DESCARTES'S

MORAL THEORY
JOHN MARSHALL
Cornell University Press
ITHACA AND LONDON
Copyright 1998 by Cornell University
All rights reserved. Except for brief quotations in a review, this book, or
parts thereof, must not be reproduced in any form without permission in
writing from the publisher. For information, address Cornell University
Press, Sage House, 512 East State Street, Ithaca, New York 14850.
First published 1998 by Cornell University Press
Printed in the United States of America
LIBRARY OF CONGRESS CATALOGING-IN-PUBLICATION DATA
Marshall, John, b. 1933
Descartes's moral theory / John Marshall.
p. cm.
Includes bibliographical references and index.
ISBN 0-8014-3567-6 (alk. paper)
I. Descartes, Ren, 1596-1650Ethics. 2. Ethics, ModernI7th
century. I. Title.
BI878.E7M37 1998
170'.92dc2I
98-28008
Cornell University Press strives to use environmentally responsible
suppliers and materials to the fullest extent possible in the publishing of its
books. Such materials include vegetable-based, low-VOC inks and acid
free papers that are recycled, totally chlorine-free, or partly composed of
nonwood fibers.
Cloth printing 10 9 8 7 6 5 4 3 2 I
-null6-

Contents
Acknowledgments
Note on Sources
Introduction I
PART ONE The Morality of the Discourse 9
i. Descartes's Morale par Provision II
2. Descartes's Second Maxim 34
3. The Third Maxim
PART Two Descartes's Final Morality 57
4. The Summum Bonum
5. The Rules of Morality
6. Gueroult's Account of Descartes's Ethics

ix
xi

48
59
71
85

7. The Moral Theory of the Passions

PART THREE Value and Generosity 115
8. A Cartesian Theory of Value
9. Morality as Generosity
Works Cited
Index

96
117
148
167
171

-viiAcknowledgments
WHEN in the early 1980s I first read Genevive Rodis-Lewis's splendid book La Morale de
Descartes, it occurred to me that there should be a similar work in English designed to
introduce Descartes's moral thought to an anglophone audience, an audiencelike myself at
the timenot inclined to think of Descartes as a moral theorist. It was only years later,
however, with the good wishes of the Dean of the Faculty at the University of Virginia,
Raymond Nelson, who generously granted me leave from teaching duties, and with the
considerable encouragement of my wife, Gail, that I came to think seriously that I might be
the author of such a book. So it was with this goal in mind that I applied to the Camargo
Foundation in Cassis for a residential grant, and it was later as a guest of the Foundation in
the spring of 1991 that I spent many happy hours working on the first draft of the present
study. I am very grateful to the Foundation and to its executive director, Michael Pretina, who
did much to make my stay agreeable as well as profitable, and much as well to keep my wife
in touch with her law practice and to guide my eleven-year-old daughter, Starling, in the ways
of the local culture. I am grateful also to three others who very early on gave me
encouragement: Martha Bolton, Jude Dougherty, and B. J. Diggs.
At a later stage Daniel Devereux, the chairman of my department, arranged for the two of us
to offer a graduate seminar on Stoic and Cartesian ethics; we were joined by our colleagues
Jorge Secada and James Cargile and, much to the delight and profit of everyone, by Lawrence
Becker of the College of William and Mary, who was at the time writing a
-ixbook on Stoic ethics. I owe a considerable debt of gratitude to them all and to three students
who attended: Michael Marshall, John Anderson, and Jay Morris. A few months later Becker
invited me to present my work to his discussion group at William and Mary. So I am doubly
indebted to him and owe special thanks to his wife, Charlotte, who produced on that
occasionwhich was also Descartes's birthdaya celebratory and suitably decorated cake.
I also express my thanks to Veronique McNelly for helping me with my translations, to Clark
Thompson for sharing with me his considerable knowledge about the period, to Marjorie
Greene for her encouragement at a late stage of the project, and finally to Cambridge
University Press for granting me permission to quote so extensively from their recent three
volume edition of Descartes's writings.
JOHN MARSHALL
Rapidan, Virginia

-xNote on Sources
CITATIONS of Descartes's works are, first, to the volume and page number of the new
French edition, Oeuvres de Descartes, ed. Ch. Adam and P. Tannery, rev. ed. ( Paris: J.
Vrin/C.N.R.S., 1964-76), and, second, to the volume and page number of The Philosophical
Writings of Descartes, trans. J. Cottingham, R. Stoothoff, and D. Murdoch, 3 vols.
( Cambridge: Cambridge University Press, 1985-91). Except where I indicate otherwise, I
adhere to these translations.
-xiIntroduction
To teach how to live without certainty, and yet without being paralyzed by hesitation, is
perhaps the chief thing that philosophy, in our age, can still do for those who study it.
BERTRAND RUSSELL,
A History of Western Philosophy
I. WHEN Leibniz was asked about Descartes's contribution to philosophy, he wrote to a
correspondent that Descartes's ethics was largely derivative from Seneca and amounted to no
more than a morality of patience without hope. "We need only inspect the incomparable
manual of Epictetus and the Epicurean of Laercia," Leibniz wrote, "to admit that Descartes
has not much advanced the practice of morality." i. The true morality, he claimed, must give
us hope and must therefore depend on our knowing that we are immortal and will be rewarded
or punished in the hereafter by a just God. Neither Descartes's God nor his immortality, he
went on, is of the right sort, his God being little different from Spinoza's and his immortality
lacking continuity of memory.
This is not a generous assessment, even if it is in one respect more accurate than Henry
Sidgwick's, expressed nearly two centuries later and shared by many anglophone philosophers
today, that Descartes did not touch on ethics proper. Descartes did write on ethics and with
much of what Descartes wrote Leibniz was evidently familiar. His misgivings about
Descartes on God and immortality, however, are, if not misplaced, in any event premature, for
in contrast to Leibniz Descartes held that neither a conception
____________________
i.
Leibniz 1989, 241. I am grateful to Larry Becker for this reference.
-iof how to live well nor the motivation to live morally should depend on assumptions
concerning the hereafter. What Leibniz saw as a weakness, therefore, Descartes claimed as a
strength. As for Leibniz's claim about Descartes's originality, I think a careful look at the
theory to be set out here will be answer enough, although in passing we could note that
Descartes's theory of morality, conceived as it was within the larger framework of Cartesian
metaphysics and an emerging scientific conception of nature, surely must contain some
originality and move beyond ancient Stoicism.

2. One place Descartes touched on morality is the letter-preface to the French edition of his
Principles of Philosophy. Since this passage, though not very helpful, is quite well known, it
deserves some comment at the start: "The whole of philosophy," he wrote, "is like a tree. The
roots are metaphysics, the trunk is physics, and the branches emerging from the trunk are all
the other sciences, which may be reduced to three principal ones, namely medicine,
mechanics and morals. By 'morals' I understand the highest and most perfect moral system,
which presupposes a complete knowledge of the other sciences and is the ultimate level of
wisdom" (9B: 14 ; 1:186). In his Principles, Descartes systematically set out his metaphysics
and the general principles of his physicsthe root and trunk of philosophy. There he claimed
to know beyond reasonable doubt "the principal attributes of God, the nonmaterial nature of
our souls ... [and] the true principles of material things" (98 14 ; I:186). This confidence,
however, has not been his legacy; few would now claim certain knowledge on Cartesian
grounds that "there is a God on whom all things depend, whose perfections are infinite," or
that our soul "subsists apart from the body, and is much nobler than the body" (4:291-92;
3:265). Nor would many claim to know with Cartesian certainty "the true principles of
material things" or "the general composition of the entire universe." What, then, is to become
of his ethics if his metaphysics and physics are not available to provide the requisite
nourishment? To be sure, our knowledge of the two branches, mechanics and medicine, not to
mention the natural science of physics, vastly exceeds Descartes's, and perhaps in ways
congenial to his cast of mind, but would this knowledge alone, unaccompanied by the proper
metaphysics, bring us close to a Cartesian perfect moral system?
On one interpretation of the tree image, it would. According to this reading, the perfected
Cartesian ethics was to be a system of Kantian hypothetical imperatives. Adhering to it would
assure health, long life, and control over nature, all combining to make us happy. This
reading fits with what Descartes had written earlier in part 6 of the Discourse on the Method,
where he admitted that he believed he could not keep his method and es
-2says on scientific topics secret "without sinning gravely against the law which obliges us to do
all in our power to secure the general welfare of mankind," since we could use scientific
knowledge to "make ourselves, as it were, the lords and masters of nature.... This is
desirable ... for the invention of innumerable devices which would facilitate our enjoyment of
the fruits of the earth and all the goods we find there" (6: 62 ; I: 143 ). And concerning
medicine specifically he added that we could apply the knowledge he hoped to secure by the
careful use of his method "for the maintenance of health, which is undoubtedly the chief good
and the foundation of all the other goods in this life. For even the mind depends so much on
the temperament and disposition of the bodily organs that if it is possible to find some means
of making men in general wiser and more skillful than they have been up till now, I believe
we must look for it in medicine" (6: 62 ; I: 143 ).
Although they are a bit overstated, the main points in these passages are both clear and
noncontroversial: assuming we have the right ends, better science will give us more reliable
knowledge about how to attain these ends, and this knowledge will therefore be a proper part
of practical wisdom, that is, of knowledge how to act and to live well. On the other hand,
these passages certainly do not imply that he conceived the whole of morality to be applied
natural science.

Nor does what he writes elsewhere on moral topics. In his correspondence contemporaneous
with the letter-preface to the Principles he emphasizes the critical role within morality of
knowledge of metaphysics and of general physics, but here he clearly is not thinking of
morality as applied natural science. If we are to live well and happily, he writes to Princess
Elizabeth, we should know that there is a God, that all things are expressly sent by God, that
the universe is vast, and that the earth enjoys no privileged status. And in a 1646 letter to
Hector-Pierre Chanut, who had written that he was planning to undertake a study of moral
philosophy, he wrote:
I agree with you entirely that the safest way to find out how we should live is to discover first
what we are, what kind of world we live in, and who is the creator of this world, or the master
of the house we live in. But I cannot at all claim or promise that all I have written is true, and
besides there is a very great distance between the general notion of heaven and earth, which I
have tried to convey in my Principles, and a detailed knowledge of the nature of man, which I
have not yet discussed. However, I do not want you to think I wish to divert you from your
plan, and so I must say in confidence that what little knowledge of physics I have tried to
acquire has been a great help to me in establishing sure
-3foundations in moral philosophy. Indeed I have found it easier to reach satisfactory
conclusions on this topic than on many others concerning medicine, on which I have spent
much more time. So instead of finding ways to preserve life, I have found another, much
easier and surer way, which is not to fear death. But this does not depress me, as it commonly
depresses those whose wisdom is drawn entirely from the teaching of others, and rests on
foundations which depend only on human prudence and authority. (4:441-42; 3:289)
Descartes does not make it clear in this letter precisely how his knowledge of physics helped
him to establish sure foundations in moral philosophy. Nor does he give a clear indication of
what he took moral philosophy to be. At about the time of this letter, however, he was at work
on a theory of the passions, of which one aim was to establish their physical causes; it may be
to this work that he was alluding in the letter. In any event, Descartes did conceive moral
wisdom to consist, in part, of knowledge of our own nature as psychophysical beings and,
more particularly, knowledge of how to get control over our passions and to get the most from
them. This is the part of moral wisdom that he offers in his Passions of the Soul.
There is, however, a further part of Cartesian moral wisdom, indeed, the principal part, that
is not in the least suggested by the tree image. This part, concerned with values and norms
and above all with moral virtue, is not applied science, nor does it depend on natural
scientific knowledge. It is the part far better represented in Descartes's dedicatory letter to
Elizabeth in the Latin edition of the Principles of 1644, where, revisiting the theme of his
morale par provision of the Discourse, he wrote: "whoever possesses the firm and powerful
resolve always to use his reasoning powers correctly, as far as he can, and to carry out
whatever he knows to be best, is truly wise, so far as his nature permits" (8A: 2 ; I:191).
It is true, nonetheless, that the moral wisdom Descartes himself had in mind does depend on
his metaphysics. To this extent, then, his famous image is faithful to his conception of a
perfect morality.

3. Although readers today find much of Cartesian metaphysics doubtful, the Cartesian cogito
holds a truth that we cannot easily escape and that is at once central to Cartesian ethics and,
I would argue, to any theoretical understanding of morality. This is the fact of our own
agency and autonomy. What kind of fact this is, is controversial. Descartes believed that what
was disclosed in the cogito could be interpreted as a substantial self a soulwith the
property of free will and the potential for immortality. But even if we doubt that what is
disclosed to us in practical deliberation (or in theoretical reflection, for that matter) can bear
such a rich metaphysical interpretation, we cannot but see ourselves as agents possessing
-4autonomy so long as we entertain the practical question, What should I do? This question is
fundamentally normative; accordingly, when we pose it we assume not only our autonomy,
since we are asking the question of ourselves, but our rationality, because when we deliberate
we are presented with a variety of considerationsstemming from desires, impulses, felt
habitual tendencies to act, and previously embraced principles of action among which we
choose, seeking of course to do so in a way that will withstand our own retrospective critical
examination. And since what we (and Descartes) call reason is essentially our capacity for
critical examination and reflectionwhatever metaphysical sense we can make of our
autonomy and rationalitythis capacity seems inescapable from our perspective as agents.
Much of Descartes's moral theory, though not all of it, can stand on this metaphysically thin
interpretation of the cogito.
Yet Descartes was acutely aware that reasons that agents consider good may in fact not be
good. From an outer perspective our putative reasons and practical judgments can be seen to
be products of our culture or other causes of which we are not aware and over which we have
no control; for when we look at ourselves as we might look at otherswhen we regard
ourselves not as we do within practical deliberation but from outsidewe are prey to the
unsettling thought that what appear to us to be good reasons from within are just common
prejudices when appraised from without. In the domain of theory, as distinct from practice,
Descartes adopted a method of inquiry that was to be proof against mere prejudice, so that
what looks compelling from the inside cannot be called into doubt from the outside. In this
way the autonomy of the theorist, the subject of pure inquiry, is secure. But for a number of
reasons, as Descartes himself was to insist, this method is not suitable for the deliberating
agent who is obliged to act without certainty and for reasons that may not hold up under
closer scrutiny. Indeed, it would not be far wrong to interpret Descartes's moral theoretical
project to be that of working out a way for us to live well and happily in the face of
uncertainty, an uncertainty that, owing to our finitude, is not fully eliminable.
4. The book falls into three main parts. The first deals with the morale par provision of the
Discourse; the second treats what I call Descartes's final morality and focuses on his account
of virtue, which is found in his correspondence and in the Passions of the Soul; the third, a
rather more speculative part, sets out a Cartesian theory of value and a system of duties.
Descartes's conception of a morale par provision addresses an important practical problem
of a general nature, not a problem peculiar to the Cartesian project of pure inquiry. A morale
par provision has the following features. (a) It is a particular person's morality, made up of
first-order moral
-5-

precepts and ideals of good character and second-order views about what one is responsible
for. Typicallyand certainly in the case of Descartes this morality is shared by a larger
community, of which for this very reason one takes oneself to be a member. When I say that a
morality is, say, Marcus's morality, I mean that it is embedded to a significant degree in his
character. Marcus not only owns up to the beliefs in question but has affective and conative
dispositions that support allegiance to the beliefs he professes. We might say that the morale
par provision has a practical grip on Marcus; Marcus defers to it in his deliberations, in his
self-criticism, and in his appraisal of the conduct and character of others (even those outside
his community). (b) If Marcus's practical grip on his morality is reasonably firm, he may
nonetheless be vulnerable to a number of doubts about it. He may have some doubts about the
correctness or validity of the first- and second-order precepts. He may also have some doubts
about the importance of morality in his life. His cognitive grip on his morality, then, may not
be ideally secure. He has lively beliefs but admits that they are short of certain knowledge
or at least that some of them are. (c) Marcus has a general cognitive procedure and is
somewhat hopeful that by following it he may be able to remove some of his doubts about his
morality and perhaps, in some as yet unforeseen way, to alter or expand its content. It might
be argued that the only morality any reflective person has in the current age is a morale par
provision and that all of us are in a situation not unlike that of Descartes in the Discourse.
Descartes's final morality is, in fact, only a more fully elaborated form of the earlier morality
of the Discourse. That this is so is a matter of controversy, one to be taken up in due course.
This final morality is not the perfect morality referred to in the letter-preface of the Principles,
a perfect morality of the distant future, one that would require a knowledge perhaps as
unimaginable to us today as our own knowledge (skepticism aside) would have been to
Descartes's contemporaries. On the other hand, this final morality contains a well-developed
account of virtue and happiness, one not to be dismissed readily as a mere relic of
seventeenth-century neo-Stoicism.
Descartes stands near the end of a great tradition of moral theory that began with the
ancients, a tradition that took its point of departure from the question of how we should live,
where "we" refers to each human being and the question is designed to take account of a full
life, one's life as a whole. This is the tradition in which the summum bonum is a central
organizing concept. The highest good is, of course, happiness, but the controversies among
theorists in this tradition concern precisely how happinessliving wellis best understood.
In the main, Descartes takes his stand with the Stoic argument that virtue is necessary and
sufficient for hap
-6piness. On the other hand, he affirms what classic Stoics deny: for human well-being and
happiness, passions are necessary. The heading of the final article of his Passions of the Soul
reads, "It is on the passions alone that all the good and evil of this life depends" (II:488;
I:404).
Descartes also stands at the beginning of modern moral theory, if we take this to have begun
more or less with Hobbesthe Hobbes who famously said that he had no use whatever for
the summum bonum as an organizing principle. Descartes evidently read some Hobbes and
once wrote that Hobbes was a good moral philosopher, if not such a good metaphysician.
What he learned from Hobbes is not clear. One thing he would not accept from Hobbes is his
theory of obligation. Descartes does not work with a theory of obligation according to which

to be obligated is to be under the threat of some sort of sanction. He mentions this common
idea in his dedicatory letter preceding the Meditations, but it does not figure at all in his
correspondence or in the Passions. Moreover, he does not tie being bound by a moral law to
his own interest. Rather, he seems to commit himself to the view that the moral "ought" is a
function of the objective value of ends.
Yet, like Hobbes, he is a modern moral philosopher in the sense that he works out a morality
in the light of the evolving modern conception of science and nature (which conceptions he
helped to initiate). He disassociates himself, in effect, from any natural law tradition that
presupposes a teleological framework. And though he does introduce a teleological
framework of his own with respect exclusively to human beings, this serves a limited and
indeed largely epistemological purpose in his moral theory. He does not read off moral laws
from this framework. What we discover in the cogito is our own freedom, the freedom of our
own reason, which sets us apart from everything else in nature. We are rational, and insofar
as we act under the guidance of our own reason, we seek the good. It is rarely within our
cognitive power, however, to achieve certainty concerning what, specifically, is the best thing
to do in particular circumstances; and even when we act with reasonable confidence, the
good we seek often turns out to have been beyond our power. We are, in short, free to reason
and to choose but also limited in our capacity to know and to control. In one respect, then, the
good we aspire to achieve is elusive. In another respect, however, it is notat least not
according to the defining line of argument of Descartes's ethics. For according to this ethics,
the highest good for us, virtue crowned with supreme contentment, is entirely in our power, so
long as we have our reason.
For many readers, as I noted above, it may come as a surprise to hear mention of Descartes's
moral theory. It is true, of course, that he did not write a single treatise on the subject, and it
is also true that his discussions
-7of moral philosophical topics over the years yielded no fully developed theory. On the other
hand, we do find illuminating discussions of important issues in moral theory, which, though
they may appear to be occasional pieces, show consistency and depth of thought and together
make up a coherent version of neo-Stoicism of sufficient interest to deserve more attention
that it has heretofore received in anglophone circles. This is the moral theory to be set out in
the present book.
In closing I should add that I have not been much concerned with such methodological
questions as whether I am writing as a historian of philosophy, as a historian of ideas, or
simply as one philosopher interested in another. I aim to be fair to Descartes and to write
sympathetically. As a philosopher, however, I am mainly interested in his arguments. While I
have taken some pains to provide a general context for his reflections on moral topics, I have
not done so in a manner that will satisfy the standards of a good historian of ideas. Whether
what remains is a good history of moral philosophy, I must leave to readers to decide. Many
of Descartes's assumptions and premises I do not accept, but for the most part I have left
criticism to one side and have tried only to make lines of argument clear. I have, however,
taken sides on several issues of interpretation. Finally, since the Cartesian texts dealing with
ethics are not generally familiar, I have been very generous with quotation.
-8-

Part I
THE MORALITY
OF THE "DISCOURSE"
IT has long been customary to take what Descartes in the Discourse calls a morale par
provision to be a provisional moralitya morality, that is to say, which Descartes felt obliged
to adopt as a stopgap measure to carry him through the period of doubt and which he expected
to replace with another and better morality at the end of inquiry. In line with this dismissive
view, commentators have not taken the morale par provision as something to be examined
seriously as a morality, or even as Descartes's morality. Some, encouraged by Descartes's
remark to Frans Burman, have gone so far as to say that he included it only as an afterthought
and then only to forestall the objection that his recommended method subverted not only
traditional philosophy and science but morality as well. i. Yet even if one grants that Descartes
had good reason to be cautious about what he said and did not say in the Discourse, we should
not take his comment to Burman to indicate that the commitments and attitudes expressed in
his morale par provision were not his commitments or his attitudes. And in fact what he wrote
later on ethics, to look no further, unequivocally bears this out. Moreover,
____________________
i.
In his conversation with Burman, Descartes is reported to have said, "The author does not
like writing on ethics, but he was compelled to include these rules [of the morale par
provision] because of people like the Schoolmen; otherwise, they would have said that he
was a man without any religion or faith and that he intended to use his method to subvert
them" ( Cottingham 1976, 49 ). Cottingham thinks that this remark provides "a fascinating
insight into ... Descartes's general attitude toward ethics" ( 118 ).
as I shall argue in Part One, once we take quite seriously Descartes's statement of his morale
and his commentary on it, we find the beginnings of a powerful moral conception, for which
he gives not a little supporting argument. There are many fine commentaries on this morale,
but none, I believe, does Descartes's morale par provision full justice.
-10CHAPTERONE
Descartes's Morale par Provision
I. ACCORDING to what I shall call the standard interpretation of the morale par provision as
Descartes sets it out in the Discourse, Descartes adopted a so-called provisional morality
simply as a prophylactic against the disorienting effect of methodic doubt on practice. This
view of the matter is not wrong, so far as it goes, but it is superficial. Not only does it not do
full justice to the text, it lacks sensitivity to the special situation Descartes found himself in
during the period in question. We need a more illuminating and more realistic account.
The standard view is this. When Descartes was twenty-three he decided to take up as a
lifelong project the search for certain knowledge in the sciences and, above all, in philosophy,
on which he took the sciences to depend. He had been disappointed by the philosophy he had
studied, complaining that he had found only controversy, preconception, and prejudice, but
nothing absolutely certain (6: 22 ; I: 122 ). What was lacking, he believed, was an effective
method of inquiryindeed, a method of the sort that was then beginning to take shape in his
own mind as he reflected on his earlier progress in mathematics. So, preparing himself

eventually to make progress in philosophy, he set about practicing and perfecting this method,
which, by the time he composed the Discourse, i. he had reduced to four rules: to break up
complex problems into simpler ones until the simplest
____________________
i.
He had made several earlier attempts to formulate this method, as we can see from the
Rules for the Direction of the Mind. For a brief account of the evolution of Descartes's
method in the Discourse, see Garber 1987.
-11elements are reached (the second rule); to recombine these simples in an orderly and
perspicuous manner in order to make the original problem clear and intelligible (the third
rule); and to make a careful review of these procedures to ensure that nothing has been
overlooked (the fourth rule). The crux of the method was the first rule, the rule of vidence:
"never to accept anything as true if I did not have evident knowledge of its truth: that is,
carefully to avoid precipitate conclusions and preconceptions, and to include nothing more in
my judgments than what presented itself to my mind so clearly and so distinctly that I had no
occasion to doubt it" (6: 18 : I: 120 ).
Descartes states emphatically, however, that he intended to restrict the scope of his first rule
to the pure search for truth, for he saw how foolish it would be to apply it to the practical
problems of everyday life, where we rarely if ever have the luxury of waiting until we have
certain knowledge before we act. For practice, as distinct from theory, he therefore suspended
the rule of evidence and followed instead his morale par provision. This, in brief, is the
standard account.
What is seldom noticed, howeverand what this brief account so far does not make clearis
that Descartes embraced this morale long before he decided to apply rigorously the rule of
evidence and to rid himself of his former opinions. Indeed, according to his own narrative, he
adopted his morale par provision at the time he dedicated himself to the search for truth,
which was in 1620. Yet he says quite explicitly that he did not at this time divest himself of
all his previous opinions (6: 15 ; I: 118 ). In his search for truth, he says, he resolved not to
proceed precipitately at the beginning. "Nor," he continues, "would I begin rejecting
completely any of the opinions which may have slipped into my mind without having been
introduced there by reason, until I had first spent enough time in planning the work I was
undertaking and in seeking the true method of attaining the knowledge of everything within
my mental capabilities" (6: 17 ; I: 119 ). So it is clear that he had his morale par provision in
hand well before he was prepared to abandon all the opinions he had hitherto accepted. Just
when this moment arrived is not clear from the Discourse, but we may suppose that it was in
1628, when he turned his attention to metaphysics. In any case, he relied on his morale not
only before he rid himself of all his earlier opinions but also afterward, and did so for the
special reason given above: the affairs of everyday life press us to decide what to do in the
absence of certain knowledge.
We suppose, then, a Descartes sufficiently practiced in his method that he is prepared to
implement the rule of vidence. He nonetheless restricts its scope specifically to pure inquiry,
where the sole aim is absolutely certain knowledge; and he continues to rely on his morale
par provision in the

-12affairs of everyday life, where the aim, broadly speaking, is to live well and happily. 2. To
illustrate the folly of not insulating practice from theory he gives the example of a traveler lost
in the middle of a forest who turns this way and that or, worse, stays in one place because he
does not know for certain which is the best way out and cannot settle on a plan of action. The
decisive traveler, by contrast, although similarly uncertain, elects to walk in a single direction
and keeps resolutely to his plan, just as if he knew for certain that his chosen path was the best
way out.
I shall return to this helpful illustration below. Less helpfully, however, Descartes first
introduces his morale using a different image:
Now, before starting to rebuild your house, it is not enough simply to pull it down, to make
provision for materials and architects (or else train yourself in architecture), and to have
carefully drawn up the plans; you must also provide yourself with some other place where
you can live comfortably while building is in progress. Accordingly, lest I should remain
indecisive in my actions while reason obliged me to be so in my judgements, and in order to
live as happily as I could during this time, I formed for myself a provisional moral code [une
morale par provision] consisting of just three or four maxims, which I should like to tell you
about. (6: 22 ; I: 122 ; my italics)
The Discourse is not the only place Descartes refers to such a morale; he mentions it again in
the letter-preface to the French edition of the Principles. If, he says there, one is to search
methodically for certain knowledge of first causes and principles, beginning only with such
ordinary wisdom as one gains from experience, conversation, and the best books, one "should
try before anything else to devise for himself a code of morals [une morale] which is sufficient
to regulate the actions of his life. For this is something which permits of no delay, since we
should endeavour above all else to live well." Referring back to the Discourse, he adds, he
has summarized there "the principal rules of logic and ... an imperfect moral code [une
morale imparfaite] which we may follow provisionally [par provision] while we do not yet
know a better one" (9B: 15 ; I 186-87). 3.
____________________
2.
Descartes repeats this point several times, often referring back to his discussion in the
Discourse. See "Second Set of Replies" ( 7: 149; 2: 106), "Fourth Set of Replies" ( 7: 248;
2: 172), and "Fifth Set of Replies" ( 7: 350; 2: 243)not to mention the letter-preface to
the Principles of Philosophy, to be discussed below. He writes in the First Meditation: "I
know that no danger or error will result from my plan, and that I cannot possibly go too
far in my distrustful attitude. This is because the task now in hand does not involve action
but merely the acquisition of knowledge" (7: 22 ; 2: 15 ).
3.
As he had earlier in the Rules for the Direction of the Mind, Descartes here says that
everyone should purge himself of his old opinions, at least on philosophical and scientific
matters,
-132. In both the Discourse and the Principles, Descartes may appear to restrict the need for a
morale par provision to those who take up pure inquiry. But the need is more general, since it
arises from doubts antecedent to pure inquiry, as I will explain below. First, however,

consider the question whether everyone who does take up inquiry must form the same morale.
In his introductory remarks to the Discourse Descartes suggests as much, since he says that
he derived (tire) his morale from his method. Yet he does not spell out how or just why the
morale he describes follows from the rules of method. Nor in the Principles does he point to a
tight connection between the rules of method and the precepts of a morale par provision;
indeed, if anything, he seems to allow that each pure inquirer might form a different morale.
It is not surprising, then, that commentators by and large agree that Descartes had to form
his morale only as a condition for getting on with pure inquiry and that he did not construct it
by following the rules of his method. 4. But more needs to be said on both these points. In the
first place, the need for a morale par provision does not arise initially from a com____________________
so that the need for a morale par provision would be universal. Moreover, the suggestion
in this passage is that it would be the same morale that each one needs.
4.
It is only in his note to the reader at the beginning of the Discourse that Descartes speaks
of deriving the morality from the method. The reader will find, he writes, "some of the
moral rules [the author] has derived from his method" (6:1; I:III). Genevive Rodis-Lewis,
the principal recent French commentator, writes of the morale that "it is not 'derived from
the method' as a consequence of the four precepts of the second part, but is rather called for
by them, since it permits their rigorous application" ( Rodis-Lewis 1970, 18.) Other
commentators who support substantially the same view include tienne Gilson and Pierre
Mesnard. Gilson writes: "As M. L. Lvy-Bruhl has pointed out ( Descartes, unpublished
lectures), morality must come last in the order of the sciences to be constructed in
accordance with the Cartesian method; this does not mean: 'A morality contructed by
means of the rules of my method,' but simply, 'It is the method that requires that we give
ourselves une morale provisoire, owing to its requirement that we place everything in
doubt, while the demands of ordinary life admit of no delay' " ( Gilson 1947, 81 ; also 23134). And Mesnard concludes: "The provisional morality draws its existence from methodic
doubt; its legitimacy has a two-part basis: it permits the mind to continue in peace the
pursuit of truth and the conquest of the sciences; it responds immediately to the urgent
character of the problems of everyday life" (Mesnard 1936, 53 ).
One who does not follow the prevailing view is Robert Cumming. Cumming's view is that
the morale par provision is Descartes's first construction of the method, first because the
rules of construction require that it be first. He distinguishes within the method the first
rule, which is a rule of verification, and the other rules, which he calls rules of
construction. Thus, according to the second rule, Descartes, as the first order of business,
divides each of the difficulties he needs to examine to see how best to resolve them. His
first difficulty, in view of his rule of verification, is to determine how best to deal with the
fact that he must act while he suspends belief in everything that is not evident. Cumming's
interpretation is highly ingenious, but not persuasive. Nonetheless, it is a move in the right
direction, since it seems that some methodologically based account of the content and
number of the rules of the morale par provision is in order. It is not enough to say that
Descartes needs some morality or other; a commentator needs to explain, if possible, why
we get the one we are given.
-14-

mitment to pure inquiry; it arises from more commonplace skeptical considerations. Moreover,
the connection between Descartes's method and the precepts that make up his morale is
tighter than his own remarks suggest.
3. Before we examine further what gives rise to this morale and how its precepts do (or do not)
depend on the rules of method, we should have them before us. Descartes says that his morale
is made up of "three or four" maxims, listing them in a way that leaves some doubt about the
status of the fourth.
The first was to obey the laws and customs of my country, holding constantly to the religion
in which by God's grace I had been instructed from my childhood, and governing myself in all
other matters according to the most moderate and least extreme opinionsthe opinions
commonly accepted in practice by the most sensible of those with whom I should have to live.
(6: 23 ; I: 122 )
My second maxim was to be as firm and decisive in my actions as I could, and to follow even
the most doubtful opinions, once I had adopted them, with no less constancy than if they had
been quite certain. (6: 24 ; I: 123 )
My third maxim was to try always to master myself rather than fortune, and change my
desires rather than the order of the world. (6: 25 ; I: 123 )
Finally, to conclude this moral code [cette morale], I decided to review the various
occupations which men have in this life, in order to try to choose the best. Without wishing to
say anything about the occupations of others, I thought I could do no better than to continue
with the very one I was engaged in, and devote my whole life to cultivating my reason and
advancing as far as I could in the knowledge of the truth, following the method I had
prescribed for myself. 5. Since beginning to use this method I had felt such extreme
contentment that I did not think one could enjoy any sweeter or purer one in this life. Every
day I discovered by its means truths which, it seemed to me, were quite important and were
generally unknown by other men; and the satisfaction they gave me so filled my mind that
nothing else mattered to me. Besides, the sole basis of the foregoing three maxims was the
plan I had to continue my self instruction. (6: 27 ; I: 124 )
____________________
5.
In part I, rather less generous about other callings, he makes the same point: "when I cast a
philosophical eye upon the various activities and undertakings of mankind, there are almost
none which I do not consider vain and useless. Nevertheless I have already reaped such
fruits from this method that I cannot but feel extremely satisfied with the progress I think I
have already made in the search for truth, and I cannot but entertain such hopes for the
future as to venture the opinion that if any purely human occupation has solid worth and
importance, it is the one I have chosen" (6: 3 ; I: 112 ).
-154. These are Descartes's precepts par provision. He compares them to a temporary dwelling
that he will occupy while he constructs a more permanent one. Fixing on this image, perhaps,
or for some other reason, commentators often refer to Descartes's morale as a morale
provisoire (provisional morality), even if Descartes himself does not. However natural morale

provisoire may seem as a gloss on Descartes's own morale par provision, it may be
misleading, 6. even if Descartes's image plays into it.
The metaphor of occupancy itself does not pose much difficulty, since it expresses the idea
that he lives within a set of rules, principles, and ideals that he has internalized and with
which he is, so to speak, at home. But owing to what having moral beliefs entails, a difficulty
arises from Descartes's suggestion that he could elect to occupy a morality on an interim basis
and as a matter of convenience, a difficulty exacerbated by the attendant image of his moving
out of his old morality and tearing it down. Of course, if one's idea of a provisional morality
were that of a body of optional precepts that one chooses to adhere to for reasons of
convenience, then Descartes's metaphor is apt. Choosing a set of precepts would be much like
deciding which one among several clubs to join. But such a body of precepts could not
properly be called one's morality, any more than could the rules of one's selected club.
To take the term fairly broadly at first, one's morality just is that constellation of beliefs,
values, and ideals to which one defers at the most basic level, both in guiding one's choices
choices of individual actions, specific plans, policies, and even styles of lifeand in
justifying these choices,
____________________
6.
From the last quarter of the nineteenth century French commentators have generally
referred to Descartes's morale as a morale provisoire, according to Michle Le Doeuff. Le
Doeuff writes: "One can establish from a seventeenth-century dictionary that 'par
provision' does not mean 'provisional.' It is a juridical term meaning 'what a judgement
awards in advance to a party'; thus one can award par provision a sum of a thousand
livres damages to a plaintiff suing for assault. The provision is not liable to be put in
question by the final judgement; it is first installment.... The expression 'provisional
morality' is a devaluing one; it designates something which is destined to be replaced,
something probably inadequate, to be invalidated once something better is found:
something which awaits its own rejection. Whereas the word provision signifies the validity
of this morality. There is in this term a positive notion of consumption.... Provision is that
which is, at most, capable of completion, but not withdrawal. Thus there is here a
difference of meaning regarding the value of this morality" ( Le Doeuff 1989, 60 - 62 ). I
am grateful to Jorge Secada for calling Le Doeuff's work to my attention.
English translations similarly skew interpretation. Among English translators we find
"provisional code of morals" ( Kemp Smith 1958), "provisional moral code" ( Descartes
1985), and "code of morals for the time being" ( Haldane and Ross 1911). To avoid
prejudicing the issue by translation, I shall continue to use morale par provision, hoping to
make clear just what Descartes's conception of this morale is.
-16both to oneself and, where called upon, to others. 7. Thus, if in some special context one were
obliged to select a particular set of rules under the rubic "provisional morality," one would
have to defer to one's basic morality to justify this choice. A provisional morality so
understood could not, therefore, be bedrock. To choose a provisional morality in this way
would presuppose one's already having a morality in the broad sense just defined.

The fact is that beyond a certain point Descartes's metaphor is inapt and does not fairly
represent the situation in which he finds himself on the occasion of taking up pure inquiry. If
we occupy a morality in the sense of having its values, goals, and precepts embedded in our
character, we cannot tear down that very morality. Tearing down our morality, it is scarcely
too much to say, would be tantamount to tearing ourselves downlosing our character and
identity. Of course, we who occupy a particular morality can understand that others "occupy"
other moralities. And we can understand how, through some voluntary efforts on our part, we
might effect changes in our morality. And perhaps this is something we can and often should
do. 8. But we can do so only a bit at a time and only under the direction either of the deeper
values in the morality we have or through some compelling new insight. Thus, whereas we
have no difficulty in understanding how and why one might move into a temporary home, we
cannot similarly understand how anyone could move out of one's morality into a temporary
one. 9. If, therefore, we are to make clear sense of the relevant texts in the Discourse as
autobiographical narrative, we need some different account of the derivation, formation, and
adoption of Descartes's morale than his occupancy metaphor suggests.
So, setting the image aside, we should take the qualifying "par provision" to signify,
unproblematically, that this morale is the one Descartes provides for himself during his search
for truth and that it is, as a morale, acknowl____________________
7.
This remark may need qualifying. I suppose that if one never looked further than one's own
subjective preferences to ground one's choices, one should not be credited with having a
morality, even in this broad sense. One who considers values as going no deeper or having
no basis other than this is one who has opted out of morality in this broad senseand it
may be in any sense. It is quite clear, however, that Descartes does not endorse such a
subjectivist view of value.
8.
It is one of Descartes's chief concerns in the Passions of the Soul to show how we can
improve our character through modifying our emotional dispositions. But he also had to
establish a normative standard of improvement.
9.
Even if we take the morale par provision to be in some relevant sense imperfect, we shall
require an intelligible narrative which builds either on the values this morality incorporates
or on new value discoveries that lead from it to a comparatively more perfect morality.
Here again, the latter cannot conceivably be embraced by a brute act of will or even on the
strength of some reasonable doubts about its being perfect.
-17edged to be in some respects doubtful and imperfect. It is not doubtful, of course, that he
needs this morale if he is to manage his everyday affairs well. This is true, however, whether
one is a pure inquirer or not. So it would seem to follow that everyone needs a morale,
although pure inquirers peculiarly sensitive to the doubtful and imperfect status of the
morale they have par provisiondiffer in this respect from less reflective members of society.
Nonetheless, from the inquirer's perspective (before the end of inquiry), everyone's morale
necessarily appears as par provision, even if those noninquirers whose morale is in view do
not themselves see it this way. Now, there is nothing odd about the notion of a morale par
provision if it is understood in this way. A morale par provision is just some person's de facto
morality seen as in some respects doubtful and imperfect. The contrasting notion is that of a
perfect morality, which is a de facto morality that is not, from this point of view, doubtful and
imperfect.

5. Unlike the question, What is the meaning of "par provision"? the question, Is the morale
par provision a morality? is seldom if ever asked. But it needs asking for at least three reasons.
First, it is not clear that a set of precepts adopted simply because one needs some rules to live
by, at least while one is ignorant of what the best rules might be, should count as a morality.
Much will depend on how this choice of precepts is itself grounded. Second, it is not obvious
that the precepts themselves are plausible candidates for the label "moral precepts." In fact,
the precepts are a mixed lot. We might well wonder, to take the most dubious case, how the
intention to take up science as one's life's work could be appropriately described as part of
one's morality. Finally, it is not clear how precepts that have no claim to generallet alone
universalvalidity or applicability can constitute a morality. In the next several sections and
in the following two chapters, I will clarify these matters and vindicate Descartes's description.
To begin with, it will be helpful to bear in mind a rough distinction, alluded to above, between
a broad and a narrow sense of morality. Taking the term in the broadest sense, one's morality
is a set of fundamental values expressed in one's character and, for the more reflective, in
one's conception of the good life. To formulate a proper conception of the good life was the
aim of philosophers within the tradition in which Descartes was himself working. 10. They
were asking, What is the good for man? What is
____________________
10.
In his Commentary, Gilson writes: "[ Descartes] remains in moral theory the heir of the
Greek ideal transmitted to the modern period by medieval scholasticism and the Christian
Stoicism of the I6th century; the final end of morality is for him, therefore, the supreme
good that alone can give us wisdom" ( Gilson 1947, 230). For an illuminating account of
the conception of morality among the ancients, see Annas 1993, 27 - 46 (a chapter entitled
"Making Sense of My Life as a Whole").
-18our supreme good? or, assuming that happiness is the supreme good, What is happiness?
What is involved in living the happy life? And they supposed that a single answer would
apply to all. Descartes came near the end of this tradition, which some judge to have come to
an abrupt end with Thomas Hobbes. We should have little difficulty in seeing how the
maxims that make up his morale par provision constitute his reflective answer to such
questions. To be sure, they are maxims formulated by someone who acknowledges that he
lacks perfect wisdom and the knowledge required to assert with absolute confidence that these
maxims define how to live as well and as happily as one possibly can, but they are
nonetheless of the right sort to constitute a morality in the broad sense.
But within this morality, broadly conceived, we find a part that comes closer to matching our
more modern conception of morality, a conception that is much narrower in scope, a
conception in which the concept of duty, for example, has a central place. This rough
distinction between a broad and narrow conception of morality was not, in fact, foreign to
ancient ethics and was well marked in the writings popular in Descartes's period. Pierre
Charron relied on it in organizing De la sagesse, and he probably took his lead from Cicero,
who wrote near the beginning of De officiis that "every treatise on duty has two parts: one,
dealing with the doctrine of the supreme good; the other, with the practical rules by which
daily life in all its bearings may be regulated." 11. In any event, although Descartes's morale
par provision is a morality in the broad sense, it contains within the first maxim a morality in

the narrower sense"the practical rules by which daily life in all its bearings may be
regulated."
6. To vindicate Descartes's description of his four maxims as a morality, I propose to take
them one by one, since what needs to be said differs in each case. I start with the fourth
maxim Descartes's resolve to spend his life cultivating his reason and searching for truth.
When Descartes introduces his morale he says that it contains three or four maxims, which
indicates that the fourth is in some general way different from the others. And it does appear
to be distinctive. First, it lacks the generality of the others, which look like rules for living
well that might be commended to all of us, whatever our particular vocation. Indeed, the
fourth maxim seems little more than Descartes's resolution to continue to devote himself to
scien____________________
11.
Cicero 1990, 9. Charron's work is filled with quotations from Cicero, among many other
ancient writers, so it is not in the least unlikely that he was influenced by the passage
quoted here. In this passage, "duty" translates officio, which is Cicero's Latin for the Greek
kathkonta and hardly has the meaning of "duty" in modern English. It means
"appropriate action" and includes strict moral requirements (in our modern sense), what
Kant called imperfect duties, and conduct that falls outside morality (again in our more
modern sense).
-19tific and philosophical inquiry as his life's work. Second, this maxim is the "sole basis" of all
the others. 12.
Descartes tells us that despite his studies and his travels, he still lacks a clear and distinct
vision of how to live with assurance. He also says that it is just such knowledge that he most
earnestly desires and that he has been encouraged by practicing his method to think that he
might, with diligence, acquire it. He is referring specifically to his state of mind during the
winter of 1619-20, but readers of the Discourse cannot but assume that this was also his state
of mind at the time of its composition. On the other hand, he is evidently not without settled
opinions about how to live and what is worthwhileopinions, moreover, that he expresses
with some warmth, as though he fully expected them to survive the critical scrutiny to which
he planed to subject them. One of these opinions is that contentment is worth pursuing for its
own sake. There are others. In justifying his choice of vocation, he refers to the value of the
knowledge he expects to obtain as well as to the value of the further fruits of this knowledge.
What brings him utmost contentment, he says, is advancing knowledge. But knowledge is
notor not justa means to contentment. In the first place, he strongly suggests that
knowledge has value for its own sake. And even if we discount this suggestion, he leaves no
doubt later, in part 6 of the Discourse, about the instrumental value of knowledge and the
value of the ends to which it is the means. Here he refers to the advances in knowledge he had
made by 1637:
they opened my eyes to the possibility of gaining knowledge which would be very useful in
life, and of discovering a practical philosophy which
____________________
12.
Now it might seem to follow, if we combine these two points, that only one who takes up

methodic inquiry as a vocation would he bound by the three maxims. And if we assume
that Descartes's choice to take up methodic inquiry is not one that we are in any way bound
to imitate, then these maxims will not have the general validity I spoke of above. However,
engagement with inquiry admits of degree. Descartes is an extreme case. So, assuming that
Descartes is correct that engagement with inquiry is a sound basis for the three maxims, it
remains to be seen what our level of engagement should be and therefore what normative
force these maxims will have. One message of the Discourse, however, although disguised
by Descartes's autobiographical presentation, is that so far as we ourselves lack a clear
and distinct understanding of how best to conduct ourselves in this life, we are all in need
of une moral par provision. This line of argument is explicit in the letter-preface to the
French edition of the Principles of Philosophy and is foreshadowed in the Rules. In the
passage from this preface quoted above Descartes describes the morale formulated in the
Discourse as "an imperfect moral code which we may follow provisionally while we do not
yet know a better one" [une morale imparfaite, qu'on peut suivre par provision pendant
qu'on n'en salt point encore de meilleure]. And if this is so, then we may infer that we
should form for ourselves the very one Descartes formed for himself. And the reason for
this, in turn, is that it is our reasonthe same in each of usthat governs this formation.
What is recommended in the letter-preface is serious methodical philosophical reflection
through (Cartesian) metaphysics, which will gain for us the most important truths we need
to know in this life. See the letter to Elizabeth, September 15, 1645 (4:290-96; 3:265-67).
-20might replace the speculative philosophy taught in the schools. Through this philosophy we
could know the power and action of fire, water, air, the stars, the heavens and all the other
bodies in our environment, as distinctly as we know the various crafts of our artisans; and we
could use this knowledgeas artisans use theirsfor all the purposes for which it is
appropriate, and thus make ourselves, as it were, the lords and masters of nature. This is
desirable not only for the invention of innumerable devices which would facilitate our
enjoyment of the fruits of the earth and all the good we find there, but also, and most
importantly, for the maintenance of health. (6: 62 ; I: 142 -43)
To be sure, in 1619, when Descartes chose his vocation, he had not yet made the discoveries
he refers to here, so we should perhaps not see humanitarian considerations as playing a major
role in his original choice. On the other hand, since he is still strongly committed to his
vocation in 1637 and to his morale par provision, we should take these unequivocal
expressions of humanitarian values very seriously. In explaining his decision to publish the
Discourse (and the accompanying essays) he says that not to publish would be to sin "gravely
against the law which obliges us to do all in our power to secure the general welfare of
mankind" (6: 61 ; I: 142 ). 13. The law that would require publishing the results of inquiry,
however, would also warrant undertaking the inquiry in the first placc. 14.
The main point is that Descartes's justifies his fourth maxim by referring us to a moral law
that he warmly embraces. It is evidently a law that belongs to Descartes's own morality.
Which morality? Certainly to his morality at the time of composing the Discourse. Is this
morality his morale par provision? It would appear that the answer must be yes in one sense
and no in another. On the one hand, since he represents his own morality at this time as une
morale par provision, we must infer that this law is part of this moral
____________________

13.

A few pages later he writes: "Every man is indeed bound to do what he can to procure the
good of others, and a man who is of no use to anyone else is strictly worthless" (6: 66 ; I:
145 ).
14.
Objecting to Gilson, who suggested that Descartes pursued philosophy basically for
philanthropic reasons, Ferdinand Alquiwrites: " Descartes seems less desirous of
spreading truth than of acquiring it, less concerned to make it known than to make use of it
himself, both to know better and to know himself better" ( Alqui1988, I:565). Of course,
Descartes may have had several motives both for doing research and publishing. To be
sure, he had a great thirst for discovery and invention, both sources of his greatest
contentment, as we have seen. But this takes nothing from Gilson's point. In the first place,
Descartes is quite explicit about the law to secure the general welfare of mankind. And if,
as he believed, by choosing his special vocation he would help to secure this worthy end,
then he could not fail to see this as a further reason for this very choice. The point I am
concerned to make in the present context, however, is that Descartes in no uncertain terms
declares his allegiance to a specific moral law to secure the welfare of mankind.
-21ity. On the other hand, since his fourth maxim is the ground of the others and this law is in
turn the ground of the fourth maxim, the law might appear at once to stand outside the morale
par provision and to serve as its moral foundation, a foundation unqualified by "par
provision." But the correct way to view the matter is that the law is neither more nor less par
provision than the other elements of Descartes's morality. "Par provision" qualifies the
morality as a whole and signals that it is a morality yet to withstand critical inquiry.
To conclude, when we look to the reasons Descartes gives for his choice of vocation, we see
that they are moral reasons. They are moral reasons in the sense that they refer us to values
that are embedded in Descartes's own morality, taking morality in both the broad sense and
the narrow. When I speak of Descartes's own morality, I mean his pre-doubt morality, the
morality he had before he committed himself specifically to his scientific and philosophic
vocation. If, therefore, the choice of vocation is, as he says, the sole basis for the three
maxims of his morale par provision, it follows that this specific morale has a foundation in
Descartes's pre-doubt morale. And if this is correct, then we have a good part of our solution
to the first problem. We can see how the formation of his morale par provision is continuous
with his own morality at the time of its formation. It is significant, moreover, that these values
are not idiosyncratic. The rationale for his choice of vocation is not only intelligible but also
sufficiently justified in terms of values he could count on his readers to share.
7. If, Descartes says, he needs some precepts to guide his practical deliberation while his
method requires him to suspend belief in theoretical matters, it is not the case that his doubt
about morality first arises from the hyperbolic doubt of the method. Nor is it this methodic
doubt alone that calls for a morale par provision, for Descartes has ample reason, prior to
methodic doubt, to question his moral opinions: a battery of skeptical arguments bequeathed
to him by Sextus Empiricus and vivified by Montaigne. So although Descartes says that he
required a morale par provision as a condition for his search for truth, we find by following
his own narrative account that the need for such a morale actually antedates the
methodological skepticism of his method. Methodic doubt is procedural and confined to the
context of theory. The pure inquirer asks what, if anything, can be known beyond the
possibility of doubting; he has doubts about matters that at the level of common sense are
morally certain. But doubts about our own moral outlook arise for us whether we are pure

inquirers or not. Indeed, the hope of the inquirer is, in the end, to put to rest those doubts that
have their origin in skeptical reflections of a more traditional, less exotic kind. And it is in
response to such skeptical arguments that the need
-22for a morale par provision initially arises, for in taking these arguments to heart, which
Descartes claims to have done in the Discourse, he also had to take a critical distance from
his moral convictions, no matter how well embedded they were in his affective and conative
dispositions and no matter how warmly he endorsed them. To be sure, Descartes says he
adopted his morale at the time he decided definitely to take up the methodical search for truth
as his life's work. 15. This does not imply, however, that the need for this morale first arose
from methodic doubt. After all, the Pyrrhonist considerations summarized in the Discourse
are presented as shaking his confidence only in areas that lacked the perspicuity and vidence
of mathematics. In his method he took the clarity of mathematical knowledge as his standard
and sought to find this same standard satisfied elsewhere. 16.
This is the context in which he places himself as he rehearses the traditional skeptic's case.
"The same man," he writes, "with the same mind, if brought up from infancy among the
French or Germans, develops otherwise than he would if he had always lived among the
Chinese or cannibals.... Thus it is custom and example that persuade us, rather than any
certain knowledge" (6: 16 ; I: 119 ). 17. Certain knowledge, for its part, is not the product of
erudition, for, as Descartes writesexpressing a sentiment he might have found in Cicero,
Montaigne, or Charron"Nothing can be imagined which is too strange or incredible to have
been said by some philosopher" (6: 16 ; I: 118 ). And nothing, he continues, can be imagined
which is too outrageous not to have been commended by some moralist. The ancients, he says,
"extol the virtues, and make them appear more estimable than anything else in the world; but
they do not adequately explain how to recognize a virtue, and often what they call by this fine
name is nothing but a case of insensibility, or pride, or despair, or parricide" (6: 8 ; I: 114 ).
18.

____________________
15.
This would imply, we may note in passing, that he had by then formulated its precepts in
something close to the form presented in 1637. Since the formulations of the maxims of his
morale are very close to those to be found in writings available to Descartes in 1619by
Michel de Montaigne, Pierre Charron, and Guillaume Du Vairthis is a plausible
conjecture.
16.
The methodic doubt described in the Discourse is not the full, hyperbolic doubt of the later
Meditations, where even the standard of vidence is called into question.
17.
From studying the great book of the world with its great variety of customs and opinions,
Descartes writes, "I learned not to believe too firmly anything of which I had been
persuaded only by example and custom" (6: 10 ; I: 116 ). Here Descartes is following the
path of Montaigne in his " Apology for Raimond Sebond," Charron in the second book of
On Wisdom, and, in general, the Pyrrhonian tradition of Sextus Empiricus, a point that
could hardly have been lost on Descartes's early readers.
18.
He is alluding to the Stoic rejection of the passions, to the pride connected with this
rejection, to the Stoic acceptance of and even glorifying of suicide, and to L.-J. Brutus's
condemning his own children to die, presiding at their execution, and to M.-J. Brutus's
killing of Caesar (see Gilson 1947, 131 -32). For a discussion of Descartes's own attitudes
toward Stoic insensibility and pride, see D'Angers 1976.

-23Far from invoking methodological doubt in these passages, Descartes professes to be moved
by the more common argument from relativity. We should, he concludes, take to heart that
reasonable individuals and peoples differ about how to live well and happily, and we should
concede that we ourselves are no more qualified to dismiss their opinions as absolutely false
than they are ours. 19.
Yet if Descartes's uncertainty echoes that of Montaigne and others, his characteristic cognitive
optimism does not; for, unlike the skeptics, he had in hand a standard of truthnamely,
videnceby which, at the end of methodic inquiry, he expected to discover a firm
foundation for a true morality, a morality beyond the threat of skepticism. And it is in the
light of this possibility that he characterizes the morality he has as une moral par provision.
Thus, to the question raised above about the sense in which Descartes's morality is par
provision, part of the answer is that a morality par provision is a morality that one keeps at a
critical distance because of the force of skeptical arguments. To describe it as par provision is
also, in part, to express optimism about eventually being able to answer these arguments.
____________________
19.
The first modern edition of Sextus Empiricus's Outlines of Pyrrhonism was brought out by
Henri Etienne in 1562; it soon became "the dominant philosophical text of the age,"
according to J. Annas and J. Barnes. Its literary influence is seen in Montaigne's "
Apology," which contains many of Sextus's skeptical arguments, and is passed on through
Charron's On Wisdom. Descartes complains in his reply to the Second Objections that
skepticism still flourishes. His own aim was not only to defeat it but also to defeat
distinctively epistemological skepticism (see Annas and Barnes 1985, 5 - 6 ).
Richard Tuck also notes that Descartes's moral skepticism does not derive from hyperbolic
doubt: "Two problems remained about Descartes's reply to the sceptic. The first was that
even if his argument against his own hyperbolical doubt was accepted, it was not clear
how far it met the ancient sceptical case, which had after all not been based on the
hyperbolical doubt. Moral relativism in particular seemed on the face of it to be untouched
by Descartes's argument, for as we have seen, in ancient scepticism it was wholly
independent of any epistemological doubt. That is why there is, as is well known, no
developed moral theory in Descartes's oeuvre, though he did say that his programme might
eventually, and in some unspecified way, issue in one. Because of the absence of any such
theory, we find in general in his writings only the sceptical premises, and not the antisceptical conclusions drawn from them which we might have expected, and which
Descartes intended to draw one day" ( Tuck 1988, 247). Tuck goes on to suggest that
Descartes remained a moral skeptic, since his resolution of hyperbolic doubt, which could
give him physics and metaphysics, could not get him ethics. "There was nothing in our
assured knowledge of the physical world which could tell us anything about the moral
world, despite Descartes's hope that the link could be made. So he remained a pretty pure
sceptic in these matters, a point underlined by the fact that the only political philosopher
he was at all drawn to was Machiavelli, just as Charron and the others had been" (248).
Although I agree with Tuck that Descartes's morale par provision is not motivated by
hyperbolic doubt, I would reject his claim that Descartes remained a moral skeptic, as well
as his claim that he was "drawn to" Machiavelli. The one moral philosopher he expresses
some grudging respect for is Hobbes, although, as I will argue later, he does not seem to
have learned much from him.

-248. We turn now to the first maxim, according which Descartes resolves (a) to obey the laws
and customs of his country; (b) to continue to hold firmly to his Catholic faith; (c) in all other
matters, 20. to govern himself "according to the most moderate and least extreme opinions
the opinions commonly accepted in practice by the most sensible of those with whom [he]
should have to live." Of these, in the present context only the third part merits much
discussion. The first part is plainly justified on prudential grounds and presumably on moral
grounds as well. 21. More difficult to explain is the second part. If Descartes exempts articles
of his faith from doubt, we may be inclined ask, does he not already have a morality sufficient
to guide him in the conduct of everyday life? 22. The answer, of course, will depend on what
range of activities he takes this morality to apply to. It will certainly control specifically
religious practice. But beyond special duties of worship, confession, and penance, Christian
moral doctrine commends other duties, as well as ideals of character. These make up the body
of common morality, a morality that is not specifically religious, much less specifically
Catholic. To simplify exposition, then, I shall distinguish between Descartes's strictly
religious activity and the nonreligious activity of everyday life. I shall restrict the scope of
part b of the first maxim to the formerfor the following reasons. First, Descartes made a
radical distinction between articles of faith and beliefs with other credentials. He held,
implicitly in the Discourse, explicitly in the later Principles of Philosophy, that faith
transcends methodic doubt and the kind of evidence it aspires to (8A: 39 ; I:221-22). 23. His
method did not, therefore, require that
____________________
20.
Where custom ends and other matters begin is not carefully defined, nor need it be for
Descartes's purposes. Other matters will certainly include areas of conduct properly
regulated by maxims of prudence and precepts of morality. Thus, in some sense, Descartes
elected to take his moral instruction from the example of the more sensible of those with
whom he had to live. One issue I examine is precisely what this sense is.
21.
In fact the public practice of the Catholic faith was proscribed in Holland, where Descartes
was living during much of the time between 1620 and 1637, but he continued to practice it
nonetheless. See Rodis-Lewis1970, 19 n. 1.
22.
As one commentator, Alfred Espinas, writes, "He sets aside religion purely and simply
with all its consequences. It is therefore useless to look for a morality; he has one." Espinas
is among those commentators who believe that all talk of a provisional morality as required
by the method is pretense. "The provisional, nearly fictional character of these rules," he
continues, "is not to be doubted. It is not without some irony that Descartes presents to his
reader this morality that lacks any duty worthy of the name; as he says in his private notes,
personam induit, he wears a mask" ( Espinas 1925, 18 ).
23.
For a thorough examination of Descartes's religious thought, see Gouhier 1924. Speaking
of the young Descartes, Gouhier endorses Gaston Milhaud's view: "Apropos of the
Descartes of 1619, Milhaud reached a conclusion full of good sense; he had 'a naively
religious cast of mind, simpler and less complex that we are generally disposed to think'"
( Gouhier 1924, 50 ). For a more recent account, see Rodis-Lewis1995, 284-97.
-25he abandon the articles of his faith qua articles of faith; it required only that he not rely on
faith in his search for truth. The second and related point is that his project is rationalist; his
aim, so far as ethics is concerned, was not knowledge based on faith and, a fortiori, not moral

knowledge based on faith; his aim was an ethics based on what can be discovered through the
use of reason. So it is fitting to give a restrictive reading to the second part of the first maxim
and an inclusive reading to the third part, to which I now turn.
In all matters not covered in the first two parts of the maxim, Descartes resolved to follow the
example of others. This part I take to cover much of the activity governed by ordinary
moralitythe normal range of duties to self, to others, and within social institutions, such as
the duty not to steal. It also includes not only matters of ordinary prudence but also the full
range of beliefs about ourselves and the world of which we can claim moral certainty. 24.
The ostensible reason Descartes offers for deferring to the example of moderate and sensible
people is that he had begun (in 1619) "to count [his] opinions as worthless." Why? "Because,"
he says, "I wished to submit them all to examination, and so I was sure I could do no better
than follow those of the most sensible men" (6: 23 ; I: 122 ). Now, this makes it look as if
Descartes actually chose to divest himself of his morally certain beliefs about the world and
of his pre-doubt moral beliefs, deferring instead, in the relevant areas, to the example of
sensible people. To be sure, both Descartes's
____________________
24.
In this first maxim, then, Descartes is, through his alleged deference to the views of
sensible people, accepting certain beliefs as morally certain. Thus, the practical domain is
insulated from the corrosive effects of methodic doubt in the theoretical domain. Referring
to the three maxims, he says toward the end of part 3: "Once I had established these
maxims and set them on one side together with the truths of faith ... I judged that I could
freely undertake to rid myself of all the rest of my opinions" (6: 28 ; I: 125 ). In part 4 he
writes: "if there are still people who are not sufficiently convinced of the existence of God
and of their soul by the arguments I have proposed [he has just given his famous
arguments], I would have them know that everything else of which they may think
themselves more suresuch as their having a body, there being stars and an earth, and the
likeis less certain. For although we have a moral certainty about these things, so that it
seems we cannot doubt them without being extravagant, nevertheless when it is a question
of metaphysical certainty, we cannot reasonably deny that there are adequate grounds for
not being entirely sure about them. We need only observe that in sleep we may imagine ..."
(6: 37 - 38 ; I: 129 -30). In the French version of the Principles he explains that "moral
certainty is certainty which is sufficient to regulate our behaviour, or which measures up to
the certainty we have on matters relating to the conduct of life which we never normally
doubt, though we know that it is possible, absolutely speaking, that they may be false"
(8A:327; I:289, 289n). Descartes's examples of moral certainty are restricted to beliefs
about the world; he does not use the expression to refer specifically to our moral beliefs.
For a good account of Descartes's notion of moral certainty, see Curley 1993, 19 - 20.
-26method and the prior argument from relativity required him to gain some critical distance
from any of his prior beliefs that lacked vidence, including his moral opinions. But holding
one's moral opinions at a critical distance is one thing, getting rid of them altogether (however
this might be done) is quite another, and not something that is in the least called for by the
desire to submit them to examination. Nor does the first maxim call for getting rid of them.

To see this, let us examine this part of the first maxim more closely. What would be involved
in identifying a group of people as moderate and sensible? Descartes describes the sensible
people whose example he chooses to follow as the most moderate in their conduct. Their
opinions, which their behavior manifests, are, he says, (a) probably the best, (b) the easiest to
act on, and (c) always closer to the right opinion than the opposing extreme. In large measure,
a, b, and c are analytic of the evaluative term "sensible." But it is Descartes who must first
determine who the sensible people are; it is he who must decide which opinions are moderate
and easiest to follow. How is he to reach such a momentous decision except by relying on his
own pre-doubt moral opinions? The sensible people, it would seemthose whose moral
beliefs are worthy of allegianceare those who by and large share with Descartes a common
morality. If this is right, Descartes's morale par provision, insofar as it includes a set of firstorder rules of right conduct, is continuous with his own pre-doubt morality.
And this has to be right, since the alternative is not credible. The alternative would require
that the following case be fully coherent, which it is not. The case is one in which there was a
notable divergence between Descartes's pre-doubt moral opinions and the moral opinions of
the (so-called) sensible people in question. Suppose most members of a group, identified in
some way as sensible, for example, through their manifest intelligence and their political and
economic power within the larger community in which Descartes livedsuppose they
rejected a precept that was embedded at a deep level in Descartes's pre-doubt morality, say,
that one ought to keep promises or that one ought to help those in need. Let us call them
sensible knaves. These sensible knaves appear to enjoy an agreeable style of life. In their view,
keeping a promise or helping someone in need would, when such an act lacks evident
prudential point, be extreme conduct (a kind of foolish moral fanaticism). Now if, as some
commentators argue, Descartes's decision to follow the example of the more sensible
members of his community were merely a matter of social conformism, then we could easily
imagine him adopting the ways of sensible knaves. But this we cannot easily imagine. And
there is nothing in the text that compels us to try.
-27Although Descartes has well-considered doubts about his own first-order morality, these are
not of the kind to undermine moral convictions or to leave him free to follow the example of
sensible knaves.
If this is a correct interpretation of Descartes's reflective understanding of his own moral
convictions, however, we confront the following interpretive problem. Why does he not
simply construct his morale par provision from his own pre-doubt moral convictions? Or, to
keep our focus on the first maxim, 25. why does he say that he will follow, not his own predoubt moral beliefs, but rather those of the most sensible men in his society?
Perhaps he proposes to do so as a preferable alternative to continuing to rely on his pre-1620
moral convictions. But since we cannot make clear sense of the proposal so construed, we
need to view his choice of moral paradigm in some other way. And what this way is should
by now be clear. What Descartes is announcing in this first maxim is that he takes himself to
be a member of a community of people bound together by a common morality. His policy is
understandably conservative, for although he doubts that this common morality is the correct
or best possible morality in all respects, he has as yet discovered no grounds for proposing or
endorsing any other. 26.

Nothing in what I have argued prevents Descartes from seeking advice from sensible people,
time permitting, when he faces a complex moral problem whose best resolution is not clear to
him. Some moral demands may be clear and have straightforward applicationspublishing
the Discourse, for
____________________
25.
I will point out below that the second maxim of the morale par provision does directly
meet the standard of the first rule of method and can be seen, therefore, to be derived from
the method in a direct and straightforward manner. It is consequently firmly established as
true.
26.
Descartes's first maxim bears an obvious resemblance to the following passage from
Montaigne's " Apology for Raimond Sebond," which comes near the end of a
comparatively sympathetic account of Pyrrhonism. "For, however much new fashions may
appeal to me, I do not readily change, for fear of losing by the exchange. And since I am
not capable of choosing for myself, I accept the choice of others, and remain in the state
wherein God has placed me. Otherwise I could not keep from perpetual rolling. Thus, by
the grace of God, I have kept wholly, without being stirred or troubled by conscience,
within the ancient tenets of our religion, amidst the many sects and divisions that our times
have brought forth" ( Montaigne 1927, 2: 14 ). I would make three points. First, Montaigne
clearly is capable of choosing, since he embraces his religion and he voluntarily accepts the
moral precepts current in his community. Second, if in so doing he is not obliged by his
reason, he is nonetheless deferring to his reason in this sense, that he is able to defend his
choice as not unreasonable and as most consonant with his character as formed so far. He
represents his practical beliefs and attitudes as at once genuinely his and as maximally
justified relative to the only standards of justification accessible to him or to anyone else.
Third, Montaigne expresses the need for such beliefs and attitudes "to keep himself from
rolling about incessantly." We could call these beliefs and attitudes, therefore, une moral
par provision and see that the need for it arises from Pyrrhonian considerations. As we
shall see, however, Descartes's position is not Pyrrhonian.
-28examplebut others may not. In such cases consideration of what others who are evidently
sensible would do or would advise is itself sensible. Taking sensible people as moral advisers
in this sense, however, is altogether different from the wild and psychologically unrealistic
suggestion that one is to uproot all one's prior moral beliefs and replace them with someone
else's. We have to have confidence in those we listen to, and this confidence must itself be
rooted in our own moral judgments. 27.
9. Up to this point I have taken Descartes's reasoning to have run along the following lines.
Before we come to doubt our moral beliefs and see them as possibly false, we are practicing
dogmatists and our beliefs have the form of precipitate conclusions and preconceptions, as
beliefs passively and uncritically acquired. We are struck, however, by the diversity of social
norms across cultures and are led to ask whether our own moral beliefs and attitudes may not
be the product of mere custom and example. 28. Taking this point to heart, we immediately
place our moral convictions at a critical distance; we cease to be dogmatists and acknowledge
our liability to error, a liability we now seek to overcome through methodic inquiry.
Specifically within the context of inquiry we refuse to assent to any claims, moral or
otherwise, that fail to meet the standard of vidence; but within the larger context of everyday

life, we retain our allegiance to the social norms we have previously embraced and do so with
the decisiveness of one who knows them to be certified by reason.
Does Descartes's reasoning go further? If we stick to what is explicitly spelled out in
Descartes's fable, perhaps not. Still, without going beyond what we can glean from relevant
texts, we can extend it in an interesting way. I begin with this question: If we are beginning
pure inquiry, are we justified in retaining allegiance to these social norms?
Suppose that we have had the full use of our reason from birth and that we have always been
guided by it. And grant that the morality we formed
____________________
27.
In his later correspondence with Elizabeth ( August 18, 1645), commenting on Seneca, he
writes: "he chides those who follow custom and example rather than reason. 'When it
comes to how to live,' he says, 'people rely on mere beliefs, never on sound judgement.'
Nevertheless he approves of our taking the advice of those whom we believe to be the
wisest persons; though he would have us also make use of our own judgement in
examining their opinions. Here I am strongly of his opinion. For although many people are
incapable of finding the right path on their own, yet there are few who cannot recognize it
well enough when somebody else clearly points it out to them. Moreover, provided we take
care to seek the advice of the most able people, instead of allowing ourselves to be guided
blindly by example, and we use all our mental powers to discover how we ought to
proceed, then however things may turn out, our consciences will be at peace and we shall
have the assurance that our opinions on morality are the best we could possibly have"
(4:272; 3:259).
28.
As Descartes engagingly puts it at one point, his opinions "may have slipped into [his]
mind without having been introduced there by reason" (6: 17 ; I: 119 ).
-29under these conditions is the perfect morality Descartes refers to in the Principles. 29. Since we
are presently in search of the perfect morality, we acknowledge that the morality we actually
have is imperfect. It is imperfect in two respects. First, even if its requirements and ideals are
those that a perfect morality would contain, we do not know this to be so. For example,
although we believe we should keep our promises, we do not know for certain that we should
do so; for when we reflect on the matter, we see that our having acquired this belief might be
explained in a manner that does not require that it be true. Second, the requirements and
ideals of our de facto morality may in fact be different in content from those of the perfect
morality. Even if we can scarcely imagine a perfect morality not containing the obligation to
keep promises, we must admit this possibility, not to mention the further possibility that it
would contain demands we have never thought of. Our question is this: how can we justify
our continuing allegiance to a morality we acknowledge to be imperfect? This is a question
Descartes himself does not expressly address in just this form. The question he addresses is
how to justify his choice to follow the example of sensible men. But as I have argued above,
what he offers in defense of this decision rests on his de facto morality. So our question
cannot be set aside.
Nor does Descartes set it aside, since he decides to follow the example of those who are not
only sensible but whose practical norms are moderate, easiest to follow, and probably the
best (6: 23 ; I: 122 -23). Putting this together with the above conclusionthat is, that

Descartes, by choosing to follow the example of sensible and moderate men, was in fact
deliberately retaining allegiance to his former moralityour question becomes: How can he
support the claim that his de facto morality has some antecedent probability of being true? I
construct a Cartesian answer to this question in three stages.
To review the relevant points, I have argued that (a) Descartes retains, as part of his morale
par provision, the substantive norms and ideals of the morality he had when he was twentythree; (b) he shares this morality with the sensible and moderate members of his society; (c)
he regards this morality as imperfect; and (d) he regards retaining allegiance to it as justified,
imperfect as he acknowledges it to be. Now, the first stage of my answer is this: we could
justify retaining our allegiance to a morality under
____________________
29.
Referring to the early reflections of 1620 or thereabouts, Descartes writes: "So, too, I
reflected that we were all children before being men and had to be governed for some time
by our appetites and our teachers, which were often opposed to each other and neither of
which, perhaps, always gave us the best advice; hence I thought it virtually impossible that
our judgements should be as unclouded and firm as they would have been if we had had
the full use of our reason from the moment of our birth, and if we had always been guided
by it alone" (6: 13 ; I: 117 ).
-30conditions a, b, and c because we have no reason to declare allegiance to another (waiving the
question of how this might be possible psychologically). That is, we might justify this
morality merely as our default morality. This is the Pyrrhonist position. The Pyrrhonist makes
no claims that his default morality has any antecedent probability of being true and is content
in his conviction that no such claims on behalf of anyone's morality can be successfully
defended either. Justified probability claims must rest on background knowledge, and, says
the Pyrrhonist, we have no such knowledge. But this does not appear to be Descartes's
position. Although he certainly could defend his continuing allegiance to his former morality
as a default morality, Descartes, unlike the Pyrrhonist, speaks credulously of antecedent
probability.
This brings me to the second stage of my answer. What Descartes has and the Pyrrhonist
lacks is confidence in the epistemic potential of human reason (6: 2 ; I:III). Descartes
attributes the diversity of opinion on matters of importance to our failure to proceed slowly
and methodically, a failure he would no doubt explain in terms of our ignorance of the proper
method, the power of desires and traditions to influence our beliefs, and our general
intellectual laziness and diffidence. Nonetheless, he takes the truths of reason to be permanent
and apprehensible by those who use their reason well. This being granted, if we could pick
out, among all the conflicting moral opinions we find in our history, some core of moral
opinion shared by the best thinkers, then, given that human reason is a genuine cognitive
power, we have some grounds for concluding that these widely shared opinions are at least
approximately truethat they have some antecedent probability.
So we come to stage three. Can we not discover thinkers in this history who have done a
creditable job of using their reason well, even if not perfectly, and who have bequeathed to us
a body of first-order moral doctrine on which there is significant consensus? The major
controversies, after all, have not been over first-order morality but over how the precepts and

ideals of first-order morality are to be explained or established. Even in the midst of

rehearsing the traditional skeptical arguments, Descartes praises the ancient moralists for their
"useful teachings and exhortations to virtue," complaining chiefly, as we have seen, that none
have provided an entirely adequate foundation for their teachings and that some have given an
implausible taxonomy of the virtues. Moreover, in the earlier Rules for the Direction of the
Mind, a work that gives us some insight into Descartes's views during the period we are
considering, he writes as the good rationalist he is:
But I am convinced that certain primary seeds of truth naturally implanted in human minds
thrived vigorously in that unsophisticated and
-31innocent ageseeds which have been stifled in us through our constantly reading and hearing
all sorts of errors. So the same light of the mind which enabled them to see (albeit without
knowing why) that virtue is preferable to pleasure, the good preferable to the useful, also
enabled them to grasp true ideas in philosophy and mathematics, although they were not yet
able fully to master such sciences. (10:376; I: 18 )
Now, assuming that Descartes's own first-order morality is part of this overlapping consensus,
I conclude that as a rationalist, he can justify his retaining allegiance to his old first-order
morality not merely faute de mieux but because, compared with others with which it not
compatible, it has a higher antecedent probability of being at least approximately true.
10. I turn finally to the question I raised in the beginning: What role do the rules of method
play in Descartes's formation of his morale par provision? In the first place, Descartes's
commitment to the method is motivated by his morality, for the method is the means he has
discovered for arriving at truth, and truth he values as an end among other ends. His prior
value commitments are expressed in the fourth maxim and elsewhere, as I have noted above.
But how does allegiance to the four rules of method determine the remaining three maxims of
the morale par provision?
The rule of evidence is the first rule of the method, but it is not always the first rule to be
applied. In the present case, the first to be applied is the second rule, which requires us to
analyze any problem or difficulty into its simplest elements. If this is not an innovative rule, it
is certainly a sensible one, as far as it goes, although it is plain that we shall need practice in
applying it to a variety of subject matters. The third and fourth rules are no more precise.
According to the third we are to proceed in an orderly way in reconstructing the original
complex or in solving the original difficulty, and according to the fourth we are to review our
effortsour analyses and step-by-step reconstructionto make sure we have left nothing out.
The first problem we face when we take up inquiry is a practical one. According to the second
rule, we need to analyze this problem. What are its elements? We have a declared practical
aim: truth in the sciences (including morality). We have a method that promises to lead to this
declared end. It is reasonable to choose the means, other things equal, to our ends. But the
methodical search for truth is itself a practical activity which can take place only if certain
other conditions are satisfiedleisure, freedom from interference, adequate food and shelter,
help and cooperation of others, and the like. The search itself, moreover, is not only
subordinate to specific moral aims, it is one among other practical activities with their own
independent aims, and all these activities must be coordinated.

-32What stands out clearly, when we survey this problem, is a distinction between the domain of
theory and the domain of practice. And what is equally clear, as Descartes emphasizes, is that
while we must scrupulously respect the rule of evidence in the domain of theory, to do so
prematurely in other areas of practice (theorizing being one practice among others) would be
self-defeating. If the analysis I have offered is correct, it follows that in the affairs of everyday
life we are to continue to act in the light of our moral beliefs as well as our beliefs about the
world forged by ordinary experience. We are to do so because that is the best we can do in our
present circumstances, that is, prior to whatever practical knowledge inquiry brings. Among
those beliefs we might imagine ourselves as having, these are the ones that we judge, on
grounds of overall coherence, the most likely to be correct, the most probable.
Let us assume, then, that this is the best we can do in our present circumstances and that we
have in accordance with the applicable rules of method solved our first problem, which was a
practical one. Then we can say that the first maxim of the morale par provision has been
derived from the method. When it comes to the second and third maxims, however, more can
be said. Take the second maxim. This maxim, as I shall argue more fully in the next chapter,
follows directly from the rule of evidence. The second maxim is vident, as Descartes writes
in the Discourse: "Since in everyday life we must often act without delay, it is a most certain
truth that when it is not in our power to discern the truest opinions, we must follow the most
probable" (6: 25 ; I: 123 ). The third maxim, as I try to make plain in Chapter 3, can be seen to
follow from the second by clear and evident steps, so that it too can bear the stamp of
approval of the rule of evidence.
-33CHAPTER TWO
Descartes's Second Maxim
I. "MY second maxim," Descartes writes, "was to be as firm and decisive in my actions as I
could, and to follow even the most doubtful opinions, once I had adopted them, with no less
constancy than if they had been quite certain." It might appear that Descartes is simply
reporting, but he goes on to justify his resolution. "It is a most certain truth that when it is not
in our power to discern the truest opinions, we must follow the most probable" (6: 25 ; I: 123 ).
This most certain truth is a second-order practical principle; we apply it only if, pressed for
time and inadequately informed as we may be, we have already adopted a considered firstorder opinion about what is best to do. Many commentators do not read the maxim in this way,
however. tienne Gilson, for example, says that this maxim is "nothing more than a practical
rule ..., only an empirical rule of thumb required for use in everyday life to which we resign
ourselves in the absence of theoretical certainty" ( Gilson 1947, 243). And Jon Elster ( 1984,
54 - 65 ) takes the maxim to be a rule of thumb, although for reasons different from those of
Gilson. In taking the maxim to be a rule of thumb Gilson and Elster make two mistakes. First,
a rule of thumb is a first-order principle, one designed typically to save time, reduce the effect
of bias, preserve valuable spontaneity, and, above all, minimize the cumulative bad effects of
our limited knowledge. Descartes's second maxim, on the contrary, is a second-order rule, and
its application presupposes that we have already reached a decision about what to do, whether
by some rule par provision or in some other way. I will examine Gilson's and Elster's views
below.

2. Second, and contrary to a common reading, this maxim is not restricted in its application,
as a rule of thumb would be, to conditions in
-34which we lack time, relevant information, or, in general, certitude about what is objectively
the best thing to do. i. Descartes is explicit on this point. What he says is that we are to be firm
and decisive even in the case of such opinions. The opinions he has in mind in the Discourse,
of course, are admittedly doubtful first-order opinions about the truth of certain moral
precepts of the first maxim and about how these apply to particular cases. But at the limit, a
first-order opinion might be quite certain. The second maxim applies in either case. Typically,
our first-order opinions are doubtful. And in the context of the Discourse it is only to doubtful
opinions that the maxim is being applied, since, ex hypothesi, doubtful opinions are the only
opinions we have; for the opinions directly in question include our first-order moral beliefs
about good and bad, right and wrongbeliefs at the center of attention in the first maxim.
What the second maxim enjoins is that we be firm, decisive, and constant, even when it is on
these beliefs, in conjunction with other probable beliefs, that we rely. But it also enjoins us to
be decisive, firm, and constant even when the beliefs we rely on are certain. Whether we are
ever in such a situation, of course, is another matter.
I shall return to consider why this maxim is commonly thought to have a restricted
application. But first we need to get clear what it means to be firm and decisive in our actions.
3. Let us begin with an ideal case, one in which we know for certain (a) what is the best thing
to do in our present circumstances and (b) that the time for action is right now. Examples that
would come close to this ideal, assuming appropriate background knowledge, would be my
taking an aspirin to alleviate an excruciating headache or my not stepping on another's gouty
toes. In such cases, on Descartes's view, we will freely do what we judge is best forthwith,
without hesitation or second thoughts. 2. Where my
____________________
2.
At the conclusion of his presentation of the morale par provision, Descartes summarizes:
"For since our will tends to pursue or avoid only what our intellect represents as good or
bad, we need only to judge well in order to act well, and to judge as well as we can in order
to do our best" (6: 27 ; I: 125 ). Commenting on this passage in a letter to Mersenne written
in late May 1637, he writes: "You reject my statement that in order to do well it is
sufficient to judge well; yet it seems to me that the common scholastic doctrine is that 'the
will does not tend towards evil except in so far as it is presented to it by the intellect under
some aspect of goodness'so that if the intellect never represented anything to the will as
good without its actually being so, the will could never go wrong in its choice. But the
intellect often represents different things to the will at the same time; and that is why they
say 'I see and praise the better, but I follow the worse,' which applies only to weak minds"
(I:366; 3:56). He expresses the same position later in his replies to the second set of
objections to his Meditations: "The will of a thinking thing is drawn voluntarily and freely
(for this is the essence of will), but nevertheless inevitably, towards a clearly known good"
(7: 166 ; 2: 117 ).
i.
In interpreting this maxim, we may ascribe to Descartes the view that there are objectively
correct answers to such questions, whether or not these answers fall within our cognitive
competence.

-35best judgment amounts to certain knowledge, my will effortlessly follows my judgment. I

judge that my taking the aspirin is the very best thing to do in the present circumstances and,
firmly and decisively, I take the aspirin.
In the ideal case there is no slippage between the intellect and the will; the will executes what
the intellect illuminates to be best. The second maxim asks that we approximate this ideal
even in the nonideal case in which full intellectual illumination is absent. We are to make up
in volitional effort what is lacking in intellectual certainty. To be firm and decisive in the
nonideal case is to judge as well as possible and to treat the resultant opinion as if it were
clear and certain. In correspondence Descartes gives the example of a man who may starve if
he does not eat but whose will is paralyzed by the worry that the only food available may be
poisoned. Suppose the food is poisoned. In this case his eating it would be quite bad. People
can dwell on possibilities of this kind and become anxious to the point of not taking decisive
action. One who adheres to the second maxim, however, is guided by his estimate of what is
probably the best thing to do eat or starve. Once he judges that the food is very probably
safe, he acts decisively, just as if he knew this for certain. What his judgment may lack in
certainty his will makes up for in resolve and constancy.
4. Being firm, decisive, and constant, however, does not rule out changing our minds. We are
to be firm, decisive, and constant in following an opinion, once we have judged as well as we
can, given our limited information and the constraints of time; but this does not imply that we
must be firm and constant in this very judgment, which, in the light of new information, might
then be seen as mistaken. Descartes makes this point in a clarifying letter for Alphonse Pollot.
If I had said without qualification that one should hold to opinions that one has once decided
to follow, even though they are doubtful, I should indeed have been no less to blame than if I
had said that one should be opinionated and stubborn; because holding to an opinion is the
same as being persistent in a judgement that one had made. But I said something quite
different. I said that one must be decisive in one's actions even when one was undecided in
one's judgements, and that one should follow the most doubtful opinions with no less
constancy than if they were quite certain. By this I meant that once one has settled on
opinions which one judges doubtfulthat is, once one has decided that there are no others
that one judges better or more certainone should act on them with no less constancy than if
one knew that they were the best, which indeed they are when so considered. There is no
danger that this constancy in action will lead us further into error or vice, since there can be
error only in the intellect which, I am supposing, remains free
-36throughout and regards what is doubtful as doubtful. Moreover, I apply this rule 3. mainly to
actions in life which admit of no delay, and I use it only provisionally [par provision],
intending to change my opinions as soon as I can find better, and to lose no opportunity of
looking for them. (2: 34 - 35 ; 3: 97 )
Indeed, in taking up inquiry he is taking advantage of the opportunity to discover a more
perfect morality than that to which he is presently, if sincerely, committed.

5. Although the second maxim applies whether or not our practical judgment is certain, as I
have remarked above, commentators often take it to apply only when our judgment is not
certain. That only doubtful moral opinions are in play in the morale par provision lends itself
to this interpretation. So too does Descartes's illustration. In adhering to the second maxim,
Descartes says, "I would be imitating a traveller who, upon finding himself lost in a forest,
should not wander about turning this way and that, and still less stay in one place, but should
keep walking as straight as he can in one direction, never changing it for slight reasons even if
mere chance made him choose it in the first place; for in this way, even if he does not go
exactly where he wishes, he will at least end up in a place where he is likely to be better off
than in the middle of a forest" (6: 24 ; I: 123 ).
The traveler chooses in a condition of ignorance, ignorance specifically about the quickest
safe way out of the forest. He is, however, intelligent and only comparatively ignorant. In
Descartes's example, on its most natural construction, he wants to survive; he believes there is
a very low probability he will survive if he remains where he is, because, among other things,
there is little chance of being rescued; he believes there is a slightly greater probability that he
will survive if he wanders about, for in this way he might by chance find himself on the edge
of the forest; he believes there is a high probability he will survive if he walks in a straight
line. If survival has a utility of I and he can reasonably assign probabilities of, say, .I to
surviving by staying put, .2 to surviving by wandering about, and .7 to surviving by walking
in a straight line, then the rational choice is clearly the last, which has an expected utility of .7,
as compared to .I or .2. The maxim does not imply that this decision is irrevocable; it holds
only that it would be irrational for the traveler to deviate from his chosen course for slight
reasons; it does not say that he should not deviate for clear and evident reasons, such as a
well-marked trail or signpost might provide. It requires solely
____________________
3.
"This rule" must be understood to refer to the specific practical rule to be followed, not to
the second maxim itself. For example, in the case of a man lost in the forest, the rule would
be to walk in a straight line.
-37and simply that he ought to continue walking in a straight line once he has deliberated and
decided that, relative to the information he has available, that is the best thing to do, the action
with the highest probability of a successful outcome. But of course he ought to be firm and
constant and to continue walking in a straight line, as if he knew for certain that the
probability of success was I. Certainty at the level of first-order opinion is just a special case.
A further reason why the second maxim might be thought to be restricted to choice not
grounded in certainty is that only when we lack certainty is it possible that we may not act as
we judge best. But at the time of the Discourse Descartes's view is represented by the
following letter to Denis Mesland of May 2, 1644, setting out the view expressed in the
Meditations and in an earlier letter to Marin Mersenne in 1637: "For it seems to me certain
that a great light in the intellect is followed by a great inclination in the will; so that if we see
very clearly that a thing is good for us, it is very difficultand, on my view, impossible, as
long as one continues in the same thoughtto stop the course of our desire." Yet the letter
continues: "But the nature of the soul is such that it hardly attends for more than a moment to
a single thing; hence, as soon as our attention turns from the reasons which show us that the
thing is good for us, and we merely keep in our memory the thought that it appeared desirable

to us, we can call up before our mind some other reason to make us doubt it, and so suspend
our judgement, and perhaps even form a contrary judgement" (4: 116 ; 3:233-34).
Descartes's view, then, is that a clear and distinct perception that some plan of action is best
does not alone assure firmness and constancy of the will beyond the moment of this
perception. 4. Even if we do not have the liberty of perversity, we still need the second maxim.
It would read as follows: Once we have deliberated and decided on the best plan of action, we
must
____________________
4.
If, however, one ascribes to Descartes a conception of free will according to which we
have the power not to do what we clearly and distinctly judge to be best, then there is even
less reason to object to the universal applicability of the second maxim. But the reason to
ascribe this conception to Descartes is found only in a single letter. In it Descartes writes:
"For it is always open to us to hold back from pursuing a clearly known good, or from
admitting a clearly perceived truth, provided we consider it a good thing to demonstrate the
freedom of our will by so doing" (4: 173 ; 3:245). Anthony Kenny argues that the precise
date and intended recipient are not known and that this fact alone is one reason for us not
to place great weight on this passage. Kenny goes on to argue that Descartes does not
strictly allow what he calls the liberty of perversity. Alqui, on the other hand, thinks that
this letter clearly indicates that Descartes changed his view about free will. My claim that
the second maxim applies universally, however, does not depend on which of these
accounts of free will we ascribe to Descartes. Both writers agree, of course, about what
conception of free will Descartes endorsed at the time of the Discourse (see Kenny 1972;
Alqui1987, chap. 14). Relevant to this topic also is a letter to Elizabeth, dated September
15, 1645, on the need for good habits.
-38adhere to this decision throughout the period of its execution, just as if at every moment we
clearly and distinctly apprehended that it was the best.
6. There is no reason to deny, then, that the maxim applies regardless of the confidence we
have in our opinion. It is, however, restricted in a different way. It applies only on the
condition that our opinion is as well-considered as time and accessible information permit.
Then and only then does it enjoin that we stick to a course of action. "Since in everyday life
we must often act without delay, it is a most certain truth that when it is not in our power to
discern the truest opinions, we must follow the most probable" (6: 25 ; I: 123 ).
A detail should be noted. As stated, the maxim requires that when necessary we act without
delay, but, presumably, only when further delay would be counterproductive. After all, many
cases admit some delay. And although assessing whether a delay is reasonable is no doubt a
delicate matter, one that requires that we be good judges of how much time we should spend
deliberating or trying to extend our informational base, we need to interpret the maxim to
allow for this complexity. Reasonable first-order judgments, then, are reasonable along two
dimensions: they combine a reasonable judgment about how much time to give to deliberation
and a reasonable judgment about what is best to do, given the information accessible within
that interval.

Now, the maxim allows reasonable decisions to be in some respects arbitrary. "Even when no
opinions appear more probable than any others, we must still adopt some; and having done so
we must then regard them not as doubtful, from a practical point of view, but as most true and
certain, on the grounds that the reason which made us adopt them is itself true and certain" (6:
25 ; I: 123 ). Although the traveler's choice of direction may be arbitrary, he is then to act just
as if this choice had been based on good evidence. The maxim itself, to be sure, gives no
guidance about how initially to decide or how long to deliberate. It says only that we must be
firm and constant once we have deliberated well and decided.
7. Descartes justifies his adopting the second maxim by an appeal to the most certain truth
"that when it is not in our power to discern the truest opinions, 5. we must follow the most
probable." Descartes's view is that our own reason demands that when we cannot know for
certain what is the best thing to do, "we must follow the most probable." The view might be
set out this way, in terms of first-order and second-order evaluation. A first order question
might be: Should I eat this food? With respect to this question, I am unable entirely to
eliminate the possibility that this food is poi____________________
5.
"That is to say, those that would be absolutely speaking the best" ( Gilson 1947, 244).
-39soned. If I could have certain knowledge, then I would freely choose to eat, my will being in
this case determined to follow my clear and evident perception of the good. As it happens, I
cannot entirely eliminate this possibility. If I cannot attain certainty at this level, I can attain
certainty at a higher level. If I cannot know for certain that x is the best thing to do here and
now, yet I reasonably believe that x is probably the best thing to do, then I can know for
certain that I ought to do x. We move from first-order uncertainty to second-order certainty.
What we may know for certain is that the best thing to do in conditions of uncertainty is to do
that which we reasonably believe is most probably the best.
Descartes's thought is that if we always knew for certain what is best in every circumstance of
life, we would unfailingly do that thing. As rational agents, we would then all be firm and
decisive in our actions. But we are not like this. Nor is Descartes, the pure inquirer, like this,
for he is a person with a life to lead while he meditates. What is a pure inquirer to do? What
are we to do? Dither, wallow in uncertainty, starve? Self-evidently not. While we may not be
able to rid ourselves of doubts at the level of our first order judgments, we can rationally
resolve the practical question, What should we do? at the second level. If I can doubt at the
first level whether x is the best thing to do, I can resolve this doubt as a practical matter at the
second level; at this level, if I reasonably judge that x is most probably the best thing to do,
then I know that x is, from a practical point of view and from the point of view of reason, the
best I can possibly do in the circumstances. Even if I should,per accidens, do what is in fact
the best thing to do while violating the second maxim, I do not escape remorse and regret, nor
do I act well.
What could stand in the way of our doing what we judge to be most probably the best thing to
do? We might dwell on the possibility that we may be making a first-order mistake, that doing
x may in fact be bad; or we might dwell on the desirable aspects of the alternative after we
have rejected it. Such second thoughts are the cause of irresolution, weakness, indecisiveness.
They are contrary to reason, since they are the dispositions of someone who does not take to

heart the truth on which Descartes's maxim is based, namely, that when certainty is not in the
offing, we must make do with probability, even in the case where the probability is quite low.
And truth is seldom if ever in the offing. First, we often lack sufficient information to
determine what is in fact the best thing to do. Second, even when sufficient information is
accessible, we often do not have the cognitive capacity to make the best use of it. Third, we
typically lack sufficient time to make the best use of what information we have ready to hand.
Finally, time itself is of the essence; the ends we seek are indexed to time and the nec
-40essary means to these ends indexed to an earlier time. So to the question, What is the best
thing to do? is added the further question, What is the optimal duration for deliberation? In
the theoretical domain of pure inquiry, we have the luxury of delaying as long as it takes to
arrive at certain knowledge. In everyday life, by contrast, we have no such luxury. And here,
Descartes argues, our own reason requires that we own up to the fact of these epistemic,
informational, and temporal limits on our practical reasoning.
8. So far I have considered the second maxim mainly as it applies to action in the immediate
present. In this case, it gives the rational response to the common human fault of irresolution
in the face of first-order uncertainty concerning what should be done here and now. But this
maxim has a more general application. It also applies when we deliberate about what to do at
some future time, or about a plan of action that will take time to carry out, or about a general
rule of action to be followed in certain types of circumstance. Significantly, Descartes's own
choice in the first maxim, to follow local custom and example, involves a precommitment to
such general rules.
Descartes's traveler lost in the forest cannot determine for certain which is the most
expeditious route to his intended destination; indeed, he cannot say of any particular route that
it is probably the most expeditious. On the other hand, among the alternative plans of staying
put, wandering around this way and that, or walking in a straight line, the last is most likely to
improve his position and move him toward his destination. Even if the direction he takes is
not selected as probably the most expeditious indeed, even if it is selected arbitrarily or
only for the weakest reasons continuing to walk in this direction is most probably the best
thing to do in the circumstances.
How, precisely, are we to understand this illustration? The correct answer, I believe, is
straightforwardly. Descartes takes himself to be such a traveler; while engaged in pure
enquiry, he is a traveler in everyday life. The general rule for those who travel in terrain
where the best path has no certain marks is to choose the most probably correct path and to
stay on it. Staying on the path is precisely what following reason in conditions of uncertainty
amounts to. Indeed, staying on a path of which the traveler knows for certain that it is best is
what following reason in conditions of certainty amounts to. But following reason in
conditions of uncertainty is rather more difficult and requires a special act of will to control
the sort of irresolution and vacillation that attends the condition of uncertainty. We are to
make up with strength of will what we lack in cognitive competence.
9. The example can, however, lead one to a different interpretation of the second maxim. At
one point in his Commentary, as I remarked above,
-41-

Gilson writes that "considered from the perspective proper to the Discourse," the second
maxim is "nothing more than a practical rule ..., only an empirical rule of thumb [procd]
required for use in everyday life to which we resign ourselves in the absence of theoretical
certainty" ( Gilson 1947, 243). While we must, if we are lost in a forest, resign ourselves to a
rule of thumb as recommended in the example, the general maxim under which we subsume
this rule is not itself a rule of thumb, nor is it based on empirical considerations. It would be a
mistake, therefore, to see a direct analogy between the traveler's rule of thumb and the second
maxim itself, both being provisional in the same sense. It is not as if Descartes's decision,
expressed in the second maxim, relies on the probabilistic judgment that resolution is the best
general policy. It is true, of course, that the rule of thumb for travelers lost in a forest could be
abandoned under conditions of certainty. If the traveler comes on a clearly marked trail out of
the forest, he may alter his present course and follow a new one. And it is also true that if
Descartes were to achieve certainty in the science of ethics, he could abandon his earlier
decision to follow custom and good example and instead follow the clearly lit path of his
reason. 6. Nonetheless, the second maxim is itself not based on empirical knowledge of
probabilities. Nor is resolution recommended strictly on the basis of its being the most
probably effective means to an agent's intended goals. It is not resolution itself, after all, that
is maximally useful to the agent in conditions of uncertainty. Resolution without good
judgment is not useful. To be sure, good judgment without the strength of will to follow it is
not maximally useful either. Both are needed. But the rule of reason, namely, to follow our
best judgment, is not itselfempirically based or justified in terms of its being the most
probably effective practical rule. It is, as Descartes says, self-evident.
10. Jon Elster, guided, it seems, by the traveler example, reads Descartes's second maxim as
an empirically justified procedure for achieving rationality by indirect means. Elster sees
Descartes as a critic of so-called "instant rationality." The question posed is this: Should we
try to decide
____________________
6.
I do not mean to suggest that even at the time of the Discourse Descartes believed such a
science of first-order ethics was possible. He certainly did not think that in matters of
ordinary prudence, where the crucial governing beliefs are beliefs about future
consequences of action, we could have clear and distinct knowledge of what is best to do.
And for the same reasons we must lack clear and distinct knowledge in the application of
some moral rules. In his letter to Hyperaspistes, dated August 1641, he expresses a view
that it is only reasonable to suppose he always held. "It would indeed be desirable to have
as much certainty for the conduct of our lives as is needed for the acquisition of
knowledge; but it is easily shown that in such matters so much is not to be sought or hoped
for. This can be shown a priori from the fact that a human, being a composite entity, is
naturally corruptible, while the mind is incorruptible and immortal. It can be shown even
more easily a posteriori from the consequences that would follow" (3:422; 3:189). Here he
gives the example, cited above, of the man who would starve if he refused to act until he
had certain knowledge that his food was not poisoned.
-42on every occasion of choice what is the best thing to dothe thesis of instant rationalityor
should we, rather, commit ourselves to follow certain rules for certain circumstances? He
interprets Descartes as advocating rules of thumb by arguing that the policy of instant
rationality is not, all things considered, cost effective. 7. There is little in the Discourse itself,

however, to suggest that Descartes's aims to reject instant rationality. 8. In his example of the
traveler in the forest Descartes does say that the traveler would be better off walking than
pondering. More generally, he says that "we must often act without delay." Now, if instant
rationality were the policy of not acting until we achieved certainty, then it would be, by
Descartes's lights, about the worst policy we could have. But if the policy of instant rationality
is that of trying to decide each particular case as it arises rather than relying on standing rules
that are insensitive to the particulars of every case, then the second maxim itself does not
favor the one policy over the other. Perhaps it is better for travelers to adopt a standard policy
for occasions when they are lost in a forest, but there are other contexts of action where we
may have no standing policies or where having standing policies would not serve us better
than trying to evaluate the situation. Standing policies no doubt save time, but it is an open
question, so far as the second maxim itself is concerned, whether the time saved might not
have been better spent evaluating the case at hand before acting.
Descartes's other principal examples, in connection with the second maxim, do not give
secondary rules a prominent role. In one case, a man has before him food of which he has not
the slightest reason to be suspicious, and the question is whether he should satisfy his hunger
by eating it. Descartes says that he should do so without vacillation, even though he believes
it is possible that the food is poisoned. In another case, a man should travel on the road that is
least frequented by robbers and do so without vacillation once he has determined that it is
probably the safest route. In both casesas in example of the traveler lost in the forestthe
point is the same: we are to do with dispatch, with firmness and resolution, what we judge to
be the best thing to do, even though we acknowledge that our judgment about what is best
may be in error. Whether we should adopt
____________________
7.
This discussion is found in Elster 1984. Elster offers nvo "Cartesian" arguments in favor of
rules of thumb. The first is that "a continuous evaluation and reevaluation requires so much
time that it can be expected to more than outweigh the time gained by the improved
direction that issues from the evaluation," and the second is that "not only [is] the time
spent getting out of the forest ... increased if the traveller constantly stops to reevaluate the
situation, but... instant rationality actually makes the path itself longer than it would have
been along an arbitrarily (or at least along an optimally) chosen straight line" ( Elster 1984,
57, 60 ).
8.
"Cartesian" arguments of these kinds can be found, however, in the Passions of the Soul.
There he argues, for example, that our having standard policies for specific sorts of
circumstances is optimal, because it saves us from lost opportunity costs.
-43standing policies for certain kinds of circumstances is a matter to be decided in the same
general way that a traveler decides to follow a straight course through the forest.
Sometimes, to be sure, we should firmly and resolutely adhere to a rule, but when this is so, it
is because we had judged earlier that, all things considered, the best thing to do if we
anticipate finding ourselves in certain situations is to adopt a general rule of conduct. In other
cases, it may be that no rule to which we are in this way precommitted applies. Then we must
try to determine, so far as time permits, what is most probably the best thing to do and then do
that thing, with the resolution of one who had certain knowledge of it. In either case, the

emphasis falls on what the maxim presupposes, namely, that the initial decision was best in
the circumstances and that these circumstances have not changed in any relevant respect.
II. In concluding his brief commentary on this maxim Descartes remarks: "By following this
maxim I could free myself from all the regrets [repentirs] and remorse [remords] which
usually trouble the consciences of those weak and faltering spirits who allow themselves to
set out on some supposedly good course of action which later, in their inconstancy, they judge
to be bad" (6: 25 ; I: 123 ). Self-evidently, Descartes says, 9. reason demands (a) that we judge
as well as we can and (b) that we act in accordance with our very best judgment (while
recognizing that this may be objectively in error). In the domain of practice, once we have
met these two conditions, we have done as well as we possibly can. Having this demand
embedded in our conative and affective dispositions is what being firm and resolute amounts
to. And this, in turn, frees us, not from sadness in general, but from the burden of remorse and
repentance. This is so because remorse and repentance, on Descartes's view, are the forms that
irresoluteness takes retrospectively; they are the passions that attend second-thoughts. If we
are firm, decisive and resolute, we do not have such second thoughts. This is not to say that in
deciding to take the road that is probably the safest we forthwith become absolutely certain
there we will meet no robbers. Not having second thoughts about how we should act does not
involve any kind of self-deception about our liability to error. Nonetheless, being firm,
decisive, and resolute, in the sense Descartes intends, is to have the same
____________________
9.
In the Discourse he says that this is a most certain truth; in the letter of 1641 to
Hyperaspistes, discussing the man who hesitates to eat on the off chance that his food is
poisoned, Descartes writes: "Suppose further that he could not obtain any food that was not
poisoned, and that his nature was such that fasting was beneficial to him; none the less, if
the food had appeared harmless and healthy, and fasting appeared likely to have its usual
harmful effects, he would be bound to eat the food and thus follow the apparently
beneficial course of action rather than the actually beneficial one. This is so self-evident to
all that I am surprised that anyone could think otherwise" (3:423; 3:189; my italics).
-44affective and conative dispositions connected to one's judgment of what is probably the best
as one would have if all one's first-order evaluations of action were les plus vraies. 10.
Contrary to some commentators, I believe we should take Descartes's remark not as
expressing the point or rationale of the second maxim but as pointing out the effect of our
taking the maxim to heart. For when we have met its demandswhen we have made firstorder decisions with care and without undue delay and then stuck to them as if we were
absolutely certain of themwe have acted as well as we possibly could; as rational agents we
cannot ask more of ourselves. And since we cannot reasonably ask more, we should be free of
regret and remorse, even if we were to discover retrospectively that our first-order judgment
was mistaken.
In the ideal case, of course, we have certain knowledge that x is the best thing to do and we
chose to do it. Here we have both perfect judgment and perfect willing, and there can be no
occasion for remorse or repentance. ii. Owing to the imperfection of our cognitive powers
(itself a reflection of our composite nature), we often cannot determine with certainty what is
objectively the best thing to do; we do not always have perfect judg-

____________________
10.
Descartes returns to examine the passion of irresolution in the Passions of the Soul: "an
excess of irresolution results from too great a desire to do well and from a weakness of the
intellect, which contains only a lot of confused notions, and none that are clear and distinct.
That is why the remedy against such excess is to become accustomed to form certain and
determinate judgements regarding everything that comes before us, and to believe that we
always do our duty when we do what we judge to be best, even though our judgement may
perhaps be a very bad one" (11:460; I:390-91).
ii.
Of course, there can be plenty of occasion for regret or sorrow in the sense that the world
may be filled with unavoidable evils. Even if I could be absolutely certain that I ought to
do x, it may still occur that x turns out to be a contributing cause of some outcome, y,
which is evil. I may judge, indeed, judge rightly, that such evil was in the circumstances
absolutely unavoidable, but it is nonetheless evil. If, in order to save my fellow spelunkers
from drowming in the rising water, I will have to blast out a member of the party who, in
no danger of drowning himself, is stuck in the only available exit from the cave, I may
rightly judge that blasting is the best thing to do in the circumstances. The death of my
innocent friend is an evil to which I cannot be indifferent; such evil is the stuff of tragedy.
But this evil of which I am the instrument, although the occasion for sorrow or sadness of
some kind, is not the proper occasion of remords and repentirs in the Cartesian sense.
"Remorse of conscience,"Descartes writes in the Passions of the Soul,"is a kind of sadness
which results from our doubting that something we are doing, or have done, is good. It
necessarily presupposes doubt" (II:464; I:392). Repentance results from our believing that
we have done some evil deed (II:472; I:396). Descartes argues that all these passions
remorse, repentance, and irresolutionhave their proper uses, but they may also be
improper or excessive. The remedy against them is the same, "to accustom ourselves to
form firm and determinate judgments ... and to believe that we always do our duty when we
do what we judge to be best, even though our judgment may perhaps be a very bad one"
(II:460; I:391). We feel remorse when we doubt that what we do is good, and we feel
repentance when we believe what we did was evil. We may feel both sentiments in
connection with the very same action, on Descartes's analysis, for so long as we are in
doubt, we see our action as at once good in one respect and evil in another and dwell in an
anxiety of irresolution. These passions are forms of anxiety that derive pointlessly from
-45ment. On the other hand we do have the power to will perfectly, as in the ideal case. So even
if we cannot overcome our finitude and achieve perfection in our judgment, we can achieve
perfection in our willing. The point of the second maxim is to set this standard; it is the
standard for both the ideal of perfect judgment and the nonideal case.
It must be understood, however, that ifin the interval between our decision to act and our
actionwe have good reason to reconsider and possibly to revise our judgment, then, time
permitting, we should do so. The rule applies only to the cases in which there is no good
reason to revise our judgment. It holds that if there is no rational basis for doing so, then there
is no rational basis (in our judgment) for revising our will or intention. In the forest example, I
am acting irrationally if I change my direction without express reason for changing my prior
judgment. From a practical point of view, I am to act as if my prior judgment were certain.
That there is a lapse of time between the time of acting and time of the relevant judgment is
not itself a reason for reconsidering or revising the judgment. What such a lapse may occasion,
however, is my inability to recall with sufficient vivacity the reasons that originally led me to

judge as I did, so that I may be tempted on the occasion of action by "some slight reason" or
impulse to think my prior judgment was flawed, which may give rise in turn to the illusion
that I should reconsider and possibly revise my intention. The rule holds simply that I am
bound by those of my past rational intentions that apply to actions in the present. 12.
We are, then, to deliberate as well as we can within the time allowed and henceforth honor
our rational prior intention. Descartes goes on to say that in acting in this way we shall avoid
remorse (remords) and regret (repentir). The terms remords and repentirs are understood by
Descartes in a special way, which the natural translations "remorse" and "regret" may not
capture. Both derive from irresolution in judgment, where the tantalizing but vague thought
that I might hit upon a better solution prevents me from acting decisively. In advance of
action, it is my fear of mistake or error that prevents my definitely resolving the matter as well
as I can, where "mistake"
____________________
our refusal to accept our limitations as human agents, particularly limitations on our
knowledge and our time, and our consequent failure to make determinate judgments in the
absence of full information. In his discussion of Descartes's second maxim Elster writes:
"It is hard to read the correspondence of Descartes without being struck by this 'grand
seigneur' aspect of his character; never explain, never apologize" ( Elster 1984, 60 ). There
is some truth in this remark, but the comment as it stands does not take into proper account
Descartes's insistence that we must also judge as well as we can.
12.
Descartes does not address the case in which my prior judgment was not my best judgment
and my prior intention not, in the defined sense, rational. If without reason I reconsider and
then revise my judgment, I may in fact judge better and, correspondingly, improve the
prospects of my course of conduct. Do I act rationally or not in such a case?
-46and "error" are given a comparatively objective or even absolute sense. Remorse and regret
refer to painful reflections arising after having made some mistake or error. Suppose I
retrospectively judge that I made an error in judgment, in consequence of which I have failed
to attain a good that (I erroneously believe) was within my grasp. For "weak and faltering
spirits" this is the occasion for regret and remorse. 13. Fear of error or mistake, prospectively,
and remorse and regret, retrospectively, derive equally from an unrealistic and effective
refusal to accept our finitude and the imperfection of our power of judgment. In sum, we act
irrationally when our intention does not match our best judgment, and our regrets are
irrational so long as we have not acted irrationally.
12. Looking back at this maxim from the viewpoint of Descartes's later work, we see it as the
nascent form of the fundamental principle of virtue. Descartes says as much in his important
letter to Elizabeth, where, expressly referring back to the morale par provision of the
Discourse, he tells her that the second rule of living the best life possible is "to have a firm
and constant resolution to carry out whatever reason recommends without being diverted by
passion or appetite. Virtue, I believe, consists precisely in firmness in this resolution; though I
do not know that anyone has ever so described it" (4:265; 3:257-58). By this time Descartes
had also argued that being virtuous and living well are the same. The point to make here,
however, is that the second maxim of the morale par provision not only survives further
critical reflection, it emerges as a central part of the principle of virtue in Descartes's final
morality. 14.

____________________
13.
As Descartes writes in the Passions of the Soul,"The weakest souls of all are those whose
will is not determined in this way to follow such judgement, but constantly allows itself to
be carried away by present passions. The latter, being often opposed to one another, pull
the will first to one side and then to the others, thus making it battle against itself and so
putting the soul in the most deplorable state possible. Thus, when fear represents death as
an extreme evil which can be avoided only by flight, while ambition on the other hand
depicts the dishonour of flight as an evil worse than death, these two passions jostle the
will in opposite ways; and since the will obeys first the one and then the other, it is
continually opposed to itself, and so it renders the soul enslaved and miserable" (II:367;
I:347).
14.
In the final statement of this morale Descartes writes in the Passions of the Soul: "Thus I
believe that true generosity, which causes a person's self-esteem to be as great as it may
legitimately be, has only two components. The first consists in his knowing that nothing
truly belongs to him but his freedom to dispose his volitions, and that he ought to be
praised or blamed for no other reason than his using this freedom well or badly. The
second consists in his feeling within himself a firm and constant resolution to use it well
that is, never to lack the will to undertake and carry out whatever he judges to be best. To
do that is to pursue virtue in a perfect manner" (II:445-46; I:384). The first part is
anticipated in the third maxim of the provisional morality, as we shall see; the second part
is a development of the second maxim. With respect to the second and third maxims, then,
there is no radical difference between Descartes's final or mature ethics and his early
ethics.
-47CHAPTER THREE
The Third Maxim
I. IT may appear from the third maxim of his morale par provision that Descartes sought to
align himself with the moral doctrines of Justus Lipsius and Guillaume Du Vair, neo-Stoics of
the period, who were developing a line of moral thought succinctly formulated by Epictetus:
"Do not seek to have events happen as you want them to, but instead want them to happen as
they do happen, and your life will go well" ( Epictetus 1983, 13 ). "My third maxim,"
Descartes writes, "was to try to master myself rather than fortune, and change my desires
rather than the order of the world. In general I would become accustomed to believing that
nothing lies entirely within our power except our thoughts, so that after doing our best in
dealing with matters external to us, whatever we fail to achieve is absolutely impossible so far
as we are concerned" (6: 25 ; I: 123 ). i. This maxim differs noticeably from the famous
Epictetan injunction, which strikes a note of quietism or resignation. Descartes's maxim is,
however, quite close Epictetus's interpreter Du Vair, who writes in The Moral Philosophy of
the Stoics:
When we doe anything, though we doe it neuer so wisely, take all aduantages, chuse all
opportunities, vse all possible diligence: yet for all this we must know that the greatest part of
the euent is altogether ouermsastered by Fortune. Wee are lords and masters of our counsels
and determinations, but all the rest dependeth vpon other matters which are not in our power.
And therefore we can doe no more but vndertake a mat____________________

i.

Descartes gives a nice commentary on this maxim in his letter to Reneri for Pollot, April or
May 1638. See also Gilson 1947, 247-48.
-48-

ter with wisedome, pursue it with hope, and be readie to suffer whateuer shall happen with
patience. If good enterprises haue bad successe, the answere which the noble man of Persia
made, may serue for an excuse to all them which are wise, but vnfortunat. One was desirous
to know of him wherefore (seeing that he knew him to bee a very wise, & valiant gentleman)
his affayres went no better forward. Because (quoth hee) in my affayres I can but giue good
counsell, that is all I can doe, the successe belongeth altogether vnto us, that is, if wee attempt
nothing but with a good end, and follow it not but by lawfull and honest means. ( Du Vair
1951, 127 )
The striking similarity between this passage (and many like it in Lipsius and Du Vair) and
Descartes's third maxim should not, however, lead us to the view that Descartes simply
borrowed his maxim from the neo-Stoic texts with which he was familiar, texts ready to hand
and suited to the temperament of a pure inquirer. In the first place, it was very far from
Descartes's style as a thinker to be tempted to put together the congenial thoughts of others in
a syncretic pastiche. Moreover, we know from the central role that Epictetus's celebrated
distinction plays in Descartes's later writings on ethics that this maxim was not one he
adopted casually, nor, I would add, primarily as an aid to his research. 2. Yet if we are not to
see this maxim as simply taken over from, say, Du Vair, do we not need some more internal
or systematic explanation for its inclusion? One might reply that this is too much to ask. As
readers of an autobiography described at one point as a fable, we should not expect the
narrative to mirror a rigorous sequence of thought. To be sure, the tone of this maxim is more
personal than that of the second maxim. Since the maxim seems more personal, it may also
seem more optional. After all, it was not in fact necessary for Descartes to become a Stoic
sage, or even some reasonable approximation of one, in order for him to get on with his
project. And while we may grant that the Stoic ideal appealed to him, this would not account
for any systematic connection between this maxim and the two preceding. Is it, then, required
to complete Descartes's morale par provision?
____________________
2.
As to the much debated question of Descartes's intellectual indebtedness to Stoic and neoStoic writers or of the degree to which he was a Stoic, it would be premature to consider it
here. For a good discussion of the question, see D'Angers 1976. It will not be my aim to
consider this question directly or as a special topic; I treat it indirectly and in the context of
other issues throughout the remainder of the book. What we shall find, in sum, is that
Descartes gives a definite rationale for Epictetus's distinction within the framework of his
metaphysics, defines virtue in terms of what is entirely in our power, and makes virtue
sufficient for happiness. On the non-Stoic side of the ledger, he rejects the traditional Stoic
theory of the emotions, and he identifies happiness with contentment, not with virtue itself.
-492. The second maxim, we have seen, is a second-order principle; it is also presented as selfevident. The third maxim is also a second-order principle; and if it is not quite self-evident, it
is systematically connected with the second maxim in such a way as to justify it and its
inclusion in the morale par provision. I say that these are second-order principles since they

bind whatever might be the content of the precepts and values of our first-order morality. The
third maxim would hold that, whatever our specific ends, we need to acknowledge that they
are never entirely in our power; if we do not realize them despite out best efforts, we are not
responsible. In this way, we might see the third maxim as a variant of the principle that ought
implies can, which principle is neutral among competing accounts of what we ought to do or
try to achieve. And being neutral, this maxim could be asserted with confidence by someone
who was, comparatively speaking, in doubt about his own first-order moral convictions.
Indeed, following this line of thought, one might be led to see the third maxim as self-evident
or at least as a principle one must accept if one accepts any morality at all.
But this line of argumenteven ifwe accept it as a plausible reconstruction of Descartes's
thinkingwould not get us all the way to the third maxim as formulated. Although it is
neutral about what specific ends are worth pursuing, the maxim nonetheless requires that we
regulate our desires in such a way that we do not, in some full-fledged or robust sense, desire
those otherwise acceptable ends we seek under the guidance of our first order morality.
3. An example may help. Suppose I become a doctor and join Les Mdecins sans Frontires
with the principal aim of bringing medical care to those in the world who are most in need.
Suppose also that I think this is worth doing because I think that human suffering is an evil
which cries out to be alleviated. These are thoughts included in my first-order morality.
Perhaps, as I go about my task, I am often aware that I risk my life by exposing myself to
disease and other dangers. Suppose I say the risk is worth it, since the end I am promoting is
more important than my own life. The focus of my attention is on the evil to be eliminated.
But whether it is eliminated is not entirely in my power. According to the third maxim, I
should discipline my desiring in such a way that my desiresor my strongest desiresfix on
what is in my power and not on the external good I am trying to achieve.
What this maxim requires is, on the face of it, paradoxical, if not self defeating. It seems to
demand that at one and the same time I desire and not desire to eliminate the evil of disease. 3.
Certainly, part of the problem
____________________
3.
The paradox is familiar from ancient Stoicism. The Stoics would classify the end, in this
case, as at once indifferent and preferred. The only true good, they would arguegood in
the sense of being in accord with nature or reasonis the manner of acting, not the state of
affairs or end sought. Critics deny that the Stoics can successfully overcome the following
diffi-50lies with the variable use of the term "desire." One line of thought is this: if I seek some end
as itself choiceworthy and justify my acting in terms of its value, I can be said to desire it.
According to another line of thought, I cannot rationally desire what I judge to be impossible.
Both are present in Descartes's commentary on the third maxim. We may follow them up to
the point where they seem to collide, starting with the second.
"Our will," Descartes writes,
naturally tends to desire only what our intellect represents to it as somehow possible; and so it
is certain that if we consider all external goods as equally beyond our power, we shall not

regret the absence of goods which seem to be our birthright when we are deprived of them
through no fault of our own, any more than we regret not possessing the kingdom of China or
of Mexico. Making a virtue of necessity, as they say, we shall not desire to be healthy when
ill or free when imprisoned, any more than we now desire to have bodies of a material as
indestructible as diamond or wings to fly like the birds. (6: 26 ; I: 124 )
In speaking specifically of the will, in this passage, Descartes limits the discussion to
outcomes we might conceive as objects of our will, that is, to abstractly possible states of the
world whose existence we could conceivably influence through carrying out some plan of
action, where the constraints on conceivability are those of our own reason and experience.
Thus, for any state of the world whose value is given, if we can trace no credible causal series
of events that would both require some voluntary actions of our own and lead to that state,
then this state is not a possible object of desire, or, more accurately, of rational desire.
Granting that it would be a good thing for me to possess China, it would not be possible, or in
any case, rational for me to desire to do so, because there is no conceivable causal series of
events that would lead me from where I am to this outcome.
To return to my example, if I am trying to alleviate the suffering of others, I certainly believe
that what I am doing increases the likelihood that
____________________
culty. If the end is not truly of positive value, then the claim that the manner of acting has
positive value is empty. In the example of the doctor, it is as if disease is not an evil, but
trying to eliminate it is nonetheless good. The objection is that this account does not make
clear sense. Looking at the matter from the point of view of the doctor, he can make sense
of his efforts only on the supposition that what he is trying to accomplish is genuinely
worthwhile; he must presuppose that a world in which he conquers disease, even to some
limited degree, is better than one in which he does not, where the difference in value is in
part simply a function of the reduced incidence of disease. If, on the other hand, he thought
the end was without any value whatever and that a world with less disease was not a better
world, he could hardly think that a world in which he sought to eliminate disease was
better than one in which he did not, the disease itself being of no value or disvalue. There
is no doubt that Descartes inherits a similar problem, which he does not successfully
resolve in the Discourse. Whether he resolves it later, using the resources of his theory of
the passions, remains to be seen.
-51those I am treating will be returned to good health and freed from pain. And my reason and
experience tell me that I may expect many of my efforts to be successful. Suppose, however,
that I come to believe in a particular case that whatever I do will have no effect on whether or
not my patient is cured or delivered from pain. Then, if my desire follows reason, I no longer
desire the good in question, while continuing to believe it is a good. It is just that under the
stipulated conditions this good can no longer be an object of my will. So, following the first
of the above lines of thought about desire, I cannot be said to desire that this patient of mine
not suffer. In this particular case, the first line of thought about desire coincides with the
second. Are there cases in which they would collide, as I have suggested?
If we follow the reasoning behind the third maxim, the answer will be that they collide only in
cases in which I am confused. Consider the case in which I believe that a patient's return to

health does depend on my continuing efforts to cure him. I acknowledge, of course, that his
health does not depend only on my own efforts; it depends in part on other factors that are
entirely beyond my control. Indeed, when I reflect on the case, I see clearly that the good in
question, if it comes about at all, will do so as a consequence of causal factors of which some
are entirely in my control and others not at all. This reflection, however, will bring the second
case into line with the first. In the first case I come to believe that whether my patient returns
to health depends entirely on factors beyond my control. In the second case I also know that
after I have done what I believe is necessary and within my power, whether health is restored
will depend on factors entirely beyond my power. Now, the good in question is a patient's
health in both cases. In the first case, we see that this outcome cannot be an object of my will
and hence cannot be an object of my desire. But the second case is essentially no different
with respect to the patient's health, because once I have done everything I can, what happens
next is not at all in my power. His health, should it be restored, is among things which happen
that are, given what I have done, entirely beyond my control. It follows that my will cannot be
directed at such objects, any more than it can be directed at health in the first case. And
according to the first line of reasoning concerning desire, it follows that I cannot desire any
objects that are not entirely in my power. I must judge things not entirely in my power as,
relative to my will, absolutely impossible and, accordingly, as goods that I cannot rationally
desire. In sum, a desire for an object not entirely in my power is an irrational desire. So far as
we are perfectly rational, therefore, we desire nothing that is not entirely in our power. And so
far as we are perfectly rational, "we shall not regret the absence of goods which seem to be
our birthright when we are deprived of them through no fault of our own."
-524. In the final analysis, things entirely in our power are restricted to our own acts of will. And
the only goods entirely in our power are good acts of will. In his commentary and defense of
the third maxim Descartes does not tell us what makes good acts of will good, although he
suggests that an act of will is good when it is willed as a necessary condition within a causal
series whose issue is something of value, say, the easing of another's pain. Moreover, he
hintsit is only a hintthat such acts of will are good not simply as means but as having
value in themselves. The hint lies in his remark that even without our achieving external
goods whose value explains our undertakings, we should find contentment simply in our
acting well; it lies also in his comment on the ideal of the Stoic sage. It takes long practice and
meditation, he says, to take the third maxim fully to heart, however compelling it is to the
intellect. In doing so
lay the secret of those philosophers who in earlier times were able to escape from the
dominion of fortune and, despite suffering and poverty, rival their gods in happiness. Through
constant reflection upon the limits prescribed for them by nature, they became perfectly
convinced that nothing was in their power but their thoughts, and this alone was sufficient to
prevent them from being attracted to other things. Their mastery over their thoughts was so
absolute that they had reason to count themselves richer, more powerful, freer and happier
than other men who, because they lack this philosophy, never achieve such mastery over all
their desires, however favoured by nature and fortune they may be. (6: 26 ; I: 124 )
The third maxim, then, is not simply borrowed from the Stoics. It is presented, rather, as the
conclusion of an argument that takes as its point of departure the notions of rational desire and
of what is impossible (relative to us). These concepts are neutral among competing first-order
moralities, and if the argument succeeds, the self-mastery it commends could be defended as a

desideratum within any morality. In this sense, this maxim could be defended as a formal
demand. And like the second maxim, I should add, it is not to be conceived as provisional in a
pejorative sense, that is, as a candidate for possible replacement with something better
something that might be discovered at the end of inquiry.
5. In addition to their being neutral among competing first-order moralities, the second and
third maxims are presented to us as demands of our own reason, commended to our bon sens
on the basis of clear conceptions and clear lines of argument. In fact, they would seem to meet
the requirements of Descartes's method, the second maxim being declared self-evident, the
third being deduced from clear and evident concepts. Beyond this, the sec
-53ond and third maxims share an important common feature; both are parasitic on the idea of a
correct first-order morality. Here they too become vulnerable to the Pyrrhonist's weapons. The
Pyrrhonist argues that although there are correct moral directives and correct answers to moral
problems, what appears correct to us is determined not by what is correct in some absolute
sense but rather by socializing forces that vary from culture to culture and that seem to swing
free from the moral facts. Whether this argument weakens the defenses of either the second or
third maxims depends on what these maxims presuppose. They do presuppose that there is a
correct first-order morality. But the Pyrrhonist argument shares this presupposition. The
question is this: Do these maxims also presuppose that we can in principle gain genuine
knowledge of the correct directives and correct answers to practical problems, at least on
some occasions? If they do, the Pyrrhonist argument is a threat even to them.
Descartes's answer to this question, I believe, would run along the following lines. First, the
Pyrrhonist argument may succeed in undermining our commonsense or naive view that our
deepest moral convictions amount to knowledge at the level of theory, but it does not or need
not succeed in undermining our confidence in them at the level of practice. Second, since we
can expect to gain epistemic access to the true moral directives at the end of methodic inquiry,
we may look forward to meeting the Pyrrhonist's argument then at the level of theory. For the
moment, however, confidence is sufficientsufficient, that is, for the second and third
maxims to be applicable. To illustrate the need for confidence, consider our doctor once more.
Even if I am enlightened and subscribe to the third maxim, so that what I will is entirely in my
own power, nonetheless what accounts for my acts of will is my conviction that another's
suffering is a real evil. If I lost all confidence in this conviction, I would have to see my own
acts of will as pointless, however much they may be entirely in my power. Moreover, if it is
true, as Descartes hints, that my own contentment follows on the heels of my doing all
entirely in my power to alleviate your suffering, this too is conditional on my conviction that
what I am doing is worthwhile, which is conditional on my confidence in my conviction that
your suffering is an evil. A point that is scarcely discernible in the Discourse but which comes
into prominence in later writings is that acting well, as I have described it, is virtue, the
supreme good in Descartes's developed theory. And he argues later that what gives the
sweetest and most durable contentment is our awareness of our own virtue. If Descartes is
right, then even amid the most trying and uncomfortable circumstances, my consciousness
will be suffused with the highest and most solid contentment that a human being can feel. Still,
the focus of my attention is on the evil to be eliminated, not on my
-54-

own virtue or on my own contentment. But it is crucial both to my contentment and to my

conception of my supreme good that the evil I seek to eliminate is a genuine evil. If truly it is
not a real evil, because in truth there is no real distinction between good and evil, then I am
deluded, and my virtue is compromisedone might be inclined to say it is bankrupt. If I am
right, on the other hand, and there is a real distinction between good and evil, then my virtue
is transparently genuine. It is important for the theory of virtue itself that this evil be real, for
it is a mark of my virtue indeed, it is constitutive of my virtuenot only that I stick to my
task but that I judge the end to be worth promoting. The point, I concede, is delicate, since we
need not insist on moral infallibility in this area in order to ground a theory of virtue. But we
do need a theory that shows that, in favorable conditions of judgment, when I take something
to be a real good or a real evil, I am at least following the scent of true good and evil.
How can I be confident that I am following the right scent? In the end, Descartes cannot keep
the levels of theory and of practice apart. The confidence that is requisite for practice looks
forward to the real possibility of our coming to know that the moral appearances accurately
mirror moral reality. At least, this is the position of the young Descartes. If I am at once a
practicing doctor and a Cartesian inquirer who has not yet met the Pyrrhonist challenge, it will
be my cognitive optimism that keeps the level of my confidence up and that validates the
second and third maxims.
6. To close this first part of the book on a further note of optimism, although Descartes did not
leave us with a full account of objective value, which his theory requires, he left enough
material for us to reconstruct a good part of such a theory on his behalf; and this theory will
show (providing we grant some of his premises) that disease and suffering are genuine evils
that I can know to be such and that I am not mistaken about the value of my work. But, to
return to the third maxim, the point of the example here is that when I seek a true external
good in my action, I should limit my desire to what is entirely in my power, namely, my
intention to help the needy. 4.
____________________
4.
As we shall see in later chapters, this third maxim is the centerpiece of Descartes's more
developed moral theory. The argument I have sketched in this chapter is fully set out, with
the aid of a theory of the passions in the Passions of the Soul, particularly in articles 143 46.
-55Part 2
DESCARTES'S FINAL
MORALITY
SINCE Descartes himself refers to the morality of the Discourse as une morale par provision
and not, as he might have done, as une morale provisoire, there is no basis in the text for our
referring to Descartes's provisional morality, a pejorative term that suggests something
makeshift which one expects later to discard. To be sure, given the particular context in which
his morale par provision is formed, however deeply embedded it is in Descartes's character
and however respectful he is of its norms and ideals, this morale is at the same time bound to
be conceived as imperfect in some respects. It is, after all, a reasonable supposition that both
metaphysics and science may have some bearing on moral questions and on moral theory, so
that any morality we embrace is likely to inherit some of the imperfections that may infect our

beliefs in these other areas. But although the morale par provision may be regarded as
imperfect in this respect, it does not follow that it is to be succeeded by something quite
different. Now, we have seen already that the Discourse is not the only place where Descartes
refers to the morality to which he subscribed in the early years of his search for truth. In the
letter-preface to the French edition of the Principles, where he describes this morality as
sufficient to regulate the actions of life, he says expressly that it is imperfect and that he
follows it only par provision until he knows a better one. We should not infer more from these
descriptions than the texts support, however, and all they support is some distinction between
une morale par provision and a "perfect moral system." But they do not give us a clear
understanding of this distinction. Nonetheless, commentators tend to argue that the morale
par provision of the Discourse is a provisional morality, where this description is readquite
naturallyin a pejorative sense.
This tendency is, no doubt, encouraged by the famous passage from the letter-preface: "Thus
the whole of philosophy is like a tree. The roots are metaphysics, the trunk is physics, and the
branches emerging from the trunk are all the other sciences, which may be reduced to three
principal ones, namely medicine, mechanics and morals. By 'morals' I understand the highest
and most perfect moral system, which presupposes a complete knowledge of the other
sciences and is the ultimate level of wisdom" (I:186; 9B: 14 ).
Here again, if we do not go beyond what the text supports, we can say only that the perfect
moral system would be one formed in the context of relevant knowledge of metaphysics,
physics, mechanics, and medicine. This leaves some very large questions: How much of this
knowledge is relevant? How is it relevant? and What would a perfect moral system look like?
Note that he does not say that we need knowledge of all the other sciences in order to derive,
much less deduce, a perfect moral system, nor does he assert anything that would, on the face
of it, compromise the autonomy of morality. In short, this famous passage is not very helpful.
The best way to interpret it, in my view, is in the light of Descartes's correspondence with
Elizabeth and his Passions of the Soul. If this is right, however, thenat least so I will
arguewe will not find any radical shift in his conception of morality between, say, 1630 and
1650, nor any radical departure from the morale par provision.
If the letter-preface should whet our appetite for the morality to come, neither in it nor
anywhere else does Descartes give even a general outline of his conception of such a system,
much less spell it out in detail. But of course, we are hardly totally in the dark. For we should
expect its main outline to be very familiar. We expect a full Cartesian moral system, like any
other, to contain the following elements: (a) a set of moral precepts of the sort familiar from
commonsense morality (or some functional equivalent), (b) a defense of these precepts
(presumably sufficient to meet skeptical doubts), (c) an account of practical deliberation and
practical judgment in which these precepts play a role, (d) an account of moral virtue, and (e)
an account of the good life in which moral virtue has its proper place.
In the correspondence and the Passions of the Soul Descartes treats systematically and at
length d and e; indeed, he gives a well-developed account of virtue and of its place in the
good life. Descartes does not anywhere treat a, b, or c with a like thoroughness. In this
second part of my study I will set out Descartes's theory of virtue and of the good or happy
life. In the third part I will construct a Cartesian theory of the good and offer some
suggestions about what a Cartesian theory of the righta first-order moralitymight look
like.

-58CHAPTERFOUR
The Summum Bonum
I. BETWEEN the morale par provision of the Discourse and the final morality of the
correspondence and the Passions lie the metaphysics and the physics of the Meditations and
the Principles. The metaphysics is important principally because it provides the framework
within which we can work out the answer to the basic ethical question of how we are to live
happily. If we fail to get our metaphysics right, Descartes implies, we are almost bound to fail
to get our morality right. The physics is important since it is through knowledge of physics
and the special sciences that we can best gain control both of the world and of our own
character and temperament. Descartes stresses the role of metaphysics in a letter to Elizabeth
of September 15, 1645. After remarking that "nobody except God knows everything perfectly,"
i.
he writes that "we have to content ourselves with knowing the truths most useful to us." He
lists the following: we should accept calmly everything that
____________________
i.
This point, perhaps obvious in itself, has special significance within Descartes's moral
theory and is one he makes repeatedly. Its significance for moral theory is brought out in a
letter to Elizabeth of October 6, 1645. "I think also that there is nothing to repent of when
we have done what we judged best at the time when we had to decide to act, even though
later, thinking it over at our leisure, we judge that we made a mistake. There would be
more ground for repentance if we had acted against our conscience, even though we
realized afterwards that we had done better than we thought. For we are responsible only
for our thoughts, and it does not belong to human nature to be omniscient, or always to
judge as well on the spur of the moment as when there is plenty of time to deliberate.... it is
true that we lack the infinite knowledge which would be necessary for a perfect
acquaintance with all the goods between which we have to choose in the various situations
of our lives. We must, I think, be contented with a modest knowledge of the most
necessary truths such as those I listed in my last letter [of September 15]" (4:307-08;
3:269).
-59happens to us "as expressly sent by God"; our soul is a substance nobler than and independent
of the body; the universe is vast, and the earth has no privileged place or status in it; and we
cannot subsist entirely alone and are essentially dependent beings, part of "the earth, the State,
the society, and the family to which we belong by our domicile, our oath of allegiance and our
birth." Although he does not spell out the connecting argument, he seems to infer from our
being dependent on these various wholes that the interest of the whole must always be
preferred to our own (4:290-96; 3:265-67). 2. Oddly, he omits from this list a fifth
metaphysical truth that is most useful to know, namely, that we have free will of unlimited
scope. This truth bears importantly on the question of what our supreme good is, for in
Descartes's view it implies that our greatest perfection, in which we approximate to the
greatest possible degree the perfection of God, is the right use of our free will. That this is so,
moreover, we can know clearly and distinctly, and among things useful to know it is perhaps
the most useful, since our happiness depends on it.

2. Leaving aside the question of how our happiness depends on our free will until later, we
may begin with Descartes's account of what he means by happiness. For the most part, what
Descartes has to say on this topic is to be found in his correspondence with Princess Elizabeth,
3.
of which nearly two-thirds is taken up with morality. Early on in their moral correspondence
( July 21, 1645) Descartes proposed that they read Seneca's De vita beata together, so that
together they might clarify their thoughts about morality and determine "the means which
philosophy provides for acquiring that supreme felicity which common souls vainly expect
from fortune, but which can be acquired only from ourselves" (4:252; 3:256). Only a few days
later (August 4), however, he expressed second thoughts about his choice of text. In choosing
it, he explained, he had taken account "only of the reputation of the author and the importance
of his topic." 4. With Seneca's open
____________________
2.
In Descartes's era this was a moral commonplace, one with ancient roots. The plausibility
of this claim, however, varies depending on which whole is in question. What foundations
it has in Descartes's theory of value I consider in Chapter 8.
3.
Princess Elizabeth ( 1618-80) was the daughter of Frederick V of Bohemia and niece of
Charles I of England. Frederick, who was king of Bohemia only for one winter, was
defeated in 1620 by the Austrians, whereupon his family went into exile in Holland,
protected by the Prince of Orange. Descartes's correspondence with Elizabeth began in
1642. Descartes dedicated the Principles of Philosophy ( 1644) to her and wrote for her his
Traiton the passions of the soul in 1646.
4.
Several writers believe that Descartes had been introduced to the writings of Seneca at La
Flche ( Mesnard 1936, 4 ; D'Angers 1954, 453-59; Sirven 1928, 38 ). As to his reputation,
Seneca was very widely read in the late sixteenth and early seventeenth centuries in
France. For a list of studies that document the influence of Seneca on the writers of this
period, see Julien Eymard D'Angers, Recherches sur le Stocisme aux XVIE et XVIIE
sicles ( New York: Georg Olms Verlag, 1976).
-60ing remark, "All men want to live happily [vivere beata], but do not see clearly what makes a
life happy," Descartes agrees. But he objects that Seneca fails to have defined happiness or
told us how to achieve it. So, postponing further comment on Seneca, 5. Descartes confidently
expresses his own views. 6.
I would translate ["vivere beata"] into French as "vivre heureusement," if there were not a
difference between l'heur and la batitude. The former depends only on outward things: we
are thought more fortunate (heureux) than wise if some good happens to us without our own
effort; but happiness (batitude) consists, it seems to me, in a perfect contentment of mind and
inner satisfaction, which is not commonly possessed by those who are most favoured by
fortune, and which is acquired by the wise without fortune's favour. So vivere beata, to live
happily, is to have a perfectly content and satisfied mind. (4:263-64; 3:257) 7.
Descartes raises two traditional issues: What is the place of external goodshealth, wealth,
beauty, friends, etc.in the happy life? and What is the relation between being wise and
being content? He responds to the second question in his next letter ( August 18, 1645). To
answer the first, he begins by distinguishing between two classes of things which "can give us
this supreme contentment, ... those which depend on us, like virtue and wisdom, and those
which do not, like honours, riches and health" (4:264; 3:257). Virtue and wisdom are

sufficient for full contentment, he says, but external goods are necessary for more perfect
contentment. It is therefore within the power of each of us to live a fully happy or contented
life even if we are not well-favored by fortune.
It is certain that a person of good birth who is not ill, and who lacks nothing, can enjoy a
more perfect contentment than another who is poor, un____________________
5.
He resumes his commentary, unfairly critical in my view, in his next letter, but there again
he grows impatient with Seneca and elects to state his own view. To be sure, Seneca's
discussion does not have the economy of Cartesian prose, and one can see why a reader
like Descartes might have become impatient. Seneca's essay is repetitious and imprecise.
On the other hand, a charitable reader could have profited from it more than Descartes
claims to have done. Indeed, I believe that had Descartes taken more pains with his chosen
text, he might have appreciated some distinctions that would have helped him to make
clearer than he does what his own view isas we shall see directly.
6.
That he does so without much hesitation, and that the views he expresses are not only
consistent with those hinted at in the Discourse but adhered to throughout his later
correspondence and in his Passions of the Soul, strongly suggest that Descartes had given
some thought to these very questions before proposing to Elizabeth that they read Seneca
together. I do not agree with Alqui, therefore, who infers from the manner in which
Descartes suggests to Elizabeth that they read Seneca that Descartes did not have firm or
clear views about ethics (see Alqui1989, 3:585n).
7.
See also Discourse 6: 26 (I: 124) and Principles 9B: 4 (I: 180).
-61healthy and deformed, provided the two are equally wise and virtuous. Nevertheless, a small
vessel may be just as full as a large one, although it contains less liquid; and similarly if we
regard each person's contentment as the full satisfaction of all his desires duly regulated by
reason, I do not doubt that the poorest people, least blest by nature and fortune, can be entirely
content and satisfied just as much as everyone else, although they do not enjoy as many good
things. It is only this sort of contentment which is here in question; to seek the other sort
would be a waste of time, since it is not in our own power. (4:264; 3:257)
It has been suggested that Descartes has two conceptions of happiness, which he cannot
combine into a single, coherent view. 8. According to one, the full-contentment conception,
virtue and wisdom are sufficient for happiness; according to the other, the perfectcontentment conception, they are not. But some of the conceptual tension might be eased in
the following way. We can distinguish between the third-person perspective and the firstperson perspective of the deliberating agent. In speaking of full contentment, Descartes adopts
the first-person point of view of an agent. From this perspective, full contentment is all that
matters. When, on the other hand, he implies that some can be more perfectly content than
others, he adopts a third-person point of view, not the point of view of a deliberating agent.
And from this third-person perspective we can very well compare the degree of contentment
of two sages, one in robust health, intellectually acute, and wealthy and the other poor, sickly,
and dull. Moreover, that we can take this point of view and make such comparative judgments
is, in fact, a corollary of the prescription to seek our own full contentment; for in seeking our
own full contentment we try to enjoy as many of the gifts of fortune as we can. What we are
not to do, however, is allow our own contentment to become a hostage to fortune. For

Descartes, as we shall see later in this same letter, living happily is essentially a matter of
having rational aims, acting with firm resolution, and taking to heart that what we do not
attain was, given our well-informed efforts, impossible for us; living happily does not entail
self-denial, nor is it a matter, for example, of a sick sage not being able to see the value of
good health. Of the two conceptions of happiness distinguished above, that of full
contentment is dominant. Indeed, in the only context in which the distinction has practical
importin
____________________
8.
I have in mind Alqui, who writes: " Descartes seems here to reject the idea that the
happiness of the sage is perfect and to grant possible degrees of his own happiness: a sage
in good health is happier than one who is sick. Things that do not depend on us, therefore,
contribute to happiness. But he is going then to invoke another principle, defining
happiness in terms of the plenitude of desires regulated by reason. One is to desire only the
possible. Accordingly, each can render himself fully content" ( Alqui1989, 3:588n).
-62the act of deliberation or during the formation of one's own character the two conceptions
are extensionally equivalent. One's own perfect contentment just is one's own full
contentment. If, then, we are to live happily and be perfectly content, we must first of all
become virtuous, but we should by no means forego the contentment of gifts of fortune so
long as we do not allow our satisfaction of mind to become a hostage to such gifts as may be
denied us. 9.
Let us grant that virtue and wisdom are necessary for happiness and turn to the second
question, How are these related to full contentment? Later, when we examine the rules we
need to follow in order to live happily, we shall see the basis of a clear distinction between
virtue and wisdom; for the moment we may treat them as the same and speak simply of virtue
as the necessary condition of happiness. Virtue, on Descartes's accountto give a brief
sketchis the disposition always to judge as well as we can what is the best plan of action
and to adhere strictly to that plan. How, then, are we to understand the connection Descartes
has in mind when he implies that virtue is a necessary condition for full contentment?
When he turns specifically to this question in his letter of August 18, 1645, his first
observation is that
there is a difference between happiness, the supreme good, and the final end or goal toward
which our actions ought to tend. For happiness is not the supreme good, but presupposes it,
being the contentment or satisfaction of mind which results from possessing it. The end of our
actions, however, can be understood to be one or the other; for the supreme good is
undoubtedly the thing we ought to set ourselves as the goal of all our actions, and the
resulting contentment of the mind is also rightly
____________________
9.
In the text on which Descartes is commenting, Seneca writes of the wise and virtuous
person's attitude toward the gifts of fortune as follows: "Who, however, can doubt that the
wise man finds in riches, rather than in poverty, this ampler material for displaying his
powers, since in poverty there is room for only one kind of virtuenot to be bowed down
and crushed by itwhile in riches moderation and liberality and diligence and orderliness

and grandeur all have a wide field? The wise man will not despise himself even if he has
the stature of a dwarf, but nevertheless he will wish to be tall. And if he is feeble in body,
or deprived of one eye, he will still be strong, but nevertheless he will prefer to have
strength of body, and this too, though he knows that there is something else in him that is
stronger than body. If his health is bad he will endure it, but he will wish for good health.
For certain things, even if they are trifles in comparison with the whole, and can be
withdrawn without destroying the essential good, nevertheless contribute something to the
perpetual joy that springs from virtue.... And besides, who among wise menI mean those
of our school, who count virtue the sole good denies that even those things which we
call 'indifferent' do have some inherent value, and that some are more desirable than
others" ( Seneca 1965, 155 -57). If we deleted Seneca's talk of one's wishing for things we
cannot have, we would have in this passage a view very close to Descartes's own. Neither
is orthodox Stoicism.
-63called our end, since it is the attraction which makes us seek the supreme good. (4:275; 3:261)
The term "end" (fin) is, then, ambiguous; it may refer to the supreme good as the end we seek
to achieve in all our actions, or it may refer to the contentment which, Descartes asserts,
results from our achieving this end. In both distinguishing the supreme good from happiness
and connecting them in this way, he hopes to reconcile two respectable lines of thought,
Stoicism and Epicureanism. He is explicit about this. Indeed, the specific occasion for his
taking up this issue is Seneca's critique of Epicurus and defense of Stoicism in The Happy Life.
Seneca is at pains to identify happiness and virtue with each other and with the supreme good,
leaving the contentment of mind that follows on the heels of virtue as a separate value that,
strictly speaking, does not add to the value of happiness. It may be that Seneca does not
represent the classic Stoic position of Zeno and his immediate successors, but he certainly
adopts a position that cannot be reconciled with that of Epicurus. For this reason, as we shall
see, we cannot agree with Descartes when he says of the views of Epicurus, Zeno, and
Aristotle that, if interpreted favorably, they can be "accepted as true and consistent with each
other."
Descartes adds to his first preliminary observation a second, "that the word 'pleasure' was
understood in a different sense by Epicurus from those who argued against him. For all his
opponents restricted the meaning of the word to the pleasures of the senses; whereas he, by
contrast, extended it to every contentment of mind, as can easily be judged from what Seneca
and others have written about him." What Descartes says here is true only in part. If many
critics of Epicurus mistakenly suppose that all pleasures were of the sort Epicurus called
kinetic pleasures, others, Seneca among them, appreciate that Epicurus sought to extend the
concept of pleasure to cover static pleasure, the state of being without bodily pain or mental
anxiety. Nonetheless, Seneca sees himself as an opponent of Epicurus. He does not fault
Epicurus for the kind of life he led, which, he concedes, was in accordance with virtue. It is
Epicurus's account the supreme good he objects to. He upbraids him for putting pleasure first
in his theory and for making of pleasure itself the final good, even in the extended sense of
tranquillity. 10. Seneca's objection is that Epicurus has made of living a virtuous life a mere
means to the final good, not the good itself. And to the sugges____________________
10.
An incidental reason for upbraiding Epicurus is that by putting pleasure at the center of his

theory, he encourages others to moral laxity under the guise of a philosopher's theory. "The
reason why your praise of pleasure is pernicious is that what is honourable in your teaching
lies hid within, what corrupts is plainly visible" ( Seneca 1965, 131 ).
-64tion of compromise, that the highest good (summum bonum) could be a union of virtue and
pleasure, Seneca responds that the supreme good can have no part that is not good. "Even the
joy that springs from virtue, although it is a good, is not nevertheless a part of the absolute
good, any more than are cheerfulness and tranquillity, although they spring from the noblest
origins; for goods they are, yet they only attend on the highest good but do not consummate
it" ( Seneca 1965, 137 ). Seneca refers to the ancient Stoic doctrine of indifferents in this
connection, arguing expressly that the pleasure that attends the consciousness of one's own
virtue is "indifferent" to one's moral goodness and, hence, to one's living happily. Where does
Descartes stand? He may appear to reject this Stoic devaluation of pleasure; indeed, he may
appear to be defending the Epicurean view that virtue is a mere means to contentment (i.e.,
happiness, batitude). In fact, however, it is not easy to interpret Descartes's letter on this
point. In the first place, not only does he not assert, with Epicurus, that virtue has value only
as a means to perfect contentment, he says explicitly that virtue has supreme value as an end.
Evidently, if there is a consistent view to be extracted from this letter, it is not easy to come
by. And earlier commentators are not much help. ii.
____________________
ii.
One writes: "How, therefore, must we characterize the Cartesian morality? It is not a form
of utilitarianism; not even perhaps a form of eudaimonism, since pleasure is reduced to the
role of means of achieving the supreme good: it is as it were a path that nature uses to
direct us to perfection" ( Hamelin 1921, 384). Another writes: "In accepting this union of
virtue and happiness, Descartes inherits the eudaimonism of the ancients. This is
accompanied, however, by an effort to make clear the basis of rather vague doctrines and
to overcome the differences, superficial according to Descartes, among the peripatetics, the
Stoics and the Epicureans. The goods of the body to be sure have their value.... But not
being at the disposal of each, these goods cannot define a happiness open to everyone. The
Stoics saw clearly that the control of our thoughts alone is in our power and that this
constitutes virtue; but they were mistaken to make virtue 'so severe and inimical to
pleasure': to make clear that batitude is obtained through virtue permits reconciling
Stoicism, to which it confines itself, and Epicureanism, which subordinates all morality to
the pursuit of pleasure" ( Rodis-Lewis 1970, 43 ). And in contrast to these, Alquisuggests
that in seeing there is no strictly analytic connection between virtue and happiness and that
these need to be joined synthetically, Descartes here anticipates Kant ( Alqui1989,
3:596n). But if Descartes anticipates Kant in denying the analytic connection, he affirms,
while Kant denies, that contentment is within reach of every virtuous person in this life. In
so doing, however, it is far from clear that Descartes's conception of happiness (batitude)
is the same as Kant's conception of Glckseligkeit, which is a rationally constructed ideal
of imagination, roughly, the notion of the optimum satisfaction of all our desires. Between
Descartes and Kant came the empiricists and with them a highly distinctive conception of
happiness (anticipated by the Cyrenaics); this was a conception of happiness, however,
that is not at the same time a conception of a final end, in the classical sense of a summum
bonum. The comparison of Descartes with Kant, on this particular point anyway, is of very
limited value.

In a letter in which, in general, he suggests Descartes is much overrated as a philosopher

and particularly as a scientist, Leibniz gives him very low marks also as a moral
philosopher. "His
-65Let us return to the letter. Descartes continues: "Aristotle considered the sovereign good of all
human nature in general, that is to say, the sovereign good that can be had by the most
accomplished of all men. Thus he had reason to compose it from all the perfections of which
human nature is capable; but that does not serve our purpose." The principal debate between
the Stoics and the Aristotelians was whether external goods such as wealth, health, friends,
and family are parts of one's own happiness. Descartes may have this question in mind here,
since it follows from Aristotle's doctrine that it is not always within one's power to attain the
supreme goodhappiness. But Descartes seems to have a further point in mind: Aristotle
projects an ideal of the supreme good as the having of all the excellences both of intellect and
character that one can have, as well as an abundance of external goods. The supreme good in
this sense is even further out of reach for most of us. In any event, it is puzzling how
Descartes thinks he can reconcile Aristotle and the others. He does, however, agree with
Aristotle that external goods can make a contribution to one's happiness, something strict
Stoics deny. On the other hand, he diverges from Aristotle and approaches the Stoics in
affirming that happiness is attainable even by the least fortunate. We see this in his image of
the vessels of different sizes.
The letter continues:
Zeno, on the contrary, considered that which each man in particular can possess; that is why
he also had very good reason for saying it consists in virtue, because, from among the goods
we can possess, only virtue entirely depends upon our free decision. But, by making all vices
equal, Zeno has represented this virtue as so stern, and as such an enemy of pleasure, that it
seems to me it must have been only melancholics, or minds entirely detached from bodies,
who were able to be his disciples. 12.
____________________
morality is a composite of the opinions of the Stoics and Epicureanssomething not very
difficult to do, for Seneca had already reconciled them quite well" ( Ariew and Garber
1989, 241).
12.
This is a common complaint, but it is not worthy of Descartes. It may indicate that he was
not an attentive reader of Seneca, Epictetus, or Cicero or that he was not speaking
seriously. But there is, in this comment, an anticipation of a deep doctrinal difference
between him and the classic Stoics. Like the Stoics, Descartes takes the mind to be single,
but in contrast to the Stoics he asserts that the passions have a vital role to play in human
happiness. This doctrinal difference is worked out on the basis of his theory of the mindbody ingenium in his Passions of the Soul. The title of the closing article expresses a view
that would be anathema to the Stoic: "It is on the passions alone that all the good and evil
of this life depends" (II:488; 1:404). A point that he ignores in the Stoics, but which Seneca
makes in the text Descartes is commenting on, is that joy supervenes on virtue. This joy
(eupatheiai) is very similar, it would seem, to the plaisirs part that Descartes posits in his
Passions of the Soul. Referring to this letter, D'Angers points out that Descartes clearly
separates himself from his contemporary Christian humanists, who held the following trio

inseparablebeatitude, supreme good, and final endbut who denied that virtue itself is
the final end. He summarizes Descartes's view as follows: "Virtue being inseparable from
pleasure, understood in its true, spiritual sense, the Stoics were seriously mistaken in
presenting virtue to us as without joy.... Thus, with
-66Lastly, when Epicurus considered what happiness consists in and to what purpose or end our
actions tend, he was not wrong to say that it is pleasure in generalthat is to say, contentment
of the mind. For although the mere knowledge of our duty might oblige us to do good actions,
yet this would not cause us to enjoy any happiness if we got no pleasure from it. But because
we often give the name "delight" to false pleasures, which are accompanied or followed by
worry, anxiety and repentance, many have believed that this view of Epicurus inculcates vice.
Indeed, it does not inculcate virtue. Suppose there is a prize for hitting a bull's-eye: you can
make people want to hit the bull's-eye by showing them the prize, but they cannot win the
prize if they do not see the bull's eye; conversely, those who see the bull's-eye are not
thereby induced to fire at it if they do not know there is a prize to be won. So too virtue,
which is the bull's-eye, does not come to be strongly desired when it is seen all on its own;
and contentment, like the prize, cannot be gained unless it is pursued. 13.
That is why I believe I can conclude that happiness consists solely in the contentment of
mindthat is to say, in contentment in general. For although some contentment depends upon
the body, and some does not, there is none anywhere but in the mind. But in order to achieve
a contentment which is solid, we need to pursue virtuethat is to say, to
____________________
Descartes, and despite his deep and sincere Catholicism, with Descartes a new spirit
appears. Man is summoned to find his contentment in himself, in the achievement pure and
simple of his duty, independently of any recompense in the beyond" ( D'Angers 1976, 46869).
13.
This is a variation on a classic Stoic image. Descartes is trying to distinguish the final end,
the supreme good, virtue, and contentment. He wishes to argue that virtue is the supreme
good, that contentment results from our attaining the supreme good, even if some forms of
true contentment can have other sources as well. He is not, therefore, a good Stoic. In De
finibus bonorum et malorum, Cicero has Cato say: "A possible source of error must be
removed here: could it not be assumed that it follows that there are two supreme goods?
Not at all, ifwe compare the supreme good to the case of a man who has set himself to aim
at some mark with a spear or arrow: in such an illustration the man has to do everything
possible to shoot straight, nevertheless the actual doing of all he can to accomplish his
purpose is in this instance his supreme good (comparable to what we call the chief good in
life), whereas his hitting of the target is 'to be chosen,' and not 'to be desired' " ( Cicero
1991, 3.22). Descartes follows Cato in arguing that virtue is the supreme good, but unlike
Cato, he takes virtue to be the target rather than the man's doing everything possible to hit
the target. The fact is that Descartes shares Cato's view that it is the striving to shoot
straight that is virtue. And we might well say that Cato has got better control of the image
than Descartes, given his understanding of virtue, for Descartes's virtuous person needs a
target other than virtue itself, say, another's well-being. In the final paragraph of the letter
Descartes alludes to the needed distinction between the target and the striving to hit it,
where the latter is what is best to do (the appropriate action). Now, Cato might agree that
there is a prize that awaits those who attain virtue. If, however, Descartes means to say

that the prize is the motive, then, in Cato's view, virtue would be lost. On the other hand, if
being virtuous involves not only trying to hit various appropriate targets but doing so for
the right reasons, and if right reasons exclude thoughts about resulting contentment, then
the distance between Cato and Descartes is considerably reduced.
-67maintain a firm and constant will to bring about everything we judge to be the best, and to use
all the power of our intellect in judging well. (4:275-77; 3:261-62)
As I noted above, it is not clear that we can extract a clear and coherent account of happiness
from this letter. The letter raises a number of questions without providing clear, definite, and
mutually consistent answers. We may begin with the question whether, on Descartes's view,
virtue is a mere means to happiness. 14. The simple and, so far as it goes, correct answerthis
the letter makes explicitis that virtue is not merely a means to contentment; virtue is,
Descartes and Zeno agree, the supreme good, "the thing we ought to set ourselves as the goal
of all our actions." As we shall see below, setting virtue as our goal amounts to following
three rules of morality (related to those of the morale par provision of the Discourse). Having
virtue as our goal, then, is not like having some specific end in mind-say, saving a friend from
drowningand adopting the means to virtue is not like choosing an action as the appropriate
means to such an end. To speak in terms of Descartes's image, aiming at virtue is not well
understood on the model of aiming at a target. Nor, so far as we can conceive virtue as a
means, is this relation to be conceived on the model of a causal relation between a specific
action and its effect. Morality is not, then, like medicine. We may conceive a science in terms
of which we can relate certain medical procedures (seen as means) to the curing of a disease
(effect), but this is not a possible model for the relationship between virtue and contentment.
It is not one's virtuous actions that procure contentment, on Descartes's view, but the
knowledge that one is virtuous that does so. 15.
Up to this point, then, we might see Descartes as aligned with Zeno. How, then, does he
accommodate himself to Epicurus? We can't say precisely, because he does not go into
enough detail concerning Epicurus's theory of pleasure. Still, he may say enough to make us
doubt that the ac____________________
14.
This is the interpretation given by Martial Gueroult. Central to his reading is his claim that
for Descartes morality is "the technique that must determine in what way I must act in this
life in order for my soul to be full of contentment.... We see immediately that morality, like
medicine, can only be a technique that makes use of a theoretical science" ( Gueroult 1985,
179). Referring to the present letter he writes: "Virtue is the goal only subsidiarily, as is the
bull's eye in a target shoot: we aim for it only to win the prize, meaning pleasure" (184). I
discuss Gueroult's interpretation in Chapter 6.
15.
This remark should be qualified. Descartes speaks of contentment in two ways.
Contentment can be a passion, the passion called joy, but it can also be a state of mind
distinct from a passion, a plaisir part, in this case a contentment that supervenes simply
on the reflective knowledge that one is virtuous. As a passion, contentment or joy is
intermittent, the contentment in question here arising from the awareness that one is
performing a virtuous action from the proper motive.
-68-

commodation he suggests is possible. The problem has to do with the precise role
contentment plays in the practical deliberation of Descartes's virtuous agent. What the letter
strongly suggests is that unless one sees supreme contentment in the offing, one will not be
motivated to become virtuous. The apparent difficulty lies in trying to combine three things: a
theory of the will according to which it is moved by a conception of the good, a theory of
value that is not hedonistic, and a theory of motivation that is hedonistic. To begin to sort
things out, we need to distinguish between two contextsa context in which someone aims at
virtue as a target and a context in which someone who is virtuous aims at some good distinct
from virtue. In the latter, the agent who is already virtuous acts for the sake of the value of the
target, this value being at once such an agent's motive or reason for acting. But how precisely
are we to conceive the former case? Perhaps the idea is that agents who lack virtue are
typically motivated primarily by thoughts of their own pleasure and that in order to put them
on the path of virtue and true happiness we must praise virtue in motivational language they
can understand and respond to. 16. In any event, the letter does not contain the doctrine that the
virtuous person is motivated by thoughts of contentment, and it is quite consistent with the
view Descartes endorses elsewhere that contentment is what follows on one's being
characteristically motivated by judgments of the good.
To the reader of John Stuart Mill this letter will give rise to another question. Is the
contentment that results from the awareness of one's own virtue of a qualitatively different
order from contentment derived from our awareness of possessing other goods? And if so, is
this contentment commensurable with other varieties? If not, is virtue the supreme good only
in that it produces more contentment, or more certain and durable contentment, than other
sources? The Cartesian answers to these questions, whose defense I hope to provide in the
following chapters, are as follows. The contentment supervenient on virtue is not only sweeter,
more certain, and more durable than contentment from gifts of fortune, 17. it is qualitatively
____________________
16.
The discussion in the Passions of the Soul sheds little light on Descartes's thinking on this
motivational matter. Joy, he says there, is the sentiment that follows directly on the belief
that we possess some good. In the case of the good of virtue, which is, in Descartes's
taxonomy, an intellectual good as distinct from a sensory good, our knowledge that we
possess virtue is itself clear and distinct, and the joy that follows is, in the first instance, an
intellectual joy. Because we are embodied and therefore passionate, however, intellectual
joy must, as a matter of fact, reverberate in our passional life. At the level of
phenomenology, Descartes assures us, this passionate joy or contentment is not only the
most durable, it is the sweetest. This account, until the passions are introduced, is in accord
with Zeno.
17.
For example, Descartes writes in his letter of November 1647 to Queen Christina: "the
peace of mind and inner satisfaction felt by those who know they always do their best to
discover what is good and to acquire it is ... a pleasure incomparably sweeter, more lasting
and more solid than all those which come from elsewhere" (5: 85 ; 3:326).
-69different and phenomenologically of a higher order, so that even at the level of experience,
there could arise no question of a trade-off. Descartes appears to be working with a theory of
value that is not, at its base, hedonistic at all. Of the goods we can possess, virtue is supreme.
But the measure of the value of goods is not the degree or kind of contentment our thought of
possessing them produces. Indeed, as I shall detail later, the level and kind of contentment is

both explained and evaluated in terms of an antecedent judgment of the value of the goods we
possess. Accordingly, in the case of virtue, it is in terms of its being supreme among goods we
can possess that the contentment it produces is a higher form of contentment. 18.
In the letter just quoted, then, Descartes should not be seen as endorsing hedonism when he
objects to the rigor of Zeno's first account of virtue. Rather, he is pointing out that virtue is
one thing, contentment another, and that virtue carries with it its own reward. There is, of
course, the strong but highly problematic suggestion that were it not for this reward, virtue
would have fewer votaries. 19. This suggestion threatens incoherence, however, if it is taken to
imply that the virtuous person's motive for pursuing moral ends is dependent on this very
contentment. For what is essential to virtue is our judging as well as we can and sticking to
the chosen plan. Our targets, if we are virtuous agents, are such goods as the well-being of
others, knowledge, and health; judging well involves our deciding how best to achieve these
and other, often competing, ends in the circumstances in which we find ourselves. As I
interpret Descartes, such ends have genuine and objective value; they are worth pursuing for
their own sakes, and our reason for acting depends on our seeing these ends as having such
value. Accordingly, the entire justification for our acting well can be told without reference to
the contentment that supervenes on our reflective awareness that we have acted well. It would
seem, then, that no true votary of virtue needs the prospects of contentment as a motive. If this
is so, and on grounds that Descartes accepts, the suggestion in this letter that the virtuous are
moved by the offer of a prize is hard to reconcile with the rest of Descartes's account of virtue
as the supreme good.
____________________
18.
In this respect Descartes's accommodation of pleasure within the good life is similar to
Aristotle's.
19.
Pierre Mesnard seems to think that, according to Descartes, virtue would have no votaries
at all if it were not for the prize. " Zeno made perhaps a greater mistake when he separated
categorically the supreme good from beatitude. Descartes reproaches him for that which
Kant was later to be reproached: that virtuous actions from pure duty have little allure and
risk being rarely performed. However, our philosopher intends to make duty as attractive
as possible: it is by pleasure that he wished to lead to virtue" ( Mesnard 1936, 155 ).
-70CHAPTER FIVE
The Rules of Morality
I. DESCARTES's use of batitude and bonheur corresponds to a distinction between goods
that depend entirely on us and those that do not. Virtue and wisdom are goods of the first kind;
riches, honors, and health are of the second kind. i. Virtue is sufficient for batitude, that is,
for full contentment, whether we are otherwise favored by fortune or not. "It seems to me,"
Descartes writes to Elizabeth in his letter of August 4, 1645,
that every man can make himself content without any external assistance, provided that he
respects three conditions, which are related to the three rules of morality which I put in the
Discourse on the Method.
The first is always to employ his mind as well as he can to discover what he should or should
not do in all the circumstances of life.

The second is to have a firm and constant resolution to carry out whatever reason
recommends without being diverted by passion or appetite. Virtue, I believe, consists
precisely in firmness in this resolution; though I do not know that anyone has ever so
described it. 2.
____________________
2.
Anthony Levi suggests, on the contrary, that Descartes's account of virtue was fairly
commonplace at the time. Without suggesting that Descartes was not sincere in his claim to
originality, Levi notes that his definition of virtue is close to Du Vair's and to Charron's
defii.
The terms batitude and bonheur do not, perhaps, have modern translations that capture the
precise sense they had for Descartes. As Descartes makes clear in the letter to Elizabeth in
which he comments on Seneca's De vita beata, the term bonheur had the sense of good
fortune, which in modern French has been largely lost. On the other hand, the term
batitude is not nowadays commonly used and has in any case the flavor of religious
ecstasy, a flavor quite foreign to Descartes's intention. What Descartes had in mind was a
blend of contentment (a positive affect) and tranquillity (an unperturbed state).
-71The third is to bear in mind that while one thus guides oneself, as far as one can, by reason, all
the good things which one does not possess are all equally outside one's power. In this way
one will accustom oneself not to desire them. Nothing can impede our contentment except
desire, regret, and repentance; but if we always do what reason tells us, even if events show us
afterwards that we were mistaken, we will never have any grounds for repentance, because it
was not our own fault. (6:265-66; 3:257-58)
2. Descartes speaks of virtue and wisdom (sagesse). Although the two are not quite equivalent
in meaning, they are in Descartes's theory inseparable and, at least within the domain of
practice as distinct from pure theory, designate different aspects of one and the same character
trait. Thus, in the second rule Descartes uses the term "virtue" to refer narrowly to firmness of
resolution. Correspondingly, we might take "wisdom" to refer narrowly to the capacity to
judge well, but Descartes elsewhere expressly denies that firmness of resolution alone is
sufficient for virtue; good judgment is also necessary. In any case, when Descartes says that
both virtue and human wisdom are alone entirely in our power, I take him to be referring to a
single, complex character trait with these distinguishable aspects. I say "human wisdom," not
the perfect wisdom that only God could have. 3.
____________________
nition of preud'hommie. As DuVair writes in the Moral Philosophy of the Stoics, "The
good ... of man will consist in the use of right reason, that is to say, in virtue, which is
nothing other than the firm disposition to do what is virtuous [honnte] and appropriate
[convenable]" ( Levi 1964, 286-87).
3.
In the first rule of his Rules for the Direction of the Mind Descartes seems to suggest that
we could achieve perfect wisdom. Since all the sciences are interconnected, he says, one
should not focus on one in particular to the exclusion of the others. "He should, rather,
consider simply how to increase the natural light of his reason, not with a view to solving
this or that scholastic problem, but in order that his intellect should show his will what
decision it ought to make in each of life's contingencies." But this passage need not to be
taken to imply that with perfect knowledge in all the sciencesthe topic of the first rule

we would be able to determine with certainty in each particular circumstance what is the
best thing to do. And since it is very obvious that scientific knowledge would not lead to
such "perfect" practical knowledge, it would be uncharitable to read the young Descartes as
if he thought it would.
In the "Dedicatory Letter to Elizabeth" Descartes speaks of "pure and genuine virtues,
which proceed solely from knowledge of what is right," and which "have one and the same
nature and are included under the single term 'wisdom.' For whoever possesses the firm
and powerful resolve always to use his reasoning powers correctly, as far as he can, and to
carry out whatever he knows to be best, is truly wise, so far as his nature permits. And
simply because of this, he will possess justice, courage, temperance, and all the other
virtues; but they will be linked in such a way that no one virtue stands out among the
others." This theme of the unity of virtue is emphasized in the later Passions of the Soul.
Descartes concedes, however, by nature some have more powerful intellectual vision than
others, yet all "achieve wisdom according to their lights" (8A: 2 - 3 ; I:191).
-72We should accordingly see the above three rules as spelling out the full conception of human
virtue or practical wisdom. The first two rules refer to the two aspects I have just mentioned.
One is the strength and firmness of will to execute the plans we judge to be best; this power to
act well is one we develop to its maximum degree through our understanding of the nature
and use of the passions. The other is the power to judge well (bien juger). The third rule
completes the conception, since we have perfect virtue only when our desires are well ordered.
Without the addition of the third rule, we would have only what Aristotle called continence,
but not perfect virtue.
3. The first rule, which has to do with judging well, applies in two contexts. In the conduct of
our daily lives, we are called upon to decide what specific actions to perform in particular
circumstances. If we go by the few examples Descartes offers of practical decision making, he
appears to assume that well-formed practical questions about what is appropriate to do have
an objectively right answer, even if we are poorly equipped to determine what this answer is.
Sometimes the appropriate thing may also be the best; at other times it may be the only thing
permitted. So it might be said that, so far as such practical choice is concerned, we are wise in
proportion as our disciplined practical judgments tend to approximate what is, objectively
speaking, appropriate. 4. Perfect wisdom in this area, however, is beyond our competence. 5.
In the second context, our judgments are not practical decisions about particular acts in
particular circumstances but decisions that bear on longer range projects, including that of
forming our own character. And here something approaching perfect knowledge is not
beyond our competence. Indeed, it is this knowledge that Descartes is concerned to impart
and to cultivate. These two domains are of course connected; to achieve wisdom in the second
way, we must be striving to achieve wisdom in the first. 6. As
____________________
4.
In speaking of "appropriate action," I have in mind the Stoic notion of kathkon, the notion
at the center of Cicero's De officiis. This is a useful concept, since it ranges over actions
that are morally required, at one extreme, through those that are morally best, to those
that are just all right or natural, and which raise no moral issue. There is no express
reference to moral rules in Descartes's discussion of good judgment. But we have seen in

the Discourse that he is not reluctant to speak of moral laws, for example, the law that we
ought to serve mankind.
5.
In his letter of September 15, 1645, he writes to Elizabeth: "In order to be always disposed
to judge well, only two things seem to me necessary. One is knowledge of the truth; the
other is practice in remembering and assenting to this knowledge whenever the occasion
demands. But because nobody except God knows everything perfectly, we have to content
ourselves with knowing the truths most useful to us" (4:291; 3:265). This is, however,
overstated. The list of these truths does not include first-order morality.
6.
If mistake in this domain is unavoidable for us, how can we any longer maintain that God
is no deceiver? The answer, first, is that full contentment depends not on our never mak-73Descartes says reassuringly in this same letter, "It is ... not necessary [for batitude] that our
reason should be free from error; it is sufficient if our conscience testifies that we have never
lacked resolution and virtue to carry out whatever we judge the best course. So virtue by itself
is sufficient to make us happy in this life."
Nevertheless, the letter may seem to betray a certain ambivalence. 7. For while he asserts quite
straightforwardly that full contentment does not presuppose freedom from practical error,
later in the letter he writes: "Virtue unenlightened by intellect can be false," and in order to
prevent virtue from being false, we require the "use of reason," which will yield a "true
knowledge of the good." 8. Only if we have true knowledge of the good, moreover, will "the
contentment which virtue brings be solid." Here Descartes might be making the familiar point
that even when we judge as well as we can, we may be mistaken, a mistake we may later
discover. Yet if we have judged as well as we could have, then even ifwe later discover error,
we know that it was not our fault. That we judged well and acted well, so far as it might affect
our contentment, could only contribute to it. It is not easy to see, therefore, what could be
meant by saying that the possibility of our making a blameless error somehow undermines our
contentment. The best to made of it, perhaps, is that we can see ourselves as causally,
although not morally, responsible for some bad outcome. In any event, Descartes's final view
is quite clear; it is the one he expresses in his letter to Queen Christina of November 20, 1647:
"I do not see how it is possible to dispose [the will] better than by a firm and constant
resolution to carry out to the letter all the things which one judges to be best, and to employ
all the powers of one's mind in finding out what these are. This by itself constitutes all the
virtues; this alone really deserves praise and glory; this alone, finally, pro____________________
ing a mistake but on our tryingso far as our finite intellect allowsto determine what is
appropriate; second, as we see from the second rule, we are not acting under the delusion
that what we so choose is clearly and distinctly right, but we act as if we had such
knowledge. In short, we can attain full contentment even if we have no clear and distinct
knowledge of what is objectively right or if we mistakenly believe we have such
knowledge; and so far as we believe we have chosen as wisely as possible, we make no
practical error.
7.
Alqui, remarking a difference between the morale par provision and the "new" rules,
writes: "Descartes seems to hesitate between two conceptions of the good use of reason:
one anticipates the good will, in the Kantian sense; the other presupposes the knowledge of
true goods" ( Alqui1989, 3:589n).
8.
He appears to express a similarly extreme position in article 49 of the Passions of the Soul,

where he says that strength of soul amounts to virtue only when our practical judgments
are based not on "some false opinion" but "solely on knowledge of the truth" (I:347;
9:368). But the context of this claim is special. He is speaking of "determinate judgments,"
rules of thumb to be applied in repeatable types of circumstances. Here what is generally
called for is not a fine-tuned judgment but a checking of the influence of our passions on
our judgment.
-74duces the greatest and most solid contentment in life. So I conclude that it is this which
constitutes the supreme good" (5: 83 ; 3:325).
4. Descartes says that these rules state conditions that relate to the first three maxims of his
morale par provision. These are, in fact, the very conditions that would be satisfied by anyone
who adhered to the earlier maxims. There is, then, no difference in content or thrust between
the earlier and the later morality. There is, however, an apparent difference, as is often
remarked, between the first maxim of the earlier morality and the first rule of the later. The
contrast is supposed to be this: according to one (the first maxim) Descartes resolved to
follow the example of others, and according to the other he enjoins us to follow our own
reason. Still, as I argued in Chapter I, the decision to follow the example of sensible members
of one's own society must itself be based on one's own reason; our decision is supported by
the best reasoning available in the context of doubt, namely, reasoning from probabilities
reasoning at lower levels of sagesse than knowledge from first causes. 9. Accordingly, in
taking local custom and the example of sensible people as his guide in this context, Descartes
was doing precisely what the "new" first rule prescribes, namely, following his own reason. In
this respect, then, Descartes does not amend the principal import of the "old" first rule.
5. Of course there is this difference: by the time of this correspondence the credentials of our
own reason have been established (at least to Descartes's and Elizabeth's satisfaction). And
with this point in mind, one might look back on Descartes's quandary in the Discourse and
question whether he had any right to consult his own reason in the way I have just indicated.
The issue is a bit complicated. In the first place, the Descartes of the Discourse did not
question the credentials of his own reason or, more
____________________
9.
Descartes distinguishes levels of wisdom in the letter-preface to the Principles. Philosophy,
or knowledge from first causes, is the fifth degree of wisdom; Descartes's four rules were
to lead us to this level. The first level, he says, "contains only notions which are so clear in
themselves that they can be acquired without meditation." The second takes in all that we
acquire through sense experience; the third and fourth embrace what we acquire from
others, either by conversation or by reading their instructive books. As a pure inquirer,
Descartes denies that wisdom at these levels amounts to certain knowledge, yet in the
practical domain, since he cannot afford to await such knowledge, he must accept the
probablethat is, the best that can be gleaned from the ordinary sources of knowledge
concerning how to act and to live as if it were certain. Some commentators argue that in
his own philosophy he never achieved the fifth level of wisdom in the practical domain. To
be sure, he did not achieve it at the level of ordinary practical decisions about what to do
and what not to do. But as I have argued, this was never in the offing at any time in
Descartes's thought, since it would require something approaching divine wisdom. On the
other hand, he did show how we could actually achieve wisdom in the practical domain

that really matters, namely the wisdom sufficient for virtue and happiness. This is the story
I am following in this chapter.
-75particularly, his standard of vidence. His complaint was that so few claims to knowledge
truly meet this standard. Our possible skepticism about the standard itself is a feature of the
Meditations, not of the Discourse. In the Discourse Descartes begins, however boldly, with
the assumption that each of us has reason and that this is a genuine cognitive power. What we
need to do to gain knowledgeor such knowledge as is within our competence is to use it
well. The first maxim of the morale par provision, therefore, is not the maxim of someone who
is skeptical about reason's claim to be a genuine cognitive power. It is the maxim of someone
who does not have certain knowledge of first-order morality. Now, even if reason's
credentials as a genuine cognitive power have in the meantime been secured, the Descartes of
the correspondence with Elizabeth does not claim to have a surer grip on first-order morality
than he did a decade or more earlier. Moreover, as we shall see shortly, he continues to show
similar respect for common public morality.
But, to return to Descartes of the Discourse and his right to defer to his own reason, we might
put the matter as follows. Call "reason" the power we exercise when we try to distinguish the
true from the false and the good from the bad, or the better from the worse. That we have
such a power or that we act as if we dois implicit in our making these distinctions, indeed,
in our asking practical questions at all. In the context of the most radical kind of doubt, to be
sure, it is an open question whether this power, so understood, is a true power, such that,
when operating as well as possible, it tends to lead us to the goals we seekthe true and the
good. 10. After the Meditations (I am assuming) reason as a true power has been validated.
Indeed, in this change of epistemic status lies one main difference
____________________
10.
In this connection, we may note that Descartes never questions our capacity to understand
certain primitive notions, even within the context of doubt. Indeed, unless we had such an
understanding, the method of doubt could not even get launched. What is more, in the
Meditations, having isolated some propositions we apprehend clearly and distinctly, he
says that so long as we attend only to these, we cannot but give them our assent. Here
Descartes discovers the perfect epistemic marriage between our best cognitive faculty and
its object, truth. The problem posed by metaphysical doubt is that of showing that our best
cognitive faculty is genuinely cognitivethat our clear and distinct ideas accurately
represent existent beings as they truly and essentially are.
To be sure, in validating reason in the Meditations Descartes seemed concerned only to
validate our clear and distinct ideas as veridically representing reality. And yet, as I have
several times emphasized, at the level of first-order decisions about what it is best to do, all
things considered, we do not have clear and distinct apprehensionat least, not in
Descartes's view. Evidently, therefore, when he speaks of using our mind as well as we can
or of relying on our reason in practical contexts, he must be authorizing its use in areas
where we have no access to clear and distinct knowledge and where, moreover, we are in
some measure under the guidance of our passions, which are intrinsically obscure. I shall
examine the role of reason in Cartesian first-order morality in Part Three.
-76-

between the old and the new morality; nonetheless, it is not a difference in content but one of
reason's own credentials. And as I have just noted, the Descartes of the correspondence is not
significantly nearer to certain knowledge in the domain of first-order morality than before.
There is a further reason why we should not take the apparent difference in content of the two
first rules too seriously. As I argued earlier, it is a bit misleading of Descartes to say that he
elected to follow the example of the most sensible members of his community, for the
decision to follow good example was necessarily of a piece with his retaining his
pretheoretical moral convictions. The point of the first rule (as I analyzed it) was to explain
how someone engaged in pure inquiry could be thoroughly critical of one's own moral
convictions while at the same time retaining them as deep moral convictions. I will not repeat
the analysis here but only recall its conclusion: throughout the project of pure enquiry,
Descartes retains both his (pretheoretical) moral convictions and his respect for custom and
good example as his most reasonable practical policy.
Now, this policy has not in the least been shown unreasonable by anything that has occurred
in the transition from Cartesian doubt to post Meditations enlightenment. What is more, in
the few texts that bear specifically on this matter, Descartes continues to express this respect
for local custom. For example, in his letter to Elizabeth of September 15, 1645, he adds to the
several things it is most useful to know the following: "I have only this to add, that one must
also examine minutely all the customs of one's place of abode to see how far they should be
followed. Though we cannot have certain demonstrations of everything, still we must take
sides, and in matters of custom embrace the opinions that seem the most probable, so that we
may never be irresolute when we need to act. For nothing causes regret and remorse except
irresolution" (4:295; 3:267).
But what, it may be asked, establishes even the antecedent probability that local customs, the
laws of one's society, and, most of all, one's deepest moral convictions are worthy of the
respect of one whose final authority is one's own reason? Why is Descartes's self-confessed
morally conservative strategy rational? About one's moral convictions I have already given
my answer in Chapter I. But two farther points may be added that cover the customs and laws
of one's society. First, there is an argument grounded in Cartesian metaphysics, specifically in
the principle of the best. Earlier in this same letter, Descartes had written that
though each of us is a person distinct from others, whose interests are accordingly in some
way different from those of the rest of the world, we must still think that none of us could
subsist alone and each one of us is
-77really one of the many parts of the universe, and more particularly a part of the earth, the State,
the society, and the family to which we belong by our domicile, our oath of allegiance, and
our birth. And the interests of the whole, of which each of us is a part, must always be
preferred to those of our individual personality. (4:293; 3:266)
We are rooted, then, not just in the physical world, but in a social world with a particular
structure and history. From the perspective of God, these factsof our being lodged in a body
not as a pilot in a ship and of our being located in a particular social worldare perfectly
ordered and for the best. We may know that the conditions in which we live are divinely
ordered and for the best even when they do not so appear. Accordingly, we have some basis

for interpreting our own social practices, including our local commonsense morality, as
naturally good, as we also interpret our passionsbut with this caution: to neither passions
nor social directives are we to subordinate our reason.
In addition to this theological argument in defense of the common laws of society, however,
Descartes offers an entirely nontheological argument. At the end of his letter to Elizabeth of
January 1646, he defends the common laws of society as tending on the whole not only to the
public good but also to the private good. Those who depart from these laws, he says, may
sometimes do better for themselves, even at the expense of others, but their occasional
success, which it is not reasonable for them to expect, is due either to the ignorance of others
or to luck. ii. As for himself, Descartes says, "the maxim I have I have observed in all my
conduct has been to follow the main road [le grand chemin], and to believe that the utmost
cleverness is not to be too clever [de croire que la principale finesse est de ne vouloir point du
tout user de finesse]" (4:357). 12. This abiding thought echoes a passage from the Discourse
where he speaks of established public institutions:
These large bodies are too difficult to raise up once overthrown, or even to hold up once they
begin to totter, and their fall cannot but be a hard one. Moreover, any imperfections they may
possessand their very diversity suffices to ensure that many do possess themhave
doubtless been much smoothed over by custom; and custom has even prevented or
imperceptibly corrected many imperfections that prudence could not so well provide against.
Finally, it is almost always easier to put up with
____________________
12.
The volume and page reference here is to Oeuvres de Descartes. This part of the letter is
not translated in The Philosophical Writings of Descartes.
ii.
I shall examine this very interesting passage in more detail in Chapter 9. Suffice it to say
here that Descartes seems to be anticipating Hobbes's reply to the fool in Leviathan, pt. I,
chap. 14.
-78their imperfections than to change them, just as it is much better to follow the main roads that
wind through the mountains, which have gradually become smooth and convenient through
frequent use, than to try to take a more direct route by clambering over rocks and descending
to the foot of precipices. (6: 14 ; I: 118 )
And in the Passions of the Soul, speaking of pride, he says that it is a "kind of joy based on
the love we have for ourselves and resulting from the belief or hope we have of being praised
by certain other persons." Pride and its opposing passion, shame, share the useful function of
moving us "to virtue, the one through hope and the other through anxiety" (II:482; I:401). It
is not good, he says, to try to rid ourselves of these passions, even if, absolutely speaking,
solid contentment does not depend on what others think of us. In sum, there are several
reasons Descartes suggests for a moral conservatism, reasons that are in place after the
validation of reason in the Meditations.
6. I conclude that we satisfy the first condition of virtue and full contentment when we adhere
to the first maxim of the morale par provision. So it is in the case of the second condition; we
satisfy it when we adhere to the second maxim of the earlier morale. This maxim, however,
should benefit from the validation of reason. If it meets the standard of vidence, as I have

argued above, the maxim gains epistemologically in that this standard has in the meantime
been validated. In this respect, the second maxim is like a truth in arithmetic; both are
indubitable so long as they alone are clearly and distinctly at the center of our cognitive focus.
The second maxim, therefore, survives the full methodic doubt of the Meditations and emerges
essentially unaltered in the final morality of the correspondence.
Here again there is a superficial change. In the correspondence the second maxim is
presented as the principle primarily constitutive of virtue. Virtue, he implies, is strength of
will, capable of withstanding the appetites and passions that oppose it. In the Discourse, to be
sure, Descartes does not speak explicitly of virtue in connection with the second maxim, and
he does not dwell on the motivational influences that oppose our steady adherence to it. It is
clear, nevertheless, that we do satisfy the first and second conditions of full contentment just
when we judge as well as we can and when we steadily adhere to our best judgments,
whatever may oppose this steady adherence. For to be virtuous, on Descartes's view, is to
have a firm and constant resolution to do what reason requiresreason demanding this very
firmnessin conditions of both certainty and uncertainty about what is appropriate. Where
we are not certain what is appropriate, however, we are especially vulnerable to the
distorting effects of our passions, whose tendency is to make objects appear better than they
actually
-79are. Here not only must we be on our guard in our practical deliberation, we need
considerable strength of will to adhere to our best judgment which is more strength than we
typically require when we know for certain what we ought to do. Strength of will is needed to
overcome irresolution, on the one hand, and undisciplined appetite and excessive fear, on the
other. In the Discourse, as we have seen in Part One, Descartes had irresolution especially
irresolution bred by cognitive uncertaintyin his sights as the enemy of constancy. In the
correspondence but especially in the Passions of the Soul, he turns his attention to the
distractions of appetite and fear. Still, the conception of virtue is essentially unchanged from
the Discourse on.
7. Descartes comments on both the first and second rules throughout the correspondence and
in several articles of the Passions of the Soul. In the later works, however, he shifts the
emphasis away from vacillation of will owing to uncertainty to vacillation owing to
unregulated passions. Yet beyond this shift of emphasis, Descartes adds a new and distinctive
element, "determinate judgments bearing upon the knowledge of good and evil" (II:367;
I:347). There are three ways, he says, we can prevent our passions from diverting us from the
right path. 13. First, we can produce an opposing passion indirectly, by directly representing
to ourselves something with which the opposing passion we wish to have is usually joined. 14.
The value of this method, however, is limited; it may simply leave us with two passions which
"jostle the will in opposite ways ... rendering the soul enslaved and miserable" (II:367; I:347).
Second, we may by an act of will resist the movement of the body to which, unresisted, the
passion would otherwise lead, as when we resolve not to run from an object that excites fear
in us. Third, knowing that we will find ourselves in circumstances of various kinds in which
our passions will likely be arousedanger caused by an insult, indignation caused by an
injusticewe may arm ourselves with "firm and determinate judgments bearing upon the
knowledge of good and evil" (II:367; I:347). Such judgments, he says, are the "proper
weapons" for the will to employ in this conflict, and they are in fact the weapon of choice of

most people. Few are so weak, he says, that they always must follow their strongest passion;
most take guidance from some determinate
____________________
13.
Among them is not the possibility of willing an opposing passion, since the concept of
willing a passion is a contradiction in terms. In the case of a weak passion, Descartes does
say that we can simply choose to ignore it by attending to something else, in the way we
can ignore a slight noise by concentration (II:363-64; I:345).
14.
"For example, in order to arouse boldness and suppress fear in ourselves, it is not sufficient
to have the volition to do so. We must apply ourselves to consider the reasons, objects, or
precedents which persuade us that the danger is not great; that there is always more
security in defence than in flight; that we shall gain glory and joy if we conquer, where we
can expect nothing but regret and shame if we flee; and so on" (II:363; I:345).
-80judgments or other. Yet just as we may err in our practical judgments about what is
appropriate here and now, we may also err in forming these more general practical judgments,
designed for occasions of certain kinds. In forming "determinate judgments," it is especially
important therefore to judge well. For these are practical judgments destined to become
embedded in our habitual responses to situations where we are most likely to err through
haste or bias; if we go wrong in forming them, we lay ourselves open to a life vulnerable to
remorse, even though we may overcome our present irresolution.
Descartes's introduction of determinate practical judgments as constituents of good
deliberative practice might seem to require that we amend our interpretation of both the
second maxim of the morale par provision and the second rule of the "new" morality. To see
whether this is so, consider Descartes's case of the man lost in the forest, a case of decision
under conditions of uncertainty. When we are uncertain about the quickest way out of the
forest and decide to walk in a straight line from where we are, we may be said to have judged
well (even if falsely). Indeed, as Descartes presents the case, the directive to follow a single
straight line out of the forest seems to be recommended as a good rule of thumb to follow in
such circumstances. In one respect, this directive is like a Cartesian determinate judgment, for
it is a general rule with sound empirical credentials. But whether it qualifies as a Cartesian
determinate judgment or not, it is not paradigmatic of what, I believe, Descartes has
specifically in mind in the Passions. Cartesian determinate judgments seem rather to be
settled policies or practical maxims having general and repeated applicability and designed
expressly as proper "weapons" of the will in its conflict with strong passions. Indeed, the
theory of the passionswhich Descartes writes as a physicist (en physicien)has its chief
practical utility in showing how we may embed determinate judgments in our own emotional
structure, a topic I shall treat below.
Let us suppose that we can formulate and then embed determinate judgments in our character
in the form of conative dispositions. We become, say, the sort of person who does not respond
inappropriately to insults, however angry they may make us. Why bind ourselves to general
rules, when in particular circumstances we may discover good grounds to make exceptions?
On our way out of the forest we may run across a well-marked trail and signpost that give us
excellent reason for changing course. We would be irrational not to respond to this new
evidence. Might not determinate judgments likewise lead to irrational action? Might they not

preclude practical deliberation leading, say, to the conclusion that in a particular case it is
appropriate to respond angrily to an insult?
-81The answer to this question will depend on what form the determinate judgment takes. As I
noted, deciding whether we should commit ourselves to general maxims or policies and what
specific form these should take requires sound judgment in the light of available information
about the relevant probabilities. But it is certainly up to us to leave our judgments somewhat
open-ended. Indeed, in one sense the model for such judgments might well be the decisions
made by someone lost in a forest, since they may take the form of policy judgments, which in
certain specific cases are defeasible. Descartes, to be sure, unfortunately for his commentators,
does not concern himself with these details, leaving such matters to be settled by our
experience and commonsense. Such is the spirit of his rather off putting response to
Princess Elizabeth, who, having asked how we are to determine when we have done enough
to promote the well-being of others, was told unhelpfully to let her conscience be her guide.
I have already noted that determinate judgments are destined to take the form of habits. So
they are a subclass, if an important one, of policy judgments. They are also designed with an
eye to our occasional liability to strong emotional reactions, which tend to distort our ability
to deliberate well. In general, they will be dispositions for dealing with anger, fear, lust, envy,
and amour-proprethe usual passions that tend to deflect us from pursuit of the right and the
good. Accordingly, many of them will be, in the narrow sense, moral judgments concerning
what is, at least presumptively, inappropriate. 15.
8. Finally, we satisfy the third condition for full contentment when we have followed the third
maxim of the earlier morale. We are to control our own emotional life so as to be unperturbed
by evils entirely beyond our power to prevent. Our happiness depends entirely on our own
virtue and on our possessing goods within our power. The enemies of full contentment are
remorse, repentance, and desires we cannot satisfy. By following the first two rules we gain
mastery over repentance and remorse, and by following the third we gain mastery over our
desires. 16. In particular, we are to rid ourselves of desires that have no instrumental value for
us and to retain only desires that do have such value, for the effect of having a desire for some
object is to make it more likely that we will try to attain that object, other things being equal.
A desire for x enhances our motivational
____________________
15.
Descartes is nowhere explicit about the form an ideal first-order morality would take.
16.
"Desire" in this context means passion in Descartes's sense. This is a somewhat restricted
use of the term. We might have conative dispositions directed toward ends without these
ends being objects of desire in this sense. Indeed, many of our actions will be directed
toward ends that it is properrationalto have without these ends being objects of desire.
I discuss Descartes's account of the passions in the next chapter.
-82tendency to try to get y. Now if, assuming we are free from remorse and repentance, we are
not fully content, this must be because we have desires that cause us impatience or sadness;
and desires of this sort are in every case desires based on some false belief, either about the
value of the object or about the possibility of our attaining it. "We do not desire to have more

arms or more tongues than we have," Descartes writes, "and yet we do desire to have more
health or more riches. The reason is simply that we imagine that the latter, unlike the former,
can be acquired by our exertions, or are due to our nature. We can rid ourselves of that
opinion by bearing in mind that since we have always followed the advice of our reason, we
have left undone nothing that was in our power; and that sickness and misfortune are no less
natural to man than prosperity and health" (4:266; 3:258).
The line of thought here is precisely that found in the Discourse (cited above), where he
writes that once we see clearly that external goods are beyond our power, we shall not regret
their absence "any more than we regret not possessing the kingdom of China or Mexico" (6:
26 ; I: 124 ). We shall meet this idea again in several articles in the Passions of the Soul. In
brief, it says that we should limit our desires to objects that are entirely in our power. In the
case of things that seem only in part to be in our power, for example, objects we can attain
only through acting, we should limit our desire to that part, leaving the rest to Providence. To
be agitated by desires that range more widely is to suffer uselessly. Suppose I judge well and
pick the safest route to a worthy destination. My desirethat is, my passion of desireshould
be limited to my taking this route; it should not be fixed on my reaching my destination. If
"Providence" decrees that I shall be set upon by robbers on this route, then I discover that my
original goal was not possible.
Reason insists that we choose the route which is usually the safer, and our desire in this case
must be fulfilled when we have followed this route, whatever evil may befall us; for, since any
such evil was inevitable from our point of view, we had no reason to wish to be exempt from it:
we had reason only to do the best that our intellect was able to recognize, as I am supposing
that we did. And it is certain that when we apply ourselves to distinguish Fatality from
Fortune in this way, we easily acquire the habit of governing our desires so that their
fulfillment depends only on us, making it possible for them always to give us complete
satisfaction. (II:440; I:381)
As here sketched, this may not seem a compelling line of thought. One might ask, somewhat in
the spirit of William James, whether we might not profit from desires in cases in which having
the desire in question is itself a
-83necessary condition of our being able to attain its object. I leave further discussion of this
matter to the next chapter. My chiefconcern here has been to demonstrate the continuity
indeed, the near identityof the earlier and later moralities. Nothing in the correspondence
hints that Descartes ever altered his conception of virtue or of what we must do to achieve it.
-84CHAPTERSIX
Gueroult's Account
of Descartes's Ethics
I. IN the second volume of his study of Descartes's philosophy, Martial Gueroult devotes two
substantial chapters to Descartes's ethics. i. In the first, he argues that according to Descartes
morality is a mere technique. I shall set out below what this interpretative claim amounts to.
In the second, having pointed out that we can identify three different conceptions of morality

in the development of Descartes's thinkingthe provisional morality of the Discourse, a

perfect morality enabling one to determine the best thing to do on each occasion of rational
choice, and a conception of morality as a striving for virtue or as a kind of morality of
intention Gueroult argues that after trying to achieve a perfect morality according to the
second conception and finding it impossible, Descartes eventually settled for a final morality
according to the third, a morality surprisingly similar in content to the provisional morality. I
think that neither interpretative claim gives us the best reading of the Cartesian texts. I do not
believe that Descartes ever conceived morality as a mere technique, nor do I think that
anything in the relevant texts obliges us to suppose that Descartes changed his conception of
an ideal morality for human beings in any essential way between 1637 and his death.
2. Let us begin with the first interpretative claimthat Cartesian morality is a mere technique.
What does this claim amount to? It has two distinct
____________________
i.
Gueroult 1985, chaps. 19, 20. Until very recently these two chapters contained the only
comprehensive discussion by an established scholar and available in English of Descartes's
treatment of moral subjects. They merit a critical assessment.
-85elements: moral virtue is a mere means, and the end that morality serves is the agent's
maximum pleasure. 2.
According to Gueroult, at least at the time of the Discourse, Descartes conceived morality as
a technique "comparable to medicine" ( Gueroult 1985, 2:178). Just as it is given to medicine
that its end is "to heal the sick and to preserve health," so it is given to morality that its end is
happiness. And just as medicine is properly conceived as a technique or applied science, so
too is Cartesian morality. To be sure, it might seem that one business of the moral theorist is
to determine the proper end of human action. But according to Gueroult, Descartes simply
dismisses this problem, or rather simply presupposes that we already know by the natural
light what our proper end is and need only to discover how to realize it. 3. Morality, then, is
comparable to medicine. Both are mere techniques and both are concerned to promote the
final ends of man as man. Medicine aims to preserve the integrity of the body, and morality
aims to promote maximum contentment of the embodied mind. For Gueroult, "Morality is
therefore the technique that must determine in what way I must act in this life in order for my
soul to be full of contentment, in spite of the fact that this soul is not only pure mind, but is
also substantially united to a body that plunges it into a natural and social world whose
vicissitudes are infinite. We see immediately that morality, like medicine, can only be a
technique that makes use of a theoretical science" (2:179). On Gueroult's interpretation, then,
Cartesian morality is to contentment as medicine is to health. And just as medicine may be
seen as applied science (physiology), so morality may be seen as the ap____________________
2.
If morality is a technique, then a perfect morality would be an infallible, perfectly effective
technique. Accordingly, one who had knowledge and command of a perfect morality in
this sense would be armed with an effective decision procedure for deciding on and
electing the very best thing to do in any particular set of circumstances in which the
practical question, What should I do? might arise. Of course, one might well conceive
morality as a technique but also believe that no such perfect morality could be within the

competence of any human being. As we shall see, Gueroult suggests that Descartes did
think at one time that he might achieve a perfect morality of this sort, just as he thought at
one time that he might achieve a perfect medicine.
3.
"Natural light, along with natural instinct (both guaranteed by divine veracity), reveal to us
immediately, without the least doubt, that our end is happiness in the present life. It
remains for morality only to furnish the technique capable of realizing this: 'There is no
person who does not desire happiness, but there are some who do not have the means'"
( Gueroult 1985, 2:178). This quoted passage is from Descartes's letter to Elizabeth,
September I, 1645, the letter in which Descartes examines Seneca's De vitae beata. I shall
give reasons below why I disagree with Gueroult's reading of Descartes. I note here only
that it does not follow from Descartes's remark in this letter to Elizabeth. No doubt all of
us desire happiness, but it does not follow that happiness is or should be the sole or final
end of all our actions. What is more, whether or not happiness is an end realizable by a
"technique" depends on how we conceive it.
-86plication of those sciences that set out the causes of contentment (mechanics, physiology, and
psychology). 4. Summarizing this part of his discussion, Gueroult concludes: "Since it is
turned toward an end external to virtue, and since it requires an action modifying material
nature as much as my own nature, asking the exact sciences to illuminate these means,
Cartesian morality is truly a technique, an applied science" (2:187).
3. As I said above, I do not think we should interpret Cartesian moralityearly or lateas a
mere technique in the sense Gueroult intends. But let us suppose that Gueroult is correct.
Then the idea naturally suggests itself that the intended successor to a provisional morality
would be a perfect scientific morality consisting of what Kant called assertoric hypothetical
imperatives (counsels of prudence) with the definiteness and effectiveness of what Kant
called problematic hypothetical imperatives (rules of skill). A perfect morality, so conceived,
would evidently be a morality in the broad sense.
Morality conceived as a mere techniqueas a set of technical imperativeswould be an
applied science. Which one? Medical science, suggests a famous passage from part 6 of the
Discourseat least a medical science such as Descartes hoped at the time he would be able to
develop. The maintenance of health, Descartes writes, "is undoubtedly the chief good and the
foundation of all the other goods in this life. For even the mind depends so much on the
temperament and disposition of the bodily organs that if it is possible to find some means of
making men in general wiser and more skillful than they have been up till now, I believe we
must look for it in medicine" (6: 62 ; I: 143 ). Descartes's expectations for rapid progress in
medicine were, of course, disappointed, and his developing anthropology also altered his view
of medicine itself; yet he did not feel required to retract what he actually says in this passage,
which is that good medical science would be a vitally important human good. For where bad
health takes away our capacity to reason, Descartes says later to Princess Elizabeth, a remedy
to restore health would be indispensable for contentment. 5.
But if improved medical science could deliver more effective counsels of prudence, only
fantasy could produce the thought that a perfect medicine will someday supply an effective
algorithm for maximum contentment whenever the question of contentment arises. The above
passage, of course,

____________________
4.
To anticipate in some measure the argument below, I should note that the complex causal
relation in Descartes's theory between moral action (in the sense of action expressive of
virtue) and contentment is not straightforwardly one of means to end. In the first place, it is
a connection necessarily mediated by a judgment of value; second, what is judged to have
value as an end is virtue itself.
5.
Letter of September I, 1645 (4:282; 3:262).
-87implies no such promise. And the same could be said of any natural science, not only of the
two other sciences on Descartes list, physics and mechanics. What is more, not even
physiological psychology, as Descartes came to conceive it, could supply such a prudential
algorithm. 6. What it can provide is important for morality as Descartes conceived it, but its
application, as we shall see, is not principally prudential casuistry. Indeed, what we are to
learn from this science, at least the part of it having to do specifically with the passions, is that
these can be very poor casuistical guides in particular cases. The plain fact is that all the
sciences together could not in principle give us knowledge of the sort that would always save
us from first-order practical errorfor the obvious reasons Kant pointed out: counsels of
prudence are not rules of skill. It would be uncharitable, then, to interpret Descartes as
holding otherwise, even should we (in my view mistakenly) read him as conceiving morality
as a technique.
4. Nevertheless, Gueroult appears to believe that it was just such a perfect morality that
Descartes was heralding in the Discourse and referring to in the letter-preface to the
Principles. Thus, he sees the question Descartes faced in his scientific work to be "whether
this morality-science (morale science) can be constituted, whether the technique of
happiness can suffice for its end, can furnish every man whatsoever a practical law indicating
to him what it is suitable for him to do or not to do in every circumstance, 'encompassing in
all cases and for all reasonable beings the same determining principle of the will' "
( Gueroult 1985, 2:191). 7. In Gueroult's reading, as we shall see shortly, Descartes
abandoned hopes for achieving a morality so conceived. But as I noted, only when we are led
to think of Cartesian morality as a technique do we ask whether a "morality-science" of this
sort can be formulated.
5. I now return to the original question of whether we should think of Cartesian morality as a
mere technique in the first place. Is virtue a mere means, on Descartes's view? Here we have
to ask, How does Descartes come down on the question of whether our proper final end is
virtue or happi____________________
6.
Gueroult makes this point, to be sure, but he does so in the context of trying to show that,
owing to this very fact, Descartes had to depart from his original ideal of a perfect
morality. In fact, on Gueroult's account, this was Descartes's second departure from his
earlier ideal.
7.
The quotation is from Kant, Critique of Practical Reason, theorem 3, scholium 3, sec. 3.
That Gueroult chooses to quote Kant in this context indicates very well how he interprets
Descartes's conception of morality as a technique. On this conception, the principles of a
morality would all be assertoric, hypothetical imperatives. On the contrary, as I shall
contend, Descartes does not construe first-order moral precepts as imperatives of precisely

this kind. What is more, contrary to Gueroult's interpretation, I read Descartes as

declaring that virtue itself is the proper end of action, not a mere means to contentment.
-88ness? Gueroult believes Descartes firmly commits himself to Epicurus's answer, not to the
Stoic's. Descartes does argue that we need to detach ourselves from "pleasures arising from
the body," to be sure, but as Gueroult reads Descartes, this is because in so doing and in
devoting our energies to "pleasures arising from the soul," we realize more pleasure in the
long run ( Gueroult 1985, 2:186). It is because our greatest contentment arises from the
consciousness of our own "strength of soul" that Descartes endorses stoic detachment. The
essential point, however, is that, for Gueroult's Descartes, contentment, not virtue itself, is the
end. "Virtue is the goal only subsidiarily, as is the bull's eye in a target shoot: we aim for it
only to win the prize, meaning pleasure" (2:184). Evidently, the question of whether virtue is
a mere means is closely connected in Gueroult's understanding to the question of Descartes's
hedonism. Let me deal briefly with this latter question and then return to the former. Did
Descartes hold a hedonistic ethical theory? He certainly sings the praises of contentment. But
the question is whether he took contentment to be the sole intrinsic good toward which our
actions should aimor do aim. As I have several times pointed out, principally in Chapter 4,
the answer is unequivocal: he did not. I have already noted several of Descartes's texts which
imply that contentment is not the only intrinsic good, much less the supreme good. Gueroult
puts Descartes squarely into Epicurus's camp, but this is correct only if, in Descartes's view,
the contentment that supervenes on virtue is the end toward which the virtuous person strives.
Although in his letter on this topic Descartes does make remarks that might lead to Gueroult's
interpretation, we are nonetheless bound to reject it, since it is at odds with the central
doctrine of Descartes's theory of value, namely, that virtue is the supreme good. Moreover, it
is at odds with the motivational view implicit in his account of virtue, according to which the
virtuous agent's primary intention is to promote goods other than his own contentment, even if
his own contentment is the crown of his achievement. Here I add one further argument against
taking Descartes to be an egoist hedonist. The contentment said to supervene on virtue
presupposes the value of virtue as an end. It would be incoherent for Descartes to argue that
virtue is a mere means. But he does not argue this way. Rather, he claims that virtue is a true
good for us (a perfection), that we can know this to be so, and that it is from our knowledge
that we are virtuous that our sweetest and most durable contentment arises. This remarks
holds both for intellectual contentment and for the passion Descartes labels gnrosit. If
virtue were not an intrinsic good, this contentment would not, on Descartes's view, be
possible. This last point, concerned specifically with virtue as a good, can be generalized.
Descartes holds that we achieve contentment, as distinct from the bodily
-89sensation of pleasure, through our believing that we are suitably joined (either by love or
possession) to some true good. The reason our highest contentment derives from our being
and knowing that we are virtuous is that being virtuous is our greatest possible perfection and
the only perfection entirely within our control. 8.
6. So much, then, for the interpretative claim that Cartesian morality is a mere technique. Let
us turn to Gueroult's second main interpretative claim. It is that, at first, Descartes had hoped
to achieve a "morality science" along the lines sketched above but that this effort finally
miscarried to a large extent; in the end, Gueroult argues, we find Descartes turning away from

a moral science conceived in terms of rules for the direction of the will, telling us what to do
and not to do in every circumstance, and toward a kind of morality of intentionfrom a
morality conceived materially to one conceived formally. "That will be the drama of Cartesian
morality, which is played across the inextricable complication of a provisional doctrine which
heralds a definitive doctrine that must replace it, and of a definitive doctrine, which, on the
contrary, confirms and consolidates, once and for all, the morality that we thought to be
provided" ( Gueroult 1985, 2:192). This passage concludes the first of Gueroult's two chapters
on ethics. The next picks up its theme in its title: "Some Consequences concerning Medicine
and Morality: Three Ideas of Medicine and Morality." Suppose we begin, as Descartes
appears to do in the Discourse, by conceiving of morality as a "science of life" which, in
Descartes's words, allows "man to distinguish the true from the false in order to see clearly
with respect to his actions, and to walk with assurance in this life" (6: 10 ; I:115). Then we
shall discover that in the strictest sense of Cartesian science, such a science of life is
impossible: on the one hand, such a science would be re____________________
8.
Two more items can be mentioned here. In his letter to Elizabeth of October 6, 1645,
Descartes clearly rejects pure hedonism, arguing that it is better to have knowledge than to
have pleasure. "If I thought joy the supreme good, I should not doubt that one should try to
make oneself joyful at any price, and I should approve the brutishness of those who drown
their sorrows in wine, or assuage them with tobacco. But I make a distinction between the
supreme goodwhich consists in the exercise of virtue, or what comes to the same, the
possession of all those goods whose acquisition depends on our free willand the
satisfaction of mind which results from that acquisition. Consequently, seeing that it is a
greater perfection to know the truth than to be ignorant of it, even when it is to our
disadvantage, I must conclude that it is better to have less cheerand more knowledge"
(4:305; 3:268). And this from the early Rules for the Direction of the Mind: "But I am
convinced that certain primary seeds of truth naturally implanted in human minds thrived
vigorously in that unsophisticated and innocent ageseeds which have been stifled in us
through our constantly reading and hearing all sorts of errors. So the same light of the mind
which enabled them to see (albeit without knowing why) that virtue is preferable to
pleasure, the good preferable to the useful, also enabled them to grasp true ideas in
philosophy and mathematics, although they were not yet able fully to master such
sciences" (10:376; I: 18 ; my italics).
-90stricted to knowledge through clear and distinct ideas; on the other, practical deliberation has
to do not with essences but with "existences and existential circumstances known by means of
the senses" ( Gueroult 1985, 193). So there can be no practical science in this strict sense.
Nonetheless, we can know, in this sense, what the role of sensation is and how sense
experience may be reliably brought to bear to help us to "walk with assurance" in this life.
Now, one science that may inform practical choice is mechanics. Another, which has to do
not with an essence but with the substantial union of mind and body, is the science of
psychology, as set out in the Passions of the Soul. From this science, we can learn that the
passions are in their nature good and, when properly regulated, serve as reliable guides to
action; and we can also learn the available techniques for regulating our passions. But this
knowledge cannot entirely satisfy the will. "For this, it would have to attain more than mere
general rules. It would have to furnish us, with respect to each case likely to occur in daily
experience, the means to discover the exactly appropriate formula for action, that is, the

knowledge of what is useful to do in order to assure both the preservation and happiness of
our life" ( Gueroult 1985, 2:197). From knowledge of such general rules we cannot derive
knowledge of what is best to do in every circumstance of our lives. If we are to act in the light
of reason, our finite reason does not give us full illumination. Although, as Descartes puts it
in the Discourse, those who judge well act well, in our practical judgments we can at best
only approximate judging well.
7. In looking over the development of Descartes's ethics, Gueroult finds three distinct
conceptions, of which two are set out in the early Discourse. According to the first, morality
is conceived as "an exact science." 9. Such a science, following on the heels of the similarly
exact sciences of mechanistic physics and medicine, would be the final fruit of the tree of
philosophy. "Wisdom, which is its aim and ultimate principle, depends on the Supreme Good.
The latter is the knowledge of the truth through first causes. This knowledge, to which the
Principles of Cartesian Philosophy gives us the means to accede, must allow us to reach the
highest degree of wisdom, meaning the fifth degree. That is the first idea of morality"
( Gueroult 1985, 2:201). The second conception is that of a provisional morality, a set of
maxims to serve us until such time as we are able to establish a definitive morality according
to the first idea. Descartes, as we know, never set out such a perfect or definitive morality.
And although some have suggested that he never abandoned the belief that such a morality
could be worked
____________________
9.
Gueroult finds this conception also in the letter-preface to the Principles. Indeed, he refers
expressly to this preface when he uses the term "exact science."
-91out, Gueroult thinks this unlikely. In the first place, a morality of this kind would "violate the
interdiction against applying the understanding alone to the substantial composite." "He must
therefore, for metaphysical reasons, and not for external reasons renounce the idea of a perfect
morality, meaning an exact morality, having for object only clear and distinct mathematicodeductive ideas drawn from physics"( Gueroult 1985, 2:201-2). In the Passions of the Soul,
however, we find, not a morality as an exact science of some pure essence, but a scientific
morality of the substantial union. In this late work, Gueroult notes, Descartes "wished to
explicate the passions not as a rhetorician nor as a moral philosopher but as a natural
philosopher" ( Gueroult 1985, 2:202). This science "assembles items of evidence of different
kinds, some purely rational ones concerning either extension alone (physiological
mechanisms), or the soul alone (various functions ofthe soul, the emotions stemming from it
alone, such as contentment, which comes solely from within), and others that are above all
sensible, concerning the sensation of passion itself" (2:203). Now, this third conception of
moralitythe morality of the Passions and of the correspondenceis one Descartes did
substantially realize and also offered as the replacement for the provisional morality.
Evidently, if this is Descartes's final morality, a morality presented, moreover, as a
"technique" sufficient for our maximum felicity in this life, it does not fit the ideal of a
scientific morality according to the first conception. In its content, furthermore, it matches the
provisional morality. Indeed, it is essentially the provisional morality placed on a secure
foundations, both metaphysical and scientific.
Morality conceived as an exact science would remove all uncertainty from practice. But we
learn from metaphysicsthe metaphysics of the substantial unionthat uncertainty in this

domain as to what is materially the best thing to do on each occasion can never be overcome.
The new or final morality, however, teaches us how to cope with and overcome this
uncertainty, or rather it teaches us how to assure maximum felicity in this life in the face of
such uncertainty. 10.
____________________
10.
Here I should mention an especially puzzling feature of Gueroult's interpretation. He
speaks of three separate conceptions of morality in Descartes's works. In fact, however, he
seems to be dealing with four conceptions of morality: (I) morality as an exact science, (2)
provisional morality, (3) morality as an exact science of the substantial union, a science
that eliminates practical uncertainty, and (4) that of a morality that teaches us how to gain
felicity in the midst of uncertainty. In his discussionand this is the puzzling featurehe
tends to run together the first and the third. The letter-preface of the Principles is the main
text which seems to express conception I. The passage from the Discourse in which
Descartes speaks of walking with assurance in this life suggests conception 3 (as does the
passage quoted in note 8 above from the Rules for the Direction of the Mind). On the
reading I prefer, there is no discontinuity between the second and fourth conceptions;
indeed, the fourth, as expressed in the correspondence and in the Passions, is simply a
more mature, better grounded form of the second. If Gueroult is indeed dealing with four
conceptions of morality, he can be seen as ar-92Since no science, neither mathematico-deductive nor psychophysical and empirical, can
reveal to us what is the true good in each case, there remains the science that furnishes a
general principle that is always within reach and that teaches us what we have to do in this
state of ignorance.... Virtue and the beatitude that accompanies itin brief, the supreme
goodno longer therefore reside in the possession of the true, from which good action would
invincibly result, but in the effort to attain it. ( Gueroult 1985, 2:206)
8. The final morality, therefore, is revealed as a kind of ethics of intention. We do not need to
know what is the best thing to do in each circumstance of this life in order to achieve
maximum contentment, we need only try to determine what is best. Gueroult puts thisvery
obscurelyas follows: "Since the supreme good does not consist for us in the possession of
the absolutely good in itself, but in the effort to attain it, the principle of the goodness of
action no longer resides in the excellence of the matter of the act, meaning in the thing it
realizes, but in the excellence of the intention that animates the will, meaning in the intention
to realize the best" (2:207). In settling for this third conception of morality, Gueroult says,
Descartes "seems to have changed its orientation radically." ii.
[Cartesian morality] was first conceived as a technique derived directly from the scientific
knowledge of the universe, and the value of an action was consequently placed wholly in its
matter, since the action is not good except in virtue of the excellence of its resultsthe
excellence of its results being guaranteed by the possession of what is true.... Virtue and good,
which until now were tightly linked to the truth of judgment, from which good action depends,
meaning the matter of the action determined by the knowledge of the true good, are now
dissociated from the nature of the matter and reported under the form of action. Whatever the
various occurrences, the possible degree of our error, our virtue remains whole once our
intention is good: the action may be good even though its matter is intrinsically bad and rests
on error. ( Gueroult 1985, 2:207-8) 12.

____________________
guing that since Descartes could realize neither the first conception nor the third, he finally
settled for the fourth as the final "substitute" for the first.
12.
Throughout, in my view, Gueroult seems never to take into account the distinction between
morality in the broad and narrow senses. When he speaks of morality as a technique, he
must mean morality in the broad sense. Here moral injunctions are hypothetical
imperativesassertoric imperatives. But the moral knowledge sought by the Cartesian
agent is not knowledge of such an imperative. The Cartesian agent strives to discover what
is the right
ii.
Later, in setting out three difficulties of Cartesian ethics, he writes: " Descartes is led to
reverse, almost completely, the fundamental principles of his morality.... A morality of the
form of the action is substituted for the technical truth of its matter, as a necessary and
sufficient condition of beatitude. In some respects the reversal is complete, since virtue is
compatible with error, and good action is compatible with a bad theoretical judgment"
(2:213).
-93We do not, of course, end up with a "morality of pure intention," because we still have to do
with morality that is conceived in terms of contentment and justified by its effectiveness as a
means to contentment. What has altered is this. Our contentment is seen to be a function not
of our doing the best thing in the circumstances but of our firm intention to do the best thing.
13.
Now, it is clear that this notion of virtue as a firm intention to do the best requires that we
keep the idea of the bestmaterially conceivedalive. "The value of the virtuous action
remains attached to the true by the intermediary of the intention that gives it its value, since
virtue is virtue only through permanent effort toward the highest approximation of the true"
(2:209).
9. Although we can accept much in Gueroult's account, we cannot assent to some of its central
theses. First and perhaps foremost, we should not accept the background assumption that, for
Descartes, morality is a mere technique and virtue a mere means. Second, we need to
highlight a vital distinction that this account leaves hidden, the distinction noted above
between morality in the broad and narrow senses. Insofar as morality is construed as a
technique, it is morality in the broad sense, where the agent's happiness is the end and
morality is made up of whatever precepts or dispositions are necessary to this end. But the
practice of virtue presupposes precepts of a radically different sort, namely, those of a firstorder morality
____________________
thing to do, where the notion of the right thing to do is a notion that figures within the
concept of morality in the narrow sense. In this context, "right" does not mean "most
optimific for me." It means, simply, right or best thing to do, objectively speaking. I
believe Descartes's view is that I will realize my own maximum contentment in trying to
discover what is the best thing to do and always have a firm resolution to do what I so
judge. But the judgment presupposes that the idea of the best thing to do is not identical to
the idea of what will bring my maximum contentment. Think of my example of the doctor
who devotes his life to promoting the well-being of those in underdeveloped countries at
great risk to his own health. The good he seeks is the good of others. Only on the condition
that their good is a genuine good or that he believes that it isdoes his conduct make
good sense. The contentment comes from his reflection that he is, to the best of his ability,

judging and acting well. This contentment would vanish with the thought that what he was
doing was maximizing his own contcntment and that the well-being of others was a mere
means.
13.
Gueroult reads Descartes as holding that our contentment is in no way compromised by our
not having knowledge of the best. Thus, on his reading, retrospectively discovered error
should not disturb our contentment. Later he identifies the fact that "moral good can
coexist with total theoretical error" as one of the inextricable difficulties of Cartesian ethics
(2:211-12). I quite agree that, in the absence of a well-founded theory of value, this would
be a serious difficulty for Cartesian ethics. But if I am right, Descartes can reply to this
criticism that we do know not only what sorts of things are objectively good but also how,
in a general way, these things can be ordered. While this knowledge is not sufficient to
preclude all theoretical error in our practical judgments, it may be sufficient to preclude
"total theoretical error" (whatever Gueroult means by this). I think it is reasonable to say
that our doctor who dedicates his life to alleviating human misery in foreign lands is not
totally off base. Indeed, if he were totally off base, then God would he a deceiver. For more
on these topics, see Chapter 8.
-94in the narrow sense. Descartes's virtuous agent is not, in deliberating, trying to discover what
course of action will maximize his own contentment. He is trying to decide, for example,
whether to publish his book, to help a friend in distress (with the possible risk of his own life),
to join the army and to fight for his sovereign (again with the risk of his life), to join Les
Mdecins sans Frontires and fight disease and poverty in foreign lands (once again with the
risk of life), and the like. These practical, everyday decisions, however, presuppose objective
values, or a belief that there are objective values in play among which one has to choose.
These are the values projected by or constitutive of first-order morality, values without which
the virtuous life would be entirely emptywithout value itself. Now, as I have said, it is
important to be aware of this distinction and of the role it plays in Descartes's account of
virtue. It is equally important to appreciate that the first-order morality that this account
presupposesunspecified and underdeveloped as it isis radically different from a set of
Kantian assertoric hypothetical imperatives. But such a set of imperatives is what Gueroult
has in mind when he speaks of a "morality-science" and claims that such a science was what
Descartes had initially hoped to achieve, a hope he later abandoned. I do not think that
Descartes ever had such a hope, for reasons I have already given. Indeed, many of these
reasons Gueroult himself has very well set out, but he draws from them the wrong conclusion.
Did Descartes in any basic way change his conception of what an ideal morality would be
after the Discourse? We have already seen that the morality he settled foror, at least, the
morality he was urging on Princess Elizabeth eight years laterwas essentially the same in
content as the earlier morale par provision, differing from it mainly in the security of its
foundations. Still, were we to think that Descartes initially conceived of the ideal morality as
a perfect scientifically based algorithm that would determine the objectively correct conduct
in every circumstance and guarantee that one could walk with assurance in this life, then it is
clear that Descartes never achieved such a morality and that what he settled for was
something not only less but altogether different. Fine points aside, it is Gueroult's view that
Descartes did have in mind such an ideal and did come to settle for less. The scientific project
was not advancing apace, the mind-body ingenium proved an obstacle to a rigorously
scientific medicine, and the theory of the passions, although scientific, could produce no such
algorithm. But if I am right, it was never Descartes's view that the path to true happiness was

through objectively right conduct; his view was always that the path went through moral
virtue. What leads me to this view, among other things already noted, is that we find in the
Passions, supported by the contemporaneous correspondence, a more thoroughly developed
form of what we met in the Discourse under the rubric une morale par provision.
-95CHAPTERSEVEN
The Moral Theory of the Passions
I. AS we should expect, the moral theory that appears in Descartes's Passions of the Soul
conforms to that of the contemporaneous correspondence and continues that of the Discourse.
The Passions develops and defines several familiar lines of argument. In its examination of
the essence and function of the passions we get for the first time an account of how we can
effect desirable changes in our passional structure. Nonetheless, the moral theory of this work
is not fully developed; it is not, therefore, the perfect morality projected in the letter-preface
to the Principles. Although it offers a much richer account of virtue than we have met so far
and culminates in a compelling analysis of gnrosit, it still lacks an explicitly formulated
first-order morality and at best only sketches a theory of value. i. I shall have more to
____________________
i.
According to Stephen Gaukroger's account of Descartes's instruction in moral philosophy
at La Fleche, little attention was given to these topics. He writes: "Aristotle's Nicomachean
Ethics, together with the detailed Coimbra commentary on his work, formed the basis for
the moral philosophy course, but there were in fact two different kinds of approach to
ethics pursued, and the Ratio studiorum was particularly explicit in its insistence on a
proper balance between the two. The first approach was that of speculative moral theology
of the sort that one finds in Aquinas' Summa Theologica. The second was practical
casuistry, which promoted a view of moral philosophy not in terms of adherence to general
and universal principles, but rather in terms of developing practical guidance for resolving
moral problems, guidance that may not be generalizable beyond the particular case. Such
an approach had its origins in the writings of the Stoics, and its first full presentation in
Cicero, and it fitted well with the kind of practical advice one might expect of a priest. The
focus of such teaching was 'cases of conscience' and there were firm strictures against
trying to resolve such cases from general principles. The cases themselves ranged from
private matters of personal conduct to very public disputes: an important treatise in the
latter category, Francisco Vitoria's Relecto de Indis et de Jure Belli (1539), for example,
looks at the question of the status and rights of the indigenous peoples of Central and
South America. There was no shortage of textbooks in the area of ca-96say about Cartesian first-order morality and the associated theory of the good in the final part
of this study.
2. Central to Descartes's theory of virtue is the Stoic distinction between what does and what
does not depend entirely on us, although in Descartes's version this is more precisely the
distinction between what does and what does not depend entirely on our own free will. This
distinction dominates Descartes's thinking about virtue and happiness, as we have seen. It is
connected, moreover, through his metaphysics, to another important distinction, that between

what happens by necessity (or by Providence) and what does not; indeed, these are shown to
be extensionally equivalent. Given these distinctions, it follows that everything we judge to be
good is either entirely in our power or not at all in our power. If someone should object that
there is a middle groundthings at least partially in our powerDescartes replies that things
that seem partially in our power are simply things in which there is a causal division of labor
between the part that is and the part that is not entirely in our powerthe part charged to
Providence. This leads to the most important question of whether our own happiness is
entirely in our power. Descartes answers that it is.
3. Descartes's instructions for attaining happiness are by now familiar; we need only
assiduously follow three rules: deliberate well, have a firm and constant resolution to carry
out what our reason recommends, and keep in mind that if we do so we shall always have all
the goods that are truly in our power. 2. It will help to secure allegiance to these rules if we
keep in mind a couple of basic truths of metaphysics: everything that is not entirely in our
power is directly sent by God, and the perfections of the mind are greater perfections than the
perfections of the body. 3. It will also help if we bring
____________________
suistry, both on specific issues and on the nature of casuistry itself, one of the most famous
texts in the latter genre being Toletus' Summma Casuum Conscientiae ( 1569), and it is
hard to believe that Descartes would not have been familiar with this, as the Jesuits made
casuistry very much their own.... We have every reason to think, given the Jesuit focus on
casuistry, that the young Descartes would have thought of morality above all in terms of
casuistry, and a training in the subject would have served him well in the kind of political,
administrative, or legal career that the Jesuits envisaged their students entering"
( Gaukroger 1995, 61 ). In this speculationthat the young Descartes thought of morality
above all as casuistryI suspect Gaukroger goes too far. Evidently reflecting on his
instruction at La Flche, the young Descartes does not speak of casuistry but complains of
the ancient moralists'the Stoics' taxonomy of the virtues and more emphatically of
their lacking good theoretical foundations. For an account of the curriculum in moral
philosophy that seems to fit better Descartes's discussion in the Discourse, see Rodis-Lewis
1995, 33 - 37.
2.
To be sure, that we can follow these rules at all presupposes that we have a first-order
morality in which we have reasonable confidence. I outline what I think such a morality
would contain in the final chapter.
3.
Alqui, among others, has complained that while Descartes makes use of the notion of
perfection, it remains very obscure. I will make some efforts to clarify the concept in Part
-97our passions under the control of our will. How we can manage this is one useful product of
the theory of the passions.
Although Descartes's account of the passions presupposes his dualistic metaphysics, it is not a
direct derivative from it. It rests on a further truth that is found directly in our own experience
of the quasi-substantial union of our own mind and body. It is the truth that we are not lodged
in our body as a pilot is in a ship, the truth vouchsafed to us, not through clear and distinct
ideas, but through two sorts of confused ideas, bodily sensations and passions. Thus it is
through bodily sensations and passions that we know ourselves to be a true union of mind and
body. And through Descartes's theory of the passions we finally meet the true subject, the

moral agent of Cartesian ethics. This is the self, whose virtue and happiness have all along
been the topic of our discussion. 4.
One final preliminary remark before turning to the theory of the passions. The theory bears on
all three of the rules of morality, although most especially on the third rule. It bears on the
first rule in showing us that because our passions tend to exaggerate the value of what they
represent to us as good, we must correct for this deceptive appearance if we are to judge well.
It bears on the second rule in showing us how we can best cope with obstacles to steadily
adhering to what we judge to be best. It bears on the third rule in showing us how to become
so thoroughly masters of our desires that our happiness cannot be held hostage to events not
entirely within our power. In this sense, but without the least suggestion that we should retreat
from an active life in the pursuit of external goods, it discloses the secret of the ancients.
4. In his treatise on the passions, Descartes seeks to answer these questions: What are
passions? How are they defined and distinguished from other psychological states? Do they
have a distinctive function or use in
____________________
Three. But it certainly seems that in using this notion, Descartes is building on intuitions
formed in the tradition of the great chain of being, which speaks of degrees of reality and
corresponding degrees of perfection (or value).
4.
Amlie Rorty ( 1986) doubts that Descartes succeeds in formulating a unified account of
this moral agent. Assuming that the will wills what the understanding represents to us as
good for us (convenable), what conception of the good, she asks, does Descartes offer that
is specifically a good for us as individual mind-body substantial unions? It is clear enough
that health and bodily strength arc perfections of the body, and it is also clear that
knowledge is a perfection of the soul. But suppose I am faced with a choice between doing
something that will achieve a good for my body and something that, while it is likely to
harm my body, will achieve a good of the soul. Which of these is good for me as a mortal
moral agent? The problem is acute in those cases in which promoting the welfare of a
friendhighly approved by Descartesrequires that we risk our lives. This is certainly not
good for the body. What is required is a unified account of what is good for us, not an
account in which what is good for the body competes with what is good for the soul; for
the moral theory in question is a moral theory for us as embodied minds, and the happiness
in question is our happiness as mortal moral agents. I go some way toward responding to
Rorty's doubts in Part Three.
-98our psychological economy? Do they admit a useful taxonomy? Why is it desirable to have a
theory of the passions?
States of the soul, Descartes begins, that is, penses or modes of thought, are of two kinds,
actions of the soul and passions of the soul. Actions of the soul are its acts of volition, which
terminate either in the soul, for example, when we voluntarily call to mind the idea of God, or
in the body, for example, when we voluntarily raise our hand. Passions of the soul are states
considered as effects, which are divided into those produced by actions of the soul and those
caused by movements within the brain. The idea of God, for example, would be a passion of
the soul, as the effect of a prior volition. Passions caused by movements in the brain are again
divided into those we refer to external objects (sensory perceptions), those we refer to states

of our body (bodily sensations), and those we do not refer to anything physical at all; these
last are the passions in the specific sense of emotion-passions, and they are, as Descartes puts
it, felt "as being in the soul itself" (9:346; I:337). 5.
In addition to the above, nominal, definition of the passions, Descartes gives an instrumental
definition and a real definition. 6. The principal effect
____________________
5.
He sums up his first definition of emotion-passion in article 27: They are "perceptions or
sensations or excitations of the soul which are referred to it in particular and which are
caused, maintained, and strengthened by some movement of the spirits" (II:349; I:338-39).
This definition calls for some comment. Although it is clear, at least in a rough, intuitive
way, what is meant by our referring sensory perceptions to external objectsthose we
naturally and spontaneously take to be their causesand by our referring bodily sensations
to regions of our body, it is not similarly clear what is meant by an emotion-passion's being
felt in the soul itself. But if the meaning of this expression is not itself clear, it is clear
enough what Descartes's view is. In the case of sensory states, we spontaneously take these
to be caused in us by the action of external objects, and the objects in turn to have the
properties they appear to us to have. In the case of bodily sensations, we take these to be
caused by states of the body and to be located in the region of these bodily states. In the
case of the passions, we do not spontaneously identify something physical as their salient,
remote cause, nor do we spontaneously take an apparent feature of the cause to be a real
feature of it. Rather, we take the passion to be a real feature of the soul itself. Of course, it
is strictly speaking an error, on Descartes's view, to ascribe either a color or a pain to
something physical, although we are not wrong in judging that something physical is the
remote, salient cause of our perception or bodily sensation, as the case may be. In the case
of the passions (in the narrow sense), while we are correct in our natural judgment that they
are states of the soul, we may be inclined to think, falsely, that they are also produced by
the soul itself. We may be so inclined simply because there is nothing in their
phenomenology that suggests to us a physical origin.
As we shall see, Descartes later introduces a different but closely related psychological
state, an intellectual emotion. These bear a phenomenological resemblance to emotionpassions, but they are not produced or maintained by physiological states. They are, in
principle, emotions that a disembodied soul could experience. Such states will be found to
play a significant role in Descartes's final account of human happiness (see especially
articles 147, 148, and 212). They do so because through the mediation of the pineal gland
they set up passional resonances.
6.
Here I follow a suggestion of Jules Vuillemin ( 1988, 19 - 23 ).
-99of the passions, he writes in article 40, is to "dispose the soul to want the things for which
they prepare the body. Thus the feeling of fear moves the soul to want to flee, that of courage
to want to fight, and similarly with the others" (11:359; I:343). As to the use of the passions,
he writes in article 52 that it "consists in this alone: they dispose the soul to will the things
nature tells us are useful and to persist in this volition" (II:372; 1:349). According to their
instrumental definition, the passions dispose us to want certain things; and according to the
real definition, these things are (in general) true goods. Although in both definitions the
emphasis falls on the instrumental role the passions play in protecting the integrity and health

of the bodythe real definition adding that the effects of the passions are true goods 7. it
later emerges that they have an even more significant role in promoting true goods of the soul.
In this connection, the passions have not only instrumental value but value as an end.
To take a standard case, 8. a passion arises in us on the occasion of our perceiving some object.
What happens when we see an animal coming toward us, for example, is that light is reflected
from its body, which in turn sets up motions in the eyes, then in the optic nerves, then in the
brain, registering finally as a set of motions in the pineal gland, these motions constituting the
image of the moving animal. If the image bears a close resemblance to images of animals that
have been harmful to us in the past, it excites the passion of apprehension and ( in some cases)
the passion of fear. These same motions in the gland that produce apprehension also propel
animal spirits both into the nerves that produce the bodily motion of running away and into
the heart, which in turn propels animal spirits back to the brain, subsequently moving the
gland in a manner that sustains the initial apprehension and fear. 9.
Just as the body has the power to affect states of the soul, so the soul has power to affect
states of the body, in both cases through motions of the pineal gland. Although I cannot
directly move the pineal gland by willing to do so, I can, merely by willing to run, make it
move in the way required
____________________
7.
The "real" definition presupposes Descartes's view that God has benevolently instituted
basic connections between movements in the pineal gland and our passions, in such a way
that the volitions these movements and their associated passions support are, in general,
beneficial, at least to the body. If Descartes held a simple preference theory of value, he
could infer directly from the instrumental definition of the passions that their effects were
(in general) good. Since, however, these connections are interpreted as expressions of
divine benevolence, the volitional effects they project are correspondingly to be interpreted
as having more than subjective value.
8.
There are nonstandard cases, as in dreams, reveries, dramatic presentations, and object
less passions. Passions also arise in connection with bodily sensations. The standard cases
are used later for taxonomical purposes.
9.
This summarizes the account Descartes gives in articles 35 and 36 (II:355-57; I:341-42).
-100for running (II:360; I:343). The effect of volition on this gland is part of my natural heritage;
some motions of my body I can naturally will, others I cannot. By habituation, however, I can
establish new connections between volition and movements of the gland. In learning to speak
a language, for example, I establish new connections between willing to say something
meaningful and appropriate motions of the mouth and tongue. Similarly, just as I cannot by
mere willing directly produce motions in the gland, I cannot by mere willing produce or
inhibit my passions. But I can do so indirectly, by willing some thought that is usually joined
to some passion I would have or opposed to present passion I would not have. "Thus, in order
to excite boldness and displace fear in oneself, it is not sufficient to have the volition to do
soone must apply oneself to attend to reasons, objects, or precedents that convince [one]
that the peril is not great, that there is always more security in defense than in flight, that one
will have glory and joy from having conquered, whereas one can expect only regret and
shame from having fled, and similar things" (II:362-63; 1:345). Calling these thoughts to
mind, however, may have only momentary and limited effect if the present passion we wish

to oppose is strong, since such a passion is being produced by physiological mechanisms that
we cannot oppose directly and which have natural momentum. What we can do in such a case
is simply refuse to act in the way our present passion inclines us. "For example, if anger
makes the hand rise in order to strike, the will can ordinarily restrain it" (II:364; I:345). The
measure of our strength of will is precisely that of the strength of the passions we can oppose
in this way, that is, by stopping the bodily motions they prompt us to. To test our strength,
however, we should use our best weapons, namely, "firm and decisive judgments concerning
the knowledge of good and evil." If these give us strength to subdue present passion, unless
they themselves are based on good judgment, they will not render us less vulnerable to regret
and repentance than we would be ifwe were weak-willed (II:367-68; I:347). To gain complete
control over our passions, therefore, we need prior judgments concerning what is beneficial,
judgments at once firm, decisive, and well founded.
In closing the first part of the Passions Descartes reminds us of a different procedure for
bringing both our passions and our conduct into line with our considered judgments about
how to live. Noting again, as he had argued in article 44, that by habituation we can alter an
established connection between particular motions of the pineal gland and particular passions
of the soul, as occurs when we learn to speak a language, he concludes:
It is useful to know that although the movementsboth of the gland and of the spirits and
brainwhich represent certain objects to the soul
-101are naturally joined with those [movements] which excite certain passions in it, they can
nevertheless by habituation be separated from them and joined with other quite different
ones.... Even those who have the weakest souls could acquire a quite absolute dominion over
all their passions if one employed enough skill in training and guiding them. (II:369-70; I:348)
There are, then, two principal ways we can gain control of our passions through firm,
decisive, and sound judgments of value and through bringing our passional dispositions
themselves into line with considered judgments concerning good and evilso that we are
appropriately passionate only about those things that truly merit the passion in question. The
critical case is that of desire, since this is the passion that directly inclines the will. What
Descartes claims is that habitual connections between the passion of desire and only those
objects truly warranting it can be effected in each of us. His argument is that we should
ideally limit the passion of desire to a single object, namely, the cultivation of the disposition
always to judge as well as we can concerning what is appropriate to do and firmly to adhere to
such judgments. This object is, in a word, virtue.
5. At the beginning of the second part of the Passions Descartes defines the passions in terms
of their use, and it is only in terms of this definition that he is able to establish their taxonomy.
I note that objects which move the senses do not excite different passions in us in proportion
to all of their diversities, but only in proportion to the different ways they can harm or profit
us or, generally, be important to us, and that the use of all the passions consists in this alone:
they dispose the soul to will the things nature tells us are useful and to persist in this volition,
just as the same agitation of spirits that usually causes them disposes the body to the
movements conducive to the execution of those things. This is why, in order to enumerate
them, one needs only to investigate, in order, in how many different ways that are important

to us our senses can be moved by their objects. I shall effect the enumeration of all the
principal passions here according to the order in which they may thus be found. (II:372; I:349)
Typically, the same motions of the gland that cause the passions also cause the muscles and
limbs to move in ways that preserve the integrity, strength, and health of the body. Thus we
may say, however obscurely, that our passions represent objects or bodily states qua
interesting, harmful, or beneficial to us. Moreover, they generally incline us freely to choose
to act in the manner in which our body is already disposed, that being in a manner beneficial
to the body. Descartes's principle of enumeration enables him
-102to identify six primitive passions: wonder, love, hatred, desire, joy, and sadness. As we shall
see, however, these passions can be aroused in us by our judgment of the interest, harm, or
benefit of objects, not with respect to our body but to ourselves as embodied souls. Thus, it
can happen that we will love something and form an appropriate desire inclining us to act in a
manner that, although good for our soul, is harmful to our body. For example, we might risk
our life for a friend or endanger our health by dedication to the search for knowledge. 10.
In subsequent articles he explains the use and potential for harm of each of the passions,
distinguishing their usefulness for the body from their usefulness for the soul. In general, they
are useful when they keep our attention focused on suitable ends and enhance our motivation
to promote them. As Descartes puts it, their use "consists only in their strengthening thoughts
which it is good that [the soul] preserve and which could otherwise easily be effaced by it,
and causing them to endure in the soul" (II:383; I:354).
The utility of wonder is that it stimulates discovery and retention in memory; those not given
to it tend to be ignorant (II:384; I:354-55). The main varieties of wonder are esteem and scorn,
generosity or pride, and humility or servility. "Esteem or Scorn is joined to Wonder according
as it is the greatness of an object or its smallness we are wondering at. And we can thus
esteem or scorn our own selves, whence come the passions and then the dispositions of
Magnanimity or Pride, and Humility or Servility" (II:373-74; I:350). ii. The other principal
passions are enumerated on the basis of their object's usefulness or harmfulness to us.
Beginning with love, Descartes observes that an object of love may be represented to us as
suit
____________________
10.
I mentioned Descartes's originality. All three definitions of the passions are original with
him, and so is this particular ordering and enumeration of the principal passions, especially
the inclusion of wonder (l'admiration). On the other hand, in many respects Descartes's
theory of the passions owes much to his predecessors. For an excellent summary account
of this, see Rodis-Lewis 1988, 21 - 32.
ii.
Themes only hinted at here are more fully set out in articles 149-61 in the third part. In
Descartes's moral theory, true generosity is identical with moral virtue. It is notable that in
the headnote Descartes uses the term "generosity" and in the text "magnanimity."
Descartes discusses his reason for using "generosity" in place of "magnanimity" in article
161. This suggests that the headnote was added to an earlier draft and that Descartes failed
to make the corresponding correction in the text. For a full discussion of this matter, see
Rodis-Lewis 1987. It is very clear, in any case, that the passion of generosity presupposes a
disinterested judgment of worth. In the later articles I have just referred to, Descartes

makes the distinction between generosity and pride on the basis of the correctness of the
value judgment. One who is proud is simply mistaken about his or her own value. The only
true basis of self-esteem, Descartes argues, is the awareness that one is using one's free will
well. Well-founded self esteem is the basis of the passion of generosity; its principal use,
moreover, is its inclining the will to cultivate the virtue of generosity.
-103able or harmful. This language is ambiguous. On the one hand, it naturally suggests that this
passion is our response to an intellectual judgment of the value, positive or negative, of some
object. This is the most natural reading of articles 56 and 61, as well as of article 79, where he
explicitly distinguishes three thingsan intellectual judgment, an intellectual emotion of love,
and the passion of loveand suggests again that judgment is a causal condition of the passion.
This would also fit the account of the passion of love he offers in his famous letter on love to
Chanut ( February I, 1647). Nevertheless, his theory requires that love and the other passions
can arise independently of any antecedent intellectual judgment of value. And in fact we find
him saying in several places, for example, in articles 91-94, that an impression of the brain
can represent something as good or bad, as novel, or as one's own. When an impression of the
brain represents something to us as suitable, we have only an obscure idea of the value of the
object. 12.
Love, according to Descartes's definition, is the passion that incites us to join ourselves in
volition (de volont) to objects that appear suitable to us, so as to form a whole of which we
are one part. The passion of love is to be distinguished from intellectual love, that is, from
judgments inclining the will in the same way, and from intellectual emotions excited by these
judgments (II:387; I:356). Several passions "participate" in love: "the passion an ambitious
person has for glory, an avaricious person for money, a drunkard for wine, a brutish man for a
woman he wants to violate, a man of honor for his friend or his mistress, and a good father for
his children." Only the last is love in a pure form, not mixed with desire for anything from the
child nor for any other mode of possession (II:388-89; I:357). Hatred is the opposite of love;
it incites us to separate ourselves in volition from an object represented to us as harmful.
Desire is the passion that disposes us to will for things represented to us as suitable. It is
useful so long as it is based on true knowledge, since then it cannot fail to move us in the
direction of a true good. 13. The passions that both presuppose and accompany desire are hope
and apprehension. We feel hopeful if we believe we are likely to get what we desire, and
apprehensive if this is unlikely. Beyond hope lies confidence, when we are certain of getting
what we desire; and beyond apprehension, despair, when we
____________________
12.
For brief discussions of Descartes's definition of love, see Alqui1989, 3:1013 n. 2 and
Voss 1989, 70 n. 30). Voss argues that relating the passion of love to a prior intellectual
judgment suffices for the enumeration of the principal passions but that, since this is not a
ubiquitous feature of all love it is not sufficient for the definition of love.
13.
This point, which will emerge as central to the moral theory of the Passions, will be
examined in some detail below.
-104-

are certain of not getting what we desire. 14. We may feel hope or apprehension concerning
things beyond our power, but when they seem within our power, we may remain apprehensive
about the best means or about our success. Here apprehension may take either the form of
irresolution, a passion inclining us not to act but to deliberate, 15. or the form of cowardice,
inclining us not to act at all for fear of failure, even when we have determined the means.
Hope, on the other hand, is a passion that inclines us to believe that what we desire is within
reach; it may be seen, therefore, as a mixture of desire and joy. When we face uncertainties of
the sort that inhibit or paralyze those who are apprehensive or cowardly, hope may take the
specific form of courage, inclining us to execute plans of action whose success we might
reasonably doubt.
In articles 177-210, Descartes analyzes passions that either are kinds of joy or sadness or have
joy or sadness as a main component. Of particular interest are remorse, envy, pity, selfsatisfaction, repentance, gratitude and ingratitude, indignation, anger, vainglory, and shame.
Remorse is a species of sadness and arises from doubt whether something we did was the best
that was in our power. It is useful insofar as it encourages us to do better in the future, but it
would be better were there no occasion for it. It is distinct from repentance, another form of
sadness, which presupposes that we have no doubt but that what we did was bad. Envy,
insofar as it is a passion, is sadness mixed with hatred; it arises when we see someone we
deem unworthy receive some benefit. So long as it arises from an offended sense of justice,
this passion is excusable, but that is not its most common form. Pity, on the other hand, is
sadness mixed with love or goodwill; it arises when we see someone receive some undeserved
harm. Those most given to envy and pity are those who tend to see the distribution of harms
and benefits as a matter of chance, not of providential necessity; those most susceptible to pity
in particular are those who, being apprehensive about their own good fortune, "represent the
misfortune of others to themselves as possibly happening to them" (II:469; I:395). In those, on
the other hand, who are not apprehensive about their own future, such sadness over
____________________
14.
About the same object, however, we may feel both hope and despair, since it may happen
that the object, viewed one way, seems very likely, but when viewed in another, very
unlikely, so that we may vacillate between these two passions.
15.
This passion has a use, but it also likely to be harmful. In article 170, Descartes writes:
"Irresolution is also a species of apprehension, which, keeping the soul balanced as it were
among many actions it is able to do, causes it to execute none of them, and thus to have
time for choosing before deciding. In this truly, it has some beneficial use. But when it
lasts longer than necessary and causes the time needed for acting to be spent deliberating, it
is extremely bad" (II:459; I:390; Voss 1989, 112 -13).
-105the misfortune of others might better be called compassion. 16. The term "self-satisfaction" can
refer to a disposition or to a passion. As a disposition, it is tranquillity or repose of conscience.
As a passion, it is the feeling that arises when we have done something we believe is good; it
is a species of joy, indeed, "the sweetest of all, because its cause depends only on ourselves"
(II:471; I:396). 17. Repentance is the opposite of self-satisfaction, arising from our belief that
what we have done is bad; it is the bitterest form of sadness, since it depends entirely on us.
Nonetheless, it has a good use, similar to the good use of remorse, since it may encourage us
to do better in the future.

Approval, Descartes says, is both a species of love and disinterested. This analysis seems on
the face of it inconsistent with his earlier account of love, which was said to be an interested
passion, its object being deemed good or suitable. "We are naturally inclined," Descartes
writes, "to love those who do things we consider good, even though no good may accrue to us
from them." I do not, however, see a deep difficulty here. In the first place, pure love, such as
the love of a father for his child, is in one clear sense disinterested, since the father asks
nothing from the child that he does not already have; he simply joins himself to the child in
volition to form a union. In approval, we can be seen to be inclined in the same way to join
ourselves in volition with the other person on account of her good (virtuous) will. In both
cases, the disposition to benevolence follows, although the disinterested love and the
benevolence are distinct from each other. The passion of gratitude is a mixture of approval of
another's goodwill and the desire to reciprocate the benefit we have received. Ingratitude is a
vice, but it is not a passion. Indignation is hatred toward those whose action expresses a bad
____________________
16.
Article 187 continues: "Now there is a difference present here: whereas the common
person has compassion for those who lament because he thinks the misfortunes they suffer
arc extremely grievous, the main object of the pity of the greatest men is the weakness of
those they see lamenting, because they consider that no possible accident could be so great
a misfortune as the cowardice of those unable to suffer it with steadfastness. And even
though they hate vices, they do not on that account hate those they see subject to them; for
these they have only pity" ( Voss 1989, 121 ). (Strictly speaking and in accordance with the
analysis of compassion given in this article, these occurrences of "pity" should be replaced
with "compassion.")
17.
The passage continues: "Nevertheless, when this cause is not just, that is, when the actions
from which one draws great satisfaction are not of great importance or are even unvirtuous,
it is ridiculous and serves only to produce a pride and an impertinent arrogance. This may
be observed in particular in those who, believing they arc Devout, are merely Bigoted and
superstitiousthat is, under cover of frequenting the Church, reciting plenty of prayers,
wearing their hair short, fasting, and giving alms, think they are entirely perfect, and
imagine that they are such friends of God that they can do nothing which would displease
him, and that everything their Passion dictates to them is righteous zeal, even though it
sometimes dictates to them the greatest crimes man can commit, such as betraying cities,
killing Princes, and exterminating whole peoples just because they do not accept their
opinions" ( Voss 1989, 121 -22).
-106will; it is the opposite of approval and might, like approval, be said to be disinterested. Anger
"contains everything indignation does," but it is directed against someone who has
intentionally harmed us; it is often mixed with the desire to avenge ourselves" (II:477; I:399).
Vainglory (la gloire) is joy founded on love of ourselves, which arises from our hope that
others will think us worthy of praise; shame is sadness founded on love of ourselves, which
arises from our apprehension that others will find us worthy of blame. Both these passions
have the same use; "they incite us to virtuethe one by hope and the other by apprehension"
(II:482-83; I:401). But Descartes adds:
We need to instruct our judgement concerning what is truly worthy of blame or praise, so that
we are not ashamed of acting well and not to be vain about our vices, as happens to others. On
the other hand, it is not good to rid ourselves entirely of these passions, in the manner of the

Cynics. For although the people [ le peuple] may judge badly, because we must live among
them and because it is important to us to be esteemed by them, we should often follow their
opinions rather than our own, at least concerning the external aspect of our actions. (II:483;
I:401)
6. A highly distinctive feature of Descartes's theory, one deriving from his radical separation
of mind and body, is that causal connections between psychological and physiological states
are all contingent and in many cases alterable, so that, for example, one could change from
desiring milk to having a strong aversion to it or, more generally, from having desires for
things not entirely in our power to having desires only for those things entirely within our
power. The desires in question here are, of course, passions. Suppose I judge (rightly) that
your health and freedom from suffering is a true good but only partly in my power. By
attending to the truth that the achievement of this good is in some measure dependent on
Providence, I can cease to desire this end with excessive passion, although I can will to do
whatever is within my power to promote it. 18.
Knowing how to realign these contingent connections between psychological states of the
soul and physiological states of the brain (pineal gland) is one thing; knowing what changes
to make is another. The former belongs strictly to the applied scientific theory of the passions;
the latter belongs to value theory. Descartes says that he writes of the passions as a physicist
(en
____________________
18.
The Stoics distinguish between ends that are good and valued ends that are indifferent.
Descartes draws a distinction that does some of the same work within his theory. He argues
that we can (and ought to) limit our desiresour desires as passionsto the desire for
virtue; yet we can have other valued ends that we seek to promote, the effort to promote
them being part of what virtue consists in.
-107physicien), by which he would seem to mean he is setting out what passions are and what
connection they have to human physiology. 19. But when he tells us what changes we ought to
make or what values we should strive to realize, he is in fact writing as a moralist.
Let us begin, then, with Descartes's view that we can take our own passions in hand and to a
significant degree shape our own emotional structureour dispositions to feel and to desire
in the way that we choose. 20. We also start with the idea that having some emotional structure
or other is a good thing, indeed, that being passionate is a good thing, so that there is no
question ab initio of trying to suppress our passions altogether. On Descartes's view, as we
have seen, there are six primitive passions: wonder (l'admiration), 21. desire, love, hatred, joy,
and sadness. Joy, 22. he says, speaking as a moralist, is the only passion that is good as an end,
and sadness the only passion that is bad as an end, relative to the soul. Of these, desire is the
only passion that leads directly to action. 23. Now if, in general, the passions are useful, not
every passion is useful in every circumstance, and, in particular, not every desire is useful;
indeed, we may have some desires that are in general harmful. A desire is useful only when it
leads us to acquire a
____________________
19.
Again, his central claim about the passions, which distinguishes his view from all previous

ones, is that they are passions precisely because their proximate causes are physical not
psychological.
20.
In her letter to Descartes of April 25, 1646, Princess Elizabeth comments on Descartes's
early draft of the Passions: "I find still less difficulty in understanding everything you say
about the passions than in practicing the remedies you order against their excesses. For
how is one to forsee all the accidents that can happen in lifeaccidents it is impossible to
enumerate? And how can we prevent ourselves from ardently desiring things that
necessarily tend to the conservation of man (like health and the means to live), which
nevertheless do not depend upon man's decision?" ( Blom 1978, 179). Descartes replies
with respect to these specific objects of desire: "I do not think that one can sin by excess in
desiring things necessary for life; it is only desires for evil or superfluous things that need
to be regulated. As for those that tend only to good, it seems to me that the stronger they
are, the better. To palliate my own faults I listed a certain irresolution as an excusable
passion, but nevertheless I esteem much more highly the diligence of those who are swift
and ardent in performing what they conceive to be their duty even when they do not expect
much profit from it" (4:411; 3:287).
21.
One original feature of Descartes's taxonomy of the passions is the inclusion of
l'admiration, which he takes to be an essential element of every passion. Wonder includes
the element of surprise, perhaps of being in some degree startled by some object, without at
the same time having any view about its being beneficial or harmful.
22.
Joy is a passion, not a bodily sensation. Typically we have joy when we possess what we
take to be good; in these cases our joy presupposes our value judgment. The full
contentment that constitutes human happiness arises from our judgment of the value of our
own virtue. In general, an action will promise joy as an outcome, then, only if it aims to
promote some good other than joy itself, say, the well-being of a friend.
23.
This is not to say that if we desire to do x, then we directly will to do x. It is simply to say
that between desire and volition there is no other passion.
-108true good; otherwise it is harmful. Since, by hypothesis, we can change our desires, the
question is, which should we keep and which should we alter?
We can sort desires according to the character of their objects. Some objects of desire are true
goods, whereas others are not. Some objects of desire are not in the least in our power to bring
about, whereas others are entirely within our power, and still others only partly in our power.
If we seek to have only beneficial desires, then we must first rid ourselves of desires for
objects that are not true goods and for objects that are not at all in our power, for a desire is
useful only if it inclines us to choose in a way that brings a true benefit. Where what we
choose to do can in no way effect the bringing about of something good, having a desire for
this object cannot be beneficial to us. On the other hand, if we desire a true good and this is
entirely within our power, then our desiring it is useful. Indeed, Descartes says that we can
hardly desire it with too much passion. What about desires for undisputed goods that are only
partly in our power? According to the analysis set out above, we should limit our desires only
to that which is entirely in our power and not desire with passion whatever part of the good is
in the hands of Providence. It should be remembered throughout that a desire (passion) is a
motivational enhancement only; we can will actions for the sake of ends, where we do not
have a desire (passion) directed to the end in question. 24. Indeed, not only can we will an end
without desiring it and yet desire the means, such willing is the condition of the ideally

virtuous person who promotes ends that have value and yet are not entirely within her control.
Let us now turn to the Cartesian texts in which this line of argument is set out.
If we consider joy, love, sadness, and hatred only as they relate to the soul (ignoring how they
may lead us into actions that are beneficial or harmful to our health), we can say that joy and
love are in themselves always good, and sadness and hatred always bad. 25. But as these relate
to desire, the matter is more complicated. Descartes begins article 144: "But because these
passions cannot lead us to perform any action except by means of the desire they produce, it is
this desire which we should take particular care to control; and here lies the chief utility of
morality" (II:436; I:379).
____________________
24.
This is evident from the definition given in article 52. The use of all the passions consists
in this: "they dispose the soul to will the things nature tells us are useful and to persist in
this volition." Nature indicates benefits to us in two main ways: first, through innate
dispositions to love and to hate and, second, through our intellectual powers to make value
judgments. The former are confined to bodily homeostasis, the latter range over all other
benefits, including that of the good use of our free will.
25.
I shall set out a Cartesian theory of value in Part Three, where I take up such value
judgments.
-109The utility of morality with respect to the passions lies in showing us what desires to acquire
and what to get rid of. It does this first, as we have just seen, by setting forth the criterion by
which we can sort desires into the beneficial and the harmful. The harmful ones are those that
promise joy and produce sadness, on the one hand, and preempt our cultivating more useful
desires, on the other. And here Descartes repeats his earlier claim.
As I have just said, desire is always good when it conforms to true knowledge; likewise it
cannot fail to be bad when based on some error. And it seems to me that the error we commit
most commonly in respect of desires is failure to distinguish adequately the things which
depend wholly on us from those which do not depend on us at all. Regarding those which
depend only on usthat is, on our free willour knowledge of their goodness ensures that we
cannot desire them with too much ardour, since the pursuit of virtue consists in doing the
good things that depend on us, and it is certain that we cannot have too ardent a desire for
virtue. Moreover, what we desire in this way cannot fail to have a happy outcome for us, since
it depends on us alone, and we always receive from it all the satisfaction we expected from it.
(II:436-37; I:379; my italics)
If we cannot desire virtue with too much ardor, we cannot desire external goods not entirely
within our power with too little ardor. Indeed, we should endeavor to suppress our desire for
such goods. If we fail to do so, we run the risk of disappointment; and if we preoccupy
ourselves with them, we run the additional risk of not attending to the good that is within our
power. One general remedy for vain desires is generosity (to be discussed later); the other is
reflection on Providence. Evidently, if we desire something that we can in no way help to
bring about, even if we believe we can do so, our desire is entirely useless, since the whole
utility of desire (as a passion) lies in its inclining us to pursue and attain some good, the
possessing of which will produce joy. 26.

____________________
26.
Article 145 continues in part: "we should reflect upon the fact that nothing can possibly
happen other than as Providence has determined from all eternity. Providence is, so to
speak, a fate or immutable necessity, which we must set against Fortune in order to expose
the latter as a chimera which arises solely from an error of our intellect. For we can desire
only what we consider in some way to be possible; and things which do not depend on us
can be considered possible only in so far as they are thought to depend on Fortunethat is
to say, in so far as we judge that they may happen and that similar things have happened at
other times. But this opinion is based solely on our not knowing all the causes which
contribute to each effect. For when a thing we considered to depend on Fortune does not
happen, this indicates that one of the causes necessary for its production was absent, and
consequently that it was absolutely impossible and that no similar thing has ever happened,
i.e., nothing for the production of which a similar cause was also absent. Had we not been
ignorant of this beforehand, we should never have considered it possible and consequently
we should never have desired it" (II:438; I:380).
-110Suppose I correctly judge that x is good but not in my power. Suppose, nonetheless, I desire x.
I can desire x only if I believe that x is possible. In this belief, of course, I may be mistaken. If
I knew x to be impossible, I would not reasonably desire it. In this case, a desire for x would
be entirely useless. There are two further cases. If my belief that x is possible is false, then
again, my desiring x is useless. For if it is not possible, my desire would be harmful both in
crowding out other more fruitful preoccupations and in leading to impatience and sadness. If,
on the other hand, x is possible, that is, if x occurs, still my desire was harmful in the former
respect and I get no additional joy that I would not have received simply from my happily
possessing x (in the desired way). Let us agree, then, that desires for outcomes entirely
beyond our control are vain and useless and that, therefore, reason counsels us to get rid of
them. But does the same hold for desires for outcomes that do, at least in part, appear to be
subject to our voluntary controlwhich embraces most of our desiresas Descartes
explicitly remarks in article 146? Here he tells us that "we must take care to pick out just what
depends only on us, so as to limit our desire to that alone" (II:439; I:380).
Again, it is a question only of the passion of desire, not of our pursuing ends we judge to be
good. So it is not a question of whether we should pursue possible outcomes in some measure
within our control, even those we know to depend also on the cooperation of other causal
factors of which we are ignorant. The question in this context has to do with the utility of the
passion of desire, where we believe that the outcome is only in some degree dependent on
what we do. This case, he argues, reduces to the one we have just examined, concerning
goods not at all in our power: "although we must consider their outcome to be wholly fated
and immutable, so as to prevent our desire from occupying itself with them, yet we must not
fail to consider the reasons which make them more or less predictable, so as to use these
reasons in governing our actions." Again it is presupposed that we may rationally pursue ends
not entirely in our control and that in so doing we ought to take the most likely route to
success. The argument is not, then, that we are not to have ends that depend only partly on us,
but that to the extent that they do not depend on us, we should not make these ends objects of
desire. Descartes gives this illustration.
Thus, for example, suppose we have business in some place to which we might travel by two
different routes, one usually much safer than the other. And suppose Providence decrees that

if we go by the route we regard as safer we shall not avoid being robbed, whereas we may
travel by the other route without any danger. Nevertheless, we should not be in
-111different as to which one we choose, or rely upon the immutable fatality of this decree.
Reason insists that we choose the route which is usually the safer, and our desire in this case
must be fulfilled when we have followed this route, whatever evil may befall us; for, since
any such evil was inevitable from our point of view, we had no reason to wish to be exempt
from it: we have reason only to do the best that our intellect was able to recognize, as I am
supposing that we did. And it is certain that when we apply ourselves to distinguish Fatality
from Fortune in this way, we easily acquire the habit of governing our desires so that their
fulfillment depends only on us, making it possible for them always to give us complete
satisfaction. (II:439-40; I:380-81)
Our desire in this case should be limited to willing the means, since the end is in large
measure in the hands of Providence, while our willing the means, as our reason directs, is
entirely in our control. We may will the end as a condition of willing the means, but we are
not to make the end an object of the passion of desire. And over this, Descartes argues, we
have voluntary control. To master our desire for ends not entirely in our power, it will help
simply to attend to the truth that such ends are in the hands of Providencethat is, to reflect
that, so far as we are concerned, the intended result does not occur because it is impossible.
Since we do not desire the impossible, this reflection should help us restrict our desires to
what is entirely within our power. Some suppose, mistakenly, that the results of action, so far
as they depend on factors other than our own will, depend on luck and that therefore they are
possible. But there is no such thing as luck. All that happens that is not entirely dependent on
us happens of necessity. 27.
To summarize, Descartes says that we commonly fail to distinguish desires whose objects are
entirely in our power and desires whose objects are not at all in our power, and he argues that
the latter desires are all vain and useless and that we should limit our desires to the former.
Now, we might agree that we cannot know of any desire that it is useful unless we can both
know the true value of its object and know that it is within our power to achieve this object.
But we may still ask why we should restrict our desires
____________________
27.
See the previous note. Descartes's example is underdescribed. A fully rational decision
would take into account the value of successfully conducting our business and the disvalue
of being robbed and their respective probabilities. In this example he points out that we
must weigh probabilities. In his letter to Elizabeth of January 1646, he emphasizes the
relative weighting of the values of the possible outcomes.
The question of what contingent ends we either may or ought to pursuecontingent in the
sense of being partly dependent on cooperating agenciesis a question to be dealt with
later, as is the further question of how our commitment to such ends is supposed to fit into
our emotional structure. At this point, we have been told only that this commitment is not
essentially to involve the passion of desire.
-112-

to those we know to be useful? After all, we might have a desire based on our belief that its
object has a certain value and that if we pursue a certain plan of action we will achieve it,
where this belief, although it does not amount to knowledge, is true. A desire so based would
be useful, especially if it were also true that we would not have pursued the end in question
had we lacked motivation enhanced by hopethat is, had we lacked this desire. If we restrict
our desires to those whose objects we know we can realize, we might miss out on possessing
true goods that, had we desired them, we would in fact have gained. In the theoretical context
Descartes is notoriously averse to risk. But from the Discourse onward he urges us to proceed
with confidence in the face of inevitable uncertainty, secure in the knowledge that if we have
judged as well as we can, any failure to attain our end was not our fault. Indeed, was not this
the message of the second maxim, to be decisive and resolute in our actions, where certain
knowledge is not to be had?
Certainly, Descartes is not to be seen as revoking the second maxim or counseling us against
proceeding with confidence where we cannot be certain of the outcome. He is trying to make
a very narrow point about our passions, arguing that we should restrict the passion of desire to
goods entirely in our power. All the same, one might object that this passion is often useful.
Without it we would be deprived of true goods we might otherwise attain, for example,
scientific knowledge and the good health and well being of others, since the persistent
effort required to attain them might well depend in some measure on our passionate
engagement, or so it would seem. To this objection, I am not certain Descartes has an
effective reply. The passion of desire, on his view, enhances motivation, even though it is not
necessary for motivation. It could, therefore, make the difference between perseverance and
giving up, where only with perseverance would some good outcome be realized. 28. Should it
happen that some of these risky desires prompt action that falls short of its goal, then that will
be the time for a good dose of reflection on Providential necessity. And it cannot be saidat
least Descartes could not saythat if this medication can be successful at all that it would fail
to come in the nick of time. Although Des____________________
28.
This line of thought is familiar from William James's colorful "Will to Believe," where in a
characteristic passage he writes: "How many women's hearts are vanquished by the mere
sanguine insistence of some man that they must love him! he will not consent to the
hypothesis that they cannot. The desire for a certain kind of truth here brings about that
special truth's existence; and so it is in innumerable cases of other sorts. Who gains
promotions, boons, appointments, but the man in whose life they are seen to play the part
of live hypotheses, who discounts them, sacrifices other things for their sake before they
have come, and takes risks for them in advance? His faith acts on the powers above him as
a claim, and creates its own verification" (James 1949, 104 ).
-113cartes has not, then, entirely made his case that we should desire only virtue with passion, he
has shown in theoretical terms that we should desire virtue with passion, and with a passion
proportional both to the value of its object and to the certainty of its being entirely in our
power. Accordingly, we might do well to follow Descartes up to the point of agreeing that
among desires that are useful to have, the desire for virtue is the most useful. This is the desire
to perfect our own will, so that we never fail to act well. In different contexts the object of this
desire is given different names: le bon sens, la bonne volont, la sagesse, virtue, and finally,
as we shall see, gnrosit. 29.

____________________
29.
Generosity, as Descartes conceives it, is a virtue, hut the term gnrositalso refers to a
particular passion, a specific form of admiration or self-esteem. But as an intrinsic good,
generosity is virtue itself, not the passionate response to the virtue.
-114Part 3
VALUE AND
GENEROSITY
IF we were to express the moral import of Descartes's theory of the passions in a single
maxim, it would read: Desire virtue and virtue only. This injunction applies only to the
passion of desire; so far as this particular passion is concerned, we are enjoined to limit it to
our acting as well as possible, that is, to our following reason to the limits of our finite
intellect, finite knowledge, and available time. The desire for virtue is always good and does
not admit of excess. This injunction does not imply that we are to limit our ends to this single
end, for obviously if we are to act at all, virtuously or otherwise, we require ends other than
virtue itself. A full moral theory would have to give an account of such ends. No such account,
at least no careful and systematic account, is set out in Descartes's works. In this significant
respect, then, his moral theory is incomplete. Although he gives a comparatively complete
account of virtue and happiness, he does not spell out the theory of value and appropriate
action that this account presupposes. My aim in Part Three is to supply such an account. The
exposition will naturally be more speculative than it has been up to this point; this is
especially true of the suggestions I make in the final chapter about what a Cartesian account
of duty might look like.
Now one might suppose, in view of Descartes's principle concern with virtue, that we should
read him simply as a virtue theorist. The idea would be, roughly, that we discover what ends
are worth promoting and what sorts of action are most appropriate by studying the paradigm
of the virtuous person and reading off from her characteristic attitudes and conduct what ends
are worth promoting and what constraints should regulate the
pursuit of such ends. Yet we cannot cast Descartes as a virtue theorist for the very reason that
has led us to search for a Cartesian first-order morality in the first place. Descartes's analysis
of virtue presupposes, and therefore cannot generate on its own, some theory of the right and
the good, without which it would be empty. For a related reason, furthermore, we cannot cast
Descartes as a precursor of Kantian formalism. On Kant's theory the prior ethical concept is
the good, to be sure, but in the special and important sense of the good that is both
unconditioned and the condition of the goodness of outcomes of action and of the rightness of
actions. It is true that the good will in Kant is the only subject of which this predicate
unconditioned goodnessholds, and it is true that, in Descartes's theory too, the good will is,
relative to us, the bonum supremum. Nevertheless, while Kant does not define the rational will
in terms of goods of the qualified kind, Descartes does define the good or virtuous will in such
terms. i. In constructing a Cartesian first-order morality, therefore, we must start with a
theory of the good.
____________________
i.
Descartes does, however, come close to Kant in his account of the attitude characteristic of
the virtuous or gnreux toward others; as he describes it, this is similar to what Kant calls

respect, a proper acknowledgment of and reverence for their free will.

-116CHAPTER EIGHT
A Cartesian Theory of Value
I. ALTHOUGH virtue is the supreme good, it is a second-order good; both its possibility and
its achievement are conditional on our believing that our ends are also true goods. Although
this belief might in fact be erroneous, it cannot be based on a fiction. There must be ends that
are truly worth choosing, even if our beliefs about them are mistaken. For example, our doctor
who volunteers to travel abroad to alleviate pain and to cure disease must suppose that his
goal merits his time, effort, and sacrifice of other goods. Even if, as a disciple of Descartes, he
should hold that virtue is a supreme good and the only good entirely in his control, he must at
the same time believe that the well-being of those he is trying to help is a good worth
pursuing for its own sake. To be sure, it might cynically be suggested from some perspective
external to the doctor's own moral outlook that he could still carry on with his work and retain
his thoughts about virtue even if the well-being of his patients had no more value, objectively,
than their suffering, so long as he believed otherwise. But whatever someone might make of
this suggestion, it is not one that the doctor himself could coherently entertain. If, as a disciple
of Descartes, he considers himself virtuous, he also believes he is resolutely trying to judge
well, and this is an activity he could not engage in on the supposition that all that matters is
that he sincerely reach some determinate decision or other about what goals to pursue. i.
____________________
i.
What I say here does not run against the grain of the second maxim of the morale par
provision. For example, although the man lost in the forest might choose which direction to
walk in entirely arbitrarily, he does so guided by the thought that his being out of the forest
and still alive is a genuine good.
-117Viewed from the perspective of Descartes's ethical theory, our doctor is not like the archer in
Stoic literature. The virtue of the archer is manifest entirely in his trying to hit the target,
where hitting the target is held to have no value in itself whatever. There might be doctors
modeled on the archer, doctors who seek difficult "targets" in order to exhibit their
extraordinary skills. Although these doctors might have the virtue of the craft of medicine,
they would not have the moral virtue of our doctor. If, therefore, we strive to be virtuous and
do what we judge to be good and worthy of our time and effort, we do so only because we
believe that our ends have real value. Moreover, we presuppose that there is somethingcall
it a practical fact of the matterthat is the objective standard of our judgment, the standard
we are striving to meet. Furthermore, we presuppose that we have a suitable level of cognitive
competence, such that when we engage in practical deliberation about what morality demands
of us, we are confident that our reflection will tend in the direction of meeting this standard.
In this way, then, Descartes's theory of virtue presupposes a theory of value. 2.
Descartes's account of virtue as the supreme good presupposes a theory of value in a further
way, for the claim that virtue is the supreme good presupposes some account of the good in
general, some account of goodness or (positive) value. There can, of course, be no doubt at all
that Descartes has a theory of value in the sense that he has settled views about the range of

things that have value and views about what judgments of the form "x is good" mean.
Following the fashion of moral theory of his time, however, he does not attend expressly to
metatheoretical issues such as the meaning of"good," the metaphysics of goodness, or the
epistemology of evaluative judgments. My project in this chapter is to indicate what I think
his views concerning these matters would have been.
2. In a letter to Chanut ( November I, 1646) Descartes says that were he to write a treatise on
moral philosophy, he would among other things undertake "to examine the right value of all
the things we can desire or fear" (4:536; 3:299). Among such things he lists health, joy,
knowledge, beautiful objects, friends, other human beings, 3. free will, 4. and God.
____________________
2.
The term "theory of value" may be taken in the narrow sense of a theory of the good,
desirable, or worthwhile or in a very broad sense as a general theory of all values, not only
of goodness but of rightness, beauty, truth, holiness, etc. Here I use the term in the narrow
sense. What I have in mind is some account of (a) what judgments of the form "x is good"
mean and (b) what grounds or reasons we have for judging of some x that it is good, better
than y, or bad or indifferent.
3.
See the relatively early letter to Mersenne, March 18, 1630, in which Descartes refers to
his Compendium Musicae. Here he tries to explain why sounds please us, and he seems to
hold that their pleasing us is what makes music beautiful and gives it value. In this letter,
which is brief, he appears to defend an intersubjectivist account of aesthetic value:
something is beautiful when it pleases most people (I: 132 -34; 3: 19 - 20 ).
4.
See especially Principles, sec. 37; Meditations, III; Passions, articles 152, 155-56.
-118Although this is scarcely an exhaustive list of Cartesian first-order values, it will go some
distance toward answering G. E. Moore's question, "What things are good (for their own
sakes)?" A full Cartesian theory of value, of course, would contain the complete enumeration
of such things. As to Moore's other question, "What does 'good' mean?" Descartes provides
no express answer. My conjecture is that he would incline toward Moore's own answer, that
the property of being good simpliciter is one that cannot yield to further analysis; for he
appears to treat such claims as that knowledge is good for its own sake, for example, as
synthetic judgments, and he treats the goodness in question as a simple nonempirical property.
5.
Thus, placed meta-ethically, in terms of the taxonomy of twentieth-century anglophone
moral philosophy, Descartes emerges as a cognitivist and nonnaturalist thinker.
3. My plan is take Descartes's list and set out Cartesian explanations of why they are true
goods, explanations designed to show in addition why we should be confident in our belief
that they are goods. I begin with the claim that health is a good. The Cartesian defense of this
claim, in barest outline, is that health is the outcome projected by the passions, that the
passions are in their nature good, having been instituted by a benevolent
____________________
5.
Descartes uses several value terms to refer to things that are desirable or good for their own
sake. Thus, virtue is the souverain bien. Two terms he frequently uses may be misleading,
at least, as they are often translated. These are bonne notre gard and convenable.
Although these are translated as "good for us" and "useful" or "agreeable," these
translations are misleading, since they suggest, not something good for its own sake, but

something good as a means. The French word convenable comes from the Latin
conveniens, which was Cicero's translation of the Greek homologia. Cicero uses the term
specifically in the context of the Stoic's view of man's chief good and what is desirable for
its own sake. "So, by the acquisition of knowledge and by reasoning, he reaches the
decision that man's chief good, praiseworthy and desirable for its own sake, rests here.
Since it does reside in what the Stoics term homologia (which, if you like, we shall call
'conformity' [convenientiam]), and in this then there is that good which is the standard of
reference for everything else" ( Cicero 1991, 3.21). For the Stoics, the standard of
goodnessthe ultimate good-making propertyis harmony or agreement with nature.
Making allowances for the differences in their respective conceptions of nature, it is not far
off the mark to say that this Stoic view is also Descartes's. To strive to use our free will
well is to live in harmony with our nature as rational agents.
There is a complication, however, in Descartes's account of value. He distinguishes
between what is good in itself or in an absolute sense and what is good for us. God, for
example, is good in the first way; virtue and knowledge, for example, are good in the
second way. This distinction is not one between two senses of "good"; Descartes predicates
the goodness of something in the second way when he considers the thing in question as
standing in a certain relation to us, namely, the relation of being in agreement with our
nature or as perfecting us in some way. Something is good for us if it is good and if it is
something we could in some way possess such that our doing so would be agreeable to our
nature. Again, this is not a distinction between types of things. God, for example, is both
good absolutely and good for us. God is good for us in this way: God is good absolutely
and, owing to our generic similarity to God in some respects, is for this reason a suitable
object of our love and, being a suitable object of our love, God is good for us.
-119deity, and that therefore any outcome they project must itself be considered good.
This line of argument has an important feature that is lacking in the Stoic argument that health
is to be preferred. The Stoic learns from nature what act-types are appropriate. Nature first
tutors us through our inclinations; we are inclined by our passions to act in health-preserving
ways, from which we can infer that nature intends for us to act in these ways, from which in
turn we can infer that so acting is appropriate (other things equal). Health, therefore, is to be
preferred. For the Stoic, however, it us not truly good. If we are strong and healthy, we are not
better than someone who is weak and sickly. Descartes follows the Stoic argument but carries
it to the conclusion that health is a true good. From the natural direction of our passions we
infer not just that our own health is to be preferred but also that, intended for us by a
benevolent God, it is a true good, the achievement of which makes our life better. Yet, again
following the Stoics, Descartes does not wish to say that we are better for being healthy. We
are good only as a function of our having free will and using it well. Nonetheless, our being
healthy is a true good for us, even if we are not better for being healthy.
In this example, we can see why metaphysics is essential to the theory of value and why
Descartes's theory is not subjectivist. Consider the ancient question, Is something good
because we prefer it, or do we prefer it because it is good? Descartes chooses the second,
although his answer is indirect and mediated by the metaphysical premise that God is
benevolent or, more specifically, nondeceiving. It is this that permits the inference from "I
prefer x" to "x is good for me (ceteris paribus)" in an objectivist sense of "good" (convenable).

The argument is not, it will be noted, that a benevolent God would not have created us to be
naturally inclined to act in ways harmful to our health. An argument of this sort would already
presuppose what it is designed to show, namely, that health is a true good. A benevolent God
wills for us what is good; he wills our health, therefore, if and only if it is a true good. The
premise that our inclinations are divinely ordained by a benevolent God serves a strictly
epistemic role in the defense of the claim that our health is good. 6. It is not just that God has
constituted us to act in health-preserving ways, however. Because our passions represent
items that conduce to our health as good for us, God would be deceiving us if these items and
the health they project were not true goods. It must be remembered that Cartesian passions
have propositional content; they rep____________________
6.
To be sure, God's benevolence is involved nonepistemically in providing us with passional
incentives to achieve what is good.
-120resent things as having value or disvalue, where the value in question is to be understood to be
objective value. It is owing to the benevolence of God that we can rely on our passion-based
judgments of value as being true judgments. Divine benevolence plays the same role with
respect to passion based value judgments as it does with respect to sensory-based
judgments about the existence of external objects. Since our passions tend to exaggerate the
value of things, however, we need to correct for this distortion in assessing their true value.
About sensible goods in general, by which I mean goods identified through our passions,
Descartes's view is this. Some of them we seem naturally drawn to; they are by nature objects
of love and of desire. Some we learn to love and to desire through experience. Indeed,
through the course of our lives the list of things we spontaneously love or spontaneously hate
tends to expand and to change. Nonetheless, according to Descartes, the general tendency of
our nature is such as to keep us from forming tastes for items of consumption that are harmful
to our health. If one should object that the common addictions to alcohol, tobacco, and drugs
are counterexamples, Descartes would reply that we have the means, namely, the good use of
our intelligence, to determine that these items are harmful and to avoid them. In the common
course of nature, after all, we do not find that animals take to things that are harmful to their
health and well-being, and the same is true of us humans at the level of our natural physiology.
If we are given to far more experimentation in the things we inhale or ingest, we are also
given the means to determine when our experiments fail. The rule for identifying true sensible
goods, therefore, at least those of the consumable sort, is to take the representations of our
passions as provisionally true as subject to correction and even to defeat in the light of our
experience.
4. So far I have sketched both the argument to support the claim that the health of our bodies
is a true good and the procedure we are to use to pick out, among things we are inclined to
consume, those that are harmful or beneficial. But many of these goods, in addition to
conducing to our health, are conducive to joy. Suppose I am hungry and I see an apple. The
desire to eat the apple may arise immediately, owing, no doubt, to a conditioning of the
connections between this passion and a constellation of physiological changes effected by my
seeing apples when I am hungry. Or the desire may arise more mediately, from my reflecting
on my past experience with apples and recalling how nourishing and how delicious they have
been. Still, if I am wise I will correct for the distorting effect of my desire for this apple and

judge that it would be good to eat, even if not quite so good as my present passion would have
me believe. I eat it and thereby conserve my health, which is itself a source of joy. I also
enjoy its taste.
-121While I can distinguish the taste sensation from my enjoying it, however, it cannot be said in
this case that my present joy follows on my believing that this taste is good. This is one of
those cases, far from rare, in which the joy, as Descartes puts it, is a confused mode of
perceiving goodness.
Let us review the role of the passions in this case, that is, the role they play in my judgment
that this apple is good. As we have seen, the effect of the passions is always to "move and
dispose the soul to want the things for which they prepare the body" (II:359; I:343). The
function of the passions is that "they dispose our soul to want the things which nature deems
useful for us, and to persist in this volition" (II:372; I:349). Mere knowledge that apples are
nourishing and delicious, unaided by the passion of desire, would leave us less likely to will
the eating of the apple. So the utility of the passion of desire, in this case, is that it enhances
our motivational tendency to take measures to acquire and to eat the apple. 7. The instrumental
utility of the passion of desire lies in its motivational force; it leads us to the passion of joy,
which is a noninstrumental value with respect to the soul and, since it is at once a sign and a
cause of good health, an instrumental value with respect to the body. Joy is not the object of
desire but rather the prize for attaining the object of a corrected desire. Joy is a thought, a
pense, a state of consciousness, a mode of a thinking substance. When it is not based on false
value judgment, joy cannot fail to be good; and in the special case of enjoyment of a
particular taste sensation, there is no room for value error and the joy cannot fail to be good.
8.

____________________
7.
In expounding Descartes's account of the role of the passion of desire in normal activity,
one repeatedly meets the problem I face here. On the one hand, as we have seen, Descartes
argues that we should so condition ourselves so that we desire only virtue (mostly by
deeply reflecting both on the role of Providence on determining outcomes not entirely in
our power and on our own nature as having free will). On the other hand, he refers to our
desires in other contexts as if these too were a good thing, since they tend to keep on the
track of true goods, as in the present case. Perhaps what we should say is that desire should
be proportionate to the value of the goods in question and that we should condition
ourselves further to react stoically to the frustration of desire, as the outcome requires.
Descartes's account of motivation, as we have seen, is complex, since he runs a series of
strictly intellectual emotions in parallel with the passions. So knowledge that the apple has
nourished us and that there are no reasons not to eat it might give rise to the intellectual
emotion of desire and hence to the volition to eat, without the passion of desire coming into
the picture at all. Yet here again, we meet the original problem, since in us, Descartes says,
any intellectual emotion tends to give rise to a corresponding passion.
8.
The passion of joy can have two causes. One is intellectual, when joy follows the judgment
of the goodness of some object and the intellectual judgment of possession. But joy may
arise independently of intellectual judgment, as it might in the case of my enjoying the
tasting of an apple. In this case, we might say, enjoying this taste just is the passional
knowledge that it is good. In article 139 Descartes says, as concerns the soul, love and
hatred result from knowledge and precede joy and sadness, "except when the latter stands

in place of the knowledge of which they are species." The knowledge that precedes is value
knowledge. In the special case of enjoying the taste of the apple, the joy just is the
knowledge of its goodncss.
-1225. According to the way I have set out the account, through our passions we come to know
that many sensible particulars are good. An apple is typical of that class of sensible particulars
of which the suitable possession consumptionpreserves bodily well-being. Suitably
possessing other sensible particulars, however, may have no direct effect and little if any
indirect effect on bodily well-being, although it may affect well-being more generally. Take
the case of seeing a beautiful flower. Here the relevant mode of possessing the valued object
may be merely seeing it. Moreover, the joy of seeing it may just be the judgment of its value.
On the Cartesian account, we should say that the flower is a good or that it has value, where
we mean not instrumental value but value as an end, in this case the value we call beauty.
This is again the kind of case in which our enjoyment is a confused idea of the value of the
object whose possession (in a suitable way) we enjoy. Just as our natural fear of, say, lions,
plays a quasi-cognitive role, according to Descartes's account, so does our enjoyment of a
flower. To put the view rather starkly, our natural or corrected enjoyment of beautiful objects
gives us epistemic access to the true value of those objects, just as the natural light gives us
epistemic access to the true nature of matter. In the case of enjoyment, however, our idea of
this value, although clear, is confused. 9.
Is the beauty of the flower an objective value? Well, clearly not, at least not in the same sense
as the flower's so-called primary qualitiesits shape and size, for example. Is it objective,
then, in the same sense as the flower's color? The Cartesian answer to this question seems also
to be no. Although Descartes hints at a theory of aesthetic response that remotely foreshadows
Kant's, he offers a fairly crude majoritarian analysis of the truth of aesthetic judgments.
Consider the following passage from an early letter to Mersenne, in which Descartes refers to
his Compendium Musicae:
You ask whether one can discover the essence of beauty. This is the same as your earlier
question, why one sound is more pleasant than another, except that the word "beauty" seems
to have a special relation to the sense of sight. But in general "beautiful" and "pleasant"
signify simply a relation between our judgment and an object; and because the judgments of
men differ so much from each other, neither beauty nor pleasantness can be said to have any
definite measure. I cannot give any better explanation than the one I gave long ago in my
treatise on music; I will quote it word for word, since I have the book before me.
"Among the objects of the senses, those most pleasing to the mind are neither those which are
easiest to perceive nor those which are hardest,
____________________
9.
I shall pick up this discussion again below, under the general rubric of the passion of love.
-123but those which are not so easy to perceive that they fail to satisfy fully the natural inclination
of the senses toward those objects nor yet so hard to perceive that they tire the senses."

To explain what I meant by "easy or difficult to perceive by the senses" I instanced the
divisions of a flower bed. If there are only one or two types of shape arranged in the same
pattern, they will be taken in more easily than if there are ten or twelve arranged in different
ways. But this does not mean that one design can be called absolutely more beautiful than
another; to some people's fancy one with three shapes will be the most beautiful, to others it
will be one with four or five and so on. The one that pleases most people can be called the
most beautiful without qualification; but which this is cannot be determined. (I: 132 -33; 3: 19
- 20 )
In the special case of beauty, therefore, Descartes begins to lose the distinction he seeks
generally to keep in place, between the joy of possessing a valued object and the value of that
object. In this letter, the beauty of the flower merges with the joy in the eye of the beholder.
Nonetheless, here as elsewhere, the joy itself is a true good.
To sum up to this point, we have identified some of ingredients of the good life, according to
Descartes: health, enjoyment, tasty food, and beautiful objects. What this discussion so far
may suggest is the following general thesis: whatever is the object of a corrected passion, of
love or desire, is a true good. I have already argued that Descartes is not a value subjectivist in
the standard sense. 10. All the same, we might model his explanation of the objective value of
sensible goods on that of recent informed-desire views that take the value of some good to
depend on a subject's desiring it. The difference is that in Descartes's case, an agent's passions,
corrected by relevant information, play a cognitive role in identifying values, having been
divinely instituted. ii.
6. I have claimed that our enjoying the possession of some good is a criterion or test of that
thing's being a true good. What of enjoyment itself? Is it also a true good? By enjoyment, I
mean the passion joy. This is not the same thing as the bodily sensation, pleasure
(chatouillement). Nor is it the same as intellectual joy, which, although an affection of the
soul, is not a
____________________
10.
His account of beauty is perhaps on the borderline.
ii.
The line of argument found in the sixth of the Meditations and developed in the Passions
of the Soul is just to this effect, that when we make appropriate corrections in the value
representations of our passions, we may then take the representations as veridical. This is
not straightforward value subjectivism, since the values are not constituted entirely from
the passions; they are rather lit up or revealed by the passions. Nonetheless, the more
sophisticated subjectivist views can be used as suitable models for this part of the Cartesian
theory of value.
-124passion. Joy (as a passion) can arise in two ways. It can derive from intellectual joy as a
resonating passion, in which case it presupposes a judgment by the intellect that I possess
some good. But as we have seen, there are important cases in which joy can arise without the
mediation of any such thought. Yet in whatever way it arises, it is itself a good. Indeed, "in
reality the soul receives no other fruit from all the goods it possesses, and while it is getting
no Joy from them it can be said that it does not enjoy them any more than if it did not possess
them at all" (II:396-97; I:360-61). I could conceivably be in a state of good health and
accordingly possesses a true good, but if I did not represent my good health as a good,

consciously or not, then I would not enjoy it; I would not feel the joy that is the fruit of good
health. My being healthy would nonetheless be a good.
Descartes gives the following account of the positive case. "So when one is in sound health
and the weather is more serene than usual, one feels a cheerfulness within oneself which does
not come from any function of the understanding, but only from impressions which the
movement of spirits forms in the brain; and one feels sad in the same way when the body is
indisposed, even though one may not know it is" (II:398: I:361). 12. The question at hand,
however, is this: What defense does Descartes offer that the passion of joy, which is the fruit
of our representing to ourselves that we possess some other true good, is itself a true good? So
far as I can see, he treats this claim as simply self-evident.
If this is right, we should not infer that, for him, health is in the final analysis a mere means.
Indeed, given his account of its origin, joy presupposes the possession of some other true
good, in the sense of something that is desirable for its own sake. The enjoyment of good
health, therefore, presupposes that health is a true good independently of whether or not it is
enjoyed. 13. Rather than health's being a mere means to joy, joy is one of nature's signs of
good health, which leads to our forming a desire for those things that conduce to maintaining
it. 14.
____________________
12.
Descartes writes in his letter to Chanut of November I, 1647, that in all likelihood joy was
the soul's very first passion, "because it is not credible that the soul was put into the body at
a time when the body was not in a good condition; and a good condition of the body
naturally gives us joy" (4:605; 3:307).
13.
Descartes makes a slight concession to hedonism, however, since he says that, other things
equal, joy based on a false opinion about one's possessing some good still has intrinsic
value. But other things are not equal, because joy gives rise to a desire to continue
possessing the item thought to be good; and since this item is not good, by hypothesis, this
desire and the joy that gave rise to it are harmful (II:435; I:378).
14.
"For the soul is immediately informed of things that harm the body only by the sensation it
has of pain, which produces in it first the passion of sadness, next hatred of what causes the
pain, and in the third place the desire to get rid of it. So also the soul is immediately
informed of things useful to the body only by some sort of titillation, which, exciting joy in
it,
-125Two things may be worth remarking in connection with joy (or contentment). The first is that
it is not the sole object, not even the typical object, of the passion of desire. For the most part,
the objects of desire or of love are goods other than contentment or joy. To be sure, joy
supervenes on our possessing such true goods in some suitable way, even in the cases in
which we discover the value of something only by means of our enjoying it, Descartes would
seem to hold that the object of love or of desire is this object itself, not the enjoyment of it.
Beauty is a case of this kind. 15.
The second thing worth remarking is that, on Descartes's view, joy or contentment can be
measured; it admits of degrees. 16. Its measure is the value of the goods on the possession of
which the joy supervenes. Thus the contentment that supervenes on our possessing goods of
the body is not so great as that which supervenes on our possessing goods of the soul. For

example, the satisfaction of making an important discovery would be greater than that of
eating a fine and nourishing meal (other things equal). The contentment that supervenes on
our achieving virtue is supreme in this respect. This contentment also has the advantage of
being entirely within our power. But here as elsewhere, the contentment presupposes that
what is possessed is a true good.
7. Let us now turn to love. According to Stephen Voss, articles 139-42 of the Passions make
it clear that, for Descartes, love has intrinsic value ( Voss 1989, 96 n. 68). I do not think this is
correct, although I agree that it is important to bring Descartes's account of love into a
discussion of his theory of value. But it is not love itself that has intrinsic value; it is the
several objects of love and the wholes that are constituted by love that have such value. Voss's
claim is based on the following passage from article 142:
____________________
next arouses love of what one believes to be its cause, and finally the desire to acquire
what can make one continue having this joy or enjoy one like it later on" (II:430; I:376).
15.
As I have pointed out from time to time, Descartes's official or standard viewfound
explicitly in the letter on love to Chanutis that joy supervenes on our belief that we
possess in some suitable way an object that is (independently) good. These objects, then,
are what love and desire target. Yet as we have also seen, for example, in the long letter in
which Descartes seeks to reconcile Zeno and Epicurus (see Chapter 4), Descartes seems to
flirt with the idea that without the thought of prospective pleasure our motivation to
promote the good on which pleasure supervenes would be diminished. His calling virtue
the supreme good and contentment the final end does not help to make his position very
clear.
16.
For example, in his letter to Elizabeth of September I, 1645, he writes: "According to the
rule of reason, each pleasure should be measured by the degree of perfection which
produces it; it is thus that we measure those whose causes are clearly known.... the true
office of reason is to examine the just value of all the goods whose acquisition seems to
depend in some way on our conduct in order that we never fail to devote our energies to
trying to procure those that are, in fact, the most desirable." We must concede, however,
that in the case where it is through joy that we learn of the true value of something we
possess, we have no way to determine that value independently of this joy.
-126it may be doubted whether or not love or joy are good when they are ... ill founded [that is,
founded on some false opinion]; and it seems to me that if one considers in abstraction
[prcisment] only what they are in themselves, with respect to the soul, it can be said that
even though joy is less solid and love less advantageous than when they have a better
foundation, they remain preferable to sadness and hatred likewise ill founded. So in life's
contingencies, in which we cannot avoid the risk of being deceived, we always do much
better to incline toward the passions that tend to the good than toward those that concern evil,
though this be only to avoid it. (II:435; I:378) 17.
While it is clear that, on Descartes's view, considered in itself and independently of its
instrumental value, joy has intrinsic value even when ill founded, I fail to see in this
passage any reason to interpret Descartes as claiming the same about love. I confess, however,
that I do not see clearly why unmerited love, considered in itself, is to be preferred to
unmerited hatred, but the reason seems to have to do with its greater utility. Descartes seems

to be saying that since we cannot always avoid error in our value judgments, we do better on
the whole if we are inclined to see the good in things than if we are inclined to see the evil. 18.
The passion of love Descartes defines as "a movement of the spirits, which impels the soul to
join itself willingly to objects that appear agreeable to it (convenable)" (II:387; I:356). This is
to be distinguished from judgments that also incline the soul to join itself de volontto objects
it judges to be good. Love is also to be distinguished from desire, although love typically
gives rise to desire when we love some future state of affairs or condition we regard as
possible to achieve through our action.
In principle, human love ranges over all objects that can in any way appear good to one. In
article 82 Descartes gives as examples of such objects: glory (of an ambitious man), money
(of an avaricious man), wine (of a drunkard), a woman (of a lustful and brutish man), a friend
or mistress (of a man of honor), children (of a good father). In the first four examples he notes
that what appears good is the possession of the object, not the ob____________________
17.
Voss's translation.
18.
In his letter on love to Chanut, Descartes writes: "Love, however immoderate, always has
the good for its object, and so it seems to me that it cannot corrupt our morals as much as
hatred, whose only object is evil. We see by experience that the best people, if they are
obliged to hate someone, become malicious by degrees; for even if their hatred is just, they
so often call to mind the evils they receive from their enemy, and the evils they wish him,
that they become gradually accustomed to malice. By contrast, those who abandon
themselves to love, even if their love is immoderate and frivolous, often become more
decent and virtuous than they would if they turned their mind to other thoughts" (4:614;
3:312).
-127ject itself. In each case, the agent in question has "desire mixed with other passions" for the
objects themselves. "On the other hand," he continues,
a good father's love for his children is so pure that he desires to have nothing from them, and
wills neither to possess them otherwise than he does nor to be joined to them more closely
than he already is; instead, considering them each as another himself, he seeks their good as
his own or with even greater solicitude, because, representing to himself that he and they
make up a whole of which he is not the best part, he often prefers their interest to his, and is
not afraid to lose himself in order to save them. The affection of people of honor have for
their friends is of this same nature, though it is rarely so perfect, and that which they have for
their mistress participates greatly in it, but also participates a little in the other. (II:389; I:357)
It is evident from the first four examples that not all the actual objects of human love are
worthy of it. That is to say, we often have love for objects that in fact lack the value that they
appear to have. Accordingly, the fact that, say, misers love money for its own sake and brutes
love sex for its own sake does not show that these objects have even prima facie value as ends,
such that in practical deliberation they ought to be given at least some weight. 19.
8. Many common objects of human love, however, do have genuine value, as Descartes
explains in article 83. As objects of the form of love he calls affection, he offers flowers, birds,

and horses; of the form he calls friendship, other people; and of the form he calls devotion,
God, our country, our prince, our city, and even a private person we esteem more highly than
we esteem ourselves.
Descartes makes it clear later that the value of flowers is their beauty; he does not comment
further on the value of animals, although we might reasonably say that in some cases their
value too is aesthetic. The objects of love and hatred, he says, are represented either by the
external senses or by what he refers to as the internal senses and our reason. "For we
commonly call something 'good' or 'evil' if our internal senses or our reason make us judge it
agreeable or contrary to our nature, but we call something 'beautiful' or 'ugly' if it is
represented as such by our external senses (chiefly by the sense of sight, of which we take
more notice than of all others" (II:391-92; I:358). Delight and abhorrence, then, are species of
love and hatred respectively. In the case of delight, what has value is, say, the visual
presentation
____________________
19.
In his letter to Elizabeth of September 15, 1645, he writes that the true object of love is
perfection and in that context (the object in question is God) he implies that the standard of
love is the objective perfection or value of the object. That is, love presupposes the value
of its object; it does not create that value.
-128of a flower or the auditory presentation of a piece of music. He remarks that these are usually
more vigorous than other species of love and hatred but also less inclined to truth, so that
"they are the most deceptive of all the passions, and the ones against which we must guard
ourselves most carefully" (II:392; I:358). This liability to being deceived about the true value
of the object of delight is dangerous, since delight leads naturally to desire and to action quite
contrary to our nature and reason. "Delight," he goes on to say in article 90, "is specially
ordained by nature to represent the enjoyment of that which attracts us as the greatest of all
the goods belonging to mankind, and so to make us have a burning desire for this enjoyment"
(II:395; I:360). Desires arising from delight vary considerably; "the beauty of flowers ... only
incites us to look at them," but the apparent beauty of a particular member of the opposite sex
can make "our soul to feel towards that one alone all the inclination which nature gives it to
pursue the good which it represents as the greatest we could possibly possess," a desire that
has a fair chance of being excessive.
9. Of particular interest in connection with delight is Descartes's claim (in the passage just
cited) that the pleasure or enjoyment supervenient on possessing a beautiful object in some
suitable manner can itself be the object of desire. This is as close to hedonism as Descartes
gets, but the comment is anomalous; it does not fit Descartes's standard pattern of analysis
according to which the object of desire is the valued good, as distinct from the pleasure or joy
supervenient on our possessing it. In the standard analysis, the object of love is the same as
the object of desire; in this case, love gives rise to a desire to possess the object in some
manner other than that of possessing it merely de volont. In the case of pure love (for
example, parental love) joy follows immediately, since the relevant mode of possession just is
that of love, no more, no less, and here it is as clear as anything could be that the object of
love is not the joy itself.

The final and decisive word on the question of whether joy or contentment is the supreme
good is to be found in the letter to Princess Elizabeth of October 6, 1645.
I have sometimes asked myself the following question: Is it better to be cheerful and content,
imagining the goods one possesses to be greater and more valuable than they are, and not
knowing or caring to consider those one lacks; or is it better to have more consideration and
knowledge, so as to know the just value of both, and thus grow sad? If I thought joy the
supreme good, I should not doubt that one should try to make oneself joyful at any price, and
I should approve the brutishness of those who drown their sorrows in wine, or assuage them
with tobacco. But I make a distinction between the supreme goodwhich consists in the
-129exercise of virtue, or what comes to the same, the possession of all those goods whose
acquisition depends on our free willand the satisfaction of mind which results from that
acquisition. Consequently, seeing that it is a greater perfection to know the truth than to be
ignorant of it, even when it is to our disadvantage, I must conclude that it is better to have less
cheer and more knowledge. (4:304-5; 3:268)
This letter may be usefully conjoined with Descartes's letter-preface to the French edition of
the Principles.
No soul, however base, is so strongly attached to the objects of the senses that it does not
sometimes turn aside and desire some other, greater good, even though it may often not know
what this good consists in. Those who are most favored by fortune and possess health, honor
and riches in abundance are no more exempt from this desire than anyone else. On the
contrary, I am convinced that it is just such people who long most ardently for another
gooda higher good than all those that they already possess. (9B: 4 : 1:180)
What I have called sensible goods bring joy when suitably possessed, but the joy they bring
may be false, since it may flow from a false estimate of their true value. In Descartes's view,
this would be so in the case of those who successfully pursue public acclaim and a luxurious
style of living as their principal aims in life, to the exclusion of friendship, family, community,
not to mention virtue and the love of God.
10. In a number of ways, indeed, God dominates Descartes's theory of value. We have already
seen how divine benevolence supports the epistemic reliability of our corrected passions as
indicators of value. But this is not the central theme of Descartes's famous letter to Chanut on
the love of God. In turning to this letter, we move to a different order of evaluation, a strictly
intellectual evaluation. Here Descartes explicitly distinguishes between intellectual or
rational love and love as a passion.
The first ... consists simply in the fact that when our soul perceives some present or absent
good, which it judges to be fitting for itself, it joined itself to it willingly [de volont], that is to
say, it considers itself and the good in question as forming two parts of a single whole. Then,
if on the one hand the good is presentthat is, if the soul possesses it, or is possessed by it, or
is joined to it not only by its will but also in fact and reality in the appropriate mannerin
that case, the movement of the will which accompanies the knowledge that this is good for it is
joy; if on the other hand the good is absent, then the movement of the will which accompanies
the knowledge that it would be a good thing to acquire it is

-130desire. All these movements of the will which constitute love, joy, sadness and desire, in so
far as they are rational thoughts and not passions, could exist in our soul even if it had no
body. For instance, if the soul perceived that there are many very fine things to be known
about nature, its will would be infallibly impelled to love the knowledge of those things, that
is, to consider it as belonging to itself. And if it was aware of having that knowledge, it would
have joy; if it observed that it lacked the knowledge, it would have sadness; and if it thought it
would be a good thing to acquire it, it would have desire. There is nothing in all these
movements of its will which would be obscure to it, or anything of which it could fail to be
perfectly aware, provided it reflected on its own thoughts. (4:601 2; 3:306)
Here is one place that Descartes unequivocally asserts that knowledge is a true good, or as he
here expresses it, a perfection of the soul, something worth having and worth seeking for its
own sake. This view of the value of knowledge, we may recall, was expressed with the same
confidence in the Discourse; and both there and in this letter to Chanut we have no reason to
doubt that he considers the claim that knowledge is good to be a clear, distinct, and certain
intuition, involving concepts that are entirely transparent to the thinking and reflecting
substance that is our soul. Descartes introduces this example, however, only to illustrate the
nature of love. Chanut's second question to Descartes was whether the natural light of our
reason instructs us also to love God, as it clearly does to love knowledge. Descartes answers
affirmatively:
In my view, the way to reach the love of God is to consider that he is a mind, or a thing that
thinks; and that our soul's nature resembles his sufficiently for us to believe that it is an
emanation of his supreme intelligence, "breath of divine spirit." Our knowledge seems to be
able to grow by degrees to infinity, and since God's knowledge is infinite, it is at the point
towards which ours strives; and if we considered nothing more than this, we might arrive at
the absurdity of wishing to be gods, and thus make the disastrous mistake of loving divinity
instead of loving God. But we must also take account of the infinity of his power, by which he
has created so many things of which we are only a tiny part; and of the extent of his
providence, which makes him see with a single thought all that has been, all that is, all that
will be and all that could be; and of the infallibility of his decrees, which are altogether
immutable even though they respect our free will. Finally, we must weight our smallness
against the greatness of the created universe, observing how all created things depend on God,
and regarding them in a manner proper to his omnipotence instead of enclosing them in a ball
as do the people who insist that
-131the world is finite. If a man meditates on these things and understands them properly, he is
filled with extreme joy. Far from being so injurious and ungrateful to God as to want to take
his place, he thinks that the knowledge with which God has honored him is enough by itself to
make his life worth while. Joining himself willingly [de volont] entirely to God, he loves him
so perfectly that he desires nothing at all except that his will should be done. Henceforth,
because he knows that nothing can befall him which God has not decreed, he no longer fears
death, pain or disgrace. He so loves this divine decree, deems it so just and so necessary, and
knows that he must be so completely subject to it that even when he expects it to bring death
or some other evil, he would not will to change it even if, per impossible, he could do so. He
does not shun evils and afflictions, because they come to him from divine providence; still less

does he eschew the permissible goods or pleasures he may enjoy in this life, since they too
come from God. He accepts them with joy, without any fear of evils, and his love makes him
perfectly happy. (4:608-9; 3:309-10)
In this letter we meet themes familiar from earlier correspondence with Elizabeth, themes
which forcibly indicate the crucial role metaphysics has in the development of Cartesian
ethics. It is not that Cartesian, first-order practical morality derives directly from
metaphysical premises or that we can somehow deduce moral precepts from our knowledge of
God. Rather, it is our knowledge of God that reveals God as a preeminently suitable object of
our love. And one direct bearing that the love of God has on Cartesian morality, aside from
the joy supervening on this love, is in helping us to adhere to the third rule of moralityto
restrict our desires to what is entirely in our powerby getting us to accept what otherwise
happens "as expressly sent to us by God," and even to "rejoice in our afflictions," to use the
language of the earlier letter to Elizabeth.
II. It may be in order to pause briefly and set out the taxonomy of value on which Descartes is
implicitly relying. Let us begin with a basic distinction, one we find, among other places, in
the letter Descartes addressed to Queen Christina, who had asked to have his views on the
supreme good. A thing, he says, can be judged to be good in two ways: without reference to
anything else and in relation to ourselves. God is the supreme good in the former sense "since
he is incomparably more perfect than any creature." God is the most perfect being, both ens
perfectissiumum and ens realissimum. When we estimate the value of a thing "without
reference to anything else," we are placing it on the great chain of ontological perfection. On
this scale of value God is the top, and we, having souls, stand well below God
-132but well above soulless matter. 20. Descartes often invokes this standard of perfection, for
example, when he says to Elizabeth that in our moral deliberations one of the most important
truths we are to keep in mind is that our soul is our most noble part and that its perfections,
knowledge, and virtue are, accordingly, higher than those of the body, such as health and
strength. So deference to our own value and the value of God is not in the least irrelevant to
the practical business of deciding how we should best live our lives. Nonetheless, the values
that figure more immediately in our practical deliberations are those considered in relation to
us. These are goods that in some suitable way we can possess, goods that Descartes otherwise
describes as suitable to our nature (convenable). These divide into three classes: those we
discover through the intellect, those we discover through the external senses, and those we
discover through the internal senses.
This basic distinction does not divide the goods themselves, however; God, for example, who
is supreme on the ontological scale, also figures among the items on this scale of goods that
are convenable. As we have just seen, God is a supremely suitable object of our love, and love
is the suitable manner in which we possess God, just as love is the suitable manner in which
we may possess friends, members of our family, and members of our society. That God is
good we know through our intellect. We know also in this way that knowledge is good and
that virtue is good; we know furthermore that free will itself is good, even the free will of
someone not disposed to judge as well as possible. Beauty is the value we discover through
our external sense, and through the internal senses we discover things conducive to our
physical well-being. Through the senses we also discover other goods that may have only a
marginal effect on our health but contribute importantly to our general sense of well-being. It

would be too dismissive to call them harmless pleasures, but they are entirely optional goods.
Certain activities fall into this classfor example, tennis or playing the piano. Some
amateurs, of course, become very serious about these activities and devote much of their time
trying to develop a high level of competence; to the degree that they succeed, they may realize
the further good of a sense of accomplishment. 21. And while the choice of one's life's work,
____________________
20.
Between us and God there might be multitudes of angels, writes Descartes in his letter to
Chanut of June 6, 1647. "When Holy Scripture speaks in many places of the innumerable
multitude of angels ... we regard the least of the angels as incomparably more perfect than
human beings" ( 5:56; 3:322).
21.
In article 95 of the Passions Descartes speaks of the pleasure we take simply in meeting
challenges and overcoming obstacles, even if the activity in question may not be profitable,
either to oneself or others, in any further way.
-133where one makes such a choice, may within a certain range also be optional, such activities as
constitute one's profession or craft may be goods known both through the internal senses and
ends known through the intellect. For reasons I will set out below, health and general wellbeing of others is a true good, not just relative to them but relative to us.
12. With this general scheme before us, I will try to place the following human goods:
friendship, community, and the state. Strictly speaking, friendship, as a form of love, is not
good for its own sake. In friendship, what is good for me is my friend and the joy that
supervenes on my simply being related de volontwith my friendsupervenes, that is, on my
making up a whole of which my friend is the other part. What makes only some people
friends? What is it about our friends that makes them appropriately different from other
persons? We do not, after all, love everyone as we love our friends. Descartes writes to
Chanut that we often love someone even before we "know their worth," by which he means
their moral character, and without knowing why. Here he must intend passional not
intellectual love; passional love is the love we discover through the joy of being related in a
single whole with the other, de volont. The explanation for our preference is sometimes
rooted in our idiosyncratic physiology.
when I was a child I loved a little girl of my own age who had a slight squint. The impression
made by sight in my brain when I looked at her cross-eyes became so closely connected to the
simultaneous impression which aroused in me the passion of love that for a long time
afterwards when I saw persons with a squint I felt a special inclination to love them simply
because they had that defect.... So, when we are inclined to love someone without knowing
the reason, we may believe that this is because he has some similarity to something in an
earlier object of our love, though we may not be able to identify it. Though it is more often a
perfection than a defect that attracts our love, yet since it can sometimes be a defect as in the
example I quoted, a wise man will not altogether yield to such a passion without having
considered the worth of the person to whom he thus feels drawn. But because we cannot love
equally all those in whom we observe equal worth, I think that our only obligation is to
esteem them equally; and since the chief good of life is friendship, we are right to prefer those
to whom we are joined by secret inclinations, provided we also see worth in them. Moreover,
when these secret inclinations are aroused by something in the mind and not by something in
the body, I think they should always be followed. (5:57- 58 ; 3:322-23)

This is one explanation of why we prefer our friends to others we equally respect. In general,
however, even our deeper personal attachments are nonetheless selective, rooted in the
complex contingencies of daily life; and
-134although they are in large part based on shared interests and shared experiences, they
nevertheless arise in ways we do not and need not understand. Of course, our friends are
human beings with free will and for this reason alone deserve our respect. But what makes
them specifically friendsthat is, what constitutes their special value for usis not a value
we can determine intellectually; what makes them friends is, although obscure, often
discovered through our passions. While respect due to all persons will determine a range of
duties to others as well as to ourselves, the special union that is friendship determines further
duties.
Descartes describes the bond of friendship in one place as sacred, 22. and in the letter just
quoted he says it is the best good in this life. He says this for two reasons. The first is the
general reason we have for respecting and therefore valuing persons above other kinds of
external goods. I shall return to this topic below. The secondand here I offer only a
conjecture is that the joy that supervenes on friendship (at least on perfect friendship, which
is mutual pure love) is directly experienced to be far more satisfying than that which
supervenes on the possession of any other external good. Giving as an example of pure love
that of a father for his child in article 82 of the Passions, Descartes points out that this love is
not mixed with a desire to possess the object of love in any other way than by that particular
kind of willing that creates a whole of which one is only a part, and often a lesser part at that.
23.
When Descartes speaks of friendship, he must be understood to intend a bond of mutual
loveat the limit, of mutual pure love. Love itself, of course, does not require love in return.
But such reciprocity would seem essential to friendship, and Descartes suggests as much in
article 83; there he also sorts the kinds of love according to the esteem we have for their
objects. But this manner of sorting presents us with a difficulty we need to deal with before we
can get clear about the distinctive value of friendship. The difficulty arises from the two
dimensions of value assessment in play, the dimension of love and that of esteem. 24. Our
____________________
22.
Princess Elizabeth had asked Descartes to give her his opinion of the views of Machiavelli.
To some commentators Descartes's comments have seemed disappointingly tolerant of
some of Machiavelli's recommendations to the prince; however this may be, Descartes
strongly denounces the abuse of friendship between princes, and not only on prudential
grounds: "But I rule out one type of deception which is so directly hostile to society that I
do not think it is ever permissible to use it, although our author approves it in several
places; and that is pretending to be a friend of those one wishes to destroy, in order to take
them by surprise. Friendship is too sacred a thing to be abused in this way; and someone
who has once feigned love for someone in order to betray him deserves to be disbelieved
and hated by those whom he afterwards genuinely wishes to love" (4:488; 3:293).
23.
See the passage from this article quoted in section 7 above.
24.
This distinction corresponds to the distinction I made earlier between the scale of
ontological perfection and the scale of things that are good relative to us.
-135-

esteem of objects is disinterested in the same sense in which, by contrast, our love of objects
is interested. Esteem places objects on the ontological scale of perfection and does not, in the
first instance, relate these objects to us, whereas love places objects on a scale of value in
which they do relate to us. How are we to combine these two kinds of assessment into a single
coherent scheme?
In his commentary on article 82, Ferdinand Alquipoints out that while it is easy to
understand how one can love some object in the sense of imagining oneself as a part of a
whole of which the object is another part, it is not easy to make sense of the idea that our
possessing the object could itself be that other part, as in the case of the brute who "loves" a
woman ( Alqui1989, 3:1015n). 25. If this case presents a problem for Descartes in his
attempt to extend his analysis this far, we may put it to one side, since it does not arise in the
case of friendship. Here we have to do, at least ideally, with pure love, that is, with love not
mixed with desire for possession of the object of love. Alqui's problem is that, on the one
hand, we regard our friends as our equals and, on the other hand, as meriting the sacrifice of
our own interests to theirs. Such sacrifice, says Descartes, would be absurd in the case of a
flower or an animal; it would, on the other hand, be justified in the case of God or an object
meriting our devotion. And the argument to this effect is just that we esteem the flower less
than we do ourselves and we esteem God more. But how, then, to justify on the basis of value
the sacrifice of our interests to those of a friend? Oddly, Descartes does not discuss this case
in this article 83; he omits it altogether.
Now, the main point Descartes seeks to make in this article is that through love we can
become members of three basic types of unions. The type will depend on the esteem we have
for the other member or members
____________________
25.
We might well doubt that this example can be fit into Descartes's definition of love in
precisely the manner he suggests. In the case of sheer lust, the other person hardly figures
as an object of love, certainly not according to Descartes's own definition; moreover, the
object of lust is in the final analysis sexual gratification in the sense of a pleasurable
release of pentup sexual appetite, in which the other figures as a mere means. Here love of
the woman drops out of the picture altogether, and we have to do rather more directly with
one's desire for one's own pleasure. It would seem, moreover, that we have to do not with
joy or contentment but with pleasurable sensation. We need to distinguish joy or
contentment from pleasure in the sense of pleasurable sensation; the opposite of the latter
is painful sensation, whereas the opposite of the former is sadness or discontent. If the case
of lust is to fit in the Cartesian analysis of love, the object of love must be the pleasurable
sensation itself, which then becomes the object of desire. Here again, even more directly
than in the case of beauty, Descartes would seem to agree that pleasurable experience can
be the object of desire, where the means to the satisfaction of desire are then valued as
mere means. If we ask what is the value of this satisfaction, the general answer must be
given in terms of the joy that supervenes on possessing this pleasurable sensation.
-136of the union, although it will remain true for any union that, as a natural effect of love, "we
feel benevolent towards itthat is, we also join to it willingly the things we believe agreeable
to it" (II:388; I:356). Whether one's interest in the whole will entail sacrificing one's own

individual interest when it conflicts with someone else's, however, will depend on one's
esteem for the other.
Strictly speaking, as we have seen, esteem is disinterested, whether it be esteem as an
intellectual judgment or esteem as a passion (which is a form of wonder). Love, on the other
hand, is interested, in the sense that it takes the object to be convenable. (Love, to be sure, is a
very special kind of interest, since it admits as particular instances the love of God, seen as
good, and the love of one's health, seen as good.) Now, the difficulty Alquifinds is that
friendship entails equality; we esteem our friends as we esteem ourselves. Esteem, of course,
is not love; esteem is one kind of assessment and love is another. Esteem places the loved
object on the ontological scale of perfection. But if esteem puts other persons on the same
level with us, presumably as beings with a free will, then although benevolence follows, it is
not clear why the disposition to sacrifice oneself for another will follow as Descartes says it
will in the ideal case of friendship.
In the letter to Chanut, Descartes writes that our love of God should be "incomparably the
greatest and most perfect of all." Alquicomplains that Descartes's view is not altogether
clear. The problem is this. First, Descartes has defined love in such a way that the love of a
bird and the love of God are instances of it. From this it follows that so long as love is based
on a correct judgment of value, the love in question will be perfect, and from this love will
follow a suitable form of benevolence. "However, within this rationalist conception of love,"
Alquipoints out, "we do not see why I ought necessarily prefer a friend to myself, since the
friend cannot be worth more than I (be more estimable than I), and why the love of God
would be more perfect, as such, than all other forms of love. These assertions seem indeed to
depend on another conception of love, one according to which the measure or perfection of
one's love is the degree to which one forgets one's own advantage. But this is a different
conception of love" ( Alqui1989, 3:720n).
Alquimakes a good point, but he is not precisely right when he says that Descartes is
working with two conceptions of love. Alquihimself does not clearly distinguish love from
esteem or appreciate that these are two distinct if overlapping kinds of valuation. Descartes,
to be sure, does not explicitly distinguish them either. Nonetheless, there do seem to be in
Descartes's theory these two kinds of valuation. Descartes speaks of both love and esteem.
Both are responsive to the value of some object, but they
-137assess the value differently. Moreover, esteem itself is represented as responsive to two scales
of value. It is responsive to values as fixed on the chain of being and to degrees of moral merit
as fixed on the scale of moral virtue. Love, on the other hand, fixes on a much broader range
of properties of objects, many obscurely picked out by our passions. Before I address Alqui's
objection, I need to point out the second difficulty, one I can formulate in the special case of
love of God. Love, as it is defined, is interested, in the sense that the judgment on which love
is based, or of which love is the expression, uses the predicate adjective convenable. Love as
a passion is correspondingly interested. The value to which esteem is responsive, on the other
hand, is disinterested.
Once we take note of this distinction between interested and disinterested evaluation, we may
be able to answer Alqui's objection. Love is one thing, esteem is another. What makes friends
of equal value is esteem, and what is esteemed is their possession of free will and their degree

of virtue. But esteem does not distinguish personal friends from other people, at least not from
other honorable people. Love, on the other hand, does so. Now, Descartes does not clearly
identify these values that we find in our friends and that distinguish them from others. But nor
does he say what it is about our children that makes them such eminently suitable candidates
for our pure love. He might be observing, however, that such love follows the course of nature
and that a child's property of simply being one's child is such a value, picked out by the
passion of love; and he might be claiming moreover that since the passion itself has a natural
strength and persistence, we may infer that it is a genuine good-making property, one that has
the endorsement of God. If this is right, then we might interpret him to be arguing
analogously in the case of friendship. When one honorable person finds herself attracted to
another on the model of pure love, then, other things being equal and the love being mutual,
we should say that the mutual value this expresses is a genuine value and, indeed, one that is
proportional to the strength of the mutual benevolence to which the mutual love gives rise. In
short, if in the case of such ideal friendship one is prepared to sacrifice one's own interest for
the other, we should take this motive to be objectively justified, that is, justified in terms of an
objective value, the value one finds in a true friend. The answer to Alqui's problem about
friendship, then, would be that although our esteem of all honorable persons is the same as is
the esteem we have of ourselves (supposing we are equally honorable), Descartes can easily
accommodate the special sacrifices we may make for a friend within his framework without
introducing a different kind of love. He can do so because the benevolence we have toward
our friends is based on love, not merely on esteem, and because this love in fact tends
-138toward benevolent self-sacrifice. What this solution brings out, however, is that esteem and
love are two distinct dimensions or forms of evaluation.
I would quickly summarize Descartes on the value of pure friendship this way: of external
goods, our friends are the highest. 26.
13. The fourth truth it is useful to know is that
though each of us is a person distinct from others, whose interests are accordingly in some
way different from those of the rest of the world, we ought still to think that none of us could
subsist alone and that each one of us is really one of the many parts of the universe, and more
particularly a part of the earth, the state, the society and the family to which we belong by our
domicile, our oath of allegiance and our birth. And the interests of the whole, of which each
of us is a part, must always be preferred to those of our own particular personwith measure,
of course, and discretion, because it would be wrong to expose ourselves to a great evil in
order to procure only a slight benefit to our kinsfolk or our country. (Indeed if someone were
worth more, by himself, than all his fellow citizens, he would have no reason to destroy
himself to save his city.) But if someone saw everything in relation to himself, he would not
hesitate to injure others greatly when he thought he could draw some slight advantage; and he
would have no true friendship, no fidelity, no virtue at all. (4:293; 3:266)
Here, in a letter to Princess Elizabeth written on September 15, 1645, Descartes expresses the
view that our state, our society, and our family are objects that merit our love and that they are
parts of the union formed by love that warrant our sacrifice, when that should be called for.
These views are more crisply and forcibly expressed in the letter on love to Chanut more than
a year later. Referring to the degrees of esteem that the objects of our love merit, he says that

even the perfect love of a flower or a building could not justify our sacrificing our lives for it;
this would be as preposterous as our sacrificing our whole body for the preservation of our
hair. "But when two human beings love each other, charity requires that each of the two
should value his friend above himself.... Similarly, when an individual is joined willingly to
his prince or his country, if his love is perfect he should regard himself as only a tiny part of
the whole which he and they constitute. He should be no more afraid to go to certain death for
their service
____________________
26.
In a letter of consolation written in 1637 to Constantijn Huygens, whose wife had recently
died, Descartes observes: "Although we cannot submit ourselves to God's law without
some pain, I value love so highly that I think that whatever we go through for the sake of it
is pleasantso much so that even those who are ready to die for the good of those they
love seem to me to be happy to their last breath" (I:632-33; 3: 54 ).
-139than one is afraid to draw a little blood from one's arm to improve the health of the rest of the
body" (4:612; 3:311).
Even if we agree with Descartes about both the value and the possible demands of friendship,
we might well find these claims about the value and demands of society and the state both
problematic and obscure. We can perhaps think of a small society made up of family and
friends as a special case of friendship, in which the other members are bound by the same
love to each other as we are to them. We can then make some sense of the value Descartes
finds in society and of the demands that membership in it makes on us, for in this conception
of society we meet no elements that were not present in simple friendship. To the extent that
the enjoyment of community is phenomenologically comparable to that of intimate friendship,
the value of community would be confirmed. But I do not think we can plausibly extend the
concept of friendship beyond a small community to a larger society, much less to a nation
state. And even if (following Aristotle) we speak of civic friendship, this is not the utopian
conception of all the members of a community being tied by reciprocal bonds of pure
friendship of the sort that tie two intimate friends.
In the letter on love to Chanut, however, Descartes does talk as if there were a hierarchy of
estimable goods, beginning with individuals and moving through community, the state, and
the prince to God as the most estimable; and he goes on to suggest, implausibly, that the
model of personal friendship can be used to interpret our membership in these different
unions. But except for the final union with God, at the one extreme, and possibly also one's
membership in a small community, this model is inapt. Although it is undeniable thatas
individuals, families, and small communitieswe are dependent on society and the state, as
Descartes says in this letter to Elizabeth, still our dependence cannot plausibly be understood
on the model of personal friendship. Our relationship to society or the state, as well as to the
individuals that constitute it, needs an altogether different and no doubt far more complex
analysis. This would be the stuff of moral political theory. The seeds of such a theoryof an
egalitarian form of such a theoryare visible in Descartes's important notion of gnrosit.
Yet, reluctant as he was to write on moral topics, Descartes was evidently even more reluctant
to develop the moral political implications of this focal concept in his theory of virtue. 27. In
any case, beyond the bare histori-

____________________
27.
There is much that could be said on this topic, but the following remark by Nannerl 0.
Keohane sums up the matter well: "Descartes humbly noted that politics is the business of
sovereigns, not private individuals; his sense of priorities for philosophy relegated politics
to a derivative status, in any case. Yet his ethics and his epistemology suggest the
possibility of a wonderful new community for human-kind, made up of rational individuals
devoting their energies to useful activities for the progress of the species, united willingly
in love and
-140cal fact that some people in Descartes's day came to love their society and devote themselves
to their prince in the way Descartes's letter to Chanut suggests, there is no good argument that
one's society or one's prince is more than an instrumental good, not a true good for its own
sake. However, in a sequence of letters following his letter of September 1645 to Elizabeth,
where he first suggested extending the model of friendship to the community and state,
Descartes backs away from this model and moves tentatively and implicitly in the direction of
Hobbes.
14. In a 1643 letter whose recipient is not known, Descartes makes the following comment on
Hobbes as a moral theorist:
All I can say about the book De Cive is that I believe its author to be the person who wrote the
Third Objections against my Meditations, and that I find him much more astute in moral
philosophy than in metaphysics or physics. Not that I could approve in any way his principles
or his maxims. They are extremely bad and quite dangerous in that he supposes all persons to
be wicked, or gives them cause to be so. His whole aim is to write in favor of the monarchy;
but one could do this more effectively and soundly by adopting maxims which are more
virtuous and solid. (4: 67 ; 3:230-31)
In letter to Princess Elizabeth a couple of years later, Descartes says that it is difficult to
determine how far reason demands that we devote ourselves to the community and that we
must rely on our conscience. He goes on to say, however, that even if we related everything
outside us to ourselvesconvinced that God has so established the order of thingsthen even
if we lacked charity we would find it in our interest to work for others as much as we could,
"and especially if [we] lived in an age in which morals were not corrupted." Here we may
detect a hint of Hobbes, although it is not evident that Descartes conceived the special
character of social cooperation as conventional in the way Hobbes did, where my duty to
cooperate is conditional on (a) my cooperation not going uncompensated and (b) my
reasonable belief that others are doing their part. But this may be the point of Descartes's
adding, "if we lived in an age in which morals were not corrupted." It does seem significant
that Descartes had read De Cive and saw in it that its author was a good moral philosopher.
But if Descartes had a glimmer of the distinction that Hobbes, Hume, and others made
between benevolence and justice, in this letter he is cer____________________
service to one another. It is impossible to be sure how much Descartes himself was aware
of these possibilities. His own scattered statements on politics are consistently
conservative, and he wanted to protect society against confusion or disorder from the

premature or ill-conceived application of his radically new methods" ( Keohane 1980,

203).
-141tainly not keeping the two clearly distinct. It is nobler and more glorious, he says, to do good
to others than to oneself, and it is the noblest souls that are so inclined. "Only weak and base
souls value themselves more than they ought, and are like small vessels that a few drops of
water can fill.... Base souls cannot be persuaded to take trouble for others unless you can
show them that they will reap some profit for themselves" (4:317; 3:273). This is his response
to Elizabeth, who was in turn responding to an earlier letter of Descartes's in which he had
emphasized our duty to the larger wholes on which we depend. She writes:
The consideration that we are part of a whole of which we ought to seek the advantage is
indeed the source of all generous actions; but I find many difficulties concerning the
conditions you prescribe for such actions. How measure the pains one takes in serving the
public against the good that comes from them, without these evils seeming greater in
proportion as their idea is the more distinct? [La considration que nous sommes une partie du
tout, dont nous devons chercher l'avantage, est bien la source de toutes les actions gnreuses;
mais je trouve beaucoup de difficults aux conditions que vous leur prescrivez. Comment
mesurer les maux qu'on se donne pour le public, contre le bien que en arrivera, sans qu'ils
nous paraissent plus grands, d'autant que leur ide est plus distincte?] And what rule shall we
have to compare things not equally known, such as our individual worth and the worth of
those with whom we live? One who is naturally arrogant will always tip the balance in his
own favor; and one who is modest will esteem himself less than his worth. ( Blom 1978, 155 )
The exchange between Descartes and Elizabeth would have profited from their clearly
distinguishing wider duties of benevolence from stricter duties defined, at least in part, in
terms of customary rules of some social practice. With respect to the former, Descartes's reply,
"Only weak and base souls value themselves more than they ought," is, although brief almost
to the point of rudeness, on the right track. On the other hand, if it is a matter of duty to do
one's part in a social practice with rules that are more or less agreed upon, then the question
might already be answered in terms of these rules. 28. Moreover, in the case of such a practice,
the worth of the other participants does not really come into view in practical deliberation,
unless it figures in the practice itself. ( It is not clear, of course, what specific kind of worth is
in question here.)
Nonetheless, all the social wholes of which Descartes speaks in his earlier letter, as I have
noted, seem to be modeled in his analysis of friendship. This
____________________
28.
Descartes does get around to this point in a later letter. See below.
-142is what allows him to speak of our being the lesser part of the whole and yetas a
consequence of our duty (or at least our right) of self-sacrifice for the sake of the otherthe
better part. Whatever one thinks of this accountAlquiraises a good point against itit
does not apply to wholes conceived along Hobbesian lines, where these are mutually
advantageous cooperative schemes that do not at all depend on the virtue of charity. The

confusion that permeates this exchange between Descartes and Elizabeth arises from neither
being clear about the kind of duty to others that is in question. Descartes's modeling of the
larger community and even the state in terms of personal friendship encourages this confusion.
If, as seems to be the case, Elizabeth is asking what are the burdens of social cooperation,
Descartes is missing the point when he replies that so long as we satisfy the demands of our
conscience, charity requires no more.
In De Cive Hobbes does not speak of charity or generosity, but he does say (the ninth law of
nature) that we are not to be useless (that is, troublesome); we are not to ask more than is
necessary but instead accommodate ourselves to the needs of others. The law of nature is that
each render himself useful to others. Since, in Hobbes's view, all these laws oblige in foro
interno, each of us is enjoined to become an accommodating person. This is far from being
Cartesian love in the form of friendship. But it is a move in the direction of benevolence and
distinct from strict justice. Justice requires only that one be trustworthy, on the condition that
others may reasonably be presumed to act in a trustworthy manner. One can be just without
being accommodating. Although Descartes says he respects Hobbes as a moral theorist, he
does not appear to have learned much from him, since he could well have brought some of the
important distinctions set out in De Cive to bear in his replies to Princess Elizabeth.
Moreover, distinctions may be made among types of friendships. At one extreme lie justice
and social convention; at the other, pure charity. In between are, in Aristotle's taxonomy,
character friendships, advantage friendships, and pleasure friendships. Descartes's answer to
Elizabeth hardly discriminates among the various contexts in which her question, How far
should we trouble ourselves in order to benefit others? might arise.
Not surprisingly, Elizabeth was not satisfied with Descartes's answer. On October 28, 1645,
she writes:
You do not believe one has need of an exact knowledge about the extent to which reason
orders us to interest ourselves in behalf of the public; for you think that, although each person
related everything to himself, he would still also work for others were he to use prudence. And
this prudence is the end-all of which I ask from you but a part. For in possessing
-143it one would not fail to do justice to others as to oneself, and want of such prudence is the
cause why a sincere mind sometimes loses the means of serving his fatherland by abandoning
himself too lightly for its interests, and similarly, why a timid person perishes along with it
instead of hazarding his well being and fortune for its conservation. ( Blom 1978, 167 )
Here she makes it clear that her question has to do with one's duty, not to friends but to the
larger public. In fact, she is following Descartes's lead from his earlier letter in which he
shifted from couching the issue in terms of friendship to phrasing it in more Hobbesian terms
of self-interest. Descartes replies:
If prudence were the mistress of events, I have no doubt that your Highness would accomplish
everything she wished to undertake; however, that would require all men to be perfectly wise,
so that, knowing what they ought to do, one could be assured what they would do. Or else it
would be required that one know in particular the humor of everyone with whom one has
some dealings; and even that would not suffice since, beyond their humor, they also have

their free decision, whose movements are known only to God. And because one ordinarily
judges concerning what others will do by reference to what one would wish to do were one in
their place, it often happens that ordinary and mediocre minds, being similar to those with
whom they deal, better penetrate their designs and are more easily successful in what they
undertake than are more elevated souls, who in dealing only with those very much inferior in
knowledge and prudence, judge very differently of things than do they. ( Blom 1978, 170-71)
Descartes here clearly has the idea of a social convention, where what one has reason to do is
conditional upon what one reasonably believes others will do, and where the question is
whether others will in fact do their part in a practice that defines what their part is.
Elizabeth responds: "As regards prudence, as far as human society is concerned, I do not
await an infallible rule; but I should be very pleased to see those rules you would give to one
who, in living only for himself, in whatever profession he has, would nevertheless not fail to
work for othersthat I should wish to see were I to dare ask of you more light after having so
badly employed that which you have already given to" ( Blom 1978, 172- 73). In his reply,
which foreshadows much of the following century's moral theory, Descartes points out the
considerable advantages of a deservedly good reputation.
One ordinarily sees it occur that those who are deemed obliging and prompt to please also
receive a quantity of good deeds from others, even
-144from people who have never been obliged to them; and these things they would not receive
did people believe them of another humor; and the pains they take to please other people are
not so great as the conveniences that the friendship of those who know them proves to be. For
others expect of us only those deeds we can render conveniently, nor do we expect more of
them; but it often happens that deeds that cost others little profit us very much, and can even
save our life. It is true that occasionally one wastes his toil in doing good and that, on the
other hand, occasionally one gains in doing evil; but that cannot change the rule of prudence
that relates only to things that happen most often. As for me, the maxim I have most often
observed in all the conduct of my life has been to follow only the grand path, and to believe
that the principle shrewdness is not to wish at all to use shrewdness. The common laws of
society, which all tend to do each other mutual good, or at least not evil, are, it seems to be, so
well established that whoever follows them frankly, without any dissimulation or artifice,
leads a life very much more assured than those who seek their utility by other ways. It is true
these other ways occasionally succeed for those who follow them, because of the ignorance of
other men and the favor of fortune; but it happens much more often that such people go
wrong and, thinking they are establishing themselves, they ruin themselves. ( Blom 1978, 176
-77) 29.
Under repeated questioning from Elizabeth, Descartes appears to have backed off from his
earlier analysis of the larger social union in terms of friendship and moved in the direction of
some sort of social convention or compact account of our duties to others. Within such an
account, one might make room for civic friendship, where virtue would consist in one's being
not only trustworthy but accommodating. But this virtue is a far cry from the ideal of pure
friendship and the virtue of charity.

What emerges from this discussion is that pure friendship is the highest of all external goods.
If Descartes suggests in one letter that our community and our state are even higher goods
(although modeled on personal friendship), not only does he not defend this view, he backs
down from it. This is not the place to explore Descartes's political theory or, rather, his
reasons for not wishing to explore the political implications of his moral theory, particularly
his view that all persons merit respect, since my present purpose is to set our his theory of
value. Here I conclude that one's community and one's state, aside from their considerable
instrumental value, are not such objects of love and joy as to be greater goods than one's
friends. 30.
____________________
29.
This is from Descartes's letter to Elizabeth of January 1646. This part of the letter is not
included in the Cambridge University Press edition of Descartes's correspondence
(Descartes 1991).
30.
As I read him, Descartes would have endorsed Montaigne's view of friendship: "There is
nothing to which Nature seems so much to have inclined us as to society. And Aristotle
says
-14515. The objective external goods that contribute to a happy life are health and strength of the
body, the experience of beautiful objects, satisfying activities and occupations, knowledge,
and above all intimate friends. Other goods that typically figure importantly in a happy life
are primarily instrumental goods such as wealth, security, and social stability. Depending on
the kind of community one lives in and on the kind of membership in the larger political
community, these wholes may also figure not only as instrumental values but also as external
goods for their own sake, that is, as suitable objects of love.
To round out this chapter, we are obliged to talk about God as an external good. In the long
passage cited above from Descartes's letter to Chanut on love, we saw that ifwe meditate
properly on the nature of Godhis omnipotence, omniscience, the infallibility of his
decreesand "weigh our smallness against the greatness of the created universe, observing
how all created things depend on God," and understand these things perfectly, we shall be
filled with extreme joy. The thought is that through perfect understanding of the nature of
God, we will join ourselves de volontto God and love God so perfectly that we will desire
"nothing at all except that his will should be done." We will neither "shun evils and
afflictions" nor "eschew permissible goods or pleasures" but accept afflictions without fear
and greet pleasures with joy, and our love of God will make us perfectly happy (4:609; 3:310).
It is obvious, he continues, "that our love for God should be, beyond all comparison, the
greatest and most perfect of all our loves." He admits that meditation and understanding of the
requisite kind is not easy, since it demands that we detach ourselves from "the traffic of the
senses"; such detachment, Descartes confesses, can be tiring and hard to sustain (4:612-13;
3:311).
That our love of God should be incomparably the greatest of our loves calls for comment in
the light of what I have already said about the value of friends. What I said was that among
external goods our friends are the greatest goods we can have. How is this to be reconciled
with that claim that, at least for those who properly understand the nature of God, the greatest
love is the love of God? Can God be our friend? The difficulty with this suggestion is that
God is not our equal, and friendship, Descartes has said, is a bond only among those who

merit the same degree of esteem. First, it must be said, God is a very special case. Certainly,
God is a proper object
____________________
that good law-givers had more respect to friendship than to justice. Now the supreme point
of its perfection is this. For, speaking generally, all those amities that are created and
nourished by pleasure or profit, public or private needs, are so much the less noble and
beautiful, and so much the less friendships, as they introduce some other cause and design
and fruit into friendship, than itself" ( Montaigne 1927, 184).
-146of veneration, devotion, and respect, as are, says Descartes, our prince and our state. Yet
while we can be said to love our prince and our state, we would not call either of these our
friends. Moreover, Descartes adds, it would be contrary to good manners to express our love
to a prince or princess. Why? "I think that the reason for this is that friendship between human
beings makes those in whom it is reciprocated in some way equal to each other, and so if,
while trying to make oneself loved by some great person, one said that one loved him, he
might think that one was treating him as an equal and so doing him wrong" (4:610: 3:310).
There is no impropriety, however, in expressing our love of God and no remote suggestion in
our so doing that we should be wronging God. Would it not still be presumptuous to suppose
that the reciprocal love that is the essence of true friendship could obtain between us and God?
While I would not insist that the answer to this question is clear, I think there is something in
the thought that we may infer from the nature of God, just as from mutual friendship, that we
may be assured of the full measure of the other's beneficence and goodwill that our own love
and esteem deserve. So it could remain true that among external goods our friends are the best
we can have in this life, and Descartes might be seen as allowing, as a very special case, that
God is not an exception but rather an exceptional friend. 31.
____________________
31.
In his letter on love to Chanut, while not exactly taking back his distinction between
friendship and devotion, he does appear to say that we can regard those to whom we are
devoted as a special kind of friend.
-147CHAPTER NINE
Morality as Generosity
I. SINCE Descartes's theory of the good is pluralistic, we are left with the task of ordering
goods effectively within an account of practical deliberation. We might approach this task
from the perspective of the virtuous moral agent, one who is disposed to judge as well as
possible and to adhere firmly and decisively to his best practical judgment. What we can say
in a very general way is that, so far as I am virtuous, I always aim at what I reasonably take to
be good. This follows from Descartes's (traditional) understanding of the object of the rational
will, namely, the good. i. And we may
____________________
i.
What we will is always chosen sub ratione boni. Descartes says that our will is free, but
there is some controversy about what his conception of free will precisely is. In a letter

taken by some to have been written to Mesland in February 1645, Descartes writes that
when an evident reason inclines us exclusively to one of two contrary actionsone that,
morally speaking, we cannot but choosewe can, absolutely speaking, nevertheless choose
the opposite (contrary to evident reason). Impressed by this letter, Alquiwrites: "It is
therefore possible, according to Descartes (and contrary to the opinion of almost all the
commentators) to reject evidence in the presence of this very evidence, to turn away from
the good even under the spell of its attractiveness. To the liberty adapted to the good and
to being is opposed the free man, and, if the degree of our liberty is determined, in the case
of fully enlightened liberty, by the ease of our choice, it is determined, in the case of pure
liberty, by putting into play the positive power which is ours to follow the worse although
we know the better" ( Alqui1987, 289). While Alqui's view merits serious consideration,
I am inclined to agree with Anthony Kenny's arguments against it. Moreover, for purposes
of grounding his claims about the value of our free will, it is not necessary for Descartes to
hold that our will is free in the sense of having the positive power to will the worse while
fully recognizing that it is the worse; what is necessary, rather, is that our choice of the
good not be mechanistically determined in the way in which the animal's actions are
determined. Thus it may be that we cannot but chose some good that is clearly and
evidently the best, just as we cannot but assent to some
-148also say that where I have several goods in view which are not jointly realizable, I will choose
the best of these. Up to this point, however, we have been given a list of goods, and although
these have been presented in a rough hierarchy, it is by no means evident how in particular
circumstances I am to determine what is best, all things considered. If we may say, in a
general way and abstracting from the context of practical deliberation, that some goods are
better than othersthat knowledge, for example, being a perfection of the soul, is better than
health, which is a perfection of the body we cannot infer that whenever I have a choice
between pursuing knowledge and pursuing health I should always choose knowledge, even to
the detriment of my health. Of course, in this case, so far as good health is an aid to the
successful pursuit of knowledge, I might find some reasonable plan of action that will give
adequate attention both to my bodily needs and to my search for knowledge. But what of the
other goodsbeauty, enjoyable activities, friendshow are these to be integrated or balanced?
Moreover, some goods that I might seek are not mine but yoursyour health or your
knowledge, for example. Goods, we need to recall, are objective goods. Accordingly, while
the value of my health gives me a reason to promote it, so does the value of your health give
me a reason to promote that as well. How do I weigh these reasons in my practical
deliberation? We have already seen that in Descartes's view the answer to this question will
depend in part on whether you are my friend, a member of my community, a member of my
state, or simply a fellow human being. It will also depend on whether my contribution is
crucial to your achieving the good in question. Perhaps someone else will provide it for you,
someone better able than I. Perhaps, moreover, it is a good that I can promote only on the
condition that others cooperate, in which case what is appropriate for me to do will depend on
how trustworthy these others are. When Elizabeth asked Descartes just how far she ought to
go in taking care of the needs and interests of others, she raised a question requiring a more
complex answer than the response she actually receivednamely, that she should do enough
to satisfy her consciencewhich, although true enough, is singularly unhelpful.
To summarize: Descartes asserts that virtue is the supreme good for us, 2. that there are other
goods, and that these, either as positive or as negative

____________________
clear and evident proposition; yet we can in both cases choose freely. Indeed, this would be
the highest form or degree of freedom, since it would be the fullest or most perfect
expression of our rational nature. See Kenny 1972; also Rodis-Lewis 1987.
2.
Descartes's most succinct expression of his argument to this effect is given in his letter to
Queen Christina ( November 20, 1647): "the supreme good of each individual... consists
only in a firm will to do well and the contentment which this produces. My reason for
saying this is that I can discover no other good which seems so great or so entirely within
each man's
-149ends, are the values we seek to promote through our actions. 3. In some circumstances, of
course, it may be clear what the best thing to do is, say, when my health requires that I take
daily medication. Perhaps other cases also will be fairly clear cut. For example, other things
equal, we are not to do or threaten bodily harm to others or to pursue goods through fraud or
deceit. These are not, of course, controversial claims, but to fill out Descartes's first order
morality we need to set out where they fit and how they are secured. The best way to do this, I
think, as well as to begin to develop a more general account of appropriate action and a more
specific account of duty, is to start with the culminating point of Descartes's developed theory
in the Passions, his conception of gnrosit.
2. What is gnrosit? 4. In one place Descartes says it is the justified good opinion we have
of ourselves (II:451; I:386). In another he writes that generosity is the self-esteem that results
from "the volition we feel within ourselves always to make good use of our free will" (II:449;
I:386). Elsewhere he says true generosity has two components, a feeling of a "firm and
constant resolution to use [our free will] well" and two items of knowledge,
____________________
power. For the goods of the body and of fortune do not depend absolutely upon us; and
those of the soul can all be reduced to two heads, the one being to know, and the other to
will, what is good. But knowledge is often beyond our powers; and so there remains only
our will, which is absolutely within our disposal. And I do not see that it is possible to
dispose it better than by a firm and constant resolution to carry out to the letter all the
things which one judges to be best, and to employ all the powers of one's mind in finding
out what these are. This by itself constitutes all the virtues; this alone really deserves praise
and glory; this alone, finally, produces the greatest and most solid contentment in life. So I
conclude that it is this which constitutes the supreme good" (5: 82 - 83 ; 3:324-25).
3.
One might refer to these as the material ends of action, distinguishing them from the single,
formal end, which is virtue. In this sense, even God could be a material end.
4.
In his note to article 153, Ferdinand Alquicomments that the term gnrositwas much
in vogue as expressing a certain moral ideal current at the time and was celebrated by the
contemporary dramatist Corneille. "Indeed, it is because he finds a notion quite commonly
used in his time, of which he wishes nonetheless to change the sense, that he speaks of true
generosity [vraie gnrosit]. But true generosity rests above all on true judgment. It
differs therefore from the glorious generosity of the Corneillian hero [la gnrosit
glorieuse de hros cornlien]. The text has the look ofa kind of lesson in morality suited to
the customs of the period" ( Alqui1989, 3: 1067n). Descartes first introduces the term
"generosity" in the title of article 54 of the Passions ( II: 373; I: 350), but in the text he
uses the term "magnanimity" to refer to the same passion. The latter term suggests a virtue

of the nobility, a virtue inaccessible to the common man. But Descartes's conception of
generosity is far more democratic or egalitarian, as we see in article 161, where he says
that by reflecting on our own free will and on the many advantages that we may expect
from a firm resolution to make good use of it, we may acquire, first, the passion, and then
the virtue of generosity. Genevive Rodis-Lewis suggests in a recent article that this more
egalitarian conception of the passion that Descartes calls "generosity" in article 161 differs
from the "magnanimity" of article 54; she suggests, furthermore, that Descartes's analysis
of generosity, both as a virtue and as a passion, in articles 153 206 was probably not
present in the first, unpublished version, which Descartes submitted to Princess Elizabeth
for comment in the spring of 1646. See Rodis-Lewis 1987, 43 - 54.
-150namely, that only our freedom to dispose our volitions belongs to us and that we are rightly
praised or blamed only for how we use our freedom. As I shall interpret Descartes, the term
"generosity," taken narrowly, refers to the firm and constant resolution to use our free will
well, that is, "never to lack the will to undertake and carry out what [we] judge to be best"
(II:446; I:384).
Taken in this narrow sense, generosity is a volitional disposition. It is a necessary but not
sufficient condition for legitimate self-esteem. Self esteem is an emotion or feeling arising
from a judgment of value of one's worth, and the necessary condition for an assessment of
one's worth is freedom of the will. Descartes assumes that if I am firmly and sincerely
committed always to choose in accordance with my best practical judgment and that,
moreover, I know that I merit praise only when this commitment itself is an expression or
product of my own freedom, then I shall have the emotion of self-satisfaction. 5. This is in
accord with his general claim that a suitable emotion of self-approbation must follow on our
awareness of having some perfection. Self-approbation or self-esteem is, here, an internal
emotion of the soul, an "emotion produced in the soul by the soul itself" (II:440; I:381). In
article 160 Descartes says that we could also use gnrositto refer to a passion, although he
points out that, strictly as a passion, generosity would arise from the same "movement of the
spirits" and involve the same mix of wonder, joy, and self-love as the passion he calls vanity.
The only difference, although it is crucial, between the passion of generosity and the passion
of vanity is that the former is justified and the latter not. But the so-called passion of
generosity is the passion that resonates with justified self-approbation. 6. In any event, it must
be clear that generosity as a virtue is a volitional disposition, an activity of free will; it is not
and could not be a passion of the soul.
Properly grounded self-esteem and the associated passion of generosity will be manifest in a
range of more specific conative and affective dispositions, notably in our dealing with and
attitude toward others. First, concerning their characteristic attitude toward others Descartes
says that the
____________________
5.
This satisfaction, as an internal emotion of the soul itself, he also calls tranquillity and
peace of mind (in articles 148 and 190). It is distinct from self-satisfaction as a passion that
arises when we perform an action we think good (article 190).
6.
The usefulness of this passion lies in its concentrating our attention on the development of
generosity as a virtue. "If one frequently occupies oneself in considering what free will is
and how great the advantages are that come from a firm resolution to use it well and also,

on the other hand, how vain and useless all the cares are that trouble the ambitious, one
may excite in oneself the passion and then acquire the virtue of generosity. And since this
is, as it were, the key to all the other virtues, and a general remedy for all the disorders of
the passions, it seems to me that this consideration is well worth noting" (II:454; 3:388;
Voss 1989, 109 ).
-151gnreux never have contempt for anyone, not even for those who exhibit some vice or who
do wrong (II:446; I:384). Why is this? He does not expressly spell out his answer, but it
would clearly be along the following lines. First, in the special case of a gnreux not only not
having contempt but having esteem for a fellow gnreux, what legitimizes self-esteem must
also legitimize esteem for others who have free will and exhibit a firm and constant resolution
to use it well. Judgments of moral worth can, after all, be universalized. The case of those
who are characteristically unjust, cruel, brutal, greedy, arrogant, deceitful, ungrateful,
bigoted, and boorish, on the other hand, would seem to be quite different. Since esteem is the
proper response only to a virtuous will, these certainly do not merit our esteem. But why do
they not merit our contempt? Descartes's answer, I believe, is that they do not merit contempt
because, vicious as they may be, they have free will and are therefore capable of becoming
gnreux themselves. Clearly, for this answer to stand, we must agree that simply in the fact
that we have free will we possess a good. As I noted in the previous chapter, the kind of
valuation that Descartes calls esteem is one that fixes, first, on ontological perfection and,
second, on something of a certain ontological kind developing its own perfection appropriate
to it. Because they possess intelligence and will, others merit our esteem as beings of a
certain kind, beings having the potential for a specific kind of development, both intellectual
and moral. The suggested line of argument is that since free will is not only a necessary
condition but also the only condition that anyone requires to achieve the supreme good, we
cannot hold anyone in contempt so long as he has this power. By "contempt," here, Descartes
must have in mind viewing others as we might view, say, animals. Even those who use their
freedom badly do not lose their potential for becoming virtuous. And freedom, as the potential
for the supreme good, Descartes seems to be arguing, always deserves a minimum of respect,
even if it does not merit the esteem of virtue achieved. 7.
3. What, then, is the characteristic attitude of the gnreux toward other persons? Although
Descartes does not use the term in this way, I think we should call this attitude that of respect.
8.
Respect looks to the value of others as rational agents with free will, and it is this freedom
that establishes a benchmark of equality among all human beings; it serves, therefore, as the
____________________
7.
We find one particularly striking passage in Descartes's letter to Queen Christina of
November 20, 1647, where he writes that "freewill is in itself the noblest thing we have
because it makes us in a certain manner equal to God and exempts us from being his
subjects" (5: 85 ; 3:326). And the heading article 37 of the Principles reads: "The supreme
perfection of man is that he acts freely or voluntarily, and it is this which makes him
deserve praise or blame" (8A:18; I:205).
8.
Descartes uses "respect" as a synonym for "veneration." See article 162 of the Passions.
-152-

valuational ground not only for not harming others but for seeking to promote their true good.
I shall return to these topics below.
Descartes next argues that in the gnreux a minimal respect for others is combined with true
(i.e., virtuous) humility. And the reason is that although we must, as gnreux, be sincerely
committed to judging and acting well, we cannot fail to be aware of "the infirmity of our
nature," and the "wrongs we may previously have done or are capable of doing," so that "we
do not prefer ourselves to anyone else and we think that since others have free will just as
much as we do, they may use it just as well as we use ours" (II:447; I:385).
To these attitudes of humility and respect is added a strong disposition to general beneficence
combined with weak self-love. The gnreux, he says, are "naturally led to do great deeds"
and "esteem nothing more highly than doing good to others" (II:447-48; I:385). At a
minimum, then, they are "courteous, gracious and obliging to everyone." Moreover, since
they have mastery over their passions, they are not generally given to excessive or
misdirected jealousy, envy, or hatred, and their anger tends to be short lived, so that the
impulses that often lead others to wrongdoing are characteristically lacking. 9. To see what
the appropriate limit on these passions is, we need to look at several articles in which they
are taken up. Jealousy he describes as a kind of anxiety about losing some possession we
value, its strength being a function of the value we set on this possession; it is necessarily
accompanied by the belief, more or less robust, that we shall not lose it. Jealousy is justified if
our valuation is correct, and it is useful in stimulating us to take appropriate care. Envythat
is, the passion of envy, which is quite different from the vice of envy Descartes describes as
"a kind of sadness mingled with hatred, which results from our seeing good coming to those
we think unworthy of it" (II:466; I:394). Envy is justified only in the case of goods of fortune
(not gifts of nature) and only when the recipients of the goods are not worthy of them.
Descartes sees justified envy as
____________________
9.
In article 156 Descartes says that the gnreux are not given to excessive anger because
"they have esteem for everyone." Now, not having contempt is not the same as having
esteem. In article 154, Descartes argues that the gnreux has no contempt for others. But
the logic of his argument would not lead to the stronger conclusion that he would have
esteem for others. Indeed, so far as esteem is legitimate only if it is based on the virtuous
will of the person esteemed, it would not be legitimate in the case of someone with a
vicious disposition. What Descartes needsand, indeed, as I have argued above, what
articles 154-56 implyis a distinction between minimum respect for persons, on the one
hand, and esteem, on the other.
Descartes also lists fear as a passion to which the gnreux is not liable. In his later
discussion of timidity and fear (or terror) he finds little use for the first and none for the
second; the gnreux is, by contrast, given to the passions of courage and boldness. He also
has the virtue (as distinct from the passion) of courage, which is essentially the confidence
and determination to do whatever he judges to be best.
-153arising from a natural love of justice; indeed, the justified passion Descartes calls envy might
for this reason also be described as hatred of injustice.

As we have seen in the previous chapter, the gnreux are effectively beyond the power of
fortune (II:470; I:395). But they do not for that reason lack compassion, for, since they have
goodwill toward everyone, and pity itself is sadness mixed with goodwill, they are made sad
by seeing another suffer undeserved deprivation of some good. The gnreux's chief object of
pity, however, "is the weakness of those whom they see complaining." "For they think that no
misfortune could be so great an evil as the timidity of those who cannot endure it with
forbearance. And although they hate vices, they do not on that account hate those whom they
see prone to them: they merely pity them." 10.
Next, we come to some special forms of love and hatred characteristic of the gnreux: favor
and gratitude, on the one hand, and indignation and anger, on the other. Favor is the special
feeling of love or approbation of another aroused in us by his performing some good deed or,
indeed, having a sincere intention to do so. The same feeling takes the form of gratitude when
we are the beneficiary. Favor, being a form of love, gives rise to the desire that some good
come to the doer of good deeds and in the case of gratitude, to the desire to reciprocate.
Indignation is a special form of hatred or aversion, directed to those who do benefit or harm
others who do not merit it. For the gnreux, however, although they cannot fail to disapprove
wrongdoing and vice, the passion of indignation is comparatively moderate; only in the case
of "the most extraordinary vices" do they "become incensed" (II:477;I:398). Anger, the form
this hatred takes when we are the victim of wrongdoing, gives rise to the desire for vengeance,
which, says Descartes, "is the most compelling of all desires" (II:477; I:399). The gnreux,
who, owing to their characteristic goodwill, will always expect the best of others, are
therefore surprised by wrongdoing. The anger they feel when, for example, they are the victim
of theft, is accordingly mixed with astonishment. As their surprise diminishes and they reflect
on the low esteem they have for such goods as may have been taken from them, and on the
satisfaction they take in the control they have over their desires, their anger soon turns to
"contempt, or at the most indignation." ii.
The passions of repentance, pride, and shame remain to be considered. Liability to the
passions of pride and shame I have not taken the to be part
____________________
10.
A similar point is made in connection with indignation; the specific object of the passion of
indignation is some vice or wrongdoing, not the person who performs it. Thus we can
consistently censure wrongdoing and vice and at the same time neither hate nor condemn
the agent.
ii.
Here, I think we should take the term "contempt" to refer to the moral judgment that the
action is bad or wrong, not to the passion of contempt.
-154of the moral sensibility of the perfectly gnreux. Nor have I mentioned repentance. Were
virtue perfected, these passions would be absent. Repentance, Descartes writes, is a kind of
sadness, the most bitter kind, which "results from our believing that we have done some evil
deed" (II:472; I:396). Since such evil deeds would be prompted typically by envy or anger,
the perfectly gnreux would lack any motive for the kind of wrongdoing that would occasion
repentance. Occasions for pride and shame would be similarly few. Descartes describes pride
as joy deriving from the expectation of the praise of others and shame as the sadness deriving
from the expectation of blame; their function is to "move us to virtue, the one through hope
and the other through anxiety." He adds, howeverand the remark would presumably apply

to the gnreuxthat it is not entirely desirable to rid ourselves of these passions, "for
although the common people are very bad judges, yet because we cannot live without them
and it is important for us to be an object of their esteem, we should often follow their opinions
rather than our own regarding the outward appearance of our actions" (II:483; I:401). We
should not read this remark as a mere counsel of prudence or as an invitation to hypocrisy. In
the first place, we must remember that in the context in which it applies, namely, our deciding
what to do in particular circumstances, we seldom if ever will have knowledge even
approximating clear and distinct perception of what is right and what wrong. In the second
place, since, as gnreux, we have virtuous humility, we will be disposed to have genuine
respect for the moral views of others. To be sure, we cannot be uncritical, but it is sometimes
a narrow line between following others' moral advice uncritically and not giving it the
hearing it merits, which is the point Descartes makes in his letter to Elizabeth of September
15, 1645: "One must ... examine minutely all the customs of one's place of abode to see how
far they should be followed. There we cannot have certain proofs of everything, still we must
take sides, and in matters of custom embrace the opinions that seem the most probable, so
that we may never be irresolute when we need to act" (4:295; 3:267).
We can say in summary, then, that although the gnreux are generally benevolent and
respectful in addition to their being philosophically enlightened, 12. their beneficence is
sensitive to justice or merit; while they are
____________________
12.
They know the several most useful things to know, which Descartes lists in his letter to
Elizabeth of September 15, 1645: that God is good, that the soul is immortal, and that the
universe is immense. They also know how to achieve self-mastery, and they know the
theory of the passions on which their knowledge of this technique is based. The knowledge
of God, of immortality, of matter, of free will, and of human nature is all knowledge at the
fifth degree of wisdom. Since knowledge at this level is philosophical knowledge,
according to Descartes's conception of philosophy, knowing how to live wellto walk
with assurance in this life and, therefore, knowing how to attain batitude, presupposes
philosophy. Thus, we should understand the following passage from the French preface to
the Principles: "Living without
-155compassionate and readily moved by indignation, they are never vengeful, and owing to their
predisposition of goodwill they are given rather to excuse than to blame others for
wrongdoing.
4. Beyond being supremely content, the gnreux is characteristically respectful, forgiving,
and charitable attitude toward others. If, then, the dominant attitude toward others of the
gnreux is due respect for persons on account of their freedom and rationality, we might
reasonably try to base a Cartesian first-order morality on the value of human freedom. We
could adopt as the ground for such a morality that freedom of persons establishes a
benchmark of equality among all human beings and provides at once a moral barrier to their
being ill-treated, on the one hand, and moral reasons for seeking their true good, on the other.
13.
It is important to keep in mind that this principle is, in one sense, overdetermined, since the
moral barrier to the ill-treatment of others is a function both of the goods that, through our
actions, others might be led to possess and of the respect or esteem that is owed to them

simply as rational beings with a free will. The general or imperfect duty to promote the wellbeing of others is similarly overdetermined.
This talk of respect for free and equal moral persons calls to mind Kant's second formulation
of the categorical imperative or, perhaps, some more recent contractualist theory of justice or,
more generally, of the right. I do not intend to follow up the idea that Cartesian first-order
morality could be developed along contractualist lines, although I believe this would be a
natural way to proceed, once we follow Descartes and take the gnreux's conception of other
persons as correct. "Cartesian contractualism" would be morally motivated, however, notas
in Hobbes's contractualismprudentially motivated. Although Descartes says that many wellestablished conventions that more or less effectively control social conduct could no doubt be
explained as a kind of unconscious resolution of the problem presented by the Hobbesian state
of nature, principles seen only in this way could not, for Descartes, be construed as
specifically moral principles, for
____________________
philosophizing is exactly like having one's eyes closed without ever trying to open them;
and the pleasure of seeing everything which our sight reveals is in no way comparable to
the satisfaction accorded by knowledge of the things which philosophy enables us to
discover. Lastly, the study of philosophy is more necessary for the regulation of our morals
and our conduct in this life than is the use of our eyes to guide our steps" (9B:3-4; I:180).
13.
In article 83 Descartes speaks of a universal friendship we are to have for all other human
beings. Speaking of others, he writes, "they are so truly the objects of this passion [love in
the form of friendship, as distinct from either mere affection or devotion that there is no
person so imperfect that we could not have for him a very perfect friendship [une amiti
trs parfaite], given that we believe ourselves loved by him and that we have a truly noble
and generous soul" (II:390; I:357).
-156they flow not from fundamental love and respect but from interest, fear, and suspicion. A
Cartesian morality of amitiof mutual love and respect could use the notion of a contract
or ideal agreement as a device to help clarify what is impermissible, permissible, or required
among persons who are (we may suppose) already morally motivated. Indeed, I believe that
Descartes might have found the idea of a contract or ideal agreement as a canon for moral
judgment congenial, although as a moral realist he could not accept the contemporary
contractualist idea that we are to define moral truth in terms of such an agreement.
In any case, what we know clearly and distinctly, in Descartes's view, is that persons have
value as being free and rational. They are therefore proper objects of our esteem, and we are
to treat them accordingly. But what, precisely, treating them accordingly amounts to in
specific cases we cannot always know for certain. As a first approximation, however, we may
say that to treat them accordingly is, on the one hand, not to harm them and, on the other, to
bestow some benefits on them. But what constitutes the true good of another, relative to
which we define harm and benefit? And is the injunction to promote the good of others to be
understood collectively or distributively? To answer the first question, we can refer back to
Descartes's theory of the good. The supreme good of each person, of course, is virtue. What
might be called the complete good of each person is virtue completed by gifts of nature and
gifts of fortune. Gifts of fortune include economic, social, and political power, opportunity to
develop native talents and to acquire useful skills, and the conditions for conserving health

and strength of the body. Up to a point, these goods are necessary for human development and
for a distinctively human mode of life. Among gifts of nature are those that concern the body:
a strong constitution, latent physical skills, and talents; and those that concern the mind:
beside le bon sens itself, intelligence, good memory, and a creative imagination. Since the use
and development of gifts of nature, including le bon sens, must depend in large part on
education, even our attaining the supreme good must inevitably owe something to good
fortune. To be sure, as we have seen in the previous chapter, Descartes argues that our
attaining virtuethe supreme goodis entirely within our power. And although this is true,
in the sense that, with proper understanding of our nature and our place in it, we can come to
see what the supreme good for us is and how we may achieve it through the proper use of our
freedom, it is nonetheless a matter of good fortune that we come to have the requisite
understanding. Descartes was understandably grateful to his tutors at La Flche and no
doubt for the other ways in which fortune paved the way for his own intellectual and moral
development. If this is not a point Descartes himself dwells on, he touches on
-157it in a different wayswhen, for example, he concedes to Princess Elizabeth that bad health
can effectively deprive us of the free use of our reason, or when he extols the value of
philosophy. Since the value of gifts of nature is thus dependent in some measure on fortune,
we may simplify our account of human well-being by saying it is partly a matter of our gifts
and partly a matter of our will. Our achieving the supreme good is essentially dependent on us,
but, as I have just pointed out, our attainment of virtue is not entirely independent of what is
given to us; indeed, in many obvious ways, it is dependent on physical and social conditions.
So we might put the matter this way. Although virtue is a matter of self-mastery and is
therefore a good that we can acquire only through the proper use of our own will, the
conditions and opportunity for self-mastery are in part a matter of fortune and, in particular, a
matter of what others do for us.
5. Descartes's older contemporary, Hugo Grotius, it has been said, was the first to systematize
commonsense morality, 14. but more thoroughly systematic still was the natural law theory of
Samuel Pufendorf, whose taxonomy Kant used (in slightly modified form) to organize his
discussion of duty in the Grundlegung zur Metaphysik der Sitten. 15. I plan to use Pufendorf's
taxonomy here. The two main distinctions are between duties to self and duties to others, on
the one hand, and between perfect and imperfect duties, on the other. A duty is perfect when it
is the duty to perform or refrain from a specific action; it is true of some perfect duties that
they can be externally enforced and of some also that they are correlated with rights of others.
16.
A duty is imperfect just when it is a duty to have a certain purpose
____________________
14.
"In 1788 Thomas Reid praised the 'immortal Hugo Grotius' as the author of the first
noteworthy attempt to systematize the commonsense morality of the human race with the
aid of the civil law's technical apparatus" ( Schneewind 1990, I: 88 ).
15.
Kant departed from this simple scheme in his later Metaphysik der Sitten. Pufendorf
writes: "Of the duties incumbent upon man in accordance with natural law the most
convenient division seems to be according to the objects in regard to which they are to be
practiced. From this standpoint they are classified under three main heads: the first of
which instructs us how, according to the dictate of sound reason alone, a man should
conduct himself toward God, the second, how toward himself, the third, how toward other
men" ( Schneewind 1990, I: 164 ). Concerning duties to self, Pufendorf writes: "But duties

of man to himself spring from religion and sociability conjointly. For the reason why he
cannot determine certain acts concerning himself in accordance with his own free will, is
partly that he may be a fit worshiper of the Deity, and partly that he may be a good and
useful member of human society" (I: 164 -65). As to absolute duties to othersduties that
do not presuppose human institutionPufendorf gives the following three in this order: the
duty not to injure another; the duty to esteem and treat every other as a natural equal; the
duty to promote the advantage of another so far as one conveniently can.
16.
Kant later called these duties "juridical" and argued not only that they can but that they
ought to be enforced through state power, fairly and impartially administered.
-158or end; such duties cannot be externally enforced, and they do not typically translate into a
specific requirement to perform a particular action. 17.
We may begin with perfect duties to oneself. As we have seen, for such duties, there will be,
in general, a twofold rationale, one grounded in the value of what perfects us and the other in
the esteem we owe ourselves as free and rational yet human beings. Since the supreme good
is virtue, we may be said to have a dutya duty to ourselvesto cultivate virtue, but on the
face of it this duty would not seem to be a perfect one, since it is a duty to have a certain end
without at the same time requiring the performance of specific actions. On the other hand, it
would be inconsistent with our having that end if we acted in ways that made its attainment
impossible or even less likely. Accordingly, other things equal, it is morally impermissible
voluntarily to take our own life or in other ways permanently to corrupt our own intellect and
will. Let us call suicide the act of voluntarily killing oneself where that is the primary
intention, thereby distinguishing suicide from taking one's own life in order to protect the
rights or promote the well-being of others, or to escape defilement or the loss of one's reason
and humanity. In the case of suicide, so understood, it would seem that the values in play
would always be decisive against it, since no value to be realized through suicide could trump
the value of virtue itself.
Now, we know from the Discourse that Descartes took a very dim view of the Stoic doctrine
that suicide was, at least for the sage, morally permissible. And we learn from his
correspondence with Princess Elizabeth some years later, when she pressed him on the matter,
that in his view, so long as our condition is not such as to deprive us of the use of our own
reason, the goods and joy available to us by choosing to live always outweigh the evils and
sorrow we would escape by ending our life. Since, however, this argument would appear to
fall short of a defense of the perfect duty never to commit suicide, we are left with the
question of whether a Cartesian first-order morality should include a categorical prohibition
of suicide.
Before meeting this question head on, let us consider some passages from Descartes's
correspondence that bear on this question, even if they do not address it directly. The first
comes from his letter to Elizabeth of September I, 1645. After summarizing a remark of
Elizabeth's"You observe very truly that there are diseases which take away the power of
reasoning and with it the power of enjoying the satisfaction proper to a rational mind.... We
cannot altogether answer for ourselves except when we are in our own
____________________
17.
The duty to promote the well-being of others is imperfect, but in special circumstances it

may demand that one render aid to this person here and now, for example, when one can
without great effort or cost save the life of another.
-159power. It is better to lose one's life than to lose the use of reason" (4:281 82; 3:262)he
continues:
There other indispositions which do no harm to one's reason but which merely alter the
humours, and make a man unusually inclined to sadness, or anger, or some other passion.
These certainly cause distress, but they can be overcome; and the harder they are to conquer,
the more satisfaction the soul can take in doing so. The same is true of all exterior handicaps,
such as the splendour of high birth, the flatteries of courts, the adversities of fortune, and also
great prosperity, which commonly does more than misfortune to hamper the would-be
philosopher. When everything goes according to our wishes we forget to think of ourselves;
when fortune changes we are the more surprised the more we trusted it. Altogether, we can
say, nothing can completely take away our power of making ourselves happy provided that it
does not trouble our reason. (4:283; 3:263)
In his later letter of October 6, 1645, speaking of those who claim to be "tired of this life" and
who would leave it, he writes that "true philosophy, on the contrary, teaches that even amid
the saddest disasters and most bitter pains a man can always be content, provided that he
knows how to use his reason" (4:315; 3:272). And in January, 1646, he enlarges upon this
idea:
there are always more good things than evil in this life; and I said this because I think we
should make little account of all things outside us that do not depend on our free will in
comparison with those that do depend on it. Provided we know how to use our will well, we
can make everything which depends on it good and thus prevent the evils that come from
elsewhere, however great they may be, from penetrating any further into our souls than the
sadness which actors arouse when they enact before us some tragic history. But I agree that to
reach such a point a man has to be very philosophical indeed. Nevertheless, I think that even
those who most give rein to their passions, really judge deep down, even if they do not
themselves perceive it, that there is more good than evil in this life. Sometimes they may call
death to help them when they feel great pain, but it is only to help them bear their burden, as
in the fable, and for all that they do not want to lose their life. And if there are some who do
want to lose it, and who kill themselves, it is due to an intellectual error and not to a wellreasoned judgement, nor to an opinion imprinted on them by nature, like the one which makes
a man prefer the goods of this life to its evils. (4:355-56; 3:283)
In summary, so long as they have no acknowledged prospects of losing their reason, those
who deliberately commit suicide must be seen as having
-160judged badly, sacrificing a lessor for a greater good entirely in their power. The texts do not
readily suggest the Kantian argument that suicide entails treating oneself as a means and must
therefore be categorically proscribed, although we might conjecture that Descartes, in view of
the distinction between two kinds of ends that we can ascribe to him, would have approved of
an argument of this kind. Nevertheless, the prohibition against suicide, in Descartes's

expressed view, derives from the imperfect duty of self-perfection, which is to cultivate virtue,
that is, to strive to achieve the ideal of gnrositdescribed in the early articles of part 3 of
the Passions of the Soul. 18. So long as we have our rational powers, virtue and the full
contentment that supervenes on it are within our power, and these are goods that must always
outweigh the loss of other external goods, even the loss of our cherished friends.
Cultivating virtue, as we have seen, is the certain route to happiness. It involves adhering
strictly to the three rules of morality and duly attending to certain truths of metaphysics. More
specifically, it involves disciplining the passions, thereby effecting courage in the pursuit of
our chosen ends and temperance in our choice of ends; acknowledging Providence and one's
place in and dependence on one's family and one's society; and choosing a mode of life,
profession, or craft that is appropriate to our natural talents and predilections and is of use to
the larger wholes on which one depends. Particularly useful for the cultivation of virtue is the
passion of gnrosit.
if we occupy ourselves frequently in considering the nature of free will and the many
advantages which proceed from a firm resolution to make good use of itwhile also
considering, on the other hand, the many vain and useless cares which trouble ambitious
peoplewe may arouse the passion of generosity in ourselves and then acquire the virtue.
Since this virtue is, as it were, the key to all the other virtues and a general remedy for every
disorder of the passions, it seems to me that this consideration deserves serious attention.
(II:454; I:388)
Two other duties are also to be included under the heading of imperfect duties to ourselves:
the duty to develop our cognitive powers, since these
____________________
18.
When Pierre Charron, in On Wisdom, speaks of the duty to self, he clearly views it as a
Kantian imperfect duty; it is, in sum, the duty to take charge of one's life as a whole and to
see to it that one is not a simple reactor to present desire and circumstance; it is, in short,
to live life seriously. When, later, Samuel Pufendorf speaks of man's duty to himself, he
says that it is a mixture of man's duty to God and man's sociability, that is, his duty to
others: "The duties of man to himself spring from religion and sociability conjointly. For
the reason why he cannot determine certain acts concerning himself in accordance with his
own free will, is partly that he may be a fit worshiper of the Deity, and partly that he may
be a good and useful memberof human society" ( Schneewind 1990, I: 164 -65). Neither
has the conception of a perfect duty to self that Kant has.
-161are the necessary means for the acquisition of knowledge and the perfection of the self, and
the duty to maintain our physical well-being, both as a perfection of our animal nature and as
a condition for the effective pursuit of other eligible ends, most especially, virtue itself.
6. The duties to others may be divided between the perfect duties of justice and the imperfect
duties of charity or benevolence. 19. One way of interpreting the formerwhich would place
Cartesian ethics squarely in the tradition in which he was educated at La Flchewould be to
follow a tradition beginning with Aristotle and divide duties of justice into those of
distributive justice and those of commutative justice. In general, justice has to do with the
distribution of goods, that is, of such goods as admit of distribution by human agents and their

institutions. Some distributions are just and others unjust. Commutative justice has to do with
the exchanges among individuals in their private capacity, as in, paradigmatically, buying and
selling. Aristotle and, following him, Aquinas distinguish between voluntary and involuntary
transactions; the latter are those in which consent is only unilateral, not mutual. Closely
following Aristotle, Thomas Aquinas lists the following types of involuntary transaction as
unjust: fraud, violence, theft, robbery, murder, assault, maiming, bearing false witness,
calumny ( Thomas Aquinas 1988, 169 ). Commutative justice is also known as rectificatory
justice. In these cases, the nonconsenting party suffers a loss while the consenting party
profits unjustly. Justice is the condition in which each party's equal loss is "rectified" by each
party's equal gain. Many involuntary transactions, however, effectively preclude meaningful
rectification, since they leave one party in no condition to receive suitable compensation.
Thus, it would be a clear violation of commutative justice to kill a person "involuntarily" in
order to save the lives of five others. In the case of voluntary transactions, where there is no
force or fraud, the presumption is that the exchange of goods leaves the parties equally well
off after the exchange as they were before. Injustice occurs when one party fails to pay her
voluntarily incurred debt to the other. The prescriptions and proscriptions of commutative
justice are grounded for a Cartesian first-order morality in the moral equality of each person
as such, an equality grounded in their rationality and free will. The ideal for all transactions
among pri____________________
19.
Descartes's occasional references to justice do not suggest that he had anything very
specific or controversial in mind. Oddly enough, perhaps the majority of his references to
justice are made in connection with God; but even here, starting from the claim that God is
perfectly just, Descartes confines himself principally to two substantive inferences, that
God would not deceive us and that all is ordered for the best. As to the human virtue of
justice, Descartes mostly leaves it to us to work out what we might expect of those who
possess this virtue. Nonetheless, we can fill in some of the demands of justice with
reasonable confidence, since these are both commonplace and well-grounded in his theory
of the value of persons as such.
-162vate individuals is for all parties to act toward other parties only in a manner that permits their
informed and uncoerced consent. The moral proscriptions against deceit and violence are, of
course, moral commonplaces. The point to be made here is just that they are well grounded in
the Cartesian conception of persons, which entails that the reason and will of each person is
owed respect, one person having no more intrinsic value than any other.
But to say that such acts as constitute theft, fraud, violence, etc., are unjust does not imply
immediately that all such acts must be morally condemned. If in an emergency I steal
critically needed drugs to save my wife's life, I act contrary to justice, but it is not so clear that
what I do is wrong in the circumstances and given my intention. Of course, in such a case, it
may be possible to restore the balance by compensating the pharmacist for her loss, so that the
injustice is adequately rectified. The case of killing one innocent to save five others, however,
does not admit of this sort of "rectification." And for this very reason, a Cartesian moralist
might well argue that unrectifiable unjust acts cannot be morally justified, no matter how
desirable the ends that they alone can attain might be. 20. So too, by the same argument, would
it be morally impermissible to kill one innocent person in order to prevent someone else from
murdering five innocent persons. On the other hand, it would be permissible to kill someone

if this were the only reasonable way to prevent him from murdering others. It would be
allowable for something like Kant's justification of coercion: it sometimes is needed to
prevent unjustified coercion; the former cancels out or "rectifies" the latter. 21.
7. Let us now turn to distributive justice, which concerns the distribution of goods by some
authority to members of some group according to their merit. 22. The group may be a family or
some voluntary association. Of
____________________
20.
Thus do Eleonore Stump and Norman Kretzman interpret Aquinas. The doctor's action of
taking the life of Smith, whose organs he uses to save the lives of five others, is morally
bad, since it is unrectifiably unjust to Smith ( Stump and Kretzman 1991, 120 ).
21.
"Any opposition that counteracts the hindrance of an effect promotes that effect and is
consistent with it. Now, everything that is unjust is a hindrance to freedom according to
universal laws. Coercion, however, is a hindrance or opposition to freedom. Consequently,
ifa certain use of freedom is itselfa hindrance to freedom according to universal laws (that
is, unjust), then the use of coercion to counteract it, inasmuch as it is the prevention of a
hindrance to freedom, is consistent with freedom according to universal laws; in other
words, this use of coercion is just" ( Kant 1965, 35 - 36 ).
22.
Aristotle distinguishes between commutative and distributive justice in terms of two kinds
of proportional equality, arithmetic and geometric. In the case of commutative justice, each
party to a transaction is taken as equal, with the moral consequence that a fair share in any
exchange among equals is an equal share. In the case of distributive justice, it is assumed
that the persons to whom shares are distributed are in a relevant sense unequal (for
example, one person may be of greater value to the community and hence deserving of a
larger share); their comparative value (to the community, in this example) will stand in a
certain ratio, and this ratio will be equal to the ratio of the value of their shares.
-163particular interest here is distributive justice at the level of the state. And here we may
distinguish between ideal distributive justice and de facto distributive justice. By the latter I
mean the legally authorized set of practices that in fact determine the distribution of goods in
a given civic society. By ideal distributive justice I mean the distribution of goods that would
be effected by legally authorized practices in a morally ideal civic society (as determined by
some moral conception of society and of persons that make it up). When Descartes remarked,
as he occasionally did, that he did not wish to write about morality, what he had especially in
mind is that he did not wish to write about ideal distributive justice, that is, about moral
politics. He did not wish to challenge the established social and political hierarchies or the
practice of distributing the benefits and burdens of such social cooperation as existed in his
timehowever unjust he may have found them. In the Discourse he expresses a view on this
issue he often repeated. Acknowledging the difficulties attendant on his private intellectual
reform, he notes that they
could not be compared with those encountered in the reform of even minor matters affecting
public institutions. These large bodies are too difficult to raise up once overthrown, or even to
hold up once they begin to totter, and their fall cannot but be a hard one. Moreover, any
imperfections they may possessand their very diversity suffices to ensure that many do
possess themhave doubtless been much smoothed over by custom; and custom has even
prevented or imperceptibly correct many imperfections that prudence could not so well

provide against.... That is why I cannot by any means approve of those meddlesome and
restless characters who, called neither by birth nor by fortune to the management of public
affairs, are yet forever thinking up some new reform. And if I thought this book contained the
slightest ground for suspecting me of such folly, I would be very reluctant to permit its
publication. (6: 14 ; I: 117 -18)
Elsewhere he observes that "as regards conduct, everyone is so full of his own wisdom that
we might find as many reformers as heads if permission to institute change in these matters
were granted to anyone other than those whom God has set up as sovereigns over his people
or those on whom he has bestowed sufficient grace and zeal to be prophets" (6: 61 ; I: 142 ).
23.
To
____________________
23.
Descartes was scarcely alone in holding such conservative views. We find in Montaigne,
for example: "Not in theory only, but in truth, the best and most excellent form of
government for every nation is that under which it has maintained itself. For its form and
essential suitability it is dependent on custom.... Nothing presses so hard upon a state as
innovation; mere change gives scope to injustice and tyranny. When some part becomes
loosened, it may be propped; we may guard against the alteration and corruption natural to
all things carrying us too far from our beginnings and principles. But to undertake to put so
great a mass back into the melting-pot, to renew the foundations of so great an edifice, is to
efface a picture in
-164have the virtue of justice, in sum, is to be disposed to respect the principle of commutative
justice and, so far as this is judged to permit it, to adhere to the positive laws and customs of
one's state and community.
8. But suppose one is born to rule or in some other way assumes political power. Are there
reasons of state that take precedence over the rules of commutative justice? Can princes lie,
dissemble, use preemptive violence? Obliged by his respect for Princess Elizabeth, Descartes
reluctantly turns to such questions. He considers first a prince who is unjust and who has
come to power through unjust means. He concedes that Machiavelli's rules for such a prince's
remaining in power are correct. But this is scarcely an endorsement of such practices, much
less a concession that, in the case of such a prince, the rules of justice may have to be
suspended. What of the just prince? Here again, Descartes makes no concessions to dirty
hands. He considers only the case in which the just prince must deal with his enemies.
Speaking of such a prince, he writes: "A distinction must be made between subjects, friends
or allies, and enemies. With regard to these last, one has a virtual license to do anything,
provided that some advantage to oneself or one's subjects ensues; and I do not think it wrong
in such a case to join the fox with the lion and use artifice as well as force" (4:488; 3:293). 24.
Here the underlying rationale is that once another threatens violence or may be suspected of
deceit, the prince may defend the state by the use of means that would, in normal contexts, be
unjust, such as violence and deceit, if such devices seem to be required. 25.
9. This brings us finally to the imperfect duty of beneficence and to the end of our study.
About the duty of beneficence little remains to be said. Descartes says of the gnreux, who
manifest beneficence in its ideal form, that "they esteem nothing more highly than doing good
to others and dis-

____________________
the cleaning, to reform particular defects by a general confusion, to cure a disease by
killing the patient, to be not so much for changing as for overturning everything ( Cicero)"
( Montaigne 1927, 2:422-23).
24.
Even in dealing with enemies, domestic or foreign, Descartes draws the line at abusing
friendship. "But I rule out one kind of deception which is so directly hostile to society that
I do not think it is ever permissible to use it, although our author [ Machiavelli] approves it
in several places; and that is pretending to be a friend of those one wishes to destroy, in
order to take them by surprise. Friendship is too sacred a thing to be abused in this way;
and someone who has once feigned love for someone in order to betray him deserves to be
disbelieved and hated by those whom he afterwards genuinely wishes to love" (4:488;
3:293).
25.
It is not my purpose here to settle this controversial issue of the degree to which Descartes
makes concessions to Machiavelli's views. In the letter just quoted he also writes that
"justice between sovereigns does not have the same bounds as justice between individuals.
It seems to me that in this instance God gives the right to those to whom he gives power.
But of course the most just actions become unjust when those who do them think them so."
I interpret this to say that a sovereign who believes he is just, who has not unjustly usurped
power, and who faces an enemy threatening the stability and common good of the state
may then use the forces at his disposal, or even deceit, to reduce the enemy's threat.
-165regarding their own self-interest" and that they are accordingly "perfectly courteous, gracious
and obliging to everyone" (II:448; I:385). They encounter only two limitations (aside from
duties of justice) on their promotion of others' well-being: first, their benevolence should not
threaten their own use of reason, which would be contrary to an imperfect duty to self; second,
they should not undertake beneficent projects they do not believe they are adequate to carry
out. If, by the standards of ordinary morality, the activities of the perfectly gnreux would
often seem supererogatory, it must be remembered that the perfectly gnreux are also
perfectly content.
-166Works Cited
Alqui, Ferdinand. 1987. La dcouverte mtaphysique de l'homme chez Descartes. Paris:
Presses universitaires de France.
. 1988. Oeuvres philosophiques de Descartes. Vol. I ( 1618- 1637). Classiques Garnier.
Paris: Bordas.
. 1989. Oeuvres philosophiques de Descartes. Vol. 3 ( 1643- 1650). Classiques Garnier.
Paris: Bordas.
Annas, Julia. 1993. The Morality of Happiness. New York: Oxford University Press.
Annas, Julia, and Jonathan Barnes. 1985. The Modes of Scepticism. Cambridge: Cambridge
University Press.

Ariew, Roger, and Daniel Garber, ed. and trans. 1989. G. W Leibniz: Philosophical Essays.
Indianapolis: Hackett.
Blom, John J. 1978. Descartes: His Moral Philosophy and Psychology. New York: New York
University Press.
Cicero. 1990. De officiis. Trans. Walter Miller. Loeb Classical Library. Cambridge: Harvard
University Press.
. 1991. On Stoic Good and Evil. De finibus bonorum et malorum. Book 3. Trans. and
commentary M. R. Wright. Warminster: Aris & Phillips.
Cottingham, John, trans. and commentary. 1976. Descartes' Conversation with Burman.
Oxford: Clarendon Press,
Cumming, Robert. 1955. " Descartes' Provisional Morality." Review of Metaphysics 9:207-35.
Curley, Edwin. 1993. " Certainty: Psychological, Moral and Metaphysical." In Essays on the
Philosophy and Science of RenDescartes, ed. Stephen Voss. New York: Oxford University
Press.
D'Angers, Julien Eymard. 1954. " Snque, pictte et le Sto
cisme dans l'oeuvre de Ren
Descartes." Revue de Thologie et de Philosophie 3:169-96.
. 1976. "Recherches sur le Stoicisme aux XVIE et XVIIE sicles". Ed. L. Antoine. Studien
und Materialien Zur Geschichte der Philosophic. New York: Georg Olms Verlag.
-167Descartes, Ren. 1911. The Philosophical Works of Descartes. Vol. I. Trans. Elizabeth S.
Haldane and G. R. T. Ross. Cambridge: Cambridge University Press.
. 1958. Descartes Philosophical Writings. Trans. Norman Kemp Smith. New York: Random
House.
. 1964-76. Oeuvres de Descartes. Edited by Ch. Adam and P. Tannery. Rev. ed. Paris: J.
Vrin/C.N.R.S.
. 1985. The Philosophical Writings of Descartes. Vols. I and 2. Trans. John Cottingham ,
Robert Stoothoff, and Dugald Murdoch. Cambridge: Cambridge University Press.
. 1991. "The Philosophical Writings of Descartes". Vol. 3, The Correspondence. Trans.
John Cottingham, Robert Stoothoff, and Dugald Murdoch. Cambridge: Cambridge University
Press.
Du Guillaume Vair. 1951. The Moral Philosophie of the Stoicks. Ed. Rudolf Kirk. Trans.
Thomas James. 1598. Reprint, New Brunswick, N.J.: Rutgers University Press.
Elster, Jon. 1984. Ulysses and the Sirens. Cambridge: Cambridge University Press.

Epictetus. 1983. The Handbook of Epictetus. Trans. Nicholas P. White. Indianapolis: Hackett.
Espinas, Alfred. 1925. Descartes et la morale. Paris: Bossard.
Garber, Daniel. 1987. " Descartes et la mthode en 1637." In Le discours et sa mthode. Ed.
N. Grimaldi. Paris: Presses Universitaires de France.
Gaukroger, Stephen. 1995. Descartes: An Intellectual Biography. Oxford: Clarendon Press.
Gilson, tienne. 1947. Discours de la mthode: texte et commentaire. Paris: J. Vrin.
Gouhier, Henri. 1924. La pense religieuse de Descartes. Paris: J. Vrin.
Gueroult, Martial. 1985. "Descartes' Philosophy Interpreted according to the Order of
Reasons". Trans. Roger Ariew. 2 vols. Minneapolis: University of Minnesota Press.
Originally published as Descartes selon l'ordre des raisons ( Paris: Aubier-Montaigne, 1968).
Hamelin, 0. 1921. Le systme de Descartes. Paris.
James, William. 1949. Essays in Pragmatism. Ed. Alburey Castel. New York: Hafner.
Kant, Immanuel. 1965. The Metaphysical Elements of Justice. Trans. John Ladd. New York:
Bobbs-Merrill.
Kenny, Anthony. 1972. " Descartes on the Will." In Cartesian Studies, ed. R J. Butler. Oxford:
Basil Blackwell.
Keohane, Nannerl 0. 1980. Philosophy and the State in France. Princeton: Princeton
University Press.
Le Michle Doeuff. 1989. The Philosophical Imaginary. Trans. Colin Gordon. London:
Athlone.
Leibniz, G. W. 1989. Philosophical Essays. Trans. Roger Ariew and Daniel Garber.
Indianapolis: Hackett.
Levi, Anthony S. J. 1964. French Moralists. Oxford: Clarendon Press.
Mesnard, Pierre. 1936. Essai sur la morale de Descartes. Paris: Boivin.
Montaigne, Michel de. 1927. The Essays of Montaigne. 2 vols. Trans. E. J. Trechmann.
London: Oxford University Press.
Rodis-Lewis, Genevive. 1970. La morale de Descartes. Paris: Presses Universitaires de
France.
. 1987" Le dernier fruit de la mtaphysique Cartsienne: La gnrosit." Les tudes
Philosophiques I (Jan-Mar):43-54.
. 1988. Introduction and notes to Descartes: Les passions de l'me. Paris: J. Vrin.

. 1995. Descartes: Biographie. Paris: Calmann-Lvy.

-168Rorty, Amlie Oksenberg. 1986. " Cartesian Passions and the Union of Mind and Body." In
Essays on Descartes's Meditations, ed. Amlie Oksenberg Rorty. Berkeley: University of
California Press.
Russell, Bertrand. 1945. A History of Western Philosophy. New York: Simon and Shuster.
Schneewind, Jerome B. 1990. Moral Philosophy from Montaigne to Kant. Vol. I. Cambridge:
Cambridge University Press.
Seneca. 1965. " On the Happy Life." In Moral Essays, trans. John W. Basore. Loeb Classical
Library. Cambridge: Harvard University Press.
Sidgwick, Henry. 1954. Outlines of the History of Ethics. New York: St. Martin's.
Sirven, J. 1928. Les annes d'apprentissage de Descartes (1596-1628). Albi: Imprimerie
Cooprative du Sud-Ouest.
Stump, Eleonore, and Norman Kretzman. 1991. " Being and Goodness." In Being and
Goodness, ed. Scott MacDonald. Ithaca: Cornell University Press.
Thomas Aquinas, Saint. 1988. On Law, Morality, and Politics. Ed. William P. Baumgarth and
Richard J. Regan. Indianapolis: Hackett.
Tuck, Richard. 1988. " Optics and Sceptics." In Conscience and Casuistry in Early Modern
Europe, ed. Edmund Leites. Cambridge: Cambridge University Press.
Voss, Stephen H., ed. and trans. 1989. The Passions of the Soul, by RenDescartes.
Introduction by G. Rodis-Lewis. Indianapolis: Hackett.
Vuillemin, J. 1988. " L'intuitionnisme moral de Descartes et le traitDes passions de l'me."
Kantstudien 79:17-32.
-169-