Meillassoux... but not all at once! (After Finitude Part 1)

I guess I've always been a Kantian. I remember being little and wondering if people saw color the same. I remember thinking that I don't even know if we do. That conclusion made me into a person who always thought that however we understood things didn't matter because the thing itself would always evoke in each that which it always does—being so completely and consistently itself. And we the same. So that no matter what we each see, we would always agree at least on the words we use to refer to the thing in itself—which is objective. We could hold proper conversations about it and never get confused. 

But what we agree on and write down is still how we write it down, say it, see it, record it—even when it's math.

But, too... in a certain sense... what the universe is for itself is part of us. We are part of it. If we could dematerialize—desubjectivize, dehumanize—our ideas, the remainder would be the universe (or in Meillassoux's terms, the ancestral). And that would mean the ancestral (and its ilk) is in our math at the very least. That conception of the ... factial, is it?... I can believe and am comfortable with. Inside the marks, in the form is the being (or perhaps the being is the form—I'm very partial to that). And I do accept that math is form. So I can in a very profound and personal way accept that thesis. Is math the one and only, though? Dunno. Maybe. 

This view is non-Kantian for many reasons, including in the sense that it gives more credence to reason. Reason isn't "just how humans think" and neither is math (or logic?). [Is it, rather, how the universe "thinks"?] It is true that we allow math to lead us, but it is also true that we allow the universe to delimit and revise our theorems. Boundary conditions and all that. It would be interesting to know whether Meillassoux is here considering pure or applied math, or if he even distinguishes the two. If applied, does that presuppose (hypostatize) pure? If pure, does that entail applied? (Which would entail the other?)

But I admit that, for me, the idea that mathematics is ontology is a seductive one.

A Grievance
Sometimes I can't quite figure out where I stand regarding this book. I love the math-as-being idea and that it challenges me as a thinker. But some of it really perplexes me too. For instance:

If one refuses to hypostatize the correlation , it is necessary to insist that the physical universe could not really have preceded the existence of humans, or at least of living creatures. 

See, here I can't figure out if he's attacking a thesis no one actually holds, or if he's saying to the correlationists: look guys, if you insist on defending the correlate, then if you follow it to its terminus, this is what you're actually defending. I've more or less accepted the correlationist perspective (though I have some trouble distinguishing Meillassoux's two types), but I've never actually believed this.

Like I said though, I'm partial to his general conception of math as ontology. (Badiou more or less held this perspective, too.) It's a damn cool idea equating primary properties with math (not altogether new) and then using that to identify the thing-in-itself. Wish I'd thought of it! ;)