New APPS

13 responses to “((Another (possibly boneheaded) question about a priori contingencies.”

1. 

   GFA

   January 23, 2014

   I’m not a philosopher of language either, so I am happy to be corrected. I think you are right and that Papineau has made a mistake (or at least not expressed himself as felicitously as possible) when he says that ‘Julius’ is DEFINED as ‘the inventor of the zip.’ The usual way of talking about Evans’ example of Julius is to say that ‘the inventor of the zip’ is a reference-fixing description, not a definition of ‘Julius.’ Since reference-fixing descriptions don’t give the meaning of the words whose referents they pick out, ‘Julius invented the zip’ is not true by definition, and thus it need not be necessary.

   LikeLike

   Reply
2. 

   Richard Baron

   January 23, 2014

   I take GFA’s point about reference-fixing descriptions, but let’s see what happens if we try to regard “the inventor of the zip” as a definition.
   Let w be a variable that ranges over worlds in which zips have been invented (overlooking the complication of non-unique inventors of zips in a single world).
   What you know a priori is:
   (All w) (inventor-in-w = inventor-in-w)
   You therefore know the following a priori, where w* is the actual world:
   inventor-in-w* = inventor-in-w*
   And that is a way of saying “Julius invented the zip in w” if you define “Julius” as “inventor of the zip in w”
   The extra bit “in w” is crucial. It stops you going back up to (All w) (inventor-in-w = Julius), and you would need to get back up to that in order to conclude that necessarily, Julius invented the zip (even on the acceptable, de dicto, reading).
   Then the point could be that use of a proper name like “Julius” pushes you to the definition as “inventor of the zip in w“, rather than allowing you to define “Julius” as “inventor of the zip in whichever world we happen to be considering” (the phrasing of the last definition is intended to avoid requiring any trans-world identity).
   The effect is pretty much the same as the one you get by taking the reference-fixing description route, but I think it exposes the point about which one could argue in a particular way. Is it the case that definitions which you attempt to substitute for proper names need to be definitions “in w“?
   It would not be the case if you regarded the proper name as a mere abbreviation for your definition, that is, if you regarded it merely as a string of characters and paid no attention to the fact that it was generally used as a proper name. But if you did pay attention to that fact, and you were effectively saying “Here I am, in w, curious about who invented the zip, and I don’t know any biographical details, but I want to say that he or she was inventive, understood textiles and metals, and so on, so I shall just call him or her Julius for the sake of having a label”, then maybe your definition would need to be a definition “in w*”.

   LikeLike

   Reply
3. 

   Eric Winsberg

   January 23, 2014

   GFA: did you mean your last word to be “a priori” rather than “nececessary?”. (If not I am even more confused!)

   LikeLike

   Reply
4. 

   tjfxh

   January 23, 2014

   What we know from the information is that a person x did y and where x happens to be a particular individual designated as “Julius,”. The form is y = f(x), where x is some person that is assigned an identifier, here a name, that is., an element of the set of people capable of satisfying the the function of inventing a zip drive. The appearance of contingency arises from specifying a particular value for x from the many possible names as identifiers. But the name doesn’t connect the name “julius” to its interpretation as a particular person that happens to be named Julius. This is not knowable based on rules alone — definitions, axioms, formation and transformation rules. Semantic interpretation of symbols signifying that which is extra-linguistic and extra-logical is a matter of experience, either direct as one’s own experience, or indirect, e.g., testimony or inference based on valid logical form and true premises. Another case of bewitchment of intelligence by language.

   LikeLike

   Reply
5. 

   GFA

   January 23, 2014

   Hi again Eric —
   Sorry to say that I did mean what I wrote. If ‘the inventor of the zip’ is a reference-fixing description (instead of a definition), then ‘Julius is the inventor of the zip’ is contingently true. (And a priori.)
   I’m not 100% clear on exactly what bit you find confusing, so I’m not sure what would be helpful to say. But the idea is that the phrase ‘the inventor of the zip (whoever that is),’ used in the actual world, singles out a particular individual in our world. Then, when we say “Let’s call this individual (whom we’ve singled out via this description) ‘Julius’,” the word ‘Julius’ refers to this same individual in all possible worlds (since proper names are rigid designators). However, in some other possible worlds, someone else invents the zip: in some other possible worlds, ‘Julius = the inventor of the zip’ is false. So ‘Julius = the inventor of the zip’ is not necessarily true.
   I apologize if I am just piling on more confusion!

   LikeLike

   Reply
6. 

   Jessica Williams

   January 23, 2014

   No–I think it is clear that GFA means “necessary.” That is, “Julius invented the Zip” is not necessary (in fact, it is contingent) because “Julius” is rigidly designating a person, and it is contingent that that person invented the Zip.
   GFA is right that “the inventor of the Zip” is a reference-fixing description, and not a definition. (Does Papineua actually say it is a definition?) If it were a definition, then we would have an a priori necessary truth when we said that “Julius invented the Zip,” and the whole point is that it’s supposed to be contingent.
   Now, what is troubling about the whole business of these a priori contingent truths is that it seems that we can know a great deal many things a priori just by stipulation. I can stipulate that “Julius” rigidly designates the person in the 80th car in front of me on the highway. And I can know a priori that that person is in the 80th car in front of me! What I know is NOT only that the 80th person in front of me is the 80th person in front of me (although I know that, too), rather, I know that THAT person is the 80th person in front of me. This is why a number of philosophers have proposed an “acquaintance constraint” on reference and singular thought.

   LikeLike

   Reply
7. 

   Eric Winsberg

   January 23, 2014

   OK. OK. Its all clear to me now.
   1)GFA: Sorry, I thought you were agreeing with me more than you are, and I was thrown off by the need not be necessary phrasing. So, in fact you are saying that “is the inventor of the zip” is a reference fixing description, and so J is both a priori and contingent.
   2) Jessica: Thank you for helping me to understand what GFA meant AND to see what it is I find puzzling about the view. It strikes me as utterly bizarre that I could fix the reference of a term in such a way that I could then come to know substantive and detailed facts about the referent a priori. So, I guess I would want to endorse an acquaintance constraint too. I gather Evans himself endorsed an acquaintance constraint?
   I can also see why an acquaintance constraint would be a hard bullet to bite… but the alternative seems crazy to me.

   LikeLike

   Reply
8. 

   GFA

   January 23, 2014

   Sorry I wasn’t clearer before… I just want to add that I don’t know what I myself think about (J) — I was only trying to articulate the orthodox view about the Julius case, and explain why what Papineau said about it is misleading.

   LikeLike

   Reply
9. 

   Eric Winsberg

   January 23, 2014

   Yes, thank you. What I really wanted to make sure was that I have the orthodox view just right, because I am teaching this next week. My guess is that DP just thought the phrase “reference-fixing description” was too obscure for his audience. The confusion was mostly mine–but Jessica diagnosed the source of it correctly: I find the orthodox view puzzling and DP’s slightly misleading paraphrase of it just amplified my puzzlement.
   I gather spelling out just how to impose an acquaintance constraint is not so easy, and subject to lots of objections.

   LikeLike

   Reply
10. 

    Nicolas McGinnis

    January 23, 2014

    There’s obviously no problem with using a reference-fixing description to pick out a unique individual at a world, I don’t think (or with the fact that the unique individual picked out will have predictable metaphysical properties (including, in most cases, that the reference-fixing description leverages a contingent property of the individual)).
    What I do object to is the use of natural language categories and their tokens instead of sticking to talk of the logical devices. It is perfectly clear, and rather trivial, that I can think of “the tallest man in Idaho” either de dicto or de re; and that if I say something like “take the tallest man in Idaho, call him, uh, John, say,—” the fact that I used “John” to simplify the discussion—or for whatever reason—doesn’t commit me to a de dicto or de re reading.
    I could easily mean ‘John’ as ‘the tallest man in Idaho, whoever that is in all possible worlds’, or I could mean ‘John’ as in “the tallest man in Idado in the current world,’ who, at some worlds, is not the tallest, or is not a man. So I don’t buy the line that because names are always rigid designators we have a case of the contingent a priori; rather, it is because individuals have specific metaphysical properties.
    So only the de re reading is a (putative) instance of the contingent a priori, and I can stipulate that this is what I mean. But it seems relatively innocuous. In particular I am confused about what problematic substantive and detailed facts can be known about de re John apart from standard metaphysical properties that accrue to any generic individual. (I’m not up on this literature, save Kripke’s strange meter-stick example; what work is the upper-case THAT doing in Jessica’s comment? Is it because we think rigid designators imply a causal theory of reference, but there is no direct acquaintance to support the implied causal link, and that’s the issue?)

    LikeLike

    Reply
11. 

    Jessica Williams

    January 24, 2014

    I take it that it is precisely NOT trivial that one could, just by taking a definite description like “the tallest man in Idaho” and making it into a reference-fixing description, have thoughts about that person (de re). One reason this is not trivial is that claims that look like they ought to be a posteriori can become a priori. The classic example from the literature (first in Kripke, but subsequently taken up others, most recently by Hawthorne and Manley in The Reference Book–though it should be noted that H and M argue against an acquaintance constraint) is taken from Leverrier’s prediction of a planet responsible for the perturbations in Uranus’ orbit. Leverrier’s claim that it a planet was responsible for the perturbations ought to be a posteriori. But if Leverrier were to simply stipulate that “Neptune” was “the unique planet that is the perturber of Uranus” then he can a priori know that Neptune is the planet responsible for the perturbations! (That is, assuming that “Neptune” is a rigid designator, that we can use reference-fixing descriptions to rigidly designate objects, and that Leverrier could grasp the proposition “If a unique planet is the perturber, it is Neptune”). Now, one might want to draw a distinction here between reference and singular thought. That is, you might think that Leverrier can successfully refer to the object (in this case planet) responsible for the perturbations, but he cannot have singular thoughts about that object. If one does grant that he can have singular thoughts, one then faces the problem of explaining the strange (lack of) cognitive significance of such thoughts.

    LikeLike

    Reply
12. 

    Richard Heck

    January 24, 2014

    Jessica, just a quick note that a very strong version of the view that there’s no acquaintance constraint has been defended by Robin Jeshion. I don’t find her view convincing, but she does such a good job with it that it’s much easier to see what the real problems with the view are.
    For what it’s worth, I think the literature has suffered greatly from a failure to distinguish questions about reference from questions about singular thought.

    LikeLike

    Reply
13. 

    Nicholas McGinnis

    January 24, 2014

    Thanks so much–that clears up the issue for me, and the Hawthorne-Manley citation was very useful!
    (I don’t know what to think about the Neptune case specifically, except that problems like these sometimes make me sympathetic to internalist semantics: meanings seem insensitive to the ontological status of externalia as a matter of routine, and only (important, useful) post-facto regimentation my philosophers brings puzzles to the fore. Too much Wittgenstein in my youth, I guess.)

    LikeLike

    Reply