Amie Thomasson

Amie Thomasson (University of Miami), “Answerable and Unanswerable Questions,” with commentary by Jason Turner (Rutgers University).  The paper, commentary, and author's reply can be found here.

Comments

[Kenny Easwaran]

OK, time for some metametaontology...

Is it correct to say that some of the disagreement here (between the paper and the comments) is about whether there can be sentences with meaningful truth-conditions that can't be put into language?

If so, then just as we can see a lot of what we do in teaching philosophy to students as clarifying the distinction between when something is the case, and when we can know it to be the case (and suitably sophisticated forms of verificationism as showing how this distinction breaks down), we might be able to see Amie's paper as clarifying the distinction between when something is the case, and when it can be stated to be the case. Just as the former distinction helps clarify the relation of metaphysics and epistemology, the latter helps clarify the relation of metaphysics and language.

[Matti Eklund]

I think I remain a bit unclear on Amie’s positive view. The following are half meant as objections, half meant as requests for clarification. Amie says that to establish that a given sortal is instantiated, it is sufficient to (a) do the requisite conceptual analysis, (b) check the empirical facts. Then consider the following two worries.

(1) It seems that there are sortals such that they cannot both be instantiated but such that it appears they both are instantiated by Amie’s (a) and (b). Both the discussion of the so-called Bad Company objection to neo-Fregean philosophy of mathematics, and some recent papers by Gabriel Uzquiano arguably provide examples of sortals such that they cannot both be instantiated, for they require different things of the cardinality of the universe. How can such cases be dealt with in Amie’s framework?

(2) Take the hypothesis that there are "Lewisian worlds" (LWs): other concrete worlds not causally or spatiotemporally related to ours. What can Amie say about this? There are two main possibilities, as far as I can see. The first is that she is committed to their existence; the second is that she can say that we should remain agnostic. The second alternative seems nicer. Let me explore that. The idea would be that the empirical facts are perfectly consistent with there being LWs; but the claim that there are LWs isn’t forced upon us. The potential trouble with this reply is: if agnosticism is the right view here, why isn’t it the right view to adopt also with respect to all pure abstracta?

[Malachai M. Nilsai]

Correct me if I'm being too free in my comparisons. But in other words (to use the language of Donnellan): the baptism of a name involves, first, certain conditions of attributive usage in order for the name to come into being, and second, certain conditions for referential usage in order to stick the name to a thing in the world.

I am actually very, very sympathetic to any effort that tries to collapse metaphysical absurdities. But nevertheless I have worries.

- Let's say I am walking down the beach and I see a thing that I've never seen before: it is like a vortex, but also a train, and perhaps a platypus with a lemon tree growing from its ear, draped in a color out of time. It is a Lovecraftian horror which I am unable to fit into any category of my experience. Can I not, then, christen it with a name? Might it not be plausible to refer to it before giving it a category?
- Does this indicate that all names (constants) should, by necessity, carry with them some predicate term or other in order to be intelligible? IE: that "a & b" is metaphysically nonsensical, but "Fa & Fb" is not? If so, should logic conform to these facts?
- Your analysis takes questions of existence and the existence-quantifier to be synonymous. But why, in general, should we expect the 'there is at least one x' quantifier to be synonymous with an assertion of existence? It seems to me that the former is best used as a provisional or conceptual claim that arises by stipulation for the purposes of examining what follows from what, while the latter involves a genuine ontological commitment.
- How is 'How long is a piece of string?’ a pseudo-question? It might be vague, but every explanation always is, and it seems absurd to say that vague answers are not truth-evaluable according to non-vague criteria. IE: if you were to specify the answer in centimetres by default, but refuse to do it in millimetres or micrometres, you should not be faulted, and your answer would be satisfying. If we can't even answer those questions, then no mundane questions can be answered either (unless you were to specify the appropriate conditions of an answer within the question).
- What exactly is the difference between the more esoteric forms of physics and metaphysics?
- Your analysis is occupied largely with the nature of objects. What about issues concerning causation?

[Amie Thomasson]

A quick response to some of the comments (with thanks to those who posted them).

Kenny's comments seem to me right on. Thanks, Kenny, for this nice way of putting things.

I'll have to come back to Matti's first point after I've had a chance to look back at those references (thanks for those). On the second point (about possible worlds), there's room for a deflationist account of what it is to be committed to possible worlds, paralleling in some ways the deflationist accounts I've given of fictional characters, etc., following Schiffer's work on pleonastic entities (see my "Ontological Minimalism", APQ 2001). That is, one may hold that "There is a possible world in which it is the case that P" is derivable simply by a hypostatizing transformation from "It is possible that P" (or "Possibly: P"), and involves us in no 'additional' commitments. (Of course this leaves open still how to understand the latter form of modal discourse and its commitments; I say something about this in Chapter 3 of _Ordinary Objects_). This is at any rate a way of handling the issue without agnosticism, and while suggesting that the first alternative--of being committed to possible worlds, in a sense--might not be too un-nice.

Yentz raises a number of questions (I'll number them in order)--here's a smaller number of answers.

#1: I find the example somewhat hard to imagine, so it's hard to know exactly what to say about it. But let me give a general kind of answer to this general kind of worry: Many people seem to share your intuition that we can refer to 'something' without having any category of 'thing' in mind to refer to. If no frame-level _application_ conditions are carried with your use of the term (that, say, would tell us whether or not your term would still refer in a situation in which it turned out that you were merely hallucinating), then I'd say we can't say whether or not the term refers or evaluate the truth of existence claims. But one may provide frame-level application conditions for a term without associating it with any identity conditions, and so without determining what category of entity is to be referred to. I'd like to remain neutral on whether we may refer in those cases (not to rule it out). Still, I'd say that to the extent that the category of entity your term is to refer to is unspecified, various claims you might make involving that term may be incomplete and not truth-evaluable, (e.g.) we can't answer questions about the conditions under which 'it' would or would not persist. (To choose another example, if I ground the term 'Sam' as a term for 'that thing' without disambiguating whether I mean 'Sam' to name an animal or a mereological sum of molecules, then questions like 'would Sam survive a fire?' are unanswerable.) And this inability isn't based in a failure of knowledge but a failure of meaning.

#4 I think you may have misunderstood the example. The "How long is a piece of string" question is just an old joke. It's not meant to be interpreted as 'how long is _this_ piece of string' (or 'the piece in my pocket'), in which vagueness would be the only concern and wouldn't make the question unanswerable. (I discuss how to handle vagueness in Chapter 5 of _Ordinary Objects_). The idea is that the question simply doesn't specify _which_ piece of string we're talking about, and so (given that pieces of string may also vary radically in length) it's ill-formed and unanswerable.

[Yentz Mahogany]

Amie,

Anything about #1 which is incomprehensible is likely due to the fact that I've been too whimsical. But your reply addresses my point spot-on, so no worries there.

I'm not sure I feel very well about this. It seems implausible for me to say that an indescribable thing which happens to be standing in front of me is bereft of membership to a stable category. Let's say this unknowable thing stayed in one place for a long time. Wouldn't its location in space-time count as a category? It seems to me that a person could then say, "The thing is staying put", and that would be truth-evaluable (and true).

The role of persistence in all this seems to be critical. Yet not all meaningful statements are about those features of a thing are ones which are taken to be persistent; nevertheless, they're truth-evaluable. That realization seems to carve out at least a minimum amount of meaning from the thing, albeit perhaps a pragmatic meaning which is confined to the present moment.

Regarding #4: Hah, thanks. I haven't heard the 'string' joke before.

[Matti Eklund]

A quick follow-up in response to Amie’s response to #2:

I’m perfectly happy to believe that Amie can find entities more lightweight than Lewisian worlds to do the work of possible worlds in the philosophy of modality. At any rate, the potential problem I wanted to pose is perfectly compatible with this. (In fact, this was the reason why I didn’t say that Amie was committed to a particular view on possible worlds, but only suggested that she may be committed to “Lewisian worlds”.)

Let me restate the original worry with this issue out of the way: What about the hypothesis that there are Lewisian worlds - other concrete worlds not causally or spatiotemporally related to ours? (Forgetting the worry about whether we need to appeal to them in the philosophy of modality.) They seem to satisfy conditions (a) and (b) [see Amie’s paper and my previous posting] as well as do abstract objects. But surely we don’t want to be committed to such objects. (Or?)

[Amie’s response sort of suggests that despite her otherwise Fregean outlook she may be unhappy about (non-“pleonastic”) abstract objects. (I haven’t read the APQ 2001 piece I’m afraid.) If so, maybe she’s not that unhappy about having to treat Lewisian worlds and (non-“pleonastic”) abstract objects on a par.]

[Amie Thomasson]

To Yentz--yes, I agree that there may be situations in which you can succeed in referring (using a term with application conditions that are fulfilled), but where you leave the associated identity conditions pretty open (thus not fully specifying a category). So suppose you use your term in such a way that you'd be happy for it to apply to an artifact, or a mere lump of stuff, or an animal, or a mereological sum of atoms... not specifying which. Some things you say may be truth-evaluable regardless (e.g. it's staying put); other claims you might make (especially modal claims), like 'it would survive a fire', 'it's the same one I saw last week', etc., won't be (unless you are tacitly associating enough identity conditions to make them evaluable).

But notice that even making claims like 'it's staying put' seem to involve presuppositions that narrow down _somewhat_ the category of entity you're referring to--e.g. not an instantaneous time slice (for there would be no persisting 'it' to stay put), and probably not a spatiotemporal region (since then the comment would be quite odd)...

[Amie Thomasson]

Hi Matti--

Thanks for the follow-up. The point of the pleonastic line is that by using (a)--the conceptual analysis--we can determine just how little such apparent commitments commit us to. E.g., given that the transformations that take us from 'it's possible that P' to 'there's a possible world in which P is the case' (and similarly from 'this apple is red' to 'there is a property of redness this apple has') we can see that the truth conditions for the latter claim are guaranteed to be met provided those for the former are. So we should allow that these conditions are met, and that there are such 'things', in this trivial sense. (And yes, as you suggested, I'm happy to treat issues of possible worlds and abstracta roughly on a par in this respect... though some different issues arise in dealing with the basic (pre-transformation) modal claims.)

In short, then, the conceptual analysis involved in revealing these pleonastic transformations is supposed to deflate the issue between the hyper-realist and the eliminativist. Given the pleonastic connections between these claims (a deflationist like me asks), what more is the hypothesis that there there 'really' are possible worlds or properties (in some 'robust' sense beyond my trivial sense) supposed to involve--such that the hyper-realist thinks these conditions are met, and the eliminativist denies it? (Mark Johston develops a nice early version of this approach in his paper in the Rea volume on Material Constitution). Both, I'd say, are going astray at the stage of conceptual analysis.

[Yentz Mahogany]

Amie,

Very interesting. Thanks. It sounds like a serious challenge to causal theories of reference.