Virtue, Knowledge and Wickedness: Ronald Milo on McDowell's internalist moral realism

“Is it possible for a person to understand that what he proposes to do is morally wrong and yet prefer to do it nonetheless?”

Ronald D. Milo opens his 1998 paper “Virtue, Knowledge and Wickedness” with this question, to which a diversity of answers appeal to a similarly diverse slew of (in several cases, incompatible) intuitions about moral phenomenology.

John McDowell’s internalist moral realist commitments—which are at the brunt of most of Milo’s critical attention—lead him (McDowell) to answer the question in the negative. Milo thinks that this response is mistaken and suggests that a flaw in McDowell’s thinking lies, in part, in McDowell’s obfuscation of moral convictions being intrinsically motivating and necessarily motivating.

First, though, I think it is important to try to get clear on what the premises are on which John McDowell relies in reaching this conclusion. Milo characterizes McDowell’s position as subscribing to the following claims (the first two representing the internalist realist framework within which he addresses the question posed at the outset):

McDowell’s internalist realist assumptions:
• The motivational force of beliefs is internal to them
• Moral properties are real, as opposed to merely “projected”


(1) A morally virtuous person exhibits two important capacities:
a. For all situations that call for a response of feeling or action on the part of the agent, the agent is able to recognize what is morally called for in that situation (Virtue and Vice, ed. Paul, Miller, Paul, p. 197).
b. The recognition in (a) provides the agent, in and of itself, with motivation sufficient to ensure that she acts accordingly (197) [(It follows from 3b that) no corresponding desire is required to motivate action; such desires might exist as upshots of the agent’s “acknowledging that her recognition that a certain act is morally called for is sufficient to motivate her”(198) whilst themselves not being necessary to motivate action.
(2) If a person is not morally virtuous, then a person cannot recognize what is morally called for in a given situation. (From 1)
(3) Those who are morally wicked cannot recognize what is morally called for in a given situation. (From 1 and 2)

The particularly strong claim in McDowell’s argument is 1b, which maintains that a virtuous person, upon recognizing some act θ to be morally required, is guaranteed to θ; this is, McDowell thinks, because this recognition will silence any competing desires. McDowell is, then, more in line with Aristotle than Kant in the sense that McDowell precludes the possibility of a morally virtuous meeting morality’s demands whilst desiring to do otherwise (while Kant finds such actions to merit the highest of praise). The recognition by a virtuous person that an action is morally required, by virtue of silencing competing desires, becomes what McDowell brands “intrinsically” motivating. This sort of move must be made for cognitivists like McDowell who must explain moral motivation without recourse to desires.

Milo, though attracted to some features of McDowell’s account, finds it unconvincing. Specifically, Milo’s problem is that McDowell (he thinks) gives no good reason to think that a conclusion about the recognition of a fact being intrinsically motivating for all people would follow from the recognition of a moral requirement’s being intrinsically motivating for a decisively virtuous person. (Indeed, McDowell does think that, for all agents, if a given agent recognizes that an action is morally required, then she is guaranteed to act accordingly. Wicked individuals, ipso facto, must have not really recognized what is morally required as such--otherwise, McDowell would offer, they would have met morality’s demands)

Milo rebuts McDowell’s reasoning with a counterexample of the following form: it depicts the recognition of something’s being the case as intrinsically motivating for a subject S, given antecedent facts about S, but not intrinsically motivating for Z, given antecedent facts about Z. Milo writes:

“Thus, one might explain a person’s desire for sweets as consisting simply int eh fact that, for this person, the conception of, e.g., candy as sweet is sufficient to motivate him to eat it. If this is so, then we can explain the difference between someone who likes the taste of bananas and someone who does not as follows: for the former, the thought that the object being offered to him is a banana is sufficient to motivate him to eat it, other things being equal, whereas for the latter this same thought fails to be motivating. For the person who likes the taste of bananas, the thought of eating one is intrinsically motivating. We need not conclude from this, however, that this thought is also necessarily motivating, in the sense that anyone who has the same thought must be motivated by it. Why not explain the difference between the virtuous person and the wicked person in the same way?” (201)

And so, Milo’s argument seems to be:

1. If the conclusion that bananas are necessarily motivating cannot be drawn from the fact that, for those who desire bananas, the thought of bananas is intrinsically motivating, then the conclusion that the recognition of an action as being morally required is necessarily motivating cannot be drawn from the fact that, for those who desire to act morally, the recognition of an act’s being morally required is intrinsically motivation.
2. The conclusion that bananas are necessarily motivating cannot be drawn from the fact that, for those who desire bananas, the thought of bananas is intrinsically motivating,
3. Therefore, the conclusion that the recognition of an action as being morally required is necessarily motivating cannot be drawn from the fact that, for those who desire to act morally, the recognition of an act’s being morally required is intrinsically motivating.

If Milo’s argument is sound, then 1b of McDowell’s argument must be rejected. Milo anticipates some responses that McDowell could give to the conclusion reached in the “banana argument.” He mentions a response by David McNaughton, who defends a breed of McDowellian internalist realism, which proffers the question: “there is something odd in the idea that an agent might recognize that he is morally required to do something and yet not believe that he has good reason to do it…for in what sense could he be said to recognize it as a requirement” (202). If the oddness of that supposition is “sufficiently odd”, then we are perhaps on our way to isolating some relevant difference about recognition of moral requirements and recognition about (say) what a banana smells like that justifies claiming that the former recognition is necessarily motivating and the latter not. Milo, however, isn’t swayed by the force of McNaughton’s appeal to oddness. “I must confess that I myself find nothing odd about this. Why should we think it odd if a wholly self-centered person sees no reason at all not to do something the he recognizes to be unjust but to his advantage?” (202)

Milo proceeds to address quite a few interesting features of McDowell’s position, all the while stressing that we have no good reason to think that a morally wicked individual could not recognize an action as morally required whilst preferring not to do it.

Milo finally won me over, but not until he pulled out the “depression” card. Milo “sneaks this card” into the deck not all at once, but by first discussing a form of temporary depression that Jonathan Dancy has called “accidie.” Dancy writes: “People who suffer from accidie are those who just don’t’ care for a while about things which would normally seem to them to be perfectly good reasons for action.” (210). Dancy adds there might, in cases of accidie, effect a loss of the normal motivational force of one’s beliefs; these are cases in which “the depressive is not deprived of the relevant beliefs by his depression, they just leave him indifferent” (210).

Milo (I think appropriately) pushes Dancy’s accidie a step further and suggests that any position is dogmatically unfounded if it rules out, a priori, the possibility of accidie being permanent. The permanency of accidie would land such a person an “amoralist” in the sense of being capable of recognizing conduct to be morally required but not caring (or at least, not caring enough to act accordingly).

Let us return now to the initial question:

“Is it possible for a person to understand that what he proposes to do is morally wrong and yet prefer to do it nonetheless?”

If there exist individuals in the state of permanent accidie (a state which, as I understand, would be a consequence of most sorts of clinical depression), then the answer to the question should be “yes.” And in answering the question “yes,” it follows that recognition of an action as being morally required is not intrinsically motivating for all. And so, in conclusion, we wouldn’t have a good reason to think that a wicked person is incapable of recognizing an action as being morally required.

Michael J. Zimmerman and the Puzzle of Moral Luck and Responsibility

In his 1987 article “Moral Luck and Responsibility,” Michael J. Zimmerman tries his hands at what he calls a “puzzle” that has, in varied forms, been tackled by Joel Feinstein, Thomas Nagel and Bernard Williams. The puzzle (labeled such by virtue of the fact that the premises appear true and the conclusion which follows from them false):

Moral Luck Puzzle:

P1. A person P is morally responsible for an event e’s occurring only if e’s occurring was not a matter of luck.
P2. No event is such that its occurring is not a matter of luck.
∴ No event is such that P is morally responsible for its occurring.

Zimmerman mentions exegetically that Feinberg provisionally accepts the conclusion; Williams accepts P2 while rejecting P1; Nagel (according to Z), in determining the puzzle to be a genuine paradox, “seems prepared to accept both premises while denying the conclusion” (Ethics, Vol. 97, No. 2 (Jan., 1987) pp. 374.

Zimmerman’s own strategy employs a preliminary breaking down of the puzzle by distinguishing two types of luck and two types of control. His luck types are:

1. Situational Luck: “Luck with respect to the situations one faces, including the nature of one’s character (inclinations, capacities and so on) as so far formed.
2. Resultant Luck: “Luck with respect to what results from one’s decisions, actions and omissions” (376).

In articulating what luck is, Zimmerman says: “Something which occurs as a matter of luck is something which occurs beyond anyone’s control”; (and, in a footnote to this definition) “More restrictively: something which occurs as a matter of luck with respect to someone P is something which occurs beyond P’s control” (fn. 376).

Zimmerman then breaks down the concept of “control” which is operative in his definitions of both situational and resultant luck. His types are:

1. Restricted control: “One may be said to enjoy restricted control with respect to some event just in case one can bring about its occurrence and can also prevent its occurrence.
2. Unrestricted control: “One may be said to enjoy unrestricted control with respect to some event just in case one enjoys or enjoyed restricted control with respect both to it and to all those events on which its occurrence is contingent. (376)

I will not go into the details of Zimmerman’s entire proposed solution to the Moral Luck Puzzle in light of these distinctions, but rather, want to focus on a particular feature of his argument. But before doing this, I will give an example that I think will capture the move he wants to make.
Suppose Susie is drowning in a pond in the dark and shouting for help. Nathan and Zeb are both standing on opposite banks, equidistant to Susie, and unaware (because of the darkness) of each other. Both jump into the pond at the same time and begin swimming toward Susie. When they are both halfway to Susie (respectively), Zeb snags his foot on an underwater branch (which he had no way of knowing was there), and by the time he frees himself, Nathan has already corralled Susie back to the bank.

If we grant what Zimmerman takes to be an undeniable premise—that an agent is not praiseworthy or blameworthy for an action unless that agent is responsible for the action—then the troublesome feature of the moral luck puzzle emerges, namely, that the responsibility for saving Susie that is a necessary condition for praise or blame for the action is perhaps threatened by the uncontrollability of getting one’s foot trapped on an underwater branch en route to the rescue.

Those who accept an unqualified reading P1 of the Moral Luck Puzzle and grant that Nathan was lucky not to have snagged his foot as Zeb did, will find themselves disinclined to praise Nathan for his action (the conclusion that would follow). Others might embrace a consequentialist account of praise or blame and thus reject P1 and praise Nathan whilst not praising Zeb. Zimmerman’s position takes the line of praising both equally. His doing so is a function of his distinction between the sort of lack of control that is relevant to precluding responsibility and the sort of lack of control that is not. And so, in essence, Zimmerman dodges the counterintuitive conclusion of the Moral Luck Puzzle by rejecting an unqualified reading of “luck” in P1.

Given this background, I want to focus on a particular aspect of Zimmerman’s position, which pertains to a principle he accepts with regard to decisively “situational” luck. Zimmerman claims that:

SL: If (i) P made decision d in what he believed to be situation s,
(ii) P* would have made d if he had been in a situation that he believed to be s, and
(iii) P*’s being in a situation that he believed to be s was not in his restricted control, then:
whatever moral credit or discredit accrues to P for making d accrues also to P*. (381)

Applying this principle to the Nathan and Zeb case I gave (which, by Zimmerman’s definition, would be an instance of “situational luck”) appears to generate Zimmerman’s position that both are equally praiseworthy.

While I am sympathetic to Zimmerman’s response to this case, I think that his principle SL runs into some troubles. This is because the conclusion that whatever moral credit or discredit accrues to P for making d accrues also to P* depends on whether the counterfactual in (ii) of SL holds true. And so, whether Zeb deserves equal praise as Nathan would depend on whether Zeb, if he believed he were in Nathan’s position (on the other side of the bank), would have decided to do what Nathan did.

In the effort of making things clear by first making them messy, I want to add a few more details to our story. Suppose the following is true about Nathan and Zeb: The bank from which Nathan jumped was the East bank. The bank from which Zeb jumped was the West bank. Prior to jumping in for Susie, Nathan believed that there were dangerous underwater stumps on the West side, but not on the East side. Additionally, if Nathan had been on the West side (where he believed that there were underwater stumps), he would not have jumped in to save Susie. Zeb, on the other hand, believed that there were dangerous underwater stumps on the East side, but not on the West side; if Zeb had been on the East side (where Nathan was), he would not have jumped in to save Susie.

The problem for Zimmerman’s view on situational luck is that, given our expanded story, he is prevented from ruling Zeb’s action as equally praiseworthy as Nathan’s because (ii) in SL is not met; (ii) in SL fails to be met because if Zeb believed he was in situation s (Nathan’s situation, on the East side of the bank), he would not have decided to do what Nathan decided to do, namely, to jump in for Susie.

Surely, though, Zimmerman would not want to accept the conclusion that Nathan is more praiseworthy than Zeb. For one thing, Both Nathan and Zeb would be prepared to jump in to save Susie from the bank that each thought was safe, and neither would have jumped in from the side each thought was unsafe; additionally, it was outside of Zeb’s restricted control that the underwater logs weren’t on the side of the pond he believed them to be.

I think that any account of moral credit that does not weigh whether Susie is in fact saved in assessing credit (as Zimmerman’s and some other deontic accounts would do) had better not rule Nathan’s action more morally praiseworthy than Zeb’s. To do so would be to disregard exactly what it is about things beyond our control that lead us to render them outside the realm of our responsibility.

Julia Driver's Egoistic Solution to the Virtue Conflation Problem

Julia Driver’s “The Conflation of Moral and Epistemic Virtues” attempts a solution to a task that, I think, begs for thorough treatment: that task is to develop a principled distinction between moral and intellectual virtues. Hume (and others) have thought that identifying demarcating features of moral and intellectual virtues—to the extent that a principled distinction could be developed—is of exaggerated importance. However, nowadays, the bulk of literature in both virtue ethics and virtue epistemology grants the supposition that there is some relevant distinction between these virtue types; thus, we are left in a position in which downplaying the task of discerning demarcating features between moral and epistemic virtues (or shirking the correlative task of determining if, in fact, a relevant distinction exists) would prevent us from coming to grips with the bulk of literature on virtue theory (ethical and epistemological) which employs these terminological distinctions as central to the arguments advanced.
Driver, in her essay, defends a view that I’ll summarize as follows:

(Initial Assumption: there does exist a distinction between moral and epistemic virtues, and a clarification of this distinction is worth pursuing.)

1. If an account that discerns moral virtues from intellectual virtues is to be adequate, then it must pick out what is distinctively valuable about the traits in question (i.e. it must pick out the respective value-conferring properties of these virtue types.)
2. “Characteristic Motivation” accounts cannot adequately meet the project in (1)
3. The only account available to successfully meet the project in (1) is an account that distinguishes moral from intellectual virtues on the basis of what goods are produced by the respective virtues.
4. Thus, the only available account is a consequentialist account.

In her defense of (2), Driver offers up, and subsequently criticizes, several characteristic-motivation accounts. There is quite a bit to be said about her treatment of these motivation accounts, however, I think what is more interesting is the way she sketches her own account in an attempt to defend (3). In doing so, Driver offers:

“The intuition I would like to explore is that intellectual virtues have—as their source of primary value—truth or, more weakly, justified belief for the person possessing the quality in question, and this is what ‘getting it right’ means for the intellectual virtues, whereas for the moral virtues the source of value is the benefit to others, the well-being of others, and for the moral virtues this is what ‘getting it right’ means. Further, no appeal to motive is needed to make the distinction at the level of value-conferring property. It is not the motive that makes the trait a given type of virtue.” (Brady and Pritchard 107).

Later in her paper, Driver clarifies her claim about the relationship between moral virtues and others, and intellectual virtues, and oneself.

“Moral virtues produce benefits to others—in particular, they promote the well-being of others—while the intellectual virtues produce epistemic good for the agent” (114).

Driver anticipates an obvious objection to this view, which is that moral virtues also produce benefits to oneself, and epistemic virtues can promote the well-being of others. Her response is that something can lead to valuable x, while its value-conferring property is nevertheless y (114). And so, it isn’t damaging to her position that some moral virtues benefit the agent, and some epistemic virtues benefit others. What would damage her view, though, is if some paragon moral virtue happens to not benefit others, or correlatively, if some paragon epistemic virtue failed to promote epistemic good for the agent. I think it’s safe to say that Driver is aware that such cases would be troublesome. This is evidenced in the final lines of her essay, when she writes:

“But note that on the account I offer, if it turns out that [for example, in the moral case] honesty does not have the good effects we think it has, then it may well be that it is not a moral virtue. This seems highly unlikely, but it is possible. Some may find this result problematic for a consequentialist account. However, it should be noted that this problem occurs for any account that weighs consequences at all” (116).

I think we should first take note that her last line here is mistaken. It doesn’t follow that the honesty-reductio that could in principle be problematic for her view would follow from an account that merely, as she puts it, “weighs consequences”; but rather, such a result would arise only if the consequences of virtues are weighed to the extent that traits are classified as virtues wholly by appeal to these weighed consequences. Not all consequence-weighting virtue accounts give such significance to consequences, and so not all consequence-weighing virtue accounts would be faced with the not-labeling-honesty-a-virtue bullet that her view could potentially swallow.

I wish to set this point aside, though, and address what I think might be most problematic about Driver’s proposal, and that is that it seems to embrace a sort of “epistemic egoism.” I say this because part of the distinction she offers between moral and epistemic virtues turns on the agency to whom the good produced by the virtue benefits. Recall that she claims:

“Moral virtues produce benefits to others—in particular, they promote the well-being of others—while the intellectual virtues produce epistemic good for the agent” (114).

The snake in the closet seems to be that her view commits her to accepting that a trait is not an intellectual virtue if it does not produce epistemic good for the agent. This is dangerously analogous to generic ethical egoism’s maxim: an act is not obligatory unless it produces good for the agent.

Rather than to criticize her view via appeal to models of criticism of egoism (i.e. by asking such questions as what is the relevant factual distinction between an agent and others that justifies a defense of the claim that an intellectual virtue’s value is agent-relative rather than relative to others outside one’s agency, etc.) I’ll try the old-fashioned style of counterexamples, which will take the form of presenting two archetypal epistemic virtues. The first, I’ll show, does not produce epistemic good for the agent that possesses it. The second intellectual virtue I’ll consider generates epistemic good for others, but not for the possessor. Either fits the model of a counterexample to her view (as both are instances of IVs that don’t produce epistemic good for the possessor of the virtue).

First case: Openmindedness: Heather Battaly on Montmarquet and the Intellectual Giants

I’m proposing openmindedness as an example of a paragon intellectual virtue that need not produce the epistemic good (which Driver offers as “truth”) for the agent that possesses it. Heather Battaly in her paper “Must the Intellectual Virtues Be Reliable” gives quite a bit of attention to openmindedness and concludes that it (perhaps among some others) requires no reliability condition. Specifically, it need not reliably lead to truth over falsehood. In her defense of this claim, she references James Montmarquet’s work in “Epistemic Virtue and Doxastic Responsibility” and highlights Montmarquet’s four most significant arguments against the claim that intellectual virtues require a reliability condition. Battaly thinks that Montmarquet has three arguments that fail and one that succeeds; she appeals to the latter as a basis for arguing that a reliability condition is unnecessary for an agent to possess the IV of openmindedness.

First, a clarifictory point: When presenting his arguments against the reliability of intellectual virtues, Montmarquet is (Battaly thinks) biting off more than he can chew in the sense that Montmarquet thinks that reliability as a necessary condition can be stripped from all intellectual virtues; it is, rather than reliability, a motivation condition that characterizes all intellectual virtues. Battally thinks this is flawed for several reasons, especially given that some IVs such as “the ability to recognize salient facts” require (obviously) some condition of reliability.

What Battaly wants to highlight, though, is that the “Intellectual Giant argument” that Montmarquet gives in his attempt to strip a reliability condition from all IVs would be successful in demonstrating a weaker but important claim, namely, that at least one intellectual virtue requires no reliability condition, and that is openmindedness.

Battaly’s amendment of Montmarquets intellectual giants argument could be stated something like this:

P1: Aristotle and Einstein both possessed the intellectual virtue of openmindedness in pursuing their respective projects in physics.
P2: Aristotle’s work turned out to be (generally) false, and Einstein’s work turned out to be (generally) true.
P3: If an agent’s possessing the intellectual virtue required that an agent be reliable in reaching the truth because of the virtue, then Aristotle would not possess the intellectual virtue of openmindedness.
∴ A reliability condition that the agent be reliable in reaching truth because of exhibiting openmindedness is not necessary for that individual to possess the intellectual virtue of openmindedness. (From 1, 2 and 3)

If this argument is sound, then Driver’s position is in trouble because we would have a case of an intellectual virtue that needn’t produce the epistemic good (which she labels as truth) for the agent. But surely we want to say that openmindedness is nonetheless an epistemic virtue.

Second case: Blabby and the virtue of “Epistemic Altruism”

I am not aware of much discussion of what I’m to propose as an epistemic virtue—the “virtue” of epistemic altruism, but I’ll try anyway to make a case for it, or at least, a case for the claim that Driver’s view would have to recognize it as an intellectual virtue (and, additionally, that it could in principle fail to produce epistemic good for the agent that possesses it.)

Suppose that there are two erudite scientists, Stingy and Blabby. Each has in his respective mental storehouse millions of useful facts about physics and chemistry. Stingy, who is arrogant and self-interestedly prudential, enjoys sponging into his memory any facts that he can gobble up, but is quite reluctant to share information with others. Stingy, although in possession of quite a few intellectual virtues which allow his own wealth of scientific knowledge to burgeon, has no desire to share his wealth of facts with others, even if such sharing would maximize net useful facts known in the world significantly. Blabby, on the other hand, is comparably as knowledgeable as Stingy, however, he is motivated to spread scientific truth as widely as possible; he is equally concerned with truths that would benefit himself (i.e. that he, and not someone else, be the first to solve some profitable equation) and truths that would benefit others. This said, he is aware about a certain fact: he happens to be in possession of more scientific facts than most others; in recognizing this, he finds that sharing information better satisfies his goal of overall truth maximizing than if he were to place value on maximizing truths that benefit himself over truths that would benefit others.

Shall we say that Blabby’s disposition to be generous with his excessive wealth of scientific facts is an intellectual virtue?

It was once written by Emerson that “the greatest homage we can pay to truth is to use it.” At the very least, it seems that the disposition to share truths it is a more praiseworthy disposition than Stingy’s sponging and misering of facts. Additionally, Blabby’s disposition contributes to what should be seen as an epistemic goal: maximizing truths in the world relevant to people’s lives, given that folks are more likely to better themselves with knowledge of the physical world than if bereft of them.

If we are prepared to grant that Blabby’s epistemic altruism is a virtue, then we have another problem for Driver’s position that a virtue is not an intellectual virtue if it does not produce the epistemic good for the agent. In Blabby’s case, it’s not clear at all that he procures more epistemic good (again, truths, on Driver’s acocunt) for himself by being epistemically altruistic; most likely, Blabby, because of his overall truth maximizing altruism, misses opportunities to silently sponge truths as Stingy does, and hence, maximize his own hoard of epistemic good.

If either of these examples correctly identifies an intellectual virtue that does not tend to produce the epistemic good for the possessor, then Driver’s project doesn’t work, and some other avenue of demarcation should be pursued.

Despite all this, I want to reiterate for the reasons I mentioned at the outset that Driver’s project is an important one. Ignoring this distinction or downplaying the importance of making it is costly to those who intend to understand contemporary arguments that cannot be adequately grasped outside the language of such distinctions. I’d like to end this post by pointing to what I think is a direction that is both potentially promising, and potentially dangerous to the prospects of VE. This proposal is one that Battaly offers in her own paper, and which has bittersweet consequences. She summarizes her conclusion in a provoking paragraph:

“Ultimately, I think that both Zagzebksi and Montmarquet have been too rigid in their analyses of intellectual virtue. Zagzebski has tried to make all of the traits that we intuitively classify as intellectual virtues fit a single mold. While, Montmarquet seems to have restricted his list of intellectual virtues to those that fit the mold he has chosen. In my view, virtues like open-mindedness require motivations for truth, but do not require reliability. In contrast, virtues like the disposition to recognize salient facts require reliability, but do not require motivations for truth. If I am correct, the virtues are a diverse lot. Consequently, there will be no single simple formula for defining knowledge or justification in terms of the virtues” (Must the Intellectual Virtues Be Reliable, 2004 INPC Session).

What is, at this point, provoking is whether accepting her “diversity thesis” (i.e. that some IV require reliability but not motivation, and others motivation but not reliability) does in fact generate the ominous prospects for VE that she thinks it does. Answering this, as I see it, would be a project of its own.

The Intellectual Virtue of Originality and the New Evil Demon Problem

I just recently read a post of Jon Kvanvig’s over at Certain Doubts which, I think, highlights a thus-far under-addressed aspect of VE, which is how to reconcile our intuitions that some virtues are proper intellectual virtues, whilst admitting that these virtues are not likely to lead possessors to a propensity of true beliefs over false ones. Kvanvig addresses a particular virtue of this sort (note: I admit I can’t think of many others), “originality,” and stipulates at the outset that it is an important intellectual virtue; however, he notes, it is not the case that a belief is more likely to be true if it has arisen out of originality than if it did not arise from originality. Kvanvig gives some suggestions for how we might be able to reconcile apparently conflicting intuitions that (1) the non-truth-conducive “virtue” of originality is an intellectual virtue, while (2) virtues are intellectual virtues only if a relationship holds between possessing the virtue and (at least to some extent) raising the probability that beliefs arising out of that virtue are more likely to be true than if they did not arise out of that virtue. Kvanvig concludes that there is a dearth of literature on (specifically) originality as an intellectual virtue, and I agree. I think a good place to begin such a project would be to try to isolate just what properties one has when one has the character trait of “originality” that leads us to want to say it falls within the domain of intellectual virtue, a domain we reserve for character traits that (unlike originality) do lead possessors to engender a higher propensity of true beliefs to false ones (than would be the case if the virtue weren’t possessed).
Rather than to attempt to address that (a question well worth addressing, I think) in this post, I want to offer some other thoughts I’ve been having about originality as an intellectual virtue—some thoughts related to aspects of originality that have nothing to do with whether the virtue is truth-conducive, but rather, whether beliefs arising out of that particular trait must, in fact, be original.
These thoughts arose while I was reading Sosa’s 1991 paper “Reliabilism and Intellectual Virtue” (reprinted in Axtell “Knowledge, Belief and Character” 2000). Sosa begins that paper with what he takes to be the three most significant problems faced by generic reliabilism, which he defines as: “S’s belief that p at t is justified if it is the outcome of a process of belief acquisition or retention that is reliable, or leads to a sufficiently high preponderance of true beliefs over false beliefs” (19).
The three problems he mentions are: (1) the generality problem, (2) the new evil-demon problem, (3) the meta-incoherence problem.
What I think has some interesting consequences for the issue of accepting originality as an intellectual virtue pertains specifically to what Sosa mentions about the New Evil Demon problem. He explicates the problem as follows:

“The evil-demon problem for reliabilism is not Descartes’ problem, of course, but it is a relative. What if twins of ours in another possible world were given mental lives just like ours down to the most minute detail of experience or thought, etc., though they were also totally in error about the nature of their surroundings, and their perceptual and inferential processes of belief acquisition accomplished very little except to sink them more and more deeply and systematically into error? Shall we say that we are justified in our beliefs and our twins are not? They are quite wrong in their beliefs, of course, but it seems somehow very implausible to suppose that they are unjustified” (20).
The thought here which drives Sosa’s (I think correct) assessment that it’s not the case that our twins aren’t justified even though they’re in the same mental state as we are whilst being justified, suggests that Sosa accepts something (I offer) like the following, on which the force of the New Evil Demon problem rests:

If A and B have lead identical mental lives, then it cannot be the case that at any time t, A is justified in a belief P at t whilst B (holding the same belief p) is unjustified at t.

The extent to which the above maxim is forceful is the extent to which it threatens reliabilism, which bestows justification on a belief by virtue of external conditions (i.e. is the belief in fact reliable) which depend on features of the world (i.e. does the world fit the belief) that can be outside the control of the epistemic agent. And in the New Evil Demon case, it is precisely because there are features outside of our twin’s control (i.e. they are, whilst mentally identical to us, being deceived while we aren’t) that lead some reliabilists to say that the twins’ beliefs aren’t justified. While there have been attempts to circumvent this problem—such as Goldman’s historical reliabilism--there are others, such as Fairweather, who find reliabilism abject for the reason that it has no good way of accounting for our intuitions that our twin is justified.

Rather than to address here different ways reliabilists have tried to answer the New Evil Demon Problem, I want to (finally!) draw what I think is a key analogy between New Evil Problem for reliabilism, and a parallel problem that crops up when we begin to consider “originality” as a candidate for intellectual virtue. The parallel is not perfectly isomorphic, but it illuminates an important feature of originality that might pose problems (not necessarily just for reliabilist minded VE theorists) for its candidacy as a proper intellectual virtue.

Consider, again, what it was that drives us to say, in the New Evil Demon case that the twins’ beliefs, though false, are nonetheless justified:

Maxim A: If A and B have lead identical mental lives, then it cannot be the case that at any time t, A is justified in a belief P at t whilst B (holding the same belief p) is unjustified at t.

Those who accept the New Evil Demon problem as problematic for reliaibilism, and do so because they accept something like Maxim A, will probably be receptive to a slightly different maxim, which has as its focus not justificatory conditions, per se, but conditions of being in a particular state of virtue:

Maxim B: If A and B have lead identical mental lives and identical dispositions of character, then it cannot be the case that at any time t, A possesses an intellectual virtue V at t whilst B (whilst possessing an identical mental life of a as well as identical dispositions of character as A possesses) does not possess intellectual virtue V at t.

A conundrum arises, though, for anyone who wants to accept Maxim B and grant that originality is an intellectual virtue. Consider the following example.
The Case of the Jailed Novelists

Jack and Phil are novelists, who are locked in separate jail cells. Each has a typewriter, a collection of the same books (for inspiration), and similarly motivated to write original novels, and each understands originality as world-relative. Jack lives on Earth, where there are only a couple million serious novelists. Phil, however, lives on planet Malthus, which hosts exactly 8 trillion google serious novelists, each which is, like Phil and Jack, motivated to write original novels. Jack knows of no other worlds but Earth, and Phil knows of no other worlds than Malthus. Additionally, the following counterfactual holds true for Jack and Phil: if Jack and Phil were writing on Earth (with the same motivation and opportunity to write original novels), each would write exactly 10 original novels.
As things unfold, Jack, on Earth, does in fact produce 10 original novels whilst in his jail cell. Phil, who writes the same amount as Jack (and exactly what he (Phil) would have written if his jail cell were on Earth), gets unlucky. The population of Malthus is so great that none of what Phil has written turns out to be original; for each book that Phil wrote, someone else on Malthus had already written a book similar to the extent relevant to preclude it from being original.
If we grant pace Aristotle (and Zagzebski) that virtues have a motivation and a reliable success condition, then we’ll be inclined to say that for both Jack and Phil, the motivation condition is met (each passionately wishes to write original novels), however Jack meets the success condition and Phil fails it. And so, the verdict would seem to be that Jack possesses the intellectual virtue of originality while Phil fails it.

This conclusion, though, is incompatible with Maxim B, which (recall) states:

Maxim B: If A and B have lead identical mental lives and identical dispositions of character, then it cannot be the case that at any time t, A possesses an intellectual virtue V at t whilst B (whilst possessing an identical mental life of a as well as identical dispositions of character as A possesses) does not possess intellectual virtue V at t.

And if we find Maxim B plausible (which, I think, we will if we find Maxim A plausible, as both are similarly motivated), we find ourselves in a stalemate of intuitions quite similar to the stalemate of intuitions that crops up when considering the dilemma of reconciling the unreliability of our twins (who we want to say are justified in their beliefs) in the demon world and Maxim A, which drives the New Evil Demon problem.

What can be learned from these stalemates? Perhaps, a close examination of reliabilist attempts to solve the New Evil Demon problem could shed some light on ways to reconcile Maxim B with the inclination to say that Phil is every bit as original as Jack. Another avenue I think might be profitable to pursue would be to note a certain property that is shared by both originality and reliability: each has an external success condition such that, for agents A and B who have identical mental and dispositional lives, A could meet the success condition (for either originality or reliability) and B could fail to do so.

Also, interestingly, the fact that the “B” analogues in our cases (i.e. our twins in the first case, and Phil in the second) fail the success condition of their respective property (reliability for the twins, originality for Phil) is out of the control of these agents. Put another way: our twins can’t help (or even know) that they are being deceived by an evil demon. And analogously, Phil can’t help (or even know) that the planet in which he desires to write original novels happens to have an astronomical number of aspirant novelists scribing away outside of his jail cell.

Now that we see the resemblance between what is going on in the New Evil Demon case and in the Jack-and-Phil originality case, we seem to be left with some lines in the sand which beg us to pick a side. On one side of the line might be those who are prepared to embrace something like the following:


1. Either both Jack and Phil have the intellectual virtue of originality, or originality isn’t an intellectual virtue.
2. Originality is an intellectual virtue.
3. Therefore, Jack and Phil both have the virtue of originality.
4. If Jack and Phil both have the virtue of originality, then whether one has the virtue of originality does not depend on externalist criteria that could hold outside the agent(s) control or awareness.
5. Therefore, whether one has the virtue of originality does not depend on externalist criteria that could hold outside the agent(s) control or awareness.

On the other side of the line, though, we might encounter the following antithetical argument:

1. If reliability and originality are intellectual virtues, then whether one possesses (either of) them will depend on whether one is, at least generally, reliable or original.
2. Whether one is reliable or original depends, at least in part, on features of the world that could be beyond the agent’s awareness of control.
3. Reliability and originality are intellectual virtues.
4. Therefore, some intellectual virtues (at least reliability and originality) are such that whether one possesses them depends in part on unknown/uncontrollable features of the world.
5. If whether one possesses a virtue depends to any extent on features of the world beyond one’s cognizance/control, then Maxim A and Maxim B are false.
6. Maxim A and Maxim B are false.
7. We have reason to doubt that (in the New Evil Demon case) we are justified and our twins aren’t and in the originality case, that Jack has the virtue of originality and Phil isn’t, only if Maxim A and B (respectively) are true.
8. Therefore, we have no reason to doubt (in the New Evil Demon case) that we are justified and our twins aren’t and in the originality case that Jack has the virtue of originality and Phil doesn’t.

Unfortunately, at this point, I have one leg on both sides of the line.

On McGinn's response to Hookway's "Affective States and Epistemic Immediacy"

I’d like to advance a certain take on the essay debate between Christopher Hookway and Marie McGinn (who commented on Hookway’s “Affective States and Epistemic Immediacy” printed in “Moral and Epistemic Virtues,” eds. Brady and Pritchard, 2002), which is as follows: McGinn’s charge that Hookway is begging the question against the skeptic in his attempted refutation of skepticism makes sense only if Hookway claimed to refute the skeptic, which he did not (I’ll attempt to show).

Hookway’s Argument (in “Affective States and Epistemic Immediacy”) proceeds something like this:

Initial presuppositions:

1. Ethics and epistemology share a common concern with evaluation, an enterprise with underlying issues whose similarity is such that an evaluative epistemic project could gain from attending to its evaluative ethical parallels.
2. Such gains (for evaluative projects in epistemology) include: (a) attaching (as most ethical theories do) importance to affective states (i.e. emotions); (b) Emphasizing the role of virtue in epistemology beyond the scope of self-proclaimed “virtue” theories.

Hookway’s Argument: (Which is an attempt to defend and clarify the second of these initial presuppositions) goes “loosely” like this:

P1. (Something like) Quine’s theory of induction is plausible/correct. This position characterizes the phenomenon of inference (which leads to our acceptance of propositions) as possessing a kind of “immediacy”; on this view, the evidential relations on which we depend aren’t available to reflection, but the “goodness” of the inferences (which lead us to accept propositions) are “felt”, and hence, “the vehicle in our confidence of inference is affective.”
P2. The best way to accommodate the “immediacy” of our evaluations is by appeal to character traits (i.e. virtues, such as “being observant”), which are simply stable patterns of identifying what is emotionally salient relative to the goal of inquiry (i.e. asking the right questions, etc.)
P3: Making such a move has the feature of paralleling the evaluative project of ethics, which pays (what Hookway appears to consider) appropriate consideration to affective states and virtues, attention lacking in contemporary epistemology’s evaluative endeavor.

It is not my understanding that a central endeavor of Hookway’s is to refute skepticism, even though Hookway does mention some implications the acceptance of his premises would have toward placing his project distinct from anti-skeptical endeavors that are (as Hookway thinks) “over-intellectualized”—by their feature of requiring reflective justification ad infinitum. Hookway’s Quinean appeal distances himself from the over-intellectualized justificatory positions, and he makes this clear on pp. 84 and 85 when pointing out how his position avoids two “troublesome regresses” (the “regress of reasons” and the “regress of justification”) that have led other positions to be hard-pressed to refute the skeptic. That Hookway’s view has this virtue, though, is better understood as a peripheral (and positive, if successful) upshot of adopting his premises, and not at all integral to the conclusion for which I take him to be arguing.
It is for this reason that I am surprised that Marie McGinn’s entire rebuttal to Hookway focused (and self-proclaimedly focused) on that one particular facet of Hookway’s position. No doubt, McGinn’s position is, in a nutshell, the claim that Hookway tries to rebut the skeptic and fails. Not only is rebutting the skeptic not a central focus of Hookway’s argument, but I think that, even the extent to which skepticism is addressed peripherally (on pp. 84 and 85) we find no solid affirmation that Hookway thinks that his position literally “defuses” the skeptic, as McGinn claims him to be positing.

My claim here needs some defense, and so, I will site verbatim the three times that Hookway explicitly mentions skepticism in his article:

(1) “And even fewer would insist that taking the role of emotions, or affective states, seriously is necessary if we are to deal in a satisfactory way with what we can think of as the central problems of epistemology—for example, the defusing of skepticism or the study of how internalist and externalist demands in the theory of justification can be integrated.” (pp. 75-76).

(2) “…it is important that education and training equip us to avoid questions that should not be addressed. In other cases (“might I be a brain in a vat?”) the explanation of why it is good not to address such questions may be different again.” (p. 84)

(3) “To answer skepticism, it seems, we need to combine these perspectives: we must be properly confident (subjective) that we are reliable (objective). The worry we now face is that we can only be properly confident that our emotional evaluations are conditionally reliable.” (86)

Fourthly, Hookway (p. 84 and 85) mentions the two “troublesome regresses” that his view can circumvent, and while these regresses are known to lead some to skepticism, Hookway never offers that his circumvention of these regresses is sufficient for defusing skepticism (rather, the much weaker claim, that he has simply dodged these particular regresses).

A careful reading of these sections will not find Hookway proposing that his position is sufficient for defusing skepticism. (1) could be read weakly as mentioning the defusing of skepticism as among the central problems of epistemology. A stronger reading of (1) would be that, few would think that an analysis of emotion would help to resolve central problems in epistemology; any substantive claim that Hookway’s position is supposed to do that job is not made explicit here. (2) merely presents the skeptical question as a “type” of question that perhaps ought not be addressed. He brings up the skeptical question (and mentions subsequently that it might be only epistemically relevant to goals of inquiry within the philosophy class) as a demonstrative analogue to another sort of question (on which he focuses) which should not be asked, namely, questions which are irrelevant to inquiry (i.e. how many grains of sand were on the beach in 1952) and which an intellectually virtuous person would not find emotionally salient. No where in (2) is any substantive claim that Hookway thinks he has refuted the skeptic; all we could make of this is, perhaps, the skeptical question ought not be asked (but not even necessarily for the same reason that the non-salient questions Hookway is discussing ought not be asked!)

(3) would be the best candidate McGinn could site in her attempt to claim that Hookway thinks he has rebutted skepticism. However, even (3) proposes only this: X is a necessary condition for rebutting skepticism. We shall see how my view meets X”. And that’s just what Hookway does; he tries to show how his position meets what he thinks are those necessary conditions; this is not a case of Hookway claiming to have met sufficient conditions.

And, as I mentioned, his discussion of the “worrisome regresses” on pp. 84-85 could lead us no further than to accept that Hookway thinks that his views have circumvented two particular regresses that tend to skepticism—a different claim than the claim that skepticism simpliciter is rebutted by his view.

Here is a bit of what McGinn has made of Hookway:

“The question I want to focus on here is whether the understanding of the nature of our epistemic practices that Hookway develops in the light of ideas that he draws from ethics provides an effective means of “defusing…skepticism” (75)

“…there is something prima facie perverse in trying to answer the philosophical sceptic by appeal to the essential role of emotional responses in our epistemic evaluations, insofar as we do not normally regard the emotions as having any special or privileged connection with veridicality.” (96).

“Moreover, given that Hookway’s account of our ordinary practice acknowledges that the relation between the emotional evaluations on which it rests and objective truth is contingent, it is hard to see how the work of resisting skepticism in a philosophical context is to be achieved.” (99).

“But [a consequence of Hookway’s view] means that I can take my current emotional evaluations as a proper ground for rejecting the sceptic’s questions only by assuming the very thing that the skeptical voice in me doubts—that is, by arguing in a circle.” (100).

There are quite a few other comments on this line because, alas, the paper McGinn’s written has skepticism as its focus.

I don’t intend to be a rabblerouser here and suggest that McGinn’s reply is in any way scandalous or malevolent. Quite the contrary! McGinn has raised a philosophically interesting question, which is what the implications of Hookway’s position would be if the mission were to rebut skepticism. Is Hookway’s position capable of rebutting skepticism? This is entirely a project worth pursuing. My intention to go to some length here was only to indicate what I think was a misattribution; rebutting skepticism was not Hookway’s goal, and so, the question of whether Hookway’s position is capable of providing a response to skepticism is a question different than whether Hookway was successful in his attempt to do so. No such attempt fell within his project, and McGinn’s response didn’t appear to recognize this.

The "Conrad the Conformist" and "Robotic Robert" cases in Fairweather's "Epistemic Motivation"

In “Epistemic Motivation,” (taken from ‘Virtue Epistemology’ eds. Zagzebski and Fairweather, 2001), Abrol Fairweather advances an argument that goes (something) like this:

1. To be able to adequately characterize what an intellectual virtue is, we must first have some understanding of what virtue (in general) is.
2. To meet the project in (1), there are three main options: Aristotle’s account of virtue as an excellence of character, Plato (Gorgias and Republic)’s account of virtue as a skill (techne); thirdly, there is Aquinas’ teleological account of virtue as the “power to bring about a certain end” (65)
3. If epistemic motivation is a necessary condition for knowledge, then the only account of virtue capable of subsuming epistemic motivation within the domain of intellectual virtues is the Aristotelian account.
4. Epistemic motivation is a necessary condition for knowledge.
5. Therefore, the only appropriate account of virtue to which VE theorist ought to give recourse in understanding the nature of intellectual virtue is the Aristotelian account.

Fairweather’s defense for P4:

1. There are (basically) three types of epistemically motivated beliefs: (a) belief with improper motivation; (b) non-motivated belief; (c) belief with good epistemic motivation.
2. If an agent is epistemically deficient in (a) and (b), then (c) is a necessary condition for knowledge (given that knowledge is incompatible with epistemic deficience on any VE account).
3. An agent is epistemically deficient in (a) and (b).
4. Therefore, epistemic motivation is necessary for knowledge.

It is now (3) in Fairweather’s defense for P4 that requires a defense. Fairweather outlines his defense just as one would expect: he argues via reductio in cases (a) and (b) to emphasize the epistemic deficiency associated with being motivated to some non-alethic (i.e. improper) end, or to have no motivation; in each instance, Fairweather hints that what is needed for the beliefs (associated with these two types of deficient motivations) to be candidates for knowledge is that the proper, alethic-ended motivation should be present. His cases are both quite clever, and do well to illustrate his point. I do want to point out a few spots in these examples, though, which appeared problematic to me, and which I think invite a bit of revision.

First off, consider Fairweather’s “Conrad, the Doxastic Conformist” case, a case intended to delineate an individual with non-alethic epistemic motivation (i.e. improper motivation), and allow the reader to see the ‘absurdity’ of deeming Conrad’s beliefs as knowledge in light of the epistemic motivation that engendered them. In short: Conrad wants to believe whatever Mr. Cool believes; his desire is significant to the extent that Conrad desires to bring his “whole belief system in to conformity of Mr. Cool’s”, and is, by virtue of this obsession, not concerned with the truth or falsity of his beliefs, per se, but simply whether they are isomorphic with Mr. Cool's. Given this, suppose that Mr. Cool believes that X will win the election in November; automatically, Conrad wishes to believe this, and would believe it independent of whether Cool offers evidence for it. Now, suppose that Cool does, in fact, articulate his evidence for believing X will win the election (i.e that the opponent Y is lagging in the poles and is involved in scandal, etc.). Conrad, in this revised situation, holds a belief we would think to be justifying evidence (i.e. that Y is lagging in the polls, etc.), but Fairweather’s contention is that Conrad is not justified; this is because the grounds Cool gave him for believing that X would win “play no role as evidence in exploaining Conrad’s acquisition of the belief…. Being supported by good evidence is a purely accidental feature of his belief since Conrad would be just as inclined to believe that X will win without possessing any evidence at all, so long as “X will win in November” is believed by Mr. Cool” (74). And so, Fairweather wishes to show, possession of good evidence for a belief is insufficient for achieving justification; the missing element is epistemic motivation. Fairweather defines the sort of epistemic motivation, necessary for justification, and lacking in the Conrad case, as follows:

EM*: A person has an epistemic motivation if and only if they have a desire (or kindred emotive state) for truth or for states whose value is derived from truth, and this desire effectively directs and controls the person’s belief formation and revision.

One point at which I take issue is Fairweather’s reasoning for why it is that Conrad lacks justification in the revised case in which Conrad is given by Mr. Cool Mr. Cool’s grounds for believing that X will win. Even in light of Conrad’s improper motivation, and even in light of the fact that he believes the grounds for the belief (i.e. that Y is lagging in the polls) decisively because Cool believes it, it seems reasonable to conclude that Conrad (insofar as he is rational, which Fairweather has given us no reason to doubt) would (importantly) “recognize” that “X will win” follows inferentially from “Y is lagging in the polls” (combined with Conrad’s assumed background knowledge that X is competing against Y). It seems to me, at least intuitively, that Conrad’s recognition of the evidential relation between the grounds of the belief (that x will win), and the belief that x will win would be sufficient (combined with Conrad’s believing the grounds of the belief, and the belief) for Conrad’s being justified in his belief that X will win in November. Now, if Fairweather had explicitly said that Conrad, due to some strange irrationality (perhaps he is inductively challenged) believed both the grounds for X will win in November, and the proposition that X will win in November, but nonetheless did not recognize that the grounds serve as evidence for the belief, then I’d agree with Fairweather that Conrad is not justified. But until Fairweather makes explicit that the recognizition of this relation I mentioned is not present in the Conrad case, then even given Conrad’s improper epistemic motivation, I’d be inclined to think he is justified (at least insofar as the evidential recognition condition I mention holds.)

Fairweather’s next step in the argument is to offer the reductio against (b) [in P3 of his defense of p4 in the initial argument]. Simply: Fairweather must show that an individual with no epistemic motivation is epistemically deficieint to the extent that beliefs arising out of this lack of epistemic motivation are not justified. Enter Robotic Robert, Fairweather’s protagonist to this end, who holds and abides by the following rule “Adopt an attitude of acceptance toward a proposition P iff I have evidence that strongly indivates that P is true.” Fairweather thinks here that Robotic Robert, who he stipulates is unable to explain why he forms beliefs according to that particular rule, rather than some other rule, is unjustified in his beliefs. Robotic Robert, Fairweather thinks, by virtue of his inability to provide a reason for adhering to his rule, has no sense that truth is the appropriate aim of belief, and hence, should not be said to be justified.
Here is my problem with this case: Fairweather indicates (p. 76) that, if Robert could have given a reason for why he formed beliefs according to that rule, then his beliefs would be epistemically motivated to the extent that they would (probably) satisfy EM* and be justified. But this is a dubious step to make. Consider this: suppose we grant Fairweather’s wish and program into Robotic Robert a response to the question Fairweather didn’t have him answer… suppose his response to the question of why he forms beliefs according to the rules to which he appeals is this: “Because I am motivated to believe what is true.” This robotic response would seem, based on what Fairweather says, to satisfy him. But what if we were to then ask, “Why do you want to believe what is true?” Robotic Robert would have no answer. And, to add here a concern at a practice al level, many epistemic agents would be left scratching their heads when asked to explain, decisively, why they wish to believe what is true.
One response Fairweather could give to my criticism here is that the chain of necessary explanation ends at the question of “why” one forms beliefs according to rules. Giving this response would help Fairweather to avoid the pragmatic problem of committing to the position that all those epistemic agents who can’t explain why they are motivated to truth (i.e. many agents we want to say are justified in a majority of their beliefs) aren’t justified; however, this response saves the practical problem at the expense of allowing Robotic Robert to be properly epistemically motivated merely because he is programmed to give an explanation for why he forms beliefs according to particular rules—a counterintuitive suggestion.
What I’ve mentioned are the two major concerns I had with Fairweather’s Conrad and Robert examples (and I’d be interested to get any opinion on these). However, I do wish to point out one more area of difficulty in the paper. In Section V, Fairweather concedes that the bold proposal of holding epistemic motivation as a necessary condition for knowledge requires an investigation into the question of: “Does a motivational requirement apply to all types of knowledge?” I’m inclined to think certainly not. (i.e. Suppose I want to believe everything Mr. Cool believes… Mr. Cool tells me that bachelors are unmarried women, however my recognition of the truth of bacherlor’s being unmarried men as analytic leads me to believe it is true, even though my epistemic motivation is merely to match my belief set with Mr. Cool’s. Shouldn’t we say I’m nonetheless justified in my belief?” And so, for a starting point, we’d probably want to say that analytic, or a priori truths, are probably exempt from any motivational requirement. All for now…

Andrew McGonigal on Eflin's "Epistemic Presuppositions and their Consequences": the question of heirarchy

I am glad I’m not paid to keep score in the essay-debate between Juli Eflin and Andrew McGonigal found in Moral and Epistemic Virtues (2003). I would give up somewhere around the point where Andrew McGonigal (in his criticism of Eflin’s essay “Epistemic Presuppositions and Their Consequences”) offers the following brief and biting criticism of Eflin’s project:

“…given that Eflin seems to hold that a combination of rules and procedures and virtue theory can provide the relevant explanations, it is difficult to see why her own theory is not both hierarchian and complete, contrary to advertisement—it looks to me as if she just takes a broader range of conceptually basic elements as necessary to provide complete explanations of problems within the epistemological domain.” (73)

Some background, after which McGonigal’s claim will become more clear…

Eflin’s position, begins with a criticism of “traditional epistemology” (her characterization of which is doubted by McGonigal) along several lines, which include that TE is “heirarchial” and “complete.” What do these terms mean? Eflin defines them as follows:

“For a theory to be heirarchial, some set of notions is taken as basic and other dlements in the theory are derived from or reduced to these basic elements” (p. 50). Also, Eflin claims that, if a theory posits self-justifying elements which explain other elements, then the theory is heirarchial. (50).

Completeness: “For a theory to be complete, everything in a theory’s domain is accounted for in terms of the basic concepts.”

Some examples of theories that are “heirarchial and complete”, according to Eflin, include Coherentism (i.e. BonJourian), as well as virtue theories such as Zagzebski’s, the latter of which, Eflin claims, posits the virtues as basic concepts from which other concepts are derivable (hence “heirarchial”) and committed to virtue as a “success term” (hence, Eflin thinks, complete).

An aside: I am skeptical of her characterization of Zagzebski’s position as heirarchial and complete. It is more clear that foundationalism and Coherentism would be candidates, though.

Anyhow—Eflin, after demarcating “traditional epistemology” as branded by being heirarchial and complete (among other criticisms, such as requiring the supposition of ideal knowers, etc.), she presents her own account—which has the virtues as well as rules as primary (central, ,but not such that other concepts can be derivable from them), but not basic—and which she thinks is not heirarchial and complete, and which has added benefits of not being “context-stripping” or stipulating ideal knowers. She describes her account as a pluralistic virtue-centered epistemology. She uses the term “virtue centered” rather than “virtue theory” because, on her view, the latter description would imply that virtue is taken as basic within the theory. Eflin’s account avoids taking virtue as basic by using what she calls a teleological position, in which understanding is identified as the end of human life (or, an end). However, says Eflin:

“We can say ‘understanding’ in the abstract, and we can discuss the various virtues on which an inquirer needs to draw to produce it, but ‘understanding’ simpliciter cannot be given an account, nor can an inquirer have it. I can only achieve particular understanding from where I am, that is, from my context. It has to be an understanding needed by an individual, achievable from her present understanding and through her epistemic virtues, or from ones she can develop.”

The problem that I saw, and which I think McGonigal (in the quote I presented initially) finds problematic here is that Eflin seems to be trading one basic concept for another. Rather than to have a basic concept of virtue from which all other concepts are derivable, she has understanding as such a concept. It seems to me that, although she doesn’t say this clearly, she thinks that because understanding simpliciter remains vacuous on her account (so that she can avoid “context-stripping”), that therefore, understanding is disqualified from being a basic concept from which other concepts are being derived, and a fortiori, she has not committed the sin of hierarchy. This seems to me a dubious inference. As McGonigal suspects of her, “it looks to me as if she just takes a broader range of conceptually basic elements as necessary to provide complete explanations of problems within the epistemological domain.”

Is that what she’s doing?

There are at least a couple questions which at this point need to be answered if we are to attempt to “deobfuscate” the issue of whether her theory avoids the charge of being heirarchial (which, to reiterate, she insists it is not).

(1) From what premises does Eflin get to the conclusion that her theory is not heirarchial, given the definition she presents for a theory’s being heirarchial.
(2) How are we to go about determining whether “understanding” fits the bill of a basic concept in her view. Additionally, what is the relationship between a concept’s being un-analyzable simpliciter and that concept’s qualifying as basic within a theory?

Without answers to these (and perhaps some more) questions, I’m inclined to think that McGonigal’s criticism is on target. Her pluralistic virtue-centered epistemology is simply invoking a broader range of concepts, and using a concept (understanding, relativized to a particular knower, in her own context) as basic.

Some clairvoyance and (perhaps?) regress problems with Goldman's "Epistemic Folkways"

In his paper “Epistemic Folkways and Scientific Epistemology” (reprinted in Axtell’s Knowledge, Belief and Character and originally published in Goldman’s own Liasons: Philosophy Meets the Cognitive and Social Sciences), Goldman undertakes the project of delineating and defending both a descriptive and a normative form of what he calls “scientific epistemology.” His term “scientific epistemology” is warranted, Goldman thinks, due to what he takes to be crucial dependences on the cognitive sciences in meeting the “two missions of epistemology” (to describe our folk concepts, and to evaluate and improve upon them).
What I found particularly interesting occurs in the first section of Goldman’s paper, in which he attempts to provide a Sosa-amended reliabilist account of justification, which he then defends against a few classic counterexamples which have plagued reliabilist positions in the past.
I take his position to claim something of the following:

S’s belief that p is justified iff S’s belief is a result of processes all of which are on the epistemic evaluator’s list of intellectual virtues. If S’s belief is a result of processes, any of which are on the epistemic evaluator’s list of intellectual vices, the belief is unjustified. If the processes which engender S’s belief are on neither list, the belief is non-justified.

How does a virtue or vice get on the epistemic evaluator’s list? At this point, (obviously) there is an appeal to reliability—processes that are deemed to produce a sufficiently high ratio of true beliefs will make the evaluator’s list of intellectual virtues.
Importantly for Goldman, because we are giving at this point a “descriptive” account of the epistemic concept of justification, we should not (he claims) stray from what the folk would agree upon. And hence, a good test question for determining whether a candidate virtue makes the evaluator’s list is to suppose we ask the folk, “How does S know that p?” Answers, Goldman thinks, would include such processes as: “Because he saw it, because he heard it, because he remembered it,” etc. These are the sort of processes Goldman has in mind.

A problem I think his account doesn’t resolve, though, is why we should not include clairvoyance on the list of intellectual virtues given his criteria for what should arrive on the evaluator’s list. I don’t say this because I think clairvoyance is, in fact, a virtue, per se, but because there seems to me no clear reason why Goldman should, given his reliability criterion, be in a position to weed it from his evaluator’s list. A good clairvoyance case, which he mentions, is the case of “Norman” (borrowed from BonJour), who has the reliable faculty of clairvoyance, but who doesn’t have any evidence for or against having this faculty. Goldman thinks that the folk intuition here is that Norman’s clairvoyance-based beliefs are not justified, and so, because Goldman seems to allow folk-approved reliable processes a place on the evaluator’s list of virtues, he has some explaining to do if he is to preserve the folk intuition that such beliefs aren’t justified.
His explanation, I think, is unsatisfying. He makes an appeal to what he takes to be the folk trait of “categorical conservatism”: the folk, he think, display a preference for “entrenched” categories; “they do not lightly supplement or revise their categorical schemes.” And so, Goldman thinks, clairvoyance wouldn’t make the list because the folk are hesitant to add it, even though the ones they do add are added for the same reason that clairvoyance would be added—namely—because it is a sufficiently reliable process.
One question we should ask here is: is Goldman right that the folk would, in fact, be inclined to take clairvoyance off the list of virtues (in light of the fact that other virtues are on their for having the process of reliability that clairvoyance possesses). I’m worried that, if he is right about this, then prospects appear dim for Goldman’s appeal to reliability as a basic concept within his theory, which is how he seems intent to present it. If justification is explained in terms of intellectual virtues, and they are explained in terms of reliability (as Goldman seems to want), then reliability must be basic. But no longer is it basic if it is not capable of explaining the justificational status of beliefs arising out of a process of clairvoyance. Such beliefs should be able to be identified by recourse to reliability (for his theory to preserve the heirarchial structure he appears to want). And so, dropping clairvoyance from the list is at the expense of the clean heirarchial structure he wants in his theory.
Another problem I’d like to explore with Goldman’s position in this paper is that it plays, I think, the dangerous game of falling into regress. My concern here might be quite misguided, to warn. It is as follows: If a criterion Goldman uses to determine whether a belief is justified is whether the folk would deem it reliable (i.e. the criterion he uses to deem clairvoyance non-evaluator-list-worthy), then it seems as though he is positing this criterion as at least a necessary condition for a candidate virtue’s making the list. I wonder, though, how broadly this criterion can be applied. If Goldman weeds virtues from the evaluator’s list by virtue of the fact that the folk’s “categorical conservativism” would not allow it, then why should we not think that folk acceptance should also be applied as a criterion at the level of “theory” acceptance, as well as process acceptance. He doesn’t give a reason to think that there is a relevant factual difference between processes and theories that would justify folk acceptance as a criterion for one but not the other. But as soon as we apply folk acceptance to the level of theory acceptance, then it seems dubious to suspect that the “folk” are going to be inclined to understand justification in terms of such technical machinery as “evaluator’s lists” and “truth ratios” and (in cases like clairvoyance) even reflection on their own categorical conservatism as needed for reference within a theory of justification. And hence, the regress is: if folk intuitions can be stipulated as criteria for ruling out processes that would make evaluator’s lists, they would also most likely rule Goldman’s own theory of justification which makes references to enough machinery to be non-folk-approved at the theory level as clairvoyance would be non-folk-approved at the process level. Of course, Goldman has a way out if he can demonstrate the relevant factual difference between processes and theories that justify folk approval as a criterion for one but not the other; however, I’m not sure what such a factual difference could be.