Virtues of The Mind An Inquiry Into The Nature of Virtue and The Ethical Foundations of Knowledge by Linda Trinkaus Zag
Virtues of The Mind An Inquiry Into The Nature of Virtue and The Ethical Foundations of Knowledge by Linda Trinkaus Zag
Virtues of The Mind An Inquiry Into The Nature of Virtue and The Ethical Foundations of Knowledge by Linda Trinkaus Zag
CAMBRIDGE
UNIVERSITY PRESS
PUBLISHED BY THE PRESS SYNDICATE OF THE UNIVERSITY OF CAMBRIDGE
The Pitt Building, Trumpington Street, Cambridge CB2 1RP, United Kingdom
Typeset in Palatino
A catalogue record for this book is available from the British Library
Acknowledgments page xi
Introduction xiii
Vlll
Contents
Bibliography 341
Name index 359
Subject index 362
IX
Acknowledgments
Los Angeles
16 September
Xll
Introduction
xm
Introduction
tween intellectual virtue and moral virtue very far. This is too bad
because interest in virtue ethics has blossomed in the last few
decades after a long period of neglect. If a virtue-based ethical
theory has advantages over an act-based theory, it ought to be
illuminating to look closely at this kind of theory for help in
developing the concepts needed in analyzing knowledge and jus-
tified belief. In any case, because the use of the act-based model in
analyzing justification has resulted in an impasse between inter-
nalists and externalists, it may now be a good time to investigate a
new approach.
The focus of the theory I will propose is the concept of intellec-
tual virtue. Unfortunately, philosophers have rarely given the
intellectual virtues much notice, and this neglect is not limited to
contemporary philosophy; it has occurred throughout the entire
history of philosophy. Of course, moral philosophers have exam-
ined virtue at length, but on those few occasions in which they
mention an intellectual virtue, it is almost always the Aristotelian
virtue of phronesis, or practical wisdom, but that has aroused
interest only because of the way Aristotle connects phronesis with
the traditional moral virtues. Other intellectual virtues are usu-
ally ignored altogether. Clearly, a true aretaic approach to epis-
temology requires a theory of virtue that gives the intellectual
virtues their proper place. I will develop here a virtue theory that
is inclusive enough to handle the intellectual as well as the moral
virtues within a single theory, and I will then show how such a
theory can be used in analyzing some of the principal concepts in
normative epistemology, including the concepts of knowledge
and of justified belief.
In the work that follows I will argue that the relationship be-
tween the evaluation of cognitive activity and the evaluation of
acts in the overt sense usually reserved to ethics is more than an
analogous one. I will argue that the intellectual virtues are so
similar to the moral virtues in Aristotle's sense of the latter that
they ought not to be treated as two different kinds of virtue.
Intellectual virtues are, in fact, forms of moral virtue. It follows
that intellectual virtue is properly the object of study of moral
philosophy. This claim is intended, not to reduce epistemic con-
cepts to moral concepts in the way that has sometimes been at-
xiv
Introduction
xvi
— Endless is the search of Truth.
prima facie case for the affirmative, arguing that responsibility for
physical acts is linked with responsibility for mental goings-on,
including beliefs, but this position is controversial, and since it
makes responsibility for cognitive activity derivative from re-
sponsibility for overt acts, it limits the scope of cognitive responsi-
bility in a way I believe seriously underrates the true extent of our
cognitive responsibility. More recently, Christopher Hookway
(1994) has argued that epistemic evaluation ought to focus on the
activity of inquiry rather than on beliefs, and that the ethics of
inquiry will show the proper place of self-controlled personal
responsibility in epistemic evaluation.2 This approach seems to
me a promising one.
The ambivalence about our responsibility for having knowl-
edge or justified belief is dramatically exemplified in the present
impasse between internalism and externalism. The internalism/
externalism distinction can be applied either to theories of justifi-
cation or to theories of knowledge. Clearly, no matter how we
define knowledge and justified belief, they will turn out to be
highly desirable and important cognitive states, and this means
there is something good about these states. But the type of good
intended is not always clear, and this confusion is at the root of
the internalist/externalist debate. Internalists about justification
require that the factors needed for justification be cognitively
accessible and internal to the believer, whereas externalists deny
this, maintaining that the believer need not be aware of the fea-
ture or features that make her belief justified. A parallel distinc-
tion has been made in the account of the normative component of
knowledge.3 Internalists think of knowledge or justified belief as
2
In this connection Kenneth W. Kemp has told me that he thinks that the root of
the difference between me and Aristotle is that the form of intellectual activity
which is my paradigm is that of inquiry, while for Aristotle the paradigm is that
of contemplation. The work that follows may serve to confirm Kemp's
appraisal.
3
There are many different ways of drawing the distinction between internalism
and externalism. Probably the most common one is to define internalism as the
view that all justifying features of a belief be cognitively accessible and internal
to the agent, whereas externalists claim that some justifying features are external
or inaccessible. But it is clear from the discussion of reliabilism, the most popu-
Using moral theory in epistemology
lar version of externalism, that externalists generally believe that the most
important or salient justifying features typically are inaccessible to the agent's
consciousness. This permits hybrid positions according to which some impor-
tant justifying features are and some are not cognitively accessible. See Lau-
rence Bonjour's contribution to Dancy and Sosa 1992, p. 132, for an explanation
of the way this distinction is used in the contemporary literature.
The methodology of epistemology
4
In "Intellectual Virtue in Perspective" (1991, p. 271) Sosa dissociates his use of
the concept of a virtue from the Aristotelian sense. In referring to the faculty of
8
Using moral theory in epistemology
sight, he says, "Is possession of such a faculty a Virtue'? Not in the narrow
Aristotelian sense, of course, since it is not a disposition to make deliberate
choices. But there is a broader sense of Virtue/ still Greek, in which anything
with a function - natural or artificial - does have virtues. The eye does, after all,
have its virtues, and so does a knife. And if we include grasping the truth about
one's environment among the proper ends of a human being, then the faculty
of sight would seem in a broad sense a virtue in human beings; and if grasping
the truth is an intellectual matter then that virtue is also in a straightforward
sense an intellectual virtue." In this passage Sosa seems to treat sight as the
virtue of the eye, but in Plato and Aristotle sight is the function of the eye, not
its virtue. Plato (Republic 352e~353c) and Aristotle (Nicomachean Ethics [hereaf-
ter abbreviated NE] 11.6.1106314-27) both accept an analogy according to
which a person's (or soul's) virtue is related to his function as the eye's virtue is
to the eye's function. The eye's virtue is a trait that enables the eye to see well.
Extending the analogy in the way Sosa desires, we should say that grasping the
truth is a function of the intellect, not its virtue. The virtues would be those
traits whereby the intellect is enabled to grasp the truth well. I thank Charles
Young for noticing this problem in the way Sosa relates the concept of virtue to
that of a function.
The methodology of epistemology
Two theories that come closer to the one I wish to develop are
those of Lorraine Code and James Montmarquet. In her book
Epistemic Responsibility (1987), Code gives a provocative account
of intellectual virtue, stressing a "socialized" approach to epis-
temology, pointing out the connections between epistemology
11
The methodology of epistemology
Persons who desire the truth would desire to have these traits,
says Montmarquet, but they are not simply traits that are truth
conducive in any straightforward way. In fact, epistemic virtues
are virtues even when or if they are not truth conducive. Further-
more, he says, since persons with such intellectual vices as closed-
mindedness or prejudice may not be deficient in the desire for
truth, some of the epistemic virtues function as regulators of that
desire (1993, p. 25). I will question later whether either of these
two claims is true. I will argue that truth conduciveness is an
essential component of intellectual virtues and I will attempt to
ground these virtues in the motivation for knowledge. But the
pertinent point for our discussion of methodological motivation
is that Montmarquet has identified some virtues that operate on
13
The methodology of epistemology
15
The methodology of epistemology
16
Using moral theory in epistemology
17
The methodology of epistemology
Philosophy (1985), and Allan Gibbard also uses this distinction in Wise Choices,
Apt Feelings (1990).
9
Martha Nussbaum takes this position in numerous places. See especially
Love's Knowledge (1990).
10
W. D. Ross is an example of a particularist, nonabsolutist, act-based theorist.
11
Although the objection that morality is not strictly rule governed has been
leveled against strict deontological theories, it has been argued that not even
Kant thought that moral reasoning is algorithmic. See O'Neill 1989, esp. chap.
1.
12
Examples would be Stocker 1976 and Blum 1980.
18
Using moral theory in epistemology
love and a motive of duty conflict, it is not obvious that the latter
ought to override the former. Although it is not impossible for an
act-based ethical theory to account for such intuitions,13 it is much
easier on a virtue approach since the perspective of the individual
agent can easily be worked into the concept of a virtue, whereas
the major act-based theories approach morality primarily from an
impersonal standpoint and can accommodate the distinctive
values of individual persons only with difficulty.
A fourth reason favoring the focus of moral theory on virtues
rather than on acts and principles is that there are virtues that are
not reducible to specifiable acts or act dispositions. Gregory Tri-
anosky (1987) has argued that there are higher-order moral vir-
tues that cannot be analyzed in terms of relations to acts. For
example, it is a virtue to have well-ordered feelings. A person
with such a virtue has positive higher-order feelings toward her
own emotions. Similarly, it is a virtue to be morally integrated, to
have a positive higher-order evaluation of one's own moral com-
mitments. These are virtues that cannot be analyzed in terms of
some relation to right action. Furthermore, although Trianosky
does not say so explicitly, such higher-order virtues are connected
to the virtue of integrity since integrity in one of its senses is the
virtue of having a morally unified self, and it is difficult to see
how such a virtue can be explicated in terms of dispositions to
perform acts of a specified kind.14
The resurgence of interest in virtue ethics in recent philosophy
is certainly not due solely to these four sets of considerations.
Nonetheless, these reasons are important, have generated con-
siderable discussion, and all have analogues in the evaluation of
cognitive activity. In fact, as we shall see, some of them are even
stronger in epistemic evaluation than in moral evaluation.
I have said that contemporary epistemology is belief-based,
just as modern ethics is act-based. Epistemic states are evaluated
in terms of properties of beliefs or belief dispositions, just as
moral evaluations are typically given in terms of properties of acts
13
Again, W. D. Ross comes to mind.
14
I surmise that the difficulty of an act-based theory in accounting for integrity
in this sense is one of the reasons integrity is rarely discussed.
19
The methodology of epistemology
15
See, e.g., Tiles and Tiles 1993; Wisdo 1993; Montmarquet 1993; and Kvanvig
1992, p. 192.
16
This case is also discussed in Wainwright 1996, chap. 3.
21
The methodology of epistemology
24
Using moral theory in epistemology
25
The methodology of epistemology
We might say that in this case what Dennis is like and what he
does is only accidentally connected to his acquiring the truth, and
his cognitive worth is the worth of the perfect computer. This
problem is structurally similar to the problem of the ideal utilitar-
ian moral agent, where the kind of person he is and even what he
does is only accidentally related to the moral evaluation of his
acts. Although Taliaferro is concerned in this example with the
concepts of omniscience and cognitive perfection, his reason for
claiming that standard accounts of omniscience are defective is
just that knowledge implies cognitive power. Clearly, Christopher
has more cognitive power than Dennis, and the analysis of
knowledge, not just omniscience, should account for this.
If Taliaferro is right, his point should hold in the case of ordi-
nary human knowledge as well, although that is not Taliaferro's
interest in this article. Compare, then, the human counterparts of
Dennis and Christopher. Let us suppose that they both believe a
large set of true propositions T. Assume that the differences in
their other beliefs are insignificant since they believe the same
number of true propositions and the same number of false propo-
sitions, and there is no difference in the importance of their re-
spective beliefs. The only epistemic difference between them is
that Christopher believes the propositions in Tby the direct use of
his own perceptual and cognitive powers, whereas Dennis be-
lieves them on the authority of Christopher. Let us assume also
that Christopher is a perfectly reliable authority, so that Dennis's
belief-forming process is just as truth conducive as Christopher's.
According to reliabilism, the normative property of the beliefs of
Dennis is equal to that of Christopher. On the assumption that
this property is also the property that converts true belief into
knowledge, Dennis and Christopher are equal in their knowledge
of T. But intuitively Christopher is superior in knowledge to
Dennis because he possesses greater cognitive power. But since
he does not know more true propositions than Dennis does, his
superiority in knowledge must involve superior quality.
Christopher's superiority might be explicated by a more
detailed description of the case. Perhaps his superiority is due to
the greater clarity or understanding that accompanies the acquisi-
tion of true beliefs by the exercise of one's own powers, but it
27
The methodology of epistemology
kind of theory, a matter that will be examined at the end of Part II,
but we can say at least this much at this stage of the inquiry: It is
an advantage in a theory if it can give a clear and unified account
of the full range of human evaluative levels, both ethically and
epistemically. The notions of virtue, vice, and their intermediate
states are intended to apply to the full range of evaluative levels,
whereas the concepts of right/wrong and justified/unjustified
are not. We will need to see whether a virtue theory can live up to
its promise of handling the full range of levels of evaluation after
such a theory has been developed.
2 DIFFICULTIES IN CONTEMPORARY
EPISTEMOLOGY
31
The methodology of epistemology
32
Difficulties in epistemology
runs into: whereas he that makes use of the light and faculties
God has given him, and seeks sincerely to discover truth, by
those helps and abilities he has, may have this satisfaction in
doing his duty as a rational creature, that though he should
miss truth, he will not miss the reward of it. For he governs his
assent right, and places it as he should, who in any case or
matter whatsoever, believes or disbelieves, according as reason
directs him. He that does otherwise, transgresses against his
own light, and misuses those faculties, which were given
him.20
34
Difficulties in epistemology
35
The methodology of epistemology
36
Difficulties in epistemology
39
The methodology of epistemology
dispute. 26 This is not to say that ethicists have not thought of the
exact analogue of the distinction in epistemology; they just have
not been as much taken by it as have epistemologists.
To see this, consider a brief but interesting argument given by
Hilary Kornblith to the effect that the same distinction is made in
ethics:
40
Difficulties in epistemology
41
The methodology of epistemology
44
Difficulties in epistemology
46
Difficulties in epistemology
47
The methodology of epistemology
49
The methodology of epistemology
The ability to see the way bits of reality fit together is important
in the conduct of our ordinary life, not just in theory construction
and other areas of formal inquiry. In our private life this ability
helps us in understanding another person. We each know in the
propositional sense an enormous number of facts about a person
with whom we are intimately acquainted, yet the knowledge of
such facts does not constitute knowing the person. To understand
a person's motivation or character we often need to be able to pick
from the profusion of information about him stored in our mem-
ory certain facts that become salient in particular contexts. The
juxtaposition of one bit of knowledge with another - say, his
susceptibility to jealousy and fearfulness of other sorts - can pro-
duce insights that extend and deepen our understanding of his
psychic makeup. And, of course, this same ability is crucially
important in our knowledge of ourselves. It is very doubtful that
an epistemology focused on the individual belief can explain the
nature of such understanding if I am right that understanding
involves the comprehension of structures of reality other than its
propositional structure.
I will end this section with a brief mention of another neglected
epistemic value that is the aim of the hedgehog rather than the
fox. That value is wisdom, and its neglect is not trivial. The nature
of wisdom may be elusive, but it is clear that whatever it is,
wisdom is an epistemic value qualitatively different from the
piling up of beliefs that have the property of justification, war-
rant, or certainty. Wisdom is neither a matter of the properties of
propositional beliefs, nor is it a matter of the relations among such
beliefs; it is a matter of grasping the whole of reality. We have
already discussed wisdom in section 1.2 as an example of a per-
sonal value analogous to love and friendship, the importance of
which moves us in the direction of a virtue theory. Wisdom may
or may not be a state distinct from knowledge, just as understand-
ing may or may not be distinct from knowledge. In any case, we
should attempt as much as possible to get a unitary account of our
epistemic life, so even if the goals of knowledge, understanding,
and wisdom are distinct, it is reasonable to expect them to be
interrelated in the way a person designs and conducts her cogni-
Believing and feeling
3 M O R E R E A S O N S T O TRY A V I R T U E
APPROACH: THE RELATIONS BETWEEN
BELIEVING AND FEELING
In the last two sections we have seen reasons for thinking a virtue
approach to epistemic evaluation might be fruitful. One set of
reasons arose within ethics and corresponded to the advantages
of virtue ethics over act-based ethics. The second set of reasons
arose from certain problems within epistemology itself. In this
section I wish to look at some considerations arising from the
philosophy of mind.
The treatment of belief as a psychic state independent of non-
cognitive states is happily nearing its demise. In fact, recently
problems have been raised with treating belief as an identifiable
psychic state of any sort, but I will not get involved with that
dispute.28 My concern in this section is simply to point out that
the psychic states conventionally assigned to epistemologists are
by no means sharply distinguished from the psychic states con-
ventionally assigned to ethicists. More specifically, I will point out
some of the numerous ways states of believing are connected
with feeling states and states of emotion. This by itself is not
sufficient to lead us into a virtue epistemology, but since moral
virtues are thought to govern the emotions while intellectual vir-
tues govern the formation of beliefs, any preexisting interest in
28
See particularly Stich 1983, where Stich argues that the concept of belief is a
useless figment of folk psychology.
The methodology of epistemology
James has numerous examples that show that the emotive state
of excitation and its forms such as desire, dread, and fear affect
belief, and that we need not resort to any new principle of choice
to explain this phenomenon: "Speaking generally, the more a
conceived object excites us, the more reality it has" (1981, vol. 2, p.
935). One of James's examples is a man who has no belief in
ghosts by daylight but temporarily believes in them when alone
at night as he feels his blood curdle at a mysterious sound or
vision. The man's fear leads to his belief. He also cites belief
arising from dread:
A hot flash seems to burn across the brain. Men in these intense
states of mind have altered history, changed for better or worse
the creed of myriads, and desolated or redeemed provinces of
ages. Nor is this intensity a sign of truth, for it is precisely
strongest in those points in which men differ most from each
other. John Knox felt it in his anti-Catholicism; Ignatius Loyola
in his anti-Protestantism; and both, I suppose, felt it as much as
it is possible to feel it. (James 1981, vol. 2, p. 936)29
What's interesting about the movie [/FX] is that it's one of the
fastest movies . . . It's like splinters to the brain. We have 2500
cuts in there, I would imagine. We had 2000 camera setups.
We're assaulting the senses . . . in a sort of new-wave tech-
nique. We admire MTV editing technique and we make no
bones about using it. We want t o . . . get to the subconscious. . .
and certainly seduce the viewer into a new perception of reality
. . . [of ] what occurred in Texas that day.31
54
Believing and feeling
the assault by images at high speed leaves little time for the
viewer to think critically. Whether it works is a matter for empiri-
cal investigation, but on the face of it, it would not be surprising if
it does. Besides raising questions about the voluntariness of be-
lief, to which we will turn in section 4, this example raises very
interesting questions about the causal connections among graphic
perceptual images, the emotional reaction to such images, and the
formation of beliefs.
Let us next look at a homey example of the way in which
believing is not clearly distinguishable from the affective states of
hoping or wanting. Suppose Molly wants to speak to a friend, but
she doubts that he is home; she somewhat disbelieves that he is
home. However, she telephones anyway. After all, the effort of
phoning is well worth the possible gain. Since she strongly hopes
that he is home, she begins to look for a reason to believe it, a
reason to make her hope reasonable. As her mind runs through
reasons why he might be home after all, she finds that the
strength of her assent to the proposition John is home increases
both because she wants to believe it and because it is supported
by the evidence she has considered in her ruminations. And these
two reasons are not independent. At the first ring her hope surges
and with it comes another surge of belief. Yes, she expects he will
answer the telephone. At the second ring both her hope and her
belief start to diminish, and at the third and fourth they diminish
even more. She now begins to think that it is unlikely he is at
home, and with each succeeding ring her belief that he is not
home increases. Her assent to the proposition John is not home
gets stronger and stronger.
This example illustrates two different things. First, it shows
that belief comes in degrees due to degrees of assent, and second,
it shows that belief is closely connected with such states as want-
ing, hoping, and expecting. Wanting and hoping are emotive or,
at least, affective states; expecting and believing are usually
thought not to be. But in the example just given, not only are there
very close causal relations between wanting and hoping on the
one hand and expecting and believing on the other, but it is not at
all easy to distinguish between the emotive state and the cogni-
tive state. Something like this might have been behind Wittgen-
55
The methodology of epistemology
32
Jenefer Robinson (1983) has argued that desires are more basic than emotions.
If so, the primary causal factor underlying both emotions a n d beliefs w o u l d b e
desire, including, presumably, the desire for truth.
33
Charles Young suggested this line from Aristotle.
57
The methodology of epistemology
4 A N O B J E C T I O N TO M O D E L I N G
EVALUATION IN EPISTEMOLOGY ON
E T H I C S : T H E D I S P U T E OVER T H E
V O L U N T A R I N E S S OF BELIEF
36
Richard Swinburne (1986, p . 127) makes such an assumption, b u t so d o m a n y
others.
37
I have been asked this question in conversation as if that were the issue.
62
Dispute over the voluntariness of belief
nition of the freedom of his own will. At that time in his life he
wrote to his father, and what is most interesting for the present
topic is that he indicated in that letter that he understood his
primary exercise of free will to be in a belief rather than in an
overt act:
63
The methodology of epistemology
64
Dispute over the voluntariness of belief
The will is one of the chief organs of belief, not that it creates
belief, but because things are true or false acording to the aspect
by which we judge them. When the will likes one aspect more
than another, it deflects the mind from considering the qualities
of the one it does not care to see. Thus the mind, keeping in step
with the will, remains looking at the aspect preferred by the
will and so judges by what it sees there. {Pensees, trans.
Krailsheimer, sec. II, p. 218)
65
The methodology of epistemology
person has false beliefs that the akratic person does not have and
that make him morally worse than the akratic individual. It takes
bad will to have certain beliefs. Edmund Pincoffs defends this
view well:
If his beliefs are his justification for being dishonest, then he is
blameable for holding a belief that, as he believes, justifies his
being dishonest. For no belief can do this; and he cannot avoid
blame by appeal to one. He is in this sense responsible for his
beliefs, in that if he "chooses what is bad" as a consequence of
them, he must either reject them or admit that they truly repre-
sent him, characterize him. (1986, p. 147)
Now there probably is something bothersome about the Aristo-
telian position that people are blameworthy for their false moral
beliefs, but it is not the lack of voluntariness of beliefs in general
that makes us worry. It is the fact that some people do not seem to
meet the minimal conditions for control, even though they
display vices, and so we may be inclined to pity them and to want
to rehabilitate them rather than to blame them. And the farther
their moral beliefs are from the norm, the more inclined we are to
attribute them to involuntary processes. But this surely cannot be
our standard response to cases of false moral beliefs, because
Pincoffs is right in calling attention to the dependency of our
responsibility for our acts on our responsibility for the beliefs
providing our reasons for acting.
I conclude that we must reject the idea that the only element of
the voluntary in acting is a distinct act of choice that occurs imme-
diately before the act. Acts that follow a process of deliberation
and choice are in a very select category, and if morality applied
only to this class of acts, it would not apply to very much. Thus,
the fact that few, if any, beliefs are the objects of choice does not
threaten the claim that many beliefs are sufficiently voluntary to
be subject to evaluation in the sense of moral evaluation.
I said earlier that beliefs and acts appear to be disanalogous at
both the extreme upper end and the extreme lower end of the
scale of voluntariness. I have considered the extreme upper end
of the scale and have agreed that probably more acts than beliefs
68
Dispute over the voluntariness of belief
are the direct objects of choice, although many acts for which we
are responsible are not in that category. At the lower end of the
scale, however, I believe the difference is largely illusory. Many
bodily acts are as involuntary as typical perceptual or memory
beliefs, such as coughing or wincing. An important difference for
the attribution of evaluative properties to acts and beliefs, how-
ever, is that epistemologists often treat involuntary perceptual or
memory beliefs as paradigms of belief, and even of justified or
rational belief. Sometimes these beliefs are treated as paradigms
of knowledge as well. These cases seem to be disanalogous to acts
as treated in ethics since nobody treats involuntary acts as the
object of moral evaluation, much less as paradigms of right ac-
tion. We will discuss perceptual and memory beliefs in Part III.
Here my concern is not to deny that true perceptual beliefs in
ordinary circumstances are rational or justified, nor do I wish to
deny that they are cases of knowledge. Instead, I want to make
the different point that it is a mistake to think of perceptual beliefs
as paradigm cases of rationality and justifiability. Rationality has
traditionally been understood as the property that makes us most
distinctively human. Perceptual beliefs, however, admit of a
number of different analyses, and on some accounts, a perceptual
belief is no different when held by a human being than when held
by many animals. If so, these beliefs are subhuman or, at least,
subpersonal. To call beliefs that occur without the agency of the
human agent "rational" or "justified" is a jarring use of such
terms, given their historical usages. Moral philosophers would
never use as paradigm cases of right acts such subpersonal invol-
untary acts as sneezing. This is not to say that there is no sense of
"right act" in which sneezing is right, but only that it would
distort the concept of a right act to begin the analysis of Tightness
with such an example as a paradigm case. Similarly, although I
am not denying that there is a sense of "rational" and "justified"
in which ordinary perceptual beliefs are rational and justified, it
does not do justice to the history of the concept of rationality to
lead off the investigation with these beliefs as paradigms.41
41
It is probably true that the main reason epistemologists begin with cases of
perceptual beliefs is their position that more complex and important beliefs
69
The methodology of epistemology
70
Dispute over the voluntariness of belief
71
The methodology of epistemology
43
See Martha Nussbaum's The Fragility of Goodness (1986), for the view that
Aristotle's acceptance of moral luck is to be contrasted with Plato's nonaccep-
tance of moral luck. Assuming that both Plato and Aristotle have virtue theo-
ries, it follows that an acceptance of moral luck is not a distinguishing
characteristic of a virtue theory but only of certain forms of it. For an interest-
ing account of the Aristotelian view on the luck of morality see also Kenny
1988.
72
Why center epistemology on the virtues?
that, I ask readers in the second group to adopt a "wait and see"
attitude about virtue theory. It is likely that contemporary virtue
theory is still too young to be the equal rival of deontological and
consequentialist theories.44 Part of its persuasiveness will un-
doubtedly be its fruitfulness in accounting for a broad range of
human life, but I do not expect to have convinced readers of this
yet. At the end of the book I will again ask the question of whether
virtue theory can do what I am now proposing it has the promise
of doing.
44
See Slote 1992b for a defense of virtue theory as the equal rival of utilitarian
and Kantian moral theories.
75
Part II
A theory of virtue and vice
79
A theory of virtue and vice
80
Types of virtue theories
well but goes on to say that a way of living counts as living well
because it is the way virtuous persons want to live.2
I am interested in one form of good-based and one form of
agent-based theory. In a common form of good-based theory the
virtues are explained either as constituents of the good life or as
means to the good life, where the good life is identified with
happiness or the Aristotelian concept of eudaimonia. I call these
theories happiness-based. Rosalind Hursthouse (1991) has a theory
that is happiness-based. She defines a virtue as "a character trait a
human being needs to flourish or live well" (p. 226). Hursthouse
explicitly maintains the derivative character of right action and
the foundational character of eudaimonia:
for the sake of that other thing, and therefore we call final without qualification
that which is always desirable in itself and never for the sake of something else.
"Now such a thing happiness, above all else, is held to be; for this we choose
always for itself and never for the sake of something else, but honour, pleasure,
reason, and every virtue we choose indeed for themselves (for if nothing re-
sulted from them we should still choose each of them) but we choose them also
for the sake of happiness, judging that by means of them we shall be happy.
Happiness, on the other hand, no one chooses for the sake of these, nor, in
general, for anything other than itself" (NE 1.7.1097330-^6).
82
Types of virtue theories
cepts of the good and of a right act from the concept of a virtue,
the virtues must be such that they are capable of having such a
function. One advantage of this type of theory is that many phi-
losophers like the concept of virtue but are suspicious of teleology
(e.g., Pincoffs 1986).
Clearly, the motivation-based approach to ethics requires some
way of determining what makes a motivation a good one and
what makes a trait a virtue. This is a daunting task, but perhaps
no harder than the task of determining what makes a state of
being a state of eudaimonia. One way of doing it, and to my mind a
forceful one, is to appeal to experience. Many of us have known
persons whose goodness shines forth from the depths of their
being, and if we have not met them in person, we may have met
them in literature, such as Dorothea Brooke in George Eliot's
Middlemarch. Such an experience raises the question of whether
we can know a person is good before we investigate her behavior,
observe the outcome of her acts, or even see how her traits pro-
duce in her a state, such as happiness, which we determine to be
good in advance. It is difficult to know the extent to which we can
do this because in actual practice we already have and are used to
using such concepts as a happy life and a right act before having
the experiences I have described. Still, I believe it is possible that
we can see the goodness of a person in this rather direct way. She
may simply exude a "glow" of nobility or fineness of character, or
as I have occasionally seen in a longtime member of a contempla-
tive religious order, there may be an inner peace that can be
perceived to be good directly, not simply because it can be ex-
plained on the theoretical level as a component of eudaimonia. If
we then attempt to find out what it is about such a person that
makes him good, we may be able to identify that goodness as
involving certain feelings or motivations such as feelings of com-
passion or of self-respect or of respect for others, or motives of
benevolence, sympathy, or love, all of which are components of
virtues. In each case we would not determine that his love, com-
passion, or benevolence is good because of its relation to anything
independently identified as good. We would simply see that these
feelings or motivations are the states whose goodness we see in
him. Alternatively, we may focus our attention directly on the
83
A theory of virtue and vice
87
A theory of virtue and vice
88
The nature of a virtue
gests that we should not be too hasty to conclude either that there
is no commonality among the virtues recognized throughout his-
tory or that the commonality is extensive enough to render the
exceptions insignificant. The evidence is neither overwhelmingly
in favor of a common substantive concept nor overwhelmingly
against it. However, the fact that writers on virtue almost always
begin by identifying a list of typical virtues suggests that the
pretheoretical notion of virtue includes the idea that certain par-
ticular traits are among the virtues. An analysis of virtue is hope-
less, I believe, unless we can assume that most of a selected list of
traits count as virtues and do so in a way that is not strictly culture
bound.
The foregoing considerations lead me to propose that an ac-
count of virtue should contain the following features: Virtue is an
excellence; virtue is a deep trait of a person; those qualities that
have appeared on the greatest number of lists of the virtues in
different places and at different times in history are, in fact, vir-
tues. These qualities would probably include such traits as wis-
dom, courage, benevolence, justice, honesty, loyalty, integrity, and
generosity. Some virtues are intellectual, others are moral, and
some may be neither intellectual nor moral. In any case, there is
nothing in either the most common lists of virtues or in the the-
oretical treatment of virtue that requires us to think of the word
"moral" in "moral virtue" as redundant. Finally, the account of a
virtue should aim for a high degree of theoretical significance
combined with practical usefulness. This would include atten-
tiveness to the way the concept of virtue is used in the history of
philosophy, as well as the way it is used in contemporary virtue
ethics. In what follows I will attempt an account of virtue that is
faithful to these constraints.
In reply he says:
We must, however, observe that as accidents and non-
subsistent forms are called beings, not as if they themselves had
being, but because things are by them, so also are they called
good or one, not by some distinct goodness or one-ness, but
because by them something is good or one. So also virtue is
called good, because by it something is good.
90
The nature of a virtue
because it makes its possessor good. I will not dwell on the on-
tological point even though it is a fascinating one, but want to call
attention to the fact that Thomas's position in this passage stresses
the idea that what is good about virtue is that it makes its possessor
good. This is interesting because it is not only inner directed but is
clearly a claim about the way in which a virtue contributes to the
admirability of its possessor. In addition, however, Aquinas
thought that virtue is connected with external good: "the Philoso-
pher says: 'Virtue is that which makes its possessor good, and his
work good likewise'" (STI-II, q. 55, a. 3). Aquinas explicitly says
that one cannot make bad use of virtue, and he quotes ap-
provingly what he calls "the definition usually given of virtue as
'a good quality of the mind, by which we live rightly, of which no
one can make bad use, which God works in us, without us'" (ST
I-II, q. 55, a. 4).5 Virtue is, then, both a good-making quality of a
person and a quality that cannot lead to bad use. Presumably
"bad use" means wrongful behavior. This could reasonably be
taken to imply that virtue is good for the world. So Aquinas's
position in this passage seems to be that virtue makes its pos-
sessor good and is good for the world, although he takes no
position on whether virtue is good for its possessor.
Both claims are controversial and deserve some attention.6 Let
us begin with the question of whether virtue is always a good-
making quality of a person (i.e., it makes its possessor admirable)
and then turn to the question of whether it is always good for the
rest of the world.
On the face of it the courage of a Nazi soldier makes him worse
overall than if he were cowardly. Gregory Trianosky (1987) in-
vents a compassionate but biased judge whose compassion for
5
In the translation by the Dominican Fathers it is noted here that this definition is
close to that of Peter Lombard, Sent. II, d. 27, chap. 5, but the form of the
definition seems to come from Peter of Poitiers, Sent. Ill, chap. 1 (PL 211,1041).
Aquinas qualifies the claim that one cannot make bad use of a virtue in the
reply to objection 5. There he says that one can make bad use of a virtue taken
as an object, as when we have an evil attitude toward virtue, but "one cannot
make bad use of virtue as principle of action, so that an act of virtue be evil."
6
Rorty (1988, pp. 314-29) denies that a person is always made better by virtue.
A theory of virtue and vice
the victims of crime makes him even more unfair than he would
have been without the compassion and, presumably, worse over-
all because of it. The idea in each case is that all other things being
equal, the addition of courage to the Nazi's character and the
addition of compassion to the judge's character result in each of
them becoming morally worse.
One response to these cases is to say that the traits exhibited by
the Nazi and the judge are not courage and compassion respec-
tively. Another is to say that these traits are courage and compas-
sion, but courage and compassion are not always virtues. A third
is to say that the Nazi and the judge do have the traits of courage
and compassion, and these traits in these cases are virtues, but
virtues are not good making in every instance. A fourth response
is to say that the traits exhibited by the Nazi and the judge are
courage and compassion and that courage and compassion are
always virtues and are always good to have, but the good-making
properties of virtues and the bad-making properties of vices do
not always add up arithmetically to yield a rating of the agent's
overall goodness. I will defend the fourth response.
I have nothing to say against those who take the first response.
If the traits of the Nazi and the judge are not courage and compas-
sion respectively, then these cases are no challenge to the position
that virtues like courage and compassion always make their pos-
sessors good. The second response is more interesting since it is a
challenge to the claim that certain specific traits always make
their possessors good, though not a challenge to the claim that
virtues always make their possessors good. Philippa Foot implies
a position of this kind when she suggests that courage does not
always act as a virtue (1978, p. 15). This position seems to me to
raise some serious difficulties about the identity of traits. If the
virtuous trait of courage is distinct from the nonvirtuous trait of
courage, this difference can have nothing to do with motivations,
beliefs, attitudes, or dispositions to act. Instead, a trait would
need to be identified by the way acting on such a trait interacts
with acting on certain other traits. Alternatively, if the virtuous
trait of courage is the same trait as the nonvirtuous trait of
courage, it follows that whether a trait is a virtue or a vice is an
92
The nature of a virtue
7
See Watson 1984, n. 6, for an argument that a virtue and a fault can be the same
trait. Watson admits he begs some questions about trait identity in that
argument.
93
A theory of virtue and vice
The argument that courage makes the Nazi worse and that compassion makes
the judge worse may be examples of what Shelly Kagan (1988) calls contrast
arguments that commit "the additive fallacy/' The moral is that we should not
expect to be able to identify the separate and independent contributions of
individual factors in, say, the evaluation of a person's character by contrasting
cases that differ in a single factor. Such factors may make their contribution as
clusters or in some other way. So even if the courageous Nazi is worse overall
than the cowardly Nazi, we cannot infer from that that it is courage that made
the Nazi worse. I thank Charles Young for calling my attention to Kagan's
article.
94
The nature of a virtue
Francis Bacon says, "In fact, as is clear, the more active and faster a
man is, the further astray he will go when he is running on the
wrong road" (Novum Organum, trans, and ed. Urbach and Gib-
son, I.61). A person who is intellectually courageous but lacks
open-mindedness may be led further astray from the truth by his
courage than he would have been without this virtue. Just as the
biased but compassionate judge is led to perform more wrongful
acts because of the combination of his unfairness and compassion
than he would have if he had not been compassionate, so too a
person might be led into more false beliefs by the combination of
her closed-mindedness and intellectual courage than she would
have been if she had not had the courage.
My position on the epistemic case is parallel to my position on
the case of the judge. It is better for a person to be intellectually
courageous even when her courage leads her to maintain beliefs
in the face of adversity that ought to be given up because they
were formed out of narrow-mindedness. She is wrong in her
beliefs, and even more wrong than she would have been if she
had not been intellectually courageous. Still, I propose that she is
closer to a high level of epistemic status with the courage than
without it.
Although a virtue is always a good thing for a person to have,
then, there is a complication in that equal degrees of a virtuous
trait are not always associated with equal degrees of internal good
in the agent. There are individual differences in the degree to
which a virtue makes an agent good. This point was noticed by
Aristotle in his discussion of the sense in which virtue is a mean
when he says that the mean is relative to a person (NE II.6.no6b).
So what counts as the mean between foolhardiness and cowar-
dice will differ from one person to another, just as what counts as
the right amount of food to eat differs from person to person.
When we look at the intellectual virtues, we see the same phe-
nomenon. Some persons are intellectually successful by careful
plodding, whereas others do their best work when they indulge
in exuberant intellectual impetuousness. These individual
differences seem to strain the distinction between virtue and vice
since the mean relative to a particular individual may lean very
far in the direction of one vice or its contrary. Still, it is fair to say
96
The nature of a virtue
I have not argued that virtues are good for their possessors in
the sense intended by Nietzsche, but it is worth considering
whether they are. Plato, of course, thought so, and in contempo-
rary philosophy Alasdair Maclntyre's well-known account of vir-
tue makes the benefit of a virtue to its possessor intrinsic to its
definition. Maclntyre defines virtue as "an acquired quality the
97
A theory of virtue and vice
98
The nature of a virtue
99
A theory of virtue and vice
100
The nature of a virtue
Even though the world is worse off as a whole because the unfair
judge is compassionate than it would be if he lacked compassion,
compassion is still a virtue in this case and is still an excellence
because the world is closer to a state of well-being if he is compas-
sionate than if he is not. His compassion makes his own character
closer to an admirable one and the world itself closer to a desir-
able one than either would be without the compassion. To con-
sider this problem more closely, consider a world full of persons
like the unfair but compassionate judge, and compare it with a
world full of persons who are both unfair and uncompassionate.
Which world is closer to the world in which persons who are both
fair and compassionate predominate? There are, of course,
different ways to measure closeness. If the normative differential
of these worlds is measured solely in terms of the Tightness/
wrongness of the acts of the judges and their external effects, then
the second world would be rated closer to the ideal for the rea-
sons Trianosky gives. But if the normative differential is mea-
sured in terms of how much a world has to be changed to turn it
into an ideally desirable world, then the first world would be
rated closer to the ideal and, hence, higher in value. That is be-
cause in the first world judges only need to learn how to be fair; in
the second world, judges need to learn both fairness and compas-
sion, and that is a much bigger project. I conclude, then, that
virtue is related to good, not by invariably increasing the good-
ness of its possessor and goodness for the world, but by invaria-
bly making its possessor closer to a high level of admirability and
the world closer to a high level of desirability.
To summarize 2.2, virtue is related to good in a number of
ways. First, a person is good through the possession of virtues.
Because a person has a virtue, both the possessor of the virtue and
the world are closer to a state of goodness than either would be
otherwise. A virtue usually results in an actual increase in the
possessor's moral worth, and it usually results in an actual in-
crease in the good for the world. That virtues are excellences that
are both praiseworthy in their possessors and beneficial to others
is an aspect of the idea of virtue that ought to count as a constraint
on any acceptable account. The question of the conceptual and
101
A theory of virtue and vice
1O2
The nature of a virtue
104
The nature of a virtue
105
A theory of virtue and vice
virtue as follows: "If, then, the virtues are neither passions nor
faculties, all that remains is that they should be states of character
[hexis\" (NE V.2.no6aio-n). We can add from the foregoing
discussion that they are states of excellence that develop over
time in a person, who could have developed the contrary state.
This suggests that intrinsic to the nature of virtue is the way in
which it is acquired. It also suggests that a virtue may be some-
thing like a skill. A discussion of these points will be taken up in
2.4 and 2.5.
107
A theory of virtue and vice
20
Robert Roberts (1984, p. 129) argues that a certain class of virtues, those of self-
control, may involve technical abilities. These virtues include courage, tem-
perance, and patience. But since Roberts also argues that the virtues in this
class are skills, it is not surprising that he maintains that they invoke tech-
niques. I claim, in contrast, that the skills of self-control are not themselves
virtues, although persons may acquire these skills in the process of acquiring
the virtue. A person who has the virtue of patience or courage, I would say, has
gone beyond the need for the technical skills a person who has trouble main-
taining his patience or his courage requires.
108
The nature of a virtue
Again, the case of the arts and that of the virtues are not
similar; for the products of the arts have their goodness in
themselves, so that it is enough that they should have a certain
character, but if the acts that are in accordance with the virtues
have themselves a certain character it does not follow that they
are done justly or temperately. The agent also must be in a
certain condition when he does them; in the first place he must
have knowledge, secondly he must choose the acts, and choose
them for their own sakes, and thirdly his action must proceed
from a firm and unchangeable character. These are not reck-
oned as conditions of the possession of the arts, except the bare
knowledge; but as a condition of the possession of the virtues
knowledge has little or no weight, while the other conditions
count not for a little but for everything, i.e., the very conditions
that result from often doing just and temperate acts.
Both Wallace (1978) and Broadie (1991, pp. 82-6) interpret Ar-
istotle's use of the term "techne," translated as "art," as meaning
skill. Wallace sees Aristotle as claiming that skills and virtues are
dissimilar in that it is enough that an act meet certain standards of
proficiency and that it be done with knowledge in order to exhibit
a skill, but that is not enough to exhibit a virtue. The conditions
for possessing a virtue include additional qualities of the agent's
will and enduring character.
Broadie extracts two additional grounds for the distinction be-
tween skills and virtues from this passage. Take first her transla-
tion of no5a2o-8:
This last point is odd if it means that the product's being in the
right condition is sufficient evidence of skill in the (immediate)
producer, since Aristotle has just said that the product could be
correct through chance or someone's instruction. Instead he
must mean that we are satisfied with things which are normally
produced by art or skill provided they are up to standard, even
when they were produced by someone without skill. If we
assess what such a doer has done by what he has made, we can
say that what he has done is good. The lack of skill implies no
defect in what he has done on this occasion, and it might rea-
sonably be claimed that the skill is of value only because who-
ever possesses it is more likely to produce acceptable articles.
Aristotle's point is that it is not like this with virtue and right
actions (hence, he implies, virtue is too different from skill for
one to be justified in drawing conclusions about virtue from
premisses about skill). (1991, p. 83)
nor vicious. If, as Aristotle says, each virtue has two contrary
vices, the point is even stronger since clearly a skill does not have
two contraries.22
Second, the behavior consisting in the exercise of a skill is not
essentially connected to anything valuable, whereas it is in the
case of a virtue. If a skill has value, that is because of features of
the situations in which it is used extrinsic to the skill itself. On the
other hand, a virtue is intrinsically valuable.
Finally, let me illustrate the distinction by example. If we com-
pare lists of virtues and skills, I think we can see that there are
both moral virtues and moral skills, and intellectual virtues and
intellectual skills. Skills serve virtues by allowing a person who is
virtuously motivated to be effective in action. Consider first the
relation between moral virtues and moral skills. Some examples
of moral virtues that have accompanying skills are compassion,
moral wisdom, fairness, self-improvement, generosity, and
courage. Examples of skills that a person with each of these vir-
tues might have are as follows:
Compassion skills: knowing what to say to the bereaved
Moral wisdom skills: being able to talk a young person into staying
in school or getting out of a street gang
Fairness skills: knowing how to fairly evaluate student papers or
papers submitted to a professional journal
Skills of self-improvement: knowing how to develop a certain talent
Skills of generosity: being effective in giving to others (e.g., in a way
that does not embarrass them)
Skills of courage: knowing how to stand up to a tormentor
Typically, moral virtues have many skills associated with them,
although there may be moral virtues that have no corresponding
skills. Perhaps humility and sincerity are in this category, as well
as the Christian virtues of faith and hope. Some moral skills may
22
The closest view to the one I have just given is Wallace's point that "[n]ot all
excellences have opposite action-states that are bad, however, and this is true
of skills" (1978, p. 50). Aristotle's point about a virtue as a mean between two
contrary vices is meant to apply only to moral virtues, not to intellectual
virtues. I will argue later that many intellectual virtues typically have two
contrary vices as well.
A theory of virtue and vice
Intellectual virtues
• the ability to recognize the salient facts; sensitivity to detail
• open-mindedness in collecting and appraising evidence
• fairness in evaluating the arguments of others
• intellectual humility
• intellectual perseverance, diligence, care, and thoroughness
• adaptability of intellect
• the detective's virtues: thinking of coherent explanations of the
facts
• being able to recognize reliable authority
• insight into persons, problems, theories
• the teaching virtues: the social virtues of being communicative,
including intellectual candor and knowing your audience and
how they respond
Intellectual skills
• verbal skills: skills of speaking and writing
• perceptual acuity skills, e.g., fact-finding skills; these are the skills
of the detective or the journalist
• logical skills: skills of performing deductive and inductive reason-
ing, the ability to think up counterexamples
• explanatory skills, e.g., the ability to think up insightful analogies
• mathematical skills and skills of quantitative reasoning
• spatial reasoning skills, e.g., skills at working puzzles
• mechanical skills, e.g., knowing how to operate and manipulate
machines and other physical objects
Many of the intellectual skills listed here are not closely associ-
ated with specific intellectual virtues, but some are. Perceptual
acuity skills, for example, probably are connected with the virtue
of sensitivity to detail and with intellectual care and thorough-
ly
The nature of a virtue
116
The nature of a virtue
that a virtue is identical with a habit if, as Gilbert Ryle has argued
(1949, pp. 42-4), habits are only one sort of "second nature," and
higher-grade dispositions such as skills and virtues are not purely
automatic but exhibit intelligence in their operation. Both pure
reflex habits and virtues are acquired through a process of repeti-
tion over time, and it should be useful to consider whether it is
intrinsic to the nature of a virtue that it be acquired in this way, or
whether the fact that a virtue requires training and develops over
time is just an accidental feature of it. Can we imagine acquiring a
virtue by some other process than by the training Aristotle
describes?
One way to test the importance of the acquisition process for
the nature of virtue is via Robert Nozick's imaginary transforma-
tion machine. This tantalizing machine is not as well known as
Nozick's experience machine, but like the latter, Nozick proposes
it in order to help us determine what really matters to us. If the
thought experiment works, it can also help us determine what it is
about the way a virtue is acquired that makes it a virtue.
24
I thank David Blake for this point.
A theory of virtue and vice
121
A theory of virtue and vice
122
The nature of a virtue
26
Ernest Dimnet was a French abbe who wrote in English and lectured in En-
glish at major universities in the United States where he was extremely popu-
lar with American audiences in the thirties. Besides The Art of Thinking, his
most popular books were the sequel, What We Live By, and a two-volume
autobiography, My Old World and My New World. Earlier Dimnet had become
embroiled in the modernist dispute, and after one of his books was placed on
the Index of forbidden books by the Catholic Church, he gave up writing on
theological matters. A major theme of The Art of Thinking is the importance of
originality and the stifling effect of habit and repetition: "Life, including such -
apparently - helpful influences as education and literature, destroys this ten-
123
A theory of virtue and vice
have made a sudden break with past habits of thought and action,
and it is precisely their genius that permitted the break. Perhaps
moral conversion can be understood equivalently as a type of
creative genius in the moral life. If so, this might permit us to
resolve the problem just addressed of how to evaluate the charac-
ter of persons who undergo sudden conversions or suddenly
perform acts that are stupendously good and also seemingly out
of character. I said earlier that the concept of a virtue and of moral
character will not permit cases of the instantaneous development
of a virtue or of an instantaneous change in character. But if there
are moral virtues that are the analogues of artistic creativity or
originality, it might be that even before the conversion, the moral
convert possessed a virtue that gave him a cast of mind that made
him ripe for insights and able to embrace them with enthusiasm.
The existence of the virtues of creativity is a problem for the
thesis that virtue must be acquired by habituation, but I will go
farther and offer the conjecture that the ability to resist habit in
general may actually be a distinguishing mark of creative people.
This exacerbates the problem that these virtues pose for the habit-
uation thesis since they are not simply an exception to the thesis;
their existence indicates a difficulty in the analysis of the other
virtues. I am suggesting that it is possible that there is something
different about even the justice, compassion, generosity, and
courage of highly original and creative persons than these same
traits in other individuals. It is possible, of course, that a person is
original only in his art or profession, but I believe that there are
people who are original in almost everything they do. It follows
that there is a strain in virtue, at least as it is exemplified in some
persons, that resists the Aristotelian account of virtue acquisition.
One response to this problem, of course, is to exclude the
qualities of originality and creativity from the class of virtues, but
in spite of the greater theoretical economy of such a response, I
dency [originality], as April frost kills blossoms, and imitation, ignoble confor-
mism, takes the place of originality. Mankind is like Herculaneum—covered
over with a hard crust under which the remains of real life lie forgotten. Poets
and philosophers never lose their way to some of the subterraneum chambers
in which childhood once lived happy without knowing it. But the millions
know nothing except the thick lava of habit and repetition" (p. 88).
124
The nature of a virtue
126
The nature of a virtue
sions. Von Wright says the virtues are all forms of self-control; the
role of a virtue is to counteract, eliminate, rule out the obscuring
effects that emotion may have on our practical judgment (1963, p.
147). It is interesting to note that this assumes that the effects of
emotion are negative. One gets the impression that if a person had
no emotions at all, there would be no problem with her practical
judgment. In the case of every specific virtue, there is some spe-
cific passion that the possessor of that virtue has learned to mas-
ter, and that is why the various virtues are so many forms of self-
control. According to von Wright, then, the goodness of the vir-
tues resides in their capacity to protect us from harm, not in their
capacity to supply us with some good.
Foot's position is similar to that of von Wright in that she says
that virtues are corrective, standing at a point at which there is
some temptation to be resisted or deficiency of motive to be made
good,28 and for this reason she does not include self-love as a
virtue since self-love, she says, is automatic (1978, p. 8). But leav-
ing aside the negative attitude toward emotion, Foot's connection
of virtue with motive seems to me to be right. Motives are impor-
tant for our study of virtue because they typically are forms of
emotion and are action-guiding. Both features have been identi-
fied by traditional writers on the virtues as important compo-
nents of virtue. I suggest that the concept of a motive is the place
at which we can see the true connection between virtues and
emotions or feelings.
129
A theory of virtue and vice
large area in between are emotions of various sorts. These are the
motives that I think are the most interesting for ethics and that are
closely connected with virtues. I will not get involved here in
debating whether all motives are forms of emotion or whether
only most of them are. The ones I will use in my examples of
virtues are all emotions. This does not mean that we have names
for all of them; in fact, we do not. In many of my examples, I begin
with the preanalytic notion of a particular virtue - say,
benevolence - and then claim that it has a motivational compo-
nent that is a type of action-directing emotion. I have no position
on the question of whether emotions are states whose phenome-
nological content is essential. Although it is generally true that
there is something that it is like to have a certain emotion, it is
possible, even likely, that we do not always feel our emotions. We
learn to identify our emotions partly by their feels and partly by
their associated circumstances. So since I will often be unable to
identify a name for the emotion itself, I may have to refer to it
indirectly as, for example, benevolent feelings or the emotion that
is characteristic of acts of courage. I will assume that when the
reader imagines such cases, she imagines being in certain emo-
tional states and is able to do so even if she and I are unable to
identify the emotion by name.
I realize that sometimes the term "motive" is used merely to
call attention to a desired end without any specification of an
emotion or emotion disposition underlying it, as when we say
that someone's motive in writing a book was to get tenure. This
way of speaking calls our attention to the fact that a motive aims
to produce a state of affairs with a certain character. Identifying a
motive with the mere aim to produce some state of affairs is
inadequate, however, for we generally feel that we have not un-
derstood what a person's motive is just by understanding at what
he aims. I may know that Booth aimed at the death of Lincoln
without knowing his motive in assassinating the president. For
the same reason, it is problematic to identify a motive with a
desire, since I do not know any more about Booth's motive if I
know only that he desires Lincoln's death. So my desire for a
lighter teaching load is sufficiently identified by identifying what
it is I want, but my motive in trying to lighten my teaching load
130
The nature of a virtue
30
See Leeper 1970, pp. 152-5, for the view that emotions are motives and that
they usually operate at persistent, low levels of intensity. Leeper writes, "At
such low levels, emotions do most of their work without the individual's
having any notable thought of being motivated, because the emotional pro-
cesses tend to be experienced as objectified or projected as perceptions of the
situation. They are usually not experienced as something special within
oneself" (p. 152). Leeper also says that excessive attention to the stronger,
flashier emotions has distorted the picture of how emotion affects our be-
havior (pp. 155-6).
132
The nature of a virtue
133
A theory of virtue and vice
partly account for a virtue's deep and lasting quality, one that in
part defines a person's identity and that leads us to think of her as
responsible for it. This means that typically a virtue is acquired
through a process of habituation, although the virtues of
creativity may be an exception.
Third, a virtue is not simply a skill. Skills have many of the
same features as virtues in their manner of acquisition and in
their area of application, and virtuous persons are expected to
have the correlative skills in order to be effective in action, but
skills do not have the intrinsic value of virtues. Other differences
have been enumerated in 2.4.
Fourth, a virtue has a component of motivation. A motivation
is a disposition to have a certain motive, and a motive is an
emotion that initiates and directs action to produce an end with
certain desired features. Motivations can become deep parts of a
person's character and provide her with a set of orientations to-
ward the world that emerge into action given the appropriate
circumstances. A motivation is best defined, not as a way of act-
ing in circumstances specifiable in advance, but in terms of the
end at which it aims and the emotion that underlies it. The easiest
way to identify a motivation is by reference to the end at which it
aims, but it also involves an emotion disposition, and that is
harder to identify by name.
This brings us to another important feature of virtue: "Virtue"
is a success term. The motivational component of a virtue means
that it has an end, whether internal or external. A person does not
have a virtue unless she is reliable at bringing about the end that
is the aim of the motivational component of the virtue. For exam-
ple, a fair person acts in a way that successfully produces a state
of affairs that has the features fair persons desire. A kind, compas-
sionate, generous, courageous, or just person aims at making the
world a certain way, and reliable success in making it that way is a
condition for having the virtue in question. For this reason virtue
requires knowledge, or at least awareness, of certain nonmoral
facts about the world. The nature of morality involves, not only
wanting certain things, but being reliable agents for bringing
those things about. The understanding that a virtue involves is
necessary for success in bringing about the aim of its motivational
136
Intellectual and moral virtues
Again, since this effort of the mind, by which the mind, in so far
as it reasons endeavors to preserve its being, is nothing but the
effort to understand., it follows., that this effort to understand is
the primary and sole foundation of virtue, and that. . . we do
not endeavor to understand things for the sake of any end, but,
on the contrary, the mind, in so far as it reasons, can conceive
nothing as being good for itself except that which conduces to
understanding. (Ethics, Pt. IV, prop. 26; parenthetical references
removed)
138
Intellectual and moral virtues
139
A theory of virtue and vice
140
Intellectual and moral virtues
straight may be, though the straight is not white, except inci-
dentally, and it is not essentially the same. (11.1.1219^27-36)
In this passage we are cautioned not to take the claim about the
soul's parts as a strong ontological claim, which, he says, is irrele-
vant to his concerns. Nonetheless, Aristotle's division of the soul
into a thinking (rational) and a feeling (nonrational or partly ra-
tional) part amounts to more than merely observing that thinking
is one thing, feeling another.32 In fact, as we will see, Aristotle's
32
In Michael Woods's commentary on this passage, he says the following: "Aris-
totle here dismisses the question whether the soul has parts in a strong sense of
'part/ as irrelevant to his present concerns. This is in line with the meth-
odological doctrine that ethics should concern itself with metaphysical ques-
tions only in so far as they are relevant. (Compare 1214312-14.) This passage is
only partly parallel to the comparable passage in E.N. The two examples given
here seem to be different from each other, and the second seems more apt for
Aristotle's purposes. What Aristotle seems to be contrasting with the case of
two physically separate parts is the case when there are two things
distinguishable only in thought. If so, Aristotle, in speaking of the rational or
desiring part of the soul, ought to regard himself as committed only to the
existence of certain capacities. In fact, however, when Aristotle speaks of parts
of the soul, here and in the psychological works, the structure is represented as
explaining the various capacities that are to be found, and thus as not simply
reducible to them. Here when he postulates that the soul has two parts he
seems to mean more than that the soul may be considered as a source of both
rational and non-rational behavior" (Eudemian Ethics [hereafter abbreviated
EE], trans. Woods, pp. 102-3).
Aquinas's commentary on the related passage in the Nicomachean Ethics
stresses the division of the parts of the soul here discussed as based on a
difference in two kinds of rationality, one by nature, the other by participation:
'Then at 'Virtue is divided/ he [Aristotle] divides virtue according to this
difference in the parts of the soul. . . . Since human virtue perfects the work of
man which is done according to reason, human virtue must consist in some-
thing reasonable. Since the reasonable is of two kinds, by nature and by
participation, it follows that there are two kinds of human virtue. One of these
is placed in what is rational by nature and is called intellectual. The other is
placed in what is rational by participation, that is, in the appetitive part of the
soul, and is called moral. Therefore, he says, we call some of the virtues
intellectual and some moral" (trans. Ross, I.L.XX.C243).
In De Anima Aristotle rejects the view that the soul has parts in a quantita-
tive sense of "part." The whole soul is present in all bodily parts: "all the same,
141
A theory of virtue and vice
142
Intellectual and moral virtues
143
A theory of virtue and vice
pleasure and pain. For, with any state of the soul, those things
whose nature it is to make it better or worse are the things to
which the nature of the state relates and with which it is con-
cerned. It is on account of pleasures and pains that we call men
bad, for pursuing or avoiding them as they should not, or those
they should not. That is why everyone actually defines virtues
in an off-hand manner as being insusceptibility and lack of
disturbance in the sphere of pleasures and pains, vices in op-
posite terms. (EE 11.4.1221^36-122235)
144
Intellectual and moral virtues
145
A theory of virtue and vice
Plato recognized the need for natural feeling and moral recti-
tude in the apprehension of truth, particularly in moral matters,
and gave a dramatic argument for their power in a letter:
One final problem with dividing the moral from the intellec-
tual virtues on the grounds that the former handle feeling states
and the latter handle thinking states is that there are states that are
actually blends of thought and feeling. Curiosity, doubt, wonder,
and awe are states of this kind, each of which can either aid or
impede the desire for truth. Curiosity is interesting because both
Augustine and Aquinas call curiosity a vice, whereas it would be
much more common these days to think of curiosity as valuable.
Feelings are involved in intellectual virtues, and intellectual
virtues are involved in handling feelings, but their operation
shows how blurry the distinction between intellectual and moral
virtues really is. Intellectual prejudice, for example, is an intellec-
tual vice, and the virtue that is its contrary is fair-mindedness, but
clearly we think of prejudice as a moral failing and fair-
mindedness as a morally good quality. It is possible that the intel-
lectual form of prejudice and the moral form are the same vice,
and the same point could apply to other cases in which an intel-
lectual trait has the same name as a moral trait, such as humility,
autonomy, integrity, perseverance, courage, and trustworthiness.
William James has said in "The Sentiment of Rationality" (1937)
that faith is the same virtue in the intellectual realm as courage is
in the moral realm: "Faith is the readiness to act in a cause the
prosperous issue of which is not certified to us in advance. It is in
148
Intellectual and moral virtues
39
Aristotle says: "The man who does not rejoice in noble actions is not even
good; since no one would call a man just who did not enjoy acting justly, nor
any man liberal who did not enjoy liberal actions; and similarly in all other
cases'7 (NE 1.8.1099316-19).
151
A theory of virtue and vice
40
Obtuseness in the moral sense is the inability to recognize the moral relevance
of certain details in a situation in which moral decision is required. Moral
theorists who put very strong emphasis on moral principle and little emphasis
on the relevance of variations in circumstances have sometimes been accused
of overlooking this vice. As I see it, the point can be extended to other forms of
obtuseness. A person who fails to see the relevance of details in the weighing
of evidence for a hypothesis or in the judgment of the reliability of an authority
would also be guilty of obtuseness, but her obtuseness is intellectual rather
than moral.
152
Intellectual and moral virtues
James says that the contrary psychological state, the state of men-
tal rest, can also be pathologically exalted:
153
A theory of virtue and vice
seem more utterly what they are, more "utterly utter" than
when we are sober. This goes to a fully unutterable extreme in
the nitrous oxide intoxication, in which a man's very soul will
sweat with conviction, and he be all the while unable to tell
what he is convinced of at all. ("Belief," para. 3)
154
Intellectual and moral virtues
155
A theory of virtue and vice
160
Intellectual and moral virtues
161
A theory of virtue and vice
may be one of the cases in which there is really only a single virtue
that operates in both the moral and the intellectual spheres.
The virtue of integrity is one of the most important of all vir-
tues and is also one of the least often discussed. Integrity in one of
its senses involves wholeness or a unity of the self. As we saw
earlier (Part I, sec. 1.2), integrity is a higher-order virtue, and it
shows more than any other example the need for an account of
the virtues that recognizes the functional unity of the person.
Moral integrity includes having a positive moral evaluation by
the self of one's own moral traits, as well as a positive evaluation
of the extent to which one has been morally successful. Analo-
gously, intellectual integrity involves a positive epistemic evalua-
tion of one's own intellectual traits, as well as a positive evalua-
tion of the results of one's cognitive efforts in the knowledge one
has obtained. But there must be a connection between moral and
intellectual integrity, for integrity is above all the virtue of being
true to oneself, and assuming that the self is singular, it is in-
coherent to be true to your moral self but not to your intellectual
self or conversely. Integrity requires, then, a functional unity not
only between moral and intellectual integrity but between moral
and intellectual virtues in general. It is beyond the scope of this
book to attempt to argue against those philosophers who have
thought that the self is divided, but it is worth remarking that a
rejection of that view leads one to reject as well any clear separa-
tion in the operation of the moral and the intellectual virtues.
Hypocrisy is one of the few vices sought out and roundly
condemned in contemporary Western society. This vice also com-
bines intellectual and moral failings, at least in one of its forms. If
the hypocrite is someone who merely pretends to have moral
beliefs that he does not uphold in his own behavior, then he is
simply a deceitful cynic. But in the more interesting form, the
hypocrite's deceit becomes self-deceit. He manages to convince
himself that he believes what he says he believes, although his
behavior expresses a deeper way in which he does not really
believe what he claims. His moral failing leads to an intellectual
failing in knowing himself. Such a person suffers a psychic split
that is opposed to integrity in the sense of self-unity. The hypo-
crite should be distinguished from the person who sincerely be-
162
Intellectual and moral virtues
1. Idols of the Tribe. These have their roots in human nature generally,
"or from its attachment to preconceived ideas, or from its narrow-
ness, or its restlessness, or from an infusion of emotions, or from
the inadequacy of the senses, or from the mode of impression"
(aphorism 52). "The human understanding is like an uneven mir-
ror that cannot reflect truly the rays from objects, but distorts and
corrupts the nature of things by mingling its own nature with it"
(aphorism 41).
2. Idols of the Cave. These arise from the individual's particular na-
ture, of both mind and body, and come also from education,
habits, and by chance (aphorisms 42, 53-8).
3. Idols of the Marketplace. These have crept into the understanding
through the alliance of words and names. "For while men believe
their reason governs words, in fact, words turn back and reflect
their power upon the understanding, and so render philosophy
and science sophistical and inactive" (aphorisms 43, 59-60).
4. Idols of the Theater. These have their source in the fashion of the
day, including theological and philosophical fashion. They "are
not innate, nor are they secretly insinuated into the understand-
ing, but are imposed and received entirely from the fictitious tales
in theories, and from wrong-headed laws of demonstration" (aph-
orisms 44, 61-2).
(a) The first is of those who seldom reason at all, but do and
think according to the example of others, whether parents,
neighbors, ministers, or whom else they are pleased to make
choice of to have an implicit faith in, for the saving of them-
selves the pains and troubles of thinking and examining for
themselves.
(b) This kind is of those who put passion in the place of
reason, and being resolved that shall govern their actions and
arguments, neither use their own, nor hearken to other peo-
ple's reason, any farther than it suits their humor, interest, or
party. . . .
164
Components of intellectual virtues
(c) This third sort is of those who readily and sincerely fol-
low reason, but for want of having that which one may call
large, sound, roundabout sense, have not a full view of all that
relates to the question. . . . They converse but with one sort of
men, they read but one sort of books, they will not come in the
hearing but of one sort of notions. (Of the Conduct of the Under-
standing, sec. 3, pp. 208-9; similar ideas are expressed in
Locke's Essay IV.20)
detail is practical wisdom, the topic of section 5,41 and I will leave
for a later project the task of giving an account of the higher-order
virtue of integrity.
In 4.11 will show how the intellectual virtues can be defined in
terms of derivatives from the motivation for knowledge and reli-
able success in attaining the ends of these motivations. In 4.2 I
explain the value of the components of intellectual virtues accord-
ing to two forms of pure virtue theory. If we were to pursue a
happiness-based theory, then the goodness of both components
of each virtue would be explained teleologically. A virtuous moti-
vation and reliable success in achieving its aims would be good
because of their relation to the good of human flourishing or
happiness. On the other hand, if we were to pursue the type of
theory I call motivation-based, we would assign intrinsic and
independent value to the motivational components of virtues and
derive the goodness of the reliability component from the good-
ness of the motivational component. I will argue that either one of
these approaches can offer a promising way to explain the nature
and value of intellectual virtues, but I find the motivation-based
theory especially exciting, perhaps because of its novelty.
166
Components of intellectual virtues
167
A theory of virtue and vice
some deeper level. I will not go very far into this matter, although
it is an interesting one and ought to be pursued in a full theory of
virtue. I have already said in section 3 that I have no position on
the question of whether intellectual virtues that share a name
with certain moral virtues are two different virtues or one. Even
within the class of intellectual virtues it is difficult to demarcate
the boundaries of the individual virtues if I am right that they all
arise out of the motivation for knowledge since that implies that
all intellectual virtues are unified by one general motivation. But,
of course, the same thing can be said about all the other moral
virtues since they also can be unified by one general motivation
for good, and knowledge is a form of good.
Let me address one more point before beginning. The defini-
tion of intellectual virtue in terms of the motivation for knowl-
edge is circular if we then go on to define knowledge in terms of
intellectual virtue, as I intend to do in Part III. For the purposes of
my entire project, then, the thesis of 4.1 must be formulated less
succinctly but without circularity as the thesis that the individual
intellectual virtues can be defined in terms of derivatives of the
motivations for truth or cognitive contact with reality, where the
motivation for understanding is assumed to be a form of the
motivation for cognitive contact with reality. In this subsection I
am formulating the position in terms of the motivation for knowl-
edge because I think that that is closer to the way people actually
think of their own motives and the way those motives are
described by others, but I am not wedded to this view. The for-
mulation in terms of knowledge motivation is simpler, and, of
course, it is only circular when the theory of virtue is combined
with the theory of knowledge; it is not a problem for the theory of
Part II, which can stand alone. When the theory of virtue and vice
presented in Part II is placed together with the account of knowl-
edge in Part III, the motivational basis for intellectual virtue needs
to be described as the motivation for truth or cognitive contact
with reality, where that is understood to include contact that is
high-quality and nonpropositional.
43
Hobbes implies here that people do not differ as much in their sensory fac-
ulties as in their virtues and vices. He also says that part of what leads us to call
a quality a virtue is that it is uncommon. The Hobbesian approach would
hesitate, then, in attributing anything virtuous to cases of simple perceptual
beliefs that are produced by normally functioning faculties.
169
A theory of virtue and vice
edge (1933, chap. 2, sec. 1). These values require the practice of
what Dewey calls "reflective thinking/' which he outlines in
some detail:
174
Components of intellectual virtues
175
A theory of virtue and vice
176
Components of intellectual virtues
177
A theory of virtue and vice
own and must be willing to withstand attack when she has good
reason to think she is right, but not otherwise. Parallel remarks
apply to the other intellectual virtues. It follows that each of these
intellectual virtues has a motivational component arising out of
the motivation to know and a component of reliable success in
achieving the aim of the motivational component.
We saw in 2.5 that most virtues are acquired by habituation and
we only consider them virtues when they are entrenched in the
agent's character. Entrenchment is a necessary feature of virtues
because they are often needed the most when they encounter
resistance. For example, the tendency to be motivated by compas-
sion does not signify the existence of the virtue of compassion in a
person who loses this motivation in the presence of physically
unattractive persons in need, even if these circumstances do not
arise very often. Similarly, the tendency to be motivated to fairly
evaluate the arguments of others does not signify the existence of
the virtue of intellectual fairness in a person who loses this moti-
vation when confronted with arguments for unappealing conclu-
sions, even if she is lucky enough not to encounter such argu-
ments very often. So the motivational component of a virtue must
be inculcated sufficiently to reliably withstand the influence of
contrary motivations when those motivations do not themselves
arise from virtues. The more that virtuous motivations and the
resulting behavior become fixed habits, the more they are able to
reliably achieve the ends of the virtue in those cases in which
there are contrary tendencies to be overcome.
One way to distinguish among the truth-conducive qualities
those that are virtues and those that are not is by the difference in
the value we place on the entrenchment of these traits. Montmar-
quet (1993, pp. 26-7) mentions that we would not want the desire
to uphold behaviorist psychology to be an entrenched trait even if
it is truth conducive, unlike the desire for the truth itself or, I
would add, the desire to be open-minded, careful in evaluating
evidence, autonomous, etc. The latter traits, when entrenched,
lead to the truth partly because of their entrenchment, whereas the
desire to uphold behaviorism is less likely to lead to the truth if it
is entrenched than if it is not. The intellectual virtues are a subset
of truth-conducive traits that are entrenched and whose entrench-
178
Components of intellectual virtues
179
A theory of virtue and vice
183
A theory of virtue and vice
184
Components of intellectual virtues
188
Components of intellectual virtues
195
A theory of virtue and vice
If Uncle Toby has intellectual vices, they are the vices of excess:
excessive attentiveness, thoroughness, diligence, perseverance.
Such traits have sometimes been called, with irony, "virtues in
excess/'51 Recall Lorraine Code's remark that one hesitates to
attribute intellectual virtue to "a voracious collector of facts"
(1987, p. 59). But interestingly, these traits do not seem to detract
from the goal of knowledge in Uncle Toby's case, and his motiva-
tion itself does not seem to be out of line, unless we think that the
proper motivation for knowledge resists the extreme specializa-
tion of Uncle Toby's endeavors. But that seems unlikely. It is more
probable that any defects arising from an overabundance of the
desire to know are not specifically intellectual but involve the
agent's general well-being, as Tristram Shandy believes to be the
case about Uncle Toby. This is not to deny that there can be vices
of excessive attentiveness, thoroughness, and perseverance that
detract from the goal of knowledge. My conjecture is that it is not
51
See Watson 1984 for an interesting discussion of puzzles involving the idea of
virtue in excess.
196
Components of intellectual virtues
flourishing life, and that means, on the account I have given, that
a person leading a flourishing life has the motivational compo-
nent of each virtue and is reliable in bringing about the aims of
these motivations. So she is characteristically motivated by com-
passion for the suffering of others and is reliably successful in
alleviating their suffering, she is motivated to evaluate the opin-
ions and arguments of others with an open mind and is reliably
successful in doing so, and so on for all the virtues. An alternative
happiness-based account would say that virtues are means to
bringing about a flourishing life. This approach would not define
human flourishing in such a way that the possession of virtues is
an intrinsic part of it, but it would argue that the possession of
virtues is a causal condition for flourishing. So a person who is
characteristically motivated by compassion for the suffering of
others and who is reliably successful in alleviating their suffering
is more apt to flourish than she would be otherwise. Similarly, a
person who is characteristically motivated to be open-minded in
evaluating the opinions and arguments of others and is reliably
successful in doing so is also more apt to flourish than she would
be otherwise, etc. I will not pursue this variant of happiness-
based ethics further, since it seems to reduce virtue theory to a
version of consequentialism, but I mention it because philoso-
phers attracted to reliabilism might find an ethical theory of this
type suitable to combine with their own epistemic theory.
The Aristotelian position that the moral virtues are all constitu-
ents of a life of eudaimonia is well known and I will not rehearse it
here. More interesting for the present discussion is the idea that
the intellectual virtues also are components of eudaimonia. Al-
though it is uncommon to find this view expressed today, lofty
claims were made about the connection between epistemic goods
and eudaimonia in both classical Greek and medieval philosophy.
The classical idea was that eudaimonia involves the fulfillment of
human nature, and knowledge is at least part of such a fulfill-
ment, perhaps the most important part. In Book X of the Nic-
omachean Ethics Aristotle characterizes eudaimonia as a life of con-
templation; Aquinas describes the ultimate end of human life as
the Beatific Vision, a state that is simultaneously the enjoyment of
perfect happiness and a perfect revelation of truth. The medieval
198
Components of intellectual virtues
52
See MacDonald 1991 for a collection of essays o n the medieval identification of
being with goodness.
53
Braeton says she prefers the concept of intellectual virtue to that of intelligence
because it is more honest. Intelligence also is evaluative, b u t less obviously so.
199
A theory of virtue and vice
54
See, e.g., Nagel 1979b for an argument that value is not and cannot be made to
be unitary. Nussbaum argues (1986) that unlike Plato, Aristotle maintained
that the attempt to harmonize the goods of a rich human life can lead to
tragedy.
2OO
Components of intellectual virtues
55
As Dewey Hoitenga has suggested to me, the essence of the story would go
something like this. A person is dependent for her happiness on the re-
ciprocating virtue of others, and that will be more forthcoming if she is vir-
tuous also. This intimates something like the Golden Rule.
2O1
A theory of virtue and vice
202
Components of intellectual virtues
203
A theory of virtue and vice
that this, most of all the senses, makes us know and brings to
light many differences between things. (Trans. Ross)
But even if the desire for knowledge were universal, that would
not be sufficient to demonstrate its value. The kind of knowledge
Locke is really interested in requires some exertion of the mind,
an effort that is accompanied by even greater pleasure. And
Locke makes it clear that the value of the drive to know is not
dependent upon its successful attainment. Let us look again at a
passage quoted in a footnote in Part I with its continuation:
and he will have reason to think his time not ill spent, even
when he cannot much boast of any great acquisition. {Essay,
"Epistle to the Reader/' beginning)
207
A theory of virtue and vice
In each case we fault the person because her attitude and be-
havior betrays a defect in her motivation for knowledge. Using an
unreliable belief-forming process is an unfortunate thing, but
what is truly bad is to have an insufficiently strong motivation for
knowledge, and what is wrong is to use procedures that a person
ought to know are unreliable in leading to the end of that motiva-
tion. It is also bad to fail to have knowledge, but the badness of a
bad motivation is not derivative from the badness of the failure to
have knowledge.
There is another reason I am deserving of criticism if I believe
as the result of guessing. If I once form a belief according to a
procedure that I should know to be unreliable, it will be easier for
me to do so again, and I may eventually acquire an unreliable
intellectual habit, one I should have known better than to let
myself acquire. Beliefs are produced by a limited number of pro-
cesses, and I tend to use the same ones over and over, so these
processes become habits. Another reason we think of the belief
formed by guessing as defective, then, is that guessing in one case
may lead to guessing in other cases and may eventually lead to a
bad intellectual habit - an intellectual vice. Once a believer has
such a vice, his problem becomes a social one since believers form
a community, and his unreliability means that others in the com-
munity cannot trust his cooperation in their own pursuit of
knowledge, nor can he trust himself in forming beliefs in the
future. Guessing, then, shows a defect not only in the guesser's
motivation to get knowledge for himself but in his motivation to
aid his epistemic community in its knowledge acquisition. The
people in his community will blame him because he does not care
enough about knowledge, either for himself or for others. From
motivation-based perspective, all of these problems in violating
proper epistemic procedures can be traced back to a fundamental
motivational defect.
The vice of being a chronic guesser has no name and that is
undoubtedly because it is so rare. But we can imagine such a
person, and the way we would evaluate him draws attention to
some of the things we expect out of persons who are epistemically
blameless, and if we follow this through, it leads us to the primary
object of blame. First, we are responsible to be aware of the
208
Components of intellectual virtues
211
A theory of virtue and vice
We have already seen how Aristotle divides the soul into two
parts. In Book VI he follows this up with a division within the
rational part itself:
We said before that there are two parts of the soul: rational and
irrational. Now we will speak of the rational part in the same
way. Let us suppose two parts of the rational soul: one by
which we consider the kind of things whose principles cannot
be otherwise; the other by which we consider contingent
things. (NE VI. 1.1139a 4-9)
[T]ruth is not the same for the practical as for the speculative
intellect. Because the truth of the speculative intellect depends
213
A theory of virtue and vice
214
The importance ofphronesis
217
A theory of virtue and vice
This passage is the closest I can find in Aquinas to the view I will
propose in 5.2.
The distinction between the speculative and the practical intel-
lect and the accompanying distinction between speculative wis-
dom (sophia) and practical wisdom (phronesis) were not always
made in Aristotle's way in Hellenistic and medieval philosophy.
Occasionally there were moves to bring sophia and phronesis closer
together, moves that could be interpreted either as emphasizing
the practical aspect of speculative wisdom or as emphasizing the
theoretical aspect of practical wisdom. The former move is more
Platonic, but the latter is closer to the view I find most promising.
There is evidence that the Stoics in particular extended the do-
main of the practical intellect in the way I wish to promote in this
book. Harry Wolfson (1970) mentions the Stoic idea of assent
(sunkatathesis), which apparently was a term proposed by the
Stoics to refer to judgments of truth as well as to judgments of
goodness, but which both arise from the practical intellect:
22O
The importance of phronesis
applies. The features of a situation that are relevant for one virtue
may not be for another. So what concerns the virtue of courage is
the nature and extent of some danger and the value and effective-
ness of facing it in various ways. But in a situation of danger,
many other features of the situation may concern other virtues,
such as generosity, fairness, or humility. Sometimes the
courageous thing to do is not the humble or generous or fair thing
to do. Suppose there are salient features of the situation that per-
tain both to courage and to humility. It might be that in this case
humility is more important than courage. That is to say, it may be
more important to act in a humble manner than to do what
courage would have dictated in the absence of considerations of
humility. Or it might be that humility is not so important in this
case, although this is not to say that humility is usually less impor-
tant than courage. The person who knows how to act (and feel) in
these cases does not simply act from a combination of humility
and courage and the knowledge of the proper function of these
virtues. If we are to avoid an excessive and unworkable fragmen-
tation of value, there needs to be some virtue that permits a per-
son to sift through all the salient features of the situation - that is,
all those features that are pertinent to any of the virtues - and to
make a judgment that is not simply the judgment of a person qua
courageous, qua generous, or qua humble but is the judgment of a
virtuous person.62 Therefore, the ability to mediate between and
among the individual moral virtues must itself be a virtue. Phro-
nesis is defined in part as the virtue that has this function.63
The mediation between individual virtues is necessary be-
tween and among the various intellectual virtues as well as the
62
I d o not m e a n to imply that there will always b e a unique right choice in every
case of virtue conflict b u t only that some w a y s of resolving conflicts are better
than others. I will discuss the possibility of moral dilemmas in 6.1.
63
Gary Watson (1984, p . 65) argues that some of these cases of apparent virtue
conflict are cases in which neither response is w r o n g b u t neither response can
express both of the virtues in conflict. So o n e response m a y express kindness
while the other m a y express honesty. Watson offers the conjecture that it is in
such indeterminate cases that persons reveal their moral individuality. The
same point could b e m a d e about the w a y s in which persons reveal their
intellectual individuality.
222
The importance ofphronesis
223
A theory of virtue and vice
65
O n the prejudices of philosophers, see Nietzsche at the beginning of Beyond
Good and Evil.
66
I have looked at the connection between N e w m a n ' s illative sense and phronesis
a n d applied it to the methodology of theology in Zagzebski 1995.
225
A theory of virtue and vice
insight into human nature and the motives that lead people to
murder. Still others stress the simplest explanation of the facts. In
all these cases the detective's ability is something that is clearly
not the following of a set of rules, at least not insofar as they can
be known and taught. Of course, a person might be able to learn
how to be a good detective by frequent and intense exposure to
good detectives. The point is that they do not learn it by learning
how to follow a procedure.
In spite of the theoretical need for phronesis in a virtue theory, it
is doubtful that anyone would be convinced of its existence if it
were not supported by our moral experience. Ultimately we must
ask whether we are reasonably confident that we know individ-
uals with phronesis or know of such people. If not, we can simply
respond to the examples I have given by refusing to admit that
they show any special quality in the detective, the historian, the
philosopher, or the ordinary newspaper reader. We could say that
the detective is simply lucky or is a plodder; if he works on a case
long enough, eventually he will hit on the truth. The brilliant flash
of discovery so deftly described in mystery novels is only fiction.
The historian and the philosopher have no business actually be-
lieving what they write since they have insufficient evidence for
their claims, and it would be better if they admitted their work is
historical fiction and philosophy fiction respectively, only not so
clever as the detective novel. As for the newspaper reader, per-
haps his defects are too evident to require comment. The most
intellectually respectable thing to do is to follow the rules of rea-
soning and evidence and refrain from forming beliefs beyond
their boundaries. Or, at least, it can be argued.
Aristotle thought that we see phronesis in some of the persons
in our community, and the contemporary experience of it has
been well described by John Casey (1990):
230
The importance ofphronesis
231
A theory of virtue and vice
233
A theory of virtue and vice
Beliefs are not acts but cognitive states that also result from the
operation of the character traits of intellectual virtues and vices.
We saw in Part I that beliefs cover a range of voluntariness but are
sufficiently voluntary to be evaluated in a sense similar to the
moral, and we have seen in section 3 of this part that the intellec-
tual virtues are forms of moral virtues. This means that if the
central evaluative property of beliefs derives from intellectual
virtues, that property derives from moral virtues. It follows that
such a property is, not just analogous to the Tightness of acts, but
very much like Tightness in its moral import. The most common
term for this property of beliefs is "justified," and I will use that
term here, even though we looked at the confusion surrounding
the notion of justification in Part I, and I said there that it is at least
arguable that Alston is right in claiming that there is no target
concept that is the object of the prevailing disputes over justifica-
tion. But it is clear that some normative property of beliefs func-
tions in epistemic evaluation in a way that is very similar to the
property of Tightness as ascribed to acts. If justification (or justi-
fiedness) is not that property, I will be happy to drop it in favor of
something else, but I doubt that there is any other candidate with
as good a claim to being the desired property as justification.
I have tied the Tightness of an act to the hypothetical behavior
of virtuous persons. It is no doubt obvious that a person can
perform a right (permissible) act who is not actually virtuous, but
the act must mimic the behavior of virtuous persons in certain
respects. A virtuous person's behavior arises out of virtuous mo-
tives and is reliably successful in achieving virtuous ends. What
makes the virtuous person reliably successful in addition to her
motive is her understanding of the moral and nonmoral facts
about the situations she encounters. The level of understanding a
virtuous person has, then, is whatever is sufficient to make her
reliably successful in producing the ends of the virtue. It follows
that to do what a virtuous person does is to do what a person
would do who has virtuous motives and such a level of under-
standing. If to perform a right act is to do what virtuous persons
might do, the conclusion is that a right act is what a virtuously
motivated person might do in certain circumstances, given the
understanding of the facts of the particular case we would expect
234
The definition ofdeontic concepts
236
The definition of deontic concepts
70
Michael Slote has said in correspondence that the motivation of a virtuous
person would lead her to find out the relevant facts of the matter before
making a decision, and so it could be claimed that the understanding compo-
nent is included in the motivation component. But although I agree that the
motivation component would lead the agent to make the right efforts to under-
stand the nonmoral facts relevant to the decision, and although I also agree
that if we reject skepticism, we can assume that moral efforts are not systemat-
ically unsuccessful, the concept of a right act depends crucially on what is
actually going on, not just on what a virtuously motivated person would believe
is going on, because a right act is morally successful.
238
The definition of deontic concepts
240
The definition of deontic concepts
242
The definition of deontic concepts
244
The definition of deontic concepts
245
A theory of virtue and vice
praise (in that sense) from the cases of failure, and the important
thing to see is that such praise is nonetheless moral praise. What's
more, it is not mere success that is praiseworthy but, rather, suc-
cess that is explained by the fact that the act has the other morally
desirable features we have identified. So accidental success is not
good enough to merit the kind of moral praise I am describing.
Since the evaluative judgment that is relevant here is a judgment
about acts, it requires us to define another category of acts that
arise from virtue or virtuous motives. I propose that it is desirable
in theoretical ethics to have a concept for an act that gets every-
thing right - in its motive, in the behavior to which it gives rise,
and in its end - and that is successful in reaching its end because
of these morally praiseworthy features of the act.
Let us call an act an act of virtue A if and only if it arises from
the motivational component of A, it is something a person with
virtue A would (probably) do in the circumstances, and it is
successful in bringing about the end (if any) of virtue A because
of these features of the act. Notice that this definition is an exten-
sion of two definitions given in 6.1: the definition of "right act"
and the definition of "is praiseworthy for doing an act." The idea
is that an act is an act of virtue A if and only if the agent is
praiseworthy for doing the act and the act is successful in bring-
ing about the end of the A motivation because of the features of it
that make the agent praiseworthy for doing it.
Let us now consider some examples and applications of this
concept. I suggest that we not call an act "an act of justice" if it
does not lead to the production of a just state of affairs. For exam-
ple, a judge or jury might reach a wildly mistaken verdict out of a
motive of justice. Such a verdict is perfectly compatible with the
judge or jury being just persons, with their acting out of the virtue
of justice in their deliberations and findings, and with their mak-
ing no errors in following proper judicial procedures. People who
criticized the jury verdicts in the first Rodney King trial almost
always based their judgment on the verdict itself, and in most
cases the judgment was not intended to impugn either the mo-
tives or the character of the jury members. The fact that there was
negligible evidence against the motives of the jury was typically
248
The definition of deontic concepts
249
A theory of virtue and vice
250
The definition of deontic concepts
The importance of the aim may vary with the virtue. I have
given examples of justice, compassion, and liberality, which are
virtues with reasonably clear external ends, and temperance,
which has a reasonably clear internal end. Intellectual virtues, as
we have seen, also have internal ends. Some virtues, such as
courage, may not have an end that is easily identifiable, and there
may be virtues, such as wisdom, that have no end of any kind. My
point is not that all virtues have easily identifiable ends, or even
that all virtues have ends at all, but only that in those cases in
which a virtue does include an end of bringing about some state
of affairs, to that extent we should not call an act an act of the
virtue in question unless that act is successful in its end.
As I have defined an "act of virtue," it is to some extent a term
of art. I realize that a great number of people, including many
who are highly reflective on these matters, use the expression
"performing an act of courage" synonymously with "acting
courageously"; and similarly for other virtue expressions. A per-
son who goes down to defeat while risking his life for a greater
good acts with courage; he acts courageously. Why not say his act
is an act of courage? My response is to say (1) the distinction I am
making is needed and important enough to be reflected in our
form of speech, and (2) we make this distinction in a limited
number of cases anyway. The latter would include some usages of
"an act of justice," such as in the Rodney King verdict just cited.
Discussions of judicial justice commonly distinguish between the
following of correct judicial procedures, regardless of outcome,
and finding the right person guilty. I suggest that the expression
"acting justly" is compatible with an outcome whereby a mur-
derer goes free, whereas the expression "an act of justice" is not. I
am also urging the generalization of this difference to other vir-
tues. The idea is that we really do make a moral distinction be-
the qualities that are the aim of virtue A, and the best explanation of the fact
that the state of affairs has those qualities is that the act has these features.
The idea for this expansion of the definition of an act of virtue comes from
John Greco's definition of knowledge. As a way of handling Gettier problems,
Greco (1994a) has proposed that in cases where S knows p, S's believing p truly
is best explained by the fact that S's believing p is the result of a stable disposi-
tion toward reliable belief.
251
A theory of virtue and vice
77
This is a central point in Beyond Optimizing (Slote 1989). The idea is first
introduced on p. 5.
254
The scope of the moral
2-55
A theory of virtue and vice
256
The scope of the moral
2-57
A theory of virtue and vice
258
Part III
The nature of knowledge
1
Plantinga and Foley are both examples of this, but in different ways. Plantinga's
main concern is knowledge, and he relegates the obviously moral concepts of
justifiability and epistemic duty to a different and less important epistemologi-
cal domain. Foley, on the other hand, prefers to focus on the egocentric concept
of rationality and the ethical concepts associated with it. He is willing to accept
the consequence that knowledge is less important than was previously thought
in epistemological history.
260
Knowledge and the ethics of belief
2
See again the pertinent comment by Thomas Nagel in the last footnote of
"Moral Luck" (1979a).
261
The nature of knowledge
262
Knowledge and the ethics of belief
from that that a state gained too cheaply does not really deserve
the accolade "knowledge," and this favors the idea that knowl-
edge should be defined relatively strictly It deserves quite a lot of
effort, and presumably, effort is required. On the other hand, we
need to consider the fact that if the requirements for knowledge
are too strict, it loses its importance in another way. People auto-
matically adjust their sights to the attainable. At least, if some-
thing is thought to be unattainable, it tends to lose importance.
This is not to say that desirability and importance are propor-
tional to attainability; they certainly are not. Nonetheless, it is
probably the case that at some critical level of unattainability, the
unattainable object loses its interest for us. Tiles and Tiles have
proposed that this is precisely the reason why Hume abandoned
the classical project of knowledge as a form of understanding and
explanation in favor of the more attainable goal of discovering
empirical regularities (1993, p. 114). Kant and the varieties of
Idealism are further evidence that philosophers are not exempt
from the very human trait of focusing interest on the attainable. If
a cognitive grasp of the thing-in-itself is unattainable, that was
reason enough for Kant to give an account of knowledge in which
the object is not the thing-in-itself. Similarly, Idealists define truth
in terms of the cognitively graspable. Redefining truth is one way
to put knowledge within our grasp. The other way is to redefine
the normative element of knowledge, making it easier to achieve.
The latter is the favored approach today.
I said at the beginning of this book that although I allow con-
temporary concerns to dictate many of the questions I address in
this work, I try to be faithful to the history of the concept of
knowledge in my analysis. This leads me to conclude that knowl-
edge ought to be defined in a way that is relatively rigorous, but it
should still be something that is reasonably attainable. I am aware
that the definition I will give is somewhat stricter than usual in
contemporary philosophy, but it is also rather elastic. There are
ways of interpreting it that ought to be acceptable to most of those
who favor looser requirements, which is a desirable feature in the
present philosophical climate. The definition is contemporary, but
it has clear classical roots.
263
The nature of knowledge
2 DEFINING KNOWLEDGE
3
See Foley 1987, pp. 168-9, f° r a discussion of chicken sexers. Foley says that it is
not implausible to suppose that the chicken sexer knows the sex of the baby
chick, although his belief may not be epistemically rational.
265
The nature of knowledge
tue. This suggests that the concept of intellectual virtue does most
of the normative work in the evaluation of cognitive processes
and states, and so it is the first place to look for a way of under-
standing the nature of knowledge. We have also seen in this sub-
section that the definition should not be too vague, that it should
not appear to be an ad hoc reaction to counterexamples, that it
should not be more precise than general agreement warrants, and
that it should be practically useful in guiding our search for
knowledge.
With these constraints in mind, let us go back to where we left
off in Part II when we defined epistemic justification. A key
difference between knowledge and justifiedness is that the latter
is a quality that even at its best only makes it likely that a belief is
true. Justifiedness is a property that a belief has in virtue of being
a member of a set of beliefs of a certain kind. Similarly, we call a
person justified in having a belief because she has a property that
(among other things) tends to lead her to true beliefs. Knowledge,
in contrast, is not essentially a matter of a belief's being like any-
thing else or tending toward anything, nor do we say that a person
has knowledge because she has a property that tends toward
anything. This is not to deny that there is a component of knowl-
edge that is something like justification, but what is crucial in
distinguishing knowing states from other good but lesser epis-
temic states is that the knowing state has a normative property
that that state has in particular, not simply qua member of a set of
states of a certain kind.
We might look at the difference between knowledge and justi-
fied belief as analogous to the difference between act and rule
utilitarianism. In rule utilitarianism an act is right because it fol-
lows a rule the following of which tends to have good conse-
quences. Similarly, a belief may be justified because it follows
epistemic rules the following of which tends to lead to the truth or
because it is an instance of a reliable belief-forming process (re-
liabilism) or on my account, because it is a belief that an intellec-
tually virtuous person might have in the circumstances. That is to
say, it is a member of the class of beliefs that a person who has
virtuous motivations and is reliable in bringing about the end of
268
Defining knowledge
269
The nature of knowledge
270
Defining knowledge
Alternatively,
Def 2: Knowledge is a state of true belief arising out of acts of
intellectual virtue.
Since the fact that a belief arises out of acts of intellectual virtue
entails that it is true, the second definition can be formulated
without redundancy as follows:
Def 3: Knowledge is a state of belief arising out of acts of
intellectual virtue.
The second definition follows the contemporary convention of
defining knowledge as true belief plus something else, but its
redundant element makes it misleading. The first definition may
be preferable since it is noncommittal on such questions as the
object of knowledge, the nature of truth, and the existence of
propositions, which are not explored in this work. It also permits
a broader interpretation of knowledge since knowledge may in-
clude cognitive contact with structures of reality other than the
propositional.
and all the way up to the concept of an act of virtue. Here we find
an exception because no concept in ethics is exactly comparable to
the concept of knowledge. The closest analogue I can find is just
the concept of an act of moral virtue itself, which we said in Part II
is the concept of a right act in the strongest sense of right. An act
of moral virtue has all the morally desirable features related to an
act - in the act, in the agent, and in the relation between the agent
and the act and its consequences. Similarly, an act of intellectual
virtue has all the morally/epistemically desirable features an in-
tellectual act can have, but our primary interest in the epistemic
case is not in the act itself but in the state produced by the act. It
may be true that to perform an act of intellectual virtue is to know,
and if so, an act of moral virtue seems to be very close to an ethical
counterpart to knowing. But knowledge is an enduring state, and
so even though the concept of knowledge may be broad enough
to encompass intellectual acts, it is certainly not limited to them. It
is for this reason that I have defined knowledge as a state that
arises from acts of intellectual virtue.
If there were an ethical counterpart to my definition of knowl-
edge, it would satisfy the following schema: x is a state of y
arising out of acts of moral virtue. As far as I can tell, there are no
x and y that satisfy this schema.5 In the definition of knowledge,
what has the position of y is the ultimate aim of intellectual vir-
tues, and knowledge is defined in part by the way we get there.
Presumably, to find the moral counterpart we would look for an
ultimate end of other groups of moral virtues to satisfy this defini-
tional schema. We have seen that some moral virtues ultimately
aim at such things as the well-being of others. But there does not
seem to be any concept that is defined as the state of an agent
when she brings about the well-being of others through acts of
moral virtue.6 One's own happiness, or eudaimonia, is probably
also an ultimate end of (at least some) moral virtues, but the
concept of eudaimonia does not seem to depend upon the way we
get there in the way knowledge does. That is, we would not deny
5
Dewey Hoitenga has suggested to me that this schema might be satisfied by the
following: Peace is a state of society arising out of acts of courage (as in war).
6
Brian Leftow has suggested to me that perhaps holiness is such a state.
272
Defining knowledge
that a woman is in a state of eudaimonia for the sole reason that she
did not bring it about even in part through proper moral be-
havior, whereas we do deny that she has knowledge if she did not
bring about that state through behavior that is proper according
to the criteria of the particular theory.
It would be interesting to speculate on the significance, if any,
of the fact that the structural similarity between normative epis-
temology and ethics breaks down at the concept of knowledge,
but at this point I will return to an examination of the definition
itself so that we can see how it can be tested and applied.
279
The nature of knowledge
she has good contrary evidence. She probably does not doubt
such introspective beliefs as that she is in pain, although we might
expect her to consider from time to time if and why such intro-
spection is trustworthy. But we would assume that most of the
time she does not doubt or even reflectively consider her percep-
tual and memory beliefs. She does not because she maintains a
presumption of truth in such cases until she is given reason to
think otherwise. Such an attitude is itself an intellectually vir-
tuous one; to act otherwise is to exhibit a form of intellectual
paranoia. So this might give us reason to think that even young
children can perform acts of intellectual virtue before they are old
enough to acquire the intellectual virtues. As long as they are old
enough to imitate the behavior of intellectually virtuous persons
in their belief-forming processes, young children (and possibly
animals) can have knowledge based on perception and memory.
Their behavior is no different from that of the intellectually vir-
tuous, and there may not even be any discernible difference in
their motives.
We must admit that there is a difference in their behavior in
counterfactual circumstances, however. Small children and ani-
mals do not doubt when it is not virtuous to do so, but they do not
doubt when it is virtuous to do so either. This consideration
would favor a compromise position on sense knowledge. True
sensory beliefs count as knowledge in normal circumstances be-
cause intellectually virtuous persons who have these beliefs do
what intellectually virtuous persons do in these cases, which does
not require them to exert themselves, but they would exert them-
selves if the circumstances warranted it.8 This gives us a view of
perceptual and memory beliefs that sees them as low-grade
knowledge as long as they satisfy certain conditions. Take the
8
Ernest Sosa (forthcoming) expresses a similar view: "No human blessed with
reason has merely animal knowledge. For even when perceptual belief derives
as directly as it ever does from sensory stimuli, it is still relevant that one lacks
contrary testimony. People automatically monitor their background informa-
tion and sensory input for contrary evidence and normally opt for coherent
hypotheses even when responding most directly to sensory stimuli. For even
when response to stimuli is most direct, credible contrary testimony would
change one's responses/'
280
Defining knowledge
9
William Blake, "Proverbs of Hell/' from The Marriage of Heaven and Hell (1790-
3).
10
Of course, an important point of Nagel's is his claim that it is also a mistake to
identify reality with reality-as-perceived-by-nobody-in-particular.
282
Gettier problems
3 GETTIER PROBLEMS
283
The nature of knowledge
three barn facades for each real barn in an effort to make them-
selves look more prosperous. Your eyesight is normal and reliable
enough in ordinary circumstances to spot a barn from the road.
But in this case the fake barns are indistinguishable from the real
barns at such a distance. As you look at a real barn you form the
belief That's a fine barn. The belief is true and justified but is not
knowledge.
In this case also the problem arises because of the combination
of two accidental features of the cognitive situation. It is only an
accident that visual faculties normally reliable in this sort of situa-
tion are not reliable in this particular situation; and it is another
accident that you happened to be looking at a real barn and hit on
the truth anyway. Again the problem arises because an accident
of bad luck is canceled out by an accident of good luck.
Gettier problems cannot be avoided by Plantinga's recent the-
ory either. Plantinga calls the property that in sufficient quantity
converts true belief into knowledge "warrant" rather than "justi-
fication." On his proposal warrant is the property a belief B has
for believer S when B is produced in S by S's faculties working
properly in the appropriate environment, according to a design
plan successfully aimed at truth (1993b, chap. 1).12 But Plantinga
does not maintain that every warranted belief is true any more
than reliabilists maintain that every reliably formed belief is true
or internalists maintain that every internally justified belief is
true. This feature allows us to form a Gettier case for Plantinga's
theory parallel to the other two cases we have considered. To do
so we look for a situation in which a person's faculties are work-
ing the way they were designed in the appropriate environment,
but she unluckily has a false belief. We can then add a second
accident that makes the belief true after all.
Suppose that Mary has very good eyesight, but it is not perfect.
It is good enough to allow her to identify her husband sitting in
his usual chair in the living room from a distance of fifteen feet in
somewhat dim light (the degree of dimness can easily be spec-
ified). She has made such an identification in these circumstances
12
A very similar wording can be found in Plantinga 1988. In that paper he calls
"positive epistemic status" what he now calls "warrant/'
285
The nature of knowledge
many times. Each time her faculties have been working properly
and the environment has been appropriate for the faculties. There
is nothing at all unusual about either her faculties or the environ-
ment in these cases. Of course, her faculties may not be function-
ing perfectly, but they are functioning well enough that if she goes
on to form the belief My husband is sitting in the living room,
that belief has enough warrant to constitute knowledge when
true and we can assume that it is almost always true.
The belief is almost always true, we say. That is because warrant
in the degree necessary for knowledge does not guarantee truth,
according to Plantinga. If it did guarantee truth, of course, the
component of truth in the analysis of knowledge would be super-
fluous; knowledge would simply be warranted belief. So it is
possible for Mary to make a mistake even though her faculties are
functioning properly enough for knowledge and the environ-
ment is normal for the faculties. Let us look at one such case.
Suppose Mary simply misidentifies the chair sitter, who is,
we'll suppose, her husband's brother, who looks very much like
him. Her faculties may be working as well as they normally do
when the belief is true and when we do not hesitate to say it is
warranted in a degree sufficient for knowledge. It is not a ques-
tion of their suddenly becoming defective or, at any rate, more
defective than usual, nor is there a mismatch between her fac-
ulties and the environment. No one is trying to surprise or fool
her or anything like that. Her husband and his brother may not
even know she is in the house, so the normal environment has not
been doctored as it is in the fake barn case. Mary does exactly
what she has done hundreds of times before. So her degree of
warrant is as high as it usually is when she correctly identifies her
husband because even in those cases it is true that she might have
misidentified the chair sitter if it had been her husband's brother
instead. Of course, she usually has no reason to suspect that it is
her husband's brother, and we can imagine that she has no reason
to suspect so in this case either. Although she knows that her
brother-in-law looks a lot like her husband, we can suppose that
she has no reason to believe that he is in the vicinity and that, in
fact, she has strong reason to believe he has gone to Australia. So
in the case we are considering, when Mary forms the false belief,
286
Gettier problems
287
The nature of knowledge
Jones's belief that Smith has virus X, and no false belief figures
causally or evidentially into her justification, nor is there any false
belief in the neighborhood. Furthermore, she would have be-
lieved that Smith had virus X in a wide range of counterfactual
situations. Nonetheless, let us suppose that the belief is false.
Smith's symptoms are due to a distinct and unknown virus Y, and
the fact that he exhibits high antibody levels to virus X is due to
idiosyncratic features of his biochemistry that cause him to main-
tain unusually high antibody levels long after a past infection. In
this case Dr. Jones's belief that Smith is presently suffering from
virus X is false, but it is both justified and undefeated. Of course,
given that the belief is false, there must be some evidence against it
accessible to her in some counterfactual circumstances, so if
defeasibility conditions are strong enough, no false empirical be-
lief passes the test. But as said above, that is to impose an unrea-
sonably strong defeasibility condition, one that makes the
justification/defeasibility condition entail truth. The most reason-
able conclusion to draw in this case, then, is that Jones's belief is
justified and undefeated but false.
Now to construct a Gettier-style example we simply add the
feature that Smith has very recently contracted virus X, but so
recently that he does not yet exhibit symptoms caused by X, nor
has there been time for a change in the antibody levels due to this
recent infection. So although the evidence upon which Dr. Jones
bases her diagnosis does make it highly probable that Smith has
X, the fact that Smith has X has nothing to do with that evidence.
In this case, then, Dr. Jones's belief that Smith has virus X is true,
justified, and undefeated, but it is not knowledge.14
It appears, then, that no account of knowledge as true belief
plus something else can withstand Gettier objections as long as
there is a small degree of independence between truth and the
other conditions of knowledge. What are the alternatives? We
have already seen that one way to solve the problem is to give up
14
Charles Young has pointed out to me that Gettier cases can generate cases of
moral luck. If Dr. Jones has the Gettier belief that Smith has virus X and he
does, then if she gives the right treatment, she is just lucky in her success. And,
of course, her treatment may have ethical implications. In general, if you act on
a Gettier belief, you act rightly only as a matter of luck.
291
The nature of knowledge
15
Robert Almeder (1992) argues that no belief sufficiently justified for knowl-
edge can be false, but his reasons are completely different from the reasons I
have given here. John Pollock (1986, pp. 183-90) defines what he calls "objec-
tive justification" in such a way that a belief is objectively justified only if it is
ultimately undefeated, and no false belief can be ultimately undefeated. Sim-
ilarly, Keith Lehrer (Clay and Lehrer 1989, p. 152) defines what he calls "un-
defeated justified acceptance" in such a way that it entails truth, and he goes
on to define knowledge as undefeated justified acceptance. See also Trenton
Merricks (1995).
292
Gettier problems
truth is acquired by luck. And if so, Gettier cases are not puzzles.
This response may not count as a solution to Gettier cases because
it amounts to simply refusing to treat them as problems. I assume
that this will not be a popular approach, and so I conclude that the
moral of the story is that the truth condition for knowledge must
be entailed by the other conditions.
Almost every contemporary theory of knowledge analyzes
knowledge as true belief that is justified or warranted, and al-
though justification or warrant puts the believer in the best posi-
tion for getting the truth, the best position is assumed to be imper-
fect, for such is life. Properly functioning faculties need not be
working perfectly, but only well enough; reliable belief-
producing mechanisms need not be perfectly reliable, only reli-
able enough; evidence for a belief need not support it con-
clusively, but only well enough, and so on. And it is important to
see that the fact that justification is not a perfect guarantor of truth
is faithful to the preanalytic notion of justification. It is not, then, a
mistake in the analysis of justification or warrant to leave open the
possibility of a false justified/warranted belief. But given the fact
that the argument of this subsection shows that the conditions for
knowledge in addition to true belief must entail truth, it follows
that JTB accounts of knowledge are in a bind. The definition of
knowledge must make truth implied by the other component(s)
of knowledge, but it must do so in a way that is plausible and is
not ad hoc. In the next subsection we will look at how a virtue-
based definition of knowledge can satisfy these requirements.
ing that everything work perfectly, both inside the agent's head
and in his cognitive hookup with reality. For whatever reason, we
are somewhat more forgiving when it comes to morally correct
behavior, perhaps because an elaborate system of rewards and
punishments is often involved. Nonetheless, I argued in Part II
(6.2) that even in ethics there are good reasons for wanting to call
attention to a category of acts that gets everything right, both in
the agent's head and in his hookup with the moral reality he is
trying to produce, and I called these acts "acts of virtue."
The distinction between exhibiting virtue and performing an
act of virtue can help us resolve Gettier-style cases in a virtue
theory. An act of virtue is virtuously motivated, is an act that a
virtuous person is apt to do in the circumstances, and successfully
leads to the ends of the virtue in question through the operation
of these features of the act. The production of the goods that the
virtuously motivated person aims to produce enhances the moral
merit of the agent, and the production of evils that the virtuously
motivated person aims to prevent detracts from the agent's merit,
in spite of the fact that the level of the agent's virtue may be no
different from what it is in the case in which his acts are morally
successful. As we have said, this is because the aim of the moral
life is not merely to be virtuous but to bring about the goods at
which virtue aims by way of the sort of actions that usually result
from virtue.
Gettier problems in virtue epistemology can be resolved by an
analogous move. If we define knowledge simply as true belief
exhibiting intellectual virtue, we will be faced with Gettier coun-
terexamples. Consider again Dr. Jones and her diagnosis of
Smith's virus. Dr. Jones may have all of the intellectual virtues a
doctor can have, as well as any moral virtues that enhance the
practice of medicine successfully. She has a true belief, and the
belief arises out of intellectual virtue and exhibits such virtue.
Nevertheless, her diagnosis is correct only by accident, and most
of us hesitate to say that she knows Smith has virus X. Certainly,
Dr. Jones has intellectual virtue and she exhibits it when she
makes her diagnosis; we may even say she acts virtuously. None-
theless, she does not reach an accurate diagnosis because of her
296
Gettier problems
intellectual virtues. She does not reach the truth through an act of
intellectual virtue.
Similarly, in the case of the belief Smith owns a Ford or Brown
is in Barcelona, the belief may exhibit intellectual virtue since it is
acquired on good evidence and there is no reason to suspect
Smith is lying about his new car, but the belief does not arise out
of an act of intellectual virtue, because the truth is not acquired
through virtuous motives or processes. The truth is not obtained
because of the virtues. The truth is acquired because by accident
Brown is in Barcelona.
Again, in the case of Mary's belief that her husband is in the
living room, she may exhibit all the relevant intellectual virtues
and no intellectual vices in the process of forming the belief, but
she is not led to the truth through those virtuous processes or
motives. So even though Mary has the belief she has because of
her virtues and the belief is true, she does not have the truth
because of her virtues.
The last statement may be puzzling since it might seem to be
the case that if Mary has a belief B because of her virtues and B is
true, it follows that Mary has the truth because of her virtues. But
this inference is invalid. We cannot infer from the fact that M
attains something S because of her trait V that she attains every
property of S because of trait V. For example, Mary may be able to
identify a fleeing figure because of her keen eyesight, and that
figure may have a small birthmark on his arm, but it does not
follow that she can identify a man with a small birthmark on his
arm because of her keen eyesight. Similarly, Mary may be able to
reach a certain belief because of her virtuous motives and acts,
and the belief may be true, but it does not follow that she has
reached the truth because of her virtuous motives and acts. There
may be no more connection between her virtues and the belief's
property of truth than there is between her keen eyesight and the
man's birthmark.
Gettier problems can be avoided if we utilize the concept of an
act of intellectual virtue. Acts of intellectual virtue are strictly
analogous to acts of moral virtue, as we saw in Part II. An act of
moral virtue is morally right in a very strong sense because it has
297
The nature of knowledge
4 RELIABILISM
reliably, even though they themselves are not aware of how they
do it, nor can observers find anything in their behavior that looks
like a process. These people simply hold up the baby chick and
the right answer comes to them - or so it appears. (Frankly, I'm
skeptical about this whole thing, but I'll let that pass for the sake
of argument.) Richard Foley (1987) says that the chicken sexer
may have knowledge, although his belief is not rational. In that
book Foley takes an internalist position on rationality but leaves
open the possibility of an externalist position on knowledge. If it
is thought that the chicken sexer has knowledge, the externalist
concludes that it must be because his belief is produced by a
reliable belief-producing mechanism. The only difference be-
tween such a mechanism and other reliable belief-forming pro-
cesses, such as simple perception, is that it is unusual.
The second case is mentioned by George Mavrodes (1988, pp.
37-8), who describes a person who wakes up in the morning with
a firm conviction that there is a God, but without any awareness
of the process by which the belief was acquired and without any
independent support. As far as she knows, the belief just popped
into her head. It could, however, have been implanted by God,
who we may presume is reliable in his belief-creating activities.
She therefore has knowledge according to the criteria of reliabil-
ism. Now what should we say about these cases?
Virtually all philosophers now and in the past have agreed that
knowledge is more than true belief. They have also agreed that
knowledge is a more valuable state than true belief. It follows that
the value of the knowing state is more than the value of the truth
that is thereby possessed. So what knowledge has in addition to
true belief has value. Let us now look at the two cases just
described.
There is obviously something valuable in the ability that the
chicken sexer has. His possession of a mechanism for determin-
ing the sex of a chick is a valuable thing. It is valuable because of
its tendency to produce beliefs that are true. Its goodness can
therefore be compared to the goodness of mechanisms for
distributing goods fairly. Such mechanisms are valuable because
of their tendency to produce fair (good) states of affairs. Now
suppose some mechanism does reliably produce a fair state of
301
The nature of knowledge
302
Reliabilism
true belief cannot be. The value of the truth obtained by a reliable
process in the absence of any conscious awareness of a connection
between the behavior of the agent and the truth he thereby ac-
quires is no better than the value of the lucky guess.
Why, then, are these cases so often accepted as examples of
knowledge? I surmise that there must be a tendency to transfer
the quite obvious value of the reliable mechanism to the product
of that mechanism, the belief. This tendency may be natural and
understandable, but I do not see that it is justified. In addition, the
motive to avoid Gettier problems has led epistemologists to think
of knowledge as nonaccidentally true belief, and there is a sense
in which beliefs produced by a reliable mechanism obtain their
truth nonaccidentally. But surely the moral of Gettier cases is not
that just any nonaccidental way of obtaining the truth will merit
the honor of being a state of knowledge. Nonaccidentality is nec-
essary for knowledge, and that is demonstrated by Gettier exam-
ples, but we have seen no reason either in Gettier problems or
elsewhere for thinking that it is sufficient.
Let us now turn to another argument that reliabilism does not
give sufficient conditions for knowledge. The attempt to demons-
trate this in the literature is usually done by examples, some of
which are highly artificial. For instance, Alvin Plantinga invents a
case of a person who is zapped by rays from outer space that give
him a brain lesion that disrupts his normal cognitive processes
but leads to a reliable process for forming beliefs in a very limited
area (e.g., the belief that he has a brain lesion) (1993a, p. 199).17
There are several reasons why we should be dubious of such an
example. In the first place, I think that an epistemic theory has the
methodological right to take for granted that the theory offers an
account of the nature of human knowledge within a normal hu-
man environment. Counterexamples based on very abnormal
conditions have much weaker force against such a theory than
those that arise from normal circumstances, although this is not to
say that they have no force at all. But more important, this exam-
ple has the defect of relying on a very narrow interpretation of the
17
Plantinga offers this example against the "old Goldman" and offers variations
on the same example against other versions of reliabilism.
304
Reliabilism
generality problem. Clearly, the brain lesion has the effect of pre-
venting the operation of what would otherwise be reliable belief-
forming processes, and only produces a reliable process in the
formation of a single belief. A reliabilist could therefore reject the
general reliability of the belief-forming process in the case of the
belief I have a brain lesion because that belief is not produced by
a process that produces a large enough set of true beliefs to count
as reliable in the sense intended by reliabilists. What is really
wrong with the brain lesion is that it interferes with the operation
of a more reliable process for forming beliefs. We will look at a way
to handle the generality problem later in this subsection, but we
can conclude that Plantinga is partly right and partly wrong. His
example no doubt misses the mark because of its excessively
narrow interpretation of what counts as a reliable belief-forming
process, but on the other hand, it calls attention to the fact that we
shirk from calling an epistemic state knowledge when it arises
from processes in which something has clearly gone wrong,
whether or not the process is reliable. I believe this insight is
important.
To show that reliabilism does not give sufficient conditions for
knowledge it is desirable to have examples in which beliefs sat-
isfy reliabilist criteria for knowledge but not other time-honored
knowledge criteria. I believe we can find such cases in literature
as well as ordinary life. Rather than to resort to such a recherche
case as that of being zapped by rays from outer space, I will
present a realistic and familiar case and will evaluate it on
grounds which appeal not only to intuition but to actual norma-
tive epistemic criteria used within the history of philosophy.
The case concerns the unreflective acceptance of the opinion of
others. This is a habit that not only is common but has been
universally condemned by careful thinkers for millennia. Let us
look at Tolstoy's description of Oblonsky in Anna Karenina:
18
Anna Karenina, trans. Rosemary Edmonds (New York: Penguin Books, 1954),
pp 18-19.
306
Relidbilism
of the ways Locke says that things can go wrong in the formation
of our beliefs, quoted in Part II (end of sec. 3.2). Like the first sort
of man Locke describes, Oblonsky seldom thinks for himself but
borrows his ideas from others. Like Locke's second sort of man,
he puts passion ahead of reason (insofar as he has passion at all)
and lets his beliefs be governed by what suits his humor, interest,
or political party. Unlike Locke's third sort of man, he does not
follow reason, but like the third sort he does converse with only
one sort of man and read only one sort of book. Yet in spite of the
fact that Oblonsky's belief-forming processes are defective, it is
possible that some significant subset of his beliefs are formed
reliably. After all, it may be that the Liberal Party is generally on
the track of the truth, although Oblonsky is in no position to know
that. It is even possible that Oblonsky is lucky enough to have a
circle of friends who are just the right sort of people from whom
to take his beliefs. So the process by which Oblonsky arrives at
some significant subset of his beliefs might actually be truth con-
ducive. The important question, however, is whether these beliefs
constitute knowledge.
Oblonsky's epistemic status would not fare well under the
scrutiny of most philosophers from antiquity, not only Locke.
Certainly his beliefs do not satisfy the conditions of Plato's Meno
of being bound by an explanatory account (98a3). But it is even
doubtful that he would satisfy the much looser conditions for
knowledge of such lenient thinkers as Thomas Reid. Reid's fa-
mous policy of setting up common sense as an authority in epis-
temic matters was sometimes charged with canonizing "the judg-
ment of the crowd," but such a criticism hardly does justice to the
subtlety of Reid's thought. His appeal to common sense was fun-
damentally a response to philosophical skepticism, and he de-
fended certain judgments that the skeptic would question by ap-
pealing to characteristics of these judgments that no crude
Oblonskian believer could satisfy.19 No, the verdict of philosophi-
cal history is against the position that Oblonsky has knowledge,
no matter who his friends and newspapers are. If reliabilism is to
19
See S. A. Grave's discussion of Reid in the Macmillan Encyclopedia of Philosophy
(1967), p. 121, for a reading of Reid that supports this interpretation.
307
The nature of knowledge
20
Plantinga recognizes that the various brain lesion examples are vulnerable to
the generality problem in 1993b, p . 29.
21
Plantinga discusses the generality problem, particularly as an objection to
309
The nature of knowledge
Goldman's paradigm reliabilism, in 1988, pp. 24-31; in 1993a, pp. 198-9; and
in 1993b, p. 29. In the last he denies that the generality problem is a problem for
his proper function theory also, but, as I have said, I am inclined to agree with
Goldman that Plantinga faces a version of the problem also.
310
Reliabilism
311
The nature of knowledge
23
The cases of the nosey neighbor and of beliefs arising from pride were sug-
gested to me by Thomas D. Sullivan.
314
Reliabilism
neighbor's trash, she will leap to the conclusion that her neighbor
is an alcoholic, and so on. It is not at all clear, then, that beliefs
formed out of envious nosiness are reliable, at least not if we
examine a sufficiently large set of beliefs formed in this way.
Beliefs formed out of pride are similar. Within a limited range,
pride may have no deleterious effect on the proportion of truths
obtained by the believer, but in the long run, prideful believing is
not reliable. Pride leads a person to have an inordinate need to be
right, and this means not giving up beliefs that have been discon-
firmed by the evidence. In the case of the medical researcher, her
desire for fame will lead her to believe in the same way as a
person with a genuine love of truth in a limited range of cases, but
eventually her desire for fame will lead her to aim to believe what
others want to hear or what will get her name in the journals, and
when that happens, her belief-forming processes will diverge
from those of the genuine truth lover. I will not comment on the
fourth example, Uncle Toby, since I am willing to say that his
belief-forming processes are reliable. His defect is something else,
which we will turn to next. My first line of defense, then, is to say
that in each of these cases, with the exception of Uncle Toby, the
person does not have knowledge because beliefs formed out of
vices are not reliable as long as the reliability test is applied to a
fairly wide set of beliefs.
But there is another problem with the epistemic states of the
persons just described. Each of them will go wrong somewhere in
his doxastic structure, even if he is epistemically successful in the
particular case. Some intellectual virtues have to do more with the
quality of the knowledge than with its quantity, and knowledge is
also something that admits of higher and lower quality. Uncle
Toby is probably lacking something doxastically even though he
has gained a lot of knowledge. He is perhaps too focused on
trivia, and although that is not such a terrible thing, it is a waste of
intellectual talent. Toby's overall doxastic structure is out of bal-
ance, unintegrated, perhaps lacking in understanding of the par-
ticular facts with which he is so fascinated. Similarly, the intellec-
tually proud person falls into ruin somewhere along the line, even
if he makes no mistakes serious enough to lead to straightforward
315
The nature of knowledge
with perseverance, etc., because those are the acts that the various
intellectual virtues require, and she is motivated by the particular
motivational components of attentiveness, carefulness, and the
rest. But these motivations do not arise out of a love of truth for its
own sake, even though she is motivated to get the truth. To her
the point of being intellectually virtuous is to become famous. She
is, then, like the person Aristotle calls good but not noble. The
question for us, however, is whether she performs acts of virtue.
The answer, I think, is yes, as long as the motivational condition
for performing acts of virtue does not preclude a second-order
ulterior motive for having virtuous motives. If so, since she satis-
fies the other conditions for knowledge, she has knowledge. The
medical researcher, then, is not an example of a person who we
intuitively believe has knowledge but who does not satisfy the
definition.
Now what about the nosey neighbor? Clearly she is not moti-
vated to get the truth out of a love of truth for its own sake, but
she cannot be compared with the Laconian, because her ulterior
motive is not a natural good. Her desire to believe ill of someone
is surely not a natural good or any sort of good. But she may also
be able to perform acts of intellectual virtue for the same reason
the medical researcher can perform such acts. As long as the
definition of an act of intellectual virtue does not preclude ulterior
second-order motives (even bad second-order motives such as
envy or pride), she can perform acts of intellectual attentiveness,
perseverance, etc. She must, of course, have the particular mo-
tives unique to each of these virtues, but there is no reason to
think she lacks these motives. Her problem is the rationale for the
motivations themselves.
Uncle Toby's case even more clearly satisfies the definition of
knowledge since he really does have a love of truth for its own
sake, and his vice, if he has one, is a matter of overlooking some-
thing of greater value. There is no reason to think that this pre-
vents him from performing acts of intellectual virtue. I conclude
that we do not yet have a case in which a belief satisfies reliabilist
conditions for knowledge but not intellectual virtue conditions
and yet is intuitively a case of knowledge.
Let us consider another type of objection to my theory. There is
318
Plantinga's theory of proper function
319
The nature of knowledge
25
Ernest Sosa (1993) has proposed his swampbaby story to defend his claim that
the design feature is an unnecessarily restrictive addition to the concept of
proper function.
323
The nature of knowledge
26
I thank John Greco for showing me that the difference between me and Plan-
tinga is not the difference between belief-based and virtue-based epistemol-
ogy but is really a difference in ethical models.
326
Plantinga's theory of proper function
ing things that just "come" to him out of the blue. If he does the
latter, the description of the process he is using in determining the
sex of the chick changes, and its reliability clearly changes with it.
The same question arises with respect to the Mavrodian believer.
If a person who wakes up with the belief that there is a God tends
to go on to form a habit of believing whatever pops into her head
in the morning, then her theistic belief is an instance of the belief-
forming process believing whatever pops into her head first
thing in the morning. On the other hand, if she does not do this
but only believes those beliefs inserted into her head by God
overnight, then her belief is an instance of the belief-forming
process believing when she wakes up whatever God has put
into her head. Clearly, the reliability of these two processes is not
the same. If I am right that the best way to solve the generality
problem is through the idea of an epistemic habit, this favors a
virtue approach since the concept of an epistemic habit is closely
associated with the concept of intellectual virtue and is not so
clearly related to the concept of proper epistemic function. So
when we explicate the idea of a belief's arising from intellectual
virtue, we are already talking about the formation of beliefs
through habits. When we explicate the idea of a belief's arising
from a properly functioning faculty, no relation to a habit is im-
plied, at least not without a longer story on the nature of proper
function than Plantinga has given us.
Let me end the discussion of Plantinga's theory of warrant by
critiquing the last clause of his definition. According to that
clause, "the more firmly S believes B the more warrant B has for
S." It is easy to interpret this clause to indicate that as long as a
belief B satisfies the conditions for warrant given in the first
clause, then S gains in degree of warrant by believing B more
firmly. So persons S and T in the same epistemic circumstances
may both be warranted in believing, say, that Claremont is to the
east of Los Angeles, but if S believes it more firmly than T, then S
is more warranted in the belief than is T. This position is implaus-
ible and Plantinga has assured me that this is not what he in-
tended. Instead, he says, his idea is that properly functioning
faculties produce not only a belief but a certain degree or firmness
of belief that is appropriate to a given set of epistemic circum-
328
Internal and external aspects of knowing
335
The nature of knowledge
336
Ethics, epistemology, and psychology
337
The nature of knowledge
explaining a broad range of human life, and that at the end of the
book I would ask whether the theory or range of theories I have
presented here look like they are capable of doing this. So now
that we are at the end, I hope that I have offered sufficient reasons
for thinking that the answer is yes, or at least that it is likely
enough to be yes to make further work on virtue theory in epis-
temology profitable. An enormous amount of work needs to be
done on the virtues, most especially the intellectual virtues, and
this includes their connection with the concepts given the most
attention in epistemology - justification and knowledge and the
neglected concept of understanding - as well as their connection
with theoretical ethics. One of the most important directions for
future work, I believe, is the broadening of ethical theory and the
inclusion of normative epistemology within ethics proper. The
benefits to both fields ought to be considerable.
Now of all things good, truth holds the first place among gods
and men alike. For him who is to know felicity and happiness,
my prayer is that he may be endowed with it from the first, that
he may live all the longer a true man. (Plato, Laws, trans. A. E.
Taylor, V.730C)
339
Bibliography
341
Bibliography
342
Bibliography
343
Bibliography
344
Bibliography
345
Bibliography
346
Bibliography
347
Bibliography
348
Bibliography
Kenny, Anthony. 1963. Action, Emotion and Will. London: Routledge &
Kegan Paul.
1983. Faith and Reason. New York: Columbia University Press,
comp. 1986. Rationalism, Empiricism, and Idealism. Oxford: Clarendon
Press.
1988. "Aristotle on Moral Luck." In Human Agency: Language, Duty,
and Value, ed. Jonathan Dancy, J. M. E. Moravcsik, and C. C. W.
Taylor. Stanford: Stanford University Press.
1992. Aristotle on the Perfect Life. Oxford: Clarendon Press.
Knaster, Stephen. 1984. "Chisholm, Deliberation, and the Free Acquisi-
tion of Belief." Philosophical Studies 46 (Nov.): 307-22.
Kornblith, Hilary. 1982. "The Psychological Turn." Australasian Journal of
Philosophy 60 (Sept.): 238-53.
1983. "Justified Belief and Epistemically Responsible Action." Philo-
sophical Review 92 (Jan.): 33-48.
1985. "Ever Since Descartes." Monist 68 no. 2 (Apr.): 264-76.
1993. "Epistemic Normativity." Synthese 94:357-76.
1994. "Naturalistic Epistemology and Its Critics." Invited paper, East-
ern Division meeting of the American Philosophical Association,
Boston, 30 Dec. Forthcoming in Philosophical Topics.
Kosman, L. A. "Being Properly Affected: Virtues and Feelings in Aris-
totle's Ethics." In Rorty 1980.
Kruschwitz, Robert and Robert C. Roberts, eds. 1987. The Virtues: Con-
temporary Essays on Moral Character. Belmont, Calif.: Wadsworth.
Kupperman, Joel J. 1991. Character. New York: Oxford University Press.
Kvanvig, Jonathan. 1992. The Intellectual Virtues and the Life of the Mind.
Lanham, Md.: Rowman & Littlefield.
Lazarus, Richard. 1991. Emotion and Adaptation. New York: Oxford Uni-
versity Press.
Leeper, Robert Ward. 1970. "The Motivational and Perceptual Properties
of Emotions as Indicating Their Fundamental Character and Role."
In Feeling and Emotions, ed. Magda B. Arnold. New York: Academic
Press.
Lehrer, Keith. 1974. Knowledge. Oxford: Clarendon Press.
1990. Theory of Knowledge. Boulder, Colo.: Westview Press.
Levy, Steven. 1977. "Defeasibility Theories of Knowledge." Canadian
Journal of Philosophy 7 (Mar.): 115-23.
Lewis, C. I. 1946. An Analysis of Knowledge and Valuation. La Salle, 111.:
Open Court.
1969. "The Rational Imperatives." In Values and Imperatives, ed. John
Lange. Stanford: Stanford University Press.
349
Bibliography
350
Bibliography
351
Bibliography
352
Bibliography
353
Bibliography
354
Bibliography
355
Bibliography
356
Bibliography
357
Bibliography
358
Name index
359
Name Index
Fine, Gail, 35, 36, 47, 49 Locke, John, 31, 33-35, 46, 86, 87, 163,
Firth, Roderick, 7, 30, 255 165, 204, 205, 262, 307
Foley, Richard, 30, 189, 260, 265, 279,
301 MacDonald, Scott, 199
Foot, Philippa, 92, 104, 105, 107, 129 Maclntyre, Alasdair, 44, 88, 97, 98, 135
Francis of Assissi, 95 Martineau, James, 211
Mavrodes, George, 62, 301, 309
Gettier, E. L., 30, 250, 253, 259, 264-267, Mayo, Bernard, 17, 86
283-285, 287-298, 304, 333 McDowell, John, 129, 156
Gibbard, Allan, 17 Meilaender, Gilbert, 107
Ginet, Carl, 41 Mencius, 211
Godlovitch, Stanley, 133 Merricks, Trenton, 292
Goldman, Alvin L, 11, 14, 43, 246, 304, Meyer, S. S., 60
309, 311 Montmarquet, James, xiii, 11, 13, 14, 21,
Goodman, Nelson, 49 63, 169, 174-176, 178, 184-190, 192,
Greco, John, 10, 11, 36, 102, 250, 326 193, 223
Moravcsik, Julius, 46-47, 139
Harman, Gilbert, 202
Hobbes, Thomas, 128, 169-171, 194 Nagel, Thomas, 70-72, 191, 261, 282
Hoitenga, Dewey, 100, 201, 272 Nelson, David, 49, 214, 215
Hookway, Christopher, 4, 65, 183 Newman, John Henry, 21, 225
Hume, David, 54, 62, 138, 139, 262, 263, Nietzsche, Friedrich, 97, 225
326 Nozick, Robert, 11, 117, 118, 120, 188
Hurka, Thomas, 80 Nussbaum, Martha, 18, 200
Hursthouse, Rosalind, 81
Hutchinson, D. S., 143 O'Neill, Onora, 18
Huxley, T. H., 325
360
Name Index
361
Subject index
act-based ethics, 8, 28, 51, 253 consequentialism, 7, 8, 10, 12, 14, 15, 25,
ad hoc definitions, 264-265 28, 42 ,43, 80, 198
agent-based theories, 81-83, 197, 202, courage, 80-96, 102, 107-109, 113, 120,
209-211, 338 124, 130-133, 138, 139, 146-151,
agent-focused theories, 16, 78-79, 232 155, 159, 160, 165, 180, 220-223,
akrasia (weakness of will), 119,150-152, 237, 244, 251, 256, 272
154-155/158 intellectual courage, 13, 60, 96, 150,
arete (see also virtue), 84, 139 174, 175, 185, 220, 223, 256, 269, 276
autonomy, 148, 159, 160, 171, 177, 191, creativity, 123-125, 136, 182-183, 187,
192, 212, 220, 221, 244, 247 190, 225
intellectual autonomy, 192, 244 curiosity, 135, 144, 148
benevolence, 82, 83, 89, 90, 130, 132, defeasibility conditions, 289-291
133, 135, 146, 165, 201-203, 2 O 9/ deontic concepts, 16, 78, 210, 231-233,
210, 252 235, 240, 271
deontological theories, 7, 8, 12, 14, 15,
carefulness, 98, 155, 177, 179, 185, 220, 18, 32, 34-37, 75/ 81, 210
221, 223, 244, 274, 293, 318 desires, 5, 9, 15, 23, 56, yj, 122, 130, 135,
character 139, 141, 143, 145-147/ 170/ 190/
moral character, 121-124, !4 2 -!44/ 206, 216, 270, 332, 337
152, 165 dogmatism, 171, 187, 191-194
intellectual character, 24, 152, 165, duty
234, 274, 275 epistemic duty, 1, 20, 30, 31, 34, 59,
trait (or state) of character, 56, 72, 73, 78, 231, 232, 241-243, 258, 260, 273,
81, 106, 129 298
cognitive contact with reality, 45, 167, moral duty, 17, 64, 232, 235, 236, 240,
168, 185, 267, 269, 270, 273, 282, 242, 243, 245, 254
298, 311, 316
cognitive integration, 24, 275 emotion, 3, 51-54, 56, 59, 106, 129-134,
compassion, 60, 83, 86, 91-94, 96, 100, 136, 137, 140, 163, 194
101, 106, 113, 116, 120, 124, 127, episteme, 46-48, 213
133-135, 146, 149, 156, 178, 198, epistemology and ethics, 2, 71, 79, 166,
239, 240, 249-252, 298 256, 267, 273
362
Subject Index
habit, 60, 103, 106, 116-119, 123-128, (the) mean, 96-97, 113, 152-153, 159-
180, 181, 196, 208, 213-215, 250, 160, 219-221, 226
276, 277, 279, 305, 306, 310, 311, moral strength, 119
327/ 328 motivation (see also knowledge, motiva-
happiness-based virtue theory, yy, 81, tion for)
197-202, 210 as component of virtue, yy, 115, 116,
honesty, 60, 85, 86, 89,115, 139, 146, 130, 133, 136, 165-167, 177, 178,
156, 158, 159, 222, 252, 257, 333 183, 198, 209, 210, 236, 237, 247,
hope, 2, 6, 16, 23, 51, ^, 58, 74, 79, 86, 248, 250, 252, 269, 270, 276, 332
113, 121, 146, 163, 166, 184, 201, in excess, 194—197
203, 231, 232, 260, 267, 274, 339 motivation-based virtue theory, yy, 82,
human nature, 12, 54, 105, 121, 164, 198, 202-211
227, 255, 278 motive, 129-132
humility, 86, 88, 113, 114, 148, 155, 159,
160, 220, 222, 223, 229 naivete, 162, 264
hypocrisy, 147, 162, 163 naturalized epistemology, 334
nosiness, 314-318
Idols of the Mind (Bacon), 306 nous, 10, 213, 217
illative sense (Newman), 225
integration, 23, 24, 275 obtuseness, 152, 155, 159
integrity, 19, 24, 25, 40, 89, 133, 137, 148, open-mindedness, 60, 96, 114, 133, 150,
155, 162, 165, 166, 174, 192, 197, 151, 155, 173, 175, 181, 185-190,
239, 247, 252 244, 253, 256, 269, 313
363
Subject Index
originality, 123-125, 155, 167,182, 183, 181, 182, 195,197, 217, 220, 224-
190 227, 237-239, 245, 268, 273, 302
parts of the soul, 140, 141, 143, 145, 212, (the) self, 28, 104, 123, 162, 255, 261, 333
213 self-control (see also temperance), 106,
pathe (see also feelings), 126, 143 108, 114, 129, 150, 151, 155, 158
perception, 34, 52, 54, 63, 153, 257, 277- self-deception, 64, 65, 147, 152, 154
283, 292, 301, 327 (the) senses, 5, 26, 54, 164, 204, 277,
perseverance, 13, 21, 60, 85, 95, 98,108, 278
114, 115, 133, 148, 150, 152, 155, sincerity, 60, 106, 113, 146, 173, 257
156, 159,174, 179, 196, 220, 221, skepticism, 229, 238, 260, 278, 307, 331
223, 274, 318 skills
phronesis, 9,10, 78, 79, 97,119,120, 140, intellectual skills, 109, 114-117, 151,
150,180, 211-213, 215, 217-230, 157, 161, 178
239, 240, 246, 247, 275 moral skills, 113, 115, 150
practical wisdom (see also phronesis), sophia, 10, 213, 217, 218
78, 97, 137, 140, 150, 166, 211-213, success (component of virtue), 137, 176
216, 218-221, 226, 228-230, 245, sunkatathesis, 218
supererogatory, 125, 155, 233, 253, 254
294, 337
praise and blame, 6, 60, 71, 125, 126 techne, 110,111,140, 213
praiseworthiness, for doing an act, 247, temperance, 85, 86,104,108,114,120,
248 138,139,146, 216, 249-251
pride, 60, 88,132, 152, 159, 163, 224, temptation, 34,104,119,129,151
314-316, 318 theoretical wisdom (see sophia)
proper function theory, 259, 299, 309, thoroughness, 98, 108, 114, 152, 155,
326, 327, 330 160, 177, 185, 196, 221, 223, 245,
prudentia (see also phronesis), 212,214,230 256, 269
pure virtue theory, 16, 77-80, 166, 197, (the) transformation machine (Nozick),
200, 210, 231, 232, 238, 241, 245, 116-122
258, 334/ 336 trust, 21, 46, 146, 150, 151, 160, 161, 208,
226, 252, 319
range of the moral, 255
types of virtue theory, JJ
rationality, 22-24, 30, 56, 69, 73, 142,
149, 190, 225, 229, 230, 255, 260, understanding (as an epistemic value),
278, 301 43-50
reliabilism, 4, 8, 10-12, 14, 25, 27, 28, 42,
62, 198, 259, 262, 268, 284, 299-301, virtue
304, 305, 307-309. 3ii/ 312, 319, act of virtue, 78, 91, 232, 248-253,
320, 323, 325, 332 268, 269, 270, 279, 295, 296, 300
reliability, 12, 25, JJ, 133, 150-152,160, Aristotle's distinction between moral
165, 166, 181, 184-187, 193, 197, and intellectual virtue, 137-158
209, 235, 247, 290, 305, 308-310, components of, 78, 136
312, 314, 315, 320, 323, 324, 328 connections between moral and intel-
responsibility, 1, 4, 11,12, 36, 59, 61, 66- lectual virtues, 158-165
68, 116, 125, 173, 174, 176, 257-258, in excess, 196
260, 319, 325 general account of virtue, 134-137
right act, 7, 16, 28, 41, 69, 77-83, 210, intellectual virtue, components of,
231-239, 241, 244, 247, 248, 258, 272 165-197
rules, 7, 18, 123, 150, 155, 167, 171, 176, unity of the virtues, 156, 163, 212, 252
364
Subject Index
365