A question re propositional (a.k.a. ex ante) warrant

Help! I have a question, and a follow-up, regarding propositional warrant – a.k.a. “ex ante warrant.” (I here follow Burge’s generic use of ‘warrant’. Beware of messing it up with Plantinga-warrant or Wright-warrant). I hope I can get some help. Here is the question:

Question: Can a subject, S, be propositionally warranted (at time t) in a proposition, p, that she is in principle incapable of thinking (at time t).

The question concerns the extent to which S’s limited conceptual resources limit the propositions that she may be propositionally warranted in. Here is the follow-up:

Follow-up: If so, what does it take for S to be propositionally warranted in p (at t)?

I am strongly inclined to answer ‘yes’ to the first question. It seems that we sometimes presuppose, at some stage our cognitive life, propositions that we cannot, at that stage, think or conceive of. And it seems that we may be rational in doing so. It is, however, not trivial that the question should be answered in the affirmative. For not only do we lack something such as a belief-generating process that may serve as the basis for an epistemic evaluation – we moreover lacks a *potential* belief-generating process. So, I am interested in hearing from people who think that our conceptual capacities set some limits on the propositions we can be propositionally warranted in.

However, my interest is mainly in the follow-up question. I am ready to assume that some psychological connection has to hold between S and p for S to be propositionally warranted in p – even when S cannot think that p. For example, we cannot automatically ascribe propositional warrant to S if someone smarter than S would believe that p had they been in her place.
To see this, consider a modestly talented mathematician, Sam, who believes that a set of axioms, A, is true. Assume moreover that a complicated theorem, T, follows from the axioms. Assume finally that the theorem T is too complex for Sam to think and, much less, prove. In such a case, it seems implausible to regard him as propositionally warranted in theorem T – even if a more talented mathematician could prove T from A (and would had she been in Sam’s shoes).

I have a couple of candidate examples of psychological connection (certain exercised competencies) that may underwrite the ascription of propositional warrant to S in a proposition, p, that she is incapable of believing. But I lack any sort of general characterization of such psychological links.

So, I would like to hear the opinions of folks with opinions about propositional warrant and the conditions under which it may be ascribed.


PS: The issue is important, I think, in its own right. But it is also important for assessing a particular kind of move that is often made in defending knowledge-accounts of assertion and action against obvious counter-examples. What I have in mind is the appeal to “excuses” or “secondary propriety” by Williamson, DeRose, Hawthorne & Stanley etc. But that is for another discussion.


A question re propositional (a.k.a. ex ante) warrant — 12 Comments

  1. Very interesting question that I need to ponder a bit more. I’m not all that familiar with Burges-warrant, but I’ve spent a good bit of time reading Plantinga’s accounts.

    My initial thought is that some theory of testimony might allow us to talk S having warrant in p (at t), where p is a proposition that S cannot think (at t). To take your mathematician as an example, Sam could be warranted in holding p if p has been presented to Sam in piecemeal fashion by the brigher mathmetician. That is, Sam can’t comprehend p as a whole (can’t think it), but has grasped p as its constituent parts and has been granted warrant for p by the brighter mathematician’s testimony that p.

    That doesn’t make much sense right now…but I had to do it quickly as I’m on my way out the door.

  2. Thanks for the suggestion Tom!

    The testimonial case presents special and very tricky issues. In the case you give, I am not sure that we want to ascribe the guy propositional warrant for the proposition that p (where p is the theorem in question). Isn’t it more plausible to ascribe to him a *doxastically* warranted belief that some proposition, p, that is some theorem he cannot think, is true.

    That said, testimony may well be one of the routes to propositional warrant in a proposition one cannot think. I’ll think more about it.


    PS: Burge uses ‘warrant’ very broadly as the genus under which internalist and externalist species fall. Thus used, ‘warrant’ is apt if one wants to side-step, at least as the outset, the internalist-externalist debates.

  3. Mikkel, interesting post here and a difficult issue. Here’s a good example of why the answer to your first question should be yes. The incapacity to entertain a proposition can arise in a number of ways, some in terms of mental limitations and some in terms of psychological limitations. Some of us are incapable of believing certain things about ourselves, not because they are false but because of ordinary functioning requires disbelief. In other people, this aversion takes the form of an incapacity to entertain the thought itself, it being too painful or disturbing to consider. In some such cases, the thought too painful to bear is one that is unfortunately warranted or justified.

    One might think of these cases in terms of “merely medical limitations,” but these psychological limitations are every bit as central to our constitution as our intelligence level. So if we want to abstract away from them, it gets harder to explain why we don’t abstract away from limitations on intelligence as well. But I think your “in principle unthinkable” clause in the question is meant to exclude finite enhancements of native endowments.

  4. Hi Jon,

    I agree that the sort of cases you gesture towards provide one reason to answer ‘yes’ to the question. It is a difficult question whether the “medical limitations” are typically “principled” enough to fall within the scope of the question. But I am, like you, inclined to think that there are propositions that we cannot think — not because of complexity but for other constitutive features of our psychology (self-preservation of some sort?). I am also inclined to think that we can be p-warranted in them. Such cases are clearly important for the theory of self-knowledge.

    But I am more interested in the cases where the limitation has to do with complexity of the proposition in question. Plausibly unsophisticated agents cannot think (I’ll use caps when *mentioning* thoughts and their constituents): I AM NOT NOW SUBJECT TO AN ILLUSION, THE EPISTEMICALLY RELEVANT CIRCUMSTANCES ARE NORMAL, THE TWO OCCURENCES OF ‘SOCRATES’ IN MY REASONING ARE UNIVOCAL or I KNOW THAT P.

    Such complex, often higher-order, propositions are sometimes said to be a subject’s cognitive presuppositions. Some theorist hold (or say things that require) that we can be p-warranted in them. Now — I think that even subject’s who is in principle unable to think these propositions may be propositionally warranted in them. If so, there is an important, I think, question as to the conditions under which such propositional warrant should be ascribed. (I certainly don’t think it comes “for free.”)

    Theorists who seem to require that a subject can be p-warranted in such complex (often higher-order) propositions have done little to account for the basis of the ascription of the warrant. (Recall the connection to the secondary propriety maneuver that I mentioned in the PS of the post.)


  5. Hi Mikkel,

    I agree with your assessment of Sam the ill-equipped mathematician. I think that we would do well to say that he is warranted in some proposition, p1, “that p is true,” than p proper. The warrant in such a case is bestowed by some model of testimony.

    I’m interested, now, in your last description of the case where a subject’s “cognitive presuppositions” are said to be p-warranted (at least by some theorists). Let’s, for our purposes, talk about Frank (Sam’s done trying to learn p, he gave up) who has warrant for a propostion, p, that the grass in the courtyard is green. Now, Frank has background knowledge (about the use of the English language, colors, etc) and a causal chain (say, visual perception) that (by whatever theory we want to play with) grants warrant for p. He also, we would presume, have warrant for all of these background beliefs, etc. But, the question you seem to be asking is whether Frank has warrant for some proposition, p’, that is a “cognitive presupposition” that is not necessarily present to his mind when he grasped p, nor part of the causal chain granting warrant to p.

    p’ might be something like the proposition that “I AM NOT NOW SUBJECT TO AN ILLUSION, THE EPISTEMICALLY RELEVANT CIRCUMSTANCES ARE NORMAL, etc, etc,” but this proposition is beyond Frank’s humble cognitive abilities, he cannot “think” it, as we’ve been saying. If this belief was not necessary for Frank to holp p (and be warranted in it), then where does it come from? Is it something like a derrivative of p, or a constituent (yet unrecognized) part of p?

    I guess what I’m getting at is something like this: p’ is not a necessary component of p in that Frank doesn’t need p’ to have warrant for p (and this is good, b/c he can’t grasp p’). So if he’s not thinking it, and doesn’t need it, how can it have warrant? Now, it’s possible that it has warrant if at some future time Frank can think it, but that warrant is not by virtue of p, nor is it existant a t1, when Frank can’t think p’.

    I may be talking in circles…but there’s something fishy to me about saying p’ (the unthinkable proposition) has warrant.

    Thanks for the fun,


  6. Hi Tom,

    I’m glad we agree on the math-case.

    Note that I have not been arguing that S needs to be propositionally warranted [henceforth: p-warranted] in the for-S-unthinkable propositions to be warranted (propositionally or doxastically) in ordinary propositions. But some philosophers, especially those upholding a Closure Principle for warrant, uphold the view. So, they owe an account of the basis of the warrant.

    Unlike you (but like Jon), I am inclined to think that S can be p-warranted in *some* for-S-unthinkable propositions. What I have been suggesting is, contra Wright et al, that there must be some *psychological* basis for this warrant. But I have only examples of cases in which there seems to be a psychological basis for ascribing S p-warrant in a for-S-unthinkable proposition. And given that we can’t even appeal to possible belief-generating processes, it is not clear what can be appealed to in many of the cases in question. So, my follow-up question is a genuine question! I’d like to hear what people who have a view – who do regard S as p-warranted in for-S-unthinkable propositions (or ‘cognitive presuppositions’) – take to be the basis for the ascription of this warrant.

    Thanks again for chipping in,

    PS: A minor terminological point: I think your last remark “” … saying that p’ (the unthinkable proposition) has warrant” is misleading. It is misleading because it is *the subject* who is propositionally warranted. It is not the proposition that is warranted “in the abstract.” Perhaps the locution ‘propositional warrant’ is a little unfortunate since it may provoke this way of thinking.

  7. On the math case, is there anything wrong with saying that the incompetent mathematician has propositional warrant to believe the mathematical theorem, but he is unable to avail himself of this justification in forming a justified belief precisely because of his mathematical incompetence, i.e. he is unable to prove it?

    I like this response because it promises a fairly sharp distinction between epistemological questions about what one has justification to believe and psychological questions about one’s ability to avail oneself of it in forming justified beliefs. This distinction comes in handy all over the place, e.g. Jim Pryor’s work on Moore’s argument, for one example.

  8. Declan,

    I like the distinction you gesture towards. But I am not sure about the application here.

    Compare Sam to another mathematician Sally – a second-year grad student – who is better equipped than Sam. She is perfectly capable of thinking the propositional content corresponding to T. In fact, she has tried, and failed, to prove T from A. She then asks her teacher who tells her that there is indeed a complicated proof from A to T but that she will not learn the techniques required for the proof until 7th semester.

    In this case, (a version of) the distinction you gesture towards seems to be applicable. I think we should regard Sally as doxatically warranted in the belief that T. But we should also recognize that the warrant is testimonial and that her epistemic position vis-à-vis T is different in kind to that of her teacher. Testimonial warrant is different in kind from warrant by understanding/reasoning. In this case, it is clearly inferior.

    On the other hand, we should also preserve the sense that she is in a better epistemic situation vis-à-vis T than Sam. I guess we can agree that Sam has doxastic warrant by testimony in whatever he believes – i.e., that some theorem that is called ‘T’ (but which he can’t wrap his mind around) is true. The question, then, is whether he *furthermore* has propositional warrant (by testimony) for the proposition corresponding to T.

    I guess my inclination towards thinking that he does not is that the difference in Sam and Sally doxastically warranted beliefs seems to account pretty well for respective strengths of their epistemic positions with regards to T. Moreover, it seems to me that what Sam gets from the testimony does not go beyond the doxastic warrant ascribed to him. So, I think a version of the distinction that you are after may be preserved w/o ascribing him p-warrant for T.

    I am nowhere as confident about this as I am of the original case which was just meant to provide a case in which no propositional warrant should be ascribed. I guess that you agree with me on that one. So, the putative disagreement is only over testimonial warrant. (And no wonder: Testimonial warrant is so bloody difficult).

    Thanks for pushing on this. Push on, if I am missing something,

  9. Hi Mikkel, I think we may be at cross purposes. I wasn’t thinking about cases involving testimony, but I did mean to disagree about the original case. You claim that no propositional warrant should be ascribed to the incompetent mathematician, which is exactly what I want to dispute. I’m skeptical about the idea that psychological limitations impose corresponding limitations on what one has propositional warrant to believe; as far as I’m concerned, they only affect which doxastically warranted beliefs one is in a position to acquire. Hence, my suggestion that in your original example, the incompetent mathematician has propositional warrant to believe the theorem, which he is unable to avail himself of in forming a doxastically warranted belief. Perhaps there are other problems with this view, but so far I’m not convinced that the math case by itself is enough to refute it.

  10. My bad Declan! I guess, I was reading your comment with the principle of charity (from my POW, of course) attributing you the less extreme view (again, from my POW).

    But I am actually more interested in what may motivate your view on the original math case, than I am in the tricky case of testimonial warrant. Do you uphold a general principle along the lines: S is propositionally warranted in all entailments of S’s doxastically warranted beliefs. Such a principle would motivate your view for the case of Sam. But I do not see how anything like the principle itself is motivated. For example, I don’t think the useful distinction between having warrant to believe and being able to form warranted belief requires it. Moreover, I worry that upholding it is to set a foot on a slippery slope towards the view that everyone is warranted in every true proposition. But maybe you have another motivation for the view that Sam the mathematician is p-warranted?

    Note that our disagreement is only partial. We agree that the psychological limitations do not impose *corresponding* limitations on p-warrant. For I agree that S can be p-warranted in some propositions that S is incapable of thinking. However, I think that S may be ascribed propositional only if there is some psychological connection — to be characterized — to the proposition. I don’t have a general characterization of the sort of psychological connection that will do the trick — only some candidate examples. The examples, suggest, I think, that propositional warrant does not come all that cheap (and certainly not “for free”).

  11. Hi Mikkel – thanks, this is helpful.

    You ask whether I’d accept the following principle: S is propositionally warranted in all entailments of S’s doxastically warranted beliefs. Perhaps I could be persuaded to change my mind on this, but for now I’m inclined to accept it. The principle is much more plausible once you’ve distinguished sharply, as I’m inclined to, between epistemic claims about which propositions you have propositional warrant to believe and psychological claims about which doxastically warranted beliefs you’re in a position to form.

    You raise a worry: isn’t this to set foot on a slippery slope towards the view that everyone is propositionally warranted in believing every true proposition? I don’t think so, although perhaps it suggests something weaker, i.e. everyone is propositionally warranted in believing every a priori proposition. But I’m happy with that, since an a priori proposition is plausibly construed as one that everyone has a priori propositional warrant to believe.

    Btw, all I need is that this very impersonal conception of justification is useful for some theoretical purposes. I don’t need to reject the claim that more psychologically constrained notions of justification might be useful for different theoretical purposes. So, there may be less at issue in this dispute than appears at first sight.

  12. Hey Declan

    Thanks – I think I get your view a little better. I don’t object to pluralism wrt. warrants (e.g., I accept that there is internalist and externalist species.) What I do find implausible is the idea of an “impersonal” warrant that is entirely divorced from subject’s cognitive life.

    Propositional warrant is partly motivated because we can evaluate a subject epistemically wrt. to propositions he does not (and, we agree, cannot) believe. But such an evaluation should always, I think, make reference to the individual’s psychology. E.g., given that a subject sees the cat on the mat, the proposition the cat is on the mat may be warranted *for her* — even if she does not believe it. Here S *acquired* propositional warrant by exercising a cognitive competence — although she didn’t form the corresponding visual belief.

    I don’t think the principle I mentioned *entails* that everyone is p-warranted in every true proposition or even every a priori proposition. (The latter also seems unattractive to me. The idea that a chimp or a toddler is propositionally warranted in a set-theoretical axiom or philosophical principle makes no sense to me.) My point was mainly that the principle is in family w. such views precisely because in that it lacks any account of the psychological basis for the warrant ascription. But of course, someone who embraces the idea of an impersonal warrant shouldn’t be concerned.

    I don’t quite know how to *argue* against the (very) idea of impersonal warrant except to suggest that we should take in the ordinary cases in which propositional warrant is clearly acquired as paradigms. Note, also, that if “impersonal warrants” are admitted, do not have mane of the normative features of ordinary warrants. For example, it is not clear that we can, on the basis of the ascription of an “impersonal warrant,” say that S (ceteris paribus) ought to believe p if she were to wonder whether p — where the ‘ought’ is one of epistemic rationality.

    Anyways, thanks again for articulating the idea of impersonal warrant. I’ll have to think more about how to argue against it rather than simply dismiss it. (A while back a freshman student suggested something like the idea of an impersonal warrant and I told him that he was engaged in a category mistake 🙂

Leave a Reply

Your email address will not be published. Required fields are marked *