What you are thinking about when you get a priori knowledge

            In the previous parts of this section I have tried to describe one way a person could acquire the kind of basic modal knowledge which is needed to power, conceivability arguments, truth table arguments and model theoretic arguments for the possibility or impossibility of various sentences being true. I imagined a person engaged in a conscious scientific process of attempting to describe the rules for their own language use, and the general laws of the world in way which fits and makes sense of the actual judgments (‘there is a white raven’, ‘the cup is on the table’) which they have made in the past. From this empirical and contingent knowledge about the workings of their own language they can then proceed to infer nessicary truths by the following kind of inference.

 

(knowledge of contingent fact) The workings of my language are such that ‘S’ will always express a truth.

(knowledge of necessary truth) S.

 

I have already discussed one kind of objection to this account of a priori knowledge (which was suggested to be by Koellner), arising from the normal phenomenology of learning basic facts about necessity and validity, in the part on ‘grasping in a flash’. Eli Chudnoff suggested another interesting kind of objection to me which arises from facts about content attribution, that is from facts about what we would say a person is (and more importantly isn’t) thinking about when they learn a necessary truth.

The objection (or, at least the version of it I want to consider here) goes like this. What normally happens when a person realizes that everything is self-identical/ this cup is self-identical or the like is just that they think about identity and perhaps the cup, and then see that the cup couldn’t fail to be self-identical. Sometimes examples may go through their head, but other times they will just do something like say the sentence to themselves and then become convinced. Normally a person doesn’t think about anything linguisitic or psychological like their ability to recognize the truth or falsity of sentences or the norms for their applications of terms. And almost no one thinks about the kind of externalist constraints which relate facts about psychological tendancies to use certain words with the norms for using them and their contributions to sentences’ truth conditions. All in all, it would be very implausible to say that the way we generally get a priori knowledge of necessity is by means of scientific inductive reflections on our actual judgments.

This is quite right. What I have aimed to give in the first part of this section is an account of how someone could use conscious occurant thought to get to knowledge of necessary truths. Most basic modal knowledge isn’t gathered this way but rather through the shorter process just described. You just sort of say the sentence or inference schema to yourself and become convinced. This raises a substantive question about how that kind of process can give people knowledge – one which I think the earlier story of how a conscious ‘over-thinker’ could gain the same kind of knowledge helps us answer.

I want to emphasize first that the challenge in question is not a matter of skeptical doubt but rather a place were naturalized epistemology is needed. What seems missing and difficult to supply is a story about what nessicary truths are and what people are doing/what is going on in them when they have this experience of sudden conviction that something is a nessicary truths, such that this experience of becoming convinced really gives them modal knowledge. While, inferentially speaking talk about ‘just seeing that X’ may be the end of the story, from the point of view of naturalized epistemology it is just the beginning.

Take for example, the fact that people can just see that a pile of books is unstable or that a dish of vegetables wont taste good (e.g. if they are overcooked) but they normally can’t just see that the pile of books was last handled by James or that the vegetables are contaminated with harmfull bacteria. Surely it is not skeptical to say that this difference needs, and can be given, a good scientific-philosophical explanation. Being neither a scientist of vision, nor a philosopher of perception I won’t aim to give this explanation in detail. But I suspect that everyone will agree that facts about how the relevant properties supervene on physical states of things you can look at (as the property of tasting overcooked does but that of having last been handled by James doesn’t, at least if James sometimes wears gloves) together with facts about something like “information loss” in the course of our perceptual interactions with various objects (e.g. looking at the same food with and without bacteria may give rise to the same changes at the level of neurons…there seems to be some finest grain of information which can make a) a difference to our behavior or b) a difference to our conscious experience). This is the kind of explanation that, I think, we would also like to have in the case of basic modal/logical knowledge.

Very crudely, when people look at things which look similar to the vegetables they found to be overcooked before they are inclined to believe that these vegetables will be bad tasting too. Similarly when they look at vegetables that look like the ones they have found to be bacteria infested in the past they will likely have an inclination to believe that these will be similarly dangerous. But only the former belief amounts to their seeing that the veg won’t taste good because, the perceptual process leading to belief formation transmits information about tasting overcooked, but looses information about bacterial infestation.

In the case of basic modal beliefs what I want to say is this. An experience of sudden conviction ‘just seeing that P’ can be inductively grounded in past experience and experiment so that one only or mostly becomes convinced of things which could be scientifically/inductively justified with reference to these experiences. As I discussed in the preceding part, the experience of ‘just seeing’ that the veg will taste overcooked is grounded in the fact that past vegitables which have looked similar have tasted over cooked. The fact that you don’t just randomly become convinced that there are vegetables on the table but have this experience of sudden conviction mostly or only in cases where this claim could be justified by induction over your past experiences is extremely important to the fact that your feeling of conviction counts as knowledge. Similarly, if your feeling of sudden conviction that a certain sentence is nessicarally true occurs not at random but pretty reliably in the same circumstances in which one could give the kind of scientific justification which I have described, this suffices for the conviction you acquire to count as knowledge.

Thus, even though the process of acquiring basic modal knowledge may be very quick and may not include any thoughts about one’s language or one’s past judgments the right naturalized-epistemology story about how our modal convictions amount to knowledge may run through the kind of scientific understand of our norms of language which I have described.