4 ms·
> You can fit all existing mathematics within it... ... > It is constructive. This is interesting because it means real numbers "do not exist" as you can not
by AnimalMuppet 5y ago
> You can fit all existing mathematics within it...
...
> It is constructive. This is interesting because it means real numbers "do not exist" as you can not write them down.
Didn't you just contradict yourself? If real numbers do not exist within HTT, then you cannot fit all of existing mathematics in HTT.
- skulk 5y agoIt's not a contradiction. In the constructive sense, we don't know the real numbers except the ones that can be talked about using concepts that are more intrinsic to our reality such as rational numbers or circles. You can express the idea that a sequence converges to a limit in constructive mathematics (if not, I'd like to know why) and you can prove that that limit is not rational, nor is it algebraic. That is to say, real analysis works in a constructive setting. At least that's my math-enthusiast understanding of it. Traditional mathematics even asserts the existence of numbers whose concrete properties we can't even possibly know, such as Chaitin's constant: https://en.wikipedia.org/wiki/Chaitin%27s_constant https://en.wikipedia.org/wiki/Chaitin%27s_constant My understanding is that these are the kinds of numbers you lose out on when you switch to being a constructivist. I guess you could say that means you can't fit "all of existing mathematics" in it, but to me that feels a bit vacuous.
- AnimalMuppet 5y agoWell, I guess I think the same about the constructivist position. > In the constructive sense, we don't know the real numbers except the ones that can be talked about using concepts that are more intrinsic to our reality such as rational numbers or circles. I accept that as being true. But then, to say that you can fit "all of existing mathematics" feels like you're saying that you can fit everything that fits, and everything that doesn't fit is defined to not be "real" in some sense. That is, it feels circular - like you're so far down the rabbit hole that you can't remember that there's a world above ground that also exists and has validity.
- skulk 5y agoI guess strictly, it is a contradiction because indeed "all of existing mathematics" cannot be derived purely constructively. The way I choose to see it, however, is constructivism is the activity of seeing how far you can go with the least possible axiomatic foundation. If, at some point, you need something like the Law of Excluded Middle, you just assert the specific instance that you need as an axiom (and not as a structural rule). In this sense, then all of mathematics is valid from a constructivist viewpoint, what changes is that now you care about where you assert certain axioms.
- R0b0t1 5y agoYou don't lose out on the fields you mentioned when you switch to constructivism. Your assertions become more precise. If anything is lost the mainstream thinking is it was not consistent to begin with.
- R0b0t1 5y agoIt is more precise to say you can fit all of existing mathematics that you can actually check. A number whose only classification is that it is real is a nonphysical entity. In constructive systems it can not be reasoned about because it can not exist. Something very close to real numbers exists, but situationally. Pi exists because I can describe a procedure to construct it, same for Euler's number.
- myWindoonn 5y agoHoTT describes any infinity-topos. When folks say "all of mathematics can be rederived constructively", they mean that all of constructive mathematics can be described relative to any topos. Classical mathematics is retrieved either by choosing a Boolean topos, or by developing Boolean algebras inside topos theory. It is a standard theorem that all constructive results are classical, and that all classical results are either constructively true or not-false. A not-false theorem can't be proven true, but it definitely would lead to a contradiction if it were false. To get a taste of what's constructive, imagine taking a subset of the finite set {0,1}. We construct the subset by enumerating its elements; a computer program might print out elements one-by-one. So, if we start a program and it prints 0, 1, ...; we know that the subset contains all of the elements and we're done. But if we start a program and it prints 0, 0, 0, ...; only printing zeros, then we have to prove that it halts in order to prove that the subset is finite. Constructively, subsets of finite sets aren't necessarily finite.
- zozbot234 5y agoStrictly speaking, there are a variety of embeddings ("translations") of classical into constructive logic and while different in their minute details, they all share the feature that non-constructive theorems are embedded as negative statements. Conversely, a proof of a negative statement in constructive logic can use proof by contradiction, hence the full power of a "classical" proof. (NOT NOT NOT p) <-> (NOT p) is a theorem in constructive logic, even though (NOT NOT p) -> p isn't.