thanks for the comment. You are indeed right that Frege does not hold identity to concern only signs.
I’d have to go back and check to be sure, but I would be surprised, though, if Badiou really adduces Frege as support for such a claim, since Badiou at this point clearly wants to give something like a Hilbertian and thus indeed formalist correction to Miller, in order to acheive a theory of number that aligns with an Althusserian conception of science rather than a psychoanalytical one. So I believe that the “provisio” mentioned above, i.e., that “the concept of identity holds only for marks”, is Badiou’s own provisio, not something he ascribes neither to Frege nor Miller.

As far as I’m concerned, though, there’s no sense in speaking of “past” or “future” with respect to Meillassoux’ time…. doing so seems inconsistent with what he says about his own “mad time”, and leaves him wide open to the kind of objections Hägglund offers. The “physical past” can change just as much as the “physical future”, i.e., eruption ex nihilo might as well happen in 1900 as in 2100, so there’s no temporal modality proper separate from the alethic modality of necessary contingency. In the terms of contemporary philosophy of time, both the “A-series” and the “B-series” of time (whichever of them actually holds) are mere contingent facts, subordinate to the absolute alethic time.

BUT, I don’t think Meillassoux is at all clear with regards to this, nor with regards to the consequences this actually has. How can we think the absolute power of this time without thinking of it in terms of a “before” and an “after”? What does it mean to say that a physical law, for instance, can be abolished if we are not to think of it in terms of it being a fact “before” and the ceasing to be a fact “afterwards”? I’m wondering whether the problem might be to define what existence means within this framework….