Acta Analytica

, 24:263

The Value of Lesser Goods: The Epistemic Value of Entitlement


    • Department of PhilosophyStockholm University

DOI: 10.1007/s12136-009-0061-5

Cite this article as:
Janvid, M. Acta Anal (2009) 24: 263. doi:10.1007/s12136-009-0061-5


The notion of entitlement plays an important role in some influential epistemologies. Often the epistemological motive for introducing the concept is to accommodate certain externalist intuitions within an internalist framework or, conversely, to incorporate internalist traits into an otherwise externalist position. In this paper two prominent philosophers will be used as examples: Tyler Burge as a representative of the first option and Fred Dretske as one of the second. However, even on the assumption that the concept of entitlement is sufficiently clarified, accomplishing these results is easier said than done – especially if we also want to ascribe positive epistemic value to entitlement. It will be shown that the epistemic value of entitlement is either granted at the expense of the epistemic value of justification or the value ends up below the level of value at which the epistemologists employing the concept of entitlement are aiming.


Epistemic valueEntitlementExternalismInternalismBurgeDretske

1 I

The current discussion of epistemic value focuses on the epistemic value of knowledge. However, the epistemic value of other lesser epistemic states, some of which are constituents of knowledge, is also interesting to investigate. Insofar as an epistemic state is a constituent of knowledge, we would thereby account for its contribution to the epistemic value of knowledge.

One of these states is entitlement that some influential epistemologists have introduced to cover unmapped terrain within the field of epistemology. Two of them are Burge and Dretske. They share little common ground concerning the nature of entitlement since, as we shall see, their motives for introducing the term are different. Nevertheless, one shared feature in their accounts that is of primary concern for this paper is that entitlement is a less demanding notion than justification. The requirements for entitlement are weaker than those for justification in at least one significant aspect. In order to determine the epistemic value of entitlement in their respective accounts, we need first to investigate what these weaker conditions consist in. Let us start with Burge.

Burge’s primary motive for introducing the concept of entitlement is to counter the epistemological trend towards “hyper-intellectualization” that Descartes started.1 Instead of a decisive break between the cognitive capacities of higher animals and small children on the one hand and mature humans on the other, Burge stresses continuity. Many philosophers have denied that higher animals and small children have cognitive capacities, and often the same philosophers exaggerate the control and understanding mature humans have over their cognitive capacities. Wishing to include more mental capacities and states as genuinely epistemic, Burge in his seminal paper “Content Preservation” distinguishes between justification and entitlement as subcategories of warrant in the following manner

Although both have positive force in rationally supporting a propositional attitude or cognitive practice, and in constituting an epistemic right to it, entitlements are epistemic rights or warrants that need not be understood or even accessible to the subject.2

Burge elaborates in a subsequent passage,

We are entitled to rely, other things equal, on perception, memory, deductive and inductive reasoning, and on [...] the word of others. The unsophisticated are entitled to rely on their perceptual beliefs. Philosophers may articulate these entitlements. But being entitled does not require being able to justify reliance on these resources, or even to conceive such a justification. Justifications, in the narrow sense, involve reasons that people have and have access to. These may include self-sufficient premises or more discursive justifications. But they must be available in the cognitive repertoire of the subject.

Even though these quotations differ in some aspects,3 they nevertheless show that the way in which entitlement is weaker than justification concerns the epistemic capacities of the subject. The constitution of the relation of the genus warrant, to which entitlement and justification are species, is a controversial topic, but let us for the sake of argument in this section follow Burge in that truth conduciveness constitutes one criterion for the relation of warrant that holds between the relata evidence E and belief that P.4 The difference between justification and entitlement, then, does not consist in that the relation of entitlement describes a lower degree of truth conduciveness, but rather in the fact that the access to the epistemic properties of her belief in general and the relation of warrant in particular that the subject is required to have in order to be entitled to believe that P is less demanding than in the case of justification. It is precisely these weaker conditions on the required access to the relevant epistemic properties that make entitlement an externalist notion.5 Although Burge seems to require that the evidential relatum (the ground) is accessible to the epistemic subject when she is entitled to the belief that P, the adequacy of the ground qua ground of the belief that P need not be accessible to the subject.6

Moreover, as the list given in the second quotation suggests, Burge thinks entitlement in this moderately externalist sense is sufficient for knowledge (together with the truth of the entitled belief of course) in many cases and can thus replace justification as the epistemic constituent of knowledge. Stronger internalist requirements for access are therefore superfluous in those cases.7 Hence, a subject that knows that P in virtue of her being entitled to her true belief that P needs only to have access to the ground or evidence E for P, but neither does she need to have access to the adequacy of the ground E, nor does she need to have access to the epistemic principle in virtue of which E constitutes an adequate ground of the belief that P.

On the basis of this brief analysis of Burge’s view of entitlement, some considerations concerning the epistemic value of entitlement will now be given. Most often the epistemic notion whose epistemic value epistemologists want to determine is knowledge. We may thus as the first question ask whether the fact that the subject’s warrant consists of mere entitlement in a case where she knows that P makes any difference in contributive value to the epistemic value of her knowledge that P compared to if her warrant had instead consisted of justification. In other words, how does entitlement contribute to the value of knowledge in comparison to the contribution that justification offers?

Well, according to Burge there can be no difference in contributive value from the point of view of veritism – the view that the final epistemic value of knowledge that P solely derives from the truth of the belief that P – in that an entitled belief that P has as high a degree of truth conduciveness as the belief would have if the belief that P was justified. Thus according to veritism, the epistemic value of knowing that P based on entitlement is the same as knowing that P based on justification.8

This preliminary verdict may lead us to question the value of the contribution justification gives over and above mere entitlement. Does this not show that the more demanding form of access that justification requires is superfluous for the epistemic value that knowledge yields? Relying on the centrality of epistemic value that many participants in the discussion ascribe knowledge might even make us question the epistemological significance of justification as an epistemic concept. This is precisely one of the objections that externalism directs against the internalist notion of justification with its stronger requirement on access. Unlike such externalists, however, Burge wants to retain the concept of justification, but the considerations alluded to here suggest that once we introduce an externalist concept like entitlement as an epistemic constituent in knowledge, the motivation behind justification becomes moot. Reaching the level of entitlement is the key epistemic achievement, and any concept depicting someone that reaches beyond that level has as little epistemological interest as any concept for reaching beyond knowledge has.9

One way of defending the superiority of internalist accounts against this objection would be to claim that a position that rests content with the epistemological significance, as well as the epistemic value, of entitlement at the expense of justification falls prey to an objection directed against (process) reliabilist versions of externalism combined with veritism. According to this objection, such positions fail to capture the epistemic value that knowledge enjoys over and above mere true belief. The true belief does not acquire any additional value by also being reliably produced, or so the so-called “swamping” objection, or problem, goes.10

In an interesting response to this objection, Erik Olsson claims that process reliabilism can indeed ascribe a higher degree of epistemic value to knowledge than mere true belief since, given some plausible empirical assumptions, knowledge acquired through reliably produced true belief ensures that the belief in question enjoys a stability that merely true beliefs lack.11 Whereas it was a matter of luck that the true belief was acquired on that occasion, misfortune may just as easily make the belief disappear again. In comparison, a reliably produced true belief ensures a stability of the belief in question that makes reliably produced true beliefs more valuable from the point of view of veritism.

That stability of true beliefs is an epistemic good will be assumed for the purposes of this paper.12 The question then becomes whether Burge’s account of entitlement can benefit from this response. In other words, can the limited form of access that entitlement entails ensure the necessary stability of beliefs that we know on the basis of entitlement?

Olsson’s response essentially consists in that the subject becomes aware of the reliability, or lack thereof, of her belief-forming mechanisms. However, if the subject becomes aware of the reliability of these processes, then her access reaches beyond merely having access to the ground E of her belief that P; in effect, she will become aware of whether the ground E that she has access to is in fact an adequate ground for her belief that P. Her disposition to preserve her belief that P thus depends upon her acquiring access to the adequacy of the ground E for her belief that P, but then her belief is no longer merely entitled but justified.13 Burge’s account of entitlement thus cannot rely on Olsson’s response in accounting for the epistemic value of entitlement since the response leads to the converse result compared to the intention by which we introduced the response in the first place. The response suggests instead that justification enjoys an epistemic value over and above mere true belief by being stable, but leaves the epistemic value of entitlement unaccounted for.

So far, we have thus not found any single value for which the following intuitively desirable relation would hold: that mere entitlement is less valuable than justification, but more valuable than mere true belief. As we saw in the case of contributive value in terms of truth conduciveness, the value of justification and entitlement is the same, whereas if we as solution to the swamping problem bring in stability as an instrumental value, then justification is ensured a higher value than entitlement, but entitlement does not yield a higher value than mere true belief in this respect.

In addition, despite the fact that entitlement and justification require the same degree of truth conduciveness, Olsson’s response entails a further problem for this account of entitlement. The plausible empirical assumptions mentioned above also seem to have the consequence that the number of merely entitled beliefs for a subject that are false are larger than the number of false beliefs that are justified for that subject since the ways in which the subject sorts out those entitled beliefs that are false from the true ones turn the surviving beliefs into justified beliefs. Insofar as the version of veritism that one prefers also attaches epistemic value to avoiding falsehood, mere entitlement is, accordingly, in this respect as well less valuable than justification.

Moving beyond veritism, the inability of this account of entitlement to employ Olsson’s response highlights anew the limited epistemic competence of the epistemic subject that entitlement entails. Those philosophers who in addition want to evaluate the performance of the epistemic subject in terms of epistemic credit should thus attach a low degree of epistemic value to an entitled believer. Now, the epistemic credit a subject deserves stands in proportion to the epistemic access that the subject has since the epistemic abilities of the subject manifests the relevant access. The epistemic credit is thus considerably lower in the case of a merely entitled believer than in the case of a justified believer. This result in turn suggests a difference in epistemic value between the two corresponding cases of knowledge since the epistemic credit of a knowing subject that merely is entitled to her belief that P is lower than the credit of a knowing subject that is justified in believing that P.14 Certainly an epistemic subject capable of meeting the stronger demands of access with its entailed conceptual abilities deserves more epistemic credit than a subject that lacks them. Whether the resulting bifurcation of the epistemic value of knowledge is a problem is left for credit theorists to ponder.

Returning to Olsson’s response at this point, we may furthermore note that although stability is an instrumental good for true beliefs, stability is not a constituent of knowledge, warrant or entitlement unlike truth conduciveness, which is a constituent of both entitlement and justification on the present account. It is true that the precise degree of truth conduciveness that a certain belief enjoys is extrinsic to that belief in that, first, what degree is required for a particular belief to be entitled is not part of the concept of entitlement and, second, whether that degree of truth conduciveness is actually fulfilled by the evidence E for the belief that P cannot be determined by analysis of the concept of entitlement alone. Nevertheless, truth conduciveness belongs to the concept of entitlement.

Now, this author suggests that it is a desideratum that the constituents of entitlement, as well as any epistemic state, determine its epistemic value – truth conduciveness on the present account. One could argue that this desideratum is fulfilled in the case of justification. In contrast to the relaxed requirement on accessibility for mere entitlement, the strong requirement on accessibility for justification entails conceptual abilities that make the stability of a justified belief constitutive of a justified subject: ceteris paribus preserving a belief that P to which a subject has access to the adequacy of the ground for that belief is constitutive for the subject qua justified believer, but not so for a merely entitled believer. The desideratum that the constituents of an epistemic state determine the epistemic value of that state is thus not fulfilled by entitlement in the case of stability.

At this point you may want to question the veritist assumption that the epistemic value of entitlement is a function of its truth conduciveness. Such a tight connection between entitlement, truth conduciveness and epistemic value may seem like too great of a concession to externalism. In order to exploit an alternative route, let us somewhat paradoxically turn to an externalist epistemologist who introduces the notion of entitlement in order to capture internalist intuitions: Dretske.

2 II

In his paper “Entitlement: Epistemic Rights Without Epistemic Duties?”, Dretske, on the one hand as the externalist he is, accepts that entitlement does not require reasons as justification does, but, on the other hand, employs the notion of entitlement to accommodate certain internalist intuitions.15 According to him, an epistemic subject can be entitled to believe that P in two ways. The first is through knowing that P.16 In that case, however, entitlement is a consequence of the knowledge that P rather than a constituent of that knowledge. Just as Dretske does himself, we shall instead focus on those cases where entitlement constitutes the warrant that the knowledge that P is based on.

The internalist intuition that Dretske wants to capture with his notion of entitlement is the case of the misfortunate brain in a vat that finds itself in the epistemically hostile environment where, despite being epistemically scrupulous and responsible, none of its beliefs about the external world are true. Many here think that there is one form of warrant according to which the poor brain in a vat is warranted in its beliefs even though none of them are true. Dretske is one of them, and he uses the notion of entitlement to depict that form of warrant. He says,

I can imagine some benighted soul – a brain in a vat will do – whose beliefs are false but whose total evidence – both the evidence he has and the evidence he can, by assiduous effort, obtain – is the same as mine. If his beliefs are false and mine true, it nonetheless strikes me that he has the right to believe whatever I have the right to believe.17

This internalist form of entitlement omits truth conduciveness as a constituent.18 The right to believe that entitlement amounts to is thus based on something else. The question is what. Dretske connects this right in the predicament that we, together with the brain in a vat, find ourselves concerning perceptual beliefs with the immediacy and irresistibility of these beliefs.19 At the same time, not everyone is entitled to every belief that presses itself onto us in those circumstances because there might also exist various defeaters that prevent the subject from being entitled. Epistemically responsible subjects, however, take these defeaters into account. Only they, therefore, enjoy the right to have entitled beliefs. In the same vein, a brain in a vat can also be more or less responsible despite the fact that all its beliefs about the external world are false.20 We are both entitled to the belief that X is sitting on the sofa in many circumstances, but not if we know, or at least are entitled to believe, that X has a twin or the light is dim. Having thus concluded the discussion of these criteria to his own satisfaction, Dretske suggests that the entitlement pertains to “the fact that the belief is unavoidable for an epistemically responsible agent.21

What epistemic value does the right to believe that entitlement amounts to depict on Dretske’s account? Two main options suggest themselves. The first and weaker alternative is to say that an entitled subject is blameless in believing what she does. Despite all its beliefs being false, the brain in a vat is, nevertheless, faultless in continuing to entertain these beliefs, and thus permitted to believe that P.22

However, even though it may be granted that the brain in a vat is blameless in some sense, from an epistemic point of view the state of the brain in a vat does not seem to be one that we should give a positive epistemic designation like being entitled, but rather take at most a neutral stance towards – just as the notion of being blameless itself suggests.23 Blamelessness per se does not have positive epistemic value, only the blamelessness of “an epstemically responsible agent” does. Thus, insofar as we want to credit the subject, as Dretske clearly wants to do, it is not for the particular belief in question, but rather for the epistemically responsible character that the subject has. That character is a general trait that does not entail that every belief that such a subject has enjoys the same positive epistemic value. Entitlement precisely depicts a class of beliefs that does not enjoy the same epistemic value as the character of such a subject, in contrast to justified beliefs where there is parity in epistemic value between the belief tokens and the epistemically responsible character. An epistemically responsible character thus enjoys a higher degree of epistemic value than an entitled belief does. The epistemic misfortune of the brain in a vat consists precisely in that, despite its epistemically responsible character, it never reaches a higher epistemic level than entitlement.

Having already realized the possibility of a higher level of warrant and, accordingly, of epistemic value that justification represents also dooms the second option of equating entitlement with the higher epistemic value of being praiseworthy.24 For precisely the very reasons that we gave in the preceding paragraph together with the relaxed conditions on the epistemic abilities of the subject, having an entitled belief that P does not seem to be worthy of praise. In perceptual cases the contribution of the subject often simply consists of having her eyes open and looking in the right direction. Dretske, as well as Burge, stresses that entitlement is defeasible, but it seems that at the very moment in which the subject succeeds in defeating a present challenge, then the subject acquires justification instead of mere entitlement precisely in virtue of being able to give a successful response to the challenge.25 Being able to defeat challenges is indeed an epistemically praiseworthy ability, but that praise attaches to the epistemic character of the subject or the resulted justification (or both),26 not to the initial entitlement.

At this point, one might object that both entitlement and justification deserve praise, but to different degrees. Entitlement is praiseworthy, but simply less so than justification. However, to respond in this way seems ad hoc for an account of epistemic value based on Dretske’s view since the properties of justification that we praise precisely are those that entitlement not only has to a lesser degree but completely lacks, like the ability to articulate one’s grounds in an epistemological vocabulary, assessing them, responding to challenges and so on. Insofar as entitlement deserves praise, it must thus be for some other reason, but we have not found any so far in Dretske’s account. In fact, we have yet to find any substitute for truth conduciveness as constituent of entitlement. Because of the lack of any substitute for truth conduciveness, a purely deontological conception of entitlement floats in the air, and we, therefore, lack a way of distinguishing entitlement from justification. Thus, we have yet to find any genuine alternative to Burge’s account of entitlement where we at least found a working conception and a clear cut difference in access that separates entitlement from justification.

In neither Burge’s nor Dretske’s account, however, have we come up with any satisfactory account of the epistemic value of entitlement. One may therefore wonder whether the conclusion to draw, then, is so much worse for entitlement as a fruitful epistemological concept or question whether the epistemological significance of an epistemic concept is exhausted by its epistemic value. What do you think?


(Burge 2003, 503).


(Burge 1993, 458).


See the careful analysis conducted in (Casullo 2007).


(Burge 1993, 470–1 and Burge 2003, 506).

Concerning perceptual entitlement Burge writes:

Entitlement is epistemically externalist inasmuch as it is warrant that need not be fully conceptually accessible, even on reflection, to the warranted individual. The individual need not have the concepts necessary to think the propositional content that formulates the warrant [...] Justification is warrant by reason that is conceptually accessible on reflection to the warranted individual. (Ibid, 504–5, emphasis in original).

See also (ibid. 528–9 and 547).


See Casullo’s lucid discussion in (Casullo 2007, especially 277–8). Casullo here draws on Alston’s fruitful distinction between access to the ground of the belief that P and access to the adequacy of the ground of the belief that P (Alston 1989a). Casullo adds a third kind of access: access to the epistemic principle governing the ground of the belief that P. The third kind is entailed by the second kind, but not conversely, which presumably amounts to that you might have access to an epistemic principle as well as the ground without having access to the fact that the ground in fact is adequate in a particular case to which the principle applies. In the same vein, one could access the adequacy of the ground of the belief that P without having access to that ground itself in a particular case: one could determine that observing the thief stealing from the cookie jar would be adequate as evidence for knowing who stole from the cookie jar without having observed the actual theft.

Besides the quotation and references given in note 5 above, further evidence for the interpretation that Burge requires, but does not go beyond, access to the ground that P is that when he describes what entitled believers lack he typically lists epistemic concepts like “epistemic, warrant, entitlement, reason, reliable, competence, entails, perception and perceptual state” (Burge 2003, 521). See also the notes Burge attaches to the referred passages in (ibid, no. 1 and 37). Casullo is therefore right when he, in turn, connects the two stronger requirements on access with second-order justification (Casullo 2007, 269–70). Note thus that justification in virtue of its stronger condition includes entitlement. Any contrast made between entitlement and justification in this section should thus be understood as between mere entitlement and justification that includes but goes beyond entitlement.


Burge, for instance, says that “[o]ur entitlement to ordinary perceptual belief is usually sufficient for perceptual knowledge” (Burge 1993, 485). In Casullo’s analysis, Burge comes out on the same side as Alston regarding the internalist requirements of knowledge despite their terminological differences (Casullo 2007, 278).


The above definition is a charitable interpretation of Olsson’s (and other’s) definition of veritism as “the view that true belief, and true belief only, has final or intrinsic epistemic value” (Olsson 2007, 344). Strictly interpreted, this definition would yield the point above concerning the same contributive value of entitlement and justification as a trivial outcome since the value of knowledge in both cases would be zero. Thanks to Jens Johanssom and Jonas Olson for this observation.


However, Pritchard considers the possibility that “there are epistemic standings that are more robust than the epistemic standing required for mere knowing, where these elevated epistemic standings raise the value of the knowing state” in (Pritchard 2007, 100).


One such critic of reliabilism, Swinburne, says in this regard: “[s]o long as the belief is true, the fact that the process which produced it usually produces true beliefs does not seem to make that belief any more worth having” (Swinburne 1999, 58).


(Olsson 2007) where further references to the epistemologists who raise the swamping objection are also given. The empirical assumptions are non-uniqueness, cross-temporal access, learning and generality (ibid, 348). More precisely the defended thesis (RST) says that “[t]he probability that S’s belief that p will stay in place is greater, conditionally upon S’s having a reliably acquired true belief that p, than it would be conditionally upon S’s having a mere true belief that p” (ibid, 347).

See also (Goldman and Olsson 2009) where the value of repeatability is offered as a response to the swamping objection; that is given the assumptions cited in the previous paragraph, a reliably produced true belief ensures a higher probability of future true beliefs than a mere true belief does. Unlike stability, as we shall see, repeatability may be an epistemic value that entitlement enjoys to some degree. The difference is due to the internalist features of stability as understood by Olsson (see below). Although this is an empirical question, this author is, however, inclined to think that justification enjoys a higher degree of repeatability and, thus, epistemic value in this respect since, just as in the case of stability, the role played by the four assumptions transform entitled beliefs into justified ones.


Even if no higher epistemic value would be secured by stability, Olsson also argues for the higher practical value of stability for reliable true beliefs over mere true beliefs in (Olsson 2007, 347–9). It should be noted that the value in both cases are instrumental rather than contributive value.

Note that Olsson’s empirical assumptions pertain to the method by which the belief is generated and that the third assumption is learning (Olsson 2007, 348), which then amounts to that the subject learns that the method is reliable. Burge says in this regard:

[t]he reliability could be inductively learned by the individual. But then the inductive connection would be the source of warrant – in fact, justification. (Burge 2003, 532).

Olsson calls the assumed ability to keep track of beliefs “a modest internalist requirement on a cognitive agent” (Olsson 2007, 352, italics in original) that is not part of the analysans of knowledge in terms of reliability, but instead figure “as essential elements of the environment in which knowledge attains its maximum worth” (ibid.).

It has been pointed out to this author that an externalist might question this internalist requirement and claim, for instance, that reliably formed beliefs are more stable simply because they are less likely to encounter defeaters. Thanks to Albert Casullo for this example. However, one should note that Olsson endorses a comparative thesis. That a reliable produced belief that fulfills this “modest internalist requirement” is more stable than mere true belief is consistent with a merely entitled belief landing in between these two cases in terms of stability. This relation of epistemic value would even be the intuitively desired result (see below), but establishing this tripartite relation goes beyond anything Olsson has shown and requires empirical investigations that fall outside the scope of this paper. It may be the case that stability has a crucial threshold over which stability ceases to add value and if that threshold is passed by mere entitlement, then we are back at the problem of justification being superfluous, where as if the threshold is only passed by justified beliefs, then the swamping problem remains. Note thus that the target in the dialectic of this paper is not a wholehearted externalist, but precisely a position, which attempts to find room, and positive epistemic value, for both for the externalist notion of entitlement and the internalist notion of justification.

Swinburne’s conviction that internalism is superior to reliabilism in accounting for epistemic value seems to rest on the fact that a subject with a justified belief “is almost always at least half-conscious” of the correct criteria of justification by which she forms her beliefs and thus has more true beliefs than a subject with a reliable produced true belief (or avoids having false beliefs about these criteria as unjustified subjects do?) (Swinburne 1998, 64). This response is not available for subjects with entitled beliefs since beliefs about such criteria concern the adequacy of the ground for the belief and such beliefs are lacking in the case of merely entitled beliefs. Nor is Kvanvig’s view that “justification is valuable independently of the value of true belief in virtue of being more accessible to reflection than is the property of truth” (Kvanvig 2003, 74) available since it is precisely in terms of accessibility that mere entitlement is inferior to justification. Insofar as truth conduciveness can be accessible on reflection, such access would determine the adequacy of the grounds for the belief that P and is thus not accessible to a merely entitled believer.


Credit theorists face the problem that many subjects do not seem to deserve any credit for easily acquired cases of knowledge (like cases where the epistemic constituent consists of entitlement). See (Pritchard 2007, 98–100) and (Lackey 2009) for discussion and further references.


(Dretske 2000). It should be mentioned that during a workshop in Frankfurt in September 2008 led by Marcus Willaschek (see acknowledgements), Dretske retreated from some of his claims concerning entitlement from that paper in the face of the consequences that this internalist concession brings. The present section has been influenced by the stimulating discussions at that event. See also note 24 below.


(Ibid. 595–6).

(Ibid. 595). Shortly below, Dretske continues:

In light of the fact that his mistake is inextricable and that he, therefore, has no way of finding out his belief is false, the fact that it is false should not count against his right to believe. He is unlucky, a victim of circumstances, and I am not. But if I am entitled to my beliefs, he is entitled to his (ibid. 595–6).

Dretske here mentions the unlucky circumstances the brain in a vat finds itself in, but does not make the move mentioned in the next note (maybe he should have).


At least in the straightforward sense. One might attempt to save truth conduciveness in the face of the example with the brain in a vat by adding a normality constraint on the contexts in which the belief should be truth conducive. The entitled belief that P would thus be truth conducive in “normal” contexts or worlds. Various externalist safety or sensitivity constraints are versions of this move. Transglobal reliabilism is another ingenious alternative. See (Henderson and Horgan 2006) and (Henderson et al. 2007). Burge provides a transcendental argument for the conclusion that normal environments are privileged in assessing entitlement where the key step in the argument is that such contexts help to determine the nature of the content of the mental states such as belief (Burge 2003, 533–7). Burge thus connects externalism within epistemology with his well known externalism about mental content. All these alternatives of course reconnect entitlement with truth conduciveness and do not therefore do justice to the initial internalist intuition that we try to capture here.

In most deontological accounts of justification, truth instead enters as the goal of an intellectual obligation like that we should strive for maximizing our true beliefs. See (Alston 1989b, especially 201). However, some epistemologists regard entitlement, or some other epistemic concept, as being constituted by both a truth conduciveness condition and some independent deontological condition. This author believes that one lesson to draw from the problems raised in this section concerning entitlement is that such a wedding yields an unhappy marriage. The right way is instead to let the normative status depend upon truth conduciveness – just as Burge does when he ties epistemic goodness to truth as representational goodness in (ibid, 505–6). See also (Janvid 2004).


(Dretske 2000, 598).


(Ibid, 601–2.)


(Ibid, 603) italics in original.


Any obligations or duties that an epistemic subject has would not pertain to any particular beliefs, but rather consist in an obligation to be(come) “an epistemically responsible agent”. See (Alston 1989c and 2005, Ch 4) for more on deontological conceptions of warrant.


It does not seem to reflect Dretske’s own position either for a number of reasons. He consistently describes entitlement in more positive terms. For instance, it seems strange to say that someone who knows that P merely is blameless in believing that P. (Recall that knowledge entails entitlement according to Dretske.) (see next note).


As this author understood him, Dretske retreated from some of his claims concerning entitlement at the workshop when he realized that his concession to internalism concerning entitlement ought to make him concede in turn that the internalist notion of justification is an even better epistemic state to be in, which squares badly with his otherwise externalist epistemology where justification is superfluous. It seems difficult to deny that an epistemic subject that exercises her epistemic abilities in defeating challenges and so on reaches a higher epistemic state, and a corresponding increase of epistemic value, than someone who does not (see next note). This concession to the internalist notion of justification is also in tension with his view that you are entitled to believe everything that you know since the entitlement seems to downgrade knowledge.

In this regard Burge says:

To have the entitlement, the individual need not, however, have a warranted belief that defeaters are not in play. Such a requirement would make entitlement unattainable for higher animals and young children. It is enough that the individual lack a reason to avoid relying on the perception (Burge 2003, 544).

One should note that, just as in the case of warrant, it may be necessary to distinguish between stronger and weaker forms of access to defeaters where being able to defeat a challenge requires a stronger form of access in terms of conceptual abilities, just as in the case of justification, than merely taking a (-n overriding) defeater into account by giving up the targeted belief. This is, however, an interesting issue that deserves a paper of its own.

Dretske stresses that entitlement is defeasible and prima facie. In his comment on Dretske’s paper, Williams points out “[a]n appropriate challenge cancels entitlement, which can be re-established only by producing reasons that defeat the challenge. Defeasability and justificational commitments go together” (Williams 2000, 610; italics added). Williams thus answers the question Dretske poses in the title of the paper, whether there are epistemic rights without duties, with a qualified no – a subject may initially enjoy a right by default, but loses this right if she cannot perform the required duties when her right is (properly) challenged.


Can a brain in a vat ever be justified in any of its beliefs about the external world according to Dretske? To answer yes seems too large of a concession to internalism (albeit more in line with the initial internalist intuition), but to answer no is in tension with the line of reasoning in note 24.



Thanks to Albert Casullo, Fred Drestske, Steven Luper, Erik Olsson and Jan Willner for helpful comments on earlier versions of this paper. Earlier versions of this paper have also been presented at the higher seminar in practical philosophy in Stockholm, at a colloquium in Frankfurt and at the Bled Conference “Epistemic Virtue and Value” in June 2009. This author is grateful to the participants at those events for comments and discussion, especially my commentator Jonas Olson, Jens Johansson and Torbjörn Tännsjö on the first and the DDD group (Claudia Blöser, Claudia Cuadra, Hannes Ole Matthiessen and Marcus Willaschek) on the second occasion. Work on this paper was funded in part by the Deutsche Forschungsgemeinschaft project “Defeasibility and Discourse Dependence” at the University of Frankfurt am Main.

Copyright information

© Springer Science+Business Media B.V. 2009