Devereux's Argument

Dan Devereux's argument about Z 3 is as follows. Consider the following claims:

A. (The Hypokeimenon Criterion) A substance is: that of which other things are predicated but which is not predicated of anything else.
B. (The 'New View') A concrete substance is predicable of its underlying matter.
C. Matter is the only substance.

According to Devereux, Z 3 aims to show that, on the assumption of B., then A. leads to C. But because C. is objectionable to Aristotle, he therefore 'demotes' the Hypokeimenon Criterion by proposing, as more important, the criteria that a substance is a 'this' and is 'separable'. Matter may satisfy the Hypokeimenon Criterion, but it fails to satisfy these other two criteria.

Devereux regards the passage we have been considering as an important step in this argument, and especially because it reveals Aristotle's commitment to B. Here again is his translation, but this time I highlight terms that he thinks are especially important:

Matter is that which in itself is neither a particular thing nor a quantity ... for there is something of which each of these is predicated, but whose being is different from the being of any of them; for while the other things are predicated of substance, this is predicated of the matter ... nor is it the negations of these, for these will belong [to it] incidentally.

According to Devereux, Aristotle's argument here is:

1. Things in the various categories of accidents ('other things') are predicated of substance.
2. Substance ('this') is predicable of the matter.
3. If X is predicable of Y, then whatever is predicated of X is predicated of Y.
   
4. Thus, things in the various categories of accidents ('each of these') are predicated of the underlying matter.
5. Matter is distinct from whatever is predicated of it.
6. Thus, matter is distinct from anything in any of the categories.
   

Devereux argues that 'substance' in 1. must indicate a concrete substance, because that's Aristotle's standard view. But 'substance' in 2. must indicate the same thing, otherwise Aristotle can't get the result that everything is predicated of the matter, which is what he wants. As Devereux puts it:

Aristotle needs to say that it is the 'same thing' that both underlies the attributes and is predicated of the matter. He makes the claim ...that there is something of which the items in the different categories are predicated, and then supports this claim ...by stating that the attributes are predicated of substance and 'this' is predicated of matter. Assuming as a tacit premise that if x is predicable of y, then whatever is predicated of x is predicated of y, we get the desired conclusion that both the attributes and the substance of which these are predicated are predicable of one thing, the matter. (176)

It's an engaging argument. But are you convinced? What, if anything, do you think is questionable about it? I see a couple of serious difficulties in it, but I wonder how what you think.