:: Article

about aboutness

Richard Marshall interviews Stephen Yablo.

[Photograph of Steve Yablo by Jon Sachs, MIT SHASS Communications]

Stephen Yablo is the Magilla Gorilla philosophikilla who thinks all the time about ontologies and metaphysics and ontologeses and metametaphysics too, about essentialism, about whether intrinsic is intrinsic to essentialism, about fictionalism and evolving to presuppositionalism, about why conceivability is a guide to possibility, why zembos are harder to get rid of than zombies, about aboutness, about subtraction and about a Wittgenstein thing and other cool stuff. This one is the Ali shuffle thought of via its opposite, only in a mouth …

3:AM: What made you become a philosopher?

Stephen Yablo: Hmmmm. I guess it was Hebrew school. The teacher said that we must never judge God, since we don’t know a thing about him. I was in love at the time with Magilla Gorilla, a cartoon character. He struck me as a higher sort of being. This sounded nutty, I realized, and I kept it to myself. Then on hearing that nothing was known about God, I inferred that in particular it wasn’t known that he was not my loveable ape. I was told on raising this question in class that one thing was known after all; God was not Magilla. This confused me enough to start me down the road to philosophy.

3:AM: One area where you work is ontology and metaphysics where you make a distinction that seems important to your approach to many issues, a distinction between the quizzicals and the curious. Carnap might be seen as the parade case quizzical and Quine the parade case curious and you seem to lean towards Carnap. Is that right and if so what’s at stake for you in this?

SY: Right, one kind of philosopher is curious about what exists and seeks a way of finding out. Another kind, the quizzicalist, thinks that at least some existence-questions are objectively moot. difficult. Moritz Schlick and Susan Stebbing in the 1930s gave “is blue more identical than music?’’ as an example. Linguists interested in the autonomy of syntax used to dig around for grammatical statements that were nevertheless not interpretable. Chomsky’s “Colorless green ideas sleep furiously” never impressed me that much. I prefer an example that came up a few years ago on the blog Language Log: “More people have been to Germany than I have.” This sounds fine until we try to evaluate it, and realize that a comparison is called for between the number of people who have been to Germany and….what? Amie Thomasson, who is no quizzicalist, suggests “Do Dell computers help you get more out of now?” Quizzicalism is apt to seem unmotivated. Many people would say that they agree either with Carnap that of course there are numbers, since there are primes over ten, or with Quine that it’s a empirical question whose answer depends on whether numbers find a permanent place in the range of our quantifiers.

The funny thing is that Carnap is speaking about a rational reconstruction of English with “framework rules” taking the place of what is actually done by habit. And Quine is talking about a first order regimentation of English. They insist on the reconstruction/regimentation because they themselves can’t make sense of “are there numbers?” as it arises in ordinary English. (Compare also Sider on Ontologese.) Quine and Carnap are really themselves quizzicalists, then, arguably, just like me.

Of course we might differ on whether regimentation allows for a useful successor question to “are there numbers?” And on how curious we propose to be about the answer to that. But that is not the question of quizzicalism as I understand it. One could also rationally reconstruct talk of heaps so that it takes exactly 4 grains of sand to make a heap, or maybe 11. But we don’t believe in a shining, resplendent question of true heapiness waiting that will reveal itself when we clean up our act. That’s how I feel about (some) existence questions.

3:AM: How far has Kripke messed up your Quinean anti-essentialism? Are you neither an
essentialist nor an anti-essentialist now?

SY: I was a fanatical anti-essentialist in my undergraduate years, under the tutelage of Danny Goldstick and Bas van Fraassen at the University of Toronto. I was a quizzicalist about it in 1986 I wrote my dissertation. The first chapter laid out all the pro- and anti-essentialist arguments I knew of and tried to answer each from the perspective of the other side. (Donald Davidson, my advisor, said, “The anti-essentialist arguments are great, but the others seem unnecessary; who any more is an essentialist?” “Saul Kripke?” I ventured, but he shrugged that off as the exception that proves the rule. “David Wiggins?” “Stop right there,” he said, “David Wiggins is my friend.”) Later, when it came time to publish stuff, I tended to emphasize the essentialist end of the balance, which seemed somehow more in need of support. (Though I can’t imagine why that would have been; it was more in need of support at Berkeley in the 1980s anyway.) This then lead by some Pascalian mechanism into a wholehearted essentialism. I have been an essentialist since 1987 or so.

3:AM: Why are mereological considerations important to getting right the notion of intrinsicness and is this the same issue as essentialism?

SY: The world “intrinsic” leads via “internal” and “inherent” to “essential,” and it is sometimes used to mean essential. There’s a history here I don’t really know. The running-together seems to go back at least to Leibniz and Kant. Rae Langton weaves a fascinating tale on this in her book Kantian Humility. See also Moore, “External and Internal Relations.” I like to think of the two distinctions as independent. Properties can be had extrinsically but essentially, and so on. I complain somewhere about Lewis’s account of intrinsicness that it puts too much weight on the notion of a natural property.

It starts indeed from the idea that extrinsic properties are never natural. Even if this was right, and the account more generally delivered the right results, the correlation would be a “necessary accident.’’ Naturalness shouldn’t figure in an account of what it is to be intrinsic. The question then becomes, are there notions constitutively related to intrinsicness that we might appeal to instead? And then I hit on was the relation of part to whole. The proposal, following up on some ideas of Peter Vallentyne’s, was that P is intrinsic just if a thing that has (lacks) it in W cannot lose (gain) it by moving to a world W’ of which W is part. I am not saying this totally works—some good objections were made recently at the Oberlin Colloquium—but it seems, every other day, like a step in the right direction.

3:AM: Is your fictionalism a way of nuancing a Quinean position regarding how we’re supposed to believe things that aren’t limning fundamental reality? How does it work and how do you answer the push-back that if the best explanation ineliminably involves abstracta then we should believe in them and not treat them as fictions? Isn’t this an argument with bite when we consider contemporary physics, for example?

SY: Maybe so. Though I am not sure the concept of further nuancing a Quinean position is one we should want to instantiate.

I think we are still not clear what it means to say that the best explanation ineliminably involves abstracta. I take a shot at this in a recent MIND article. The upshot was: on some interpretations, explanations can ineliminably involve things nobody believes in or would feel committed to, such as infinitely deep oceans (an example of Penelope Maddy’s) or physical points at infinity. (This is a point Quine makes himself; it’s part of the reason for regimentation.) . On other interpretations, ineliminable involvement might carry more ontological weight, but then questions arise as to whether such explanations are really given. One thing that makes me suspicious is that “algebraic” math—group theory, vector algebra, etc— seems just as indispensable as “quasi-categorical” math—arithmetic, theory of the reals, set theory.

Yet algebraic theories aren’t even in the market for truth, because they don’t have a standard mathematical model. If we’re to believe in 0 because of arithmetic’s applications, shouldn’t the applicability of Boolean algebra argue for the existence of “the maximal element”? Let’s try to identify the non-truth-y features that underwrite algebraic applications, before insisting that truth is the only possible underwriter in the case of number theory.

3:AM: So according to your fictionalism what are abstract objects? Just metaphors, figurative aids to help us describe concrete reality so, for example, so the eight in eight apples is about apples, not eights? But ‘she’s got butterflies in her tummy’ even though it shouldn’t be taken literally could be. But how can a number have that option, say, of being literally understood or metaphorically understood, if they are just metaphorical. (Is that right?) Why doesn’t this make you a nominalist?

SY: I will answer in the guise of the fictionalist/figuralist I once was, before becoming a presuppositionalist. (Let me take this opportunity to apologize for “evolving” a few too many times on the issue of abstract objects.)

So, to begin at the end, I’m not a nominalist for the same reason that it wouldn’t make me a disbeliever in angels to use the word non-committally on occasion, as in, “you’re on the side of the angels.’’ It doesn’t make me a disbeliever in goats that I’m not talking about them when I say, “that really gets my goat.” The number case is admittedly trickier, because—this may be what you are getting at with “how can a number have that option?” —it is harder to make out a contrast between literal and non-literal uses.

One possible view is that certain terms lack literal meaning altogether. Kendall Walton used to say this about “exists,” that its whole reason for being is to enable us to speak in metaphorical terms about which terms refer. “Metaphorical” might indeed be a misnomer, Walton thinks, in the absence of a contrasting literal use. Davidson too maintained that metaphorical meaning, or usage, is always playing off literal meaning.

I don’t agree that fictional or figurative meaning rests on a felt contrast with literal meaning. It seems so only because one is assuming a Gricean conception of non-literal meaning as a kind of implicature; the search for a second meaning is triggered by our unhappiness with the first. I give the example somewhere of “gravid liquid,” posited in The Third Policeman as the sole intrinsic possessor of weight; it confers weight on other objects by being “subtly disseminated” through them. (It can’t be distilled a beaker because It would break through the bottom.) “Gravid” does turn out to have a literal meaning—-pregnant—but that is not what is guiding us when we read the novel.

A more figurative sort of example is “smarts,” as in, “She has a lot of smarts,” which has a literal meaning qua adjective but not noun, or “welcome” in “All three of you have worn out your welcomes,” which again has meaning as a verb but not noun.

That being said, I don’t (or didn’t) treat number-talk as straightforwardly metaphorical. Sometimes when we launch a sentence into the world our confidence in the sentence outruns our sense of how it is best interpreted, in particular whether it is best assigned a literal reading or a metaphorical one. There’s an implicit FINESS operator: construe it literally if possible, Figuratively If NeceSSary. These things can take a while to sort themselves out. And sometimes they never are sorted out.

3:AM: A key question you’ve wrestled with is how we can know modal facts. Can you give an example of one, why it’s significant and how you think we know the fact? And does this approach make a difference to how we think about the relation between conceivability and possibility? Why doesn’t your account of conceivability help refute physicalism, for example, when we conceive of zombies as a counterexample?

SY: Chalmers has a paper called “Does Conceivability Entail Possibility?” My earlier title was “Is Conceivability a Guide to Possibility?” The difference is important. Perception is a guide to actuality, but not a proof that, say, there are tables. It is only the idealist who thinks perceptual appearances cannot, when all the data is in, be wrong. Not that Chalmers is an idealist! He is talking about ideal conceivers.

Most of us think perception sometimes leads us astray, and that the reasons why this happens need not be perceptually apparent. That’s pretty much how I think about conceivability. Just as we go with the perceptual appearances, unless there’s a reason not to and/or they can be explained away, likewise we should treat conceivable scenarios as possible, unless there is a reason not to and/or they can be explained away. Here comes the perverse part. Chalmers has emphasized zombie intuitions: same physical base, no mental life. Descartes was fonder of what I somewhere call zembo (for “disembodied”) intuitions: same mental life, no physical basis, or any other kind of basis. Zombie intuitions are easier to explain away than zembo, I feel. So on the one hand I am suspicious of physicalism. But on the other I don’t think conceivability appearances support the possibility of zombies.

3:AM: Causality is another thing you refine in a way that seems to work between rivals rather than taking a side. It’s a counterfactual account that nevertheless isn’t compatibilist, which counterfactualists usually are. How does your approach deal with mental causation?

SY: It’s compatibilist about some causal relations—those built on straight counterfactual dependence, say— but not, or anyway less so, about “causation proper.” Causation proper is subject to a proportionality constraint, which tells us to leave out irrelevant extras. So, the death of Socrates was caused by his drinking the hemlock, and not his guzzling it, because the death would still have occurred if the drinking had occurred without the guzzling. More generally the less determinate candidate C wins out if the effect still would have occurred, had C occurred without its determinate C+. Pain is a determinable (!) of the underlying brain state, so it’s the pain that causes you to wince, if you’d still have winced had the pain been differently neurally implemented. That was the view in 1993, anyway. The landscape is different now— contrastivists have another way of explaining the judgments, as do causal network fans—so it might not be the view I would hold today.

3:AM: You’re finding the subtle nuances in logical space again with your new book ‘Aboutness’ and you start with a great example to help us see the issue. Your daughter tells you that you never get her ice-cream, you respond by saying that you got her some only last week at her birthday and she says “I wasn’t talking about that.’ So she seems to be dismissing a clear counterexample to her claim. You’re saying there’s something right in what she did. How so, and why does it matter?

SY: Sometimes we judge a statement on what it says about the subject matter under discussion. Zina was talking about what normally happens, special events aside. This cannot always be written off as laziness or loose talk. Maybe the best way to formulate our point is to overshoot it a bit, and then cut back. One way of cutting back is to zero in on what our statement says about the matter under discussion. Another is to strip the sentence of one of its implications. This is what happens with so-called exceptive constructions—Everything is packed but the food. Logical subtraction is meant to be a generalization of this. Triangles are similar if they’re congruent, except possibly in respect of being the same size.

3:AM: Subtraction was supposed to have been dealt with by the Wittgenstein question: what’s left over if I subtract my arm’s going up from my raising it? But subtraction has a big role in your theory of aboutness doesn’t it? Is it part of your idea that presupposition failure is not a problem but an opportunity natural languages take advantage of? And what’s the idea of proportionality you use and why is it so crucial?

SY: Let’s build up to the Wittgenstein thing in stages.

Start with “what A says about a certain subject matter M.’’ This is true in a world W just in case A can be made true by mucking with W in a way that preserves how matters stand where M is concerned. (I take from Lewis the idea of identifying a subject matter with the relation of being alike where it is concerned.) What does Al is 5’ 6” say about the matter of height to the nearest inch? That Al is between 5’ 5 1/2” and 5’ 6 1/2”. That’s because that is the range within which we can make Al 5 ‘ 6” exactly without laying a hand on how tall he is to the nearest inch. About height to the nearest half inch, it says he’s between is between 5’ 5 3/4” and 5’ 6 1⁄4”.

Then there’s “what A says that is not (at all!) about subject matter M.’’ This is trickier to define so l’ve got to ask you to grant me it. Now we can explain A~B like so: it is what A says that is not at all about the matter of whether B is or is not the case. So, what is left over when we subtract My arm went up from I raised my arm? I’ll tell you what: it’s what I raised my arm says that is silent on the issue of whether my arm went up. When you work through all the definitions you get that A~B is true in W when B ⊃A is true for the right kind of reason. The right kind of reason is the kind that doesn’t trade on B being false in W or A true there, targeting instead the gap between them. That’s probably more than you wanted to know, so I’ll stop there. Except to say, what may have been Wittgenstein’s point, that you never get the right kind of reason in worlds where my arm stays down. That my arm goes up is inextricable from my raising it in much the way that red is inextricable from scarlet. Another way to put it, using Williamson’s term, is that arm-raising does not factorize into arm-rising and some other thing.

3:AM: Does aboutness help in efforts to reduce types of knowledge into one another,
closure violations and other issues regarding knowing?

SY: T otally!!

3:AM: And for those here at 3:AM wanting to follow you into this philosophical world, are there five books you can recommend to us?

SY: Robert Stalnaker, Context.

Penelope Maddy, The Logical Must:
Wittgenstein On Logic,
Allan Gibbard, Meaning and Normativity.
Angelika Kratzer, Modals and Conditionals.
Amie Thomasson, Ontology Made Easy.

Richard Marshall is still biding his time.

Buy the book here to keep him biding!



First published in 3:AM Magazine: Friday, October 10th, 2014.