Guy Axtell Work

Virtue Epistemology:
Essays on Epistemic
Virtue and Responsibility
Abrol Fairweather
Linda Zagzebski,
Editors
OXFORD UNIVERSITY PRESS

10
epistemic luck in light
of the virtues
Guy Axtell
The problem of restoring integration and cooperation between man’s

beliefs about the world in which he lives and his beliefs about the
values and purposes that should direct his conduct is the deepest
problem of modern life. It is the problem of any philosophy that is not
isolated from that life.
— John Dewey
I. Luck, Control, and Virtue Theory

In its broadest context, as part of contemporary virtue theory, virtue epistemology
can be understood as one side of an integrated account of the human being as both
a knower and a valuer. What attracts me to it, personally, is my sense of the frag-
mented state of reason in contemporary western culture, a condition that is to some
extent an inheritance of modern philosophy. Virtue theory promises a response to
what Dewey, quoted above from The Reconstruction of Philosophy, presents as a cen-
tral problem of modern life and even as the theoretical chasm for philosophy to
bridge: the chasm between our beliefs about the world and our beliefs about the
virtues, rules, and purposes that should direct our actions, both individually and col-
lectively. I use these metaphors of bridge and chasm in connection with Dewey’s
passage, because this is an essay about bridges: bridges, firstly, between epistemology
and ethics; and secondly, between virtue epistemologists themselves since they, like
others, invariably bring with them to the discussion of knowledge their own diver-
gent interests in explanation.
Our approach will be through an analysis of the various possible meanings of
“epistemic luck” and their impact on epistemology. Luck presents a complex of
problems that impact both ethics and epistemology, problems that must be balanced
with and related to our conceptions of responsible agency. The presence of luck in
158
our cognitive as in our moral lives shows that the quality of our intellectual charac-
ter may not be entirely up to us as individuals, and that our motivation and even our
ability to desire the truth, like our moral goodness, is fragile. But the fragility of our
character makes it a more, not a less, important topic for philosophical inquiry. As
Claudia Card and Martha Nussbaum have both pointed out, the impact of luck on
our lives can also add depth to our understanding of responsibility and increase our
sense of the worth of the virtues.1
Luck also affects our understanding of the reliability of our belief-forming cog-
nitive processes (hereafter BCPs). The reliability of these processes is in one sense
radically subject to luck, to a “natural lottery” as some would describe it, or to the
control not of ourselves but of nature. To the extent that our intellectual habits and
dispositions can be influenced by choices under our control, however, their reliabil-
ity might be taken as an achievement. But to what extent is this? There is keen de-
bate among epistemologists about whether our intellectual dispositions— those that
affect our BCPs— are less subject to voluntary control than are the moral disposi-
tions that shape our actions. Views on this matter are bound to impact our concep-
tions of both ethical and epistemic agency. It is a commonplace to hold that the more
externalist an epistemologist’s orientation, the more tolerance they have for luck in
our cognitive lives.
Here is a first indication of differences among virtue epistemologists: those in-
clined toward reliabilist views tend on the whole to be more skeptical concerning is-
sues of responsibility for one’s intellectual character than those inclined toward what
has recently come to be called “responsibilist” views.2 For instance, in a recent essay,
“Moral and Epistemic Virtue,” Julia Driver, who takes a stance she calls objective
consequentialism3 allied with “the externalist extreme” of a thoroughgoing reliabil-
ism, writes that “Luck is a fact of life, and it may be a form of moral and intellectual
virtue to humbly accept that there will be such limits to success no matter how well
the agent is justified.” Those who think epistemic agents have little control, or only
very indirect control over their intellectual dispositions and habits, lean toward re-
liabilism even if, as Linda Zagzebski has suggested, there may be few necessary con-
nections to be found between externality and lack of control, or between internality
(awareness) and control.4
The issue of responsibility for character (which would appear to follow from our
capacity to control or affect our own traits) is a difficult one, and all the more so
when we extend the term “character” to include intellectual habits and dispositions
in addition to moral ones. The analogies between ethics and epistemology are un-
doubtedly on stronger ground when one compares the degree of control we enjoy
over ethical and intellectual character traits than when one attempts to compare our
control over our actions with our control over our beliefs. But terms of character such
as “disposition” and “habit” are used to cover a wide range of attributes not well de-
marcated by their openness to conscious or to voluntary control. Moreover, episte-
mologists often invoke different vocabularies when issues arise about control. Some
speak without worry about a “weak doxastic voluntarism,” from which control and
hence responsibility flows, while others speak as if epistemic agents are about equally
responsible or non-responsible for each of their beliefs (certainly not something we
epistemic luck in light of the virtues 159

hear said about each of our actions!). Both vocabularies can contribute to conflation
of the descriptive and evaluative tasks of epistemology. The former is odd if it as-
sumes one needs to support doxastic voluntarism in order to speak about control over
traits of intellectual character. The latter is odd in light of the generally accepted
points that “S believes p” is attributable to an agent S (with the possible exception of
perceptual beliefs) only on the basis of a prior assumption that the agent S does (or
would) accept the proposition that p, and that the “justification” for acts of accept-
ance is not an all-or-nothing affair, but rather comes in degrees.
Hilary Kornblith avoids this conflation between the normative and the descrip-
tive by distinguishing between “actions” and “processes,” that is, between actions that
we ought to perform in order to make the processes by which we arrive at our beliefs
more conducive to truth, from those processes (BCPs) themselves. A related sug-
gestion, to which I will try to adhere, is that in order to demarcate the proper appli-
cation of normative concepts of agency like “rational,” “responsible,” “praiseworthy,”
etc., we should distinguish between the acceptance of a proposition and the state of
belief. Whichever distinction is preferred, formulating criteria for evaluating an
agent’s acceptance of a proposition involves these normative concepts of rationality
and responsibility.
In general I find differences over the issue of control over intellectual habits and
dispositions to run less deep among virtue epistemologists than among epistemolo-
gists at large. Part of the reason for this may be that virtue epistemologists share a
broad range of assumptions that bring them together with interest in issues— both
the descriptive and evaluative — affecting our understanding of epistemic agency.
Moreover, one opinion virtue epistemologists widely hold is the present need for a
“dual component” conception of knowledge, sometimes also called a “mixed” ex-
ternalist account of knowledge.5 This is one that integrates constraints on an agent’s
faculty reliability with constraints on the agent’s responsibility in gathering and pro-
cessing evidence. Kornblith divides the latter into “internal coherence” (desire for
epistemic integration) and “action-theoretic” constraints (characterized in terms of
desire for and effort in attaining the truth). To briefly cite three further instances,
Ernest Sosa writes that in his virtue perspectivism, a proposition is evident or known
(from the K point of view) to a subject “only if he is both rationally justified in be-
lieving it and is in a position to know (from K’s point of view) whether it is true.”6 In
her Virtues of the Mind (1996), Linda Zagzebski, uses the term “dual-component” to
describe her mixed account of believing from virtue, which includes both a motiva-
tional and a success component. And John Greco says that his account is intended to
satisfy both the “subjective justification” and the “no accident” conditions on knowl-
edge by drawing on the resources of virtue theory.7
Mixed accounts, I think, present distinct advantages over their competitors be-
cause they insist on the complementarity of two concerns that have driven recent
developments in epistemology: what it means to have beliefs formed by reliable
cognitive processes, and what it means for agents to be responsible and well-
motivated in their acquired cognitive habits and dispositions. But differences per-
sist among virtue epistemologists as elsewhere in philosophy and can at times be
quite pronounced. Ernest Sosa, John Greco, and Alvin Goldman describe their
160 virtue epistemology

virtue epistemologies as specific forms of reliabilism. Lorraine Code formulated
the distinction and contrast between reliabilist and responsibilist virtue episte-
mology in her Epistemic Responsibility (1987), arguing that “the concept ‘responsi-
bility’ can allow emphasis upon the active nature of knowers/believers, whereas
the concept ‘reliability’ cannot.”8 I will draw a distinction, as I have elsewhere, be-
tween those inclined toward virtue reliabilism and those inclined toward virtue re-
sponsibilism.9 The differences that these labels identify should not be exaggerated,
since these two hardly represent anything like the extremes of externalism and in-
ternalism of the recent past in epistemology. Indeed the issues of first-person con-
straints such as “internal access” to the grounds of one’s belief seems not to be the
dividing line between virtue reliabilist and responsibilist accounts of justification.
Neither thinks of the nature of justification as turning on that issue, so closely as-
sociated with the received definitions of internalism and externalism.10 But I will
take the descriptions of an author as a virtue responsibilist or virtue reliabilist to
have a useful application in pinpointing differences of interest and attitude, and to
often be applicable to particular authors even where a “mixed” account may be ex-
plicitly advocated.
I hope to show how an analysis of epistemic luck can lend substantial support to
mixed externalist epistemologies. But the analysis of luck that will be offered is
unique, and first I want to show how problems that surround epistemic luck have
in fact served to divide reliabilist and responsibilist accounts in the recent past. If
we concern ourselves first with the issue of the role of the will and responsibility in
propositional acceptance, we find that it is the responsibilists who are most keen to
explain the ground for evaluative judgments of epistemic agents: They remind us,
for instance, that the Aristotelian account (which allows attributions of responsibil-
ity for intellectual habits and dispositions) is not committed to a strong voluntarism
(Hookway); that only weak voluntariness is at issue in responsibility (Montmar-
quet)11; and that that there is a range of voluntariness in belief as in action—where
“accident and intentional action are two ends of a spectrum of conscious control”
(Zagzebski).12
Reliabilists for their own part are likely to point out, as does Driver, that “there
is a built-in responsibility for action” that does not seem to be paralleled in episte-
mology, where she believes we are “more forgiving” of defects. But this contrasting
emphasis is by no means to reject responsibility for character tout court, nor by itself
does it signal any deep division between our two sub-groups within virtue episte-
mology. Underlying the mixed account of knowledge is what Thomas Nagel would
term a “compatibilist account” of cognitive freedom and causality. This is the epi-
stemic analogue of the view in ethics that one may be responsible for what one does,
even if what one does depends in important ways on factors not within one’s direct
control. In epistemology as in ethics, Nagel tells us, compatabilism “would leave
room for the ordinary conditions of responsibility— the absence of coercion, igno-
rance, or involuntary [actions]” without excluding from the analysis all influence of
factors that are a matter of luck.
Let me clarify this last point. In his influential essay “Moral Luck,” Nagel frames
the relevant epistemic analogue to his compatibilist position in ethics:

The corresponding position in epistemology would be that knowledge consists of
true beliefs formed in certain ways, and that it does not require all aspects of the
process to be under the knower’s control, actually or potentially. Both the correct-
ness of these beliefs and the process by which they are arrived at would therefore
be importantly subject to luck. The Nobel Prize is not awarded to people who
turn out to be wrong, no matter how brilliant their reasoning.13
Such a compatibilist stance in epistemology, to continue the analogy, is simply one

that would leave room for the ordinary conditions of epistemic responsibility—
primarily the absence of coercion—without excluding all influence of factors that are
a matter of epistemic luck. Yet this shared compatibilism aside, we can certainly
identify issues where the reliabilist and the responsibilist are more sharply divided.
Let us examine these. One such issue is how far the analogy between ethical and
epistemic luck can or should be pushed. Zagzebski has commented that the recent
interest in epistemic luck “makes the attempt to model epistemic evaluation on
moral evaluation easier to do,” because it indicates that “it is much too facile to dis-
tinguish evaluation in the two areas on the ground that we control the one but not
the other.”14 Our intellectual habits and dispositions, she holds, are generally not less
subject to voluntary control than are their ethical counterparts, although again a
range of voluntariness must be acknowledged in regard to both. Not everyone
would share these responsibilist views, and I use them to illustrate that it is much
easier to show how the theoretical differences between virtue reliabilists and re-
sponsibilists are manifested than it is to locate and sort out the interests in explana-
tion that drive their divergent epistemological accounts.
In the most widely discussed example, overt differences are manifested in
sharply divergent definitions of the intellectual virtues themselves. Reliabilists define
them by their conduciveness to the production of true beliefs, allowing genetically
endowed faculties and powers to be counted among the virtues; responsibilists tend
to take them more restrictively in terms of acquired habits and dispositions— traits
internal to agency that are the proper object of praise and blame. Responsibilists may
also reject the attempt to define the virtues consequentially by what they do, in favor
of viewing them as conceptually prior to and as partly constitutive of the epistemic
goal itself.15
The differences between reliabilism and responsibilism manifest in another way
of which we should take notice: in debate over the possibility and desirability of a
unified account of ethical and epistemic virtue. Julia Driver usefully distinguishes be-
tween the claim that an account of the virtues is unified and the claim that the virtues
themselves are unified. What sense should we give to the latter, more contentious
claim? If it is understood as the robust classic Socratic/Aristotelian claim that a per-
son cannot have the intellectual virtues without the moral virtues (and conversely),
then it probably has few supporters.16 But there are various senses of a unified ac-
count of the virtues that many responsibilists, and only responsibilists, want to de-
fend. The strongest of these may be the subsumption thesis that Zagzebski argues for
in Virtues of the Mind (1996), where she “subsumes the intellectual virtues under the
general category of the moral virtues, or aretai ethikai, roughly as Aristotle under-
stands the latter.”17

Less strong, Montmarquet provides an account of the relationship between the eth-
ical and epistemic virtues that is critical, on the one hand, of Zagzebski’s strong “as-
similationist position,” and on the other, of the strong “externalist anti-assimilationist
position” which Driver may be thought to hold. The model Montmarquet develops
for the relationship between the virtues does not arrange them by content or do-
main, but rather by the susceptibility of each disposition to direct control. Central to
his understanding of control is the agent’s efficacy in exemplifying a particular
virtue at will, i.e., where “trying” to exemplify the virtue is sufficient for exemplify-
ing it. This innovative approach issues in a model reflecting a threefold distinction
and “a hierarchy of evaluatively relevant qualities of persons, which largely cuts
across the supposed epistemic/moral divide”:
At the lowest level will be those “externalist” traits subject only to our indirect
control. At the middle level will be the “internalist” epistemic virtues and perhaps
certain broadly moral virtues. At the summit, will be certain narrowly moral
traits [picked out by the fact that we are able both to exemplify them and to ex-
emplify their contraries at will].18
A unified account of the virtues may be associated with a still-weaker thesis, but
one that like the preceding claims also invites “cutting across” some of the discipli-
nary boundaries between ethics and epistemology. This thesis is one that associates
virtue epistemology with the development of a general theory of value. Montmarquet
and Zagzebski have both supported a sense of this thesis in their book-length treat-
ments of virtue epistemology. I have as well, in previous essays that connect the ap-
peal of virtue epistemology with a broader metaphilosophical reorientation in phi-
losophy. Responsibilists, then, side with Zagzebski’s general claim in Virtues of the
Mind that “the unification of moral and epistemic evaluation is a welcome advan-
tage.” They also tend to resist the drift of reliabilism toward a view of epistemology
as a disparate field that studies various but unrelated ways in which beliefs can be
justified, and thereby side with Paul Bloomfield’s general claim that “it would be
preferable to find ‘justification’ to be helpfully univocal.”19 By contrast, those of re-
liabilist orientation tend to be more sanguine about the prospects of unification in
any of the senses we have discussed, and so we find Driver arguing “that a unified
account of virtue is neither doable nor desirable.”20
Let me give just one substantial example of how this debate plays out, using
Zagzebski and Driver to exemplify sharply divided responsibilist and reliabilist
viewpoints. Zagzebski rejects any sharp distinction between moral and other sorts
of evaluation, which she thinks is a residue of the faulty Kantian view that the good
will (seen as radically free in its noumenal existence) is the only proper object of
moral evaluation. But her reliabilist critic may perceive her [in a sense analogous to
Bernard Williams’s critical perspective in Moral Luck (1981)], as actually involved in
a neo-Kantian attempt to immunize epistemology from luck by exaggerating the
freedom and control of the agent (in this case the epistemic agent). While Zagzeb-
ski, like Williams, criticizes the Kantian attempt to ground moral evaluation in
something that is luck-free, she also takes it as a definite advantage of an epistemic

theory that it “reduces the component of luck.” Taking the reduction of the impact
of luck as an advantage may seem unobjectionable, but in the context of Zagzebski’s
approach it reflects her attempt to develop a “motivation-based virtue theory” that
places goodness within motivational structure and identifies certain motives as in-
trinsically good. Seeing Zagzebski’s responsibilist account in light of this project
makes it difficult to separate Driver’s complaint that a motivation-based virtue the-
ory “doesn’t provide a criterion” and leaves “mysterious” what makes a motive good
or bad, from Williams’s notable objection addressed to the Kantian, that “the dis-
positions of morality, however far back they are placed in the direction of motive
and intention, are as ‘conditioned’ as anything else.”21
To summarize, we have seen that virtue epistemologists often explicitly disagree
over the understanding and identification of the virtues, over the issues of responsi-
bility for character, over the strength of the analogy between ethical and epistemic
evaluation, over the prospects for a unified account of the virtues, and over the pos-
sibility of a univocal sense of epistemic justification. But if these issues, as we said,
merely mark spots where the pot boils over—where explicit theoretical differences
persist and main lines of division are manifested among virtue epistemologists—
then we should now be enticed to look further and to seek the source of these dif-
ferences in their divergent interests in explanation. The problem of epistemic luck is
an especially promising area to explore in seeking these deeper sources.
II. Forging Bridges: Riggs’s

Conjunctive Approach
Is knowledge compatible with epistemic luck? An initial step in answering this fun-
damental query is to distinguish questions about the extent of epistemic luck from
questions about the kinds of epistemic luck thought to have an impact on human
cognition. Luck in all its senses appears to involve discontinuities or lack of control
either over a process or over its outcome. And as Nagel points out, “epistemological
skepticism arises from consideration of the respects in which our beliefs and their re-
lation to reality depend on factors beyond our control.”22 Epistemic luck is a generic
notion, and some kinds of luck may simply be irrelevant to an account of justifi-
cation. But it is a highly contested issue which types present a relevant challenge to
knowledge claims. Both groups of virtue epistemologists concede that we cannot al-
ways avoid error without luck, but they typically see the relevant discontinuity, and
hence the kind of luck involved, very differently.
In an innovative recent essay, “What Are the Chances of Being Justified?”
Wayne Riggs argues “that both truth-conducivist and responsibilist conceptions of
epistemic justification are directed at disallowing a particular kind of chance (true)
belief from counting as epistemically sanctioned.”23 Many of Riggs’s examples are
drawn from virtue epistemologists, making his analysis especially pertinent here.
“Chance” is selected as the common term characterizing conditions that might pre-
clude instances of true belief from constituting knowledge, and the author then goes
on to explicate and define two relevant kinds. A virtue reliabilist (Riggs uses the

term “truth conducivist”) might define the justification condition on knowledge in
such a way as to preclude the kind of chance that Riggs simply calls epistemic luck:
“S’s coming to hold a true belief, p, is a matter of epistemic luck for S to the extent that
p is unlikely, given that it was produced by process R.”24 I will use Riggs’s terminol-
ogy in this section, except that I will refer to this first kind of chance as coincidence,
allowing us to preserve a broader and more conventional use for the term “epistemic
luck” in the sections that follow. Riggs’s definition of coincidence reflects the objec-
tive link to truth that is characteristic of the truth-conducivist account of justifi-
cation. The motivation behind reliabilism centrally involves disallowing beliefs that
are merely “lucky guesses” or coincidental truths from counting as knowledge.
Consideration of Gettier cases, for instance, has strongly pushed toward external
constraints on knowledge that can prevent a justified, true belief from counting as
an instance of knowledge if its truth is merely a coincidence and is not “linked” with
its causal ground in an appropriate (epistemizing) way.
By contrast, the kind of chance that responsibilists want to preclude from count-
ing as knowledge is what Riggs refers to as “epistemic accident.”25 Citing John
Greco and Hilary Kornblith as examples, he takes their accounts of justification as
primarily concerned with the evaluation of epistemic agents or their practices. The
responsibilist begins from the agent’s motives, and so insists that a belief is justified
only where the intentions to have true beliefs and to avoid false ones have played ap-
propriate grounding roles in the agent’s belief-forming practices. What the respon-
sibilist means by an appropriate grounding role Riggs takes to be a certain kind of
causal role, and so “S’s coming to hold a true belief, p, is an epistemic accident for S if
coming to hold p was not (sufficiently) caused in an appropriate way by S’s intention
to have true beliefs.”26
The responsibilists’ motivation for placing such a constraint precluding epi-
stemic “accidents” from counting as knowledge is partly explained by their intuition
that the justification condition demands beliefs held (or propositions accepted) not
only in conformance with our epistemic norms, but from or in light of such norms.27
An agent’s consciously held belief that satisfies this constraint will be one that is
guided by his taking a support relation to hold between reason state r and his accept-
ance of p. This concern to preclude epistemic accidents, as Riggs points out, also ex-
plains the enthusiasm that responsibilists have shown for cases of deviant causal
chains, where a reliable BCP has been in operation, yet a belief produced by it fails
to “link up” with the agent’s intentions or subjective justification in an appropriate
(epistemizing) way. Summarizing Riggs’s analysis briefly, coincidence (or epistemic
luck for Riggs) is primarily a matter of success despite low-likelihood, given the
agent’s actual process, while accident is primarily a matter of success despite inten-
tional inefficacy, given the agent’s intentions.
Riggs’s most challenging claim in assessing the upshot of his analysis is that “The
responsibilist and truth-conducivist conceptions of justification define distinct epi-
stemic evaluations and so are not in any interesting sense rival notions.”28 He views
their divergent constraints on chance as mutually consistent, deriving from “an
identical epistemological concern.” From this view of their consistency, he proposes
an intriguing conjunctive formula for the justification condition on knowledge. The

necessary and sufficient conditions for knowledge are according to this formula true
belief, and a revised third (justification) condition, where the conjunction of the for-
mer two conditions “is not a matter of chance” in either of the two senses we have
specified. This conjunctive definition is interesting in its own right and may prove
more resistant to the usual roundup of counter-example cases than definitions that
do not include both the reliabilist and responsibilist constraints. He proposes a new
focus for epistemology in line with his approach, that of working out “an adequate
account of epistemically responsible belief, as well as an account of sufficiently truth-
conducive belief.” As we have previously seen, such a twofold approach is also
indicative of the mixed account of justifications widely advocated among virtue
epistemologists.
The compatibility that Riggs asserts to obtain between reliabilism and responsi-
bilism is quite appealing; it must be correct in some sense for a mixed account of
justification to be plausible. The motivation for his conjunctive strategy also seems
clear enough: reliabilist justification prevents a true belief that will constitute knowl-
edge from being a coincidence, but not from being an accident; and responsibilist
justification prevents a true belief that will constitute knowledge from being an
accident, but not from being a coincidence. However, at the same time I have reser-
vations about Riggs’s conjunctive strategy for defining knowledge — that is, the
simple adding of the one constraint to the other in order to render the account of
justification more adequate or ‘complete.’ This strategy may be shown to be prob-
lematic if it invokes assumptions that make it more rather than less difficult to re-
spond to the challenge of skepticism. The one assumption I would focus on here is
that of the incompatibility of luck and knowledge. We should notice, after all, that
the analysis he presents puts us nowhere closer to explaining the commonplace be-
lief with which we began, that the reliabilist takes herself to be more amenable to
epistemic luck than the responsibilist. This point should help us to see that using
stipulative definitions excluding certain forms of chance cannot provide a suitable
answer to the entire complex of problems associated with epistemic luck. As discus-
sions in ethics have rightly pointed out, luck is not entirely a conceptual matter, but
depends upon the make-up of ourselves, our world, and our relationship to it. If we
all rely on luck in our intellectual lives, then we cannot merely stipulate that knowl-
edge excludes it— not, at least, without simply begging the question against the
skeptic by assuming from the outset that we know what we think we know.
I do not mean to burden Riggs with holding the incompatibility thesis in its most
objectionable form, the claim that knowledge is incompatible with luck simpliciter;
he has defined the specific senses to be excluded, and it may be only a limitation of
his focus that he does not address other senses of luck which may be compatible with
knowledge. But an adequate account must go beyond showing that excluding epi-
stemic coincidences is compatible with excluding epistemic accidents, to a positively
stated account both of this compatibility and of epistemic justification. This would
be one that explains in positive terms what it means to be “in a position to know”
and what it means to be a “well-motivated and responsible” agent. Riggs hints at
such a positive account in saying that the constraints against the different kinds of
chance “arise from an identical epistemological concern,” and that the parties to the

debate share a common conception of what epistemic justification is for (that is, a
kind of filter of knowledge claims). Related claims that Riggs makes— that the di-
vergent constraints on justification represent value-charged demands for the intelligi-
bility of knowledge, and that neither can claim logical priority over the other— also
hint at arguments for their compatibility. Yet these comments far outstep the argu-
ments Riggs actually provides in his article.
Such limitations to his analysis are not surprising, I want to say, because there is
another whole side to the issue of luck we have not yet explored. We begin to engage
it in the next section by directing critical attention to the thesis of the incompatibil-
ity of knowledge with luck. There I will briefly lay out an argument against the in-
compatibility thesis, showing its self-defeating nature. Then in Section IV, I will fur-
ther elaborate certain senses of “luck” to which the reliabilist and the responsibilist
must each respectively acquiesce once the incompatibility thesis is abandoned. In ac-
knowledging kinds of luck that cannot be excluded but must instead be part of any
adequate account of knowledge, I hope to build upon the idea of constraints on
justification as value-charged demands for the intelligibility of knowledge, and
thereby to clarify the basis for a positive account of the compatibility of reliabilism
with responsibilism.
III. Skepticism and the Incompatibility Thesis

In this section I make use of another thought-provoking essay on epistemic luck, “Is
Luck Compatible with Knowledge?” in which Mylan Engel argues that “both in-
ternalist and externalist epistemologies lead directly to skepticism when they are
coupled with the incompatibility thesis.”29 Why must both internalist and external-
ist epistemologies reject the incompatibility thesis? Because the view that luck sim-
pliciter is incompatible with knowledge entails skepticism. But why in turn is this
the case? Because each of these theories can in fact be shown to entail a kind of luck.
Luck, as Nagel pointed out for us, has always provided the skeptic with a foothold;
unrealistically denying luck’s role in knowledge may provide him a ladder! For the
sake of clarity I will distinguish the kinds of epistemic “chance” asserted in the pre-
vious section to be incompatible with knowledge from forms of epistemic “luck”
whose nature we have yet to explore.
Engel’s essay utilizes the distinction between internalist and externalist episte-
mologies rather than the more subtle distinction between responsibilist and relia-
bilist epistemologies, which is our real concern. We will need to follow Engel in this
for the present, and in order to underline differences, I will typically mean by the
term “externalism” its more “pure” or non-mixed versions unless I specifically refer
to a “mixed” version. We can also pick up the metaphors with which we began, by
examining these kinds of luck in terms of the ideas of perceived crucial discontinu-
ities or inferential lacuna (“gaps”), and responding to these, attempt to build theo-
retical connections (“bridges”).
Gaps are well-acknowledged in ethics, a prominent example being the gap be-
tween being virtuous and living well, according to Aristotle: Uncontrolled happen-

ing (moral luck) can step into this gap, impeding the person of arete from finding its
proper fulfillment in eudaimonia. Now it is clear that for internalist accounts in epis-
temology, there must remain a gap between justified belief and truth. Internalists
maintain that epistemic justifiedness is exclusively a function of the cognizer’s in-
ternal states— belief, memory, or perceptual states. No internalist theory can pro-
vide a conceptual bridge or connection between justification and truth, as Gettier-
cases purport to show. According to Engel then, “no internalist theory can eliminate
the role luck plays in a person’s coming to have a true belief. So if luck really is
incompatible with knowledge, then no internalist epistemology can give rise to
knowledge.”30 Since the experience of those in a demon world is subjectively in-
distinguishable from our own, the internalist is content to construe demon world
inhabitants justified in their beliefs if the quality of their efforts to attain truth
and avoid error have been beyond impeachment. The internalist takes beliefs as
virtuously held on the basis of evidence available to a situated agent, and when it
is revealed that those beliefs are nevertheless false due to the demon’s systematic
deception, she can merely acquiesce to a form of epistemic luck, conceding that
luck must intervene to turn our justified beliefs into true beliefs, and hence into
knowledge.
Externalist theories may appear to be on much stronger ground, but as Engel
points out, if this is so, the reason cannot be that the externalists need not admit luck
into their accounts. On the contrary, “externalist epistemologies with truth-connected
theories of justification simply replace one kind of epistemic luck with another, for
while it is not a matter of luck when a [process-reliabilist]-justified belief turns out to
be true, it is a matter of luck when a belief turns out to be [process-reliabilist]-
justified.”31 Engel terms the kind of luck to which the internalist must acquiesce
veritic luck, and that to which the externalist must acquiesce evidential luck. Veritic
luck might most simply be understood as luck with respect to the output of our
BCPs and evidential luck as luck with respect to the empirical or evidential input that
our BCPs must work from and with.
A demon-world case can also be used to illustrate evidential luck. Consider your-
self in relationship to your doppelganger in the demon world and reflect on the fact
that the two of you live phenomenologically indistinguishable cognitive lives. Since
there is no discernible difference between your worlds, you must conclude that it could
have just as easily been you who were the one with the unreliable belief-forming
cognitive processes. Given your situated place within this scenario, there is no evi-
dential basis that could possibly serve you as a guide from which to reach any other
conclusion. If this is so, then it is still a matter of luck that, in the context of this sce-
nario, it is you and not she who has the process-reliabilist true beliefs.
According to the foregoing argument, the externalist must acknowledge that it
is always a matter of evidential luck when reliably produced beliefs turn out to meet
the subjective conditions that internalists would place upon knowledge. And the in-
ternalist must acknowledge that it is always a matter of veritic luck when our beliefs
turn out to meet the objective conditions an externalist would place upon knowl-
edge. But is this a cogent line of argumentation? Clearly Engel has allowed the very
basis for the characterization of luck to differ in his demon-world scenarios. But

why should he not? Perspective matters! If it takes an externalist perspective to
point out the kind of luck which internalism entails, it seems valid to allow the same
in reverse, by allowing the evidential factors rather than matters of objective fact to
be the ones that may vary in the examples. The perspective focusing on the evil-
demon’s revelation to me illustrates that it must always be a matter of luck from an
external or factual perspective when an internally justified belief turns out to be true.
The perspective focusing upon my relationship with my doppelganger illustrates
that it must always be a matter of luck from an internal or evidential perspective that
one of us, myself or my double, should in fact be the one possessing sound evidence
for normally functioning BCPs to use as input.
But are these kinds of luck of equally significant import for epistemology? In
Gettier and other types of cases which might have been discussed in connection to
the kinds of “chance” examined in the previous section, our intuitions tell us that an
agent’s justification is undermined by certain facts about his situation that are be-
yond her ken. In the eyes of their devisers, at any rate, such examples exploit factual
defeaters to an agent’s knowledge claims that, once they have done their dirty work,
leave the would-be knower only with what Matthias Steup calls nonepistemizing
justification. But is that also the case with the kinds of demon-world scenarios we are
now considering? That seems not as clear. We are now encountering forms of epi-
stemic luck that have more relevance to the general challenge of philosophical skep-
ticism than to the statement of specific conditions on knowledge. This is partly why
I separate them here from Riggs’s two forms of chance discussed in the previous
section.
Our two forms of luck both have important epistemic implications, even if they
differ from the forms of chance we earlier investigated. Engel himself argues that
only veritic luck has epistemological import, thereby giving the strong advantage to
externalism. I disagree here: Evidential luck, too, is a kind of luck that concerns the
crucial relationship between the agent and the known fact; it is not, any more than
is veritic luck, a kind that concerns merely the existence of the known fact or the ex-
istence or abilities of the man who knows. I will need to say more on this later, but
I take input and output luck to be on generally equal footing in this regard. It takes
the other’s perspective to point out the type of epistemic luck that each account must
admit, and internalists and externalists both think that once pointed out, the kind
of luck systematically implied in their adversary’s account represents a deep fault
in their approach. Moreover, both perspectives contribute to explicating the pre-
theoretical notion of luck as referring to factors affecting that crucial relationship
(between the agent and the known fact) that remain beyond our ken or, in other
words, beyond human control.
Since internalism and externalism are defined by Engel in the usual mutually ex-
clusive and exhaustive way (exhaustive except for the option that would radically
eliminate justification as a condition on knowledge altogether), we are left with the
implication that luck is deeply involved in knowledge on either account. So to re-
turn to our main line of argument, if luck simpliciter really were incompatible with
knowledge, then both theories would make knowledge impossible and hence lead
directly to skepticism. This argument I believe is cogent, and it leaves us with the

following options: (1) embrace skepticism, (2) embrace a non-justificationist account
of knowledge, or (3) reject the incompatibility thesis.
If we accept this scenario, then the last option appears most promising. This
means that, again following Engel’s lead, we must try to “reconcile the rather strong
intuition that epistemic luck [chance] is not compatible with knowledge with the
equally evident observation that it must be.”32 Both accounts have been only partially
successful in shielding knowledge from a kind of luck that its adversaries see as un-
dermining its claim to adequacy. Of course, there are certainly kinds of luck that are
simply irrelevant to the question of knowledge. But neither of the types of luck
Engel discusses appears to be of such a type, and the route of a non-justification
form of externalism is unappealing. Nor should we assume that the presence in any
degree of one or the other kind of epistemic luck is sufficient to undermine knowl-
edge, an assumption which might lead us to embrace skepticism. Consistent with
Engel’s third option of rejecting the incompatibility thesis, the point we should stress
is rather that on both internalism and externalism, understood as mutually exclusive
accounts, it is necessary for luck to step into a perceived “gap” in order for justified
belief and truth to link up in an epistemizing way. Both theories, in this sense, re-
main essentially “incomplete” in their response to skepticism.
We have now argued that the incompatibility thesis is mistaken. A strictly con-
junctive account, we should now be able to see, is unrealistic as an attempt at bridge
building. It requires too much of human cognizers, because it excludes from the
analysis the influence of factors that are a matter of luck. An adequate account of
epistemizing justification cannot be stated in Riggs’s negative fashion of merely ex-
cluding luck from knowledge, even where that statement is in fact a conjunctive one
expressing both internalist and externalist chance-precluding constraints. We have
also argued that pure internalism and externalism— theories that state their condi-
tions for knowledge in mutually exclusive terms, must each by their own respective
logic “acquiesce” to an important role for epistemic luck. Since each is premised
upon the incompatibility of luck with knowledge, this acknowledgment paralyzes
their ability to state conditions of knowledge in a positive manner. The upshot (here
in line with Riggs) is that internalism and externalism as so understood are neces-
sarily incomplete accounts, and that neither alone can be an adequate response to the
skeptic. Taking this as background, we are now in a position to discuss the advan-
tages of the mixed externalist approach to justification and perhaps outline the de-
mands upon the positive account of epistemic justification we are seeking.
IV. The Gaps Problem: Dealing with the

Incompleteness of Epistemic Theories
It is in their acquiescence to different forms of luck that I want to say we find the
deepest source of the divergence between externalists and internalists, and by exten-
sion, between reliabilists and responsibilists. In the context of our earlier discussion,
we said that gaps represent perceived inferential lacunae or discontinuities in the
human cognitive process. Gaps are present to us existentially simply as forms of luck

that demand our recognition because they impact our lives. Philosophically, and for
better or worse, we tend to perceive these demands differently one to another. If it
weren’t obvious to the reader already, the discussion of the previous section serves
to show how epistemologists routinely view the forms of luck logically implied by
their own favored position as epistemologically innocuous, and their adversary’s
forms as nothing short of a plague upon their house. This disemblance cannot long
be maintained, and the oft-noted “stalemate” between these mutually exclusive ad-
versaries is easy to recognize. By admitting any gap, any discontinuity in the theo-
retical account of human cognition, internalist and externalist epistemologies reveal
themselves as incomplete. Since the forms of luck that the internalist and external-
ist each must acquiesce to are the direct target of their adversaries, it becomes obvi-
ous to those who are party to the debate that neither approach, freestanding, is ade-
quate to respond to, let alone capture both sets of intuitions.
Conceptualizing this incompleteness in terms of “gaps” is simply meant to for-
malize this problem. Gaps now correspond to acknowledged forms of luck impact-
ing human cognition and potentially blocking epistemizing justification. Bridges
are now theoretical attempts, either conceptual or empirical, to respond to gaps through
our philosophical reconstructions, and to successfully carry through the claim that
present justification is indeed epistemizing. To clarify this, we must first try to for-
malize our conception of the gaps themselves. Internalist theories, we’ve shown,
leave us with a gap between (1) truth and (2) justified belief. We can simply call this
1–2 gap the veritic gap to correlate with the form of luck to which internalists must
acquiesce. Externalist theories, however, do provide a conceptual link between
justified belief and truth; that is to say, a conception of a necessary connection be-
tween knowledge and its object. But for them a gap re-opens between belief (or
more strictly, propositional acceptance) and (3) good reasons or adequate evidence.
This is a token of a discontinuity between the reliabilist’s strong conceptual claims
and their lack of a supporting theory of evidence.33 Borrowing a Greek term to bal-
ance against the Latin veritas, we can call this 2–3 gap the zetetic gap. This comes
from zetesis, referring to the inquiring quality of an investigator, or of a method of
investigation, and was used by the Greeks in the context of a common search for an
unknown truth. Again this is intended to match the sense of evidential luck discussed
earlier.
Our characterization of gaps follows the analysis of our previous section. But we
should add at least one more, that between (3) reasons and (4) (proper) motivation.
We can call this 3–4 gap the enkratic gap, for the Greek term for continence. Do rea-
sons have to motivate? For instance, if I accept that I have reason to doubt/believe
some proposition, must I be motivated to some degree to doubt/believe that propo-
sition? What Christopher Hookway calls (motivational) internalism answers “yes”;
he sees the normative impact of reasons as inseparable from our motivational ten-
dencies, at least in rational agents. For the internalist, epistemic rationality requires
us to be rightly motivated: Evidence doesn’t provide me with reasons, unless it is in-
tegrated in an appropriate way with my motivational tendencies, and “it seems that
the existence of a gap between my normative judgments and my motivations signals
a kind of irrationality.”34

Now if we hold this model of three gaps between four terms [1–2–3–4] in mind,
we can say that the amalgam of views associated with internalism allows it to pre-
sent us with bridges for both the zetetic and enkratic gaps, but at the cost of leaving
the veritic gap radically open. Conversely, it is often said that externalism presents a
great departure from “traditional epistemology” with its predominantly internalist
character. In the terms we have used in this essay, we might then say that external-
ism has the advantage of being able to radically bridge the veritic gap, but at the cost
of leaving us with suspect resources for addressing the other two.
To conclude our brief discussion in this section, I would point out that each of the
gaps and associated kinds of epistemic luck we have discussed have strong ana-
logues in Nagel’s discussion of moral luck. The veritic gap (and with it Engel’s veritic
luck) bears resemblance to what Nagel termed consequential luck; the zetetic gap
(and with it Engel’s evidential luck) can be profitably compared with Nagel’s cir-
cumstantial luck; and finally, the enkratic gap, as exemplified in the possibility of a
kind of epistemic akrasia that Hookway examines, betokens largely unexplored
comparisons with Nagel’s constitutive luck. What I will call the gaps problem is the
problem of providing a positive account of knowledge and justification, one that ad-
dresses each of the discontinuities under a single unified perspective, without ignor-
ing or negating the influence of factors that are a matter of luck. Thus the parallels
suggested between acknowledged forms of moral luck and the forms of epistemic
luck we have examined are further indication of the import of the gaps problem for
epistemology today.
V. Mutual Assistance
We have now described two specific forms of “chance” (section II) that, though
sharply contested among reliabilists and responsibilists, are each claimed by some to
preclude true beliefs from constituting knowledge; we have also described forms of
“luck” (Section III) that must be acknowledged rather than precluded and that,
while also impacting epistemology, do so primarily in terms of their relevancy to a
general skeptical challenge that virtue epistemologists, whether those we have called
virtue reliabilists or responsibilists, must confront together. In section IV we have
seen that the presence of veritic and evidential luck indicates an incompleteness in
internalist and externalist epistemologies and, most important, we have gone on to
connect each form of luck with a theoretical “gap” whose place in any positive ac-
count of justification demands acknowledgment. In this final section we will discuss
objections to the present analysis of epistemic luck and clarify the demands on a
mixed externalist account of justification in addressing what we have come to call
the gaps problem.
One objection that might be brought against the discussion of epistemic luck in
the previous two sections is simply that it seems to bear little direct implication for
virtue epistemology. After all, the arguments in those sections lean heavily upon the
dichotomy between internalism and externalism, understood as mutually exclusive
and exhaustive accounts of justification. But we had previously been at pains to

show how and why virtue epistemologists have been attracted to a “mixed” exter-
nalist account, one that is capable of consistently integrating responsibility con-
straints on justification that ensure the guiding role of good reasons in reflective
human knowledge. To hold a mixed account is already to agree that purely internal
or purely external constraints cannot provide the right conditions for epistemizing
justification. Even if there are serious differences among those who advocate mixed
accounts, those we began by characterizing as virtue responsibilists are not really in-
ternalists, since they certainly do not think that the conditions on justification are ex-
clusively internal; and those we characterized as virtue reliabilists do not claim, as
some externalists might, that we can do without an account of subjective justifi-
cation for reflective knowledge. So the discussion became skewed when we slipped
back into talking in terms of a dichotomy between internalism and externalism that
most of us agree has outlived its usefulness. For one who accepts a more compro-
mising mixed account, the problem of epistemic luck need not have either the deci-
sive importance or the divisive character I have suggested it has.
In response, I would first have the reader note that I have been concerned with an
indirect effect of the problem of epistemic luck on mixed externalist accounts. We
have considered how the understanding of epistemic luck in virtue epistemology
tends still to be divided along lines that reflect a backdrop of externalist/internalist
debate in epistemology over the past three decades. I agree that this distinction is be-
coming outmoded and hope that the distinction between reliabilist and responsi-
bilist virtue epistemology will before long also outlive its usefulness. But that should
not preclude our notice of real and present tensions among those working in the
field.
Second, two common claims made by virtue epistemologists of all stripes are (1)
that their approach has something unique to offer in the way of overcoming the op-
position between internalist and externalist conceptions of justification, and (2) that
it provides substantial resources for addressing the challenge of skepticism. I can
find no better way to put these two common claims to the test than to inquire into
the reception and understanding of epistemic luck. A virtue-based account of a
mixed character should show that justification does not turn on the issue of “inter-
nal accessibility” to the ground of one’s belief, the issue that the distinction between
internalism and externalism was build around. While the various strategies of virtue
epistemologists who advocate mixed accounts of justification suggest a displacement
of the access issue, we have found that the issues have been transformed but not re-
solved, and that serious disagreements persist between virtue reliabilists and virtue
responsibilists.
Third and most important, I disagree with Driver when she says that mixed ac-
counts represent inherently unstable compromises and that “the superficial plausi-
bility of the mixed account is purchased at the cost of significant theoretical advan-
tages” presented by the “pure extremes” of internalism and externalism. On the
contrary, I have tried to argue that the advantages reside in a view that takes seri-
ously both reliability and responsibility constraints on justification, because this mit-
igates the necessary incompleteness of ‘pure’ internalism and externalism and pro-
vides that best basis for responding to skepticism.

But I would make a partial concession to Driver’s complaint: Mixed accounts cer-
tainly are not immune to the gaps problem or to more generalized problems con-
cerning epistemic luck; they remain subject, as Driver claims, to many of the same
theoretical problems that afflict the internalist and externalist extremes. This con-
cession may indeed be implied by my view that one simply can’t bridge all three of
our recognized gaps simultaneously under one perspective (as the notion of adding
or conjoining responsibilist to reliabilist conditions on justification would suggest).
This is philosophically unrealistic because it would produce a conception of the
human agent with cognitive powers far in abundance of what the cognitive sciences
and our own lived experience reveal to us. To say as we did before, that Nagel’s three
kinds of ethical luck can be seen to have important epistemic analogues, is to say that
a mixed account is one that will seek to understand the epistemic import of each of
the three gaps we have sketched and to produce a positive account of epistemic
justification in light of them.
Acknowledgment of the gaps problem motivates rather than prevents the tasks of
theoretical bridging. But my conception is one on which there are serious trade-offs
involved in such theoretical endeavors: to emphasize a conceptual connection here is
to allow a potentially larger lacuna elsewhere in one’s system of thought. And if this
is correct, then there are undoubtedly a variety of different ways that epistemic luck
can be theoretically addressed. One might say that our theoretical bridges are make-
shift, especially in this era of new empirical studies in the cognitive sciences, and that
they will need to be constantly adjusted one to another as we learn more about
human cognition and motivation structures and as we advance our theories of
knowledge and evidence in light of them.
It is simply not clear at present how well mixed accounts can handle this task, but
we have shown reason to think that mixed accounts are on better footing than non-
mixed accounts in this regard. While sometimes lauded for a certain perceived rigor
or inner consistency, we should now be able to see that non-mixed approaches place
themselves in a position of having to argue, most implausibly, that the type of luck to
which their account must acquiesce lies outside the relationship between the knower
and the known and is of no real epistemological significance. To return to Driver’s
objections, I cannot resist playing her claim of inherent instability off against Jonathan
Dancy’s similar claim. An irony is present because Dancy uses similar reasoning to
prescribe a move away from a consequentialist/reliabilist conception of the virtues
toward a virtue epistemology based upon Aristotle’s account of the moral virtues:
Non-consequentialists also are unwilling to admit that the consequentialists are

right about anything, because they feel that consequentialism is like a cancer—
once one has let it in at all it will grow until it has taken over completely. The cru-
cial question is whether the two camps are right at least about this, that no com-
promise is intellectually acceptable. And I think that they are.35
Though the conclusions that Driver and Dancy reach are highly antithetical, their
‘inexorable’ logic on this score is strongly analogous, since both argue that compro-
mise solutions are theoretically unstable and intellectually unacceptable.36

Such claims about theoretical instability really amount to little more than predic-
tions concerning the fate of mixed accounts. If mixed accounts prove only super-
ficially plausible as both think, virtue reliabilists and responsibilists can be expected
to soon part company and to do their research under separate self-descriptions
rather than accepting their shared description as virtue epistemologists. With such
predictions I need not agree. If this prediction is not to prove correct, however, there
is a burden on the virtue epistemologist to present a theoretically consistent, positive
account of epistemic justification, one that reliabilists and responsibilists both sub-
stantially agree with and contribute to. At the end of a lengthy essay, however, it may
come as some relief to the reader to learn that I do not have such a theoretical ac-
count ready to offer!
I would, however, put forward a practical point in closing. The stability of any
research program is often as much a practical matter as a theoretical one. The easi-
est way to exhibit instability is to lose the valuable benefits of mutual assistance. Re-
liabilists have focused on the demands of philosophic naturalism and on conceptions
of the supervenience of normative properties; they have been especially attuned to
the role of the cognitive sciences in understanding both reflective and non-reflective
(including animal) knowledge, and to the social and communal dimensions of epi-
stemic evaluation. Responsibilists have focused on concerns with active agency in the
context of reflective knowledge and with studies and thick-descriptions of particu-
lar epistemic virtues and vices. They have been especially attuned to the intercon-
nections between ethical and epistemological disposition and to the dynamics of in-
dividual psychology. With a mixed externalist account of justification progressing as
a shared project among virtue epistemologists, I am hopeful that the complemen-
tarity of these research foci will achieve greater recognition and that the interests
that shape research within virtue epistemology will merge significantly further.
Surely this mutual-acknowledgment and integration of interests is practically nec-
essary if virtue epistemologies of a “mixed” externalist character are to confront the
very serious doubts about their theoretical stability that we have heard expressed
from positions representing internalist and externalist extremes.
Notes
1. Claudia Card, The Unnatural Lottery: Character and Moral Luck (Philadelphia: Tem-
ple University Press, 1996). See also Daniel Statman (ed.), Moral Luck (New York: SUNY
Press, 1993).
2. For a discussion of this division in virtue epistemology in light of debate over Aris-
totle’s own distinction between the moral and intellectual virtues, see my “The Role of the
Intellectual Virtues in the Reunification of Epistemology,” The Monist, 81, 3 (1998): 488–
508.
3. Julia Driver, “Moral and Epistemic Virtue,” in Guy Axtell (ed.), Knowledge, Belief
and Character: Readings in Virtue Epistemology (Lanham, Md.: Rowman & Littlefield, 2000),
123–34. Compare Driver’s description of her view with Firth’s term “epistemic rule-
utilitarianism,” which Sosa has more recently used to describe reliabilist-oriented virtue
epistemology.
4. Linda Zagzebski, “Religious Knowledge and the Virtues of the Mind,” in Rational

Faith: Catholic Responses to Reformed Epistemology, Linda Zagzebski, ed. (Notre Dame,
Ind.: University of Notre Dame Press, 1993), 202.
5. See also Ernest Sosa’s Knowledge in Perspective (Cambridge: Cambridge University
Press, 1991) and Linda Zagzebski’s Virtues of the Mind (Cambridge: Cambridge University
Press, 1996). By received definitions, internalism insists that the conditions on justification
are exclusively internal, and externalism denies this, so any mixed account will by definition
be classed as externalist.
6. Sosa, Knowledge in Perspective, 28.
7. Stanford Online Encyclopedia, entry for “Virtue Epistemology.”
8. Lorraine Code, Epistemic Responsibility (Hanover, N.H.: University Press of New
England, 1987), 51.
9. For more on this distinction, see my “Introduction” to Knowledge, Belief and Char-
acter, xi–xxix.
10. For another article defending “mixed” accounts of justification that makes some
closely related points, see Paul Bloomfield, “Virtue Epistemology and the Epistemology of
Virtue,” Philosophy and Phenomenological Research 60, 1 (2000): 23–43.
11. Montmarquet’s thesis of direct as opposed to indirect doxastic responsibility disputes
the reliabilist assumption that control over beliefs can be achieved only directly, through
mental actions of one sort or another. Montmarquet rejects the acceptance/belief distinction
and argues for (soft) doxastic voluntarism. See his Epistemic Virtue and Doxastic Responsibil-
ity (Lanham, Md.: Rowman & Littlefield, 1993), 90–95. For a pragmatist account support-
ing the acceptance/belief distinction, see D. S. Clarke, Jr., Rational Acceptance and Purpose
(Lanham, Md.: Rowman & Littlefield, 1989), especially 33–36.
12. Compare Zagzebski, Virtues of the Mind, 59–73.
13. Thomas Nagel, “Moral Luck,” in Mortal Questions. (Cambridge: Cambridge Uni-
versity Press, 1979), 24–38.
14. Zagzebski, Virtues of the Mind, 71.
15. Zagzebski, Virtues of the Mind, and Montmarquet, Epistemic Virtue are again prime
examples of this responsibilist perspective. For more on the issue of defining the intellectual
virtues, see my “Virtue Theory and the Fact/Value Problem,” in Knowledge, Belief and
Character, 177–203.
16. Richard Paul may be a holder of the strong thesis: “The problems of education for
fair-minded independence of thought, for genuine moral integrity, and for responsible cit-
izenship are not three separate issues but one complex task.” Richard Paul, “The Contri-
bution of Philosophy to Thinking,” in Critical Thinking (Sonoma, Calif.: Foundation for
Critical Thinking, 1993), 424–25.
17. Zagzebski, Virtues of the Mind, 255: “I think of this move as expansionist rather than
reductionist since it would be more accurately described as expanding the range of ordinary
moral evaluation to include epistemic evaluation, rather than reducing the latter to the
former.” Note that Greco describes Zagzebski’s account as “‘neo-Aristotelian’ rather than
‘Aristotelian,’ because Aristotle did not hold that the moral and intellectual virtues are
unified in this way” (Stanford Online Encyclopedia). But Driver’s attempt to frame exam-
ples where moral goodness is incompatible with epistemic virtue poses no apparent prob-
lem for Zagzebski, her explicit target, who explicitly refrains from taking a position on the
classic Socratic/Aristotelian unity thesis (Virtues of the Mind, 156–157).
18. James Montmarquet, “An ‘Internalist’ Conception of Epistemic Virtue,” in Knowl-
edge, Belief and Character, 135–48.
19. Zagzebski, Virtues of the Mind, 258; Bloomfield, “Virtue Epistemology,” 35.
20. Driver, “Moral and Epistemic Virtue,” 123.
21. Bernard Williams, Moral Luck (Cambridge: Cambridge University Press, 1981), 26.
For a related critique of neo-Kantian elements in Zagzebski’s thought, see Amelie Rorty’s

“Distinctive Measures of Epistemic Evaluation” and Zagzebski’s response in the book sym-
posium on Virtues of the Mind, in Philosophy and Phenomenological Research 60, 1 (2000):
203–206, 216–19.
22. Nagel, “Moral Luck,” 467.
23. Wayne D. Riggs, “What Are the ‘Chances’ of Being Justified?,” The Monist, 81, 3:
452–72, 463.
24. Riggs 465. For background see also Peter Unger’s “An Analysis of Factual Knowl-
edge,” Journal of Philosophy 65, 6 (1968): 157–70.
25. Riggs, “What Are the ‘Chances,’” 467.
26. Ibid.
27. If I can use Robert Audi’s guide/effect distinction to connect Riggs’s sense of causal
efficacy with believing for a reason, it seems that a constraint against accidents is one that
should disallow knowledge in cases where an agent S’s belief p may be merely the effect of
a reason state r, but not also guided by reason state r. “But if r is a reason in the light of which
S believes that p, guides S’s belief formation (or retention), and is discriminative, then surely
the explanatory relation between the belief that p and the basis and connecting beliefs is not
accidental.” Robert Audi, “Belief, Reason, and Inference” in Philosophical Topics 14, 1 (1986):
47.
28. Riggs, “What Are the ‘Chances,’” 453.
29. Mylan Engel, “Is Epistemic Luck Compatible with Knowledge?,” Southern Journal
of Philosophy 30, 2 (1992): 59–75. For a response to Engel quite different from my own, see
Barbara Hall’s “On Epistemic Luck,” Southern Journal of Philosophy 32 (1994): 79–85.
30. Engel, “Is Epistemic Luck Compatible?,” 61–62.
31. Ibid., 63.
32. Ibid., 67.
33. Ibid.
34. Christopher Hookway, “Regulating Inquiry: Virtue, Doubt, and Sentiment,” in
Knowledge, Belief and Character, 149–160.
35. Jonathan Dancy, “Supervenience, Virtues and Consequences,” reprinted in Knowl-
edge, Belief and Character, 73–86.
36. Dancy continues: “Consequentialists are wise to seek to give a unified account of all
the virtues, because otherwise they will find themselves saying that of the virtues, some are
virtues for one sort of reason and others are virtues for another. This position is theoretically
unstable, and will always be vulnerable to one which manages to give the same account of
why this or that feature is a virtue throughout. Similarly, virtue theorists are right to resist
the irruption of a second form of explanation of the status of a character trait as a virtue.
Their standard form, which asks how the virtues together contribute to a good epistemic
life, is perfectly capable already of capturing the nature and role of the consequence-related
virtues. That they are consequence-related does nothing to show that we should accept a
consequentialist understanding of them” (82).

Just the Right Thickness:
A Defense of Second-Wave Virtue Epistemology
Guy Axtell & J. Adam Carter
Abstract: Do the central aims of epistemology, like those of moral philosophy,

require that we designate some important place for those concepts located
between the thin-normative and the non-normative? Put another way, does
epistemology need ‘thick’ evaluative concepts? There are inveterate traditions in
analytic epistemology which, having legitimized a certain way of viewing the
nature and scope of epistemology’s subject matter, give this question a negative
verdict; further, they have carried with them a tacit commitment to what we argue
to be an epistemic analogue of the reductionistic centralist thesis that Bernard
Williams in our view successfully challenged in ethics. In this essay, we challenge
these traditional dogmas and in doing so align ourselves with what has been
recently called the “Value Turn” in epistemology. From this perspective, we
defend that, contrary to tradition, epistemology does need thick evaluative
concepts. Further, the sort of theories that will be able to give thick evaluative
concepts a deservedly central role in both belief and agent evaluation are those
non-centralist projects that fall within what we call the second-wave of virtue
epistemology. We recognize that, in breaking from centralism, there is a worry
that a resulting anti-centralist theory will be reductionistic in the other direction
—making the thick primary. We contend however that second-wave virtue
epistemologies should be thought to provide the wave of the right thickness, and
as such, constitute the most promising approaches within a field that has become
increasingly more normative, diverse and expansive than was the traditional set of
problems from which it emerged.
1. Introduction
In a paper on norms and rationality, Pascal Engel notes that within the domain of normative
concepts, it is common to distinguish between two classes: deontic and evaluative. Deontic Page15
concepts, such as 'right', 'ought', 'permissible', and 'forbidden,' are by definition thin concepts.
But within the terrain of evaluative concepts, philosophers distinguish between those that are
thin--such as 'good', 'bad', 'desirable', etc.--and those thought to be thick, examples of which
include trait concepts like 'courageous’ and ‘open-minded,’ and affective concepts like 'rude',
'lewd', etc. A mark of thick concepts is, as Bernard Williams has suggested, that they seem to be
simultaneously both evaluative and descriptive. For the purposes of taxonomy, these concepts
must be, as Jonathan Dancy thinks, engrafted distinctly between, rather than within, the non-
normative (which he places at the 'bottom') and the thin-normative, at the top.
According to Dancy, Bernard Williams, Hilary Putnam and others who are as Putnam
describes them, ‘friends of entanglement,’ moral agents and moral philosophy both need the
thick, intermediate layer of concepts. After all, moral theory should not be concerned only with
what is good and bad, right and wrong, all things considered, but also with how agents ought to
be (i.e. 'brave,' 'prudent', etc.) in order to best meet, from an ethical point of view, the demands of
the world.1
If this much is correct, we may ask: does epistemology need thick concepts also? Do the
central aims of epistemology, like those of moral philosophy, require that we designate some
important place for those concepts located between the thin-normative and the non-normative?
What would it mean to say that for epistemic as well as moral agents, our conceptual language is
and ought to be ‘entangled’ (Putnam)?
Upon first glance, it isn’t obvious that epistemology actually does need thick concepts.
After all, the central tasks of epistemology have in the past half-century typically been thought to
lie within the specific project of elucidating the nature and scope of knowledge and justification--
a “doxastic paradigm” that doesn't obviously wear, like the subject matter of moral philosophy
does, normativity on its sleeve.
2. Epistemic Normativity
2.1 The Nature and Scope Argument
The most obvious case to be made for the claim that the issues of central concern in
epistemology are conceptually non-normative would be to reason along the lines of what we call
the Nature and Scope Argument—an argument that has been implicitly endorsed in epistemology
over the past half-century, and despite the waning of the Fact/Value dichotomy, which it
presupposes.
Nature and Scope Argument
1. Epistemology should be concerned with, first and foremost, the nature and scope of
knowledge and justification.
2. Knowledge and justification are fundamentally non-normative concepts used to pick
out particular epistemic standings.
Page15
3. Therefore, epistemology should be concerned with, first and foremost, the nature and
scope of what is conceptually non-normative.
If the Nature and Scope argument is sound, it's not hard to see why thick concepts, such
as intellectual virtues, which bridge the evaluative and non-evaluative would not be central to
epistemological theory. A refusal to recognize any distinctly normative dimension within the
subject matter of epistemology pervades several prominent philosophical research programs,
including naturalized epistemology, counterfactual epistemology and externalist, reliabilist
analyses of knowledge. Subsequently, thick evaluative concepts—for these and other projects
premised upon the sort tradition captured by the Nature and Scope Argument—are left without
any important theoretical role.
2.2 Epistemic Scope and Epistemic Practice
Concerning premise 1, we may ask: are knowledge and justification really more important than
other positive epistemic states and standings? We might be inclined to think so if we simply take
the attention they have enjoyed as a mark of epistemic importance. But this would be a mistake.
There’s no good reason to think the good by way of agency is less important for the purposes of
inquiry than the good way of belief, and further, it’s not clear why states of theoretical
understanding, for example, would not represent some of our highest cognitive achievements,
and be at least as important as knowledge and justified belief. With this said, the conclusion
should be that the privileged role the analysis of propositional knowledge and epistemic
justification have enjoyed in post-Gettier thought is simply unmerited. And by claiming this, we
need not at the same time think that knowledge and justification aren’t valuable epistemic states;
the problem is that within the tradition that has artificially inflated their epistemic significance,
only a portion of what is important from the epistemic point of view has been come to presented
as the whole. After all, from what William Alston calls the “epistemic point of view,”' 2 a variety
of things are important (Compare Battaly, this volume). Given the diversity of cognitive
successes and the similar diversity of those traits that lead us to them, we should think that what
matters, epistemically, is more broad than is captured by pointing to beliefs in isolation from the
agents who, some more appropriately than others, hold them.
These initial considerations are themselves enough to suggest that while (1) may have
tradition on its side, it is under closer inspection, implausibly restrictive. But even if we were to
follow the inveterate precedent of reducing epistemology to a theory of knowledge and
justification (and thus charitably side with tradition and accept Premise 1), the Nature and Scope
argument is flawed for a separate and more interesting reason. The reason is that Premise (2) of
the argument—the dogma that knowledge and justification are non-normative—is also false.
One way in which to criticize (2) is to question a combination of background views that
make (2) appear plausible: these are a veritist axiology and an instrumentalist conception of
epistemic normativity—a combinatory position that is strongly shared by internalist and
externalists, like Chisholm and Goldman, alike. Let’s take instrumentalism first: for an
instrumentalist, epistemic normativity is a species of instrumental normativity in the service of
an epistemic goal. In Goldman and Olsson (2008) perhaps most explicitly, instrumentalism as an
account of epistemic normativity is premised on epistemic value monism, as an adequate account
of epistemological axiology. For Goldman and Olsson, veritism is “a form of epistemic value Page15
monism that holds that true belief is the only fundamental epistemic value” (12).
The appeal of veritism's very minimal epistemological axiology is quite strong amongst
those who hold themselves heirs to Quine’s influential call for the naturalization of
epistemology, and for replacement topics for traditionally normative concerns. To Quine our
normative concerns can easily be naturalized; epistemic normativity “is a matter of efficacy for
an ulterior end, truth….The normative here…becomes descriptive when the terminal parameter
is expressed (1986).” Such a view is premised on there being no serious problem about
entanglement of fact and value, and no need of sharing an evaluative perspective in order to fix
the extensions of epistemic concepts. A result for the veritist-instrumentalist is that the central
concepts in epistemology such as knowledge and justification could be thought non-normative
just like any other epistemic concept, but this result comes at the cost of meeting the burden of
proof that comes with its reductionistic thesis. After all, understanding (a theory, for example)
does not require that theory be true, and as Riggs and Kvanvig point out, nothing about the idea
that truth is valuable undermines the idea that understanding is an independently epistemically
valuable.
Leaving behind the narrowly-construed veritist-instrumentalist picture of epistemic
normativity upon which the non-normative picture of epistemic concepts such as knowledge and
justification rests, we are led to two further commitments: pluralism in our conception of
epistemic axiology, and non-centralism in our treatment of epistemic normativity.
To be clear, pluralism becomes the consequence of rejecting the claim that the
multiplicity of epistemic values can be reduced to the value of truth (or some other single or
“core” epistemic good).3 And on the pluralist picture, according to which thick evaluative
concepts, and not just truth, can be useful in articulating our epistemic aims, there becomes a
need to deny epistemic versions of what Bernard Williams calls "centralism," the claim of the
explanatory priority of thin deontic and evaluative ethical concepts over thick evaluative ethical
concepts.
Williams takes centralism to be "a doctrine about language and linguistic practice," a
doctrine holding "that very general ethical truths were logically prior to more specific ones"
(1995, 184; see also Alan Thomas, this volume). Epistemological non-centralism, if we apply the
definition of Susan Hurley as Williams does, would reject the view that the general concepts like
“justified” and “ought” are logically prior to and independent of specific reason-giving thick
epistemic concepts of virtue and vice.
Centralism is at least a contested thesis in philosophy, and in our view it is false in

respect both to ethical and to epistemic normativity. There are no adequate defenses of epistemic
centralism, despite the burden of proof centralists bear for its reductionist thesis, a thesis that
looses its footing in the absence of veritism. Without centralism, optimism about the reducibility
of the thick epistemic normativity to thin epistemic normativity appears no more warranted than
the reducibility of all other epistemic values to the value of truth.
By rejecting the dogmas that (1) epistemology is the theory of knowledge and Page15
justification, and (2) that these central epistemic concepts are non-normative in the way we have,
we not only have rejected assumptions that undermine the thought that epistemology needs thick
evaluative concepts but also offer a presumption in favor of (1) pluralism over monism as a
theory of epistemic axiology, and (2) non-centralism over centralism as a theory of epistemic
normativity.
Parting ways with the dogmas hostile to the thick evaluative, we in so doing bring
ourselves into alignment with what has come to be known as the “Value Turn” in Epistemology
—a viewpoint from which thick evaluative concepts are indispensible rather than peripheral.
2.3. Dictates of the Value Turn
The “Value Turn” in epistemology, as Wayne Riggs has called it, is characterized by the
thought that epistemology is a normative domain of inquiry the central tasks of which are framed
by considerations of epistemic value. The epistemic point of view will accordingly be a point of
view from which knowledge and justification are important, but unless we have already assumed
veritism, so are understanding, wisdom, responsible believing, as well as whatever goals govern
good epistemic practice. For epistemic value pluralists, “[E]pistemic goals include knowledge,
understanding, wisdom, rationality, justification, sense-making, and empirically adequate
theories…. (Kvanvig 2005, 287).
It has been a positive implication of the value-of-knowledge debate that we “shift the
focus on contemporary epistemological theorizing away from the merely minimal conditions for
knowledge—a focus that has arisen largely in response to the Gettier problem—and move it
towards higher epistemic standings” (Pritchard 2007, 23). Pressing the case for epistemology’s
expansion still further, Kvanvig argues that concepts that should be deemed epistemically central
concern not only the nature but also the value of different epistemic standings; correspondingly,
we are led to study epistemic goals like understanding, wisdom, rationality, justification, sense-
making, and empirically adequate theories. As Kvanvig puts it, “[T]he class of epistemic goods
is manifold, as wide as the class of cognitive successes” (2005).
The dictates of the Value Turn, we can see, just are, more or less, the denials of the
dogmas that undermine the need for thick evaluative concepts. We might express these dictates
as follows:
(1) An epistemological theory should expand its focus beyond merely knowledge and
justification to include other things that are epistemically important, and in particular, it
should give a central importance to the matter of how agents themselves ought to be in
order to best meet, from an epistemic point of view, the cognitive demands of the world.
(2) An epistemological theory must specify the nature and sources of epistemic value; in
doing so, the theory will have provided an evaluative background against which a given
belief will be better than another—for example, “known” as opposed to merely
“justified.”
Zagzebski’s virtue epistemology is an example of a theory that clearly meets the requirement in
(2); against the background of epistemic flourishing, she explains what sort of thing knowledge Page15
and justification must be (to note, less robust virtue epistemologies also analyze knowledge only
against an antecedent background against which the traits said to give rise to knowledge are
evaluated as good—i.e. virtues.)
There is a third dogma, however, along with the first two, which the Value Turn should
be taken as denying. In an important and slightly different way, this third dogma also has served
to marginalize thick evaluative concepts: this is the long-standing dogma that act-based ethics
should serve as the structural model for epistemic evaluation, a dogma that has propagated the
belief-based paradigm—that conceptually thin properties of beliefs are logically prior to
conceptually thick epistemic concepts such as virtues. The rejection of this dogmatic paradigm
will be then a third dictate of the Value Turn, a dictate bringing the problem with centralism
mentioned earlier to the forefront.
(3) An adequate epistemological theory must be non-centralist; that is, it must reject the
conceptual priority the belief-based paradigm gives to thin epistemic concepts over the
epistemic concepts of virtue and vice, and the kinds of agent evaluations that such
concepts provide.
In support of the idea captured in (3), the ethicist Peter Railton writes, “We should
probably try more often to work from the inside of agents, from their centers of mass as agents
and moral beings. For such an approach, questions of normative guidance become questions
about how normative guidance occurs within the agent, what gives norms their life, and how
they enter into the shape and meaning of the agent’s experience, thought, feeling, and action”
(2008, 3). This seems true in respect to epistemic appraisal as well, where thick descriptions of
particular virtues like self-trust, conscientiousness, intellectual honesty, intellectual humility,
perseverance, etc. provide insight for agents themselves into how to develop habits that extend
their competence and improve their reliability as cognitive agents. On the doxastic paradigm,
such appraisals could function in a theory only as tag-alongs to the thin notions traditionally
applied to beliefs.
The three dictates of the Value Turn not only capture what a theory must do to avoid
what we’ve taken to be indefensible dogmas, but also stand as independently supportable by
some of the diverse and illuminating projects found in the recent literature—projects within in
which both thick evaluative concepts and epistemic normativity more generally are central.
We view Greco’s agent reliabilist approach as offering further support of our claim in (1).
In Knowledge as Success from Ability (2009, forthcoming), John Greco states argues that “If
knowledge has an evaluative dimension—if epistemology is a normative discipline—then a
1
Then there are moral agents themselves who these authors argue also require the rich resources
of the language of character traits and affective concepts in order to facilitate their own moral
reflections, judgments, and decisions. The ‘middle’ layer of evaluative concepts is arguably vital
to reflective morality, by bringing to bear conceptual resources for the kinds of reflection and
emotional response without which moral character development and integration might indeed be
impossible.
22
What Alston means here is, put generally, that we can think about the value of something not
Page15
from the point of view in which all considerations are weighed, but rather, from a point of view
in which those considerations that are weighed are properly epistemic. In this sense, we can
understand an “epistemic point of view” in the same way we can evaluate a house from an
‘aesthetic point of view’ only, apart from how we might evaluate the house from a pragmatic
point of view, or an all-things-considered point of view.
3
Thanks to John Greco for pointing out the need to distinguish the thesis of epistemic value
monism as a reductionistic thesis, from the somewhat separate question of the monist’s ability to
acknowledge a multiplicity of epistemically valuable states and value-driven epistemological
concerns.
central task of epistemology is to provide an account of the normativity involved” (CH1: 2). The
kind of “credit for success” a knower merits on Greco’s account indicates a kind of normativity
quite ubiquitous: we credit athletic and other forms of achievement to agents, under normal
conditions, in much the same way. Both by allowing us to resituate our practices of epistemic
appraisal within a broader and more varied set of practices—a more general normative
phenomena of merit for achievement—virtue reliabilism’s ability to account for knowledge as
credit for true belief brings virtuous agency of the sort that known true beliefs are creditable to
the forefront. Further for Greco, what knowledge is and the normative status that knowledge
requires will not be sharply divided, since an account of knowledge as credit for true belief “is
intended as both an account of knowledge and an account of epistemic normativity.”
If (1) is thought of as bolstering the Value Turn through a sort of “ecumenical”
requirement (2) might be thought as a sort of inverse: a requirement that we specify the nature
and sources of epistemic value, a task that itself will require that our project be ecumenical. In
support of the requirement in (2), that “Determining… (epistemic) values is itself one of the
tasks proper to…epistemology”, Jonathan Kvanvig writes: “...there is a presumption in favor of
holding an epistemological theory responsible to two criteria. A correct account of the nature of
knowledge must resist counterexample, but it also ought to be amenable to an account of the
value of knowledge (2003).” This is translatable into the requirement that in saying what
knowledge is we must also say what makes it valuable. And so to say what knowledge is, we
must first have an antecedent theory of epistemic axiology.
Wayne Riggs takes the Value Turn to imply that when we “put the need to find the
bearers of epistemic value at the fore of…epistemological theorizing” (Riggs 2007), we firstly
render the study of the nature and sources of epistemic value across all of our positive epistemic
standings a more central and continuous endeavor. Doing this will, in no small way, require that
we evaluate the wide scope of character traits and cognitive faculties from the epistemic point of
view made up by, as Riggs says, the fundamental values “at the fore of epistemological
theorizing.”
(1) and (2) reflect a conception of an expanded field of epistemology, a conception
informed by our understanding of the Value Turn, which its proponents argue “present[s] a
theoretical foundation for greater diversity of interest in epistemology” (Kvanvig 2003, 188) and
supports quite directly the view that “epistemological inquiry deserves at least some enlargement
in the direction of concepts other than knowledge” (2005, xvi).
Our third requirement the Value Turn demands of an adequate epistemology brings us
back in the most direct way to the challenge of epistemological ‘thickies’ to the predominance of Page15
‘thin-focused’ epistemology during the post-Gettier era. We might begin supporting (3) by
noting that Bernard Williams held that thick moral concepts are “characteristically related to
reasons for action” (ELP). But Williams never implied that the two aspects of thick concepts—
world-guidedness and action-guidingness—are necessarily found together in an individual’s
psychology: a person may be perfectly able to apply the concept accurately to features of the
world, without its application being action-guiding. As he puts it, ‘An insightful observer can
indeed come to understand and anticipate the use of the concept without actually sharing the
values of the people who use it’ (1985, 141-2). In this way, thick moral concepts being
“characteristically” related to reasons means they are so related for the agent who fully
embraces, or is “fully engaged” with the thick affective or aretaic concept, and this is likely to be
seen as an achievement and as a manifestation of integrated moral character (Moore; Goldie).
Analogously, we suggest this analogy applies to epistemology: thick epistemic concepts like
“open-mindedness,” “conscientiousness,” etc., when held in the engaged way such that one
shares its evaluative valence (or perspective), can characteristically be associated with reasons to
pursue inquiry, or to pursue it through one particular method/strategy rather than another.
Likewise, decision-oriented virtues such as prudence, as Adam Morton notes, perhaps even more
directly have epistemic relevance: “in planning and carrying out a belief-acquisition strategy one
has to look forward as carefully as in any other activity” (Morton, 117).
Christopher Hookway has still more explicitly stressed the importance of these
conceptually thick features of inquiry in a way consonant with the American pragmatist tradition,
in which (in Dewey’s view) the identification of all virtues and vices are derived from concerns
with inquiry and there is no very sharp separation between moral and epistemic virtues (see also
Axtell 2008b). The “doxastic paradigm” he identifies with post-Gettier epistemology essentially
characterizes what we have called epistemology undertaken on analogy with act-based ethics.
The alternative conception of epistemology as a theory of “how we are able to carry out our
inquiries and theoretical deliberations in a well-regulated manner” (2006) fulfills (3)’s demand
for a model that gives agents their due place as objects of epistemic evaluation, and provides a
comfortable home for both the thick reflective virtue concepts as well as the thick faculty-based
concepts, such as those at the fore of Greco’s view.
So epistemological ‘thickies’ ask: “How it is possible to be good at inquiry rather than,
more simply, what it is to have justified beliefs or knowledge” (Hookway 2006, 101). It should
be noted that, on these non-centralist approaches, epistemic inquiry is best construed as inquiry
extending over time. To believe truly has no diachronic component, but inquiry is intimately
connected with the diachronic goal of maintaining a stable stock of belief. On this view, the
concepts we use for the purposes of epistemic evaluation—knowledge, justification and the like
—serve to evaluate epistemic standings against the background of some commitment to such
extra-synchronic goals, goals that our intellectual virtues allow us to recognize as goals, as well
as to promote. Morton argues, “Most intellectual virtues have essential connections to capacities
to search in some particular manner, and capacities to know when that kind of search is a good
idea. These are the capacities we makes names for, because they are the ones we need names
for…multi-purpose virtues of intelligent activity” (2006). Whatever else inquiry is, it is a
problem-solving activity, governed by norms. But while our ability to reason rests upon a system
of capacities and subdoxastic processes that the special sciences of cognition study, it also rests
on skills and attitudes, acquired habits, emotional, moral, and intellectual dispositions which
serve as enabling conditions for reflection (103). Inquiry methods and strategies which
systematically underwrite success-conducive theses deserve to be credited with a significant Page15
measure of rational warrant. Here’s Morton: “Nearly all the intellectual virtues that we have
everyday names for are virtues of intelligent activity generally [making and carrying out plans],
and not specifically of belief formation, decision, or some other category of thought…So the
epistemic virtues, in particular, are pointless unless they coincide with or cooperate with virtues
of epistemic strategy….” (117-118). Epistemology as a theory of inquiry will be, therefore, at the
same time a theory of epistemic values toward which inquiry should be thought to aim, as well
as a theory about what intellectual practices and traits best achieve this aim or aims.
Having outlined a way of thinking about the Value Turn that issues ‘thick-friendly’
theoretical requirements on an adequate epistemology, we have gone some distance towards
identifying the sorts of theories that will best meet the requirements that we have tried to
establish. But we are still left with lingering questions: do the three 'thick-friendly' theoretical
requirements implied by the Value Turn count in favor of any particular type of epistemological
theory? From what has been implied, they will count in clear favor of some variety of virtue
epistemology. However, some virtue epistemologies conceive of virtues quite differently than
others. The virtue-responsibilist/virtue-reliabilist divide, for example, is clear evidence. This
invites us to ask just what role thick evaluative concepts should be thought to play within the sort
of theory that does justice to them.
3. The Second Wave of Virtue Epistemology
Our aim in this section shall be first to maintain that the epistemological projects most
comfortably at home within the bounds of the theoretical requirements implied by the Value
Turn will include those pursued by what we call “second-wave” virtue epistemologists.
Secondly, we make the case that second-wave virtue epistemologists not only locate thick
evaluative concepts within the core of their theories, but also that, by doing so, simultaneously
serve to expand epistemology in a way that unites two areas previously disconnected: that about
which epistemology's central tasks are concerned, and that which is epistemically important.
In order to best understand the context within which the “second wave” of virtue
epistemology has emerged, it will be helpful to draw attention to some parallels between
successive ``waves'' of interest in virtue-theoretic perspectives in ethics and epistemology. The
resurgent interest in ethical virtue is often dated to a half-century ago, 1958, when influential
papers in ethics by G.E.M. Anscombe and Philippa Foot were published and served to, in
different ways, promote the idea that ethics would benefit by turning focus from the thin to the
thick. Over the past half-century, frustration with act-based ethics and its preoccupation either
with thin deontological concepts like ``right'' and `òught'' or with thin axiological concepts like
``good,'' has given rise to extensive studies of ethical character traits, their relationship to
emotional dispositions, and their role in explanation and evaluation of agents.
In its first wave, virtue ethics remained focused on general, largely metaethical questions,
especially with the defense of cognitivism against non-cognitivism. Only in the second wave of
virtue ethics, as it turned to more constructive projects in the '70s and '80s, did virtue theorists
and other friends of the descriptive-evaluative entanglement have those consequences that Page15
Putnam, in The Collapse of the Fact/Value Dichotomy, associates with them: consequences of
inspiring a more pluralistic account of the good, the resurgence of first-order ethics as a subject,
and a parallel resurrection of political philosophy. For virtue ethics, the move from thin to thick
was clearly also a call to expand.
In its first wave—consisting in the virtue epistemologies offered in the 80s until roughly
the mid-90s—the debate within virtue epistemology still revolved around traditional issues like
scepticism and the nature of justification. Although intellectual virtues were imbedded in these
theories, interest in intellectual virtue took a back seat to the traditional preoccupations. It's not
clear then that the first wave of virtue epistemology was expansive enough in its theoretical aims
to meet Requirement (1), nor non-centralist enough in its theoretical methodology to meet
Requirement (3).
While the first wave is responsible for making thick evaluative concepts at least
something we should think of knowledge and justification in terms of, it is the second-wave of
virtue epistemology, and especially the sort of projects that have arisen after Zagzebski’s (1996)
landmark work, that have more clearly rebelled against the dogmas we’ve denied—the dogmas
suggestive of taking the narrow set of traditional problems to be as important as the first-wave
took them to be. The first-wave of virtue epistemology, having pre-dated much of the literature
on epistemic normativity, gave us little reason to worry about the connections between epistemic
virtue and epistemic value that have been a standard theme within the past decade; after all, the
first wave of projects would lead us to think the epistemically thick concepts used within
analyses of knowledge are epistemically important just because knowledge is important, which is
perhaps why it has only been the more recent ‘thickie’ or non-centralist projects that have taken
these concepts to be a subject for research in their own right, apart from whatever role they might
have in explaining knowledge.
Focusing on these thick concepts rather than or in addition to concern with the
paradigmatically thin concepts of knowledge and justification, the second wave, moreso than the
first, has thus put the need to find the bearers of epistemic value at the fore of epistemological
theorizing. Recently, thick epistemic concepts have figured centrally in the zetetic context, the
first-personal or deliberator's context of active inquiry as a means to settle a doubt or improve
his/her epistemic circumstances. The zetetic context informs us not primarily about evidence in
the internalists' sense—synchronic rationality—but about diachronic rationality in regard to
actions-in-inquiry. The necessary inclusion of thick concepts in practices of agent appraisal and
epistemic credit follows from value considerations in credit-for-achievement, but also from the
necessity of evaluating the agent's zetetic activity, and therefore diachronic, as opposed to merely
synchronic rationality.
Unlike the first wave, then, the present wave in virtue epistemology finds the study of the
reflective virtues expanding well beyond the bounds of those projects framed by thin concepts.
And by proceeding in this way, in the absence of the dogmas that determined what was
important in the previous two decades, the second-wave of virtue epistemology has legitimized
the importance of analyzing specific evaluatively thick virtue concepts that we think are
important to inquiry, regardless of whether they are important for knowledge. Recent examples
of thick descriptions of intellectual virtue might begin with Bernard Williams' seminal book
Truth and Truthfulness (2002), and extend to work on particular intellectual virtues such as
intellectual curiosity (Miscevic), trustworthiness (Elgin), self-trust (Lehrer; Foley), Page15
conscientiousness (Montmarquet), open-mindedness (W. Hare; Riggs; Baehr; Arpaly; Kvanvig);
insight (Riggs), and humility (Roberts and Wood). Correspondingly, thick descriptions of
intellectual virtue have led to an investigation of epistemic vices such as epistemic akrasia
(Hookway), malevolence (Battaly) and self-indulgence (Baehr).
It should be clear that the contrast between those philosophers Simon Blackburn calls
“thinnies” and “thickies” is not one that is limited to ethics, where the distinction was first
thought relevant. The present second-wave in virtue epistemology is indeed increasingly a wave
ridden by epistemological “thickies,” and we have endeavored to add philosophical weight to the
interests behind it. Sometimes the thickening agent is an interest in epistemic standings other
than knowledge and justified belief; at other times it is concern with the epistemic, practical, and
sometimes even ethical value or disvalue of inquiry, and with methods and strategies to go about
it.
Extended case studies and even fiction literature help us to see the profound impact in our
intellectual lives of the diachronic acquisition and development of intellectual virtue and vice.
Social epistemic perspectives also expand upon the primarily synchronic focus on propositional
knowledge and justification, and second wavers have concerned themselves with what
epistemological projects may lie at the intersection of collective and character epistemology. As
Lahroodi puts it, “Character epistemology teaches that the cognitive character traits of
individuals are proper subjects of epistemological inquiry. Collective epistemology…teaches that
collectives can be genuine cognizers like individuals. Combined together, they suggest the
possibility that some collectives too may have cognitive character traits and be worthy of
epistemic assessment as virtuous or vicious” (2007b, 282; compare Fricker). At still other times
this thickening gets expressed as understanding our moral and intellectual competencies in the
light of the special sciences of human cognition. Virtue epistemologists have begun to explore
how to evaluate the psychological literature on need on “metacognition” and on “need for
cognition” from the standpoint of virtue theory (Lahroodi, 2007a; Morton; Lepock). Thicker
descriptions of agents’ activities at inquiry are clearly required where strategy-selection is a
feature of the study.
4. Conclusion
The ‘second-wave’ of virtue epistemology is, we have shown, inseparable from a movement
towards the ‘thickening’ of epistemology; its proponents are all of them ‘friends of
entanglement’ in some sense that puts them at odds both with pure internalism as traditional
armchair epistemology, and with pure reliabilism, causal theories of knowing or other radical
attempts to naturalize epistemology. A maturing second-wave in virtue epistemology seeks to
understand the rich interconnections between thick and thin epistemic concepts in permitting the
performance of the tasks that we have argued should be taken as central from the epistemic point
of view.
Unless our language and linguistic practice related to epistemic credit attributions run
quite contrary to the language and practice of ethical evaluation, the collapse of the fact/value
dichotomy goes quite some distance against the 'centralism' in epistemology as well as in ethics.
So we reject that non-symmetrical relation between the thick and the thin that centralists or
'thinnies' assert, and with it, we would add, many key shared assumptions in the old
Page15
internalist/externalist debate that often go un-noticed as such. We have described epistemological
thickies as pluralists and non-centralists, but some might characterize themselves as anti-
centralist. There may be different views about the proper import of the Value Turn, but it is
important not to overlook important asymmetries between the results of ethical reflection and
judgment, and epistemic judgment and reflection. As Alan Thomas (this volume) well notes,
Williams himself held a rather more "vindicatory" account of epistemic concepts and virtues than
of ethical concepts and virtues, and this is not without import for those who would carry the
comparisons between epistemic and ethical concepts too far.
It has explicitly not been our position that we should maintain an asymmetry between the
thin and the thick, only reverse it. We acknowledge that a “thickening” of epistemology could go
so far as to constitute a re-conceptualization rather than the expansion of epistemology we
envision continuous with the analytic tradition. There is the possibility of radically constructivist
versions of ethno-epistemology, or of an analogue of ethical particularism and its anti-theory,
attaching itself to “thick” epistemology. To ride the wave of appropriate thickness, however, we
must not follow these, nor pursue what have rightly been called the ‘imperialist ambitions’ of
some over-strong reliabilist and neo-Aristotelian versions of virtue theory. At times during the
resurgence of interest in virtue ethics, some of its proponents appear demand a shift “from” thin
to thick philosophic concerns, and to identify this with the wholesale shift of focus from the
rightness of actions to the appraisal of agents. This is a contentious matter that arguably leads to
virtue ethics as a misleading category. But these are questions that needn’t concern us in arguing
as we have for the epistemic centrality of the concepts of intellectual virtues and vice that occupy
the ‘middle-thick’ level of discourse. We maintain that epistemology is not exhausted by the
tasks of elucidating the nature and scope of knowledge and justification, and this is consistent
with holding both beliefs and agents as central foci for epistemic evaluation.
So while our approach has been one we hold consistent with various accounts of
philosophic normativity that are “virtue-focused” (and perhaps even “virtue-based” Miscevic,
2008), the strength of the second wave of virtue epistemology as we take it here comes in part
from rejecting not only the dogmas of the past, but also any need to move ‘from’ thin to thick
concepts in epistemology; the thick/thin distinction is one of degree, and we need concepts at
various levels of abstraction for the different resources and different normative and explanatory
interests that they bring with them. An anti-centralist stance in epistemology, in other words,
would be quite a sea-change, but may not produce the ‘wave of the right thickness,’ which is the
one we want to carve. We here take the moderate stance of de-motivating reductionist projects of
both kinds, and set aside these questions about the difference between centralism, non-centralism
and anti-centralism for the focused and extended discussion they deserve.
We are content, then, if we have motivated a particular way to think about the value-turn
in epistemology, and through it to have made epistemology safe for any plurality of research
programs into the 'thick' of intellectual/epistemic responsibility. Such a set of research interests
gets the comfortable home it deserves in our expanded but still naturalistic conception of
epistemology—and whether or not individuals most interested in pursuing the study of
intellectual character and value-driven epistemology consider themselves to be strongly
naturalistic in their philosophic orientation. The manner in which our prescribed approach might
unite concerns about the quality of inquiry with concerns about the role of thick concepts in
epistemology is challenging, but still we think quite inviting for epistemologists—a wave of just Page15
the right thickness!4
4
?
Much thanks from the authors go out to organizers and participants at the 2008 Episteme
Conference “Epistemic Agency,” Pascal Engel, Julien Dutant and the University of Geneve; to
John Greco for his comments, and his own Epistemic Normativity Lectures; to Jason Baehr,
Heather Battaly, Matthew Chrisman, Kate Elgin, Peter Goldie and Alan Thomas for related
discussion; and to Ben Kotzee, Jeremy Wanderer, and an anonymous referee at Philosophical
Papers for helpful comments on an earlier draft.
Works Cited
Anscombe, G.E.M. 1958. “Modern Moral Philosophy,” Philosophy 33, No. 124; reprinted in
Axtell, 2010. “Intersections: Character-Trait Ascriptions in Ethics and Epistemology,”
Metaphilosophy, forthcoming.
Axtell. 2008a. “Expanding Epistemology: A Responsibilist Approach,” Philosophical Papers 37,
1, 2008, 51-88.
Axtell. 2008b. “Thick and Thin Concepts: responses to Battaly, Elgin and Goldie” [transcript].
Battaly, H. 2008. “Intellectual Virtue Through Thick and Thin: Disagreement and Combinatorial
Vagueness,” this volume
Carter, J. Adam. 2008. “A Virtue Comparativist Account of Knowledge, Luck and Ability,”
typescript.
Carter, J. Adam. 2007a. “A Happy Triumvirate: Virtue Epistemology, Contextualism, and the
Safety Principle,” typescript.
Carter, J. Adam. 2007b. “The Meno Problem and the Meno Requirement” typescript.
Elgin, C. 2008. “Trustworthiness” [draft for APA Pacific 2008, “Thick and Thin Concepts”
symposium].
Engel, P. “Epistemic Norms and Rationality” [2007, typescript].
Goldie. P. 2008. “Thick Concepts and Emotion,” APA Pacific symposium paper [typescript].
Goldman, A. and Olsson, E. 2008. “Reliabilism and the Value of Knowledge,” in A. Haddock et.
Al. (eds.), Epistemic Value, forthcoming.
Greco, J. “The Nature of Ability and the Purpose of Knowledge,” [typescript]

Hookway, C. 2000. “‘Regulating Inquiry’, in Axtell (ed.) (2000), 149-160.
Hookway, C. 2006. ‘Epistemology and Inquiry: The Primacy of Practice,’ in S. Heatherington Page15
(ed.) Epistemology Futures, 95-110.
Kornblith. H. 2002. Knowledge and its Place in Nature. Oxford: OUP.Kvanvig, J. 2005. “Truth
is not the Primary Epistemic Aim,” in M. Steup and E. Sosa (eds.) Contemporary
Debates in Epistemology, 285-295.
Kvanvig, J. 2003. The Value of Knowledge and the Pursuit of Understanding. Cambridge:
Cambridge University Press.
Lahroodi, R. 2007a. “Evaluating Need for Cognition,” Philosophical Psychology 20, 2: 227-245.
Lahroodi, R. 2007b. “Collective Epistemic Virtues,” Social Epistemology, 21, 3: 281-297.
McDowell, J. (1998). Virtue and Reason. Mind, Value, and Reality. Cambridge, MA,
Harvard University Press: 50-76.
Miscevic, N. 2006. “Moral Concepts: From Thickness to Response-Dependence,” Acta analytica
21, 1: 3-32.
Miscevic, N. 2007. “Virtue-Based Epistemology and the Centrality of Truth: Towards a Strong
Virtue Epistemology,” Acta analytica 22: 239-266.
Moore, A. “Maxims and thick ethical concepts”, Ratio (new series) 19, 129-47 (2006).
Morton, A. 2006. “Knowing What to Think About,” in S. Hetherington eds. Epistemology
Futures, 111-130.
Putnam, H. 2002. The Collapse of the Fact/Value Dichotomy. Boston: Harvard University Press.
Pritchard, D. 2007. “Recent Work on Epistemic Value,” American Philosophical Quarterly,
forthcoming.
Railton, P. 2008. “Normative Guidance.” (transcript)
Rassmussen and Rabinowicz (2000).
Riggs, W. 2007b. “The Value Turn in Epistemology,” in New Waves in Epistemology, V.
Hendricks & D. H. Pritchard (eds.). Ashgate, forthcoming.
Steup, M. and Sosa, E., (eds.) Contemporary Debates in Epistemology. Oxford: Blackwell.
Williams, B. 2002. Truth and Truthfulness: An Essay in Genealogy. Oxford: Oxford University
Press.
Williams, B. 1995. Making Sense of Humanity. Cambridge: Cambridge
University Press.
Williams, B. 1985. Ethics and the Limits of Philosophy. Boston: Harvard University Press.
Page15
Expanding Epistemology: A Responsibilist Approach
Guy Axtell
Introduction.
If the recent but quickly-growing body of literature on epistemological axiology
can serve as an indication, there are changes brewing in the conception of
epistemology and its central tasks. By a literature on epistemological axiology, I
mean not just the value problem and its impetus to newfound interest in the
nature and sources of epistemic value, or the ‘value-driven’ approach to
epistemology that has lately been proposed. I mean as well a rising debate
between epistemic value monists, dualists, and pluralists, which addresses still
more basic philosophical questions. In periods in any field of research where a
long-standing paradigm or dominant mode of practice is being questioned and
alternatives explored, it is no longer only theories that are on the table for serious
debate. At such times theories, methods and aims are likely to be viewed as more
closely interconnected, the prime example being that problems originally posed
to particular theories are viewed more and more as calling for re-evaluation of
methodological and axiological assumptions animating that field of study.
Such is the description I would start with of our present state of affairs in
the field of epistemology. Our position I see as significantly different than at any
time during the four-decade period often referred to as the post-Gettier era,
which saw lively debate over theories (foundationalism vs. coherentism;
internalism vs. externalism), but a comparatively strong consensus, relative to

today, over methodological and axiological assumptions. The new externalist
theories functioned at the start of this era to expand epistemology in important
ways. They reminded philosophers that it is always by favor of nature’s grace and
not just by one’s own good works that one knows something; that there are
thence a plethora of ways in which true beliefs might be epistemized; and that
knowledge-attributions and discursive entitlements need to be considered in
social perspective. Externalism also put to work the potent resources of modal
thinking in epistemology. In order to fly these modal kites, however, certain other
assumptions were held steady as inviolable, such as a semantic notion of truth
and a veritist axiology that drew from it. Especially when epistemology came to
be treated as the theory of propositional knowledge and the focus was either on
giving a theory of justification or on immunizing a proposed analysis of
propositional knowing from the kind of epistemic luck involved in Gettier cases,
concerns about epistemic goods other than knowledge and true belief were
largely pushed aside, or even treated as conceptual confusions. 1
In this paper I treat newfound interest in epistemological axiology as a clear
example of non-allegiance to long-standing assumptions about methods and
aims. I look for expansions of epistemology that are continuous with the analytic
tradition, but that axiologically re-cast what is still of value the
coherentism/foundationalism and internalism/externalism debates. The
expansions I envision would share some of experimental philosophy’s criticisms of
the analytic tradition’s a prioristic tendencies and ‘armchair’ methods, without
succumbing to what Michael Williams (this volume) refers to as ‘radical
reliabilism.’ My challenge to readers of this paper is to think seriously both about
2
the present need for an expansion in the field of epistemology, and about what
particular directions that expansion should take.
I divide the paper simply into two Parts. Part 1 addresses why a significantly
broader conception of epistemology’s central tasks is today called for. Here I
provide an overview of three converging arguments to that effect, arguments
each of which constitutes an ‘internal’ critique of the analytic tradition and the
place of the project of analysis of knowledge within it. Each of the arguments to
be examined concludes that the ‘received view’s’ insular conception of
epistemology’s central tasks inhibits philosophers from accomplishing certain of
their own intended purposes. While there are many other authors who offer
arguments that might also be construed as internal critiques of the project of
analysis, these three carry especial force. Moreover, I think their arguments
provide impetus to just the right kind of expansion in our field. As a primary
instance of this, all three arguments affect, in Jonathan Kvanvig’s words, a
“loosening of the shackles fastened on epistemology by skepticism.” Part 2
revisits each of our three focal authors, trying to deepen the inter-connections
between them, and augmenting their own explicit suggestions for new research
projects in epistemology with some of my own. The suggestions for new
epistemological futures that I make in Part 2 needn’t be taken as ones that my
focal authors would necessarily be on board with. But I will try to show how they
are floated by the kind of “responsibilist” approach that I favor.
Part 1. Motivating an expansion of epistemology’s central tasks
3
The argument that I want to highlight from my first focal author, Duncan
Pritchard, explicitly targets the ill-effects of a kind of “philosophical platitude.” He
terms it the epistemic luck platitude: undiscriminating adherence to the thesis
that knowing is incompatible with luck. By analogy, we might think of the
arguments made by the other two authors as similarly intended to unburden us of
other platitudinously held commitments. Jonathan Kvanvig, my second author,
could then be taken to target the platitude that ‘analysis of knowledge is analysis
enough,’—to target, that is, the trite assumption of the centrality of knowledge in
relation to other epistemic aims within our overall cognitive economy. My third
author, Michael Williams, might be seen targeting platitudinously-held
assumption that, in the debate over radical scepticism, the real or imagined
sceptic has the right to issue ‘blind’ and ‘indefinitely iterable’ challenges to the
epistemic entitlement of any would-be knower.
Let me provide a brief but somewhat more formal description of the specific
argument to be highlighted from each of these three authors:
1.1 The “anti-luck” argument. Pritchard’s anti-luck argument leads to
abandoning any project that presupposes uncritical adherence to the
epistemic luck platitude; less strongly, this argument leads to rejecting as self-
undermining any project that takes knowing (or other epistemic standing) as
incompatible with luck without first carefully distinguishing between the
various specific kinds of luck that claim may be taken to cover.
4
1.2 The “epistemic value” argument. Kvanvig’s epistemic value argument
leads to abandoning veritism and other forms of epistemic value monism; less
strongly, this argument leads to rejecting as self-undermining any project that
aims to provide an account for the nature of a positive epistemic standing
(such as knowing or theoretical understanding) without coordinating concerns
in regards to the value of that epistemic standing.
1.3 The “epistemic compatibility” argument. Williams’ epistemic compatibility
argument leads to abandoning assumptions that have made it appear
impossible to preserve the integral contribution of agent responsibility to
agent reliability; less strongly, this argument leads to renouncing a genre of
epistemology tied to a “prior grounding” model of justification (and its
assumption of a fundamental epistemic claimant-challenger asymmetry), in
favor of a new one that throws its lot in with a “default and challenge” model.
1.1. In his book Epistemic Luck (2005), Duncan Pritchard agrees with
epistemologists who regard the elimination of epistemic luck as being a
fundamental adequacy condition on their theory. Still, he argues that the received
view of the project of analysis adheres to an over-generalized and merely
platitudinous version of the claim that knowing excludes luckily true belief.
This is our first argument supporting an expansion of epistemology’s central
tasks: the high cost of assent to what Pritchard describes as the epistemic luck
platitude. Pritchard insists that discriminating the kind or kinds of epistemic luck
at play in radical sceptic arguments is requisite to developing a more
5
philosophically satisfying diagnosis and response to scepticism. We need better
taxonomies and theories of epistemic luck, he argues, but we need a change of
direction, too, or rather a substantial broadening of our anti-luck interests.
Externalist luck has been the over-whelming focus in epistemology in recent
decades, starting with discussion of the structure of Gettier cases. But if, given
only what the agent is able to know by reflection alone, it is lucky that her belief
is true, then a kind of epistemic luck has interceded that is it neither that of
Gettier cases (veritic externalist luck), nor fake barn cases (environmental
externalist luck), yet is a kind that could be thought to inhibit our ability to
responsibly discount sceptical alternatives to our belief. Pritchard calls this
reflective luck, and traces its role to radical (or Cartesian) scepticism. It is the
anti-luck concern with reflective luck that most directly stokes the passions of the
sceptic, yet this isn’t well-acknowledged in a literature that focuses around the
Gettier problem.
Addressing the latter problem and others (e.g. fake barn cases) that turn on
externalist luck remains an important concern of epistemologists into the future,
Pritchard maintains, since the intuition is strong that luck of a kind that can
intervene ‘betwixt belief and fact’ will preclude an agent from knowing, even
though her belief may be true and personally justified. 2 But fallibilists about
knowledge will be wary of the assumption that progress in epistemology demands
its complete exclusion. Moreover, what is at minimum right about the demand for
the exclusion of luck is captured in the intuition that if one knows then it ought
not to be the case that one could easily have been wrong. A level of “safety”
must characterize the process through which a target belief is acquired, and such
6
a demand goes together with an externalist or truth-linked account of our
capacity for knowing. But what else might be a relevant concern? We rarely show
awareness of the specific kinds of epistemic luck our favorite thought-
experiments turn upon, and when we do pay attention to this it becomes clear
that not all anti-luck demands are equally valid, and that their validity depends
upon the particular epistemic standing—for instance justification, knowledge or
understanding—one is concerned with. So repudiating our robust or undiscerning
epistemic luck platitude is still consistent with describing oneself, as Pritchard
does, as an “anti-luck” epistemologist.
The upshot of the anti-luck argument is that without first redressing the post-
Gettier era’s tunnel-vision on veritic luck, the project of analysis simply cannot
serve the anti-sceptical tasks it has widely been expected to serve. But to admit
the need for a more comprehensive approach to epistemic luck and its bearing on
our epistemic aims, severally, would already constitute a very significant
expansion of epistemology. While their bearing on our capacity for knowledge
and other valued epistemic standings may turn out to vary, both forms of
epistemic luck, the veritic and the non-veritic, Pritchard concludes, “need to be
responded to in any adequate epistemological theory” (2005). This conclusion
and its import for epistemologists is one we will discuss further when we return to
Pritchard in the second part of the paper.
1.2. Jonathan Kvanvig’s book The Value of Knowledge and the Pursuit of
Understanding (2003) and more recent papers can be used to illustrate another
and substantially overlapping set of challenges to the prevailing conception of
7
epistemology’s central tasks. Given that we seem right to value knowledge more
highly than mere true belief, there being no very obvious explanation of this leads
to the value problem. Epistemologists have been focusing myopically upon the
“nature of knowledge” question without giving due consideration to the “value of
knowledge” question which is its natural complement. According to Kvanvig,
accounting for the nature and the value of knowledge should be taken as
something like “twin desiderata on a theory of knowledge.” But he also insists
that when we first do so, we encounter a paradox of sorts. For what we find by
looking at the literature is “a repeated pattern in which progress with respect to
one desideratum is balanced by greater weakness with respect to the other”
(117). If this problem were endemic to it, then these twin desiderata might
become a ‘Scylla and Charibdis’ that sink the project of analysis. Showing why
this shouldn’t be the case is one challenge he proposes as meeting the need for a
significant expansion of epistemology.
This aspect of Kvanvig’s epistemic value argument has been influential enough
to spur examination of the structure of the value problem, of which extant
approaches are best-suited to deal with it, and of what further changes in
epistemology might be driven by it. As he concludes, “questions about the value
of knowledge and other epistemic aims are important in their own right and can
help evaluate the adequacy of proposals concerning the nature of knowledge or
other epistemic aims” (2008).
The other and stronger way in which we described Kvanvig’s epistemic value
argument highlights his rejection both of the analytic tradition’s platitudinous
adherence to epistemic value monism, and of its subsequent focus on
8
propositional knowledge as its focal concept. These two aspects of Kvanvig’s
epistemic value argument, “the slogan that truth is the epistemic goal” and the
ensuing focus on propositional knowing (or “knowing that”), are clearly connected
as can be seen in the recent exchange between Kvanvig and Marian David
(2005). There David’s monist (or veritist) stance leads him explicitly to adopt a
conception of epistemology as the analysis of knowledge. Kvanvig, to the
contrary, argues both that epistemic value monism is “too narrow a view of the
variety of epistemic values and goods,” and by extension that platitudinous
allegiance to epistemology as the analysis of knowledge is too narrow a view of
the epistemological project.
For the pluralist, “epistemic goals include knowledge, understanding, wisdom,
rationality, justification, sense-making, and empirically adequate theories….[T]he
class of epistemic goods is manifold, as wide as the class of cognitive successes”
(2005, 287). A pluralist axiology suggests the need to rethink many further
assumptions about epistemology’s central tasks and to readdress the issue of the
nature of epistemic normativity. Can an instrumentalist conception of epistemic

1
Compare Kornblith (2002) who also argues that we today need an approach that will “lend
substance” to debates over epistemological axiology, “rather than merely dismissing them as
conceptual confusions” as he says the Quine-Goldman veritist approach tended to due during the
post-Gettier era . The ‘received view’ that dismisses them is reflected by those who hold that “as far
as epistemology is concerned, belief and truth are given”; that “epistemology is the theory of
knowledge”; and that “once an account of knowledge is in hand, the task of epistemology pretty
much reduces to the task of giving a theory of justification.” (David, 2005; Kvanvig, 2005).
2
“Veritic Luck,” Pritchard clearly explains, “is the kind at issue in the counter-examples to the
classical tripartite account of knowledge that were famously advanced by Edmund Gettier (1963)”
(2005, 148). It is a major goal of Pritchard’s approach to be able to offer “a more general account of
what is required to eliminate veritic epistemic luck” (151). Although Pritchard has recently
distinguished between veritic epistemic luck and another kind of externalist luck, environmental
epistemic luck, the same implication holds. If neither of these is the kind that most directly motivates
radical skepticism, then we need to rethink the place of such externalist conditions within the overall
project of analysis of knowledge.
9
normativity survive apart from the support it receives from taking truth,
semantically conceived, as the single telos or ‘core’ aim of the life of the intellect?
We can look more closely at this question when we return to Kvanvig’s argument
in Part 2 of the paper. But to summarize, Kvanvig urges an expansion of
epistemology through study the nature and sources of epistemic value (295). If
the epistemic value argument functions as he intends, it “present[s] a theoretical
foundation for greater diversity of interest in epistemology” (2003, 188), showing,
among other things, that “epistemological inquiry deserves at least some
enlargement in the direction of concepts other than knowledge” (xvi). The
prescription flowing out of this second and stronger aspect of the epistemic value
argument, as we’ll see more fully when we return to Kvanvig’s work in Part 2, is
for a “value driven” epistemological axiology, or for what Pritchard calls a “value-
theoretic epistemology.”
1.3. Michael Williams, my third focal author, argues in Unnatural Doubts (1991)
and Problems of Knowledge (2001) that improving our philosophical responses to
radical skepticism first requires improving the “theoretical diagnosis” we give of
the sceptic’s motivations. Pritchard and Kvanvig I think would easily agree; but in
holding that we should identify and scrutinize the sceptic’s own philosophical
assumptions, Williams wants more than that we drop platitudinous acceptance of
skepticism as a purely logical problem. He strikes me as challenging numerous
other widely-presupposed views about the project of analysis and the role that a
reductive analysis of knowing is widely held to play in responding to radical
scepticism. Most importantly, Williams says that in conceiving our relationship to
10
a radical sceptical interlocutor, we must let drop our adherence to a genre of
philosophy shaped by a “Prior Grounding Model” (PG Model) of justification.
A good diagnosis of radical scepticism should not lead to a highly concessive
response, yet a response to scepticism is bound to be highly concessive when it
allows, as the PG Model does, a grossly asymmetrical view of our justificational
responsibilities and epistemic entitlements. Williams characterizes this model as
composed of four inter-related theses of an internalist and evidentialist
orientation.3 As Williams and Robert Brandom (2008) both argue, adoption of a
Default and Challenge Model (DC Model) constitutes a shift that levels the
playing field substantially between the philosopher and a radical sceptical
interlocutor, by making it apparent why challengers and claimants more
symmetrically share these justificational responsibilities. On the DC Model, “No
move in the game of giving and asking for reasons is presuppositionless” (2001,
150). The sceptical interlocutor inherits no right to unlimited demands for
justification; rather, “questions of justification arise in a definite justificational
context, constituted by a complex and in general largely tacit background of
entitlements, some of which will be default” (158).
3
Williams (2001, 147) analyzes the Prior Grounding Model into four interconnected theses that taken
together are exemplified in the internalism of Chisholm and many others, (arguably including
contemporary internalist evidentialists like Conee and Feldman): “(PG1) No Free Lunch Principle.
Epistemic entitlement—personal justification—does not just accrue to us: it must be earned by
epistemically responsible behavior. (PG2) Priority Principle. It is never epistemically responsible to
believe a proposition true when one’s grounds for believing it true are less than adequate. (PG3)
Evidentialism. Grounds are evidence: propositions that count in favor of the truth of the proposition
believed. (PG4) Possession Principle. For a person’s belief to be adequately grounded…the believer
himself or herself must possess (and make proper use of) evidence that makes the proposition
believed (very) likely to be true.” The possibility of rejecting PG 2-4 and of re-situating PG 1 in
relationship to the DC Model suggests to me that contra Feldman and Conee, there really are no good
arguments from epistemic responsibility to epistemic internalism. Bonjour (2003, 175-6) accepts this
point in his recent exchange with Sosa.
11
The demand for prior grounding reveals many of the ways in which internalist
ideas underlie certain arguments for neo-Pyrrhonian as well as Cartesian (or
radical) scepticism; indeed tacit adoption of the PG Model helps explain the
appeal of the most serious arguments for radical scepticism. In response to the
demand for a “completely general” understanding of the human capacity for
knowledge, Williams argues that if there are a plurality of ways in which
epistemizing justification might arise, then “the sceptic’s hyper-general questions
may be deeply flawed” (2005, p. 214).
By contrast with the internalism and tacit commitment to a PG Model that
drives that demand, one of the genuine insights of the externalist turn in
epistemology has been that there are a plurality of kinds of justification that may
serve to ‘epistemize’ true belief. If an externalist response to scepticism is to
carry force, it should be able to use this plurality of kinds of epistemizing
justification to its advantage. Yet tacit adherence to the PG Model in most extent
forms of externalism undermines its ability to do so, by making this plurality
impossible to square with the demand for a minimal set of necessary and
sufficient conditions on knowing free from serious counter-example —that is to
say, for a reductive analysis of knowledge.
Williams makes it clear that the PG Model is presupposed not only among
mentalist and access internalists, but also among externalists, in their doubtful
manner of answering certain kinds of questions. That this may weaken externalist
responses to scepticism is also pointed out by John Greco, when he describes the
inconsistencies of externalists who “go internalist” in treating the question of how
to understand defeating evidence to one’s belief (2004).4 So for Williams, trying
12
explicitly to effect an externalist turn in epistemology while tacitly holding onto a
PG Model creates no small degree of havoc; the cost of platitudinous commitment
to the PG Model is only failure to see how it “both generates the threat of skepticism and
constrains our responses to that threat” (2001, 188).
Adherence to the Prior Grounding platitude constitutes a profound dialectical
mis-step for anti-sceptical philosophy. Williams’ epistemic compatibility argument
indicates that anti-sceptical philosophers are better-served to reconcile than to
bifurcate between their explanatory interests in epistemic reliability and
epistemic responsibility. Bifurcation occurs regularly in the genre of epistemology
his critique targets, and it leads to just the kind of inconsistency between the
treatment of supporting and defeating evidence just mentioned. The PG model
cannot perform this marriage operation between first and third-personal interests
in agent-appraisal. The expansion of epistemology that Williams looks for occurs
when detachment from it allows us to see justification “as exhibiting a Default and Challenge
structure, where constraints on the reasonableness of challenges and the appropriateness of
justifications are contextually variable along several dimensions” (245). This isn’t just ‘changing
the topic,’ as sceptics contend. As I’ll later want to argue, the support that
epistemic compatibilism is able to provide for substituting the PG with the DC
Model, is the main issue that today needs to be considered by epistemologists
concerned to improve the adequacy of philosophical responses to scepticism.

4
In “Holding Defeat to the Fire,” Greco (2004a) writes, “But when reliabilists go internalist regarding
defeaters, this must result in a schizophrenic approach to justification and evidence. After all,
reliabilism insists on a reliabilist account of evidence in favor of a belief. But then how can the same
theory plausibly understand evidence against belief differently? Such a strategy seems at best ad hoc.
At worst, it is theoretically incoherent” (3). Greco’s claim that this asymmetrical treatment of
counter-evidence is an “under-recognized problem for reliabilism,” seems very illustrative of
Williams’ point that most reliabilists remain tacitly committed to aspects of the PGM, and that
‘foundationalism’ and ‘internalism’ are often two terms for the same thing.
13
Part 2. New directions in epistemology
In this second Part of the paper, lets go back over each of the three internal
critiques of the project of analysis we described, looking more closely at how they
inter-relate, and trying to draw out more specific suggestions for fruitful new
directions in epistemology.
2.1. We earlier identified a central theme in Pritchard’s book Epistemic Luck as
the philosophical costs of uncritical adherence to the epistemic luck platitude.
Pritchard’s rebuke of the platitude indicates a need to avoid pre-occupation with
veritic luck (in the Gettier problem and its offshoots). Now responsibilists might
readily agree that the anti-luck argument exposes an undue narrowness of
professional interest in the externalist kind or kinds of epistemic luck. Luck’s
intervention into the veritic gap—the gap ‘betwixt belief and fact’—is not the only
area of anti-luck concern, and on closer inspection is shown not as relevant to our
best diagnoses of radical scepticism as once thought. It has been a mistake to
judge the adequacy of our response to scepticism on the actual or even the
possible complete exclusion of veritic luck in our analysis of knowing, a mistake
engendered by a dubious understanding of Gettier’s challenge. In one ill-founded
reading, Gettier’s challenge is to secure a “fourth condition” on knowledge to
prevent re-gettierization, while essentially holding in place the older internalist-
evidentialist understanding of the third (or personal justification/achievement)
condition; on another, the search for such an externalist principle leads
epistemologists to deny the need for personal justification altogether, and to opt
14
for radical reliabilism. But even if readers think radical sceptical arguments
deserve to be the central concern in epistemology as they have been in recent
decades, Pritchard still argues that viewing the problem of radical scepticism
through the left lens of anti-luck concerns and the right lens of epistemic value
concerns, serves to highlight for us a quite different challenge (2008b).5
One key difference between myself and Pritchard is that I see exposing the
narrowness of interest in externalist luck as being corrected by interest in
evidential epistemic luck as the kind most directly concerned with the quality of
the agent’s efforts in inquiry—her zetetic efforts in pursuit of virtue-relevant
goals. While we are both neo-Moorean in our responses to scepticism, I see
epistemic compatibilism as crucial to the support of this response, and direct
concern with the quality of the agent’s efforts at inquiry as crucial to the support
of epistemic compatibilism. The ancient zététicós was the seeker who “proceeds
by inquiry.” Zetetic is a Greek term that indicates direct concern with the quality
of an agent’s motives and efforts in pursuit of intellectual ends. The very notions
of evidential and doxastic epistemic luck may be unthinkable apart from
assessment of the quality of the agent’s zetetic activity: her active, inquiry-
directed strategies and activities. Evidential epistemic luck intercedes when “it is
lucky that the agent acquires the evidence that she has in favor of her belief”;
doxastic (and possibly circumstantial) luck intercedes when “it is lucky that the
agent believes the proposition” (2005). Pritchard treats both of these as “benign”
5
I do not mean to say, however, that responsibilists are particularly interested in the kind of luck that
Pritchard calls reflective luck and sees as a key motivation for scepticism. To the contrary as will be
argued in section 3, a responsibilist approach might make it easier to identify what is misguided in
the sceptic’s blanket demand for its exclusion.
15
kinds of luck as far as knowledge and understanding go, and sets them aside
early in his book. But what can it mean to call the issues raised by kinds of luck,
so defined, as epistemically ‘innocuous’ and ‘insignificant’? (2005, 137). If we
agree with John Greco that in respect to epistemic responsibility, “etiology
matters,” then epistemic responsibility as it pertains to knowing is not captured
by synchronic epistemic rationality: that time-slice assessment of an agent
performance purely ‘by their own lights,’—that is, by evidence they presently
possess and epistemic norms they presently accept at that reconstructed
moment, whatever their validity. By contrast with this ideal of pure epistemic
rationality keyed to deontic duties with regard to things the agent can assess by
her own lights, as we find it in internalists such as R. Feldman, appraisals of
epistemic responsibility I hold must factor in the quality of the zetetic activities of
the agent in appropriately seeking, acquiring, weighing, and updating evidence
for what they believe.
Responsibilists can clearly find advantage in Pritchard’s anti-luck argument,
but might resist his dismissive treatment of the non-veritic kinds of luck (apart
from reflective luck). I would further argue that dismissing evidential and doxastic
luck as “benign” functions to impede Pritchard’s intended “final value” (or ‘value-
of-achievements’) response to the Meno problem. Credit theories require that we
get clear on the sense or senses of luck capable of undermining attributions of
creditable success to agents over a range of virtue-relevant aims. But this
suggests the need for a significant expansion of epistemology to elucidate, as
Wayne Riggs proposes, “the conceptual connections among a family of concepts
that include credit, responsibility, attribution, and luck” (2007a); or again as
16
another responsibilist, Jason Baehr, proposes, “the proper role of the ‘character’
(or reflective) virtues in epistemology.”6 The acquired intellectual virtues are
habits and dispositions that meet the reliabilist criterion of promoting true beliefs
and minimizing false ones. Why, then, they ask, are issues of epistemic
creditability in respect to the agent’s acquired habits and dispositions so rarely
deemed to connect directly with the reliabilists’ preferred epistemological
projects? (Baehr, 2006; Dalmiya, 2002).
Since theoretical understanding turns out to be a distinctively valuable epistemic standing
on Pritchard’s approach as well as on Kvanvig’s, we find in both of their arguments the clear
implication of a need to study both it and all other epistemic standings that are
bearers of value. So on responsibilist accounts, in my way of putting it, the etiology
issue and the axiology issue both matter, philosophically: To ask, ‘What the
bearers of epistemic reliability?’ is to ask after the right kinds of process type and
actions-in-inquiry; to ask, ‘What the bearers of epistemic value?’ is to ask after
the right value type for the kinds of agent appraisal we are undertaking. I want to
retain these concerns with the bearers of reliability and of value as two ‘master
intuitions’ that need to be accommodated (compare Pritchard 2008c). But
seeking to uphold the first plunges us into the generality problem, where the
difficulty is describing at the right level of generality the type of process or
6
Baehr argues that “virtue reliabilists do not make a principled exclusion of intellectual character
virtues from their repertoire of intellectual virtues. There is nothing in their formal definitions of an
intellectual virtue that would prevent character virtues from qualifying as intellectual virtues in the
relevant sense. Nevertheless, when they go on to develop their accounts of intellectual virtue and its
role in knowledge, they tend to focus exclusively on cognitive faculties and abilities” (2006b, 199-
200; see also his 2006a). Since Baehr argues that this neglect “leaves virtue reliabilists unable to
account for some of the most valued kinds or instances of knowledge,” his argument could easily be
construed as another forceful ‘internal critique’ exposing how an insular conception of
epistemology’s central tasks winds up undermining epistemologists’ own best interests.
17
actions-in-inquiry whose reliability determines whether a token of a process type
produces a safe belief. And seeking to uphold the second pulls us into the value
problem, where the difficulty is to identify the kind or kinds of value associated
with the epistemic standings we hold valuable. On the approach taken by virtue
epistemologies, we want to identify the right kinds of process with those that
derive from the agent’s stable abilities and competences; and on the approach
taken by credit theorists, we want to identify the right kind of value with what
Pritchard terms final value, the kind attaching to cognitive achievements qua
achievements.
On these matters, those I call virtue reliabilists and responsibilists largely
agree. But virtue responsibilists go on to ask, ‘Why should we treat the
intellectual virtues as epistemically important or central concepts only if they
contribute to a successful theory of propositional knowing, or again, if they are
concepts that can be conceived in terms of truth, semantically conceived?’
Responsibilists think that the intellectual virtues and their role in the intellectual
life are “epistemologically interesting in their own right” (Baehr), and are
furthermore concerned that the received view’s conception of epistemology
truncates the concept of epistemic virtues. Neither the internalist’s focus on
synchronic evidential rationality nor the externalist’s focus on what Sosa calls the
“aptness” of belief can adequately capture the role of the reflective virtues, which
readily extend beyond belief acquisition to its maintenance, transmission, and
change as other essential functions of our overall cognitive economy. The
reduction of epistemology to the theory of knowledge—a shared assumption of
veritisms of both the left and the right—also functions to inhibit our conception of
18
the role of the reflective virtues in the intellectual life, since firstly we seek
interesting truths, and secondly our virtue-relevant goals include not just knowing
and believing truly, but also higher or more distinctively human epistemic
achievements such as theoretical understanding.
Due to concerns such as these, it has sometimes been suggested that a
responsibilist research program into the reflective virtues can only take shape
after being relocated outside of analysis of knowledge proper. Following this line
of reasoning, we come to attempts to expand epistemology through one or
another version of a “Two-Project Proposal.” One could call for a “trial separation
between the theory of knowledge and the theory of justified belief” or something
similar, while at the same time insisting upon a conception of epistemology broad
enough to acknowledge the importance or ‘centrality’ of both projects.’ 7 This kind
of proposal is typically made by internalists (Foley, 2004), and sometimes taken
seriously by sympathetic externalists as well (see Pritchard 2006, 55); moreover
it is often cast in terms of a pessimism regarding our ability to contain our
different anti-luck concerns in a single view.
Connected with this, Baehr writes that “moderate” forms of virtue
epistemology take issue with “conservative” forms by insisting that there are
“substantive theoretical issues and questions pertaining to intellectual virtue that
7
Pritchard sometimes appears to endorse a version of the Two Projects proposal. As he states them,
“The first is to analyze those epistemic concepts that are closely tied to responsibility—namely,
epistemic rationality and justification. The second is to analyze knowledge” (2006, 55). This
characterization of the two projects has the virtues playing a role outside of analysis of knowing, but
a nonetheless significant role in providing attendant explanatory stories about how we acquire
knowledge. However, the position Pritchard takes in favor of an “anti-luck virtue epistemology” in
2008b and 2008c is quite inconsistent with this, and seems to indicate a significant shift in his
position towards what I call the DATA-form analysis (Axtell 2007a; 2008a), where an areteic
condition isn’t taken to be ‘swamped’ by an anti-veritic luck condition.
19
can be addressed in relative isolation from traditional epistemology” (2008b).
While I agree with the point as here stated, I have elsewhere argued that the Two
Project Proposal Foley makes isn’t, on closer inspection, an attractive option for
responsibilists to pursue (Axtell 2007a). While it does appear to secure a kind of
co-existence and compatibility between ‘externalist’ and ‘internalist’ research
programs after a period of prolonged antagonism, the Two Projects Proposal is as
Williams (this volume) puts it, “a bifurcating strategy” best avoided due to its
steep hidden costs with respect to our ability to side-step scepticism. I will later
argue that a more robust epistemic compatibilism—a compatibilism made
possible by convergences between virtue theory, credit theory, and the Default
and Challenge Model—is the real ‘road to Larissa.’
2.2. In Section Two we drew attention to Kvanvig’s proposal that the desiderata
of competing accounts of a positive epistemic standing such as propositional
“knowing” or theoretical “understanding” should include their ability to address
together both the nature and the value of that standing. This proposal he
explicitly makes in order to present “a theoretical foundation for greater diversity
of interest in epistemology.” The post-Gettier era focused on providing a
reductive analysis of propositional knowing primarily out a concern to respond to
radical scepticism. What we have called his epistemic value argument explains
how this era’s focus devolved, leading into an unsustainably narrow conception of
epistemology’s central tasks.
Kvanvig’s argument issues into his call for greater diversity of interests, and
this as we mentioned has provoked a growing literature on epistemological
20
axiology. Are some kinds of knowledge more distinctively valuable than other
kinds? Are there epistemic standings of greater value than propositional knowing,
and if so, should they thereby occupy a more central place in our studies? One of
Kvanvig’s primary theses is that “understanding” has a special kind of value that
other epistemic states such as knowledge, simpliciter, do not; this fact, he holds,
threatens the rationale for the focus on knowledge that the history of
epistemology displays (2008a; 2008b). Kvanvig’s and Pritchard’s epistemologies
are both explicitly “revisionist” of the intuitions that drive the Meno Problem,
since knowledge ‘simpliciter’ turns out not to have a value always greater than
that of its component parts. But even without endorsing such a revisionist stance,
Elgin, Riggs, Zagzebski, and others have each made arguments to a similar
effect: if epistemology follows the trail of epistemic value, then “understanding”
should be accorded a more central place.
The epistemic value argument also leads to questioning whether analytic
epistemologists’ concern should remain focused on the framing of a ‘minimal set’
of conditions on propositional knowing, or extend to higher, distinctively human
epistemic standings irrespective to their connections to the project of analysis.
But the expansion of epistemology proposed on the basis of the epistemic value
argument typically runs considerably deeper than this. Kvanvig is also a critic of
the epistemic value monism that stands as a key shared assumption in the staled
debates between internalists and externalists. Epistemic value monism is the
thesis that all other epistemic values derive their status as epistemic values from
the value of truth as the fundamental and non-derivative epistemic good. This has
been the predominant account of our epistemological axiology throughout the
21
post-Gettier era, and is perhaps best exemplified in what Alvin Goldman terms
veritism.8 Its wide-spread influence is evident by the fact that commitment to an
instrumentalist conception of epistemic normativity (or epistemic rationality) is
prevalent not only among reliabilists like Goldman, but also among ardent
internalists like Conee and Feldman (2004) who agree with the reliabilists on little
else. Advocates of epistemic value pluralism reject this shared veritist axiology,
insisting that in addition to true belief and knowledge there are further, relatively
independent, basic epistemic values such as understanding, wisdom, rationality,
justification, and empirically-adequate theories (Kvanvig, 2005). To the extent
that Conee, Feldman and Goldman continue to base commitment to
instrumentalism on assumption of epistemic value monism, then the consensus
that has held during the post-Gettier era in favor of the adequacy of an
instrumentalist account of epistemic normativity will also be challenged by the
epistemic value argument.
A turn from epistemic value monism to pluralism potentially has the deeper
effect of rendering epistemological axiology a continuous concern: study of the
nature and sources of epistemic value across all of our positive epistemic
standings becomes a more central and continuous endeavor. Similar points are
made among responsibilists in Baehr’s (2007) “value pluralism conception” of the
value problem, in Hookway’s call for a shift from the “doxastic paradigm” to
“epistemology-as-inquiry,” and more generally in Riggs’ call for a “value turn” in
epistemology. As Riggs puts the proposal that has been behind much recent
8
See Kvanvig (2008) for his reply to Goldman and Olsson’s (2008) defense of process reliabilism
against the swamping problem, and to Pritchard’s claim that the swamping problem for reliabilism
turns on a failure to recognize a distinction between intrinsic value and final value.
22
interest among responsibilists in a pluralist axiology, we should “put the need to
find the bearers of epistemic value at the fore of…epistemological theorizing”
(2007).9
Many forms of both internalism and externalism ignore the social dimension of
knowledge (Williams 2001, 36). Lorraine Code's responsibilism, by contrast,
remains connected with feminism and with social epistemology, as is evident in
her recent work on "the power of ignorance" in shaping epistemic practices that
result in exploitation and oppression. How are epistemic goals, norms or desiderata (e.g.,
knowledge, truth, warrant, justification, rationality, and consensus) promoted or impeded by various
social practices? Linda Zagzebski, Miranda Fricker, and Christopher Hookway each
also develop versions of responsibilism that provide responses to questions such
as this. Zagzebski’s stated reasons for seeking a virtue epistemology include not
only the impasse between internalist and externalist accounts of epistemic
justification, and a neglect of understanding and wisdom as other epistemic ends
worthy of study, but also a lack of direct concern with the practical ends of
theoretical reasoning, and a certain blindness towards social dimensions of
epistemic practice. Miranda Fricker's recent book Epistemic Injustice: Power and
the Ethics of Knowing, elaborates further convergences between virtue theory
and social epistemology. Fricker’s book studies the ethical and political aspects of
our epistemic conduct, and works to support these as proper concerns for
epistemologists. Social identity and relations of power impact our collective

9
Pritchard agrees that it is unfortunate that so much work on the value of truth/true belief “has taken
place independently from the debate regarding the value of other epistemic standings like knowing
and understanding” (2007b, 28). It has been a positive implication of the “value of knowing” debate
that we “shift the focus on contemporary epistemological theorizing away from the merely minimal
conditions for knowledge—a focus that has arisen largely in response to the Gettier problem—and
move it towards higher epistemic standings” (23).
23
epistemic practices, and especially the assessments we make of testifiers and
their testimony. I agree with Fricker’s call for “a truly social epistemology,” and
think that such a proposal naturally finds support in the responsibilist approach
within virtue theory.
While Fricker is critical of veritist axiology, she is equally critical of neo-
Aristotelian virtue epistemologies like Zagzebski’s, which draw upon an
epistemological analogue of an Ideal Observer Theory to anchor the normative
concepts of epistemology. What Zagzebski calls “exemplarist Ideal Agent theory”
is a radical form of responsibilism in which the exemplarist approach is held to be
“as useful in epistemology as in moral theory” (133). I describe this as phronomic
virtue responsibilism in order to capture the primacy it accords to the judgments
of ideal agents as exemplars of full virtue. However, Zagzebski’s phronomic virtue
responsibilism is, on my view, given to an exaggerated conception of what
epistemically responsible believing consists in, and is subsequently unable to take
proper advantage of the genuine insights of externalism. As Fricker rightly
objects,
If credit accounts are to work for knowledge-in-general, it is essential that a certain
'moralism' is avoided. The moralism I have in mind is that of forcing all kinds of credit into a
moral mould, namely that of credit one incurs from having a good ‘motive’ (and which
consequently renders praise and blame responses appropriate). I think Zagzebski's view is
vulnerable to this charge of moralism, and the problem stems directly from embracing a
certain fully ethical conception of virtue and applying it uniformly to epistemology.10
24
What I term inquiry-focused or zetetic virtue responsibilism provides
another counterpoint to Zagzebski’s view. Phronomic virtue responsibilism insists
as fervently on an “internal” connection between ethical and epistemic norms as
others internalists like Feldman do on a kind of purely epistemic appraisal
untouched by practical and ethical norms. Both extremes are to be avoided.
Inquiry-focused responsibilism, as an alternative to both, is more Deweyan than
Aristotelian, in the sense that it locates the role of the reflective virtues in relation
to an agent’s zetetic activities within particular contexts of inquiry. 11 Perhaps all
forms of responsibilism concern, as Christopher Hookway puts it, “how it is
possible to be good at inquiry rather than, more simply, what it is to have justified
beliefs or knowledge” (2006, 101).12 But developing the model of “epistemology

10
Fricker 2007b. She continues, "The natural order of explanation marks a point of disanalogy between
virtue epistemology and virtue ethics. In virtue ethics, it is a thoroughly plausible idea that the value
of the various goods that virtues aim at cannot be specified independently from the values of the good
motives animating the virtues. It is entirely plausible to say there is a non-vicious circularity in how
we characterize these values—the virtuous agent is motivated towards the good, and the good cannot
be specified independently from what motivates the virtuous person. But this becomes, at best, a far
less plausible idea when transferred to the field of epistemological value. For it is all too easy to
specify the value of truth, and thereby the knowledge that captures it for us, in purely practical terms
without reference to our epistemic motives...."
11
A concept of epistemic responsibility or justification entirely unconcerned with the zetetic appraisals
—that is, with quality of evidence resulting from the agent’s responsible efforts at inquiry—is what
Goldman terms ‘own lights internalism’ and points out that it is fully compatible with ‘epistemic
sloth.’ While this kind of account has defenders at least in Conee and Feldman, on the present
account, this kind of deontic justification turns out to be an answer in search of a question.
12
For zetetic responsibilists like Hookway, Olson and I, it is useful to describe the kinds of intellectual
habits around which epistemological appraisal turns as those that are “inquiry focused” (Axtell,
2007a). Thick descriptions of particular virtues like self-trust, conscientiousness, intellectual honesty,
humility, perseverance, etc. provide insight into how agents developmentally acquire habits that
extend their competence as cognitive agents and problem-solvers. The inquiry-directed functions of
the virtues are one reason the classical pragmatists so often seized on the issue of habit and its
inculcation to explain the manner in which critical intelligence is applied to establish and improve
social practice. The pragmatist who says that knowledge is that upon which a person is prepared to
act, is effectively saying that knowing is of the nature of habit, such that any sharp division between
knowing and acting can result only in what Dewey described as a ‘spectator theory’ of knowledge.
25
as theory of inquiry,” zetetic virtue responsibilists such as Hookway think that,
“The target of epistemic evaluations lies in our ability to carry out inquiries, to
reason effectively and solve problems” (98). On the account of Dewey and other
pragmatists, even theoretical inquiry is a practice, an activity. While a veritist
axiology sharply divides knowing and acting, proponents of zetetic responsibilism
or epistemology-as-inquiry reject any radical dichotomy between the ‘theoretical’
and the ‘practical,’ and instead re-instate the ‘primacy of practice’: “[S]ince
reasoning is a goal-directed activity, the norms that govern reasoning will include
norms of practical reason: we are evaluating strategies for solving problems, the
effectiveness of agents at putting their strategies into effect, and so on”
(Hookway 100).
So utilizing my new taxonomic distinction between phronomic and zetetic
forms of virtue responsibilism, I note that both have the potential to support a
closer integration of social epistemology with projects deemed central in the field,
but that zetetic or “inquiry-focused” responsibilism isn’t given to those problems
Fricker (and Olson, 2006) identifies in Zagzebski’s neo-Aristotelian or phronomic
virtue responsibilism. Zetetic responsibilism seems to me better able to account
for the genuine insights of reliabilism and to provide support for social
epistemology. It has a further advantage if it makes it easier to acknowledge
what Williams calls “ought-to-be rules” as genuine rules of criticism, and the
agent’s zetetic activities in conformity with them as conferring epistemic credit or
value on the agent even when not followed self-consciously or “deliberately.”
Credit, like achievement, is clearly something that comes in degrees; this
should not be obscured by the point that credit theorists associate achievements
26
with a specific kind of value: final value in Pritchard’s technical vocabulary.
Understanding and reflective knowing are valuable because achievements are valuable, and
successes that cannot be explained as due to the agent’s creditable exercise of a cognitive ability of any
kind cannot plausibly be thought of as achievements. But how does this relate to actions that are near
‘automatic’ because highly habitual or even dispositional? Here the advantages of zetetic over neo-
Aristotelian or phronomic responsibilism, due to its more restrained account of epistemic
responsibility, become readily apparent. Our reliability-knowledge typically requires
conformity with ought-to-be rules, and not necessarily also with more demanding
“ought-to-do” rules, the latter being rules that are imperatival in form (Williams,
this volume). For an agent to accrue epistemic credit for acting virtuously, it
mustn’t be thought necessary that she be aware of all the situational cues to
which she responds, or even that she be consciously aware of her epistemic aims.
Habitual actions are actions repeated so frequently that they are performed
with little or no conscious attention or deliberation. This may be so with respect to
our reliability knowledge as well while we operate under default entitlements, and
why Williams describes its role in epistemic appraisal as primarily “negative,”
entering typically in response to a specific challenge to an attribution of
creditable success. Always having a reflective ‘perspective’ on our epistemic
reliability (Sosa) seems too strenuous a demand. What matters, philosophically, is
that agents are sensitive to the limits of their reliability in different epistemic
situations, and to a “relevant” range of possible defeaters to default entitlements.
To the extent that agents indeed are in fact sensitive in some degree, reflective
reason remains a ‘silent partner’ in our distinctively human mode of knowing. This
is so even when the target belief may appear to be an instance of merely ‘brute
27
externalist knowledge’ (Pritchard) or ‘animal knowledge’ (Sosa) that results from
an innate faculty virtue like keen eyesight. Habitual actions are easy to construe
as ‘mere acts,’ obscuring the distinction between habitual behaviors which are
virtuous, and which are not. Agents are still the authors of their habitual actions,
and these actions are explained not as one would compulsions or bodily
processes, but as conforming or failing to conform to genuine rules of criticism.
In order that some degree of credit for achievement accrues to the agent
for cognitive success with respect to virtue-relevant goals, the virtues need only
figure among salient factors in explanation of the agent’s success. Achievements
are creditable successes, and these in turn are successes through ability. A
belief’s being true ‘because of’ virtue is the way the issue is typically framed in
virtue epistemologies, but we should be wary that this language also carries an
ambiguity: the ‘because of’ relationship will vary with the agent’s context. Credit
attribution will, when one speaks of success, in some contexts presuppose a
causal explanation, and in others an explanation matching the epistemologist’s
interest in conformity with particular norms, standards, or ‘rules.’ The only way
not to conflate between these is to be sure that we explicitly take the notion of
success being ‘because of’ virtue as a flexible notion from the start. The undue
restriction of attributions of achievement or final value just to what conforms with
the more demanding ought-to-do rules—rules in imperatival form—, in particular,
turns out to be but another way of making an inflated demand on epistemic
responsibility.
When success with respect to true belief or any other epistemic aim is
attributable to one’s abilities, or to performance with respect to a competence,
28
innate or acquired, it is a cognitive achievement and a bearer of value in the
sense required for a final value response to the Meno Problem (Kvanvig;
Pritchard). Williams’ use of Sellars’ distinction between these two types of rules
may be thought to support the credit theorist’s final value response to the Meno
problem. For what is of chief importance isn’t the degree of credit or how that might relate to what
factor is singularly most salient in rational reconstructive explanations (Lackey), but that the value, in
whatever degree it comes, be of the right kind. And this is what is carried when Williams points out
that both kinds of rules in question are genuine rules of criticism. So habitual and even apparently
‘automatic’ zetetic activities, as I will argue more fully in the final section, can be seen
to merit the agent a degree of that distinctive kind of value—final value—that
attaches to cognitive achievements as achievements.
2.3. We previously mentioned the critique of the Prior Grounding Model of
justification, a critique made by the several accomplished heirs of Wilfred Sellars,
including Michael Williams and Robert Brandom (2001; 2008). Williams identifies
this model as the prevalent conception both among internalists and, perhaps
more intriguingly, among many externalists as their way of retaining the
importance of personal justification or epistemic responsibility. Our platitudinous
commitment to a prior grounding paradigm of philosophical reason must be
exploded. “It is easy to miss the fact that the practice of justifying is only activated by finding
oneself in the context of a properly motivated challenge” (2001, 150). Adopting the PG Model
turns out to be scepticism-inviting, but Williams wants to show us that it can
reasonably be set aside. We can maintain that knowing is essentially connected with
justification, but only if we detach ourselves “not only from the classical demonstrative ideal and its
29
infallibilist descendants, but also from all conceptions of justification that insist on respecting the Prior
Grounding [Model]” (245). With the dialectical shift from the PG Model to a Default
and Challenge (or DC) Model, the problem of infallibilism, closure, and even
underdetermination-based arguments for radical scepticism is more effectively
side-stepped.
Williams argues that the PG and DC Models “set different standards for
epistemic responsibility, hence for epistemic entitlement” (170). In other papers I
have taken this point and tried to develop a virtue epistemology that can
contribute to and take advantage of Williams’ and Brandom’s prescribed change
of venue (2007a; 2008a). This shift is one that facilitates what Sven Bernecker
(2006) aptly describes as “epistemic compatibilism,” the thesis that kinds of
epistemic appraisal that drive internalism and externalism about justification can
be part of a single theory. 13 Williams’ distinctive version of epistemic
compatibilism is developed in Problems of Knowledge as the claim that “We can
preserve the link between knowledge and justification without accepting the Prior
Grounding Requirement” (2001, 148). From a responsibilist position,
compatibilism is both an advantageous and a plausible thesis, even though
radical reliabilists like Bernecker judge its prospects to be poor. The burden on
the compatibilist is not to argue in support some grand reconciliation of
‘internalism’ and ‘externalism’ as competing accounts of the very same thing,
which of course would be non-sensical; it is rather (and merely) to make sound

13
Sven Bernecker (2006) characterizes as “epistemic compatibilists” those who hold that it is possible
to “combine satisfactorily internalist and externalist features in a single theory.” I regard Williams’
account as an advantageous form of epistemic compatibilism, and a cogent response to Bernecker’s
dour assessment of the “prospects of epistemic compatibilism.” In these respects, the Sellarsian
tradition is like the descriptions of virtue epistemologies that Sosa, Greco and Zagzebski all provide,
as “mixed theory.”
30
philosophic sense of the conceptual connections between our explanatory
interests in agent responsibility and in agent reliability, especially as they relate
to cognitive success with respect to various epistemic goods. 14
I will end with four specific suggestions for developing the philosophic
import of Williams’ epistemic compatibility argument. Firstly, I want to formalize
my earlier claim that the distinction between ought-to-be and ought-to-do rules is
pertinent to the issues of epistemic value as previously discussed, and can even
be utilized directly in support of the “final value response” to the value problem
(as currently favored by a number of credit theorists; see also Carter 2007). We
earlier noted that appraisal of an agent’s epistemic responsibility and reliability
may in some cases focus on behavior that is habitual and even near-automatic.
But why should this be a strike against crediting the agent who performs them?
When no situational features have prompted an agent to bring goals and
strategies for pursuing them to the level of conscious deliberation, an accredited
agent remains in a Default context, where her habitual zetetic activities may still
be taken as purposive, and as rational in the sense of being done for a reason.
Some authors allege deep problems in credit theory, however, by arguing that
knowing is sometimes less than an achievement. To apply Williams’ distinction
against this objection, and in support of a final value version of the credit theory,
it needs to ward off worries about purported “K-SANS’ cases—cases of knowing-
14
Internalists in the past typically argued that epistemic responsibility implies
compliance with epistemic norms which must be accessible upon reflection for
the agent. But this route from responsibility to a general internalist condition is
difficult to maintain, as some self-described internalists now concede (Feldman,
281; Bonjour, 175-6): There are no good arguments either from epistemic
responsibility or from the notion of creditable success, to any sort of general
(access or mentalist) internalist condition.
31
without-achievement (Lackey 2007; Pritchard 2008b). Plausibly, one can argue as
follows:
1. Final value accrues to the agent’s acts/actions whenever they are in conformity
with genuine rules of criticism.
2. Ought-to-be rules (and not just ought-to-do rules) are genuine rules of
criticism.
____________
with ought-to-be rules (and not just ought-to-do rules).
with ought-to-be rules (and not just ought-to-do rules).
5. The agent’s acts/actions in K-SANS cases are in conformity with ought-to-be
rules.
____________
6. Final value accrues to the agent’s acts/actions in K-SANS cases.
Secondly, I want to allege that the connection between the sceptic’s anti-luck
demands and the prior grounding requirement is closer than has been recognized
in the literature. While Pritchard’s diagnosis locates the primary motivations for
underdetermination-based arguments for scepticism in concern over the
32
pervasiveness of “reflective luck” over our human epistemic condition, Williams
locates these motivations in “spurious generalizations” that give rise to a prior
grounding conception of justification. But these I see as two ways of putting
almost the same diagnostic point: Our reasonableness in refusing the PG Model is
directly connected to our ability to decline the sceptic’s demand for insulation
from reflective epistemic luck. Both ways of putting the sceptic’s demand bear
upon epistemically desirable conditions, but they support the premises of
(Cartesian or other) underdetermination-based arguments for radical scepticism
only if justification and knowledge aren’t to be had without them. If we can
instead accept Williams’ diagnosis and reasonably decline the Prior Grounding
requirement and the right to ‘naked challenges,’ then since this is what gives rise
to the demand for the blanket exclusion of reflective luck, we can reasonably
decline that demand as well.
Thirdly, virtue epistemologies not only address the issues of epistemic
value discussed earlier, but also support epistemic compatibilism. It is precisely
for this reason that they are described by Greco, Sosa, and Zagzebski (whatever
their other differences) as “mixed theories.” Williams’ key claim that we can
preserve conceptual ties between knowledge and personal justification without
accepting the Prior Grounding Requirement appears to mean several things. It
means first to support some connections recognized in the standard or JTB
account of knowing and arguably reflected in the intuition that knowing is
generally more valuable than merely believing truly. As previously said, it also
means that discarding the PG in favor of the DC Model is properly seen as a key
upshot of the externalist turn in epistemology. This allows us also to resist
33
austere externalisms: those that Williams terms “radical” and that Brandom
describes as yielding to the ‘naturalistic temptation’: that is, “to suppose that the
concept of reliability of belief-forming processes can simply replace the concept
of having good reasons for belief” (2000, 373). Compatibilists maintain that while
on fallibilist premises it can be no ‘guarantor’ of success, the epistemic
responsibility of agents is nevertheless conceptually connected with epistemic
entitlements and evaluations. When the ‘change of venue’ to the DC Model
occurs, we find to our advantage that the conceptual connections between our
first and third-personal perspectives on epistemic appraisal can be more easily
supported and developed. The ‘What is knowledge?’ question (related to the
project of analysis) and the ‘What are we doing?’ question (related to epistemic
entitlements and implicatures) are more easily separable as distinct concerns.
Since proponents of credit theories, virtue epistemologies, and the DC Model
have each in various ways explicitly championed epistemic compatibilism, the
relationships between these three mutually-reinforcing ‘legs’ of the epistemic
compatibility argument need to be more closely studied.
The final suggestion regards a more specific area where the synergies
between the DC Model and virtue epistemologies seem to me especially strong.
This regards the manner in which the DC Model seems to require distinguishing
between contexts in which “thick” and “thin” terms of epistemic appraisal are
called for. Virtue-theoretic concepts, I argue, provide the right kinds of conceptual
resource to service that need. So if the DC Model is to be more widely embraced,
epistemologists will need to think carefully about the thin/thick distinction as it
relates to the epistemic appraisal of agents. Both thick and thin descriptions of
34
intellectual character find an important role in the explanations needed to
discharge our obligations to a sceptical interlocutor. This is not a proposal like
those debated recently in ethics, for a move either “from” thin to thick theory, or
vice-versa. It is simply the claim that realizing the resources of the DC Model will
require that we find the proper role of both thickly and thinly-described character
traits in the logic of explanation.
The inadvertent ‘thickening’ of our accounts of justification has often led to
under-motivated debate and miscommunication, according to Heather Battaly. “A
concept is maximally thick when all of the necessary and sufficient conditions for
its application are fixed—when all of its boundaries are precise” (105) Given this
definition, it is clear that knowledge and justification are typically approached as
maximally thick concepts, and the project of analysis is typically conceived in this
way. What Heather Battaly describes as “thin concept analysis” by contrast
argues instead for a flexible set of conditions on terms such as “justification” and
“epistemic virtue,” and I argue that on the DC Model this is all that is owed the
radical sceptical interlocutor unless and until a motivated challenge arises to an
attribution made about a particular agent in a particular case. Flexibility as the
Taoists say is the companion of life, rigidity the companion of death. A ‘thin’ or
merely formal areteic condition on knowledge, I want to argue, may also be all
that the DC Model requires to support the bearers of value, or what Pritchard puts
better as our “master intuition” regarding knowledge-as-achievement (2008c).
If positive epistemic standings like “knows” and “is justified” are ones that
have multiple independent conditions of application, then philosophers should
rightly decline the demand for a thickly-stated yet ‘completely general’ analysis,
35
an analysis that would specify ahead of all challenges just what combinations are
necessary, or which are sufficient, for proper attribution. The demand for
reductive definition of knowledge in terms of maximally thick conditions isn’t
reasonable in the kind of situation where, as William Alston thinks, we are
working with a “family relations concept” with no fixed conditions at all. Nor is it
reasonable where, as Battaly alternatively thinks, the parties to the dispute do
agree to some fixed functions of justification, but what they share is still only a
“thin conception” of these functions. The demand for thickly-stated sufficient
conditions on knowing appears on her diagnosis to arise from spurious
generalization: the sceptic pines for a maximally thick characterization of
epistemic conditions, when the diagnosis instead reveals that only a rather thin
and formal one is reasonably demanded of human knowers acting in their Default
context. Thickening the areteic (or any other achievement-related) condition can
be a way of placing a general internalist condition on knowing, inadvertently or
intentionally, and assuming a potentially inflated conception of our epistemic
responsibility. By contrast, acknowledging the “combinatorial vagueness” of
epistemic terms best characterized as thin concepts merely anticipates that same
plurality of possible combinations for epistemizing justification that Williams and
Brandom both insist is one of the guiding insights of the externalist turn in
epistemology. The proposal for thin concept analysis might therefore accomplish
in less objectionable (because more “unified”) form what Sosa tries to accomplish
in a “two-tiered” model of justification that divides between more minimal
conditions on “animal knowledge” and more demanding conditions on “reflective
knowledge.”
36
This proposal for thin concept analysis by virtue responsibilists like Battaly
and myself might also be thought of as supporting Williams’ attempts to show
that, “The conception of epistemic responsibility that emerges from adopting the
default and challenge model is considerably less demanding” than that
associated with the PG Model. In the Default context it is just the plurality of
possible sources of epistemizing justification that we want to maintain. But when
a defense commitment is engaged about an attribution in a particular case, then
our default analysis of knowing based upon a thin conception of achievement-
through-competence must give way to agent-specific explanations. In a
successful defense we would expect the virtues, whether faculty or reflective, to
be salient aspects of the best explanation for the agent’s success. If an attribution
of knowing is to be upheld, we generally want to be able to appeal to the agent’s
virtuous epistemic agency in order to preserve what would be the normal
attribution of credit had no motivated challenge come to light. A particular
agent’s success with respect to an epistemic aim being ‘because of’ or ‘due to’
her cognitive ability will (in cases at the high end of the spectrum) need to be
fleshed out by appealing to thick characterological features of that agent.
If this is correct, then we have shown that virtue-theoretical explanations
may well be what are needed by proponents of the DC Model in order to meet the
fuller accounting of the anti-sceptical obligations we incur on each of its two
recognized contexts, Default and Challenge. Proponents of the DC Model need a
fuller account of epistemic responsibility, and can surely find a resource in the
flexibility of virtue-theoretic concepts that can be ‘bent’ towards either the faculty
or the reflective virtues (or ought-to-be or ought-to-do rules) as called for by the
37
triggering of a Defense commitment in a particular, real-world agent and case. My
view is that the benefit of potential synergies works in both directions: Adopting
the DC Model has been seen in this paper as providing epistemologists a way to
better realize the resources of characterological explanations, and as Kvanvig
desired, a way to ‘loosen the shackles fastened on epistemology by skepticism’
during the post-Gettier era.
These are my present suggestions for expanding epistemology, developed
as far as space allows. Readers are, of course, welcome and invited to draw their
own conclusions about new projects and directions in epistemology.
Responsibilism as I have thought of it focuses on developing a research program
into the reflective intellectual virtues, and I have here used three overlapping
internal critiques of the analytic tradition in order to help clear the path for such a
program. A responsibilist approach, as it has here been described, is continuous
with the tradition of analytic epistemology, yet highlights numerous new
challenges. It also, I think, leaves us with an optimistic assessment of our
epistemological future(s.)15
15
I would like to thank Kelly Becker, and others who participated in or helped sponsor the “Expanding
Epistemology” session at the 2007 APA Pacific meeting. Comments on my draft made by Miranda
Fricker, Michael Williams, Robert Brandom, Jason Baehr, Adam Carter, Ward Jones and Wayne
Riggs were quite helpful, as were further comments on the “expanding epistemology” thread made by
participants at the JanusBlog (http://janusblog.squarespace.com).
38
Works Cited
Axtell, G. 2008a. “Virtue-Theoretic Responses to Skepticism,” in J. Greco, (ed.)

Oxford Handbook of Scepticism. Oxford: Oxford University Press,
forthcoming.
Axtell, G. 2008b. “From Internalist Evidentialism to Virtue Responsibilism,” in T.
Dougherty (ed.) Evidentialism and its Discontents. Oxford: Oxford University
Press, forthcoming.
Axtell, G. 2007a. “Two for the Show: Anti-luck and Virtue Epistemologies in
Consonance,” Synthese 156.
Axtell, G. 2007b. “Epistemic Virtue,” in M. Steup (ed.) Blackwell Companion to
Epistemology, 2nd ed. Oxford: Blackwell.
Axtell, G. (ed.). 2000. Knowledge, Belief, and Character: Readings in Virtue
Epistemology. Lanham, Md: Rowman & Littlefield.
Baehr, J. 2007. “Is There a Value Problem?,” in The Value of Knowledge, eds.
Adrian Haddock, Alan Millar, and Duncan Pritchard (Oxford: Oxford UP),
forthcoming.
Baehr, J. 2006a. ‘Character in Epistemology’, Philosophical Studies 128, 479-514.
Baehr, J. 2006b. “Character, Reliability, and Virtue Epistemology”, Philosophical
Quarterly, 56: 193-212.
Bernecker, S. 2006. ‘Prospects for Epistemic Compatibilism’, Philosophical Studies Vol. 20, no. 1: 81-
104.
Bonjour, L. 2003. “Reply to Sosa,” in Bonjour and Sosa (eds.) Epistemic
Justification. Malden, Ma.: Blackwell, 173-200.
Brandom, R. 2008. Between Saying and Doing. Oxford: Oxford University Press.
Brandom, R. 2000. Articulating Reasons. Cambridge, Ma.: Harvard University
Press.
Carter, Adam J. 2007. “A Happy Triumvirate: Virtue Epistemology, Contextualism,
and the Safety Principle,” typescript.
Code, L. 1987. Epistemic Responsibility. Hanover, N.H.: University Press of New
England.
Conee, E. and Feldman, R. 2004. Evidentialism. Oxford: Oxford University Press.
David, Marian. 2001. “Truth as the Epistemic Goal,” in M. Steup (ed.) Knowledge,
Truth, and Duty, 285-296.
Dalmiya, V. 2002. “Why Should a Knower Care?,” Hypatia 17, 1: 34-52.
Elgin, C. 2005. “From Knowledge to Understanding,” in S. Heatherington (ed.)
Epistemology Futures, 199-215.
Feldman, R. 2005. “Justification is Internal,” in Sosa and Steup (eds.), 270-284.
Foley, R. 2004. “A Trial Separation between the Theory of Knowledge and the
Theory of Justified Belief”, in J. Greco (ed.), Sosa and His Critics, 59-72.
Fricker, M. 2008. “Scepticism And The Genealogy of Knowledge: Situating Epistemology In
Time,” this volume.
Fricker, M. 2007a. Epistemic Injustice Power and the Ethics of Knowing. Oxford;
Oxford University Press.
Fricker, M. 2007b. “The Value of Knowledge and the Test of Time,” typescript.
Fricker, M. 1998.“Rational Authority and Social Power: Towards a Truly Social
Epistemology,” Proceedings of the Aristotelian Society, Vol. XCVIII, 2, 159-177.
Goldman, A. 2002. Pathways to Knowledge. New York: Oxford University Press.
39
Goldman, A. and Olsson, E. 2008. “Reliabilism and the Value of Knowledge,” in A.
Haddock et. Al. (eds.), Epistemic Value, forthcoming.
Greco, J. (ed.). 2004. Sosa and his Critics. Oxford: Blackwell.
Greco, J. 2004. “Holding Defeat to the Fire,” typescript.
Greco, J. “Virtue Epistemology,” in Stanford Encyclopedia of Philosophy, Online.
Haddock A., Millar, A. and D.H. Pritchard (eds.). 2007. Epistemic Value. Oxford:
Oxford University Press, forthcoming.
Heatherington, S. 2005. Epistemological Future(s). Oxford; Oxford University
Press.
Hookway, C. 2000. “‘Regulating Inquiry’, in Axtell (ed.) (2000), 149-160.
Hookway, C. 2006. ‘Epistemology and Inquiry: The Primacy of Practice,’ in S. Heatherington (ed.)
Epistemology Futures, 95-110.
Kornblith, H. 2002. Knowledge and its Place in Nature. Oxford: Oxford University
Press.
Kvanvig, J. 2008a. “Further Thoughts on the Swamping Problem,” typescript. APA
Central, 2007.
Kvanvig, J. 2008b. “The Value of Understanding,” typescript. APA Pacific, 2007.
Kvanvig, J. 2005. “Truth is not the Primary Epistemic Aim,” in M. Steup and E.
Sosa (eds.) Contemporary Debates in Epistemology, 285-295.
Kvanvig, J. 2003. The Value of Knowledge and the Pursuit of Understanding.
Cambridge: Cambridge University Press.
Kvanvig. J. 1992. The Intellectual Virtues and the Life of the Mind. Lanham, Md:
Rowman & Littlefield.
Lackey, J. 2007. “Why We Don’t Deserve Credit for Everything We Know,”
Synthese 156.
Olson, P. 2006. ‘Flourishing Rationality: A Pragmatic Approach to Virtue and
Reason’. Unpublished dissertation.
Pritchard, D. 2008a. 'Sensitivity, Safety, and Anti-Luck Epistemology,' in J. Greco
(ed.) The Oxford Handbook of Scepticism, forthcoming.
Pritchard, D. 2008b. “Knowledge, Understanding and Epistemic Value,” in
Epistemology, (ed.) A O’Hear (Cambridge University Press).
Pritchard, D. 2008c. “The Value of Knowledge” (2007 APA Pacific symposium
paper and handout)
Pritchard, D. 2007a. “Anti-Luck Epistemology,” Synthese 156.
Pritchard, D. 2007b. “Recent Work on Epistemic Value,” American Philosophical
Quarterly, forthcoming.
Pritchard, D. 2006. What is this Thing Called Knowledge. New York: Routledge.
Pritchard, D. 2005. Epistemic Luck. Oxford: Oxford University Press.
Riggs, W. 2007a. “Why Epistemologists are so Down on their Luck”, Synthese,
forthcoming.
Riggs, W. 2007b. “The Value Turn in Epistemology,” in New Waves in
Epistemology, V. Hendricks & D. H. Pritchard (eds.). Ashgate, forthcoming.
Sinnott-Armstrong, W. (ed.). 2005. Pyrrhonian Skepticism. New York: Oxford.
Steup, M. and Sosa, E., (eds.) Contemporary Debates in Epistemology. Oxford:
Blackwell.
Williams, M. 2008. “Reliability and Responsibility,” this volume.
Williams, M. 2005. “The Agrippan Argument and Two Forms of Skepticism," in W.
Sinnott-Armstrong (ed.) Pyrrhonian Skepticism, 121-45.
40
Williams, M. 2004. “Scepticism and the Context of Philosophy,” Philosophical
Issues, 14: 456-475.
Williams, M. 2001. Problems of Knowledge. Oxford: Oxford University Press.
Williams, M. 1991. Unnatural Doubts. Oxford: Blackwell.
Zagzebski, L. 2005. “Ideal Agents and Ideal Observers in Epistemology,” in S.
Heatherington (ed.) Epistemology Futures, 131-148.
Zagzebski, L. 2003. ‘Epistemic Value and the Primacy of What We Care About’,
Philosophical Studies 33: 3, 353-377.
41
Notes
42
Draft version. Forthcoming (2012) in an special topics issue of Journal of the Philosophy of
History, on intersections of historicism, naturalism, and virtue epistemology, edited by Mark
Bevir and Herman Paul. All rights reserved. Contact author: gsaxtell@radford.edu
The Dialectics of Objectivity
Guy Axtell
Abstract:
This paper develops under-recognized connections between moderate historicist
methodology and character (or virtue) epistemology, and goes on to argue that their
combination supports a “dialectical” conception of objectivity. Considerations
stemming from underdetermination problems motivate our claim that historicism
requires agent-focused rather than merely belief-focused epistemology; embracing
this point helps historicists avoid the charge of relativism. Considerations stemming
from the genealogy of epistemic virtue concepts motivate our claim that character
epistemologies are strengthened by moderate historicism about the epistemic virtues
and values at work in communities of inquiry; embracing this point helps character
epistemologists avoid the charge of objectivism.
Keywords: historiography, virtue epistemology, naturalism, objectivity, thick
concepts, underdetermination problem

1. Introduction: Thick Concepts and a Thicket of Problems
Debate over historiography and its methods is already witness to a number of instances
where historicist thinkers have appealed to one or another form of character epistemology.
But it is still not widely recognized that historicism and virtue theory are mutually
supportive in ways that make attempts to combine them philosophically appealing or
advantageous. To show this it will be argued that considerations stemming from
underdetermination problems show that historicism requires agent-focused rather than
merely belief-focused epistemology. It will also be argued that considerations stemming
from the genealogy of epistemic virtue concepts show that character epistemologies are
strengthened by moderate historicism about the epistemic virtues and values at work in
communities of scientific inquiry.
So the advantage of the admixture of moderate historicism and character
epistemology for philosophers, and for philosophers of history in particular, is this paper’s
constructive thesis. Historicism has been much-discussed by philosophers of history, but
what does character (or virtue) epistemology bring to the table? Among the general
advantages that character epistemologies have over their belief-focused rivals is their ability
to explain the important roles that “thick” concepts play in guidance of research strategies
and in theory confirmation across the sciences and humanities. I will attempt to show why
this is important for philosophers of history and for everyone concerned with the
relationship between the physical and the social sciences.
Section 2 introduces the constructive thesis by identifying a dialectical conception of

objectivity as a shared theme of authors who have already partially combined their moderate
historicism with some version of virtue epistemology. I provide four somewhat different
examples of such an approach. The constructive thesis is picked up again in Section 5,
where we examine more closely at the thick concepts on which the dialectical model
revolves. “Calling them ‘virtues’ rather than ‘values’ draws attention to their status as
attributes at once objective and desirable”,1 and we develop both this Janus-faced conception
of virtue theory and how moderate historicism aids in the achievement the tasks that it sets
for itself. Yet as we’ll see the methodological functions thick concepts commonly play in
heuristic advise and in theory confirmation vary substantially with one’s field or discipline;
they do not appear to track the traditional distinctions between “hard” and soft” sciences,
“explanatory” and “interpretive” theories, etc.
To see what they do track we will first need to ask why some historiographers appeal
primarily to impersonal theory virtues as ampliative (non-deductive) desiderata of theory
choice, while others instead appeal to the personal character or bon sens of researchers
themselves. Are we more tempted to dispense with this distinction between theory virtues
and personal virtues in some fields of research than in others? This is the burden of Sections
3 and 4, and the diagnostic thesis I develop in these two sections answers these questions in
a way that further supports to the constructive thesis when we return to it. Ernan McMullin
rightly notes that “Discussion of theory virtues exposes a fault-line in philosophy of science
that [separates] very different visions of what the natural sciences are all about”. The
1
E. McMullin, “The Virtues of a Good Theory”, The Routledge Companion to Philosophy of Science,
eds. M. Curd and S. Psillos (Routledge, 2009), 501.

diagnostic thesis might be seen as developing this point to expose fault-lines both narrower
(within historiography) and wider (between disciplines) than McMullin intended.2 My thesis
regarding the latter debates is that the weight that authors accord to a) empirical testing, b)
impersonal theory virtues, and c) the personal intellectual traits of inquirers typically reflects
the conception that they have of what constitutes objectivity within their own discipline. The
diagnostic thesis also and for the same reason has something to say about the nature of the
fault lines between historians themselves over how to understand historiographic methods. It
more specifically explains why philosophically-inclined historians who antecedently hold
substantially different conceptions of what constitutes historiographic objectivity have
tended to adopt and support distinct versions of virtue epistemology. Historiographers today
appeal to different versions of character epistemology for reasons that we can articulate, and
once articulated this provides some new tools for mediating their disputes, which is just
what we should constructively want to do.
2. First Outlines of a Dialectics of Objectivity
Authors who exemplify a combination of historicist assumptions and virtue theory appear to
maintain that historiographic objectivity as they understand it entails a “condition of character”3
of some weaker or stronger sort. Their accounts, insofar as they involve themselves in questions
about what constitutes norms of historiographic objectivity, all appear to be variations of a
dialectical conception, something that sets them off from accounts of disciplinary objectivity that
2
McMullin, “The Virtues of a Good Theory”, 506.
3
M. Bevir, The Logic of the History of Ideas (Cambridge University Press, 1999), 104.
are not part of this family.
Now objectivity has often been denatured by both universalizing proponents and reactionary
critics. But for any character epistemologist, objectivity is an achievement concept. It is an ideal
with widely varying applications in the scholarly domains in which it plays a methodological
role, yet with abiding value. The achievement that objectivity represents, and that other terms
like “neutral”, and “value free”, fail to capture, partly explains its enduring value of the ideal as I
see it. A “dialectics” of objectivity suggests a negotiative conception of the means and ends of
inquiry, one with clear reference to the active, adjectival sense of the objective inquirer. In
historiography and in praxis-oriented accounts of method in some other fields as well, objectivity
has sometimes even as we’ll see been developed primarily as a virtue concept, one attended by
various sub-virtues — “virtues of self-distanciation” — as Thomas Haskell and Herman Paul
both refer to them.4
We can utilize Allan Megill to give this thesis some initial credibility, showing that the term
is already recognized in the literature. In Rethinking Objectivity (1994) Megill outlines four
competing senses of objectivity: the absolute, the disciplinary, the dialectical, and the procedural:
A striking feature of both absolute and disciplinary conceptions of objectivity is
their negative relation to subjectivity. Absolute objectivity seeks to exclude subjectivity;
disciplinary objectivity seeks to contain it…. Phrases like ‘aperspectival objectivity’ and
4
See for instance H. Paul, “Distance and Self-Distanciation: Intellectual Virtue and Historical Method
Around 1900”, History and Theory, Theme Issue 50 (2011), 104-116; also T. Haskell, Objectivity Is Not
Neutrality: Explanatory Schemes in History (Baltimore: John Hopkins University Press, 2000).
‘view from nowhere’ really draw attention to… negativity. In contrast, dialectical
objectivity involves a positive attitude toward subjectivity. The defining feature of
dialectical objectivity is the claim that subjectivity is indispensable to constitute the
objects. Associated with this feature is a preference for ‘doing’ over ‘viewing’... in other
words, subjectivity is needed for objectivity; or, as Nietzsche put it, ‘objectivity is
required, but is a positive quality.’5
While other approaches to objectivity may also be able to recognize the interplay with
contrary conceptions of method, Megill takes the special emphasis of a dialectical approach as
how something is constituted as an object for inquiry through interplay between researchers and
that which they study. Objectivity, proponents of a dialectical model hold, is not a property of a
right method or a steady state, but rather describes a process carried out actively through
communicative interaction and comparison. Now my claim would be that researchers who
combine their historicist assumptions with some form of virtue epistemology are among the
fittest champions of the dialectical model which Megill has introduced for us. In order to show
this, let me now give brief descriptions of four such authors. While the emphasis in this section is
on the many substantial similarities between them, these descriptions will also serve to introduce
certain differences that will be our direct focus in the Section which follows.
Our first example of a dialectical conception is Mark Bevir’s chapter, “Objectivity” in his
Logic of the History of Ideas (1999). Bevir calls for epistemology to take an anthropological
turn: “We must define objectivity as a human practice based on intellectual virtues. When people
5
A. Megill (ed.), Rethinking Objectivity (Durham, NC: Duke University Press, 1994), 8.
debate the merits of rival theories, they engage in a human practice governed by rules of thumb
which define a standard of intellectual honesty”. Objectivity in the history of ideas, Bevir’s
special focus, he thinks “rests on a combination of agreement on certain facts, an extensive use
of criticism, and a comparison of rival views in relation to clearly defined criteria. Historians
cannot pronounce their particular theories to be decisively true or false, but they can make
rational decisions between rival webs of theories, and thereby pronounce their theories to be the
best currently available to us.’6 Again later he writes,
Our logic of comparison contains a form of justification appropriate to the history of
ideas. Historians can justify their theories by showing them to be objective, where
objectivity arises not out of a method, nor a test against pure facts, but rather a
comparison with rival theories. Historians can justify their theories by relating them, in
a comparison with their rivals, to criteria of accuracy, comprehensiveness, consistency,
progressiveness, fruitfulness, and openness.7
Bevir’s criteria for theory choice among competing histories is a set of thick concepts, epistemic
virtues that he thinks are clearly defined and accepted by most practitioners in the history of
ideas. The majority of them are cognitive values or theory virtues (virtues of theories and
hypotheses) framed in impersonal language.8 We will simply term them theory virtues when we
6
Bevir, The Logic of the History of Ideas, 112-113.
7
Bevir, The Logic of the History of Ideas, 143.
8
Ernan McMullin says that “Calling them ‘virtues’ rather than ‘values’ draws attention to their status as
return to look more closely at Bevir’s account.
A second example of a dialectical conception of objectivity is Jon A. Levisohn’s paper,
“Negotiating Historical Narratives" (2010). There he writes that,
Historiographical inquiry is appropriately characterized as a negotiation among
narratives; the historical narratives, rather than emerging from the inventive mind of the
historian, are generated by a process of negotiations; and …this conceptualization
enables us to escape from the picture of historical narratives being imposed by the
historian on an unnarrativised past… Moreover, [this] descriptive account also contains
the seeds of the normative account of what makes one narrative better rather than
worse. The quality of the narrative is a function of the quality of the negotiation, of how
successfully—artfully, seamlessly, elegantly, insightfully—the historian negotiates
among in integrates the various elements that she encounters or introduces into the
inquiry. As we conceptualize the goals of history education, the cultivation of these
interpretive virtues is a good place to start.9
attributes at once objective and desirable”. McMullin, “The Virtues of a Good Theory”, 501.
9
J.A. Levisohn, “Negotiating Historical Narratives: An Epistemology of History for History Education”,
Journal of Philosophy of Education, 44:1 (2010), 18. “Each of the qualities just mentioned—
responsibility, creativity, openness to disconfirmatory evidence, boldness, modesty-- can be understood as
virtues of historical inquiry” (17). For a fuller development, see his The Interpretive Virtues: A
Philosophical Inquiry into the Teaching and Learning of Historical Narratives (Wiley-Blackwell, 2011
forthcoming).
Levisohn views historical narratives as inevitably evolving over time, since every generation
brings new questions and perspectives to bear on historical personages, events, or eras of the
past. In asking whether we have gotten the historical story right, “we implicitly are asking
whether we carried out the task of negotiating among narratives with responsibility, with the
right kind of creativity, with openness to disconfirmatory evidence, and with that particular
combination of boldness and modesty that marks good historiography—boldness that
accompanies a story that we believe others ought to endorse, together with the modesty that
derives from our knowledge of the ways that historical interpretations change over time” (17).
Herman Paul is our third example; “Inspired by a ‘performative turn’ in the history
and philosophy of science”, his paper “Performing History” “focuses on the historian’s
‘doings’ and proposes to analyze these performances in terms of epistemic virtue”.10 He
clearly sees virtue-based evaluations as a post-positivist approach to professional conduct,
and to quality assessment for historiographers. Paul also argues that that “[W]hereas
historical scholarship is shaped by epistemic virtues, such virtues, in turn, are shaped by
historical contexts”.11 His work elaborates this dialectic between professional norms and
historical contexts and goes on to suggest strong connections between the evaluation of
historiographic practices or “doings” and evaluating the traits of character of actual agents
who perform the research.
10
Paul, “Performing History”, 1.
11
H. Paul, “Performing History: How Historical Scholarship is Shaped by Epistemic Virtues”, History
and Theory, 50 (2011), 1–19, 11.

While Levisohn emphasizes that exemplars of good research, and even personal
traits (including internal motivations) of ideal inquirers taken more abstractly, are
important pedagogically for students of history-writing, Paul more explicitly presents them
as playing a “constitutive role” in the acquisition of scholarly knowledge. For Paul, virtue
theory of a sort that focuses on personal traits of researchers adds depth and specificity to
the “virtuous performances” that historians and others already recognize as normative for
professional conduct. Paul also wants to insist that the value of virtue concepts for
historiographical practice is not only retrospective (reconstructive), but prospective
(heuristic) as well. When we return to further examine Paul’s account, we will refer to the
kind of epistemic virtue he and Levisohn focus upon as personal virtues in order to
contrast it with the emphasis on impersonally-framed theory virtues we earlier saw in
Bevir’s approach.
Our final example of a dialectical conception of objectivity is Lorraine Daston and Peter
Galison’s recent book Objectivity (2007). They argue that concept of scientific objectivity has a
history, but that the epistemic norms that have informed scientific practice can be historicized
without leading to relativism. Only lack of historical awareness among philosophers of science
allows us to overlook how shifts transitions between explicit decision criteria and expert
judgment have occurred in the past. The authors’ historical and comparative approach to a study
of scientific atlases reveals more specifically a slow negotiation and transition between what they
claim are at least five distinct scientific “ways of seeing”, five competing conceptions of
objectivity: truth-to-nature, mechanical objectivity, structural objectivity, trained judgment, and
presentation. Part of the authors’ thesis is that alternative conceptions of objectivity abound, but
that neither logic nor history can make the choice among different conceptions of objectivity
together with their supporting epistemic virtues. Still, they hold, the choice exists, so that it’s
important to see what it hinges upon. While the authors do not argue explicitly for any one of the
main models that they see exemplified in science, the dialectical overtones in their treatment of
objectivity emerge from their descriptions of the pairing of each epistemic virtue with a
particular, value-charged conception of the “scientific self” of objectivity. Each recognized virtue
is one that compensates for or helps combat a cognitive or motivational “danger” for this
particular scientific self—one that this self is thought to be especially susceptible to. The rise of
the ideal of objectivity thus goes hand-in-hand with the acknowledgment of particular epistemic
virtues, virtues that are on the one hand practiced in order to know the world, but which on closer
examination “turn out to be literal, not metaphorical, virtues” in the sense that forms of scientific
self and epistemic strategies are mutually supportive.12
Let us summarize before moving on. Bevir is focused on history of ideas, Levisohn on
history education, Paul on the “doings” of historians, and Daston and Galison on a genealogy of
epistemic virtues concepts through comparative case studies. Yet each combines historicist
assumptions with appeal to virtue epistemology, and all four share key aspects of the dialectical
account of disciplinary objectivity that Megill initially described for us. Paul, along with Daston
and Galison, are the most concerned to show that what we deem to be epistemic virtues change
over time—that objectivity has a history, and that specific epistemic virtues, whether of the
personal or the theory virtue type, are valued more at one time than at another. Bevir and
12
L. Daston & P. Galison, Objectivity (New York: Zone Books, 2007), 39.
Levisohn, along with others including Frank Ankersmit13 develop “objectivity through
comparison” of competing extant theories, while Daston and Galison’s form of comparison
engages distinct images of the scientific self reflected over broad spans in the history of science.
But again, all of these authors assert view the thick concepts upon which they focus as enabling
objectivity, and appeal to them as helping historicists to avoid relativism. Each in this sense also
develops a further point that Megill described as a basic feature of a dialectical objectivity: A
central concern with epistemological (including meta-methodological and axiological) questions
of how objects of investigation are produced and how research programs are criticized and
compared.
Our four authors have each expressed what Paul calls historicist awareness of how
intellectual virtues are shaped by their historical contexts. But far from painting them with a
single brush, I now want to elaborate upon certain difference between some of them. So the next
section asks the very odd but I hope thought-provoking question why Paul (and Levisohn)
appeals primarily to desirable personal virtues of the writers of biographies and histories as
providing the forward-looking guidance and backwards-looking criteria of choice or evaluation,
13
The dialectical model is recognized by Frank Ankersmit when he writes that “I fully agree with
[Bevir’s] comparativist view of historical truth as expounded in the section of Chapter 2 on 'objectivity
through comparison': historical truth is not a matter of a correspondence between historical writing on the
one hand and past reality on the other, but requires us to compare a set of rival representation of the past
with each other and of seeing, next, which is the most satisfactory one of the set”. “In my book on
narrative logic of some twenty years ago I came to exactly the same conclusion - and on the basis of
much the same argument as Bevir's (Ankersmit 1983: pp. 235 – 47).
while Bevir’s instead appeals primarily to impersonally-framed theory virtues as the shared
desiderata: “criteria of accuracy, comprehensiveness, consistency, progressiveness, fruitfulness,
and openness”. Both authors engage central issues of historiographic method, yet they
intriguingly do so through appealing to quite distinct versions of virtue epistemology. Why is
this? It is a question that invites us to frame a diagnostic thesis about what makes these different
appeals attractive to each author.
3. Fault Lines: A Diagnostic Thesis
The difference between Bevir on the one hand and Paul and Levisohn on the other appears to be
explainable by differences in how they view objectivity in their fields. But I do not need or
intend to pick a side in the debate within historiography between emulationist, narrativist,
interpretivist, tropological, etc. conceptions of methodology. My purpose again is diagnostic, and
the diagnostic thesis holds that the different ways Bevir and Paul integrate epistemic virtues (and
virtue epistemology more generally), into their respective accounts are reflective of divergent
conceptions they hold of the objectivity that history-writing can aspire to.
Bevir appeals to theory virtues as “clearly defined criteria” helping us compare webs of
theories. “Objective knowledge”, Bevir writes, “arises from our exercising the virtue of
intellectual honesty within a practice of comparing rival webs of theories. If we disagree about
the relative merits of different views, we should draw back from the point of disagreement until
we can agree upon a platform from which to compare them, where the platform at which we
arrive will consist of agreed facts, standards of evidence, and ways of reasoning”.14 While the
14
Bevir, The Logic of the History of Ideas, 153.
ability to “draw back” clearly calls upon praiseworthy self-distancing habits of the inquirer and
ultimately on their intellectual honesty, it is the agreed theory virtues that the actual comparison
of theories proceeds upon.
By contrast, this layer of criteria, the theory virtues, goes virtually unmentioned by Paul
and Levisohn, two narrativists who I suspect would find Bevir’s appeal to it superfluous or
unworkable in the discipline of historiography. On the other hand, Paul’s and Levisohn’s
apparent collapse of the distinction between theory virtues and personal virtues might be thought
to invite relativism, as some forms of strong historicism do. Paul’s position as we’ll later see is
influenced by White’s “tropological” view of the nature of historical narration. Whether or not
that is so, Levisohn still more explicitly claims that success in the negotiation of narratives “must
necessarily be a quality of the activity of historical inquiry rather than the quality of the
historiographical product”.15
As we develop these focal contrasts, let us stipulate, in order to avoid an unnecessarily
demanding or restrictive account that impersonally-framed theory virtues, and inquiry-directed
personal virtues and their associated motivations, will be taken sub-species of epistemic virtues.
For the connection between theory virtues and virtue epistemology is not merely ornamental.
Both are directly concerned with ampliative (non-deductive) reasoning in the sciences.
Moreover, application of theory virtues to choose among competitors involves weighing these
theory virtues against one another, and thereby calls upon the judgment or what Pierre Duhem
would call the bon sens of the scientist. Hence a condition of character is clearly apparent even
where the explicit appeal is only to impersonally-framed theory virtues. Indeed both the appeal
15
Levisohn, “Negotiating Narratives”, 17.
to theory virtues and the appeal to bon sens remind us that the scientist qua scientist makes value
judgments.
But differences remain. Theory virtues—accuracy, comprehensiveness, consistency,
progressiveness, fruitfulness, etc— are impersonally described, and seem clearly to be
“cognitive” values or virtues, as distinct from ethical concerns. 16 The personal virtues by
contrast describe real or ideal excellences of inquirers, not of theories or hypotheses per se.
Perhaps for this very reason, they are less purely intellectual, and indeed those who emphasize
their role in inquiry, from Pierre Duhem to Daston and Galison, and to Levisohn and Paul, often
want to insist that they are or include character traits in the full Aristotelian sense; they reflect
norms internalized by appeal to ethical values and so cross boundaries between epistemic and
ethical evaluation.
To look more closely, there is good reason to think that strong narrativists will find most
appealing the character epistemologies which focus on personal intellectual virtues, and tend to
allow or even to insist upon a blurring of the distinction between the intellectual and ethical
dispositions manifested in good or praiseworthy research. In other papers I have called this
phronomic character epistemology (after the Aristotelian phronimos or person of practical
wisdom). What I think of as an extreme example of it is the character epistemology of Lynn Holt
in Apprehension: Reason in the Absence of Rules, a form of virtue epistemology in which the
apprehensive reasoning and sensitivities of experts in particular domains is explicitly claimed to
16
This does not mean to imply they are ‘purely epistemic.’ Pragmatism would be wary of that notion.
Some cognitive values like “simplicity” are often seen to be theory virtues, yet as much aesthetic and
pragmatic as an indicator of truth.

be constitutive of knowledge in their field. Holt’s calls this his “Apprehensionist” virtue
epistemology, and develops its theme of reason in the absence of rules in sharp contrast with
“Methodism” about reason and epistemic status. Highlighting the change in the direction of
analysis that has for better or worse come to be associated with character epistemology, he
argues that “To make a discovery is to be justified, precisely because the discovery flows from
virtuous activity”.17
Now I do not mean to suppose that Levisohn or Paul are committed to any so radical
version of character epistemology, but I do hope this example shows that what scientific and
historiographic objectivity mean may already be in play when, as desiderata of theory choice,
authors appeal to those personally or impersonally- framed thick concepts that we here designate
by the common term, epistemic virtue. Moreover, if I interpret them correctly, Paul and Levisohn
do indeed hold that the personal virtues of exemplary historians are not only instrumentally
valuable to their research practices, but partly constitutive of historiographic objectivity
(knowledge of the past) and of the positive epistemic status of historical explanations.
Now I want to argue that the viability of the distinction between personal and theory
virtues may as a general rule wane or sharpen as we move from one field of research to another,
depending on how prevalent are worries about underdetermination. Studying this question can
help us make better sense of divisions among scientific fields. To approach it we engage a recent
17
“In an apprehensive account of reason, the problem of conceiving how scientific discovery could be
rational is not the problem of supplying a logic or methodology of discovery. It is rather the problem of
determining the appropriate intellectual virtue”. L. Holt, Apprehension: Reason in the Absence of Rules
(Farnham, UK: Ashgate, 2002), 88.

debate among philosophers of science over Pierre Duhem’s account of bon sens (personal virtues
of good sense) of the scientist as ampliative desiderata of theory choice in his own field of
physics.
4. Duhemian Bon Sens and Differences between Disciplines
Considerations stemming from underdetermination and holism support our claim that moderate
historicism require complementation with agent rather than merely belief-focused epistemology.
It also helps us see why different authors appeal to distinct versions of virtue epistemology.
In David Stump’s paper “Pierre Duhem’s Virtue Epistemology”, the author takes
Duhem (not-mistakenly) to recognize an important problem of theory choice, one that would
later take on the name of the underdetermination problem and play a role in the downfall of the
account of metascience advocated by logical empiricists. He sees Duhem resolving
underdetermination worries by appealing to a cluster of virtues he calls bon sens, or ‘good
sense.’ Stump writes, “Duhem’s concept of ‘good sense’ is central to his philosophy of science,
given that it is what allows scientists to decide between competing theories. Scientists must use
good sense and have intellectual and moral virtues in order to be neutral arbiters of scientific
theories, especially when choosing between empirically adequate theories”.18 In Duhem's account
18
D. Stump, “Pierre Duhem’s Virtue Epistemology”, Studies in History and Philosophy of Science, 38,
149-159, 149-150. Duhem’s position seems close to many readers to conventionalism, and to the view
that there is simply no cognitive way to decide between empirically equivalent theories. Stump argues
that despite the fact that Duhem is sometimes read as a conventionalist, closer examination reveals
adequate support for the view that a scientist with good sense will choose between competing theories.
of scientific theory choice, there is openness, since strict rules and crucial experiments rarely
apply, but there is also objectivity. The source of this objectivity is the epistemic agent—the
scientist who acts as an impartial judge and makes a final decision: As Stump describes it,
“holism threatens to make testing impossible, yet Duhem believes that scientific consensus will
emerge. While the pure logic of the testing situation leaves theory choice open, good sense does
not”.19 Stump makes the strong claim that “Duhem maintains that the objectivity of the theory
selection rests with the scientist as an epistemic agent, not with nature”. So Stump’s neo-
Duhemian philosophy of science seems exemplary of Apprehensionist (or Phronomic) character
epistemology.20 The choices made by the virtuous scientist working under conditions of
underdetermination confers genuine though fallibilistic epistemic standing upon his or her actual
choice. This position, however, seems to stray far from naturalism.21 Stump’s critic, Milena
Consensus does emerge, and not for purely conventional reasons.

19
Stump, “Pierre Duhem’s Virtue Epistemology”, 154, emphasis added.
20
The personal virtues in question—impartiality, sobriety, intellectual courage, humility, rectitude, and
probity—Stump thinks (mirroring own thinking of the virtues of bon sens) are partly moral and well as
intellectual. Responsibilist epistemologists generally think that focusing on epistemic agency rather than
merely beliefs blurs the distinction between epistemic and intellectual evaluation.
21
It is of course controversial that some share of the epistemic value of a scientific theory should be
located in the traits of scientists themselves; this seems to reverse the usual direction of analysis. But
many have thought this is precisely what a virtue epistemology should do. Abrol Fairweather calls this the
“direction of analysis criterion”, noting that taking the agent rather than the belief as the ‘seat of
justification’ has been indicative of much responsibilist (if not also reliabilist) virtue epistemology.
Fairweather, “The Epistemic Value of Good Sense”, Studies in the History and Philosophy of Science,
Part A, Vol. 43,1 (2012), 139-146.
Ivanova is intent on showing that treating Duhem as a proto-character epistemologist is a
mistake; she counters with passage by Duhem suggesting that the role of bon sens regards only
theory pursuit and that it is only later revealed facts that confers epistemic status on the chosen
theory. This better satisfies Ivanova’s own empiricist views, on which only testing and empirical
fit decides the epistemic standing of a theory. Good sense “can be descriptive” of a basis of
choice, she thinks, “but it cannot provide us with a solution when faced with the problem of
theory choice”22; some (later-found) evidential support alone confers epistemic status on a belief
or theory. For her, then, neither form of ampliative desiderata, ampliative theory virtues (as
separated out from empirical adequacy) nor personal virtues pertain to theory confirmation, but
only to psychology and pursuit.
A third author who enters this debate, Abrol Fairweather in “The Epistemic Value of Good
Sense” (2012), re-introduces the distinction between theory virtues and personal virtues that
seems missing in the exchange between Stump and Ivanova. Also useful for our own purposes,
he ties it carefully to an overlapping distinction between contexts of UD inquiry and contexts of
non-UD inquiry. For our purposes we should understand contexts of UD inquiry to refer to a
local instances where theory choice is underdetermined even by a shared methodological
standards or a shared list of theory virtues. This makes sense of Fairweather’s claim that in
circumstances of UD inquiry the scientist’s “methodological cognitive character is not a
22
M. Ivanova, “Pierre Duhem’s God Sense as a Guide to Theory Choice”, Studies in History and
Philosophy of Science Part A 41(1):58-64, 63.

constitutive element”.23 We need not engage here the broader debate over how rare or common-
place UD inquiry actually is. Fairweather’s claim is that in UD inquiry it’s not at all clear that
appeals to the methodological or even the personal character of the researcher is not also
contributory to the epistemic status accorded to a theory. Agreeing with this point, like
Fairweather I think one can take a via media between Stump’s and Ivanova’s “unmixed”
responsibilist and reliabilist accounts of epistemic normativity. Fairweather argues that, “The
virtues of good sense do not have a constitutive role in generating the epistemic standing of
theories in non-UD inquiry. Method and evidence reign when they can, but epistemic
normativity becomes aretaic in UD inquiry with the express purpose of resolving
underdetermination. The success condition for the relevant virtue or ability is simply to break the
empirical stalemate in an appropriate way, where good sense supplies the relevant sense of
appropriateness….24
With this interesting thesis that epistemic normativity becomes aretaic in UD inquiry, and
the further claim that the virtue theoretic reading provides continuity between these two contexts
of inquiry, providing needed constraints on admissible resolutions of underdetermination, we
23
Although a scientist’s methodological character is developed through non-UD inquiry, he sees it as
having sensitivities to what David Henderson terms morphological content—the need to respond to global
features of an agent’s cognitive system to avoid problems associated with otherwise unmanageable
computational complexity. “[R]easoning with theory virtues will involve global morphological features of
theories, cognitive dispositions sensitive to these morphological features and non-rule-governed cognitive
transitions” (Fairweather, “The Epistemic Value of Good Sense”, 141).

24
Fairweather, “The Epistemic Value of Good Sense”, 141.
have come as far as we need to. Reducing desiderata of theory choice to personal virtues,
Fairweather rightly worries, invites the unmotivated introduction of radically new epistemic
values or virtues. But at the other extreme, if only evidence can “determine” theory choice as
Ivanova thinks then there should be no need for good sense at all; we seem to be ignoring
underdetermination problems because they don’t fit with a preconceived model of rationality,
rather than engaging with how such problems are actually resolved in scientific practice.25
The lesson for us, it seems to me, is not Ivanova’s, that only evidential factors can confer
epistemic status upon a theory or theory-choice. Rather it is that even moderate historicism about
the epistemic virtues at work in situations of underdetermination throws into question the ideals
of epistemic purity. The ethical connotation of these virtues is only a further mark of this
historical conditioning.26 Whether one accepts that virtue concepts play a constitutive role in
generating the epistemic standing of theories depends upon one’s particular field and whether or
25
John Zammito finds that, “Duhem’s concreteness ties in profoundly with recent efforts to get at science
as ‘practice’”; when extricated from Quine’s views, the lessons of Duhem’s moderate holism are easily
relatable to Zammito’s ‘moderate historicist methodology.’ A Fine Derangement of Epistemes: Post-
positivism in the Study of Science from Quine to Latour (Chicago, IL: University of Chicago Press, 2004),
22, 228.
26
One way in which the impurities are present, according to social and feminist epistemologist Heidi
Grasswick, is in the manner in which “attention to issues of epistemic responsibility inevitably blurs the
boundaries of ethics and epistemology”. H. Grasswick, “The Impurities of Epistemic Responsibility:
Developing a Practice-oriented Epistemology” in H. Nelson and R. Fiore (eds.), Recognition,
Responsibility and Rights: Feminist Ethics and Social Theory (Lanham, MD: Rowman and Littlefield,
2003), 101.
not it is regularly beset by worries about underdetermination. Fairweather argues cogently, I
think, that “the virtues of good sense do not have a constitutive role in generating the epistemic
standing of theories in non-UD inquiry” but that the case is not clear in respect to UD inquiry,
where epistemic normativity (aims of justification, rationality, truth and understanding) is in an
important sense aretaic.
My thesis with respect to Levisohn and Paul’s emphasis on character virtues over theory virtues
should now also be clear: Both start with such a narrative conception of historiography and this
leads them to develop the specific form of virtue epistemology that complements that conception
of their field. This will tend to see underdetermination of theory-choice by methodological norms
common-place in their field, and in such an interpretive field the character virtues are appealed to
as desideratum of the quality of professional works or ‘doings.’
An additional influence over Paul that makes an Apprehensionist character epistemology
appealing is Hayden White’s tropological conception of historiography. White argued that there
are very good reasons why history cannot be rendered a science without losing its identity as
history. These reasons have directly to do with rhetorical strategizing and with the impossibility
of excising the “imaginative element” from historical writing, as it might be excised in some
other types of academic writing. Historical writing is “more tropological than logical in kind”
(from rhetorical “tropes”), there being “an ineluctable poetic-rhetorical component” necessary to
the construction of its narratives.27 The historian, as perhaps others do not, needs to select or
“choose” the techniques of explanation and interpretation, and the choice is made to actively suit
27
H. White, “An Old Question Raised Again: Is Historiography Art or Science?” Rethinking History 4:3
(2000): 391-406, 391.

rhetorical purposes: hence the “tropological” character of historical narration.28 Originating from
what I think is a conception of historiography strongly influenced by White, Paul’s favoring of
personal virtues and relative neglect of the theory virtues Bevir appeals to is easily
understandable.
But now that our diagnostic thesis has been described and supported, we must ask what support
it in turn can add to the development of our constructive thesis. I would draw three main lessons
from it. First is the “local” and contextual nature of underdetermination worries. It is tempting to
over-generalize by thinking either that the theory virtue/personal virtue distinction tracks
differences between scientific and non-scientific research, or that epistemic normativity
“becoming aretaic” is as true in the sciences as in non-scientific fields. Sociologists of
knowledge have been sometimes promoted such overgeneralizations in countering those of
positivist metascience.29 By contrast with both, underdetermination worries as the diagnostic
28
White writes, “Thus, a tropological approach to the study of historical discourses seems imminently
justified if not required by the differences between historical and scientific discourses, on the one side,
and the similarities between historical and literary writing, on the other” (393-394). But one qualification
is that neither these writers nor White himself need to be seen as prescribing rhetorically-strategized
history. Ankersmit points out that White’s “tropological” view is primarily diagnostic or descriptive,
trying to make historians and philosophers alike more aware of the stumbling blocks facing would-be
scientizers of the field of history. If history-writing is more akin to journalism, then noticing rhetorical
dimensions is a wake-up call.

29
Porter issues a useful counterpoint, however: “[Trevor] Pinch implies that this is universal, that all of
science depends on judgments of character and skill. No doubt he is right. But rarely is informal, personal
knowledge so dominant as among the gentlemanly geologists and high-energy physicists. There are ways
thesis takes them do not necessarily track the distinction between hard and soft science, or
between nomothetic and ideographic disciplines: Duhem found UD inquiry and its attendant
worries about desiderata to be common-place in physics, often considered the epitome of hard
science. Our thesis thus lends support to independent arguments such as those of Thomas Nickles
“that methodology no longer will be a single, unitary subject but will, at the more interesting
levels of detail, breakdown into domain and context specific rules, practices, and advice”.30
The second way that our diagnostic thesis might serve our broader development of synergies
between historicism and character epistemologies is simply by supporting the dialectical
approach to objectivity. If there is any strength in its ability to expose and articulate “fault-lines”
within historiography, it becomes an invitation to talks about those differences in conceptions of
the field, just as on a broader scale it introduces a new way of mediating the old debate about the
relationship between methodology in the physical and social sciences. The third way that it might
serve is by highlighting for us the essential roles played by thick concepts both prospectively and
retrospectively—both as “desiderata” and “exemplars” for the practitioner, and at times as
“criteria” in the ampliative kinds of inferences that theory-confirmation often calls for.
5. Some Advantages of Character Epistemologies
Character epistemologies arguably have distinct advantages over their reliabilist and evidentialist
of making knowledge more rigid, standardized, and objective, and these have gone a long way to reduce
the need for personal trust”. T. Porter in Trust in Numbers: The Pursuit of Objectivity in Science and
Public Life (Princeton University Press, 1995), 223.

30
T. Nickles, “Good Science as Bad History”, 126.
rivals that are of direct relevance to the concerns of philosophers of history. They conceive
responsibility, a diachronic or longitudinal concept, to be an epistemically central concern, and
so have the advantage of being consonant with contemporary social epistemology, inviting both
genealogies and normative reconstructions of epistemic concepts like objectivity, rationality,
responsibility, etc.31
Are character epistemologies also better able to explain the intellectual creditworthiness
of scientists of bygone eras? “Newton was probably the greatest scientist who ever lived (the one
whose intellectual acts merited the greatest degree of praise); still, he was wrong wholesale––as
we now know”.32 Mainstream epistemologies, externalist and internalist, have a surprisingly
difficulty even squaring with our intuitions about the creditworthiness of the work past scientists
whose theories or methods are now overstepped. Let us briefly consider the reasons why, and
then look at concerns with how we ascribing to them and to others praiseworthy intellectual
motivations and habits of inquiry. This will lead to further connections between thick concepts
and historical explanation.
On reliabilist epistemologies, which focus on knowledge attributions, if a scientist’s work
31
Stephen Napier writes that, “One theoretical fall out of responsibilism is that it marks a shift away from
analyzing epistemic concepts (e.g. knowledge) in terms of other epistemic concepts (e.g., justification) to
analyzing epistemic concepts with reference to kinds of human activity…Much of analytic epistemology
focuses on epistemic concepts, whereas the responsibilist focus is on epistemic activity”. S. Napier, Virtue
Epistemology: Motivation and Knowledge (London: Continuum, 2008), 144.

32
R. Lockie, “Problems for Virtue Theories in Epistemology”, Philosophical Studies 138:2 (2008), 169 -
191, 174.
on x did not result in their believing truly that x, there is apparently no “success” in the sense of a
positive epistemic standing to credit them with. Reliability is taken as a truth-linked success
concept, and with respect to correspondence-truth there are no gradations of success; it is all or
nothing and the beliefs of scientists whose theories have been displaced seem simply to have
been collections of beliefs unreliably produced. This is not to say that objective success is
necessary for rationality on the externalist view; it is not, but neither is this approach very
interested in questions of rationality and responsibility. While both may be seen as contributing
to agent reliability, both are generally treated as superfluous for the kind of epistemic appraisal in
which they are primarily interested, and the only kind they take to be directly connected with the
central issue in the field of epistemology, the analysis of knowledge.33 Insofar as reliabilists have
been dragged by responsibilists to include “understanding” and not just knowledge as a valuable
epistemic aim, reliabilism in only able to treat understanding as knowledge or true belief about
causes. Responsibilists argue that this “factive” conception of understanding is unduly restrictive.
“It neither reflects our practices in ascribing understanding nor does justice to contemporary
science”.34 Given especially what Robert Brandom calls the ‘naturalistic temptation’ to think that
(naturalized) epistemology can do without “good reasons” explanations by supplanting concerns
with rationality with wholly causal ones, the concepts of rationality and responsibility play little
33
Lorraine Code writes that “[A] ‘reliable’ knower could simply be an accurate, and relatively passive,
recorder of experience. ...An evaluation of human knowledge-seeking in terms of responsibility is
instructive precisely because of the active, creative nature of that endeavour (L. Code, Epistemic
Responsibility, Brown University Press, 1987, p. 51).

34
Catherine Elgin, “Understanding and the Facts” (Philosophical Studies 132(1) (2007): 33-42).
role in the reliabilist’s approach. Reliabilist theories have trouble explaining a) the place of
theoretical understanding in scientific reasoning and in human cognitive economy more
generally, b) how theoretical models and scientific instruments “extend” cognition, c) how to
understand agent reliability when dealing with cases of misleading evidence, and even d) why
we should value acquiring interesting true beliefs over simply more true beliefs.
Epistemological internalism would initially appear to be in a better position to support the
common-sense ascriptions of creditworthiness (and intellectual virtue) to Isaac Newton or other
noted scientists of yore. Internalists focus on propositional or evidential justification, and could
point out how our past masters were evidentially justified by the tests they ran and the evidential
good reasons they had available to them. Evidential justification can be had in many cases even
where the belief is not true, and Newton might be intellectually praiseworthy in this sense. But
on closer inspection, the leading internalist theory, evidentialism, isn’t able to account for a full
range of discovery heuristics and intellectual virtues valued in scientific practice. Take for
instance a scientist’s tenacity in holding onto budding research programs. Such newbie programs
Imre Lakatos famously described as ‘born refuted,’ since they as yet enjoy no advantage in
empirical adequacy over extant competitors, and typically start out immersed in a ‘sea of
anomalies’ that only subsides over time as they build a track record of experimental and
theoretical success. Inquiry is by definition diachronic or extended over time; but the aspect of
rationality that leading evidentialists are exclusively interested in is wholly synchronic. One is a
reasonable or rational epistemic agent at time T1 if and only if one is proportioning the strength
of belief that x to the total evidence supporting x at that time-slice.35 Hence no evidentialist could
35
The distinction between synchronically and diachronically-focused epistemology is closely entwined
endorse the reasonableness of accepting a theory that doesn’t have a decisive advantage in
empirical adequacy over its competitors (and budding research programs seldom do). One who
accepted it or pursued it on bon sens, but without a preponderance of evidence, would appear
synchronically irrational, their degree of belief out-stepping their current evidence.36
Evidentialist epistemologies would appear to surrender what McMullin calls the “diachronic
dividend”, and to render paradigm instances of scientific discovery and of “tenacity”
epistemically blameworthy behavior.
6. Historicism, Naturalism, and the Functions of Thick Concepts
Herman Paul writes that virtue theory “encourages thick description and careful
contextualization, so as to take into account the peculiarities of practices and epistemic cultures
in which historians find themselves working”.37 Let us develop these points through the work of
Bernard Williams and Thomas Haskell. For Williams, thick characterological and affective
with the distinction between an epistemology being belief-focused and being agent-focused, a debate the
virtue epistemologists have especially been concerned with.

36
I have elsewhere critiqued the reductive claim by leading evidentialist philosophers that only the
synchronic mental event of “fit” between one’s beliefs and one’s available evidence is a source of
epistemic value. An agent’s conduct in inquiry—how they acquire what they or their community consider
evidence—is often an issue of legitimate epistemological concern, and not always ‘nothing but’ a
pragmatic or moral concern. See G. Axtell, “Recovering Responsibility” (Logos and Episteme, II, 3
(2011): 429-454); also “From Internalist Evidentialism to Virtue Responsibilism”, in T. Dougherty (ed.),
Evidentialism and its Discontents, (Oxford University Press, 2011).

37
Paul, “Performing History”, 15.
concepts are Janus-faced, having both descriptive and evaluative functions. “Normativity and
naturalism cannot be very sharply divided because the primary element in all judgments of
responsibility, moral or intellectual, is causation…The other issues can arise only in relation to
the fact that some agent is the cause of what has come about. Without this, there is not concept of
responsibility at all”.38 According to Williams, “these four elements, cause, intention, state, and
response…are the basic elements of any conception of responsibility”.39 Responsibility for the
virtue theorists then is primarily a diachronic, a historical concept, and not a synchronic one. 40
Now explanations and what Paul calls thick descriptions, especially those attributing
intentions and motivations to agents, are selective and value-charged. The explanatory salience
of any explanatory scheme depends upon its fit with facts, but also upon the kinds of interests in
explanation we have. One might say then that adequacy of an explanatory scheme must be
judged in terms both of naturalistic and normative concerns. Starting with our naturalistic
concerns, it is readily apparent that historians, social scientists, and philosophers often invoke
character traits in the course of explaining or interpreting the reasons for actions and decisions. A
38
B. Williams, Shame and Necessity (Berkeley: University of California Press, 1993), 56.
39
Williams, Shame and Necessity, 55.
40
“Certain phenomena or notions are historical, and, in contrast, certain phenomena are current time-slice
notions. A current time-slice notion is something that does not depend in a crucial way on its history
being a certain way rather than another. The current time-slice notions supervene on snapshot properties –
properties that one could in principle take a snapshot of at a given time….So there are current, time slice
notions and historical notions. My claim is the moral responsibility is an historical notion”. John Martin
Fischer, “Responsibility, History, and Manipulation”, The Journal of Ethics, Vol. 4(4), 385-391, 386.
concern with rational agency is typically expressed through a common-sense or folk-
psychological framework of intentions, beliefs, desires, character-traits, etc. This framework of
thick description is fraught with difficulties, but it allows encourages empathetic consideration of
what people might have thought and felt under past circumstances. But let’s look closer at how
historians utilize character-trait ascriptions when writing about historical agents, and what
concerns might arise with it.
Thomas Haskell argues in Objectivity Is Not Neutrality that “Historians are not
shy about assigning responsibility or imputing the causal status upon which responsibility
rests” nor should they be.41 There is a substantial role of causal reasoning in history as in
everyday life; historians can appropriately pass judgments of praise, blame,
responsibility, liability, deliberateness, etc. Haskell like Williams asserts that all such
judgments “ride piggyback on perceptions of cause and effect”.42 But here he also points
out lies a daunting set of problems for working historians because—or so Haskell argues
—practicing historians are often deeply confused about the role of causal reasoning in the
writing of history. Most notably he finds that their penchant for attributing responsibility
to people for particular actions and events often run in tension with their own skepticism
about the language of causes. This skepticism in turn is in no small degree the result of
“polarizing dynamics of the contest between narrativists and Hempelians”. Heated
debates over historiographic methods have led many working historians to neglect the
interest-laden character of intentional explanations, and unwisely to abdicate all talk of
41
Haskell, Objectivity Is Not Neutrality, 12.
42
Haskell, Objectivity Is Not Neutrality, 11.
causes to the Hempelian “nomological-deductive” model of causation.
Haskell is highly critical of the assumptions many historians and philosophers of history
make, “that narration and causation are polar opposites” and that “in spite of all appearances to
the contrary, causal reasoning plays only a peripheral role in history”. These assumptions feed
further polarization over methodology, and encourage the ill-considered stance that “what we
want from history is not explanation, but something entirely different, ‘understanding.’”43 To
counter-act what he characterizes as a maladaptive re-positioning of history and historiographical
methods, Haskell argues that not all causal inference is characterized by the Hempelian notion of
subsumption of the particular event under a nomological or law-like regularity. The ‘unity of
science’ ideal that prompted Hempel’s insistence that history could be no exception to the search
for general laws disintegrates, but not because of historiography’s lack of need of the concept of
causation. It disintegrates rather because of the plurality of modes of causal inference that post-
positivists have identified. This plurality is a further sign of how interest-dependent the notion of
explanatory salience is. I think we should agree with Haskell firstly that historians, in tracking
change, are employing explanatory schemes; and secondly, that the “narration” needed to link
historical happenings into ordered and meaningful sequences is not really an exception to, but
only “an especially supple form” of causal reasoning.
These arguments do not deter Haskell from recognizing that historiographic projects are
quite diverse, and have an equally substantial rhetorical dimension. Histories, biographies, etc.
are written with particular audiences and effective persuasive strategies in mind. Some are “gray,
meticulous, and patiently documentary”, but others reflect quite different literary intentions. The
43
Haskell, Objectivity Is Not Neutrality, 13-14 & 17.
rhetorical triangle of Pathos-Ethos-Logos and the rhetorical posture of the person doing the
explaining are ever-present concerns. So if we think about problems with character trait-
ascriptions that historians should avoid, one is the charge of “fundamental attribution error”. Not
only virtue ethics but also virtue epistemology has been targeted as empirically inadequate by
situationist social psychologists who argue that morally or epistemologically irrelevant
situational factors and the agent’s temporary mood are by-and-large better explainers than are
deep-seated traits of moral or intellectual character. Should historians on this account avoid
attributing any robust or global traits to particular persons? I think not. But natural scientific
study should qualify the use of trait-ascriptions in certain ways. Awareness of the heuristic of
leveling and sharpening seems important. People leave out what seems unimportant or
uninteresting and sharpen or highlight what seems interesting or explanatorily salient to them,
and this is a known source of bias and error; any extended testimony-chain game play in a
classroom easily reveals this. Do historians like other people ‘leveling out’ what might have been
important situational factors for an agent? Do they ‘sharpen’ what is of interest to them by over-
attributing robust character traits to people who are key agents of historical change, and by
exaggerating personal responsibility? And do our biographies and histories sometimes reflect the
further common bias of attributing personal traits in an asymmetrical fashion, such that the
ethical failures of those we’re well-disposed towards are treated as due to situational
happenstance, while the failures of those we’re ill-disposed towards are more often treated as due
to their own personal intellectual or moral flaws or vices?
While it is useful to reflect on lessons that the social scientific studies of biases and
heuristics may have for the ways we employ character-trait attributions, the wholesale skepticism
about stable personality and character traits that some situationists espouse should be rejected44;
situationism is based upon a radical re-interpretation of long-standing social-psychological
studies. Neither character-traits nor responsibility for actions are things than have or can be
empirically discredited, and historians needn’t shy away from them on the basis of situationist
theory. Moreover, explanatory schemes that attribute responsibility (in one of Williams’ senses)
to an individual for an action they perform arguably remains crucial to our ability to confer
meaning and intelligibility upon events. An historical narrative that explains important social
change a being primarily due just to happenstance, including the mood the person was in, etc.,
would not appear to offer a means of conferring much intelligibility upon past events.
Let us now consider the normative face of virtue theory by looking at the evaluative
aspects of thick characterological and affective concepts. These concepts have a certain
“valence” that marks them as value-charged concepts; when we attribute a virtue or vice to a
person, we are offering a judgment, an evaluation. Thus we might worry that the trait is being
applied moralistically or anachronistically. This is especially worrisome when the attributor also
takes that trait to be a salient explainer of actions or events within an historical narrative. But this
is relatively easy to avoid if one is careful, and is certainly no strike against the value of thick
concepts generally, since thin ones (right/wrong; good/bad) are subject to the same criticisms.
Virtue theory supports the Janus-faced, descriptive and normative project Williams embarks
44
For a fuller discussion see also my “Virtue-focused Epistemologies, Defended: A Reply to Doris and
Alfano” a typescript at JanusBlog: The virtue theory discussion forum, http://janusblog.squarespace.com.

upon by fostering a “thickening” of the language of normative concepts mainstream reliabilist
and internalist epistemologies focus upon (justification, warrant, knowledge, etc.).45 Character
epistemologists have called for a ‘thickening’ of epistemology, serving to bring mainstream
approaches into more direct contact with ‘fringe’ movements of social and feminist
epistemology. Character epistemologist Catherine Elgin comments,
If Williams is right about the relation of a thick concept to its thin descriptive precursor,
then thick concepts are not a mere convenience. Without such normatively loaded
concepts, we could not partition the world as we do. If epistemic concepts are thick, we
should expect them to display the same pattern. They should be anchored in, but not
supervene on, a descriptive core. They should mark out extensions that we have no
evaluatively neutral way to demarcate.46
45
On recent calls for the “thickening” of epistemology, see the papers in the 2008 Philosophical Papers
edition, Epistemology through Thick and Thin, 37(3). These include Catherine Elgin, “Trustworthiness”,
371-387; Guy Axtell and Adam Carter, “Just the Right Thickness”, 413-434; and Alan Thomas, “The
Genealogy of Epistemic Virtue Concepts”, 345-369. Thomas argues explicitly that “genealogy is an
important contribution to the project of virtue epistemology” (2008, 362).

46
C. Elgin, “Trustworthiness”, 374-375.
We have briefly looked at naturalistic and normative concerns with character-trait
ascriptions, whether by a historian or by someone else. But what special naturalistic and
normative concerns with thick concepts might philosophers themselves have? Consider
Williams’ own efforts to provide a “genealogy” of central epistemic concepts. Williams was
likely the first to develop genealogy along virtue-theoretic lines, so this can also inform us
further about synergies between historicism and virtue theory. In Truth and Truthfulness, his
last major book, Williams contrasts genealogy with the project of conceptual analysis aimed
at identifying a complete set of necessary and sufficient conditions for the application of a
concept. Despite the normative valences of thick concepts and the fictive character of a
‘state of nature’ narrative, Williams intimates that his social-functional or genealogical
accounts of epistemic concepts is "intended to serve the aims of naturalism”.47 Thick virtue
concepts, having both descriptive and evaluative meanings, are crucial to Williams in that
they allow him to fulfill without conflation the Janus-faced (normative and naturalistic)
aspects of his genealogical account.48

47
Bernard Williams, Truth and Truthfulness: An Essay on Genealogy (Princeton University Press, 2002),
22. My own version of pragmatic naturalism shares with Williams the thought that any apparent tension
with naturalism is ameliorated when we move away with preoccupation with reductionism as a hallmark
of naturalism. "Questions about naturalism, like questions about individualism in the social sciences, are
questions not about reduction but about explanation” (23).

48
“A genealogy is a narrative that tries to explain a cultural phenomenon by describing the way in which
it came about, or could have come about, or might be imagined to have come out. Some of the narrative
will consist of real history, which to some extent must aim to be, as Foucault put it, ‘gray, meticulous, and
patiently documentary’… However, genealogy is not simply matter of what I have called real history.
The collaboration between historicism and virtue theory is thus well-displayed both in Williams’
and in other philosophers’ attempts to provide a genealogy of epistemic virtue concepts. History
and virtue theory share a commitment to the value of the diachronic for evaluating both agents
and their beliefs. For as Colin Koopman writes,
A developmental perspective is the one thing afforded by history [and
genealogy] which other forms of inquiry do not, at least not by themselves, have a
handle on. This is why it can sometimes be helpful to contrast historical inquiry with
strict or pure conceptions of empirical and theoretical inquiry…. History takes its
subject matter as diachronic. It is concerned with development. As such, history offers a
perspective which empirical and philosophical approaches cannot mount without
explicitly involving themselves in history. This unique perspective is one which faces in
both normative and descriptive directions simultaneously.49
There is also role for a fictional narrative, and imagined developmental story, which helps explain the
concept or value or institutions by showing ways in which it could have come about in the simplified
environment containing certain kinds of human interests or capacities, which, relative to the story, are
taken as given”. B. Williams, Truth and Truthfulness, 20-21.

49
C. Koopman, “Bernard Williams on Philosophy’s Need for History”, Review of Metaphysics 64:1, 3-30
(2010a), 17-18. As an inquiry-focused philosophy, the American pragmatist tradition is also more inclined
to moderate historicism and to a Janus-faced conception of philosophy’s tasks. On the historicism of the
classical pragmatists James and Dewey, see Koopman, “Historicism in Pragmatism: Lessons in
Historiography and Philosophy”, Metaphilosophy 41:5, 690-713 (2010b): “In claiming pragmatism as a
kind of historicism, my idea is that pragmatism locates each of its central concepts (practice, inquiry,
To conclude this fuller development of our constructive thesis, our dependence on thick
concepts to achieve the tasks we set ourselves — whether as historians or as philosophers of
history — indicates again the strong mutual support between historicism and virtue theory
that we have argued for. This dependence is also an acknowledgement of the complex
entanglement of fact and value involved in historiography and in many other fields of
research, and the synergies alleged here grow in importance just to the extent that we
acknowledge this collapse of the fact/value dichotomy.50
7. Conclusion
If my approach has been on target, then character epistemologies can play a useful role in
the meta-methodologies of various scientific fields. But they can play an especially robust
role wherever researchers either see their field as tropological, in the sense of White’s claim
that “historical studies remain both rhetorical and literary”; or more generally where they
find comparing competing theories to be routinely underdetermined by agreed facts.

experience, etc.) as transitional processes…Experience is not a presence with its own substantial identity
—it is rather wholly constituted by its relations to past and future” (692).
50
A fuller development of this last point would lead us to further connections between the present view
and with what I above called pragmatic naturalism, which I suggest is a third component to the needed
metascientific account along with moderate historicism and inquiry-focused character epistemology. See
Hilary Putnam, The Collapse of the Fact/Value Dichotomy. (Boston: Harvard University Press, 2002). See
also my own paper “Thickies: The Pragmatist Friends of Entanglement” (typescript).

Underdetermination of theory by facts is no guarantee of under-determination of theory
choice by methodological norms such as an accepted list of theory virtues (though radical
historicists tend to conflate these different senses of the underdetermination problem). But
even the milder form construal of the underdetermination problem highlights the centrality
of non-deductive or ampliative processes for theory choice, and the thick concepts that
function as desiderata in these cases. Where theory-choice is underdetermined even by
accepted theory virtues then the evaluation of competing works is not necessarily stymied,
but turns to the determinacy-restoring bon sens of researchers themselves.
To summarize some of our steps to these conclusions, I have developed our
diagnostic thesis which is an elaboration of McMullin’s point that “Discussion of theory
virtues exposes a fault-line in philosophy of science that [separates] very different visions of
what the natural sciences are all about”. We have used this idea to pursue differences
between conceptions of history and historiographic methods, and we have also used it to
pursue a more formal way of understanding the relationship between different fields of
scientific research. Thus have we found that the question of how to parse the differences
between empirical testing, theory virtues, and the personal bon sens of researchers
themselves admits of no general answer, but depends crucially upon local issues about the
relative normality of inquiry being pursued under conditions of underdetermination.
Endorsing the diagnostic thesis did not force us to take a definitive view about the status of
historiography vis-à-vis science. But it helped us to distinguish some extant versions of
character epistemology and thus also to recognize the somewhat different ways of restoring
determinacy that they each support. It also helped us to further develop our constructive
thesis, the thesis of the advantages of integrating historicism and character epistemology.
We have now also seen that an account that mirrors Bernard Williams’ approach to truth and the
virtues of truthfulness will treat objectivity not only as an idealized state but also as an agential
virtue. It entails, contra its standard presentation, a condition of character not recognized in
conceptions of objectivity that prescribe an essentially negative relationship to subjectivity,
activity, and perspective. But neither should we accept, as the position that a dialectics of
objectivity leaves us with, any account that plies upon false dichotomizes between method and
apprehension, internal and external history of science, or the rational and the social. Indeed, part
of the dialectic which philosophy must account for and attempt to effectively mediate is just
these crude dichotomies.
Methodism (logicism), which draws such false dichotomies in order to prioritize the
former items—method, logic, rational reconstructions, etc.—and to preclude the latter from
scientific practice, may rightly be accused of objectivism. Some reliabilist virtue epistemologies
may be subject to this objection as well, although we have endeavored to show that combining
inquiry-focused character epistemology with moderate historicism, and encouraging genealogies
of central epistemic values, allows us to sidestep these problems. On the other hand, radical
historicisms that exaggerate problems associated with underdetermination or holism may also
succumb to the same false dichotomies, just in order to prioritize the latter items —
apprehensions, sociological explanations, etc. These radical historicisms have often been accused
of flirting with relativism, although we have endeavored to show that character epistemologies
allow us a clear path to historicize epistemic virtues and values in ways that avoid relativism and
that indeed helps us explain the intellectual credit still due to past master’s whose theories have
become obsolete due to the growth of knowledge.
Philosophers of science who have tried to undo the deleterious effects of the
aforementioned dichotomies have rightly urged us to disassemble the myth of universally valid
methodological and epistemological standards, the rational-social divide, and the distinction
between internal and external history of science that rational reconstructionists appeal to.51 They
have rightly suggested that pure versions of logicism and historicism are unequal to the task of
elucidating the reasoning involved in theory choice, but that methodism and historicism may still
converge, insofar as we are able to construct a philosophy of science that draws both upon
history of science and upon a more general logical and epistemological framework. These ideas
about metascience beyond the polarity of objectivism and relativism have certainly been raised
in the literature, though rarely with much specificity. Our thesis of the advantages of integrating
moderate historicism and character epistemology are meant to give these suggestions more
philosophic substance, and a clear direction for further development.52
51
See for instance Helen Longino, The Fate of Knowledge (Princeton University Press, 1990) and
Richard Burian, “More than a Marriage of Convenience: On the Inextricability of History and Philosophy
of Science”, Philosophy of Science, 44, 1 (1977): 1-42. R. Keat, “Relativism, Value Freedom, and the
Sociology of Science”, in M. Krausz (ed.) Relativism: Interpretation and Confrontation (West Bend:
University of Notre Dame Press, 1989). For a further development of a normative naturalistic alternative
to the rational reconstructionist projects see my “Bridging a Fault Line: On Underdetermination and the
Ampliative Adequacy of Competing Theories”, Synthese (2012, forthcoming).

52
Special thanks to the Society for Philosophy of History, to Mark Bevir, Herman Paul, and the Journal’s
editors and staff; also to my co-contributors for insightful comments on an earlier draft.
Guy Axtell, Philosophy & Religious Studies, P.O. Box 6943, Radford University, Radford, VA
24060 USA gsaxtell@radford.edu

Forthcoming as Chapter 38 in The Routledge Handbook of the Philosophy and Psychology of
Luck, Ian Church and Robert Hartman (eds.), 2019. An expanded version appears as chapter 2 of
G. Axtell, Problems of Religious Luck: Assessing the Limits of Reasonable Religious
Disagreement (Dec. 2018/Jan. 2019). DRAFT: Cite either published version or contact author:
gsaxtell@radford.edu. https://rowman.com/ISBN/9781498550178/Problems-of-Religious-Luck-
Assessing-the-Limits-of-Reasonable-Religious-Disagreement#
The New Problem of Religious Luck
Qualifying and Refocusing Etiological Challenges Based on Contingency
As we earlier saw with our example of J.S. Mill, contingency arguments allege a kind of
accidental quality to most people’s culturally nurtured beliefs. Our obvious psychographic
differences – differences in such things as attitudes, values, and most importantly for this study,
what John Rawls terms comprehensive conceptions of the good – are strongly conditioned by our
historic, geographic, and demographic location. Religious identity and whatever faith-based
beliefs one has are a prime example of this familial or cultural inheritance, and the apparent
contingency of such nurtured beliefs. Although certainly not the only example, the proximate
causes of one’s religious identity and the formation of attendant beliefs are, for most people, a
matter of their epistemic location, which in turn appears to be an accident of birth.
Michel de Montaigne gave a version of a ‘contingency’ or epistemic location argument
when he wrote,
[W]e receive our religion in our own way and by our own hands, and no
differently from the way other religions are received. We happen to find ourselves in
the country where it has been practiced; or we value its antiquity or the people who
have supported it; or we fear the threats it attaches to wrongdoers, or we follow its
promises… By the same means another country, other witnesses, similar promises
and threats, could in the same way imprint in us a contrary belief.1
Arguments like Montaigne’s and Mill’s have strengths and weaknesses that are rarely
carefully noted. They can be wildly over-ambitious if (mis)interpreted to have the implication of
reducing to epistemic irrelevance the evidence for all beliefs that are even tinged by cultural
contingency (Muscat 2015). Nathan Ballantyne (2015) is correct to find that in arguments from
contingency both ancient and modern, “pointing is typically all we get—worked-out arguments
based on variability are uncommon.” 2 Although religious beliefs are usually the ones singled
out, epistemic location has a conditioning or shaping effect over all of what in the epistemology
of disagreement are termed our “nurtured beliefs.” George Sher (2001) re-ignited philosophical
interest in nurtured belief; Sher asks us to study “the implications of the fact that even our most
deeply held moral beliefs have been profoundly affected by our upbringing and experience—that
if any of us had had a sufficiently different upbringing and set of experiences, he almost certainly
would now have a very different set of moral beliefs and very different habits of moral
judgment.”
As Sher indicates, etiological challenges and the contingency anxiety they arouse in those
who take them seriously may attach to a far broader group than just religious beliefs. 3 They
affect the epistemic assessment of culturally nurtured beliefs more generally. Several
philosophers have given thoughtful treatment to contingency arguments in recent papers. 4
Among them there is much agreement that to improve the epistemology of controversial views,
philosophers need to focus more carefully on the right target, proper scope, and epistemological
force of arguments of the sort that Montaigne and Mill share.5 We will follow J. Adam Carter
(2017) and others who use the term controversial views to refer to the broader class of nurtured
belief/opinions which include our moral, political, and philosophical views.6 But each of these
domains have unique characteristics such that they cannot be treated indifferently. 7 Retaining our
central focus on religious beliefs, one main etiological challenge starts from what are essentially
thought experiments about how different our beliefs in the religious-domain we easily be had we
been raised in a different geographic or demographic location. The challenge is that if raised
elsewhere demographically or by family, you'd likely accept different such beliefs than you
actually do. Is this true? Empirical data is clearly relevant. I think the specific question that needs
answering here is, “What percentage of adult religious adherents are adherents of the religion in
which they were raised as children?”
It would be good to see experimental philosophers fill in the gaps if sociological surveys
do not adequately engage the empirical questions like this one, that epistemic location arguments
lean upon.8 But sociologists of religion are quite interested in the study of “religious mobility,”
and while I can’t claim to have exhausted the literature, from studies in the America’s it appears
that there is solid survey data on this question: The relevant percentage stands right around 90 or
91%.9 This figure if correct surely does provide formally strong empirical support for the
generalization behind John Hick’s argument, similar to Mill’s and Montaigne’s, that in “a great
majority of cases” religious people (presuming adults) practice that particular religion in which
they were raised, or had “instilled” in them from an early age.
More than the obvious fact of inheriting aspects of collective identity from family, or from
the culture or sub-culture in which they are raised, this strongly suggests that whatever
affiliation-changes people may make later in life, they tend strongly to initially reach out to a
religious option most readily available to them in their epistemic location. These known facts are
reflected in the thesis of descriptive fideism developed in Chapter 3. They are sufficient for
motivating the seriousness of contingency or epistemic location arguments, even if, as we have
seen, the normative force of etiological challenges is debated in the literature.
Etiological challenges to controversial views often bear marks of contingency and of
what John K. Davis (2009) terms trait-dependence. Davis writes that trait-dependent belief is
widely acknowledged in domains where informal logic, interpretation, weighing factors or
norms, and the exercise of judgment are normal aspects of epistemic assessment. In the
acquisition of trait-dependent beliefs, “the subject reasons competently from justifying
considerations to a belief, but would not do so if he or she lacked some trait that appears to be
epistemically irrelevant, even if the subject and the situation were the same in all other respects.”
The agent “would not take those considerations to justify that belief if she had a different
socioeconomic background, religious affiliation, temperament, political ideology, or the like,
even if she were otherwise in the same epistemic circumstances” 10
The epistemic location problem is clearly involved in the psychographic differences
Davis discusses as trait-dependent. I will not in this book be able to develop a full epistemology
of disagreement for controversial views. In the recent literature on disagreement, the focus
questions seems to be whether disagreement should undermine one’s confidence in some target
belief or proposition. While I will remain neutral on some of these issues, I do not think there is a
one-size-fits-all answer either to this question, or to an over-broad question about how people
should rationally respond to recognized peer disagreement. Our approach holds that the
normative upshot of the prevalence of trait-dependent beliefs in the domains of controversial
views is not that they are never well-founded. Although many nurtured beliefs may be biased, we
cannot assume that all are without begging the interesting philosophical questions.11 It is better to
say (as seems to be the intent of both Montaigne and Mill) that what normative upshot
philosophers should draw crucially depends upon the varying degree of blind spot bias that
actual agent’s exhibit. It may exhibit it in various ways, through over-estimation of the epistemic
status of their views, through rhetorical and unsupported asymmetries of explanation, and more
defensively through rhetorical peer denial. It is the dogmatic ways in which nurtured beliefs are
sometimes held that motivates a serious de jure challenge.
So our inductive risk account starts out in agreement with Rawlsian reasonable
pluralism, and with Davis’ permissivist view that recognized trait-dependent over-determination
of a belief in domain of controversial views does not undermine the basing relationship in any
sweeping sense: it undermines that relationship only if and when that trait-dependence manifests
or mirrors known personal or social biases. We will return to develop specific, scalar markers of
this in Part II. But when it does not undermine the basing relationship the agent reasons
competently, and the influence of personal traits need only be regarded as one of the many
sources of the faultless cognitive diversity that John Rawls explained as grounds for reasonable
pluralism. Davis makes the connection between trait-dependence and Rawlsian reasonable
pluralism explicit by quoting the deservedly famous “burdens of judgment” section of Political
Liberalism: “To some extent (how great we cannot tell) the way we assess evidence and weigh
moral and political values is shaped by our total experience, our whole course of life up to now;
and our total experiences must always differ” (25).
The epistemic location problem is one of the most unavoidable of these sources of
contrariety, being basic to the human condition, and closely related to the evidential ambiguity
that affects so many views we hold on morals, politics, philosophy, and religion. The epistemic
location problem might suggest grounds for challenging especially our culturally nurtured
beliefs, and these range far beyond the domain of religious ideas. But these epistemic worries
that contingency arguments raise do not necessarily fall evenly across cases, let alone domains.
This is partly what philosophy of luck and our inductive risk approach to the limits of reasonable
disagreement can help us to see. They need not fall evenly across the four mentioned
controversial domains. Indeed part of what motivates the etiological challenge I am about to lay
out is that the sharply asymmetric value ascriptions associated with religious exclusivism are
much more prominent in some models of religious faith and responses to religious multiplicity
than in others. It is the philosophy of luck that brings the implications of asymmetric trait-
ascriptions into focus, and that will help us identify the more and less morally/epistemically
plausible models. I will argue that focusing on asymmetric religious value ascriptions and the
specific conditions that make them morally, epistemologically, or theologically doubtful helps to
illuminate the limits of reasonable religious disagreement.
Since we need to think carefully about epistemic location if we are to explicate the force
of the epistemology-related religious luck problems, the New Problem tries to present a qualified
and re-focused set of concerns. It carefully delimits the scope of its de jure challenge. In general
terms, a de jure challenge implicates dereliction of epistemic duty, intellectual viciousness, or
some other sense of epistemic unacceptability to some target class of religious belief, narrow or
broad (Plantinga 1995). By engaging facts of religious multiplicity and understanding an
epistemological anti-luck condition in terms of the degrees of inductive risk entailed by a mode
of belief uptake, we can state the specific challenge of the New Problem as a challenge to the
reasonableness of religious exclusivist a) conceptions of faith and b) responses to religious
multiplicity.
Let us clarify some terms before going further. Inductive risk is the study of the chance or
possibility of getting it wrong in an inductive context.12 Debate in the second half of the
twentieth-century over the role of values in scientific research introduced the concept of
inductive risk. Philosophers of science like Carl Hempel reasoned that because no evidence can
establish an explanatory hypothesis with certainty, “acceptance … carries with it the ‘inductive
risk’ that the hypothesis turns out to be false.”13 Based on this admirably broad definition, the
conversation about risk-management with respect to scientific research and the applications of
technical knowledge expanded in later decades. Government or corporate science policy-makers
dialogue over risk-assessment together with stakeholders: all those potentially impacted
positively or negatively by the decisions made. The concept of inductive risk, if portable beyond
philosophy of science and epistemology, might serve equally well as common-ground for
discussions between philosophers, theologians, and researchers in cognitive science of religion.
There are a number of concepts in the risk-related family of concepts –risk, probability, chance,
luck, fortune, fate or destiny, etc. Risk is an especially useful diagnostic concept in epistemology
because it is connected with what is calculable, in contrast to concepts like destiny or fate.
Counter-inductive thinking entails the highest degrees of inductive risk. The definition of
counter-inductive thinking is independent of any special concern with a special domain of
inquiry. Counter-induction is defined in dictionaries as a strategy that whether self-consciously
or not reverses the normal logic of induction. One example would be ignoring the normativity of
a strong pattern of how things are within our past and present experience, while making a
prediction about the future. For instance I predict something will happen that is quite counter-
point to what the evidence and inductive reasoning would suggest. Another example, closer to
our concern, is a strategy of taking a mode of belief-uptake that we ourselves assert or are
committed to holding is unsafe/unreliable for the production of true beliefs in other people, as a
truth-conducive mode of belief-uptake in our own case. For our purposes counter-inductive
thinking primarily describes a commonplace logical failing: a failing to see reason to apply to
oneself our ingroup a judgment that one readily and even eagerly applies to others (and their
epistemic situations); or a failure to supply sufficiently independent and non-circular epistemic
grounds for one’s self-exemption from an inductive norm.
We are now in position to lay out the New Problem formally. It will consist of three
theses, and a fourth which follows from the first three. The first two theses are about formal
symmetry, symmetry in the belief-forming cognitive process. The first thesis is:
Familial-Cultural Displacement Symmetry
(DS) For the great majority of religious adherents, had s/he been raised in a
family or culture with a different predominant religious identification than
that in which s/he was actually raised, but with the same natural capacities
and intellectual temperament, s/he very likely would come to identify
herself with that religion, with roughly the same degree of personal
conviction.
Next we propose a belief-forming cognitive process-type salient in the formation of
religious beliefs of widely divergent content. The process-type should be wide, but our purpose
is again not to challenge as unsafe the etiology of all religious beliefs, but the etiology of beliefs
associated with a fideistic uptake of a testimonial tradition. Max Baker-Hytch (2014) points out
that what a modal reliability (or safety-based) version of the contingency argument should do to
serve its purpose is not to pick out a process-type whose specification involves particular
religious texts or particular testimonial chains. Rather, “selecting a wider process type, whose
specification includes no such particulars… achieve[s] the desired result: locating a process type
a significant proportion of whose outputs are mutually inconsistent, thus whose overall truth-
ratio cannot, even given the truth of one or other consistent subset of the process’s outputs, be
high enough to satisfy a process type reliability condition.”14 In line with this useful point, our
second symmetry thesis is:
Etiological Symmetry
(AS) The epistemologically-relevant level of generality at which to characterize
the belief-forming cognitive process by which persons in a great majority
of cases, and across epistemic locations, acquire their religion,
denomination, or sect-specific beliefs is a level of testimonial authority-
assumption.
Testimonial authority-assumption is our proposed “wider process type, whose
specification includes no such particulars.” We can all recognize that accepting the unique
authority of a purported revelation is a common way to acquire a religious identity, and often in
testimonial traditions such acceptance by the individual is tantamount to what they are taught
from an early age that faith consists in. So now we follow up these two (difficult to deny) claims
about symmetrical processes of testimony uptake with a thesis about exclusivist asymmetric
ascriptions of truth and falsity to the beliefs acquired through this mode of belief-uptake. (EA)
makes a qualified contingency argument specific to religious exclusivists:

Exclusivist Asymmetries
(EA) Religious believers of exclusivist orientation in the actual world whose
original mode of belief acquisition is aptly described by (DS) and (AS)
would likely, if raised in a different epistemic location with a different but
still exclusivist dominant testimonial tradition, a) adopt as uniquely true
and salvific beliefs that in the actual world they hold to be erroneous; and
b) ascribe falsity and error to beliefs that in the actual world they hold to
be true.
Finally, (CIT) draws out the key implication that prior concession to all three of (DS), (AS), and
(EA) logically commits one to:
Counter-inductive Thinking
(CIT) Religious believers in the actual world whose mode of belief acquisition
and maintenance is aptly described by (DS), (AS), and (EA) ascribe to a
mode of belief acquisition and maintenance that should by their own lights
be acknowledged as counter-inductive thinking.
The intention of this argument is to demonstrate how the mind-set of religious
exclusivism is enabled only through counter-inductive thinking. The exclusivist ascribes
falsehood to the theological systems of religious aliens, but truth to their own, through what any
non-committed party would judge to be a common mode of belief uptake: testimonial authority
assumption. (CIT) shows this to be an unmotivated self-exemption from the normal logic of
induction, where if a form of belief-uptake is judged unreliable for others to use it is very
probably unreliable for ourselves. Religious exclusivists are among those who are most likely to
deny that religious aliens are peers, to dismiss the epistemic significance of religious
disagreement, and to maintain themselves as fully rational in what Griffiths refers to as “non-
accommodationist” attitudes towards religious aliens. But (CIT) strongly implies that peerhood is
not easy to dismiss; it indeed suggests that dismissing it proceeds the same way —by a self-
serving self-exemption from inductive norms (i.e., by counter-inductive thinking). Denial of the
epistemic significance of religious disagreement is shown vain if the fact of persistent
disagreement directly affects the assessment of whether the luck concerns in people’s epistemic
situations are concerns about evidential or about environmental luck.
In order to deny acknowledging (CIT) exclusivists must deny one or more of (DS), (AS),
or (EA), but this is not easy to do. The primary challenge that follows from being logically
constrained to accept (CIT) is that the foundedness of the exclusivist’s belief will then be shown
impacted by malign environmental luck not just in the counter-factual but also in the actual
world. Your beliefs in both the counter-factual and the actual world are acquired on an unsafe
basis, under conditions of very high inductive risk, which in turn should be recognized as a
strong de jure challenge to their epistemic standing. How can one respect the morality and
rationality of a mode of belief-uptake that generates sharply asymmetric moral, epistemic, and
theological trait-ascriptions directed by some its users against other of its users? The de jure
challenge stemming from the New Problem of religious luck is that that the mode of belief
uptake characterizable as testimonial authority assumption is from an externalist viewpoint
unsafe.
Are they also from a responsibilist viewpoint intellectual vicious? That is a somewhat
different question, and would involve us more deeply in discussion not just of the safety but of
the sensitivity of belief. The epistemic location problem, and the way we have articulated its
force in the New Problem, plausibly shows that the contingency of so many of our nurtured, or
testimony-dependent beliefs, involves a question about the basing relationship. I believe this is
correct, but sensitivity is often criticized as a strong demand and I am not assuming that it is a
condition of knowing.15 Still, it seems to track some relevant aspects of reasonableness, and to be
pertinent to an epistemology for controversial views.16 For a culturally nurtured belief to be
insensitive, we do not have to imagine radical deception scenarios.17 We only have to make some
quite modally close changes, such as: growing up in a politically or religiously conservative
family instead of a liberal one, growing up in our same society but in a different religious
tradition; growing up in a different society that has a different majority religious tradition, etc.
Whatever we can say about the truth-aptness of beliefs in domains of controversial view, and
about trusting putative moral or religious experts, it is clear that beliefs in these domains are
exceptionally insensitive.18 But their insensitivity and their causally over-determined etiology are
almost indistinguishable.
The Exceptionalist Dilemma
“Genuine tragedies in the world are not conflicts between right and wrong. They
are conflicts between two rights.” –Georg Wilhelm Friedrich Hegel

The question just asked leads me to a complementary description of the New Problem in
terms of a dilemma for religious exclusivists: The Exceptionalist Dilemma. The exclusivist, it
holds, is caught between embracing two responses to the New Problem, neither of which proves
philosophically adequate. The first horn says that the exclusivist cannot without great cost just
concede that the formal mode of belief-uptake for adherents of the home religion is of the same
general type as that of religious others. For one who grabs the first horn and makes a “same
process response” will still want to maintain that the faith-based believing of adherents of the
home religion results in true belief. But given that they themselves hold that it this general
process-type results in untruth and error in a vast majority of cases, they certainly cannot say that
the process is safe, as epistemologists use that term. The cost of grabbing this first horn is being
saddled with recognition that one’s mode of belief uptake is unsafe. The claim seems to be that
the symmetry of proximate causes of belief just doesn’t matter so long as truth is claimed. What
was an inductive pattern of causal conditions quite saliently applied to religious aliens in judging
their falsehood does not apply at home. The same proximate causes at work in those other cases
produces true belief in their own case.
Note that lack of safety is entailed by an inference being counter-inductive, since (in
more internalist terms) a counter-inductively derived conclusion is locatable at the Weak pole on
the standard scale of inductive arguments, running from strong inductive arguments where the
premises make the conclusion probable or highly likely, to weak inductive arguments where the
premises fail to make the conclusion probably. So we should think of weak inductive inferences
are unsafe by definition. Inductive weakness and inductively risky belief immediately re-raises
problems of religious luck: If the religious insider’s beliefs, like that of the all the religious aliens
who got theology wrong, are grounded on counter-inductive thinking, then doesn’t it follow that
the religious insider’s beliefs are, if true, only luckily true from the epistemic point of view? It is
lucky that in some way –for whatever reasons— the inductive finger points only outwards and
does not point back at you. But the moral, theological, scientific, and philosophic disciplines
each have a right and obligation to press the counter-inductive conclusion further, asking why
not.
The safety or reliability of a belief-forming process is a condition on knowing and related
epistemic standings, and we should conclude that we lack personal doxastic responsibility if we
believe on the basis of a mode of belief-uptake that we allow is unsafe.19 Grabbing the first horn
then comes back on the initial asymmetrical positing of truth: The truth status the exclusivist
confers on their own specific theological system is now counter-inductive to the falsity they
claim for all other theological systems.
It is likely a more appealing option for exclusivists to make a “unique process response,”
thus denying the charge of counter-inductive thinking that comes with affirming that their beliefs
are generated by the same general process of adherents of other testimonial traditions. But the
second horn of our dilemma is that all attempts to justify the uniqueness of the mode of belief-
uptake ascribed to adherents of the home religion will either be empty or will implicitly rely on
self-favouring ascriptions of good religious luck.
To effectively grab the second horn the uniqueness asserted must genuinely be of process
type, not just of content. One cannot just appeal to content by saying the process type is a ‘one
off,’ for instance, by claiming the input to it is the one true revelation of God, or the process must
be reliable because it outputs the one true creed. By strength of analogical inference, the
seriousness of the charge of engaging in counter-inductive self-exemption is not avoided by such
form/content conflating responses, but actually re-enforced. To avoid vicious circularity, a

response that grabs the second horn of our dilemma needs to argue for uniqueness of process-
type in an adequately formal sense. The challenge I put forth is that no such unique process
response can be given to the Exceptionalist Dilemma that is not empty and cannot be shown to
lean once again on a self-serving religious luck ascription.
There are different responses that might be tried out, and numerous counter-analogies or
disanalogies that might be raised to try to dislodge our challenge. So rather than claiming that the
New Problem or the Exceptionalist Dilemma are strictly inescapable I present them primarily as
a useful focus for a new discussion between philosophers of luck, theologians, and proponents of
CSR. But if neither ‘same process’ nor ‘unique process’ responses to the dilemma prove
satisfactory, then the adherents of our different religious faith traditions should concede that the
de jure objection reveals the intellectual poverty of exclusivist conceptions of faith and responses
to religious multiplicity.
Explanatory Asymmetries and Critical Reasoning Dispositions
Attributional asymmetries shed a good deal of light on explanation, and what makes for a
well-founded causal explanation. In the social as in the natural science where asymmetry,
understood basically as ‘the arrow of causation,’ has played a central role in many debates about
explanation.20 So before returning to discuss the New Problem or even religion specifically, let’s
look at asymmetry in the sense of the arrow of causation. Daniele Molinini introduces the
importance of asymmetry for scientific reasoning this way: “Explanatory asymmetries appear
when we have pairs of deductively valid arguments which rely on the same law(s) but which
differ radically in explanatory potential.” The classical example of such a pair is that of the
flagpole and the shadow. Let Figures A and B, below, represent these arguments.
[Insert Table 1 here, centered]
Figure A Figure B
Figure B’s argument represents a reasonable casual explanation of why the shadow has that
particular length, but Figure A’s argument does not:
The problem with this second derivation, in the context of explanation, is that
it seems nonsense to say that the length of the shadow explains why the flagpole
has that particular height… In the tower and the shadow example we intuitively
recognize that there is a direction which gets the explanation right…. The height
of the flagpole causes the length of the shadow, while the converse is not true.21
Besides comparing how asymmetry is understood in empirical and in mathematical cases,
Molinini draws a more general lesson that we should pay attention to when genuine causal
explanations are hypothesized in any domain, or when competing explanations are weighed:
There are pragmatic constraints which operate in scientific explanation. These
pragmatic constraints are specific abilities to reason (for instance, the ability to
reason analogically, or the ability to reason causally) which are used in non-
scientific explanations as well…. Our ability to reason causally acts as a
pragmatic constraint in the case of the flagpole and the shadow, thus permitting us
to discriminate between the two putative explanations.22
[Insert Table 2 here, centered]

Figure A Figure B
That there are constraints on a well-founded explanation both inside and outside the
empirical sciences is a point worth highlighting. Abilities to reason inductively and conceptual
resources combine in the judgment that B but not A is a good explanation. All cases of ‘getting it
right’ in a putative causal explanation involve constraints and criteria. Molinini indeed gives
examples from multiple fields to show “how explanatory asymmetries shed light on explanation”
in scientific and non-scientific domains alike. The light this sheds is that the constraints on
acceptable explanations are critical thinking dispositions that help us to assess one explanation as
better than another. To do so we must acknowledge the force of inductive norms, compensate for
our own cognitive and motivational biases, and avoid relevant fallacies. Beyond this, the abilities
needed in the scientific context are especially abilities to employ cause-effect, generalization,
and analogical reasoning. But these same abilities are widely used in all domains, for in all
domains we need to be able to distinguish strong from weak causal claims, strong from weak
analogies, or generalizations. For us, these sometimes quite general and at other times very
specific abilities for assess inductive strength are simply the tools in a person has in their
inductive risk toolkit.
From a psychological perspective, Keith Stanovich, Richard West, and Maggie Toplak
take Myside bias to occur “when people evaluate evidence, generate evidence, and test
hypotheses in a manner biased toward their own prior opinions and attitudes.”23 Avoiding myside
bias or other known biases has little to do with intelligence, but far more with habituation to
higher-order critical reasoning skills and dispositions. The cultivation of general intellectual
virtues that help attenuate cognitive biases and inappropriate heuristic responses to problems
calling for reasoning leads Stanovich and his colleagues to study how well or poorly people are
able to engage their fluid rationality.24 When lacking thinking tends to be dominated by what
Stanovich terms “crystalized inhibitors,” or just plain “miserly” processing. These inhibitors as
Stanovich describes them include egocentric processing, over-reliance on one’s own
introspective powers, and assuming superiority of intuition. Effective problem-solving seems
inextricably connected with the agent’s fluid rationality, which in turn involves utilization of
skills of inductive reason. How an individual performs on an assigned task involves skills with
inductive reason that give power and portability to fluid rationality.
But why is this relevant to the asymmetric attribution of religious truth or value? Why
should it matter to religious practitioners, and to philosophical theologians? How is the centrality
of inductive norms for sound inferences involving cause and effect, generalization, and analogy/
disanalogy supposed to place suspicion on any putative explanation that a theology offers for
religious difference? How is it supposed to ‘reveal’ all, most, or many such theologically-cast
asymmetric trait ascriptions as the products of crystalized inhibitors, and inductive illogic?
Our argument is not brought forth on the basis that every favoring of prior opinions and
attitudes deviates from an ideal reasoned. As a pragmatist account it is one that can respect
background beliefs and the important senses in which we are children of time and can only start
from where we are, and that we ‘live forward’ and therefore face all kinds of pragmatic
constraints. But as a pragmatist account it is also one that preserves and supports the functions of
criticism, and the amply resources that inquires have available to them. Religious thinking at the
level of practice and theology are both shot through with cause-and-effect talk, with the drawing
of generalizations about events and populations, and the application of such generalizations to
particular actions, persons, or groups; and with analogies and disanalogies between events and
populations. These are the specific focus of our inductive risk account in what it criticizes and in
how it takes studies of moral and cognitive to apply to philosophical concerns with assessment
and guidance. So to turn the previous questions around, we should start from clear
acknowledgment that there are well-established inductive norms and skills that fuel success with
problems of inquiry involving inductive risk. But then how are we to understand the positing of
final in contrast to efficient or proximate cause explanations? Are there experts on extra-
scientific cause and effect?25 If so, on what skills and dispositions does this expertise depend to
distinguish good from bad explanations about religiously-relevant traits, if not skills articulated
in terms of inductive reason? If our best cognitive assessments in all other walks of life indicate
and utilize these skills, then what exactly would lead a religious believer to hold that processing
religious experiences and religious testimony should ask the believer to follow a different logic
entirely?
Our interests are primarily epistemological rather than metaphysical. The proximate
causes of religious belief appear symmetrical with beliefs in other domains of controversial
views, in that epistemic location and nurturing importantly help explain differences in belief. But
in a theologically-cast final cause explanation all those formally symmetrical proximate causes –
everyday cause and effect – are typically left in place, while the asymmetry in God’s action, as a
supernatural event, is conceived as quite independent of them. Were a religious disagreement
neutrally describable as the exchange of contrary putatively causal explanations without
observable causal effects, what should the real inference to the best explanation be? How could
this radical asymmetry among unfalsifiable final cause explanations between one religion and the
next not have implications for broader debates over realism or fictionalism about religious
language? In what ways could a neutral observer affirm, under such conditions, even confirm
that final cause explanations are genuine explanations in contrast to the misrepresentation of
products of religious imagination?
I mention these questions really only in order to set them aside. Such questions best
pertain to de facto challenges, and are therefore outside the purview of our de jure focus. The
New Problem targets the reasonableness of religious exclusivist responses to religious
multiplicity, not realism about religious language, so I will occasionally connect the inductive
risk account to motivation for so-called ‘de-bunking’ theses, but take no stance on them. Chapter
6 will involve us in cognitive science of religion, and there we will examine more closely
different view about the relationship between (weaker) de jure and (stronger) de facto
challenges, but again without taking any strong stance on the matter. What we can ask is, why
should the domain matter this much? Why should we think that resistance to miserly processing,
overconfidence, myside bias, and inductive illogic are in principle less pertinent understanding
religious ideas and theologically-cast explanations than they are to beliefs and explanation with
non-religious content?
At the start of this book we posed the question of what is going on when persons,
theologies, or purported revelations ascribe various kinds of religiously-relevant traits to insiders
and outsiders of a faith tradition in sharply asymmetric fashion. This is a diagnostic question,
and in these terms the New Problem offers what one college critical thinking textbook terms
diagnostic arguments:
These are arguments in which we reason by explaining certain data we encounter,
and when explanation becomes a tool in our reasoning, it provides a huge increase
both in the vocabulary we may exploit in talking about reasons and in the
apparatus we may employ in analysis.26
Moral, epistemic, and also religious luck are diagnostic concepts – concepts that play an
important role in psychological, philosophical, and theological answers to diagnostic questions.
The concept of religious luck grants us significant insights into the epistemological significance
of religious disagreement. Pervasive disagreement and epistemic risk make the epistemic context
an inductive context for agents: contexts in which a person cannot assess their own reliability
apart from considering how reliable others would likely be using a formally similar method, and
how reliable they themselves would likely be still using their same method but under a range of
modally-varied circumstances. So a philosophy of luck should necessarily be concerned with
inductive risk. To say that a religious ascription appears to lean on luck or that a particular way
of thinking aggravates problems of lucky belief or luckily true belief, is almost synonymous with
saying that the group or agent makes the ascription in a way that violates default norms of
symmetry and induction. Our argument in this chapter shows that the degree of inductive risk
implicit in an epistemic strategy is a formal and sometimes readily assessable issue. It also
suggests that the high degree of inductive risk in contexts in which testimony-based claims to
religious knowledge are typically made is a fully sufficient criterion for determining it to be a
context of malign environmental, rather than benign evidential luck.
Finally, degrees of incurred inductive risk may not only be a sound criterion for taking
the etiology of a belief to be affected by malign epistemic luck; they may also prove a useful
means of measuring where a particular theology falls on the scale between religious rationalism
and religious fideism.27 Rationalists claim that to be entitled to a religious belief, there should be
no more risk involved in it than in other more everyday beliefs. The denial of this claim holds
that at least for religions-specific beliefs, it is necessary to believe by faith. Further, the fideist
may go on to valorise supra or even counter-evidential ‘belief,’ on the grounds that a
commitment matched only to the degree of rational confidence in their evidence provides at best
a conditional and hence unstable faith. So our approach suggests that these same concerns with
inductive risk and safety-principle violation also supply a way to distinguish moderate and strong
(sometimes called radical or counter-evidential) fideism on the basis of recognizable formal
markers—something that might be quite useful in religious studies and CSR.
By circumscribing the de jure challenge of the New Problem rather narrowly, it should be
clear that I am not using this or any of the other problems of religious luck as a blunt instrument
to delegitimize religious faith en toto. But family resemblances among historical faith traditions,
as we have noted, are important to take notice of. Moreover, that the Abrahamic religions have
often given quarter to religious exclusivism is, in good part, a function of shared characteristics
that feature in religions of special revelation. Religion-as-revelation seems both to generate
variety, and at the same time to aggravate problems of religious luck in ways that will bear
further attention in the Part II of this book. But to reiterate, the New Problem as presented here
targets only the reasonableness of strongly counter-inductive models of faith, and of the
exclusivist response to religious diversity.
A key approach we have taken to the issue of religious rationality is to re-cast safety and
sensitivity concerns as indications of high inductive risk entailed by the cognitive strategies
people rely upon in acquiring or maintaining their belief.28 The concept of inductive risk is one
we will continue to develop throughout the book, but one debate that seems to bear upon it is
debate over whether rationally maintaining one’s belief in the face of genuine peer disagreement
requires one to offer “independent” reasons. Independence requires that, when you assess
whether to revise your belief in P upon discovering that someone disagrees with you, you
shouldn't rely on the reasoning that led you to your initial credence in P. Baldwin and Thune
(2008) for example discuss religious multiplicity as a defeater for religious exclusivism, and
attempt to rule out “defeater-defeaters” that are not independent of the initial basis of the belief
whose defeat is at issue.
David Christiansen writes that an independence requirement “essentially gives us a way
— to use Mill’s terms — of taking precautions against our own fallibility.” This seems especially
important when theological claims are asserted in the face of polarized contrariety. Again I do
not think the epistemology of disagreement literature has paid enough attention to the
significance of polaralized contrariety, and how it part of what explains it is a fideistic denial of
any independence requirement on the true believer. Christiansen adds that J.S. Mill’s position on
free speech and diversity in the marketplace of ideas “is straightforwardly epistemic: Mill would
have us treat the opinions and arguments of those who disagree with us as an epistemic
resource.” The need for such constraints “stems from the realization that we often err by
overlooking, or failing to attend carefully to, arguments opposing what we’re inclined to believe.
And engaging carefully and honestly with those who disagree is the natural precaution against
this sort of fallibility.” Christiansen further notes that his own version of an independence
requirement “flows from the realization that our fallibility runs a bit deeper than Mill suggests:
we also often err in coming to settled, all-things-considered judgments, even after the most
careful reflection and conscientious engagement with our opponents.”29
While an independence principle is seen, by those who acknowledge it, as applying to the
reasoned defence of all of our controversial views, the special problem in regard to religious
disagreement specifically is the tactic of responding to epistemological objections with
theological, hence metaphysical, answers. So as a stab at a principle of explanatory constraint
that takes account of our special concerns with asymmetric trait-attributions, consider EGO, or
“Epistemically-Grounded Only” Principle:
(EGO) Reasonable religious disagreement recognizes that trait-ascriptions applied
asymmetrically to religious insiders and outsiders — that is, ascriptions of
asymmetrical religious status or value explanations — require independent
epistemic justification, and not merely moral, or metaphysical, or
theological, or home-scriptural justification.
Independence is in the first instance an attempt to determine common principles of reasoned
dialogue. I am sensitive to critics who hold that strict independence may be an impossible ideal
for disagreement with respect to domains of controversial views. Accordingly, I will not treat
adherence to an independence principle as a strict requirement of reasonableness.30 The force that
it has is a matter of degree, not all-or-nothing, and for us this means that radical denials of
Independence are direct indicators of high inductive risk, and only through this, of agent
unreasonableness. (EGO) still supports Baldwin and Thune’s (2008) argument for the
inadequacy of the reformed theological uses of “internal” and “external rationality,” and of the
circularity of their ways of discounting defeaters.31 Against these philosophers’ recognition of
epistemological limits of experience-based exclusive religious belief are negative apologetic
works such as Jeroen de Ridder’s aptly titled article, “Religious Exclusivism Unlimited.” de
Ridder (2011) uses the notion of external rationality to claim that,

[A]n exclusivist can rationally hold religious beliefs in the basic way without
having anything by way of a reason or argument to defeat the defeater of religious
pluralism. The exclusivist would simply have to find herself with a firm
conviction that her beliefs really are right, in spite of counter-testimony from
seemingly trustworthy sources.32
Once an exclusivist has “defeated the defeater of pluralism” by a simple reiteration of the same
basic or non-inferential cognitive process s/he claims originally produced the beliefs, her
“structure now includes a belief that adherents of other religions are epistemically less fortunate
than she is; this belief being inferred from her reproduced properly basic first-order religious
beliefs” (455).
This ‘iterative’ response, would clearly fail an independence principle like (EGO). The
Reformed view is re-iterative, but Jerome Gellman’s much more subjectivist-internalist response,
or others that appeal to a subjective method would clearly fail it as well.33 The asymmetry
between religious aliens “epistemically less fortunate” than adherents of the home religion is
diagnosed, in our inductive risk account, as a simple iteration of the author’s prior appeal to
constitutive religious luck for the unique truth of his own religion-specific beliefs. It is
epistemically circular to attempt to justify one’ framework through reference to that very
framework.
How vicious or virtuous should we take this circularity to be? It is not as if we are not
permitted a worldview. (EGO)’s standard of reasonableness I think is not asking too much
in a world where few but the most isolated tribes do not have regular exposure to religious
diversity and religious unbelief. Tess, in the basic Tess Case of Chapter 1, wasn’t highly
blameworthy that she gained so little perspective on news sources. Even if we can’t say she
only manifested a virtuous agency and credit for her true belief if she had confirmed or
challenged it by looking beyond the Blue newsstand publications, her belief in its stories
are the actual history of Land of Lakes County deserves is not well-founded. Tess could
have and should have done better since she had other boxes available to her to supply some
perspective independent of what her uncle and the Blue newsstand publication told her.
She knew there was deep disagreement between their contents, as this was announced in
bold letters on the back of the paper: “Remember, trust only the news from this box! All of
the other boxes contain fake news.” But since the possibility and even the desirability of
meeting an independence principle are said to be debatable matters, we can leave (EGO)
up for debate. I conclude that in the first instance, the radical rejection of any independence
principle, and what will be more closely argued to be the associated logical circularity of
exclusivist responses to religious multiplicity, are among numerous indicators of the high-
risk doxastic strategies prescribed in strongly fideistic conceptions of religious faith.
An Objection and Reply
Consider this objection: The claim crucial to the New Problem, that as a doxastic strategy
home-religion testimonial uptake is unsafe, unfairly totalizes over religions. It is easy for the
exclusivist, whether on epistemologically internalist or externalist assumptions, to deny this.
More particularly they can deny (AS), the Etiological Symmetry thesis, which seems to say that
testimonial authority-assumption is all one needs to know about the causal etiology of belief to
determine both its ubiquity across religions, and its epistemic viciousness. But the analysis is too
course-grained. For these same reasons it is therefore easy to grab the second horn of the
Exceptionalist Dilemma: to maintain that the uniqueness of my religion’s testimonial tradition
qualify its transmission as a “different process” than that used by adherents of other faith
traditions. The safety of testimonial chains needs to be assessed singly, not in the totalizing way
posited by the dilemma. Looked at that way, it can be safe even as I, as an exclusivist, hold that
testimonial authority assumption indeed results in false belief for adherents of alien religions.
In response, let me first remind the objector of the requirements of grabbing the second
horn: “To effectively grab the second horn the uniqueness asserted must genuinely be of process
type, not just of content. One cannot reasonably just appeal to content by saying the process type
is a ‘one off,’ for instance, by claiming that the experiential or testimonial input to it is self-
authenticating, or that the process must be reliable because it outputs the content of the one true
creed.” As this passage hints, religious exclusivism has champions who defend it through an
internalist apologetic, and others who use an externalist apologetic. If the conditions I place on a
“different process” response are correct, the question is whether either apologetic strategy can
escape circularity well-enough to establish reasonable debate-independence for its supporting
reasons.
Starting with internalism, the appeal to unique phenomenology, as de Ridder (2014)
points out, is an appeal to internal evidences, evidences such as what it feels like to read the Bible
and feel its inspiration, or to participate in a ritual or other communal practice. These factors are
vitally important to one’s religious identity. Indeed, that every religion comes with unique
concepts and experiences is an important aspect of my permissivist ethics of belief. Religious
narratives place readers where they can feel what it is like to have certain sorts of experience, but
they appear not to situate them well for making truth claims, let alone exclusivist ones, on their
basis. The question at hand is not whether the inner evidences count as personal or subjective
justification —they do— but whether the exclusivist can infer epistemological differences from
phenomenological ones.
Appeals to a first-personal perspective and phenomenology seem ill-equipped to break
the default symmetry among people as epistemic peers on a certain matter. They can be used to
deny peerhood, but only, I think, in a rhetorical way, by the insulating moves of saying the
evidence-sharing condition on peerhood is never met across traditions, and that religious
disagreements therefore are not peer disagreements. Adherents of alien religious traditions can
easily mirror a self-guaranteeing claim that, ‘I feel it so strongly, it can’t be false.’ The next
move, again shared by all, is to ‘logically’ infer the falsity of any view contrary to one’s own
orthodoxy. The claim of the self-guaranteeing authority of the home scripture or experience is
something religious studies scholars associate with strongly fideistic models of faith, but these
claims have never settled any actual disagreement. At their worst, exclusivists about the truth and
worth of purported special revelations generate their own ‘enemies in the mirror.’ Let us use the
resources of a philosophy of risk and luck to erase, as I have tried to do, the phenomena of self-
constructed enemies in the mirror. Doing so it seems to me is genuine religiosity—at least, it
requires no anti-realism about religious language.
Those who appeal instead to externalist epistemology in order to defend an attitude of
religious exclusivism, such as Alvin Plantinga, may also often appeal to unique phenomenology.
But they would have us accept that we can deny etiological and peer symmetries by starting from
the other end, from the posited truth of their religious worldview and from the religious
knowledge they purport to have. Unfortunately, jumping ship on the epistemic level and the
requirement of independent reasons, by telling a metaphysical/theological story about the causes

of one’s experience that set it apart as especially veritic, has never resolved religious
disagreement either.
Despite the twists in Plantinga’s account of a divine design plan, I hold the basic belief
apologetic to be no less fideistic and circular than a phenomenal conservative apologetic
strategy. Both make religious knowledge very easy, but falter over why views contrary to their
own do not also count as properly basic, warranted, and safe. All that either apologetic strategy
ultimately does is the say that warrant (externalism) or undefeated evidence (internalism) only
follows the true, and beliefs running contrary to the Christian’s aren’t true. Here again
epistemological questions are given metaphysical answers. On Plantinga’s externalist apologetic
strategy, a reliable etiology (safety or warrant) is assured for general theistic and also for
Christian teachings infused irresistibly by the work of the Holy Spirit. But ask him about the
logical possibility on which the Holy Spirit and God do not exist and we find Plantinga forced to
concede that neither his religion-specific nor his theistic beliefs would in that case be warranted,
either. Their warrant depends upon supposition of their truth: no truth, no warrant. This explains
why the potential mirroring of contrary views that also assert properly basic beliefs in the Great
Pumpkin-believers cases, for instance, became such a serious problem for Plantinga.
One might think (AS) paints religious belief acquisition with a broad brushstroke, and that
when you look in finer detail, the similarity and formal symmetry of the grounds of belief are
very different. It is true that (AS) is framed broadly, but this is a reasonable level of generality at
which to describe a good deal of belief uptake, and the location-switching thought experiment
works to confirm that something very like this is the causal etiology of the religious beliefs of
most people who grew up in a testimonial faith tradition.

This is of course a claim only about observable proximate causes. That the perspective on
causal etiology in (AS) attends only to proximate causes is a methodological limitation, but it
does not beg the question by presupposing that all religions are epistemically on a par. The
reader should not confuse the charitable defeasible presumption of peerhood when faced with
disagreement, with question-begging. I am not committed to the metaphysical claim that no
religious perspective or experience is especially veritic. Similarly, there is nothing in (AS)’s
identification of a common proximate cause (home religion testimonial authority assumption) in
the acquisition of religious belief that is blind to the claims that some persons use a safer method
than others, and that whatever tradition they acquire their belief from, some persons work harder
than others to improve their epistemic positions. The New Problem encourages attempts to make
good on symmetry-breaking reasons, just in order to understand them more clearly and analyze
their formal structure. Default symmetry, a weaker notion than parity or peerhood, is a defeasible
posit, but one that allows us to focus directly upon particular attempts to break symmetry, and on
the inductive risk explicitly or implicitly accepted in these attempts.
So contrary to the objection, I think that the inductive risk approach presupposes qualitative
or epistemic differences, and invites looking for “finer detail.” It looks for this especially in
symmetry-breaking reasoning, and in the various models of faith that lead adherents to their
response to religious multiplicity. But “finer detail” should not be reduced to an opportunity to
substitute a sacred narrative in place of a call for epistemic situation-improving reasons. I this
closer look reveals that the purportedly unique process mirrors know group biases and inductive
fallacies, then far from its uniqueness being established, this would it would support (AS) being
a good description of it. So this goes case-by-case. If the reasons offered in attempts to grab the
second horn are reasonably independent of the disagreement then symmetry-breaking may be on
inductively strong grounds. But if these reasons turn out to be circular or to lean explicitly or
tacitly on asymmetric attribution of religious luck, then it starts to look like there is far less
genuine uniqueness in claims to religious uniqueness than exclusivists would like us to think.
Conclusion
Chapter 2 argued that the concept of religious luck grants us significant insights into the
moral, theological, and epistemological adequacy of different models of religious faith and the
responses to religious multiplicity that they motivate. The concept of religious luck has to earn
its keep by the value of the diagnostic questions it raises, but it is well able to do so. This is in
part because it draws needed attention to variations among belief policies or doxastic methods,
“theological methods,” as some religious studies call them, prescribed by different actual
traditions of religious faith. This diversity among conceptions of models of faith allows keener
insight into attempts of theologians to produce ‘luck-free’ soteriologies, and into attempts by
philosophers and religious studies scholars to study the logic and illogic of radically asymmetric
religious trait-ascriptions.
The study of symmetry-breaking attributions, especially through a philosophy of luck or
risk, provides theologians, ethicists, epistemologists, and cognitive science of religion new
insights into the causes of religious disagreement and the limits of reasonable religious
disagreement. The New Problem of religious luck presented reason to think the denial of
epistemic symmetry, and the assertion of asymmetric value in God’s eye between adherents of
the home religion and all others, will always on close examination be seen to self-attribute one or
another type of religious luck. It also began a process of translating the rather vague concepts of
moral, epistemic, and religious luck into more terms of risk that are more scalable and open not
just to thought experiments, but to close psychological study. This translation will be developed
through Part II.
The presence of pervasive disagreement and epistemic risk make one’s epistemic context
inductive. So our argument in this chapter shows that a philosophy of luck applied to religious
epistemics must necessarily be concerned with inductive risk, and that the degree of inductive
risk in an epistemic strategy is a formal matter, and potentially measurable. Our study further
suggests that the high degree of inductive risk in contexts in which testimony-based claims to
religious knowledge are typically made is a fully sufficient criterion for determining it to be a
context of malign environmental, rather than benign evidential luck. Testimonial authority
assumption is on this standard a strategy of high inductive risk, even if one holds themselves to
have attained truth through it. For it is counter-inductive that one’s own testimonial authority
assumption should succeed in delivering the epistemic goods, while it is conceded that all
symmetrical authority assumptions on the part of persons in a different testimonial faith
traditions would fail.
Finally, Part I has given initial reason to think that degrees of incurred inductive risk may
not only be a sound criterion for taking the etiology of a belief to be affected by malign epistemic
luck. If so, this suggests new means for measuring where a particular model or conception of
faith falls on the scale between religious rationalism and religious fideism.34 Over-strong
rationalism (religious or sceptical) and over-strong fideism both function to undermine the
possibility of reasonable religious disagreement, as well as to obscure a theory of its moral and
intellectual limits. Religious rationalism in the form of natural theology provides arguments for
God’s existence that do not depend upon scriptural sources.35 Religious fideists by contrast hold
that faith is necessary because a person whose assent to theistic beliefs is matched only to the
degree of rational confidence in their evidence has only an unstable faith. So our approach
suggests that these same concerns with inductive risk and safety principle violation also supply a
way to distinguish moderate and strong (sometimes called radical or counter-evidential) fideism
on the basis of recognizable formal markers. These markers are the implications and applications
of the theoretical issues regarding luck/risk introduced here in Part I. In Part II we will develop
their more specific connections with comparative study of religious fundamentalisms (Chapter 3
and 4), with religious exclusivism (Chapter 5), and with cognitive science of religion (Chapter
6).
1
Montaigne, Apology for Raimond Sebond [1580], 6.
2
It is hard to say, for instance, just what the conclusion of Mill’s argument about the “London
Churchman” in On Liberty is. Ballantyne (2015) starts from this same “pervasive and disconcerting
worry about intellectual life: our controversial beliefs regarding morals, politics, religion, and
philosophy depend on facts about our personal history.” Bogardus (2013) very plausibly suggests
that the alleged wrong—and the basic reason for contingency anxiety— is a violation of safety.
“For Mill, there are nearby possibilities in which one forms religious beliefs via the same method
she actually used, and yet in which she would believe something which is, by her own lights, false.”
But it also seems to involve not-insignificant insensitivity. See also Nathan Ballantyne (2013) and
Baker-Hitch (2014).
3
The truth-aptness of philosophical judgments have been challenged in similar ways. See Amia
Srinivasan (2015) for a treatment of “genealogical skepticism” about philosophical judgments.
Srinivasan’s account overlaps mine where she provides discussion of different arguments for
genealogical skepticism and in response to it; more especially, in her noting that, “Epistemologists
differ over the extent to which luck plays a role in the acquisition of knowledge. All epistemologists
will agree that luck has some role to play... Where epistemologists disagree is on just how much
knowledge we can acquire through good luck” (347). The Tess case argued that this is a matter of
the kinds of luck in play, and not merely the domain.
4
See Mogensen 2017a; Ballantyne 2013; Vavova 2014; DiPaoli and Simpson 2016. Sometimes
these same contingency arguments are called arguments from evidentially irrelevant causes of
belief, or debunking arguments, because they threaten to explain the etiology of these beliefs in a
way that has nothing to do with their likelihood of being truth. Vavova 2018 also argues that
evidence of irrelevant belief influence is sometimes, but not always, undermining. It is surprising
how little this quite plausible thesis has been systematically explored, but epistemologists of
disagreement have enamored of universalist answers.
5
See Axtell 2019 where I develop etiological challenges to well-founded belief as directly related to
a belief’s being highly overdetermined by trait-dependent factors. Although I will not explore them
here, my understanding of these issues has benefitted from papers (see Bibliography) by Baker-
Hitch, Ballantyne, Bogardus, DiPaoli, Mawson, Mogensen, Simpson and others. Religious mobility
sometimes occurs for practical reasons of marriage, and another big reason why young adults aged
16-30 are far and away the most mobile or identity-shifting age-group, is suggested in multiple
studies to be parental divorce. The data on religious mobility show some trends, but that people are
mobile in both directions between religious and secular identities is a basic fact. The overall rise of
religious mobility, or what are called “total mobility” levels, compared with decades ago, suggests
to me a national trends towards greater reflection, and personal responsibility and active,
autonomous pursuit of options in a marketplace of ideas. This trend towards increased religious
mobility would clearly suit Mill, a strong proponent of “experiments of living,” and James. It suits
this book as well, supporting the openness not only to choice, but also to criticism of every
historical conception of faith, every doxastic method, and every degenerative research program in
any domain.
6
According to J. Adam Carter (2017), recognition of peer disagreement implies that “we are
rationally obligated to withhold judgment about a large portion of our beliefs in controversial
subject areas, such as philosophy, religion, morality and politics.” He recognizes that a thorough-
going agnostic suspension or the kind recommended by some epistemologists is open to
objectionable consequences of ‘spinelessness,’ and impracticability—un-liveability. So he distances
his version of controversial view agnosticism from these worries, qualifying it such that it allows
for ‘suspecting that’ but not ‘believing that.’ In a nutshell I would draw the lines differently, and
would not accept a non-permissivist view. In defence of permissivism, see Kelly (2014), Kopec and
Titelbaum (2016), Booth and Peels (2014), and Schoenfield (2014).

7
Philip Quinn identifies a crucial shared assumption between evidentialists and the others like
phenomenal conservatives: “One should demand no more, and no less, by way of justification for
beliefs in one area of inquiry than one does in another. Equally stringent standards of rationality
should apply in all cognitive domains” (1991, 317). The first group thinks religious belief fails these
standards, and the latter that they meet them. But by setting aside this faulty shared assumption of
religious evidentialism in favor of pragmatism, we can make a similar reply to epistemic parity
arguments like those of Plantinga. The project of a broader epistemology for our controversial
views is indeed one that I am interested in, but will require another book. But Quinn finds the
argument unsound: “It may be that there is no way to prove that the skeptic is mistaken, and so
perhaps the only escape from skepticism is by means of a leap of faith. But even if this is so, it does
not follow that all leaps of faith are, so to speak, equal in size. Some leaps of faith may be, as it
were, bigger than others” (340; compare Penelhum 1983).
8
A recent PEW study (2009) put the number of adult respondents “who have changed from one
major religious tradition to another” at 9%, a figure that closely corroborates earlier surveys I have
seen that asked this question. This group includes “converts from a variety of different
backgrounds, including converts to Catholicism and converts from or to religions other than
Catholicism or Protestantism.” Somewhat contrary to the contingency argument assumption, the
study finds that “Americans change religious affiliation early and often. In total, about half of
American adults have changed religious affiliation at least once during their lives. Most people who
change their religion leave their childhood faith before age 24, and many of those who change
religion do so more than once….” But at the same time, the authors of the study also say, “The
group that has grown the most in recent years due to religious change is the unaffiliated
population.”
9
Studies of religious mobility certainly undermine the simplistic nurturist claim that culture is
destiny. For recent studies of religious mobility, see also T. Smith (2005). The PEW study
published in 2009 used a re-contact survey years after the original religious Landscape 2007 survey,
found only 56% of respondents were in the same Christian denomination as that in which they were
raised. “The 2007 survey found that more than one-in-four American adults (28%) have changed
their religious affiliation from that in which they were raised. That number includes people
changing denominations—often local churches. Thus the 56% would increase very substantially to
just above 90%, and corroborate the 1988 figure, if remaining in the same “affiliation” were taken
as remaining Christian, etc. These are strong numbers, but I suggest we ask more directly what the
empirical presuppositions of properly-qualified contingency-of-belief arguments are, and whether
there are extant studies that bear upon it. If not, what would an empirical investigation on this
matter look like--what specific questions would it need to ask? How would this questions differ
between the domains of controversial views?
10
Davis (2009, 23). See Axtell (2019) for development. Davis defines ‘trait’ broadly to include
“not only personal traits such as gender or features of one’s personality, but also such properties as
socioeconomic background, rigorous training, exposure to certain individuals or groups, subscribing
to a certain ideology or religion, or having a certain personal history.”
11
Davis does not think it is correct to say that traits consist of holding particular beliefs, though the
trait may involve holding some beliefs. (24) A bias in cognition “is a tendency towards a certain
kind of distortion in one’s process of reasoning and belief formation.” But in trait-dependence cases
as Davis wants to understand them, ex hypothesi “there are no such distortions – just the
dependence relation… Trait-dependence is not a cognitive distortion unless trait-dependence
defeats the basing relationship – and that is the question before us; we cannot assume an answer to
it.” (24)
12
See especially Bishop 2007a and b on how moderate fideism can constrain doxastic ventures.
Note that my account does not depend on testimonial knowledge being inductive or inferential; it is
neutral with respect to debates between reductionists and anti-reductionists. But if testimonial
uptake is inferential, then this still further substantiates logical connections between belief-
acquisition through testimonial authority assumption and high inductive risk.
13
Hempel 1965, 92. See Douglas 2000 for discussion.
14
Baker-Hytch, 190.
15
This, despite the fact that the belief of the victim of a malin genie that he has a physical body is as
insensitive also. My entitlement to hold fast to my metaphysical beliefs in a physical universe, other
minds, etc. where all sources of empirical evidence support my causal story does not extend to an
entitlement to hold steadfast in the case of more controversial views. My thought here, which
relates to epistemic parity arguments, is roughly that sensitivity tracks reasonableness when the
closest error-possibilities are modally close, but it doesn’t track reasonableness when the closest
error-possibilities are distant. Thanks to Patrick Bondy for helping me articulate this suggestion.
16
Kevin Wallbridge (2018) argues that while inductive knowledge may not be strongly sensitive, it
is weakly sensitive. The point is supportive of the importance of the insensitivity of counter-
inductive thinking to epistemic rationality or reasonableness, even though I do not take sensitivity
as a general necessary condition on knowing.
17
Insensitive belief that is not based upon shared facts but rather on private intuitions or other
subjective factors cannot claim the same reasonableness as the belief that I am not radically
deceived. My entitlement for the latter does not imply entitlement to the former, potentially much
more temperamental choice.

18
James Fritz (2018) writes, “Many writers have recently argued that there is something distinctively
problematic about sustaining moral beliefs on the basis of others’ moral views. Call this claim
pessimism about moral deference. Pessimism about moral deference, if true, seems to provide an
attractive way to argue for a bold conclusion about moral disagreement: moral disagreement
generally does not require belief revision. Call this claim steadfastness about moral disagreement.”
Fritz then goes on to argue that there no easy or compelling root from pessimism about moral
deference to steadfastness about moral disagreement.
19
Relatedly, Rik Peels (2016) argues that responsible belief “is compatible with its being a matter of
luck that we hold that belief.” The present analysis, however, shows that it very much matters what
kind of luck is salient, and whether an appeal to religious luck for holding the beliefs one does is
morally as well as epistemologically concerning. This does not mean that I disagree with Peels;
indeed I very much agree that if we want to act responsibly, we should believe responsibly, and that
doxastic responsibility has diachronic and not merely synchronic aspects. Peels develops the
interconnections between ethics and epistemology through the problem of doxastic involuntarism,
which I agree with him gives in many of its forms a distorted view of belief. Peels goes on to give a
detailed account of how we can still believe responsibly “if we are excused for our belief by force,
ignorance, or luck.” My term constitutive epistemic luck seems to be broken down further by Peels
into mechanism and aretaic luck.
20
For a thorough introduction to this important topic, see Molinini, Pataut, and Sereni (2016).
21
Molinini 2012.
22
Tables 1 and 2 reprinted complements of Daniele Molinini.
23
Stanovich, West, and Toplak 2013, 259. See also the authors’ 2008. There many studies show that
the magnitude of the myside bias shows very little relation to intelligence (or to whatever IQ tests,
tests). Studies demonstrate quite substantial individual differences on tasks normed to commit or
avoid myside bias; these demonstrated differences, cognitive science has sufficiently established,
are largely is due largely to differences in critical reasoning skills and dispositions.
24
Fluid rationality describes a range of available critical reasoning dispositions, together with
ability for sustained de-coupling of their default, heuristic thinking. When utilized by the agent, s/he
exhibits resistance to misery informational processing and my-side bias.
25
Carter (2017) usefully develops conditions of the “centrality” and “symmetry” of disagreement in
a domain as a way a criterion for challenges to the well-foundedness of belief in domains of
controversial views. In our terms, when these conditions are met, the discourse is likely only
minimally truth-apt, and extrinsically adequate theories need to take account of this fact that
propositions can only be ‘quasi-asserted.’
26
Wright, Critical Thinking, 206. OUP, 2013.
27
According to David Hume, “All those opinions and notions of things, to which we have been
accustomed from our infancy, take deep root, that ‘tis impossible for us, by all powers of reason and
experience, to eradicate them.” (A Treatise of Human Nature). D.J. Hanson tries to show that
Hume is the first English utilizer of fideism, and that his ‘fideism’ was a solution which he found to
extricate himself from his scepticism (5). Bayle was “the chief Pyrrhonist influence on Hume,” and
led Hume to be a fideist in both his philosophy of religion and his metaphysics. Penelhum puts
Bayle along with Catholics Montaigne and Erasmus, in the skeptical rather than the evangelical
fideism camp. Penelhum defines prescriptive fideism as “the insistence that faith needs no
justification from reason, but is the judge of reason and its pretensions” (quoted by Hanson, 10).
Richard Popkin suggests that if a person accepts “metaphysical doctrines of cause, personal
identity, substance, or external objects (common sense beliefs), or even religious doctrines of
miracles, immortality, or the existence of God on non-rational basis, we should judge him a fideist
rather than a skeptic. For he then accepts non-rationally what he rejected on a rational basis”
(quoted by Hanson, 5). But Bayle like Erasmus encourages his readers to turn away from doctrinal
disputes altogether. This separates him from Pascal and Kierkegaard as evangelical fideists, where
most of the other Protestants appear. So Penelhum puts Bayle along with Catholics Montaigne and
Erasmus in a skeptical rather than evangelical fideism camp.
28
Safety and sensitivity are both externalist conditions, so that to the extent that it is more plausible
to take them as failing than as being met in an exclusivist belief, they rebut the kind of claim de
Ridder and Plantinga make, that defeaters are defeated by a reiterative application of how strongly
one “feels” one’s belief to be true. That is an appeal to phenomenal consciousness. J. J. Ichikawa
and M. Steup (2014) point out that sensitivity, like safety, is “an externalist condition on knowledge
in the ‘access’ sense. It is also externalist in the ‘state’ sense, since the truth of the relevant
counterfactuals will depend on features outside the subject.”
29
Christiansen (2014). Kelly discusses another passage by Christiansen (in Christiansen and Lackey
eds. 2013, 37): “The motivation behind the principle is obvious: it’s intended to prevent blatantly
question-begging dismissals of the evidence provided by the disagreement of the others. It attempts
to capture what would be wrong with a P-believer saying, for example, ‘Well, so and so disagrees
with me about p. But since P is true, she’s wrong about p. So, however reliable she may generally
be, I needn’t take her disagreement about p as any reason at all to change my belief.’ There is
clearly something worrisome about this sort of response to the disagreement of others. Used as a
general tactic, it would seem to allow a non-expert to dismiss even the disagreement of large
numbers of those he took to be experts in the field.” While Christiansen treats acceptance or
rejection of an independence principle as a crucial divide between conciliationism and non-
conciliationism, Kelly suggests certain qualifications to this equivalence, and to Christiansen’s
proper formulation of the principle itself.
30
The combination of (DS) and (AS) poses a defeater for the radically asymmetrical trait-
attributions and explanations of religious status that are part and parcel of attitudes of religious
exclusivism, and (EGO) places the reasonable constraint that one cannot just appeal to the notion
that ‘everyone has an equal right to an opinion,’ or that ‘my religious phenomenology or my
religious authority is unique,’ is the end of the story and sufficient rational grounds to dismiss the
epistemic significance of geographic – psychographic epistemic location problem.
31
I here focus on Plantinga’s claim of external rationality, but Baldwin and Thune nicely articulate
related objections to reformed epistemologists’ use of internal rationality. An iterative response to a
potential defeater, a response of simply saying that a defeater is itself defeated by reapplication of
the ‘If properly produced, then warranted’ claim, is viciously circular because it just privileges itself
to the truth of the antecedent of this conditional, even though that was precisely what the potential
defeater challenged. “Plantinga’s external account of warrant includes the stipulation that a belief
must be internally rational if it is to be warranted. And he tells us that ‘[i]nternal rationality
includes, in the first place, forming or holding the appropriate beliefs in response to experience.’”
The appeal to experience cannot be watered down so much that anything firmly believed is
internally rational. That just confuses psychology and epistemology; “it is hard to see how a belief
could come to be indefeasible in this way” (446-447).
32
De Ridder 455. Plantinga appeals to what he calls “impulsional evidence,” something that might
be present in cases of necessary truths but which he applies far more generally. Todd Long (2010)
argues that Plantinga assumes (1) one’s impulsion to believe (i.e. the experience of the impulse
alone) causes the proposition to seem true to one; thus, (2) ‘impulsional evidence’ is evidence for
the proposition believed. He continues, “But, the inference from (1) to (2) is doubtful. Note first that
feeling impelled to believe a proposition is not always epistemic. A proposition such as ‘everything
will turn out all right’ may be so attractive to a person in a desperate situation – because it is
reassuring – that the person feels impelled to believe it. But its being reassuring does not itself
provide any epistemic reason to believe the
proposition. An epistemic reason to believe a proposition just is an indication that the proposition is
true. The desperate-situation example shows that one may feel impelled to believe a proposition
even if one has no epistemic reason to believe it. So, if one does have an epistemic reason to believe
a proposition, that reason is something other than the impulsion to believe. The upshot is that an
impulsion to believe is not itself evidence. Second, reflection suggests that Plantinga has backward
the explanatory relation between seeing the truth and feeling impelled to believe. One doesn’t see
the truth of a proposition because (in part) one feels impelled to believe it, but rather one feels
impelled to believe the proposition because one sees its truth” (11).
33
Gellman’s “contented” religious exclusivist also clearly fail an independence principle like
(EGO). Gellman (2008, 381) concedes that “My defence of a rational exclusivism, therefore, cannot
rationally be invoked as a protective strategy by just any want-to-be exclusivist.” Gellman’s thesis
is that “religious beliefs may go rationally unreflective for the religious believer in the face of other
religions. A believer may rationally invoke her unreflective religious belief to defeat opposing
religious claims, without having to consider the question any further” (375). But then the notion of
rationality which Gellman employs for showing when exclusivist responses to religious multiplicity
are not a protective strategy, is a psychological notion: whether for the individual, her religious
beliefs “squeak in the face of religious diversity.” But then this subjective method cannot answer
the question, which was, ‘Is the contented exclusivist derelict in any epistemic obligations?’ It can
only change the topic, and if that is not a merely evasive maneuver it is a kind of teleological
suspension.
34
On religious rationalism compared with fideism, and the relationship of natural theology to
particular faith ventures, see Bishop 2007 a and b.
35
Natural theology presents proofs for the existence of God, and aims to provide reasoned support
for monotheism, through non-revealed sources. For a CSR perspective on the “enduring appeal” of
natural theological arguments, see de Cruz (2014). Blanshard (1974, Chapter vii), a skeptical
rationalist, writes that “The rationalist in philosophy need not, so far as I can see, deny revelation in
religion. But he will say with the best will in the world, he cannot accept a revelation that
contravenes his reason, because such a revelation he could make no sense of…. [O]nce reason is in
abeyance, the best of us are not safe against confusing our preconceptions with leadings from on
high.” But that is a high mark since all religions shock and contest our everyday reason to some
extent. So I take “contravenes” and reason “in abeyance” in the strong sense of a kind of
teleological suspension of moral, epistemic, or logical norms.

Cover page
Bridging a fault line: On underdetermination and the ampliative adequacy of competing

theories
Guy Axtell, Philosophy, P.O. Box 6943, Radford University, Radford, VA, USA 24142
gsaxtell@radford.edu
Abstract This paper pursues Ernan McMullin’s claim that talk of theory virtues exposes and
perhaps helps to bridge a fault-line in philosophy of science separating “very different visions”
of scientific theorizing. It argues that connections between theory virtues and virtue
epistemology are substantive rather than ornamental, since both address underdetermination
problems in science, helping us to understand the objectivity of theory choice and more
specifically what I term the ampliative adequacy of scientific theories. The paper argues
therefore that virtue epistemologies can make substantial contributions to the epistemology and
methodology of the sciences, helping to bridge the gulf between realists and anti-realists, and to
re-enforce moderation over claims about the implications of underdetermination problems for
scientific inquiry. It finally makes and develops the suggestion that virtue epistemologies, at least
of the kind developed here, offer support to the position that philosophers of science know as
normative naturalism.
Keywords underdetermination; virtue epistemology; theory-choice; empirical adequacy;

cognitive values; normative naturalism
1
Bridging a fault line: On underdetermination and the ampliative adequacy of competing
theories
Guy Axtell
Abstract This paper pursues Ernan McMullin’s claim that talk of theory virtues exposes and
perhaps helps to bridge a fault-line in philosophy of science separating “very different visions”
of scientific theorizing. It argues that connections between theory virtues and virtue
epistemology are substantive rather than ornamental, since both address underdetermination
problems in science, helping us to understand the objectivity of theory choice and more
specifically what I term the ampliative adequacy of scientific theories. The paper argues
therefore that virtue epistemologies can make substantial contributions to the epistemology and
methodology of the sciences, helping to bridge the gulf between realists and anti-realists, and to
re-enforce moderation over claims about the implications of underdetermination problems for
scientific inquiry. It finally makes and develops the suggestion that virtue epistemologies, at least
of the kind developed here, offer support to the position that philosophers of science know as
normative naturalism.
Keywords underdetermination; virtue epistemology; theory-choice; empirical adequacy;

cognitive values; normative naturalism; Pierre Duhem
“Discussion of theory virtues exposes a fault-line in philosophy of science that

[separates] very different visions of what the natural sciences are all about.” ---
Ernan McMullin1
1 Kinds of underdetermination and the worries they arouse
Logical empiricists posited a singular logic of inquiry across all sciences, and pied-pipered social
scientists to follow them by emulating the methods of “hard” science. Post-positivist thinkers
denounced the pied-piper, but themselves often imposed an equally rigid egalitarianism among
1
“The Virtues of a Good Theory” (2009), 506.
2
academic fields and cognitive styles as sources of knowledge.2 More careful attention to
problems related to underdetermination, however, indicates a relationship between the sciences
at once both less neat but far more intriguing than either of the foregoing views: A more dappled
relationship where some bits of history or sociology may be more reliable than some bits of
physics; where underdetermination problems are not expected to track any traditional distinction
between hard and soft sciences, and where worries about local underdetermination are felt as
regularly in such areas of natural science as contemporary theoretical physics and cosmology, as
in the social sciences like economics where we usually locate them.
Attention to the underdetermination problem in the sciences is also one among a number
of bridges between philosophies of science and virtue epistemologies (hereafter VE), or so I will
argue. As Ernan McMullin writes in “The Virtues of a Good Theory” (2009), “The assessment of
theory is a form of inference quite different from [the Newtonian account of] induction over a set
of observation reports resulting in a law-like generalization” (501). The verificationist
conception of objectivity, he shows, moved under criticism (around mid-century) to a fall-back
position closer to the hypothetico-deductive account; yet this could not satisfy critics of
positivism who pointed out that assessment of theory is more often a comparison of extant rivals,
and less often a sheer encounter between a stand-alone theory and one or more experimental test.
Rare at best is the Popperian “crucial experiment” or the associated notion of a quick kill of a
theory faced with recalcitrant evidence from a disappointing test result.
2
Connected with this is Sankey’s point that “while empiricists explain consensus but have a hard time
with disagreement, post-empiricists emphasize dissensus at the cost of being unable to explain how
agreement is arrived at. But [any] adequate philosophical model of scientific rationality must explain both
consensus-formation and the existence of widespread disagreement” (1996, 1).
3
The deeper reasons for these criticisms of the received view of theory choice involve a
long discussion of methodological holism in response to recognition of certain kinds of worries
about the underdetermination of theory by data, and again in a still deeper worry, about the
underdetermination of theory choice, in one case or many, by the methodological principles and
norms of the scientific community. The former kind, sometimes called logical
underdetermination or Humean underdetermination, I think of as a ‘global’ but at the same time
a rather weak thesis that no theory is strictly-speaking proven or entailed by its confirming
instances/predictions.
This is how Larry Laudan formalizes the first (‘Humean’) kind in “Underdetermination
Demystified” (1990a, 323):
(HUD) For any finite body of observable evidence, there are indefinitely many mutually
contrary theories or hypotheses, each of which together with auxiliary assumptions
logically entails that evidence.
Laudan characterizes (HUD) as deductive underdetermination, “Quinean holism” (QUD), by
contrast with (HUD), is most evident in the early Quine, along with his writings on meaning and
incommensurability. (QUD) or (QUD)-related claims make stronger claims than (HUD) about
the underdetermination of theory by data:
(QUD) Any theory can be reconciled with any recalcitrant observable evidence by
making suitable adjustments in our other assumptions.
But fortunately for my readers and me both, we won’t need to go there. As Laudan points out,
Quine in "Two Dogmas of Empiricism" "propounded [but did not give any good reasons for
4
believing] a thesis of normative, ampliative, egalitarian underdetermination" (334). He later
conceded in writing that his early views about holism were both stronger than was needed in
order to challenge the dogmas of empiricism, and stronger than he wished he would have made.3
From the direction of Quine scholars, the so-called Duhem-Quine Thesis is now widely regarded
as a mistaken designation, since Duhem’s views were considerably milder than Quine’s; but for
present purposes we needn’t go into these matters, either.4
(HUD), nevertheless, already shows us a quite substantial sense in which theory-choice
turns upon non-deductive or ampliative desiderata, which primarily include the virtues of a good
theories (theory virtues). Often in the literature these are called cognitive values, a term that like
theory virtues can cover both axiological and evaluative concepts. That scientists qua scientists
make value judgments is arguably a key tenant of all post-positivist philosophy of science, and
reasoning or assessing theories and hypotheses through the cognitive values shows ‘values at
work’ in science, as the Kuhnian phrase goes. But the virtue epistemologist agrees with
3
Laudan argues effectively against Quinean underdetermination, which isn’t to say that there aren’t
important and serious issues of meaning, but that Quinean meaning incommensurability is not as different
as Quine likely imagined from Carnapian ‘external questions,’ Wittgensteinian language games or forms
of life, etc. Quine maintained his acknowledgment of theory virtues in scientific reasoning even while he
repudiated much of his meaning holism (and pseudo-pragmatism). But the version of naturalized
epistemology he developed never discovered the roots or fruits of a more genuine pragmatism. The issues
of the meaning and use of thin and thick axiological and evaluative concepts is accounted for more
successfully in our own pragmatist or inquiry-focused —zetetic-pragmatist— virtue epistemology,
through its primary focus on practices of inquiry and what can be learned and emulated about associated
traits of good inquirers. Meaning is use, to the classical pragmatists, after all, and aren’t matters of use
matters of habit and therefore of practice?
4
For an admirably clear explanation of the differences between Duhem and Quine, and the Quinean
history of retracting (QUD) and his own mistaken conception of pragmatism, see Massey (2011).
5
McMullin when he writes that “Calling them ‘virtues’ rather than ‘values’ draws attention to
their status as attributes at once objective and desirable”.5 As an aside to be developed later,
virtue theorists of all kinds present a Janus-faced (or compatibilist) conception of the relationship
between naturalism and normativity. Whether one prefers to characterize “fertility,” “explanatory
power,” “coherence,” “simplicity,” etc. as cognitive values or theory virtues, the recognition that
reasoning about the ampliative adequacy of scientific theories relies upon a combination of
descriptive/evaluative concepts such as these might be thought to suggest the need for a
‘thickened’ conception of epistemology such as virtue epistemologists have often argued for.6
There is another thesis different than and stronger than (HUD) that must concern us, one
that more directly fuels the shared worry of relativists and objectivists that if method doesn’t
compel one theory choice over another, the normativity involved in theory choice is so troubling
as to imply that science isn’t a rational enterprise. By contrast with deductive
underdetermination, Laudan identifies this worry with a thesis about ampliative
underdetermination, or the underdetermination of theory choice by methodological rules or
standards. It is a worry that ampliative desiderata in the form of theory virtues and the good
sense of the researchers themselves fail in providing a rationally satisfying preference weighting
in any or all situations. Regarding ‘any or all,’ Alex Rosenberg (2012) argues, and I think rightly,
that “The problem of empirically equivalent but logically incompatible theories becomes
5
E. McMullin, 501.
6
On virtue epistemology’s role in respect to recent calls for the “thickening” of epistemology, see the
papers in the 2008 Philosophical Papers edition, Epistemology Through Thick and Thin, 37(3). These
include Guy Axtell and Adam Carter, “Just the Right Thickness,” which identifies and challenges an
epistemological analogue of the (ethical) “centralist” thesis (of the primacy of thin concepts over thick)
that Bernard Williams criticized.
6
especially serious as science becomes more theoretical.”7 But this does not in itself support the
global nature of the worry about ampliative underdetermination. Its globality and association
with a thesis of “non-uniqueness” should be captured in our definition of it, in order to show the
strong claim that it makes with respect to the rationality of science. So let us define ampliative
underdetermination like this:
(AU) Ampliative reasoning, or reasoning through thick evaluative and/or axiological
concepts, always results in a rationally-unsatisfying non-unique choice of theory or
hypothesis.
(AU) is a contentious but not completely implausible thesis, and our way of framing it
does seem to capture a number of worries about norms governance in science. While I will argue
that (AU) and related worries about the non-uniqueness of resolutions to local
underdetermination problems are over-stated—as indeed much discussion of underdetermination
and holism has been over-wrought—it is a thesis that will need to be scrutinized carefully in later
sections of the paper.
It may be that the kind of intractable, irresolvable theoretical disputes that ampliative, in
contrast to deductive underdetermination seems to make possible are almost never actual in
scientific practice. Still, the logicist conception of objectivity is already shattered by recognition
that ampliative reasoning is ever-present in scientific reasoning, and is inescapably normative.

7
Rosenberg, 212. Any attempt at a more directly empiricist justification for the methodological rules we
employ in theory choice, he concedes, “is circular as an argument against the threat of
underdetermination,” (214) and appeals to them as a priori are unavailable to empiricists, who would be
right to doubt them anyway. Thus he concludes that neither the rationalist nor the empiricist props for an
algorithmic account of theory choice is at all satisfactory (214). I lean on Rosenberg especially here
because he seems to acknowledge that even strong empiricists like himself can accommodate a
substantial role for theory virtues and for ampliative reasoning more generally in theory assessment.
7
The fact/value dichotomy collapsed along with logical positivism, under the weight that
the ‘language of science’ was a companion in guilt, as Putnam put it, with the ‘language of
ethics,’ to which the positivists had hoped to banish all problematic normativity. A new scandal
of reason imploded positivism from within—the gig of normative science and positive ethics was
up—if as McMullin claims, “[B]esides the test of observation, theories are also judged on other
criteria: simplicity, economy, explanatory unification, precision in prediction…consistency with
other already adopted theories…amount of allowable experimental error, etc.”
For philosophers of science who show proper concern for underdetermination problems
without going so far as to invoke radical or relativistic implications from them, theory-
confirmation and disconfirmation is describable as utilizing a “toolbox” of theoretical virtues:
“Theory choice is a continual process of iterative applications of this same toolbox of
considerations in order to assess the implications of empirical observation in making theory
choices” (Rosenberg, 214). Empiricists like Rosenberg (214) concede that while there are
disagreements and sometimes very great disagreements among theorists, yet over time these
disagreements are settled, and a new consensus emerges, by reasoning about ampliative
adequacy. But the same acknowledgment will not come from those who find hold (AU), since
their conceptions of the rationality of science do not allow that non-unique or non-logically-
deduced choices can ever be rationally satisfactory ones. The terms on which (AU) employs are
slippery ones that we will have to examine closely.
The connections we are beginning to see here between talk of “theory virtues” and virtue
epistemology are not merely ornamental. Both directly address underdetermination worries by
developing norms for ampliative (non-deductive) reasoning in the sciences. Moreover,
application of theory virtues to choice among competitors involves weighing these theory virtues
8
against one another, and thereby calls upon the judgment –or what Pierre Duhem would call the
bon sens—of the scientist as a qualified expert in his or her field. If so, then a condition of
character is implied even where the explicit appeal may only be to impersonally-framed theory
virtues. Indeed both the appeal to impersonal theory virtues and the appeal to personal bon sens
remind us again that the scientist qua scientist makes value judgments. This is all the more so
when theory assessment is a comparative choice among extant rivals. So what we will here term
the ampliative adequacy of a theory is conceived largely as a matter of comparison of extant
rival theories, and not as a question only of a single theory and its relationship to observational
data.
Moreover, virtue epistemology as will be argued further is what helps us to find
continuity between contexts of scientific assessment that are not beset by localized
underdetermination worries, and those that are. This continuity and the attendant sense of how
theory virtues and the intellectual virtues of researchers themselves aid objectivity in science,
shows us how virtue epistemology serves philosophy of science by allowing us to accommodate
the turn to practice, while also avoiding relativism. Indeed the present view, like McMullin’s,
directly responds to the shared assumption of logicists and their radical historicist critics, that if
theory choice isn’t fixed by observations or some kind of observation-linked algorithm, then
when it is fixed it must be by non-epistemic factors, like personal bias, desire for authority, fame
and fortune, etc. Virtue theory can also show us quite straightforwardly and well, I think, how to
say ‘Neither/Nor’ to this bogus kind of ‘Either/Or.’8

8
Like Daston and Galison in their book Objectivity ( 2007), I would argue that the concept of scientific
objectivity has a history, but that the epistemic norms that have informed scientific practice can be
historicized without leading to relativism. I have elsewhere argued that considerations stemming from
underdetermination problems motivate the claim that historicism requires agent-focused rather than
merely belief-focused epistemology, and that this is partly what makes it possible to distinguish weak or
9
In this paper I urge a virtue epistemology extricated from any overly-strong interpretation
of holism or of underdetermination, and one that isn’t strongly committed to a particular position
on the realism/anti-realism debate but which will take issue with supporters of (AU)’s dismissal
of the rational acceptability of ‘non-unique’ theory-choices in science. We will address different
concerns about underdetermination as well as different versions of virtue epistemology. I will
have one (relatively independent) thesis for each of the four sections. Section 1 develops a
taxonomy of theory virtues that addresses both prescriptive guidance and normative assessment.
I argue for conceiving “epistemic fit” as a virtue concept, and for the relevance to scientific
meta-methodology of a whole range of ampliative desiderata that fill-out and complement the
central scientific virtue of epistemic fit. Section 2 discusses a thesis that Richard Dawid argues
for, which describes what he sees as a “substantial shift” he sees occurring in contemporary
fundamental physics: “the increasing importance of assessments of scientific
underdetermination” (2011, 2).
Dawid’s thesis is a reasonable one, nicely descriptive of problems of theory choice in
high energy physics and the localized reliance there on standards of ampliative adequacy. I argue
that it thereby also indicates a need in philosophy of science to utilize a kind of virtue
epistemology. But what kind of virtue epistemology, specifically? One that essentially stops with
theory virtues shared by a community of inquiry or one that appeals as well to the virtues or bon
sense of good researchers themselves? That raises the issues of what we’ll call ampliative
underdetermination. Section 3 develops these further connections between ampliative reasoning
moderate historicism from radical historicism about the epistemic values recognized in science. See my
“The Dialectics of Objectivity,” (2012) for a development of my moderate historicism and inquiry-
pragmatism, in a recent topical issue of Journal of the Philosophy of History on “intersections between
historicism, and naturalized/virtue epistemologies.”
10
and theory choice by joining a recent debate among philosophers of science over Pierre Duhem's
account of the bon sens or good sense of scientific practitioners. I work out my differences from
Abrol Fairweather, David Stump, and Milena Ivanova in their respective interpretations of
Duhem, and from Ivanova in her adherence to what Longino describes as van Fraassen’s ‘bundle
and reject’ approach to the super-empirical virtues, theory virtues other than logical consistency,
empirical adequacy, and empirical strength. Section 4 further develops what we termed a Janus-
faced conception of the descriptive and normative functions of thick axiological and evaluative
concepts that ampliative inference is based upon. This final section argues against rational
reconstructionist conceptions of scientific meta-methodology, and in favor of conceiving virtue
epistemology as cohering with and supporting a “normative naturalist” alternative. With one
major exception, I present normative naturalism much as Laudan (1987b; 1990; 1996) earlier
developed it.
2 The virtues of empirical and ampliative adequacy
McMullin provides a useful taxonomy of theory virtues, a taxonomy that as one commenter puts
it, “preserves the epistemic character of scientific theory without confining the epistemic values
merely to first-order ‘empirical adequacy’ as van Fraassen understands it.”9 Is empirical
adequacy always the ‘thin’ notion associated with a test, or does it in function in scientific
debates function more like set of virtue concepts? McMullin argues for the latte view. The
association of theory confirmation with deductive implications of observations and tests should
be restricted to the synchronic and retrospective virtue of empirical fit. Empirical Adequacy
actually refers to a more over-arching class of cognitive virtues than does empirical fit, and on
9
Allan, 81.
11
close inspection contains some forward-looking sub-virtues. “Empirical fit should be
distinguished from empirical adequacy, as this is defined in van Fraassen’s constructive
empiricism. Empirical adequacy refers to all of the consequences of the theory, regardless of
whether they have ever actually been drawn or checked against observation” (502).
McMullin sub-divides what I call the theoretical virtues of Ampliative Adequacy into
internal, contextual and diachronic virtues. These virtues he presents as complementary to the
central theoretical virtue of epistemic fit. This claim is one I would agree with, for the present
view can accept testability as a hallmark of science; the idea that theory virtues function to
complement empirical fit already supports a potentially robust contribution of virtue
epistemology to philosophy of science. Resolving a localized situation of underdetermination
through a new test providing decisive empirical advantage of one theory is the best outcome.10
But where it is not to be had, then with McMullin we must “argue for the relevance of a whole
series of confirmatory virtues that complement the central virtue of epistemic fit, transforming
natural science from a mere saving of the phenomena to a genuinely explanatory and
ontologically expansive enterprise” (2009, 502).
“Internal” virtues like internal consistency are well-acknowledged while others also are
important in limiting the degree of allowable ad hocness. “Contextual” virtues include external
consistency and consonance, which address consistency with background knowledge, and
optimality, which involves not only retroduction but also the comparative merit of a theory. But
the most unusual and useful feature of McMullin’s taxonomy is his close attention to
10
“In scientific research one always hopes for determination: that the world should determine the
observations we make of it; that evidence should determine the theories we adopt; that the practice of
science should determine results independent of the sort of society in which that practice takes place”
(McMullin, 1995, 233).
12
“Diachronic” cognitive values, including especially fertility, along with consilience and
durability. Fertility, almost an executive virtue for McMullin, is Janus-faced, looking backwards
to novel facts predicted and confirmed, as well as forwards to potential for the hypothesis to
issue new, bold predictions. Unlike “logicist metascience,” as McMullin describes it, defenders
of the relevance of diachronic virtues like fertility, consilience, and durability to theory
confirmation can easily maintain a lively understanding of the importance of history of science to
philosophy of science. McMullin’s emphasis on fertility as epistemic desiderata re-opens the
diachronic aspects of theory assessment that were lost in the logicist account of scientific
objectivity. But we earlier claimed that the more serious underdetermination worries are
localized ones, and that these do not track the conventional distinction between soft and hard
sciences but rather emerge in a more dappled way wherever a field of study becomes more
theoretical and less directly experimental. Let us now look at how this occurs today in theoretical
physics and cosmology, and at what implications this may have for the centrality of ampliative
reasoning in our conception of scientific objectivity.11
3 Underdetermination and theory choice: The case of string theory
The notion that theory confirmation should be strictly rule-governed, and that accordance with
this logic constitutes the objectivity of science or the rationality of particular scientists, invites
underdetermination and more broadly, skeptical problems. In actual practice, underdetermination
problems are not ‘solved,’ but they typically are resolved after a period of time. They often need
to be if researches are to continue to identify and pursue successful research strategies. It is left
to philosophers and others to debate the epistemic status of the chosen theory and to consider
what distinctions should be drawn, for instance, between distinction between the normativity of
11
Rosenberg 2012, 212.
13
heuristic advice (guidance) and of theory assessment, between theory-pursuit and propositional
acceptance, or again between propositional acceptance and belief etc. With these self-appointed
tasks philosophers have not always done very well. As a case study of this, let us look at
contemporary String Theory, and the quandary in which standard accounts of theory
confirmation leave it.
Richard Dawid writes that,
The canonical understanding of scientific progress…strictly distinguishes
assessments of scientific underdetermination from the core elements of scientific
progress, which are (1) the development of a scientific hypothesis and (2) the empirical
testing of that hypothesis… Assessments of scientific underdetermination, to the
contrary, are taken to constitute mere instances of auxiliary reasoning that may be of
some relevance by channeling scientific activity towards more promising investigations
but do not directly contribute to the generation of scientific knowledge. Put in terms of
the old conceptual dichotomy between context of discovery and context of justification,
one may say that assessments of scientific underdetermination were always
acknowledged as playing some role in the context of discovery but were denied any role
in the context of justification.12
While philosophers and non-practioners tend to accept the canonical view, among high
energy physicists themselves Dawid claims there is a substantial “shift” taking place as they
increasingly question whether the canonical understanding of theory assessment is adequate for
12
Dawid 2011, 4. “Assessments as to how likely it is that no or few alternative theories can be fit to the
available data thus lie at the root of all considerations regarding the prospective viability of a so far
empirically unconfirmed or insufficiently confirmed theory. We want to call such assessments
‘assessments of scientific underdetermination’ (2011, 3).
14
grasping String Theory’s merits. He argues that theoretical virtues must supersede strict
dependence on empirical fit, and that appeals to desiderata of ampliative adequacy “amount to
assertions of limitations to scientific underdetermination”; they should be taken to moderate the
demand that a scientist be agnostic about the parts of her theory not open to direct observation.
Dawid of course is not suggesting that assessments of scientific underdetermination can ever
replace empirical confirmation, but rather that we need an epistemology for the sciences that
“can establish an intermediate epistemic status for theories that lies between ‘empirically
confirmed’ and ‘pure hypothesis.’”13 Dawid more constructively sees the old dichotomy between
empirical confirmation and mere speculation replaced “by a continuum of degrees of credibility,
where the available elements of empirical corroboration and non-empirical theory assessment
jointly contribute to an overall evaluation of theory’s chances of being viable” (2011, 19).
Rejecting the canonical view associated with empiricism and the hypothetico-deductive
model means turning in certain fields of research from objectivity through direct testing to
objectivity through ampliative reasoning. Dawid’s proposal helps make sense out of problems of
theory choice in contemporary theoretical physics, and I hold that this shift is one that virtue
epistemologies help us to articulate and implement. If we do need an altered conception of theory
assessment in high energy physics and scientific cosmology, it is a conception in which
reasoning through theoretical virtues plays a more direct role in theory assessment.
13
“The emerging new paradigm moves away from an understanding…that attributes the status of mere
hypotheses to scientific theories which have found no empirical confirmation.” But Dawid also qualifies
his claim in important ways: “Non-empirical theory assessment thus crucially relies on empirical testing
and can never fully replace it. Nor does non-empirical theory assessment award the same status to a
theory as strong empirical confirmation. It is vaguer and less conclusive than the testing of theories by
empirical data. Its vagueness induces the risk that its deployment might be overstretched….” (2011, 18-
19).
15
But perhaps the reader will think that this way of thinking doesn’t adequately distinguish between
theory pursuit and confirmation, the former being a partly practical, and the latter a purely
epistemological affair; or that it doesn’t address the deeper problems of whether appeal to theory virtues
really helps to provide a basis for uniqueness of the choice of theory. Let’s take these matters up more
directly in the next section.
4 Duhem and the role of bon sens in scientific practice
Thus far we have associated a virtue epistemology for the philosophy of science with the study of
ampliative reasoning utilizing impersonal theory virtues, basically according to McMullin’s helpful
taxonomy. But to what extent will a virtue epistemology draw us also into study of the personal
intellectual virtues of scientists themselves—the good sense or bon sens of the inquirer? I would argue
that scientists’ ‘doings’ are always relevant when we take a practice-focused approach to scientific
reasoning, which includes “inquiry-focused” versions of virtue epistemology such as my own. Certainly
there could be interesting research programs into the personal traits and habits of good scientific
inquirers.
But could the personal bon sens of the scientist ever directly contribute to the epistemic status of
the theory which that scientist chooses? Abrol Fairweather holds that what he terms the “methodological
cognitive character” of the scientist—“the set of abilities, skills and dispositions a scientist acquires and
expresses through the structured forms of inquiry involved in applying scientific methods” (141)— plays
a role supportive of the prowess or reliability of scientists in interpreting evidence and in deductive and
ampliative reasoning. To this extent I would think their study might be of interest more in a sociology
than in an epistemology of the sciences. But there may be exception cases where personal virtues either
descriptively or normatively help fix views about the epistemic status of a theory or hypothesis. If so,
such cases could also likely to be identified by the types of underdetermination worries that dog inquirers
16
in that instance. What, for instance, about the Kuhnian claims that scientists basically share the same list
of theoretical virtues? Is this conservative claim true? Is it true in some sciences, but not in others? And
what if, as Kuhn also claimed, these thick concepts can mean different things to different scientists, can
be seen as applying differently by them to particular to the case, and can also be weighed differently one
scientist from another in inferences to the best explanation? If or when worries such as these arise then
inquiry is taking place under another level or type of underdetermination problem. Our primary focus
becomes underdetermination of theory choice by methodological standards, including the theory virtues.
Let us call this type ampliative underdetermination, and move to consider now the philosophical concerns
it raises and the resources that virtue epistemologies have for responding to them.
Paralleling our treatment of logical underdetermination, I want to say with respect to
ampliative underdetermination that the sheer possibility that ampliative criteria will not result in
a unique choice is not much of a worry, since it really only restates how we got to this point:
ampliative reasoning by definition does not meet deductive standards of entailment; if we cannot
read theories off of their empirical consequences, the notion of an algorithm for theory choice is
off the table and so to re-impose a logicist conception of rationality or objectivity is simply
inappropriate for beings such as we are. Further, just as it is always desirable but not always
possible that theory choice be based on experimental findings that confer empirical adequacy
uniquely upon one theory, so I would hold that it is always desirable but perhaps not always
possible to distinguish sharply between the desiderata of impersonally-framed ‘theory virtues’
and personal intellectual traits of good researchers themselves.
So once again, could the personal bon sens of the scientist ever directly contribute to the
epistemic status of the theory which that scientist chooses? Well, conceivably, but this goes far
beyond the kind of “shift” I agree with Dawid is needed for fields like high energy physics beset
by underdetermination worries. I am not sure what other examples there might be of it. With
17
McMullan and against van Fraassen we have held that ampliative reasoning clearly contributes
to epistemic status. And we have held that personal virtues and vices (probably both intellectual
and ethical) are active and implicated within ampliative reasoning—most obviously in the kind
of weighing that inference to the best explanation demands. But also on the present view, we
must remain wary of “collapsing” the theory virtues into a set of personal virtues of scientists
themselves. My view isn’t shared by all self-described virtue epistemologist, however, and this is
why I bring it up. Since there are a number of different extant versions of virtue epistemology it
is unsurprising to find them running the full gamut of views in relationship to the
underdetermination problem. Some authors neglect the theory virtue/personal virtue distinction
by not recognizing the importance of the researcher’s character and “doings,” while others
collapse the theory virtues into personal virtues, attaching no real importance to impersonally-
framed theory virtues.
Perhaps the clearest example of a view that collapses the distinction between impersonal
theory virtues and personal bon sens is Lynn Holt’s Apprehension: Reason in the Absence of
Rules (2002). Holt’s Apprehensionist virtue epistemology takes “Methodism” as it’s opposite and
as its stalking horse. He contrasts the apprehensive virtues of understanding (nous) and of the
practical wisdom (phronesis) of the qualified expert with “the non-apprehensive elements of
expertise—calculative reasoning, technical skills” (44). But theory virtues don’t fit well within
this dichotomous picture, and indeed are almost entirely left out of Holt’s account of theory
choice. His view of theory assessment and the epistemic status of theories is basically that it is
whatever reflects the judgments of the experts, those who possess the phronesis relevant to their
field.14 This might then be the form of virtue epistemology appealing to someone who takes a
14
One of the most common objections to apprehensionism is that it is circular; another is that it is simply
intuitionism in new garb—apprehension or bon sens as ‘the Emperor’s new intuitions.’ Holt
18
strong stance on holism or on ampliative underdetermination, and sees it as motivating an either-
or choice between Apprehensionism and Methodism. Its result is the very strong claims that “the
way to adjudicate between rival traditions is to ask the wise” (72) and that “a genuine
epistemology ought properly to be regarded as virtuoso epistemology: an account of who is best
able to judge truth from falsity in virtue of his or her possession of wisdom” (73).
While he neglects to examine Duhem’s account of theory choice, I take it that Holt’s
central distinction between apprehensive and non-apprehensive expertise and the personal traits
that constitutive the former strongly overlaps with Duhem’s distinction between “intuitive” and
“mathematical” reasoning. Not surprisingly then, some neo-Aristotelian virtue epistemologists
find substantial interest in Duhem’s account of bon sens. David Stump’s paper “Pierre Duhem’s
Virtue Epistemology” offers a virtue-theoretic account of the role of bon sens in Duhem’s
philosophy of science. Stump argues that despite the fact that Duhem is sometimes read as a
conventionalist arguing that there is simply no cognitive way to decide between empirically
equivalent theories, closer examination reveals that through good sense of practitioners
consensus typically does emerge, and not for purely conventional or epistemically irrelevant
reasons.15
acknowledges these criticisms of his position by Hintikka (2002), who argues that intuitionism is a failed
view in the philosophy of science, and “apprehensionism” just an unsuccessful re-working of
intuitionism. Holt’s Chapter 3, “Apprehension and the Apprehensive Virtues”, offers his direct reply.
15
Stump (2007), 149-159, 149-150. The personal habits that comprise the scientist’s methodological
cognitive character describe real or ideal excellences of inquirers, not of theories or hypotheses per se.
Perhaps for this very reason, they are less purely intellectual, and indeed those who emphasize their role
in inquiry, from Pierre Duhem to Daston and Galison, often want to insist that they are or include
character traits in the full Aristotelian sense, engaging motivations and sometimes crossing boundaries
between the epistemic and the ethical. See Stump (2011) and the case studies of scientific reasoning in
this collection for further developments of this view.
19
Whether rightly or not, Milena Ivanova seems to associate Stump’s reading of Duhem
with an apprehensionist variety of virtue epistemology. She associates it with a very strong
“change in the direction of analysis” thesis that she takes all form of virtue epistemology to be
committed to, where the merits of the agent’s character determine the epistemic standing of
particular beliefs.16 Although I don’t think that thesis is held by reliabilist or ‘mixed’ forms of
virtue epistemology, and don’t see Stump himself as making all the strong apprehensionist
claims she seeks to refute, given what we have said above I do agree with much of her criticisms
of Apprehensionist virtue epistemologies. She objects that it seems to negate the need for
scientists to look for future evidence to evidentially distinguish the theory chosen by good sense.
If this were correct it is easy to see why calling Duhemian good sense a virtue theoretic solution
to underdetermination will arouse suspicions. As Fairweather puts it, it will be controversial to
locate some share of the epistemic value of our currently accepted scientific theories in
properties of the scientist, rather in properties of the science itself” (2012, 140). Even
independently of whether Stump is right to see Duhem’s account as proto-virtue epistemological,
Ivanova finds Stump’s own views about the epistemic value of good sense unsatisfactory. She
also argues for a different reading of Duhem’s account of good sense. I will not have space to go
very far into her or Fairweather’s interesting responses to Stump, but would like to draw out a
few general points and to try to straighten out what I see as some misconceptions in the debate.
Ivanova writes that “[Duhemian] good sense does not determine the construction of a
theory and it is not what justifies a belief in the truth of a theory. It determines the scientist’s
choice, but not uniquely. It restricts the scientist’s choice by excluding some of the
16
In fact it represents only one strong form of virtue epistemology that would define justified true belief
in terms of what an intellectually virtuous person would believe. Stump (2011) does come close to
endorsing such a view. The ethical analogue is called Qualified Agent or Pure Virtue Theory.
20
possibilities with which he is faced in theory choice. It does not lead to justified true belief,
but simply to a temporary acceptance of a theory” (62). While I can largely agree, I have
some non-trivial concerns. Developing a distinction Fairweather suggests, I would say that
virtue epistemology helps us recognize the contribution to epistemic value of the scientist’s
methodological cognitive character, but that it leaves open a range of views about the
epistemic status of the theory that good sense selects. But there is also something correct in
Fairweather’s point that “Method and evidence reign when they can, but epistemic
normativity becomes aretaic in UD inquiry with the express purpose of resolving
underdetermination” (141). In my own terms, the question of how to parse the differences
between empirical testing, theory virtues, and the personal bon sens of researchers
themselves admits of no general answer; rightly seen, the question depends crucially upon
local issues about the relative normality of inquiry being pursued under conditions of
underdetermination.
The success condition of Duhemian good sense seems merely to be its breaking of
the empirical stalemate in an appropriate way, not in a way that necessarily confers
uniqueness (across competent scientists or the epistemic community as a whole) on the
choice made. The work of ampliative reasoning addresses underdetermination by
eliminating some alternatives, and this is promotive of objectivity and especially helpful
with when a choice is between extant competitors, even if it can’t identify and so can’t
eliminate every possible theoretical explanation that saves the phenomena, save one.
Uniqueness as tied to this latter ideal is no measure of reasonableness of choice, unless
logicism be assumed. The close connection Ivanova makes between uniqueness and
epistemic merit or status arguably only arises under assumption of too strong a claim about
21
ampliative underdetermination worries in the sciences. While respecting the distinction
between theory pursuit and acceptance or belief, I also think this distinction should not be
made too rigid, like the psychology/logic or discovery/justification dichotomies on which
logicist metascience depended. A virtue-theoretic account better ties theory and practice,
bringing what Fairweather calls “axiological continuity” between heuristics and assessment,
scientists’ ‘doings’ and their stated values. It brings continuity to the movements to and from
UD (underdetermination) and non-UD inquiry:
In UD inquiry we are trying to resolve the problem of theory choice, whereas in
non-UD inquiry we either have not yet faced the problem, or have resolved it for the
time being. The virtues of good sense do not have a constitutive role in generating the
epistemic standing of theories in non-UD inquiry…The virtue theoretic reading exhibits
axiological continuity between the two contexts of inquiry and thus provides a
constraint on admissible resolutions to underdetermination by precluding the
introduction of radically new epistemic values (141).
A further and related concern with Ivanova’s account is about her own claims and
assumptions about non-uniqueness. In her discussions of theory virtues, she jumps too quickly
from the possibility of ampliative underdetermination or non-uniqueness to the generalization
that a unique choice is never indicted by the criteria of ampliative adequacy. “Even though
criteria to describe theory choice can be found, they cannot determine the choice uniquely” (60);
“These criteria can help us to describe, explain and justify the scientist’s decision, but they do
not do so uniquely” (63). Cannot? Do not? This was precisely Laudan’s point, that such a move
relies upon too great a generalization. The mere possibility of differential weighting applied to
the theory virtues, etc., does not mean that judgments based upon them must differ so
22
significantly; it would be wrong to presuppose that every time two theories are empirically
equivalent (i.e., logical underdetermination prevails) there must also be underdetermination of
theory choice by methodological standards (i.e., ampliative underdetermination) (Laudan 1990a).
Both the local nature of ampliative underdetermination and the very real possibility that
ampliative reasoning through internal, contextual, and diachronic theory virtues does sometimes
indicate a unique preference among two or more extant competitors, should serve to deflect this
kind of claim. If we are not implicitly identifying a unique choice with a logically or evidentially
forced choice, the claims Ivanova repeatedly makes that virtue epistemological treatments must
be “unsatisfactory” because they fail of “solve” the underdetermination problem but only move it
to a new level, lose all motivation. They cannot lead to the bogey of the global non-rationality of
theory-choice. “Solving” underdetermination problems was never really in the cards, and one
who bases “satisfactoriness” of resolutions on that measure will always be disappointed. The
sense of uniqueness that should be in play is that of consensus within a scientific community, not
that of “conclusiveness” as she uses it. It is quite possible are arguably also common that a
consensus emerges that one of a pair of competing empirically equivalent theories is preferable
on the basis of its ampliative adequacy, though of course any such judgment is fallible. We need
not take forced beliefs as paradigmatic of rationality or objectivity in philosophy of science any
more than with other practices. So Ivanova’s demand for uniqueness and her claim that
epistemic virtues fail to provide a satisfactory ‘solution’ in terms of uniqueness, if
premised on this second sense of logical ‘conclusiveness,’ is either off-target or
simple question-begging.. There just is no basis in the alleged failure of ampliative reasoning
to deliver uniqueness either for dichotomizing between pragmatic and epistemic dimensions of
23
theory appraisal, or for insisting that “the only belief involved in acceptance…is the belief that
the theory is empirically adequate” (van Fraassen 88; emphasis added).17
This leads to my final point of criticism.18 I think Ivanova goes too far in the direction of
reading Duhem as holding definitively like van Fraassen that only a later experiment that gives
unique advantage in empirical adequacy provides grounds for justified belief or acceptance. This
returns us to McMullin claim that debate over the deductive and ampliative reasoning in the
sciences “usually masks a deeper difference about the epistemic function of theory itself” (507).
As he explains it, the truth of a theory entails the reality of the causes it postulates, and so
regrettably the issue of realism vs. instrumentalism is linked with the logic of theory-
assessment.19
17
To an extent van Fraassen can willingly concede that theory acceptance has a pragmatic dimension, and
that “It is a mistake to think that the terms in which a scientific theory is appraised are purely hygienic,
and have nothing to do with the other sort of appraisal, or with the persons and circumstances involved.”
But what is given with one hand is taken by the other, as van Fraasen nevertheless wants to bundle the
theory virtues as “pragmatic” instead of properly scientific, inserting a sharp contrast between virtues that
do, and that do not “concern the relation between the theory and world.” Given McMullin’s articulation of
theory virtues, below, I do not think this characterization of or strategy for dealing with ampliative
reasoning in science can be maintained: The super-empirical virtues McMullin discusses are not merely
“a function of our interests and pleasures” as van Fraassen defines the pragmatic, nor are they properly all
characterized as providing only “reasons for using a theory, or contemplating it, whether or not we think it
true” (1980, p. 87 and 96, emphasis added).
18
On my view Ivanova presents an obvious straw-man version of McMullin’s stance in numbering him
among those who supposedly “believe that we can always choose a unique theory from a set of
empirically equivalent rivals by simply pointing to the amount of virtues the chosen theory possesses”.
19
“Those who deny the ability of theory to reveal underlying structure will also tend to see empirical fit as
the only feature of theory worth worrying about, with possible pragmatic concession for such features as
lend themselves to convenience of use or utility of application…[whereas] those who see in theory the
way to discover real underlying causes of macroscopic regularities are likely to stress a variety of
24
On instrumentalist assumptions where the matter of the underlying causes is left in
abeyance, theory assessment is more likely to be treated as a fairly straightforward affair in
which explanatory success and longitudinal concerns like how competing theories develop over
time, to what extent responses to anomaly appear ad hoc, etc. are of lesser importance than they
are to the realist. Empirical fit and explanatory success tend to be separable issues for the realist,
while an instrumentalist or constructive empiricist might deny any real distinction. A strong
realist theory, in turn, might hold that “the global excellence of theory is the ultimate measure of
truth and ontology at all levels of cognition” (Paul Churchland, 1985). While it remains open for
particular theorists to involve ampliative desiderata in an argument for or against scientific
realism, I do not think that proper recognition of ampliative reasoning in science in commits one
to a position like Churchland’s. But if his is one extreme exhibiting the tension that McMullin
speaks of, another is the constructive empiricist’s resistance to allowing even impersonal theory
virtues any real connections with theory choice or assessment. Although too much or too little
deference to underdetermination problems or to ampliative reasoning in science should be
pointed out and criticized, a virtue theory that allows for a range of views on the realism/anti-
realism spectrum offers a chance to help mediate this debate or to significantly recast it debate.
This is the primary position into which I would like to place inquiry-focused virtue
epistemology.
epistemic virtues and to insist that saving the phenomena is not enough” (McMullin 1996, 17). Indeed
McMullin sees this as extending to the work of personal traits of scientists themselves: “Theory
assessment involves the faculty of good judgment (bon sens) which permits disagreement between
competent scientists.…What tends to decide the issue between competing theories is how they develop
over time, to what extent their response to anomaly appears ad hoc, and so forth” (17; compare Duhem
1954, 216-218).
25
To return to Duhem, his account of bon sens in the resolution of local underdetermination
is not well-developed, but he was always walking a thin blue line between empiricist
conventionalism and realism, and my reading of him has him walking a similar line here. It is
true that as Ivanova points out, Duhem did in one passage characterize ampliative criteria or at
least the criteria of bon sens as “essentially subjective, contingent, and variable with time, with
schools, and with persons” (1954, 288). But I think she leans too hard on this passage as
definitive of Duhem’s stance on ampliative reasoning in science, and as raising the bogey of
historicism-cum-relativism. Firstly it would cast doubt on Duhem, since the claim made seems to
be an obvious over-generalization. Secondly, it is not difficult to find counterpoint passages in
which Duhem articulates the eliminative abductive reasonings of scientists that eventually bring
resolution to the situation of underdetermination where direct testing and hypothetic-deductive
reasoning cannot. For instance Duhem writes, “Pure logic is not the only rule for our judgments:
certain opinions which do not fall under the hammer of the principle of contradiction are in any
case perfectly unreasonable” (1906, 217). Perhaps the difference is that Duhem, as we see,
distinguishes between reasonable choice and logically forced choice, while the constructive
empiricist’s treatment of the relationship between evidence and theory appears to resist any such
distinction.
5 Rational reconstructionism meets normative naturalism
So what implications for the epistemology and methodology of the sciences might follow from
our present mode of bridge-building? We haven’t the space to develop detailed answers to this
question, but I make three suggestions here from the perspective of “inquiry focused” VE,
recognizing that proponents of other versions might draw somewhat different implications. The
26
first is the one we began with, the idea of a more varied or “dappled” conception of the
relationship between the sciences, which it can be argued is an implication of our thesis of the
localized nature of the most worrisome kinds of underdetermination. I purposely alluded to a
term from Nancy Cartwright’s A Dappled World (1999) to describe this thesis, in part because I
anticipate substantial lines of support can be developed between virtue epistemology and the
thesis of metascientific pluralism she argues for. Pluralism as a metascientific level thesis
presents an alternative to both the “unity of method” that Hempelian logical empiricists
demanded, and to the epistemological relativism of some of the post-positivists.20 Indeed the
connection with a theory of epistemic virtues has already been made from the other direction by
the editors of a notable collection, Scientific Pluralism (Kellert, Longino and Waters eds., 2006),
when they explain,
Philosophers of science have begun to advance pluralism at the metascientific level,
most notably with respect to epistemic virtues. A variety of views regarding the role,
status, and identity of scientific or epistemic virtues has been advanced in the
philosophical literature…[some pluralists claim] that which virtues should hold what
degree of regulative status in any given research project is a function of features specific
to the problem and of the particular aims of the research (2006, x).
My second implication is a quite different way of approaching questions of demarcation
and of the relationship between disciplines or fields of research. This is the view of John Dupre,
who more explicitly than other pluralists has suggested that “we try to replace the kind of
epistemology that unites pure descriptivism and scientistic apologetics with something more like
20
Moderate historicism, according to which the ‘units of selection’ in theoretical enterprises of all types
are historical research programs, and a moderate confirmation holism, seem well-suited to provide this
kind of pluralism, but more radical versions of historicism and holism do not.
27
a virtue epistemology.” If “no strong version of scientific unity of the kind advocated by classical
reductionists can be sustained” (1993, 242), then “the successor to the quest for demarcation
criteria between science and non-science may be an account of theory virtues that characterize
scientific reasoning…[W]e are much better off to think in terms of epistemic virtues, features of
an investigative practice that confer credibility. No doubt the cardinal empirical virtue is a proper
connection with empirical evidence, which is the large grain of truth in the criterion of
falsificationism” (2010). The implication is to abandon of the hierarchy of the sciences ideal, but
without invoking any rigid egalitarianism either: “Many plausible epistemic virtues will be
exemplified as much by practices not traditionally included within science as by paradigmatic
scientific disciplines…No sharp distinction between science and lesser forms of knowledge
production can survive this re-conception of epistemic merit. It might fairly be said, if
paradoxically, that with the disunity of science comes a kind of unity of knowledge” (1993, 243).
These two purported bridges between philosophy of science and virtue epistemology are
not very original, I am afraid, having been drawn by others. But I would like to end by
developing something that I think isn’t already found in the literature, which is a relationship of
mutual support between virtue theory — often described as “Janus-faced” — and normative
naturalism. This also ties in nicely with our previous discussion of the debate over Duhem’s
bon sens, where Ivanova appears to assume that if the personal virtues of good sense do any
serious lifting, it must be in terms of aiding a project of rational reconstruction; if they cannot
serve this role, they are lacking in epistemic significance. But firstly, rational reconstruction
appears far from the Duhemian meaning of good sense, and Duhem’s contemporary interpreters
should not steep him in the assumptions of the positivist era. Secondly and more to the point,
normative naturalists like Laudan and myself explicitly reject the project of rational
28
reconstructionism as a central task of the epistemology of the sciences: “The requirement of
rational reconstructibility is neither wanted nor needed” (1987, 21). Theory virtues and the
personal traits of good inquirers shouldn’t be sharply contrasted because they are not sharply
divided in scientific practices. But the epistemological contributions of axiological and
evaluative thick concepts to the logic of the sciences needn’t be viewed as tied to any traditional
rational reconstructionist project.
Positivists and many post-positivists alike misconstrue the underdetermination problem
because they either mistakenly assume that theories possessing the same positive instances must
be regarded as equally-well confirmed, or because “they assume that the only rational basis for
rejecting [dismissing] a theory or hypothesis is if it has been definitely refuted.” Epistemic
values and virtues (scientific axiology) may change somewhat as science develops, but we are
still able to view rules possessing normative force as grounded in factual means-end relations.21
Methodological rules are fixed by means-end relations, but our conception of ends—scientific
axiology—is neither given nor timeless. We need also an axiology of inquiry whose function is
to certify or decertify certain aims as legitimate, for “methodology gets nowhere without
axiology.” Against rational reconstructionism Laudan proposes normative naturalism:
[E]pistemology can both discharge its traditional normative role and nonetheless claim a
21
Methodology so conceived is basically “restricted to the study of means and ends,”; they are “best
understood as relativized to a particular aim” and judged by whether they guide inquiry to its
achievement. But far from the Quinean version of epistemology naturalized qua replacement thesis for
normative epistemology, Laudan holds that “methodology gets nowhere without axiology,” and that “We
thus need to supplement methodology” with an investigation into an axiology of inquiry (1987, 29).
Axiology in turn is multi-faceted, and while generally naturalistic also “preserves an important critical
and prescriptive role for the philosopher of science” (29).
29
sensitivity to empirical evidence ... normative naturalists hold that the best methods for
inquiry are those which produce the most impressive results ... the naturalist uses the
simple method of induction to ‘bootstrap’ his way to more subtle and demanding rules
of evaluation which, in their turn, become the license for subsequent and yet more
highly refined rules and standards …. (1990b, 44, 58).
For the normative naturalist, as another of its proponents puts it, “there have got to be other
criteria, coherence, simplicity, predictive fertility, explanatory power, that an epistemology, like a
scientific theory, must meet, and it must meet them, not because they are intrinsic goals of
science, but because they are instrumental ones, instrumental to the goal of attaining
knowledge.”22 Thus responsibilist and reliabilist concerns combine in the present view, which fits
better the Janus-faced understanding of virtue theory as both a descriptive account and one
aiming to provide prescriptions for the improvement of practice.23
In this more naturalistic alternative to the overt or ‘closet’ intuitionism of the rational
reconstructionists, we need not preoccupy ourselves with the question of whether we can always
replicate the choices of past scientists as rational. This is no grand mark of the adequacy of a
methodology of science anyway. The normative naturalist will instead simply “inquire about
which methods have promoted, or failed to promote, which sorts of cognitive ends.” History of
science still plays a key role here, and indeed may be center stage in the evaluation of proposed
methodological standards. But the one major difference in my account of normative naturalism
22
Rosenberg 1990, 42-43.
23
Sankey (1996) nicely points out overlaps but also important differences between Laudan’s normative
naturalism and reliabilist epistemology. Normative naturalism is not committed to truth and realism in a
way that would be objectionable to instrumentalists.
30
from Laudan’s development of it is that I hold that a repudiation of logicist rational
reconstructionism must also entail a repudiation of the dichotomy between internal and external
history of science that he and Imre Lakatos each appealed to. This is but another version of the
dichotomy between the rational and the social that an appropriately thickened conception of
epistemology should have the effect of sweeping away. As I have argued further elsewhere,
normative naturalism should be seen as demanding that we disassemble (along with the myth of
universally valid methodological and epistemological standards), the rational-social dichotomy,
the logic/psychology dichotomy, and the dichotomy between internal and external history of
science, as false crutches that rational reconstructionists appeal to in trying to fulfill what is
really the same old untenable project of insulating epistemic purity from practice.24
Dichotomies like those between the rational and the social have much staying power,
however. One main reason for this is that because underdetermination problems leave us with
questions about how scientific theories are chosen when empirical evidence fails to determine
one theory as uniquely choice worthy, they also appear to present us with a referendum on the
rationality of science. They do not. Only a philosophy of science in the service of rational
reconstructionism, or a radical historicism that uncritically assumes the same dichotomies in
order to take the opposite, relativistic side of the issue, pushes us towards any such referendum.
So for instance we hear that to ground theory choice in anything else but hard data impugns the
objectivity of the theory chosen, and of science itself; we must perforce seek sociological
explanations of scientists’ cognitive choices! Or we hear that if assumptions of some sort are
required to mediate the relation between data and hypothesis, these assumptions “can be the
24
See Axtell 2012, which updates my older (post-dissertation) but fuller critique of the “objective
historicism” of Laudan and Lakatos.
31
vehicles on which cultural ideology or social values ride ‘right into’ the rest of science”.25 Or
again we hear that because appeal to theory virtues are a means of “persuasion” between
advocates of different empirically equivalent systems, these concepts reveal a rhetorical
dimension that should be entirely foreign to science were it really objective.
But the correct response to each of these claims, it seems to me, is to reject from the
outset the notion that scientific practices should be characterized by a sort of epistemic purity
that social practices of other sorts lack. Yes, we want our epistemic values to push others out of
our standards of theory assessment, but rational reconstruction as the received view demands it is
not needed to defend the rationality of science generally, or of particular episodes.
To try to mark such a distinction between the rational and the social or between internal and
external questions, where there is none, represents another remnant of positivist dogma.26
Rhetorical strategies of persuasion, like underdetermination worries, are ever-present in some
areas of research yet rare in others, but their absence or presence has never neatly divided
scientific from other forms of inquiry. To the extent that we take the advice of Longino and
others to “disassemble the rational-social divide” we will subject our cognitive values to
continual social and genealogical criticism; but the global referendum notion around which so
much discussion of holism and incommensurability have been situated should fade further away,
while local problems where guidance is genuinely needed, and thick descriptions or case studies
25
Longino, 1990.
26
See Zammito 2012, “The ‘Last Dogma’ of Positivism: Historicist Naturalism and the Fact/Value
Dichotomy,” Journal of the Philosophy of History 6 (2012) 305–338. Owen Flanagan goes further when
he writes, “[I]t seems simply stipulative to suggest the values that guide science and normative
epistemology, as we know them, are and problematically ‘ulterior ends’-- some sort of ‘descriptive
terminal parameter’ specified by nature …. We do science and epistemology in ways that show every sign
of being driven by socially specific values.” (1988, 549).
32
of ethical and intellectual virtues at work in scientific practice should come more clearly into
focus. I would hope in this paper to clear some conceptual ground for detailed case studies of this
kind.
To conclude, then, rather than setting the rational and the social, or again epistemology
and history, or discovery and justification against one another as the legacy of logicism and its
project of rational reconstructionism requires, I argue that a virtue epistemology tied to
normative naturalism puts us on a different and more advantageous path. Ernan McMullin hoped
to mediate the realism/instrumentalism issue, and also urged that “We may still be able to
construct a philosophy of science that derives both from the learning that has gone on in history
and from a more general logical and epistemological framework” (1984, 57). While the present
account of the contributions of virtue theory to the philosophy and methodology of the sciences
shares this hope and certainly finds it consistent with preserving important “critical roles” for
philosophers of science (Laudan 1987, 29), I also maintain that to help drive its realization virtue
epistemology must first be placed in the service of more naturalistic and practice-focused
conceptions of scientific meta-methodology, and therefore more clearly disassociated from the
dogmas of traditional rational reconstructionism.
Acknowledgements
Special thanks go out to Abrol Fairweather for comments on an earlier draft, as well as to Lynn
33
Holt, Milena Ivanova, James Kidd and David Stump for comments and discussion on related
2010-2011 posts at JanusBlog: The Virtue Theory Discussion Forum,
http://janusblog.squarespace.com.
34
References
Allan, P.L. (2006). Ernan McMullin and critical realism in the science-theology dialogue.
Ashgate.
Axtell, G. (2012). The dialectics of objectivity,” Journal of the Philosophy of History, 6,
339–368.
Axtell, G. and Carter, J.A. (2008). Just the right thickness: A defense of second-wave virtue
epistemologies, Philosophical Papers (topical edition on Epistemology through thick
and thin), 37(3), 413-434.
Churchland, P. (1986). Scientific realism and the plasticity of mind. Cambridge University
Press.
Dawid, R. (2011). Theory assessment and final theory claim in string theory, Foundations of
Physics (Online First). DOI 10.1007/s10701-011-9592-x.
Daston, L. and Galison, P. (2007). Objectivity. New York: Zone Books.
Duhem, P. (1954[1906]). The aim and structure of physical theory. (P. P. Wiener, Trans.)
Princeton: Princeton University Press.
Duhem, P. (1991[1915]). German science: Some reflections on German science: German science
and German virtues (John Lyon, Trans.). La Salle, IL: Open Court.
Dupre, J. (2010). The disunity of science. Interview with P. Newall at http://www.galilean-
library.org/site/index.php/topic/3302-john-dupre-the-disunity-of-science/
Dupre, J. (1993). The Disunity of Things. Harvard University Press.
35
Fairweather, A. (2012). The epistemic value of good sense. Studies in the History and Philosophy
of Science 43, 139-146.
Flanagan, O. “Pragmatism, Ethics, and Correspondence Truth: Response to Gibson and Quine,” Ethics
98 (3): 541-549 (1988).
Hintikka, J. (2002).The emperor’s new intuitions, The Journal of Philosophy 96(3): 124-47.
Holt, L. (2002). Apprehension: Reason in the absence of rules. Farnham, UK: Ashgate.
Ivanova M. (2011). Good sense in context: A response to Kidd, Studies in the History and
Philosophy of Science, 42, 610-612.
Ivanova M. (2010). Pierre Duhem’s good sense as a guide to theory choice, Studies in the
History and Philosophy of Science, 41, 58-64.
Kellert, H., Longino, H. and Waters, C.K. (2006). Scientific Pluralism. University of Minnesota
Press.
Kuhn, T. (1977). Objectivity, value judgment, and theory choice. In T. Kuhn The Essential
Tension. University of Chicago Press: 320-353.
Laudan, L. (1996). Sins of the fathers...: Positivist origins of post-positivist relativisms, in
Laudan, Beyond Relativism. Westview Press: 77-87.
Laudan, L. (1990a). Demystifying underdetermination. In C. W. Savage, ed., Scientific theories.
Minnesota Studies in the Philosophy of Science, 14, 267–297. Minneapolis: University of
Minnesota Press.
Laudan, L. (1990b). Normative naturalism. Philosophy of Science 57, 44-59.
Laudan, L. (1987). Progress or rationality? The prospects for normative naturalism.” American
Philosophical Quarterly 24(1), 19-31.
Longino, H. (1990). The fate of knowledge. Princeton University Press.
Massey, J. (2011). Quine and Duhem on holistic hypothesis testing, American Philosophical
36
Quarterly, 48(3), 239-265.
McMullin, E. (2009). The virtues of a good theory. In M. Curd and S. Psillos eds., The Routledge
companion to philosophy of science. Routledge.
McMullin, E. (1995). Underdetermination, Journal of Medicine and Philosophy 20 (3), 233-252.
McMullin, E. (1984). The goals of natural science, Proceedings and Addresses of the American
Philosophical Association, 58(1), 37-64.
McMullin, E. (1996). Epistemic virtue and theory appraisal. In I. Douven and L. Horsten (eds.),
Realism in the Sciences, Leuven: Leuven University Press, pp. 13–34.
Rosenberg, A. (2012). Philosophy of science, 3rd ed. Routledge.
Sankey, H. (1996). Normative naturalism and the challenge of relativism, International Studies
in the Philosophy of Science 10(1), 37-51.
Stump D. (2011). The scientist as impartial judge: moral values in Duhem’s philosophy of
science. New perspectives on Pierre Duhem’s The aim and structure of physical theory
(book sympisium), Metascience, Vol. 20, pp. 1–25.
Stump, D. (2007). Pierre Duhem’s virtue epistemology, Studies in History and Philosophy of
Science, 38, 149-159.
Van Fraassen, B.C. (1980). The Scientific Image. Oxford: Oxford University Press.
37
Forthcoming in Well Founded Belief: New Essays on the Epistemic Basing Relation,
Patrick Bondy and J. Adam Carter (eds.) Routledge, 2019.
Well-Founded Belief and the Contingencies of Epistemic Location
Guy Axtell
Abstract. A growing number of philosophers are concerned with the epistemic
status of culturally nurtured beliefs, beliefs found especially in domains of
morals, politics, philosophy, and religion. Plausibly, worries about the deep
impact of cultural contingencies on beliefs in these domains of controversial
views is a question about well-foundedness: Does it defeat well-foundedness if
the agent is rationally convinced that she would take her own reasons for
belief as insufficiently well-founded, or would take her own belief as biased,
had she been nurtured in a different psychographic community? This chapter
will examine the proper scope and force of this epistemic location problem. It
sketches an account of well and ill-founded nurtured belief based upon the
many markers of doxastic strategies exhibiting low to high degrees of
inductive risk: the moral and epistemic risk of ‘getting it wrong’ in an
inductive context of inquiry.
1 Epistemic Location and the Epistemology of Nurtured Beliefs
Many of the beliefs that people hold dearest to their sense of personal and social
identity are beliefs profoundly shaped by their own upbringing. The impact of a person’s
1
place in time, their familial, communal, cultural, and geographic/demographic setting are
some of these contributory causes, but to a varying degree an agent’s controversial view
might be highly conditioned by personal temperamental factors or other pragmatic constraints
on an agent’s beliefs.
We will follow J. Adam Carter (2018) and others who use the term controversial
views to refer not just to one’s religious or irreligious views, but also our substantial views in
the domains (at least) of morals, politics, and philosophy.1 Nurtured beliefs/opinions are not a
‘domain,’ but the best examples of contrariety among culturally nurtured beliefs fall within
these domains. Let a person’s epistemic location refer us to how the individual is located
demographically (family; broader culture; class, etc.), in addition to geographically and
historically. We will use the term epistemic location problem to highlight etiological
challenges to controversial views that bear marks of contingency and of what John K. Davis
(2009) terms the impact of trait-dependence upon our cognitive judgments.2 In its most
general sense, the epistemic location problem is the problem that our obvious psychographic
differences – differences in such things as attitudes, values, and most importantly for this
study, what John Rawls terms comprehensive conceptions of the good – are strongly
conditioned by contingent matters of the individual’s historic, geographic, and demographic
location.
Religious identity and whatever beliefs one has stemming from a testimonial faith
tradition are a prime example of this familial or cultural inheritance, and the apparent
contingency of such nurtured beliefs. Although certainly not the only example, the
proximate causes of one’s religious identity and the formation of attendant beliefs are,
for most people, a matter of their epistemic location, which in turn appears to be an
accident of birth.3 Michel de Montaigne gave a version of a ‘contingency’ or epistemic
location argument when he wrote,
2
[W]e receive our religion in our own way and by our own hands, and no
differently from the way other religions are received. We happen to find
ourselves in the country where it has been practiced; or we value its antiquity or
the people who have supported it; or we fear the threats it attaches to wrongdoers,
or we follow its promises… By the same means another country, other witnesses,
similar promises and threats, could in the same way imprint in us a contrary
belief.4
Perhaps motivated by a resurgence of social epistemology and epistemology of
testimony, and by the increasing importance of understanding group dynamics and the causes
of psychographic diversity, philosophers have become increasingly concerned with the
epistemology of culturally nurtured beliefs. So there is much agreement that to improve the
epistemology of controversial views, philosophers need to focus more carefully on the right
target, proper scope, and epistemological force of arguments of the sort that Montaigne and
Mill share. Davis’s work takes steps in this direction, as he has plausibly argued (2009) that
the concern about the contingency of so many of our testimony-dependent beliefs involves a
question about the basing relationship: Does it defeat proper basing if the agent is rationally
convinced that she very likely would see her own actual nurtured belief as both false and
tainted by unrecognized bias, had she been nurtured in a different culture or epistemic
community?
George Sher (2001) and Gerald Cohen (2000a; 2000b) are also often credited with
igniting a more careful philosophical interest in nurtured belief, and with it, arguments from
the contingency of epistemic location. Sher asked us to study “the implications of the fact that
even our most deeply held moral beliefs have been profoundly affected by our upbringing
3
and experience—that if any of us had had a sufficiently different upbringing and set of
experiences, he almost certainly would now have a very different set of moral beliefs and
very different habits of moral judgment.”5 As Sher here indicates, etiological challenges and
the contingency anxiety they arouse in those who take them seriously may attach to a far
broader group than just religious beliefs.6 Cohen similarly describes ‘paradoxes of
conviction,’ paradoxes that he thinks we face in assenting to propositions that we realize, or
should realize, are produced and maintained through cultural influences, such that if one had
a sufficiently different upbringing they would likely hold different beliefs, attested by
different justificatory reasons.
Nathan Ballantyne (2015) has also worked on this “pervasive and disconcerting worry
about intellectual life: our controversial beliefs regarding morals, politics, religion, and
philosophy depend on facts about our personal history.” Yet he is correct to find that in
arguments from contingency both ancient and modern, “pointing is typically all we get—
worked-out arguments based on variability are uncommon.”7 To move beyond this,
Ballantyne connects serious engagement with the problems posed by Montaigne and Mill not
to a generalized scepticism or to dogmatism, but to more concerted attempts to “debunk
biased thinkers (including ourselves).” Joshua DiPaolo and Robert Mark Simpson (2016)
similarly investigate the epistemic location problem not just with cases described from the
armchair, but together with utilization of affects that psychologists are studying, including
contingency anxiety and indoctrination anxiety.8 This is an approach that we will pursue, one
that marshals not just considerations of epistemic luck/risk, but also the now-vast literature on
studies of individual/social biases and heuristics.
Before going further, however, I note that Ballantyne draws two general lessons from
his study of the impact of the epistemic location problem on the epistemology of
controversial domains, including especially nurtured controversial views: “[W]e should hold
4
some of our controversial beliefs with less confidence… [and] we need better methods to
make judgments about biases.”9 This will be a good starting point for us, since I generally
agree with the spirit of both lessons, but will work them out somewhat differently.
I begin by locating the historic, geographic, and demographic contingency of people’s
nurtured controversial views within a much broader set of recognized sources of cognitive
diversity. A person’s epistemic location is one of the most unavoidable of these sources of
contrariety, being basic to the human condition. Arguably the contingencies of belief that
derive from people’s geographic and demographic diversity are closely related to the
evidential ambiguity that affects so many views we hold on morals, politics, philosophy, and
religion. Such contingency makes for no sweeping indictment of a belief’s reasonableness:
this would be to dichotomize between the rational and the social, which our understanding of
the epistemic location problem most expressly should not. Instead I want to develop
Ballantyne’s second lesson, the need for more principled and fine-grained application of bias
studies to well-motivated etiological challenges. In doing so we will have recourse to work
on the differences between benign and malign epistemic luck, to Patrick Bondy and Duncan
Pritchard’s discussion of the close connection between malign luck and epistemic risk, and to
Ian Kidd’s work on the differences between “robust” and merely “rhetorical” vice charging.
This chapter aims to develop several sides of an inductive risk-based account of what
it means to motivate serious etiological challenges and to support them empirically through
markers of bias. More fully, this chapter sketches an inductive risk-based account of
assessments of the doxastic states (belief/alief/credence) and epistemic standings (knowledge,
understanding, subjectively/objectively justified belief, rational belief, etc.). Inductive risk is
the study of the chance or possibility of getting it wrong in an inductive context. The concept
of inductive risk is widely used in science, and the possibility of getting it wrong is there
acknowledged to often raise moral as well as epistemic concerns. If portable beyond
5
philosophy of science, the concept of inductive risk might serve equally well as common-
ground for wide-ranging discussions over doxastic responsibility. Counter-inductive thinking
entails the highest degrees of inductive risk.10
I will relate motivated etiological challenges to well-founded belief as directly related
to its being highly overdetermined by trait-dependent factors. We do indeed need better
methods to make judgments about biases and other temperamental factors which negatively
affect well-foundedness, in terms both of the agents’ creditworthy exercise of a genuine
cognitive ability, and of their personal justification.11 But we do not contend that the
normative upshot of the prevalence of trait-dependent beliefs in the domains of controversial
views is that they are never well-founded. As seems to be the case with Montaigne and Mill,
what normative upshot philosophers should draw is more contextual than that; assessment
cannot be uniform because it crucially depends upon the varying degree of blind spot bias
that actual agent’s exhibit. Agents may exhibit the bias in various ways, through over-
estimation of the epistemic status of their views, through rhetorical and unsupported
asymmetries of explanation, and more defensively through rhetorical peer denial. It is the
dogmatic ways in which nurtured beliefs are sometimes held that motivates a serious de jure
challenge, rather than the amorphous line between properly doxastic and sub-doxastic
attitudes.
This returns us to Ballantyne’s first lesson, a lesson about guidance. Ballantyne like
many others conclude that we should hold “with less confidence” our controversial views.
Conformists make this a universal prescription, moving from epistemic assessment to
guidance. Even Carter’s qualified version of controversial view agnosticism, which opens up
sub-doxastic attitudes such as “suspecting that,” which he rightly points out that Richard
Feldman Triad model of doxastic attitudes neglects, speak in terms of downgrading the
degree of confidence in a proposition being the prescriptive upshot of conditions of genuine
6
peer disagreement. Under conditions that cover a large portion of our beliefs in controversial
subject areas, Carter’s principled controversial view agnostic asserts, “we are rationally
obligated to withhold judgment.”12
Carter argues effectively that “a tacit commitment to the Triad View, with its
deontological categories of belief, suspension of belief, and disbelief, has the effect of
artificially restricting the range of reasonable attitudes we might take up in controversial
areas….” (15). By opening up other sub-doxastic attitudes besides “suspension,” Carter
develops resources for modifying the conformist thesis to make it more ‘liveable,’ while
holding on to its guiding principle.13 Carter’s version of conformism we can term principled
agnosticism about domains of controversial views. While I appreciate Carter’s risk-focused
account of controversial views, principled agnosticism is still impermissivist.
Let me say something about this in order mostly to set it aside. I will return to this in
my conclusion, but I hold that the same detail that the bias-studies and inductive risk
approaches bring to the table to distinguish motivated from unmotivated etiological
challenges, shows how overgeneralized are the prescriptions that conformists and steadfasters
each ask us to accept as the upshot of genuine peer disagreement. Epistemic assessment and
epistemic guidance need to be more carefully distinguished than has been the case in this
debate. These are primarily questions of guidance and thus also of the ethics of belief. On
these questions I have written in defense of permissivism.14 I see no easy path, either from
moral or epistemic evidentialism, to the kind of universal guidance issued either by the equal-
weight view, or by principled agnostics.15 If we should discern a more diverse set of doxastic
attitudes than the Triad model allows us to see, I would argue that we should also discern a
more diverse set of permissible responses to genuine peer disagreement. More specifically,
the importance of the reliable etiology of belief for doxastic justification seems from my
7
pragmatist or inquiry-focused epistemology to cast doubt on why doxastic responsibility and
guidance-prescriptions should take a primarily synchronic form.16
So while I won’t try to provide a fuller account of guidance, I just want to state my
resistance to any and all of the universalized prescriptions on offer from dogmatists,
phenomenological foundationalists, equal-weight conformists, and principled agnostics.
Instead I will agree with Ian Church and Justin Barrett (2016) that “psychological
dynamics… suggest that belief firmness, or a belief’s resilience to revision or relinquishment,
are not the only or best relevant metrics for intellectual humility.”17 This allows that there
may be different levels at which to exhibit epistemic deference, not just at the level of one's
credences but also on various levels of one's reasonings (Pittard 2014).
Pragmatists like Davis, Susan Haack, and Susanna Rinard insist that we distinguish
more carefully between norms for guidance-giving and those for epistemic assessment.18 At
the same time, permissivists like virtue theorists hold that “the gap between the ways in
which we are meant to normatively assess belief and action may not be as wide as has been
thought,"19 and that responsible actions are often not ‘obliged,’ but merely ‘permissible’
actions. An area for merely permissible belief seems missing on the impermissivist view. So
before moving on I just want to leave it in the mind of readers that a pragmatist and
permissivist ethic of belief – in so far as it is a permissivism ‘with teeth’ – may prove itself
more effective than the principled agnostic/evidentialist demand for synchronic
‘downgrading,’ in challenging a dogmatic thinker’s faulted attempts to epistemically
privilege their own or their ingroup’s nurtured beliefs.20 For if epistemologists, as I believe,
have useful guidance to give agents in regard to the limits of reasonable disagreement, that
guidance should not assume an ideal or atemporal agent without pragmatic interests or
constraints. It should be guidance consistent with Montaigne’s two points: 1) that “we are all
of the common herd,” a thought commensurable with psychological studies of biases and
8
heuristics, and 2) that due to our directional thinking especially in matters we care a great
deal about, we are constantly guilty of confusing affectively-conditioned commitments with
possessing truth, warrant, proper basing, and right epistemic motives.21
2 Environmental Luck-based Etiological Challenges: A Tess Case
This section aims to elaborate the epistemic significance of the distinction between
contexts of inquiry dependent only on benign evidential luck, and contexts of inquiry
impacted by malign environmental epistemic luck. Pritchard’s post-2005 splitting of veritic
luck into environmental and intervening types has strong implications for the epistemology of
nurtured beliefs. So might Bondy and Pritchard’s recent identification of propositional luck
as a malign form of epistemic luck in addition to veritic luck, but given our limited space I
must pass over discussion of propositional luck.22 What these authors’ do that more directly
concerns us is to translate questions about epistemic luck into questions about the modal
riskiness of a belief-forming cognitive strategy. 23 With this in mind, what I want to do is to
discuss the importance for well-founded belief of the distinction between evidential (as
benign) luck and environmental (as malign) luck. Can we always correctly distinguish them
when presented with a case, and if so, how?
Note first that intervening and environmental luck, while subtly different, are both
forms of what epistemologists refer to as veritic luck. In cases of veritic luck, it’s a matter of
luck if the belief one holds is true, –viz., one very easily could have believed incorrectly. The
intervening form is the form of luck that we find in standard Gettier cases such as the famous
sheep-in-the-field case. Environmental luck by contrast is the kind that we find in barn facade
cases. Environmental epistemic luck, understood as veritic luck and hence as distinct from
9
simple evidential luck, is not compatible with knowledge, most epistemologists hold. If they
are correct in this, it is because Barney’s belief, considered modally, appears to be unsafe.
By contrast, Pritchard’s taxonomy of forms of epistemic luck recognized several benign
kinds. I will treat only evidential luck, since it is the complicated relationship between malign
environmental luck and benign evidential luck that I want to get at. Simple evidential luck
does not violate the safety principle; it is the luck of being situated in a way that others might
not be to have supporting evidence for a true belief. Ernest Sosa uses the simple paradigm
example of coming to hold the true belief that there is a crow in the yard, but only because
one happened to glance out the window at that particular moment it flew by. By contrast,
environmental luck does violate the safety principle. It is the luck that one’s belief is true,
given a set of modal or other epistemic circumstances that are inhospitable to the reliability
of the utilized doxastic strategy (mode of belief-uptake). What is importantly different
between intervening (Gettier) and environmental luck cases is that in the former it is no
matter of ability or competence or achievement that a true belief is acquired, whereas in
environmental luck cases the agent’s beliefs are the product of the exercise of a cognitive
ability that in more cooperative epistemic circumstances might provide more positive
epistemic status to their beliefs. The concepts of luck and risk helps us analyse how agents
achieve or fall short of more valuable epistemic states or standings –rationality, personal
justification, knowledge, understanding, etc. But they may not apply in quite the same way
across domains of controversial views.24
With that much said, I now want to argue that it is not difficult to construct
testimonial environmental luck cases, cases in which our intuitions about epistemic
status basically parallel those that people report about Barney cases, where the agent’s
visual perception is the primary source of the target belief. The predominance of visual
perception cases in epistemology is partly due to their relative simplicity, but partly
10
also to many decades where methodological individualism was assumed. If so,
testimonial cases allow the philosophy of luck/risk to better engage contemporary
social epistemology. Here is such a case.
The Basic Tess Case
Imagine Tess, a good friend of Barney, travelling to visit relatives in Land of Lakes
County. In the base case, this is Tess’s first visit, and she does not know that many
others refer to this county as ‘Fake News County.’ Scattered about on corners of the
town and the whole county are brightly-coloured metal or plastic, free publication
newsstands, each advertising its wares in its small front window. Sometimes there were
several such boxes at the same corner, but most often just one. Tess, who knew none of
this, is met at the train station by her uncle Sal, and before they get to his ride they pass
a corner outside the station with a blue metal newsstand. Tess had just asked her uncle
a question about the history of the county, and Sal goes to the box and gets them each a
copy. “Blue-box publications. Yes, this one you can trust!” says Sal, and to emphasize
his point he flips the paper over and taps the large printed warning on its back page:
“Remember, trust only the news from this box! All of the other boxes contain fake
news.”
Tess finds this a bit quizzical, in part because she has not encountered other
boxes; but they have much to talk about, and the conversation quickly takes another
direction. But that night when she retires to the guest bedroom, she finds the paper on
her dresser, and reads it in bed. It contains many tales about the county and its
founding citizens that Tess finds quite moving and even profound. Although there were
seemingly fantastical elements to these stories, and some of them drew strong moral
11
lessons that clearly went beyond factual information, Tess remembers her uncle’s
assurance of the paper’s trustworthiness, and she accepts the content of the paper
pretty much at face value.
While being driven back to the station after her pleasant weekend visit, Tess
notices for the first time a different coloured newsstand, then another, then another.
Indeed walking into the station she comes upon a veritable array of such boxes in a
row. Having been so enamoured of the first, Tess starts walking up to a yellow plastic
one to get an issue of it for some reading on the way home. Picking it up, she is
surprised to see the same strong warning against trusting other papers that Sal had
called attention to on her blue-box paper. But immediately Sal stops her, saying, “All
these other boxes are from different publishers, and they give only fake news. Return
them. They are worthless –only trust the papers in a blue-box. They tell you all you
need to know.”
It saddens Tess a bit that she won’t get more such stories, but out of respect for
her uncle she puts it back and refrains from gathering more papers. On the train,
though, she pulls out her blue box paper and reads it again. It is growing on her, and
when she tells her sister about her trip, what she relates as factual about the history of
Land of Lakes County and its founding citizens is what she says any blue-boxer like
herself holds as true.
Now we can imagine multiple variations on this base Tess Case. Perhaps Tess learns
that had she listened to Sal’s neighbour, she would have been introduced only to a red-
box paper, and been told that that was the reliable one. Perhaps all the people in Sal’s
family trust the blue-box paper, but most people in the county trust the yellow, or vice
versa. Perhaps Tess knows that she is in Fake News County (Enlightened Tess) or
12
perhaps she does not (as in the base case). In each such case, although it is testimonial
transmission rather than visual perception that is the source of belief in Tess cases, it
must be acknowledged as an environmental veritic luck-impacted context of inquiry if
Tess was veritically lucky (that is, lucky that she came to acquire a true rather than
false belief) given the doxastic method she employed in her specific epistemic
environment.
I hold that these conditions are fulfilled in the Tess case, and that her belief fails
to be knowledge even if there was one wholly true newsstand and it was the one she
vested authority in. True, this argument requires modal closeness, and relevant
similarity of basis, but these conditions seem to be fulfilled and I do not see other ways
to pry the Barney and Tess cases apart. This does not imply that multiple pieces of
independent evidence might not mitigate the risks that in Tess’ case constitute an
environment of malign veritic luck. Not all testimonial transfer is unsafe, and not all
testimonially-based beliefs are insensitive in the way that Tess’ are. I will discuss the
compounding of benign “evidential” by malign “environmental” luck below; but my
claim is about Tess, as described, and not about all persons who have invested authority
in a testimonial source under conditions of contestation. So neither do I think that
recognizing the impact of malign luck on Tess’s beliefs about the history of the county
must inevitably lead us some much broader scepticism about testimonial knowledge
generally. The way that safety and sensitivity are here construed does not invite, but
will indeed I think provide grounds for rejecting the broad “parity” response popular in
religious apologetics: the response that to be sceptical about the epistemic status of
Tess’ testimonial beliefs will result in excessive scepticism about a much wider range
of ordinary testimony cases.25
13
Another reason why Tess’ and other agents’ beliefs in the narratives of one or
another Fake News County newsstand must be seen as the product of a highly risky
doxastic strategy is the relationship between the testimonies that the papers provide
about the history of the county: contrariety of content itself. Part of the intuition that
there are propositional defeaters to Tess’ personal justification for her testimonial
beliefs is that the base case describes significant contrariety of content. Further, it
describes what we will term symmetrical contrariety, in that each publisher claims all
other publisher’s publications are untrustworthy. Now had there not been such actual or
reported contrariety to their contents, would Tess’ beliefs, if we assume them true, be
less impacted by malign environmental luck? If we answer to this question ‘Yes,’ as I
want to argue that we should, then why do epistemologists of testimony seem so often
to ignore such factors, and think only in terms of the reliability of the single testimonial
chain an agent “trusts”?26 The inductive risk account shows as epistemically significant
not just the diversity of beliefs in a domain, but contrariety of those beliefs. Not all
testimonies or testifiers are as polemical as the described news publishers, demanding
counter-inductive inference to the unique or complete truth of just one. These are things
that compound Tess’ situation with malign luck. As an agent acquires more information
about the contents of the papers in the different coloured boxes, the degree of
contrariety and mutual vice-charging further impacts the well-foundedness of a belief
acquired on the basis of acquaintance with just one of the numerous publishers.
Epistemologists refer to an epistemic environment as “hostile” if it is one that is
unsafe for the doxastic method employed, and this is a key characteristic that
distinguishes mere evidential luck from a condition where it is compounded by
environmental luck. Barney’s method of coming to believe ‘That is a red barn’ is
unsafe, because easily could he have gotten it wrong in his driving environment,
14
trusting only to his eyesight from the roadway. If Barney looks out the window a
minute earlier or later, he acquires a belief with relevantly similar content, yet false.
Barney’s belief is also insensitive since what Barney affirms as a barn he would have
affirmed, even if he was not lucky enough to have come across one of the few real
barns. Analogously, I argue that Tess’ method of coming to believe that the blue box
described true history of Land of Lakes County is unsafe, because easily could she have
gotten it wrong in her news reporting environment, trusting only to the testimony of a
friend or family member, and to the vivid phenomenal seemings she has in reading blue
box stories.
Tess’ belief is also insensitive because we have to surmise that if the publisher
was not reliable, Tess would have still believed that it was. If she would trust the news
of just the first box she came to, when it might be a small and/or unrepresentative
sample, or because it is uniquely recommended by one among many disagreeing
residents, or by someone she considers reliable because a kinsman, then Tess would
believe the same thing even were it false. Insensitive beliefs are not typically a by-
product of a hostile epistemic environment, as Pritchard describes unsafe belief, but of
what I propose to term a beguiling environment. I will develop connections between a
beguiling evidential situation and epistemic responsibility in the final section. But as a
general point, when trait-dependent factors are salient in one’s belief, it is
commonplace to find the agent reasoning backwards from the assumption that their
belief is true, to the benign nature of any luck they may have had in coming to that
belief. It is commonplace to find them confusing (presumed) truth with objective
justification, which in turn beguiles them into thinking that their (obviously good) luck
of having grounds for a true belief also guarantees the benign nature of luck’s impact
on their epistemic situation. But the assumption is naïve: the malign/benign epistemic
15
luck distinction is instead a modal one, and one that is inextricable, for agents
performing inquiry, from recognition of the centrality of inductive norms to epistemic
assessment.
But perhaps the most interesting conclusion that might be drawn from the preceding is
that the positive epistemic status of beliefs based on testimonial transmissions is not
guaranteed, even if it is maintained that the particular testimonial chain that sources the
beliefs to be assessed is a trustworthy testimonial chain.27 It is not guaranteed any more than
that Barney’s true belief that he sees a barn has positive epistemic status.28 That depends not
just on the object, and his process, but upon a third factor that situates his inference in an
inductive context. What in a simple, non-fake barn country scenario would certainly seem to
have positive epistemic status, is far more problematic in Barn County. Whether recognized
or not, Tess like Barney is in an epistemic environment in which there are defeaters to
personal justification. An inductive context implies inductive epistemic risk. High epistemic
risk derives from epistemic situations inhospitable to the epistemic strategy one is employing.
Modal riskiness marks epistemic luck as veritic and malign. Environmental luck is veritic
luck, and the epistemic standing of Barney’s luckily true belief is doubtful because of the
inductive norms Barney violated in forming his belief.29
Epistemic success arguably requires an agent’s doxastic strategy being modally safe,
for modal riskiness marks epistemic luck as veritic and malign. This may be an externalist
perspective, but if we think in terms of argument structures we can translate these concerns
into ones of defeaters and defeat. Tess like Barney has a propositional defeater of the
undercutting variety, which is a most serious matter. Propositional defeaters are conditions
external to the perspective of the cognizer that prevent even a personally justified true belief
from counting as knowledge. The pertinent external fact is that our two agents Tess and
Barney are in such conditions that there is no level of generality that a reliable belief-forming
16
process plausibly explains the truth of their beliefs. Our account can certainly be flexible
enough to allow partial defeaters. But these bare facts of Tess and Barney’s inductive
contexts of inquiry are arguably each as much a propositional defeater of the undercutting
variety as that the wall in front of me is being irradiated with a red light is an undercutting
defeater for my belief that the wall in front of me is painted red because it visually appears
red to me. It could still be true that it is painted red and not white or some other color; but my
trusting my eyes as the rational basis for that belief is undercut by this further fact of which I
was unaware when I formed my belief.
Summarizing, Tess like Barney does an inductive ‘fail,’ although an agent’s culpability
for being ignorant of their inductive context of course depends upon details of the case
described.30 My point is that environmental luck is a serious worry about the well-
foundedness of belief whether one is aware or ignorant of their inductive context. That
environmental luck threatens to impact the well-foundedness of an agent’s belief, and that the
agent’s context of inquiry is properly describable as an inductive context, are nearly
synonymous.
We should say that environmental luck, when it affects an agent’s epistemic situation,
compounds evidential luck. It is not as if evidential luck went away and a malign kind just
‘replaced’ it. An epistemic context can change by degrees, much as the assessment of the
strength of an inductive argument can change by degrees. Evidential luck, as the only way we
are in a position to know anything beyond the analytic and a priori, is ever-present to the
human condition, but its benign status is upset when malign conditions change its demeanor.
This compounding thesis suggests a more complex relationship than one where
epistemologists treat the benign/malign distinction more as separate buckets than as
dialectically negotiated borders.31
17
This negotiation is quite apparent when we recognize how closely the epistemology of
nurtured controversial views depends on testimony and testimonial transmission. As Rachel
E. Fraser points out, “Recent epistemological history has inclined towards ‘testimonial
optimism,’ keen to stress the division of epistemic labour and the ubiquity of our dependence
upon the words of others.”32 Not incidentally, testimonial optimism is associated with
Christian evidentialist apologetics, phenomenological conservativism, and the unmovably
steadfast position its proponents justly describe as “dogmatism.” These view seem at opposite
extremes from the broad skepticism about knowledge in domains of controversial views
mentioned earlier (Cohen and Sher). I would like to think of our inductive risk-based account
as a third option in what Fraser seems right to see as an important emerging debate between
testimonial optimists and testimonial pessimists.33 But as a third option it is not completely
neutral between an account that makes for ‘easy knowledge’ even of religion-specific claims
so long as one thinks their purported special revelation is more special than other purported
special revelations, and a view that is skeptical of that. We should all be skeptical of that, and
the specialness of the home religion’s special revelation is an article of faith, not a premise in
an argument that those not already predisposed to should accept on the basis of its epistemic
merits. Every religious testimonial tradition’s revelation is reliable to its adherents, just as
every theology or sect is orthodox unto itself.34 The self-reassurance of religious knowledge
here becomes an article of faith. But there is good philosophical criteria for when a
testimonial environment is impacted by malign environmental luck. Cultural chauvinism,
gender chauvinism and racism manifest attitudes the commit epistemic injustice to outsiders,
and this must be what each religious exclusivist community thinks of all others.35 Even
setting aside apologetic motivations for testimonial optimism, phenomenological
conservativism, etc., in testimonial cases generally we have to look not just at the source, but
at agents engaged in inquiry, at how to naturalistically describe their belief-forming cognitive
18
strategies, and at their objectively-described inductive context. Only in this way can we
assess whether the kind of epistemic luck operating in particular real or imagined cases is
benign, or instead malign, i.e., undercutting of positive epistemic status.
3 Trait-dependent Overdetermination, Risk, and Well-founded Etiological Challenges:
The Psychology and Epistemology of our Importunate Presumptions
“Our eyes see nothing behind us. A hundred times a day we make fun in the
person of our neighbour, and detest in others, defects which are more clearly
present in ourselves, and we marvel at them with prodigious impudence and
heedlessness. Oh, importunate presumption!” – Montaigne36
Montaigne’s passage describing our ‘importunate presumptions’ captures quite well
the contemporary recognition of our common bias blind spot. Our obvious psychographic
diversity, and the polemical ground dynamics involved in our ‘culture wars’ are compounded
on the agential side by the invisibility of our biases to ourselves. The judgments we make in
ignorance of our own biases Montaigne calls our importunate presumptions, and he suggests
a host of practical factors that make them appealing. Along with its denial in favor of
exceptionalism, Montaigne points out that the cost of these ego, ethnic, and anthropocentric
presumptions is that, sadly, “it comes to pass that nothing is more firmly believed than things
least well-known.”
This is an ironic caricature of dogmatism and bias to be sure, but Montaigne is
noticing persons and groups, and that while the impact of directional thinking on nurtured
controversial views is very significant, it’s bearing on well-foundedness is not all-or-nothing.
Which of our beliefs can claim to be free from underdetermination/overdetermination? So I
19
agree with Davis (2009) and other permissivists that recognized trait-dependence in the
aetiology of belief does not undermine the basing relationship in any sweeping sense: it
undermines that relationship only if and when that trait-dependence takes the form of
personal or social bias. We will return to specific, scalar markers of this shortly. But when it
does not undermine the basing relationship then agent reasons competently, and the influence
of personal traits need only be regarded as one of the many sources of the faultless cognitive
diversity that John Rawls explained as grounds for reasonable pluralism. Davis makes the
connection between trait-dependence and Rawlsian reasonable pluralism explicit by quoting
the deservedly famous “burdens of judgment” section of Political Liberalism: “To some
extent (how great we cannot tell) the way we assess evidence and weigh moral and political
values is shaped by our total experience, our whole course of life up to now; and our total
experiences must always differ” (25).
Although many nurtured beliefs may be biased, we cannot assume that all are without
begging the interesting philosophical questions.37 The ‘trait-basing question,’ which asks
whether and when trait dependence defeats the basing relationship, requires investigation. In
order to investigate it, Davis thinks we should first adequately distinguish simple trait-
dependence from bias, in order to compare them. He defines ‘trait’ broadly to include “not
only personal traits such as gender or features of one’s personality, but also such properties as
socioeconomic background, rigorous training, exposure to certain individuals or groups,
subscribing to a certain ideology or religion, or having a certain personal history.”
Differences in people’s experiences, background beliefs, and available testimonial
evidences can make robust the evidential ambiguity that affects so many views we hold on
morals, politics, philosophy, and religion. The robustness of evidential ambiguity and the
resultant need for holistic judgments on the part of agents, in turn impacts both epistemic
assessment. We accordingly describe epistemic location not straight away either as bias or as
20
the intrusion of epistemically-irrelevant influences, but rather as a source of (sometimes but
not always) faultless disagreement.38 Consistent with Davis’ divergentism (2015), I take
faultless disagreement and responsibility in doxastic as well as sub-doxastic ventures as the
charitable default assumption about controversial views. This kind of faultlessness does not
imply relativized truth, or the idea of both parties being right.39
So censure on the basis of an agent’s doxastic irresponsibility is the exception, and
has the burden of evidence upon it.40 But there are many exceptions where censure is well-
founded because the agent’s beliefs are not well-founded, and these exceptions may readily
occur in any of the four domains of controversial views. An etiological challenge has to be
mounted domain-by-domain, and case-by-case. Those cases where well-foundedness is
especially challengeable are one’s where belief-formation or maintenance flow from a risky
doxastic strategy, and/or where the agent exhibits psychological marks of undue influence by
subjective factors. The agents may “mirror” known biases, engage in rhetorical vice-
charging, and/or exhibit certain psychological affects like contingency or indoctrination
anxiety, or confabulation.
Philosophically, some thought experiments that heighten these effects can be helpful
for agents to gain perspective on their nurtured controversial views. What if the agent is
rationally convinced that she very likely would see her own actual nurtured belief as false had
she been nurtured in a different culture or epistemic community? What if the agent concedes
she would likely see it as a product of unrecognized bias? Would these outcomes of the
thought experiment be defeaters to proper basing? Sensitivity is often criticized as a strong
demand, and I am not assuming that it is a condition of knowing. But it seems to track some
relevant aspects of reasonableness, despite the fact that the belief of the victim of a malin
genie that he has a physical body is as insensitive also. My entitlement to hold fast to these
metaphysical beliefs in a physical universe, other minds, etc. where all sources of empirical
21
evidence support my causal story may not extend to an entitlement to hold steadfast in the
case of more controversial views.41
For a culturally nurtured belief to be insensitive, we do not have to imagine radical
deception scenarios.42 We only have to make some quite modally close changes, such as:
growing up in a politically or religiously conservative family instead of a liberal one, growing
up in our same society but in a different religious tradition; growing up in a different society
that has a different majority religious tradition, etc. So it is plausible that sensitivity tracks
reasonableness when the closest error-possibilities are nearby, and it doesn’t track
reasonableness when the closest error-possibilities are distant. 43 Whatever we can say about
the truth-aptness of beliefs in domains of controversial view, and about trusting putative
moral or religious experts, it is clear that beliefs in these domains are exceptionally
insensitive.44 But their insensitivity and their causally over-determined aetiology are almost
indistinguishable.
Now even if the mentioned thought experiments regarding the safe and sensitive
founding of our beliefs are indeed epistemologically significant, does the bias blind spot
allow agents to see the implications? Does it render them able to see when the attributions of
bias or intellectual vice which they readily apply to others, apply as well or better to
themselves? The approach taken here is far from defeatist, because I think there is much to be
said that can redress the bias blind spot. The numerous indications of bias are a resource for
epistemologists, just as they are for psychologists. True, epistemologists are always going to
be censuring those who are least likely to acknowledge their importunate presumptions, or to
be motivated to re-evaluate their beliefs. Indeed that is why we censure what we perceive as
bias and circular reasoning: for without a mirror to hold oneself up against, it is almost
impossible to see that the ‘inductive finger’ points not just outwards at holders of contrary
views, but frequently back at them.45 It would be impossible to understand that people with
22
contrary views to ours may still be made in our same image; instead they become trapped in
seeing their deviance from our opinions as confirmation of their bias.
So the problem of motivation to de-bias oneself is one in which philosophers can seek
the aid of psychology. The horses I am familiar with do drink when led to water, and if they
don’t then I suggest thereafter riding them much harder. But our project here is much
concerned with what philosophers can contribute to the assessment of bias and other
defeaters to well-founded belief. Here I see a lot of untapped resources. Epistemology can
show the enemy in the mirror to those who need most to see it, although the act of
recognition, since it requires proper motivation, has to come from within.
We have started to sketch an account of well and ill-founded belief based upon low
and high inductive risk. Our account says that agents mitigate epistemic risk by
acknowledging an inductive context and abiding by inductive norms. It says that agents
exacerbate moral and epistemic risk by asymmetrically positing themselves or their sources
of belief as exemptions to a recognized pattern. In one sense this is really just the
philosophical analysis of what psychologists call my-side, or belief-bias. Complementary to
discussion of degrees of trait-dependence, I want to introduce overdetermination theory. The
problematic sort of overdetermination stems from finding multiple trait-dependent factors
each sufficient to produce the target belief. In such cases we have trouble isolating which of
these processes actually caused the belief. If the belief somehow is true, we have lost the
connection with creditworthiness on the part of the agent.
To develop the inductive risk account further, let’s very briefly take a closer look at
two further inductive risk-indicators: confabulation, and merely rhetorical or self-deceived
bias-charging. Confabulation is counter-point to contingency anxiety, though they could be
seen as two different ways to deal with the cognitive or moral dissonance. As Andreas
Mogensen (2017) explains, “Etiological Challenges encourage us to pay attention to notable
23
facts about our belief-forming processes that would otherwise be ignored.” Mogensen
usefully gives a name —contingency anxiety— to the anxiety that a person might have who
rationally concedes to counter-factual statements indicating that they would in other
circumstances have come to hold beliefs that are by their own lights wrong, or more to the
point to reject as false beliefs that are by their own lights true. DiPaolo and Simpson focus on
a close cousin:
Indoctrination Anxiety, on our usage, is something narrower than
Genealogical Anxiety, in which an individual is caused to ‘worry that the
origins of her beliefs will turn out to be a source of discredit not vindication,’
and also narrower than a more general feeling of Contingency Anxiety, in
which an individual is led into ‘a feeling of unease due to discovering that she
holds certain beliefs because of arbitrary factors in her background.’
Indoctrination Anxiety, rather, is the distinctive sense of unease a person
experiences when she’s led to suspect that her beliefs resulted from a
systematic program of doctrinal inculcation.”46
Where one or another form of anxiety and attendant epistemic humility is appropriate
yet lacking in an agent, we can hypothesize that she will be quick to engage in confabulatory
explanation. Confabulation is counter-point to contingency anxiety, though they could be
seen as two different ways to deal with the cognitive or moral dissonance. William Hirstein
writes, “Confabulation involves absence of doubt about something one should doubt: one’s
memory, one’s ability to move one’s arm, one’s ability to see, etc. It is a sort of pathological
certainty about ill-grounded thoughts and evidences.” More than simple rationalization,
“Confabulators don’t know that they don’t know what they claim.”47 Hirstein gives these
conditions:
24
“Jan confabulates if and only if:
1) Jan claims that p (e.g., Jan claims that her left arm is fine).
2) Jan believes that p.
3) Jan’s thought that p is ill-grounded.
4) Jan does not know that her thought is ill-grounded.
5) Jan should know that her thought is ill-grounded.
6) Jan is confident that p.”48
Sharp and apparently unprincipled explanatory asymmetries are another key marker
of bias. In “You Don't Know me, but I Know You: The Illusion of Asymmetric Insight,”
Pronin, Kruger, Savitsky, and Ross discuss psychological studies confirming that people
often exhibit “an asymmetry in assessing their own interpersonal and intrapersonal
knowledge relative to that of their peers.”49 We tend to judge others as biased, especially
when they disagree with our controversial views. Yet we are more likely to take our own
views (and our own attributions of bias) to be bias-free. So perhaps ironically, one place to
look for such asymmetries is in “bias-charging” behavior itself. Peer denial through ill-
founded bias-charging is a very common but highly dogmatic way to insulate particular
beliefs from rational criticism. Emily Pronin and Lee Ross in particular have also suggested
broader application of their findings by describing more of the psychology of “naïve realism”
that biased trait-attributions often presupposes: “although this blind spot regarding one's own
biases may serve familiar self-enhancement motives, it is also a product of the
25
phenomenological stance of naive realism.”50 Naïve realism as psychologists discuss it is
connected with what philosophers such as Lisa Bortolotti & Matthew Broome (2009) refer to
as failures of belief ownership and authorship.51 So psychologists, and philosophers who
utilize psychological research are both interested in “the relevance of these phenomena to
naïve realism and to conflict, misunderstanding, and dispute resolution.”52
Ian Kidd (2016) relatedly points out that vice-charging can either be an
encouragement for needed self-awareness and doxastic responsibility in the agent who is
criticized, or it can be a strategy of protecting oneself or one’s beliefs from criticism. Kidd
distinguishes rhetorical complaints and robust charges, where only the latter qualify as
legitimate modes of criticism: “A rhetorical vice charge involves an agent expressing a
negative attitude, opinion, or evaluation of some other agent… but not the presentation of any
reasons, evidence, or feelings in support of them, so they do not do any real critical
work….”53 Robust vice-charges require a clear concept of epistemic responsibility, and so
“should be sensitive to the aetiology of vice and the ecological conditions of epistemic
socialisation.”
Each of these psychological effects might motivate a strong etiological challenge to
the well-foundedness of belief. Each suggests the salient causes of belief to be temperamental
factors that, if they do not exhibit, at least must be acknowledged to mirror known personal
or social biases. To summarize this section, the study of trait-dependence is vital to the
epistemology of controversial views. Trait-dependence appears to be indicative of the
overdetermination of belief by temperamental factors.54
The beguiling nature of evidence in environmental veritic luck cases is exacerbated by
the trait-dependent overdetermination of belief. This appears to be especially so with
culturally nurtured controversial views. In cases of causally overdetermined belief, an agent’s
26
belief might have been adopted on any one of several different trait-dependent bases,
although on an ex ante basis the belief is underdetermined by the agent’s actual evidence,
evidence which would not rationally convince persons not already disposed toward the belief.
Although we rarely treat them this way, evidential underdetermination and causal
overdetermination are paired theses. The first is a philosophical concept and the second is a
scientific one, but the two are conceptually linked. Overdetermination theory is still a largely
unexplored approach in debates over the basing relationship. But it is motivated by the
holistic nature of people’s reasoning about worldview beliefs, and under conditions of
uncertainty and other pragmatic constraints, as Rawls alerted us to. It is motivated also, we
have now seen, by some specific psychological studies, research that illuminates how trait-
dependent judgment contributes to psychographic contrariety.
4 Conclusion
This chapter has argued for a fairly common-sense view: That our nurtured beliefs being
exposed to the epistemic location problem need not undercut their reasonableness or our right
to hold them. The bearing of epistemic location on nurtured controversial views is very
significant, but need not fall evenly across domains. Nor does it fall on all agents the same, as
we need to know more specifics about the sensitivity of an agent’s reasons for her belief in
order to assess how seriously to take an etiological challenge. People are not necessarily
intellectually vicious for accepting nurtured beliefs and holding them without a great deal of
reflection. But neither does such a permissivist account rationalize dogmatism or imply the
reasonability, tout court, of holding to what we are taught. Permissionism should sharpen
reasoned criticism rather than lead to its abandonment, and the turn to risk and inductive risk
in particular, I have been arguing, shows us how. There is faultless or reasonable
27
disagreement aplenty on the present view, but it occurs primarily in minimally truth-apt
discourses where the parties to the disagreement recognize their discourse as minimally truth-
apt. The disputants then take commitments in the domain of their disagreement as requiring a
greater degree of epistemic and moral humility than disagreements over straight-forwardly
empirical claims or questions. Faultless disagreement does not occur under conditions of self-
deception or of bias mirroring. Ways of acquiring or maintaining a belief dependent upon
apparent violations of inductive norms are far from faultless, and the beliefs of agents who
rely on such methods are especially exposed to serious etiological challenge.
To conclude, nurtured beliefs cannot all be evaluated in a uniform way as conformists
and dogmatists have assumed.55 Our commitments in domains of controversial views, do not
deserve to be assessed as ill-founded simply because conditioned by temperament and
epistemic location. But our approach suggests that neither do they deserve the free pass (as
personally justified and as enjoying positive epistemic status) that epistemic conservatives
and epistemic dogmatists issue them on the basis of an agent’s phenomenal seemings.56
28
Bibliography
Aiken, Scott and Robert B. Talisse. “The Will-to-Believe is Immoral.” In William James,
Moral Philosophy, and the Ethical Life: The Cries of the Wounded, edited by Jacob
Goodson, 143-160. Lanham, MD: Lexington Books, 2018.
Axtell, Guy. Problems of Religious Luck: Assessing the Limits of Reasonable Religious
Disagreement. Lanham, MD: Lexington Books, 2019.
-------. “William James on Pragmatism and Religion.” In William James, Moral Philosophy,
and the Ethical Life: The Cries of the Wounded, edited by Jacob Goodson, 317-336.
Lanham, MD: Lexington Books, 2018.
-------. “Possibility and Permission? Intellectual Character, Inquiry, and the Ethics of Belief.”
In William James on Religion, edited by S. Pihlstrom and H. Rydenfelt, 165-198.
London, United Kingdom: Palgrave MacMillan UK, 2013.
-------. “From Internalist Evidentialism to Virtue Responsibilism: Reasonable Disagreement
and the Ethics of Belief.” In Evidentialism and its Discontents, edited by Trent
Dougherty, 71-87. Oxford, United Kingdom: Oxford University Press, 2011.
Baker-Hytch, Max. “Religious Diversity and Epistemic Luck.” International Journal for
Philosophy of Religion 76, no. 2 (2014): 171-191.
------. “Testimony amidst Diversity.” In Knowledge, Belief, and God: New Insights in
Religious Epistemology, edited by Matthew A. Benton, John Hawthorne, and Dani
Rabinowitz. Oxford, United Kingdom: Oxford University Press, 2018.
doi:10.1093/oso/9780198798705.003.0010.
Baldwin, Erik and Michael Thune. “The Epistemological Limits of Experience-Based
Exclusive Religious Belief.” Religious Studies 44, no. 4 (2008): 445-455. doi:
10.1017/S0034412508009530.
29
Ballantyne, Nathan. “De-Biasing Biased Thinkers (Including Ourselves).” Journal of the
American Philosophical Association 1, no. 1 (2015): 141-162.
-------. “The Problem of Historical Variability.” In Disagreement and Skepticism, edited by
D. Machuca, 239-259. New York, NY: Routledge Press, 2013.
Basu, Rima and Schroeder, Mark (eds.). “Can Beliefs Wrong?” special edition, Philosophical
Topics 46 (1), 2018a.
Basu, Rima and Schroeder, Mark. “Epistemic Wronging.” In Pragmatic Encroachment in
Epistemology, edited by Brian Kim and Matthew McGrath. London: Routledge,
2018b.
Bishop, John. “Trusting Others, Trusting in God, Trusting the World.” In Religious Faith and
Intellectual Virtue, edited by Lauren F. Callahan and Timothy O’Connor, 159-173.
New York, NY: Oxford University Press, 2014.
-------. “How a Modest Fideism May Constrain Theistic Commitments: Exploring an
Alternative to Classical Theism.” Philosophia 35, no. 3-4 (2007a): 387–402.
-------. Believing by Faith: An Essay in the Epistemology and Ethics of Religious Belief. New
York, NY: Clarendon Press, 2007b.
Bondy, Patrick and Duncan Pritchard. “Propositional Epistemic Luck, Epistemic Risk, and
Epistemic Justification.” Synthese (2016): 1-10. https://doi.org/10.1007/s11229-016-
1262-2.
Booth, Anthony and Rik Peels. “Why Responsible Belief is Permissible Belief.” Analytic
Philosophy 55, no. 1 (2014): 75-88.
Booth, A.R. “The Theory of Epistemic Justification and the Theory of Knowledge: A
Divorce,” Erkenntnis 75 (1) (2011): 37-43.
30
Bortolotti, Lisa and Matthew Broome. “A Role for Ownership and Authorship in the
Analysis of Thought Insertion.” Phenomenology and the Cognitive Sciences 8, no. 2
(2009): 205-224.
Bortolotti, Lisa. “Stranger than Fiction: Costs and Benefits of Everyday Confabulation.” Rev.
Phil.Psych 9 (2018): 227–249. https://doi.org/10.1007/s13164-017-0367-y.
Carter, J. Adam. “On Behalf of Controversial View Agnosticism.” European Journal of
Philosophy. (2018): 1- 13. https://doi.org/10.1111/ejop.12333.
-----. “A Modal Account of Luck Revisited.” Synthese 194, no. 6 (2017): 2175-2184.
Church, Ian M. and Robert J. Hartman. The Routledge Handbook of the Philosophy and
Psychology of Luck. London, United Kingdom: Routledge Press, 2019.
Cohen, G. A. “Paradoxes of Conviction.” In If You’re An Egalitarian, How Come You’re So
Rich?, edited by G. A. Choen, 7-19. Cambridge, MA: Harvard University Press,
2000a.
-----. If You’re an Egalitarian, How Come You’re So Rich? Cambridge, MA: Harvard
University Press, 2000b.
Davis, John K. “Faultless Disagreement, Cognitive Command, and Epistemic Peers.”
Synthese 192, no. 1 (2015): 1–24.
-----. “Subjectivity, Judgment, and the Basing Relationship.” Pacific Philosophical Quarterly
90, no. 1 (2009): 21–40.
DiPaolo, Joshua and Robert M. Simpson. “Indoctrination Anxiety and the Etiology of
Belief.” Synthese, 193, no. 10 (2016): 3079–3098. doi: 10.1007/s11229-015-0919-6.
Engel Jr., Mylan. “Epistemic Luck.” Internet Encyclopedia of Philosophy.
http://www.iep.utm.edu/epi-luck/.
-----. “Is Epistemic Luck Compatible with Knowledge?” Southern Journal of Philosophy 30,
no. 2 (1992): 59-75.
31
Fantl, Jeremy. Limitations of the Open Mind. Oxford, United Kingdom: Oxford University
Press, 2018.
Faulkner, P. and Simpson, T. (eds.). The Philosophy of Trust. Oxford: Oxford University
Press, 2017.
Fraser, Rachel E. “Testimonial Pessimism.” In Knowledge, Belief, and God: New Insights in
Religious Epistemology, edited by Matthew A. Benton, John Hawthorne, and Dani
Rabinowitz, 203-227. Oxford, United Kingdom: Oxford University Press, 2018.
Freedman, Karyn L. “Testimony and Epistemic Risk: The Dependence Account.” Social
Epistemology 29, no. 3 (2015): 251-269. doi: 10.1080/02691728.2014.884183.
Haack, Susan. “‘The Ethics of Belief’ Reconsidered.” In The Philosophy of Roderick M.
Chisholm, edited by Lewis E. Hahn, 129-144. Chicago, IL: Open Court, 1997.
Haidt, Jonathan. The Righteous Mind: Why Good People Are Divided by Politics and
Religion. New York, NY: Vintage Books, 2012.
Harman, Gilbert. “Knowledge, Reasons, and Causes.” Journal of Philosophy 67, no. 21
(1970): 841-855.
Hirstein, William. Brain Fiction: Self-Deception and the Riddle of Confabulation. Denver,
CO: A Bradford Book, 2005.
Howell, Robert J. “Google Morals, Virtue, and the Asymmetry of Deference.” Noûs 48, no. 3
(2014): 389-415.
Kahan, D. M. (2017), “The Expressive Rationality of Inaccurate Perceptions,” Behavioral
and Brain Sciences 40.
----- (2013), “Ideology, Motivated reasoning, and Cognitive Reflection,” Judgment and
Decision Making 8: 407–24.
32
Kelly, Thomas. “Evidence Can Be Permissive.” In Contemporary Debates in Epistemology,
2nd Edition, edited by Matthias Steup, John Turri, and Ernest Sosa, 298-311.
Hoboken, NJ: Wiley-Blackwell, 2013.
Kidd, Ian J. “Charging Others with Epistemic Vice.” The Monist 99, no. 3 (2016): 181–197.
-----. “A Phenomenological Challenge to ‘Enlightened Secularism.’” Religious Studies 49,
no. 3 (2013): 377-398.
Kölbel, Max. “Faultless Disagreement.” Proceedings of the Aristotelian Society, 104 (2004):
53-73.
Kopec, Matthew and Michael Titelbaum. “The Uniqueness Thesis.” Philosophy Compass 11,
no. 4 (2016):189-200.
Lackey, Jennifer. “Experts and Peer Disagreement.” In Knowledge, Belief, and God: New
Insights in Religious Epistemology, edited by Matthew A. Benton, John Hawthorne &
Dani Rabinowitz, 228-245. Oxford, United Kingdom: Oxford University Press, 2018.
Mawson, T. J. “Mill's Argument against Religious Knowledge.” Religious Studies 45, no. 4
(2009): 417-434.
McCain, Kevin. “The Virtues of Epistemic Conservatism.” Synthese 164, no. 2 (2008):
185-200.
Mill, John S. On Liberty. BLTC Research. https://www.utilitarianism.com/ol/three.html
(accessed October 30, 2017).
Mogensen, Andreas L. “Contingency Anxiety and the Epistemology of Disagreement.”
Pacific Philosophical Quarterly 98, no. 1 (2017a): 590-611.
------. “Moral Testimony Pessimism and the Uncertain Value of Authenticity.” Pacific
Philosophical Quarterly 95, no. 2 (2017b): 261-284.
33
Montaigne, Michel de. The Essays of Montaigne, translated by Charles Cotton, edited by
William C. Hazlitt. e-books@Adelaide. https://ebooks.adelaide.edu.au/m/montaigne/
michel/essays/contents.html.
------. The Complete Essays of Montaigne, translated by Donald Frame. Redwood City, CA:
Stanford University Press, 1958.
------. Apology for Raymond Sebond, translated by Roger Ariew and Marjorie Grene.
Indianapolis, IN: Hackett Publishing, 2003.
Pelling, C. “Assertion, Telling, and Epistemic Norms,” Australasian Journal of
Philosophy 92 (2) (2014): 335-348.
----- . “Assertion and Safety,” Synthese 190 (17) (2013): 3777-3796.
Plantinga, Alvin. “Pluralism: A Defense of Religious Exclusivism.” In The Rationality of
Belief and the Plurality of Faith, edited by T.D. Senor. Ithaca, NY: Cornell University
Press, 1995.
Pritchard, Duncan. Epistemic Luck. Oxford, United Kingdom: Oxford University Press, 2005.
-----. “Epistemic Risk,” The Journal of Philosophy 113(11) (2016a): 550-571.
-----. “Anti-Luck Virtue Epistemology and Epistemic Defeat,” Synthese 195, no. 7 (2016b):
3065-3077. https://doi.org/10.1007/s11229-016-1074-4
Pronin, Emily, Thomas Gilovich, and Lee Ross. “Objectivity in the Eye of the Beholder:
Divergent Perceptions of Bias in Self versus Others.” Psychological Review 111, no.
3 (2004): 781-799.
Pronin Emily, Justin Kruger, Kenneth Savitsky, and Lee Ross. “You Don't Know Me, But I
Know You: The Illusion of Asymmetric Insight.” Journal of Personality and Social
Psychology 81, no. 4 (2001): 639-656.
34
Pronin, Emily, Daniel Y. Lin, and Lee Ross. “The Bias Blind Spot: Perceptions of Bias in
Self versus Others.” Personality and Social Psychology Bulletin 28, no. 3 (2002):
369-381.
Quinn, Philip. “Fideism.” In Oxford Companion to Philosophy, 2nd ed., edited by Ted
Honderich. Oxford, United Kingdom: Oxford University Press, 2007.
------. “Epistemic Parity and Religious Argument.” Philosophical Perspectives 5 (1991): 317-
341.
Rawls, John. Political Liberalism. New York: Columbia University Press, 1993.
Rinard, Susanna. “Believing for Practical Reasons.” Noûs (2018a): 1-22.
-------. “Pragmatic Skepticism.” Philosophy and Phenomenological Research, forthcoming
(2018b).
------. “Equal Treatment for Belief,” Philosophical Studies (2018c): 1-28.
Sher, George. “But I Could be Wrong.” Social Philosophy and Policy 18, no. 2 (2001): 64-
78.
Silva, Paul. “Knowing How to Put Knowledge First in the Theory of Justification,” Episteme
14 (4) 2017): 393-412.
Simpson, Robert M. “Permissivism and the Arbitrariness Objection.” Episteme 14, no. 4
(2017): 519-538.
Srinivasan, A. “The Archimedean Urge,” Philosophical Perspectives 29 (2015): 325-362.
Tucker, Chris. (ed.) Seemings and Justification: New Essays on Dogmatism and Phenomenal
Conservatism. Oxford, United Kingdom: Oxford University Press, 2013.
------. “Phenomenal Conservatism and Evidentialism in Religious Epistemology.”
In Evidence and Religious Belief, K. J. Clark & R. J. VanArragon (eds.), pp. 52-73.
Oxford: Oxford University Press, 2011.
35
Vavova, Katia. “Irrelevant Influences.” Philosophy and Phenomenological Research 96, no.1
(2018): 134-152.
Wallbridge, K. “Sensitivity, Induction, and Miracles,” Australasian Journal of Philosophy
96, no. 1 (2018): 18-126.
White, R. “Evidence Cannot Be Permissive.” In Contemporary Debates in Epistemology,
Matthias Steup & John Turri (eds.), 312-323. Oxford, U.K.: Blackwell Publishers,
2013.
Notes
36
1
According to J. Adam Carter (2018), recognition of peer disagreement implies that “we are
rationally obligated to withhold judgment about a large portion of our beliefs in controversial
subject areas, such as philosophy, religion, morality and politics.” He recognizes that a thorough-
going agnostic suspension or the kind recommended by some epistemologists is open to
objectionable consequences of ‘spinelessness,’ and impracticability —the un-livability objection.
So he distances his version of controversial view agnosticism from these worries, qualifying it such
that it allows for ‘suspecting that’ but not ‘believing that.’

2
Trait-dependent belief is widely acknowledged in domains where informal logic, interpretation,
weighing factors or norms, and the exercise of judgment are normal aspects of epistemic
assessment. In the acquisition of trait-dependent beliefs, “the subject reasons competently from
justifying considerations to a belief, but would not do so if he or she lacked some trait that appears
to be epistemically irrelevant, even if the subject and the situation were the same in all other
respects.” He agent “would not take those considerations to justify that belief if she had a different
socioeconomic back-ground, religious affiliation, temperament, political ideology, or the like, even
if she were otherwise in the same epistemic circumstances” (23).
3
See Dan M. Kahan (2013) for a discussion of empirical work on motivated reasoning, and (2017)
on how identity-protective cognition can generate inaccurate perceptions.

4
Montaigne, Apology for Raimond Sebond, 6.
5
Sher 2000. See also Mogensen 2017a; Ballantyne 2013; Vavova 2014; DiPaoli and Simpson 2016.
Vavova 2018 also argues that evidence of irrelevant belief influence is sometimes, but not always,
undermining. It is surprising how little this quite plausible thesis has been systematically explored,
but epistemologists of disagreement have been enamored of universalist answers.

6
The truth-aptness of philosophical judgments have been challenged in similar ways. See Amia
Srinivasan (2015) for a treatment of “genealogical skepticism” about philosophical judgments.
Srinivasan’s account overlaps mine where she provides discussion of different arguments for
genealogical skepticism and in response to it; more especially, in her noting that, “Epistemologists
differ over the extent to which luck plays a role in the acquisition of knowledge. All epistemologists
will agree that luck has some role to play... Where epistemologists disagree is on just how much
knowledge we can acquire through good luck” (347). The Tess case (below) will argue that this is a
matter of the kinds of luck in play, and not merely the domain. I thank Adam Carter for bringing
similarities with Srinivasan’s learned paper to my attention.

7
It is hard to say, for instance, just what the conclusion of Mill’s argument about the “London
Churchman” in On Liberty is. Bogardus (2013) very plausibly suggests that the alleged wrong—and
the basic reason for contingency anxiety— is a violation of safety. “For Mill, there are nearby
possibilities in which one forms religious beliefs via the same method she actually used, and yet in
which she would believe something which is, by her own lights, false.” But it also seems to involve
failure of sensitivity. See also Ballantyne (2013) and Baker-Hitch (2014).

8
DiPaolo and Simpson (2016). They bid philosophers to more carefully investigate the question,
“How does recognition of the contingent cultural etiology of one’s beliefs affect their epistemic
standing?” See also Simpson (2017).

9
Ballantyne 2015, 159.

10
Counter-induction is defined in dictionaries as a strategy that whether self-consciously or not
reverses the normal logic of induction. In its most formal sense, counter-inductive thinking is
something much more specific than weak inductive reasoning. It is not just weak analogy, weak
causal inference, or faulty generalization, to refer to the three forms of inductive reasoning,
analogical, causal, and generalization. Rather, counter-inductive inference is the logically illicit
move of reasoning oppositely to what induction suggests. For our purposes it is more simply a
logical failing to apply to one’s self (or to one’s own epistemic situation) an explanation that one
recognizes as applying to others (and others’ epistemic situations).

11
I prefer to interpret well-foundedness broadly enough to indicate plural kinds of epistemic
assessment. Where it might indicate merely subjective justification, or a sense of rationality or
personal justification related deriving from ‘synchronic evidential fit,’ objective factors are
obscured from view and I can no longer find it a very interesting concept. Well-foundedness cannot
mean such a subjective status as personal or subjective justification, cut off from causal etiology
and objective justification, without losing its philosophical interest. Thanks to Ru Ye for pointing
out the need for me to say more about well-foundedness. See Axtell (2011) for a fuller critique of
the evidentialist way of construing grounding, which fixates on ‘synchronic evidential fit’ as
answering questions both about agent rationality and about epistemic assessment. Actually that
concept can answer neither of these concerns. I portray it as an answer in search of a question.
12
It may be that the traditional project of analysis shows that people “know” far less than they
imagine. I suspect this is true, but it is a question in the project of analysis, and I do not see the
connections which others would make that one therefore should not believe. The language of being
rationally obligated to withhold judgment but to adopt only a lesser doxastic attitude than belief, is a
language that I suspect is objectionably voluntaristic.
13
To accommodate the unlivability objection to principled agnosticism, Carter expands the
connotation of “agnosticism” to include the sub-doxastic attitude of ‘suspecting that,’ when
conditions are right. But in this prescription, much like Feldman, there is still assumed a single right
response to revealed peer disagreement among controversial views: agnosticism. Like Feldman it
appears that Carter’s categories of doxastic attitudes are still essentially treated deontologically,
since they line up with epistemic duties or entitlements. These are things denied by permissivists
like myself. See especially the work of Thomas Kelly, and Matthew Kopec and Michael Titelbaum.
14
In defence of permissivism, see especially Kelly (2013), Booth and Peels (2014), and Kopec and
Titelbaum (2016). Virtue theory in contrast to internalist evidentialism I take to be champion of
diachronic norms, viz., the axiological-etiological or forward/backwards spectrum. For my own
virtue-theoretic account of permissivism, which I term doxastic responsibilism, see Axtell (2013
and 2018).
15
Principled agnosticism has seemed to its critics to prohibit actions where an act is forced. I do not
want to stray very far into questions of guidance-giving, since I take the norms that should inform it
to not be highly commensurate with norms that inform epistemic assessment, or the project of
analysis of propositional knowledge. I do not see these projects as very commensurate and wonder
what concept of rationality can bear the burden of guidance that one should always ‘split the
difference’ with our disagreeing peers, or again that one should ‘not traffic’ in belief at all in the
domains of controversial views.

16
I have elsewhere (2011) argued that the norms that inform an ethic of belief are typically more
diachronic than synchronic, and that guidance-giving takes place in the context of ecological
rationality, not ideal agency where the order of acquired evidence should make no rational
difference as all. Note that the objections I present to Feldman and Conee’s explicitly epistemic
evidentialism are meant to be complemented by my direct response (2018) to the over-weaning
moral evidentialism of Scott Aikin and Rob Talisse (2018). Both parties I think mis-apply the
rational uniqueness thesis to the epistemology of controversial views.
17
It is admittedly difficult to say what appropriate deference in one’s reasoning is, and this
difficulty is made worse by belief-focused epistemology where synchronic measures of evidential

fit are privileged due to the ex ante approach taken to personal justification. The highly influential
Protestant conceptions of religious faith that are prescriptively anti-evidentialist yet also identify
faith with assent to belief I take as sufficient to show why this ‘degree lowering’ kind of guidance is
of little value in the debate over the rationality of religious belief. Church and Barrett’s (2016)
alternative is welcome. They are unhappy with extent accounts of intellectual humility, and so
propose a doxastic account on which intellectual humility “is the virtue of accurately tracking what
one could non-culpably take to be the positive epistemic status of one’s own beliefs.”
18
I do not want to say too much about the nature of our control over belief, but I agree with Rinard
that practical considerations to serve as motivating reasons for belief. See her critique of Feldman
and the rational uniqueness thesis. Contrasting that evidential principle with the more pragmatism-
friendly Equal Treatment, Rinard correctly argues that “Insofar as we have control over these
beliefs (be it direct or indirect), Equal Treatment acknowledges the moral dimension as highly
relevant to the question of what we should believe” (“Equal Treatment for Belief,” 2018c). See also
Rinard 2018a and b.

19
Booth and Peele, 2014.

20
As children of time, we deserve respect for background beliefs and for many other effects of
culture. Guidance must be consistent with psychological acknowledgment of pragmatism about
reasons and of the ecological rationality of human agents. I would not presume to say that belief
may never be permissibly responsive to non-epistemic reasons. We must not forget that we rightly
reason holistically, and that as creatures of time as well as of place, we so inevitably ‘live forward.’
Looking backwards, as Montaigne correctly says, is much more difficult for us, and this is where
philosophy and the sciences help the most.

21
These points are arguably recognized in Rawls’ guidance of expecting much faultless
disagreement as an implication of conditions of free inquiry in a democracy. The burdens of
judgment are coordinate with sources of cognitive diversity over which we have little control. But
they appear to be denied by those whose self-ascriptions of knowledge distain the burdens of
judgment, if not also by impermissivists as a consequence of allowing no place for doxastic
permissions/invitations not reducible to epistemic duties to believe, suspend belief, or disbelieve.

22
Propositional luck is not a form of veritic luck because it is not a type that comes ‘betwixt the agent
and the world,’ as is the case in Gettier-type (intervening veritic luck) and barn-façade-type
(environmental veritic luck) cases. In Bondy and Pritchard’s explication of propositional epistemic
luck (PEL), they write, “S’s belief B is propositionally epistemically lucky iff S has a good reason
R (and therefore, propositional justification) for B, but it is only a matter of luck that she does…..
All cases of propositional epistemic luck are cases where a subject has a belief which is
propositionally but not doxastically justified (though, as we will shortly see, not all cases of beliefs
which enjoy propositional but not doxastic justification will involve propositional epistemic luck).
There are two ways in which a belief that is propositionally justified can fail to be doxastically
justified: it can be held on the basis of a bad reason, or it can be held on the basis of a good reason
but in a bad way.”

23
For other important recent work on the turn to epistemic risk, see Carter (2017) Freedman (2015),
and Riggs (2008).

24
This differs somewhat from Carter’s treatment; Carter sets up helpful conditions of “centrality” and
“symmetry” of disagreement in a domain as a way to distinguish less and more well-foundedness.
But he then seems to generalize to these conditions being met in all four domains of controversial
views, without recourse to more specifics that might distinguish them. It is unclear to me why a
person’s knowing that they are not in the market for knowledge, is thereby rationally from being in
the market for belief. Belief that p entails believing that p is true in some sense, but does not
without absurdity entail believing that one has everything else required of knowledge over and
above true belief.
25
Baker-Hitch (2018, 189) uses a parity approach to argue that contingency arguments like Mill’s
will “result in excessive skepticism concerning a range of ordinary testimony cases.” But Mill
already mentioned political ideologies in his argument, and we have already conceded that the
epistemic location problems affect domains of controversial views. For a deep and provocative
treatment of broad parity responses to etiological challenges to knowledge in a testimonial religious
tradition, see Quinn (1991 and 2007).
26
Contrariety is something largely lacking is perceptual cases, and this is part of why Barney cases
do not aid philosophers in analyzing environmental luck in ways that adequately distinguish it from
benign evidential luck. Why is ‘actual trust plus posited truth’ supposed to be a defeater for the
importance of testimonial diversity, and not ‘testimony diversity with symmetrical contrariety’ a
defeater for rational trust and the right to claim truth?

27
I agree also with John Bishop (and others including Jennifer Lackey, 2018) who argue “against the
proposal that faith’s similarities to interpersonal trust merit its being considered reasonable or
virtuous.” Bishop argues that “there is an important analogy between faith and trust that is crucial to
understanding the content of faith. But the disanalogies between the two sever the attempt to justify
faith along the same lines as trust” (Bishop 2014). For recent work on trust, see Faulkner and
Simpson (eds.) (2017).

28
As Rachel E. Fraser points out, “Recent epistemological history has inclined towards ‘testimonial
optimism,’ keen to stress the division of epistemic labour and the ubiquity of our dependence upon
the words of others” (2018, 204). Several authors have explored testimonial pessimism, which
takes a more dour view of testimonial transmission in domains of controversial views. See also
Howell 2014, and Mogensen 2017b. As both Fraser and Mogensen argue, our dealings with
testimonial reception bring in tow an ideal of authenticity that places special demands upon us. I
understand the key demand as primarily one to symmetrically apply inductive norms, rather than to
think counter-inductively. Charlie Pelling’s (2013) “Assertion and Safety” offers an account that
connects a safety condition on knowing with “a safety account of assertion, according to which one
asserts p properly only if one asserts p safely. The central idea is that an assertion’s propriety
depends on whether one could easily have asserted falsely in a similar case.” This kind of
translation between 3rd and 2nd personal (roughly, externalist and internalist) perspectives on
epistemic assessment is always welcomed.

29
On Pritchard’s ‘turn to risk,’ see his 2017 and recent papers on epistemic dependence and ALVE
papers. As a proponent of ALVE, Pritchard does not make sensitivity a generally necessary
condition on knowing. Indeed a strong sensitivity condition for everyday beliefs may invite radical
skepticism. But I will be interested to try to explain why it does mark out belief formation that,
viewed in terms of high epistemic risk, sometimes motivates a serious etiological challenge. It also
motivates an inference that contradicts the agent’s own account of reasons for belief: the inference
that the belief is insensitive because biased. The agent’s bias blind spot prevented them from seeing
that they have no good grounds for denying that the inductive finger points back at them, and that
their own belief was about as strongly overdetermined by cultural and temperamental factors as was
the contrary belief of their peers.
30
Questions of agent culpability I take to go with a theory of doxastic responsibility or epistemic
rationality. That theory may well connect with aims of guidance and perhaps censure, not with the
project of analysis. These I see as separate projects. Indeed I think there are three not two sorts of
normativity epistemologists are interested in: “personal” justification (synchronic and diachronic),
“epistemic” justification (epistemic ability and any further anti-luck conditions), and “guidance”
which may censure, but is purely ideal unless it respects human ecological rationality and pragmatic
constraints on doxastic attitudes. So evidentialists like Feldman are mistaken in the first place to
treat synchronic rationality as basic to analysis of knowledge, and mistaken again when taking
epistemic evidentialism as grounds for his evidentialist ethics of belief. See Booth (2011) in
support of the separate projects idea, the ‘divorce’ between the theory of rationality and the analysis
of knowledge earlier proposed by Richard Foley.
31
What I present is certainly more contextual an account than one where religious apologists and
philosophical theologians treat the benign/malign and the evidential/environmental as lining up, but
just determinable by whether one thinks that good religious epistemic luck has given them a
uniquely reliable testimonial source. This is circular reasoning that privileges the ‘home’ religion
without acknowledging the extent that adherents of contrary testimonial faith traditions do just the
same. Arguably its social consequences, far from respecting the Rawlsian burdens of judgments,
promotes rhetorical vice-charging. I argue in Problems of Religious Luck (Axtell, 2019) that it
appears to recommend embracing us/them group polemics rather than facing down our bias blind
spot or allowing any possibility, consistent with faith, that as far as religious epistemology goes, the
‘inductive finger’ points back at them.
32
Fraser 2018, 204. Charlie Pelling (2014; 2013) relatedly argues that there are significant
differences between the norms of “assertion” and of “telling,” differences that are often overlooked
among testimonial optimists. They need to be treated separately because the norms of telling are not
reducible to those of assertion.
33
Several authors besides Fraser have explored testimonial pessimism, which takes a more dour
view of testimonial transmission. See also Howell 2014, and Mogensen 2017b. My own Problems
of Religious Luck (2019) also engages this dispute.

34
Post-liberals maintain that all discussion should proceed “intra-textually.” Plantinga holds that if
Christianity is true then that are no propositional defeaters to warranted Christian belief. But these
arguments are both examples of the epistemically circular attempt to justify one’ framework with
reference to that very framework.

35
Consistent with work on epistemic injustice, I hold that attitudes and beliefs about others can
wrong others. But this claim is not uncontroversial. For recent work on this question of doxastic
responsibility and its limits, see the journal special edition edited by Rima Basu and Mark
Schroeder (2018a), and their paper “Epistemic Wronging” (2018b) which (like Axtell 2013)
appears to defend Susan Haack’s (1997) moral-epistemic “overlap” account.

36
Montaigne, “Of the Art of Discussion,” in Frame (ed.), 709.
37
Davis does not think it is correct to say that traits consist of holding particular beliefs, though the
trait may involve holding some beliefs. (24) A bias in cognition “is a tendency towards a certain
kind of distortion in one’s process of reasoning and belief formation.” But in trait-dependence cases
as Davis wants to understand them, ex hypothesi “there are no such distortions – just the
dependence relation… Trait-dependence is not a cognitive distortion unless trait-dependence
defeats the basing relationship – and that is the question before us; we cannot assume an answer to
it.” (24)
38
On the one hand, Davis reasons, it is implausible to present all controversial views as merely
biased. Agents can be quite sincere and competent in offering justificatory reasons for their nurtured
beliefs, and we often cannot detect overt bias or cognitive distortions in the reasons they offer.
Davis (2009) thinks that judgment can be based on the considerations the agent claims as her
reasons or justification even when those reasons depend on personal traits. On the other hand, Davis
points out, there remains a deeply worrisome relation between a person’s actually-held belief and
the counter-factual consideration that the agent would not take those considerations to justify that
belief if his or her ‘epistemic location’ were different than it actually is. I agree with Davis that it is
implausible that either on a moral evidentialist or epistemic evidentialist basis, guidance on doxastic
responsibility given to agents should demand strict suspension of nurtured beliefs. The treatment
both of epistemic assessment and of guidance-giving needs to be more contextual than this, and as I
will add, these two forms of normativity need to be more carefully distinguished.
39
Thanks to Adam Carter for bring Max Kolbel’s work (2004) and the relativist/contextualist view
to my attention. It may be that all domains of controversial views are only minimally truth apt, but
even if this is it does not imply relativism. I am using ‘faultless’ in a sense of the bounds of
reasonableness. I could concede that that must be some error when anyone comes to a false belief,
since we should not reduce error to culpable fault.

40
See especially Bishop 2007a and b on how moderate fideism can constrain doxastic ventures. Some
degree of failure to apply inductive norms is common-place in our thinking. See also Erik Baldwin
and Michael Thune (2016) on the relationship between experiential and testimonial transmission
account of warranted religious belief.
41
Kevin Wallbridge (2018) argues that while inductive knowledge may not be strongly sensitive, it
is weakly sensitive. The point is supportive of the importance of the insensitivity of counter-
inductive thinking to epistemic rationality or reasonableness, even though I do not take sensitivity
as a general necessary condition on knowing.
42
Insensitive belief that is not based upon shared facts but rather on private intuitions or other
subjective factors cannot claim the same reasonableness as the belief that I am not radically
deceived. My entitlement for the latter does not imply entitlement to the former, potentially much
more temperamental choice.
43
I thank Patrick Bondy for suggesting this formulation of the difference.

44
James Fritz (2018) argues that there no easy or compelling root from pessimism about moral
deference to steadfastness about moral disagreement.
45
Some will reply that abiding by inductive norms is the council of caution, but that a council of
courage is just as or more legitimate in some domains like that of religion. But our account already
acknowledges pragmatic reasons for beliefs in CV domains; it targets only self-deception and overt
bias mirroring, while allowing some influence of personal temperament, as quite reasonable and
permissible. To the principled agnostic it says, ‘Seek your ataraxia your own way’; to the political
ideologue, the moralist, and the religious enthusiast it says, ‘Seek your personal perfection, but
don’t let your faith venture risk me or others unjustly.’ Interestingly, Carter utilizes the dual-aims
point, but neglects responding to Kelly (2013) who had earlier used it as a central reason to support
permissivism over impermissivism.
46
DiPaolo and Simpson continue, “Indoctrination consists in the use of educational practices that
serve to impart something like absolute and inflexible acceptance. To indoctrinate is to ‘teach
someone to fully accept the ideas, opinions, and beliefs of a particular group and to not consider
other ideas, opinions, and beliefs.’”
47
Confabulation is arguably of special philosophic concern when it manifests in connection with the
holding of controversial views for which there are strong aetiological challenges. If we fill in that
that “believes that p” in the above is a belief about a trait asymmetry between Jan and another
person or persons, then we can see that asymmetry and confabulation are often found combined.
Rationalizing an asymmetric ascription or explanation on weak rational grounds invites overt

confabulation on the part of the agent, and perhaps the more so as it incites psychological
contingency anxiety or another form of cognitive dissonance.

48
William Hirstein (2005), 209 and 187. Compare Lisa Bortolotti: “When people confabulate they
ignore some of the psychological processes responsible for the formation of their attitudes or the
making of their choices, and produce an ill-grounded causal claim when asked for an explanation”
(2018, 235).
49
Pronin, Kruger, Savitsky, and Ross (2001), 639. See also Pronin et. al. 2002; 2004.
50
Pronin, Gilovich, and Ross (2004, 781).
51
The strong connections between failure of belief ownership and appeals to luck should be
obvious. Bortolotti and Broom argue that “by appealing to a failure of ownership and authorship we
can describe more accurately the phenomenology of thought insertion, and distinguish it from that
of non-delusional beliefs that have not been deliberated about, and of other delusions of passivity.”
Breyer and Greco (2008), in their account of the epistemological importance of cognitive
integration and the ownership of belief, hold that a belief is well-integrated in the way that brings
abillity and epistemic credit to the agent, not only if the subject owns the belief, but also only if the
(real or putative) process or ability is not subject to any defeaters to which the agent has access.
But contrary to these authors, I argue that counter-inductive methods of belief-formation have
defeaters of which the agent has access. Disowning grounds for the truth or justification of belief of
one’s belief, against inductive pattern, is itself a most serious violation of the second condition of
the absence of accessible defeaters.

52
Pronin, Lin, and Ross (2002), 369.

53
Kidd (2016), 183.
54
If we were to determine that this is the case with all extent beliefs in the domain, then even
though evidence always plays some significant role, it might suggest to neutral observers that we
are dealing with a domain of discourse that, unlike straightforward empirical domains, is only
minimally truth-apt. So trait-dependent overdetermination and the grounds for anti-realism about
discourse in a domain, are also interestingly linked. Axtell (2019) also looks at this in terms of what
conditions might lead one to legitimately move from a de jure to a stronger de facto objection. For
example, when might one move from a Humean criticism of the rationality of testimonially-
grounded belief in one purported miracle event, to a stronger metaphysical or de facto
(‘debunking’) claim that the best explanation of the epistemic shortcomings of many/all miracle
beliefs is that there are no miracles. But there as here I leave the legitimacy of these shifts as open
questions, since my point is that there is no direct inference to be made, but only an argument based
on inference to the best explanation. So my aim is merely to articulate conditions under which they
gain in their plausibility.
55
See Haidt (2012). Davis distinguishes broad and narrow traits, and argues that one’s belief “is
properly based when such traits operate just like broader traits whose basing relationship is not
disputed” (2009, 36). Broad traits help one form many uncontroversially true beliefs, and when they
do so reliably “are good candidates for faculties, capacities, or epistemic virtues” (30). Getting
credit for true belief through well-integrated cognitive abilities is indeed earned credit. It is not
credit on the cheap as we find in cases so insensitive and unsafe that “warrant” (positive epistemic
status) for the agent would disappear if the belief were not presumed true. Plantinga for example
concedes that what he terms “warranted” Christian belief would not be warranted, by his own
account of the term, were the content of the claim not true. So warrant simply follows the purport of
truth in metaphysical claims, and that truth itself is a felix culpa. In my (2019) I develop Plantinga’s
concession as indicating a self-described externalist apologetic that ‘leans on luck,’ and fails to
accept the full implications of externalism for epistemic assessment.
56
Examples of philosophers taking these positions are McCain (2008) for phenomenological
conservativism, and Fantl (2018) and Tucker (2010) for dogmatism.

Synthese (2007) 158:363–383
DOI 10.1007/s11229-006-9045-9
O R I G I NA L PA P E R
Two for the show: Anti-luck and virtue epistemologies

in consonance
Guy Axtell
Published online: 10 October 2006

© Springer Science+Business Media B.V. 2006
Abstract This essay extends my side of a discussion begun earlier with Duncan
Pritchard, the recent author of Epistemic Luck. Pritchard’s work contributes signifi-
cantly to improving the “diagnostic appeal” of a neo-Moorean philosophical response
to radical scepticism. While agreeing with Pritchard in many respects, the paper ques-
tions the need for his concession to the sceptic that the neo-Moorean is capable at
best of recovering “‘brute’ externalist knowledge”. The paper discusses and directly
responds to a dilemma that Pritchard poses for virtue epistemologies (VE). It also
takes issue with Pritchard’s “merely safety-based” alternative. Ultimately, however,
the criticisms made here of Pritchard’s dilemma and its underlying contrast of “anti-
luck” and “virtue” epistemologies are intended to help realize his own aspirations for
a better diagnosis of radical scepticism to inform a still better neo-Moorean response.
Keywords Epistemic luck · Virtue epistemology · Virtue responsibilism ·

Skepticism · Neo-Moorean
“Who will unravel this tangle? Nature confutes the sceptics, and reason confutes
the dogmatists…[such that a person] can neither avoid these two sects, nor yet
hold fast to either one of them!”
Blaise Pascal, Pensees1
1 Introductory remarks
Over the past few years and culminating in his remarkable book Epistemic Luck
(2005), Duncan Pritchard has worked on various aspects of the problem of radical
1 Pascal’s Pensees #434, my own translation.
G. Axtell (B)
Department of Philosophy, University of Nevada,
Mail Stop 102, Reno, NY 89557 USA
e-mail: axtell@unr.nevada.edu
364 Synthese (2007) 158:363–383
scepticism, and in particular on improving the “diagnostic appeal” of a neo-Moorean

response to the sceptic.2 Pritchard shows his readers convincingly how discussions
not just over scepticism directly, but also over an indirectly-connected panoply of
Gettier, lottery, closure, etc., problems currently popular in mainstream analytic epis-
temology, often treat epistemic luck vaguely or ambiguously. Still more to the point,
these discussions misstate the motivations for radical scepticism, by focusing more
around “veritic” than around “reflective” epistemic luck; so for philosophers who
engage them in order to confront the sceptical challenge, these issues can become
stumbling blocks rather than footholds. Pritchard sees philosophers creating certain
unnecessary stumbling blocks for themselves especially through individual and com-
munal adherence to what he calls the “Epistemic Luck Platitude” (hereafter ELP).
The “robust” version of this platitude avers a blanket incompatibility of knowledge
and luck; while deeply unsound in this form, Pritchard shows how explicit or tacit
loyalty to it underlies many of our analytic practices and discussions.
Some years ago I wrote about epistemic luck as a virtue epistemologist prior to
encountering Pritchard’s work.3 While agreeing with Pritchard in what has been said
thus far and much else in regards to his neo-Moorean approach to scepticism, the
encounter solidifies my sense of luck (ethical or epistemic) as a natural focus of study
in virtue theory. Yet this outcome will be ironic for him, since it chafes against a direct
challenge to the motivations of virtue epistemology (VE) that Pritchard claims to flow
out of his book’s study of epistemic luck. These differences may have a metaphilo-
sophical aspect over and above their epistemological ones, since it seems certainly
the case that interest in moral luck and a resurgence of virtue (or areteic) ethics have
been substantially coextensive over the past several decades. While some epistemic
analogues both of moral luck and of virtue-based ethical theory have made significant
inroads in contemporary analytic epistemology, the ability of philosophers to respond
satisfactorily to the challenge of scepticism would be enhanced by carrying these anal-
ogies further. Or so I argued in “Felix Culpa” (2003), but this second and far fuller
response to Pritchard’s work on radical scepticism reflects my view as now informed
by the reading of his stimulating book.
The main argumentative thesis of the paper is that what Pritchard calls the diagnos-
tic appeal of his account of the motivations for radical scepticism would be significantly
enhanced had he added that there is a second major epistemological “platitude” that
should be exploded at the same time as we explode the ELP. This second, equally
uncritical and debilitating commitment among epistemologists, is to what will here be
termed the Platonic Definition Platitude (hereafter PDP). If ELP is the unqualified
assumption that knowledge-possession is incompatible with epistemic luck, then PDP
is the similarly unqualified assumption that the two-fold goal of “philosophical” (qua
anti-sceptical) analysis of knowledge is its complete definition, preliminary to showing
that that definition allows creatures like ourselves to know most of the things we think
we know.
By a complete definition is meant the attempt to non-circularly state a completely
general as well as counter-example free set of necessary and sufficient conditions for
knowledge. Just as Pritchard examines what purposes are thought to be served by
2 See Axtell (2003) for the first response, which was to Pritchard (2003). See also Pritchard (2002)
for more about recent work on radical scepticism.
3 Axtell (2001a). My view also acknowledged Internalism and Externalism as directed towards
distinct kinds of luck, and argued an epistemic “compatibilist” position on that basis.
Synthese (2007) 158:363–383 365
holding what he calls the “robust” version of ELP, I want to question the purposes
thought to be served by allegiance to a robust version of PDP.4 With an eye towards
radical scepticism, my main source of concern is the wisdom behind the intention
of using Platonic definition as a bulwark against radical scepticism. Sections 1 and 2
develop the basic claim that we should disavow PDP for many of the same reasons
of philosophical advantage that lead Pritchard to think that we should disavow ELP.
While it should be made clear that Pritchard’s continued commitment to PDP isn’t
here taken to be any stronger than that of most other contemporary epistemologists,
it will be argued that we can find such a commitment evident in his writings. It is
discernable, firstly, in the conspicuous absence of recognition, within his diagnosis, of
a certain important kind of scepticism: the distinctively “philosophical” kind which
Stroud (2004a, 2004b) and Fogelin (2004) both write extensively about. After argu-
ing this in Sect. 1, Sect. 2 argues that the same dubious commitment to PDP is also
apparent in the more glaring misdiagnosis Pritchard gives of the motivations for VE.
The general upshot of these early sections is that we will be better off in our
debates with the radical sceptic by inviting the dialectical repositioning that results
from abjuring both platitudes together. Later sections extend this claim by responding
to a specific dilemma that Pritchard’s poses to dilemma for VE, explaining why dis-
abusing ourselves of the twin platitudes—the one about the blanket incompatibility of
luck with knowledge and the other about Platonic definition of knowledge being the
goal of analysis—might be seen as serving Pritchard’s own ends by allowing improve-
ment to the neo-Moorean argument. Pritchard doesn’t well account for the fact that
the neo-Moorean position and a safety principle are already part of the position of
epistemologists such as Ernest Sosa and John Greco, whom he criticizes. In the last
section of his book, Pritchard expresses deep doubts about whether the neo-Moorean
argument as he develops it can be successful in recovering from the shadow of scep-
ticism more than a “‘brute’ externalist knowledge”. This concession may be one that
flows from his adherence to the “merely safety-based” form of epistemological exter-
nalism that he develops, and a sharper thesis running through both parts of the paper
is that it needn’t be made: neo-Mooreans can better rebut the sceptical challenge if
they preserve the resources of VE and maintain, as Pritchard does not, its distinctive
“compatibilist” stance on the relationship between those interests in explanation we
typically label as “internalist” and “externalist”.
2 From the Sampson principle to PDP
This section briefly pursues one particular concern about the diagnostic appeal of
Pritchard’s account of the motivations for radical scepticism. That concern is that his
diagnosis can’t be correct or even close to correct, if it overlooks a type of radical scep-
ticism prominent in the literature. I refer to philosophical scepticism, and will want
to show that Pritchard’s overlooking it reveals the extent to which he is committed to
the robust version of PDP.
4 The two platitudes are in fact intimately connected on my view. It is because the ELP serves the goal
of Platonic definition, as witnessed especially in search of a Gettier condition free from counterexam-
ple, that we witness as T. Williamson puts it, “the more or less ad hoc sprawl that analyses have had
to become” (2000, p. 31). Williamson also argues that knowledge cannot be analyzed into a complete
set of necessary and sufficient conditions, and that the post-Gettier literature strongly suggests that
there is no good reason to expect non-circular necessary and sufficient conditions for knowledge in
terms of truth, belief, and other factors.
366 Synthese (2007) 158:363–383
Fogelin (2004) writes of what he calls the “Samson Principle”, and how it frames
a central tenet of the neo-Pyrrhonism that he himself accepts. This is, “The sugges-
tion…that the epistemological enterprise, when relentlessly pursued, not only fails in
its efforts, but also, Samson-like, brings down the entire edifice of knowledge around
it”. (2004, p. 164) The Sampson Principle is strongly seconded by Stroud (2004a), who
writes that Fogelin’s “updated Pyrrhonism”, like his own, is “a form of philosophi-
cal scepticism. That is, it is a sceptical or negative response to something that arises
in philosophy” (2004a, p. 175; italics in original). It arises Stroud says, especially in
“philosophical epistemology”, that
…tries to account for human knowledge of the world in general, or at least
for as much of it as can be accounted for in completely general terms. It is the
attempt to explain how…our beliefs about the world are in general justified or
warranted or well supported on the basis of the grounds we have for holding
them. (p. 175)
To further explain this as a key characteristic of his updated Pyrrhonism, Stroud goes
on to say that the Agrippan Trilemma is invoked not against all belief, but rather
“against the epistemological enterprise of showing how our beliefs in general are
justified on the assumed basis”. (p. 175) The ‘assumed basis’ as Stroud uses this term
mentions no analysis of knowledge in particular, which we must think makes his
intended target something closer to a ‘platitudinous’ commitment associated with an
anti-sceptical import attached to a fully general definition of knowledge.
Pritchard’s account therefore seems suspect to the extent that he misses the
importance of philosophical scepticism to some of the best-known proponents of
neo-Pyrrhonism today. But the same thought should suggest that Pritchard pays too
much deference to PDP and in so doing commits the dialectical mis-step alluded to
earlier. Now philosophers may of course hold themselves to be primarily interested
in “conceptual analysis”, that is, in pursuing a complete set of (individually necessary
and jointly sufficient) general conditions in order to gain greater clarity regarding
terms such as knowledge and justification. But while the interest in conceptual anal-
ysis might be thought initially independent of an anti-sceptical benefit, our epistemic
practices themselves testify to the strong ties holding between them. One may be
initially interested in general conditions on knowledge for the sake of concept clari-
fication, employing a sceptical interlocutor only in order to ‘push oneself’; but when
being unable to provide an exception-free definition comes to be taken as indication
of a failed analysis, then the thinker is shading quickly into the pursuit of Platonic
definition as a bulwark against radical scepticism.
This kind of subtle shading in PDP from something that seems innocuous to some-
thing that is in fact highly debilitating is an obvious aspect of ELP as well, and further
testifies to the symmetries between them. The PDP has its “dissatisfied” in addition to
its “robust” version, just as Pritchard shows us ELP does.5 In this case the dissatisfied
or disappointed version is simply the argument that the recurrent failure of episte-
mologists to provide Platonic definition of knowledge is tantamount to the triumph of
scepticism. Even subtle shading from conceptual analyses to anti-sceptical bulwarks
seems to be motive enough for our neo-Pyrrhonists to counter-act the philosophers’
5 I emphasize more strongly than Pritchard does that the sceptic is committed to a version of the
EL platitude as well: Not of course the “robust” version associated with anti-sceptical thought, but
what we may view as its alter-ego, its disappointed or “dissatisfied” foil. I will show how the sceptic is
committed to a “dissatisfied” form of the PD platitude as well.
Synthese (2007) 158:363–383 367
robust version of PDP with arguments that support its (equally platitudinous) mirror
image, the dissatisfied version. It is useful to remind ourselves then that the converse of
interest in conceptual analysis is disinterest in conceptual analysis, while the converse
of commitment to robust versions of ELP and PDP is commitment precisely to the
dissatisfied versions of each platitude as a key motivation behind radical scepticism.
It is clearly strong anti-sceptics who hold each of their robust versions, and when it
is ELP that is held in this way Pritchard says that it supports “infallibilism”, and so
also drives forward the well-known infallibilist arguments for radical scepticism. This
is quite right, but seems also to connect with and even to damn the motivation for
epistemologists to produce an “infallibilist analysis” of knowledge such as Pritchard
himself seeks, i.e., one on which the conditions in addition to true belief are intended
to entail the truth of the target belief. Even a tacit adherence to robust PDP has a
similar inflammatory effect than an endorsement of robust ELP: for nothing more
inflames the sceptic’s incredulousness, than hearing once again from his “philosoph-
ical friend” that the promissory notes he was issued are fulfilled at last in his newest
analysis.
Pritchard’s lesson of disabusing ourselves of uncritical acceptance of the ELP, so
as to get away from the Gettier problem’s tunnel-vision on veritic luck and move
on to more subtle investigation of different kinds of luck and their differential bear-
ing on knowledge-possession, is extremely valuable. But PDP seems to be just as
deeply ensconced in analytic epistemology as ELP. When Pritchard fails to address
“philosophical” scepticism as a key motivation of the sceptic, and ends with the final
position of offering (with apologies) a “pragmatic” rather than an “epistemological”
response to radical scepticism (Chapter 9), the consequences are obvious: This can
only serve to confirm to Stroud or Fogelin that he like so many others who practice
analysis of knowledge and engage the Gettier-game for its anti-sceptical value, has
only succeeded in bringing his citadel of knowledge down around him.
The argument is not that it is terribly difficult to provide the sceptic with a ‘general
understanding’ of the human capacity to know, but rather that we do ourselves a
grave disservice to assimilate giving a philosophically satisfactory response to scep-
ticism with providing a Platonic definition of knowledge as characterized above. We
wouldn’t constantly ‘fail’ that task of non-circular Platonic definition if it were one
that could be reasonably declined, and that is where the present criticism of Pritchard
begins. For he is one of many who might have faired better in their response to radical
scepticism had he heeded Michael Williams’ (2005, p. 214) point that if there are a
plurality of epistemic aims, or a plurality of ways in which justification might arise,
then “the sceptic’s hyper-general questions may be deeply flawed”.
Pritchard is right that in order to make better headway, we first need to get more
bang for the buck out of a study of epistemic luck and how it is related to analysis of
knowledge on the one hand, and to motivations for radical scepticism on the other.
But in seeking to show that there is more to this than he allows, I am led to argue that
despite his achievement Pritchard falls just exactly one platitude short. If epistemolo-
gists are intent on practicing analysis of knowledge for its value as a bulwark against
scepticism, then Pritchard’s rebuke of ELP still constitutes a step forward: It allows
us to acknowledge the distinctive kind of epistemic luck that most impassions the
sceptic—reflective luck—and therefore to replace our constant pre-occupation with
veritic luck (in the Gettier problem and its offshoots) with issues that directly engage
the more genuine (and more potent) sceptical concerns. I imagine Pritchard, claiming
to have solved the Gettier-problem at long last by his modalized safety principle and
368 Synthese (2007) 158:363–383
thereby advanced the quest for Platonic definition, metaphorically striking the chord,
its ‘One (platitude) for the Money’! But if instead we would be witness to a transfor-
mation in the theory of knowledge such that the anti-sceptical philosopher no longer
confirms the Sampson Principle against his own best intentions, then I hold we must
choose to strike a second chord, crooning that its ‘Two for the Show’!
3 Virtue responsibilist externalism: A declaration of independence
Epistemic luck plays the central role in a dilemma Pritchard puts to virtue epistemol-
ogies both of a “reliabilist” (Ernest Sosa’s virtue perspectivism; John Greco’s agent
reliabilism) and of a “neo-Aristotelian” character (Linda Zagzebski’s “moral model”
or “pure virtue theory”). In this section his challenge to the motivations behind VE is
examined and a response to it is begun. Pritchard thinks the motivations for VE are
placed “on ice” as a result of his distinction between veritic and reflective luck, and
subsequent division of everyone in terms of it. One horn of his dilemma is addressed
to each, with the first horn addressed to the reliabilists:
If one has externalist intuitions about knowledge, then one should seek a mere
safety-based theory of knowledge that will, no doubt, be supplemented by a
further explanatory story concerning the epistemic virtues and cognitive fac-
ulties that explains how agents gain safe beliefs that are not veritically lucky
(p. 198).
The dilemma thus far depends upon characterizing externalist VE as concerned

exclusively or primarily with the exclusion of veritic luck, though that as we will see
may stem from his own bifurcation of explanatory interests, rather than theirs. But
let’s first set the full dilemma before us by stating its second horn:
[I]f one has internalist intuitions about knowledge, then one should seek an in-
ternalist safety-based theory of knowledge that will, no doubt, be supplemented
by a further explanatory story concerning the epistemic virtues that explains
how agents gain safe and internalistically justified beliefs that are neither veri-
tically nor reflectively lucky. Either way, one is left with a non-virtue-theoretic
account of knowledge and, far from motivating the virtue-theoretic position in
this regard, reflection on the role of epistemic luck merely highlights the juncture
at which the virtue epistemological thesis goes awry. (p. 198)
By “virtue epistemological thesis” we should be clear that Pritchard addresses this

dilemma specifically to those who place an areteic condition into their analyses of
knowledge; it challenges just the contention that the concept of intellectual virtue
belongs inside analysis of knowledge. For the sake of clarity and development, let’s
refer to any such defence of an areteic condition as “Strong VE”. Virtue reliabi-
lists like Greco and Sosa have championed Strong VE, and Zagzebski (1996) has as
well, though from a perspective she describes as non-reliabilist and which Pritchard
describes as internalist. I will later make my own argument for Strong VE by relating it
to “mixed externalism”, or as can be more simply and elegantly put, to “epistemic com-
patibilism”. But the point here is that Pritchard leaves the door open (and elsewhere
appears sympathetic) to “Weak VE”, if that describes a thesis that virtue-theoretical
Synthese (2007) 158:363–383 369
concepts maintain an epistemically central role insofar as they provide explanatory

stories that run supplemental to philosophical analysis proper. Between analysis of
knowledge and such attendant explanatory chores Pritchard, whether justifiably or
not, maintains a sharp distinction.
Pritchard’s notion of “an internalist safety-based theory of knowledge” is another
thing that appears problematic in his statement of the dilemma, and so bears closer
scrutiny. We aren’t menaced with a sharp horn if the point is unclear, one might say;
but in saying that it catches Zagzebski and other “neo-Aristotelians”, Pritchard seems
to mean two things: the first is that because her view is one that supports the ‘fuller
sense’ of cognitive achievement involved in reflective knowledge that he associates
with epistemic internalism, “We should expect the motivation for most virtue epis-
temologies of this sort to be susceptible to the same diagnosis”; secondly, Pritchard
means that Zagzebski can produce a self-consistent analysis of knowledge only if she
adopts his safety principle (thereby fully eliminating veritically lucky knowledge),
and then insists further upon “the necessity of adding via the focus on the epistemic
virtues alone, an internal epistemic condition to the view” (p. 195).6 By this second
claim Pritchard appears to mean that neo-Aristotelian VE necessarily becomes for-
mally internalist because it asserts ‘additionally’ something very much like a general
condition demanding exclusion of reflective luck (luck from the first-personal per-
spective) at any and all points along the high-low or reflective-brute spectrum of
human knowledge. The reasoning seems to be that virtue epistemologists of this
stripe will routinely reject the “sufficiency” for knowledge of analyses that lack such
an additional condition, which when included render it internalist.
But if adding such a general internalist condition is not a demand shared by the
form of Strong VE advocated here, then these and related points can be used to argue
that its motivations aren’t susceptible to the same diagnosis as Pritchard applies to
Zagzebski. So the task before us is set: like a Spanish forcado we are challenged to face
the charging horns of a bull and vault directly between them, landing still back upon
our feet if we do so with any grace. Pritchard’s dilemma constructs a divide between
externalist and neo-Aristotelian VE, and says that everything on the former side of
that divide essentially aims to incorporate “a role for responsibility by demanding
internalist justification,” which means in his diagnostic approach that they are simply
motivated by the impossible desire to exclude reflective luck. Those on the other side
are concerned only to exclude veritic luck. I grant that it is easy to confuse the concern
with personal justification cashed out in terms of motivation and responsible modes
of inquiry, with the demands that internalists and sceptics make for the exclusion of
reflective luck; perhaps none of us have been immune to such confusions. I share the
numerous criticisms of Zagzebski that she ‘thickens’ her conception of intellectual
virtue in such a way as to demand internalist justification, though it is of course an
inconsistency rather than an overt “motive” on her part to do so since she explicitly
6 Here is the full quotation:
“We should expect the motivation for most virtue epistemologies of this sort to be suscepti-
ble to the same diagnosis, in that their underlying concern regarding agent reliabilism is that
whereas the reliabilist element of the view deals with veritic epistemic luck…it won’t capture
the fuller sense in which knowledge is a cognitive achievement of the agent. This ‘fuller sense’
of cognitive achievement involves, of course, not just the elimination of veritic epistemic luck
but also the elimination of reflective epistemic luck—hence the necessity of adding, via the
focus on the epistemic virtues alone, an internal epistemic condition to the view” (p. 195).
370 Synthese (2007) 158:363–383
presents her view as “mixed externalism.”7 Such an inconsistency, however, is slim

grounds for Pritchard to base his claims about the motivations of even her form of VE,
let alone those of the virtue responsibilists. He largely overlooks the focus on proper
motivations and quality of effort to procure evidence in their accounts of epistemic
responsibility, something that quite a number of writers have correctly pointed out
that access internalism itself tends to be blind to. Because internalists often describe
justification in terms of a time-slice view of what reflectively good reasons for belief the
agent has by her ‘own lights,’ they often ignore altogether the crucial “responsibilist”
focus with the quality of that agent’s evidence-gathering and effort.8
To check the footing before our forcado makes his vault, it should also be pointed
out that in framing the dilemma Pritchard employs the standard stipulative definitions
of internalism and externalism, wherein they are mutually exclusive and exhaustive
accounts of the same epistemic concept. Yet I’ve always thought that one of the selling
points of VE in all of its versions is that it empowers what Foley (2004) calls “more
interesting readings” of the dispute: that instead of being competitors in the stan-
dard sense, the interests in explanation that we (mis?) label internalist and externalist
“intuitions” are better construed revisionistically as concerned with different issues,
or again with different aspects of a common set of issues. If stipulative definitions
fail to capture what is really at issue between actual disputants, their usefulness fades
quickly and they become roadblocks to inquiry. Pritchard should perhaps attend to
Dewey when he writes,
[T]he conviction persists … that all the questions that the human mind has asked
are questions that can be answered in terms of the alternatives that the questions
themselves present. But in fact intellectual progress usually occurs through sheer
abandonment of questions together with both of the alternatives they assume?an
abandonment that results from their decreasing vitality and change of urgent
interest. We do not solve them: we get over them. Old questions are solved by
disappearing, evaporating, while new questions corresponding to the changed
attitude of endeavor and preference take their place. (1970, p. 402)
So if it is a useful exercise to take internalism and externalism as ultimate categories
into which various forms of VE can be tossed, we need an argument for this, especially
where, as in this case, the figures Pritchard discusses are all quite explicit supporters of
epistemic compatibilism. In lieu of such argument, invoking the power of stipulative
definitions borders upon a merely dismissive treatment of their views.9
An antidote to this might be to extend our alternative taxonomy, and I would
frame it in terms of competing Conflict, Independence and Integration models of
the relationship between the pertinent interests in explanation. Once we see that
all three of these models are represented in the post-Gettier literature, it becomes
7 Greco (2003) is quite correct in criticism of Zagzebski that while the moral model is well-suited
to an account of understanding, “it is a mistake to generalize from such concerns to an account of
knowledge, per se,” since it results in conditions on knowledge that are too strong and thereby invite
skepticism.
8 Goldman points out that the construal of internalism as a matter of the subject’s “having rea-
sons” (by his own lights) for his belief is inconsistent with this focus on conscientious effort in ethics.
Kornblith and Montmarquet hence view virtue responsibilism as consistent at best with only specially
qualified forms of internalism.
9 Indeed as seems apparent, the virtue epistemologists as epistemic compatibilists really care only
for some remainder of the old internalist/externalist debate, most notably the relationship between
epistemic reliability and epistemic responsibility in human knowledge.
Synthese (2007) 158:363–383 371
obvious that unquestioned employment of the standard definitions of internalism and

externalism actually favours the proponents of one of these models in particular, the
Conflict model. This is so because the Conflict model among those who manifest it
in their writing, such as Bernecker (2006), explicitly rejects the thesis of epistemic
compatibilism.10
If what we call the internalist/externalist debate manifests these three different
models, Pritchard may well want to join Bernecker in arguing explicitly in favour
of Conflict; but again we should not allow his stipulative definitions to beg the sub-
stantive issues of those holding one of the two compatibilist models, “Independence”
and “Integration”. Isn’t it clear that the virtue epistemologists are epistemic compa-
tibilists, and that this is to a great extent the distinctiveness of their approach? Sosa
(2004) calls internalism and externalism, and coherentism and foundationalism “two
false dichotomies”, and Greco and Zagzebski, whatever their other differences, both
emphasize that the “mixed” character of VE is philosophically crucial rather than
detrimental to its ability to respond to the sceptical challenge.11 Strong VE can now
be seen as associated with strong compatibilism, i.e., with an Integrationist stance, and
in my own case quite explicitly so since I hold that our prospects of giving a satisfac-
tory philosophical response to radical scepticism are inextricable from our ability to
maintain the philosophic stability of epistemic compatibilism. But even discounting
that claim, it should be apparent from what has been said that Pritchard’s fault in
ignoring a concern for personal justification even among virtue reliabilists stems at
least in part from his own expressed temptation towards the view that knowledge is
nothing but safe true belief.
Pritchard’s dilemma is drawn in terms of a distinction between forms of VE taken
as exhaustive, with others besides the figures discussed judged “susceptible to the
same diagnosis.” Although we have cast doubt on the sharp dichotomy between
externalist and internalist VE on which it is based, I still intend to present a full
response by articulating motivations for a “responsibilist” VE with the right stuff
to pass ‘between the horns’, even if it were right to say that it presents a genuine
problem for other forms of VE. Those we can call the “virtue responsibilist posse”12
have substantially different concerns about extant forms of virtue reliabilism than
those that Pritchard attributes to Zagzebski. As a prime example, if the requirement
of specifically acquired or reflective virtue is what makes Zagzebski an internalist in
Pritchard’s view, this is something the virtue responsibilists have been explicit critics
of her over (Axtell, 2001a, b; Baehr, 2006a; Battaly, 2001), arguing that it works to the
detriment of the responsibilists’ concern to lay the foundations for a unified research
program for the reflective virtues. In my own papers, the responsibilist orientation
reflects a special but non-exclusive concern with active epistemic agency, and with the
“zetetic context” of an agent’s reflective and investigative activity; to engage these
interests in intellectual responsibility and the quality of zetetic activity, even insofar
as they do naturally bring up questions about cognitive achievement in knowledge,
10 We should add quickly however that rejection of compatibilism can easily stem from internalism
as well, so that the Conflict model can be motivated by internalists quite as much as by externalists
like Bernecker.
11 For a further development of this argument, see Axtell (2007a, b).
12 When I speak of the posse, I have in mind not only those who adopt the terms explicitly, but
those who seem to me (somewhat vaguely I concede) to bear its stamp, authors such as Jason Baehr,
Heather Battaly, Kelly Becker, Lorraine Code, Juli Elfin, Catherine Elgin, Chris Hookway, Charlotte
Katzoff, Adam Leite, James Montmarquet, Robert Roberts, and Jay Wood.
372 Synthese (2007) 158:363–383
does not for that reason imply commitment to some general internalist condition on
knowledge.13
Yet if I am right that my “responsibilist externalism” is a version of Strong VE with
substantially different motivations than those Pritchard attributes to Zagzebski, then
Pritchard can be counted on to object that it avoids the second horn only to impale
itself on the first. His argument against the motivations for virtue epistemologies
of Sosa and Greco’s reliabilist orientation now catches up to our would-be forcado,
unless this new responsibilist externalism can equally-well distinguish its motivations
from those of extant forms of virtue reliabilism. It was to this horn of the dilemma
I earlier responded in “Felix Culpa”, arguing that there is little more to Pritchard’s
charge that virtue reliabilisms are “radical”, than that he thinks the safety principle
can go it alone in “a merely safety-based theory”, in contrast to Sosa and Greco, who
do not.14 But if we are to allow for the sake of argument that Pritchard does present a
problem for the virtue reliabilists such that ‘grabbing’ this first horn isn’t useful after
all, then more needs to be said regarding how a responsibilist externalism might set
itself apart from ‘reliabilist VE’ just as it did from ‘internalist VE’.
Greco’s term “agent reliabilism” seems congenial enough to responsibilist interests
in explanation, since neither of his two terms needs to be taken as primary. Yet admit-
tedly VE’s short history has been one in which “faculty virtue theory” has remained
largely unconversant with “reflective virtue theory”. A reliabilist focus on causal con-
ceptions of epistemological grounding works best with what Pritchard describes as
‘brute’ externalist knowledge, and comports with the idea that knowledge must either
be ‘easy’ or ‘impossible’. Conceiving knowledge as a spectrum, causation doesn’t come
to an end when we move further up the scale from animal to reflective knowledge, yet
our ability to speak to the issue of process-reliability must quite apparently become
curtailed. So surely it is too easy for us to say that things like divergent interests in
explanation and selectivity biases towards cases at different ends of the spectrum of
human knowledge can explain the continuation of something like the internalist/ex-
ternalist debate even within VE. Whether the Conflict model gets motivated from the
one side or the other, the philosophical systems, even if starting out as little more than
professional biases, soon enough grow tentacles to present themselves as far more
endemic incompatibilities.
To persuasively preach the via media, then, virtue responsibilists must make clear
the philosophic advantages of epistemic compatibilism, asserting as its precondition
the fulfilment of what we can call the comfortable home demand. This demand is
that their own broadly normative interests in explanation find legitimacy and sup-
port within a naturalistic account of human action, knowledge, and understanding. It
has a substantial and fairly direct analogue in metaethics. While responsibilists can
be epistemic internalists or externalists, those of the latter sort will comport with
reliabilist externalism only on condition that it is able to meet this comfortable home
13 Concerns which virtue responsibilists share with Zagzebski include wanting conditions on knowl-
edge “that are both theoretically illuminating and practically useful” (1996, p. 264); also in her taking
the term knowledge broadly, “to cover a multitude of states, from the simplest case of ordinary per-
ceptual contact with the physical world, requiring no cognitive effort or skill wherever, to the most
impressive cognitive achievements” (p. 264). See Elfin (2003) for a virtue responsibilist development
of these themes.
14 Becker (2006 forthcoming) has cogently argued that some forms of the safety principle implic-
itly entail a kind of reliability while other forms avoid counter-examples only when combined with
reliabilism.
Synthese (2007) 158:363–383 373
demand. While the externalist turn in epistemology currently presents obstacles to

the continued recognition of the conceptual relevance of active agency and “moti-
vation” to knowledge possession, the virtue responsibilist is an optimist who thinks
these roadblocks will prove only temporary. But optimism is strained if the inter-
nalist/externalist debate is only re-hashed within VE as Pritchard appears to hold;
and indeed, my own impatience with reliabilist externalism meeting the demand is
compounded by recent criticisms such as those of Bernecker to extant forms of virtue
reliabilism.15 Bernecker (2006) rejects Sosa’s and Greco’s virtue-based analyses of
knowledge in just the sense that Greco (2005) describes VE as having a distinctive
character as “mixed theory”.16 Perhaps then, as Bernecker has put it, the “general
lesson to be learned from the critique of virtue perspectivism is that internalism and
externalism cannot be combined by bifurcating justification and knowledge into an
object-level and a meta-level and assigning externalism and internalism to different
levels”.17
This means that potentially there is a deeper dialectical breach between respon-
sibilist externalism and epistemologies of a reliabilist sort. Perhaps approaches such
as those of Sosa, and Greco cannot meet the comfortable home demand because the
virtue reliabilists’ own home is a house-divided. The “two-level” structure of Sosa’s
virtue perspectivism appears, not incidentally, as perhaps the most widely-voiced crit-
icism among contributors to the recent Ernest Sosa and his Critics collection, and
Bernecker pushes quite similar concerns against Greco (despite Greco’s (2004) own
criticism of Sosa’s requirement of epistemic “perspective” for reflective knowledge).
Must the virtue responsibilists, then, find the resolve to step two-footed into this
dialectical breech in order to maintain their own optimism over the prospects of epi-
stemic compatibilism? There is much that might be said on both sides of this issue, and
I think we can simply leave it as an open question whether responsibilist externalism
needs to distinguish itself very sharply from the virtue reliabilisms in order to respond
satisfactorily to Bernecker’s concerns and to Pritchard’s dilemma.
To summarize, if reliabilist forms of VE cannot support epistemic compatibilism
because they construct roadblocks to inquiry into the reflective virtues and fail the
comfortable home demand, then it would be consistent with my view to further sep-
arate my responsibilist externalism from them by arguing that their problems stem
from stronger remaining loyalties to the EL and PD platitudes. This means that
even if Bernecker’s criticisms of Sosa and Greco are on target, one needn’t share
his pessimism about the prospects of the compatibility thesis itself. Unless epistemic
compatibilism is judged dead in the water, another answer can be made: that the
‘torch’ of epistemic compatibilism in the sense of Sosa’s initial profound vision of
VE as an approach to reconcile foundationalism and coherentism, externalism and
internalism, passes on from its reliabilist to its responsibilist wing. But put in this way,
perhaps there is a serious dilemma for reliabilist VE even if it is not the one Pritchard
15 See Baehr (2006a, b), Bernecker (2006), and Foley (2004).

16 See also Greco (2004) for his own criticism of Sosa. See Baehr’s (2006b) responsibilist criticism of
virtue reliabilism that bears some overlap with Bernecker’s objections.
17 A fuller response to Bernecker is coming in Axtell (2007b). See Leite (2004) for criticism of
the “spectatorial theory” assumed in certain forms of reliabilist externalism, criticism that suggests
overlap with Dewey and other pragmatists like Sandra B. Rosenthal on the “spectator theory”.
374 Synthese (2007) 158:363–383
suggests: This more subtle dilemma is whether its proponents can answer the charges
of the instability of a “bi-level” analysis of knowledge, on the one hand, and yet meet
the responsibilists’ demand of a ‘comfortable home’ for reflective virtue theory and
epistemic normativity, on the other.
One might also consider at this point that the attempt to secure an epistemically
central role for the character virtues has led a number of self-described virtue re-
sponsibilists to turn towards “Weak VE,” wherein the study of the virtues is still
taken to be central in epistemology, though relocated outside of analysis of knowl-
edge proper. Weak VE sidesteps Pritchard’s dilemma for Strong VE, constituting
a safe and saving relocation for the study of what Baehr calls the “character” vir-
tues. So it would be a mistake to say that a responsibilist orientation in epistemology
entails my own endorsement of Strong VE. Weak VE still reflects a kind of epistemic
compatibilism, but of the Independence rather than the stronger Integration kind. It
accepts, perhaps, Foley’s repeated call for a “trial separation” between the theory of
knowledge and the theory of justified belief. Among Independence-minded authors,
the top candidates for a relocated theory of the reflective virtues today appear to be
(a) an account of how we gain safe belief as opposed to what knowledge is (Pritchard;
Umbers); (b) a theory of rational/justified belief (Foley) (c) a theory of knowledge
attribution (Reed); or (d) a theory of ‘epistemic value pluralism’ where it is falls under
an account of “understanding” or some other epistemic goal distinct from knowledge
(Fairweather; DePaul).
If I dissent from all of these Weak VE or Independence-minded forms of compatib-
ilism, it is because such compartmentalization strategies are oftentimes unstable and
ill-adaptive ways of dealing with a problem. One worry is that the study of the faculty
virtues and of the reflective virtues will be pushed into separate closets rather brought
together, and another is that these proposals, not having scepticism in view, appear
indifferent to the project of re-tooling the neo-Moorean argument as a response to
radical scepticism. So perhaps we could stop here, alleging to have described a re-
sponsibilist externalism not susceptible to the diagnosis Pritchard gives of other extant
versions of Strong VE. But given the admittedly hard line being marked out, Pritchard
may demand a fuller explanation of underlying ‘motivations.’ Secondly, if epistemic
compatibilists propose moving away from preoccupation with veritic luck and the
Gettier game, they had better have “replacement topics” to offer. And thirdly, the
integrationist ideas of responsibilist externalism that have been highlighted stand at
odds with certain assumptions commonplace in mainstream analytic thought, so that
my stance likely requires a sharper critique of ‘analysis as usual’ than proponents
of Weak VE might concur with. I prefer to voice such a sharp critique, joining Leite
(2004) in criticism of what (with shades of Dewey) he refers to as a “spectatorial” con-
ception of knowledge, than to countenance what appears to me as attachment among
many epistemologists today to a fallaciously sharp dichotomy between knowledge
and action.
In addition to suggesting the nature of such a critique, Sects. 4–6 also provide an
opportunity to mediate between Pritchard’s anti-luck or tucheic condition on knowl-
edge and the areteic condition of Strong VE. Section 4 initially resorts to Plato’s
technique of telling a mythos to help bring into life a idea that can then be explored
philosophically. In our case it is a story about where anti-sceptical philosophy has
been focused, and what might result for debates between sceptical and anti-sceptical
philosophers from the dialectical repositioning that follows upon discarding both the
PD and the EL “platitudes”.
Synthese (2007) 158:363–383 375
4 Has the Gettier game become philosophical Jumanji?
The Gettier problem arose in the mid-1960’s as a challenge to the JTB model of
knowledge, and the search for a “fourth condition” on knowledge led quickly in an
externalist direction. Externalist theories of justification were established when phi-
losophers began to take notice that the kinds of conditions best suited to responding
to Gettier’s challenge were ones where, as Greco (2004) puts it, “etiology matters”.
The etiology or causal history of a belief being something known from a third-per-
son perspective, it follows that not all of the conditions necessary and sufficient to
epistemicize true belief need be ones internally available to the agent upon reflection.
The externalist turn in epistemology therefore invites “mixed” theories, but does
not of itself necessitate them. Some externalists opted for just three general con-
ditions, replacing subjective and objective justification with a singular term such as
“warrant” that has a suitably monolithic connotation. These forms of pure as opposed
to mixed externalism are easily associable with the most radical versions of the Conflict
model. Brandom (1998) well explains the abandonment of epistemology’s traditional
normative tasks in favour of ‘eliminative’ or ‘pure’ externalisms:
The primary insights of externalist reliabilism lead to a ‘temptation’ to suppose
that the concept of reliability of belief-forming processes can simply replace the
concept of having good reasons for belief—that all the explanatory work for
which we have been accustomed to call upon the latter can be performed as well
or better by the former. (p. 373)
Those who reject this temptation will likely find themselves critics of the ‘Gettier-
game’, its preoccupation with luck ‘upstream’ of agency, and its incentive towards
infallibilist analyses of knowledge. Although the critics of debates over the Gettier
problem are a diverse lot, they are sometimes painted with the same brush-stroke as
tender-minded reactionaries uncomfortable with a more rigorous and scientific con-
ception of epistemology’s central tasks. What might aid the critics of the Gettier-game
and its allied style of philosophizing is if they were to try just a bit harder to express the
kind of game it has become, and the reasons why training in it as exemplary of good
philosophical analysis has in recent decades done little but to re-enforce sympathies
with radical scepticism.
The tag line for the board-game version of Jumanji, based on an award-winning
children’s book, is ‘A Game for Those Who Seek to Find… a Way to Leave Their
World Behind’. As the book’s author only slightly differently puts it, “a game espe-
cially designed for the bored and restless” (Van Allsburg, 1981). It’s a story that begins
with children who find an old and unusual board game stuffed into a tree-hollow in a
park. They take it home and open the box. Inside they discover the ornate bone dice,
the elaborate jungle-themed board, and of course the beautiful, enticing golden city
of Jumanji that lies at its endpoint. Of course they read on the box its engraved first
rule, that “Once you begin, you must finish the game”. But represented by the author
as moved to alleviate their own ennui more than anything else, they start play without
taking seriously the potential implications for them of the game’s rules.
This proves unwise; things do not go as expected in ‘normal worlds.’ In the film
version, with the first child’s roll of the dice he is immediately pulled into the Jungle,
where his own most worrisome inner imaginations, his own subliminal Children of the
Night as it were, take shape as natural calamities or as fictive but formidable adver-
saries that would hunt him to extinction. A person may be stranded in the Jungle for
376 Synthese (2007) 158:363–383
years, as this boy comes to be, but for however long it takes once the game has been
commenced, it must continue until someone—anyone—makes that special role of the
dice to land them upon the golden city. Then, shouting out its magical name will permit
the children to finally bring to an end the game that has aroused but also tormented
them; only then, as well (or so they believe), will their ‘real’ world, their normal world,
finally be restored. This is another reason why once entered upon, those who started
the game must play on to a finish; it is why in our story the boy who becomes caught
in the jungle for twenty-four years until freed by the roll of other children who have
since stumbled across the box, immediately conscribes this younger generation of its
victims to follow through with the primary directive, “You must finish the game”. And
isn’t it also why we can anticipate that the game-players, when they think the game
over, will attempt, however vainly, to vouchsafe the continuation of their restored
normal world by closing its case tightly, and rushing to find some deep hole or dark
tree hollow into which the game can be stuck way, ‘safe’?
5 Strategizing a “thin concept rescue” of the game’s stranded players
I allege that insofar as they presuppose commitment to PDP, the implied rules of
the Gettier-game are strongly analogous to those of a game of philosophical Jumanji.
But what I now want to ask is what we might offer to counter-act its allure; what
alternative practices and motivations for those neo-Mooreans or others who want to
‘bring their real world back’? For philosophically, it is our own freewheeling accep-
tance of the two platitudes that drags anti-sceptical epistemologists into such a mug’s
game. Having once rolled the dice on giving a “complete” and “completely general”
definition of knowledge, the possibility of re-gettierization (indicated by logical gaps
where epistemic luck can intervene to pull truth and justification apart) is bound to
be perceived as devastating.
But if there are readers sympathetic to the claim that adherence to PDP is a dia-
lectical mis-step for anti-sceptical philosophers, as of yet little impression has been
provided as to how analysis of knowledge might proceed apart from it. So adventure
surely awaits those who find that our little mythos aptly depicts practices and tacit
motivations in the Gettier game. If these sojourners are to charge themselves with
effecting rescue of their philosophical friends stranded in the jungles of philosophical
Jumanji, then they must consign themselves to the fact that there is no alternative but
to roll the dice and enter upon the game.
So pick up the old bones; James is right that there isn’t an experiment or textbook
that may not be a mistake. It’s only by risking our whole persons from one hour to
the next that we live at all! But let’s hold the dice awhile to talk strategy, lest the
game merely make more of us its victims. We need firstly to resist the temptation of
falling back upon the two platitudes, of course, but we need a more positive strategy
as well, and I will now argue that the virtue responsibilists have one. This section
discusses proposals made by Battaly (2001) and Williams (2000), and then in the final
section a bit of flesh is added to the golem of my own fashion, the DATA analysis
(for doxastic, alethic, tucheic and areteic conditions), in order to argue that it is at
least the kind of analysis of knowledge we should hope to establish in post-ELP/PDP
practice.
It will aid us enormously if we can establish a contrast between a “minimal set”
of conditions that one hopes to be nevertheless individually necessary and jointly
Synthese (2007) 158:363–383 377
sufficient for knowledge, and a “thin set” of conditions. These are alternative goals, in
part because in the latter the conditions aren’t intended to guarantee sufficiency for
knowledge as they must when analysis of knowledge is under the sway of the PDP. Its
proponents are those who hold that the notion of ‘epistemizing justification’ doesn’t
exhibit the degree of unitary essence needed for it to be susceptible of Platonic defini-
tion. To elaborate, well-acknowledged family-relations concepts (like “religion”) are
ones that are standardly allowed can be picked out by different combinations of factors
with very little required in the way of common denominators between them. Battaly
(2001) suggests this is true of the concept of justified belief as well: “The meaning
‘justified belief’ does not determine which combination of conditions… is necessary
for its application, or which, short of the whole, is sufficient”.
Battaly’s claim that internalists and externalists are somewhat unconsciously thick-
ening a thin concept of justification, I would maintain, is an entirely complementary
way of putting Pritchard’s own claim that they are somewhat unconsciously directing
themselves to the exclusion of different kinds of epistemic luck. The latter is merely
the manner in which the former becomes manifest in the literature. Her recommen-
dation is to have us analyze justification in terms of a thinly-stated areteic condition,
but one backed up by a broad list of possible meanings relating to subjective and
objective justification—items ranging perhaps from simple ‘aptness’ at the lower end
of the scale, to reflective sensitivity to potential defeaters to our inquisitive efforts
and methodologies at the high end. Each meaning of justification may characterize a
combination of factors sufficient to epistemize an agent’s true belief in some epistemic
context. But to support this possibility, Battaly says we need to rely upon the ability of
“thin concepts” to bring needed flexibility to our analyses: the best and perhaps only
feasible way to approach concepts that have such a plethora of conditions of applica-
tion is to treat their analysis like “a roughly drawn sketch that can be completed in
different ways”.
Perhaps because she follows William Alston in parts of her argument, Battaly
limits her claim to “justification” and “intellectual virtue” being naturally construed
as thin concepts. I doubt there are sound reasons to prevent extending this claim to
“knowledge” as well. To attempt stating general conditions on knowledge guaranteed
to be sufficient for knowledge at all points along any high/low spectrum can now be
seen as paradoxical, and we can give it a name: the paradox of ‘general sufficiency’.
Avoiding this paradox is crucial for anti-sceptical philosophers; but doing so implies
letting go of certain essentialisms and thoroughly rethinking whether we have really
understood what Gettier’s challenge is.
The irony about those reluctant to let go of the search for a fully general account
of sufficient conditions is that they are failing to see the incompatibility between their
essentialism about knowledge and the externalism they purport to espouse: for hasn’t
it has been a very upshot of the externalist turn in epistemology that what lists we
make of forms of epistemizing justification will be quite miscellaneous? Externalists
who want to remain self-consistently such would be better served to stand with Wil-
liams (2000), who holds that a major lesson of the externalist turn is the replacement
of the “Prior Grounding” model (hereafter PGM) with a “Default and Challenge”
model (hereafter DCM) of discursive obligations.18
18 Williams defines the PGM as comprised of four interconnected theses (2000, p. 147):
“(PG1) No Free Lunch Principle. Epistemic entitlement—personal justification—does not just

accrue to us: it must be earned by epistemically responsible behavior. (PG2) Priority Principle.
378 Synthese (2007) 158:363–383
To develop this last claim, Williams’ DCM is one that takes challengers as well
as claimants to be saddled with justificational obligations. In an environment where
the DCM was both genuinely embraced and consistently employed, Battaly’s concern
for the flexibleness of our conditions on knowledge, and subsequent turn to “thin
concept” analysis for key epistemological terms, would be seen as a quite practical
proposal. Moreover, if the DCM is, as Williams argues, what self-consistent external-
ists must endorse, then the virtue epistemologists might be seen as enabling this when
they say that the concept of intellectual virtue is useful because it can be ‘bent’ either
way, towards reliabilist or responsibilist connotations. Thin concept analyses and the
DCM—these two proposals, in short, are made for one another, and even if reliabilist
VE and Zagzebski’s pure virtue theory both sometimes neglect this, our responsibilist
strategy will try to take full advantage of it.
We have now learned that philosophical Jumanji is in large part generated by epi-
stemic externalism insufficiently detached from the internalist underpinnings of the
PGM. By marrying our two proposals we have now put in place an overarching strat-
egy of “thin concept rescue” for those held captive by it, and it is time to finally make
our roll, attempting to put this strategy into effect. In the final section of the essay a
brief sketch of the “DATA analysis” must serve as the attempt.
6 Attempting rescue: A sketch of the “data” analysis
DATA is my proposal for a four-condition analysis where the doxastic and alethic
conditions (i.e., ones to discriminate true belief), of what Williams calls the “standard
analysis” are bolstered by a thin areteic condition (demanding aetiology out of an
intellectual virtue), and a thin tucheic (or anti-luck) condition. Modifying an analysis
that Riggs (1998) stated some years ago to include an independent areteic condition,
the DATA analysis might be put this way:
S knows that p only iff:
Doxastic (1) S believes that p
Alethic (2) p is true
Areteic (3) S’s belief that p is grounded in an intellectual virtue of S
Tucheic (4) the conjunction of (1), (2), and (3) is not a matter attributable to luck
The kind of analysis suitable to us after eschewing ELP and PDP is one that takes
knowledge to be a range-concept. We have supplied the sketch for such an analysis by
‘marrying’ the thin-concept proposal of Battaly to the Default and Challenge model
of our dialogical obligations to the sceptic, and I have elsewhere argued that this mar-
riage substantially magnifies the anti-sceptical force of each of these two proposals,
taken separately (Axtell, 2007b). So in the DATA account above, truth, virtue, and
luck are all represented in a merely formal or ‘deflated’ manner such that while their
individual necessity is asserted, their collective sufficiency for knowledge as repre-
Footnote 18 continued
It is never epistemically responsible to believe a proposition true when one’s grounds for believ-
ing it true are less than adequate. (PG3) Evidentialism. Grounds are evidence: propositions
that count in favour of the truth of the proposition believed. (PG4) Possession Principle For a
person’s belief to be adequately grounded…the believer himself or herself must possess (and
make proper use of) evidence that makes the proposition believed (very) likely to be true.”
Synthese (2007) 158:363–383 379
sented by the “iff” clause has only the status of formal presumption: what I intend
that clause to imply is that fulfilment of the areteic and tucheic conditions is indeed
necessary for knowledge, and but that their joint sufficiency to ‘epistemize’ true belief
is something only to be cashed out in the context of particular, motivated challenges
to the truth of the agent’s belief being of proper epistemic credit to her as an agent.
DATA is thus penultimately fallibilist, allowing us to maintain, as players of philo-
sophical Jumanji apparently cannot, that infallibilism is no part of Gettier’s legacy.
That prima facie sufficiency is all that the “iff” clause should be taken to imply, seems
to me nothing more than to accept the logical implications of Williams’ proposal to
adopt the DCM. When faced with such a challenge to epistemic credit or related
concerns of knowledge attribution, this model imposes an obligation on the part of at-
tributors to thicken their formal conditions in order to state the explanatorily salient
features of the case. But to state general sufficient conditions on knowledge is to
commit to the “universalizability” of those conditions (Vahid, 2001), and universaliz-
ability becomes lethally seductive under the pall of the combined ELP/PDP, leading
us directly into the paradox of ‘general sufficiency.’ That paradox is avoidable on the
present approach.
If we can distinguish what is legitimate in the demand for a “general understand-
ing” of our capacity for knowledge from the paradoxical demands that flow from
robust adherence to the EL and PD platitudes, then we can reasonably decline the
sceptic’s hyper-general demand for a set of thick conditions universalizable across all
points along the knowledge spectrum. We avoid the temptation to continue playing
that game by accepting that, as a range concept, what appears sufficient for knowl-
edge for cases on the low end of the knowledge spectrum can’t be universalizable as
sufficient for knowledge at the high end, and similarly that what appears necessary
for the justification of reflective knowledge can’t be universalized as necessary at the
low end of the spectrum. The ‘thinness’ of our terms allows us to say that the areteic
condition can be met by the instancing of even basic faculty virtues, allowing us to
attribute epistemic credit to the agent in a minimal though sufficient sense, since active
agency itself is present in only a minimal sense. Our Default and Challenge model of
our dialogical obligations to a sceptical interlocutor doesn’t require of us that it be
otherwise.
DATA’s inclusion of both areteic and tucheic conditions is not merely a recommen-
dation for compromise between Pritchard and the virtue epistemologists he criticizes,
though it does call into question in a very direct way why “anti-luck” epistemologists
like Engel, Riggs, and Pritchard, and “virtue” epistemologists like Sosa, Greco, and
Zagzebski, typically take their accounts to be mutually antagonistic. It is an alternative
with distinct advantages, wherein instead of seeing these as alternative “anti-luck” and
“virtue-based” analyses, we come to regard them as agent-based accounts where aret-
eic and tucheic conditions can serve distinct even if overlapping functions of support.
It is natural enough, I suppose, to see the reduction to three conditions as advanta-
geous, but it should also be pointed out that the ‘minimum set’ goal for philosophical
analyses is typically driven by robust adherence to the Platonic Definition platitude,
and is somewhat at odds with the ‘thin set’ goal of an agent-based epistemology set
within the context of the DCM. The latter stresses the importance of the flexibility
of our account in facing motivated challenges that may come at any point along the
knowledge spectrum. Especially given that the anti-luck philosophers think that there
are cases virtue epistemologists don’t handle well, and vice versa, why should we
rush to reduce our conditions to three? Putting them together will seem to players of
380 Synthese (2007) 158:363–383
philosophical Jumanji as making the account necessarily ‘too strong,’ just as anti-luck
and virtue epistemologists today each see each other’s accounts as ‘too weak.’ But
this is to miss the distinctive advantages of the present proposal, which indeed largely
sets those kinds of worries aside. What is directly pertinent instead is the flexibility
of one’s analysis to deal with a range concept, including and especially the manner in
which having multiple conditions allows us to better sort out the normative from the
descriptive aspects of agent evaluation.
Thin and thick descriptions can both be seen as helpful in analysis of knowledge,
though in quite different respects. I would suggest that it is these differences as much as
those between kinds of epistemic luck that need to be understood if the anti-sceptical
force of the neo-Moorean argument is to be enhanced. In respect to radical scepticism,
DATA’s advantage comes in the repudiation of the PGM in favour of the DCM. Thus
to tie DATA further into a neo-Moorean argument, one might argue that the formal
analysis attempts to provide the sceptic the ‘general understanding’ he demands, with-
out accepting his “hyper-general” questions or motivating his ‘disappointed’ versions
of the two epistemological platitudes.
In respect to particular or domain-specific challenges, its advantage comes in how
its ‘twin thin’ areteic and tucheic conditions allow for the effective sorting of the
descriptive and normative aspects of our evaluation of a particular agent in a particu-
lar context.19 DATA, one might say, turns Janus-faced in the face of the unavoidable
problem of relating naturalism and normativity to our philosophic chores. There is no
paradox here; unlike the paradox of ‘general sufficiency’, the having and relating of
our normative and descriptive posits is unavoidable, and native to the philosophical
enterprise itself. To do it well is our only free option, so that it may behoove us to
contrast a positive condition calling for the instancing of intellectual virtue with a
negative one calling for the exclusion of specific kinds or effects of epistemic luck.
The contrast reaffirms the mutual independence of these conditions, and the myriad
of ways in they might be materially satisfied in the states of actual epistemic agents.
An areteic condition when met by the instancing of a faculty virtue provides a nat-
uralistic ground for agent evaluation not directly present in certain other analyses of
knowledge, such as counter-factual and indefeasibility analyses. Requiring it reminds
us that we need to examine agents and not just propositional contents. Some will
argue that including powers and faculties among the virtues stretches the notion of
a virtue, or that the instancing of such faculties, even if reliable, is no basis to attri-
bute “credit” to the agent (Lackey, this volume). But let us stretch the concept just
as thinly as we can, and we can still maintain as Sosa does that reflective reason is
always a “silent partner” in our distinctively human mode of knowledge, so that even
if the vast majority of interaction with our environment reflects what may be deemed
‘low-grade’ knowledge, it isn’t really passive in ways that should have ever tempted
us to assimilate it with knowledge of a merely ‘brutish’ sort.
A tucheic condition also seems like a good candidate for a necessary condition
on knowledge, to the extent that there clearly are types of epistemic luck that are
knowledge-precluding, and others that may not be, yet are nevertheless epistemically
undesirable and thereby still caught up with the sceptical challenge. So I agree with
Riggs, 2007, DOI: 10.1007/s11229-006-9043-y who describes the role of a tucheic con-
dition as providing help in clarifying “the conceptual connections among a family of
concepts that include credit, responsibility, attribution, and luck”; indeed given the
19 For a virtue-responsibilist treatment of religious scepticism, see Axtell (2006a).

Synthese (2007) 158:363–383 381
normative status of this family of concepts, I agree with Riggs that we not only need
such an anti-luck condition, but need it to be a “distinct” or independent one. Riggs
like myself seems content to take it thinly, waiting as it must upon further research
for “a more determinate rendering” of the various kinds of epistemic luck and their
benign or malign effects upon human epistemic agency (1998, p. 282). But that a tuc-
heic condition typically serves a more overtly normative role than that of the areteic
condition indicates to me again that aside merely from the desire for reduction and
simplicity, there is little reason why they need to be seen as competitors. In lieu of
specific arguments to the contrary, the defender of our ‘thin concept’ rescue plan can
continue to maintain that this is a false dichotomy, and that the two conditions instead
function “in consonance.”20
But one key objection to our approach is likely to be that by leaving sufficient con-
ditions on knowledge a merely prima facie affair, we commit ourselves to accepting
certain sharp limitations on the analysis of knowledge. In response let me say that one
reason we fall short of the ideal of Platonic definition is indeed principled: the paradox
of general sufficiency is irresolvable and must therefore be avoided. This we addressed
by criticism of the Platonic Definition Platitude itself, pointing out that analogously
to the Epistemic Luck Platitude, its “robust” version is mirrored by a “disappointed”
version that becomes a key motivation to the neo-Pyhrronian’s “philosophical scep-
ticism.” But another and simpler reason is that we haven’t finished learning about the
aetiology of reliable human belief, nor about the kinds of luck that may impugn the
epistemic credit typically due us for the truth of our beliefs—indeed that we’re just
beginning in the present century to address these questions in a naturalistically sound
way. As Dewey correctly pointed out, “The place for an accurate definition of a sub-
ject is at the end of inquiry rather than at its beginning” (1989, p. 9). That project will
challenge epistemologists to turn away from preoccupation with the Gettier problem
and other “veritic” luck concerns, and to significantly expand epistemology through
attention to the “reflective” intellectual virtues and vices, and the complex ways in
which factors of motivation and the quality of habits or methods of inquiry contribute
to an agent’s success or failure.
Far from being a pessimistic conclusion, then, we have every reason to think that
the important questions about when our twin conditions are materially satisfied, and
about what it is that ties together various instances of knowledge at the high and low
ends of the knowledge spectrum, are ones that epistemologists can make significant
progress with, even if they cease to address them in the essentialistic and hyper-general
way in which the sceptic wants to pose these questions.
It was argued earlier in this paper that for the kind of transformation needed to put
anti-sceptical philosophy on better footing, “It takes two for the show”—a rejection
of what Pritchard calls the Epistemic Luck platitude, but also of a Platonic Definition
platitude with equally deleterious effects upon epistemological practice today. I can
in conclusion only imp myself, harping that line again. For we have taken a stance
unique in the literature, first by supporting the areteic condition with the Default and
Challenge model that Williams argues all self-consistent externalists should adopt,
and second by arguing that when taken ‘thinly,’ we can stop thinking of areteic and
tucheic conditions on knowledge as competitors, and instead avail ourselves of the
20 It is worth pointing out that reflective virtues like “intellectual self-trust” are quite normative and
perhaps ‘folk-psychological.’ Could it be that at the high end of the spectrum, the natural and norma-
tive status of the two conditions is reversed, such that the anti-luck condition becomes the naturalistic
ground of our attribution of particular virtues to the agent?
382 Synthese (2007) 158:363–383
resources that each brings to the table. DATA, and more generally the ‘responsibilist
externalism’ we have argued for is a version of Strong VE, yet its motivations seem
irreducible to those Pritchard refers to in framing the dilemma for VE that has been
a central focus in this paper.21
References
Axtell, G. (2001a). Epistemic luck in light of the virtues. In A. Fairweather, & L. Zagzebski (Eds.),
Virtue epistemology: Essays on epistemic virtue and responsibility (pp. 158–177). Oxford: Oxford
University Press.
Axtell, G. (2001b). Introduction. In Axtell (Ed.), Knowledge, belief and character: Readings in virtue
epistemology (pp. xi–xxix). Totowa, NJ: Rowman and Littlefield.
Axtell, G. (2003). Felix Culpa: Luck in ethics and epistemology. In M. S. Brady, & D. H. Pritchard
(Eds.), Moral and epistemic virtues (pp. 235–257). Oxford: Blackwell.
Axtell, G. (2006). Blind man’s bluff: The basic belief apologetic as anti-sceptical stratagem. Philo-
sophical Studies (forthcoming).
Axtell, G. (2007a). The present dilemma in philosophy. manuscript.
Axtell, G. (2007b). Virtue-theoretic responses to scepticism. In J. Greco (Ed.), Oxford handbook on
scepticism. Oxford: Oxford University Press (forthcoming).
Baehr, J. (2006a). Character in epistemology. Philosophical Studies, 128(3), 479–514.
Baehr, J. (2006b). Character, reliability, and virtue epistemology. Philosophical Quarterly (forthcom-
ing).
Battaly, H. (2001). Thin concepts to the rescue: Thinning the concepts of epistemic justification and
intellectual virtue. In A. Fairweather, & L. Zagzebski (Eds.), Virtue epistemology: Essays on
epistemic virtue and responsibility (pp. 98–116). Oxford: Oxford University Press.
Becker, K. (2006). Reliabilism and safety. Metaphilosophy (forthcoming).
Bernecker, S. (2006). Prospects for epistemic compatibilism. Philosophical Studies, forthcoming.
Brandom, R. (1998). Insights and blind spots of reliabilism. The Monist, 81, 371–392.
Dewey, J. (1989 [1932]). The later works (vol. 7). Carbondale: Southern Illinois University Press.
Dewey, J. (1909 [1970]). The influence of Darwin on philosophy. In P. Appleman (Ed.), Darwin: A
Norton critical edition. New York: W.W. Norton and Co.
Elfin, J. (2003). Epistemic presuppositions and their consequences. In: M. S. Brady & D. H. Pritchard
(Eds.), Moral and epistemic virtues (pp. 47–66). Oxford: Blackwell.
Fogelin, R. (2004). The Skeptics are Coming. The Sceptics are Coming! In: W. Sinnott-Armstrong
(Ed.), Pyrrhonian skepticism (pp. 161–173). New York: Oxford University Press.
Foley, R. (2004). A trial separation between the theory of knowledge and the theory of justified belief.
In J. Greco (Ed.), Ernest Sosa and his critics (pp. 59–72). Oxford: Blackwell.
Gettier, E. (1963). Is justified true belief knowledge? Analysis, 23, 121–123.
Greco, J. (2004). How to preserve your virtue while losing your perspective. In J. Greco (Ed.), Sosa
and his critics (pp. 86–95). Oxford: Blackwell.
Greco, J. (2005). Virtue epistemology. Stanford Encyclopaedia of Philosophy, http://plato.stanford.
edu/entries/epistemology-virtue/
Leite, A. (2004). On justifying and being justified. Philosophical Issues, 14, 219–253.
Pritchard, D. (2002). Recent work on radical skepticism. American Philosophical Quarterly, 39,
215–257.
Pritchard, D. (2003). Virtue epistemology and epistemic luck. Metaphilosophy, 34, 106–130.
Pritchard, D. (2005). Epistemic luck. Oxford: Oxford University Press.
Riggs, W. (1998). What are the ‘chances’ of being justified? The Monist, 81, 452–472.
Riggs, W. (2007). Why epistemologists are so down on their luck. Synthese. DOI: 10.1007/s11229-006-
9043-y (this issue).
Sosa, E. (2004). Two false dichotomies: Foundationalism/Coherentism and internalism/external-
ism. In W. Sinnott-Armstrong (Ed.), Pyrrhonian skepticism (pp. 146–160). New York: Oxford
University Press.
21 I thank Duncan Pritchard enormously for discussion, editing, and suggestions. I thank Jason Baehr,
Juli Elfin and John Greco for additional comments, and Richard Greene and the Society for Skeptical
Studies for organizing the originating session on Pritchard’s book, Epistemic Luck, at the 2005 APA
Pacific.
Synthese (2007) 158:363–383 383
Stroud, B. (2004a). Contemporary pyrrhonism. In W. Sinnott-Armstrong (Ed.), Pyrrhonian skepticism

(pp. 174–187). New York: Oxford University Press.
Stroud, B. (2004b). Perceptual knowledge and epistemological satisfaction. In J. Greco (Ed.), Sosa
and his critics (pp. 165–173). Oxford: Blackwell.
Vahid, H. (2001). Knowledge and varieties of epistemic luck. Dialectica, 55, 351–362.
Van Allsburg, C. (1981). Jumanji. Boston: Houghton Mifflin Co.
Williams, M. (2000). Problems of knowledge. Oxford: Oxford University Press.
Williams, M. (2005). Doing without immediate justification. In M. Steup and E. Sosa (Eds.), Contem-
porary debates is epistemology (pp. 202–216). Oxford: Blackwell.
Williamson, T. (2000). Knowledge and its limits. Oxford: Clarendon Press.
Zagzebski, L. (1996). Virtues of the mind. Cambridge: Cambridge University Press.
RECOVERING RESPONSIBILITY
Guy AXTELL
ABSTRACT: This paper defends the epistemological importance of ‘diachronic’ or cross-

temporal evaluation of epistemic agents against an interesting dilemma posed for this view
in Trent Dougherty’s recent paper “Reducing Responsibility.” This is primarily a debate
between evidentialists and character epistemologists, and key issues of contention that the
paper treats include the divergent functions of synchronic and diachronic (longitudinal)
evaluations of agents and their beliefs, the nature and sources of epistemic normativity,
and the advantages versus the costs of the evidentialists’ reductionism about sources of
epistemic normativity.
KEYWORDS: epistemic normativity, character epistemology, evidentialism,
synchronic and diachronic
1. Epistemic Normativity and the Synchronic/Diachronic Divide

Trent Dougherty’s recent article, “Reducing Responsibility: An Evidentialist
Account of Epistemic Blame,”1 raises quite interesting issues about the epistemic
appraisal of agents and their beliefs. Dougherty, editor of a new collection,
Evidentialism and its Discontents,2 constructs and discusses a case similar to those
that authors he terms the ‘core responsibilists’ have utilized as counter-examples to
evidentialism. But Dougherty’s purpose in constructing his case – the ‘Craig Case’ as
we’ll call it – is to defend the analysis of epistemic justification advocated by
evidentialist thinkers like Earl Conee and Richard Feldman, according to which
‘evidential fit’ (or ‘synchronic epistemic rationality’) is the sole source of properly
epistemic norms. This he terms the evidentialists’ ‘reductionist’ account of
epistemic normativity, in sharp relief from the ‘non-reductionism’ of the core
responsibilists, who allow and indeed insist upon other sources of epistemic
normativity besides evidential fit.
As technical background, Conee and Feldman hold that a proposition being
justified for an agent is just a question of that agent having at any and every given
moment of time that singular doxastic attitude towards that proposition – belief,
1 Trent Dougherty, “Reducing Responsibility: An Evidentialist Account of Epistemic Blame,”

European Journal of Philosophy, 2011, doi: 10.1111/j.1468-0378.2010.00422.x (accessed August 2011).
2 Trent Dougherty, ed., Evidentialism and its Discontents (Oxford: Oxford University Press,
forthcoming 2011).
© LOGOS & EPISTEME II, 3 (2011): 429-454

Guy Axtell
suspension, or disbelief – that uniquely ‘fits’ the total evidence bearing upon it that
she possesses. Propositional justification is thus a matter of the right sort of
cognitive response to evidence (mentally-accessible reasons), and this is wholly a
‘synchronic’ (present time-slice) evaluation. The evidentialists furthermore holds
the propositionalist view that epistemic justification (justification that epistemizes
true belief, i.e., renders it knowledge) is just the having of propositional justification
for some proposition, plus the basing of one’s actual belief upon those reasons.3
Their view of ‘epistemic normativity’ derives directly this propositionalist account,
although it is a contentious thesis not shared by non-internalists, who typically
begin with a conception of the basing relationship and of doxastic justification that
does not entail propositional justification as a necessary condition on knowing.
The case Dougherty has us focus on is that of an imaginary friend and peer,
Craig, a short earth creationist who basically ignores his (Dougherty’s) introduction
of counter-evidence to this belief as well as a well-meaning recommendation that
Craig consider and investigate this counter-evidence by reading a book he offers
him on subjects of geological and evolutionary science. The worry in the Craig case
and in many cases like it is that the agent is and remains personally justified
according to the evidentialist standard of epistemic fit, but ironically only because
that agent’s evidence-base is so extremely narrow; indeed, Craig nurtures this
ignorance of alternatives by simply dismissing or otherwise failing to pursue inquiry
into this potentially undermining counter-evidence to his belief. “[E]videntialism
requires that a change in epistemic status of belief issue from a change in evidential
status.”4 But the Craig Case is one where the personal justification of the agent
appears to drop with or after the encounter with his friend, and for reasons other
than lack of evidential fit.5 A responsibilist at first gloss might suggest that the drop
3 Jonathan Kvanvig, “Propositionalism and the Metaphysics of Experience,” Philosophical Issues

17 (2007): 165-78.
4 Dougherty, “Reducing Responsibility,” 5.
5 I insert ‘drop’ along with simple ‘loss’ of status in recognition of the important point that
externalists and holders of ‘mixed’ accounts beleive that Craig never had the positive status of
being epistemically justified (doxastically justified) in the first place, by virtue of meeting the
evidentialist standards of doxastic justification (having evidence that propositionally justifies,
plus basing one’s belief on that evidence). Where diachronic justification (competent or
responsible evidence-gathering) isn’t a further requirement, agent reliability is by no means
ensured, and doxastic justification or warrant as externalists understand it, won’t be present.
Yet, as Baehr writes, “Perhaps there is some epistemic value simply in having a belief that fits
one’s evidence – regardless of whether this evidence is the result of defective inquiry. Such
beliefs might be said to involve a kind of logical coherence or consistency, which indeed is
often regarded as an epistemic desideratum.” (Jason Baehr, “Evidentialism, Vice, and Virtue,”
Philosophy and Phenomenological Research, 78 (2009): 549.)
430
Recovering Responsability
in personal justification for his belief is due to Craig subsequently manifesting poor
intellectual motivations (strong confirmation bias) and his displaying intellectual
irresponsibility diachronically (cross-temporally) through failing to engage or ‘deal
with’ the counter-evidence to his belief in ways that we would reasonably demand
of any intellectually virtuous agent similarly situated.
Dougherty readily agrees that the evidentialist must take account of such
cases. However, he argues, on closer inspection evidentialism is well-able to handle
them by offering a kind of error theory. “My position is that all instances of epistemic
irresponsibility are in fact either forms of instrumental rationality or moral
irresponsibility in so far as there is anything amiss that goes beyond one's belief
fitting the evidence one has at the time.”6 Despite first appearances, then, there is
nothing distinctly epistemic about what the responsibilists call ‘intellectual
responsibility’ in inquiry: being diachronic or longitudinal evaluations, these are
misconstrued “moral and prudential evaluations of behavior related to the formation
of beliefs.”7 To show this Dougherty offers three different responses which an
evidentialist might make, which he thinks don't require us to recognize anything
distinctively epistemic about Craig’s failures and which therefore don’t require us
to recognize any exception to the evidentialist account of epistemic normativity.
At issue between the evidentialists Dougherty defends and the self-described
responsibilists he targets with his error theory is not only the role of time-slice
(synchronic) and longitudinal (diachronic) evaluations in epistemology, but also,
equally important to him, the right attitude that a philosopher should hold towards
reductionism about the sources of epistemic normativity. The intended upshot of
his treatment of the Craig Case is to re-instate the evidentialists’ reductionist
account of epistemic value by replying directly to the best responsibilist counter-
examples. His reductionist thesis as localized to the responsibilist’s person-level or
diachronic picture of epistemic evaluation is formalized as the Identity Thesis:
IT: Each instance of epistemic irresponsibility is just an instance of purely
nonepistemic irresponsibility /irrationality (either moral or instrumental).8
But this is just one application of a more general reductionist approach to

epistemic normativity, captured in the statements that “any normativity concerning
6 Dougherty, “Reducing Responsibility,” 3. “Like Feldman, my position is that when one’s belief
fits the evidence, all other forms of evaluation concerning the belief are either moral or
instrumental.” (Dougherty, “Reducing Responsibility,” 5.)
7 Dougherty, “Reducing Responsibility,” 3. Compare Richard Feldman, “The Ethics of Belief,” in
Evidentialism: Essays in Epistemology, ed. Earl Conee and Richard Feldman (Oxford: Oxford
University Press, 2004), 190.
431
Guy Axtell
belief that goes beyond fitting the evidence, and in particular epistemic responsibility,
is either moral or instrumental”9, and that “Like Feldman, my position is that when
one’s belief fits the evidence, all other forms of evaluation concerning the belief are
either moral or instrumental.”10
As one of the ‘core responsibilist’ authors targeted by Dougherty in his paper,
I want to make my replies to this reductionist thesis and to his application of it in
treatment of the Craig Case, while recognizing that I cannot speak for any other
author treated in this paper. Other thinkers critical of the Identity Thesis might
take a significantly different tact in their own responses. There are a number of
interesting questions that arise in the above quotations, but I first want to point to
two less obvious ones, questions that Dougherty’s implicit answers to already
inform the manner in which he glosses the concept of epistemic normativity in his
paper. One of these questions is whether epistemic normativity applies only to
evaluation concerning beliefs, or also to evaluations concerning agents and their
habits and dispositions. As Roger Pouivet aptly notes, “Virtue epistemologists generally
agree that, more than anything, good intellectual habits ground our pretensions to
warranted beliefs, and to knowledge. And habits are properties of persons, not of
beliefs.”11 Another is whether epistemic normativity, even granting it the narrower
extension of ‘normativity concerning belief,’ is best construed as restricted to
knowledge-relevant epistemic status, or includes theoretic understanding among
primary positive epistemic goods or standings. Responsibilists have been among the
foremost proponents of conceiving the telos of the life of the mind to include (at
least) understanding, and this question of epistemological axiology is directly
relevant to the question as to whether diachronic assessments of agents can be
properly epistemological. What Raymond Nickerson in Aspects of Rationality calls
‘active fair-mindedness’ – an interpretation that would require one to put
“significant effort into seeking evidence, including evidence that goes against a
favored hypothesis as well as evidence that supports it”12 – may not seem to be
required for more passive sorts of propositional acceptance, yet seems prerequisite for
achieving the epistemic good of understanding. Virtue-relevant epistemic standings
such as theoretical understanding are thus more naturally given to a dynamic and
developmental account than is true belief, often conceived (or misconceived) as a
static cognitive state. We cannot directly pursue these matters here, but we can
9 Dougherty, “Reducing Responsibility,” 1, emphasis added.

10 Dougherty, “Reducing Responsibility,” 5, emphasis added.
11 Roger Pouivet, “Moral and Epistemic Virtues: A Thomistic and Analytical Perspective,” Forum
Philosophicum: International Journal for Philosophy, 15, 1 (2010).

12 Raymond Nickerson, Aspects of Rationality (New York: Taylor and Francis Group, 2008), 140.
432
return to consider their place in this debate after we attend more directly to
explicating Dougherty’s thesis and argument. Put in more positive terms, the
reductionism of the evidentialists that he argues in support of is that evidential fit is
the only proper source of epistemic normativity, and that this is wholly a matter of
synchronic evaluation.
Michael Williams’ term ‘due care and diligence,’ from his paper “Responsibility
and Reliability”13 will be used here to characterize the responsibilist position on the
properly epistemological standing of evaluations of an agent’s motivations and
habits that bear directly upon the conduct of inquiry.14 Dougherty writes, “[T]he
core responsibilists all say that there is more to epistemic responsibility than
evidential fit” and that if evidentialism is false, then the ethics of belief goes well
beyond this consideration of synchronic rationality.15 This is quite correct; there is
considerably more to epistemological evaluation – and still more especially to the
ethics of belief – than is allowed for by a theory that takes believing according to
one’s present total evidence (however gotten or ill-gotten) as the only source and
measure of epistemic value. I think the issues that his paper raises are misdescribed
as issues about the ethics of belief, because that is a question of what one should
believe ‘all things considered,’ which is not Dougherty’s focus. The ethics of belief,
taken this way, explicitly involves ethically-guided considerations of the
consequences our beliefs have on our actions, and so obviously goes well beyond
considerations of synchronic rationality or evidential (propositional) justification.
(10) What Dougherty seems to have in mind is something closer to ‘belief’s own’
ethics – doxastic norms. Having argued elsewhere against both sides of Feldman’s
evidentialist progam – his account of ‘epistemic’ justification since the driving idea
is, and what he describes as ethics of belief16 – I here only want to argue that there
13 Michael Williams, “Responsibility and Reliability,” Philosophical Papers 37, 1 (2008): 1-27.
14 Williams distinguishes two senses of epistemic responsibility, as a matter of accountability
(responsibility for belief – we are responsible for what we believe in the sense of accountable)
and as it contrasts with irresponsibility: failure to exercise due care or diligence in what I term
our inquiry-directed or zetetic habits and strategies (sometimes as attested-to by our overt
activities or behaviors). The antonym of the first sense is not responsible, and of the second
sense is irresponsible. Williams writes that “The two kinds of responsibility are related in an
obvious way: We are accountable for our beliefs precisely because of our obligation to manage
them properly.” (Williams, “Responsibility and Reliability,” 2.) The issue between Dougherty
and the responsibilists is primarily over this second sense. The responsibilists hold that
knowledge involves reliability, and that justified belief is justified (responsible) believing (i.e.,
doxastic justification or warrant).
16 Guy Axtell, “From Internalist Evidentialism to Virtue Responsibilism,” in Evidentialism and its
Discontents, ed. Trent Dougherty (Oxford: Oxford University Press, forthcoming).
433
Guy Axtell
is more to ‘epistemic normativity’ and to a proper understanding of responsibility-

relevant norms than evidential fit and the agent’s having the sort of rationality that
it is characterized by. The agent with unconscientiously acquired or maintained
true belief is not necessarily doing fine qua epistemic agent, even when their ascent
to a proposition is finely in line with their evidence. Unconscientiously held beliefs
will definitely affect one’s understanding, but I think there is more to it even if this
term ‘epistemic normativity’ is restricted just to ‘knowledge-relevant’ normativity.
Evaluation of persons (agents), as contrasted with the evaluation only of beliefs
(cognitive states), is properly part of the theory of knowledge, and evaluation of
persons is never wholly a synchronic affair abstractable from motivational factors
and from habits of inquiry. In short then, my stance is that epistemic fit is not a
sufficient or even a very useful measure of personal justification, epistemic value, or
agent reliability when set apart, as evidentialists intend, from diachronic
considerations of how well or poorly motivated and conducted were the inquiries
that provided the agent with just that set of what the evidentialist calls an agent’s
‘total evidence at time t.’17
Dougherty to his credit provides very clear characterizations of the issues and
of the stances various philosophers have taken, and one thing I agree with fully in
his paper is that for the responsibilists’ explicitly non-reductionist account to be
plausible, its proponents need to show that there are nontrivial cases of what they
call epistemic irresponsibility that cannot be better explained in one of the three
explanations he offers in the Craig Case. That case presents a common (all-too-
common) situation, and (with one notable exception to be discussed below) is as
good as most cases might be in serving as a focal point for this debate. The thrust of
my reply is therefore a straightforward one: None of the alternative explanations
Dougherty offers in the Craig Case is as plausible as he thinks they are; neither
individually nor collectively do they provide evaluations as plausible or informative
as the one according to which the personal justification Craig enjoyed for his
creationist belief does indeed drop, and drops not merely due to the discovered fact of
peer disagreement, but due to Craig’s own subsequent failure to meet an expected
(diachronic) standard of due care and diligence in treating the counter-evidence his
friend and peer presents him with.
17 The virtue responsibilists can certainly agree with their virtue reliabilist counterparts that the
etiology of belief matters in knowledge-yielding doxastic justification, and in the achievement
of other epistemic goods as like theoretical understanding as well. Synchronic rationality qua
‘fit’ of the strength of belief with one’s evidence at any point in time is thus viewed as at best one
among a number of sources of epistemic standards, which include agent reliability as well as
reliability–enhancing intellectual motivations and habits of inquiry. (Guy Axtell, Philip Olson,
“Three Independent Factors in Epistemology,” Contemporary Pragmatism 6, 2 (2009): 89-109.)
434
2. Reductionism, Non-Reductionism, and the Unification Ideal

Before proceeding to the Craig Case, the three different possible evidentialist
explanations of it, and my responses to the adequacy of each, something more
should be said about Dougherty’s contrast of reductionism and non-reductionism.
He acknowledges that there are a range of positions that might support non-
reductionism as he employs that term, some but not all being versions of virtue
epistemology. He cites the more explicitly non-reductionist authors like Lorraine
Code, James Montmarquet and I, among those who develop the analogies between
ethical and epistemic evaluations. For the virtue responsibilists non-reductionism
does not preclude, as one might initially think, but even supports aspirations to
greater theoretical unity between philosophy’s main normative sub-fields.18 It is a
possible path to properly circumspect theoretical unity, and as such not to be
dismissed as an obstruction to it or a merely eclectic view. Dougherty rightly
notices that the non-reductionist approaches of certain self-described responsibilists
are considerably different than the approach taken by Linda Zagzebski, who in
Virtues of the Mind tries to “subsume the intellectual virtues under the general
category for moral virtues.”19 This Dougherty is right to point out is a subsumption
thesis, and a reductionist thesis albeit one running in a direction quite contrary to
the evidentialists’ own intended reduction.
Yet we should draw attention to more positions available to non-reductionists
than Dougherty distinguishes; Susan Haack20 helpfully identifies five distinct views
in the literature about the relationship between ethical an epistemic appraisal:
[1] that ethical appraisal is strictly inapplicable where epistemological appraisal is
relevant (the independence thesis);
[2] that epistemic appraisal is distinct from, but analogous to, ethical appraisal (the
analogy thesis)
[3] that positive/negative epistemic appraisal is distinct from, but invariably
associated with, positive/negative ethical appraisal (correlation thesis);
[4] that there is not invariable correlation, but partial overlap, where
positive/negative epistemic appraisal is associated with positive/negative ethical
appraisal (the overlap thesis)
18 Although similarly confused identifications of unification and reductionism do sometimes occur

even in the sciences, most philosophers of science acknowledged that these are separable goals;
it seems possible then to hold that a virtue theory or other general theory of value provides a
non-trivial degree of unification to epistemology and ethics.
19 Linda Zagzebski, Virtues of the Mind (Cambridge: Cambridge University Press, 1996), 255.
20 Susan Haack, “The Ethics of Belief Reconsidered,” in Knowledge, Truth, and Duty, ed. Matthias
Steup (Oxford: Oxford University Press, 2001), 21-34.
435
Guy Axtell
[5] that epistemic appraisal is a subspecies of ethical appraisal (the special-case

thesis).
On this taxonomy, Dougherty holds Model [1], and Zagzebski’s claim in

Virtues of the Mind “that epistemic evaluation is a form of moral evaluation”21 is an
instance of Model [5]. Non-reductionists might hold any of Models [2]-[4], some of
which it should be noted make considerably stronger claims than others; yet each
of [2]-[4] are able to consistently maintain basic distinctions between theoretical
and practical reason, as well as between epistemic and moral evaluation (as Model
[1] arguably is not). This broadened taxonomy is helpful in correcting one mistake
it appears Dougherty makes in characterizing the non-reductionism of the core
responsibilists and the burden of proof issues that lie between them. For at one
point he strays from his initial, correct characterization and attributes to them a
stronger thesis than I think its defenders need to or in fact do make. This comes
when he claims that they need to demonstrate “a purely epistemic category of
evaluation which does not concern fit with one’s evidence,” or again, “a purely
epistemic evaluation” or kind of irrationality, “over and above synchronic
irrationality.”22 It is quite unclear what ‘purely’ is meant to indicate in these
passages, and credit-worthy intellectual habits and problem-solving strategies are
anyway surely connected not only with concerns of agent reliability in the external
sense, but also with concerns of how evidence, including counter-evidence, is
internally processed by agents engaged in inquiry. But Dougherty’s claim about
what the responsibilists must demonstrate in order to maintain their non-
reductionist stance seems to function as a burden shifting move that they needn't
accept. Even the strongest of the three models ([2]-[4]) available to them asserts
only a strong kind of ‘entanglement’ of epistemic and ethical evaluation (Model [4],
the overlap thesis, which happens to be Haack’s own view).23 Yet this strong kind
of entanglement, and even the weaker kinds of it in Models [2] and [3], the analogy
21 Zagzebski, Virtues of the Mind, 6.

23 This should be familiar from the history of ethics in the 20th century, where for instance the
position of pragmatists from Dewey through Putnam, in opposition to the influential schemes
of reduction offered by C.L. Stevenson, and typically endorsed by the logical positivists, took
the form of entanglement thesis (see Putnam’s The Collapse of the Fact/Value Dichotomy and
Other Essays (Cambridge, MA: Harvard University Press, 2004)). Defenders of the irreducibility
of ‘thick’ affective and characterological concepts to ‘thin’ deontological ones is another pertinent
example; in writers from Bernard Williams to John McDowell this irreducibility is similarly
couched as an entanglement thesis (quite contrary to asserting a ‘pure x’ idea). See also Putnam,
The Collapse of the Fact/Value Dichotomy, for an extended discussion of return of philosophical
interest in thick concepts and its relationship with “the collapse of the fact/value dichotomy.”
436
thesis and the correlation thesis, appear substantial enough to motivate certain
main tenets of responsibilist thought, including especially the inclusion of ‘character
epistemology’ (Baehr’s term) into epistemology’s purview and the properly
epistemic standing of evaluations of intellectual motives and habits. This is why I
thus think it is mistaken to claim that it is incumbent upon critics of the Identity
Thesis to demonstrate “a purely epistemic category of evaluation which does not
concern fit with one's evidence.”24 This is misguided because the responsibilist can
hardly be expected to argue for entanglement (strong or weak) and for ‘purity’ at
the same time. Better to acknowledge, as social and feminist epistemologists like Heidi
Grasswick do, the “impurities of epistemic responsibility,” and sharing their demand
“that a viable concept of epistemic responsibility must be consistent with the
impurities of epistemic agency … as we make decisions regarding how to know.”25
The only direct burden of a non-reductionist as Dougherty uses that term is to
show that proposed reductions don’t work. Yet the same point also underlines that
it is clearly incumbent upon Feldman and other reductionists to demonstrate that
every instance of purported epistemic irresponsibility in the kinds of cases before us
is really “just an instance of purely non-epistemic irresponsibility/irrationality” –
since that is just what Identity claims.
3. Competing Explanations in the Craig Case

With these notes about the Identity Thesis and the nature of the responsibilists’
opposition to it out of the way, we can precede directly to Dougherty’s case of Craig
the short-earth special creationist. The author acknowledges finding Craig’s mind-
set troubling:
The problem, though, didn’t seem to be that his beliefs didn’t fit his evidence—
they did seem to fit his evidence, for he had read very narrowly on the subject and
had been raised and schooled all his life in an apparently reliable community
which sustained this belief in the usual social ways, and which had reasonable-
sounding stories for why people deny their views. Rather, the problem seemed to
be precisely that he only had the very limited evidence he had, since I’d often
recommended books challenging his views. In language that is becoming more
common, his belief seemed to satisfy the standards of synchronic rationality: it
seemed to fit the evidence he had at the time; but it didn’t appear to meet the

25 Heidi E. Grasswick, “The Impurities of Epistemic Responsibility: Developing a Practice-
Oriented Epistemology,” in Recognition, Responsibility and Rights: Feminist Ethics and Social
Theory, eds. Hilde Nelson and Robin Fiore (Lanham: Rowman and Littlefield, 2003), 90.
437
Guy Axtell
standards of diachronic rationality, which is a cross-temporal assessment of

rationality.26
This case does indeed typify ones that responsibilists have raised.27
Dougherty even sharpens this challenge by imagining that he suggests to Craig
“that he read a couple books which I say show that the arguments of young-Earth
creationists are seriously flawed,” and that “he refuses to do so.” 28 If we allow the
dynamic aspect of evaluating Craig over some interval of time, this refusal to
inquire into the matter further by consulting readily available books and arguments
concerning geologic time, etc. seems intellectually irresponsible and something an
intellectually virtuous agent would not do. If we further trust the intuition that
there is a change (a drop) in our evaluation of either or both of Craig himself, qua
epistemic agent, or of the warrant he has for his belief, then the case appears to
undermine the Identity Thesis. For again, “evidentialism requires the change in
epistemic status of belief issue from a change in evidential status,"29 and here it
seems Craig has no substantial new evidence, having ignored rather than
investigated the challenges raised by his friend’s testimony.
But Dougherty, having now nicely identified the challenge that the Craig
Case presents to evidentialism, goes on to give evidentialist explanations of it,
explanations that proceed without a need for the ‘diachronic picture,’ and that
apply a kind of error theory to those attributions of intellectual irresponsibility
which on first glance had looked formidable. Initially we were supposing Craig’s
belief to be justified by the evidentialist standard, even if only because the
homogenous culture he was raised and educated confirmed this belief at every step
and he never sought out any potentially disconfirming evidence. But on the first of
these three alternative interpretations we are mistaken to think that there is no
change to Craig’s evidential status. There is a loss or drop: After Craig’s encounter
his belief no longer fits the evidence as it previously did. Recognition of serious
peer disagreement eventuates in the loss of whatever positive epistemic status of
Craig’s belief that evidential fit is supposed to supply, but explains this by a
corresponding drop in his evidential status or ‘fit.’ “Evidence of evidence is
evidence,” so that “at the time at which Craig became aware that he had testimony
from a known reliable source that some of his beliefs were false, at that moment he
had evidence that his beliefs were false.”30 Thus, this loss of positive epistemic status

27 For others see especially Baehr, “Evidentialism, Vice, and Virtue.”
29 Dougherty, “Reducing Responsibility,” 5, emphasis original.
438
of his belief “is perfectly explicable in terms of synchronic justification”31 because

Craig is interpreted as having gone from synchronically justified (‘epistemically
rational’) to synchronically unjustified (‘not epistemically rational’) in holding his
belief. Hence by this first explanation, “If his beliefs don’t change at all we have
something to explain our inclination to condemn his belief in the lack of
synchronic justification.”32
This first explanation functions so as not to allow responsibilists to run cases
in the way they want in order to directly challenge the Identity Thesis, and
Dougherty believes that in many cases this explanation may offer the most natural
interpretation. He is nevertheless willing to concede that responsibilists would be
able to describe the case so that it is clearly assumable, or is stipulated, that Craig
remains synchronically justified in accepting the creationist proposition after the
encounter with his friend. Thus Dougherty allows his first response to be treated as
something of an aside, acknowledging that dialectically the greater weight of
argument over issues between the evidentialists and the responsibilists falls upon
the success or failure of the latter two of his three evidentialist explanations of the
Craig Case.
These second and third explanations take a markedly different tact than the
first. According to these, Craig maintains the ‘fit’ between his level of belief and his
total evidence even post-encounter, and so long as this is true of him then he is
doing perfectly well qua epistemic agent.33 Craig is indeed culpable in some sense
for failing to pursue inquiry and/or discounting the testimony of the epistemic peer
who disagrees with him, but the second explanation is that what he displays is only
an ‘instrumental’ or ‘prudential’ irrationality. “It is instrumentally irrational for
Craig not to read the books, since if he wants to get the truth, and we supposed he
did, he ought to read the books, since he has evidence that reading the books will
get him to the truth. The cost is low, the potential payoff, we may assume, is
high.”34 The second explanation leans heavily upon the point that there is no
general philosophical principle as to what determines whether acquiring some true
belief is worth our effort attention; expending effort on inquiry and taking an
attitude towards a proposition always involves personal costs and trade-offs, and
since there are many truths not worth knowing, any answer to what determines
whether some truth is worth our attention engages personal/practical interests
and/or moral concerns. Although he doesn’t use these terms, Dougherty might be

33 Earl Conee and Richard Feldman, ed., Evidentialism: Essays in Epistemology.
439
Guy Axtell
thought to be accusing the responsibilists of conflating theoretical and practical

reason, or again of presupposing a strong voluntarism about belief in ascribing
‘intellectual’ responsibility and irresponsibility to agents.
But what if it is responded that this explanation isn’t plausible because truth
is, after all, not some merely pragmatic goal but a prime (perhaps even intrinsic)
epistemic good, and it looks as though Craig’s faulty way of maintaining his
challenged belief shows that he lacks appropriate motivation to believe truly? Then
there is a third explanation which an evidentialist might invoke: “But this also is
either a matter of instrumental irrationality if he has a sufficiently strong desire to
believe truly on this matter or a purely moral one if he does not have such a
desire.”35 There is a kind of irresponsibility operating, but it is purely moral and not
intellectual irresponsibility. This relates back to the earlier point about ‘culpable
ignorance’ when there are costs to information. Dougherty doesn’t take himself to
be arguing that there is no such thing as culpable ignorance in matters of belief, but
rather that “being in a state of ignorance is, when irresponsible, morally irresponsible
or instrumentally irrational.”36
The evidentialist reductionist does not need to argue that these three
explanations are mutually exclusive, but they do need to show that they are all that
are needed in the Craig Case, and that it’s unlikely that we will find other real-
world cases that cannot also be better explained in at least one of the ways
indicated. Dougherty holds that simplicity argues for reducing normative categories
wherever possible, and that if epistemic responsibility can be adequately explained
through any one of the three proposed strategies then we needn’t go looking for a
different kind of explanation.
4. Interpreting the Craig Case

Let's now turn to critically evaluate each of Dougherty’s proposed explanations of
the Craig Case in turn. His first explanatory strategy allows that there is a loss37 in
the epistemic status of Craig's creationist belief after the encounter with his friend
and epistemic peer, but it explains this loss in terms of a corresponding loss of the
evidential fit that his belief enjoyed prior to the encounter. Hence it explains the
drop in status wholly in terms of loss of synchronic rationality or evidential fit.

37 Dougherty appears to differ from Feldman in allowing that there are degrees of belief, but it
seems to me that he applies this inconsistently to his epistemology of disagreement if he still

assumes as Feldman does that a person’s justification for a belief is fully defeated by the
awareness of disagreement, or that no belief is defeated by this awareness.
440
I would concede that there are in principle certain cases for which each of
the three explanatory strategies might work. But in what kinds of cases, specifically,
and if in the Craig Case, then why? Dougherty makes no effort to clearly identify
what the moral fault or the prudential aim would be construed to be in the Craig
Case, for instance. With the first explanatory strategy, it is clearly the evidentialist’s
burden to provide not a mere sketch or suggestion but a full accounting, a theory of
evidence illuminating the actual basis for claiming that Craig’s belief was
synchronically rational up to this one encounter, but synchronically irrational
thereafter. Otherwise invoking this explanation appears ad hoc: What is the
independent motivation for viewing this one particular piece of testimony or this
one encounter with an epistemic peer as being what ‘tips the scale’ or ‘crosses the
threshold’ between synchronic rationality and irrationality?
Dougherty and I would agree, I think, in taking recognition of serious peer
disagreement as having epistemic consequences for agents. Dougherty does not
seem entirely comfortable with those currently-popular epistemologies of
disagreement that allow only the three doxastic attitudes of full belief, disbelief, or
suspension, and that consequently tend to leave only the options of a ‘no defeater’
or a ‘full defeater’ view of the epistemological significance of serious peer
disagreement. I would concur with rejecting these assumptions, if that is indeed
what Dougherty intends. But allowing that epistemic-peer disagreement often
results in a partial defeater,38 while likely making the description of the focus-case
more realistic, would correspondingly also make it more difficult to support the
assertion that Craig passes over this threshold from being synchronically justified to
being synchronically unjustified in holding his creationist belief, at just the point at
which he receives his peer’s contesting testimony.39 As a further aside, epistemic
‘peerhood’ even seems impossible to define apart from admirable shared habits of
inquiry and a normal level of intellectual motivation to achieve a range of socially
shared and acknowledged epistemic goods. Thus I think that attributions of
evidential fit with ‘total evidence’ are implicitly parasitic on attributions of
adequate investigatory habits, suggesting again that bare synchronic rationality
maximization is not a fundamental criterion of epistemic value. Note that
responsibilists also often say much the same thing about bare or ‘thin’ attributions
38 As does Michael Thune, “‘Partial Defeaters’ and the Epistemology of Disagreement,” The
Philosophical Quarterly 60, 239 (2010): 355-372.
39 Under the influence of evidentialist assumptions, “the debate about the epistemological
significance of disagreement is largely focused on whether a belief’s justification is retained or
lost wholesale,” yet a more adequate “account of the epistemological significance of
disagreement should leave plenty of room for cases which result in the partial defeat of beliefs.”
(Thune, “‘Partial Defeaters,’” 356 & 372.)
441
Guy Axtell
of reliable genetically-endowed processes (faculty virtues) made by austere versions

of reliabilist externalism that neglect the importance of personal justification, both
synchronic and diachronic.40
This suggests a further line of argument. The first explanatory strategy
illustrates for us how evidentialists want to treat Craig's belief on the model of
acceptance of an atomistic proposition, and want to treat the total evidence he has
for that proposition in the way we would treat the evidence for any everyday, flatly
empirical assertion. Yet in most matters of any importance in relation to a person’s
moral, political, philosophical, or religious beliefs the evidence is diverse, subtle and
complex; what the agent finds salient in this diverse evidence-base, and therefore
what constitutes the total evidence he ‘has’ that bear upon it, will vary over time.
The reasoning that supports propositional acceptance based upon such evidence is
cumulative in form: It is not just a matter ‘total’ evidence, but of totalizing or on-
balance reasoning. Cumulative case arguments are ones that agents approach with
cognitive strategies that vary greatly in their sophistication and in their
appropriateness to the subject matter before them. For example, a figure like Craig
(who hasn’t examined what we would call ‘hard’ empirical evidence) likely relies
pretty heavily upon Paley-esque arguments by analogy and arguments by appeal to
Biblical authority. It is these I imagine he would cite if pressed to offer defense of
his special creationist belief. But both of these inductive strategies are notoriously
difficult to provide objective/probative measures for. Logic tells us that with
analogical argument and appeal to authority, we need to carefully distinguish our
evaluation of the argument’s inductive strength from our evaluation of the truth of
its premises; yet many or most people are not logically proficient enough to notice
let alone to carefully apply this distinction between inductive strength and cogency
in their reflective weightings of these kinds of arguments.
The upshot here is that an evidentialist attributor of the first explanation
must first be in a position to know or reasonably judge that Craig’s encounter with
the friend and epistemic peer really does tip the scale from synchronically justified
to unjustified – from on-balance fitting to on-balance not fitting his total evidence;
but the cumulative or all-things-considered nature of the agent’s inference in such
cases strongly suggests that the evidentialists are here setting themselves a task that
cannot be met. The problem of operationalizing ‘evidential fit’ to a degree that can
evade the ad hoc objection will be still more difficult whenever broad-scale
hypotheses or theories confront one another; now the choice between them is
generally characterizable as inference to the best explanation. Cases of inference to
40 See George Streeter, “Virtues of Inquiry and the Limits of Reliabilism,” Social Epistemology 20,
1 (2006): 117–128.
442
the best explanation are again common in science, philosophy and religious
metaphysics, but they place a further special burden on the evidentialist who wants
to apply Dougherty’s first explanatory strategy to the Craig Case: produce the
further weightings that allow us to support the claim that an agent like Craig was
really on-balance synchronically justified in his belief at time T1 but not so at T2,
post-encounter with a proponent of a contrary large-scale hypothesis or theory.
Another major worry about the satisfactoriness of the first evidentialist
explanatory strategy also arises from Dougherty’s rather artificial presentation of
Craig as an agent previously situated in near complete cultural isolation from the
scientific tradition of our day. The author describes this as a ‘harmless idealization’41
and says his account could easily be adapted to more realistic ones and still run this
first explanation. I doubt that this is true, however. For Craig’s epistemic situation
to be one that genuinely invites interpretation along the lines of the first
explanation, such an idealization seem needed.42 In a case where the individual has
already been aware of scientific ideas which challenge his special creationism, it
will be all the more difficult to motivate the claim that this one further peer-
encounter or one additional piece of testimonial evidence results in that agent’s
belief sliding from properly to improperly fitting his total body of evidence.43
The second and third explanations that Dougherty offers, by allowing us
interpretations on which Craig continuously fulfills the evidentialist’s condition of
being synchronically rational, are better keyed to the issues at stake between
evidentialist supporters of Identity and their critics. The responsibilists now appear
in a better position to argue that the deserved criticisms of Craig qua epistemic
agent aren’t captured on assumption that synchronic epistemic rationality is the
sole source of legitimately epistemic norms, but rather by something like a failure
in due care and diligence on Craig’s part. But in reply to this charge, Dougherty is
arguing that the evidentialist might succeed in saving the Identity Thesis by
41 Dougherty, “Reducing Responsibility,” footnote 7.

42 The unrealisticness of this idealization is the one small exception I mentioned earlier to
allowing the Craig Case as a shared focus for debate.
43 Note that in the page 5 quote, above, Dougherty writes that “I’d often recommended books
challenging his views” in the past. In this more realistic glossing of the case, I’d hold, the first
explanation is clearly implausible because the newest encounter is unlikely to dramatically
result in moving the agent’s belief from on-balance fitting to on-balance not fitting total
evidence. The encounter with a disagreeing peer is then clearly an important piece of evidence
(and serious peer disagreement, I agree with Dougherty, deserves to be taken seriously). But I
am skeptical of an ideal attributor providing the single objective weighting of these diverse
evidences needed to say that an agent was synchronically rational at time T1, but
synchronically rational at time T2, post-encounter.
443
Guy Axtell
sticking to their guns and saying, first, that in remaining synchronically justified
Craig is ‘doing fine’ as an epistemic agent, and second, that what the responsibilists
describe as Craig’s intellectual irresponsibility will always on closer analysis be seen
to be an extra-epistemological evaluation. More formally, the suggestion that the
responsibilists are confusing intellectual responsibility with moral blameworthiness
is first broached by Dougherty, as we noted earlier, through the question of what
determines whether a particular truth is worth one’s attention. From there, his line
of reasoning leading to the second and third explanations is formalized as a
constructive dilemma:
Either there is some interest at stake in knowing or there is not. If there is not,
then there is no irresponsibility. If there is, it is either the inquirer's interest – or
someone else's interests are at stake – in which case it is a moral shortcoming.44
A constructive dilemma is a valid argument form.45 This dilemma I’ll argue is

unsound, and I’ll approach it by ‘grasping’ the second of its two horns.46 The
dilemma as just quoted requires further clarification, however, since its second horn
is itself intended to be disjunctive, leading directly to Dougherty’s second and third
explanations:
2nd Horn: if it's the inquirer's interests at stake then the irresponsibility or
blameworthiness is really just a matter of instrumental irrationality, and if it's
someone else's interests at stake then it’s really a matter of moral irresponsibility.
First, some general comments regarding assumptions underlying Dougherty’s

constructive dilemma. One mistake that I think is made here is to cast these issues
in such an individualistic philosophical idiom. Knowledge and knowledge

45 A constructive dilemma is a valid argument of the form (if P then Q) and (if R then S); P or R;
therefore, Q or S. There is actually an embedded second dilemma attached to the second horn:
“If it concerns the interests of others, then either I have a duty to promote their interests or I
don't. If I don't, then I'm doing nothing irresponsible in not being scrupulous. If I do, then the
irresponsibility is clearly moral.” So if it concerns the interests of others, then either I'm doing
nothing irresponsible, or my irresponsibility is clearly moral. My response, which consists of
grasping the horns of the constructive dilemma, encompasses a response to this second,
embedded dilemma as well.
46 The first horn – that where there is no interest at stake in a particular proposition being
believed, there can be no question of irresponsibility, intellectual or ethical – I won’t challenge.

But there is a worry about providing a mantle for vices of inattention, which I do not consider
to be all and only moral vices. The issue is properly that one can never be intellectually
blameworthy for not believing or not disbelieving a proposition regarding which they have no
evidence that a morally and intellectually virtuous agent would be presumed to have; it is not
that lack of attention entirely insulates an agent from epistemic appraisal.
444
attributions serve social functions, such that agents are members of communities of
inquirers, and knowledge-attributions ‘genealogically’-considered serve social
functions of marking reliable testifiers. Another problem that needs to be avoided
in the framing of the dilemma is treating assumptions stemming from the fact/value
dichotomy of Dougherty’s Model [1] as if they were analytically true (true by
definition). This cannot be done without begging some of the most interesting
questions at issue in the present debate. I will highlight what I see as mistakes of
this kind as they crop up with the nothing but ‘instrumental irrationality’ and the
nothing but ‘moral irresponsibility’ reductions.
Let’s start with the third, or ‘moral irresponsibility’ strategy. Dougherty's
point that we sometimes confuse moral with epistemic appraisal is again a valid
one. But our burden is only to say why in the Craig Case and similar cases this third
explanation does not look very plausible or philosophically satisfying. Firstly, the
responsibilists think that we should consider Craig intellectually irresponsible for
dismissing his peer’s disagreeing testimony and suggestion to re-open inquiry into
the grounds for his belief, but not necessarily or obviously morally irresponsible for
doing so. The evidentialist who appropriates the third explanation will be saying
asserting exactly the opposite; I doubt that evidentialists have consensus intuitions
in this matter on their side. This is a minor point in itself – what the evidentialist is
offering is, after all, an error theory, and could encompass the error of even what
seem to me strong common-sense intuitions to the contrary. What it does is only to
raise the bar on the need to show the clear superiority of their way of explaining
the case. Secondly, it seems that far too much of the weight of argument with the
third explanation flows merely from identification of the ‘interests at stake’ in a
particular instance of knowing with interests of ‘others’ besides the agent him or
herself. That every self-regarding consideration is a non-moral consideration, and
every other-regarding consideration is a moral one, is not an assumption that I
think many ethicists share, whatever normative ethical account they hold.
Secondly, far too much of the weight of argument with the third explanation flows
merely from association of the dynamic aspects of belief acquisition and
maintenance with overt ‘behavior related to the formation of belief.’47 This term
‘behavior’ threatens to carry practical or moral assessment by definition, potentially
placing it all within the realm of practical reason with one fell swoop. But of course
much of what the defenders of character epistemology and a diachronic picture of
epistemic evaluation are concerned with are intellectual dispositions, habits and
problem-solving strategies, of which overt behaviors are outward manifestations.
Conduct and character are strictly correlative, as Dewey puts it; virtue theorists
47 Dougherty, “Reducing Responsibility,” 3, Feldman, “The Ethics of Belief,” 90.
445
Guy Axtell
would anyway be among the least likely fellows to focus on conduct to the
exclusion of character. So this term ‘behavior,’ while convenient for the
evidentialist to employ, misses its mark by a good distance in describing the
responsibilist approach.
Thirdly, if I claim that an agent lacks the motivational set needed to
distinguish aesthetic objects from mechanical ones, my judgment of the agent is not
itself an aesthetic judgment but a judgment of agential competence with respect to
this target domain. Relatedly, an evidentialist characterization of Craig’s
blameworthiness as relating to his moral rather than intellectual agency seems to
confuse two things. It may well be moral considerations, such as the impact that an
agent's highly ethnocentric belief might have on others when he acts upon it, that
initially draws philosophic or social scientific researchers to an interest in the case,
and to scrutinize habits of inquiry that the agent manifested in acquiring or
maintaining this belief. But once so attending, the focus easily shifts to epistemic
evaluation simpliciter, where the standards utilized are intellectual standards
involving consideration of whether the agent manifested normal intellectual
motivations, utilized effective or ineffective cognitive strategies for problem-
solving, avoided known cognitive biases and fallacious tendencies in reasoning, etc.
This latter evaluation will be clearly epistemic to the extent that what the presence
(or absence) of normal desire for true belief and strategic efforts at inquiry is salient
in explaining isn’t the agent’s blameworthiness in acting upon the belief, but simply
why the agent in this instance was or wasn’t successful in achieving distinctively
epistemic aims such as true belief, etc. This is then a distinct evaluation of the
quality of the agent's inquiry and of the competence and performance of the agent
qua inquirer. If this weren’t the case, then presumably the proponent of the third
explanatory strategy would have to hold that all attributions of ‘epistemic credit’ on
credit accounts of knowing are disguised moral assessments as well.
Enough has been said regarding the claim that if the interests at stake in
knowing in a particular case are not the agent’s own, then evaluation of that agent’s
habits of inquiry is really only disguised moral evaluation. But let's look at the other
conjunct of the second horn of Dougherty’s dilemma, the one which states, “If it
pertains to one's own interests, then the irresponsibility at hand is easily explained
in terms of practical irrationality.”48
In my view the Craig Case and cases like it will not plausibly be adequately
explained on the other two explanatory strategies, and so the ability of this
‘practical’ or ‘instrumental irrationality’ explanation to catch all remaining is crucial
to the support of the Identity Thesis. Explaining the agent’s failure of due care and
446
diligence with counter-evidence will be necessary if we are to be convinced that

the Identity Thesis is true and that there aren’t real-world cases of apparent
intellectual irresponsibility that cannot be explained in one of these three ways.49
Yet Dougherty’s reasons for claiming that diachronic assessments of the intellectual
irresponsibility of the agent amount only to assessments of ‘practical’ or ‘means/end’
or ‘instrumental’ rationality/irrationality, strike me as obscure and dubious. Yes, our
intellectual lives are always caught up in the balancing of the goals of accumulating
interesting true beliefs and avoiding false ones; this involves both valuation and
risk, a point that pragmatists like Haack have often highlighted. There are also, to
be sure, issues concerning tradeoffs between the cost of information in terms of
energy and attention, and the constant need for active fair-mindedness. I agree
with Dougherty that his points about how practical interests and values may inform
issues of attention, etc. serve as a corrective to any veritists who would claim “a
general, impersonal duty to seek truth as such.”50 But these points I see going
nowhere in regards to showing that all choices of attention and all investigative
strategies in pursuit of true belief, knowledge, understanding, etc. are nothing but
moral directives or else matters of prudential/instrumental rationality. If doing fine
as an epistemic agent were really the passive affair the synchronic account seems to
suggest, there would be no response to the ‘updating problem’ with inductive
knowledge, which in turn would be a steep concession to scepticism. There may be
no general answer to the question of how long you remain justified in your belief
that the car you parked on the street yesterday, or the child you parked in the
playroom before the game started, is there yet. But the reasonableness of
occasionally getting off the coach and checking (‘updating one’s evidence’) is not
merely pragmatic/moral – it’s that, to be sure, but one’s level of active searching for
updating information is clearly also a matter of what rational confidence you can
have for those inductive beliefs, and what good reasons you could offer if
challenged in them. Evidential fit is presumably supposed to supply reflectively
‘good reasons’ that one can discursively offer as grounds for the belief. But if the
agent hasn’t been active in updating information when needed in order to maintain
rational confidence, or put any effort into inquiring into counter-evidence to their
belief once presented with it, then that agent’s reasons aren’t going to wash when
someone asks them why they (still) believe it, are they?
49 The language of intellectual vices seems well-adapted to providing detailed characterizations of

the cognitive heuristics as cognitive studies are revealing to us, without treating agents who fall
afoul of the standard of evidential fit as simply ‘dysfunctional’ agents, as Dougherty appears to
want to treat them.
50 Dougherty, “Reducing Responsibility,”10.
447
Guy Axtell
It is noteworthy that Dougherty uses ‘prudential,’ ‘instrumental’ and

‘practical’ synonymously, in order to mark a sharp separation from matters
epistemic. But why can’t it be both instrumentally and epistemologically evaluable,
where true belief is valued intrinsically and true beliefs serve many a practical goal?
There may well always be the possibility of additive personal value from having
true beliefs that are relevant to action, or to satisfaction of personal desires, but this
doesn’t show that their personal value excludes their status as an epistemic good. A
decision to refrain from some activity “does not constitute a rejection or a denial of
the norms that govern it”51; neither does choosing to engage in that activity render
its assessment wholly pragmatic.52 An agent who while engaged in inquiry forms
reflective beliefs by way of methods or strategies that aren’t reliably truth-
conducive seems to be engaging poorly in epistemic activities. Efforts to be actively
fair-minded are obviously intimately involved in the improvement of agent
reliability in areas of contested belief. It seems that if the efforts to be actively fair-
minded towards counter-evidence to one’s belief is ‘easily explained in terms of
practical irrationality’ it is only because instrumental or means-end redescriptions
are always easily available when aim-pursuit is involved. Its availability in and of
itself does little to show that such instrumentalist descriptions are adequate to their
subject-matter, however. Their availability as descriptions doesn’t mean they
capture all of the relevant aspects of rationality involved. Cases where the means
are partly constitutive of the end, as seems to be the case with intellectual as with
moral virtues, are among the many cases in which instrumentalist redescription
provides too narrow a conception of the rationality-connected concerns. Nickerson
identifies at least eight distinct conceptions of rationality, which he sees as ‘aspects’
dependent upon explanatory interests. These are rationality as:
• consistency with self-interest
• pragmatic adaptiveness
• consistency of actions with preferences or goals
• optimal analytic choice behavior
51 Michael-John Turp, “Naturalized Epistemology and the Normative,” Forum Philosophicum 13

(2008): 343-355.
52 “We do, after all, decline to investigate the truth of a great number of propositions without this
implying that we thereby violate in the norms of judgement. We might do so because the
propositions are trivial, or because the investigation would be inappropriately time-consuming,
or for moral, prudential, political or aesthetic reasons …. However, a decision to refrain from
some activity does not constitute a rejection or a denial of the norms that govern it. If one
decided that [some project] were a reasonable project, then quite plausibly one would be
governed by the applicable epistemic norms, as in all one's epistemic endeavors.” (Turp,
“Naturalized Epistemology,” 352.)
448
• satisfycing
• conformity to norms
• reflectiveness
• responsiveness to reasons.
Those most clearly epistemological among the eight aspects of rationality would
be rationality as reflectiveness, and as responsiveness to reasons. The responsibilist’s
interests most clearly fall under rationality as reflectiveness – “a matter of attitude
and intent – love of truth, willingness to examine issues from various points of
view, active fair-mindedness”53; but the responsibilist makes no pretension that this
is the whole of rationality as related to the achievement of epistemic goods. The
evidentialist’s interests most clearly fall under rationality as responsiveness to
reasons if we restrict ‘responsiveness’ to the passive and synchronic, as they would.
They do claim that this is the whole of epistemic rationality, and that rationality as
reflectiveness, because it often involves active and not merely passive fair-
mindedness, reduces to rationality as consistency of actions with preferences or
goals. This seems unmotivated. How much effort rationality requires that an agent
expend is an open question; clearly the rational requirement to seek evidence must
be tempered by the cost of obtaining information and by the recognition of
practical limits on what people can be expected to do. But it is far from clear that
‘epistemic rationality’ consists exclusively in being responsive or fair with whatever
evidence one happens to encounter. “A more active interpretation would have
rationality require that one put significant effort into seeking evidence … active
search for evidence, especially counterindicative evidence, is a key aspect of some
conceptions of what it means to reason well.”54
Recalling a passage noted earlier, “maybe the problem is that we think Craig
doesn’t care enough about the truth. But this also is either a matter of instrumental
irrationality if he has a sufficiently strong desire to believe truly on this matter or a
purely moral one if he does not have such a desire.”55 This strikes me as another
false dilemma. Briefly, if the agent’s performance in gathering and weighing evidence
is thought to be irresponsible for want of skills rather than sound intellectual
motivation, then the sense of ‘instrumental’ irrationality that could be applied
would clearly be consonant with epistemic evaluations rather than representing
merely practical or pragmatic irrationality.56 Yet the case only appears to get worse
53 Nickerson, Aspects of Rationality, 25.

54 Nickerson, Aspects of Rationality, 141.
56 To illustrate further, there are epistemologists who conceive first-personal epistemic rationality
as a kind of instrumental rationality in the service of one’s distinctively cognitive or epistemic
449
Guy Axtell
for this explanatory strategy if we hold true belief to be an aim of intrinsic value.
Nor need we agree, as the second horn of the newest dilemma supposes, that if
Craig’s intellectual motivations are what is in questions – that is to say, if indeed we
find that lack of a normal level of intellectual motivation leads to his unreliability
either generally or in a particular case, that this is necessarily to make a moral
assessment of him. The previous objections to the second explanatory strategy I
think already work to dispel this contention.
In summary, if the ad hocness worry was strong with the employment of
Dougherty’s first explanatory strategy, that worry is greatly magnified with this
third strategy, which purports to show that evaluations of an agent that aren’t
moral evaluations but that do go beyond ‘evidential fit’, can always be treated as
some ‘other sort of normative failing’. ‘Instrumentally’ and ‘prudentially’ irrational
are the catch-all terms, but it appears that pursuit of perfectly respectable epistemic
ends are simply being re-described in this manner in order to save the reductionist
theory. Yet the real point of contention may not be these issues about differences
between epistemic and prudential assessments, but about sharp differences in what
Kvanvig57 calls the ‘idealizations’ embodied in different epistemological theories. In
the way that virtue responsibilists idealize personal justification, it depends not just
on the information the agent in fact recognizes, or has available to her, but also
information that person would have obtained given sound epistemic motivation
and a baseline degree of cognitive competence on that person's part. The standard
of competent due care and diligence that an agent’s performance is judged against is
not restricted to a passive conception of fair-mindedness, but it does not need to
goals; Thomas Kelly dubs this the instrumentalist conception of epistemic rationality, and it is
exemplified in a framework of naturalized epistemology by thinkers like Hilary Kornblith. This
view has some plausibility, but assessing Craig’s shortcomings as a kind of instrumental
irrationality using this model would confirm rather than deny its standing as an epistemological
assessment of the agent. If this isn’t what Dougherty intends, then I suspect he is just smuggling
in at this point the quite radical Conee/Feldman stance “if there is an aim for belief, or a norm
for belief, it is evidence, not truth … if a person has strong evidence for a false proposition F she
should believe that falsehood” (Conee, Feldman, Evidentialism: Essays in Epistemology, 184),
and that “a person who irrationally believes a lot of truths is not doing well epistemically. In
contrast a person who forms a lot of false beliefs rationally is doing well epistemically.” (184)
But, as Pascal Engel has noted, to purport to supplant knowledge, and understanding as the telos
of the life of the mind with constant, perfect synchronic rationality is a radical doctrine that
finds no motivation aside from a host of assumptions not shared by evidentialism's critics
(Pascal Engel, “Review of The Architecture of Reason, by Robert Audi / Evidentialism, by Earl
Conee and Richard Feldman,” Disputatio 20, 1 (2006): 349-358, http://disputatio.com/articles/
020-5.pdf, accessed October 2010).
57 Kvanvig, “Propositionalism and the Metaphysics of Experience.”
450
insist upon unrealistically large efforts at inquiry either. That level is determined by
the context of inquiry itself, including the nature of the claim, and the kinds of
investigative and reasoning strategies that bear upon it.58
5. Closing Remarks
Trent Dougherty’s paper highlights important issues that deserve considerably more
attention. Issues about active fair-mindedness and the cost of information is one
such question responsibilists would like to see discussion expanded upon. This paper
has replied directly to his dilemma for responsibilism, and to his key reductionist
claim that one or another of his three proposed explanatory strategies will always
better-explain cases like the Craig Case than will any diachronic or dynamic
concept of epistemic irresponsibility. I conclude that none of the three explanations
Dougherty offers in the Craig Case is as plausible as he thinks they are, and that
neither individually nor collectively do they provide evaluations as plausible or
informative as one invoking person-level motivation and/or habits of inquiry. It
isn’t an error or illusion: Craig really isn’t doing as well qua epistemic agent as
Dougherty and other evidentialists would have us suppose. Moreover, the merits of
a reductionist urge in epistemology, as Dougherty presents it, is considerably over-
wrought. Reductionist philosophical approaches that claim motivation from a desire
for theoretical unification often not only fail at their task, but actually preclude the
possible realization of greater theoretical unification. Evidentialists like Feldman
claim to want to develop as purely internalist a conception of epistemic justification
as possible. The evidentialist insistence on reducing epistemic normativity to a
purely synchronic standard, and the perpetuation of the longstanding debate over
internalism and externalist about knowledge, I submit, is a case in point.
The objections we have made here to Foley’s and Dougherty’s Identity Thesis
and its confusions of evidential status with epistemic status largely agree with those
that Richard Foley has previously made; this makes relevant the lesson Foley draws
from his own discussions of internalism, the lesson of the need to properly separate
the theory of knowledge from the internalist account of justification. The key
58 Dougherty could argue directly against such agent-focused idealizations, but the plausibility of
the explanation in terms of Craig having exhibited instrumental/practical instead of
intellectual/epistemic blameworthiness cannot lean on the assumption that propositionalism is
true without begging all of the interesting questions at stake here. Propositional and doxastic
justification are indeed difficult to reconcile, whatever epistemological orientation one adopts.
But virtue epistemologists do not think the evidentialists are fair in imposing the self-serving
propositionalist view over the debate. Etiology matters in epistemology, and diachronic factors
are substantially a matter of a belief’s causal etiology.
451
Guy Axtell
lesson for Foley is “the corrupting consequences of the assumption that there is a
conceptual tie between epistemic justification and knowledge.”59 This “Unfortunate
Assumption,” as he calls it – the assumption that the conditions that make a belief
justified are by definition conditions that turn a true belief into a good candidate for
knowledge – is “needlessly limiting”: “It discourages the idea that there are
different, equally legitimate projects for epistemologists to pursue” and it inevitably
distorts both the project of trying to provide and analysis of knowledge, and the
project of understand epistemic responsibility or rationality its importance to us.
Now, what Dougherty calls the Identity Thesis is a product of the Unfortunate
Assumption and appears completely unmotivated apart from it. What commitment
to that assumption produces in the present instance is an artificial separation
between the standard of evidential fit or synchronic rationality and our everyday
assessments of each other's opinions, which as Foley points out “tend to emphasize
whether we have been responsible in forming our beliefs rather than whether we
have satisfied the prerequisites for knowledge."
‘The remedy’ which Foley prescribes, and which I here suggest we prescribe
for defenders of the Identity Thesis, is that they not resist the philosophical reasons
pushing us to a ‘separation’ of the two projects. The prescribed remedy is for them to
jettison the Unfortunate Assumption and all associated attempts to forge a necessary
link between evidential justification, and knowledge possession, and then to develop an
account of synchronic rationality as one distinctive sense of personal justification or
responsibility, and one among several epistemic factors or ‘springs’ of epistemic value.
John Greco’s recent book, Achieving Knowledge: A Virtue-Theoretic
Account of Epistemic Normativity,60 provides quite more direct arguments against
the evidentialist understanding of epistemic justification and epistemic normativity
than I have tried to give here. Knowledge-relevant normative status is not exhausted
by the facts about one’s evidence. The view that it is exhausted by evidential fit, he
argues, suffers from a “psychological plausibility problem” and is undermined by
contemporary cognitive science, and “the prospects for evidentialism about knowledge
look bleak.”61
But in order not to conclude with this criticism of reductionism in epistemology,
let me simply propose a model of diachronic evaluation of the competence and
59 Richard Foley, “Justified Belief as Responsible Belief,” in Contemporary Debates in Epistemology,

eds. Matthias Steup, Ernest Sosa (Malden, Oxford, Carlton: Blackwell, 2008), 313-326. All
quotes from pages 314-315.
60 John Greco, Achieving Knowledge: A Virtue-Theoretic Account of Epistemic Normativity
(New York: Cambridge University Press, 2010).

61 Greco, Achieving Knowledge, 68.
452
performance of inquirers in the pursuit of epistemic aims or goods. The issue

between Dougherty and the core responsibilists is not about what has been called
‘pragmatic encroachment’ upon epistemic norms; but if like Feldman one tries to
construct as an account of epistemic justification as internalist as possible, the issue
at hand might well be seen as a challenge of ‘diachronic encroachment.’ On the
following chart reflects key concerns with diachronic evaluation that I think are
today shared among social, feminist, and character epistemologists, and that
constitute responsibilist research programs in the sense of programs that put
epistemic responsibility into a central place in the theory of knowledge. “Normative
epistemological assessment need not be restricted to judgments of whether
particular claims or beliefs are good or bad, epistemically speaking. Normative
epistemological assessment can also take the form of developing models of how to
practice good inquiry … action and inquiry are central to the concept of an
epistemically responsible agent.” 62 The diachronic are evaluable in at least three
basic ways: morally, pragmatically, and intellectually. We begin with our interests
in diachronic evaluations, and then examine them ‘in light of’ moral, pragmatic, or
intellectual and epistemic concerns. We do not precategorize synchronic and
diachronic norms as wholly one or another, but allow the agent’s context and our
own interests as evaluators to focus our assessment in one or more directions.
Diachronic evaluations are sometimes backwards and other times forward-looking,
and this is true of each of the three ways just mentioned. So this model yields six
basic types of evaluation, to which on the chart I have added basic ‘prompting
questions,’ to give you a feel for how I’m thinking about each. In some cases I have
further separated out prompting questions that motivate social epistemological
evaluations from those that motivate individual-agent epistemological evaluations. I
use my usual terminology in this, referring to these as zetetic evaluations, because
the diachronic as here understood identifies inquiry-directed motives, habits, and
activities, or zetetic considerations.63
62 Grasswick, “The Impurities of Epistemic Responsibility,” 91-92.

63 Thanks especially to Trent Dougherty, Phil Olson, Heidi Grasswick, Nancy Daukas, Christian
Miller and Timothy Chappell for stimulating discussion, as well as to many other discussants at
JanusBlog: The Virtue Theory Discussion Forum, http://janusblog.squarespace.com.
453
Guy Axtell
Responsibility Unpacked: Doxastic Norms & the Logic of Diachronic Evaluations
Epistemological Practical Moral
Individual: The etiology of belief Individual zetetic prompting Individual zetetic prompting
as relevant to doxastic justification question #1: “What did I do to question #1: “What did I do to
of beliefs, and hence to epistemic ensure that the beliefs and ensure that the beliefs and
credit attributions and to final attitudes upon which my actions attitudes upon which my actions
value (the value associated with were based were formed with a were based were formed with a
achievements of all kinds). degree of responsibility fitting the degree of responsibility fitting the
gravity of the decision?” gravity of the decision?”
Individual zetetic prompting
question: “What did I do that my Individual zetetic prompting Individual zetetic prompting
belief was grounded on solid and question #2: “Did I exhibit the question #2: “Were my personal
adequate evidence?” kinds of non-epistemic interests and motivations
deliberative virtues – e.g. consistent with ethical virtue and
Social: Genealogical narratives friendliness or sincerity that with a good life?”
illuminating the axiological conduce to a healthy deliberative
(normative/value or authority- environment and good epistemic Social zetetic prompting question
conferring) and social functions of practice?” #1: ”Was there epistemic justice
epistemic concepts and or injustice displayed in the group
attributions within our overall Social zetetic prompting question institution’s deliberative practice
cognitive ecology. #1: “Was a healthy deliberative in question?”
climate present during the time of
Social zetetic prompting question: my inquiries and deliberations?” Social zetetic prompting question
“What functions, descriptive #2: ”Was the division of epistemic
and/or normative, does this Social zetetic prompting question labor fair, and were varied
concept serve within epistemic #2: – “Were the background perspectives fairly included in this
practices and communities, and conditions during the time of my group or institution’s deliberative
how does its history illuminate inquiry and deliberation such that practice?”
these functions?” good epistemic practices could be
pursued?”
Backward-looking
Forward-looking
v
Epistemological Practical Moral
Improving one’s or one’s group’s Improving the deliberative Increasing my/our ethical
epistemic situation, intellectual climate and nurturing the non- awareness and considering what
motivations, and/or problem- directly epistemic deliberative ethical aims and projects to
solving strategies. virtues in myself and others. pursue. Delineating ideals against
which current practices and
Individual zetetic prompting Individual zetetic prompting institutions may be criticized, and
question: ”Is my epistemic question: ”How can I acquire or future practices designed.
situation good enough for me to apply deliberative virtues that
understand the matter, or to make will improve my own Individual zetetic prompting
a sound judgment in accepting or deliberations (and those of others question: ”What special
rejection a proposition; or should I in my community of inquiry)?” responsibility for pursuing further
first pursue further inquiries or inquiry, given the gravity of the
adopt new strategies in order to Social zetetic prompting question: situation on which my actions
improve my epistemic situation?” ”How should cognitive labor be will bear, and the social roles I
divided in order to promote have (qua professional, parent,
Social zetetic prompting question: shared cognitive goals? Also, etc.)?”
”Are the institutional or social what part of that labor is it my
group practices conducive to truth responsibility to take on, and how Social zetetic prompting question:
and understanding, or should they do I responsibly assess the ”What socially ‘transformative’
be altered in light of the present credibility of inquiry performed aims should I endorse, and what
problem-situation?” by others?” individual, social and collective
‘new’ virtues should I posit as
reflective of those aims?”
454
THINKING TWICE ABOUT
VIRTUE AND VICE:
PHILOSOPHICAL SITUATIONISM AND
THE VICIOUS MINDS HYPOTHESIS
Guy AXTELL
ABSTRACT: This paper provides an empirical defense of credit theories of knowing

against Mark Alfano’s challenges to them based on his theses of inferential cognitive
situationism and of epistemic situationism. In order to support the claim that credit
theories can treat many cases of cognitive success through heuristic cognitive strategies
as credit-conferring, the paper develops the compatibility between virtue epistemologies
qua credit theories, and dual-process theories in cognitive psychology. It also provides a
response to Lauren Olin and John Doris’ ‘vicious minds’ thesis, and their ‘tradeoff
problem’ for virtue theories. A genuine convergence between virtue epistemology and
dual-process theory is called for, while acknowledging that this effort may demand new
and more empirically well-informed projects on both sides of the division between
Conservative virtue epistemology (including the credit theory of knowing) and
Autonomous virtue epistemology (including projects for providing guidance to
epistemic agents).
KEYWORDS: bounded rationality, dual-process theory, ecological rationality,
heuristic reasoning, situationism, virtue epistemology
1. The Great Rationality Debate

One goal of this paper is to defend virtue epistemology (hereafter VE) against a
number of charges that Mark Alfano brings against it based upon the
incompatibility of its claims with epistemic situationism, and that Lauren Olin and
John Doris bring against it based upon an objection we can call the ‘trade-off’
problem.1 The latter problem alleges a dilemma in the form of a necessary tradeoff
with ill-consequences for VE, a trade-off between the normative appeal of a
virtue-theoretic (ability) condition on knowing, and the empirical adequacy of
such a condition. Doris presented a short version of the trade-off problem in his
1 Mark Alfano, Character as Moral Fiction (Cambridge: Cambridge University Press, 2013);
Lauren Olin and John M. Doris, “Vicious Minds,” Philosophical Studies 168, 3 (2014): 665-692;
John Doris, Lack of Character (Cambridge: Cambridge University Press, 2002).
© LOGOS & EPISTEME, VIII, 1 (2017): 7-39

Guy Axtell
critique of virtue ethics in Lack of Character; Olin and Doris develop the problem,
extending it into a challenge for virtue epistemologies as well.
At the same time, however, the paper will have more constructive goals
than that of defending VE against the challenge of philosophical situationists and
their close cousins. Epistemologists have not paid enough attention to what
psychologists call the normative-descriptive gap, or to the ‘bounded’ or ‘ecological’
nature of human cognition. Our human susceptibilities to motivational and
cognitive biases may be well-recognized by today’s more naturalistically-inclined
philosophers. But there remains a worry about normative theories in ethics and
epistemology, shared by many in the social scientific community, which must be a
concern for virtue theories as well: philosophers must be warned of repeating the
errors of the past by permitting discussions of ethical and epistemic normativity to
continue as a ‘separate culture’ from the social and cognitive sciences. As Appiah
puts the point, “The questions we put to the social scientists and physiologists are
not normative questions. But their answers are not therefore irrelevant to
normative questions.”2
Naturalistic virtue theory agrees: It is a dubious non-naturalism in
philosophy, and in particular an intellectualist tradition associated with the
autonomy of philosophy from the human and natural sciences, that motivates the
separate cultures notion. Virtue theories have arguably been at the forefront of the
movement to integrate the scientific image of humans into normative ethics and
epistemology. But my approach suggests that much more attention needs to be
paid by ethicists and epistemologists both to the normative-descriptive gap and
more particularly to how to conceive the relationships between ‘reasoning and
thinking,’ ‘competence and performance,’ and ‘assessment and guidance.’ In terms
of positions in the great debate over human rationality which has raged over the
past half-century, I will try to show virtue theorists as Meliorists standing against
the situationist’s Skepticism. But avoiding the traps of intellectualism and the
philosophy-as-autonomy view also means distancing one’s stance from those
proponents of the Apologist and Panglossian positions who Daniel Kahneman
chastised decades ago as acknowledging only two categories of errors, “pardonable
errors by subjects and unpardonable ones by psychologists.”3
2 Kwame Anthony Appiah, Experiments in Ethics (Cambridge: Harvard University Press, 2008),
62.
3 Daniel Kahneman quoted from Keith Stanovich and Richard West, “Individual Differences in
Reasoning: Implications for the Rationality Debate?” in Heuristics and Biases: The Psychology of
Intuitive Judgment, eds. Thomas Gilovich, Dale W. Griffin, and Daniel Kahneman (Cambridge:
Cambridge University Press, 2002), 421.
8
Thinking Twice about Virtue and Vice
Let us be somewhat clearer about what will be the critical and constructive
goals of the paper. On the critical side I take the approach of empirical rebuttal
and argue that Alfano’s two key theses, which he terms inferential cognitive
situationist (ICS) and epistemic situationist (ES) (sections 2 and 3, below) are
generalizations that are not strongly supported by the selection of studies in
cognitive and social psychology (respectively) he bases them on. Generalization as
a form of inductive arguments is assessed as either strong or weak, and weak
inductive arguments are akin to invalid deductive arguments in falling short of
logical compulsion. If such generalizations as philosophical situationism depends
upon are not as strong or cogent as they claim, then of course they do not provide
a benchmark (as their challenges assume) against which the empirical adequacy of
an epistemology should be measured. Sections 2 and 3 respond to challenges
directed against the more reliabilist and the more responsibilist versions of VE,
respectively. But my weak generalization claim about the status of Alfano’s two
key theses regarding human cognitive agency also extends to Olin and Doris’
‘vicious mind’ thesis. These authors claim to infer from their discussion of select
studies to the ‘enormous’ variability in human cognitive functioning due to our
sensitivity to even slight situational variables. Section 4 offers a fuller response to
their ‘trade-off’ problem for virtue theories than I have previously given, allowing
me a chance to more fully develop the Narrow-Broad Spectrum of Agency-
Ascriptions I alluded to in an earlier exchange.4
On the constructive side, I concede that few extant versions of VE have
tried explicitly to square themselves with bounded rationality, or with dual-
process theory (hereafter DPT), as I will argue that they should. Indeed, there has
been more work on accommodating DPT among ethicists than among
epistemologists.5 Bounded rationality, extending from seminal work by Herbert
Simon, asks and studies how real people make decisions with limited time,
information, and computation. Gerd Gigerenzer writes that the science of
4Guy Axtell, “Agency Ascriptions in Ethics and Epistemology,” in Virtue and Vice, Moral and
Epistemic, ed. Heather Battaly (Oxford: Wiley/Broadview Press), 73-94. See also Christopher
Lepock, “Unifying the Intellectual Virtues,” Philosophy and Phenomenological Inquiry 83, 1
(2011): 106-128.
5 Daniel Lapsley and Patrick Hill, “On Dual Processing and Heuristic Approaches to Moral
Cognition,” Journal of Moral Education 37, 3 (2008): 313-332; Holly M. Smith, “Dual-Process
Theory and Moral Responsibility,” in The Nature of Moral Responsibility: New Essays, eds.
Randolph Clarke, Michael McKenna, and Angela M. Smith (Oxford: Oxford University Press,
2015), 176-208; Nancy Snow, Virtue as Social Intelligence: An Empirically Grounded Theory
(London: Routledge, 2009); Snow, “Habitual Virtuous Actions and Automaticity,” Ethical
Theory and Moral Practice 9, 5 (2006): 545-561.
9
Guy Axtell
heuristics “Asserts that ecologically valid decisions often do not require exhaustive
analysis of all causal variables or an analysis of all possible actions – and –
consequences. The best decisions do not always result from such effortful,
reflective, calculations, but instead rely on ‘frugal,’ incomplete and truncated
inquiry.”6 Ecological rationality challenges expectations that human reasoners are
rational or justified only when they meet normative standards derived
independently of empirical and social psychology. It suggests that demands upon
rationality be perfectly feasible for agents, computationally speaking, and that
norms of epistemic assessment, while still truth connected, not be ‘free-floating’
impositions.7 That cognition is so heavily ecological means that norms of epistemic
rationality and responsibility bump up against pragmatic constraints and inborn
limitations in ways that challenge ideal observer and maximizing conceptions of
reasoning.8
Gigerenzer’s approach has substantial differences from those versions of
ecological rationality that I want to highlight in this paper, versions that fall under
the umbrella term of dual process theories. No less a pioneer of the biases and
heuristics studies than Daniel Kahneman notes that “Tversky and I always thought
of the heuristics and biases approach as a two-process theory.”9 What he describes
in his recent book Thinking, Fast and Slow as ‘Type 1’ and ‘Type 2’ thinking
(hereafter T1/T2) merely modifies terms he acknowledges were introduced first by
Keith Stanovich, Richard West and Jonathan St. B.T. Evans, whose work we will
focus on in the next section.
6 Gerd Gigerenzer, Ecological Rationality: Intelligence in the World (Oxford: Oxford University
Press, 2012), 20.
7 See Rysiew’s helpful conditions on psychology-sensitive philosophical norms of rationality, in
Patrick Rysiew, “Rationality Disputes – Psychology and Epistemology,” Philosophy Compass 3,

6 (2008): 1153-1176.
8 Gigerenzer sees the facts of ecological rationality challenging what he terms the ‘classical
conception of rationality,’ a conception with ‘appealing but often unrealistic goals’ that he
thinks is anti-naturalistic in its tenor yet remains still deeply-rooted in philosophy, economics,
and decision theory. The standard view in the cognitive sciences associated with unbounded
rationality Gigerenzer blames for the institutionalized division of labor between principles
based upon the ‘is’ and ‘ought' division. “Until recently, the study of cognitive heuristics has
been seen as a solely descriptive enterprise, explaining how people actually make decisions. The
study of logic and probability, by contrast, has been seen as answering the normative question of
how one should make decisions.” (Gigerenzer, Ecological Rationality, 496). This split or schism
Gigerenzer thinks has served to wrongly elevate logic and probability above heuristics; the
result is “contrasting the pure and rational way people should reason with the dirty and
irrational way people in fact do reason.”
9 Daniel Kahneman, Thinking, Fast and Slow (New York: Farrar, Straus & Giroux, 2012).
10
Since few attempts have already been made to square virtue theory with
ecological rationality or with dual-process theory, this paper is one with
prescriptive implications for the direction of epistemological research. My title
should be taken to reflect the idea that virtue theorists both in ethics and
epistemology do indeed need to think twice if their appeal to person-level abilities
and to characterological concepts more generally is to be empirically-informed
and naturalistically sound. Adapting Evans’ book title Thinking Twice: Two Minds
in One Brain (2010),10 the real and direct challenge to virtue epistemologists as I
conceive it, is to rethink the epistemic credit-related family of concepts in light of
the distinction between (T1) fast, automatic, holistic, and intuitive ways of
thinking that require relatively little cognitive effort, and (T2) slow, deliberative,
and serial or analytic ones that require substantially more sustained cognitive
effort [see Table 1 reprinted from Evans and Stanovich].11 Thinking fast/frugally
and slow/effortfully are both adaptive; both are routinely successful, although
both can also fail. But far too often an effortful, maximizing conception of
rationality has been taken as paradigmatic, with heuristic reasoning viewed only
as a source of error and dysrationalia. So my thesis of possible convergences
between normative epistemology and DPT prescribes substantial rethinking on
the part of epistemologists, including proponents of VE.
10 Jonathan St. B. T. Evans, Thinking Twice: Two Minds in One Brain (Oxford: Oxford
University Press, 2010).
11 Reprinted from Jonathan St. B.T. Evans and Keith Stanovich, “Dual Process Theories of
Higher Cognition: Advancing the Debate,” Perspectives on Psychological Science 8, 3 (2013):

223-242, 224.
11
Guy Axtell
2. Abilities, Heuristic Reasoning, and Epistemic Credit

In his book Character as Moral Fiction and papers leading up to it, Mark Alfano
poses the following dilemma for reliabilist and mixed virtue epistemologies:
First Horn: If they say that heuristics are not intellectual virtues, skepticism
looms: If most people use non-virtuous heuristics, then most people have a large
number of unjustified beliefs, which do not count as knowledge.
Second Horn: If, however, they say that heuristics are intellectual virtues, then
they need to explain how these dispositions are to be construed as reliable.
The dilemma tries to force a choice between holding absolutely either that
heuristic reasoning is or isn’t virtuous. Virtue epistemologists do not want to set
conditions on knowledge so high that knowledge becomes scarce. But embracing
the second horn Alfano thinks is barred by the thesis of inferential cognitive
situationism (ICS):
(ICS) People acquire and retain most of their inferential beliefs through
heuristics rather than intellectual virtues.12
Arguably, there is a problem with Alfano’s statement of the problem since

(ICS), by presupposing that heuristics are not virtues (the either/or language of
‘rather than’) begs the question against someone who wants to ‘grab the second
horn’ of the dilemma. But more charitably, since all of Alfano’s and Olin and
Doris’ challenges focus specifically around the aretaic or ability condition that
credit (or achievement) theories13 place upon knowledge possession, I will
construe the demand for clarity about the status of heuristics to be about whether
virtue epistemologists can accommodate them in a credit theory of knowing. If
knowledge is an achievement creditable in significant measure to an agent’s
manifestation of ability/virtue, how can inferences that apply heuristic reasoning
instead of patterns that we recognize as sound and reliable inferential practice,
count as virtuous?
The issue Alfano raises is indeed important, but the demand he places upon
reliabilist and mixed VE seems to me a false dichotomy, so I will go ‘Between the
Horns’ of his dilemma in reply. For Alfano’s dilemma clearly takes ‘heuristics’ and
‘cognitive virtues related to deductive and inductive reasoning’ out of context
from bounded rationality theories. Although bringing concern with heuristic
reasoning to the forefront of epistemology can be applauded as likely to spur
12Alfano, Character as Moral Fiction, 201 and 191.

13I here take ‘credit theory’ as a general type of analysis broad enough to include both Robust
and Anti-luck VE.
12
progress in the field, virtue epistemologists can surely avail themselves of leading
theories in cognitive psychology, theories on which it would be absurd a demand
a choice between the reliability or unreliability of heuristic strategies. Cognitive
and social psychology indeed militates against pitting facts about thinking against
norms of reasoning in the way that Alfano’s dilemma pits them. To briefly digress,
DPT and Gigerenzer’s ecological rationality theory have some sharp differences,14
but it is fundamental to the broader ‘new paradigm’ in cognitive psychology
which they share that intuitive and reflective thinking can each be highly reliable
when well-matched to the agent’s problem situation. In Evans’ terms, “both Type
1 and Type 2 processing can lead to ‘good’ or normative answers.”15
Although logocentrism, and a deductivist bias, still casts its shadow on
philosophy’s ways of approaching norm governance, most epistemologists, and I
think all of those associated with VE, are today concerned with human thinking,
not just ‘reasoning,’ and with inference, not just ‘argument.’ Concerns with success
on cognitive tasks through heuristic strategies and T1 processing should indeed
prompt a more minimal account of epistemic credit than an epistemology could
offer if it remained locked into understanding reasoning and inference only on the
model of argument. As Paul Thagard argues, a naturalistic epistemology consistent
with the successes of ecologically rational agents needs to maintain that
“rationality should be understood as a matter of making effective inferences, not
just good arguments.”16 But assumptions that might restrict epistemic credit to the
14 Gigerenzer’s ecological rationality theory finds unmotivated the division into two types,
systems, or ‘minds,’ arguing instead that intuitive and deliberate judgments are based on
common principles such that a unitary rather than a dual-process account can be given of them.
But the overlap of shared lessons from cognitive psychology is nevertheless broad. See Arie W.
Kruglanski and Gerd Gigerenzer, “Intuitive and Deliberate Judgments are Based on Common
Principles,” Psychological Review 18, 1 (2011): 97-109.
15 Jonathan St. B.T. Evans, “Dual-Process Theories of Reasoning: Facts and Fallacies,” in The
Oxford Handbook of Thinking and Reasoning, eds. Keith J. Holyoak and Robert G. Morrison
(Oxford: Oxford University Press, 2012), 23. See also Evans, “Questions and Challenges for the
New Psychology of Reasoning,” Thinking & Reasoning 18, 1 (2012): 5-31; David Over, “New
Paradigm Psychology of Reasoning,” Thinking & Reasoning 15 (2009): 431-438.
16 Paul Thagard, “Critical Thinking and Informal Logic: Neuropsychological Perspectives,”
Informal Logic 31, 3 (2011), 152. Manktelow (Thinking and Reasoning) and Over (“The New
Paradigm”) both discuss how seminal to psychology the gap problem and the distinction
between reasoning and thinking have been. Compare Thagard, writing as a naturalistic
epistemologist critiquing the costs of confusing inference and argument: “If inference were the
same as argument, it would have the same serial, linguistic structure. However, there’s ample
evidence from cognitive psychology and neuroscience the human inference is actually parallel
13
Guy Axtell
latter are hardly to be associated with reliabilist VE, where naturalistic orientation
runs highest. Epistemic credit associated with knowledge-attributions generally
demands only weak cognitive achievements unless we are talking about
specifically reflective knowledge. If this is correct then it can credit successes that
might be easily had in untaxing ways in epistemically friendly environments.17
One would think that the turn in epistemology from states and standings to
epistemic agency and inquiry would underline these points. Evans writes,
From a pragmatic viewpoint, even if people fall prey to certain biases, it does not
mean they are irrational [or generally unreliable, or ‘vicious’]. Making mistakes
can still be part of a rational, or a reliable, or an intellectually virtuous agent’s
repertoire. As some leading dual process psychologists have argued more
explicitly, errors of thinking occur because of, rather than in spite of, the nature
of our intelligence. In other words, they are an inevitable consequence of the
way in which we think and a price to be paid for the extraordinary effectiveness
with which we routinely deal with the massive information processing
requirements of everyday life.18
Another way of putting my point is that having an account of inferential

knowledge and recognizing virtues connected with sound deduction and
abduction does not commit epistemologists to what Adam Morton, in his recent
book Bounded Thinking: Intellectual Virtues for Limited Agents calls an N-theory
of rationality. N-theories derive their norms independently of ecological
considerations of time, information, and energy, and so often maintain some
computationally demanding conception of rationality. Another aspect of the
appeal of the naturalistic turn in epistemology, which virtue theory provides an
interpretation of, is that we should not dichotomize (as non-naturalistic theories
sometimes have) between norms of epistemic assessment and the aim of providing
agents with guidance. While naturalistic approaches in epistemology will still
recognize that the normativity of assessment and guidance should be
distinguished, questions of psychology cannot be treated today as they sometimes
rather than serial, multimodal rather than just language-based, and as much emotional as
cognitive” (Thagard, “Critical Thinking,” 152).
17 Compare Pritchard, who identifies strong cognitive achievements with overcoming a
significant obstacle to cognitive success, or with the manifestation of high levels of cognitive
skill. Weak cognitive successes by contrast are those where it is very easy to attain the relevant
cognitive success. Here “one will meet the rubric for cognitive achievements pretty easily.”
(Duncan Pritchard, “Virtue Epistemology and the Epistemology of Education,” Journal of
Philosophy of Education 47, 2 (2013): 236-247, 240.)
18 Evans quoted in Robert Sternberg and Talia Ben-Zeev, Complex Cognition: The Psychology of
Human Thought, (Oxford: Oxford University Press, 2001), 194.
14
have in the past as merely pragmatic or as only guidance but not assessment-
related.19 By focusing on how our intellectual habits, whether genetically-based or
acquired, ground our pretentions to knowledge, understanding, or other epistemic
goods, virtue epistemologies it seems to me bypass the worries about N-theories
and armchair epistemologies that always demand of agents lots of self-reflection
and high effort or ‘load-heavy’ cognitive processing.
It is only when an agent applies a cognitive strategy (such as a heuristic
pattern of reasoning) outside its known range of reliability, or perhaps where
System 2 or metacognitive ‘over-ride’ skills and sensitivities were expected of
virtuous agents but were not manifested, that epistemologists are likely to balk at
credit for success. In these instances error possibilities are high, and the agent’s
employed strategy is said to have a low ecological validity. Judgments of these
sorts may rightly be used to deny epistemic credit (and hence knowledge or
understanding). Of course, the instances where T2 over-ride failures occur in
humans are many and not few. Philosophers have not come to grips with what
Kahneman simply describes as ‘the quirkiness of System 1 and the laziness of
System 2.’ There are numerous causes of dysrationalia in heuristics and biases task
experiments that DPT recognizes, including failure to detect the need to override
a heuristic response, lack of acquired ‘mindware’ available to carry out override,
and inability for ‘sustained decoupling’ that allows hypothetical reasoning and
other powerful metacognitive aptitudes to engage.
To summarize thus far, I think there is ample opportunity to go ‘Between
the Horns’ of Alfano’s dilemma for reliabilist and mixed virtue epistemologies. No
one need deny the valuable functions that N-theories serve;20 but neither should a
credit theory of knowing place the agent under any assumed burden of following
strategies that maximize cognitive load. VE can and should agree with DPT that,
Since the fast, automatic, and evolutionarily older system requires little cognitive
capacity, everyone has the capacity to deal rationally with many reasoning and
decision making problems that were important in the environment in which we
evolved. Moreover, since the new, slow, rule-based system can be significantly
affected by education, there is reason to hope that better educational strategies
19 Michael Bishop, who clearly holds that bounded rationality should deeply impact our
approach to epistemic normativity, finds that the received internalist notion of responsibility is
what heuristic reasoning especially challenges. (Michael Bishop, “In Praise of Epistemic
Irresponsibility: How Lazy and Ignorant Can You Be?” Synthese 122, 1 (2008): 179-208, 179. See
also Michael Bishop and J.D. Trout, Epistemology and the Psychology of Human Judgment
(Oxford: Oxford University Press, 2005.)
20 See Adam Morton, Bounded Thinking: Intellectual Virtues for Limited Agents (Oxford:
Oxford University Press, 2013).
15
Guy Axtell
will improve people’s performance on those problems that the old system was
not designed to deal with.21
Situationists are skeptics about character-traits, and so also of the claims

about individual differences among agents that character-traits help determine.
But in order to go beyond arguing for a basic consistency of VE with cognitive
psychology’s recognized continuum between fast and slow, or intuitive and
reflective ways of processing, I offer reflections on three specific guiding themes
in the work of Evans and Stanovich that I think point out potential convergences
between them:
Individual differences. According to proponents of DPT there are few continuous

individual differences among people with respect to autonomous mind, or Type 1
processing. We are all energy economizers and want to fit strategies to problems
ecologically when we can, rather than doing the ‘expensive’ reasoning of ideal
inquirers qua unbounded reasoners. But they also insist that “the intelligence of
the new mind is quite variable across individuals.”22 DPT both predicts and finds
confirmed substantial individual differences in what Stanovich terms ‘rational
thinking dispositions’ and therefore recognizes “individual differences as essential
components of heuristics and biases research.”23 The cultivation of general
intellectual virtues that help attenuate cognitive biases and inappropriate heuristic
responses strongly overlaps with what Stanovich and his colleagues term the
cultivation of our fluid rationality. Fluid rationality describes a range of available
critical reasoning dispositions. Research shows that those who do poorly on
cognitive task tend to be ‘cognitive misers’ in the way that they process, while
those who do better than average exhibit a more desirable collaboration between
their T1 and T2 thinking, and as Stanovich has it [Table 2], between our
crystalized and fluid rationality. Some of the imperfect but powerful aptitudes
fluid rationality describes are ‘resistance to miserly information processing,’ and
‘absence of irrelevant context effects in decision making.’
21Richard Samuels and Stephen Stich, “Rationality and Psychology,” in The Oxford Handbook
of Rationality, eds. Alfred R. Mele and Piers Rawling (Oxford: Oxford University Press, 2004):
279-300.
22 Evans, Thinking Twice, 209.
23 See also Keith E. Stanovich, Richard F. West, and Maggie Toplak, “Individual Differences as
Essential Components of Heuristics and Biases Research,” in The Science of Reason, eds. Ken
Manktelow, David Over, and Shira Elqayam (New York: Psychology Press, 2012), 335-396.
16
Table 2. The Conceptual Structure of Rational Thought24
24Reprinted from Keith E. Stanovich, Richard F. West, and Maggie Toplak, “Intelligence and
Rationality,” in Cambridge Handbook of Intelligence, eds. Robert Sternberg and Scott Barry
Kaufman (Cambridge: Cambridge University Press, 2011), 784-826, 799.
17
Guy Axtell
Critical reasoning dispositions differ from innate IQ and are trainable. In

explaining the significant individual differences in cognitive performance that the
heuristics and biases literature evidences, Stanovich urges us to distinguish
intelligence from rationality in order to give the proper share to each. For
instance, intelligence tests miss much; the magnitude of the myside bias in
individuals shows very little relation to intelligence, and avoiding it is one of
numerous rational thinking skills that are not assessed by IQ. “[I]ntelligence and
rationality occupy different conceptual locations in models of cognition,”25 and
must be measured using different tasks and operations. Individual differences in
biases and heuristics tasks are “more related to rationality than intelligence.”26
Measures of rational thinking dispositions are positively correlated with normative
performance on tasks and often predict unique outcomes. Of course motivational
factors are already in play when we speak of rational thinking dispositions. As
Evans explains,
For the kinds of problems where reflective reasoning is required, you also need
to have the disposition to apply effortful reasoning, rather than to rely on
intuitions and feelings. This disposition is partly a matter of personality, but is
also influenced by culture and context.27
So it is very important from a virtue-theoretic perspective that Stanovich

describes many response differences as stemming from differences such as these
rather than what IQ tests test for:
[M]ost importantly, IQ is a resource that you have to apply to a problem for it to
be of any use to you… so one cause of dysrationalia is that while a person of high
IQ could reason well, they're actually failing to engage their reflective abilities.
Instead, they are inclined by their personality or circumstances to rely on gut
feelings and intuitions, to be strongly influenced by prior beliefs (which may be
false) or prone to social influences by peers (who might not be bright).28
So central is this distinction between intelligence and rationality to

Stanovich’s version of DPT that he quite recently makes a ‘tripartite proposal’ on
which Type 2 reasoning subdivides into the algorithmic mind and the reflective
mind. This subdivision captures the platitude that being intelligent does not
25 Research shows “there is a positive correlation between IQ and rational thinking, but it is
relatively modest in size…people vary not only in their cognitive ability but also in their
disposition to think critically about problems they face…” (Keith Stanovich, Rationality and the
Reflective Mind (Oxford: Oxford University Press, 2011), 206).
26 Stanovich, Rationality and the Reflective Mind, 206.
28 Stanovich, Rationality and the Reflective Mind, 206.
18
guarantee being smart [see Table 3]. If you lack the required critical reasoning
dispositions or mindware, you will find it hard to escape Type 1 intuitive,
heuristic responses, even if highly intelligent.
Table 3. The tripartite structure and the locus of individual differences 29
Meliorism and a balance of inner and outer. VE and DPT motivate a moderately
‘Meliorist’30 position in what Stanovich and West call the Great Rationality
Debate. Meliorism contrasts with overtly Skeptical automaticity, ‘vicious mind,’ or
situationist claims on the one hand, and Apologist/Panglossian views on the other.
Stanovich writes that, “What has been ignored in the Great Rationality Debate is
individual differences,” something which he cites as devaluing the Meliorist
position that proponents of DPT support. Skeptics and Panglossians share an
unfounded bent towards underestimating or ignoring the degree of difference
29 Reprinted from Keith Stanovich, “On the Distinction between Rationality and Intelligence:
Implications for Understanding Individual Differences in Reasoning,” in The Oxford Handbook
of Thinking and Reasoning, 352.
30 See especially Stanovich, West and Toplak, “Intelligence and Rationality”; and Stanovich and
West, “On the Relative Independence of Thinking Biases and Cognitive Ability,” Journal of
Personality and Social Psychology 94 (2008): 672-695.
19
Guy Axtell
found in subject responses in the heuristics and biases studies. Melioristic attitudes
of piecemeal improvability of human reasoners are made more plausible by the
family of dual processing theories. The point is not unrelated to how Nancy Snow
concedes that virtues “might start out by being local,” while holding that “they
need not remain so,” and how making knowledge and skills chronically accessible
through training and habituation is often possible.31 Kahneman and Frederick for
example explore how “complex cognitive operations eventually migrate from
System 2 to System 1 as proficiency and skill are acquired.”32 The automation or
chronic accessibility of what was once a slow, effortful process he illustrates
through the ability of chess masters to very quickly and without great effort assess
the merit of chess moves.
Relatedly, Meliorists are generally moderate in respect to how to improve
cognitive performance, balancing what might be termed cognitive change and
environmental change. DPT’s prescriptive upshot is that of a balance of “teachable
reasoning strategies and environmental fixes”; improve the environment where
that helps to improve rationality, and improve individual skills and competences
directly through practice and education.33 Skeptics about character by contrast
criticize character education efforts. In discussing the prescriptive upshot of (ES),
Alfano proposes retaining virtue-talk in education primarily as something of a
motivational white lie that can enhance short–term cognitive performance
primarily by raising the effect of mood. Doris’ conclusion in Lack of Character still
more pessimistically suggests turning away from attempts to develop pedagogy for
integrated character to something like enlightened situation-management.34
In summary, DPT highlights an empirically well-grounded balance between
the inner and the outer. It explains the complex interactions necessary for the
successful exercise of different kinds of reasoning, and the trainability of T2
rational thinking dispositions. Behavior is seen as a complex function of the two
systems or types reasoning working in cooperation and competition with each
other. The importance of T2 skills and dispositions for decision making is evident
31 Snow, Virtue as Social Intelligence, 37.

32 Daniel Kahneman and Shane Frederick, “Representativeness Revisited,” in Heuristics and
Biases, 49-81, 51.
33 Stanovich, “Distinction,” 359.
34 Epistemic and moral paternalism rather than character cultivation is appealing to Skeptics.
Virtue theorists would reject this but could agree that there needs to be a balance between
managing situations and acquiring virtue (inner/outer management). Of course, managing
situations is partly the individual’s job, as the virtuous are often those that wisely avoid
temptations, etc. they know they are susceptible to, rather than those who show outstanding
continence or willpower.
20
from the way they monitor T1 responses and potentially correct for the biases that
fast, intuitive processes are most vulnerable to. According to Evans,
We actually know quite a lot, from the experiments conducted by psychologists,
about when the reflective mind will intervene in decision-making. This will
happen more often when people are given strong instructions for rational
thinking, for example to engage in logical reasoning and disregard prior beliefs.
There will be less intervention when people are given little time to think about
the problem, or are required to carry out another task at the same time that
requires their attention.35
Stanovich puts much the same point by saying that “mindware gaps most often
arise from lack of education and experience.”36
I conclude that reliabilist VE is not substantially out of accord with DPT,
either in terms of individual differences as anticipated and found in test results on
heuristics tasks, or in terms of their explanation. The cautious optimism that
character epistemology shares with moderate Meliorist social psychology supports
the possibilities of substantially improving education for individual rational
thinking dispositions. It therefore finds quite congenial the argument that grounds
this possibility empirically in the basic distinction between largely innate IQ and
largely acquired intellectual habits and skills. Dual process theory can help
philosophers address the question of credit worthiness for success of heuristic
strategies and T1 processes. At the same time, DPT allows us to reject Alfano’s
dilemma regarding VE and inferential knowledge as based upon the false
dichotomy, the forced choice between treating heuristics en toto as either virtuous
or non-virtuous. Finally, together with the other considerations we have raised, it
reveals his first key thesis, (ICS), as a generalization based upon that false
dichotomy (i.e., a principle fallaciously claiming that people employ heuristics
‘rather than’ manifesting reliabilist virtues). The next section turns attention to
studies that Alfano uses from social psychology to support a substantially different
challenge, one aimed at responsibilist and inquiry-focused forms of VE.
3. Responsibility, Reflective Virtues, and Social Environment

When situationists criticize ‘classical’ (neo-Aristotelian) and ‘inquiry-focused’
(Peircean/Deweyan) virtue responsibilism as empirically inadequate, they
naturally enough utilize studies from social psychology.37 The previous section on

36 Stanovich, “Distinction,” 356.
37 John Dewey was very concerned not just in formal reasoning, but in how we think. His
account of habit is highly attentive to the philosophical importance of entrenched aspects of our
21
Guy Axtell
Alfano’s dilemma for virtue reliabilists was cast in terms of epistemic credit
restricted to achievement of epistemic goods through an individual’s ability, but
Alfano’s suggestion is especially pertinent to testimonial knowledge and to
discussion of reflective intellectual virtues like intellectual courage, humility, and
trust.
One proposal I appreciate in Alfano’s Character as Moral Fiction is a need to
move away from treating individuals as the bearers of virtue, to thinking of them
as “a triadic relation among the agent, a social milieu, and an environment.”38 But
apart from the attempts by some thinkers to put demanding intellectual
motivation conditions on knowledge,39 the suggestion to conceive reflective
virtues in this ‘triadic’ way could be related to a wealth of recent work among
responsibilists on collective and group virtues, and beyond this to strong overlaps
one finds among character epistemology, social epistemology, and feminist
epistemology today.40 As inquiry-focused VE does not partake of what Alfano
describes as Linda Zagzebski’s ‘classical responsibilism,’ I will not defend her
conditions on knowing.41 We will instead attend to Alfano’s more general claim
that “the intellectual virtues traditionally countenanced by classical responsibilism
cognitive architecture, as well as situational factors within our environment. Habit is the fixed
routine of activity which normally predominates, often manifesting in behavior in which
consciousness may play only a token role. He states that people often know more with their
habits, not with their consciousness; Action may take place with or without an end-in-view,
and in the latter case, there is simply settled habit. It is only if a problematic situation arises that
habit is disrupted and impulse proves inadequate. At this point, if we have the needed flexibility
and metacognitive wherewithal, more effortful thinking intervenes to help resolve the
problematic situation, or it does not, and the result is likely to be unsatisfactory. The
adjustments are only successful as we have the flexibility of mind to apply a strategy of inquiry
well-adapted to the particular situation.
38 Alfano, Virtue and Moral Fiction, 146.
39 Brendel distinguishes between holding that virtues that function as means of acquiring
knowledge are rightly described as dispositions, and the attempt to define knowledge in terms
of these virtuous dispositions. I agree with her description of Linda Zagzebski’s form of virtue
responsibilism as leading to ‘a counter – intuitive and intellectually over-loaded concept of
knowledge.’ See Elke Brendel, “The Epistemic Function of Virtuous Dispositions,” in Debating
Dispositions: Issues in Metaphysics, Epistemology and Philosophy of Mind, eds. Gregor
Damschen, Robert Schnepf, and Karsten Stüber (Berlin: De Gruyter, 2014). I have elsewhere
described inquiry-focused VE as holding an ‘overlap’ model of ethical and intellectual norms, in
contrast to Linda Zagzebski’s ‘reduction’ model. Alfano seems to mean something else by
inquiry-focused VE than I do.
40 Work on civic and collective/group virtues and on theory virtues in science apply aretaic
concepts without supposing these to be personal traits.

41 Linda Zagzebski, Virtues of the Mind (Cambridge: Cambridge University Press, 1996).
22
and inquiry responsible are empirically inadequate.”42 The support he offers for
this empirical inadequacy charge is the force of the thesis he calls Epistemic
Situationism (ES):
(ES) Most people’s conative intellectual traits are not virtues because they are
highly sensitive to seemingly trivial and epistemically irrelevant situational
influences.43
Alfano’s key theses (ICS) and (ES) are broad generalizations that are
purported scientific conclusions from psychological experiments. The main studies
that Alfano uses to support his generalization (ES) are the famous Asch Line Task
studies of the early 1950’s, and its follow-ups. Staying with the approach of strong
empirical rebuttal, I now want to show that Alfano’s use of the Line Task studies
depends upon a misleading re-interpretation of them. They do not strongly
support the lesson he wants and needs to draw from them. Firstly, Alfano does not
mention that the original Asch studies on conformity took place on American
subjects exclusively, and during the McCarthy Era when lack of conformity was
often identified with lack of patriotism and socially ostracized on that basis.
Alfano’s interpretation of the results of these studies is furthermore substantially
at odds with the experimenters’ own conclusions. I will now argue that his
interpretation is at odds with their own statements both about the variability of
responses found on the Line Task, and about the explanations of these differences
in response, where Alfano blatantly ignores Asch’s and Milgram’s explanation
involving cultural variances.
Variability of responses: Asch did conclude that a majority can influence a

minority even in an unambiguous situation in which the correct answer is
obvious, confirming (versus M. Sherif) that majority influence is stronger than had
been thought. But he saw the studies as clearly showing that people are capable of
greater or lesser ‘strength’ in resisting peer pressure. “Among the independent
individuals were many who held fast because of staunch confidence in their own
judgment.”44 What Asch and his colleagues found to be ‘the most significant fact’
42 Alfano, Virtue as Moral Fiction, 185.

43 Alfano, Virtue as Moral Fiction, 162. See also Alfano’s earlier “Expanding the Situationist
Challenge to Responsibilist Virtue Epistemology, Philosophical Quarterly 62, 247 (2012): 223-
249.
44 Quotations taken from S. E. Asch, “Opinions and Social Pressure,” Scientific American 193, 5
(1955): 31-35. Retrieved July 5, 2016, from http://www.columbia.edu/cu/psychology/terrace

/w1001/readings/asch.pdf. In a later publication, Asch does again say that “errors increased
strikingly” among subjects as majority increased, this being measured against control group. Yet
23
Guy Axtell
about the fully 25% best-performers who were consistently independent “was not
absence of responsiveness to the majority but a capacity to recover from doubt and
to reestablish their equilibrium.” They also found that the degree of independence
increases with the deviation of the majority from the truth and with “decreased
clarity in a situation.” They held that the study “establishes conclusively that the
performances of individual subjects were highly consistent”: those who showed
independence from or conformity with a majority over which of three presented
lines was longest continued the same pattern of response over time. This last
conclusion from the data deserves special mention since it seems quite
inconsistent with the claim of (ES) that most people are highly sensitive to
situational variables.45
What degree of independence is needed to demonstrate significant
individual differences is a matter of much dispute among psychologists, but
Alfano’s exclusive emphasis on the degree of conformity relative to a control
group not exposed to peer pressure presents a one-sided reading, substantially at
odds with Asch’s own conclusions emphasizing very substantial individual
differences in motivation or ability to resist peer pressure, and high individual
consistency over time. Alfano emphasizes that unanimity among confederates
‘produced striking conformity’46 in test subjects, while Asch himself emphasizes
‘startling differences’ and strong ‘consistency’ among these same subjects. Neither
insight is strictly wrong, but neither one is the whole story.
Explanation of the data: Foremost among questions that Asch says his studies raise
is the question of the extent to which these ‘startling differences’ can be attributed
to ‘sociological or cultural conditions.’ Asch concluded by suggesting that cultural
values are among the primary explainers, and that
the tendency to conformity in our society…raises questions about our ways of
education and about the values that guide our conduct…anyone inclined to draw
he also concludes that “Individuals responded in fundamentally different ways to the opposition
of the majority,” and that “Despite the effect of the majority the preponderance of estimates
was, under the present conditions, independent of the majority.” (“Studies of Independence and
Conformity: A Minority of One against a Unanimous Majority,” Psychological Monographs:
General and Applied 70 (1956): 1-70, DOI: 10.1037/h0093718. Retrieved March 24, 2016,
from http://psyc604.stasson.org/Asch1956.pdf (12).) Note also that if the line task is a ‘rapid
response’ task, recent DPT research predicts that this ‘taxing’ effect on subjects will result in
lowered performance, even independently of the ‘majority effect’ that Alfano tries to say
undermines the robustness of intellectual courage as a character trait.
45 Asch, “Studies,” 20.
46 Alfano, Virtue as Moral Fiction, 183.
24
too pessimistic conclusions from this report would do well to remind himself that
the capacities for independence are not to be underestimated. 47
Stanley Milgram in 1961 conducted follow-up comparative studies in part

to see if Asch was right about cultural values deeply impacting response to peer
pressure. Comparing French and Norwegian subjects, he found the hypothesis of
cultural variation to be corroborated by his study. While his conclusions were
phrased tentatively (“These findings are by no means conclusive”), he reported
that “No matter how the data are examined they point to greater independence
among the French than among the Norwegians.” Yet when Alfano does mention
Milgram’s “Nationality and Conformity,” he leaves its guiding hypothesis of
cultural variation totally out of his presentation, instead re-interpreting this study
in a self-serving way as just re-confirming a lesson about humans taken
collectively: that character-traits are routinely swamped by epistemically-
irrelevant situational variables like majority effect. So he writes that “As in the
original Asch study, unanimity produced striking conformity.” But while Alfano
cites situational factors like majority size as salient, Milgram’s own conclusion
actually highlighted the salience of cultural influences supported by the finding
that “in every one of the five experiments performed in both countries the French
showed themselves to be the more resistant to group pressure.”48 The situationist
interpretation of Line Task studies fails to account for the importance of cultural
influence on conformity, as tested across time (Americans in the Fifties vs. the
Sixties, say), or across cultures (French versus Norwegians). Alfano also omits any
mention of the prominent 1996 meta-study of Asch-type line judgment tasks,
which actually drew upon 133 studies in 17 countries. The Bond and Smith meta-
study found that “levels of conformity had steadily declined since Asch’s studies in
the early 1950s”; in respect to just the U.S. studies it found “that the date of study
was significantly negatively related to effect size, indicating that there has been a
47 Asch, “Opinions,” 34: “At one extreme, about one quarter of the subjects were completely
independent and never agreed with the erroneous judgments of the majority. At the other
extreme, some individuals went with the majority nearly all the time.”
48 “Twelve per cent of the Norwegian students conformed to the group on every one of the 16
critical trials, while only 1 per cent of the French conformed on every occasion. Forty-one per
cent of the French students but only 25 per cent of the Norwegians displayed strong
independence. And in every one of the five experiments performed in both countries the
French showed themselves to be the more resistant to group pressure.” (Stanley Milgram,
“Nationality and Conformity, Scientific American 205, 6 (1961). Retrieved July 3, 2016
https://www.scientificamerican.com/article.cfm?id=milgram-nationality-conformity&print=
true.)
25
Guy Axtell
decline in the level of conformity.”49 The section comparing national and national
trans-national studies concluded that
the impact of the cultural variables was greater than any other, including those
moderator variables such as majority size typically identified as being important
factors. Cultural values, it would seem, are significant mediators of response in
group pressure experiments.
Here again, what these (non-situationist) authors conclude to be the variable

whose impact on task response was ‘greater than any other’ is allowed no
significant role in the situationist re-interpretation of these same studies.
These are all reasons to think that the Asch Line Task studies do not provide
the evidence needed to make Alfano’s thesis (ES) a strong generalization from
social psychological experiments. We should for the same reasons of weak
empirical support reject Alfano’s associated claims that “virtues identified by
inquiry responsibilism are not the sorts of traits that many people possess,” and
that “rather than being intellectually courageous, people are at best intellectually
courageous unless faced with unanimous dissent of at least three other people.” It
appears that theoretical orientation deeply affects the interpretation that
situationists like Alfano apply to the studies they cite and discuss. Add to this that
the studies they cite are highly selective and often quite dated ones pulled from a
much larger literature on biases and heuristics. Both points greatly diminish the
support these studies can provide to the situationist’s generalizations about human
behavior.
To digress for a moment, one point that critics have made against
behaviorist and situationist methodology is that individual differences in
personality or character-related effects routinely get ‘construed as noise’ and
therefore deprived of due consideration.50 Recognition of quite significant
49 Rod Bond and Peter Smith, “Culture and Conformity: A Meta-Analysis of Studies Using
Asch's Line Judgment Task,” Psychological Bulletin 119, 1 (1996): 111-137, 124-125,
http://psycnet.apa.org/?&fa=main.doiLanding&doi=10.1037/0033-2909.119.1.111.
50 John Kihlstrom argues that understood historically, it is because of what situationism inherits
from behaviorism that situationists are predisposed to interpret social psychological studies
largely in terms of behavioral reflexes upon a set human nature, or less strongly, in terms of
social behavior varying as a function of features of the external environment. Kihlstrom
observes that “In such research, the effects of individual differences in personality are generally
construed as ‘noise.’ This view is captured by what might be called the Doctrine of Situationism:
‘Social behavior varies as a function of features of the external environment, particularly the
social situation, that elicit behavior directly, or that communicate social expectations, demands,
and incentives’…. Situationism has its obvious origins in stimulus-response behaviorism.” (John
26
differences in performance, whether attributable more to individual or to cultural

character, threatens their conception of human nature as varying largely only as a
function of features of the external environment. Alfano for his part calls it a
bogeyman to identify situationists with a crude behavioristic model of conduct; he
claims that it holds appropriate place for deliberation and sensitivity to reasons in
the determination of conduct in an agent. But that Alfano ‘blinks’ over all matters
related to the salience of cultural values in the explanation of differences in
response among test subjects suggests otherwise. I argue that it clearly shows an
inadequate acknowledgment of social milieu, and in this sense betrays their
avowed interactionist, triadic relationship between agent, social milieu, and
environment. To cross the line between social and cognitive psychology for a
moment, we earlier discussed DPT as defending the reality and importance of
individual differences (Section 2, above). But the role of cultural values in
explaining the variability of responses is easily accommodated and indeed
supported in DPT. As Manktelow summarizes in Thinking and Reasoning, studies
affirm that the new mind is more heavily influenced by culture and formal
education. In Stanovich’s terms, people can acquire new ‘mindware,’ and the
mindware that people employ is influenced by their culture.
Concern with interpretations of social psychological experiments that
ignore the impact of cultural values on task performance highlights the fact that
the situationist generalization draws upon a base of mostly latitudinal or ‘single
pass’ studies. The lessons that must be learned between psychology and
philosophy, this should remind us, run both directions. Methodologies that
purport to draw broad generalizations but are based mainly on single-pass
heuristic task studies seem to me as open to critique as are studies of moral
reasoning narrowly dependent upon intuition-pumping and ‘Trolley-ology,’ or
traumatic, dilemma-focused tests. Philosophers and psychologists have both been
at fault, and only working together will philosophers and psychologists better
address the Descriptive-Normative Gap problem introduced at the start of this
paper. But before discussing that problem directly in the final section, let me say
more about virtue epistemologies themselves, and different forms of they take,
since everyone agrees that the situationist challenge affects different forms of VE
differently, and some more strongly than others.51
F. Kihlstrom, “The Person-Situation Interaction,” in Oxford Handbook of Social Cognition, ed.

Donal E. Carlston (Oxford: Oxford University Press, 2013), 2-3.)
51 I am taking ‘credit theory’ broadly enough to include both Robust (RVE) and anti-Luck
(ALVE) virtue epistemologies. Responsibilist and inquiry-focused VE cut across Jason Baehr’s
distinction between autonomous and conservative VE; Inquiry-focused VE may still offer an
27
Guy Axtell
Robust, global character-traits and the responsibilist forms of VE that are

most concerned with these ‘thick’ characterological and affective concepts are a
natural first target of situationist critique. I think of the specific form of VE I
support – inquiry-focused VE, or as I have sometimes also termed it, zetetic
responsibilism in contrast with Zagzebski’s phronomic responsibilism – as
champion of things diachronic. This means it supports and investigates both the
backwards-looking concern with the etiology of particular belief, and forward-
looking – let us call this axiological – concern with the need to reflect upon or
improve one’s cognitive strategy and/or evidential situation and to fit one’s
epistemic goals, etc.52 The first is a primary matter for virtue reliabilists and any
account of doxastic justification; the second is a concern for both reliabilists and
responsibilists, but is not closely connected with epistemic assessment or the
project of the analysis of knowledge.
There is a close if not perfect connection between philosophical concern
with the diachronic as crucial to both ethical and epistemic normativity, and
dissatisfaction with situationist social psychology as generalizing latitudinal
studies. There is no dearth of longitudinal studies, but much of it exists over the
gap between situationist and automaticity theory, on the one hand, and positive
psychology on the other. Blaine Fowers and his colleagues for example try to
operationalize Aristotelian virtue theory. It is not incidental to this that in his
book Psychology and Virtue, Fowers argues that taking latitudinal studies as a
basis for ‘judging timelessly’ is highly problematic. If humans don’t just encounter,
but also help construct their environments, then situational and agential factors
are not as easily sorted out as it might appear when we draw only from latitudinal
studies.
If we are serious about exploring whether character strengths actually manifest
in behavior, a very different approach to the research [than latitudinal studies
take] is necessary. Investigators have to assess three essential components of a
trait or character strength. First, there must be individual differences on the
account of knowing even while re-envisioning epistemology in the Deweyan fashion of theory
of inquiry. For opposing views on whether virtue epistemologies really build more empirical
assumptions into their accounts than do other, more generic forms of reliabilist externalism see
David Henderson and Terrance Horgan, “Epistemic Virtues and Cognitive Dispositions,” in
Debating Dispositions, 296-319. Also Christian Nimtz, “Knowledge, Abilities, and ‘Because’-
Clauses,” in Knowledge, Virtue, and Action: Putting Epistemic Virtues to Work, eds. Tim
Henning and David P. Schweikard (London: Routledge, 2013).
52 See Guy Axtell, “Recovering Responsibility,” Logos & Episteme II, 3 (2011): 429-454 for
defense of the epistemic centrality of diachronic and not just synchronic evaluations, as
internalist evidentialists insist.
28
characteristic. Second, evidence for consistency in trait-associated behaviors is
needed across situations. Third, there must be consistency in trait-associated
behavior over time.53
Connecting with Fowers second and third point, Jesse Prinz points out that
“Factors external to a person can influence behavior in two different ways:
synchronically or diachronically,” but that latitudinal studies largely capture only
synchronic influence.54 I find each of these distinctions especially pertinent to the
prospects of tying philosophy and the social sciences closer together. For example,
sociologist Gabriel Abend argues that if the contemporary science of morality
focuses on situations or courses of action judged as ‘right,’ ‘okay,’ or ‘permissible,’
recorded through the push of a button, then it “is not a science of morality, but of
thin morality only.” Much as social scientists are rightly skeptical of philosophical
ethics presented as dilemma-cases and Trolley-problems, virtue theorists are
skeptical of a long tradition in social psychology focused on a ‘thin’ conception of
moral and cognitive reasoning.55
We can end this section with a passage from Prinz, who argues that the
difference between synchronic and diachronic influence matters vitally, because
the two forms of influence hold quite different theoretical lessons and
implications:
If we were swayed only by synchronic factors, then all people would be the
same: put two people in the same situation, and they will probably do the same
thing. But, if diachronic influences are possible, then people can internalize
social norms, and, as a result, people with different backgrounds will behave
differently in the exact same situations. If all people behaved alike in the same
situations, character based ethical theories would be in trouble: it would be
impossible to cultivate character. At best, we could do what Doris recommends:
try to put ourselves in situations that promote good behavior. But, if diachronic
influence is possible, the cultivation of character is possible… Cultivating
53 Blaine Fowers, Virtue and Psychology: Pursuing Excellence in Ordinary Practices

(Washington DC: American Psychological Association, 2005), 40.
54 Jesse Prinz, “The Normativity Challenge: Why Empirically Real Traits Won’t Save Virtue
Ethics,” Journal of Ethics 13 (2009): 117-144, 132.

55 I have sometimes described my fellow virtue responsibilists as epistemological ‘thickies.’
Abend points out that “Unlike thin predicates, thick predicates have institutional and cultural
preconditions or presuppositions … [T]he moral concepts and properties expressed by those
predicates – e.g., the concepts and properties of humanness, gentlemanliness, piousness – are
partly constituted by institutional and cultural facts.” (Gabriel Abend, "Thick Concepts and the
Moral Brain,” Archive of European Sociology LII (2011): 143-172.) For a thin-centered account,
see Thomas Hurka, “Virtuous Act, Virtuous Disposition, Analysis 66, 1 (2006): 69-76.
29
Guy Axtell
virtuous traits that do not get overwhelmed by synchronic situational variables
may be difficult, but there is no reason to think it’s impossible. 56
4. The Normative-Descriptive Gap and the Olin/Doris ‘Trade-off’ Thesis

According to Olin and Doris in their “Vicious Minds” paper (2013), it is “insofar as
reliability is a condition on epistemic virtue” that we have reason to doubt that
human organisms possess such virtue. The unreliability of agents follows from lack
of ‘stable’ dispositions, due to our enormous vulnerability to irrelevant context
effects. In order to formalize their challenge to VE they present the following
dilemma involving a forced trade-off:
[V]irtue epistemologists encounter a dilemma: they can either formulate their
theories to successfully accommodate the empirical challenge and lose the
normative appeal derived from virtue ethics, or retain the normative appeal
derived from virtue ethics and fall prey to the empirical challenge.57
We can call this the trade-off problem, because it is based on a thesis about
a forced option between maintaining normative appeal and empirical adequacy.
An embedded assumption I want to focus on is the trade-off thesis:
pressures on theory building in virtue epistemology are hydraulic: increase
empirical adequacy at the expense of normative appeal, or increase normative
appeal at the expense of empirical adequacy.58
Olin and Doris argue that virtue epistemology seems skepticism-inviting,

but notice first that if the challenge to agent-reliability affects externalist and
internalist epistemologies both, it may be that their own view is the one that
invites skepticism. At least they say nothing to indicate what conditions would set
a reasonable bar so that knowledge is not fleetingly rare. It looks like our modes of
belief production on the vicious mind thesis are neither safe nor sensitive, so an
analysis of knowledge involving tracking conditions or other sorts of counter-
factuals may not pass muster either.59 But more charitably we can take their
56 Prinz, “The Normativity Challenge,” 132.

57 Lauren Olin and John Doris, “Vicious Minds,” Philosophical Studies 168, 3 (2013): 665-692.
58 Olin and Doris, “Vicious Minds,” 32.
59 It is hard to separate these issues of skepticism and anti-skepticism from concern with
epistemic luck, which the authors do not mention; the distinction between Robust (RVE) an
Anti-luck virtue epistemologies (ALVE) is quite as pertinent as the distinction between virtue
reliabilists and responsibilists (and maybe more so, if the latter is largely a matter of emphasis
and any VE must account for both knowledge-constitutive and auxiliary virtues). But luck isn’t
treated in Olin and Doris’s paper. Since I am interested in virtue-theoretic responses to
skepticism, I would argue for the advantages that allowing ALVE’s independent anti-luck
30
concerns as focusing on credit theories, and more specifically on problems with an

aretaic condition, including its best-known formulations in terms of a
dispositional condition (Sosa) or an explanatory ‘because’ (Greco; Zagzebski).
Like Alfano, Olin and Doris do acknowledge my earlier comparison
between the generality problem in epistemology and the situationist challenge to
character-traits.60 While they write that they “are sympathetic to the observation
that the search for virtues and the search for reliable belief-forming process types
may be partly coextensive,” they deny my argument that this attenuates the
situationist challenge to the credit theory; they anyway find humans to be without
many stable belief-forming process types. Since my claim of partial co-extension is
granted but Olin and Doris do not comment on how I try to resolve this problem
with aNarrow-Broad Spectrum of Agency-Ascriptions [Table 4], I will use this
opportunity to provide a fuller development of how a properly modeled spectrum
of agency-ascriptions assuages the concerns that Olin and Doris raise.
condition has over RVE attempts to have a single aretaic condition serve to deal with both the
value problem and the epistemic luck problem. In addition to this key advantage I have
elsewhere argue for, ALVE affords us a substantially smaller ‘empirical footprint’ than RVE. I
cannot here explain my ‘smaller empirical footprint of anti-luck VE relative to Robust VE’
claim, but see Jesper Kallestrup and Duncan Pritchard, “The Power, and Limitations, of Virtue
Epistemology,” in Powers and Capacities in Philosophy: The New Aristotelianism , eds. Ruth
Groff and John Greco (London: Routledge, 2012), 248-269 and “Robust Virtue Epistemology and
Epistemic Anti-Individualism,” Pacific Philosophical Quarterly 93, 1 (2012): 84-103 on
advantages of ALVE (sometimes called ‘Weak VE’) in accounting for veritic luck cases especially
in regard to what they term epistemic dependence, and the implausibility of any version of VE
that cannot account for it. Pritchard also recognizes a spectrum of weak and strong epistemic
achievements, where in many instances one can meet the rubric for cognitive achievements
pretty easily, while in other instances the agent must overcome a significant obstacle to
cognitive success, or manifest high levels of cognitive skill. See Guy Axtell, “Felix Culpa: Luck in
Ethics and Epistemology,” Metaphilosophy 34, 3 (2003): 331-352; and “Two for the Show: Anti-
Luck and Virtue Epistemologies in Consonance,” Synthese 158, 3 (2007): 363-383, for my first
forays into a version of ALVE, which I ironically proposed in responding to Pritchard’s early
Robust Anti-luck epistemology (RAL), the position he took in this first book, Epistemic Luck
(Oxford: Oxford University Press, 2005).
60 Olin and Doris, "Vicious Minds," note 16.
31
Guy Axtell
Table 4: The Narrow – Broad Spectrum of Agency-Ascriptions61
Narrowly-typed Abilities (NTA) Broadly-typed Abilities (BTA)
— Low-level virtues (faculty virtues). — High-level virtues (reflective virtues). Best

Dispositions construed as genetically- suited to tell about the agent’s abilities and
endowed abilities or cognitive practices in a certain domain/area.
capacities.
— Best suited to holistic evaluation of agents,
—Best suited to evaluating the etiology of a including the quality of their activities of
single belief or narrow range of beliefs; inquiry.
—Their ascription is not primarily to doxastic

—Their ascription is often keyed to doxastic justification of particular beliefs. Their
justification of particular beliefs ascription is not to the ideology of
actually held by an agent. ‘processes,’ but to the axiology of inquiry or
‘ideal-types.’
—Their ascription is primarily occurrent,

—Their ascription answers to the Generality describing personal habits or counter-factual
Problem by fixing the “narrowest, states of motivation an agent who performs a
content-neutral process that is certain act-type in the given situation (and
operative in belief production” for an ethically or intellectually virtuous agent)
actually held belief. 62 would have.
— The value of high-level virtues attaches directly

— The value of low-level virtues is to their possessor but only tenuously to their
transmitted directly to their products products.
and only indirectly to the agents who
have them.
To develop the role of this N-B Spectrum in responding to the dilemma, I

first want to distinguish between two senses of the trade-off thesis that drives Olin
and Doris’ dilemma, replying to each sense in turn. The formal sense of the trade-
off thesis indicates a general trade-off between pursuing normative assessment and
pursuing descriptive adequacy. The substantive sense makes as I understand it a
stronger and more directed claim: Proponents of an aretaic or person-level ability
61 See also Lepock “Unifying the Intellectual Virtues,” from which my present chart and that of
Axtell, “Character Trait Ascriptions,” draws many key points. Thanks to Christopher for
discussion of his distinction and chart.
62 Alvin I. Goldman, Epistemology and Cognition (Cambridge: Harvard University Press, 1988),
363.
32
condition face an especially difficult or dire version of this problem (relative to

other (unspecified) potentially successful anti-skeptical epistemologies). They
force this problem because whenever an aretaic condition is formulated so as to be
empirically adequate (psychologically realizable) it will turn out to lack normative
appeal, and whenever such a condition is formulated to have strong normative
appeal, it will turn out to be empirically inadequate (psychologically unrealizable).
The formal sense of the trade-off thesis is a general warning about
conflating explanatory and normative posits; if this is correct, it is largely a matter
of logic. In response to this formal sense of the ‘trade-off’ problem, the Narrow-
Broad Spectrum distinguishes different ways of ascribing epistemic agency to
people. This formal sense makes no great worry for normative epistemologies
because it is already acknowledged by the distinctions that the spectrum uses to
avoid conflations that otherwise occur. Allowing that agency ascriptions serve a
variety of purposes, as our spectrum does, means that what has the most empirical
content need not have much normative appeal, and what has the most normative
appeal need not have much empirical skin in the game.
The formal sense of the trade-off thesis locates it on the familiar grounds of
the normative-descriptive gap. In so saying, however, familiar resources for
responding also come to mind. The dilemma reflects the much-discussed and
troublesome relationship between the normative appeal of unbounded rationality,
with standards that ignore time, information, and computation of agents, and
bounded rationality, where the scientific image, the real world, rushes back in.
That we will be forced to judge humans as broadly irrational if we judge
rationality by an ‘Enlightenment picture’ of unbounded agency when this is
untrue of our actual way of being in the world, has been a common argument in
the great rationality debate.63 This is a rubbing point between N-theories and
ecological rationality. But virtue epistemologies are not committed to any
particular view on these broad issues. For them the formal sense of our dilemma
just seems to be this well-known problem, re-directed towards conditions on
knowledge rather than on rationality. There always is and needs to be a contrast of
performance with competence, but standards of competence that arise entirely
independently of psychology from ideals of pure logic or ideal agency, may lead us
to doubt human competence universally.
Responding to the substantive sense of the ‘trade-off’ thesis is where our
real work lies. Here there is more than a general warning, but a real risk of ‘costs,’
63See David Matheson, “Bounded Rationality, Epistemic Externalism and the Enlightenment
Picture of Cognitive Virtue,” in Contemporary Debates in Cognitive Science, ed. Robert J.
Stainton (Malden: Blackwell, 2006), 134-144.
33
Guy Axtell
in the present case to a credit theory of knowing. If we cannot say how and why
this theory can avoid such conflations, or explain what legitimate epistemic
concerns broadly-typed reflective virtues serve in an analysis of knowledge except
to be causal – explanatory posits, then the theory looks to be in trouble. This
requires a more detailed exposition of our spectrum. Responding to the
substantive sense of the trade-off thesis requires making several further
distinctions, including especially that between epistemological concerns with the
etiology of belief (also called doxastic justification or ex post justification), on the
one hand, and epistemological axiology together with norms of agential
rationality/responsibility on the other.
At the Narrow end of the spectrum lie ascriptions of particular mental
faculties and strategies, and at the Broad end lie more ‘occurrent’ ascriptions
associated with agency under normal or ideal circumstances. It is only NT traits
that directly address the generality problem, and hence doxastic justification, the
core question of an account of knowing on all but purely internalist
epistemologies.64 This point basically follows from the normal reliabilist line on
the generality problem, which is that the pertinent process-type to try to specify
for doxastic justification is the “narrowest, content-neutral process that is
operative in belief production.”65 NT traits are best ascribed to a particular belief of
an agent, to reconstruct its reliable etiology in a way that excludes all but modally
remote error-possibilities. Agency ascriptions serving this etiological function, we
can see, simply don’t require much normative ‘appeal,’ at least on the fairly
minimal account of credit we discussed in connection with DPT in section 2.
On the other hand, the Broad end of the spectrum is directly concerned
with praise; it is high praise, but does not in the same way entail epistemic credit,
since there are plenty of ways to be personally justified (to be both synchronically
rational as the evidentialist demands, and diachronically rational in our zetetic
activities as the virtues would guide us to be), yet still come out with a false belief,
or with a true belief not related in the right way to the agent’s ability, as in Gettier
cases. Thus, if by attacking the ‘empirical adequacy’ of BT attributions Olin and
Doris are assuming that such attributions always imply some robust disposition,
this is simply mistaken. BT (broadly-typed) virtues are typically ascribed in
assessment of an agent’s actions-at-inquiry – what I call zetetic activities – rather
than in order to credit an agent for coming to have a particular a non-luckily true
64 I do agree with some criticisms of a methodological individualist conception of knowledge,

and I take VE to have strong overlaps with social epistemologists.
65 See Goldman, Epistemology and Cognition, 363.
34
belief.66 The ‘normative appeal’ of agency ascriptions and descriptions of the BT

kind, I will need to argue, does not require them to have much empirical
exposure. Since broad reflective virtues do not consist in a single neurological state
of a person, that personal’s behavior may be our best or only overt reason for
ascribing such a dispositional property to them. The personal praise for broadly-
typed virtues can be dispositional or occurrent. Occurrent attributions associate an
agent with actions a virtuous or vicious agent would perform, or with motives
they would have. We can call them occurrent because these kinds of
counterfactuals demand only conformity with norms, and on the present view, are
substantially weaker than causal-explanatory attributions.
Here we bump up against one of the major differences between ‘inquiry-
focused’ VE and ‘classical’ responsibilism. I agree entirely with Olin and Doris’
point that ‘mixed’ accounts need to be careful what reliability and responsibility
conditions are accepted and how they are framed. Inquiry-focused VE argues that
intellectual virtues thought of as character-traits make us good at inquiry, but they
contribute to a formal account of knowledge only in indirect ways. Hookway
rightly notes that “Virtues regulate inquiries and deliberations, and only indirectly
regulate beliefs.”67 Since habits of mind acknowledged as intellectual virtues are
rather abstract and complex traits, they accordingly have “a broad variety of
possible manifestations in the intellectual activities.”68 The motivations of
reflectively virtuous inquirers normally do not regularly predict anything very
specific by way of beliefs or activities.
It is clear that occurrent attributions needn’t be explanatory in the way that
dispositional ones are, that the normativity of epistemic assessment and doxastic
guidance differ substantially, and that the third-personal concerns with the
etiology of belief (that the project of analysis requires) should be clearly
distinguished from the first-personal concerns of the inquirer, which norm to
what her epistemic community identifies and values as theoretical and/or personal
virtues. What is less well-recognized is that virtue theories in ethics and
epistemology can easily accommodate a narrow/broad, or causal-
explanatory/normative spectrum of agency ascriptions, where posits at the
66 Alfano, Character as Moral Fiction, 171, allows responsibilisms that “stand as a purely
normative theory” without ‘explanatory-cum-evaluative’ ambitions. But this is in tension with
claims he also makes that even Autonomous VE as Baehr defines it has conditions of
realizability for virtues that it cannot meet. This may be a problem of semantics over what is
designated by ‘character-traits.’
67 Christopher Hookway “How to be a Virtue Epistemologist,” in Intellectual Virtue, eds.
Michael DePaul and Linda Zagzebski (Oxford: Oxford University Press, 2003), 183-202, 197.
68 Elke Brendel, “Function of Virtuous Dispositions,” 330.
35
Guy Axtell
different ends of the spectrum are recognized as serving substantially different

albeit equally legitimate explanatory and evaluative functions.
Given this, it also follows that we should carefully distinguish not only
dispositional from occurrent attribution, but overlapping with it, epistemic credit
from praise. The Narrow end of our spectrum is concerned with epistemic credit,
but not necessarily praise, because it may not be fitting to praise or blame someone
for faculty virtues or ‘automatic’ responses, however successfully employed. Thus,
if by ‘normative appeal’ Olin and Doris mean praise and censure, then its absence
in NT ascriptions on the present account comes at no real ‘expense.’
In most cases the concerns with doxastic responsibility and other aspects of
BT virtues should be carried on apart from the project of analysis of knowledge.
This is the general lesson of the failure of internalist evidentialism, and of
intellectualist conceptions of mind, more generally.69 Doxastic responsibility, like
related concepts of rationality and personal justification, is taken as a generally
necessary condition on knowledge only by more internalistically-oriented
analyses. If one is going this route, the condition in question should be expressed
negatively and occurrently, in the sense that meeting the condition merely means
that the condition is met if we need not attribute to the agent motivations that an
intellectually virtuous agent would not have, or actions or omissions that a
virtuous agent would not perform or omit. Negative conditions on knowing (the
agent’s motivations were not vicious or her actions and omissions were not
irresponsible, for instance) obviously are judged in the way occurrent ascriptions
are, not by actual global-trait manifestation). And, of course, the more externalist
versions of VE are still more cautious about a right motivational condition on
knowledge. The focus for reliabilist VE is doxastic justification, which is what the
generality problem relates to. For externalists the broad end of the spectrum just is
not ‘truth-linked’ in the way that doxastic justification is.70
69See Axtell, “Recovering Responsibility.”

70 That responsibility conditions can easily be included in a formal analysis as negative rather
than positive conditions is something overlooked when ‘mixed’ epistemology is treated as the
epistemic situationist do, as always just adding another contentious generally necessary
condition on knowing. Knowledge-attributions arguably range from animal (‘brute’/externalist)
to reflective (high-end/internalist) cognition. The beauty of negative conditions is that they
allow sufficiency on knowledge in a plurality ways, and that they therefore fit with 'range
concepts.' This is why situationism challenges Zagzebski’s RVE more than ALVE. Since what
classical responsibilists really care about is conditions on ‘high-end’ or specifically reflective
knowledge, this is better put in a negative way of saying that the agent is ‘not ill-motivated’ or
that her efforts at inquiry and evidence-gathering were ‘not irresponsible,’ etc., than by a
general necessary condition on knowledge. My account thus accords with and adds support to
36
Some proponents of VE like Zagzebski who view the relationship between

epistemology and ethics to be tighter than I do, may disagree. For those who try to
reduce epistemic to ethical evaluation, or to treat personal responsibility as
equivalent to doxastic justification, I reiterate the demand for a more general
separation between the theory of knowledge and accounts of personal justification
(synchronic and diachronic rationality).71 We need to avoid the confusion of
treating traits and states at the Broad end of the spectrum as doing the work of
positive conditions on doxastic justification or knowledge, whose place is really
with the generality problem and at the Narrow end. Critics of ‘classical
responsibilism’ are probably right that trying to model epistemological
normativity on ethical normativity (the neo-Aristotelian approach) contributes to
these conflations.72
Our distinction between the formal and substantive senses in which the
trade-off thesis can be construed has directly informed our response to Olin and
Doris’ dilemma for VE. If theory construction and VE is ‘hydraulic,’ as they
contend, then is it any more so than for other philosophical analyses of
knowledge, and if so, why? Even granting that it is more so and there is a
substantive worry about virtue epistemologies in particular, it surely is unfair not
to allow virtue epistemologists the fluidity and flexibility of a spectrum of agency
ascriptions in order to clarify the specific functions and levels of empirical
exposure presupposed in different kinds of agency ascriptions. When we do, I
Kallestrup and Pritchard’s, which argues that ALVE fairs better on several dimensions,
including empirical adequacy concerns, than does RVE.
71 So consistent with my argument in “Recovering Responsibility,” that evidentialists like
Feldman should accept the ‘separation,’ ‘divorce,’ or ‘two project’ proposal (that R. Foley and
Anthony Booth persuasively argue for), I here place much the same demand upon phronomic
virtue responsibilism. This result favors my more moderate stance of zetetic or inquiry-focused
VE, which holds that reflective virtues are things that make us good at inquiry, and sometimes
part of a rational reconstruction of belief acquisition or maintenance, but not for that reason a
necessary condition on knowing.
72 While I think avoiding these conflations is best addressed by inquiry-focused forms of VE, our
account still agrees with Peter Samuelson and Ian Church when they write, “Presumably, any
robust, full account of intellectual virtue will have to account for both cognitive faculty-virtues
as well as character trait-virtues; whether one does this along agent-reliabilist lines or neo-
Aristotelian lines could be, to some extent, a matter of emphasis. For a person can hold a belief
more strongly (or weakly) than warranted due to biases inherent in one’s cognitive systems, or
due to some lack of character, just as a person can exhibit virtuous knowing via the proper
functioning of one’s cognitive system or through the exercise of a virtuous character.” (Peter
Samuelson and Ian Church, “When Cognition Turns Vicious: Heuristics and Biases in Light of
Virtue Epistemology,” Philosophical Psychology 28, 8 (2014): 1095-1113.)
37
Guy Axtell
believe that VE has advantages over other epistemologies in helping us address the
Normative-Descriptive Gap problem, which is just what philosophers and
psychologists working together should want.
A continuum or spectrum of agency-ascriptions helps to clarify what
normative and explanatory concerns we have with different cases and different
kinds of agency. But nothing we have said about agency ascriptions as they relate
to knowledge possession implies the unreality of global and robust reflective
virtues, or the idea that they are purely descriptive and never part of salient
explanations for why we know. The reality of BT character – traits needn’t be
called into question by acknowledging the spectrum’s range covering both
dispositional and occurrent attributions. Being responsible is a great way to
achieve epistemic goods like true belief, knowledge, and understanding. Indeed,
without a lot of luck, this may be my only way, or at least my best shot at it. And
the habits of responsible inquiry that I display today in coming to hold a true
belief will, when an assessor retrospects on it tomorrow, be part of that belief's
reconstructed etiology. But these contributions of responsibility to knowing still
should not lead us to confuse etiology and axiology, or to think of personal
justification (synchronic or diachronic) as guaranteeing these epistemic goods.
Believing truly rather than falsely does not follow from my having unquestionable
motives and giving my best cognitive effort. That is why Zagzebski’s ‘‘because’
condition is too strong, and why conforming to norms of motivation that a
virtuous agent would have, or performing the inquiries a virtuous agent would
perform, is generally sufficient to meet Broad-end norms associated with
epistemic responsibility.
To conclude, these concerns with the formal and substantive senses of the
‘trade-off’ thesis are serious ones for virtue epistemologists, Olin and Doris are
right to contend. But tying back to Section 2, DPT on my view also has the
philosophical implication of supporting both the reality and value of global
reflective virtues, moderating what can be claimed on empirical grounds about the
modularity, or localness of character-traits, and the lack of robustness of traits of
intellectual character. From a normative perspective, optimizing coordination of
T1 and T2 within our natural limits is of crucial philosophical and pedagogical
concern, especially since the parallel nature of T1 and T2 means they not only
cooperate, but also both routinely operate at the same time and quite often
compete in determining an agent's cognitive or ethical judgment.73 These latter
73Compare Lisa Grover’s argument that “We should accept the psychological reality of narrow,
localized character traits, while retaining the thick evaluative discourse required by virtue
ethics….Thick global concepts are necessary for a theory of localized character traits and
38
facts about how we process make the metacognitive prowess that comes with
acquiring rational thinking dispositions more necessary and more efficacious than
they appear to be on either situationist or automaticity ('System 1’) theory.
Habituating ourselves to rational thinking dispositions remains perhaps the most
powerful tool within our adaptive toolbox.74
situation management to make sense. Without evaluative integration of different local traits
under thick evaluative concepts we cannot identify which local traits to develop, and which
situations to seek out, or avoid.” (Lisa Grover, “The Evaluative Integration of Local Character
Traits,” Journal of Value Inquiry 46 (2012): 25-37, 36.
74 Thanks to Mark Alfano, J. Adam Carter, Abrol Fairweather, Lauren Olin, John Doris,
Anthony Booth, John Kihlstrom, Christopher Lepock, Christian Miller, and Holly Smith for
comments and discussion. Lauren in particular provided a thoughtful and detailed set of
comments on an early draft.
39
American Philosophical Quarterly
Volume 49, Number 3, July 2012
Recent Work
in Applied Virtue Ethics
Guy Axtell and Philip Olson
I. Introduction ethics, it contrasts especially with theories
T he use of the term “applied ethics” to

denote a particular field of moral inquiry
emphasizing acting in accord with universal
rules or duties, or acting in order to bring
about good consequences, and so on. Virtue
(distinct from but related to both normative ethicists highlight the moral importance of
ethics and meta-ethics) is a relatively new cultivating habits or dispositions such as
phenomenon. The individuation of applied generosity, courage, humility, friendship,
ethics as a special division of moral inves- love, and honesty, along with their associated
tigation gathered momentum in the 1970s moral sensitivities. From increasing coverage
and 1980s, largely as a response to early- in textbooks in one or another area of applied
twentieth-century moral philosophy’s over- ethics to a growing number of essays, edited
whelming concentration on moral semantics collections, and monographs, applied virtue
and its apparent inattention to practical moral theory has clearly become a vibrant area of
problems that arose in the wake of significant philosophical research. Some would find this
social and technological transformations. a surprising development because they have
The field of applied ethics is now a well es- been antecedently convinced by its detrac-
tablished, professional domain sustained by tors that virtue ethics stands conceptually ill
institutional research centers, professional equipped to offer practicable moral guidance.
academic appointments, and devoted jour- (For direct discussion, see Louden 1984,
nals. As the field of applied ethics grew and Hursthouse 1995, Tiberius 2002, and Zyl
developed, its contributors predominantly 2009.)
advocated consequentialist and deontological Sheer numbers of researchers and publica-
approaches to the problems they address; but tions, of course, do not make the case against
lately a significant number of moral philoso- virtue ethics’ distracters for the importance of
phers have begun to bring the resources of virtue ethicists’ contributions to the methods
virtue ethics to bear upon the ever-evolving and results of applied ethical inquiry. It is
subject matters of applied ethics. therefore timely, we think, to make a more
Virtue ethics is an approach emphasizing comprehensive review of the state of research
the centrality of the role of character traits in applied virtue ethics today. The present
(virtues), the possession of which is needed essay tries to take an inclusive view of virtue
for persons to be good and to live well. As ethics, and therefore of applied virtue ethics,
a distinctive approach within normative admitting almost all methods and models
©2012 by the Board of Trustees of the University of Illinois
APQ 49_3 text.indd 183 5/14/12 9:29 AM

184 / American Philosophical Quarterly
that highlight the importance of personal, Oriented Approach to Environmental Ethics

social, and/or civic virtues for addressing (2007), in which he defends a virtue-theoretic
the problems that applied ethicists study. But approach to environmental ethics, providing
some of the more exemplary recent works an account of the nature of environmental
we will address include those in high-profile virtue and applying the resulting ethic to
collections as Virtue Ethics and Professional practical moral concerns (e.g., the genetic
Roles (Oakley and Cocking 2001), Virtue modification of crops). Most recently, the
Ethics and Moral Education (Carr and Steutel Journal of Agricultural and Environmental
1999), Working Virtue (Walker and Ivanhoe Ethics published a special edition, edited by
2007), Virtue Jurisprudence (Farrelly and Cafaro, devoted to environmental virtue eth-
Solum 2007), Environmental Virtue Ethics ics (Cafaro 2010).1
(Sandler and Cafaro 2005), and Sex and Eth- Before reviewing the research more closely,
ics (Halwani 2007). it should be noted that the relatively recent re-
Walker and Ivanhoe’s Working Virtue vival of virtue ethics, beginning a half-centu-
(2007) brings together thirteen essays by ry ago, poses some special problems, for our
many of the foremost moral philosophers understanding of the goals and methodology
of our time, covering a wide range of proof applied ethics have arguably been largely
fessions; bioethics; environmental ethics; formulated independently of the contributions
animal, legal, racial, and sexual ethics; that virtue ethics make. Sensitivity to the
global citizenship; and the role of civic and possible contributions of virtue ethics need
deliberative virtues in democratic societies. not require a radical rethinking of the aims of
In Farrelly and Solum’s Virtue Jurisprudence applied ethics, which remain centered on the
(2007), eight prominent authors working in use of moral norms or theories as resources
the areas of legal theory and moral philoso- for addressing practical moral problems that
phy contribute essays that campaign for an individuals and communities face (or may
Aristotelian (or neo-Aristotelian) alternative be expected to face) in their daily lives. But
to preference-based (consequentialist) and applied virtue ethicists do call into question
rights-based (deontological) normative legal certain fundamental suppositions that have
theories. Guided by Aristotle’s idea that “a guided earlier work in applied ethics, as-
state exists for the sake of a good life” (Poli- sumptions about what is applied in applied
tics III.9: 1280a, 32), the contributors to Vir- ethics (only general rules or principles?) and
tue Jurisprudence argue that the law should about the subject matters to which moral
have human flourishing as its ultimate end. norms and theories are to be applied (always
The rapidly growing field of environmental overt actions and states of affairs external to
virtue ethicists has recently seen a number moral agents themselves?). Applied virtue
of significant contributions. Sandler and ethicists provide normative resources that can
Cafaro’s Environmental Virtue Ethics (2005) in some instances be applied to actions but
offers thirteen essays by some prominent in other instances are better seen as address-
environmental ethicists with the explicit aim ing features internal to moral agents (e.g.,
of determining “the norms [of character] that affective states, deliberative processes, and
should govern our interactions with [nature]” perceptual sensitivities). If the suppositions
(p. 1) and providing “guidance on what at- about applied ethics that virtue ethics would
titudes and dispositions we ought and ought challenge (suppositions we will return to
not to have regarding the environment” (p. clarify further in our concluding section) were
2). Sandler has since published a manuscript allowed to define what applied ethics is and
titled Character and Environment: A Virtue must be, then virtue ethics would certainly
APQ 49_3 text.indd 184 5/14/12 9:29 AM

recent work in applied virtue ethics / 185
be at a disadvantage in comparison with act- scrutiny by broad-based ethical theories such

based approaches. But the point at present is as utilitarianism and Kantianism,” which led
only that if we acknowledge that taking virtue to the neglect of some practice-based pro-
ethics seriously may call upon us to transform fessional ethical norms in favor of judging
or expand our understanding of what it means professional behavior directly in terms of a
for ethical norms or theories to be applied, standard provided by one or another ethical
and of the range of subject matters to which theory (pp. 2–3; cf. Banks and Gallagher
those norms or theories are appropriately 2008, p. 41). Virtue theory’s development
applied, then the value of its contributions to has gained some strength by the impression
the field may be easier to recognize. that it offers a more contextually sensitive
The following sections survey recent and detailed treatment of professional roles
work in multiple subfields of applied ethics. and practices than do other leading ethical
The survey will begin with recent work in theories, better supporting the bottom-up
professional ethics and in education before approach of practitioners.2 This need not ren-
proceeding on to environmental ethics and der virtue ethics conservative. Virtue theory
other issues of public concern that focus less also has the means to criticize practices and
around specific professions. The final section to prescribe new virtues. But the bottom-up
attempts to draw a number of methodological approach does not overlook normatively
concerns out of this review of literature and significant differences between various pro-
to suggest some new directions for applied fessional contexts; instead, it begins with
virtue ethics going forward. practice-relative values, which may or may
not coincide with practice-neutral values.3
II. Professional Ethics This last point will be taken up more fully in
As moral education and the study of prob- the final section of this essay.
lems of practice became more explicit parts The top-down and bottom-up approaches
of the curriculum in medical and business here described are complementary, and
schools in recent decades, practitioner- although their closer integration remains a
academics have increasing come to value difficult task, it is aided both by thorough
educational and training strategies that in- acquaintance with professional roles and by
clude aspects of character education—even practice-centered accounts of virtue like that
in the absence of any robust commitments to of MacIntyre (1984). According to Oakley
virtue ethics as a normative theory (see also (2009), virtue ethics offers a “regulative
Horner 2000, Clark 2006, Radden 2007, and ideal,” a way for the agent to internalize the
Swanton 2007). Indeed, in many core areas relation between a favored ethical tradition
of applied ethics, the strongest impetus for a or theory’s criterion of rightness and its ac-
focus on character traits appears not to have count of how agents are to be guided by this
come top-down from virtue theory, but rather criterion (p. 1). But virtue ethics emphasizes
bottom-up, from the concerns of practitioner- that the regulative ideal of an ethical theory—
academics confronting difficulties that arise eudaimonia in some versions of virtue eth-
within practices or within a setting of training ics, for instance—need not take the form of
for new professions. codifiable principles or rules in order to play
In Virtue Ethics and Professional Roles a motivating role in moral deliberation and
(2001), Oakley and Cocking explain that “the judgment.4 Still, certain ends and principles
rise of systematic approaches to professional that govern practices such as medicine or law
ethics in the 1970s saw traditional practices often can, in fact, be codified, rendering them
in various professions subjected to critical more readily communicable and assessable
APQ 49_3 text.indd 185 5/14/12 9:29 AM

among members of a professional commu- activity specific to each profession. Each profes-
nity. Such is the case with the “statement sion has its own “act” of promise which invites
of values” in many businesses and codes of trust and entails certain virtue as a result. Each
ethics for various professions, which often has a morality internal to its end and the kind
of activity it is. (2007, p. 78)
directly incorporate virtue language in articu-
lating service ideals and expectations for best Some other authors see this appeal to each
professional practices. The choice of such profession’s telos as problematic on the
language perhaps reflects a view in which grounds that teleology itself “has become
“the ‘living morality’ of the ‘virtuous com- harder and harder to defend” as a source
munity of practitioners’ must be the starting of moral justification (Bertland 2009, p.
point to elaborate any code of [professional] 25). Although Bertland does not himself
conduct that stands a chance of being inter- explain what is problematic about teleology,
nalized by its members and, therefore, truly other authors have identified increased sen-
achieving its purpose” (Consoli 2008, pp. sitivity to apparently irreconcilable conflicts
241–242).5 among teloi as the root of the problem (e.g.,
Pellegrino’s discussion of medicine in Louden 1984; Zagzebski 1996, p. 200).
“Professing Medicine, Virtue-Based Ethics, Yet Pellegrino’s passage helps explain the
and the Retrieval of Professionalism” (2007) appeal that a practice-oriented approach to
attempts to contain any worry about uncodifi- the virtues has had for recent authors who
ability of virtue-ethical norms by employing emphasize the goods internal to practices
an end-based “schema.” This schema has such as doctoring and nursing (Armstrong
been especially influential in professional 2007, Barilan 2009, Holland 2010a, Holland
ethics (see also Oakley and Cocking 2001). 2010b, Walker 2005), psychiatry (Radden and
Though Pellegrino argues that virtue theories Sadler 2008, Robertson and Walter 2007),
in ethics cannot stand entirely on their own, counseling (Stewart-Sicking 2008), public
independent of principles and duties, he health care (Horner 2000; Larkin, Iserson,
still supports a fairly robust virtue ethics in Kassutto et. al. 2009), public administration
which virtue theory has more to offer profes- (Lynch 2004), and social work (Adams 2009,
sional ethics than do other moral theories. McBeath and Webb 2002, Pellegrino 2007,
Pellegrino anchors his position in the idea Pullen-Sansfacon 2010).
that professions are distinct human activi- Conceiving virtues as goods internal to
ties that link virtues to the “defining ends” practices heightens the contrast between the
of professional practice.6 With the medical external moral guidance provided by “thin”
profession, the “act of profession” which institutional norms instantiated in master
initiates the physician-patient relationship principles or strict decision procedures, on the
entails certain virtues that foster a moral one hand, and on the other, those resources for
medical community. This conceptual schema, moral deliberation that we find in culturally
Pellegrino thinks, has applicability to helping “thick” conceptions of character traits that
professions beyond medicine: provide psychologically internalized moral
With proper definition of the ends, peculiar motivations. These partly descriptive, partly
to each profession, this schema also defines evaluative character traits may be conceived as
the good of the lawyer’s client, the teacher’s general traits or as those more specific or local
student, and the minister’s penitent or parish- traits that we expect of good or bad lawyers,
ioner. As with medicine, the ends of these other physicians, patients, nurses, parents, accoun-
helping professions are linked to a particular tants, public servants, eco-citizens, and so on.
APQ 49_3 text.indd 186 5/14/12 9:29 AM

We have now seen how practitioner- virtue of ends, virtue of service, virtue of
academics have availed themselves of the belief, virtue of dialogue, virtue of action,
resources of virtue theory by regarding many and virtue of will. Athanassoulis and Ross
or most professions as moral communities (2009) offer a sophisticated virtue ethical
containing notions of the good that derive account of making decisions about risk that
from their core purpose or service ideal, bears especially on engineering.
and which can serve as reference points
for identifying the virtues within particular III. Education
professional contexts (Banks and Gallagher and Developmental Psychology
2008, p. 40). Business and management is Perhaps the most appropriate work on
another place where we find an abundance of which to focus first is the earliest of the area-
work underlining this turn away from “ideal focused collections here reviewed: namely,
theory” (Sandler’s term) in favor of conceiv- Carr and Steutel’s Virtue Ethics and Moral
ing virtues as ways to realize goods intrinsic Education (1999). This edited collection was
to practices. Gregg and Stoner’s collection, pioneering in updating and applying virtue-
Profit, Prudence and Virtue (2009), on the theoretic ideas and analyses to the practical
financial crisis of 2009 illustrates that virtue problems and concerns of contemporary
theory need not take practices as givens, but educators. Most of the contributors share
may be used to sharply critique practices as the view that a virtue ethical conception of
well as to prescribe reformed practices and moral education best accounts for the ways
new attendant virtues. Dobson (2007), Moore in which motivational factors must enter into
(2008), Oakley and White (2005), Solomon any truly effective appreciation for the value
(2003), and Vogel (2005) are among the most of moral principles. The claim is not that a
prolific virtue ethicists focusing on business virtues-focused approach to moral education
ethics. IT, advertising, media ethics, and is original, but rather that it reflects a basically
journalism have received recent treatments correct view of the nature of human moral
by Volkman (2010); Steiner and Okrusch psychology and moral development, allow-
(2006); Murphy (1999); Wyatt (2008); and ing educators to trade less in deduction from
Adams, Craft, and Cohen (2004). general intellectual principles and more in the
The turn toward practice-centered virtues cultivation of sensitivity to the particularities
can also be found within engineering ethics in of experience.
the essays of Crawford-Brown (1997), Harris While reflex psychological and character
(2008), and Frey (2010). Viewing engineering education approaches focus mainly on be-
practices as inseparable from the personal havior shaping or training, the ethics of care
life of the engineer, Crawford-Brown argues concentrates on emotional development, and
that the cultivation of virtue simultaneously liberal education and cognitive developmental
enriches the life of the engineer and improves approaches dwell primarily on the rational
the norms governing engineering practices intellectual aspects of moral understanding,
by helping engineers to negotiate potentially virtue ethics regards moral development as a
competing loyalties to clients, colleagues, matter of crucial interplay between all these
family, self, and the public. The virtuous dimensions of human being and [attempts to]
engineer is aware of the wider value-context give a coherent account of this interplay. (Carr
and Steutel 1999, p. 252)
within which the norms of engineering oper-
ate and is motivated to sustain an engineering The importance of exemplars and narratives
culture that deeply values six kinds of virtue: for the formation of personal and cultural
APQ 49_3 text.indd 187 5/14/12 9:29 AM

moral identity is accorded strong importance now evident that important new insights about
by several contributors to the collection, character and character education will only be
while others draw special attention to the possible when there is sustained exploration at
interdependencies between the intellectual the interface of these disciplines. (p. 3)
and moral virtues in Aristotelian thought and Grasp of the history of character education
their significance for moral education (see is important, too. Reflection on strategies
esp. Curren 1999; see also Chappell 2006). that failed in the past may reveal “how not
Empirical work in developmental psychology to educate character” (Kupperman 2005, p.
is of course a crucial consideration, and virtue 201); but the collection also takes positive
ethicists have themselves disagreed about steps to uncover the art and science of ef-
how to understand some of the connections fective character education. This includes
between psychology and moral education, inattention to the role of self-identity in the
cluding the relation between virtue, emotion, formation of character and how it helps us
and attention (see also Brady 2010, Goldie to understand moral motivation, commit-
2004, and Sherman 2010). ment, and self-worth. Kupperman states that
Another excellent collection dealing with “what is required is historical (and perhaps
many of these issues is Character Psychol- sociological and anthropological) analysis
ogy and Character Education (2005), ed- to remind us of how situated is our notion
ited by Lapsley and Power. This collection of character . . . and its relational functions”
primarily takes up the question of character (p. 337; see, in addition, McKinnon’s essays,
education in schools, families/parenting, also in Lapsley and Power 2005). Additional
and sports (see also McDougall 2007 and education-related research includes a focus
Austin 2009). The selections by thirteen on multicultural understanding (Katayama
contributing authors aptly demonstrate 2003), military virtues (Sandin 2007, Sher-
the editors’ abiding theme of the need for man 2007), and leadership virtues (Wilson
a multidisciplinary and interdisciplinary 2008).
focus on moral character, one that explores Welchman, herself the editor of The Prac-
several different psychological literatures for tice of Virtue (2006), argues in “Virtue Ethics
insights about moral character. This volume and Human Development: A Pragmatist Ap-
includes perspectives from a number of dis- proach” that contemporary virtue theorists to
ciplines, including philosophy, personality their own detriment often focus only on the
and development research, and educational same group of agents that principle-based
theory. According to the editors, theories do. There is “near universal tendency
we are clearly at a point where important work to treat moral agents as if they spring into be-
in moral psychology and ethical theory is ing as full adults. . . . By contrast, dispositions
reaching a common juncture. Indeed, increased closely associated with periods of depen-
attention devoted to moral selfhood, character, dency . . . however serviceable they may be to
and identity is the result of movement from ourselves or others, are either ignored outright
two directions. It results from the desire both to or grudgingly allowed an inferior status”
expand the explanatory reach of moral psychol- (Lapsley and Power 2005, p. 142). In order
ogy beyond structures-of-justice reasoning and to bring philosophy of education into closer
to ground ethical theory in a defensible account
proximity with our best theories of moral
of moral psychology. Both trends, then, from
development and aging, as well as to allay
within moral psychology and philosophical
ethics, point toward greater interest in virtues, this methodological bias in favor of only the
character, and moral identity. Moreover, it is most independent and autonomous of moral
APQ 49_3 text.indd 188 5/14/12 9:29 AM

reasoners, Welchman focuses on virtues that Applied virtue ethics has made an espe-
are properly included among what MacIntyre cially large impact in books and journals
(1999) termed “the virtues of acknowledged on environmental philosophy. Many of the
dependence” (p. 119). Practical concerns authors that Holly (2006) and Hull (2005)
related to the earliest and latest phases of identify in their review articles as founders
dependent life frequently overlap with some of this branch—EVE (environmental virtue
of the central concerns of bioethicists (e.g., ethics)—of applied virtue ethics have also
abortion, health care, senescence, death, and contributed newly commissioned essays in
dying). These intersections generate oppor- either Sandler and Cafaro’s Environmental
tunities yet to be fully seized upon for build- Virtue Ethics (2005) or in the recent spe-
ing dialogue between educational theorists cial edition of Journal of Agricultural and
and practitioners, social and developmental Environmental Ethics (Cafaro 2010). These
psychologists, and bioethicists. notable proponents of EVE include Frasz
(2005) and Wensveen (2000, 2005), Hill
IV. Bioethics (2005), Shaw (2005), Welchman (2008),
and Environmental Philosophy Westra (2005), and Welchman (2008), along
Walker and Ivanhoe rightly note in the with Cafaro and Sandler themselves. New
introduction to Working Virtue that “an un- voices such as Kawall (2010), Throop (2011),
derstanding of bioethics as incorporating the Treanor (2010), and numerous others are also
ethics of how we treat non-human animals contributing to the conversation over EVE.
as well as aspects of environmental ethics is In the introduction to Sandler and Cafaro’s
much closer to the original meaning of the Environmental Virtue Ethics, Sandler (2005)
term . . . than is the narrow focus on medical writes that “once the need for environmental
ethics that is currently fashionable” (2007, p. virtue ethics is recognized two questions im-
24). A broader conception of bioethics may mediately present themselves. First, what are
increase and deepen conversations among the attitudes and dispositions that constitute
virtue ethicists across the subfields of medi- environmental virtue? Second, what is the
cal ethics, environmental ethics, and animal proper role of an ethic of character in an
ethics—as some utilitarian and Kantian moral environmental ethic? These two issues . . .
philosophers have done (e.g., Singer and Re- are central to environmental virtue ethics and
gan). For applied virtue-ethical approaches to largely orient the philosophical work that ap-
our relationship with nonhuman animals, see pears in this collection” (p. 3).
Hursthouse 2006a and Bryant 2009. Virtue The two questions are distinct but also
ethics has recently been applied to further closely connected in Sandler’s thought. He
bioethical topics, including pragmatism and outlines four distinct strategies sometimes
virtue ethics in clinical research (Goldberg used to identify environmental virtue: “exten-
2008); genetic modification of crops (Sandler sionism, considerations of benefit to the agent,
2005, Farrelly 2007b), Confucian bioethics considerations of human excellence, and
(Fan 2006); euthanasia (Zyl 2004); abortion the study of role models” (pp. 5–6). Sandler
(Rovie 2002); hunting (Lovering 2006); views these strategies as delineating a range
divergent views of genetic selection and of complementary roles, from instrumental to
enhancement (Farrelly 2007b, Oakley 2009, foundational, that environmental virtue might
Saenz 2010); and post- or transhuman chal- play within a complete environmental ethic.
lenges to virtue ethics’ emphasis on “human” Proponents of environmental virtue eth-
flourishing (Cherry 2009). ics develop the concept of eco-citizenship,
APQ 49_3 text.indd 189 5/14/12 9:29 AM

arguing that environmental virtues such Virtue, and the Good Life (2007), are notable
as gratitude, respect, solidarity, and caring examples. In his introduction to Sex and Eth-
potentially play an important role in our ics, Halwani tries to show that a more open
responses to both local and global environ- discussion of sexuality may challenge the
mental problems. EVE’s advocates uniformly conservative outlook found in some virtue ethi-
agree that moral development and education cists’ treatments of particular virtues (2007,
are crucial for a sustainable future (Cafaro p. 3). In his own books and articles, Halwani
2010, p. 4) and that character education plays argues that a virtue ethics may allow for certain
an important role here. A good deal of work sexual lifestyles that are often deemed wrong
has also gone into developing the connection by traditional mores: for example, promiscuity,
between human flourishing and that of other open relationships, and even sex work. Nev-
animals and into the rationale for preserv- ertheless, a number of sexually conservative
ing wild nature (Jamal 2004). While these perspectives are also represented.
connections may well be fruitful, several It is hard to imagine how our erotic and
critics of EVE share Ralston’s concern that sexual nature could be quarantined or ex-
a eudaimonic virtue ethics, as Sandler puts orcized from the workplace, school, civil
it, could be “dangerous to the extent that society, or more private association. Thus, it
its focus on human flourishing distracts us is not surprising to find that dialogue across
from the intrinsic value of natural entities subfields of applied ethics is very lively in the
that makes environmental virtue possible” philosophy of love and sex, which embraces
(Sandler and Cafaro 2005, p. 8; see also topics in education, bioethics, political and
Hursthouse 2006b, Walker and Ivanhoe 2007, legal ethics, and feminist ethics.
and Wenz 2005). In recognition of this chal-
lenge, some defenders of strong, agent-based VI. Virtue Jurisprudence
accounts of ethics, such as Sandler, take EVE Aretaic approaches have made inroads in
in a nonanthropocentric direction, sometimes theories of law and jurisprudence, with the
looking to virtue theorists other than Aristotle approach called “virtue jurisprudence” (here-
in order to redefine virtue with reference both after VJ). In the introduction to their collec-
to the agent’s and the patient’s good (compare tion of that title, Farrelly and Solum (2007)
Zyl 2002). Others, like Walker and Ivanhoe present VJ as an alternative to both realist
(2007), stay closer to traditional eudaimonism and neoformalist theories of law. “In moral
but argue for similarities between human theory,” they say, “virtue ethics offers a third
and nonhuman animal flourishing that make way—an alternative to the deontological and
nonhuman animal flourishing normative for consequentialist approaches that dominated
humans. modern moral theory until very recently.
What would happen if we transplanted virtue
V. Philosophy of Love and Sex ethics into normative legal theory?” ( p. 1; see
One area in which virtue theorists have high- also Koller 2007).
lighted the study of certain animal aspects of The virtue ethics tradition is cited by the
specifically human nature is the philosophy of authors as offering insights into the legal
love and sex. Halwani’s two authored books, profession, criminal liability, judging, and is-
Philosophy of Love, Sex, and Marriage (2010) sues such as the legitimacy of judicial review.
and Virtuous Liaisons: Care, Love, Sex, and VJ offers distinctive answers to certain basic
Virtue Ethics (2003), as well as his edited col- questions of law: What is the aim of law?
lection Sex and Ethics: Essays on Sexuality, How are we to understand the relationship
APQ 49_3 text.indd 190 5/14/12 9:29 AM

between law and morality? How can legal to this volume turns attention to the profes-
institutions do their job of resolving disputes? sional roles of lawyers and to the difficult
The aretaic turn, according to the authors, moral dilemmas those roles can present. She
moves toward the reintegration of legal theory applies virtue ethics to the Lake Pleasant
and practice and away from efforts to discon- Bodies Case, in which the defense lawyers
nect the academy from the bench and bar knew the location of the victims’ bodies but
(by Posner, for example). VJ does not accept were bound by confidentiality not to reveal
that the central function of law is to prevent the information to the grieving family that
actions that harm others or that the purpose sought closure. Her treatment of how we
of the law is to protect property. Rather, ac- would expect the virtuous lawyer to act and
cording to Farrelly and Solum (2007), VJ “is feel in situations colored by special profes-
naturally inclined to the view that the law sional responsibilities is seen as applying also
should enable and sustain the material and to a much broader set of professions.
social conditions that would enable each and
every individual to achieve the highest level VII. Civic Virtues
of human function that is consistent with a and Deliberative Democracy
similar level of functioning for all” (p. 2). At In his book Justice, Democracy, and Rea-
its strongest, VJ challenges preference-based sonable Agreement (2007a), Farrelly further
and rights-based normative legal theories applies virtue theory to a range of issues in
that identify welfare efficiency, autonomy, or political philosophy, including constitutional
equality as the fundamental concepts of legal design, economic incentives, free speech,
philosophy, urging that the central notions of and reasonable pluralism (see also Wiggins
legal theory should be virtue, excellence, and 2004). A somewhat overlapping collection,
the promotion of human flourishing. Aristotle’s Politics Today, edited by Goodman
Legal scholars disagree on the criteria for and Talisse (2007), asks how Aristotle’s moral
good legal decision. Furthermore, the role and political insights might bear on pressing
political ideology plays in the appointment problems in contemporary liberal politics.
of judges makes it likely that people will dis- One question that several contributors address
agree about which judges are excellent. None- is whether contemporary liberal theorists
theless, Farrelly and Solum (2007) attempt to have something to gain from setting aside
identify and articulate a set of uncontested Rawlsian neutralism and embracing substan-
judicial virtues over which there is likely to be tive moral discourse (as Solum thinks). The
widespread consensus: “By ‘virtue,’ we mean study of deliberative virtues and the condi-
a dispositional quality of mind or will that tions for their vitality has been lively, raising
is constitutive of human excellence, and the concerns that intersect with epistemology and
‘judicial virtues’ include both the human vir- ethics. Talisse focuses on some specific epis-
tues that are relevant to judging and particular temic virtues that deliberation requires, while
virtues that are associated with the social role Goodman highlights the role that phronesis
of judge” (p. 7). Solum has argued that VJ plays in law and politics. Contributors to this
provides the best contemporary expression of collection also discuss other-regarding moral
the natural law thesis that there is an essential virtues such as generosity, friendship, justice,
connection between law and justice; his own and, drawing from the work of Confucius,
contributions to Virtue Jurisprudence suggest filial piety (see also Ivanhoe’s “Filial Piety
that practical wisdom and justice are the key as Virtue” in Walker and Ivanhoe, Working
judicial virtues. Hursthouse’s contribution Virtue [2007]).
APQ 49_3 text.indd 191 5/14/12 9:29 AM

From work on civic virtues (Audi 1998) ently public or accessible” (Weithman 2010,
to more recent discussions of deliberative p. 65). Weithman draws attention to the dan-
virtues and how they illuminate moral dis- ger that members of a privileged class may
agreement, several authors have considered seek to normatively elevate their discourse
whether and to what extent democracy by dubbing their reasons “public.” As an al-
requires not only institutions like consti- ternative to strong deliberativism, Weithman
tutionalism and the rule of law, but also a offers that “mutual translation” of arguments
more active and educated citizenry (see esp. across discursive traditions “can go some way
Gutmann and Thompson 1996, Misak 2000, toward mitigating the problems that are said
Misak 2009, and Aiken and Clanton 2010). to be posed by citizens’ ineliminable reliance
The editors of Deliberative Democracy on their conceptions of the good in political
in Practice (Kahane, Weinstock, Leydet, argument” (p. 68). To be effective, this al-
and Williams, 2010) bring together eleven ternative requires that schools try to educate
essays devoted to exploring the prospects students about the rhetorical conventions and
and obstacles that lie ahead for deliberative the historical and cultural backgrounds that
democratic theory. Deliberative democracy inform different discursive traditions.
aims to provide more than just a norma- Weithman’s essay foreshadows Ivison,
tive philosophical perspective on political Coulthard, and Valadez’s shared concern
engagement; it also prescribes the active re- that the “virtues” of liberal democrats may
design of political processes and institutions privilege the political interest of certain par-
in order to increase and to improve citizen ticipants while marginalizing indigenous,
participation and to help bring about moral colonized citizens. Each of these authors rec-
or political agreement among citizens. ognizes the demands of deliberative democ-
The political concerns of deliberative racy in relation to matters of global justice
democracy clearly overlap with interests in concerns, rather than merely local or national
virtue education (see section III, above). One ones. Relatedly, Tessman’s Burdened Virtue:
central question addressed by the contributors Virtue Ethics for Liberatory Struggles (2005)
to Deliberative Democracy in Practice con- is a reflection upon oppression and liberation
cerns “which traits of character . . . the ideal and the uneasy role played by “burdened vir-
deliberator possess[es], and what should the tues”—that is, traits of character that allow
role of the state, via the institution of public one to endure or to resist political oppression
schools, be in inculcating them?” (Kahane et as well as other, less overt kinds of oppres-
al. 2010, p. 7). Endorsing a “weak” version sion (see also Friedman 2008). Works such as
of Rawls’s theory of legitimacy, Brighouse these can also raise the issue of the distinction
hypothesizes that reasonable religious and between a virtue ethics and a “virtue politics”
nonreligious persons are more likely to and the need to differentiate between them.
endorse together the “constitutional es-
sentials” of a democratic polity if the state VIII. Prospects and Challenges
seeks “to collaborate with religious parents for Applied Virtue Ethics
in the provision and regulation of schooling” This essay concludes with some general
(Brighouse, 2010, p. 52). Weithman questions observations about the state of applied virtue
“strong deliberativism’s” claim that “public ethics (and applied ethics more generally),
deliberation can serve its legitimating func- along with some suggestions for future re-
tion only if participants in public deliberation search. As was indicated in the introduction
are prepared to offer one another, and are to this essay, virtue ethics challenges certain
responsive to, a class of reasons that is inher- assumptions that have dominated applied eth-
APQ 49_3 text.indd 192 5/14/12 9:29 AM

ics, and these challenges, if taken seriously, agents’ actions, attitudes, and deliberations.
affect our very conception of the field of What distinguishes virtue-ethical treatment
applied ethics and of the methods appropri- of agents has to do with the place that these
ate to it. More specifically, virtue ethics first features occupy in the description and evalua-
challenges the idea that what is applied in tion of moral conduct and moral life, although
applied ethics must always be general rules again, virtue ethicists are not the only moral
or principles. Second, virtue ethics seeks to philosophers who consider these features of
expand the range of subject matters to which agents to be important.
moral norms or theories are appropriately Second, the focus on features of agents
applied. Third, it seeks to refashion our un- expands the purview of applied ethics to
derstanding of what it can mean for ethical include problems directly related to agents
norms or theories to be applied. themselves. Traditionally, applied ethics has
The foregoing review of literature clearly sought to propose solutions to problems that
demonstrates that work in applied virtue exist within real or imagined states of affairs
ethics can be motivated by a number of that exist, as it were, “outside” of the moral
theoretical paths that reveal no unanimity, agent. If the subject matters of applied ethics
including over questions about whether vir- are thought to be exclusively these sorts of
tue ethics offers (1) an independent, “third” problems, then it is understandable that the
moral theory, (2) a theory of right action, norms appropriate for responding to practical
(3) a theory of moral decision making, and moral concerns will always be action-guiding
(4) whether virtues qua character traits are norms aimed at providing answers to the
properly conceived as practice- or role-based question, “What must I do?” Contributors
traits or as more global traits. Nevertheless, to applied virtue ethics draw attention to
recent work in applied virtue ethics is dis- problems concerning characteristics of agents
tinctive in a number of important ways. In themselves (i.e., their attitudes, motives,
what follows, three distinctive features of feelings, emotions, deliberative strategies,
applied virtue ethics will be discussed. As conceptions of the good, etc.), each of which
we shall see, one of these features supports may affect moral development, reasoning,
a pluralist perspective, which weakens the and conduct. But even when applied virtue
bite of theoretical disputes that otherwise ethicists stress the importance of the question,
threaten to sidetrack applied ethicists’ shared “What kind of person should I be?” they need
pursuit of practicable solutions to concrete not be arguing for the universal explanatory or
moral concerns. Finally, this essay will con- evaluative priority of character-based norms
clude with a reasoned recommendation for over act-based norms.
increased dialogue across subfields within Third, work in applied virtue ethics is
applied ethics. distinctive on account of its diachronic or
First, contributions to applied virtue eth- longitudinal approach to the study of moral
ics involve a distinctive focus on features agents and moral agency. Whether or not
of the moral agent herself. These include virtue ethics is appropriately understood as
psychological, characterological, norma- offering a theory of right action, it does seek
tive, and social features that inform agents’ to understand particular actions and particular
conceptions of the good, as well as agents’ decisions within the context of the agent’s life
moral deliberations. These features (whether as a whole. It is this interest in the diachronic
they are conceived globally or locally) may that most clearly distinguishes virtue ethical
serve as independent sources of value, and treatments of moral agents from act-based
thus as criteria for the evaluation of moral perspectives on moral agency. This feature of
APQ 49_3 text.indd 193 5/14/12 9:29 AM

virtue ethics most often leads to its associa- way, a diachronic or longitudinal account of
tion with eudaimonism. But that connection moral agency and moral judgment undercuts
seems to be less essential to a viable virtue the theoretical turf battles that drive a great
ethics than is a strong connection with moral deal of contemporary discourse in ethics.
development and moral education, for even A more pluralistic approach to applied eth-
those virtue ethicists who oppose happiness- ics, one that recognizes multiple sources of
based conceptions of virtue (e.g., Zagzebski moral value, could very well have a practi-
1996) insist upon the central role of education cal advantage. When the different emphases
for the inculcation and exercise of the virtues. of “ethics to do” and “ethics to be” harden
Although discussions of moral growth and into dichotomies, the ability of virtue ethics
development are not entirely absent from act- to provide action guidance becomes much
based approaches, virtue ethics has clearly more problematic. Emphasizing theoretical
heightened attention to a longitudinal view of turf battles in the moral education of nurses,
moral agents and to the diachronic as opposed social workers, and other professionals may
to merely synchronic considerations in moral distract and frustrate students, leading them
judgment. As Kupperman (2009) points out, to lose sight of real-world problems and
in numerous versions of virtue ethics (from perhaps encouraging a crude relativism or
Aristotle and Confucius to contemporary skepticism about ethics. In the context of
authors) “there is great attention to a longi- moral education, aretaic norms can work
tudinal view of virtues, with emphasis both in conjunction with deontological and con-
on how people can come to be virtuous and sequentialist norms to inform and facilitate
on the rewards of a life that centers on be- responsible deliberation and action. A virtues
ing virtuous. This longitudinal view sharply aspect to education may help to inculcate
separates virtue ethics from much in contem- appropriate responsiveness to salient moral
porary philosophical ethics, especially the principles or rules and appropriate sensitivity
emphasis on dramatic cases (e.g., the trolley to significant conditions antecedent to and
problem) that lend themselves to atomistic consequent of particular actions. In effect, an
consideration” (p. 250). Surely we want to overemphasis on theoretical differences and
retain norms and procedures that allow us to rivalries undermines a natural (and perhaps
evaluate human conduct synchronically at well-founded) inclination to view different
the level of acts. But virtue ethics reminds normative ethical theories as offering com-
us that the aims of moral inquiry are not all plementary tools and resources, all of which
reducible to the distribution of praise or blame are needed to address serious, shared social
for particular actions. problems. By transcending the reductionistic
One important theme for virtue ethicists to spirit so evident in moral theory today, ap-
consider is how a longitudinal view of moral plied virtue ethicists—indeed, all applied
agency and moral judgment makes important ethicists—will better succeed in fulfilling a
contributions to the field of applied ethics, socially useful role.7
and again how these longitudinal consider- If the subject matters of applied ethics are
ations potentially change our understanding to include problems concerning the growth
of how that field should be approached. This and development of moral agents over the
third distinctive feature of virtue ethics can course of their lives as a whole (including
easily be taken as recognizing the mutually infancy, childhood, and elder years), then a
supportive roles of rules, consequences, and diachronic approach will help us to appraise
virtues in the assessment of agents’ moral and evaluate the moral trajectory of agents.
growth and development over time. In this Kamtekar (2004) reminds us that an over-
APQ 49_3 text.indd 194 5/14/12 9:29 AM

emphasis on particular actions and behaviors moral agents. It may not be so much that the
can overlook “information about subjects’ importance of particular political or public
feelings about their actions,” which is “in- virtues is heightened by the severity of the
formation that might further the understand- challenges of one’s historical context, as
ing of why people act as they do” (p. 474). that the importance of imaginatively rede-
Moreover, Kametar (2010) says, practically fining and then internalizing the virtues is
wise agents use this information to “improve” needed to successfully deal with present and
upon their capacity to order goods and to act foreseeable problems. By prescribing, for
rightly (p. 157). Agents engaged in serious instance, new conceptions of eco-citizenship
moral reflection can make sound, practical or a new set of public environmental virtues,
use of the “thick” affective and character- environmental virtue ethics clarifies our
ological concepts that virtue ethics provides. environmental choices and responsibilities.
This is true even if human character and the This reminds us that there is something ir-
virtues are more fragmented and less unified, reducibly pragmatic in virtue ethics’ focus
more “modular” or local and less global, than on actual practices. The practices with which
some versions of virtue ethics take them to one starts may be approached critically
be. Virtue ethicists are right to challenge the by applied researchers, and the goods and
assumption that there is little more to moral virtues required by the professions may be
motivation and decision than a procedure persuasively redefined in light of evolving
of applying a criterion of rightness to the problems of practice.
choices of action with which one is faced. This last observation about the need to
What kind of person (or what kind of doctor, rethink virtue and character, with regard to
patient, parent, etc.) one wants to become is ongoing changes in social and environmental
an important consideration in one’s moral problem contexts, leads into a final recom-
decisions. And becoming a good person, mendation for future work in applied virtue
professional, and so on does not require that ethics. The foregoing review of literature re-
one know in advance all that it is required to peatedly draws attention to the ways in which
be what one wants to become. the problems of applied virtue ethics intersect
The long-acknowledged importance of with one another in important ways. Yet
context in virtue ethics may also illustrate there remains a conspicuous lack of explicit
how changing social problems fundamentally dialogue across the various subfields within
change the kinds of persons we need to be in applied virtue ethics. Clearly, several authors
order to flourish. While this may affect per- have participated in discussions within a
sonal virtue, it perhaps has a deeper impact number of different areas of applied ethics;
on group and public virtues (see MacIntyre but even if an author’s distinct contributions
1984, pp. 181–203). A good illustration may are grounded in a single theoretical perspec-
be the call for “new” environmental virtues tive, this general theoretical unity cannot
in light of “environmental changes—the replace the sort of practical cooperation be-
realities of global warming for example . . . tween subfields from which applied ethicists,
[which] can bear upon the environmental as well as the persons and communities they
virtues, having effects not only on the condi- aim to serve, would more greatly benefit. The
tions of their application but also altering the force and appeal of applied virtue ethics will
concepts themselves” (Thompson 2010, p. strengthen if increased attention is given to
56; see also Hursthouse 2006b and Treanor the shared perspectives, themes, and practi-
2010). Changes in social and environmental cal counsels that virtue ethics provide across
contexts challenge us to grow and develop as all areas of practical moral concern. Some
APQ 49_3 text.indd 195 5/14/12 9:29 AM

authors are beginning to recognize the need necting practice-localized virtues with more
for greater exchange across subfields. For general traits of character.
example, in their lengthy and rich introduc- In A Theory of Virtue, Adams (2006) ar-
tion to Working Virtue, Walker and Ivanhoe gues that virtues local to a type of situation
(2007) seek to provide “a general sense of (what he calls “modular virtues”) can serve
. . . how [the essays] relate to one another . . . as a moral starting point from which agents
how they can be located in the bigger picture may develop virtues that exhibit “sufficient
of virtue ethics as practical ethics,” and how generality and consistency across situations
the essays “relate to some of the broader to count as traits of character” (p. 120). Rad-
themes in contemporary virtue ethics” (p. den shares with Adams what we might call an
36). Still, the essays collected in this wide- “expansionist” account of virtue acquisition.
ranging volume are not themselves explicitly In “Virtue Ethics as Professional Ethics: The
in conversation with one another. Perhaps Case of Psychiatry” (2007), Radden argues
the demands of professional specialization that there is a practical need to identify and
partly explain why these intersections remain sanction, in addition to general moral vir-
largely underexplored. tues, traits of character “which, outside the
Another plausible explanation has to do context of professional practice, are morally
with the place of social roles and practices neutral—neither virtues nor vices—or are
within contemporary applied ethical dis- at most prudential and intellectual virtues,
course. As indicated in the introduction to this rather than moral ones” (p. 114). These traits
essay, applied virtue ethics has been strongly deserve more than an honorific status as vir-
influenced by practitioner-academics, whose tues because, Radden believes, they advance
focus on problems of practice often seeks to the goods internal to professional practices
identify virtues that are “local” to particular and because they “might be expected to
professional roles and practices. Role eth- spread, eventually affecting the rest of [the
ics, and what Radden (2007) calls “role- practitioner’s] non-professional life” (p. 130).
constituted virtues,” have received a good Thus, role-constituted virtues can facilitate
deal of attention from contemporary applied the development of general virtues, thereby
virtue ethicists. Philosophers and social contributing to the agent’s self-unity and
psychologists who sympathize with recent moral integrity. Whereas Radden’s “weak”
“situationist” critiques of virtue ethics might role ethics presumes a hierarchy of moral ob-
find role-constituted virtues more plausible ligations and permissions, according to which
than the “global” conceptions of virtue that the more exacting demands of professional
certain empirical studies are thought to render roles cannot override more general moral
untenable (see Doris 2002). Yet even role- obligations, Swanton (2007) advocates “a
constituted virtues might not be local enough genuine pluralism of ends” (p. 208) that does
to fully satisfy those who argue that traits of not subordinate role virtues to more general,
character do not even manifest consistency traditional virtues. Following Slote (2001),
across different concrete situations within Swanton claims that the value of role-defined
practical domains. Most moral philosophers virtues does not necessarily derive from any
now agree that the “situationist challenge” contribution that those virtues might make
has brought into focus some serious obstacles to the good of society at large or to the good
to the acquisition of virtue. But contemporary of the person possessing those virtues. Rec-
virtue ethicists do not see these obstacles ognizing that the goods internal to particular
as insurmountable; and even advocates of practices may conflict with the human good
role-constituted virtues show interest in con- in general, Swanton contends that both sorts
APQ 49_3 text.indd 196 5/14/12 9:29 AM

of good come into focus together: role virtues to demand the concerted efforts of persons
offer greater action-guidance than do thinner, occupying a number of roles and practices.
“prototype virtues” (e.g., justice, honesty, For these reasons, applied virtue ethics could
etc.), while the prototypes help to temper the become stronger by focusing more broadly on
pursuit of professional goals. virtues that facilitate the resolution of shared
Adams, Radden, and Swanton each make practical problems instead of on the narrower
some use of the bottom-up, practice-based problems of practice.
approach to virtue acquisition that practi- In practice and in effect, our decisions and
tioner-academics advocate. But a bottom-up actions inevitably have implications that are
approach to applied ethics need not locate not confined to the purview of any one role
roles or practices at the bottom. Indeed, by or practice. This fact alone should provide
focusing upon problems rather than practices sufficient impetus for developing more fully
or roles, applied virtue ethicists could take the kinds of deliberative virtues, processes,
the lead in fostering greater dialog across any and institutions that take these intersections
number of subfields within applied ethics. into account. Moreover, we can expect to find
Bottom-up approaches that identify roles and more circumspect and more effective solu-
practices as the basis for virtue acquisition tions to concrete moral problems by drawing
and action-guidance face the dilemma of simultaneously upon the resources provided
either having to reestablish connections be- by theorists and practitioner-academics in
tween various roles and practices (as Adams, both education and bioethics; in EVE and
Radden, and Swanton try to do) or having to virtue jurisprudence and business ethics; in
accept a compartmentalized conception of education and legal and political ethics and
virtue and human functioning. But a problem- feminist ethics and the ethics of love and
based approach avoids this difficulty, while sex. It is unwise to pretend to be able, in
also respecting the practitioner goals and advance, to define the problems that lie at
norms that top-down approaches disregard. these intersections. It is a task to be under-
The various fields of practical moral inquiry taken through processes of cooperative dialog
are themselves organizations of the conclu- among practitioners and theorists working
sions of past inquiries. The goals, methods, upon shared (or overlapping) concerns across
and norms that govern these various fields subfields of applied ethics. Some applied
may facilitate intelligent problem solving by virtue ethicists have begun to turn in this
providing funds of settled knowledge that direction, prescribing public virtues, which
serve as resources for addressing the problems denote actions, characteristics, or dispositions
of applied ethics. These resources can serve as that benefit the community rather than the
preconditions for applied ethical inquiry only individual (Holland 2010b). Treanor (2010)
insofar as they actually help us to define and is probably right to point out that “while
resolve concrete moral concerns. But different both personal and public virtues ultimately
roles and practices do not provide criteria for contribute to one’s flourishing, virtue ethics,
determining what our problems are or how including environmental virtue ethics, has
best to resolve them. Rather, the value of the tended to focus on the former to the neglect
norms that govern particular practices should of the latter” (p. 26). And Lapsley and Power
be assessed in terms of how well those norms (2005) wisely urge that “more emphasis is
help us to resolve the moral difficulties that required on notions of community, on civic
we currently face or which we can reasonably virtues proper to democratic citizenship, and
expect to face. Furthermore, lasting solutions on the interpersonal basis of character and
to our most pressing moral problems are likely its relational functions” (p. 337; see also the
APQ 49_3 text.indd 197 5/14/12 9:29 AM

exemplary work of Aiken and Clanton [2010] individualism so strong in twentieth-century

on group deliberative virtues). Public virtues analytic philosophy.
always contribute essentially to eudaimonia,
and applied virtue ethicists will do well to Radford University
try to balance individualistic notions of trait Virginia Polytechnic Institute
possession with a more social approach, dis- and State University (Virginia Tech)
tancing virtue ethics from the methodological
Notes
1. While many applied ethicists specialize in one subfield or another, there are others who demonstrate
wide-ranging interest, participating in conversations across theoretical and applied virtue ethics, as
well as within a number of different areas in applied philosophy. Credit for the development of ap-
plied virtue ethics (a lion’s share of it going “down under” to research centers in applied philosophy
and public ethics in New Zealand and Australia) goes especially to authors like Hursthouse; Swanton;
Zyl; van Hooft; Solomon; and, of course, to those editors—Carr, Cocking, Oakley, Steutel, and others
already mentioned—whose collections serve as first-of-their-kind models of research and resources
for work in applied virtue ethics.
2. As Rhodes earlier put it, “a virtue-based ethics seems particularly appropriate to professions, be-
cause the ethical issues often focus on the nature of the relationships and our responsibility in those
relationships—to the client, other colleagues, our supervisors, the agency itself. What sort of person is
a ‘professional’ social worker to be? What is human excellence in that context?” (Banks and Gallagher
2008, p. 41).
3. The distinction between practice-relative and practice-neutral values is an adaptation of Quinn’s
distinction between “agent-relative” and “agent-neutral” values (Quinn 2007).
4. As McDowell (1998), too, contends, undertaking a particular behavior “as constituent means to
eudaimonia . . . [specifies] a distinctive sort of reason an agent can have for behaving as he does. . . . It
is the sort of reason for which someone acts when he does what he does because that seems to him to
be what a human being, circumstanced as he is, should do” (p. 10).
5. Although the more deontological language of some codes of professional conduct can make them
more onerous among professionals themselves, these may be appropriate in some circumstances of
public trust and may also occasionally give rise to recognition of “new” professional virtues.
6. “Professions have identifiable and defining ends, that is, each serves certain universal human
needs . . . in each of these professions, the end or telos is the welfare of a human being and particular
existential state, in need of a specific kind of help. . . . This is the meaning of the very first sentence of
Aristotle’s Nicomachean Ethics—’every art and every inquiry, and similarly every action and choice,
is thought to aim at some good, and for this reason the good has really been declared that at which all
things aim’” (Pellegrino 2007, p. 64).
7. “Critics of virtue ethics have argued that its focus on character rather than action, as well as its
rejection of universal rules of right action renders virtue ethics unable to shed much light on the ques-
tion of what ought and ought not to be done in specific situations” (Zyl 2002, abstract).
APQ 49_3 text.indd 198 5/14/12 9:29 AM

References
Adam, S., S. Craft, and E. Cohen. 2004. “Three Essays on Journalism and Virtue,” Journal of Mass
Media Ethics, vol. 19, nos. 3–4, pp. 247–275.
Adams, P. 2009. “Ethics with Character: Virtues and the Ethical Social Worker,” Journal of Sociology
& Social Welfare, vol. 36, no. 3, pp. 83–105.
Adams, R. M. 2006. A Theory of Virtue (Oxford: Oxford University Press).
Aiken, S., and J. C. Clanton. 2010. “Developing Group-Deliberative Virtues,” Journal of Applied Eth-
ics, vol. 27, no. 4, pp. 327–431.
Armstrong, A. 2007. Nursing Ethics: A Virtue-based Approach (New York: Palgrave).
Aristotle. 1996. The Politics and the Constitution of Athens, trans. J. Barnes, ed. S. Everson. Cambridge
Texts in the History of Political Thought (Cambridge: Cambridge University Press).
Athanassoulis, N., and A. Ross. 2009. “A Virtue Ethical Account of Making Decisions about Risk,”
Journal of Risk Research, vol. 16, no. 1, pp. 217–230.
Audi, R. 1998. “A Liberal Theory of Civic Virtue,” Social Philosophy and Policy, vol. 15, no. 1, pp.
149–170.
Austin, M. 2009. “Magnanimity, Athletic Excellence, and Performance-Enhancing Drugs,” Journal of
Applied Philosophy, vol. 26, no. 1, pp. 46–53.
Banks, S., and A. Gallagher. 2008. Ethics in Professional Life: Virtues for Health and Social Care (New
York: Palgrave Macmillan).
Barilan, Y. M. 2009. “Responsibility as a Meta-Virtue: Truth-Telling, Deliberation, and Wisdom in
Medical Professionalism,” Journal of Medical Ethics, vol. 35, no. 3, pp. 153–158.
Battaly, H., ed. 2010. Virtue and Vice, Moral and Epistemic (Oxford: Wiley-Blackwell).
Bertland, A. 2009. “Virtue Ethics in Business and the Capabilities Approach,” Journal of Business
Ethics, vol. 84, pp. 25–32.
Brady, M. 2010. “Virtue, Emotion, and Attention,” in Virtue and Vice, Moral and Epistemic, ed. H.
Battaly (Oxford: Wiley-Blackwell), pp. 115–132.
Brighouse, H. 2010. “Religious Belief, Religious Schooling, and the Demands of Reciprocity,” in
Deliberative Democracy in Practice, ed. D. Kahane et al. (Vancouver, Canada: University of British
Colombia Press).
Bryant, T. 2009. “Virtue Ethical Animal Law Reform.” Conference Papers—Law & Society (2009 An-
nual Meeting): Academic Search Complete, EBSCOhost (accessed May 25, 2010).
Cafaro, P., ed. 2010. “Introduction.” Special issue on Environmental Virtue Ethics, Journal of Agri-
cultural and Environmental Ethics, vol. 23, nos. 1–2, pp. 3–7. DOI: 10.1007/s10806–009–9204–3.
Carr, D., and J. W. Steutel, eds. 1999. Virtue Ethics and Moral Education (New York: Routledge).
Chappell, D., ed.. 2010. Values and Virtues: Aristotelianism in Contemporary Ethics (Oxford: Oxford
University Press).
Cherry, R. 2009. “The Posthumanist Challenge to a Partly Naturalized Virtue Ethics,” in The Normativity
of the Natural: Human Goods, Human Virtues, and Human Flourishing, ed. R. Cherry. Philosophical
Studies in Contemporary Culture, vol. 16 (New York: Springer), pp. 153–172.
Clark, C. 2006. “Moral Character in Social Work,” British Journal of Social Work, vol. 36, pp. 75–89.
Consoli, L. 2008. “The Intertwining of Ethics and Methodology in Science and Engineering: A Virtue-
Ethical Approach,” Interdisciplinary Science Reviews, vol. 33, no. 3, pp. 233–242.
Crawford-Brown, D. 1997. “Virtue as the Basis of Engineering Ethics,” Science and Engineering Eth-
ics, vol. 3, no. 4, pp. 481–489.
Curren, R. 1999. “Cultivating the Intellectual and Moral Virtues,” in Virtue Theory and Moral Educa-
tion, ed. D. Carr and J. Steutel (London: Routledge), pp. 67–81.
Dobson, J. 2007. “Applying Virtue Ethics to Business: The Agent-Based Approach,” Business and
Organization Ethics Network, at http://ejbo.jyu.fi/articles/0901_3.html
APQ 49_3 text.indd 199 5/14/12 9:29 AM

Doris, J. 2002. Lack of Character: Personality and Moral Behavior (Cambridge: Cambridge University
Press).
Fan, R. 2006. “Towards a Confucian Virtue Bioethics: Reframing Chinese Medical Ethics in a Market
Economy,” Theoretical Medicine and Bioethics, vol. 27, no. 6, pp. 675–682.
Farrelly, C. 2007a. Justice, Democracy, and Reasonable Agreement (New York: Palgrave Macmillan).
________. 2007b. “Virtue Ethics and Prenatal Genetic Enhancement,” Studies in Ethics, Law, and Tech-
nology, vol. 1, no. 1, pp. 1–13.
Farrelly, C., and L. Solum. 2007. “An Introduction to Aretaic Theories of Law,” in Virtue Jurisprudence,
ed. C. Farrelly and L. Solum (New York: Palgrave Macmillan).
Frasz, G. 2005. “Benevolence as an Environmental Virtue,” in Environmental Virtue Ethics, ed. R. Sandler
and P. Cafaro (Lanham, Md.: Rowman & Littlefield), pp. 121–134.
Friedman, M. 2008. “Virtues and Oppression: A Complicated Relationship,” Hypatia, vol. 23, pp.
189–196.
Frey, W. 2010. “Teaching Virtue: Pedagogical Implications of Moral Psychology,” Science and Engi-
neering Ethics, vol. 16, no. 3, pp. 611–628.
Goldberg, D. 2008. “Pragmatism and Virtue Ethics in Clinical Research,” American Journal of Bioeth-
ics, vol. 8, no. 4, pp. 43–44.
Goldie, P. 2004. On Personality (New York: Routledge).
Goodman, L. E., and R. B. Talisse, eds. 2007. Aristotle’s Politics Today (Albany: State University of
New York Press).
Gregg, S., and J. Stoner, eds. 2009. Profit, Prudence, and Virtue: Essays in Ethics, Business, and Man-
agement (Charlottesville, Va.: Imprint Academic).
Gutmann, A., and D. Thompson. 1996. Democracy and Disagreement (Cambridge, Mass.: The President
and Fellows of Harvard College).
Halwani, R. 2003. Virtuous Liaisons: Care, Love, Sex, and Virtue Ethics (Chicago: Open Court).
________. 2007. Sex and Ethics: Essays on Sexuality, Virtue, and the Good Life (New York: Palgrave
Macmillan).
________. 2010. Philosophy of Love, Sex, and Marriage: An Introduction (New York: Routledge).
Harris, C. 2008. “The Good Engineer: Giving Virtue Ethics Its Due in Engineering Ethics,” Science
and Engineering Ethics, vol. 14, no. 2, pp. 153–164.
Hill Jr., T. 2005. “Ideals of Human Excellence and Preserving Natural Environments,” in Environmen-
tal Virtue Ethics, ed. R. Sandler and P. Cafaro (Landham, Md.: Rowman & Littlefield), pp. 47–60.
Holland, S. 2010a. “Scepticism about the Virtue Ethics Approach to Nursing Ethics,” Nursing Philoso-
phy, vol. 11, no. 3, pp. 151–158.
________. 2010b. “The Virtue Ethics Approach to Bioethics,” Bioethics, at http://onlinelibrary.wiley
.com/doi/10.1111/j.1467–8519.2009.01758.x/full.
Holly, M. 2006. “Environmental Virtue Ethics, a Review of Some Current Work,” Journal of Agricultural
and Environmental Ethics, vol. 19, no. 4, pp. 391–424.
Horner, S. 2000. “For Debate: The Virtuous Public Health Physician,” Journal of Public Health Medi-
cine, vol. 22, no. 1, pp. 48–53.
Hull, R. 2005. “All About EVE: A Report on Environmental Virtue Ethics Today,” Ethics and the En-
vironment, vol. 10, no. 1, pp. 89–110.
Hursthouse, R. 1995. “Applying Virtue Ethics,” in Virtues and Reasons: Philippa Foot and Moral Theory,
ed. R. Hursthouse, G. Lawrence, and W. Quinn (Oxford: Oxford University Press).
________. 2006a. “Applying Virtue Ethics to Our Treatment of the Other Animals,” in The Practice of
Virtue, ed. J. Welchman (Indianapolis, Ind.: Hackett).
________. 2006b. “Environmental Virtue Ethics,” in Working Virtue: Virtue Ethics and Contemporary
Moral Problems, ed. P. J. Ivanhoe and R. L. Walker (Oxford: Oxford University Press).
________. 2008. “Two Ways of Doing the Right Thing,” in Virtue Jurisprudence, ed. C. Farrelly and L.
Solum (New York: Palgrave Macmillan).
APQ 49_3 text.indd 200 5/14/12 9:29 AM

Jamal, T. B. 2004. “Virtue Ethics and Sustainable Tourism Pedagogy: Phronesis, Principles and Prac-
tice,” Journal of Sustainable Tourism, vol. 12, no. 6, pp. 530–545.
Kahane, D., D. Weinstock, D. Leydet, and M Williams, eds. 2010. Deliberative Democracy in Practice
(Vancouver, Canada: University of British Columbia Press).
Kamtekar, R. 2004. “Situationism and Virtue Ethics on the Content of Our Character,” Ethics, vol.
114, pp. 458–491.
________. 2010. “Comments on Robert Adams’ A Theory of Virtue,” Philosophical Studies, vol. 148,
no. 1, pp. 147–158.
Katayama, K. 2003. “Is the Virtue Approach to Moral Education Viable in a Plural Society?” Journal
of Philosophy of Education, vol. 37, no. 2, pp. 325–338.
Kawall, J. 2006. “Environmental Virtue Ethics,” Environmental Ethics, vol. 28, no. 4, pp. 429–432.
________. 2010. “The Epistemic Demands of Environmental Virtue,” Journal of Agricultural and Envi-
ronmental Ethics, vol. 23, nos. 1–2, pp. 109–128.
Koller, P. 2007. “Law, Morality, and Virtue,” in Working Virtue, ed. P. J. Ivanhoe and R. L. Walker
(Oxford: Oxford University Press), pp. 191–128.
Kupperman, J. 2005. “How Not to Educate Character,” in Character Psychology and Character Edu-
cation, ed. D. Lapsley and F. Clark Power (Notre Dame, Ind.: University of Notre Dame Press), pp.
201–217.
________. 2009. “Virtue in Virtue Ethics,” Journal of Ethics, vol. 13, nos. 2–3, pp. 243–255.
Lapsley, D., and F. C. Power, eds. 2005. Character Psychology and Character Education (Notre Dame,
Ind.: University of Notre Dame Press).
Larkin G., K. Iserson, Z. Kassutto et al. 2009. “Virtue in Emergency Medicine,” Academic Emergency
Medicine, vol. 16, no. 1, pp. 51–55.
Louden, R. 1984. “On Some Vices of Virtue Ethics,” American Philosophical Quarterly, vol. 21, pp.
227–236.
Lovering, R. 2006. “The Virtues of Hunting: A Reply to Jensen,” Philosophy in the Contemporary
World, vol. 13, no. 1, pp. 68–76.
Lynch, T. D. 2004. “Virtue Ethics, Public Administration, and Telos,” Global Virtue Ethics Review,
vol. 5, no. 4, pp. 32–49.
MacIntyre, A. 1984. After Virtue: A Study in Moral Theory, 2nd ed. (West Bend, Ind.: University of
Notre Dame Press).
________. 1999. Dependent Rational Animals: Why Human Beings Need the Virtues (Chicago: Open
Court).
McBeath, G., and S. Webb. 2002. “Virtue Ethics and Social Work,” British Journal of Social Work,
vol. 32, pp. 1015–1036.
McDougall, R. 2007. “Parental Virtues,” Bioethics, vol. 21, no. 4, pp.181–190.
McDowell, J. 1998. Mind, Value, and Reality (Cambridge, Mass.: Harvard University Press).
McKinnon, C. 2005. “Character Possession and Human Flourishing,” in Character Psychology and
Character Education, ed. D. Lapsley and C. Power (Notre Dame, Ind.: University of Notre Dame
Press), pp. 33–66.
Misak, C. 2000. Truth, Politics, Morality: Pragmatism and Deliberation (London: Routledge).
________. 2009. “Truth and Democracy: Pragmatism and the Deliberative Virtues,” in Does Truth Mat-
ter? Democracy and Public Space, ed. R. Geenens and R. Tinnevelt (New York: Springer), pp. 29–39.
Moore, G. 2008. “Re-Imaging the Morality of Management: A Modern Virtue Ethics Approach,” Busi-
ness Ethics Quarterly, vol. 18, no. 4, pp. 483–511.
Murphy. P. 1999. “Character and Virtue Ethics in International Marketing: An Agenda for Manag-
ers, Researchers, and Educators,” Journal of Business Ethics, vol. 18, no.1, pp. 107–124. DOI:
10.1023/A:1006072413165.
Oakley, J. 2009. “A Virtue Ethics Approach,” in A Companion to Bioethics, ed. H. Huhse and P. Singer,
2nd ed. (Oxford: Blackwell).
APQ 49_3 text.indd 201 5/14/12 9:29 AM

________, and D. Cocking. 2001. Virtue Ethics and Professional Roles (Cambridge: Cambridge Uni-
versity Press).
________, and L. White. 2005. “Whistleblowing, Virtue, and Accountability in an Age of Precarious
Employment,” Just Policy, vol. 37, pp. 64–70.
Pellegrino, E. 2007. “Professing Medicine, Virtue Based Ethics, and the Retrieval of Professionalism,”
in Working Virtue, ed. P. J. Ivanhoe and R. L. Walker (Oxford: Oxford University Press), pp. 61–85.
Pullen-Sansfacon, A. 2010. “Virtue Ethics for Social Work: A New Pedagogy for Practical Reasoning,”
Social Work Education, vol. 29, no. 4, pp. 402–415.
Quinn, A. 2007. “Moral Virtues for Journalists,” Journal of Mass Media Ethics, vol. 22, nos. 2–3, pp.
168–186.
Radden, J. 2007. “Virtue Ethics as Professional Ethics: The Case of Psychiatry,” in Working Virtue, ed.
P. J. Ivanhoe and R. L. Walker (Oxford: Oxford University Press), pp. 113–134.
________, and J. Sadler. 2008. “Character Virtues in Psychiatric Practice,” Harvard Review of Psychiatry,
vol. 16, no. 6, pp. 373–380.
Robertson, M., and G. Walter. 2007. “Overview of Psychiatric Ethics II: Virtue Ethics and the Ethics
of Care,” Australasian Psychiatry, vol. 15, no. 3, pp. 207–211.
Rovie, E. M. 2002. “Abortion: Approaches from Virtue,” Auslegung: A Journal of Philosophy, vol. 25,
no. 2, pp. 137–150.
Saenz, C. 2010. “Virtue Ethics and the Selection of Children with Impairments: A Reply to Rosalind
McDougall,” Bioethics, vol. 24, pp. 499–506.
Sandin, P. 2007. “Collective Military Virtues,” Journal of Military Ethics, vol. 6, no. 4, pp. 303–314.
Sandler, R.. 2005. “A Virtue Ethics Perspective on Genetically Modified Crops,” in Environmental Virtue
Ethics, ed. R. Sandler and P. Cafaro (Lanham, Md.: Rowman & Littlefield), pp. 215–232.
________. 2007. Character and Environment: A Virtue-Oriented Approach to Environmental Ethics (New
York: Columbia University Press).
________, and P. Cafaro. 2005. Environmental Virtue Ethics (Lanham, Md.: Rowman & Littlefield).
Shaw, B. 2005. “A Virtue Ethics Approach to Aldo Leopard’s Land Ethics,” in Environmental Virtue
Ethics, ed. R. Sandler and P. Cafaro (Landham, Md.: Rowman & Littlefield), pp. 93–106.
Sherman, N. 2007. “Virtue and a Warrior’s Anger,” in Working Virtue: Virtue Ethics and Contemporary
Moral Problems, ed. R. L. Walker and P. J. Ivanhoe (Oxford University Press).
________. 2010. “Virtue and Emotional Demeanor,” in Feelings and Emotions: Interdisciplinary Ex-
plorations, ed. A. Manstead, N. Frijda, and A. Fischer (Cambridge: Cambridge University Press).
Slote, M. 2001. Morals from Motives (Oxford: Oxford University Press).
Steiner, L., and C. M. Okrusch. 2006. “Care as a Virtue for Journalists,” Journal of Mass Media Ethics,
vol. 21, nos. 2–3, pp. 102–122.
Solomon, R. C. 2003. “Victims of Circumstances? A Defense of Virtue Ethics in Business,” Business
Ethics Quarterly, vol. 13, pp. 43–62.
Stewart-Sicking, J. 2008. “Virtues, Values, and the Good Life: Alasdair MacIntyre’s Virtue Ethics and
Its Implications for Counseling,” Counseling and Values, vol. 52, no. 2, pp. 156–171.
Swanton, C. 2007. “Virtue Ethics, Role Ethics, and Business Ethics,” in Working Virtue, ed. P. J. Ivanhoe
and R. L. Walker (Oxford: Oxford University Press), pp. 207–224.
Tessman, L. 2005. Burdened Virtues: Virtue Ethics for Liberatory Struggles (Oxford: Oxford University
Press).
Thompson, A. 2010. “Radical Hope for Living Well in a Warmer World,” Journal of Agricultural En-
vironmental Ethics, vol. 23, pp. 43–59.
Throop, W. 2012. “Environmental Virtues and the Aims of Restoration,” in Environmental Virtues and
the Aims of Restoration, ed. A. Thompson and J. Bendik-Keymer (Cambridge, Mass.: MIT Press).
Tiberius, V. 2002. “Virtue and Practical Deliberation,” Philosophical Studies, vol. 111, no. 2, pp. 147–172.
Treanor, B. 2010. “Environmentalism and Public Virtue,” Agricultural and Environmental Ethics, vol.
23, nos. 1–2, pp. 9–28.
APQ 49_3 text.indd 202 5/14/12 9:29 AM

Vogel, D. 2005. The Market for Virtue (Washington, D.C.: Brookings Institution Press).
Volkman, R. 2010. “Why Information Ethics Must Begin with Virtue Ethics,” Metaphilosophy, vol.
41, no. 3, pp. 380–401.
Walker, R. 2005. “Living Well with End Stage Renal Disease: Patients’ Narratives Interpreted from a
Virtue Perspective,” Ethical Theory and Moral Practice, vol. 8, no. 5, pp. 485–506.
Walker, R., and P. Ivanhoe. 2007. Working Virtue: Virtue Ethics and Contemporary Moral Problems
(Oxford: Oxford University Press).
Weithman, P. 2010. “Religious Education and Democratic Character,” in Deliberative Democracy in
Practice, ed. D. Kahane, D. Weinstock, D. Leydet, and M Williams (Vancouver, Canada: University
of British Columbia Press).
Welchman, J. 2005. “Virtue Ethics and Human Development: A Pragmatist Approach,” in Virtue Ethics
Old and New, ed. S. M. Gardiner (Ithaca, N.Y.: Cornell University Press), pp. 142–158.
________. 2006. The Practice of Virtue: Classic and Contemporary Readings in Virtue Ethics (India-
napolis, Ind.: Hackett).
________. 2008. “Environmental Virtue Ethics,” Journal of Applied Philosophy, vol. 25, no. 1, pp. 77–83.
Wensveen, L. van. 2000. Dirty Virtues: The Emergence of Ecological Virtue Ethics (Amherst, N.Y.:
Humanity Books).
________. 2005. “The Emergence of Ecological Virtue Language,” in Environmental Virtue Ethics, ed.
R. Sandler and P. Cafaro (Lanham, Md.: Rowman & Littlefield), pp. 15–30.
Wenz, P. 2005. “Synergistic Environmental Virtues: Consumerism and Human Flourishing,” in Envi-
ronmental Virtue Ethics, ed. R. Sandler and P. Cafaro (Lanham, Md.: Rowman & Littlefield).
Westra, L. 2005. “Virtue Ethics as Foundational for a Global Ethic,” in Environmental Virtue Ethics,
ed. R. Sandler and P. Cafaro (Lanham, Md.: Rowman & Littlefield), pp. 79–92.
Wiggins, D. 2004. “Neo-Aristotelian Reflections on Justice,” Mind, vol. 113, no. 451, pp. 477–512.
Wilson, M. D. 2008. “Applied Virtue Ethics: A Philosophic Perspective on Authentic School Leadership,”
typescript conference paper, Annual Philosophy of Education Society of Great Britain Conference,
New College, Oxford, March 28–30, 2008, at http://www.philosophy-of-education.org/conferences/
pdfs/Michael_Wilson.pdf
Wyatt, W. 2008. “Being Aristotelian: Using Virtue Ethics in an Applied Media Ethics Course,” Mass
Media Ethics, vol. 23, no. 4, pp. 296–307.
Zagzebski, L. 1996. Virtues of the Mind (Cambridge: Cambridge University Press).
Zyl, Liezl van. 2002. “Virtue Theory and Applied Ethics,” South African Journal of Philosophy, vol.
21, pp. 133–144.
________. 2004. “Virtuous Motives, Moral Luck, and Assisted Death,” South African Journal of Phi-
losophy, vol. 23, no. 1, pp. 20–33.
________. 2009. “Agent-based Virtue Ethics and the Problem of Action Guidance,” Journal of Moral
Philosophy, vol. 6, pp. 50–69.
APQ 49_3 text.indd 203 5/14/12 9:29 AM

r Metaphilosophy LLC and Blackwell Publishing Ltd. 2003.
Published by Blackwell Publishing, 9600 Garsington Road, Oxford OX4 2DQ, UK and
350 Main Street, Malden, MA 02148, USA
METAPHILOSOPHY
Vol. 34, No. 3, April 2003
0026-1068
FELIX CULPA: LUCK IN ETHICS AND EPISTEMOLOGY
GUY AXTELL
ABSTRACT: Luck threatens in similar ways our conceptions of both moral and
epistemic evaluation. This essay examines the problem of luck as a metaphilo-
sophical problem spanning the division between subfields in philosophy. I first
explore the analogies between ethical and epistemic luck by comparing influential
attempts to expunge luck from our conceptions of agency in these two subfields. I
then focus upon Duncan Pritchard’s challenge to the motivations underlying
virtue epistemology, based specifically on its handling of the problem of epistemic
luck. I argue that (1) consideration of the multifold nature of the problem of
epistemic luck to an adequate account of human knowledge drives us to a mixed
externalist epistemology; and (2) the virtue-theoretical approach presents a
particularly advantageous way of framing and developing a mixed externalist
epistemology.
Keywords: virtue epistemology, virtue theory, moral and epistemic luck,

internalism and externalism, metaphilosophy.
Felix Culpa! [Oh, most fortunate fault!]

– St. Augustine
Luck has been a topic of philosophical interest dating back to the

ancients. Aristotle’s Nichomachean Ethics gave qualified acceptance to
the role of luck (tuche) - and ‘‘external goods’’ in attaining complete
happiness, thereby retaining connections with the archaic roots of the
term eudaimonia, which literally means having a good spirit watching
over oneself from birth. But following the Socratic tendencies in
Aristotle’s own thought, the later Stoics found more inviting his claim
that ‘‘to enlist to tuche- what is greatest and most noble would be quite
inappropriate’’ (NE 1099b 10–25; 2000, 15).1 They insulated their
accounts of moral agency from the perceived threat posed by luck by
reducing the domain of legitimate moral judgment to the inner life of
moral agents, emphasizing self-sufficiency and claiming that one’s
complete happiness depends only upon oneself. Under Stoicism the
1
For Aristotle’s claim that arete is largely within our control, but that ‘‘happiness needs
the presence of external goods as well,’’ see NE 1099a 31–33; 2000, 15.
r Metaphilosophy LLC and Blackwell Publishing Ltd. 2003

332 GUY AXTELL
virtues would shift from being the major components of happiness to

being exhaustively constitutive of it; the ‘‘indifferents,’’ or external goods,
were denied as constituents of complete happiness (Sherman and White
forthcoming).
The Stoics also tried to extend this barrier against luck into the realm
of cognition, overrunning along the way certain distinctions that
Aristotle conceived of between the virtues of character and the
intellectual virtues. Nancy Sherman and Heath White (forthcoming) tell
us that in trying to make this parallel the Stoics conjectured that ‘‘just as
the sage’s moral virtue is sufficient for his happiness, so too his epistemic
virtue is sufficient for the truth.’’ There need be no luck involved in
believing the truth. This requires, however, a conception of the knower as
essentially infallible, holding that cognitive skills honed the right way are
sufficient for truth, and ‘‘that if we are non-careless and non-precipitous,
non-casual and non-random in our assent to appearances, we can be
infallible in our grasp of the truth’’ (Sherman and White forthcoming).
We can easily discern from this how the temptation of a Stoic view of
knowledge has been reflected in the internalist and deontological strains
of modern epistemology since Descartes’s time. ‘‘On the Cartesian view
as well as the Stoics’, the availability of an infallible means for discerning
the truth places knowledge squarely within the sphere of our own
control’’ (Sherman and White forthcoming). Or as Daniel Statman (1991,
150) puts it rather more simply, ‘‘Epistemic justification’s immunity to
luck is gained by viewing it as a deontological notion.’’ But over the past
half-century philosophy has undergone a remarkable ‘‘naturalistic turn,’’
leading to the reassessment of many cornerstone assumptions of the
modernist tradition. For the present purposes of this essay, the most
relevant challenges are those that have led philosophers to reflect on the
role of luck, and to reexamine the relationship between ethical and
epistemic evaluation. I refer in particular to Thomas Nagel’s ‘‘Moral
Luck’’ (1979) and Edmund Gettier’s ‘‘Is Justified True Belief Knowl-
edge?’’ (1963).
Since Nagel framed the question of moral luck and indicated four
distinct types – constitutive, circumstantial, causal, and resultant – much
effort has gone into refuting his muted acceptance that luck plays a role
in our assessments of moral responsibility. Following instead basic Stoic
and Kantian intuitions, many philosophers have deemed it unacceptable
that moral differences between agents should be affected by luck. In
epistemology, consideration of Gettier cases has strongly pushed toward
objective, external constraints on knowledge that can prevent a justified,
true belief from counting as an instance of knowledge if its truth is merely
a coincidence and is not appropriately ‘‘linked’’ with its causal ground.
This externalist or truth-conducivist understanding of epistemic
justification has had far-reaching implications for how philosophers
respond to the challenge of radical skepticism. Eschewing the circular

FELIX CULPA: LUCK IN ETHICS AND EPISTEMOLOGY 333
guarantee upon which Descartes pinned his return out of deep subjective
doubt, philosophers have largely given up on the idea that there is any
necessary connection between subjective justification and truth. We have
turned fallibilist, no longer basing our response to skepticism on our
ability to find a method to secure certitude for all of our beliefs. If a true
belief can be unjustified and a justified belief can fail to be true, we may
need to take another direction in looking for conditions that mark
knowledge off from (mere) true belief, a direction that gives greater (or
even exclusive) place to factors of which the knowing agent may not be
internally aware. In epistemology this externalist path is called a ‘‘truth-
linked’’ analysis of knowledge, because it starts from the external,
antecedently specified goal of maximizing interesting true beliefs and
minimizing false ones. As Martha Nussbaum points out, the Greek
tradition already contained resources for such an alternative means of
subduing tuche, - for the identification of ‘‘something external, clear, and
antecedently specifiable that counts as an end result’’ of an activity
provides a clear measure of that activity’s success (2002, 165).
Andrew Latus helps to clarify further the connection between Nagel’s
challenge in ethics and Gettier’s challenge in epistemology when he writes
in ‘‘Moral and Epistemic Luck’’ (2000, 161–62) that, ‘‘just as we are
inclined to believe that luck cannot play a role in determining a person’s
moral standing, so epistemologists have been inclined to think that luck
cannot play a role in instances of knowledge . . . . It is the incompatibility
thesis that drives the antagonism between epistemological internalists and
externalists.’’ When, for instance, Gettier-style counterexamples are
routinely taken to falsify an author’s analysis, it is in large part dependent
upon an unquestioned assumption of the incompatibility of luck with
knowledge.
In the Gettier literature each writer tries to do to his or her opponent’s
analysis of knowledge what Gettier is perceived to have done to the
standard justified true belief (JTB) analysis. Counterexamples using the
Gettier technique of double luck are constructed and routinely taken to
falsify an author’s analysis of necessary and sufficient conditions on
knowledge. Meeting ‘Gettier’s challenge’ has typically been presumed to
mean identifying a ‘something extra,’ a ‘property x’ that together with
true belief would provide a set of necessary and sufficient conditions for
knowledge. But here a deeper question emerges about whether this is
really what Gettier’s challenge should be taken to be. While the Gettier
literature has been a battleground for debate between epistemological
externalists and internalists, epistemologists have not always been
sufficiently self-critical about the manner in which the rules of
engagement presupposed a strong thesis about the incompatibility of
knowledge with luck.
We expect internalistic theories to come with a denial of luck, in that
they understand justification in ways that demand internal access to the

334 GUY AXTELL
grounds of our beliefs and therefore presuppose a high degree of control

over whether or not our beliefs are justified. Yet, as has been pointed out
by Mylan Engel and others, there has in fact been a strong ‘‘antiluck’’ or
‘‘incompatibilist’’ characteristic to the Gettier literature on both sides of
the internalist/externalist divide. Carolyn Morillo (1984, 109) makes this
point well by saying that ‘‘‘no accident,’ nomological analyses are
characteristically externalist, naturalistic, and anti-Cartesian’’ in spirit,
yet – evoking Nagel’s conflict between one’s self-image and one’s actual
practices of evaluation – often present to us the author’s own too-
Cartesian image of the human cognizer. If this is correct, then perhaps, as
she suggests, the approach to constructing and criticizing an analysis of
knowledge (which gave rise to this now vast Gettier literature) has not
been naturalistic enough.
The reservation that I express here is in line with the deep ambivalence
that Latus gives voice to concerning the incompatibility thesis and the
role that it has played in the countless attempts to meet Gettier’s
challenge. Latus points out that this ambivalence has been widely felt of
late and helps explain why what we might call the ‘driving force’ of
externalist views about knowledge and justification has never quite
surmounted the ‘resistance’ of internalism. Indeed, as he suggests, the
lesson that we should take from these debates is not the unbridled victory
of the one side. Rather, failure on both sides to meet the demands that
they commonly presuppose in conformance with some version of the
incompatibility thesis has apparently worked to raise the appeal of
skepticism.
This problem of how to give constructive meaning to Gettier’s
challenge has been heightened by philosophers who have more recently
taken notice that Gettier-type counterexamples appear to be framable for
‘‘any fallibilist theory’’ (Harper 1996, 278), or, again, for any analysis in
which the proposed additional conditions are based on ‘‘defeasible
probability statements’’ (Craig 1990, 146), or ‘‘do not entail truth’’
(Zagzebski 1996, 288).2 If this is correct, then surely our assumption that
we meet Gettier’s challenge only by simple expulsion of epistemic luck
from our analyses of knowledge needs to be reassessed. For any account
not subject to such problems, widely construed, will then become subject
to a dilemma: Either it will be one that risks returning us an all-too-
Cartesian image of the human cognizer, or else it will be one that risks
setting standards for knowledge so high that we would all agree they can
only rarely be met.
For a variety of reasons, then, the pursuit of an analysis of knowledge
providing necessary and sufficient conditions on an ‘‘incompatibilist’’
2
For example, Harper (1996, 278) writes, ‘‘No fallibilist theory of justification can rule
out unacceptable luck, regardless of whether one construes the concept of justification
internalistically or externalistically’’.

basis has been charged not merely with becoming a trivial pursuit but
actually a dangerous one in the philosophical sense. Skepticism has
always gained a foothold in the ‘‘gaps’’ between subjective justification
and truth through which luck can flow (Axtell 2001); but by committing
ourselves on principle to discounting luck and then failing to deliver, we –
all of us, externalists and internalists alike – invite the skeptic in through
a broader breach. We have, as Latus vividly puts it, engaged in a mug’s
game of mutual refutation played by rules too accepting of a largely
unexplicated thesis of the incompatibility of knowledge with luck. Since
the most basic rule of this game is that the contrasting intuitions of
externalists and internalists make them mutually exclusive competitors
over a single ‘‘correct’’ analysis of justification, the skeptic has had only
to sit back bemused while the strongest of polemical objections and
skeptical arguments are unfolded one against another by the contestants
themselves.
I think there is some truth in this parody of the heyday of the
internalist/externalist debate, but in contrast I find philosophers doing
something far more interesting in their analyses today. Some are
challenging the import of abstract counterexamples to the adequacy of
an analysis of knowledge, and some are questioning the role assigned to
intuitions in adjudicating claims to knowledge. Many now validate the
intuitions on both sides of the internalist/externalist debate and are
seeking for ways to move beyond it by constructing a consistent, ‘‘mixed’’
account. It is here that we can start to make connections with virtue
epistemology.
In the final section of this essay I shall return to the thesis of the
incompatibility of luck with knowledge, and the question of how various
positions on the issue of epistemic luck parallel or differ from those
treated in respect to the problem of luck in ethics. First, however, I want
to explore further the topic of epistemic luck on its own terms, and to
make a special case for a virtue-epistemic approach to the problem. I can
better frame this approach by referring to Duncan Pritchard’s (2003)
‘‘Virtue Epistemology and Epistemic Luck.’’
Pritchard raises what he believes to be a fundamental difficulty for
aretaic, or virtue-based, epistemological theories. While generally quite
sympathetic to virtue theory, he wonders why a virtue-epistemic
approach is necessary for responding to the problem of epistemic luck.
There are different groups of virtue epistemologists with divergent
accounts of warrant or justification, and ‘‘each is focusing upon a
different species of epistemic luck’’ (107). The ‘externalist’ group demands
that we eliminate what Pritchard characterizes as ‘‘veritic epistemic luck,’’
and the ‘internalist’ group insists also on eliminating ‘‘reflective epistemic
luck.’’ But, Pritchard objects, eliminating veritic luck can be equally or
better achieved outside virtue epistemology, by modalized forms of
process reliabilism, and ‘‘there is no fully adequate way of eliminating

336 GUY AXTELL
reflective epistemic luck’’ (106). Hence, either way virtue epistemology’s

key claim, that one cannot account for knowledge and other central
epistemic concepts without appeal to the virtues, is itself dicey: ‘‘Far from
motivating the adoption of virtue-based theories, consideration of the
role of epistemic luck, in both its salient forms, actually reveals that virtue
epistemology is curiously ill-motivated as it stands’’ (127).
I must first readily acknowledge that Pritchard’s objections, if sound,
hit home with me. For I argue in an earlier article, ‘‘Epistemic Luck in
Light of the Virtues,’’ that a strong motivation for virtue epistemology
emerges when we take epistemic luck (as Pritchard himself suggests we
should) as a central epistemic concern. Can the problem of epistemic luck
motivate a virtue-theoretical response, or does it instead fail to motivate
and perhaps even undermine such an approach? To argue the affirmative
side of this question against his negative, I restate the central argument of
that earlier article in two steps, and in the course of supporting it shall try
to show it capable of repelling Pritchard’s challenge. The two-step
argument that I want to develop is the following:
1. Consideration of the multifold nature of the problem of epistemic
luck to an adequate account of human knowledge drives us to a
mixed externalist epistemology.
2. The virtue-theoretical approach presents a particularly advanta-
geous way of framing and developing a mixed externalist
epistemology.
I shall develop these two claims in the ensuing two sections. The third
section gives Pritchard’s challenge a further short, direct response, and to
conclude I reconsider attitudes toward the problem of luck as presented
in both its ethical and epistemological guises.
1. From the Problem of Luck to Mixed Externalism

The Latin expression felix culpa derives from St. Augustine’s famous
allusion to one unfortunate event, the Fall of Man, precipitating a second
event of good auspice, the coming of the Redeemer. Aside from the high
stakes in Augustine’s allusion, it displays a degree of formal symmetry
with Gettier cases, which as Pritchard points out are also constructed
from instances of double luck. In Gettier cases an agent’s belief
eventually comes out true but purportedly fails to constitute knowledge
because its truth is ‘‘lucky,’’ having been already cut off by the first
instance of luck from the evidential basis that provided its justification
for the agent.
Luck is ‘‘double’’ in another, quite different sense, which Pritchard
captures in saying that a key to understanding externalism and
internalism is that ‘‘each is focusing upon a different species of epistemic
luck’’ (107). As Wayne Riggs (1999, 463) similarly argued, ‘‘Both

truth-conducivist and responsibilist conceptions of epistemic justification

are directed at disallowing a particular kind of chance (true) belief from
counting as epistemically sanctioned.’’ Several authors in recent years
have made directed attempts to resolve the clash of intuitions involved in
the externalist/internalist debate by delineating the different species of
epistemic luck that inform their contrasting intuitions.3 While there is
certainly room to debate both how various forms of epistemic luck are to
be distinguished and whether and when they are indeed knowledge
precluding, I generally find this approach to epistemic evaluation through
the problem of luck illuminating, and I agree with Pritchard that ‘‘both
forms of epistemic luck need to be responded to in any adequate
epistemological theory’’ (106). We both attempt to frame externalist and
internalist intuitions about justification in terms of the kind of luck that
each side takes to be knowledge precluding.
On my own account, this task is made easier if we first attend to the
‘‘pure’’ or ‘‘simple’’ or ‘‘unmixed’’ forms of these theories, since these are
the approaches most susceptible to skeptical challenge. I shall use
Pritchard’s terms veritic and reflective luck for the sake of simplicity,
though the discussion that follows differs in significant ways and we
might just as easily substitute Hamid Vahid’s terms truth-oriented and
justification-oriented luck (2001). This susceptibility to skeptical challenge
is easiest to see with the internalist view, so we can start by characterizing
the type of luck that it sacrifices all else to exclude; we can then better
understand how skeptical problems arise for epistemic internalism from
another type of luck, to which it in consequence renders itself defenseless.
Focusing upon a zetetic context, a context concerned with the quality
of an agent’s inquiry into truths as yet unknown to her, internalists
emphasize a first-person perspective on epistemic evaluation. Following
from this, they have been concerned to exclude reflective luck, or luck
with respect to the upstream (or inputs) of a belief-forming cognitive
process. Within the context of zetetic inquiry several concerns with luck
can be raised that draw our attention to the agent’s motivational states
(dispositions to believe) and information-gathering behavior.4 In terms of
Nagel’s four types, these would certainly include epistemic analogues of
3
These authors include Engel, Vahid, Riggs, and Pritchard. The attempt to frame an
explicit no-luck condition on knowledge is typically traced to Peter Unger (1968), and
includes essays cited here by Harper and Riggs.
4
Sherman and White (forthcoming) view the emphasis on the emotions in Aristotle’s
thought (and even to some extent in Stoicism) as ‘‘a resource for contemporary virtue
epistemology.’’ Emotion is important in both moral and intellectual virtue, since ‘‘full virtue
of either sort is never just a disposition or capacity, it is a way of ‘standing toward’
dispositions and faculties that involve conscious shaping, regulation, and valuing as
component parts of living well.’’ Emotions are constitutive of character for Sherman and
White, and ‘‘the Aristotelian claim that virtue is characterized by apt emotions holds in the
intellectual sphere as well.’’

338 GUY AXTELL
what he termed both ‘‘constitutive’’ and ‘‘circumstantial’’ luck. The

problem of constitutive reflective luck points to the affective and
motivational dimensions of cognition, insisting that justification and
hence knowledge depend upon an agent who is both properly motivated
and properly affected by her cognitive environment. The problem of
circumstantial reflective luck points to a demand that beliefs be ‘‘based
upon’’ or propositions accepted ‘‘in light of’’ good epistemic reasons.
Internalists want to eliminate from knowledge all cases such as where S
truly believes p, but the belief is ‘‘not prompted by’’ S’s evidence. Pure
internalists typically confine themselves to a strictly subjective sense of
justification in which the evaluation of justifying factors is restricted to
evidence available to the agent (and other beliefs she holds) at a given
point.5 This is because, following intuitions commonplace in ethics,
internalists find the highest degree of luck in consequences, and hence
want to restrict justification to factors one has the most control over or
reflective ‘‘access’’ to. If my best defense against reflective luck is to base
my belief in normatively ‘‘good reasons’’ of which I can be reflectively
aware, she reasons, then we can exclude from knowledge all cases in
which it is mere (reflective) luck that the belief is true, given the evidence
available to me.
Yet in the broader context of zetetic inquiry it clearly also matters
what evidence may be uncovered with further effort and inquiry, and why
the agent has or doesn’t have the means to bring that ‘‘objective’’
evidence, including potential factual defeaters, to her attention. This
possibility of objective defeaters of my belief appears from the subjective
perspective on justification to be something of a matter of luck. If one
concedes that objective justification is necessary, then such defeaters may
exist to undermine the justification of my belief no matter how well I have
conducted my own mind. This objective aspect of justification also raises
important social dimensions of knowledge. Are available sources of
information generally reliable? Are they reliable in a particular case? Has
the subject ‘blinked’ information generally known within her community,
or accepted evidence on false testimony or dubious authority? How do
such factors affect the quality of the agent’s inquiries and our evaluations
of whether or not her true belief constitutes knowledge?6
A second and still broader gap for the emergence of luck accrues from
noting that even beliefs that have these forms of justification are not
necessarily true. For the internalist such questions about the ‘‘output’’ of
the process must be left as radical contingencies. I can have everything
5
For a further rationale for the internalist perspective, see the account of epistemic luck
entailed by Richard Foley’s ‘‘Cartesian approach to epistemology’’ (Foley 1987, chapter 4,
especially 199 and 206).
6
Vahid (2001, 360), for instance, suggests that ‘‘the more inaccessible the defeaters are,
the more compatible the resulting justification-oriented luck with knowledge is.’’

right at my end, that is, in how well I comport my mind, but whether or
not the world complies with the output of my best cognitive efforts must
ultimately be deemed, from this first-person perspective, as being beyond
our ken. Hence veritic luck or luck of the output remains a big problem
for any such account, as witnessed in the charge that the internalist
analysis of knowledge still leaves us quite open to radical (or brain-in-a-
vat) skepticism.
Focusing upon a reconstructive context for epistemology, on the other
hand, the more ‘‘pure’’ forms of externalism, such as simple reliabilism
and counterfactual analyses, take a third-person perspective on epistemic
evaluation. Following from this, they have been concerned to exclude the
veritic-luck characteristic of Gettier’s cases, a kind of luck analogous to
Nagel’s ‘‘resultant’’ luck because it concerns the downstream (or outputs)
of a belief-forming cognitive process. This context, as we saw, draws
attention to serious objections to the JTB analysis of knowledge, and
pure or radical externalism attempts to leave subjective justification
entirely out of the analysis of knowledge. It is the fact of p, rather, that
must help us to explain the belief that p; this third-person perspective
provides for a reliably truth-conducive dependency or ‘‘link’’ between the
knower and the object known. Knowledge is true belief that meets certain
objective conditions, such as counterfactual conditionals or having been
produced through a reliably truth-conducive causal mechanism. This
may even engender the replacing of justification conditions, placing it
irrevocably beyond the agent’s ken as to whether or not she meets the
conditions for having knowledge.
Of course, the identification of cognitive faculties as reliable proceeds
on a probabilistic basis, and is relative to the field over which the faculty
ranges and the normal conditions in which it operates. One must
therefore avoid committing the fallacy of division that would occur from
attributing success to an agent in any particular instance based only on
such general reliability of her faculties. Notice also that in such an
externalist approach it does not matter how fortunate we are in the
cognitive faculties we come equipped with or how dependable the
information or evidence is that these faculties must rely upon as inputs in
any particular instance. Such ‘‘upstream’’ matters, including any role
played by an agent’s motivational states and evidence gathering and
weighing, are of no direct relevance in a pure externalist analysis of
knowledge; looking at knowledge from a third-person perspective, these
‘‘inner’’ intensional and qualitative aspects of a particular person’s
cognitive agency must remain beyond our ken. This is to say that as pure
externalism positions itself to expel veritic luck, it also opens itself to
reflective luck.
So, in consequence, externalists maintain that internalism allows too
much room for the operation of veritic luck, and internalists maintain
that externalism has the same fault with respect to reflective luck. Both, I

340 GUY AXTELL
would argue, are right, since what is revealed here is merely a result of the
limitations of their respective starting points. As each approach seeks to
exclude a particular form of epistemic luck, it becomes susceptible to
another kind. Little wonder, then, that Nagel views the general structure
of the problem of luck as inherent in the tension between a subjective and
an objective view of action. Statman and Latus both offer explanations of
this susceptibility in terms of the logic of constructing counterexample
cases in both ethics and epistemology. According to Latus (2000, 166), the
skeptic has many options to find a foothold for luck in the process of
belief formation, for ‘‘it is always possible to simply pass the entrance of
luck down a step in the chain, from something like Nagel’s resultant luck,
to causal luck, to circumstantial luck, to constitutive luck, wherein even
one’s basic dispositions and traits of character can be presented as if
acquired in some manner outside all personal control.’’7
The upshot of our discussion (here in line with Engel, Harper, Latus,
Riggs, Vahid, and Axtell [2001]), is that pure externalism and internalism
are necessarily ‘‘incomplete’’ accounts of knowledge and justification.
Along with the strengths of each ‘‘pure’’ externalist and internalist
analysis of knowledge come certain inherent weaknesses, and paramount
among these weaknesses is their respective susceptibility to a kind of luck
that their opponent believes to be knowledge precluding. This is by no
means a ‘mysterious’ outcome: that the strength of each account should
appear the weakness of the other is just what we might expect if the
conflict of intuitions is ultimately based more on divergent interests than
on strictly incompatible logics.
But of course ‘susceptibility’ is not a term that either side would
choose, except to characterize its opponent’s position. In the rhetoric of
the externalist/internalist debate, ‘acquiescence’ would be a preferred
term. For neither side denies that it would be a ‘good thing’ or ‘desirable’
to meet the conditions that the other places on justification, but, simply
because its logic cannot allow it, each denies it as a valid condition on
knowledge. Each presents the form of luck to which the other is
susceptible as undermining, but presents the form from which its its own
account suffers as ‘ineliminable’ and at any rate relatively innocuous;
instead, as Pritchard says (126), it is simply ‘‘part of what we might call
the ‘epistemic condition,’’’ and something to which we should humbly
7
So according to Latus (2000, 161), ‘‘Attempts to avoid the existence of moral luck fail
because the conditions they appeal to that will supposedly nonluckily determine the moral
worth of the person in question turn out to be able to be satisfied themselves by luck.’’
Statman (1991, 152) adds an important point to this when he says that in ethics it is the
relative independence of forms of luck that makes it so easy for luck to be ‘‘passed down the
line’’ in constructing counterexample cases: ‘‘The independence of the various ways in which
luck influences morality strengthens the skepticism engendered by luck. Thus, any attempt
to reject moral luck and the skepticism it raises must show that each and all of these kinds of
luck don’t have an effect on morality.’’

acquiesce. But without further argument this logic is circular, since it

merely reiterates respective claims of each side that we should accept its
intuitions and reject those of its opponents. On the orientation we are
urging, then, such remarks as these are typically part of the rhetoric of the
externalist/internalist debate, and they are best understood as implicit
concessions to the weaknesses of a theorist’s own analysis.
On the other hand, recognizing that simple or pure externalism and
internalism are each deeply susceptible to epistemic luck does not imply
that the kinds of luck they are respectively susceptible to are necessarily on
epistemically equal footing. We must leave this as a matter of contention,
requiring detailed argument for why and in what degree each is taken by
its opponent to be knowledge precluding. But the point I have argued (so
far only in a negative fashion) – the advantages of a mixed externalist
strategy over both pure externalism and internalism – does not require any
such thesis of strict parity in their intuitions regarding epistemic luck.
Each we have found is susceptible to a form of luck that has potential
knowledge-precluding implications, and precisely because each is equally
committed to the incompatibility of knowledge with luck, they forfeit any
possible consensus on the best way to respond to skepticism.
From these considerations I believe we can clearly infer the advantage,
with respect to the conjoined problems of luck and skepticism, of a mixed
externalist strategy. Such accounts are externalist in the standard sense
that not all the constraints they place on justification need to be internally
accessible. The ‘‘hybrid character’’ of such accounts sets them apart both
from internalist theories and from the more strongly stated externalist
theories in the contemporary literature.8 In the next section we can
begin to give a positive development of mixed externalist epistemology
and explain why an aretaic, or virtue-theoretical, approach presents a
particularly advantageous way of developing it.
2. From Mixed Externalism to Virtue Epistemology

In the introduction to their recent collection Virtue Epistemology: Essays
on Epistemic Virtue and Responsibility, Abrol Fairweather and Linda
Zagzebski (2001, 12–13) write,
It happens that the dispute between internalists and externalists can be framed
nicely within a virtue-based framework. Since some virtue epistemologists
maintain that both the causal history and efficacy of a person and her
motivational states are important in conferring virtue, both internalist and
externalist requirements must be satisfied in the possession of virtue. Rather
than ending up with radically opposed epistemic theories, this form of VE
[virtue epistemology] proposes a unified framework that admits the value of
both criteria.
8
See Zagzebski 1996, 299.

342 GUY AXTELL
This passage illustrates a ‘‘mixed’’ externalist account of justification,

sometimes also called a ‘‘dual-component’’ account. This is one that
integrates constraints on the reliability of an agent’s cognitive faculties
with constraints on the agent’s zetetic responsibility and motivational
states (or dispositions to believe). The passage above is qualified by
saying that such a mixed account of justification is proposed in ‘‘this form
of VE,’’ but it has in fact been a widely shared view among self-described
virtue epistemologists. Of course, most epistemologists today avoid the
extremes of pure externalism and internalism by espousing some admixture
of internally inaccessible and accessible constraints on justification. So I
am not claiming that the mixed externalistic approach is a distinguishing
characteristic of virtue epistemology. Our point is rather that the manner
in which this approach is given philosophical and psychological grounding
in the virtues makes VE’s defense of it unique and advantageous.
We can begin to see this by noting that on the Aristotelian view neither
the motivation nor the success component is singly sufficient for the
presence of virtue. Greco and Zagzebski (authors sharply contrasted in
Pritchard’s essay) both point out that Aristotle’s account of the moral
virtues combines considerations of inner motivation and objective
success:
Every virtue causes that of which it is a virtue to be in a good state, and to
perform its characteristic activity well. . . . If this is so in all cases, then the
virtue of a human being too will be the state that makes a human being good
and makes him perform his characteristic activity well. (NE 1106a, 16–20;
2000, 29)
The Aristotelian conception of virtue provides an integrative naturalistic

ground for affirming the complementarity of reliability and responsibility
constraints on justified belief. We can further illustrate how perceived
advantages of mixed externalism are brought to light through an aretaic
approach by adducing representative statements from a few of VE’s best-
known authors.
Ernest Sosa writes,
Our virtue epistemology and virtue ethics focus . . . on the agent and cognizer.
When the agent’s actions are said to be right and the cognizer’s beliefs to be
knowledge, we speak implicitly of the virtues, practical or intellectual, seated in
that subject. . . . Knowledge requires not only internal justification or
coherence and rationality, but also external warrant or aptness. We must be
both in good internal order and in appropriate relation to the external world.
(1997, 194 and 203)
John Greco writes,

The main idea [behind agent reliabilism] is to define knowledge in terms of
virtuous cognitive character, and to define virtuous cognitive character in

terms of proper motivation and reliable success. This takes care of the ‘no
accident’ condition on knowledge, in that true belief which is formed through
an agent’s reliable character is not an accident in any relevant sense. It takes
care of the ‘subjective justification’ condition as well, since there is proper
motivation, and as Aristotle would say, ‘the moving principle is within the
agent.’ Roughly, a belief is both subjectively and objectively justified, in the
sense required for knowledge, when it is produced by a properly motivated,
reliable cognitive agent. (2000, 5)
Linda Zagzebski writes,
It is important to see that while the account of virtue in terms of motivation
plus success is my own, the concept of virtue has almost always combined
internally accessible and (potentially) internally inaccessible elements. So the
blend of internal and external aspects is something that comes with the concept
of virtue and that gives it an enormous advantage in epistemology, where it is
becoming apparent that it is desirable to avoid both extreme externalism and
extreme internalism. (1996, 332)
Without obscuring real differences that exist among these authors, I
would like very briefly to outline three features of mixed externalist virtue
epistemology that are viewed as advantages by its proponents:
1. Integrating Reliabilist and Responsibilist Constraints on Justification
The turn toward externalism allows us to take the problem of skepticism

seriously, while opening new avenues to support knowledge against the
skeptical challenge. Grounding mixed externalism in the motivation and
success components built into the concept of a virtue allows us to avoid
the standard approach that ties these intuitions to mutually exclusive and
exhaustive logics. It views these components instead as falling out from
an analysis of intellectual virtues as relatively stable dispositions and
habits of belief formation. If something like Greco’s distinction is correct,
between subjective and objective justification as each is related to a
common aretaic-basing condition, then this will be a further advantage in
reconciling the legitimate insights that externalists and internalists have
about knowledge-precluding forms of luck.
2. Agent-Basing, or Turning to the Agent as the Seat of Justification
The failure of process reliabilism to account for the serious problem of

strange and fleeting processes is one way of indicating the need for agent-
basing, that is, for taking the agent himself, in his stable dispositions, as,
to use Sosa’s phrase, ‘‘the seat of justification.’’ Virtue epistemology and
virtue ethics both affect such a ‘‘reversal in the direction of analysis’’
(Greco), where evaluation of beliefs is tied in with the evaluation of the
agents who hold them. As Sosa puts it, ‘‘To praise a performance as

344 GUY AXTELL
skillful or an action as right, or a judgment as wise or apt, accordingly, is

to assess not only the action or the judgment, but also the reflected
aptitude or character or intelligence. This is a distinctive view with
versions both in epistemology and ethics’’(1997, 194).
3. Addressing the ‘‘Value Problem’’
The difficulty of explaining what makes knowledge more valuable than

true belief is what is called the ‘‘value problem.’’ Knowledge could not be
epistemically better than mere true belief if true belief is the only
epistemic good (DePaul 2001). The independent source of value or good
needed to account for the superior value that we accord to knowledge
over (mere) true belief is the agent herself (Greco, Riggs, Zagzebski). The
virtues add to the subject’s worth as agent or cognizer, and their value is
not purely that of instrumental means to right action or true belief.
Knowing, as the authors mentioned above all generally agree, ‘‘has
something to do with the agent getting credit for the truth, that she gets
to the truth because of something about her as a knowing agent – her
virtues or virtuous acts’’ (Zagzebski forthcoming).9
Each of these tenets of a virtue-theoretic approach in epistemology can

be viewed from the perspective of advantages in dealing with the problem
of epistemic luck. Connections with the first two tenets should already be
clear from our foregoing discussions. The agent-basing of epistemic
evaluation provides an integrated and naturalistically sound way of
balancing ‘first-person’ and ‘third-person’ constraints on knowledge.
Zagzebski (forthcoming) has recently referred to this as the idea that
knowledge is an ‘‘organic unity,’’ a unity of true beliefs arising from
virtuous intellectual performances. Regarding the third tenet, Riggs
rightly points out that ‘‘the problem of accounting for the value of
knowledge and the problem of moral luck are conceptually intertwined.
Each raises the issue of what degree of credit we are due when the
outcome in question is due, to a large degree, to chance’’ (2002, 95). Of
course, such connections need not imply that the problem of luck is more
than one way of motivating mixed externalist virtue epistemology.
Yet virtue epistemologists are sometimes divided along lines between
‘‘virtue reliabilists’’ and ‘‘virtue responsibilists,’’ a division that brings
into focus differences over such issues as the supervenience of normative
properties, responsibility for character, and prospects for a ‘unified’
account of ethical and intellectual virtue. These philosophical differences
9
‘‘The difference that makes a value difference here is the variation in the degree to
which a person’s abilities, powers, and skills are causally responsible for the outcome,
believing truly that p’’ (Riggs 2002, 94). See also Greco (forthcoming a) and Sosa
(forthcoming b).

are interesting, but Pritchard’s manner of defining the internalism/

externalism distinction is suspect if it leads one to disregard the strong
consensus about the advantages of mixed externalism that underlies these
differences.10 The problem in defending mixed externalism is not that
concerns for faculty reliability are somehow incompatible with concerns
for agent responsibility, but that the temptation to recast differences
between subinterests within VE in terms of the dichotomous categories of
the older externalist/internalist debate works to obscure the common
ground of these concerns. It is important, in particular, to avoid
conflating virtue responsibilism with epistemic internalism in any of its
standard senses, because the key constraints that virtue responsibilism
places on justification regard motivation and agency, not ‘‘access.’’ Vrinda
Dalmiya (2001, 232) makes this point well:
When the knowing-self moves to center stage, epistemic evaluation, whether it
is of beliefs or of character, cannot function within the constraints of a strict
internalism. The relaxation of internalist criteria occurs on two fronts. First,
consideration of reliability and success in achieving truth become relevant, and
second, a social dimension is introduced to rupture the isolationism of purely
‘internal’ looks ‘within’. . . . Epistemic responsibility now is not a function of
either not violating epistemic obligations (deontology) nor of factors purely
transparent to the knower (internalism).
Aside from helping us avoid the polarizing logic of the externalist/

internalist debate, an additional and often overlooked advantage of
mixed externalist virtue epistemology is the response to skepticism that it
allows. If the failed attempts on the part of internalists and externalists
for a noncircular and ‘‘fully general theory of knowledge’’ has only
strengthened the skeptical challenge (Sosa 1999a, 93), then a more
bootstrapping approach is called for. And a bootstrapping response, in
turn, is strengthened to the extent that our judgments concerning the
reliability of our cognitive faculties and the responsibility of our zetetic
efforts can be shown to be not merely logically compatible but actually
mutually supportive or complementary. Mixed accounts that are based
specifically in the concept of intellectual virtues again seem uniquely well
equipped to reveal the ground of this complementarity, and in so doing to
support a claim of philosophical stability in the face of the skeptical
challenge. An ideal epistemic agent has beliefs formed by reliable
processes, is cognitively well integrated because ‘properly affected’ or
attuned to her environment, and seeks to acquire virtuous habits or
10
The critics of mixed externalism, including some from within the field, allege that
‘hybrid’ accounts are philosophically unstable and that ‘pure’ internalism or externalism
have the best or only chance of redressing the challenge of skepticism. See Axtell 2001,
158–64, for a discussion of the differences between virtue reliabilists and virtue responsibi-
lists, and 173–75 for a critique of Driver’s and Dancy’s arguments for the instability of
mixed accounts.

346 GUY AXTELL
dispositions both for their own sake and as a means of appropriately

basing her reflective beliefs on good reasons.
3. A Response to Pritchard: Optimism on (Thin) Ice?

Duncan Pritchard offers a keen analysis of how distinctions concerning
epistemic luck inform epistemological accounts, including those of virtue
epistemologists. His challenge to the motivations of virtue epistemology
takes the form of a dilemma. Until it is met, Pritchard suggests that virtue
epistemologists keep optimism regarding their approach to questions of
knowledge ‘‘on ice.’’ The first horn of his dilemma is meant to pose a
problem for those virtue epistemologists Pritchard identifies as reliabi-
lists: ‘‘If veritic epistemic luck is the problem, then a safety-based account
of knowledge is the answer’’ (127). The second horn is meant to pose a
problem specifically for those virtue epistemologists, including Linda
Zagzebski, whom Pritchard wants to characterize as ‘‘internalists’’: if
reflective epistemic luck is legitimately a supplementary problem, ‘‘then a
safety-based account of knowledge coupled with the demand for
internalist justification is the answer, and, again, this need make no
essential reference to the epistemic virtues’’ (127). My manner of
responding to Pritchard is essentially an attempt to grab the first horn
of the dilemma by showing that the problem that is supposed to accrue
for the virtue reliabilist is a nonstarter. I shall not have space to address
the second horn directly, but given what I have just said about the
mistake of confusing virtue responsibilism with epistemic internalism, I
believe that this second horn of the dilemma could be similarly rebutted.
Pritchard wants to argue that ‘‘the best accounts of knowledge
available which respond to the problem of veritic epistemic luck make no
essential mention of the epistemic virtues at all’’ (107). By ‘‘best
accounts’’ he means those that have a modally framed safety principle
to preclude Gettier-type cases. Yet Ernest Sosa’s much-discussed safety
principle and neo-Moorean response to skepticism (1999b) – both of
which are shared by Pritchard – are closely integrated into Sosa’s broader
virtue epistemology. Although Pritchard mentions at one point that Sosa
employs a safety principle, he does not examine the connections that
actually exist in Sosa’s thought between the safety principle and his
virtue-basing conditions, nor does he offer any independent reason to
think that Sosa’s employment of the principle is in some way inconsistent
with commitments entailed by his broader virtue epistemology. Rather,
Pritchard attempts to put a modalized version of process reliabilism into
competition with agent reliabilism, and to argue that if veritic luck is the
motivating problem then a modal ‘‘safety-based’’ theory can supplant
both, since it alone completely eliminates it.
It may still be instructive to see why Sosa has held that ‘Cartesian
tracking’, or ‘C- tracking’, and the safety principle that he derives from it

is only a necessary condition on knowledge, and one which – far from

standing sufficiently on its own – entails essential reference to the virtues
to be properly understood. Unaided, tracking accounts reduce justifica-
tion to little more than a claim about that belief’s counterfactual relation
to the truth of what is believed. For Sosa, ‘‘what is required for a belief to
be safe is not just that it would be held only if true, but rather that it be
based on a reliable indication’’ (2000b, note 10, 17–18). This account is
given fuller development in Sosa’s ‘‘Tracking, Competence, and Knowl-
edge’’ (forthcoming a), and here we find again that
the safety requirement is after all just a necessary condition. So the mere fact
that a belief satisfies that requirement does not commit us to counting it a case
of knowledge. On the contrary, whether or not one knows . . . will plausibly
also depend also on why it is that one believes as one does. And here one must
look to the ‘‘habit’’ of thought that [guides one’s belief]. (Sosa forthcoming a)
So Sosa’s fuller view, of which safety is one part, is what he terms a

‘‘tracking-through-a-virtue view’’: for a proper understanding of tracking,
it is essential and not merely ancillary that it is the agent’s constitution and
positioning via-à-vis a fact p, and the field in which that fact obtains, that
determines whether a subject will or will not track the truth as to p. In
broader view, as expressed in ‘‘The Place of Truth in Epistemology,’’
eudaimonist virtue epistemology gives pride of place to truth, as Aristotle
explains. But the truth that matters is not just the truth of true propositions in
splendid isolation. What matters is your having the truth. And what matters
most importantly, the chief good, is your grasping the truth attributably to
your intellectual virtues acting in concert conducted by reason, and thus
attributably to you as epistemic agent. (Sosa forthcoming b)
Although we do not have space here to evaluate all aspects of Sosa’s

account, these arguments to the effect that a safety principle is necessary
but not sufficient for knowledge directly answer the problem that
Pritchard’s poses as the first horn of his dilemma, and they show that
it does not present a serious problem to the motivations for reliabilist
versions of virtue epistemology. These lead to other, interrelated issues on
which virtue epistemology sets itself apart, such as the value problem
defined above, where it appears that an adequate response requires a turn
to agent-basing. I would suggest that Pritchard’s decision to contrast his
safety-based analysis with both reliabilism and virtue epistemology leaves
him without resources to address this problem. But if the value problem
is a serious one, then mustn’t what helps us to resolve it be present as part
of the very conditions for knowledge?
Epistemologists have all too often written as if the particular feature
on which they themselves focus excludes the insights of others and
thus exhausts the analysis of knowledge and justification. Virtue
epistemologists can claim no general exemption from this common

348 GUY AXTELL
academic fault, but their culpability in it seems less than that of writers
generally during the heyday of the internalist/externalist debate. Given
our discussion in the previous sections, I hold it as erroneous to
characterize the outstanding differences among Sosa, Greco, and
Zagzebski in terms of the division between externalist and internalist
accounts of justification. Mixed externalism rejects the position that sees
these groups as mutually exclusive competitors over a single ‘‘right’’
concept of justification.11
This can lead us to compare Pritchard and Greco more directly.
Pritchard regards the problems that have arisen for process reliabilist views
as pushing us in the direction of a safety-based account, while Greco (1999)
holds that this same movement has been one toward ‘‘agent reliabilism.’’
Aiming at minimalism, each is in this sense a ‘‘three-condition’’
epistemologist. Pritchard leans toward the view that true belief which
meets the safety principle is the essence of knowing, and that a further
demand for aretaic grounding of belief is therefore superfluous. Greco,
conversely, advocates an aretaic-grounding condition on knowledge, and
since this ‘‘takes care of the ‘no accident’ condition on knowledge,’’ a
distinct safety principle would seem to be superfluous. Ironically, in Putting
Skeptics in Their Place (2000), Greco does not claim sufficiency for the
conditions he puts on knowledge. He stays with ‘‘S knows p only if’’
throughout the book because he concedes there that his aretaic-grounding
condition is ‘‘not adequate for addressing Gettier problems’’ (251). Sosa’s
tracking principle is seen as one attempt to provide direct aid against
Gettier problems, yet Greco remains uncommitted to such a strategy. But
in ‘‘Virtues in Epistemology’’ Greco (forthcoming b) offers essentially the
same set of conditions, though now with ‘‘S has knowledge regarding p if
and only if S believes the truth regarding p because S believes p out of
intellectual virtue.’’12 This considerably stronger stance he supports by
arguing that ‘‘agent reliabilism has the resources to address a wide range of
‘Gettier problems’’’ (Greco forthcoming).
Neither of these ‘‘three-condition’’ analyses secures my confidence,
and I must simply fall back on a rather mundane four-condition strategy,
true belief plus both an aretaic-grounding condition and a c-tracking or
other condition specifying forms of knowledge-precluding luck. We can
simply call this the DATA account, for its doxastic, alethic, tucheic, and
aretaic conditions. But this suggestion to combine some version of the
conditions that, if I have read them correctly, Pritchard and Greco
independently take as sufficient for knowledge will then seem redundant
11
For more on the rejection of value monism and the defense of value pluralism, see
Riggs (1998), DePaul (2001), and Zagzebski (forthcoming).
12
For Greco this condition implies distinct but interrelated subjective and objective
justification conditions both framed in aretaic terms; but he says that if we further stipulate
that intellectual virtues involve a motivation to believe the truth (the subjective-justification
idea), we may collapse the account into this more simplified condition.

to both authors. For the one, a modal no-luck provision makes any
reference to virtues or processes of the agent superfluous. For the other, a
correct attribution of the grounding of one’s true belief in an intellectual
virtue already implies nonaccidentality. I must, however, leave defense of
the alternative four-condition approach to another time.
4. Conclusions
In a ‘‘Postscript’’ to his influential paper on moral luck, Bernard Williams
(1993, 252) writes that
the resistance to luck is not an ambition gratuitously tacked on to morality: it
is built into it, and that is why morality is inevitably open to skeptical doubts
about its capacity to fulfill this ambition. In this respect there is an analogy to
the idea (which I also accept) that there are intrinsic features of the concept of
knowledge that invite skepticism . . . the concept of knowledge is itself
involved in discounting luck.
Williams is here concurring not only with Nagel’s earlier analogy between
the problem of moral luck and ongoing debates in epistemology but also
with the idea that skeptical arguments do not depend on the imposition
of arbitrarily stringent standards of knowledge but rather, as Nagel
(1979, 467) put it, ‘‘appear to grow inevitably from the consistent
application of ordinary standards.’’ Whether or not Williams and Nagel
are correct in this point of mutual agreement, our discussions have
contributed to showing how the problem of luck is by its nature a
metaphilosophical one, because luck threatens in similar ways our
conceptions both of moral and of epistemic evaluation (Statman 1991).
Adequately addressing the problem of luck as a metaphilosophical
problem – that is, as a problem that ranges across philosophic subfields –
requires avoiding certain extremes that we have touched upon. We need
to eschew both the Stoics’ over-brave synthesis of the good and the true
as well as the positivists’ prosaic dichotomy between moral and epistemic
evaluation. As we have seen, carrying over into epistemology an
essentially Stoic conception of ethical agency proves unrealistic by
setting the bar for knowledge too high. We began with Sherman and
White’s discussion of the connection between Stoicism and epistemic
internalism, and we should now turn to their own conclusion that
humans are somewhat asymmetrically related to the goals of happiness
and truth, since ‘‘happiness can be to a larger degree a matter of our own
making’’: ‘‘Though the Stoics themselves argue for a unified thesis that
demands a parallel between truth and happiness, we suggest that at least
on this issue, we do well to separate virtue epistemology from virtue
ethics’’ (Sherman and White forthcoming).
On the other hand, neither these authors’ greater deference to the role
of luck in the epistemic domain nor our own call for reexamining the

350 GUY AXTELL
cogency of the ‘‘incompatibility thesis’’ in the analysis of knowledge

implies a radical lowering of the bar, or the claim that humans have to be
content with doxa and eschew episteme. The ‘‘gaps’’ between what one
believes to be true, relying on the evidence available to one, and what is
really true have always been a motivating source of skepticism. But our
discussion of how this problem is addressed within mixed externalist
virtue epistemology has also pointed out its unique advantages for
responding to skeptical arguments.
Department of Philosophy/102
University of Nevada, Reno
Reno, NV 89557-0056
USA
axtell@unr.nevada.edu
References
Aristotle. 2000. Nichomachean Ethics. Translated by Roger Crisp.
Audi, Robert. 2000. ‘‘Philosophical Naturalism at the Turn of the
Century.’’ Journal of Philosophical Research 25: 27–46.
Axtell, Guy. 2001. ‘‘Epistemic Luck in Light of the Virtues.’’ In
Fairweather and Zagzebski 2001, 158–77. Oxford: Oxford University
Press.
FFF, ed. 2000. Knowledge, Belief, and Character: Readings in Virtue
Epistemology. Lanham, Md.: Rowman and Littlefield.
Battaly, Heather. 2001. ‘‘Thin Concepts to the Rescue: Thinning the
Concepts of Epistemic Justification and Intellectual Virtue.’’ In
Fairweather and Zagzebski 2001, 98–116.
Craig, Edward. 1990. Knowledge and the State of Nature. Oxford: Oxford
University Press.
Dalmiya, Vrinda. 2001. ‘‘Knowing People.’’ In Steup 2001b, 221–34.
Dancy, Jonathan. 2000. ‘‘Supervenience, Virtues and Consequences.’’ In
Axtell 2000, 73–86.
DePaul, Michael. 2001. ‘‘Value Monism in Epistemology.’’ In Steup
2001b, 170–83.
DePaul, Michael, and Linda Zagzebski. Forthcoming. Intellectual Virtue:
Perspectives from Ethics and Epistemology. Oxford: Oxford University
Press.
Driver, Julia. 2000. ‘‘Moral and Epistemic Virtue’’ In Axtell 2000,
123–34.
Eflin, Juli. 2000. ‘‘The Structure of Virtue Centered Epistemology.’’ Acta
Analytica 15: 73–87.
Engel, Mylan. 1992. ‘‘Is Epistemic Luck Compatible with Knowledge?’’
Southern Journal of Philosophy 30, no. 2: 59–75.

Fairweather, Abrol, and Linda Zagzebski, eds. 2001. Virtue Epistemol-

ogy: Essays on Epistemic Virtue and Responsibility. Oxford: Oxford
University Press.
Foley, Richard. 1987. The Theory of Epistemic Rationality. Cambridge,
Mass.: Harvard University Press.
Gettier, Edmund. 1963. ‘‘Is Justified True Belief Knowledge?’’ Analysis
23: 121–23.
Greco, John. 1999. ‘‘Agent Reliabilism.’’ In Philosophical Perspectives 13:
273–96.
FFF. 2002. Putting Skeptics in Their Place: The Nature of Skeptical
Arguments and Their Role in Philosophical Inquiry. Cambridge:
FFF. Forthcoming a. ‘‘Knowledge as Credit for True Belief.’’ In
DePaul and Zagzebski forthcoming.
FFF. Forthcoming b. ‘‘Virtues in Epistemology.’’ In Moser forthcoming.
FFF, ed. Forthcoming c. Sosa and His Critics. Oxford: Basil Blackwell.
Harper, William. 1996. ‘‘Knowledge and Luck.’’ Southern Journal of
Philosophy 34: 271–83.
Katzoff, Charlotte. 2001. ‘‘Epistemic Virtue and Epistemic Responsi-
bility.’’ Dialectica 55, no. 2: 105–18.
Latus, Andrew. 2000. ‘‘Moral and Epistemic Luck.’’ Journal of Philoso-
phical Research 24: 149–72.
McGinn, Colin. 1984. ‘‘The Concept of Knowledge.’’ Midwest Studies in
Philosophy 9: 529–54.
Millgram, Elijah, ed. 2002. Varieties of Practical Reasoning. Cambridge,
Mass.: MIT Press.
Morillo, Carolyn. 1984. ‘‘Epistemic Luck, Naturalistic Epistemology,
and the Ecology of Knowledge.’’ Philosophical Studies 46: 109–29.
Moser, Paul, ed. Forthcoming. Oxford Handbook of Epistemology.
Oxford: Oxford University Press.
Nagel, Thomas. 1979. ‘‘Moral Luck.’’ In Mortal Questions, 24–38. Cam-
bridge: Cambridge University Press.
Nussbaum, Martha. 2002. ‘‘The Protagoras: A Science of Practical
Reasoning.’’ In Millgram 2002, 153–202.
Pritchard, Duncan 2002. ‘‘Recent Work on Radical Skepticism.’’
American Philosophical Quarterly 39, no. 3: 215–57.
FFF. 2003. ‘‘Virtue Epistemology and Epistemic Luck.’’ Metaphilo-
sophy 34, no. 1/2: 106–30.
FFF. Forthcoming. ‘‘Epistemic Luck.’’ Journal of Philosophical
Research 28.
Riggs, Wayne. 1998. ‘‘What Are the ‘Chances’ of Being Justified?’’
Monist 81, no. 3: 452–72.
FFF. 2002. ‘‘Reliability and the Value of Knowledge.’’ Philosophy and
Phenomenological Research 64, no. 1: 79–96.

352 GUY AXTELL
FFF. Forthcoming. ‘‘Understanding ‘Virtue’ and the Virtue of Under-

standing.’’ In DePaul and Zagzebski, forthcoming.
Sherman, Nancy, and Heath White. Forthcoming. ‘‘Intellectual Virtue:
Emotions, Luck, and the Ancients.’’ In DePaul and Zagzebski forth-
coming.
Sosa, Ernest. 1997. ‘‘Reflective Knowledge in the Best Circles.’’ In Steup
2001b, 187–203.
FFF. 1999a. ‘‘Philosophical Scepticism and Epistemic Circularity.’’ In
Skepticism, edited by Keith DeRose and Ted Warfield, 93–114.
FFF. 1999b. ‘‘How to Defeat Opposition to Moore.’’ Philosophical
Perspectives 13: 141–53.
FFF. 2000a. ‘‘Skepticism and Contextualism.’’ In Philosophical Issues
10: 1–18.
FFF. 2000b. ‘‘Replies.’’ In Philosophical Issues 10: 38–42.
FFF. Forthcoming a. ‘‘Tracking, Competence, and Knowledge.’’ In
Moser forthcoming.
FFF. Forthcoming b. ‘‘The Place of Truth in Epistemology.’’ In DePaul
and Zagzebski forthcoming.
Statman, Daniel. 1991. ‘‘Moral and Epistemic Luck.’’ Ratio 4, no. 2: 146–56.
FFF. 1993. Moral Luck. New York: SUNY Press.
Steup, Matthius. 2001a. ‘‘Epistemic Duty, Evidence, and Internality.’’ In
Steup 2001b, 134–48.
FFF, ed. 2001b. Knowledge, Truth, and Duty. Oxford: Oxford
University Press.
Umbers, Richard. 2002. ‘‘Virtue Epistemology in Linda Zagzebski.’’
Unpublished dissertation, University of Navarre.
Unger, Peter. 1968. ‘‘An Analysis of Factual Knowledge.’’ Journal of
Philosophy 65, no. 6: 157–70.
Vahid, Hamid. 2001. ‘‘Knowledge and Varieties of Epistemic Luck.’’
Dialectica 55, no. 4: 351–62.
Walker, Margaret U. 1993. ‘‘Moral Luck and the Virtues of Impure
Agency.’’ In Statman 1993, 235–50.
Williams, Bernard. 1993. ‘‘Postscript.’’ In Statman 1993, 251–58.
Zagzebski, Linda. 1994. ‘‘The Inescapability of Gettier Problems.’’
Philosophical Quarterly 44, no. 174: 65–73.
FFF. 1996. Virtues of the Mind. Cambridge: Cambridge University Press.
FFF. 2001. ‘‘Recovering Understanding.’’ In Steup 2001b, 235–52.
FFF. 2003. ‘‘The Search for the Source of Epistemic Good.’’
Metaphilosophy 34, nos. 1/2: 12–28.
FFF. Forthcoming. ‘‘Epistemic Value Monism.’’ In Greco forth-
coming.

r 2010 The Author
Journal compilation r 2010 Metaphilosophy LLC and Blackwell Publishing Ltd
Published by Blackwell Publishing Ltd, 9600 Garsington Road, Oxford OX4 2DQ, UK, and
350 Main Street, Malden, MA 02148, USA
METAPHILOSOPHY
Vol. 41, Nos. 1–2, January 2010
0026-1068
AGENCY ASCRIPTIONS IN ETHICS AND EPISTEMOLOGY:

OR, NAVIGATING INTERSECTIONS, NARROW AND BROAD
GUY AXTELL
Abstract: In this article, the logic and functions of character-trait ascriptions in

ethics and epistemology is compared, and two major problems, the ‘‘generality
problem’’ for virtue epistemologies and the ‘‘global trait problem’’ for virtue
ethics, are shown to be far more similar in structure than is commonly acknowl-
edged. Beyond the aporia of character-trait ascription and between the Scylla and
Charybdis that virtue theories are faced with in each field of philosophy, we find
our passage by making full and explicit use of the ‘‘narrow-broad spectrum of trait
ascription,’’ and by accounting for the various uses of it in an inquiry-pragmatist
account. In virtue theories informed by inquiry pragmatism, the agential habits
and abilities deemed salient in explanations/evaluations of agents in particular
cases, and the determination of the relevant domains and conditions that an
agent’s habit or ability is reliably efficacious in, is determined by pragmatic
concerns related to our evaluative epistemic practices.
Keywords: generality problem, situationism, virtue, vice, virtue theory, character,

metacognition.
The sea is still the aporetic place par excellence, and it is still the best metaphor
for the aporia of discourse.
—Sarah Kofman, ‘‘Beyond Aporia?’’ (1988, 78)
1. Introduction: Major Problems with Trait Ascriptions

Character-trait ascriptions serve a variety of purposes, philosophical and non-
philosophical. The appeal to human character and personality traits in folk
psychology is arguably something of a mishmash of explanatory and evalua-
tive intentions. Philosophers often try to render commonsense character psy-
chology self-consistent and informative. There are, however, major problems
they encounter on this journey, and for virtue theorists in particular these are
often presented as a Scylla and Charybdis that cannot both be evaded.1
1
For other recent and related treatments of character-trait ascriptions see Fricker 2007
and 2008 and Upton 2005.
r 2010 The Author

74 GUY AXTELL
A major problem concerning the ascription of intellectual or epistemic

character traits to an agent is the Generality Problem. Is there a non–ad
hoc way to select the proper level of generality at which to describe a
belief-forming process in order to evaluate its reliability?2 If every token
of a belief-forming process belongs to many different types of such
processes, there may be no principled way to select the proper level of
generality to describe the process token that produced the belief. The
Generality Problem has sometimes been presented as an objection to
reliabilist theories of justification (Conee and Feldman 2004, chap. 6), but
it is more accurately a problem that must be a concern for any theory that
has even a reliability component. This is in part why Linda Zagzebski, in
Virtues of the Mind (1996), acknowledges that there is relatedly a problem
for virtue epistemologies in setting the level of generality at which
epistemic virtues are described, and this will be true whether the virtue
epistemology is one that acknowledges ‘‘faculty’’ virtues, as agent reliabil-
ist theories do, or restricts the virtues to acquired habits in the way that
Zagzebski and other neo-Aristotelian accounts do.
A major problem concerning the ascription of moral character traits is
what can here be called the Global Trait Problem, the problem often
alternatively described (in a large and growing literature) as the ‘‘situa-
tionist challenge’’ to character theory. When we ascribe a global character
trait like honesty or kindness to someone, we typically think of this trait
as robustly held such that it resists undermining, and as one so settled or
habitual that the agent will manifest it not just in a few situations that
invite it, but in many. Situationists say that character-trait ascriptions are
poor explainers and that ‘‘minor and seemingly irrelevant differences in
the perceived situation appear more readily to explain behavior and
behavioral differences than character traits’’ (Harman 2000, 223).3
Gilbert Harman (1999, 2000) claims that ascriptions of character traits
to individuals in folk psychology and philosophical ethics is subject to a
‘‘fundamental attribution error’’—we far too often jump to conclusions
about underlying personality and character traits from the behavior we
observe in people. Ultimately, Harman holds that there is in fact ‘‘no
character or personality’’ (2000). John Doris (2002) holds that there are
2
‘‘For example, suppose I form a true belief based on a coin toss. One would normally
think that this process is unreliable. But one could always cite a more fine-grained
individuation of the type to undermine that verdict, say, for example, forming beliefs about
who will win the 2007 NCAA basketball tournament based on flipping this coin on the
Monday afternoon before the championship game. If the belief is true, this ‘process’ is
correct 100% of the time, hence adequately truth-conducive, hence reliable’’ (Becker 2008,
354).
3
The thesis of the explanatory salience of robust and cross-situationally manifested
moral dispositions is claimed to come open to empirical refutation by studies, many of them
well known and even infamous, like the Milgram and the Stanford Prison experiments. See
Doris 2002 for extended discussion and interpretation.
r 2010 The Author

AGENCY ASCRIPTIONS IN ETHICS AND EPISTEMOLOGY 75
moral character traits, but that the only kinds of moral character traits
that can be rightly attributed to persons are local or narrowly construed
dispositions. Both Harman and Doris make philosophic hay from
empirical psychology, arguing that ‘‘Aristotelian-style virtue ethics’’ can
have little empirical content—little value for predictive and explanatory
purposes—since they ‘‘share with folk psychology a commitment to
broad-based character traits of a sort that people simply do not have’’
(Harman 2000, 7).
The Global Trait Problem, I want to argue, is largely a kind of
generality problem, a problem about the right level of generality at which
moral trait ascriptions best serve the explanatory and normative interests
involved in moral evaluation.4 But my aim here is comparative and
methodological; it is not an attempt to solve these problems in the
abstract but an attempt to address them as problems of practice faced
in a variety of fields cutting across philosophy and the social and
behavioral sciences. By relating these two problems directly and revealing
their shared logic, we might hope to reap the benefits of an improved
understanding of the explanatory and evaluative practices that make use
of character-trait ascriptions. One further aim of this essay is to show that
whereas the Global Trait Problem is often held captive to stale debates in
metaethics over which elements in a theory should be taken as concep-
tually primary or foundational, it could more profitably be treated in the
way that an inquiry-focused or ‘‘inquiry-pragmatist’’ epistemology would
suggest we treat the Generality Problem. On a pragmatist account, facts
alone do not determine relevant types (Kappel 2006). Rather, in any
particular case, the relevant reliability is determined by the agent’s
epistemic competence and performance, and the ways to settle on what
field (or domain) and conditions a character trait is efficacious in turns
upon pragmatic concerns related to our evaluative epistemic practices
(Kappel 2006, 539–40). There is no question of identifying the total cause
of an agent’s action or belief, and the partial causes we select and deem
salient and usefully generalized upon, whether triggering or configuring
causes, situational or agential, are contextual and have much to do with
the interests-in-explanation of the persons providing the disposition-citing
explanation.
First I will explore what common structure or logic our two problems
share, insofar as they each necessitate distinguishing the functions of
narrowly and broadly typed trait ascriptions, and the explanatory
purposes that each of these is good for. In my view the narrow and
broad kinds of trait ascription are strongly interconnected in our practices
of agent evaluation, whether moral or epistemic, and can only be
separated in theory. Philosophers have noted these interconnections,
but when they have tried to explain them, they haven’t done a very
4
I thank John Greco for discussion of this point.
r 2010 The Author

76 GUY AXTELL
good job. To do a better job, we need to think seriously about how to

apply a sliding scale, allowing some cases to be best addressed by trait
ascription of a narrow sort, with other cases best addressed by more
global or broadly typed trait ascriptions.
More study of this scale—of the narrow-broad spectrum of trait
ascription—is needed if we are to meet the burdens of the two problems.
The present account recognizes that both narrowly and broadly typed
traits serve indispensable functions and that narrowly and broadly typed
traits often are interconnected in agent and act evaluation. But I also view
this, if correct, as leading away from or having a ‘‘deflating’’ effect on
some of the debate surrounding whether we should be ‘‘internalists’’ or
‘‘externalists’’ (Feldman, Goldman) in epistemology and ‘‘localists’’ or
‘‘globalists’’ (Doris, Annas) or ‘‘occurrentists’’ or ‘‘dispositionalists’’
(Hurka, Zagzebski) in our accounts of moral character. Thus I will treat
especially the debate between equally reductionistic occurrentist accounts
like Thomas Hurka’s and dispositionalist accounts like Linda Zagzebski’s
(both in this collection) as subordinate to our shared need for a more
comprehensive and flexible account of trait attributions. Setting aside
such unmotivated ‘‘priority’’ debates is needed today if we are to get a
clearer view of what really lies in the ‘‘intersections’’ between ethics and
epistemology.
2. The Logic of Intellectual Trait Ascription and the Generality Problem

The Generality Problem is, as we briefly described it, the problem that
any process token is an instance of several process types, and it is not clear
which process type is relevant for evaluating reliability. Most responses to
the problem came from reliabilists, and followed the lead of Alvin
Goldman, who holds that the best way to navigate through the dual
potential pitfalls of defining the belief-forming process too narrowly or
too broadly, is to try to locate the narrowest (content-neutral) process
that is causally operative in belief production (Becker 2008, 363).5
But according to Christopher Lepock in ‘‘How to Make the Generality
Problem Work for You’’ (forthcoming a), different types of appraisals
pick out processes at different levels of generality, even when appraising
the same belief. While many epistemologies and even some virtue
epistemologies are reductionist in this way, granting conceptual or
explanatory primacy to narrowly or broadly typed trait ascriptions,
Lepock develops the logic of a matter of gradations or levels from
5
Conee and Feldman relate the Single Case, No Distinction, and Generality Problems
by saying that ‘‘the problem for defenders of the reliability theory, then, is to provide an
account of relevant types that is broad enough to avoid The Single Case Problem but not so
broad as to encounter The No-Distinction Problem. Let us call the problem of finding such
an account ‘The Problem of Generality’’’ (2004, 144). See also Beebe 2004 for a sound
overview.
r 2010 The Author

narrowest to broadest, as an interpretation of the distinction between the

faculty virtues that are the especial focus of the causal/explanatory
interests of externalist epistemology, and the virtues and vices that bear
upon the conduct of inquiry. An antireductionist account able to properly
acknowledge the distinction between and relationships among trait
ascriptions at various levels of generality emerges from ‘‘mixed’’ accounts
such as Christopher Hookway’s, which the present work on the narrow-
broad spectrum helps motivate. The philosophic importance of virtue-
theoretic concepts goes beyond what contribution they may make to an
analysis of knowing; at the broad end of the spectrum reside, according to
inquiry-pragmatist forms of epistemology like Hookway’s (2006) Peir-
cean form, thickly describable inquiry and deliberation-regulating cogni-
tive character traits, that is, the reflective virtues (cf. Putnam 2002; Axtell
and Carter 2008).
The problem we are addressing is fundamentally a problem of practice,
since we can and likely should continue to use both narrow- and broad-
type ascriptions but lack, in current theories, the resources to properly
relate them. Recognition that trait ascriptions necessarily run along a
spectrum from narrow to broad, together with acknowledgment that our
explanatory interests shape the determination of what situational and
agential factors we deem explanatorily salient, provides us a fresh start.
We can discern at least two distinct ways of attributing abilities or
other efficacious character traits to agents, each underlying a different
sort of appraisal of beliefs and believers: ‘‘One is based on the reliability
of the process narrowly construed, and appraises the status of the
particular belief. Another makes use of reliability of a more broadly
construed process, and describes the creditworthiness of the agent in
having formed the belief’’ (Lepock forthcoming a).
So when we look at how we actually ascribe or attribute epistemic
traits to agents, we find indications that there isn’t a single level of
generality at which the ascriptions are directed. ‘‘When a [belief-forming]
process is reliable at the narrow end of the scale, it is NTR [narrow-type
reliable]; when it is reliable at the broad end, it is BTR [broad-type
reliable]’’ (Lepock forthcoming a). A narrow or local trait is one that
yields its evaluatively relevant behavioral outputs in a relatively narrow or
local set of circumstances (compare Sosa 2008). Broad or global traits
also support desirable behaviors, but their applications go beyond even
behaviors, to the agent’s internal life. But while it might seem that
distinguishing between two distinct uses of reliability only complicates
the problem of generality, Lepock argues that we should exploit this
situation, and try to ‘‘put the generality problem to work’’ for us.
To this end Lepock asks what sorts of appraisals make use, respec-
tively, of narrowly typed and broadly typed reliability (NTR and BTR for
short). Trait ascriptions range from ‘‘low-level’’ to ‘‘high-level’’ virtues,
from those dispositions most directly involved in perceptual knowledge to
r 2010 The Author

78 GUY AXTELL
the acquired virtues as capacities for metacognitive control. ‘‘There are

prima facie important differences between these two categories and the
sort of evaluations they are involved in. . . . It appears that the value of
low-level virtues is transmitted directly to their products and only
indirectly to the agents who have them, while the value of high-level
virtues attaches directly to their possessor but only tenuously to their
products’’ (Lepock forthcoming b).
To compare them more closely, narrowly typed epistemic trait ascrip-
tions primarily serve to appraise the status of a particular belief, and it
was primarily in this context that the generality problem was framed. Says
Lepock,
NTR tells us a great deal about the etiology of a particular belief or narrow
range thereof. . . . It tells us whether we can trust the particular belief in
question, or whether that particular belief is appropriately grounded. How-
ever, the epistemic status identified by NTR does not necessarily accrue to the
agent. NTR tells us little about the agent’s overall capacities or cognitive
practices, because it reflects only the etiology of such a narrow range of beliefs
on such a narrow range of occasions. Thus while NTR seems central to
assessing the status of a single belief, it is less important for assigning credit or
blame to believers. (Forthcoming a)
Broadly typed reliability, by comparison, does not convey very specific

information about the particular belief at hand, yet it can convey a great
deal about the believer’s abilities and practices in the general area.
Whereas the main function of narrow trait ascriptions may be belief
evaluation, the main function and natural home of broad trait ascriptions
is in the evaluation of agents themselves and the quality of their motives
and efforts at inquiry. Also, when we are not in a position to evaluate
whether someone’s belief is knowledge but we do have some experience of
his or her belief-forming practices, we typically fall back on BTR
evaluations. ‘‘We are thus more willing to praise or blame believers for
BTR, since it says more about their status as cognitive agents’’ (Lepock
forthcoming a). We can see then that the two kinds of epistemic reliability
serve different basic explanatory functions. But it is also highly useful to
see that they are often intimately connected in epistemic evaluation of
agents and their beliefs. Broadly typed reliability may exculpate agents
from blame for beliefs formed in non-NTR ways. It can work the
opposite way as well: ‘‘Using non-BTR processes prevents agents from
receiving credit for their NTR beliefs’’ (Lepock forthcoming a).
NTR and BTR evaluations can come apart, as they do in several cases
Lepock discusses, with the effect of undermining an initial intuition we
may have had about the creditability to the agent for a particular true
belief. There is an important role of credit in knowledge ascriptions, and
creditability for true belief is most straightforward when an NTR success
is backed by a BTR.
r 2010 The Author

This way of framing the narrow-broad spectrum of trait ascription is

necessary if we are to make sense of the challenges of the two problems to
virtue theory. The flexibility of the spectrum or range account also helps
us to recognize and to cut across certain unmotivated debates, and to
recast the distinctions between propositional, doxastic, and personal
justification. Some writers treat virtue epistemology as little more than
an innovation within generic reliabilism. But ‘‘the trouble with treating
virtues as belief-forming processes is that it seems to rule out any
possibility of uniting the high-level and low-level virtues’’ (Lepock forth-
coming a). The focus on processes neglects the importance in the upper
range of cases with problem-solving strategies, and with the social and
communal nature of the norms of inquiry. The notion of a process’s
‘‘excellence’’ involves BTR; ‘‘achievement’’ and ‘‘reliability’’ are both
diachronic concepts, terms denoting agential success across time, and as
diachronic concepts they too involve BTR. Epistemic credit as credit for
‘‘getting it right’’ carries the externalist’s acknowledgment of the dia-
chronic and of personal justification through sound motivation and
efforts at inquiry, as an important source of epistemic value. This
recognition of the contribution to epistemic value made by stable
intellectual character traits is part of the insight of reliabilism, and
provides mixed theories with a substantial advantage over internalist
epistemologies like Conee and Feldman’s (see 2004). These accounts
vainly attempt to cash out epistemic justification exclusively in terms of
synchronic considerations of the agent and his or her present evidence,
ignoring altogether the quality of the inquiry leading the agent to have
just that total evidence to work with (for extended defense of the
epistemic value of diachronic epistemic rationality see Axtell forthcom-
ing, and Axtell and Olson forthcoming). The latter theories retain the
importance of personal justification but are committed to viewing
diachronic traits, including the intellectual virtues, as regulators of
inquiry, as strictly nonepistemic. This, as I argue elsewhere, denatures
the centrality of these virtues in explanations attributing epistemic credit
to the agent for the truth of his or her belief.6
Lepock thinks it is better to conceive of intellectual virtues as capacities
for metacognitive control of our actions of inquiry and methods and
strategies of problem solving. ‘‘Open-mindedness, intellectual courage,
and the like are not dispositions to form beliefs, though they are (speaking
loosely) dispositions . . . [to] engage in inquiry in certain ways’’ (2008a,
17). Most intellectual virtues have essential connections to capacities to
search in some particular manner, and capacities to know when that kind
of search is a good idea (Morton 2006). This way allows for the centrality
in epistemology of the analysis of doxastic justification and what
6
See Axtell 2009; see also Crisp 2010 on the centrality of diachronic traits in virtue
epistemology as well as virtue ethics.
r 2010 The Author

80 GUY AXTELL
Feldman calls synchronic epistemic rationality but also for what Hook-
way (2006) says contrasts with the doxastic paradigm, an inquiry-focused
epistemology. Even if, as Lepock holds, BTR isn’t as directly involved in
knowledge as it is in other important epistemic standings, such as
theoretical understanding, intellectual virtues like intellectual humility
and open-mindedness guard against certain biases and promote the
agent’s epistemic reliability in a variety of ways. They are praiseworthy
traits that can be seen as configuring causes of belief or of action. The
manner in which they are valued reflects diachronic goals of maintaining
a stable set of beliefs over time. But the argument is not that there is any
neat mapping of narrow-type ascription onto belief evaluation, or of
broad-type ascription onto agent evaluation. Nor is it the need to
prioritize one over the other for the assuaging of reductionist aspirations.
The proper way to distinguish and relate the different levels of generality
is in terms of the driving interests-in-explanation that philosophers have in
a particular case; and of course, the nature of the case itself partly
determines this although our own special purposes matter as well. Thus
the present account both draws from and supports virtue-based con-
textualism (see Sarah Wright 2010, Greco 2009, Thomas 2008, and Upton
2005).
Recognition of the different, even if overlapping functions of NTR and
BTR ascriptions (see table 1), together with the view that reflective
intellectual virtues should properly be conceived as capacities for meta-
cognitive control of inquiry, offers other potential benefits. Lepock argues
that his view is needed to make good sense of the commonsense ascription
of intellectual virtue to figures like Newton and Aristotle. For we do so
without presupposing that many of their particular scientific beliefs are in
fact true. But this ascription is rendered senseless if virtues are identified
strictly with reliably truth-conducive processes, as austere forms of
reliabilism appear committed to. Such persons are commonly regarded
as exemplars of scientific reasoning, despite the fact that the reliability,
power, and portability of their faculties are unexceptional compared to
those of educated people today.7
So by way of review of our distinction between levels, intellectual-
character-trait attributions vary across a narrow-broad spectrum. The
approach recommended here, as we have seen, begins by asking why we
have such different ways of ascribing intellectual traits to persons, and
what distinctive functions are served by ascriptions made at different
levels of generality. Knowledge is a collective good, and this kind of
‘‘genealogical’’ question prevents us from divorcing epistemological
7
Lepock 2007, 167; Riggs 2003, 210–13. Lepock argues that ‘‘what makes it possible to
explain our appraisals here is the fact that a virtuous trait can be valuable in part because of
its own power and portability, even though it does not typically have enough effect on belief-
formation to make one’s beliefs into knowledge’’ (2007, 167).
r 2010 The Author

TABLE 1. Lepock’s narrow-broad spectrum of intellectual-character-trait attributions
NTR: Narrowly typed reliability

—Low-level virtues (faculty virtues). Dispositions construed as genetically endowed
cognitive capacities.
—Best suited to evaluating the etiology of a single belief or narrow range of beliefs; tells us
nothing about an agent’s other beliefs.
—The value of low-level virtues is transmitted directly to their products and only indirectly
to the agents who have them.
BTR: Broadly typed reliability
—High-level virtues (reflective virtues). Best suited to explaining the agent’s intellectual
abilities and methods/strategies in a certain domain/area of inquiry.
—Best suited to holistic evaluation of agents, including the quality of their activities of
inquiry.
—The value of high-level virtues attaches directly to their possessor but only tenuously to
their products.
concerns from the realities of social interaction. Inquiry-focused virtue

epistemologists argue that the uses of trait concepts at different levels of
generality have quite distinctive functions, yet are also clearly connected
in many instances of epistemic evaluation. NTR and BTR can both be
seen as playing a justificatory role, or perhaps even as supplying different
important kinds of justification, doxastic and personal. We will later
return to develop more fully the suggestion that we might put such a
solution to the Generality Problem ‘‘to work’’ in epistemology by taking
it as an opportunity to utilize the resources of the narrow-broad spectrum
of attributions. But let’s now turn to another problem where narrow and
broad construal of character traits is central, the situationist challenge to
virtue ethics.
3. The Logic of Moral Trait Ascription and the Global Trait Problem
When we look at our commonsense ascriptions of moral traits to agents,
we find again that there is no single level of generality at which they are
directed. Doris describes the most important differences as those between
‘‘local’’ and ‘‘global’’ trait ascription, and he rejects ‘‘theory of character’’
insofar as it aims to be a theory with substantial empirical content (a
theory that is ‘‘realist’’ in contrast to antirealist, or fictionalist) about the
efficacy of global traits like moral virtues and vices. We will briefly look at
the local/global distinction as Doris employs it, and then at Hurka’s
(2001, 2009) equally suspect treatment of that same distinction.
Resurgent interest in ethical virtue is often dated to a half-century ago,
1958, when influential papers in ethics by G. E. M. Anscombe and
Philippa Foot were published. Anscombe’s ‘‘Modern Moral Philosophy’’
criticized a thin-focused ‘‘law conception of ethics’’ in British moral
r 2010 The Author

82 GUY AXTELL
philosophy, for its neglect of individual and social psychology. Anscombe

suggested that ethical theory would not advance unless and until
philosophers attended to the neglected psychology of character and
emotion. But the rallying cries of many self-described virtue ethicists,
‘‘moral psychology,’’ ‘‘practical wisdom,’’ and ‘‘the resources of thick
concepts,’’ all find them caught up in various ways with the globalist
assumptions that Doris, Harman, and other situationists find so problem-
atic in light of empirical psychology (see Goldie 2000). However dis-
connected virtue ethics may have been from empirical psychology in
previous decades, a general climate of trading armchair philosophy for
empirically informed theory, as well as Doris’s spirited polemic against
globalist virtue ethics, has today brought the global trait problem to
center stage.
Commonsense morality posits persons of good character who are not
easily swayed by circumstance. Aristotelian good character is supposed to
be an integrated association of robust traits and evaluatively consistent
personality structures. But, wrote one situationist author recently, ‘‘Ex-
periments show that efficacious traits are not global, and allegedly global
traits are not efficacious . . . character traits are either narrow and
efficacious or broad and inert. Either way, the conception of traits
favored by Aristotelian virtue ethics finds little empirical confirmation
in these studies’’ (Prinz 2009, 118).
Doris thinks we must restrict ethical trait ascriptions to the local, or
situational, like ‘‘dime-finding,’’ or ‘‘dropped-paper’’ compassionate
(Doris 1998; see Webber 2006, 2007, 2008 for criticism). We should
altogether eliminate or at least treat as very error prone ‘‘highly general
trait ascriptions like ‘honest’ or ‘compassionate’’’ (Doris 2002, 112).
Doris holds that one of the key upshots of the social psychological
research is a ‘‘‘fragmented’ conception of character, which countenances
a plethora of situation-specific ‘narrow’ or ‘local’ traits’’ that aren’t
unified with other traits (2005, 665). Doris’s use of the local/global trait
distinction reflects his contrast of his ‘‘fragmentary’’ account of person-
ality with the Aristotelians’ ‘‘evaluative consistency thesis,’’ and his
localism with their globalism. Doris’s fragmentary account holds ‘‘that
systematically observed behavior, rather than suggesting evaluatively
consistent structures, suggests instead fragmented personality struc-
tures—evaluatively inconsistent associations of large numbers of local
traits’’ (1998, 508). So the opposition as Doris constructs it is between
globalists, who posit character as ‘‘an integrated association of robust
traits and evaluatively consistent personality structures,’’ and localists or
situationists, who posit character only as ‘‘evaluatively inconsistent
associations of large numbers of local traits’’ (1998, 508).
Now there have been many responses made to Doris’s book, evoking a
variety of stances. Many philosophers do think the situationist studies
should be a wake-up call, to virtue theorists in particular. Ernest Sosa
r 2010 The Author

thus concedes that ‘‘it seems incumbent on virtue theory to grant that the
experiments do raise legitimate doubt as to how global and robust is
human practical wisdom, and how global and robust are its more specific
component virtues, such as kindness . . . honesty, [and] courage’’ (2008,
281). But Sosa and Webber both argue that holding Doris’s ‘‘fragmentary
account’’ isn’t mandatory. Some philosophic defenders of broadly based
moral trait ascriptions think that a theory of virtue can get by without as
much empirical content as others expect it to have. The most obvious
move in this regard is what Doris calls the ‘‘rarity argument’’ many virtue
ethicists give: it matters little to virtue ethics how rare full virtue is, so long
as it is not an impossible goal; even if full virtue or integrated character is
rare, we all have some capacity and opportunity to inculcate virtue in
ourselves. We still have reason to take character education as important,
too, and to resist the prescriptive implication situationists draw, that we
ditch it in favor of (presumably) enlightened ‘‘situation management.’’
Other defenders of broadly typed moral traits note that Doris takes
latitudinal or cross-situational studies to supply all the data needed to
reject a unitary account of character in favor of a fragmentary one,
whereas our best defense of our characterological intuitions is our
longitudinal acquaintance with the individuals around us—our knowing
them over an extended period of time. What allows us to attribute global
traits like honesty to an agent may be our long-term or longitudinal
acquaintance with that person, but for practical reasons there is very little
useful experimental study of this available. We can acknowledge that, as
one defender of global traits put it, there are ‘‘contingent difficulties that
often beset the ascription of traits, particularly their attribution to
strangers and loose acquaintances, but also that ‘‘there is no fundamental
attribution error that impugns [global] attribution itself, [though there is]
. . . a range of attribution difficulties that account for the ways in which
[global] attribution can go wrong’’ (Webber 2007, 90–91, 102–3). There
need be nothing wrong with ascribing stable and robust traits to people of
whom we have long acquaintance, though the empirical studies should
impact folk practices so as to improve them. The kind of character traits
of interest as one moves toward the global end of the local/global
spectrum include fundamental motives, desires, and goals, much as I
earlier said that the broad end of intellectual trait ascription brings in
diachronic as opposed to merely synchronic considerations.
Thomas Hurka’s (2001, 2006, 2009) treatment of the local and global
trait distinction also dichotomizes, turning it into fodder for debate
between philosophical analysis of virtue deemed mutually exclusive and
exhaustive. In ‘‘Virtuous Act, Virtuous Disposition,’’ Hurka’s initial
thesis is actually strongly analogous to Lepock’s: ‘‘Everyday moral
thought uses the concepts of virtue and vice at two different levels’’
(Hurka 2006, 69). At the global level, says Hurka, it applies these concepts
to persons or to stable character traits or dispositions. In contrast to these
r 2010 The Author

84 GUY AXTELL
standing traits, the local applications of concepts of virtue and vice are
applications to specific acts or mental states, such as occurrent desires or
feelings. Hurka writes that ‘‘the global and local uses of the virtue-
concepts are clearly connected, in that we expect virtuous persons to
perform and have, and virtuous traits to issue in, particular virtuous acts,
desires, and feelings. A philosophical account of virtue should explain this
connection, but there are two different ways of doing so. Each takes one
of the two uses to be primary and treats the other as derivative, but they
disagree about which is the primary use’’ (2006, 69).
The Dispositional View, says Hurka, ‘‘takes the global use to be
primary and identifies virtuous acts, desires, and feelings in part as ones
that issue from virtuous dispositions. Aristotle famously took this view.
In the NE he said that for an act to be virtuous it must meet some initial
conditions, including about its occurrent motivation, but must also
‘proceed from a firm and unchangeable character’; if it does not, it may
be such as a brave or generous person would perform, but is not itself
brave or generous. . . . The dispositional view . . . treats virtuous disposi-
tions as primary and defines virtuous occurrent states derivatively, as ones
that proceed from such dispositions’’ (2006, 70). The Occurrent-State
View, says Hurka, ‘‘takes the local use to be primary and identifies
virtuous dispositions as ones to perform virtuous acts and to have
virtuous desires and feelings. . . . It applies the virtue concepts
first to such states and then defines virtuous dispositions derivatively’’
(2006, 70).
Hurka identifies the local and global uses of virtue concepts with two
competing accounts of virtue, each supporting a different, albeit equally
reductive, systematic account of the nature and value of the virtues (see
also Hurka and Epstein 2009). He thus polarizes the debate strongly in at
least two ways: first, by accepting an occurrentist definition of virtue
against Slote’s and Zagzebski’s incompatible but equally contentious
dispositionalist definitions; and second, by prioritizing aims over rules
and virtues among the basic elements of moral theory. He describes the
dispositional view as ‘‘overwhelmingly dominant,’’ but argues against it
on a number of scores, including by claiming that the contemporary
commonsense understanding of virtue is actually the occurrent-state one:
‘‘When everyday moral thought applies the virtue-concepts, it is primarily
to occurrent states considered on their own’’ (2006, 70).
One example Hurka gives is of the soldier who is awarded a Medal of
Honor for throwing himself on a hand grenade to save others. The
members of a military committee are considering whether or not to give
the soldier a medal for bravery: ‘‘Would they say, ‘We know he threw
himself on a grenade despite knowing it would cost him his life and in
order to save the lives of his comrades. But we cannot give him a medal
for bravery because we do not know whether his act issued from a stable
disposition or was, on the contrary, out of character’?’’ ‘‘They would say
r 2010 The Author

no such thing,’’ Hurka judges, ‘‘and they would be obnoxious if they did’’
(2006, 72).
The folk certainly make global judgments about virtue, but they treat
those judgments ‘‘as derivative from local judgments about the virtuous-
ness of particular acts, desires, and feelings, and takes those states’
virtuousness to be independent of any tie to dispositions.’’ ‘‘Moreover,’’
Hurka goes on to contend against dispositionalists, ‘‘it is right to do so:
an act of helping another from a desire for his welfare is no less admirable
when out of character than when dispositionally based’’ (2006, 74).
In summary, this section has described two of the main uses of
narrowly and broadly typed moral trait ascriptions in the literature
today: Doris’s contrast between the local and the global construal of
traits, and the localist/globalist debate it engenders; and Hurka’s own
treatment of local and global uses of virtue concepts, and the occurrentist/
dispositionalist debate it engenders. I compared briefly the importance of
latitudinal studies and longitudinal acquaintance in defense of the
legitimate functions of broadly typed moral trait ascriptions. Hurka
constructs this either/or choice (between dispositionalism and occurent-
ism) much as Doris constructs that between globalists and localists, and
traits and situations. While I said little in direct criticism of either author,
directing attention to the rhetorical strategy of dichotomizing between the
two kinds of trait ascription sets off our own nonreductionist approach
by contrast. This sets us up, then, for a closer comparison between our
two problems regarding trait ascription, and for arguing that such
dichotomization is uncalled for and that our handling of the Global
Trait Problem would benefit from following the same sort of approach we
found Lepock bringing to the Generality Problem.
4. The Common Structure of the Two Problems

We have taken our two problems about trait ascription as serious
problems in ethics and epistemology. But an approach by way of putting
the narrow-broad spectrum to work rejects the primacy-granting reduc-
tive ‘‘higher-level accounts’’ offered by Hurka and by Zagzebski, in favor
of allowing our actual evaluative practices to be our focus and our guide.
As Nicholas Rescher notes, ‘‘The understanding/explanation orientation
is much less atomistic and more social than the certainty/justification
orientation’’ (2001, 237). An epistemology that does not make inquiry
central will likely also be unable to satisfactorily integrate the findings and
perspectives of social and collective epistemology. Why try to reduce one
kind of attribution to the other, when it is possible through a more flexible
model to accommodate both? The main reason why I have been attracted
by Lepock’s development of the narrow-broad spectrum, as his own
interpretation of Hookway’s distinction between faculty and reflective
virtues, is that it facilitates such a nonreductive account and undercuts
r 2010 The Author

86 GUY AXTELL
unmotivated debates such as those between localists and globalists,

occurrentists and dispositionalists. Narrowly and broadly typed traits
are thus seen as ascribed in response to different, even if overlapping,
explanatory interests, with neither being primary over or reducible to the
other. Our actual epistemic appraisals may not just be measures of NTR
or BTR; often added to these are other considerations, from anti-Gettier
conditions to internalist conditions, which additions make the quest for
reducibility that much more difficult. Mostly, the approach entreats us to
carefully distinguish different sorts of appraisal, and to use this informa-
tion to help determine the relevant process type or relevant level of
generality that responds to the explanatory interests we have in any
particular case. As a problem of practice, it is easier to be clear about the
different types of appraisal or evaluation we employ—for instance, even
reliabilist and internalist appraisal—when we treat them as concerned
with things addressed at different levels of generality.
The forms of virtue epistemology that are distinguished both from
virtue reliabilist and neo-Aristotelian forms by a strongly antireductionist
stance are those most closely associated with inquiry pragmatists, or those
that Hookway, citing Peirce, describes as supporting ‘‘epistemology-as-
inquiry’’ (Hookway 2006, 95). It is here also that I locate my own
approach.8
If we consider what virtue theories the global trait problem is most
serious for, they are globalist virtue ethical theories, including Stoic and
neo-Aristotelian versions of virtue theories, since these make the strongest
claims about the causal efficacy and philosophical importance of broadly
typed traits of character. The focus on habits of responsible and
successful inquiry among the pragmatist virtue theories helps them resist
the flaws of globalism, while addressing empirical challenges to folk
epistemological practices as well. If there is no single level of generality at
which trait ascriptions in either area aim, and especially if, as I’ve
suggested but not set out to prove here, narrowly and broadly typed
traits are interconnected in evaluation of agents and their actions, then we
need a nonreductionistic approach, which is what we are denied by
engagement in undermotivated debates like those between the situationist
localist and neo-Aristotelian globalist or again between the occurrentist
and dispositionalist analyses of moral virtue. If we want to facilitate more
constructive work at the intersections of ethics and epistemology, we need
to see that our ability to do so is improved by our adopting a more
deflationary attitude toward a number of contentious debates over the
8
This approach is akin to that suggested in Dewey’s account of reflective morality, in
which at whichever end we begin we find ourselves intellectually compelled to consider the
other end. Prioritizing or reductionistic definitions of virtue such as those we find the
occurrentist and dispositionalist employing are immediately suspect if in fact ‘‘we are not
dealing with two different things but with two poles of the same thing’’ (Dewey 1989, 7:173).
r 2010 The Author

conceptual primacy of the narrow or the broad, the local or the global,
the occurrent or the dispositional, the ‘‘thin’’ or the ‘‘thick’’ (on the latter,
see Hurka and Elstein forthcoming, Axtell and Carter 2008, Battaly 2008,
and Elgin 2008). With this deflationary attitude in place, I suggest we
place ourselves on a more constructive footing to put salience contextu-
alist approaches to the Global Trait Problem and the Generality Problem
to work for us.9 We might draw the further upshot that Zagzebski’s virtue
theory, sometimes called neo-Aristotelian or ‘‘pure’’ virtue theory on
account of some of its other commitments, comes out looking especially
problematic (see also 2003, 2006). This is not because it attempts a unified
account of virtue across ethics and epistemology, for that is something
that opposing inquiry-focused accounts can also endorse. It is acutely
problematic because it is structured in ways that leave it especially
challenged by each of the two specific problems we’ve examined.
Zagzebski’s view is:
1. especially challenged by the Generality Problem, because of the
robust (or ‘‘motive reliabilist’’ [Levin]) use of the ‘‘because of virtue’’
idea to entail truth and preclude Gettier and environmental forms of
epistemic luck; and
2. especially challenged by the Global Trait Problem (and directly
targeted by Doris) because of her neo-Aristotelian demand that the
sources of virtuous actions be ‘‘entrenched’’ moral virtues, and that
moral agents be motivationally self-sufficient.10
Of course, these are not faults of Zagzebski’s approach alone. Extant

theories of virtue do not do a very good job of unifying low- and high-
level virtues. Here as in science, unification and reductionism are not the
same thing, though a zealous reductive spirit wants to treat them that way
and limits itself by creating but another ‘‘great divide’’ in ethics, this time
9
This reply to Hurka, then, is not so much different from Robert Adams’s reply when he
writes of dispositional and occurrent uses of virtue concepts that ‘‘both types of conception
have their uses,’’ and that there is little harm with being pluralist about the sources of virtue,
and with operating with both, ‘‘provided we are clear about what we are doing’’ (Adams
2009, 124).
10
While Doris has to date restricted the situationist challenge to ethical theory, he at one
point in the book gives reason to think that it could be generalized to target at least some
versions of virtue epistemology. He writes: ‘‘[T]he contextual variability of cognitive
functioning may problematize globalist, highly general, accounts of intelligence. A wealth
of empirical work indicates that people experience remarkable difficulty ‘transferring’
cognitive skills across even closely related domains; they may perform well in one context
and poorly in other, seemingly very similar situations, rather like the case of moral behavior.
If the ‘contextualism’ about cognitive ability this empirical work inspires is right (see Ceci
1996), it would be a nuisance, not only for conceptions of practical reasoning emphasizing
reliable flexibility but also for recent ‘virtue epistemologies,’ that import globalist psycho-
logical theories from the ethics literature in an attempt to elucidate central epistemic notions
(e.g., Zagzebski 1996, esp. 178)’’ (Doris 2005, 670).
r 2010 The Author

88 GUY AXTELL
between concern with what we ought to ‘‘do’’ and concern with who we
want to ‘‘be.’’
The case for a unified virtue-theoretic account of normativity need not
and should not be cast, as Roger Crisp rightly warns, as ‘‘the question of
how one should live or what kind of person one should be rather than the
question of how one should act’’ (2010). What we can say is that the
broad end of the spectrum of trait ascriptions in both ethics and
epistemology is concerned with diachronically described traits, and with
more holistic evaluations of agency than with narrower concerns about
the rightness of a particular act or the justification of a particular belief.11
So a final implication I want to draw from our comparison of the two
problems is that the dichotomizing and reductive spirit of Doris’s and
Hurka’s approaches to moral trait ascription obscures the need to
recognize the interrelatedness of the narrowly and broadly-typed traits
in moral evaluation. Both authors, as we’ve seen (section 3), turn the
differences between narrowly and broadly typed trait ascriptions into
fodder for philosophical debates motivated only by prioritizing the one or
the other. Doris’s use of the local/global distinction threatens to turn it
into a false dichotomy wherever and whenever he uses it to insinuate an
inexorable choice between (realistic) fragmentary and (utopian) global
conceptions of personality; Hurka’s use of the distinction follows a
similar dichotomizing strategy, but in the service of a consequentialist
‘‘higher-level’’ reduction of dispositionalist to occurrentist moral theory.
The choice he presents as inexorable is between virtues treated and
defined atomistically and virtues treated and defined holistically—that
is, between a view that privileges ‘‘specific acts or mental states such as
occurrent desires of feelings,’’ and one that privileges ‘‘persons, stable
character traits or dispositions’’ (2006, 70). But from the present per-
spective I think we can see that the intended forced choice is wholly
factitious.
Hurka’s account, which acknowledges that systems of ethics proceed
from emphasizing one pole or the other, is refreshingly blunt. Most
philosophers try much more actively to ‘‘sink’’ the fact of such privileging
or selective emphasis. For instance, should internalists and externalists
about epistemic justification acknowledge that what motivates their
debate is merely their respective selecting of one meaning of justification
11
Gregory Pappas’s book John Dewey’s Ethics gives us another and still more
challenging response, in terms of which Dewey’s thought should not be assimilated to
that of self-described virtue ethicists, despite the criticisms they share of deontology and
consequentialism: ‘‘It has been assumed that the great divide in ethics is between act-centered
views, ethics of doing, and character-centered views, ethics of being; in other words, morality
should be conceived as a matter of doing good or being good. . . . John Dewey anticipated it
and evaluated its legitimacy. Dewey undermines the grounds for the divide issue, and he
proposes a way to move beyond the debates between character-centered and act-centered
ethics’’ (Pappas 2008, 129).
r 2010 The Author

from among several, that by their taking one instance of knowledge,

‘‘brute’’ or ‘‘reflective,’’ as paradigmatic they would necessarily be giving
up the claim that their account constitutes a ‘‘complete’’ account of
epistemic justification. If the situationist and the neo-Aristotelian virtue
ethicist acknowledge that they mean quite different things by behavior, or
the occurrentist and the dispositionalist that the facts pertaining to
occurrent states and to diachronic traits are potentially independent
sources of value, then the contrary systems of thought they generate
could still be resources for the active agent engaged in moral deliberation
and reflection, but the conflict between such systems couldn’t really be
expressed. But would that be a bad or a good thing?
To be sure, the systems or theoretical perspectives in question do fit
particular cases better or worse, and the Global Trait Problem isn’t
skirted merely by pointing out that appeal to dispositions serves an
indispensable role in ethics. Preserving the philosophical importance of
explanations involving the broadly typed character virtues may well
require distancing ourselves from certain aspects of what Doris targets
in his critique of globalism, especially the conception of the morally
virtuous agent as motivationally self-sufficient.
To move toward conclusion, it is useful to point out that up to a point
there are some quite strong analogies between Hurka’s and Lepock’s
understandings of the narrow-broad spectrum of trait ascriptions. What
is both shared and highly useful in Hurka’s and Lepock’s approaches may
be summarized this in a series of steps:
(a) there is no single level of generality at which trait ascriptions in
their respective subfields of philosophy are aimed;
(b) ascription varies across a narrow-broad spectrum;
(c) the uses of trait concepts at different levels are clearly connected;
(d) a philosophical account of virtue should explain this connec-
tion;
(e) there are different ways of doing this; and
(f) this often fuels debate between competing accounts of virtue
based on the primacy of the concepts at one end of the spectrum
or the other.
I find these points to be an excellent start for a logic of trait ascription that
can help us navigate the intersections of ethics and epistemology. There,
however, the approaches of Lepock and Hurka begin to diverge dramati-
cally, as Lepock goes on to argue that:
(g) both levels of trait ascription are often involved in epistemic
appraisal of agents and their beliefs;
(h) there is no general answer to the question of which end of the
spectrum is logically or conceptually prior to the other;
r 2010 The Author

90 GUY AXTELL
(i) we should therefore take nonreductionism as our default posi-

tion and avoid contentious debates motivated only by the
privileging of one end of the narrow-broad spectrum; and
(j) this allows us to put the narrow-broad distinction (and indeed
the Generality Problem and the Global Trait Problems) to work
for us in concrete ways by seeing how such potential barriers to
trait ascription get addressed in our evaluative practices, and in
particular cases.
It is these latter steps, missing in Hurka’s treatment of these same issues,

that I think aid us in navigating clear of the Scylla and Charybdis so
often claimed to shipwreck philosophical attempts to render character-
ological attributions both informative and philosophically consistent.
The epistemological contextualism that ensues from putting the narrow-
broad spectrum to work is just a demand for better awareness of the
different roles played by our own explanatory interests as we try to say
what method, field, and conditions should be employed when assessing
the epistemically relevant reliability of any particular belief. Salience
contextualism argues for the need to nonreductively balance and
utilize the resources of trait concepts that function at different levels of
generality.12 Can we, by following Lepock’s rather than Hurka’s example
here, apply this same approach to the Global Trait Problem? At this stage
I hope that readers will share the conviction that we can and should, but
also see that both Doris’s and Hurka’s approaches militate against it.
Although Hurka, in his manner of treating the Global Trait Problem,
appears to confer with our present approach in (a) to (f) he takes a quite
contrary stance with respect to (g) to (j). The dispositionalist (Zagzebski,
Slote) argues from the idea that it can never be right to sacrifice virtue to
the view that virtue is more valuable than other goods, even infinitely
more valuable. Hurka has ‘‘argued against this view, holding to the
contrary that virtue is always a lesser good’’ (2010). So Hurka’s stance
implicitly holds (gn) to (jn):
(g*) the levels of trait ascription cannot be independent sources of

value, but one or the other must be deemed primary;
(h*) this primacy of the local or occurrent end of the spectrum
explains the value of the other end;
(i*) we should therefore view this and similarly structured primacy
debates and the choices they entail as inevitable in systematic
theories in ethics; and
12
Indeed, without studying the shifts in explanatory frames that occur as we move from
trait ascriptions of one level to those of another, it would arguably be impossible to
determine when two explanations as occurrentists and dispositionalists typically present
them are consistent or inconsistent with or irrelevant to one another.
r 2010 The Author

(j*) we should reject the pluralism inherent in the spectrum

approach in favor of one or another reductivism—either that
of the occurrentist or that of the dispositionalist.
From what has been said, it should be clear that Hurka and Zagzebski, if
not epitomizing the occurrentist/dispositionalist debate, are at least the
most sharply opposed of the contributors over this issue in the present
collection, opposed over what elements of a moral theory—goods,
virtues, and duties—are conceptually primary or foundational. While I
have benefited from the richness of thought in both authors, I do want to
suggest that one upshot of our study is to cast doubt on the usefulness of
their respective ways of turning the distinction between dispositional and
occurrent uses of virtue terms into a clash of competing philosophical
systems.
This comes with some qualification. Hurka’s ‘‘higher-level account’’
(2010) provides a consequentialist model that is in an important sense less
reductionistic than other and better known consequentialist models. To
be sure, Hurka makes an insightful critique of virtue ethics. He says that a
moral philosophy should not make global dispositions a condition for the
value of occurrent attitudes, and I agree. He also says that we should not
let the importance accorded by virtue ethicists to thick concepts in recent
years be used to deride the importance of thin normativity and act
evaluation, and I agree as well. But these are not best accomplished by
taking sides on opposed Occurrentist and Dispositionalist systems of
ethics.13 They are instead best accomplished by respecting the quite
different functions that local and global trait ascriptions serve, and
acknowledging their interconnectedness in the explanatory and normative
interests that we have in the moral evaluation.
Department of Philosophy and Religious Studies

Box 6943
Radford University
Radford, VA 24142
USA
gsaxtell@radford.edu
13
Compare Appiah’s discussion of how virtue ethics became distorted when during the
1960s and 1970s it repurposed itself to that narrower conception of morality that is the
subject of mainstream moral philosophy. This is the shortcoming of virtue ethicists
themselves, of course, but also of Kantians and consequentialists who would reduce an
account of excellence to a cluster of duties or consequences, and procedures for fulfilling
them. Once this is the case, the eudaemonist approach is lost: ‘‘Virtue ethics rather loses its
point. And its way’’ (Appiah 2008, 63).
r 2010 The Author

92 GUY AXTELL
Acknowledgments
I thank John Greco, David Kaspar, Christopher Lepock, and Ken Lucey
for discussion of earlier versions of this essay, and Heather Battaly, Amy
Coplan, Roger Crisp, Thomas Hurka, and other ‘‘Virtue and Vice: Moral
and Epistemic’’ conference participants, June 26–27, 2008, for helpful
discussion. Heather, Christopher, and the editors of Metaphilosophy
helped immensely with editing suggestions.
References
Adams, Robert M. 2009. A Theory of Virtue: Excellence in Being for the
Good. Oxford: Oxford University Press.
Anscombe, G. E. M. 1958. ‘‘Modern Moral Philosophy.’’ Philosophy 33,
no. 124:1–19.
Appiah, Kwame Anthony. 2008. Experiments in Ethics. Cambridge,
Mass.: Harvard University Press.
Axtell, Guy. 2009. ‘‘Diachronic Rationality and Epistemic Value.’’
Unpublished manuscript under review.
Axtell, Guy. Forthcoming. From Internalist Evidentialism to Virtue
Responsibilism.’’ In Evidentialism and Its Discontents, edited by Trent
Dougherty. Oxford: Oxford University Press.
Axtell, Guy, and J. Adam Carter. 2008. ‘‘Just the Right Thickness: A
Defense of Second-Wave Virtue Epistemologies.’’ Philosophical
Papers issue entitled ‘‘Epistemology Through Thick and Thin,’’ 37,
no. 3:413–34.
Axtell, Guy, and Phillip Olson. Forthcoming. ‘‘Three Independent
Factors in Epistemology.’’ Contemporary Pragmatism.
Battaly, Heather. 2008. ‘‘Metaethics Meets Virtue Epistemology: Salva-
ging Disagreement About the Epistemically Thick.’’ Philosophical
Papers 37, 3:435–54.
Becker, Kelly. 2008. ‘‘Epistemic Luck and the Generality Problem.’’
Philosophical Papers 139:353–66.
Beebe, James. 2004. ‘‘The Generality Problem, Statistical Relevance and
the Tri-Level Hypothesis.’’ Nous 38, 1:177–95.
Conee, Earl, and Richard Feldman. 2004. Evidentialism: Essays in
Epistemology. Oxford: Oxford University Press.
Crisp, Roger. 2010. ‘‘Virtue Ethics and Virtue Epistemology.’’ Included
in this collection.
Dewey, John. 1989. The Later Works. Carbondale: Southern Illinois
University Press.
Doris, John. 1998. ‘‘Persons, Situations, and Virtue Ethics.’’ Nous 32,
no. 4:504–30.
———. 2002. Lack of Character: Personality and Moral Behavior.
r 2010 The Author

———. 2005. ‘‘Replies: Evidence and Sensibility.’’ Philosophy and

Phenomenological Research 71, no. 3:656–77.
Elgin, Catherine. 2008. ‘‘Trustworthiness.’’ Philosophical Papers 37,
no. 3:371–87.
Foot, Philippa. 1958. ‘‘Moral Arguments.’’ Mind 67:502–13.
Fricker, Miranda. 2007. Epistemic Injustice: Power and the Ethics of
Knowing. Oxford: Oxford University Press.
———. 2008. ‘‘Scepticism and the Genealogy of Knowledge: Situating
Epistemology in Time.’’ Philosophical Papers 37, no. 1:27–50.
Greco, John, ed. 2004. Sosa and His Critics. Oxford: Blackwell.
———. 2009. Achieving Knowledge. Cambridge: Cambridge University Press.
Goldie, Peter. 2000. The Emotions. Oxford: Clarendon Press.
Harman, Gilbert. 1999. ‘‘Moral Philosophy Meets Social Psychology:
Virtue Ethics and the Fundamental Attribution Error.’’ Proceedings of
the Aristotelian Society 99:315–31.
———. 2000. ‘‘The Nonexistence of Character Traits.’’ Proceedings of the
Aristotelian Society 100:223–26.
Hetherington, Stephen (ed.). 2006. Epistemology Futures. Oxford: Oxford
University Press.
Hookway, Christopher. 2006. ‘‘Epistemology and Inquiry: The Primacy
of Practice.’’ In Heatherington 2006, 95–110.
Hurka, Thomas. 2001. Virtue, Vice, and Value. Oxford: Oxford
University Press.
———. 2006. ‘‘Virtuous Act, Virtuous Dispositions.’’ Analysis 66, 1:69–
76.
———. 2010. ‘‘Right Act, Virtuous Motive.’’ Included in this collection.
Hurka, Thomas, and Daniel Elstein. Forthcoming. ‘‘From Thick to Thin:
Two Easy Reduction Plans.’’ Canadian Journal of Philosophy.
Kappel, Klemens. 2006. ‘‘A Diagnosis and Resolution to the Generality
Problem.’’ Philosophical Studies 127:525–60.
Kofman, Sarah. 1988. ‘‘Beyond Aporia?’’ In Post-Structuralist Classics,
edited by Andrew Benjamin, 77–88. London: Routledge.
Lepock, Christopher. 2006. ‘‘Adaptability and Perspective.’’ Philosophi-
cal Studies 129, no. 2:377–91.
———. 2007. ‘‘Metacognition and Intellectual Virtue.’’ Ph.D. dissertation,
University of Alberta.
———. 2008. ‘‘Metacognition and Epistemic Virtue,’’ available at http://
www.ualberta.ca/ clepock/lepock%20-%20metacognition%20& %20
epistemic%20virtue.doc (last accessed 22 September 2009).
———. Forthcoming a. ‘‘How to Make the Generality Problem Work for
You.’’ Acta Analytica.
———. Forthcoming b. ‘‘Unifying the Intellectual Virtues.’’ Philosophy
and Phenomenological Research.
Levin, Michael. 2004. ‘‘Virtue Epistemology: No New Cures.’’ Philosophy
and Phenomenological Research 69, no. 2:397–410.
r 2010 The Author

94 GUY AXTELL
Lockie, Robert. 2008. ‘‘Problems for Virtue Theories in Epistemology.’’

Philosophical Studies 138, no. 2:169–91.
Morton, Adam. ‘‘Knowing What to Think About: When Epistemology
Meets the Theory of Choice.’’ In Heatherington 2006, 111–28.
Pappas, Gregory. 2008. John Dewey’s Ethics. 2008. Bloomington: Indiana
University Press.
Prinz, Jesse. 2009. ‘‘The Normativity Challenge: Why Empirically Real
Traits Won’t Save Virtue Ethics.’’ In Virtue Ethics, Character, and
Moral Psychology, edited by Candace L. Upton, 117–44. Oxford:
Oxford University Press.
Putnam, Hilary. 2002. The Collapse of the Fact/Value Dichotomy.
Cambridge, Mass.: Harvard University Press.
Rescher, Nicholas. 2001. Epistemology. Oxford: Oxford University Press.
Sosa, Ernest. 2008. ‘‘Situations Against Virtues: The Situationist Attack
on Virtue Theory.’’ In Philosophy of the Social Sciences: Philosophical
Theory and Scientific Practice, edited by C. Mantzavinos, 274–91.
Thomas, Alan. 2008. ‘‘The Genealogy of Epistemic Virtue Concepts.’’
Philosophical Papers 37, no. 3:345–69.
Upton, Candace. 2005. ‘‘A Contextual Account of Character Traits.’’
Philosophical Studies 122:133–51.
Webber, Jonathan. 2006a. ‘‘Character, Consistency, and Classification.’’
Mind 115, no. 459:651–58.
———. 2006b. ‘‘Virtue, Character and Situation.’’ Journal of Moral
Philosophy 3, no. 2:193–213.
———. 2007. ‘‘Character, Global and Local.’’ Utilitas 19, no. 4:430–34.
Wright, Sarah. 2010. ‘‘Virtues, Social Roles, and Contextualism.’’ Meta-
philosophy 41, nos. 1–2:95–114.
Zagzebski, Linda. 1996. Virtues of the Mind. Cambridge: Cambridge
University Press.
———. 2003. ‘‘Epistemic Value and the Primacy of What We Care
About.’’ Philosophical Studies 33, no. 3:353–77.
———. 2006. ‘‘Ideal Agents and Ideal Observers in Epistemology.’’ In
Heatherington 2006, 131–48.
r 2010 The Author

Forthcoming 2007, in Oxford Handbook of Skepticism, John Greco (ed.) OUP.
All rights reserved.
Virtue Theoretic Responses to Skepticism
Guy Axtell
1. Introduction
Virtue theory concerns philosophical evaluation of human agents in their interaction with the
world. Being embedded in a world, both natural and social, the interactions of human agents give
rise to ethical and to epistemic contexts, which in turn become the primary locus for the
descriptive, explanatory and normative tasks with which philosophy concerns itself. The central
focus in virtue theory on understanding agents and the habits and dispositions through which
their interactions in the world unfold allows for the acknowledgement of analogies between
epistemic and ethical evaluation, but for the acknowledgement of disanalogies as well.
Virtue theory has been applied to numerous domains of philosophic study, and its roots
arguably go back to some of the earliest Greek discussion of the motivations to philosophize. Its
resurgence over the past four decades has taken place first in the subfield of ethics, and since the
1980‘s in the subfield of epistemology as well.1 Questions about how to proceed in the study of
core epistemic concepts like justification, knowledge, and understanding are ones that
contemporary virtue epistemologists quite often engage. Skeptical arguments are another, and
indeed are closely caught up in discussion of the former concepts and the problems that surround
them. Even where no ‗true skeptics‘ are present to support them, skeptical arguments, whether
those of ―global‖ (radical) or ―local‖ (domain-specific) import, help to uncover the depth in a
conception of epistemic agency, and allow us to reflect more carefully upon our human
epistemic condition. As John Greco puts it, the study of skeptical arguments ―drives positive
epistemology.‖
This paper focuses on the responses that proponents of VE make to radical skepticism,
and particularly to two related forms of it, Pyrrhonian skepticism and the ―underdetermination-
based‖ argument, both of which are receiving widening attention in recent debate. Section 2 of
the paper briefly articulates these two skeptical arguments and their inter-relationship, while
section 3 explains the close connection between a virtue theoretic and a neo-Moorean response to
them. As I cannot fully canvas the growing field of VE, the focus will primarily be on leading
figures such as Ernest Sosa, who develops ―virtue perspectivism‖ in a series of papers and in his
2004 John Locke Lectures, and John Greco, who develops ―agent reliabilism‖ in Putting Skeptics
in Their Place (2000) and recent papers. In Sections 4-5 I advance my arguments for improving
the prospects of virtue-theoretic responses, sketching a particular version of VE that seeks to
recast somewhat how we understand the ―externalist turn in epistemology,‖ thereby suggesting
ways of improving the adequacy of philosophical responses to skepticism.
2. Pyrrhonian Skepticism and the Underdetermination Argument
One type of skepticism that current debate takes especially seriously is expressed in
underdetermination-based arguments. According to such arguments, one‘s total evidence, at
least as considered internalistically from a first-person perspective, underdetermines one‘s
judgment in favor of common-sense realism over alternative ‗hypotheses‘ that depict us as
victims of systematic deception. Here is how Pritchard (2005) more formally articulates for study
UA, the ―template underdetermination-based skeptical argument‖:
2
(U1) If my evidence does not favour my belief in everyday propositions over the known to be
incompatible skeptical hypotheses, then I am not internalistically justified in believing
everyday propositions.
(U2) My evidence does not favour my belief in everyday propositions over the known to be
incompatible skeptical hypotheses.
(UC) I am not internalistically justified in believing everyday propositions (and thus I lack
knowledge of everyday propositions). (2005, 205)
Pritchard takes the underdetermination argument to reflect especially well the
motivations behind contemporary neo-Pyrrhonism (see also Sinnott-Armstrong (ed.), 2004). He
rightly points out that (the explicit appeal to what is ―internalistically justified‖ not withstanding)
we can find versions of this argument among the ancients as well. Indeed I would connect the
underdetermination principle2 with the principle of the equipollence (roughly, ‗justificational
equivalence‘) of theoretical suppositions or ―judgments‖ in the thought of the second-century
A.D. skeptic Sextus Empiricus. In Outlines of Pyrrhonism, his most basic characterization of
skepticism is this:
Skepticism is an ability, or mental attitude, which opposes appearances to judgments in
any way whatsoever, with the result that, owing to the equipollence of the objects and
reasons thus opposed, we are brought firstly to a state of mental suspense and next to a
state of ‗unperturbedness‘ or quietude [ataraxia] (1, 8).
In Sextus and Agrippa‘s Pyrrhonism, the attempt of anti-skeptical philosophers to escape
the equipollence of theoretical judgments is found to land them upon one or another point of a
trilemma, where the philosophic reconstruction of one‘s ability to know (or alternatively, one‘s
3
prerogative to claim rational justification for one‘s belief), is a) viciously circular, b) endlessly
regressive, or c) ultimately arbitrary. ‗Agrippa‘s Trilemma‘ remains a matter of deep concern in
contemporary epistemology. As Pritchard points out in Epistemic Luck (2005) it centrally
supports the view ―that any claim to know can be called into question via the skeptical
techniques the Pyrrhonian skeptics have identified, and [that] this highlights the ultimately
‗brute‘ nature of our epistemic position‖ (220).
Another reason why Pyrrhonian skepticism and the underdetermination argument are
chosen as the focus of our examination is that they allow us to engage the issue of how ―the
externalist turn in epistemology‖ over the past several decades impacts the prospects of the anti-
skeptic‘s path in philosophy. It is clear that the externalist turn and with it, the rejection of access
internalism, has recast the role of positive epistemologists in dialogue with skeptical challengers.
But do externalist responses merely change the topic and avoid the real challenge of skepticism,
as its critics allege? Or are they as their proponents claim, highly advantageous by allowing us to
see how arguments like UA can come can appear insurmountable when they are not? These
questions are pertinent here because, while it would be an overstatement to say that there is a
single distinctive virtue theoretical approach to skepticism, the best-known proponents of virtue
epistemology clearly take advantage of externalist responses to skepticism, believing that it
provides strong resources both for diagnosing the motivations to skepticism, and for responding
to the skeptic‘s strongest arguments. In the next section we can canvas some of their views.
3. Nature and Reason in the Common-Sense Tradition
Philosophers as diverse as Pascal and Hume have seen the primary tension inclining us to radical
skepticism as that between ―nature‖ and ―reason.‖ ―Who will unravel this tangle?‖ asks Pascal.
4
―Nature confutes the skeptics, and reason confutes the dogmatists […such that a person] can
neither avoid these two sects, nor hold fast to either one of them!‖3 Nature inclines us to
common-sense beliefs (such as the existence of an external world, of other minds, of causal
regularities in nature, etc.) while reason inclines us to take seriously skeptical arguments that
would impugn our capacity for knowledge even of these things. The underdetermination
argument stated in Section 2 expresses one key form of the skeptic‘s worry. Doesn‘t our apparent
inability to eliminate known-to-be incompatible radical skeptical hypotheses by reflectively good
reasons demonstrate at least that our human condition is lacking in qualities that are
epistemically desirable? Doesn‘t it, further, impugn our ability to know and our epistemic
responsibility in attributing justified belief to ourselves and others?
G. E. Moore‘s ―common-sense realist‖ reply is to defend our natural confidence in our
everyday knowledge and our ability to claim it for ourselves. But critics allege that his reply is
question-begging, and therefore hangs upon the circularity horn of the above mentioned skeptical
trilemma. Contemporary ―neo-Moorean‖ responses argue that Moore‘s response is actually quite
illuminating, but only after it gets a retrofit to reflect a reliabilist stream in modern
epistemological thought stretching back to Thomas Reid. Sosa (2000b) and Greco (2002a) both
discuss how Reid, too, is part of the commonsense tradition, and how certain aspects of his
reliabilist orientation are subtly suggested in Moore‘s better known ―defense of common sense‖
(1925).4
For the virtue epistemologists we are considering, developing a neo-Moorean argument
with the kind of naturalized conception of reason that can take advantage of the resources of
reliabilist externalism would be quite advantageous. Hence Greco finds it illuminating to study
the substantive and methodological continuities between Reid and Moore in his ―How to Reid
5
Moore.‖ Greco argues that neither is ‗simply insisting‘ that we know what the skeptic denies, or
whether each is in his own way an astute critic of skeptical arguments. Rather, both authors held
that the evidence of sense is no less reasonable than that of demonstration, and for Reid at least,
as for contemporary reliabilists and other epistemic externalists, introspective consciousness,
perception, memory, testimony, and inductive reasoning are all possible sources of knowledge in
addition to deductive or demonstrative reasoning.
In ―How to Defeat Opposition to Moore,‖5 Sosa‘s articulation of a key area of
contemporary debate focuses upon the skeptical ―argument from ignorance,‖ hereafter AI. This
begins with two definitions:
h I am a handless brain in a vat being fed experiences as if I were normally
embodied and situated.
o I now have hands.
Then follows Sosa‘s version of the argument:
1. I don‘t know that not-h
2. If I don‘t know that not-h, then I don‘t know that o.
So, C. I don‘t know that o
Sosa characterizes the three main positions that have been adopted on AI:
Skeptic: 1, 2; therefore C
Nozick et al.: 1, Not C; therefore Not 2
6
Moore: 2, Not C; therefore Not 1
The skeptic holds the Moorean escape from skeptical conclusion C to be viciously
circular, but the neo-Moorean responds that the logic of the argument cuts two directions, such
that if I do know that I have hands, then I do know that skeptical hypotheses inconsistent with
this knowledge are false, even if I haven‘t directly considered them all.6
Sosa‘s basic description of knowledge is ―apt performance‖ in the way of belief. For a
belief to be apt, as Sosa puts it in his Locke Lectures, A Virtue Epistemology, is for it to be
―correct in a way creditable to the believer, as determined by how salient is the believer‘s
competence in the explanation of his being right‖ (2007, 60; see also his 2002a and 2002b). On
Sosa‘s view, epistemic competence is socially as well as individually seated, and virtue-theoretic
terms better describe this competence than other approaches. Whether considering a correct
belief due to intellectual virtue or a right action due to practical virtue, the competence and
performance of agents are central to our explanations of virtuous success, and a virtuous
performance will involve both the agent‘s constitution and situation.
Since the aptness of the agent‘s belief entails that its correctness is attributable to a
competence exercised in appropriate conditions, the concept of aptness is an externalist one, and
the condition an externalist condition. Yet Sosa accommodates certain intuitions traditionally
associated with epistemic internalism by retaining the importance of ―reflective coherence‖ in
the individual, and a strong sense of human knowledge even at its lowest rungs as an
achievement. These concerns with the lasting significance of reflective coherence are developed
in the fuller account he describes as ―virtue perspectivism,‖ which requires that agents achieve a
7
degree of epistemic ―ascent‖: ―reflective knowledge goes beyond animal knowledge, and
requires also an apt meta-apprehension that the object-level perceptual belief is apt‖ (2007, 81).
It is important to point out at this juncture that none of the authors mentioned here sees a
need to ‗go internalist‘ in order to acknowledge and respect ―the understanding and coherence
dear to intellectuals.‖7 Sosa, Greco, Zagzebski, Riggs, Axtell and others each adopt what has
been called a ―compatibilist‖ view of the relationship between our internalist and externalist
interests in explanation. They each articulate a distinction between personal and objective
justification that they take to be miscast as a distinction between internalism and externalism
conceived, as the typically are, as mutually exclusive and exhaustive accounts of epistemic
justification. Epistemic compatibilism and incompatibilism will be more fully discussed later,
but Greco‘s (2005) description of VE as ―mixed theory‖ is one expression of it, where he writes,
―The main idea is that an adequate account of knowledge ought to contain both a responsibility
condition and a reliability condition. Moreover, a virtue account can explain how the two are tied
together. In cases of knowledge, objective reliability is grounded in epistemically responsible
action.‖
Still, each has distinctive views in this area, and Greco (2004b) worries that Sosa‘s meta-
requirement of epistemic perspective makes broader concessions to internalism than is necessary,
perhaps hampering rather than improving VE‘s anti-skeptical force.8 The concern with an
epistemically relevant distinction between reflective and animal knowledge, and his stratified or
‗two tiered‘ account of justification seems to be a recurring theme in contributors including
Greco to the watershed Ernest Sosa and his Critics volume (Greco, ed. 2004a; see Kornblith, for
example). But Sosa‘s ―Replies‖ there, and the 2004 Locke Lectures, A Virtue Epistemology
engage these concerns quite directly.
8
In Putting Skeptics in Their Place Greco examines a wide range of skeptical problems,
and argues that responding to them drives us to a form of reliabilist externalism centered on the
cognitive abilities of epistemic agents (compare also Audi 2004). A key thesis in Greco‘s ―agent
reliabilism‖ is that the core ―relevant alternatives intuition‖ and this form of VE are mutually-
supportive. The relevant alternatives intuition as Greco describes it is the common-sense
suggestion that modally far-off error possibilities do not present relevant philosophical
challenges to our ability to know most of the things we think we know. Philosophers can remain
true to this intuition and harness its anti-skeptical import by relating it to the settled dispositions
and competencies through which agents strive for truth and effective agency in the world in
which they find themselves. If the kinds of cognitive dispositions a person must manifest in order
for her to meet a normative requirement of reliability are sustained by her thinking
conscientiously, then this latter concept is important as well and shows us that issues of
motivation and habituation remain important factors in any sound philosophic understanding of
our epistemic agency.9
Greco thinks that agent reliabilism provides some unique resources for redressing the
faulty assumption that for agents to have knowledge they must be able to discriminate the truth
of their beliefs from every alternative skeptical scenario regardless of its modal distance from the
actual world. Firstly, we need to understand relevant possibility in terms of the ―possible worlds‖
semantics of modal logic: A possibility is relevant if it is true in some close-by possible world,
and closeness is to be understood in terms of overall world similarity (2000, 206). Secondly, it is
virtue theory which best captures our reasoning about which possibilities are relevant: the very
concept of knowledge, and therefore of the relevant possibilities that must be excluded for its
possession, involves reference to cognitive abilities and dispositions:
9
In the language of possible worlds, someone has an ability to achieve some result
under relevant conditions only if the person is very likely to achieve that result across
close possible worlds. But if knowledge essentially involves having cognitive abilities,
and if abilities are dispositions to achieve results across close possible worlds, then this
explains why possibilities are relevant only when they are true in some close possible
world. Specifically, only such possibilities as these can undermine one‘s cognitive
abilities. In an environment where deception by demons is actual or probable, I lack the
ability to reliably form true beliefs and avoid false beliefs. But if no such demons exist
in this world or similar ones, they do not affect my cognitive faculties and habits ( 207).
4. Three Competing Anti-Skeptical Strategies
We have focused primarily upon the work of Sosa and Greco because of the direct contributions
each has made to a philosophically adequate response to radical skepticism. Turning in this
section to points about VE made by its critics, we will describe and engage Sven Bernecker‘s
objections informed by a kind of austere externalism that rejects epistemic compatibilism, and
Richard Foley‘s objections from an internalist perspective on justification that result in his
proposal for ―a trial separation between the theory of knowledge and the theory of justified
belief.‖ This serves my later purposes, since I want first to defend epistemic compatibilism in
this general way before going on to develop philosophical support for its specifically virtue-
theoretic versions.
A thorough-going externalist, Bernecker considers and rejects Sosa‘s and Greco‘s
epistemologies as forms of what he characterizes more generally as epistemic compatibilism.
This is a useful term but defined only vaguely by the claim that it is possible to ―combine
10
satisfactorily internalist and externalist features in a single theory‖ (2006, 81). The target
indicated by his critique is quite wide, but Bernecker‘s ―Prospects of Epistemic Compatibilism‖
provides a useful taxonomy of different arguments in the literature supporting distinct versions
of compatibilism. Sosa‘s and Greco‘s accounts are singled out for extended criticism (the latter
in Bernecker 2007) for the reason that the virtue epistemologists offer some of the best-sustained
defenses in contemporary epistemology of views associated with epistemic compatibilism.
Of course, nobody could hold that what needs to be reconciled are internalism and
externalism as understood in the usual stipulative definitions of them. Internalism about
justification is standardly defined for introductory purposes as the thesis that all of the elements
that justify a belief must be accessible to the mind upon reflection, with externalism being
defined as the negation of that thesis. Notice that this style of definition excludes a middle and
leaves the terrain of theories of justification asymmetrically shaped: Internalism is a positive
thesis of specified meaning, while externalism, by its purely negative character, is left to a wide
parcel that might admit of a plurality of positions once developed as positive epistemologies.
Externalism invites ‗mixed‘ theory but does not necessitate it: Compatibilists are those
who accept the invitation, while ‗incompatibilists‘ is what we can call those (at either end of the
scale running from pure internalism to pure externalism) who spurn the invitation to mixed
theory. Hence Bernecker‘s incompatibilist stance is not simply the rejection of internalism,
something which is true of all forms of externalism by the definitions he employs. His stance,
rather, is that externalists should reject any concerns about epistemic evaluation arising from the
first-person perspective. Contrary to the thrust of VE, Bernecker‘s proposal is that we give up
efforts at reconciliation and embrace instead a conception of naturalized epistemology on which
only third-personal concerns play a central explanatory role.
11
Besides being able to distinguish different kinds of compatibilism, we should also want a
taxonomy that acknowledges as ―incompatibilist‖ such eliminative forms of externalism as
Bernecker maintains. In order to get a proper handle on the range of positions in the debate, I
propose adopting the broadest taxonomy possible, which we might do by adapting that familiar
trichotomy between theorists who view themselves as ―Enemies,‖ ―Strangers,‖ or ―Partners‖ in
any given debate. More formally, we‘ll describe these as the ―Conflict,‖ ―Independence,‖ and
―Integration‖ models of the interests in explanation engaged by the internalist/externalist debate.
This three-fold taxonomy allows us to see three quite distinct models of the relationship of
internalist and externalist interests in explanation, each well-represented in present-day debates
in analytic epistemology. Moreover, when we extend this taxonomy to inform us about divergent
forms of VE specifically, we find a confirmation that few if any existing forms of VE adopt the
Conflict model, but instead support one or another form of epistemic compatibilism.10
Finally, we may further explore differences within the ―compatibilist‖ camp. Here
positions can be usefully divided between those adopting the Integration model and those
adopting the Independence model. This choice between the two models supporting epistemic
compatibilism overlaps with a fairly clear distinction between what I will term ―Strong‖ and
―Weak‖ forms of VE. First, let‘s refer to any analysis of knowledge that defends an areteic (i.e.,
virtue theoretic) condition on knowledge as ―Strong VE‖. This term is broad enough to include
Sosa, Greco and Zagzebski as proponents, despite their outstanding differences. The Integrative
model suggests that a correct account of knowledge is one that finds epistemic reliability and
epistemic responsibility not as sources of antithetical third and first-personal ‗logics,‘ but as
normative concepts mutually presupposed in a proper understanding of self-reflective agents like
us. Second, there are today quite a number of authors who can be described as virtue
12
epistemologists because they acknowledge an epistemically-central place for the study of the
intellectual virtues, yet see its place lying entirely outside the analysis of knowledge proper. 11
These are, on our taxonomy, proponents of ―Weak VE,‖ and they have sometimes been the most
explicit critics of the preoccupation with analysis of knowledge and justification in epistemology
and in Strong VE in particular. When used to illuminate differences within the family of virtue-
theoretic epistemologies, our taxonomy clarifies the point that proponents of Strong VE typically
maintain Integrationist models for epistemic compatibilism, while proponents of Weak VE
typically subscribe to Independence models.12
In summary thus far, our proposed taxonomy allows for the fact that Independence and
Integration models are often employed by epistemologists who are not proponents of VE in
either the strong or weak sense; but it also identifies under-noticed sources of tension between
epistemologists, and a veiled debate with quite direct implications for how best to respond to
radical skepticism. As ―models‖ rather than ―theories,‖ Conflict, Independence and Integration
represent in our taxonomy three different discursive strategies (Table 1), and the choice between
them demands that we think closely about which one best serves anti-skeptical philosophy.
13
Table 1: Three Competing Anti-Skeptical Discursive Strategies
Metaphor Model Description
Enemies Conflict Standard access internalism as excluding externalist
(or incompatibilism) elements; also ‗pure‘ or ‗eliminative‘ externalism as
in Bernecker‘s stance against all forms of
compatibilism.
Strangers Independence Foley‘s compromising ―separation proposal‖ which
(or weak compatibilism) severs the link between knowledge and justification,
while maintaining that internalism remains correct
when understood as something like a theory of
personal justification tied to a general theory of
rationality; includes ―Weak‖ VE where belief out of
intellectual virtue is epistemically desirable, but isn‘t
studied for any connection to knowledge-possession.
Partners Integration ‗Mixed theories‘ of justification that are externalist
(or strong compatibilism) in character but maintain requirements for personal
justification/proper motivation as well as for agent
reliability; includes ―Strong VE‖ where the
14
requirement of successful belief out of intellectual
virtue carries both kinds of demands.
We can now utilize this taxonomy to state a stronger thesis: that as an anti-skeptical
strategy, adoption of the Conflict model constitutes a dialectical mis-step. One reason for this
that I have already suggested but will continue to develop is that a neo-Moorean argument well
serves the interests of an externalist response to skepticism, yet the neo-Moorean approach
thrives in an environment of epistemic compatibilism while withering apart from it. More
specifically, when starting from incompatibilist assumptions, externalists are tempted to go
beyond acknowledging instances of unreflective ‗brute‘ knowledge, to supposing that most if not
all human knowledge is merely of this nature. Here the concerns shared predominantly but not
exclusively by internalists and skeptics, concerns about our cognitive responsibility in the human
modes of inquiry we conduct and the evidences we use when we provide reflective reasons for
our beliefs, appears lost or simply given up. The ―brute nature,‖ of the human epistemic
condition is conceded as we learn to leave first-personal perspectives out of the field of the
theory of knowledge. Perhaps reflective of such an austere form of externalism, Bernecker
explicitly rejects Sosa‘s concern with ―so-called reflective knowledge,‖ denying that it amounts
to a central concern of epistemology as he defines it, and countering it with the suggestion that
unreflective ‗brute‘ knowledge is simply paradigmatic of what we must take knowledge to be.
According to our own compatibilist approach, however, contemporary neo-Pyrrhonists such as
Barry Stroud (2004a, 2004b) are actually correct to see the kind of externalist response to
15
skepticism afforded by ‗unmixed‘ or ‗eliminative‘ forms of externalism, as philosophically
unsatisfying.13
Having considered and briefly responded to eliminative externalist objections to VE, let
us briefly look at the problem of the stability of epistemic compatibilism in light of certain
objections stemming from an internalist conception of justification. Externalists and internalists
are in agreement that questions about agent responsibility in inquiry are distinct from questions
about the agent‘s internal access to reasons. Alvin Goldman, an externalist, rightly points out that
access internalism is quite compatible with the ‗epistemic sloth‘ of an agent who is subjectively
justified only because she shirks her responsibility to carefully investigate or weigh potential
counter-evidence to her belief. Lawrence Bonjour, an internalist/coherentist, recently conceded
that despite having earlier conflated these issues, ―being epistemically responsible‖ is neither
necessary nor sufficient for ―internalist justification‖ as he continues to use that term (2003,
176). This broad agreement suggests that compatibilists can respond to Bonjour‘s internalism by
drawing more fully on the distinction between a) VE‘s interest in concerns with epistemic
responsibility in personal justification, and b) the strong demand that Bonjour allows the skeptic
to make, that there be available an ―internalist justification‖ with which to respond to the
Underdetermination Argument.14
Another related criticism of Strong VE coming from an internalist perspective is captured
by Foley‘s proposal for a ―trial separation between the theory of knowledge and the theory of
justified belief.‖ Foley claims that externalism and internalism ―need not be competitors at all.‖
He finds their conflict to be due to a major false assumption they share, ―that the properties
which make a belief justified are by definition such that when a true belief has those properties, it
is a good candidate to be an instance of knowledge‖ (2005, 314). It was this assumption of
16
logical connection between knowing and having justified belief, Foley charges, that prompted
many externalists who until fairly recently were only reacting against justification-driven
accounts of knowledge, to seek to reconceive justification externalistically as well.
Foley does indeed provide an interesting reading of the debate by articulating and
questioning what he simply calls ―the unfortunate assumption‖ that would have us assess the
satisfactoriness of a theory of justification by the service it provides in improving one‘s analysis
of knowledge. The adequacy of an account of justification, at the very least, certainly needn‘t be
restricted to its contribution to the analysis of propositional knowledge; but Foley‘s proposal
demands more than this, a full (though trial) separation of the two theories. Sosa, he thinks, still
makes the unfortunate assumption. But Foley overstates his argument that externalists and
internalists ―are principally concerned with different issues,‖ by claiming that the one is
concerned with the theory of knowledge, and the other with the theory of justified belief (2004,
60). It is this division that I take as providing Foley‘s Independence model for epistemic
compatibilism.15
Although Bernecker‘s and Foley‘s objections are both directed against Sosa, it should be
clear that they reflect some otherwise quite antagonistic philosophical motivations. In contrast to
the stance against Strong VE that both authors‘ exhibit, the position I will argue for in section 5
remains conservative enough to retain a conceptual connection between knowledge and personal
justification. In response to Bernecker I hold that adoption of the Conflict model constitutes a
dialectical mis-step for anti-skeptical philosophy; in response to Foley I point out that there are
numerous such ‗separation proposals‘ in epistemology today, and many of them upon closer
inspection undermine rather than improve our ability to respond to radical skepticism. The
advantageousness of the Integrationist stance in this regard should be readily apparent, and
17
though I have yet to make a positive case to show its philosophical stability in the face of the
skeptical challenge, we can at least conclude thus far that the ―prospects‖ for epistemic
compatibilism are not something that should be easily dismissed, at least if a philosophically
satisfying response to radical skepticism is part of what any theory of knowledge should strive
for.
5. Improving the Prospects of Externalist Responses to Skepticism
Edmund Gettier‘s objections in ―Is Justified True Belief Knowledge? (1963) to the ‗standard‘ or
Justified True Belief (JTB) analysis of knowledge provided a spur to what has since come to be
termed the ―externalist turn in epistemology.‖ But the received view of what I will call Gettier‘s
Challenge, wherein the task is to produce an analysis of knowledge with truth and justification
as conditions, plus some further condition specifically to handle Gettier cases, is multiply
ambiguous. It presupposes that the justification condition remains in place unaffected through
the externalist turn in epistemology, with an externalist ―Gettier condition‖ coming only by way
of simple addition—as ‗fourth condition‘ suggests. It also seems to push at every instance
towards ―infallibilism,‖ the view that for the challenge to be met, the other conditions in our
analysis won‘t suffice unless they actually entail the truth of the target belief.
If this is the way matters stand, they can easily be seen to motivate just the kinds of radical
or eliminative externalism that we previously identified and found wanting. But fortunately, this
motivation for infallibilism in epistemology depends on a dubious conception of Gettier‘s
Challenge. On my view, the externalist turn does not spell the end of the legitimate concern of
epistemologists with personal justification, but does necessitate its thorough reconceptualization.
To help motivate this claim I want to suggest an alternative construal of Gettier‘s challenge that
18
accepts its impetus to externalism as a correct implication, but provides a quite different
understanding of how epistemology needs to change on account of it. On this construal we can
affirm that infallibilism is no part of Gettier‘s legacy, and take the impetus to externalism as
genuinely empowering a new mode of response to radical skepticism.
The argument in support of these claims will proceed in three steps. First, contrary to the
demand that the received view triggers, for a statement of ‗thick‘ conditions to inform us what
will be sufficient for knowledge in any case whatever, I argue that we need only include ‗thin‘ or
deflated conditions in our analysis, conditions capable of being flexible enough to take on very
different content in response to different cases at hand. This first step is best exemplified in the
literature in the proposal made by Heather Battaly (2001) in ―Thin Concepts to the Rescue:
Thinning the Concepts of Epistemic Justification and Intellectual Virtue.‖ Second, our claim that
personal justification remains important, but cannot retain its original, internalist construal
through the course of the externalist turn, will be supported through the work of Michael
Williams (2001, 2005). I construe Williams as an Integrationist whose form of strong epistemic
compatibilism is embodied in his ―Default and Challenge‖ model of epistemic justification
(hereafter DCM). Third and finally, we develop the symmetries between these two proposals:
we spell out the philosophical advantages of ‗marrying‘ the thin concept of an epistemic virtue as
a condition on knowledge to the DCM as embodying the kind of analysis that should be sought
once we‘ve fully eschewed attachment to the tradition and internalistically-motivated Prior
Grounding model. It is this third step, I will argue, that provides epistemology with new
resources for a more self-consistent form of compatibilism, and through it a more philosophically
satisfying response to radical skepticism.
19
a. Thin concepts of epistemic justification and intellectual virtue.
In her essay ―Thin Concepts to the Rescue,‖ Battaly argues that we can circumvent much ill-
motivated debate ―by recognizing that the concepts of justification and intellectual virtue are
thin. Each is thin because it has multiple conditions of application […such that] there is no
definite answer as to which of these combinations is necessary, or which is sufficient, for its
application‖ (99). Internalists and externalists (and sadly to some extent virtue epistemologists
too) drag out unsupported debates by ‗picking the poison apple‘ of tacitly thickening these thin
concepts in different ways. A ―combinatorial vagueness‖ ensues that spurs heated exchanges and
seems to invite still further thickening of a condition as a way to make an analysis more precise
and to avoid counter-examples to the necessity and sufficiency of its stated conditions. But just
what is this debate over, she asks, if there is ―no sharp distinction between the combinations of
conditions that are, and those that are not, necessary and sufficient for its application‖? (104)
To avoid picking the apple, Battaly prescribes leaving the concepts of justification and
intellectual virtue with little pre-given meaning. A thin areteic condition is a condition that
remains largely formal in character, but is for that same reason highly flexible, being able to take
on different meanings in different contexts. To place it into one‘s analysis of knowledge is
thereby merely to present the skeptical interrogator with ―a roughly drawn sketch that can be
completed in different ways‖ (107).
If this proposal is useful it is because what epistemologists want to analyze through terms
like ‗justified‘ and ‗knows‘ actually occurs along a range or spectrum.16 Left in a formal or
deflated manner, an areteic condition can be ‗bent‘ in several directions, allowing it to stand in
for a diverse list of possible meanings of justification—items ranging from simple ‗aptness‘ that
might invite reference to only the faculty virtues in some instances, to the complex reason-giving
20
and sensitivity to counter-evidence that we associate with the application of critical reflective
intelligence, and with sound reflective intellectual habits and inquisitive methods.
But can this first step of the argument lead us anywhere worth going? By suggesting we can
treat justification this way, aren‘t we conceding that we fail to meet a reasonable demand by the
skeptic for a single set of ―thick‖ necessary and sufficient conditions that define knowledge
across its entire knowledge spectrum? If so, we must explain both why skeptical arguments have
the initial plausibility to deserve serious attention, and why they lose their force or don‘t arise in
the same way when we proceed by our own premises. In order to meet this burden, we will need
to take a second step involving a more direct examination of just what our dialogical obligations
to the skeptic actually are.
b. The “Default and Challenge” model of epistemic justificationn.
Williams‘ form of epistemic compatibilism makes a crucial contrast between the Prior
Grounding conception of knowledge (hereafter PGM), and the Default and Challenge model
(hereafter DCM).
[T]o take account of externalist insights, we have to detach the idea that knowledge is
essentially connected with justification, not only from the classical demonstrative ideal
and its infallibilist descendants, but also from all conceptions of justification that insist
on respecting the Prior Grounding [Model]. Effecting this detachment leads us to see
justification as exhibiting a Default and Challenge structure, where constraints on the
reasonableness of challenges and the appropriateness of justifications are contextually
variable along several dimensions.17
21
Not only the Agrippan Trilemma, but also the traditional internalist conception of
justification builds in the PGM, which Williams characterizes as composed of four inter-related
theses of an internalist and evidentialist orientation.18 If we implicitly adopt the PGM from the
outset, then we accept an asymmetry of justificational obligations and an unrestricted
commitment on the part of claimants to demonstrate entitlement to opinion; ―If all reasonable
believing is believing-on-evidence, the skeptic is entitled to ask for the evidence to be produced,‖
generating a vicious regress. But absent this requirement, the skeptic holds no right to issue such
―naked challenges.‖
Many authors have noticed substantial connections between epistemic internalism and
motivations to skepticism. Williams explains this connection by noting that the PGM ―both
generates the threat of skepticism and constrains our responses to that threat‖ (2001, 188). On the
DCM, by contrast, ―questions of justification arise in a definite justificational context,
constituted by a complex and in general largely tacit background of entitlements, some of which
will be default‖ (158). What Williams calls our default entitlements are not ―mere assumptions,‖
because they are always provisional and backed up by a defense commitment. The most
important point of Williams‘ approach for us to develop is a different conception of our
discursive obligations. Adopting the DCM means that challengers and claimants share
justificational responsibilities, and hence that one needn‘t concede to the gross asymmetry that
the PGM instantiates: ―no move in the game of giving and asking for reasons is
presuppositionless. On a Default and Challenge conception of justification, there is no room for
either the skeptic‘s global doubts or the traditional epistemologist‘s global reassurances‖ (150).
22
c. Marrying thin concepts and the default and challenge model.
From the one side, the DCM is not only compatible with, but seems to require for its completion,
two distinct targets for philosophical analysis: one for the default mode of inquiry and the other
for particular motivated challenges in particular cases. The thin and thick description of
intellectual virtues supply the DCM with just the kind of explanations one might intuitively think
the ‗default‘ and ‗challenge‘ contexts call for: Thin descriptions for our default philosophical
mode of life, and thick descriptions for more skeptical philosophic mode of life, or simply,
whenever a defense commitment is engaged for the attributor of knowledge to a particular
agent.19
In the default situation , no particular challenge is set before us and so one or more ‗thin‘
conditions on knowledge suffices; areteic and anti-luck conditions (one or both) are good
candidates for this since they aim to provide what is needed while keeping the focus on
naturalistically-grounded talk of the agent‘s cognitive habits and dispositions. ‗Thin concept
analysis‘ featuring virtue-theoretic terms, in particular, underscores a naturalistic approach to
knowledge (one grounded in our habits and dispositions) that nevertheless preserves the link
Williams wants with issues of personal justification (or responsibility). Prior to or in lieu of a
motivated challenge to truth and virtue having come apart in a particular case, a thin areteic
condition suffices for the expectation that the truth of the agent‘s belief will be of epistemic
credit to her as an agent. By contrast, the dialectical context in which there has arisen a
motivated challenge over a particular case calls for a different kind of analysis, and in this
contest thickly describable dispositions and acquired virtues again seem to provide adherents of
the DCM with just the kind of explanation their model calls for.
23
Viewed from the other side, proponents of VE might come to see adoption of the DCM
as advantageous if they find that it helps them clarify the very different explanatory roles that
thick and thin concepts play in our field. If the DCM supports a more stable form of epistemic
compatibilism, as Williams clearly holds, then its adoption and the consequent relinquishing of
preconceptions about justification bound up in the PGM may aid inquiries into the relationship
between our normative and naturalistic posits, and into what Riggs (2007) describes as ―the
conceptual connections among a family of concepts that include credit, responsibility,
attribution, and luck.‖ For in responding to a particular motivated challenge to our attribution of
knowledge to an agent, we are essentially responding to concerns that the agent either has a false
belief despite her best cognitive effort, or else to concerns that the agent should not be accorded
the epistemic credit that we normally would accord her for her cognitive success — that is, for
the truth of her belief (for Greco‘s account of epistemic credit, see his 2003a and 2003b).
Adoption of the DCM also aids VE my providing another way of understanding the
importance of the ―relevant alternatives intuition.‖ ―It is easy to miss the fact that the practice of
justifying is only activated by finding oneself in the context of a properly motivated challenge,‖
and when we miss this we allow the skeptic ―to transform the ever-present possibility of
contextually appropriate demands for evidence into a unrestricted insistence on grounds,
encouraging us to move from fallibilism to radical skepticism‖ (Williams 2001, 150). After
Williams‘ proposed dialectical re-positioning, the anti-skeptical philosopher‘s inability to answer
the skeptic premised upon conditions set forth by the PGM can be acknowledged, but the
demand itself reasonably set aside.20
The view we have arrived at through our three-step argument underlines the pragmatic
wisdom behind Battaly‘s proposal for harnessing the resources of thin concept analyses of
24
central epistemic terms, and acknowledges the force of Williams‘ demand that self-consistent
externalists stop answering to demands whose rationale lies only in the Prior Grounding model
of our discursive obligations to the skeptic. Moreover, we have taken a step somewhat unique in
the literature by intentionally tying the Battaly and Williams proposals closely together, arguing
that doing so multiplies the anti-skeptical force of each taken separately. In summary of the
arguments in sections 4-5, my intention has been to point out how the force of virtue theoretic
responses to skepticism might be substantially enhanced by realizing their symmetries with the
Default and Challenge model of the discursive obligations holding between skeptical
interrogators and anti-skeptical interlocutors. It isn‘t merely circular reasoning or arbitrary
assumption to hold that modally far-off possibilities do not present philosophical objections to
our ordinary practices of knowledge attribution. Adoption of the DCM allows us to see this,
while leading to an understanding of Gettier‘s Challenge that invites neo-Moorean
epistemologies that draw upon the core relevant alternatives intuition.21
25
Notes
26
Works Cited
Audi, R. 2004. ―Intellectual Virtue and Epistemic Power,‖ in J. Greco (ed.) Sosa and his Critics,
3-16.
Axtell, G. 2000. Knowledge, Belief and Character: Readings in Virtue Epistemology. Totowa,
NJ: Rowman and Littlefield.
Axtell, G. 2001. ―Epistemic Luck in Light of the Virtues‖, in A. Fairweather and L. Zagzebski
(eds.), Virtue Epistemology: Essays on Epistemic Virtue and Responsibility, Oxford
University Press, Oxford, 158-77.
Axtell, G. 2003. ―Felix Culpa: Luck in Ethics and Epistemology‖, in M. S. Brady and D. H.
Pritchard (eds.), Moral and Epistemic Virtues, Blackwell, Oxford, 235-57.
Axtell, G. 2006a. ―Blind Man‘s Bluff: The Basic Belief Apologetic as Anti-Skeptical
Stratagem‖, Philosophical Studies Vol. 30, No. 1 (2006): 131-152.
Axtell, G. 2006b. ―The Present Dilemma in Philosophy‖, manuscript.
Axtell, G. 2007. ―Two for the Show: Anti-luck and Virtue Epistemologies in Consonance,‖
Synthese special edition on epistemic luck (D. Pritchard, ed.), forthcoming.
Baehr, J. 2006a. ―Character in Epistemology‖, Philosophical Studies, forthcoming.
Baehr, J. 2006b. ―Character, Reliability, and Virtue Epistemology‖, Philosophical Quarterly,
forthcoming.
Battaly, H. 2001. ―Thin Concepts to the Rescue: Thinning the Concepts of Epistemic
Justification and Intellectual Virtue‖, in A. Fairweather and L. Zagzebski (eds.), Virtue
Epistemology: Essays on Epistemic Virtue and Responsibility, Oxford University Press,
Oxford, 98-116.
27
Bernecker, S. 2006. ―Prospects for Epistemic Compatibilism‖, Philosophical Studies Vol. 20, no.
1: 81-104.
Bernecker, S. 2007. ―Agent Reliabilism and the Problem of Clairvoyance,‖ Philosophy and
Phenomenological Research, forthcoming.
Bonjour, L. 2003. ―Reply to Sosa,‖ in Bonjour and Sosa (eds.) Epistemic Justification. Malden,
Ma.: Blackwell, 173-200.
DePaul, M. and Zagzebski, L. 2002. Intellectual Virtue: Perspectives From Ethics and
Epistemology. Oxford: Oxford University Press.
Foley, R. 2004. ―A Trial Separation between the Theory of Knowledge and the Theory of
Justified Belief‖, in J. Greco (ed.), Sosa and His Critics, 59-72.
Foley, R. 2005. ―Justified Belief as Responsible Belief.‖ In Matthias Steup and Ernest Sosa,
(eds.), Contemporary Debates in Epistemology. Oxford: Blackwell: 313-26.
Gettier, E.: 1963. ―Is Justified True Belief Knowledge?,‖ Analysis 23: 121-3.
Greco, J. 2000. Putting Skeptics in Their Place: The Nature of Skeptical Arguments and Their
Role in Philosophical Inquiry. New York: Cambridge University Press.
Greco, J.: 2002a. ―How to Reid Moore,‖ Philosophical Quarterly 52, 544-63.
Greco, J. 2002b. ―Virtues in Epistemology.‖ In Paul Moser (ed.), The Oxford Handbook of
Epistemology. Oxford: Oxford University Press, 287-315.
Greco, J. 2003a. ―Knowledge as Credit for True Belief‖. M. DePaul and L. Zagzebski (eds.),
Intellectual Virtue, 111-34.
Greco, J. 2003b. ―Virtue and Luck, Epistemic and Otherwise,‖ Metaphilosophy 34: 353-66.
Greco, J. (ed.) 2004a. Ernest Sosa and his Critics. Oxford, UK: Blackwell.
28
Greco, J.: 2004b. ―How to Preserve Your Virtue while Losing Your Perspective‖, in J. Greco
(ed.) Sosa and His Critics, 86-95.
Greco, J.: 2005. ―Virtue Epistemology‖, Stanford Encyclopedia of Philosophy,
http://plato.stanford.edu/entries/epistemology-virtue/.
Hookway, C. 1990. Skepticism. London: Routledge.
Hookway, C. 2003. ―How to be a Virtue Epistemologist,‖ in M. DePaul and L. Zagzebski (eds.),
Intellectual Virtue: Perspectives from Ethics and Epistemology: 183-202.
Kornblith, H. 2004. ―Sosa on Human and Animal Knowledge‖, in J. Greco (ed.) Ernest Sosa and
His Critics, 126-134.
Kvanvig, J. 2003. The Value of Knowledge and the Pursuit of Understanding. Cambridge:
Leite, A. 2004. ―On Justifying and Being Justified.‖ Philosophical Issues 14: 219-253.
Montmarquet, James. ―Pure vs. Practical Epistemic Justification,‖ manuscript.
Moore, G. E. 1925. ―A Defence of Common Sense‖, in Contemporary British Philosophy (2nd
Series), ed. J. H. Muirhead (London: Allen and Unwin).
Moore, G.E. 1939. ―Proof of an External World‖, Proceedings of the British Academy, 25, 273-
300.
Pritchard, D. 2005. Epistemic Luck. Oxford: Oxford University Press.
Pritchard, D. 2006. ―Greco on Reliabilism and Epistemic Luck,‖ Phil. Studies Vol. 30, No. 1: 35-
45.
Pritchard, D. 2007. ―Anti-Luck Epistemology,‖ Synthese, forthcoming.
Reed, B. 2001. ―Epistemic Agency and the Intellectual Virtues,‖ Southern Journal of Philosophy
39: 507-26.
29
Riggs, W. 1998. ―What are the ‗Chances‘ of Being Justified?, The Monist 81: 452-72.
Riggs, W. 2002. ―Reliability and the Value of Knowledge.‖ Philosophy and Phenomenological
Research 64 (1): 79–96.
Riggs, W. 2007. ―Why Epistemologists are so Down on their Luck‖, Synthese, forthcoming.
Roberts, Robert and Wood, Jay W. 2007. Regulative Epistemology. OUP, forthcoming.
Sextus Empiricus. 1933-1939. Outlines of Pyrrhonism. R. G. Bury translation. Cambridge, Ma.:
Harvard University Press.
Sosa, E. 1999b. ―How to Defeat Opposition to Moore,‖ Philosophical Perspectives 13: 141-53.
Sosa, E. 2000a. ―Reflective Knowledge in the Best Circles,‖ in Sosa and J. Kim (eds.),
Epistemology: An Anthology. Malden, Ma: Blackwell, 274-87.
Sosa, E. 2000b. ―Thomas Reid‖ (co-authored with James Van Cleve), in S. Emmanuel, ed., The
Modern Philosophers. Oxford: Blackwell.
Sosa, E. 2002a. ―Tracking, Competence, and Knowledge.‖ In The Oxford Handbook on
Epistemology, edited by Paul. Moser. Oxford: Oxford University Press.
Sosa, E. 2002b. ―The Place of Truth in Epistemology.‖ In DePaul and Zagzebski (eds.)
Intellectual Virtue, 155-80.
Sosa, E. 2004a. ―Replies,‖ in Greco (ed.) Sosa and his Critics. Oxford: Blackwell, 275-325.
Sosa, E. 2004b, ―Two False Dichotomies: Foundationalism/Coherentism and
Internalism/Externalism‖, in W. Sinnott-Armstrong (ed.), Pyrrhonian Skepticism, Oxford
University Press, New York, 146-60.
Sosa, E. 2007. A Virtue Epistemology. Volume one: Apt Belief and Reflective Knowledge.
Forthcoming.
30
Stroud, B.: 2004a. ―Contemporary Pyrrhonism‖, in W. Sinnott-Armstrong (ed.), Pyrrhonian
Skepticism. New York: Oxford University Press, 174-87.
Stroud, B.: 2004b. ―Perceptual Knowledge and Epistemological Satisfaction‖, in J. Greco (ed.),
Sosa and His Critics. Oxford: Blackwell, 165-73.
Williams, M.: 2001. Problems of Knowledge, Oxford: Oxford University Press.
Williams, M.: 2005. ―The Agrippan Argument and Two Forms of Skepticism," in W. Sinnott-
Armstrong (ed.) Pyrrhonian Skepticism. New York: Oxford, 121-45.
Zagzebski, L. 1996. Virtues of the Mind. Cambridge: Cambridge University Press.
1
For a fuller account of the emergence of virtue epistemologies, see the introduction to Axtell
(2000).
2
―For all S, φ, ψ, if S‘s evidence for believing φ does not favour φ over some hypothesis ψ that
S knows to be incompatible with φ, then S is not internalistically justified in believing φ‖
(2005, 108). Throughout the paper, I transpose ―scepticism‖ into ―skepticism‖ for the sake of
continuity.
3
Pascal‘s Pensees #434, my translation.
4
Perhaps the earliest of the virtue epistemologists to draw from Reid was Christopher Hookway
in Skepticism (1990, 2003), where he argues that Reid‘s approach allows us to recognize the
contingency of our confidence in our common-sense beliefs, without denying the legitimacy of
that confidence (240

5
Sosa 1999; compare his ―Replies‖ in Greco ed. (2004a), 276.
6
Sosa 2000a. He allows that his virtue perspectivism is thus ―structurally‖ Cartesian while
pointing out that ―in content it is not‖ (281). For the agent‘s epistemic ascent can be understood
31
naturalistically, without Descartes‘ invocation of a creator-God who benevolently guarantees
the truth of what we most clearly and distinctly conceive.

7
―Replies‖ in Greco (ed.) 2004a, 290-291.
8
Sosa‘s virtue perspectivism has been characterized over the years by its two-tiered or
―stratified‖ conception of justification (1991, 189), where a key distinction is between
"externalist, reliability-bound aptness and internalist, rationality-bound justification." The
version of Strong VE I will sketch is partly intended to show that a troublesome stratification
needn‘t be posited.
9 As Sherman and White (2002) put the point somewhat more generally, Aristotle‘s emphasis on
the emotions remains a resource for contemporary VE.

10
For example, Sosa (2004b) argues that internalism/externalism and
coherentism/foundationalism are ―two false dichotomies‖ overcome by taking the agent as the
‗seat of justification,‘ while Greco and Linda Zagzebski, whatever their other differences, both
emphasize that the ‗mixed‘ character of VE is philosophically crucial rather than detrimental to
its ability to respond to the skeptical challenge. James Montmarquet locates internalist concerns
as reflecting interest in ―pragmatic epistemic justification‖ (that is, on acting in a morally
satisfactory way on our beliefs) , rather than in ―pure epistemic justification‖ and the question of
the simple possession of knowledge. John Turri argues that externalist intuitions track ―doxastic
justification‖ while internalist intuitions track ―propositional justification.‖ Heather Battaly
(2001, 109) argues that the debate between internalists and externalists is misguided insofar as
authors fail to see that they are ―thickening‖ a naturally ―thin‖ concept of justification. Wayne
Riggs (1998, 453) writes that ―The responsibilist and truth-conducivist conceptions of
justification define distinct epistemic evaluations, and so are not in any interesting sense rival
32
notions‖ because each is addressed to a different kind of epistemic luck. Axtell (2001) argues
that the truck that internalists and externalists have with filtering-out different kinds of epistemic
luck shows each to hold an incomplete account, reflecting in part divergent value-charged
demands for the intelligibility of all knowledge.
11
These authors are engaged in such projects as the study of ―understanding‖ (Elgin, Kvanvig,
Riggs), explanatory stories about the acquisition of knowledge (Eflin, Pritchard, Reed), or
theories of pragmatic rationality as contrasted with purely epistemic rationality (Foley,
Montmarquet). Alternatively they describe themselves as ―agnostic‖ about a role for the virtues
in analysis of knowledge, and simply turn their attention to what they think are other interesting
roles that the virtues can play (Roberts and Wood, 2007).
12
On my earlier taxonomic distinction between ―reliabilist‖ and ―responsibilist‖ VE, a
responsibilist is free to advocate either Weak or Strong VE.

13
For a fuller defense of this claim and presentation of Stroud‘s Challenge to what he terms
―scientific externalism, see Axtell 2006a.

14
Having made his concession, why does Bonjour go on to embrace ―internalist justification‖ as
the one of epistemic centrality? If Bonjour allows that internalist justification and personal
justification (qua responsibility) can come apart, ―responsibilists‖ in epistemology might be
taken as those who instead choose epistemic responsibility, with its ties to natural dispositions
and to acquired habits of inquiry (hexis for the Greeks), as the more epistemologically interesting
concept. This makes it most plausible to hold that issues of personal justification and of
distinctively reflective knowledge and reflective intellectual virtues maintain a continuing
importance, even after the externalist turn and indeed partly because of it. For responsibilist
33
approaches to problems about religious belief and religious skepticism, see Axtell (2006a) and
Roberts and Wood (2007). For more on ―responsibilist VE,‖ see Baehr 2006a, 2006b and the
work of other Editors of JanusBlog at http://janusblog.squarespace.com.

15
Foley and Montmarquet both adopt such an Independence model, and with it the Weak VE
concessionary strategy of endorsing the austere externalist account of knowledge as merely true
belief reliably produced. Foley says he sees no reason ―to read back into the account of
knowledge some duly externalized notion of justified belief‖ but rather locates reason-giving
explanations elsewhere, in a ―general theory of rationality‖ (2005, 315). Montmarquet
alternatively re-locates his role for the intellectual virtues in the study of ―pragmatic epistemic
justification,‖ distinct from ―pure epistemic justification‖; this makes it a study about action, or
more specifically about how to evaluate agents with respect to their acting in morally satisfactory
or unsatisfactory ways on their beliefs. Montmarquet argues that Sosa‘s influential distinction
between animal and reflective knowledge should give way to a distinction between knowledge
(animal, mechanical, or human) and the distinctively human form of responsibility (epistemic as
well as moral) involved in acting on one‘s beliefs.

16
Zagzebski reflects this idea of thin concept analysis in writing that she wants to construct a
virtue-theoretic approach that takes knowledge broadly enough ―to cover a multitude of states,
from the simplest case of ordinary perceptual contact with the physical world, requiring no
cognitive effort or skill wherever, to the most impressive cognitive achievements‖ (1996, 264).
17
Williams 2001, 245. There is ―contextualism‖ in Williams‘ thought, but it is one that
distinguishes between two different philosophical contexts for dialogue, rather than between an
‗everyday‘ and a ‗skeptical‘ context, as self-described contextualists typically do.
34
18
Williams (2001, 147) analyzes the PGM into four interconnected theses: ―(PG1) No Free
Lunch Principle. Epistemic entitlement—personal justification—does not just accrue to us: it
must be earned by epistemically responsible behavior. (PG2) Priority Principle. It is never
epistemically responsible to believe a proposition true when one‘s grounds for believing it true
are less than adequate. (PG3) Evidentialism. Grounds are evidence: propositions that count in
favour of the truth of the proposition believed. (PG4) Possession Principle. For a person‘s
belief to be adequately grounded…the believer himself or herself must possess (and make
proper use of) evidence that makes the proposition believed (very) likely to be true.‖
19
In doing so, my form of VE stands in opposition to those who hold that we need to either
endorse infallibilism (Zagzebski) or reject the epistemic closure principle (Dretske, Nozick).
This adds more substance to the claim that we are offering a genuine reinterpretation of Gettier‘s
Challenge.
20
Compare this strategy with Greco, who until recently has proposed only a ―partial‖ account of
knowledge, restricted to stating necessary conditions, but who more recently strengthened his
requirements in order to be able to maintain their general sufficiency as well. The present
proposal responds to many of the same tensions that previously made Greco reluctant to assert
the general sufficiency of his conditions, yet does so by the quite different approach of adopting
the DCM and rejecting as unmotivated the claim that an adequate analysis is one that states
‗general sufficient‘ conditions on knowledge.

21
Special thanks go to Jason Baehr, Sven Bernecker, Juli Eflin, John Greco and Ernest Sosa for
useful and helpful suggestions on this paper, and to participants of JanusBlog: The Virtue
Theory Discussion Group for extended discussion of many of the pertinent issues.
35
OUP CORRECTED PROOF – FINAL, 8/8/2011, SPi
PART II
Virtue Critiques: Evidence
and Inquiry
Evidentialism appears to be a normative theory. It says what one ought to believe. In

Evidentialism Conee and Feldman say ‘We hold the general view that one epistemically
ought to have the doxastic attitudes that fit one’s evidence. We think that being
epistemically obligatory is equivalent to being epistemically justified. There are in
the literature two other sorts of view about epistemic obligations’ (2004: 88). Feldman
has written a number of things concerning this—including ‘The Ethics of Belief’,
reprinted in Evidentialism:
Feldman, Richard. 1988. ‘Epistemic Obligations,’ in James Tomberlin (ed.)
Philosophical Perspectives 2, Epistemology. Ridgeview Publishing Co., pp. 235–56.
——. 1988. ‘Subjective and Objective Justification in Ethics and Epistemology,’
The Monist 71: 405–19.
——. 1998. ‘Epistemology and Ethics,’ Routledge Encyclopedia of Philosophy, edited
by Peter Klein and Richard Foley. Routledge.
——. 2001. ‘Voluntary Belief and Epistemic Evaluation,’ in Matthias Steup (ed.)
Knowledge, Truth, and Duty: Essays on Epistemic Justification, Responsibility, and
Virtue. Oxford: Oxford University Press, pp. 77–92.
——. 2002. ‘Epistemological Duties,’ in Paul Moser (ed.) The Oxford Handbook of
Epistemology. Oxford: Oxford University Press, pp. 361–84.
——. 2008. ‘Modest Deontologism in Epistemology,’ Synthese, 161: 339–55.
Evidentialism was in fact intended largely as a response to denials of evidentialism which
stemmed from normative theories concerned with human limitations. Conee and
Feldman think that practical concerns should be kept separate from epistemic concerns.
Guy Axtell also offers a forceful call to unite theory and practice. He clearly
thinks epistemic normativity should be tied to the process of inquiry. He also thinks
evidentialist views are too narrow in a number of ways related to this. First, he argues
that an adequate notion of epistemic normativity should be diachronic, not synchronic.
70 V I RT U E C R I T I Q U E S : E V I D E N C E A N D I N Q U I RY
Second, he claims that evidentialists are committed to a form of epistemic value monism
called veritism: that the only epistemic value is avoiding believing falsehoods. Finally,
Axtell is writing from firmly within the Pragmatist camp, so he thinks that the social
consequences of a theory pertain to its acceptabilitiy. He clearly thinks that evidential-
ism has bad consequences for democracy.
Jason Baehr revives concerns with this aspect of evidentialism’s normative compo-
nent. He offers cases which purport to show that though some proposition p fits one’s
evidence, one shouldn’t believe p. These cases all involve some kind of vice displayed
in the subject whether that be because of lazy investigation or actual malicious intent.
He suggests that evidentialists add a virtue component to their view in order to
accommodate such cases. He worries that if they don’t, they risk focusing on a property
that lacks any real value.
Keith DeRose’s essay at the beginning of Part IV also raises concerns concerning
inquiry which are part of what motivate his skepticism about evidentialism. Thus it is
an honorary member of this section.
4
From Internalist Evidentialism
to Virtue Responsibilism
Guy Axtell
The question has not yet been generally accepted in the Academy—let alone any
answer agreed upon—whether the twin concepts of ‘rationality’ and ‘reasonableness’
are not interdependent ideas, of comparable authority and philosophical interest.
Indeed, it is not always recognized that the two ideas can be distinguished.
Stephen Toulmin, Return to Reason (2001, 2)
4.1 Introduction
Evidentialism as Earl Conee and Richard Feldman present it is a philosophy with
distinct aspects or sides: Evidentialism as a conceptual analysis of epistemic justification,
and as a prescriptive ethics of belief—an account of what one ‘ought to believe’ under
different epistemic circumstances. This chapter will focus especially on the evidentialist
ethics of belief and an associated set of views in a growing literature usually dubbed the
epistemology of disagreement. I find myself a discontent with evidentialism in both of
its main senses however, and will begin with an undercutting critique of evidentialism
in its first and primary sense. Effectively rebutting their evidentialist account of episte-
mic responsibility would serve to undercut Conee and Feldman’s ethics of belief
because they are explicit in endorsing only cognitive and not also moral evidentialism;
the two sides of the evidentialist research program are asymmetrically related, their
epistemological account of what it means to be a successful epistemic agent being the
whole of the philosophic underpinning they provide for the extension of evidentialism
into an ethics of belief. It is cognitive evidentialism that the undercutting criticism of
Part II is primarily concerned with.
It is perhaps too usual a practice in epistemology today for writers to concern
themselves directly only with matters such as these—only, that is, with the purported
theoretical pedigree of an account—and to accept on that basis whatever consequences
for actual practice it may turn out to have. Yet any proposed ethics of belief, I begin by
noting, can have quite profound consequences on how we perceive and treat others
whose beliefs and/or attitudes differ from our own. Views about the ethics of belief
affect our judgment of the reasonableness and even the sanity of people who differ
from us in their beliefs and attitudes. Such judgments in turn have a sometimes deep
impact on the great cultural debates of one’s day and age. Any philosophical ethic of
belief, let us therefore say, proposes a kind of ‘accommodation’ between the individual
and the community of inquirers, an accommodation that can and should be evaluated
pragmatically.
Adherence to the evidentialist accommodation is considered by many to be the very
meaning of the Enlightenment Challenge to pre-modern modes of thinking. Yet as
Feldman concedes with some reluctance in recent papers, his account renders reason-
able disagreement among public evidence-sharing epistemic peers impossible, or nearly
so. As Feldman puts it in ‘Reasonable Religious Disagreements’, ‘open and honest
discussions seem to have the puzzling effect of making reasonable disagreement
impossible’ (2007: 202). My discontent derives in part from the ‘Victorian’ austerity
of the evidentialist ethic of belief, but also and more pertinently for this paper, from
concern that Feldman’s puzzlement at what he concedes is an ‘unfortunate’ implication
of his approach far understates the deleterious practical consequences of his claim.
Essentially then, the critical claims of this chapter are that Conee and Feldman’s
evidentialist philosophy has both ‘weak roots’ (Section 4.2) and ‘sour fruits’ (Section 4.3).
A concern with pragmatic consequences will at points lead us away from epistemology
proper and into a broader discussion of the importance of a sound ethic of belief as a
philosopher’s tool of mediation in our ongoing and often volatile ‘culture wars’ over reason
and faith, science and religion, etc. Thinkers can mean different things by ‘peer’ and
‘public-evidence sharing’ of course, and this may well be an area where Conee and
Feldman are not in as close agreement as they are on the more general contours of the
two sides or aspects of evidentialism (Conee 2010). In offering a critique of the practical
consequences of the evidentialist accommodation in Section 4.3, I will focus primarily on
Feldman’s recent papers (2006, 2007) which directly contest the possibility of reasonable
disagreement.
In the quote at the start of this chapter, Stephen Toulmin asks fellow philosophers if
they should not be concerned to delineate the different extensions and normative
meanings of ‘rationality’ and ‘reasonableness’. In his sweeping critique of intellectual-
ism, Return to Reason, from which this passage is drawn, he calls upon us to correct
‘a current imbalance between our ideas of “rationality” and “reasonableness” ’ (2001:
12). The evidentialist approach not only doesn’t allow for this distinction, it indeed
drives the reduction of the latter concept to an idiosyncratically internalist construal of
the former. But with a different epistemological approach we are also likely to arrive at
an ethics of belief with a substantially different character; with a different enough
approach we may advantage ourselves to ‘recover reasonableness’ and ‘return to
reason’. In order to answer this call, the concluding section (4.4) will briefly suggest
why what seems profoundly dissatisfying in the evidentialist epistemology and ethics of
belief is avoidable when we begin instead from a virtue epistemology. What I propose
F RO M I N T E R N A L I S T E V I D E N T I A L I S M T O V I RT U E R E S P O N S I B I L I S M 73
as an alternative responsibilist accommodation conceives the standards of intellectual

responsibility or reasonableness pertinent to the ethics of belief to primarily derive
from evaluation of intellectually virtuous/vicious motivations and habits of inquiry—
diachronic or longitudinal ‘reasonableness’—rather than deriving from the narrow,
technical sort of epistemic significance they call ‘synchronic rationality’. In this way we
recover the senses of reasonableness and responsibility that naturally befit philosophic
concern with a sound and civic ethic of belief, from the evidentialists’ reduction of
intellectual reasonableness to that obviously responsibility-eschewing sense of it that
‘synchronic rationality’ (or evidential fit) denotes.
4.2 Weak roots

Internalist evidentialism, this section will argue, is highly incomplete as an account of
epistemic responsibility: it offers too narrow a conception of the epistemic, of the
epistemic goal, and of what it means to maximize epistemic value or to be a successful
epistemic agent. The critique of internalist evidentialism in this section will be less
severe than John Greco’s (2005) claim that internalism is false as a thesis about any
interesting or important sort of epistemic valuation and any corresponding sort of
epistemic normativity, yet substantially more robust than Jason Baehr’s claim that
internalist evidentialism only needs mending with a ‘virtue patch’ (this volume).
If internalism was ever a consensus view about epistemic justification, this is no
longer true in contemporary epistemology, and the centrality of doxastic justification
in non-internalist epistemologies, and in virtue epistemologies in particular, makes
necessary a thorough reconceptualization of the relationship between doxastic justifi-
cation (concerned with belief and knowledge), and personal justification (concerned
with agency and responsibility). There does seem to be a near consensus view among
internalists, however, first conceding the need to take account of doxastic justification
in an analysis of knowledge, and secondly asserting the adequacy of addressing this need
by describing the agent as doxastically justified whenever they possess normal eviden-
tial or propositional justification that p, and ‘base’ their belief that p on the evidence
which propositionally justifies it.
This way of bringing the basing relationship into the analysis of knowing through a
back door will not do for the non-internalist, however, who may allow internalism
about evidential justification, but will not equate evidential justification with epistemic
justification, nor accept any general internalist condition on knowing. The internalist
still fails to give the importance of the etiology of belief its due, conceiving a justifiable
belief and a belief in fact justified for an agent, as separated only by a synchronic, time-
slice act of ‘believing on the basis’. But virtue epistemologists like John Turri object,
arguing that ‘doxastic justification cannot be understood simply as propositional
justification plus basing’ (2010: 13).
Doxastic norms (norms concerning practices of belief-formation and maintenance)
can be diachronic or synchronic; that the diachronic norms arise from cross-temporal
or longitudinal evaluations does not of itself undermine their status as genuinely

epistemic. Indeed, if the relationship between knowledge-possession and the etiology
of belief runs deeper than the internalist analysis allows for, then either propositional
and doxastic justification are independent sources of epistemic value (see Axtell and
Olson 2009), or ‘we should explain propositional justification in terms of doxastic
justification’ (Turri 2010, 1; compare Greco 2005). In either case the internalist fails to
acknowledge what I want to insist is a robust inter-relationship between epistemic
responsibility of the sort that results in doxastic justification, and the etiology of belief.
Epistemic rationality or personal justification has two aspects, diachronic and synchron-
ic, and while neither is directly truth-linked it is arguably the former that is the more
salient prefiguring cause of doxastic justification and of knowledge in most cases.
A proper understanding of epistemic responsibility therefore shows it to integrally
involve knowledge-attributors in considerations of the etiology of an agent’s target
belief (Greco 2005). ‘Poor utilization of even the best reasons for believing p will
prevent you from justifiably believing or knowing that p. . . . The way in which the
subject performs, the manner in which she makes use of her reasons fundamentally
determines whether her belief is doxastically justified’ (Turri 2010: 10).
So what happens when we don’t accept the internalist account of epistemic
justification qua evidential justification in the first place? What if we think with
Michael Williams that the account presupposes a ‘Prior Grounding Model’ of
justification that is skepticism-inviting, and further that as Williams writes in ‘Res-
ponsibility and Reliability’ (2008), the debate over internalist and externalist views of
justification lingers on after it has ceased to be useful owing largely to a tendency on
the one side, ‘to adopt an overly demanding, hyper-intellectualized conception of
what epistemic responsibility demands’ (2008: 1)? In that case I think we’ll see the
externalist turn in epistemology as undercutting or at least problematizing claims that
function to privilege propositional justification as conceptually more basic than
doxastic justification, and as the only epistemic good that diachronic responsibility—
responsibility in inquiry—is good for. We will challenge and I think reject what I call
the Well-foundedness Formula, that doxastic justification is just synchronic proposition-
al justification ‘plus basing’. Conee and Feldman help justify the claim that only
synchronic epistemic rationality—one’s present time-slice response to evidence—matters
in epistemic evaluation of agent rationality and justification, in part by insisting that
across-time responsibility or irresponsibility in inquiry raises ‘moral or prudential
questions rather than epistemic questions’ (2004: 178). The ‘too narrow objection’
to this view of the epistemic has previously been made by Alvin Plantinga, and
Feldman’s reply toes a Chisholmian line, where anything except raw synchronic
rationality, the rationality of evidential fit, is ‘irrelevant to the central notion of
epistemic justification’ (1988: 249). Also irrelevant to this judgment are said to be
the long-term epistemic consequences of adopting the belief, as well as any consid-
eration of how the agent came to hold what they consider the evidence bearing on
that target proposition, ‘whether by conscientious enquiry or by avoiding potentially
troublesome information’ (2004: 189–90). Evaluation of the agent’s inquiry-directed

motivations and conduct across time is possible, but is irrelevant to what one ought to
believe at any given moment; synchronic evaluation is the only genuinely epistemo-
logical concern, or the only one they allow to factor into their calculation of epistemic
value maximization.
I suggest that the claim that Conee and Feldman repeatedly make, that evidentialism
provides no guidance about what an agent should do (2004: 189), but only what she
should believe at any moment about any given proposition, sets up a classic false
dichotomy, and that this teaches us more about the shortcomings of internalist
evidentialism than about the irrelevance of inquiry-directed habits and activities to
epistemic responsibility and epistemic justification. Dichotomizing as the evidentialist
wants to between the synchronic and diachronic rationality, in order to define only the
former as arising from a general intellectual requirement and the latter as always having
a non-epistemological source out of moral or prudential considerations, reflects a
doubtful strategy. The virtue-theoretic response to this Chisholmian deontology, for
the obvious reason it takes intellectually virtuous habits and abilities/competencies to
determine doxastic justification, provides what I think is strong support for the force of
the ‘too narrow’ objection to Conee and Feldman’s account of what epistemic
obligation and success consist in.
In a ‘salience contextualist’ or ‘inquiry pragmatist’ version of virtue responsibilism, as
one alternative, judgments about what constitutes an agent’s total evidence in respect
of comprehensive religious, philosophic, and political beliefs are viewed as ‘entangled’
with qualitative or value-laden judgments about prudential, and even at times moral
factors that affect the agent’s efforts at inquiry. ‘Intellectual responsibility’ should
accordingly be acknowledged as the ‘thick’ concept common language takes it to be.
In lieu of a cogent reduction plan, we should not assume that the sense or senses of
responsibility that inform our ethics of belief are of a nature radically different than this.
If there is a valid philosophical interest in synchronic epistemic justification—and
I have not denied that this is the case—it is not in a direct sense the interest in what
epistemic justification is, nor is it the interest in the sense or senses of responsibility that
bear most directly on the ethics of belief. Especially when we attend to what we
actually have meta-cognitive control over, our most fundamental doxastic responsi-
bilities are diachronic; synchronic rationality is underwritten by responsibility in
cultivating and exercising virtuous doxastic habits.
Admittedly I offer no direct argument here for salience contextualism and for the
entangled view—pragmatic/diachronic encroachment upon what the evidentialist
understands as the purely epistemic (see Axtell 2010 for such a view). But the burden
of proof does seem to me largely to fall on those whose claim it is that the reduction is
possible: that we can always clearly factor out the purely epistemic considerations of ‘fit’
(as a descriptive state of affairs) in attributions of knowledge, from agent/attributor
interests and from evaluation of the agent’s own selected investigative methods and
strategies. Perhaps when readers receive from Feldman his long-anticipated theory of
evidence we will see just how this reduction is supposed to proceed in application to
cases like the theism/non-theism debate. But at present I would characterize Feldman’s
account of these issues in the ‘Ethics of Belief ’ chapter of Evidentialism and in essays like
‘Epistemic Obligations’ (1988) and ‘Epistemic Duties’ (2002) to be carried largely by
unargued assumption of a fact/value dichotomy, together with persuasive definition of
key concepts like ‘epistemic responsibility’ and ‘reasonable belief ’, persuasive defini-
tion that would have us relocate these concepts quite a distance from their broader
everyday connotations.
But to go further into the substance of the matter, our ‘too narrow’ objection to
their proposed restriction of epistemic value to maximizing synchronically rational
belief can be framed as a challenge to the conception of epistemological axiology with
which the evidentialists would have us begin. Conee and Feldman restrict the episte-
mic to the relation of beliefs to evidence. For them ‘the fundamental epistemic goal is
just to have reasonable beliefs’, and taking this to be the case they further posit that
‘nothing about evidence gathering or the like follows as a means to that goal’ (2004:
188). This approach to justification has deep roots in a Chisholmian conception of the
tasks of epistemology. But virtue epistemologies contest that conception, routed in
single-source epistemic value monism. They often posit that understanding, and not just
the veritistic goals of true belief and knowledge or the internalist goal of maximizing
synchronically rational beliefs, ranks highly in our epistemological axiology. If this
rebuttal is sound, this has direct bearing on the cogency of Feldman’s claim that
evaluation of the conduct of inquiry is a topic for a theory of practical reasoning but
‘irrelevant’ to a theory of epistemic justification. In any conception of epistemological
axiology that recognizes a more diverse range of epistemic goods, or that simply gives
pride of place to understanding rather than to propositional knowledge within our
cognitive economy, epistemic value will not be measured by synchronic rationality
alone. Although we have here been largely limited to a critique of Feldman’s eviden-
tialist epistemology, I have argued elsewhere more constructively for the genuine
contributions to epistemic value of the intellectual virtues, illustrating how the virtues,
as diachronic traits that bear upon the quality of our motivations and efforts at inquiry,
prefigure doxastic justification, and why they remain central to appraisals of agents that
matter most to us, philosophically and scientifically (Axtell and Olson 2009).
Arguments from the paucity of the evidentialist’s conception of epistemic axiology
and from the positive contributions of diachronic rationality to epistemic value will also
bear on Feldman’s and other philosophers’ endorsement of the Rational Uniqueness
Thesis (hereafter RUT). This is the claim that given a body of evidence, there is only
one reasonable doxastic attitude to take towards any proposition, where possible
doxastic attitudes include believing, disbelieving, and suspending judgment. Once
both degrees-of-belief and commitment-qualification through epistemic humility are
set aside, RUT becomes almost a foregone conclusion. In terms of fit or doxastic
attitude, there will always be one porridge that is too cold, another too hot, and a third
that is just right.
With RUT and the role it has recently played in assertion of the impossibility of
reasonable disagreement, we have to be cognizant of how its application already
presupposes cognitive evidentialism. Thus the explicit defense of RUT in Evidentialism
derives from thesis O2: that an agent ought always to have just exactly that attitude
towards a proposition supported by his/her evidence at that particular moment. This in
turn is best supported by V3: that being synchronically rational at every moment is
uniformly what it is to constitute epistemic success and to maximize epistemic value
(2004: 185, 258). But we have already given sufficient reason to doubt V3, namely its
dependence upon orthodox Chisholmian epistemic value monism. So to the extent
that we adopt the richer conception of epistemological axiology that I argue elsewhere
is a key lesson of recent work on the ‘value problem’ and value-driven epistemology
(see Axtell 2008), then V3 can be counted false and no longer provides the needed
support for O2. Thus RUT in turn, simply lacks the cogency with which much on the
recent literature on epistemic disagreement invests it. This would be a good thing, as
I’ll elaborate in Section 4.3.
So from the present perspective it can be concluded that at two levels—that of how
to maximize epistemic value when adopting an attitude towards a proposition, and that
of how the ethics of belief is claimed to be governed only by consideration of the
synchronic rationality of the agent’s doxastic attitudes—the evidentialist account looks
quite doubtful. If this is the case then as we began by saying, it can’t provide the
philosophic support that cognitive evidentialists rely upon it to provide in order to
motivate evidentialism in its other main aspect—as a normative ethics of belief and/or a
normative account of proper responses to peer disagreement/diversity.
4.3 Sour fruits

Thus far I have followed an approach of undercutting evidentialism as an ethics of
belief by criticizing its roots in the internalist evidentialist account of epistemic justifi-
cation. In this section we’ll ask whether the ethic of belief Feldman defends in
‘Reasonable Religious Disagreements’ and other recent papers is likely to be of value
or disvalue as a tool of mediation in our present-day culture wars, including especially
clashes of religious and secular culture.
In the paper we are focusing on, Feldman reflects upon a course he recently taught
in philosophy of religion, and attempts to show the falsity of ‘the tolerant and
supportive view’ (hereafter TSV) he says a good many of his students had, both towards
disagreement between believers and skeptics, and between (presumably moderate) ad-
herents of different world faith traditions. He frames his argument around two questions:
Q1 Can epistemic peers who have shared their evidence have reasonable disagree-
ments?
Q2 Can epistemic peers who have shared their evidence reasonably maintain their
own belief yet also think that the other party to the disagreement is also reasonable?
(2007: 201).
The burden Feldman places on his students to support TSV is not to argue just that
religious beliefs can be held by ‘generally reasonable folks’ but that people are ‘episte-
mically reasonable with respect to their specific beliefs’. He argues that while reason-
able disagreement (Q1), even with mutual acknowledgment (Q2), may appear initially
plausible, we must ultimately come to negative answers to both questions. Epistemic
peers who have shared their evidence cannot reasonably come to different conclusions:
That is, ‘there cannot be reasonable disagreements of the sort I was investigating’, and
‘I cannot make good sense of the supportive and tolerant attitude my students
displayed’ (2007: 213).
I want to defend the students’ intuitions, while also exploring the possibility of
turning Feldman’s or other RUT-defenders’ ‘unfortunate’ dismissal of reasonable
disagreement back into a pragmatic test of the value or disvalue of his ethics of belief
for mediating our most serious cultural debates. An old adage going back to Hume is
that errors in religion are dangerous; those in philosophy only ridiculous. I’ll suggest
that this is not so in the present case of using cognitive evidentialism plus RUT
to undermine reasonable disagreement among evidence-sharing epistemic peers!
My critique of the practical implications of accepting the position Feldman defends
begins with William James’ characterization of ‘faith-tendencies as extremely active
psychological forces constantly outstripping evidence’ (1979b [1911]: 112). The ‘faith-
ladder’, thought James, was no logical chain of inferences, but neither is it usefully judged
by the intellectualist standards of skeptical rationalism, that both demands establishing a
threshold of sufficient evidence for reasonable belief, and denies that that posited thresh-
old is ever met. Moral evidentialists like William Clifford (1999 [1877]) are discontents as
well, not of the evidentialist accommodation but of what Richard Rorty calls the Jeffersonian
Compromise: mutual acquiescence to viewing religious beliefs as private in a sense of
having little direct bearing on social/political order, but as still relevant to, and possibly
essential for, individual perfection. Jefferson ‘set the tone for American liberal politics’
when he wrote in Notes on Virginia that ‘It does me no injury for my neighbor to say that
there are twenty Gods or no God. It neither picks my pocket nor breaks my leg.’ Citizens
of Jeffersonian democracy, the compromise holds, ‘can be as religious or irreligious as they
please so long as they are not “fanatical”’ (Rorty 1991: 175).
The problem is that making this important qualification—‘so long as’—involves
noticing the differences between radical and moderate forms of religious fideism in
terms of the faith ventures they respectively authorize. Yet a cognitive evidentialist like
Feldman can no more acknowledge a difference between the epistemic responsibility
of moderate and radical fideists than can he acknowledge a difference between
reasonable and synchronically rational belief. There is no more principled basis in his
philosophy for the one distinction than for the other.
It is a practical and oftentimes a political decision explicitly to reject the Jeffersonian

compromise, whether in the direction of hegemony of the ‘home religion’ or in that of an
aggressive campaign of secularization or atheism. Conee and Feldman present themselves
as holding the view ‘suggested by the words of Locke and Clifford that
our epistemological duty is to believe as the evidence we have dictates’ (Feldman
2002: 364). But Locke and Clifford on closer examination are not nearly so closely aligned
as this suggests. Clifford’s argument was basically that ‘Belief, that sacred faculty which
prompts the decisions of our will, and knits into harmonious working all the compacted
energies of our being, is ours not for ourselves, but for humanity’ (Clifford 1999: 74).
Quite incompatibly with this claim, Locke’s conceptual pluralism is exemplified in his
Essay Concerning Human Understanding (1997 [1686]), where he writes that, ‘Since . . . it is
unavoidable to the greatest part of men, if not all, to have several opinions, without certain
and indubitable proofs of their truths, it would, methinks, become all men to maintain
peace and the common offices of humanity and friendship in the diversity of opinions.’
There are quite difficult issues that the epistemology of disagreement/diversity raises,
and it is to Conee and Feldman’s credit that they helped bring these issues to center-
stage in epistemology in recent years. But it is an ironic outcome if an approach
Feldman describes as in part motivated by Locke’s philosophy should come around
to problematize central values of Lockean reciprocal amity—the affirmation of ‘common
offices of humanity and friendship in the diversity of opinions’.
Let us investigate further whether this might indeed be the case, by focusing more
closely on Feldman’s denial of reasonable disagreement. The unfortunate fruits of
Feldman’s evidentialist account of the epistemology of disagreement might be pointed
out by noting certain literatures that RUT and its associated conception of reasonable
disagreement appear to have negative implications for. I will survey these literatures
briefly first, and then focus my strongest arguments around the third.
The first literature that evidentialism compromises is the work of philosophical and
religious pluralists who explicitly base pluralism upon the ambiguity of total public evidence
supporting religious and naturalistic ‘hypotheses’ (viewable as the ‘hard cores’ of very
large scale alternative research programs). This includes leading proponents of religious
pluralism (Basinger 2002; Bishop 2007), as well as philosophic pluralists who draw
upon the same assumption (McKim 2001; Bernstein 2003). Religious pluralism is a
theistic account, and I am not saying the evidentialist needs to accept its premises, but it
is certainly a perspective that might be informing the perspective of some of Feldman’s
students. For example, it is often possible to counter a leaning of intolerance or
disrespect by a ‘moral case’ against salvific exclusivism. ‘The idea that the most devout
and ethical persons of non-Christian faiths might find themselves in hell simply because
they do not believe one of the essential tenets of Christianity seems difficult to
reconcile with God’s moral perfection’, and motivates inclusivist and pluralist sympa-
thies (Himma 2002: 1). Now if we are epistemically tied to RUT, what Kenneth
Himma calls this ‘moral high ground case for salvific pluralism’ appears to be undercut,
the former working against the latter. The constructive work that recognition of an
epistemic situation of ‘religious ambiguity’ is supposed to serve in a pluralist conception
of religious truth and/or salvation falters, for the reason that the only entitled attitude to
hold towards a proposition one acknowledges isn’t unambiguous in evidence is,
according to Feldman, the attitude of strict suspension (as we’ll see more fully below).
The second literature potentially problematized by Feldman’s position in the epis-
temology of disagreement is the theory of deliberative democracy, by which I mean a
whole family of contemporary democratic theories that make central use of the
Rawlsian thesis of reasonable pluralism among people with respect to their comprehen-
sive religious, philosophical, moral, and political doctrines. Rawls describes reasonable
pluralism not only as a ‘fact’ about a democratic polity but also as the ‘long-run
outcome of the work of human reason under enduring free institutions’ (Rawls
1996: 129). This Rawlsian position called reasonable pluralism is a modern extension
of Lockean reciprocal amity. That the burdens of judgment arguments are ones Rawls
describes as epistemologically-driven indicates that deliberative democratists do make
demands upon the epistemology of disagreement/diversity and the ethics of belief. So
there is at least a deep tension here, one that Alan Hazlett (2009) puts in a quite useful
way by defining ‘Epistemic Liberalism’ (closely akin to the TSV claim of Feldman’s
students) as a contrary to RUT, and then simply asking whether post-Rawlsian
theories of political liberalism depend upon there being a connection between their
political theories and Epistemic Liberalism. If so, then it appears no self-respecting
deliberative democratist can be an evidentialist.
One might object that while there is tension, there is no genuine contradiction here:
Our commitments to toleration, as civil or political commitments are of a different order,
and are not impugned even if the epistemological underpinnings Rawls thinks it impor-
tant to give to reasonable pluralism turn out to be unsound. But for the sake of
argument, Jean Jacques Rousseau considered our problem and wrote, ‘Those who
distinguish civil from theological intolerance are, to my mind, mistaken. The two
forms are inseparable. . . . Wherever theological intolerance is admitted, it must inevi-
tably have some civil effect’ (1987 [1762]: 134). Evidentialists should not be flippant
about the impact of accidentally undermining epistemological arguments that much
modern deliberative democracy appears to lean upon. The tension, as I think attention
to our third literature shows, exists not just between RUT and these Rawlsian
arguments, but between RUT and proper recognition of deliberative virtues such as
‘reciprocity’, ‘magnanimity’, ‘openness’, themselves. But these are difficult questions
that bear further study, as does the epistemology of liberal democracy more generally.
The final literature and the one I want to argue Feldman’s epistemology of disagree-
ment/diversity most clearly compromises, is a literature focused around ‘friendly the-
ism/atheism’, and religious toleration though epistemic humility (see Greco 2008;
Kraft and Basinger 2008). These authors argue in favor of ‘friendly theism’ and ‘friendly
atheism’ supported through epistemic humility, against the non-reciprocating ‘un-
friendly’ versions of each that tend to predominate in the polemical discourse of our
present-day ‘culture wars’ over reason and faith, science and religion. This is more
straightforwardly a matter of prescribed attitude towards disagreement among peers, and is not
complicated by the distance between one’s epistemology and one’s political life. The
clearest support I could give to my claim that Feldman’s ethics of belief is of disvalue in
mediating these culture wars would be reasons to think that the combination of
cognitive evidentialism and RUT in fact destabilizes the ‘friendly’ versions of both
theism and atheism while empowering the alternate, ‘unfriendly’ versions of each. To
show this let’s now return to look more directly at the reasons behind the negative
answers Feldman provides to his own Q1 and Q2.
The argument that Feldman gives for his skepticism about reasonable disagreement
refers us directly back to RUT, and to that internalist account of epistemic justification
and value which we directly sought to undercut in the previous section. The crucial
claim Feldman derives from it is that there are ‘really only two’ potentially philosophi-
cally respectable responses to serious peer disagreement, those which he terms the Hard
Line model (hereafter HL), and the Modest Skeptical model (hereafter MS). Let’s
consider these two evidentialist-approved ‘responses’ to peer disagreement, and the
forced option Feldman alleges between them.
HL: The Hard Line view
The first view or model that Feldman says may be the distinctly rational one to take
towards your disagreement with another individual is the model that sanctions the
response of, ‘I’m reasonable; you’re not’. Disagreement under conditions of shared
public evidence indicates not just error, but indeed the epistemic irrationality or
unreasonableness of one of the disputants. ‘The hard line says that the evidence they
share really must support one view or the other, and the one whose belief fits the
evidence is the rational one’; hence ‘Either the theists or the atheists are rational, but
not both. There can be no reasonable disagreements’ (2007: 211).
The HL model is quintessential evidentialism, and while Feldman might not
approve (Feldman describes himself as a ‘complacent atheist’, not an ‘aggressive’
one), a prominent example of the HL model put to work in popular culture wars is
Sam Harris’s Letter to a Christian Nation, where at the outset he sets aside all Christian
liberalism (as inconsistent non-sense), in order to pick out as his target audience just
such people as really know what Christianity is:
We agree, for instance, that if one of us is right, the other is wrong. The Bible is either the
word of God, or it isn’t. Either Jesus offers humanity the one, true path to salvation, or he does
not. We agree that to be a true Christian is to believe that all other faiths are mistaken, and
profoundly so. (Harris 2006: 3).
Briefly, the main problem with HL that we’ll return to is the one Peter van Inwagen
(2010) worries over in ‘We’re Right; They’re Wrong’, that sanctioning this model/
attitude towards disagreement as the default view and as a kind of dictate of reason may
sound good in theory, but in practice lends itself to rationalizing dismissals of the
reasonableness of those we disagree with (and all the better with moderates summarily
dismissed). With the sanction of the HL attitude behind us, do we not often come to
expect more intellectual virtue of others than we do of ourselves? Another worry is that
Harris’ employment of an HL rhetorical strategy arguably encourages its Christian
evidentialist mirror image, characterized by equally aggressive intelligent design theory
and other examples of overt religious rationalism such as that by disinterested fact-
weighing it is ‘overwhelmingly probable’ that Jesus was God incarnate (Swinburne
2003). Moreover, is it even possible to define concepts like religious ‘fundamentalism’
and religious ‘exclusivism’ cross-culturally—that is in terms of a shared manner or form
of belief despite divergent belief content—without reference to an assumed HL attitude
towards disagreement and towards outsiders to the home religion?
MS: The Moderate Skeptical view
While Feldman allows no principled distinction between strictly empirical debates and
those over comprehensive doctrines, he does to his credit take note of his TSV
students’ objections that the ‘fit’ of the HL model will be harder to defend in cases
in which people on both sides of a debate have been reflective and have openly
discussed (shared) their reasoning and experiences. Perhaps, then, ‘In these moral,
political, scientific, and religious disputes, it is implausible to think that one side is
simply unreasonable’ (2007: 211). Feldman is not here conceding that he has personally
become convinced to move off of his HL atheism, but in the event that he were to be
convinced of it, the move is not first towards a ‘friendly’ atheism though epistemic
humility. Rather, he now claims, the Moderate Skeptical view becomes the singularly
intellectually responsible and hence obligatory response to adopt.
This second of the two evidentialist-approved ways of responding to peer disagree-
ment is the one a person is intellectually obliged to conform to whenever there is
recognition or even serious suspicion of vagueness or ambiguity of total public evidence
on the claim in question. Since RUT tells us it is epistemically wrong to believe a
proposition when one’s total evidence fails to decisively support that proposition, ‘the
right thing for both of us to do is to suspend judgment’ (2007: 212). Because ‘the right
thing’ here is presented as singular and obligatory, MS runs directly in conflict with
literatures one and two in the philosophy of religion, noted above (while HL appears to
conflict with all three). Suspending judgments we are given to believe is the only way to
be epistemically ‘successful’ in the moment, where evidential ambiguity presents itself.
So if ‘there are really only two’ options, HL and MS, because reasonable belief is just
evidentially justified belief and there always either is or isn’t ‘sufficient’ shared evidence
for evidentially justified acceptance of whatever proposition is in question, then surely
we have to answer ‘No’ to Q2, and then to Q1 as well. It may be that MS and not HL
holds in the case of comprehensive doctrines, but TSV, the tolerant and supportive
view, is nevertheless now exposed as a confused attempt to place sentiment and
political correctness above plain reason.
I want to suggest that we ‘return to reason’ not when we deny this logic, but when
we actually stand it on its head: Rather than accepting the forced option he presents to
his students between the HL and MS responses to deliberative disagreement, we should
straightforwardly re-describe this ‘forced option’ as thehorns of a dilemma that Feldman

and other cognitive evidentialists impale themselves upon when they try to apply their epistemic
principles to the real world of comprehensive philosophical, moral and religious doctrines. The
Feldmanian over-extension of the RUT thesis places the evidentialist (and would if we
allow, place all of us) in just those states of affairs often found precursory to war, when
‘Things fall apart; the centre cannot hold’; his ethic of belief evokes such circumstances
where, as the poet W. B. Yeats bemusedly put it, ‘The best lack all conviction [MS],
while the worst are full of passionate intensity’ [HL].
This Rational Uniqueness Dilemma, as we can term it, depicts the defenders of RUT
and its close cousins as displaying a self-defeating intellectualism, one in which they
find themselves needing to embrace one or the other of two quite unwelcome
consequences: (a) submitting to what seems the intellectual thanatos of applying the
Hard Line’s logic of exclusivism across the board to disagreements over religious, philo-
sophical, political, and moral ‘comprehensive’ doctrines (to the destabilization of
Lockean reciprocal amity and ‘friendly’ theism/atheism) or (b) submitting themselves
to the ‘spinelessness’ of strict suspension in all cases where evidential ambiguity or
conceptual vagueness is apparent (as well as disappointing the Rawlsian expectation
that a kind of conceptual pluralism is the natural outcome of an advancing liberal society).
Consider more closely what it means to grab the first horn. The unwelcome
consequence here is essentially the same that McKim articulates when he writes that,
‘Advocates of large-scale systems of beliefs that include discrediting mechanisms are not
in a position to appreciate the appeal to systems of belief they think to be discredited.
They are not likely to be able to give them a fair or sympathetic hearing’ (2001: 152).
The HL’s logic of exclusivism is one that liberalism’s discontents known well, one that
allows easy discrediting of non-true-believers. Gutmann and Thompson argue in Why
Deliberative Democracy that ‘Deliberation cannot make incompatible values compatible,
but it can help participants recognize the moral merit in the opponent’s claims when
those claims have merit’ (2004: 11). By contrast, prescribing HL as the proper response
for disputants to take in disagreement among comprehensive doctrines just appears to
crown as the height of reasonableness such judgments of the culpable failures of others
that Rawls described as typically the result of ‘prejudice and bias, self and group
interest, blindness and willfulness’ (Rawls 1993: 249).
So yes, of course, we can follow the evidentialist’s and religious fundamentalist’s shared
logic of exclusivism and just ‘bite the bullet’ of this first horn; the question is whether we
can do so without shooting ourselves in the head. But perhaps I am presumptuous in
asserting this ‘we’. For the discontents of the Evidentialist Accommodation and of the
Jeffersonian Compromise we are now seeing, are many and varied, and here make strange
bedfellows. Part of what they have in common is that many seem not to really want the
Jeffersonian middle to hold, except perhaps (as one would hope) in the thinnest, civic way,
or until conditions ripen for their alternative view of the ‘new order’, whether religious,
secularist, or secular humanist. To be a discontent as I intend it, however, is to be repelled
by the whole state of our cultural debate over reason and faith dominated by just such
extremes of religious and skeptical rationalism. It is to be dismayed with the aggressive

secularizers like Harris as well as with their Christian evidentialist mirror image.
But grabbing the second horn, encompassing not increased epistemic humility about
our belief but a change from full belief to suspension, is arguably even less appealing
if it is indeed psychologically plausible to begin with (extricatable from strong volun-
tarist assumptions). ‘Do you have any convictions on controversial political, philosophical,
or scientific matters? The equal weight view seems to say: kiss them goodbye’ (Elga 2007:
484). There are different forms of skepticism, and this one appears risk-averse in such a
radical way as to be open to the charge of being unlivable. Instead of describing this ‘equal
weight’ model as ‘moderate’, as Feldman does, its critics seem to me more apt in their
description of it as demanding ‘dispiriting spinelessness’ (Sosa 2008) and/or ‘lack of self-
trust’. ‘It is implausible that rationality requires such spinelessness’ (Elga 2007: 484).
I think this equal weight view in the epistemology of disagreement results from
disregard for the role of doubt in the history of social philosophy, especially in the
early-modern period of Erasmus, Montaigne, and Locke where the theory of tolera-
tion was re-born. The nouveau Pyrrhonisme of Montaigne and Erasmus, Toulmin
reminds us in Cosmopolis, ‘was no “negative dogmatism” which systematically refuses
to accept everyone’s right to opinions arrived at by honest reflection on first-hand
experience’ (1992: 50). Doubt had to be a multi-edged sword for those who lived
through the roughest centuries of intra-Christian warfare, serving the pragmatic aims of
moderation and mediation against extremes on all sides.
Feldman’s students or other supporters of TSV, I conclude, have plenty of resources
to defend reasonable disagreement. In making our argument we have not impugned
the importance of intellectual responsibility with respect to evidence. We have how-
ever applied our earlier critique of the internalist evidentialism’s claim to provide an
adequate account of intellectual responsibility. If his students aren’t convinced of
internalist evidentialism as an account of epistemic responsibility, they won’t accept
his account of epistemic obligation and the forced option he presents it as issuing in.
They won’t equate reasonable belief with synchronically rational beliefs, the diachron-
ic activities of the agent set aside or bracketed. While they needn’t assume any kind of
relativism or deny genuine disagreement, they will reject all attempts to apply RUT
unqualifiedly to debates like those between theists, atheists, and agnostics. Accordingly
they will neither conceive of the Hard Line view over such issues as the sin qua non of
rationality, nor accept the claim that universal suspension of belief is the only reasonable
response to the acknowledged ambiguity of total public evidence.
4.4 Conclusion: towards a responsibilist

accommodation
I have argued that the best-known version of evidentialism as an ethics of belief and
position in the epistemology of disagreement/diversity has both weak roots and sour
fruits. The pragmatic critique of Section 4.3 was intended only to supplement the
epistemological objections of Section 4.2. The dilemma just presented to Feldman is
meant to illustrate just how far his evidentialist philosophy, purportedly initially
motivated by important insights of John Locke, has come back around to problematize
two key themes of Locke’s social philosophy: his conceptual pluralism and his defense
of toleration (reciprocal amity) as a value and a virtue.
The lack of any genuine positive epistemic invitations or entitlements to believe that
are not merely obligations to believe, disbelieve, or suspend, has been another source of
my own discontent with the evidentialist accommodation: its lack of recognition of
what I think of as our need of and intellectual right to various ‘doxastic ventures’,
philosophic, religious, or political; its evident lack of that ‘spirit of inner tolerance’ for
another’s mental freedom, without which William James warned that our outer (social,
political) tolerance is likely to be or to become unstable.
Should the evidentialist accommodation between private persons and the collective
community of inquiry also appear overly restrictive or ‘Victorian’ on this score, the
resources of virtue theory and the prospects of an alternative responsibilist accommodation
remain available to explore. On this alternative accommodation as I propose it, it is
diachronic rather than synchronic evaluation of agents that most directly informs a sound
and civic ethics of belief. Diachronic rationality’s contribution to epistemic value was
the upshot of our epistemological critique, but has quite direct implications for our
understanding of the ethics of belief as well. It implies that there may well be invitations
of sorts—simply put, things an intellectually virtuous believer might believe, that are
things not all virtuous believers in this self-same evidential situation necessarily would
believe. But any right to be choosers of our risk is not on the virtue-theoretic view pre-
critically given to us. We must earn our intellectual right to vary from the evidentialist
norm of belief, though it comes far more readily in respect to weltanschaulich questions
and beliefs as they arise in fields like politics, philosophy, and religion. We earn this
right to maintain old or initiate new doxastic ventures—what Mill and James called our
various ‘experiments in living’—through intellectually conscientious efforts at inquiry, and
through an habituated sensitivity to potential defeaters, defeaters that enliven an agent
to the kind or level of confidence that they have in their belief, and to diachronic
responsibilities to extend, revisit, or update the reflectively good reasons they could
muster to support that confidence.
References
Axtell, G. 2008. ‘Expanding Epistemology: A Responsibilist Approach,’ Philosophical Papers, 37,
1: 51–88.
——. 2010. ‘Agency Ascriptions in Ethics and Epistemology,’ Metaphilosophy, 41, 1–2: 73–94.
——. and J. Adam Carter. 2008. ‘Just the Right Thickness: A Defense of Second-Wave Virtue
Epistemologies,’ Philosophical Papers, 37, 3: 413–34.
——. and P. Olson. 2009. ‘Three Independent Factors in Epistemology,’ Contemporary Pragmatism,
6, 2: 89–109.
Basinger, D. 2002. Religious Diversity: A Philosophical Assessment. Farnham, Surrey: Ashgate.
Bernstein, R. 2003. ‘Pragmatism’s Common Faith,’ in S. Rosenbaum (ed.) Pragmatism and
Religion. La Salle: Illinois University Press, 129–41.
Bishop, J. 2007. Believing by Faith: An Essay in the Epistemology and Ethics of Religious Belief.
Chappell, T. 2009. ‘On Not Proportioning your Belief to the Evidence,’ (typescript).
Clifford, W. K. 1999 [1877]. ‘The Ethics of Belief,’ reprinted in The Ethics of Belief and Other
Essays. Amherst, NY: Prometheus Books.
Conee, E. 2010. ‘Rational Disagreement Defended,’ in R. Feldman and T. Warfield (eds),
Disagreement. Oxford: Oxford University Press, 69–90.
—— and Richard Feldman. 2004. Evidentialism: Essays in Epistemology. Oxford: Clarendon.
Elga, A. 2007. ‘Reflection and Disagreement,’ Noûs 41, 3: 478–502.
Feldman, R. 1988. ‘Epistemic Obligations,’ Philosophical Perspectives 2: 235–56.
——. 2002. ‘Epistemological Duties,’ in Paul Moser (ed.), Oxford Handbook of Epistemology.
Oxford: Oxford University Press, 362–84.
——. 2006. ‘Epistemological Puzzles about Disagreement,’ in S. Heatherington (ed.) Epistemo-
logical Futures. Oxford: Oxford University Press, 216–36.
——. 2007. ‘Reasonable Religious Disagreements,’ in L. Antony (ed.), Philosophers without Gods.
Oxford: Oxford University Press, 194–214.
Greco, J. 2005. ‘Justification is not Internal,’ in M. Steup (ed.), Contemporary Debates in Episte-
mology. Oxford: Blackwell, 257–70.
——. 2008. ‘Friendly Theism,’ in J. Kraft and D. Basinger (eds), Religious Tolerance through
Humility. Burlington, VT: Ashgate, 51–60.
Gutmann, A. and Thompson, D. 2004. Why Deliberative Democracy? Princeton: Princeton
University Press.
Harris, S. 2006. Letter to a Christian Nation. New York: Alfred A. Knopf.
Hazlett, A. 2009. ‘Faith and Liberalism.’ [typescript]
Himma, K. 2002. ‘Finding a High Road: The Moral Case for Salvific Pluralism,’ International
Journal for Philosophy of Religion, 52: 1–33.
James, J. 1979a. [1908] The Will to Believe and other Essays in Popular Philosophy. Cambridge:
Harvard University Press.
——. 1979b [1911]. The Problems of Philosophy. Cambridge: Harvard University Press.
Kraft, J. 2006. ‘Religious Tolerance through Religious Diversity and Epistemic Humility,’
Sophia 45, 2, 101–16.
Kraft, J. and Basinger, D. 2008. Religious Tolerance through Humility. Farnham, UK: Ashgate.
Locke, J. 1997 [1686]. Essay Concerning Human Understanding. London: Penguin Classics.
McKim, R. 2001. Religious Ambiguity and Religious Diversity. Oxford: Oxford University Press.
Pappas, G. 1994. ‘William James Virtuous Believer’ Transactions of the Charles Peirce Society, 30, 1:
77–110.
Rawls, J. 1993. Political Liberalism. New York: Columbia University Press.
Rawls J. 1996. Political Liberalism. New York: Columbia University Press.
Rorty, R. 1991. Objectivity, Relativism, and Truth: Philosophical Papers, Volume 1. Cambridge:
Rousseau, Jean Jacques. 1987 [1762]. On the Social Contract. Indianapolis, IN: Hackett Publishers.
Sosa, E. 2008. ‘The Epistemology of Disagreement,’ in A. Haddock, A. Millar, and
D. Pritchard (eds), Social Epistemology. Oxford: Oxford University Press, 278–97.
Swinburne, R. 2003. The Resurrection of God Incarnate. Oxford: Oxford University Press.
Toulmin, S. 1992. Cosmopolis: The Hidden Agenda of Modernity. Chicago: University of Chicago
Press.
——. 2001. Return to Reason. Cambridge, MA: Harvard University Press.
Turri, J. 2010. ‘On the Relationship Between Propositional and Doxastic Justification,’ Philoso-
phy and Phenomenological Research, 80, 2: 312–26.
Van Inwagen, P. 2010. ‘We’re Right. They’re Wrong,’ in R. Feldman and T. Warfield (eds),
Disagreement. Oxford: Oxford University Press, 10–28.
Williams, M. 2008. ‘Responsibility and Reliability,’ Philosophical Papers, 37, 1: 1–26.
Wood, J. 2006. ‘Intellectual Virtues and the Prospects of a Christian Epistemology,’ in
M. Cherry (ed.), The Death of Metaphysics; the Death of Culture. Dordrecht: Springer, 41–69.
Zagzebski, L. 2007. Philosophy of Religion. Oxford: Blackwell.

Guy Axtell Work

Uploaded by

Copyright:

Available Formats

Guy Axtell Work

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Guy Axtell Work

Uploaded by

Copyright:

Available Formats

Virtue Epistemology:

OXFORD UNIVERSITY PRESS

The problem of restoring integration and cooperation between man’s

I. Luck, Control, and Virtue Theory

epistemic luck in light of the virtues 159

160 virtue epistemology

epistemic luck in light of the virtues 161

Such a compatibilist stance in epistemology, to continue the analogy, is simply one

162 virtue epistemology

epistemic luck in light of the virtues 163

II. Forging Bridges: Riggs’s

164 virtue epistemology

epistemic luck in light of the virtues 165

166 virtue epistemology

III. Skepticism and the Incompatibility Thesis

epistemic luck in light of the virtues 167

168 virtue epistemology

epistemic luck in light of the virtues 169

IV. The Gaps Problem: Dealing with the

170 virtue epistemology

epistemic luck in light of the virtues 171

172 virtue epistemology

epistemic luck in light of the virtues 173

Non-consequentialists also are unwilling to admit that the consequentialists are

174 virtue epistemology

epistemic luck in light of the virtues 175

176 virtue epistemology

epistemic luck in light of the virtues 177

A Defense of Second-Wave Virtue Epistemology

Guy Axtell & J. Adam Carter

Abstract: Do the central aims of epistemology, like those of moral philosophy,

2.1 The Nature and Scope Argument

Nature and Scope Argument

Centralism is at least a contested thesis in philosophy, and in our view it is false in

3. The Second Wave of Virtue Epistemology

Engel, P. “Epistemic Norms and Rationality” [2007, typescript].

Greco, J. “The Nature of Ability and the Purpose of Knowledge,” [typescript]

If the recent but quickly-growing body of literature on epistemological axiology

can serve as an indication, there are changes brewing in the conception of

epistemology and its central tasks. By a literature on epistemological axiology, I

nature and sources of epistemic value, or the ‘value-driven’ approach to

more basic philosophical questions. In periods in any field of research where a

long-standing paradigm or dominant mode of practice is being questioned and

methodological and axiological assumptions animating that field of study.

which saw lively debate over theories (foundationalism vs. coherentism;

internalism vs. externalism), but a comparatively strong consensus, relative to

theories functioned at the start of this era to expand epistemology in important

knowledge-attributions and discursive entitlements need to be considered in

assumptions were held steady as inviolable, such as a semantic notion of truth

giving a theory of justification or on immunizing a proposed analysis of

largely pushed aside, or even treated as conceptual confusions. 1

In this paper I treat newfound interest in epistemological axiology as a clear

example of non-allegiance to long-standing assumptions about methods and

tradition, but that axiologically re-cast what is still of value the

coherentism/foundationalism and internalism/externalism debates. The

expansions I envision would share some of experimental philosophy’s criticisms of

the analytic tradition’s a prioristic tendencies and ‘armchair’ methods, without

succumbing to what Michael Williams (this volume) refers to as ‘radical

reliabilism.’ My challenge to readers of this paper is to think seriously both about

particular directions that expansion should take.