Devitt on the A Priori

So I have been reading Devitt’s paper “No Place for the A Priori” where he lays out his case against a prori knowledge. His claim is that there is only one way of knowing, and that is the empirical way. His strategy in the paper is to first argue that the kinds of knowledge that people usually claim to be examples of the a priori (i.e. math, logic, and philosophy) can all be explained on the emprical model. This means that there are two options for how we come to have knowledge of (say) the logical truths and so we would then need to see if each option is equally viable. This lead him to the second part of his strategy, which is to argue that the notion of a priori knowledge is so mysterious and obscure that the relatively well worked out empirical model is to be preferred. He even goes so far as to suggest that if he is right, “it is not rational to believe in the a priori.” Oh irony of ironies!

What then is the empirical model? He says,

An answer starts from the metaphysical assumption that the worldly fact that p would make the belief that p true. The empirical justification of the belief is then to be found in its relationship to experiences that the worldly fact would cause. Justified beliefs are produced and/or sustained by experiences in a way that is appropriately sensitive to the way the world is. This is very brief and we shall return to the question later. Still it is hard to say much more.

So the empirical model looks like it boils down to the correspondance theory of truth. There is though a lot more to say. For instance ‘worldly fact that p’ is ambiguous as between (in Armstrong’s sense) a world of states of affairs (or, in Russell’s older terms ‘facts’), and a world of things. But we can leave that aside. His main claim is that even this sketchy characterization of the empirical model is more worked out than any account of a priori knowledge and can also account for our knowledge that seems to us a priori (math, logic, etc).  

Now I am generally sympathetic to this claim, being a naturalist myself, but the way that he puts it certainly seems acceptable to some ‘a-prioraphiles’. Might not it be the case that the ‘worldly fact’ that modus ponens is a valid logical form play the kind of role that p does in the characterization above? Rational intuition is often characterized as a kind of ‘grasping’ with the mind. It is an intellectual kind of experience where one sees or appreciates some necessary fact about reality. Thus my beliefs about modus ponens would be ‘appropriately sensitive to the way the world is’. Of course, one couldn’t be a physicalist (like I am), again in Armstrong’s sense, but one could still be a naturalist…Or is that the a-priorafile can’t allow that modus ponens is a non-physical natural phenomena (i.e. a worldly phenomena that exists in space and time)?

17 thoughts on “Devitt on the A Priori

  1. His strategy in the paper is to first argue that the kinds of knowledge that people usually claim to be examples of the a priori (i.e. math, logic, and philosophy) can all be explained on the emprical model.

    I think any plausible empirical explanation of why we think that math and logic are obvious will have to use at least some logic and math (evolutionary theory without math?, physics without math?, argument without logic?)?
    So, the explanation will depend on the acceptance of logic and math, hence will have to be circular.

  2. Hi Tanasije,

    He address this concern in the paper. Here is what he says

    We must start by distinguishing “premise-circularity” from “rule-circularity.” An argument is premise-circular if it aims to establish a conclusion that is assumed as a premise in that very argument. Premise-circularity is obviously reprehensible. But my argument for [the rules of logic being empircally justified and so that they are good rules (he calls this TS)] is not guilty of it because it does not use TS as a premise. An argument is rule-circular if it aims to establish a conclusion that asserts that the rules used in that very argument are justified. My argument tries to establish TS which asserts the justification of S, the system used in that argument to establish TS. So the argument is certainly rule-circular. Guided by the Neurath image, the argument accepts the non-epistemological part of our web for the moment and seeks to justify the epistemological part, TS. And that justification is governed by just the same rules that govern the justification of anything, the rules of S.

    He then goes on to argue that even the a-prioraphile is committed to a kind of rule-circularity and so each party is in the same boat. Which he takes to support his side because of the second part of his strategy which aims to show that the whole idea of a priori knowledge is so mysterious and obscure that the empirical way wins in abduction…

  3. Hi Richard,

    I’m not so concerned about explicit premises, but about the possibility of an argument without any rules of inference (like modus ponens) implicitly accepted.
    Also, when I mention math, I wasn’t thinking about really accepting the truth of mathematical theorems, but implicitly accepting, when we are talking about e.g. evolution, that when a bird lays 4 eggs, it is one more egg than 3 eggs.

    Of course it might be just the limit of my reasoning, that I can’t imagine a detailed explanation of our thinking that we know logical and math claims a priori, without accepting at least this kind of minimal math and logic. I see the paper is available online, so I guess I should go and read the paper 🙂

  4. Yeah, neither is he. The point is that the fact that the arguemtn relies on a rule in the set of rules the argument is trying to justify is going to happen no matter which route you take so it is not a point against his claim that his is.

    As for the stuff about implicitly accepting math, I think that that is right, but his point is about the justification of teh implicit acceptance. His claim is that the implicit belief is justified by our experience. The explanation starts with the way that the world behaves and then explains our belief in terms of that. But, they hold, it is possible that the world doen’t really behave that way. This means that we may someday have enough evidence to believe that somekind of non-classical logic is correct (say one that rehects the law of non-contradiction, as some have argued that quantuum mechanics suggests)…this is taken care of on his veiw because he requires that the rules for revision are part of the belief system itself…

    but yeah, if you read the paper get back to me and let me know what you think….

  5. Hi Richard,

    I read the paper, and it doesn’t seem to me that he at addresses satisfactory the circularity issue. Seems more his response is “apriorists have same problem too, so who cares” strategy.

    1. I agree with Devitt on his attack on given justifications of the a priori knowledge. But, I agree because no justification is needed besides the understanding of the proposition itself for a priori knowledge. Or, I think the rationalist needs just to say – the reason why it is true IS the reason why I think it is true.

    2. Connected to the previous point, the apriorist doesn’t need to claim that instances of modus ponens are justified by the rule that modus ponens is valid inference. On contrary I think that a person can understand an instance of modus ponens without every having heard about modus ponens, or even having thought about some general form to which that claim belongs. (People did think ‘logically’ even before Aristotle gave syllogisms). So again, there is no need to go from an instance of a priori knowledge to some kind of rule, or further justification.

    So, to me it seems that Devitt is not in the same boat as the apriorists re justification. The apriorists can just deny the need for any further justification, while Devitt can’t. He needs to address it, but he doesn’t in the paper, especially doesn’t give answers to the points raised by Boghossian cited in the paper.

    BTW, why does Devitt worry about circularity? Wouldn’t the rule that arguments shouldn’t be circular (be it in premise-circular or rule-circular way) be just part of the rules which are open to change?

  6. yes, but that does address the issue…each side ultimately has to resort something like rule-circularity so it can’t be a problem for Devitt’s view that his does.

    Re 1 & 2: but that’s just the point. When you say ‘the reason why it is true is the reason why I think it is true’ you haven’t said what the reason is. If you say ‘I know a priori that it is’ then you are applying the rule ‘if it seems to me that it is a priori then conclude that it is’….rule-circularity….

    I agree that people encounter and use modus ponens without knowing it, but that not the point. The point is what do you say to the person who thinks that modus ponens is not a good logical rule? The only thing you could do is to point out how the world is and how this seems to be the way the world works…that is you ise the empiral model.

    The idea that the denial of the need for jusitification is a legitimate move is very scary to me…doesn’t it licence anyone to believe anything they wany? I mean, doesn’t the Nazi say with just as much conviction that it is self-evident that the Jews are evil? They just understand that the proposition is true, can’t you?

    Re the last thing, yes I suppose that rule, like all others, could be revised, but it would take A LOT of evidence to make us even begin to think of doing so and so far we don’t have any reason to think that we should.

  7. Richard,

    Right, I’m not stating the justification, and in fact denying that in the case of a priori knowledge we need further justification. So, I’m not justifying it with a rule ‘if it seems to me that it is a priori then conclude that it is’. In fact I would even think that something might seem a priori true, but that in fact it isn’t. There is at least one case in math for some theorem to be held true (so to seem as a priori knowledge) for a long time, only to be proven wrong (for a priori reasons, showing a mistake in the previous reasoning). So that something seems a priori, is not justification for it’s acceptance.

    As for the Nazi’s, I believe that in that case there is not much sense in discussing of justification. I’m sure with the hate present in the society, it would be easy to “find” empirical support for claims about Jews. Probably lot of “scientists” did “find” empirical support about superiority of German race. And that they could pick the ‘a priori’ arguments, at least select those which go nicely with their cause. But, I don’t think that we can blame neither empiricism nor a priori thinking for this kind of misuse.

    Re the non-circularity rule, yes, I understanding that it would be one of the most-central rules of the web, but it does seem to be consistent with Devitt’s view that in principle all rules can be changed. And as Devitt is only arguing about “much weaker claim that we now have no good reason to think that such a justification could not be given”, it seems that he can use that in his argument. (Of course, it doesn’t make sense to me , but seems it would be consistent 🙂 )

  8. A priori knowledge as you describe is so insane that I really don’t know what to make of it…I gues I just don’t know what you mean when you say that we don’t need any justificaion for a priori knowledge and yet some stuff we think a priori might turn out not to be…this just sounds like a plain contradiction to me…it sounds like what you are saying is “I have a belief that is just true but I don’t know why its true, and in fact it might be false”…this can’t be what you mean, can it?

    as for the empirical stuff, I just don’t agree. I don’t dispute that the nazis did in fact make up empirical sounding justification but these claims do not hold up to testing. Even if the claims did suport the claim that the nazis were superior (I don’t for a second believe this, but let’s say…) that still wouldn’t licences the kind of behavior exhibited by the nazis..whereas the a priori belief that Jews are evil has no standards to meet…it just is true…how could you ever prove that this belief was wrong?

    I suppose he is in part arguing that way when he says that rule circularity (properly circimscribed) should be accepted…but as for premise circularity, there is absolutrly no reason to to revise that.

  9. Yeah, I think I wasn’t clear in my claims…

    As I think that knowing p a priori, is in fact understanding that p, my claims are, that:
    a. one can understand that p
    b. if one understands that p, he doesn’t need further justification that p
    c. one can think that he understands that p, but be wrong. however that doesn’t negate possibility of (a)

    As for justifying the claim in relation to others, I think it is a separate issue. I may understand how Pythagorean theorem is true in Euclidean space, and to convince somebody else I can give proof of it, ‘unpacking’ it to the claims that that other person understands to be true.
    And for sure, the claim that one understands that p, in those complex cases is not to be taken uncritically, but (a priori) proof should be asked for.

    The case with Nazi’s claims about Jews would fall under this case. As good empiricism will require good testing, good rationalism would require good explanation , so that I as a rationalist will require to understand (v.s.blindly accept) that p.

    Simple a priori claims like ‘when there are two things there is one and one more thing’ or ‘if a is longer than b,then b is shorter than a’ or ‘if every color is property of a thing, than red is property of a thing’ can be understood without further explanation. So, in terms of justification of their truth, I don’t see need to provide one, as they are so simple that can be understood as true as they are.

  10. yeah but you still haven’t said anything about what it could possibly mean to ‘understand’ something.

    Do you mean that the sentences are analytic and that is supposed to explain our understanding? Or something else all together?

  11. in devitts there is no a priori essay, his obscurity argument(as i understand it) is that since empiricism explains knowledge in the simplest way, then, using abduction, it is the best explanation for knowledge.
    im writing a paper on this and i just dont get how he can use abduction to disprove a priori knowledge. i just dont see how because something is obscure it doesnt exist. but i am having a hard time trying to argue against abduction.
    any ideas would help.

  12. Hi Cavan,

    Sorry about the delay in getting back to you!!

    I don’t think that his argument is that since its obscure it doesn’t exist. I think that what he means is that we have no reason to believe that it exists. No one has ever given us any idea what it would mean for there to be a prioiri knowledge (besides demonstrating some alleged examples of it). Also, we havea perfectly intelligble account of knowing (the empirical way) that does offer us an account of how knowledge is possible and it also looks like it could be used to explain the alleged examples of a priori truth. So, his overall argument (as I see it) is that given these two things we should reject a priori knowledge. This does not mean that it doesn’t exist, it is just that it looks unlikely (no one is able to really say what it is) and uneeded (we have a perfectly good account iun the form of the empirical account)

    Hope that helps!

Leave a Reply to cavan Erskine-smith Cancel reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s