Do Determinables Exist? (2)

In a previous post, I expressed scepticism about a recent argument by Gillett and Rives to the effect that determinable properties don't exist: only determinate properties do. Yesterday, we discussed Gillett and Rives' paper in the NEH Summer Seminar on Mind and Metaphysics. Curiously, John Heil (who probably doesn't read my blog) expressed the same criticism that I had made in my post (though unlike me, Heil has little sympathy for Shoemaker's "subset view").

I'll take this as an opportunity to look at a reply kindly sent to me by Brad Rives:

"Thanks for your comments on the paper. I'm not sure that I understand your response to the parsimony argument. You say: "Determinables exist because the causal powers that constitute them exist; it's just that there are other relevant causal powers beyond them. Hence, there is neither double-counting of causal powers nor causal overdetermination."
I don't see how this follows. Suppose determinables are constituted by powers that are subsets of those that constitute their determinates. It's true that on this view particulars will have the powers that individuate determinable, but the point of the simplicity argument is that the determinables won't be contributing any causal powers to particulars that aren't contributed by some or other determinate. We can thus account for ALL the powers of particulars simply by attributing determinates to them, whereas this isn't true of determinables. Assuming we should only posit those properties needed to account for the causal powers of particualrs, the argument concludes that we shouldn't posit determinables. If you suppose that both determinate and determinables are instantiated, it's hard to see how there won't be overdetermination of powers. Since some of the powers that individuate a determinate also individuate the determinable, those powers will be contributed by two distinct properties, which just is overdetermination. Convinced?"

No.

The worry about overdetermination arises only if a property is something over and above the powers that it "contributes" and most importantly, the relation bewteen a determinable property and its determinates is not analogous to the part-whole relation that holds between the powers they "contribute".

If all there is to properties is powers and the powers of determinables are a subset of the powers of determinates, then there is no double counting and no overdetermination. But even if properties are something more than the powers they "contribute," there is still no overdetermination provided that determinables stand in a part-whole relation to their determinates.