Author Information: Rik Peels, Vrije Universiteit Amsterdam, firstname.lastname@example.org
Peels, Rik. “Let’s Bite the Bullet on Deontological Epistemic Justification: A Response to Robert Lockie.”  Social Epistemology Review and Reply Collective 4, no. 12 (2015): 42-50.
Please refer to:
- Lockie, Robert. “Perspectivism, Deontologism and Epistemic Poverty.” Social Epistemology (2015): 1-17. doi: 10.1080/02691728.2014.990281.
- Elqayam, Shira. “Instrumental Bounded and Grounded Rationality: Comments on Lockie.” Social Epistemology Review and Reply Collective 4, no. 11 (2015): 47-51.
Image credit: Anita Ritenour, via flickr
In his paper, Robert Lockie points out that adherents of the deontological conception of epistemic justification, in meeting the epistemic poverty objection, have argued that there is also such a thing as subjective epistemic justification. Lockie argues that subjective accounts of epistemic justification must take cultural limits seriously and that the relevant subjectivism is, therefore, at least in part cultural. I fully agree with Lockie on this point, but argue that his arguments in favor of his view are not convincing. Moreover, I argue that adherents of the deontological conception might make a bolder move. Rather than claiming that the deontological conception is a conception of a more subjective notion of epistemic justification, one could make the strong assertion that the deontological conception of epistemic justification is the right conception of epistemic justification, because epistemic justification is a subjective notion. One has resources to argue that those who level the epistemic poverty objection against the deontological conception are working with an idealized concept of rationality that has little to do with epistemic justification.
On the deontological conception of epistemic justification—the word derives from the Greek deomai, ‘ought to’—one’s being epistemically justified in holding a particular belief should be understood in terms of obligations, responsibility, praise, blame, and other similar notions. One of the most important objections that has been leveled against the deontological conception of epistemic justification is the so-called epistemic poverty objection.
The main point of it is that the following seems possible. A cognitive subject is in an epistemically bad position and her cognition, at least regarding some specific proposition p that she believes, is, therefore, not truth-conducive. However, she has not violated any obligations in believing that p and is, therefore, blameless and, thus, deontologically justified in believing that p. Hence, one can be epistemically justified while believing in a way that is not truth-conducive. But if one’s belief is not formed in a truth-conducive way, then it cannot be epistemically justified. Hence, the deontological conception of epistemic justification is mistaken. Adherents of deontologism have, as a response to this, embraced bounded (perspectival) notions of epistemic justification, on which epistemic justification is significantly more subjective than often thought.
Robert Lockie, in his illuminating essay, argues that such subjective accounts of epistemic justification must take cultural limits seriously and that the relevant perspectivism is, therefore, at least in part cultural perspectivism. I fully agree with Lockie on this point. However, I think that his arguments in favor of his view are not entirely convincing. After an exposition of the epistemic poverty objection (§2), I explain why I think they aren’t (§3). Moreover, I argue that the adherent of the deontological conception of epistemic justification might actually make a bolder move than Lockie makes. Rather than claiming that the deontological conception is a conception of a more subjective notion of epistemic justification, one could make the strong assertion that the deontological conception of epistemic justification is the right conception of epistemic justification, because epistemic justification is a subjective notion. One has resources to argue that those who level the epistemic poverty objection against the deontological conception are working with an idealized concept of rationality that has little to do with epistemic justification (§4). I conclude that we should bite the bullet and explore whether the deontological conception of epistemic justification is the right conception of epistemic justification.
The Epistemic Poverty Objection
One of the main advocates of the epistemic poverty objection is William Alston. As Alston rightly points out, it is possible that one has met all one’s epistemic obligations—that one has done everything that could rightly be expected of one and that one is, therefore, blameless—and, yet, one believes that p on a ground that is not truth-conducive. It is not truth-conducive because, due to no fault of one’s own, one is in an epistemically bad situation or an epistemically bad condition. In such a case, the fact that one’s belief is based on that particular ground does not render it likely that the belief is true. One well-known example that Alston uses to illustrate this runs as follows:
Tribesman case. S has lived all his life in an isolated primitive community where everyone unhesitatingly accepts the traditions of the tribe as authoritative. These have to do with alleged events distant in time and space, about which S and his fellows have no chance to gather independent evidence. S has never encountered anyone who questions the traditions, and these traditions play a key role in the communal life of the tribe. Under these conditions it seems clear to me that S is in no way to blame for forming beliefs on the basis of the traditions. He has not failed to do anything he could reasonably be expected to do. His beliefs about, for example, the origins of the tribe, stem from what, so far as he can see, are the best grounds one could have for such beliefs. And yet, let us suppose, the traditions have not been formed in such a way as to be a reliable indication of their own truth. S is deontologically justified, but he is not believing in a truth-conducive way.
Laurence BonJour was the first one to call situations like these cases of epistemic poverty: one is epistemically justified in holding one’s belief on the deontological conception of epistemic justification because one has not violated any epistemic duties. Still, one holds a belief on a radically inadequate ground. The point is: even if one does everything that can reasonably expected of one from an epistemic point of view, one may still be in a position that is not at all likely to lead to true belief. Since one is epistemically justified in believing a proposition only if it was formed in a truth-conducive way—the fact that it was based on the ground that it was based on renders it likely that the belief is true—the deontological conception of epistemic justification is untenable.
Grounded and Bounded Epistemic Rationality
A common response to the epistemic poverty objection has been that in cases of epistemic poverty, one is subjectively epistemically justified even though there may also be a more objective sense of epistemic justification on which one is not epistemically justified. This is also Lockie’s response in his paper. In giving this reply, he uses the expressions ‘epistemic justification’ and ‘epistemic rationality’ synonymously. In what follows, I focus on epistemic rationality. In the next section, I return to the distinction and argue that it might be more accurate to distinguish between rationality and justification.
As Lockie rightly points out, the idea that epistemic rationality can plausibly be understood subjectively squares well with lines of thoughts in other fields: there was also a development in the psychological and economical literature on rationality that resembles this in a way: rationality may be understood subjectively in the sense that the rational standards we should hold people to need attenuation in the light of people’s constraints in specific circumstances (so-called grounded rationality) and human limitations generally (so-called bounded rationality).
This raises the question of how grounded rationality is. According to Lockie, “[p]hilosophers commonly underestimate how culturally situated, how educationally laden, and how restrictive is their conception of the rationally acceptable variance here.” As far as I can see, Lockie gives three arguments in support of this claim. In this section, I critically discuss each of them.
First, Lockie appeals to fascinating empirical research by Aleksander Luria. Luria presented illiterate rural Uzbek peasants with the following two propositions:
(1) In the far north, where it snows, the bears are white.
(2) Nova Zembla is in the far north, and it is always snowy there.
After that, he asked them the following question:
What color are the bears in Nova Zembla?
Most peasants replied that they didn’t know, since they had never been there, or that they hadn’t travelled that far so one shouldn’t ask them about it, or that there are different kinds of bears and that if a bear was born red, it would stay that way, or that they didn’t know what kinds of bears there are, since they could only judge what they had seen. Such responses are, of course, remarkable, for the answer that the bears in Nova Zembla are white obviously deductively follows from propositions (1) and (2). Even leading questions, such as the question what one’s (Luria’s) words suggested, drew little response. Most peasants simply tried to avoid solving the task.
According to Lockie, it is implausible that all these peasants were simply epistemically irrational. According to him, what follows, then, is that epistemic rationality is highly perspectival and that cultural limitations are among the core subjective features that determine whether an individual is epistemically rational.
There is a fairly simple question that Lockie fails to address, though: exactly with regard to what were these peasants rational? After all, the issue under discussion in Lockie’s paper and in my reply is rationality in believing certain propositions. So, which propositional beliefs these Uzbek peasant held does Lockie take to be rational? Is it the belief that they shouldn’t be asked about bears in Nova Zembla because they haven’t travelled that far? Or their belief that they don’t know the answer to the question of what color the bears in Nova Zembla have? Or is rather not their belief, but rather their suspension of judgment on the proposition that the bears in Nova Zembla are white that Lockie takes to be rational? It isn’t easy to assess Lockie’s claim as long as he doesn’t specify what the relevant proposition is. Nobody would deny, for instance, that the peasants’ belief that they didn’t know the answer was rational, given that, apparently, they didn’t know the answer.
But even apart from this worry, it isn’t entirely clear that the evidence from this empirical study suggests that these peasants were rational. After all, some peasants did know the right answer to the question. More precisely, out of the 15 test subjects, 9 were able to solve syllogisms associated with experience, 2 were able to solve syllogisms not associated with experience, and 4 after were able to solve syllogisms not associated with experience on conditional assumptions like ‘from your words I can gather…’. If they knew it, then why couldn’t we say that many of those who didn’t know were irrational, given that they had the same information as those who did know the answer? After all, they were all in the same cultural-historical circumstances. This is especially important because Lockie appeals several times to the ought-implies-can principle and suggests that the Uzbek peasants didn’t fall short of a rationality norm because, apparently, they couldn’t meet it (p. 10). The fact that at least some of them made a valid deduction shows that they could meet the norm.
Second, Lockie points out that we shouldn’t forget that we, in Western democratic societies, have been trained immensely to decontextualize and that people may be entirely rational in failing to decontextualize if they haven’t been taught how to do so:
Because we have been so exhaustively inculcated, literally from birth, into such decontextualized contexts, we fail to appreciate how extraordinarily immersive and intensive our acquisition of this context (of decontext) has been (…)—or how recent its arrival in human cultural and intellectual history (7).
Lockie is right that we have, of course, been trained much more thoroughly in logical reasoning and drawing conclusions about decontextualized situations than these peasants. If it is (largely) due to their lack of training that they didn’t come to the relevant conclusions, then, given that it wasn’t up to them that they didn’t receive more education, they are clearly blameless for not coming to the relevant conclusions. However, why should we think that it follows from that, that they were rational? Why not think that they were blamelessly irrational or that their attitude was blamelessly non-rational (beyond the domain of rationality and irrationality)?
Third, according to Lockie, “a ‘reified’ rational agent, idealized beyond all bounds, is no rational agent at all” (p. 8). We, therefore, have to adopt a more subjective notion of rationality. However, I find it hard to see why someone who is perfectly rational in all regards wouldn’t count as an agent. (In fact, if that was clear, we would have an easy argument against the existence of God, who is, on most definitions, perfectly rational.) In the same way as nobody is morally perfect, but someone who is morally perfect would still be an agent, no human being is epistemically perfect, but if there were somebody who is epistemically perfect, that person would still be an agent.
I conclude that Lockie’s three arguments are unconvincing. This is not to deny that there may well be a subjective or perspectival notion of rationality. In fact, I think most philosophers would agree that there is, even though there is also a more objective and even an idealized notion of rationality.
Defending the Deontological Conception of Epistemic Justification
Now, let’s return to epistemic justification. As I said, Lockie equivocates between ‘rationality’ and ‘justification’. But it is not clear that we should do so. For one thing, ‘rationality’, even though it is clearly a normative term, is not obviously a deontological term—I mean, a term that has to do with responsibility and blame—whereas ‘justification’ does seem to have a deontological connotation. As William Alston, who himself criticizes the deontological conception of epistemic justification, acknowledges, the term ‘justified’ is most naturally understood deontologically.
Should we indeed understand epistemic justification deontologically? Elsewhere, I have argued in more detail that there is good reason to think that we should. Here, I would like to make just three brief points about epistemic justification that give us some reason to take this suggestion seriously.
First, it is widely thought that any plausible conception of epistemic justification ought to show a concern with the twofold epistemic goal that William James famously discussed, namely believing the truth and not believing falsehood. This criterion distinguishes epistemic justification from, say, moral or prudential justification. Now, epistemic justification, deontologically conceived, clearly has to do with the Jamesian goal of acquiring true rather than false beliefs. The person who violates no obligations in coming to hold or in maintaining a belief shows a concern with the goal of having true rather than false beliefs, especially if that person reflects on the issue in question, as, for instance, the Uzbek peasants did.
Of course, that one’s belief is deontologically justified does not mean that the belief in question is likely to be true. The tribesman’s beliefs that are based on tradition, for instance, aren’t formed in a truth-conducive way, even though they are deontologically justified. However, we should note that the idea that a belief is epistemically justified only if it is likely to be true is problematic, for that idea would rule out from the very start all sorts of accounts of epistemic justification that are being defended in the literature. Here, we should think, for example, of internalist accounts that require merely that the belief in question be based on a ground that is adequate from the subject’s perspective.
According to Richard Feldman and Earl Conee, for instance, one is epistemically justified in believing a proposition if and only if believing that proposition fits one’s evidence (which is why they call their view ‘evidentialism’). Of course, being justified in this sense doesn’t imply that one’s ground is actually truth-conducive. The tribesman, for instance, may well be epistemically justified on this view. But it seems we can’t rule out evidentialism merely because epistemic justification isn’t truth-conducive on this view.
What seems required for a plausible conception of epistemic justification is that it shows the relevant connection with the Jamesian goal of having true rather than false beliefs. And believing in a way that fits one’s evidence is generally an adequate way of having true rather than false beliefs. Hence, it seems the deontological conception of epistemic justification meets one of the most important criteria that one might put forward for an account of epistemic justification: it shows a relevant concern with the Jamesian goal.
Second, one might think that a plausible conception of epistemic justification ought to be analogous to conceptions of justification in other realms of life. What immediately comes to mind is, of course, actions’ being justified. What features do justified actions usually display? Well, at least two of them come to mind. The first one is that if an action is justified, then in performing that action, one does not violate any rules or obligations or principles, where these can be, say, moral, prudential, or legal. An action is, thus, justified if it’s permitted by the relevant rules. Second, usually—or maybe even always—when an action is justified, one is blameless for performing it. If Henry was justified in pulling the trigger, he is blameless for it. Mutatis mutandis, the same seems to apply to other things for which one can be responsible: character traits, emotions, intentions, and so on. If they are justified, then one doesn’t violate any obligations in having them and one is blameless for having them. If epistemic justification for beliefs is analogous to justification in other realms—which seems prima facie plausible, that is, plausible unless we have good reason to think otherwise—it meets the core conditions of what it is to be justified.
Third, one might think that epistemic justification is a necessary condition for knowledge. This is controversial: several philosophers have argued that epistemic justification is not necessary for knowledge. In fact, Alston is one of them. But let’s assume that epistemic justification is necessary for knowledge. Could deontological justification be necessary for knowledge? It seems to me there is good reason to think that it is. For, if one is not deontologically justified, one’s belief is either beyond justification (e.g., the belief of a psychopath) or unjustified because one has violated certain duties that have to do with the aim of holding true rather than false beliefs. It’s hard to see how someone who has violated certain duties regarding p when it comes to holding true rather than false beliefs could know that p. Of course, much depends here on how exactly these duties are spelled out. But, intuitively, someone who does not care for the truth of p, who believes p merely because she wants p to be true, or who has avoided counter-evidence regarding p, does not seem to be deontologically justified in holding p, and it also seems that such a person doesn’t know that p, even if p is true.
Of course, each of these points needs further elaboration and defense. However, they jointly suggest that, rather than taking a step back and claiming that deontological justification is one way, namely a perspectival and subjective way, of being justified, one might simply bite the bullet of the epistemic poverty objection and say that, even though there is such a thing as being based on an adequate ground or being truth-conducive, epistemic justification requires no such thing and the most plausible conception of epistemic justification is a deontological one. After all, it meets three important conditions on epistemic justification that other conceptions of epistemic justification do not meet (e.g., most forms of reliabilism wouldn’t meet the second condition). In this way, the epistemic poverty objection could be met head-on. One could, then, discuss separately how best to understand epistemic rationality, if there is any single such phenomenon.
In his paper “Perspectivism, Deontologism and Epistemic Poverty” Robert Lockie rightly points out that the epistemic poverty objection against the deontological conception of epistemic justification fails: epistemic justification might well be understood more subjectively. In addressing the poverty objection, though, Lockie makes the problematic assumption that epistemic justification and epistemic rationality are identical. Moreover, his arguments in favor of his view that epistemic rationality is perspectival, where the relevant perspective includes cultural-historical circumstances, aren’t entirely convincing. It is wiser to keep epistemic justification and epistemic rationality apart. Moreover, it’s worthwhile to bite the bullet and explore the idea that the deontological conception of epistemic justification is not merely a conception of a particular kind of epistemic justification but the right account of epistemic justification simpliciter.
Alston, William P. “Justification and Knowledge.” In Epistemic Justification: Essays in the Theory of Knowledge, 172-182. Ithaca: Cornell University Press, 1989a.
Alston, William P. “The Deontological Conception of Epistemic Justification.” In Epistemic Justification: Essays in the Theory of Knowledge, 115-152. Ithaca: Cornell University Press, 1989b.
Audi, Robert. (2011). Epistemology: A Contemporary Introduction to the Theory of Knowledge, 3rd ed. (London: Routledge).
BonJour, Laurence. “Externalist Theories of Empirical Knowledge.” In Midwest Studies in Philosophy, Vol. V: Studies in Epistemology, edited by Peter A. French, Theodore E. Ueling Jr., and Howard K. Wettstein. 53-73. Minneapolis: University of Minnesota Press, 1980.
BonJour, Laurence “Reply to Sosa.” In Epistemic Justification: Internalism vs. Externalism, Foundations vs. Virtues, edited by Laurence BonJour and Ernest Sosa, 173-200. Oxford: Blackwell, 2003.
Feldman, Richard, and Earl Conee. Evidentialism: Essays in Epistemology Oxford: Oxford University Press, 2004.
Foley, Richard. “Justified Belief as Responsible Belief.” In Contemporary Debates in Epistemology, edited by Matthias Steup and Ernest Sosa Oxford, 313-326: Oxford: Blackwell, 2005.
James, William J. “The Will to Believe.” in The Will to Believe and Other Essays in Popular Philosophy, 13-33. Cambridge, MA: Harvard University Press, 1979.
Lockie, Robert. “Perspectivism, Deontologism and Epistemic Poverty.” Social Epistemology,(2015): 1-17. doi: 10.1080/02691728.2014.990281
Luria, Aleksander Romanovič. Cognitive Development: Its Cultural and Social Foundations. Cambridge, MA: Harvard University Press, 1976.
Nottelmann, Nikolaj. Blameworthy Belief: A Study in Epistemic Deontologism Dordrecht: Springer), 2007.
Peels, Rik. “Responsible Belief and Epistemic Justification.” Synthese, forthcoming, 2016a.
Peels, Rik. Responsible Belief: An Essay at the Intersection of Ethics and Epistemology. New York: Oxford University Press, forthcoming, 2016b.
 This publication was made possible through the support of a grant from the Templeton World Charity Foundation. The opinions expressed in this publication are those of the author and do not necessarily reflect the views of the Templeton World Charity Foundation.
 Alston 1989b, 145.
 BonJour 2003, 176.
 See Luria 1976, 107-109.
 Luria 1976, 116.
 Another worry one may have is this: couldn’t it be that the Uzbek peasants, or at least a fair number of them, were able to make the valid deduction, but didn’t do so, not because of a lack of insight into the validity of deduction in unfamiliar contexts, but because in their society there is a rule that says that one should assert something only if one has seen (heard, felt, etc.) it oneself. One of the peasants, for instance, says: “we always speak only of what we see; we don’t talk about what we haven’t seen.” (Luria 1976, 109) In that case, the absence of asserting the conclusion that follows logically from the premises wouldn’t be a cognitive failure, but the result of a social barrier that psychologically prevents one from making such assertions.
 It is not unusual to distinguish rationality from justification. See, for instance, Feldman and Conee 2004.
 See Alston 1989b, 115-116, 143.
 See Peels 2016a; 2016b.
 See James 1979, 24.
 E.g. BonJour 1980, 54; Foley 2005, 317; Nottelmann 2007, 55.
 See Feldman and Conee 2004.
 See Alston 1989a, 172-182; 1989b, 144. For the same view, see Audi 2011, 270-282.