James Schmidt.
Cambridge: Cambridge University Press.
Kant, Immanuel (2013a). “Of the Different Human Races,” in Kant and the Concept
of Race: Eighteenth-Century Writings, trans. and ed. Jon M. Mikkelsen. Albany:
State University of New York Press.
Kant, Immanuel (2013b). “Determination of the Concept of Human Race,” in Kant
and the Concept of Race: Eighteenth-Century Writings, trans. and ed. Jon M.
Mikkelsen. Albany: State University of New York Press.
Kant, Immanuel (2018). Religion within the Boundaries of Mere Reason, eds.
Allen W. Wood and George Di Giovanni, foreword Robert Merrihew Adams.
Cambridge: Cambridge University Press.
Kavka, Gregory (1986). Hobbesian Moral and Political Theory. Princeton:
Princeton University Press.
Kavka, Gregory (1995). “The Rationality of Rule-Following: Hobbes’s Dispute
with the Foole,” Law and Philosophy 4: 5–34.
King, Peter (1830). The Life of John Locke, with Extracts from His Correspondence,
Journals, and Common-Place Books, 2 vols. London: Henry Colburn & Richard
Bentley.
Kirk, Linda (1987). Richard Cumberland and Natural Law: Secularisation of
Thought in Seventeenth-Century England. Cambridge: James Clarke.
Köhler, Wolfgang (1938). The Place of Value in a World of Facts. New York:
Liverlight Publishing Corp.
Korsgaard, Christine (1996a). The Sources of Normativity. Cambridge: Cambridge
University Press.
Korsgaard, Christine (1996b). Creating the Kingdom of Ends. Cambridge:
Cambridge University Press.
Korsgaard, Christine (1996c). “Morality as Freedom,” in Korsgaard 1996b.
Kosch, Michelle (2006). Freedom and Reason in Kant, Schelling, and Kierkegaard.
Oxford: Clarendon Press.
Kosch, Michelle (2018). Fichte’s Ethics. Oxford: Oxford University Press.
Kripke, Saul (1981). Naming and Necessity. Malden, MA: Blackwell Publishing.
Langton, Rae (1992). “Duty and Desolation,” Philosophy 67: 481–505.
Law, Edmund (1824). “A Defence of Mr. Locke’s Opinion Concerning Personal
Identity,” in Locke (1824), III, 179–198.
LeBuffe, Michael (2003). “Hobbes on the Origin of Obligation,” British Journal of
the History of Philosophy 11: 15–39.

Published online by Cambridge University Press

 works cited 355

Leibniz, Gottfied Wilhelm (1969). Gottfried Wilhelm Leibniz: Philosophical Papers

and Letters, 2nd ed., ed. and trans. L. E. Loemker. Dordrecht: Reidel.
Leibniz, Gottfried Wilhelm (1985a). Textes inédits, ed. Gaston Grua. New York:
Garland.
Leibniz, Gottfried Wilhelm (1985b). Theodicy, trans. E. M. Huggard. La Salle, IL:
Open Court.
Leibniz, Gottfried Wilhelm (1988). Political Writings, ed. Patrick Riley. Cambridge:
Cambridge University Press.
Leisinger, Matthew A. (2019). “The Inner Work of Liberty: Cudworth on Desire
and Attention,” International Journal of Philosophical Studies 27: 649–667.
Leisinger, Matthew A. (2021). “Cudworth on Freewill,” Philosophers’ Imprint, https://
quod.lib.umich.edu/p/phimp/3521354.0021.001/1/--cudworth-on-freewill?
​rgn=full+text;view=image
Leiter, Brian (2015). Nietzsche on Morality, 2nd ed. New York: Routledge.
Lerner, Adam (unpublished). “Empathy as Evidence.”
Lipscomb, Benjamin J. B. (2021). The Women Are Up to Something. Oxford: Oxford
University Press.
Lloyd, S. A. (2009). Morality in the Philosophy of Thomas Hobbes: Cases in the Law
of Nature. Cambridge: Cambridge University Press.
Locke, John (1824). The Works of John Locke, 12th ed., 9 vols. London: C. and J.
Rivington.
Locke, John (1975). An Essay Concerning Human Understanding, ed. Peter H.
Nidditch. Oxford: Oxford University Press. First edition, 1690; final, fourth
edition, 1700,
Locke, John (1988). Two Treatises of Government, ed. Peter Laslett. Cambridge:
Cambridge University Press. Originally published in 1689.
Locke, John (1990). Drafts for the Essay Concerning Human Understanding and
Other Philosophical Writings, eds. Peter H. Nidditch and G. A. J. Rogers.
Oxford: Oxford University Press.
Locke, John (1997). Political Essays, ed. Mark Goldie. Cambridge: Cambridge
University Press.
Locke, John (2002). Essays on the Law of Nature: The Latin Text with a Translation,
Introduction and Notes, Together with Transcripts of Locke’s Shorthand in His
Journal for 1676, ed. W. von Leyden. Oxford: Oxford University Press.
Løgstrup, K. E. (2007). Beyond the Ethical Demand, trans. Susan Dew and Heidi
Flegal, intro. Kees van Kooten Niekerk. Notre Dame, IN: University of Notre
Dame Press.
Løgstrup, K. E. (2020). The Ethical Demand, trans., intro., notes Bjørn Rabjerg and
Robert Stern. Oxford: Oxford University Press.
Long, A. A. (1986). Hellenistic Philosophy: Stoics, Epicureans, Sceptics. 2nd ed.
Berkeley and Los Angeles: University of California Press.
Lovejoy, Arthur O. (1908). “Kant and the English Platonists,” in Essays Philosophical
and Psychological in Honor of Williams James, ed. members of the Columbia
University Philosophy Department. New York: Longmans, Green.

Published online by Cambridge University Press

 356 works cited

Lu-Adler, Huaping (2022a). “Kant and Slavery: Or Why He Never Became a

Radical Egalitarian,” Critical Philosophy of Race 10.2: 263–294.
Lu-Adler, Huaping (2022b). “Kant on Lazy Savagery: Racialized,” Journal of the
History of Philosophy 60: 253–275.
MacIntyre, Alasdair (1981). After Virtue: A Study in Moral Theory. Notre Dame,
IN: University of Notre Dame Press.
Mackie, J. L. (1977). Ethics: Inventing Right and Wrong. New York: Penguin Books.
Mandeville, Bernard (1988). The Fable of the Bees or Private Vices, Publick Benefits,
2 vols. with a commentary critical, historical, and explanatory by F. B. Kaye.
Indianapolis, IN: Liberty Fund, (1988). Originally published in 1714.
Martineau, James (1901). Types of Ethical Theory, 2 vols. Oxford: Oxford University
Press.
Mautner, Thomas (1999). “Pufendorf and the Correlativity Theory of Rights,” in
Grotius, Pufendorf and Modern Natural Law, ed. Knud Haakonssen. Ashgate
Dartmouth: Aldershot, pp. 159–181.
McIntyre, Jane (1994). “Hume: Second Newton of the Moral Sciences,” Hume
Studies 20: 3–18.
Mill, John Stuart (1924). Autobiography. London: Oxford University Press.
Mill, John Stuart (2002). Utilitarianism, ed. George Sher. Indianapolis, IN: Hackett
Publishing Co., Inc. References to chapter and paragraph numbers.
Miller, Fred D., Jr. (1995). Nature, Justice, and Rights in Aristotle’s Politics. Oxford:
Clarendon Press.
Mills, Charles W. (1997). The Racial Contract. Ithaca, NY: Cornell University
Press.
Mills, Charles W. (2005a). “Ideal Theory as Ideology,” Hypatia 20: 165–184.
Mills, Charles W. (2005b). “Kant’s Üntermenschen,” in Race and Racism in
Modern Philosophy, ed. Andrew Walls. Ithaca, NY: Cornell University Press.
Mills, Charles W. (2017). “Black Radical Kantianism,” Res Philosophia 95.1: 1–33.
Moore, G. E. (1993). Principia Ethica, rev. ed. with the preface to the (projected)
2nd ed. and other papers, ed. with an intro. Thomas Baldwin. Cambridge:
Cambridge University Press.
Moran, R. (2005). “Getting Told and Being Believed,” Philosophers' Imprint 5: 1–29.
Muller, Jerry Z. (1995). Adam Smith in His Time and Ours: Designing the Decent
Society. Princeton: Princeton University Press.
Murdoch, Iris (2001). The Sovereignty of the Good. London and New York:
Routledge.
Nadler, Steven (2006). Spinoza’s Ethics: An Introduction. Cambridge: Cambridge
University Press.
Nadler, Steven (2018). Spinoza: A Life, 2nd ed. Cambridge: Cambridge University
Press.
Nagel, Thomas (2022). “What Is Rude?” London Review of Books. www.lrb.co.uk/
the-paper/v44/n03/thomas-nagel/what-is-rude
Neuhouser, Frederick (2008). Roussesau’s Theodicy of Self-Love: Evil, Rationality,
and the Drive for Recognition. Oxford: Oxford University Press.

Published online by Cambridge University Press

 works cited 357

Nietzsche, Friedrich (2007). On the Genealogy of Morality, rev. student ed., ed.
Keith Ansell-Pearson, trans. Carol Diethe. Cambridge: Cambridge University
Press.
Nisbett, Richard (2018). Culture of Honor. New York: Routlege.
Norton, David Fate (1982). David Hume: Common-Sense Moralist, Sceptical
Metaphysician. Princeton: Princeton University Press.
Nozick, Robert (1974). Anarchy, State, and Utopia. New York: Basic Books.
Oakley, Francis (2005). Natural Law, Laws of Nature, Natural Rights: Continuity
and Discontinuity in the History of Ideas. New York: Bloomsbury Academic.
OED (2021). Oxford English Dictionary. Oxford: Oxford University Press.
www-oed-com
Olson, Mancur (1971). The Logic of Collective Action. Cambridge, MA: Harvard
University Press.
Parfit, Derek (1986). Reasons and Persons. Oxford: Oxford University Press.
Parfit, Derek (2011). On What Matters, 2 vol. Oxford: Oxford University Press.
Pashukanis, Evgeny (2003). The General Theory of Law and Marxism, intro. by
Dragan Milovanovic. New Brunswick and London: Transaction Publishers.
Passmore, John (1951). Ralph Cudworth: An Interpretation. Cambridge: Cambridge
University Press.
Passmore, John (1968). Hume’s Intentions, 2nd rev. ed. New York: Basic Books.
Paul, Joanne (2020). Counsel and Command in Early Modern English Thought.
Cambridge: Cambridge University Press.
Pettit, Philip (1993). The Common Mind: An Essay on Society, Psychology, and
Politics. Oxford: Oxford University Press.
Pettit, Philip (2009). Made with Words: Hobbes on Language, Mind, and Politics.
Princeton: Princeton University Press.
Pettit, Philip and Michael Smith (1990). “Backgrounding Desire,” The Philosophical
Review 99: 565–592.
Pettit, Philip and Michael Smith (1996). “Freedom in Belief and Desire,” Journal of
Philosophy 93: 429–449.
Pink, Thomas (2007). “Normativity and Reason,” Journal of Moral Philosophy 4:
406–431.
Plato (2001). The Republic, ed. G. R. F. Ferrari, trans. Tom Griffin. Cambridge:
Cambridge University Press.
Platts, Mark (1979). Ways of Meaning. London: Routledge & Kegan Paul.
Price, Richard (1974). A Review of the Principal Questions in Morals (1758), et. D. D.
Raphael. Oxford: Clarendon Press.
Prichard, H. A. (2002). “Does Moral Philosophy Rest on a Mistake?,” in Moral
Writings, ed. Jim MacAdam. Oxford: Oxford University Press.
Pufendorf, Samuel (1672). De Jure Naturae et Gentium. Lund: Adam Junghans.
Pufendorf, Samuel (1710). On the Law of Nature and Nations (De Jure Naturae et
Gentium), trans. Basil Kennett. Oxford: A. and J. Churchill.
Pufendorf, Samuel (1718). Les devoirs du citoien, trans. Jean Barbeyrac, 4th ed.
Amsterdam: Chez P. de Coup.

Published online by Cambridge University Press

 358 works cited

Pufendorf, Samuel (1934). On the Law of Nature and Nations, trans. C. H. Oldfather
and W. A. Oldfather. Oxford: Clarendon Press. Originally published in 1672.
Pufendorf, Samuel (1991). On the Duty of Man and Citizen According to Natural
Law. trans. Michael Silverthorne, ed. James Tulley. Cambridge: Cambridge
University Press.
Pufendorf, Samuel (2009). Two Books of the Elements of Universal Jurisprudence,
ed. Thomas Behne and trans. W. A. Oldfather, with revisions by Thomas
Behne. Indianapolis, IN: Liberty Fund. Originally published in 1660.
Railton, Peter (1986). “Moral Realism,” The Philosophical Review 95: 163–207.
Raphael, D. D. (1969). British Moralists: 1650–1800. Oxford: Clarendon Press.
Rawls, John (1955). “Two Concepts of Rules,” The Philosophical Review 64: 3–32.
Rawls, John (1971). A Theory of Justice. Cambridge, MA: Harvard University Press.
Rawls, John (1980). “Kantian Constructivism in Moral Theory,” The Journal of
Philosophy 77: 515–572.
Rawls, John (1993). Political Liberalism. New York: Columbia University Press.
Rawls, John (2000). Lectures on the History of Moral Philosophy, ed. Barbara
Herman. Cambridge, MA: Harvard University Press.
Rawls, John (2007). Lectures on the History of Political Philosophy, ed. Samuel
Freeman. Cambridge, MA: Harvard University Press.
Raz, Joseph (1972). “Voluntary Obligations and Normative Powers,” Proceedings of
the Aristotelian Society 46: 79-101.
Raz, Joseph (1977). “Promises and Obligations,” in Law, Morality, and Society:
Essays in Honor of H. L. A. Hart, eds. P. M. S. Hacker and Joseph Raz. Oxford:
Oxford University Press, pp. 210–228.
Raz, Joseph (1986). The Morality of Freedom. Oxford: Clarendon Press.
Raz, Joseph (2002a). Practical Reason and Norms. Oxford: Oxford University
Press.
Raz, Joseph (2002b). The Authority of Law: Essays on Law and Morality. Oxford:
Oxford University Press.
Reid, Thomas (1969). Essays on the Active Powers of the Human Mind. Cambridge,
MA; MIT Press. Originally published in 1788.
Reid, Thomas (2007). Thomas Reid on Practical Ethics: Lectures and Papers on
Natural Religion, Self-Government, Natural Jurisprudence, and the Law of
Nations, ed. Knud Haakonssen. Edinburgh: Edinburgh University Press.
Ripstein, Arthur (2009). Force and Freedom: Kant’s Legal and Political Philosophy.
Cambridge, MA: Harvard University Press.
Roberts, John R. (unpublished). “Virtue Has No Master: Cudworth’s Metaphysics
of Free Will.”
Roberts, Robert (2003). Emotions: An Essay in Aid of Moral Psychology. Cambridge:
Cambridge University Press.
Rosenkoetter, Timothy (2003). “A Semantic Approach to Kant’s Practical Philosophy,”
presented at the APA Central Division 2003 meetings in Cleveland, Ohio.
Ross, W. D. (2003). The Right and the Good, 2nd ed., ed. Philip Stratton-Lake.
Oxford: Clarendon Press.

Published online by Cambridge University Press

 works cited 359

Rousseau, Jean Jacques (2003). The Social Contract and Other Later Political
Writings, ed. Victor Gourevitch. Cambridge: Cambridge University Press.
Rousseau, Jean Jacques (2019). The Discourses and Other Early Political Writings,
ed. and trans. Victor Gourevitch. Cambridge: Cambridge University Press.
Sartre, Jean-Paul (1992). Being and Nothingness, trans. Hazel Barnes. New York:
Washington Square Press.
Scanlon, T. M. (1998). What We Owe to Each Other. Cambridge, MA: Harvard
University Press.
Scanlon, T. M. (2014). Being Realistic about Reasons. Oxford: Oxford University
Press.
Schapiro, Tamar (2021). Feeling Like It: A Theory of Inclination and Will. Oxford:
Oxford University Press.
Schiffman, Kelley (2015). Rethinking Reason and Right: Moral Rationalism in
Eighteenth-Century Britain. Ph.D. Diss., Yale University.
Schliesser, Eric (2019). “Sophie de Grouchy, Adam Smith, and the Politics of
Sympathy,” in Feminist History of Philosophy: The Recovery and Evaluation
of Women’s Philosophical Thought, eds. Eileen O’Neill and Marcy P. Lascano.
Springer: Springer Nature, Switzerland.
Schneewind, J. B. (1977). Sidgwick’s Ethics and Victorian Moral Philosophy.
Oxford: Oxford University Press, 1977.
Schneewind, J. B. (1990). Moral Philosophy from Montaigne to Kant, v. i.
Cambridge: Cambridge University Press.
Schneewind, J. B. (1998). The Invention of Autonomy. Cambridge: Cambridge
University Press.
Schroeder, Mark (2005a). “Cudworth and Normative Explanations,” Journal of
Ethics & Social Philosophy 1.3: 1–27.
Schroeder, Mark (2005b). “The Hypothetical Imperative?” Australasian Journal of
Philosophy 83: 357–372.
Selby-Bigge, L. A. (1964). British Moralists, Being Selections from Writers
Principally of the Eighteenth Century, 2 vols. Reprinted in one volume with a
new introduction by Bernard H. Baumrin. Indianapolis, IN: Bobbs-Merrill.
Sellars, Wilfrid (1956). “Empiricism and the Philosophy of Mind,” in Minnesota
Studies in the Philosophy of Science 1: 253–329.
Sensen, Oliver (2011). Kant on Human Dignity. Berlin: DeGruyter.
Shafer-Landau, Russ (2003). Moral Realism: A Defense. New York: Oxford
University Press.
Shaftesbury, Anthony Ashley Cooper, 3rd Earl of (1698). “Preface,” in Select
Sermons of Dr. Whichcot, ed. Anthony Ashley Cooper, 3rd Earl of Shaftesbury.
London: A. & J. Churchill.
Shaftesbury, Anthony Ashley Cooper, 3rd Earl of (1900). The Life, Unpublished
Letters, and Philosophical Regimen of Anthony, Earl of Shaftesbury, ed.
Benjamin Rand. London: Swan Sonnenschein.
Shaftesbury, Anthony Ashley Cooper, 3rd Earl of (2001). Characteristics of Men,
Manners, Opinions, Times, 3 vols., ed. Douglas J. Den Uyl. Indianapolis, IN:

Published online by Cambridge University Press

 360 works cited

Liberty Fund. (Based on 1732 edition; first edition, 1711.) References are to vol-
ume and page number from the 1732 edition.
Sharp, Franc Chapman (1912). “The Ethical System of Richard Cumberland and
Its Place in the History of British Ethics,” Mind 21: 371–398.
Shelby, Tommie (2005). We Who Are Dark: The Philosophical Foundations of Black
Solidarity. Cambridge, MA: Harvard University Press.
Shelby, Tommie (2016). Dark Ghettos: Injustice, Dissent, and Reform. Cambridge,
MA: Harvard University Press.
Sheridan, Patricia (2007). “Pirates, Kings, and Reasons to Act: Moral Motivation
and the Role of Sanctions in Locke’s Moral Theory,” Canadian Journal of
Philosophy 37: 35–48.
Sidgwick, Henry (1964). Outlines of the History of Ethics for English Readers, 6th
ed. Boston: Beacon Press.
Sidgwick, Henry (1967). The Methods of Ethics, 7th ed. London: Macmillan.
Silk, Alex (2015). “Nietzschean Constructivism: Ethics and Metaethics for All and
None,” Inquiry 58: 244–280.
Skorupski, John (1999). Ethical Explorations. Oxford: Oxford University Press.
Slote, Michael (1992). From Morality to Virtue. Oxford: Oxford University Press.
Smith, Adam (1976). An Inquiry into the Nature and Causes of the Wealth of
Nations, eds. R. H. Campbell and A. S. Skinner. Oxford: Clarendon Press.
Originally published in 1776.
Smith, Adam (1982). The Theory of Moral Sentiments, eds. D. D. Raphael and A. L.
MacFie. Indianapolis: LibertyClassics. Originally published in 1759.
Smith, Michael (1989). “Dispositional Theories of Value,” Proceedings of the
Aristotelian Society 63: 89–112.
Smith, Michael (1994). The Moral Problem. Oxford: Blackwell.
Sorabji, Richard (2014). Moral Conscience through the Ages: Fifth Century BCE to
the Present. Chicago: University of Chicago Press.
Spinoza, Benedict (1985a). Short Treatise on God, Man, and His Well-Being, in
Collected Works, v. i, ed. Edwin Curley. Princeton: Princeton University Press.
Spinoza, Benedict (1985b). “Ethics,” in Collected Works, v. i, ed. Edwin Curley.
Princeton: Princeton University Press.
Spinoza, Benedict (2016). “Theological-Political Treatise,” in Collected Works, v. ii,
ed. Edwin Curley. Princeton: Princeton University Press.
Spinoza, Benedict (2007). Theological-Political Treatise, ed. with intro. Jonathan
Israel. Cambridge: Cambridge University Press.
Spivak, Gayatri Chakravorty (1999). A Critique of Postcolonial Reason: Toward A
History of the Vanishing Present. Cambridge, MA: Harvard University Press.
Stevenson, Charles (1937). “The Emotive Meaning of Ethical Terms,” Mind 46:
14–31.
Strawson, P. F. (1968). “Freedom and Resentment,” in Studies in the Philosophy of
Thought and Action. London: Oxford University Press.
Street, Sharon (2008). “Constructivism about Reasons,” Oxford Studies in
Metaethics 3: 207–245.

Published online by Cambridge University Press

 works cited 361

Suarez, Francisco (1944). A Treatise on Laws and God the Lawgiver, in Selections
from Three Works of Francisco Suarez, S.J., v. 2, trans. Gwladys L. Williams,
Ammi Brown, and John Waldron, rev. Henry Davis, S.J., intro. James Brown
Scott. Oxford: Clarendon Press, 1944. Originally published in 1612.
Thompson, Michael (2004). “What Is It to Wrong Someone? A Puzzle about
Justice,” in Reason and Value: Themes from the Philosophy of Joseph Raz, eds.
R. Jay Wallace, Philip Pettit, Samuel Scheffler, and Michael Smith. Oxford:
Oxford University Press.
Tierney, Brian (1997). The Idea of Natural Rights. Grand Rapids, MI: Wm. B.
Eerdmans Publishing Co.
Tomasello, Michael (2018). A Natural History of Human Morality. Cambridge,
MA: Harvard University Press.
Trianosky, Gregory (1978). “On the Obligation to Be Virtuous: Shaftesbury and
the Question, Why Be Moral?” Journal of the History of Philosophy 16: 289–300.
Tschentscher, Axel (2012). The Basic Law (Grundgesetz) 2012: The Constitution
of the Federal Republic of Germany. http://papers.ssrn.com/sol3/papers​
.cfm?abstract_id=1501131
Tuck, Richard (1981). Natural Rights Theories: Their Origin and Development.
Cambridge: Cambridge University Press.
Tuck, Richard (1993). Philosophy and Government: 1572–1651. Cambridge:
Cambridge University Press.
Tuck, Richard (2001). The Rights of War and Peace: Political Thought and the
International Order from Grotius to Kant. Oxford: Oxford University Press.
Tucker, Robert C. (1969). The Marxian Revolutionary Idea. New York: W. W.
Norton & Company.
Varden, Helga (2020). Sex, Love, and Gender: A Kantian Theory. Oxford: Oxford
University Press.
Villey, Michel (1982). Philosophie du Droit: 1. Definitions et fins du droit, 3rd ed.
Paris: Galloz.
Villey, Michel (1984). Philosophie du Droit: 2. Les moyens du droit, 2nd ed. Paris:
Galloz.
Voitle, Robert (1984). The Third Earl of Shaftesbury. Baton Rouge: Louisiana State
University Press.
Wahrman, Dror (2006). The Making of the Modern Self: Identity and Culture in
Eighteenth-Century England. New Haven, CT: Yale University Press.
Watson, Gary (1975). “Free Agency,” Journal of Philosophy 72: 205–220.
Watson, Gary (1996). “Two Faces of Responsibility,” Philosophical Topics 24:
227–248.
Watson, Gary (2009). “Promises, Reasons, and Normative Powers,” in Reasons for
Action, eds. David Sobel and Steven Wall. Cambridge: Cambridge University
Press, pp. 155–178.
Williams, Bernard (1981a). Moral Luck. Cambridge: Cambridge University Press.
Williams, Bernard (1981b). “Internal and External Reasons,” in Moral Luck.
Cambridge: Cambridge University Press.

Published online by Cambridge University Press

 362 works cited

Williams, Bernard (1985). Ethics and the Limits of Philosophy. Cambridge, MA:
Harvard University Press.
Williams, Bernard (1995). “Internal Reasons and the Obscurity of Blame,” in
Making Sense of Humanity. Cambridge: Cambridge University Press.
Winkler, Kenneth P. (1985). “Hutcheson’s Alleged Moral Realism,” Journal of the
History of Philosophy 23: 179–194.
Wisdom, John (1948). “Things and Persons,” Proceedings of the Aristotelian Society
22: 202–215.
Wolf, Susan (1987). “Sanity and Moral Responsibility,” in Responsibility, Character,
and the Emotions: New Essays in Moral Psychology, ed. F. D. Schoeman.
Cambridge: Cambridge University Press.
Wolff, Robert Paul (1970). In Defense of Anarchism. New York: Harper & Row.
Wollaston, William (1978). The Religion of Nature Delineated. New York: Garland
Publishing Co., Inc. Originally published in 1722.
Wood, Allen W. (1990). Hegel’s Ethical Thought. Cambridge: Cambridge
University Press.
Wood, Allen W. (1999). Kant’s Ethical Thought. Cambridge: Cambridge University
Press.
Wood, Allen W. (2016). Fichte’s Ethical Thought. Oxford: Oxford University Press.
Woolhouse, Richard (2007). Locke: A Biography. Cambridge: Cambridge
University Press.
Worsnip, Alex (2015). “Hobbes and Normative Egoism,” Archiv für Geschichte der
Philosophie 97: 481–512.
Yaffe, Gideon (2000). Liberty Worth the Name: Locke on Free Agency. Princeton:
Princeton University Press.
Young, Julian (2010). Friedrich Nietzsche: A Philosophical Biography. Cambridge:
Cambridge University Press.
Yovel, Yirmiyahu (1992). Spinoza and Other Heretics, Volume 1: The Marrano of
Reason. Princeton: Princeton University Press.

Published online by Cambridge University Press

 INDEX

Albee, Ernest, 100 Butler, Joseph, 248–249

Allais, Lucy, 12, 341 Clarke, Samuel, 237–239, 247–250,
Allison, Henry, 280 287
Anderson, Elizabeth, 14 Hobbes, Thomas, 287
Anscombe, Elizabeth Hutcheson, Francis, 248–249,
challenge, xi, 1–13, 16–20, 24, 35, 38, 287
101 Price, Richard, 237–239, 287
consequentialism, critique of, 1 Suárez, Francisco, 250
law conception, 1, 17–20, 25, 29 moral sense theory, rejection of, 252
legislation, divine, 2 morality, deontic conception of,
Aquinas, Thomas, 11, 16, 17, 26, 64, 102 238–239, 256
Bonum honestum, 20 motivation, 249–250
eudaimonism, 26 necessity, 249
law obligation
eternal, 26–27, 92 as normative relation, 242
natural, 16–20, 26–27, 155 internal, 249–250
Naturalism, Aristotelian, 17, 26 practical and theoretical reasoning,
perfectionism, 114 parallel between, 249–250
Aristotle, 4, 5, 11, 15, 23, 36, 167, 204 rationalism, 113, 229, 237–238,
justice, distributive, 29 248–250
naturalism, 7, 11, 12, 16, 17, 19, 25, 26, reason as source of moral
28, 30, 40, 92, 97, 101, 102, 122, motivation, 238, 248–250
147, 151, 152, 163, 168, 189 “the reason of the thing and the
perfectionism, 114–115 rectitude of the action,” 283,
wisdom, practical, 108 287
virtue as rational determination of
Baier, Annette, 9 the mind, 248
Balguy, John Barbeyrac, Jean, 27, 34, 87
accountability, 248, 287 on Grotius, Hugo, 15, 22–24, 32, 35
agency, 248–249 on Leibniz, Gottfried Wilhelm,
akrasia, 249 163–164
approbation, 249 on Pufendorf, Samuel von, 87
benevolence as insufficient and Bentham, Jeremy, 162, 205
unnecessary for moral in relation to
goodness, 248 Hume, David, 205
fitness, 249–250 Hutcheson, Francis, 177
in relation to Bratman, Michael, 260

363

https://doi.org/10.1017/9781139025065.011 Published online by Cambridge University Press

 364 index

Butler, Joseph Leibniz, Gottfried Wilhelm, 166

accountability, 169–172, 194–197, Locke, John, 100, 169–170
200, 239, 246, 256 Pufendorf, Samuel von, 194
action in circumstances as object of Shaftesbury (Anthony Ashley
reflective judgment, 190 Cooper, Third Earl of), 142,
agency, 168–172, 185–198, 200, 222, 147, 151, 165–166, 169–170, 185,
241, 253, 256 189, 192, 195, 199, 222, 238
self-reflective, 165 judgment
Anglicanism, 184 deontic moral, 188, 191–192
approbation, reflex, 142, 170 normative practical, 171–172,
authority and mere power, distinction 186–197
between, 185, 195, 209 morality, deontic conception of,
autonomy, 169–170, 187–188, 206, 168–172
221–222 nature, 168, 187–189, 192–193
benevolence, 185, 186 obligation, 172
conscience, 147, 165–166, 168–172, passions, private and public,
183–199, 200, 241, 248, 249, distinction between, 185–186
255, 257, 259, 270 proto-utilitarianism, rejection of,
and self-love, dualism between, 166, 199, 206
168–169 psyche
authority of, 142, 151, 168–172, constitution of, 169, 185–188,
184, 186–188, 192–197, 205, 194–195
222, 238, 240, 246, 253, 259, economy, 169, 185–188, 191–193
261–263 psychology, moral, 165–166, 185–188,
teleological argument for, 200
192–194 reason, 127
theological argument for, 192–194 reflection, principle of, 165, 171–172,
transcendental argument for, 186–197, 222, 249, 255, 263,
193–197 270
in relation to self-love. See Butler, Rolls Chapel, 184
Joseph:Self-love:In relation self-deception, 188
to conscience self-love, 168–169, 184–193, 199–200,
“cool hour” passage, 56, 189 261–262
egoism, rejection of, 185, 199–200 in relation to conscience, 186–188,
eudaimonism, rejection of, 172, 197
189–190, 192 self-reflection, 190–192
“everything is what it is…,” 172, 185 sermons, overview, 183–184
God as designer, 193–194 utilitarianism, indirect, 198, 200
good and right, 197–199 vice as misery, 169
happiness, 186, 189 virtue
hierarchy, normative, 186–188, 255 as conscientious self-guidance,
in relation to 165, 166, 206, 241
Aristotle, 189 reduction to benevolence,
Clarke, Samuel, 196 rejection of, 166, 170–171,
Cudworth, Ralph, 169–170 205, 257
Cumberland, Richard, 170–171 watch analogy, 194
Hutcheson, Francis, 165–166,
169–171, 185–192, 195, 197–199, Cassirer, Ernst, 140
205–206, 221–222, 238 Cicero, Marcus Tullius, 11, 156–157

https://doi.org/10.1017/9781139025065.011 Published online by Cambridge University Press

 index 365

Clarke, Samuel relations, natural, 244–245

accountability, 246–247, 273 supervenience, 245
agency, 10, 243, 245–247, 256, 273 voluntarism, rejection of, 243, 248
benevolence, universal, 246 will, 247, 249
Boyle Lectures, 196, 243 Cudworth, Ralph, 10
circumstances, 243–246 accountability, 112, 128–129, 135–139,
conscience, 196 145–146, 148, 169–170
authority of, 246–247 agency, 112, 148, 158, 169–170
differences of things, 244–246 anti-positivism, 111, See also
equity, 243, 246 Voluntarism, rejection of
fitness, 208, 241, 243–246 atheism, rejection of, 128
as founded in the nature of things, autexousy, 136–139, 261
243, 244 autonomy, 169–170
as synonymous with “right” and believing and judging, distinction
“reasonable,” 243 between, 136–137
golden rule, 243, 287 benevolence, 134–135
good, universal, 243 blame, 128, 135–136
in relation to charity, 112, See Cudworth,
Balguy, John, 237–239, 247–250, Ralph:Love
287 conscience, 139
Butler, Joseph, 196, 243, 246–247 determination, practical, 132–139
Cudworth, Ralph, 243, 248 eudaimonism, 169–170
Fordyce, David, 240–246 fault and sin, 137
Hobbes, Thomas, 287 God, 117
Hutcheson, Francis, 245–246 as ideal of moral agency, 133–135,
Leibniz, Gottfried Wilhelm, 157, 139
248 goodness, natural and positive,
Price, Richard, 237–239, 273, 287 distinction between, 131
Reid, Thomas, 273 government, divine, 128
Shaftesbury (Anthony Ashley guilt, 128, 135–136
Cooper, Third Earl of), 142 honesty, 138, 139
Suárez, Francisco, 250 idealism
internalism, motivational, 247, 255 ethical, 131–134
judgment, 244–247 metaphysical, 130–131
love, 243, 246 in relation to
mathematics and morality, analogy Hobbes, Thomas, 129, 130,
between, 244 132–133, 136, 138, 152
morality, deontic conception of, Leibniz, Gottfried Wilhelm, 111
238–239, 246–247, 256 Locke, John, 112, 128–129, 131,
obligation, 196, 241, 244–246, 250 169–170
as “eternal reason of things,” 246 Plato, 130, 134
as normative relation, 242, 246 Plotinus, 130
rationalism, 113, 229, 237–238, 241, Pufendorf, Samuel von, 128–129,
243–248 131, 138
founder of British tradition, 243 Shaftesbury (Anthony Ashley
reason as the source of moral Cooper, Third Earl of), 111,
motivation, 238 151, 169–170
reasoning, practical and theoretical, Spinoza, Baruch, 111, 128
parallel between, 247, 249 internalism, autonomist, 133

https://doi.org/10.1017/9781139025065.011 Published online by Cambridge University Press

 366 index

Cudworth, Ralph (cont.) goods, natural, 106

judgment, practical, 112, 135–137 happiness, 102–104, 107–108
justice, 134–135 in relation to
knowledge, 130–131 Aquinas, Thomas, 102
life, “stories of,” “animal” and Aristotle, 101, 102, 104
“divine,” 137, 139 Grotius, Hugo, 87–90
“light” in which agents view options, Hobbes, Thomas, 88–90, 101–102,
137 104–105, 107, 108
love, 112, 133–135, 138, 139, 141, 146, 243 Hume, David, 90
metaphysics, 128, 135 Hutcheson, Francis, 90
mind, 130–135 Leibniz, Gottfried Wilhelm, 90
modifications of, 131–134 Locke, John, 88–90, 103–106, 109
of God, 131, 133–134 Pufendorf, Samuel von, 87–90,
morality, eternal and immutable 103–104, 109
character of, 111, 129–134, Shaftesbury (Anthony Ashley
208, 237, 252 Cooper, Third Earl of), 90
motivation. See determination, Suárez, Francisco, 146
practical instrumentalism, 105–108
nature, necessary, 137 internalism, 90
obligation, 112, 131–133, 135–139 law, natural, 88–89, 102–105, 108–111,
rationalism, 270 141, 146
Sui-Potestas. See Cudworth, love, 104, 105, 134
Ralph:autexousy morality as science, 87, 88
virtue, ethics of, 113 naturalism, 87–90, 101–110
voluntarism, rejection of, 42, 70, nonaggression, mutual, 105
111–112, 128–134, 152, 248 obligation, 89, 103–104
will, 260 as epiphenomenal, 108–110
as “last practical judgment,” peace, 105, 106
136–137 propositions, practical, 102–103
free, 112, 128, 135–139 punishments and rewards, 103–104
animal and moral, 112, 136–139, reason, practical, 107–109, 171
169 standard of, 107–108
Culverwell, Nathaniel, 92 reductionism, 87–89, 106, 108–110, 171
Cumberland, Richard right, the, 106
agency, 88, 89, 101–108 as epiphenomenal, 108–110
authority, divine, 103–104 sanctions. See punishments and
as epiphenomenal, 109 rewards
benevolence, 88, 101, 102–105, 171 utilitarianism, 88–89, 100, 106, 109
best end as the greatest good of all, voluntarism, theological, 90, 101,
103, 106–110, 171 108–109
effects, 102–103, 105–108 will, consequence-directed rational,
egoism, rejection of, 101 105–107
empiricism, 87–89, 101, 102, 104–105
envy and hatred, 105 Descartes, René, 120
eudaimonism, rejection of, 101 dualism, 114
good
as single normative standard, Fichte, Johann Gottlieb, 334
109–110 in relation to Reid, Thomas, 265
common, 88, 89, 103–108, 134, 177 summons, xii

https://doi.org/10.1017/9781139025065.011 Published online by Cambridge University Press

 index 367

Florentinus, 21, 22 in relation to

Fordyce, David Aquinas, Thomas, 19, 20, 24
accountability, 242, 246, 258 Aristotle, 23
affections, 240 Florentinus, 21, 22
agency, 239–242, 258 Seneca, 21
conscience Suárez, Francisco, 17–21, 23–25,
authority of, 240–242 292
conscientiousness and moral individualism, 40
goodness, connection ius, 20–25
between, 241, 258 law, moral, 34
empiricism, 239–241 law, natural, 11, 16, 19, 24, 29–30,
in relation to 36–38, 63–64, 111
Butler, Joseph, 240–242 liberty, 32
Clarke, Samuel, 240–246, 251 minimalism, 35, 36, 40
Hume, David, 240–241 morality as science, 34–36
Hutcheson, Francis, 239–241, “morality,” use of word, 87
258 Naturalism, Aristotelian, 11, 16, 25
Locke, John, 239, 258 necessity, moral, 155
Price, Richard, 239, 258 obligation, 24–25, 29, 32–38
Shaftesbury (Anthony Ashley perfect and imperfect, 62–63
Cooper, Third Earl of), 240 principles, general, 37, 38
morality, deontic conception of, punishment, 31, 33–34
239–242, 246, 258 recognition, mutual, 64, 91
motivation, 239–242 reparation, 32
obligation, 239–243, 245, 246, 258 restitution, 32
as indefinable, 239–241, 251 right, fountain of. See Grotius,
as normative relation, 242 Hugo:Sociability
reason, 239–241 rights
sense, moral, 239–242, 245, 258 claim, 31–33
senses as source of ideas, 239–241 natural, 24–26, 31–33, 35
sentimentalism, 239–246 perfect and imperfect, 16, 22–23,
“state of moral obligation,” 242 38, 62–63
“state of the case,” 241, 244 sociability, 19, 21–22, 24, 29, 34,
supervenience, 245 36–38, 40, 63–64, 73, 91, 265,
Fricker, Miranda, 14 268, 269, 274, 277, 278
sovereign, 31
Galileo, 45, 120 speech, 37
Grotius, Hugo superiority, right of, 21, 38, 64
accountability, 23, 31, 34–38, 91, 155 voluntarism, rejection of, 19, 24
autonomy, 35 war and peace, 42–43
Carneades, challenge of, 16, 25–31, Will, 31
36, 156–157, 163–164, 197, 279 Guyer, Paul, 288, 301, 306, 334
collective action problem, 27–28, 37
consentium gentium, 36, 40 Haakonssen, Knud, 15
dignity, 33 Hare, R.M., 162, 245
Dutch trading companies, 31 Hart, H.L.A., 31, 33, 50
equality, right of, 21, 38, 64 point of view, internal, 216
eudaimonism, 25 rules, external and internal aspects
golden rule, 36 of, distinction between, 219

https://doi.org/10.1017/9781139025065.011 Published online by Cambridge University Press

 368 index

Hegel, Georg Wilhelm Friedrich, 10, 12, and the good, 52–54
265, 334 metaethics, 49
Hobbes, Thomas obligation, natural, 49–50
absurdity and injury, 51 passions, 44, 45
accountability, 36, 55, 56 power, normative, 41–42, 44, 61, 215
attitudes, reactive (Strawsonian), 59–61 projectivism, 60, 61, 119–120,
color analogy, 45–48, 89, 120 See also Hobbes,
conscience, 58 Thomas:expressivism
consequence, 39, 43–44 reasonable, the, 54–56
contract and covenant, 41–44, reciprocity theorem, 49, 54
48–54, 61, 125, 215, 216, 219 reductionism, 44
violation of, 51–59 sanctions, divine, 141
creatures lacking conception of right scarcity, natural, 105
and wrong, 59 self-preservation, 41, 48, 49, 53, 55,
deliberation, 47–51, 60–61 59
desire and the good, 41, 43–49, 219 sense, 39
and the right, 48 slaves and servants, 57–58
metaethics of, 48, 59–61 sovereign, 49, 54–59, 125, 215, 216
egoism, 30 subjectivism, 44–45
empiricism, 39–40, 44, 59, 61 utility, expected, 52
Epicureanism, 89, 104 war and peace, 41–45
evil, 45, 46 War, English Civil, 40
expressivism, 45–48, 210 will, 47, 51, 60, 120, 136
fool, 25, 30, 56, 52–59, 279 Hohfeld, Wesley Newcomb, 65, 159
God, power of, 50 rights, claim, 32
Golden Rule, 36, 54, 59, 73, 87, 287 rights, liberty, 32, 33, 159
in relation to Hume, David
Grotius, Hugo, 30, 40–43 aesthetic and moral assessment,
Suárez, Francisco, 18, 146, 292 analogy between, 143, 221,
individualism, 40 230
instrumentalism, 49 approbation, 249, 264
judgmentalism, 60 as anatomist, 203
justice, 43 autonomy, 206
law, 54, 55 beauty, moral, 230
natural, 41, 44, 48, 52–59, 146 benevolence, 204–205, 212–215,
liberty, 41, 42, 49–51, 53 218
blameless, 33, 49, 55, 59, 75, 83 bigot, seditious, 205, 214
materialism, 39–40, 44, 51, 59, 61 causation, 252
minimalism, 40 collective action problems, 204
morality as science, 39–40, 43–44, color analogy, 69, 237
47, 51 consequentialism, 211
“morality,” use of word, 87 constructionism, social, 212
nature convention, 205, 213–221
right of, 41, 49, 53, 55, 75, 159 creatures, rational, 220–221, 232
state of, 41, 52, 53, 55, 57, 58, 105, desires, 201, 207
213 empiricism, 151, 241, 252, 259
nominalism, 39, 43, 130, 132 Enlightenment, Scottish, 201
obligation and the right, 40–44, 54, expressivism, 210
48–61, 68, 219, 239 fact, matters of, 209–210

https://doi.org/10.1017/9781139025065.011 Published online by Cambridge University Press

 index 369

free rider. See Hume, David:knave, motives as object of approbation,

sensible 214, 218, 222, 229
ideas, association of, 175, 204–205, naturalism, empiricist, 201, 203–204
209, 211–212 Newtonianism, 203–204, 211
ideas, relations of, 207, 209, 244 obligation, 89, 166, 171, 266–267
in relation to interested, 141, 181, See Hume,
Balguy, John, 237–238 David:Obligation:Natural
Butler, Joseph, 184, 190–191, moral, 181, 205, 213, 217–220, 232,
205–206, 214, 221–222, 229, 238 240
Clarke, Samuel, 208, 237–238, natural, 141, 181, 213, 217–220, 240
243–247 rule, 219–220
Cudworth, Ralph, 208 passion and reason, opposition of,
Cumberland, Richard, 89–90 rejection of, 206–208
Grotius, Hugo, 30 passions, calm and violent, 127
Hobbes, Thomas, 213–214, 216–217 performance, external, as lacking
Horace, 211 merit, 167, 213, 214, 220
Hutcheson, Francis, 166–167, perspective, observer’s, 190, 203, 218,
175, 177–179, 181–182, 201, 221–223, 230
203–206, 210–211, 213–214, philosophy as science, 139
218, 220–224, 229–230, 232, promising, 214–218, 266–267
237–241, 251, 254, 282 property, 214–219
Price, Richard, 237–238 rationalism, rejection of, 201,
Shaftesbury (Anthony Ashley 206–211
Cooper, Third Earl of), 113, reason as “slave of the passions,”
124, 140–144, 151, 172, 213, 238 127, 316
Smith, Adam, 201–203, 220–224, reason as motivationally inert, 166,
229–238 201, 206–210, 247, 255
Spinoza, Baruch, 115, 122, 124, 127 reasoning
“is”/“ought,” 201, 209–210, 244 demonstrative, 207
judgments, ethical, 201–202, probabilistic, 207, 209
208–213, 222, 229–231 resentment, 221
justice, 203–205 rowing example, 216
artificial, 167, 204–205, 213–221, seditious bigot, 37
232–233, 265 self-interest, 205, 213–214, 217–220
motive of, 218–220 self-love, 214–216
practices, rule-governed sense, moral, 113, 210, 259
cooperative, 205, 213–221 rejection of, 203–204, 210–211,
knave, sensible, 25, 30, 205, 217–218, 264
279 sentimentalism, 113, 143, 167, 190,
mind, secret springs and principles 201, 206–212, 214, 218–220,
of, 203 222–223, 229, 230, 237–241,
“moral problem,” 201 243–246, 263–264, 274
moral virtues and natural abilities, subjectivism, 210
distinction between as purely sympathy, 175, 204–205, 209, 211–
verbal, 167, 177, 203–204, 211, 212, 220–224, 229–232, 264
236 as emotional contagion, 211, 224
morality, deontic, rejection of, 167 utilitarianism, 205, 211
motivation, theory of, 60, 206–212, utility as foundation of justice, 203,
220 205

https://doi.org/10.1017/9781139025065.011 Published online by Cambridge University Press

 370 index

Hume, David (cont.) expressivism, 210

virtue God, 170
“in rags,” 212 goodness, 173–181
amiable and awful, distinction moral, 167–168
between, 234 moral and natural, distinction
artificial and natural, distinction between, 167–168, 173–176,
between, 213 179–181
ethics of, 167, 177, 204–205, natural, 166
213–215, 251, 254 public, 122, 168, 178
will, 206–210, 220 greatest happiness principle, 166, 177,
Hursthouse, Rosalind, 9 179–180, 204, 205, 213, 232
Huseyinzadegan, Dilek, 12, 341 in relation to
Hutcheson, Francis, 9 Aristotle, 167–168, 177
agency, 248 Balguy, John, 237–238
appetitus rationalis. See Hutcheson, Butler, Joseph, 165–166, 169–171,
Francis:will 180, 183–192, 195, 197–199,
apprehension, rational, 179–180 205–206, 221–222, 229, 238
approbation and condemnation, Clarke, Samuel, 237–238, 243,
167–168, 174–177, 182, 204, 245–246
210, 211, 249, 251, 258, 263, 264 Cumberland, Richard, 89–90, 101,
moral, and nonmoral esteem, 105, 151, 168, 173, 177
distinction between, 167 Hobbes, Thomas, 173, 174
beauty, 175 Hume, David, 177–179, 213,
benevolence, 133, 147, 151, 175–178, 220–224, 229–230, 232,
199, 204–206, 211, 213, 214, 237–241, 251, 254, 282
248, 254, 258, 261, 282–283 Leibniz, Gottfried Wilhelm, 124,
calm extensive. See Hutcheson, 153, 166
Francis:benevolence, Locke, John, 167, 173–174, 239, 251
universal Mandeville, Bernard, 173
particular, 180 Price, Richard, 237–238
universal, 166, 176–181, 184, 204, Shaftesbury (Anthony Ashley
232, 246 Cooper, Third Earl of), 113,
universal, and self-love, dualism 124, 140–144, 147, 151, 165–
between, 168, 170, 181, 186 166, 172–173, 177, 180–182,
choiceworthiness, moral, 133, 167, 204, 213, 238
168, 174, 177–178, 204, 205, Spinoza, Baruch, 122, 124, 127
213, 218 instinct, 179
color analogy, 69, 176 judgments, moral, 222
condemnation. See Hutcheson, justice, 168, 232
Francis:approbation and love, 166–167, 175
condemnation moral virtues and natural abilities,
desires, calm, 121, 127, 179–181 distinction between, 211
desires/affections and appetites/ morality, deontic, rejection of,
passions, distinction 166–168, 177, 204
between, 178–181 motives, contemplation of, 174–177,
egoism, rejection of, 173, 229 198, 213, 214, 218, 222, 229,
Enlightenment, Scottish, 201 251, 258
eudaimonism, rejection of, 168, 173, naturalism, empiricist, 151, 166,
178–181 173–176, 201, 240, 241, 251, 259

https://doi.org/10.1017/9781139025065.011 Published online by Cambridge University Press

 index 371

obligation, 89, 171, 173, 177–178, finite rational, 10, 273–274,

181–183 278–281, 284–286, 290–292,
of moral sense, 181–182, 213 324–326, 329–333, 339–340
of self-interest, 181–182, 240 perfect rational, 10, 273, 284, 285,
“ought” as “unlucky,” 181 289–290, 324, 329–331, 334,
passions, calm and violent, 127 340
perspective, observer’s, 165, 176–177, autonomy, 10, 136, 153, 221, 313, 322
190, 192, 198, 222–223, 232, and heteronomy, distinction
251, 258 between, 271–272
proto-utilitarianism, 177–178, 199 freedom, positive, 271, 309–312
psychology, moral, 165–166 of the will, 271–277, 280, 287, 300,
rationalism, rejection of, 201, 229 302, 308–315, 317–323, 325,
reason as motivationally inert, 168, 333–334, 336, 337, 339
179 benevolence, 282–283, 285, 309, 314,
reason, practical, 178–181, 184 322, 324, 328
reflection, calm rational, 168 “categories of freedom,” 271–272
right, the, 232 challenge, morality as chimerical, 25,
rights 30, 279–280, 288, 301, 309,
alienable and inalienable, 178 318, 321–322
perfect and imperfect, 174, 178 common humble person example,
self-reflection as interfering with 326–329
moral virtue, 165, 180 conscience, 330–331, 333
sense, moral, 113, 165–168, 173–183, constraint, 284
190–191, 198, 203–204, 210, external, 335–339
213, 218, 222–223, 229–230, self, 314, 335–340
239–240, 248, 251, 259, 263 deliberation, 10, 273–274, 280, 285,
sentimentalism, 113, 143, 167, 201, 294, 300, 309–316, 321, 322,
203–204, 218, 230, 237–241, 327, 339–340
243–246, 248, 263, 274 desire, 208, 286, 289, 299, 306–307,
utilitarianism, 162, 204, 205 309, 312–314, 316–321,
virtue 323–324, 328, 331
as coinciding with self-interest, dignity, 12, 203, 234, 321, 322, 327, 341
181 as an end in oneself, 299–308
ethics of, 166–168, 173, 176–183, disposition, morally good, 307
204–206, 213, 251, 254 duty, 273–274, 278–288, 290,
will, 179 293–295, 301–302, 307, 326,
329–338
Irwin, Terence, xiii, 10–12, 15, 16, 19, as the necessity of an action from
20, 28, 30, 64, 92, 97, 147, 151, respect for law, 283, 330
162–164, 189 ethical, 335–336
Israel, Jonathan, 115, 122 good will, 281–288
of right, 335–338
Johns, Christopher, 159 perfect and imperfect, distinction
between, 295
Kant, Immanuel, 10 strict and wide, distinction
accountability, 273, 280, 283, between. See Kant,
285–288, 292–294, 309, 314, Immanuel:Duty:Perfect
324–325, 331, 333, 336 and imperfect, distinction
agency, 208, 269, 322, 278–340 between

https://doi.org/10.1017/9781139025065.011 Published online by Cambridge University Press

 372 index

Kant, Immanuel (cont.) in relation to

to oneself, 335 Balguy, John, 283, 287
to others, 295, 302, 335, 338 Butler, Joseph, 221
empirical, the, 271, 278, 316–320, Clarke, Samuel, 273–274, 287
325–326, 332 Cudworth, Ralph, 270
false testimony example, 323–325 Grotius, Hugo, 30, 277, 279, 338
force. See constraint Hobbes, Thomas, 279, 287, 292
freedom, 10 Hume, David, 221, 279
as ratio essendi of the moral law, Hutcheson, Francis, 221, 282–283
314, 323 Leibniz, Gottfried Wilhelm, 125,
autonomy. See Kant, 153–154, 157, 160, 270
Immanuel:autonomy Locke, John, 100
external, 272–273, 336–339 Paul the Apostle, 282–283
laws of, 66, 127, 153, 278, 288–289, Price, Richard, 252, 270, 273–274,
311 287
negative, 309–311 Pufendorf, Samuel von, 300, 309
of choice, 272–273, 284, 306, 334, Rousseau, Jean-Jacques, 269,
336 274–278, 300, 304, 308, 336
positive. See Kant, Shaftesbury (Anthony Ashley
Immanuel:autonomy: Cooper, Third Earl of), 142,
freedom, positive 151
goodness, 279–293, 307–308, Smith, Adam, 203, 221, 300
311–313, 317, 319–322, 325, Suárez, Francisco, 18, 292–293
330–334, 336 Wolff, Christian, 153, 270
hindrance, 283, 337 incentives, 325–328
idealism, transcendental, 273, 300, 314 inclinations, 271, 273, 284–286,
ignorance, veil of, 304–306 289–290, 313, 314, 316–317,
imperative, categorical, 36, 208, 273, 323–324, 329–330, 334,
279–281, 285, 287, 288, 291, 339–340
298, 309, 315, 317, 320–322 insincere debtor example, 295–299,
Formula of Autonomy (FA), 301–302
302–303, 306, 308, 321, 334 kingdom of ends, 304–306
Formula of Humanity (FH), 295, law
299–303, 306, 308, 321, 334 moral, 271–275, 278–281, 284,
Formula of the Kingdom of Ends 286–291, 303–310, 323–335,
(FKE), 206, 306, 308, 321 337–340
Formula of the Law of Nature as ratio cognoscendi of freedom,
(FLN), 294 314, 323
procedure, 294–299, 303–306, 320 awareness of being bound by,
universal law formula (FUL), 272, 280, 314–315, 323–324,
285–290, 293–299, 301–303, 333
306, 308, 321, 334 moral and practical, distinction
imperatives, categorical and between, 271, 273, 279,
hypothetical, distinction 286, 290–291, 303–304,
between, 271, 288–291 331, 339
imperatives, hypothetical, 208, 271, practical, 271–273, 279, 281,
288–292 286–287, 289–290, 293–295,
assertoric, 292 299, 303–304, 311, 314–320,
problematic, 292 322, 330–332, 334, 339

https://doi.org/10.1017/9781139025065.011 Published online by Cambridge University Press

 index 373

lawgiving, 306, 308–309 prudent shopkeeper example, 281

external, 334–339 racism, 12–13, 340–342
internal, 334–340 rational egoist example, 297–298
logic, transcendental, 271 rationalism, 10, 269–272
love, 277, 282–283, 285, 314, 322, 324, critique of earlier rationalists, 270,
335 272, 320
self, 319, 328 realm of ends. See kingdom of ends
lust and gallows example, 323–324 reason, objective law of, 284–286,
matter, 271, 316–319 290
maxims, 271, 275, 280–282, 285–289, reason, practical, critique of, 272,
293–299, 302–303, 306, 270–274, 276, 279–280, 284,
315–320, 334 288, 294, 307–309, 314–323,
morality 333
deontic conception of, 269, reason, pure practical, 208, 271–274,
273–274, 278–288, 290, 293, 276, 278–281, 283–284,
301, 307, 308–311, 314–315, 287–290, 301, 309, 314–315,
318, 320–340 317–318, 322, 325–326,
deontic, normative 330–334
epiphenomenality of, fundamental law of, 273, 315,
273, 280–281, 283–285, 317–318, 322, 332
309, 333 synthetic use of, 279, 288
foundations of, a priori, 279 reason, pure, antinomies and
supreme principle of. See Kant, paralogisms of, 270
Immanuel:Imperative, reciprocity thesis, 280, 288, 322
categorical respect, 283–285, 290, 299–303, 314,
nature, laws of, 66, 127, 272, 278, 325–330
288–289, 311 and humiliation, 328–329
necessitation, 273–274, 280–281, appraisal, 307–308, 327–329
283–285, 289–290, 302–304, as a priori feeling, 326
307, 309, 314, 325, 329–331, as phenomenal feeling, 273–274
334–340 in relation to dignity, 300,
necessity, inner, 289–290 306–308, 321
necessity, natural, 289, 292 observantia, 307
obligation, 153, 273–274, 278, 281, recognition, 307–308, 326–328
283–287, 290, 293, 314, reverentia, 307
324–325, 330–333, 335–337 Revolution, Copernican, 270, 320
authoring of, 302–303 right, the, 273, 287, 300–302,
narrow and wide. See Kant, 306–308, 325, 330–339
Immanuel:duty:perfect Right, Universal Principle of (UPR),
and imperfect, distinction 272, 301, 306, 337–338
between sciences, 278
ought self-conceit, 203, 326, 328–329
narrow and wide scope readings, sexism, 12, 13, 340–342
291–292 sociability, 276–277
principles, material, 314–320 status, social, 326–329
principles, material and formal, sympathy, 282
distinction between, 271, thinking, form of, 270–271, 312
294 thought, theoretical, 272, 310–314,
prudence, 281, 292–293, 320 320

https://doi.org/10.1017/9781139025065.011 Published online by Cambridge University Press

 374 index

Kant, Immanuel (cont.) Greatest Happiness Principle, 153,

truth, synthetic a priori, 318 156, 166
value, priceable and dignity, harmonizer, as, 152
distinction between, 306–308 in relation to
volition, objects of, 153, 271, 272, 276, Aquinas, Thomas, 155
279, 308, 311, 317–319, 325 Cicero, Marcus Tullius, 156–157
will Clarke, Samuel, 157
determining ground of the, 271, Cudworth, Ralph, 111–113, 152,
316–319, 326 158
form of the, 271, 272, 312, 317, Grotius, Hugo, 155–157, 163–164
319–320, 322, 325 Locke, John, 158
goodness of the, 280–286, 288, Pufendorf, Samuel von, 112–113,
290, 293, 307, 308, 320, 327, 152, 154–155, 163
329, 332 Shaftesbury (Anthony Ashley
and duty. See Kant, Cooper, Third Earl of),
Immanuel:duty:good will 111–113, 152–155
hindrance, 339 Spinoza, Baruch, 111, 122, 124, 156
holy, 153, 290 judge, sentencing criminal, 162
world, phenomenal, 284 justice, 152–157, 159–164
worth, moral, 142, 151, 206, 278, as “the charity of the wise,” 125,
280–284, 325, 329 152, 155, 156, 159, 161
Kennett, Basil, 67 as balanced relationship between
Kierkegaard, Søren, 10 self-love and love of others,
Korsgaard, Christine, 300 159
agency, deliberative, 313 as universal benevolence, 152–154
autonomy, appeal to, 321 “Kingdom of Grace,” 157
contructivism, 320–321 love, 152, 156, 163
normativity, source of, 8, 172, 201 metaphysics, 152, 155, 157–158
Kosch, Michelle, xiii teleology, 158
moral quality, twofold, 159
Leibniz, Gottfried Wilhelm Naturalism, Aristotelian, 163
agency, free, 157–158 necessity
anti-positivism, 112–113 moral and metaphysical,
beneficence, 163 distinction between, 157–158
benevolence, 124, 152–157, 163, 166 moral and natural, 124, 152, 153,
charity, 124, 152, 155 155, 159
compatibilism, 158 obligation and the right, 152–155,
complaint, reasons of, 161 158–164, 171
demands, 153, 160–162 “science of right,” 153, 155, 158–159
eudaimonism, 124, 156, 163 claim rights, 160
God, 117, 152 modes of right, 155
as creator, 157 obligations, bipolar, 160
as exemplar, perfectly virtuous, 152 precepts, deontic, 160–161
wise and benevolent nature of, equity, 163, 164
154–159 piety, 160
Golden Rule, 160–161 perfectionism, 153, 155–157
good, the, 152–159 pleasure as knowledge of perfection,
dependence of moral goodness on 156
motives, 154 possibility, moral, 155

https://doi.org/10.1017/9781139025065.011 Published online by Cambridge University Press

 index 375

rationalism, 270 Cumberland, Richard, 88–90

reciprocity, principle of. See Grotius, Hugo, 34, 88–91
Leibniz, Gottfried Hobbes, Thomas, 90, 98
Wilhelm:obligation Hume, David, 90
and the right:precepts, Hutcheson, Francis, 90
deontic:equity Leibniz, Gottfried Wilhelm, 90
reductionism, 124 Pufendorf, Samuel von, 88–92,
respect, 153, 162, 163, 166 95–100
sanctions, eternal, 157 Shaftesbury (Anthony Ashley
theodicy, 152 Cooper, Third Earl of), 90,
“Universal Republic of Spirits,” 157 111, 139–140, 169–170
utilitarianism, 153, 156, 162–163 Suárez, Francisco, 93, 95, 97, 98
virtue, ethics of, 154–158, 163 internalism, 90, 96
voluntarism, rejection of, 111–112, knowledge, three-fold division of
152, 154–155, 248 logic and semantics, 97
wisdom, 155–157 natural philosophy (theoretical
Lloyd, Sharon, 54 knowledge), 97
Locke, John, 10 practical knowledge, 97
accountability, 90, 95, 98–100, 145, ethics, 97–98
148, 169–170, 239 law
agency, 2, 98–100, 148, 158, 169–170, distinction between divine, civil,
261 and law of “Opinion or
archetypal and ectypal concepts, Reputation,” 96
distinction between, 91, 97 natural, 88–89, 91–95, 97–100
authority, divine, 91, 90–94, 98, 103, morality as science, 88, 91
129 “morality,” use of word, 87
autonomy, 98–100, 112, 169–170 motivation, moral, 95–98, 100
believing and judging, distinction naturalism, 89
between, 137 nominal and real essences,
bonum honestum. See Good, distinction between, 91
common obligation, 90, 95–96
collective action problem, 93–95, 103 person, 2, 35, 87, 99
debitum naturale, 98 property, 94
egoism, 150 punishment, 90, 93, 95, 96, 99
empiricism, 167, 174 sanctions, 91
Epicureanism, 104 divine, 98, 91–100, 103, 141, 170
eudaimonism, 169–170 scarcity, natural, 94
externalism, 90, 96–98, 100 self-determination. See Locke,
Golden Rule, 87 John:autonomy
good, common, 95 souls, immortality of, 93–94
good, the, and the right, 92–100 utility, 92
hedonism, rational egoistic, 91, voluntarism, theological, 111, 117
93–97, 100, 170 will, 96, 260
imputation, 99 divine, 103
in relation to freedom of, 90, 98–100
Aristotle, 97
Confucius, 97 MacIntyre, Alasdair, 9
Cudworth, Ralph, 100, 111, 112, 136, Mandeville, Bernard, 173
169–170 Marx, Karl, 341

https://doi.org/10.1017/9781139025065.011 Published online by Cambridge University Press

 376 index

Masham, Damaris, 13, 111 Butler, Joseph, 253–257

Mill, John Stuart, 8, 162, 177, 183, 197, Clarke, Samuel, 237–239, 243, 247,
205, 229, 257 250, 255–256, 273, 274, 287
Mills, Charles, 14 Cudworth, Ralph, 113, 252, 256
Moore, G.E., 6–9 Fordyce, David, 251
egoism, refutation of, 107 Hobbes, Thomas, 287
fallacy, naturalistic, 209 Hume, David, 251, 254, 255, 274
in relation to Hutcheson, Francis, 251–252, 254,
Butler, Joseph, 185 257, 274
Cumberland, Richard, 107 Plato, 252
supervenience, 245 Reid, Thomas, 237, 273
value, intrinsic, 122, 319 Shaftesbury (Anthony Ashley
Cooper, Third Earl of), 142
Nadler, Steven, 121, 125 Suárez, Francisco, 250
Nietzsche, Friedrich, 3, 6–10 internalism, motivational, 253–256
in relation to intuitionism, 113, 252–253, 257, 263
Rousseau, Jean-Jacques, 274 liberty, 253–254
Spinoza, Baruch, 114, 115, 118–119, moral sense theory, rejection of,
121, 127 251–252, 254, 258, 259
morality, deontic, as illness, 115 morality
ressentiment, 118 as objective, necessary, and
unconscious, “dark workshop” of, 118 universally binding,
will to power, 121 252–253
deontic conception of, 113,
Oldenburg, Henry, 117 238–239, 251, 253–256, 258
obligation, 242, 250–253, 256–258
Parfit, Derek, 6, 223 as indefinable, 250–253
Pashukanis, Evgeny, 341 rationalism, 113, 229, 237–239,
Paul the Apostle, 282–283 250–258, 270
Pettit, Philip, 267 reason, 250–258
Plato, 4, 5, 127, 130, 134 as the source of moral ideas, 251,
Socrates on justice, 4, 11, 16, 28, 217 257, 258
Platonists, Cambridge, 113, 139, 144, 173 as the source of moral motivation,
Plotinus, 130 238, 255–256
Price, Richard Rectitude, 253–256
accountability, 250–251, 253–254, self-determination through
256, 273 conscience, moral goodness
agency, 251, 253–256, 273 as, 253–257
benevolence, instinctive and understanding, the, 251–253, 258
rational, distinction between, virtue
254 abstract and practical, distinction
capacity, moral, 253–254, 256 between, 256–257
fitness, 250–251, 253–257 heads of, 257–258
golden rule, 287 Prichard, H.A., 4–8, 219, 257
ideas, normative, as indefinable, 239 Pufendorf, Samuel von
ideas, simple, 209, 250 accountability, 61–62, 67, 66–69, 71,
in relation to 73, 80, 86, 91, 96, 164, 194,
Balguy, John, 250–252, 255–256, 202, 219, 239, 256, 309
274, 287 agency, 79, 82–86, 239, 256

https://doi.org/10.1017/9781139025065.011 Published online by Cambridge University Press

 index 377

authority, divine, 62, 63, 65, 70, pacts, 65, 66, 68, 79–81
76–82, 129 powers
imposition of, 41, 42, 61, 62, 64, natural, 68, 75, 83, 86
66–72 normative, 41–43, 62–75, 78–83,
authority, mutual human, 76 86, 90
causes, moral, 68 divine, 62, 83, 86
color analogy, 69 promising, 65–66
compensation, 77, 78 punishment, 77–78
compulsion and obligation, recognition, 61, 62
distinction between, 69, 75, mutual, 62, 64, 67, 68, 72–73, 91
82–86, 91, 96 sanctions, 66, 67, 69, 75, 77, 80,
conscience, 83 83–86, 96
and shame, distinction between, self-defense, 76
83–84 sociability, 22, 62–65, 71–75, 78–79,
consent, 81–82 83, 86, 91, 265, 268, 269, 274
creation, divine, 66–67, 70, 72 space, moral, 62, 66–67
demands, authoritative, 62, 66 voluntarism, theological, 41, 62, 111,
dignity, equal human, 42, 62, 64, 68, 117, 164
69, 74–76, 78, 79, 86, 234, 300 war and peace, 42–43, 72, 74
effects, moral, 66–68, 75 will, 67, 79, 80, 84–85
empiricism, 41
entities, moral and physical, 65–67, Rawls, John, 9, 60, 116, 257
69, 72, 90 categorical Imperative Procedure,
esteem, 73–74, 76, 78–79 294–299, 305, 320
goodness, natural, 69 constructivism, 320–321, 334
imputation, 62, 68–69 ignorance, veil of, 304–306
in relation to justice as fairness, 306, 331
Grotius, Hugo, 38, 42–43, 62–65, justice, conditions of, 215
73, 78 needs, true human, 305
Hobbes, Thomas, 41–43, 61, 65, nullishing and telishment example,
69, 83 296–297
Locke, John, 96 practice and summary conception of
Suárez, Francisco, 62, 63 rules, 217
injury, 74 rational, the, 54
intercourse, 73–74, 83 right, priority of, 331, 332
law, natural, 42, 62–65, 70–71, 73, 76–77 “self-originating source of valid
“morality,” use of word, 87 claims,” 42, 300
motivation, moral, 79–81, 84–86, 96 Raz, Joseph, 65
nature, state of, 83 authority, theory of, 339–340
necessity, moral, 155 Reid, Thomas
obligations and rights, 62, 68–70, accountability, 260–262, 266–268,
75–86, 239 273
bipolar, 64–66, 69, 70 agency, 259–262, 273
mutual, 42, 74, 76, 86 approbation, 263–265
natural, 160 attention, 261
perfect and imperfect, 62–64, 67, benevolence, 261, 262
71–73, 76–79, 81 children, 266–268
to god, 42, 64, 67, 71, 76 cognitivism and noncognitivism,
violation of, 78 contrast between, 263–265

https://doi.org/10.1017/9781139025065.011 Published online by Cambridge University Press

 378 index

Reid, Thomas (cont.) self-government, capacity of,

common sense, moral, 259 259–263
conscience, 261–263 self-love. see conscience
conscience and self-regard, and self-regard,
relationship between, relationship between
261–263 sense, moral, 258–259
deliberation, 261, 263 sentimentalism, empiricist, rejection
empiricism, critique of, 259 of, 263–265
faculty, moral, 259, 262 sociality, 265–269
favor and injury, 265–266 subjectivism, rejection of, 259,
Golden Rule, 263 264–265
gratitude, 265 testimony, 267
in relation to the animal, the mechanical, and the
Balguy, John, 237 rational, distinction between,
Butler, Joseph, 259, 261–263 260
Clarke, Samuel, 237, 273 trust and trustworthiness, 268
Cudworth, Ralph, 260, 261 virtue
Fordyce, David, 258 and the right, 259–262
Grotius, Hugo, 265, 268 as conscientious conduct, 259–262
Hume, David, 259, 263–267 will, 260–262
Hutcheson, Francis, 259, 261, Ripstein, Arthur, 336
263–265 Ross, W.D., 185, 208, 247, 257, 258
Kant, Immanuel, 260 Rousseau, Jean-Jacques
Locke, John, 260, 261 Amour de soi, 277
Price, Richard, 237, 258–260, 263, Amour propre, 277–278
273 association, political society as, 269,
Pufendorf, Samuel von, 265, 274–275, 304
268 contract, social, 274–277
Smith, Adam, 265, 268 dignity, equal human, 269, 274,
incitements to the will and volition, 277–278, 300, 308
distinction between, 260 freedom, 274–278
indignation, 265 common, 275–276
judgment, 259, 261–265 moral, 275–276, 278
justice, 260, 262, 265–268 in relation to
mind, power of, moral, 258–259, Grotius, Hugo, 269, 274, 277, 278
262 Locke, John, 100
obligation, 259–263, 265–268 Pufendorf, Samuel von, 269, 274
power, active, 259–263 Reid, Thomas, 269
promising, 266–268 Smith, Adam, 269
prudence, 260, 262–263 law, 274–278
purpose, fixed. see resolution slavery, 275–276
rationalist/sentimentalist debate, sociability, 269, 274, 276–277
reluctance to enter into, 237, will, general, 274–277
259 as compared with the “will of all,”
resolution, 260–261 275
self-condemnation, 261–262
self-determination. see reid, Saint Augustine, 124
thomas:self-government, Scanlon, T.M., 38
capacity of reasonable, the, 54

https://doi.org/10.1017/9781139025065.011 Published online by Cambridge University Press

 index 379

Schapiro, Tamar, 313 Platonists, Cambridge, 173

Schneewind, J.B., 16, 152, 154, 163–164 Spinoza, Baruch, 111, 115, 122, 124
Seneca, 21 Suárez, Francisco, 141, 146
Shaftesbury (Anthony Ashley Cooper, Whichcote, Benjamin, 113
Third Earl of) inspiration, 144–145
accountability, 146 inspiring disease, 149
“advice to authors,” 147, 149 interest, public, 141, 143, 145
aestheticism, 142–145 love, 112, 142, 151
affections, 141–147 mind, 142, 144–146
degrees of, 142 natural order, teleology of, 141–145, 151
agency, 151, 169–170 Naturalism, Aristotelian. See
anti-positivism, 112–113 Shaftesbury (Anthony
autonomy, 169–170, see Shaftesbury, Ashley Cooper, Third Earl
Anthony Ashley Cooper, of):eudaimonism
Third Earl of:self- obligation to virtue, 89, 140–141,
determination, self-critical 146–147, 171, 173, 180, 181, 213,
beauty. see Shaftesbury, Anthony 228, 240
Ashley Cooper:third earl powers, creative and practical, 113
of):goodness:as beauty psyche as system, 147
benevolence, 142, 151, 180 rationalism, 144
deontic conception of morality, reductionism, 181, 182
absence of, 140–141 self, the, 140
desires, “sly” and “insinuating,” self-converse. see Shaftesbury,
148–150 (Anthony Ashley Cooper,
education, liberal, 143 Third Earl of):soliloquy
egoism, rational, 147, 150 self-determination, self-critical, 142,
enthusiasm, noble, 144 145–151, 165, 180, 238
eudaimonism, 141, 147, 151, 173, 181, 180 self-government, 150–151
God as unifying creative mind, 144 self-love, 142
goodness, 113, 141–145 sense, moral, 113, 140, 142–145, 151,
as beauty, 142–145 173, 181
first order of, 145 sentimentalism, 124
as natural, 141–143, 151 soliloquy, 147–151
hedonism, rejection of, 150 systems, scorn of, 140
honesty, 149 taste, cultivated, 143
ignorance, motivated, 148–149 theism, 144
in relation to therapy, self-analytic reality, 148
Butler, Joseph, 238 virtue, 140–148
Cudworth, Ralph, 111–113, 139– as intrinsic good, 151
140, 144–148, 150, 151, 169–170 ethics of, 113, 140–142, 151, 173, 177
Cumberland, Richard, 89 voluntarism, rejection of, 111, 112,
Descartes, René, 146, 147 139, 151, 152
Epictetus, 146, 147 will, 151
Hobbes, Thomas, 139 Shelby, Tommie, 14
Leibniz, Gottfried Wilhelm, 111 Sidgwick, Henry, 161, 185, 197, 257
Locke, John, 100, 112–113, 138–140, ancient and modern ethics, contrast
145–148, 150–151, 169–170 between, 4–7, 11–12, 19–20,
Locke, John, 152 24–25, 28, 30, 62, 101, 151, 170,
Mandeville, Bernard, 173 200, 204, 217, 330–332

https://doi.org/10.1017/9781139025065.011 Published online by Cambridge University Press

 380 index

Sidgwick, Henry (cont.) Shaftesbury (Anthony Ashley

benevolence, rational, principle of, 107 Cooper, Third Earl of), 147,
“good from the point of view of the 222, 228, 234, 238
universe,” 122 inductivism, 229
in relation to judgment, aesthetic, 230–231
Butler, Joseph, 168, 184, 186, 189, 200 justice, 221, 231–234, 265, 268
Cumberland, Richard, 101, 107, magnanimity, 235
168, 184 merit and demerit, judgments of,
Hutcheson, Francis, 168, 184, 204 231–232
Shaftesbury(Anthony Ashley person outside society example, 230
Cooper, Third Earl of), 140 projection, imaginative, 202, 221–233
“ought,” irreducibility of, 209 epistemic function, 225–226
reason, practical, dualism of, 168, normative function, 222–223,
182, 184, 186 225–229
utilitarianism, 162 projectivism, 231
Slote, Michael, 9 propriety, 202, 221–236, 243
Smith, Adam prudence, 229
accountability, 202–203, 221, punishment and reward, 231–234
233–234, 236, 268 pyschology, moral, 202
Adam Smith Problem, 202 rationalism, rejection of, 201
adoption and approval of mental resentment, 221, 231–234, 265
states, 225–226 respect, 202–203, 234–236
autonomy, 221–222, 229 self-command, 222, 234–236, 238
benevolence, 229 excellent exercise as perfect virtue,
condolence, 227 222
dignity, equal human, 202–203, self-government. see smith,
233–236, 300 adam:self-command
economics, 202, 234 self-love, 234, 235
empiricism, 201, 231 sentimentalism, 201, 202, 230,
Enlightenment, Scottish, 201 237–239, 244–246
expressivism, 231 simulation, 224
fellow-feeling, 202, 225, 226, 231 situation, 202, 223–230, 244
fittingness. see smith, slavery, 235
adam:propriety sociability, 269
gratitude, 231, 265 spectator, impartial, 202, 227–228,
hand, invisible, 202 232–233, 235
in relation to sympathy, 202, 224–225, 227–231
Balguy, John, 229, 237–238 virtue, 222
Butler, Joseph, 184, 221–222, 229, amiable and respectable,
234, 238 distinction between, 234–236
Clarke, Samuel, 229, 237–238, ethics of, 228–231
243–244 Smith, Michael, 201, 206, 209, 267
Grotius, Hugo, 233 Spinoza, Baruch
Hume, David, 201–203, 220–224, accountability, 117–118
227–238 action, 117–118
Hutcheson, Francis, 201, 221–24, activity, 121, 125
227–230, 232, 234, 237 Amsterdam, Portuguese Jewish, 116,
Price, Richard, 229, 237–238 126
Rousseau, Jean-Jacques, 235 atheism, 116

https://doi.org/10.1017/9781139025065.011 Published online by Cambridge University Press

 index 381

badness as privation, 118, 119, 121 lens grinder, occupation as, 126
bearing calmly, 123, 125–127 liberalism, 116
blame, 118–119 love, 118, 119, 126–127
blessedness. see spinoza, metaphysics, 116–118, 123
baruch:bearing calmly ethics as, 114–115
conatus, 115, 118–121 monism, 114, 122, 125
consciousness, 114 morality, deontic, rejection of,
consequence, causal, 118 115–119
demands, 123 nature, 113–126
doctrine of free will, rejection of, 115 human, model of, 121–122
duty. see spinoza, baruch:right, the necessity, natural, 117–119, 123–124
Enlightenment, Radical, 116, 122 passions, 117–119
eudaimonism, perfectionist, 114, emotions, taxonomy of, 126–127
119–128 perfection, 114–115, 119–126
extension, 114 politics, democratic, 115–116, 125
fatalism, 117 power, 118, 121, 122, 125, 126
fear, motivation by, 118 projectivism, 119–121
freedom, 114, 121, 125, 127 reality, perfection as, 118
God reductionism, 121, 124–125
as substance, 116–119 remorse, 118
personal, rejection of, 115 repentance, 118
good, the, 118–126 right, the, 114, 122–125
apparent good and objective good, sadness, 119, 126–127
distinction between, 120–121 sanctions, 117
common, 114, 122, 124, 125 striving. see conatus
happiness. see Spinoza, virtue, 113–115, 121, 125
Baruch:bearing calmly voluntarism, rejection of, 111
hate, 118–119, 126–127 well-being, 114
ignorance, 118 will, 120–121
in relation to appetites/desires and will/action,
Aquinas, Thomas, 114 distinction between, 120
Aristotle, 114–115, 122 doctrine of free will, rejection of,
Cudworth, Ralph, 111, 117, 118 117–118, 127
Descarte, René, 114 Stoics, 5, 11, 16, 28, 37
Hobbes, Thomas, 114, 119–120, Strawson, P.F.
125, 126 attitudes, reactive, 30, 59, 196, 231,
Leibniz, Gottfried Wilhelm, 111, 239, 287
117, 122, 124 Suárez, Francisco, 26
Locke, John, 117 accountability, xii, 17–18, 164, 270
Pufendorf, Samuel von, 117, 118 agency, xii
Shaftesbury (Anthony Ashley Bonum honestum, 20, 92
Cooper, Third Earl of), 111, eudaimonism, 26, 30
122, 124 externalism, 29
indignation, 118, 126–127 in relation to aquinas, thomas, 17,
inquisition, the, 116 19–20, 92
intellect, 120, 126–127 right, the, 68
jealousy, 119 law and counsel, xii, 8, 11, 17–19, 23,
joy, 126–127 29, 54, 141, 146, 155, 159, 164,
knowledge, 114, 120, 121, 125 171, 181, 238, 250, 292–293

https://doi.org/10.1017/9781139025065.011 Published online by Cambridge University Press

 382 index

Suárez, Francisco (cont.) Vázquez, Fernando, 32

law, natural, 16–20, 26, 62, 63, 92
Naturalism, Aristotelian, 19, 26 Whichcote, Benjamin, 112, 113
voluntarism, 18–19, 24 Williams, Bernard, 3, 9, 165
voluntarism, theological, 164 morality system, critique of, 269
will, 18 Wolff, Christian, 153, 270
Wollaston, William, 113, 237, 243
Thompson, Michael, 159 Wood, Allen, xiii, 300, 302
Tuck, Richard, 19, 31, 33, 63
Tyrell, James, 87 Yovel, Yirmiyahu, 116

https://doi.org/10.1017/9781139025065.011 Published online by Cambridge University Press