Saturday, November 3, 2007

Justin D's Chapter 4 Post:

In chapter four, Williamson argues against those who try hold onto philosophical exceptionalism by defending epistemic analyticity. One of these arguments in particular is focused on rejecting the construal of epistemic analyticity in terms of (UAl’LRule): necessarily, whoever understands the meaning of the logical particles contained in a logical rule assents to that rule. This is supposed to be a case where (the UAl’ version of) epistemic analyticity is on its firmest footing. If Williamson can successfully argue against cases like these, it’s back to the drawing board for defenders of epistemic analyticity. He argues as follows:

1. If there is someone who understands the meaning of the logical particles that occur within a rule but dissents from that rule, then there is a counterexample to (UAl’LRule).
2. If there is a counterexample to (UAl’LRule), then (UAl’LRule) fails.
3. McGee is someone who understands the meaning of logical particles that occur within modus ponens (MP) but dissents from MP.
4. If (3), then there is a counterexample to (UAl’LRule).
5. If (3) and (4), then (6) (UAl’LRule) fails.
6. (UAl’LRule) fails.

I would now like to canvass opinions on the following Weathersonian counterargument to Williamson:

7. “[T]he meaning of a denoting term is the most natural object, property or relation that satisfies most of our usage dispositions” (Weatherson, 2003: 23).
8. If (7), then one can be a competent language user and also be wrong about the meaning of a denoting term, in the case where one's dispositions to use the term fail to coincide with the most natural object, property or relation in the vicinity of said disposition.
9. If (7) and (8), then (10).
10. If X’s disposition to use a term t fails to coincide with the most natural object, property or relation in the vicinity of t, then X is wrong about the meaning of t but can still be a competent language user.
11. McGee’s disposition to use ‘if’ fails to coincide with the most natural object, property or relation in the vicinity of ‘if. (Since, the most natural meaning of ‘if’ is the classical meaning of ‘if’).
12. If (10) and (11), then (13).
13. McGee could still be a competent language user but McGee is wrong about the meaning of ‘if’.
14. If McGee and those with parallel intuitions are wrong about the meaning of ‘if’, then it would be unsurprising that they deny MP and their denial of MP would not count as a counterexample to (UAl’LRule).
15. If (13) and (14), then (16) the McGee counterexample to (UAl’LRule) fails.
16. The McGee counterexample to (UAl’LRule) fails.

Moreover, if a strategy of this sort works in this case, then it will also work against some of Williamson’s other arguments against an epistemological conception of analyticity.
Does anyone buy this argument? Why or why not?
In my opinion Weatherson’s theory of meaning puts too much weight on naturalness, especially given that he takes naturalness to be basic/irreducible. Does his theory of meaning license too much? Can we get clear enough on what it is for something to be more natural than something else in order make use of his theory?

6 comments:

Anonymous said...

I've got a couple of questions:

From what you've said here, it seems like Weatherson has something a bit different in mind from the David Lewis/David Armstrong sort of 'naturalness,' where being a natural property is akin to something like being a property that 'carves the world at its joints' or whatever. On this view, the natural properties are going to be like a small subset of all the properties, and these are going to be the kinds of properties we need to explain basic physical laws, etc. So: being-a-postitively-charged- electron (or whatever) = natural; being-the-first-person-on-the-bus = non-natural.

Does Weatherson have something different in mind?

If he has in mind something like the generic meaning of a term, then can he not just give an account of x being more-natural-than-y just in terms of the former being the property, object or relation that does a better job, on balance, of satisfying most of our usage dispositions?

The premise that strikes me as standing in need of clarification is (8). If someone is consistently wrong about the meaning of a denoting term, then, on Weatherson's view, it seems like that is going to mean that they are not able to identify the most natural object, property or relation satisfying usage dispositions. But if this is the case, how could such an individual also be a 'competent' language user, according to Weatherson?(Isn't competency going to consist of being able to identify the meanings of denoting terms?)

Justin D said...

Hi Adam, good post.
To answer your questions in order:
1. Weatherson does claim to hold that natural properties are 'whatever carves nature at its joints'. So he could certainly give some justification for 'if' being the truth functional 'if' on the grounds that it is the most natural connective that satisfies most of our usage dispositions.
2.As for the 'competency' worry, there are a couple of things to say.
-If McGee is incompetent then he will not count as the counterexample Williamson needs. Williamson argues against taking this route by pointing out that McGee is an expert on conditionals and if he is incompetent, then it looks like almost everyone turns out to be an incompetent language user in this respect. But no matter how this falls, Williamson is counterexampleless.
-I'm not sure what Weatherson would say about the identity between being competent and being able to identify the meanings of denoting terms. But again, if he goes this route, it looks like there are very few competent users of English...
I should also point out that the argument I provided is only in the spirit of Weatherson, NOT Weatherson's. So all errors of reasoning contained within are my own.

Justin D said...

It was pointed out to me that premise 14) is unclear, and I think that this relates to Adam’s post, so I will try to clarify.
I write (14)‘If McGee and those with parallel intuitions are wrong about the meaning of ‘if’, then it would be unsurprising that they deny MP and their denial of MP would not count as a counterexample to (UAl’LRule)’. But this is confusing, for if one is a competent language user, like I admit that McGee could be in premise (13), then doesn’t it follow that McGee understands the relevant meaning of ‘if’? And if so, can’t he still be used as a counterexample by Williamson?
-Ok. Admittedly, if indeed McGee is a competent language user, it is intuitive that McGee understands the relevant meaning of ‘if’. Call this conditional (CU). But even assuming (CU) holds, I still take the Weathersonian argument to show that McGee does not understand the relevant meaning of ‘if’, i.e. the second conjunct of premise (13) still holds. However, if we assume (CU) holds then the first conjunct of premise (13), [‘that McGee could still be a competent language user’], needs to be deleted since by modus tollens on (CU) it is false. Hopefully now it should be clear that even if (CU) holds McGee will not be a counterexample to (UAl’LRule) (given that the rest of the Weathersonian argument is sound). Note: to be a counterexample requires that that there be someone who understands the relevant meaning of ‘if’ and dissents from MP. Given the Weathersonian argument, McGee satisfies the latter condition but not the former.
-On the other hand, if the stated conditional (CU) is false then it is still plausible to assume that the Weathersonian argument shows McGee to be wrong about the meaning of ‘if’. In this case too the needed counterexample fails to obtain.
I hope this clarifies what I was getting at :)

Dan said...

Great post Justin.
I saw a red flag in here though. If Mcgee's use of 'if' fails to refer to the same object/property/relation as the rest of english speakers, then Mcgee has no genuine dissagreement with them. However, I'm inclined to believe that Mcgee's dissagreement is genuine, since he is in fact an expert on conditionals and is making an argument against MP. I think the following quote from Williamson triggered the red flag:
"Peter and Stephen are emphatic that they intend their words to be understood as words of our common language, with their standard English senses... Each of them believes that his semantic theory is correct for English as spoken by others, not just by himself... Giving and incorrect theory of the meaning of a word is not the same as using the word with an idiosyncratic sense" p.18-19
The point applies to Mcgee as well, and I believe the point is this. Mcgee is well aware of the classical meaning of 'if' as well as his own meaning of 'if'. The dispute is over which meaning of 'if' applies to the english word. However, Williamson would deny that anybody concerned does not understand 'if' as it occurs in english, and therefore deny (15).
I think Williamson has looser standards for what it takes to understand an english expression than Weatherson. If a disposition to sometimes stray from the common usage is enough for non-understanding, then the point that very few speakers understand anything will arise once again. You preserve UAl'L rule at the cost of understanding english. As Williamson said, failure to fully know the meaning of a word is different from using the word idiosynchratically.

Justin D said...

Thanks for the comment Dan.
I think that Weatherson would push back by saying "sure, if my view is right, then very few people understand the meaning of most english words. However, this isn't such a big deal since most people can still be competent users of the language on my view. Their use can still overlap correct use most of the time." He would take his theory to be a novel explanation of how people can use language while still being wrong about the meanings of most words.

Dan said...

Weatherson may very well take that retreat, but how would this new notion of understanding fit with UAl'?

UAl': necessarily, whoever understands the meaning of the logical particles contained in a logical rule assents to that rule

It seems like it would work well for stipulative definitional analytic statements, but not for natural language analyticity. After all, by Weathersonian understanding, most people won't understand "all vixens are female foxes".