Sunday, January 28, 2007

MM Hawthorne: doubts, doubts, doubts...

Hi all,

I have to confess that I have had a lot of difficulties to understand the text. Now, I have a lot of doubts. I expound them here. I guess I have not enough background to follow the discussion; my fault, sorry.


At a certain point Hawthorne asks: wouldn’t it be more charitable to interpret allegedly Bold Gabriel as Timid Gabriel? And he answers: Gabriel’s commitment to Ref. puts considerable pressure on us to interpret him as Bold.

Ref. Sentences of the form ‘That is F’ as uttered by Michael, are true only if Michael refers to something by ‘that’.

My doubt: If we accept that one can elaborate a semantic theory about a foreign language (in the relevant sense of foreign language that is in play here), what are the restrictions in our theorizing? For example, why cannot I use something like Ref* instead of Ref:

Ref*. Sentences of the form ‘That is F’ as uttered by Michael are true only if Michael refers to something-on-Michaels-domain-of-objects by ‘that’.

This way here there would be no reason to interpret allegedly Bold Gabriel as Bold Gabriel and not as Timid Gabriel.


Hawthorne describes his Convention Lover as saying that: when thoughts were conceived of hyperintensionally, neither Gabriel nor Michael could express the thoughts of the other on account of the fact that the quantifiers of each were semantically alien to the other.
This leads to the non-acceptance of Ref 2. by a Convention Lover.
Ref 2. If Gabriel utters a truth by a sentence of the form ‘That is F’ then E(m)x(m) (‘that’ refers to x(m))
Doubt: But even if in general the Convention Lover might say this kind of things I do not see why in the specific case Hawthorne postulates he cannot say that Michael will be able to express the thoughts of the other because Gabriel’s language seems to be just a sublanguage of Michael’s language; a restriction of it.

Hawthorne says that: The Convention Lover will happily speak of the truth and falsity of sentences with superficially more restrictive ontologies. But she will not use the familiar kinds of apparatus to describe how those sentences get to be true; she will not use the concepts of domain, reference, extension, property, and so on in this connection. One normally thinks of the concept of sentential truth as forming part of a family, linked integrally to such concepts as reference, being true of, and so on. Retain the family and one will inevitably favour the Plenitude Lover over the Convention Lover.
Doubt: But then, it seems like HaWthorne has chosen the Convention Lover easy to fight with. What are his reasons not to choose a Convention Lover of the kind that defends there is no transcendent truth predicate (he says Quine and Carnap were of that sort.)


Hawthorne says: Suppose we take the Convention Lover is speaking a language in which she is right to say of the central claims of her mereology –formulated in her language- that they are analytic. How would she then be situated vis-à-vis the Plenitude Lover?
And he adds: Let us suppose that the Plenitude Lover is speaking a language in which quantifiers and variables are deployed in such a way that the central tenets of his mereological theorizing are neither analytically true nor analytically false.
Doubt: it is my fault, but I have the following basic doubt: if one is a Plenitude Lover, is there some reason not to claim that the central tenets of mereology are analytic?

Sunday, January 07, 2007

Is it possible to have two dossiers with exactly the same information?

This is the question that raised a bitter discussion in the last session of the reading group on propositional attitudes. We were reading Kripke’s (1979) A puzzle about belief, and we end up talking about the antidescriptivist argument in pp. 260-2. According to Kripke: “The puzzle can arise even if Pierre associates exactly the same identifying properties with both names” (p. 260). Now, this is hardly convincing for a modern description, someone who would accept a version of metalinguistic causal descriptivism (the reason being that the identifying properties for a name will contain the name itself; for a name N we will always associate the property of being the bearer of N.)

The discussion began when someone conceded that it’s possible something close to that: that it’s possible to associate to a given proper name two mental dossiers with the same information. Part of what’s at stake here is what a dossier is, and I don’t want to start begging the question in my favour. But what I can say is that the dossier for a given proper name comprises the information we associate to that name and that we have more than one dossier for those proper names we use to refer to more than one object (like ‘Aristotle’.)

So is it possible to find examples of a proper name with two dossiers containing the same data about the referent? Manuel Perez Otero came up with one of the best examples. Let suppose that 20 years ago you hear about a Brazilian football player called ‘Socrates’ and about a physician football player with the same name. They happen to be the same person, and so we got a case in which there are two dossiers for the same name and person. Later you forget some of these things; some information is eliminated from these two dossiers. You forget that the first ‘Socrates’ was Brazilian and that the second one was a physician. Now, do we still have two dossiers?

My answer is that none. After we forget those things we just believe that there were two football players called ‘Socrates’ and we cannot meaningfully say anything about none of them. We cannot say ‘Socrates was a football player’ because we just belief that there were to Socrates and we don’t know anything that would possibly distinguish one from the other. This would have been different if at the beginning we just had one dossier…

This answer makes perfect sense if you treat dossiers as individuated by the information they contain. A dossier will be just a set of beliefs that are believed to be by the subject about the same object. But, is it possible, for those who defend that we still have two dossiers, to provide a plausible elucidation of the notion of dossier that would render their claims true?