Why Must the Facts about Parthood be Systematizable?

I’m writing a review of Horgan and Potrč’s new book, Austere Realism, and I thought I’d float some of the ideas here first. Their big project is to defend a totally stripped-down ontology, in which (among other things) there are no tables, mountains, or other ordinary composites, but all the while insisting that ordinary utterances like ‘there are tables’ are strictly and literally true. I’m planning to have a second (and possibly third) post on their strategy for reconciling their revisionary metaphysics with ordinary discourse, but first I want to talk about one of their arguments against “naive commonsense ontology.”

The idea behind their “argument from arbitrariness” (§2.3, if you have the book) is that naive commonsense ontology is too messy and unsystematizable to possibly be right. Here I think is a fair reconstruction of the argument:

(P1) The correct answer to the special composition question, “under what conditions are some things the parts of a further thing?”, must be general and systematic.
(P2) There is no general and systematic answer to this question that accommodates the bulk of our intuitive judgments about when composition does and does not occur.
(C) So a great many of our intuitive judgments about when composition does and does not occur must be incorrect.

Anyone who’s attempted to give even a rough-and-ready answer to the special composition question that accommodates our intuitive judgments about composition is going to feel the force of P2. But why accept P1? Horgan and Potrč suggest that if indeed the compositional facts do not conform to some general and systematic answer to the special composition question, then they would have to be explanatorily basic (or “brute”), which surely they are not.  It is far from obvious, though, that one who embraces the bruteness of composition—i.e., that there is no general and systematic account of the conditions under which composition occurs—is thereby committed to the brutality of compositional facts—i.e., that facts about whether composition occurs are explanatorily basic. To deny that there are interesting general principles governing the conditions under which composition occurs seems entirely compatible with maintaining that, in each case, there are noncompositional facts in virtue of which the compositional facts are as they are.

Now for the ad hominem: Horgan and Potrč themselves defend just such a particularist claim later in the book. For reasons having to do with their revisionary semantic theory, they wind up embracing a semantic particularist thesis on which there are no general and systematic answers to a certain range of questions about the semantic facts (§6.4). But of course they don’t think that the unsystematizable semantic facts are explanatorily basic; rather, they’re explained by specific (nonsystematizable) facts about the speaker’s context together with facts about intentional mental states (§7.4.3).

One last thing: It’s worth comparing the special composition question to the “special inheritance question,” that is, under what conditions does a composite inherit properties from its proper parts? (I’m borrowing this case, and most of the examples, from Dave Liebesman’s very cool paper on generics.) Hard question. In order to have the property of touching the wall a chair only has to have a single part that’s touching the wall. But, in order to have the property of being plastic, the vast majority of its parts have to be plastic. And in order to have the property of being red, only the parts on its surface have to be red. And it may be 5.7 pounds even though none of its parts are 5.7 pounds. In other words, good luck trying to find a general and systematic answer to the special inheritance question! But we don’t find that intolerable; we don’t take that as grounds for skepticism about our “naive” intuitive judgments about inheritance. We don’t then become universalists about inheritance (all of thing’s parts have to be F in order for it to be F) or nihilists (inheritance never occurs) or god forbid organicists (composites only inherit properties from their living parts), for fear of being stuck with an unsystematizable mess. And notice that here, the bruteness of inheritance certainly doesn’t entail the brutality of inheritance facts: the table inherits the property of being red from its parts because the parts on its surface are all red.

First question: is there any reason why one who denies the bruteness of composition also has to deny the brutality of compositional facts?

Second question: is there any reason to think that bruteness of composition (without brutality of compositional facts) is worse than bruteness of semantics (without brutality of semantic facts)?

Third question: why is the lack of a systematic answer to the special composition question any more reason for skepticism about our intuitive judgments than the lack of a systematic answer to the special inheritance question?

Comments

[Kenny Easwaran]

How general and systematic does something have to be to satisfy 2? Are the facts about motion general and systematic enough to satisfy them? After all, these facts are extremely complicated and involve four separate types of force, each involving a different relation between distance and some other properties of various objects. In fact, most physicists still don't think our understanding is systematic enough, which is why they've built the Large Hadron Collider (which just started up yesterday!). But this hasn't turned anyone into skeptics about motion. But still, most people would probably adhere to something like 2, I would think.

[Dan Korman]

Hi Kenny,

I guess the reason people don’t become skeptics when they see how difficult it is to systematize the facts about motion is out of a (reasonable) optimism that we’ll eventually find a systematic account. I say it’s a reasonable optimism, because we do seem to at least be making progress towards an empirically adequate systematic account of the motion facts, whereas there’s been nothing resembling progress in our search for an intuitively adequate account of the compositional facts.

But the analogy to motion does raise some interesting questions (in addition to your question: just how systematic does the account need to be?) For instance, suppose that physicists ultimately conclude that there is no systematic account: we can explain why things move the way they do in each case, but there’s no one principle that predicts what happens at every level or for every combination of interacting fundamental particles. Certainly there would be no temptation whatsoever to adopt a revisionary view of motion that cannot accommodate all of the empirical data. So why should the lack of a unified theory of parthood (assuming that we can explain, in any given case, why composition does or does not occur) be grounds for skepticism about the intuitive data?

[Bryan Pickel]

So don't we just expect generalizations about semantics and pragmatics and for that matter, economics, psychology, literature, etc to be ridiculously complicated? Why? Well, perhaps b/c their supervenience base includes like a billion neurons (I actually have no idea about this, it seems to me that whenever we cut into someone's head we just find a pile of goo, but I'm told that there's more to it). But when I consider what it takes for two things to make something, I don't expect the supervenience base to include too much beyond these two things, so I expect it to be reasonably systematic. Another way to put the point: if facts about the fundamental physical ontology (including parthood) aren't relatively clean, lawlike and sytematic, then how could we expect ever to get anything like a generalization over more complex things like psychology and semantics? Anyhow, that's just a reason to expect the semantic laws to be more complicated then the parthood laws. There is an alternative which I'm actually somewhat sympathetic to. My argument crucially seemed to rely on the view that the facts about parthood are among the basic facts in terms of which all others are explained. This may have been wrong. If it is wrong, and facts about parthood are derivative - perhaps even on facts about those billion neurons, then there's no particular reason to expect them to have relatively clean laws.

[Dan Korman]

Bryan,

I think you understate how complex the supervenience base for parthood facts may have to be. After all, the same supervenience base that’s going to determine whether your two particles compose something is also going to have to have enough in it to determine whether composition occurs for any number of things in any one of the infinitely many arrangements that some things can be in. Additionally (but very controversially) intentional facts are going to affect the facts about composition. Whether a suitably arranged steel ball and steel rod compose something – specifically, a token of the lower-case letter ‘i’ – is (arguably but plausibly) going to depend in part upon whether anyone intended to make a letter ‘i’. If so, then we’ll need those same messy intentional facts in the supervenience base as we have in the base for the semantic facts.

Also, you say that if the facts about parthood aren’t among the basic facts, then “there’s no particular reason to expect them to have relatively clean laws.” What’s interesting is that this is in direct opposition to Horgan and Potrc, who say that the parthood facts aren’t basic and *therefore* must have relatively clean laws. I’m tempted to agree with you, though really I’m not sure why we should expect relatively clean laws either way: whether those facts are basic or whether they’re derivative.

[Dan Korman]

Terry Horgan sent me this email, and I'm posting it on his behalf (I'll post a reply soon):

"Very nice post! My initial reaction is not to shoot off some from-the-hip reply, but rather to acknowledge the depth and difficulty of the questions you raise. Broadly similar in kind, it seems to me, to questions about vagueness that arise for those of us (and there are many) who think there is not, and/or cannot be, ontological vagueness. How can there be real mental states/properties, really instantiated in the world, that on one hand have vague intentional content, and on the other hand are not ontologically vague? That is a question that Matjaz and I pose and address in the book, but it seems to him and me there is a lot more that needs doing to address it adequately. We had a further stab at it at a recent conference on consciousness and thought in Dubrovnik this past August, co-organized by Zdravko Radman and Dave Chalmers.

Your questions/worries are similar in spirit. Here too, it seems to me that they potentially arise for many metaphysicians other than Matjaz and me--since many folks think that brute-composition facts are anathema.

My first reaction, then, is to acknowledge how good and pressing your questions are, and to put them in my mental pipe to smoke. And to urge that many metaphysicians face such questions, not just Matjaz and me.

That said, here are a few (very preliminary) thoughts. You ask (your first question), "Is there any reason why one who denies the bruteness of composition also has to deny the brutality of compositional facts?" Well: Familiar, and widespread, conceptions of explanation think of explanation as a matter of subsuming specific phenomena under general laws/principles. Here we come up against the pressing question of what *other* pertinent forms of explanation might look like. How might one explain compositional facts (so that such facts are non-brute), without appeal to general laws/principles? Damn good question. (And: might there be particularistic explanatory formats that work in the realm of non-ontologically-committing semantic standards for what Matjaz and I call "indirect correspondence," but *do not* work within serious ontology? I think our intuition, Matjaz and mine, was that the answer here is "Yes," but certainly a story needs to be told that we did not really tell in the book.)

I'll leave it at that. Raising good questions is a very valuable and very serious philosophical enterprise, and you're doing so. This is food for serious thought, not for quick/glib responses.

Feel free to post this. I myself am a low-tech dinosaur, so I'm just emailing it
to you.

Best,
Terry"

[viagra online pharmacy]

Hey great information! I like this post 'cause is just what I was looking for so please keep sharing things like this and keep up the good work.