Hey Internalists, Which Experiences Justify and Why?

Here is my impression: it is very popular to allow certain kinds of experiences to provide (prima facie propositional) justification for certain propositions.  Which propositions might an experience justify?  The most straightforward thing to say is that certain experiences provide justification for their propositional contents; but if experiences don’t have propositional contents, some more elaborate story will need to be told.  For the sake of simplicity, I’ll assume that an experience justifies a proposition only if it has a propositional content.

Which experiences provide justification for their propositional contents?  Here are some popular answers: perceptual experiences, memorial experiences, a priori intuitions, or some combination thereof.  (I am thinking of a priori intuitions as experiences.  Intuitions about morality, math, modality, and philosophy often count as a priori intuitions.)  It is not popular, however, to say that just any experience (or just any experience with a representational content) can justify its conclusion.  But this raises the following questions: if some experiences justify their contents, why don’t all experiences justify their contents?  What principled criterion is there for allowing only certain experiences (with contents) to justify their conclusions?

Reliabilists have a straightforward answer: only experiences that are caused in a reliable way can justify.  Other externalists will have similar stories to tell.  But what about the internalist?  What criterion can he use to distinguish between experiences that justify and those that don’t?

One answer is available to those who allow a priori intuitions to justify.  They can say something like this: an experience that P justifies P just in case it is the result of possessing, or understanding, the concepts involved in P.  The intuition that torture is wrong justifies because it results from possessing the concepts ‘torture’, ‘is’, ‘wrong’.  The experience that the Cubs will win the World Series does not justify because it does not result possessing the concepts involved.

I’ve never found this explanation very appealing for two reasons.  The first is essentially this argument: (i) perceptual experiences and a priori experiences justify; (ii) concept possession does not explain why perceptual experiences justify (at least not all of them); (iii) there is one explanation for why experiences justify (so if concept possession doesn’t explain why perceptual experiences justify, then it doesn’t explain why a priori intuitions justify); (iv) therefore, concept possession doesn’t explain why a priori intuitions justify.  The second reason is that if you are going to give concept possession such a prominent role in a priori justification, a priori intuition doesn’t seem to do any work—it just seems along for the ride.

My own view on the matter is phenomenal conservatism: necessarily, if it seems to S that P, then S has (prima facie propositional) justification that P.  A seeming that P is an experience with propositional content P and a special kind of phenomenal character.  The phenomenal character “assures” the subject that its content is true.  Perceptual experiences, memorial experiences, and a priori intuitions are all plausible candidates for seemings.  Phenomenal conservatism is unpopular because many find it implausible that all seemings provide justification.  Surely, some argue, a seeming produced by wishful thinking has no justificatory power.

Here is another way of pressing the question in which I’m interested: if some seemings justify, why don’t all seemings justify?  It seems that many internalists allow perceptual and a priori seemings to justify, but they don’t allow wishfully-produced seemings to justify.  What principled, internalist criterion can allow only some (and the right ones) to justify?


Hey Internalists, Which Experiences Justify and Why? — 17 Comments

  1. Hey Chris,

    I think it’s sort of strange to think of memorial justification and apriori justification as involving apriori experience or memory experience, but I guess someone could go that way. I was interested in this argument:

    (i) perceptual experiences and a priori experiences justify;
    (ii) concept possession does not explain why perceptual experiences justify (at least not all of them);
    (iii) there is one explanation for why experiences justify (so if concept possession doesn’t explain why perceptual experiences justify, then it doesn’t explain why a priori intuitions justify);
    (iv) therefore, concept possession doesn’t explain why a priori intuitions justify.

    (iii) seems to assume that there couldn’t be a kind of pluralism in epistemology that seems attractive in ethics. It seems that the grounds of right action are diverse and any attempt to specify some single feature by virtue of which all right acts count as right will fail. Why should we assume monism in epistemology? Perhaps someone who was convinced of (iii) would want to try to build an account of memorial and apriori justification on, say, the model of perceptual justification [like the phenomenal conservative view on which justified belief in p involves the proposition/thought that p and a shiny mental thing] but perhaps the reason that this talk of apriori experience and memorial experience seems strange is that I’m not at all certain why I think we should assume (iii).

  2. Hey Clayton,

    I wasn’t assuming a kind of monism in epistemology. I was assuming only a kind of monism regarding the sort of justification provided by experience. The assumption is motivated by simplicity, but of course that sort of concern can be outweighed by other considerations.

    I actually do endorse a limited plurality for justification–some beliefs are appropriate because of experience and some beliefs are appropriate because of other beliefs. If testimonial justification were a kind of non-inferential justification that wasn’t experiential, then (iii) would be compatible with there being a different explanation for why my testimonial beliefs are justified. What seems a little strange to me is saying two experiences, E1 and E2, both provide justification for their contents, but they provide justification for different reasons.

    You say: “Perhaps someone who was convinced of (iii) would want to try to build an account of memorial and apriori justification on, say, the model of perceptual justification…” The people who think that a priori intuitions provide justification don’t normally take themselves to be modeling a priori justification on perceptual justification (so far as I know). In fact, some (e.g. Bonjour) allow a priori intuition to justify but not ordinary perceptual experiences.

  3. It seems to me that there is no problem with having all seemings play a justificatory role. However, a seeming can only play a justifictory role just in case the propositional content of the belief fits the phenomenal character of the experience.

    Here are two exmples to demonstrate my point:

    (a) Beliefs that have propositional content about my current perceptual experience are justified by seemings that originate from my current perceptual experience. For example, the belief: ‘I am experiencing a red square’ is justified by the presence of the red square in my perceptual experience and my awareness that the propositional content of my belief fits the character of my perceptual experience.

    (b) Beliefs that have popositional content about what I am wishfully thinking are justified by my experience of wishful thinking. For example, the belief: ‘I wish my grade was an A instead of a C’ is justified by my having the experience of wishully thinking A instead of C.

    Thus, the internal criterion that makes some seemings justfiable and others not is the awareness on the part of the believer that the propositonal content of the belief fits the propositional content of the experience*.

    This takes care of wishful thinking falsely justifying beliefs that are not about wishful thinking.

    * I used your terminology that an experience has propositional content, however, I have a problem with perceptual experience having propositional content. Rather, I think that perceptual experience is a concious mental state and thus perceptual experience is made up of sense data and an awareness about the sense data that is built into the concious mental state. Justfication of a perceptual belief, then, comes from the comparison of fit between the the propositional content of the belief and the “built-in” awareness of the features of the sense data. (Most of this comes from Laurance Bonjour)

    (I tried, but I may misunderstand what you mean by “seemings”, “wishful thinking”, or some other key idea that you have presented and am therefore way off base)

  4. Hey John,

    If I understand your picture, it is something like this: S has experiential justification for P when S has an experience that is the truthmaker for P and it seems to S that P fits the experience. This view provides the sort of criterion I was looking for. It says an experience justifies P only if the experience is the truthmaker for P.

    I have two concerns about this picture, but I’m pretty sure that Bojour doesn’t share them. First, I think a perceptual experience that there is a tree in front of one justifies a belief that a tree is in front of one. But your view doesn’t explain why this is so.

    Second, suppose, as Williamson and Fumerton seem to allow, that one can be mistaken about whether one is in pain. For example, one might mistake a mere itch for a mild pain. Now suppose S has an itch instead of a mild pain, but it seems utterly obvious to S that he has a mild pain. I say S is justified in (falsely) believing that he has a mild pain, but your view doesn’t allow that.

    Also, I think I need to clarify the sense in which wishful thinking is a problem for phenomenal conservatism. Suppose it seems to S that the object is a gold nugget. If the seeming is caused by S’s wanting the object to be gold, then Markie and others say that the seeming can’t justify S’s belief that the object is gold.

  5. “I was assuming only a kind of monism regarding the sort of justification provided by experience.”

    That was the sort of monism I was thinking of. Looking at your argument again, it might be that anyone who accepts (i) will be somewhat sympathetic to (iii) given simplicity considerations. I think part of my skepticism concerning (iii) is really skepticism concerning (i). I don’t know what I’d be committed to by saying that apriori experiences justify beliefs that are apriori justified.

    Let me ask a slightly different question about the apriori. Suppose someone held the following view: when S’s intuitions aren’t the result of S’s understanding, S’s intuitions don’t justify; when S’s intuitions are the result of S’s understanding or the grasp of the truth, the beliefs that are intuitively correct to S are justified.

    You might say that on this model, the intuitions/experiences don’t do any justificatory work but are only along for the ride. Possibly, but the nice thing is that S can find P intuitive only because S is confused. It’s not clear that this should involve any sort of justification for S to believe p. It might be that S believes p unreflectively and having never given the matter careful consideration, it has never been that P seems intuitive to S. Still, S knows p apriori because S understands that p is true/S’s grasp is sufficient for S to know p.

    What’s wrong with that sort of view? It seems that whatever is happening when I have an intuition, it’s separable from when I grasp a truth/see something is true based on the understanding alone and when that intuition is the result of reflecting on some matter while being conceptually confused, it adds nothing positive. The idea that the ‘apriori experience’ doesn’t justify on its own regardless of whether someone suffers from a confusion should be a feature, not a bug. It’s kind of like when someone finds it intuitive that some conclusion is supported by premises that don’t support it. In determining whether and the degree to which the conclusion is justified, I’d look at the supporting premises and not whether the bad reasoner also just thinks they see that the conclusion has been established. That doesn’t count for anything. (Just imagine a logic class where students were given extra points because unlike the other students who committed fallacies, they found the conclusions of their fallacious reasoning really intuitive and were committed conservatives!)

  6. Hey Clayton,

    If you had in mind the experiential-justification-monism, then I don’t understand the disanalogy you were trying to draw. Here’s how I was understanding pluralism in ethics. There is no single feature that explains why all right acts are right. Nonetheless, many actions are made right by the same feature. E.g. acts 1-40 were right because they shared the feature ‘keeping a promise’, acts 41-60 were right because they shared the feature ‘helping someone in need’, etc. If this is correct, then my view in epistemology is analogous: beliefs 1-40 are right because they are based on appropriate experiences and beliefs 41-60 are right because they are based on good arguments. Maybe the epistemic kinds of rightness are smaller in number, but that seems right to me: moral properties seem sensitive to a greater variety of goods than epistemic properties.

    On the view you describe in the third paragraph (which is the sort of view I tried to describe in my original post), I agree that it would be better to say that a priori intuitions/experiences don’t do any justificatory work. That was the sort of point I was trying to make with the “second reason” I didn’t find the view appealing.

    Suppose we consider the understanding based view without trying to give a priori intuitions work to do. The internalist in me thinks that if you have understanding without an intuition (or some sort of awareness), you are in a Bonjour-styled Norman case and so don’t have justification. Also, on any understanding based view, I worry that it is impossible to have (non-inferential) a priori justification for a necessary falsehood. To me, that is a drawback of the view (but based on what you say, maybe you don’t agree).

    The phenomenal conservative can get the right result in the logic class case. Logic teachers grade for correctness and perhaps understanding. People shouldn’t get credit for having justified, but mistaken beliefs.

    I think your analogy with higher-level seemings (e.g. a seeming that X supports P) is a bit distracting. Phenomenal conservatism says, basically, that all seemings constitute good evidence. Suppose X is not good evidence for P. Phenomenal conservatism does NOT say X can be turned into good evidence for P if it seems to the subject that X supports P. (But I do think higher-level seemings and beliefs that E supports P have an important role to play in acquiring inferential justification for P. I’ll leave that story for another day, though.)

  7. The logic class example was intended to be a quasi-humorous poke at the view. I don’t really have a view on the apriori, but my immediate reaction to the suggestion that apriori justification works in the way that perceptual justification is thought to work on the phenomenal conservative view of that kind of justification is not positive.

    Suppose we have three thinkers. The first good apriori justification for the premises and reasons correctly to conclusions without thinking about doing that and without any intuitive sense that the conclusion is true. We might say that this thinker knows how to draw the appropriate conclusions from the premises. The second is just like the first but is prompted to reflect on the matter and finds it intuitive to say that the conclusions she draws from her premises non-fallaciously are properly drawn. Our third is like the second, but all the reasoning is fallacious and she reasons that way because she is badly confused. So, it intuitively seems her her conclusions are correct to her and her premises are all justified (at least, according to phenomenal conservatism).

    I’ll grant that our second thinker might be in the best shape, but I think the first is better off than the third. I don’t think it’s seeming to you that such and such a conclusion follows from such and such a premise when it seems that way because you are confused about which inferences are appropriate provides any justification for the conclusion. I don’t see how the phenomenal conservative avoids that. One view is to limit justification to our second subject (we need the phenomenal whatever plus understanding or the absence of confusion). Another is to say that our first two subjects are justified. I think the view we ought to avoid is on which the 2nd and 3rd or just the 3rd is justified in her conclusions, but I don’t see how the phenomenal conservative can do that.

    I’m not imagining a view where 3rd subject treats E as evidence for the conclusion, C, and E is turned into evidence by the phenomenal state. I’m imagining a view on which the phenomenal state is the very thing that does what E can’t do, justify belief in the conclusion. That view seems really, really weird to me. So, is it a weird view I should accept or is it a view the phenomenal conservative can wiggle out of? If it seems to our 3rd subject that her C follows from established truths, on PC, isn’t that seeming state “evidence” for C?

    [Full disclosure–I don’t have a view on the apriori, I’m just trying to figure out how the PC among us deal with cases and I’m not afraid of BonJour’s Norman.]

  8. Chris,

    Im thinking that the more interesting aspects of a philosophy of Concepts is in its relationship to necessity, not intuition. At some point one needs to ask the question what it means to grasp a concept? This begins the road to a robust internalism.

    Assuming for a moment that Beliefs are Propositional Attitudes (whatever that means), it would simply be odd to have the belief (1) that all unmarried men are flif, if you dont have a grasp on the concept of flif. It would be even more odd to suggest that (1) was causally formed in you, regardless of whether or not they were formed through reliable functioning cognitive faculties.

    The content of the proposition (i.e. the concepts in play) are not vacuous in regards to the justification of the belief, rather they are critical to it.

  9. Hey Adrian,

    I’m not sure I understand your line of reasoning. But I worry that you are conflating these two questions:

    1. Is possessing (or some particular degree of possessing) a concept C required to believe a proposition involving C?
    2. Does possessing the concepts involved in P (or possessing them to some sufficiently high degree) constitute justification for P?

    Presumably, the answer to 1 is yes, and I take it that your second paragraph tries to make that point. But the discussion between Clayton and I is over the second question, which so far as I can tell you haven’t really addressed.

    Am I misunderstanding you?

  10. Chris,

    Great question. It is not a conflation. I merely suggest, in response to your original post, that a robust internalism begins with a robust notion of concepts.

    I think the answer to (1) is Yes as you point out. And I suppose I’m prepared to defend a Yes to (2) as well. Let me qualify that with this: grasp of a concept means knowing (loosely) both the conditions and the consequences of application. In that regard, concepts are counterfactually robust.

  11. Hey Clayton,

    Let’s divide the third thinker into, 3a and 3b. 3a mistakenly thinks X supports C. This mistaken belief causes C to seem true to 3a. PC entails that the seeming that C provides S with (prima facie propositional) justification for C. (If 3a bases her belief solely on X and not the seeming, then she wouldn’t have doxastic justification). Provided that seemings really do provide justification, that is the correct result. In principle, it is possible for some possible subject to arrive at a correct understanding of the concepts involved in C by mistakenly thinking that X supports C. So understanding-based views still have to say that, in certain cases, one gains justification for C via misunderstanding the connection between X and C.

    3b is different. It seems utterly obvious to him that X supports C, but C does not seem true to him (but nor does it necessarily seem false either). PC entails that S has justification that X supports C, but it doesn’t follow that X has justification for C (unless you are working with a different notion of support than I am). Suppose it also seems to 3b that X justifies him in believing C. And suppose that justification that one has justification for P is itself justification for P (I think a more refined version of this principle is probably true). Although X doesn’t justify 3b’s belief in C, given these additional suppositions and PC, its seeming that X justifies C does provide S with justification for C.

    Given how we have spelled out 3b, PC entails that 3b has justification for C. Is this a problem for PC? I doubt it. Imagine that S justifiably (but falsely) believes that X justifies C. He has this belief on the basis of expert testimony. Even though X doesn’t justify C, the justified higher-level belief that X justifies C still plausibly justifies C.

    If your examples identify a problem with PC, it is probably because we have intuitions that seemings can’t justify when they are caused in inappropriate ways. If so, I think the Markie-styled wishful thinking examples provide a more straightforward way of pressing the point.

    PS- I’m inclined to think that S bases her belief in Q on her belief in P only if S takes P to support Q. Given what you say about the first thinker, then, I’m not sure the first thinker has reasoned, i.e. based a conclusion on a premise.

  12. Hey Adrian,

    You say: “grasp of a concept means knowing (loosely) both the conditions and the consequences of application.” The knowledge at issue is to be understood in terms of knowledge that or knowledge how. If it is to be understood in terms of the former, then possessing a concept requires a great deal of sophistication, sophistication that children probably lack. If it involves the latter, the view you describe isn’t, as you seem to think, obviously internalist.

  13. Thanks, Chris, that’s helpful. I think the nub of the worry is essentially the same as the Markie worry. (I don’t think it’s a bad thing, Markie’s points often seem right to me, so you must think they sort of justify me a little…)

    Here’s a version of the example inspired by one of Dancy’s examples. Having read some recent articles on the issue, Sam realizes that most injuries in train accidents take place in the last car of the train and so Sam finds it intuitive to think that the way to reduce injuries is just to remove the last car from every train.

    No justification there. Sam’s just confused. But, Sam believes what he does because it seems to him intuitively that it’s correct and because of a bunch of other stuff he knows having read the paper.

  14. Hey Clayton,

    Although I am a proponent of PC and have a response to the Markie-styled objections, I am a little worried that it is too permissive. Of course, I think every view has problems or counterintuitive implications, so we just have to pick the view that seems to have best combination of virtues and vices. To me, PC seems the best so far. But perhaps I’m still young and foolish 🙂

  15. Chris,

    I do not think it is that simple. If one acknowledges, as I think we should, that one can not or should not have a belief without understanding the concepts in play, then one is already committed to an advance cognitive dimension. I think most psychologist in early childhood development would say that children develop the use of concepts at an early age. Never mind the fact that most psychologist lack a robust notion of concepts.

    Grasping the conditions of application in terms of being able to respond differentially to environmental stimuli in the way that parrots and thermostats can, is insufficient for grasping a concept. One also has to understand what one is committed to by deploying x (i.e. the consequences). This is the sense in which concepts are counterfactually robust.

    I have a 5 year old and a 3 year old and it seems obvious to me that there was a transition from responding differentially to external stimuli to becoming conceptually aware.

  16. Chris, thanks for your reply and the original post. I realized shortly after making my post that I focused on one portion of your original post and inadvertantly forgot the bigger picture of what you were getting at (I even blantently ignored some presuppositions that you listed!).

    So, thanks for making me aware of phenomenal conservatism and giving a thoughful reply to my post.

    (Sorry it took longer than normal to write this. I wanted to wait until the post calmed down so I didn’t detract from the conversation)

Leave a Reply

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