First let me change the example from cookies to a loaf of bread (this shouldn't matter). There are two ways of looking at what happens in the oven. First, we can say that we put something into the oven and when it -- that thing -- came out, it was quite different, qualitatively speaking, from how it was when it went in. In this sense nothing came into being or passed away. But we can also say that we put one thing into the oven and took another thing out. When we talk like this, we say: the loaf of bread came into being and the uncooked lump of dough no longer exists -- it has become something else which did not exist before. This latter is Clayton's view: as he put it, the [loaf of bread] is numerically distinct from the hunk of dough that went into the oven.
Is one of these ways of speaking the correct one and the other incorrect? That is, does one speak truly of what "really happened" while the other is loose or elliptical or downright wrong? If so, which is correct? If not, how can this be?
In my original post, I mangled the concept of substance, causing me to fail to see the point of Clayton's view (I called it "a mistake"). Leaving out at least three more definitions (isn't metaphysics grand?), let's distinguish between substance (1) = bearer of properties or substratum; substance (2) = entity or concrete individual; and substance (3) = type of stuff.
If concrete individuals can be the bearers of properties, which it seems they would have to be, then senses (1) and (2) of "substance" coincide. The only sense of (1) which is independent, is, I would think, is the extremely abstract one which we are not talking about here, so let's fold them together for now. The other one, (3), might seem simply colloquial, in that being (made up of) a type of stuff is a property, not a substance, which is the bearer of properties. On the other hand, some individual things are such that being made up of a particular type of stuff is essential to them, so that if they lost that property, they would thereby pass out of existence. In this case, we might say that senses (2) and (3) coincide, or at least overlap.
Here's the one view, then, in these terms. It is essential to [the substances (2) which are] loaves of bread that they are made up of [the substance (3)] bread; so if something is made up not of bread but of uncooked dough, then it is not (yet) a loaf of bread. At the (vague) point at which dough is cooked enough to be properly called bread (if still perhaps undercooked), then the lump of uncooked dough passes out of existence (because of course being made up of X is essential for something to be a "lump of X"), and a numerically distinct entity, the loaf of bread, comes into being. The thing which went into the oven underwent, as Clayton calls it, "substantial change" (and thereby passed out of existence).
Fair enough; I grant that this is a coherent way of speaking and not just the mistake I called it before. On the other hand, and this was my first point, I hardly need to be committed to the classical metaphysical notion of substance as pure substratum (an accusation, I should point out, has never been made, to my knowledge, here or elsewhere) in order to say instead, or as well, that something (some concrete individual) went into the oven, underwent a qualitative change therein (which it survived), and then emerged from the oven in a newly crusty state.
In other words:
I put the bread into the oven at 3:00, and when it came out at 5:00, it was done.
If that's not enough, here's a real example, not made up (although it's hard to imagine anyone doubting that people actually do say what I just did). My mom's been making bread in a bread machine, and in the manual it says the following:
For the French bread cycle you can expect the following things to happen as the timer counts down to zero.
To begin: The dough is kneaded for the first time. (18 minutes)
At 3:32: The dough begins to rise (40 minutes)
At 2:52: The dough is kneaded for the second time. (22 minutes)
At 2:30: The dough continues to rise. (20 minutes)
At 2:10: The dough is "punched down." (30 seconds)
At 2:10: The dough rises for the final time. (65 minutes)
At 1:05: The bread begins to bake. (65 minutes)
At 0:00: The bread is finished.
If at time = 1:05 to go, the bread machine contains either a lump of uncooked dough x-or a loaf of bread (which is just now beginning to bake), then these two accounts are at odds. Which is right? On one view of the matter, it doesn't matter what people actually say. All that matters is what is really the case, whether or not that's what we say in ordinary cases. If we are to get things right, as philosophers, we must discern the actual ontology and modify our (philosophical) language to track it. (Of course there's nothing wrong with speaking loosely "outside the study," as we say, or in the kitchen in this case.) If this is right, I have hardly helped my case by citing an actual example of what people say. Maybe that we do (sometimes) talk that way just means that (sometimes) we get things wrong (and need philosophers to help us out).
In this context consider two ways of understanding what has come to be known as "ordinary language philosophy." Obviously the "ordinary language philospher" (let's call him "Austin") wants us to take ordinary talk seriously. But does this mean that ordinary talk gets things right where "metaphysics" fails? That is, is the correct ontology picked out by ordinary talk while "metaphysics," in its misguided attempt at rigor, is simply false? Or is it rather that "metaphysical" talk and ordinary talk, although giving different accounts of the world for different purposes, are equally correct?
Neither of these sounds particularly attractive to the traditional ear. The first sounds like simple nihilism, as if there could be no point in speaking rigorously. This is worse than positivism, as while the positivists also rejected "metaphysics," they at least substituted a rigorous empiricism, where "Austin" just encourages the sort of loose talk we would engage in if there were no such thing as philosophical inquiry into the real. (This was Russell's attitude toward "ordinary language philosophy.") The second is no better. If both ways of talking are correct, then this implies that how things are depends in some way on how we talk; while if neither is correct, then this is a different sort of nihilism, of a skeptical sort, as if our language were essentially inadequate for describing reality.
Of course another possibility is that as it happens in this case, what we ordinarily say points us to the correct ontology, where another, more clearly "philosophical" account gets it wrong. If this were my view, I wouldn't have brought up ordinary language in the first place, seeing as I would need a traditional metaphysical argument anyway. (In my original post I left this possibility open. Oops.) Neither is the first conception of ordinary language philosophy my view. Ordinary language has no special ontological status, as if it were purer or uncontrived. This leaves the (two versions of) the second conception. I reject the latter version (that neither way of talking is correct) for the reasons suggested: this does indeed sound too skeptical (or non-cognitivist) for me. Our language is perfectly adequate for doing what it does. The former version -- that how things are depends in some way on how we talk -- gets the closest, I suppose. I do claim that in speaking normally we speak the literal truth; and I have also conceded to Clayton that his way of talking is not necessarily false, depending on what it is that one was thinking of doing in talking this way.
But that's not the same thing as saying, flatly, that "both are correct" (or, not so flatly but equally lamely, "... relative to our interest, or conceptual scheme, or whatever"). Clayton naturally denies that things change simply in being thought of differently for different purposes. Although he does not use these terms, in warning of the "untold havoc" that would result from this view, it sounds like he is defending "realism" against an "idealistic" threat to the idea of an independent reality. It is this that provoked me to deliver myself, neither for the first time nor the last, of my anti-dualistic rant (neither realism nor anti-realism is acceptable, etc.). I won't repeat it here, except once again to recommend Cavell, who I should point out doesn't necessarily put things in any way like I have here, or even in general for that matter; nor does Austin escape criticism. (Read Cavell.) On the other hand I can be provoked to talk this way in other contexts too (Nietzsche, Kant, Davidson, Hegel, or, well, anyone you like).
Naturally this settles nothing. All I've done, if that, is clear the space for a better account -- not of the bread, but of how we (should) speak of the bread, and of how we should speak of how we speak of the bread, and so on -- or in other words, of how we should speak. And of course for the most part we speak perfectly well. When we do philosophy we can lose sight of this, and fall into error. But that does not mean that doing philosophy -- moving out of the kitchen and into the study -- thereby amounts to falling into error, as some have claimed on behalf of (or in supposed response to) Austin and Wittgenstein alike. Nor does it mean that we are condemned to speaking not of the world but of language (another common complaint about linguistic philosophy). The two are inextricably linked -- that's the whole point, and one reason for bringing in Austin in the first place.
Let me finish with Austin's own words on the matter (from "A Plea for Excuses," p. 182):
When we examine what we should say when, what words we should use in what situations, we are looking again not merely at words (or 'meanings', whatever they may be) but also at the realities we use the words to talk about: we are using a sharpened awareness of words to sharpen our perception of, though not as the final arbiter of, the phenomena.
Again, though, this is not simply a defense against (accusations of) linguistic idealism. Taken together with what he actually says about our use of words -- what, that is, he takes this "sharpened awareness" to consist in -- the point cuts both ways (as does, again, Cavell's chapter title, in The Claim of Reason: "What a Thing Is (Called)"). That, after all, is what made this denial of idealism necessary in the first place. (Note, finally, the qualifier "though not as the final arbiter of" the phenomena; that's what rules out the dogmatically nihilist reading (Russell's, above) of his views.)