I think you are mistaken. The idea that math proofs are a social construct relates to, in my view, much deeper ideas than you seem to think [1].
It is not just that convincing other mathematicians that a proof is correct is a social process, but also that the reasoning on which any proof relies, even if it seems unassailable, even if built into an automated checker, is still a product of the human mind. Usually there is a level of logic that can challenge even what seems so basic as to be fool-proof.
Take the proof that the square root of 2 is irrational. The proof relies on a contradiction that arises if one assumes the root is rational, but one can imagine a logic system where such a contradiction does not imply that the original assumption is false. How possible, you say? It's all math, where one is allowed any starting assumptions, and works out the implications of those.
But, there is something deeply satisfying about thinking that contradictions are (or should be) impossible in our universe, and so this "proof" seems solid.
I think you are mistaken.
I used to be very troubled by the notion that no single set of axioms really can be agreed on to do mathematics, but I have been convinced finally that the truth of the matter is a very subtle point; that the truth of mathematics is absolute; it merely is not finitely-axiomatisable.
With a sufficiently weak proof system, clearly we can conceive of a system where the irrationality of the square root of 2 is not provably true, but no consistent proof system can prove that the square root of 2 is rational. Certain mathematical constructs, indeed most(*) of known mathematics, that which is constructible by constructivist methods, are irrefutably there in any consistent mathematical universe, and in that sense, true.
Yet, no single axiom system can encompass all mathematical truth, as is well known from Goedel's theorems, but neither does that mean that the set of axioms to be worked with can be arbitrarily chosen. The chosen set of axioms must be consistent. The question, then, is whether consistency of axioms can possibly be an objective fact; and even though for any sufficiently strong set of axioms, its consistency cannot be proven in of itself, it consistency can in fact be objectively established - objectively established, but not finitely established.
My evidence for this perspective is Scott Aaronson's construction of a Turing-Machine encoding of the ZFC axioms. What the construction of this encoding implies, is that the revelation of the uncomputable busy-beaver (BB) function for value 8000, which is a finite, well defined, and an objectively irrefutable, albeit unthinkably massive, number, constitutes a proof of the consistency of ZFC axioms. A similar procedure I believe can be applied to any set of axioms that one wishes to work with.
The part where the magic occurs, I believe, is in the uncomputable nature of the BB function, by which it is possible to finally and objectively establish consistency of sets of axioms. Uncomputability amounts to the acknowledgement that though something may always be well-defined, there is no finite method to encompass its values; that is the nature I now take of mathematics as well.
So the very foundations of mathematics, the logic we use behind our proofs, is inherently a social construct that arose out of Greek philosophy as it was adapted in the west.
I think you need to separate the process of developing mathematics, from the self-consistency and validity of the logical argument or mathematical structure itself. The process is social but the validity and mathematical structure itself is not.
Computer science has led to an explosion in different logics. We've never had more logics/formal systems than we do today, but whether any given formal system actually is consistent is not dependent on consensus, it is a fact, either true or false, completely independent of consensus.
The definition of "consistent" seems completely entangled with a given social group's ideas of "rational". You might imply that our word for it hints at a Platonic ideal of "consistent", but if that's true, then you're caught in an infinite cascade of which nuances of meaning between the Platonic Ideal and our concrete reality are actually reflections of the truth or corruption of it.
Many do not seems to see the fundamental issue at play here, and another way to think of them is what you have hinted at: the role of language. There is no objective way to nail down the meaning of words, like "consistent", "proof", "equal", etc.
Suppose one wanted a maximally rigorous definition of "equal". Does it mean two things that cause people to think of the same thing when they are mentioned? Does it mean two things that occupy the same position in space at all times? It is actually a difficult concept to define rigorously.
This is not to deny that there is an objective reality. But that reality is highly contextual and multi-faceted. We cannot be 100% exact in defining that reality using language (even a math language), and this is where the social nature of that reality becomes apparent.
The role of proofs are in creating, as far as possible, as rigorous a shared context for the reality being described.
1) x = x
2) x = y => P[x] => P[y]
In my opinion, there is a mathematical reality, which is shared by everyone, even by those who don't believe in it :-) For example, a logical system exists in that reality, and you can either derive a theorem in that system or not in this reality. I don't think it is possible that there is a third possibility. I don't think it is possible that I have a different reality from you in that respect. This reality is not socially constructed, it just is. Intuitionism will tell you that because you don't know if a certain theorem is derivable, it is in some sort of hybrid state until we know for sure via a intuitionistic proof or a counter example. I think that is bullocks. Either there is a proof or not. Either there is a counter example, or not.
Beyond that, extending this mathematical reality, there is a wider, not as easily accessible reality. We can try to understand that reality by modelling it via certain assumptions, and then applying our mathematical reality to those assumptions. I believe the mathematical conclusions we draw from this will be real to the extent that the assumptions are true; but of course you cannot ever be sure about those assumptions, and so you cannot be sure about the conclusions. But if you notice that your conclusions do not hold, you need to challenge your assumptions, not your mathematics.
In short what it usually means (the exact meaning depends on the model under consideration) is that there is a binary operation "=", such that "x = x" is a theorem, that is evaluating "x = x" will evaluate to "true" for any object x in the mathematical universe. Furthermore, for any unary proper operator "P", and any two objects x and y of the mathematical universe, the expression "(x = y) => (P[x] = P[y])" will also evaluate to "true". Here "=>" is another binary operation called implication, which has some special properties outlined in the link. P[x] denotes the application of the operator P to the object x.
Edit: Oh, forgot to add the third axiom for equality:
3) A => (A = true)
What this means is that for any object A of the mathematical universe, if you evaluate "A => (A = true)", you obtain the value "true".
I think you're conflating universality and objectivity. Those terms you list all have objective definitions, but the specific characteristics they have in any given logic may differ. That means they are not universal, but that doesn't make them non-objective. Objective typically means "mind independent".
Your example of equality already demonstrates you understand equality's objective definition: you implicitly operate on the notion that "equality" means some form of equivalence, some ability to substitute B for C in a specific context that results in no observable/expressible change. That is an informal but objective understanding of equality.
What you're recognizing is that equality can have different logical properties in different contexts, where "context" can be understood as the formal language we're using, ie. it's not universal. But it's role in any given logic is always the same and not dependent on the provers mind state or his surrounding culture, ie. it is objective.
Godel showed that there is no such thing as a universal logic in our current approach to formal systems, but that didn't suddenly make logic non-objective. It simply means that there is no Ur-logic that can subsume all other logics (which is why most assert that Godel ended Hilbert's program).
So what logics a culture or species may use or find interesting, and the process by which they explore these systems are socially contextual, but the structures themselves and their internal consistency is not socially constructed. A culture can certainly believe a formal system they use to be logically consistent, but that's no more interesting a statement than that some cultures believed that Thor caused lightning. In other words, they could just be wrong about the consistency of their arguments.
Could you elaborate on that? Any references?
This is why Hilbert's program to finitely axiomatize mathematics can't be completed. The "escape hatch" here is simply that not all propositions are actually interesting, so finite axiomatizations are still very useful, and we can extend the axiomatic basis as needed given satisfactory justification. This last part is the only place where social consensus sometimes comes into play (continuum hypothesis, etc).
Edit: there is another possible escape hatch that hasn't been fully explored IMO, and that's some variant of finitism. All these impossibility proofs depend on infinite structures to derive incompleteness or contradiction, but if infinite structures are not expressible...
This is different what I would understand under an Ur-Logic: Just a logical system that can express anything you want, given you are free to add axioms. Obviously there are a few choices for that.
Could you elaborate on what such a formalism where infinite structures are not expressible might look like? It sounds an intriguing notion, though I'm unable to quite get at it.
A Defense of Strict Finitism, http://www.jeanpaulvanbendegem.be/strict%20finitism.pdf
> This is not to deny that there is an objective reality...
So I do agree that object reality exists, contrary to what you imply in your response.
However, there is no word in human language that has an objective definition. If you think this is so, maybe you have not thought or read enough about the subject.
On the subject of equality, you said: > Your example of equality already demonstrates you understand equality's objective definition: you implicitly operate on the notion that "equality" means some form of equivalence, some ability to substitute B for C in a specific context that results in no observable/expressible change.
You're making my point. Consider the vagueness of what you have written: "...some form of equivalence, some ability to substitute B for C in a specific context that results in no observable/expressible change." That doesn't sound very rigorous to me, and in fact I could argue this is not what I understand by "equality" at all.
Now you could make the point the definition can be made rigorous for a very specific context, but this is still not 100% possible. To show this, you can attempt to give me a specific definition of equality of a specific context, and I will show that it still admits of multiple possible meanings.
Even in our commonly used logic system, what does false actually mean?
false = (∀x. x)
So in this case, it derives its meaning from whatever the meaning of the ∀ operator is. Its meaning is the result of applying the ∀ operator to the identity operation.Nobody is questioning that there are different possibilities for logical systems, and different possible semantics for them. The choice of logical system is up to you, and ultimately, part of your assumptions. A proof assistant uses a fixed logical system, and your mathematics needs to fit into this system to be expressible in the proof assistant. Apart from my favourite logic, there are several well-known logical systems commonly assumed to be able to represent all of mathematics more or less faithfully: first-order logic, simply-typed higher-order logic, and dependent type theory.
What I like about my favourite logic is that I believe that all other logics, even quantum logics, can be expressed in it in a straightforward way. This has the advantage that once you expressed your logic in this way, you get the meaning of it for free. Of course, you still need to make sure that this is the intended meaning.
That is kind of obvious, isn't it? But some people need more convincing than others :-)
Personally, I don't think intuitionism makes much sense on a fundamental level beyond being an elegant mechanism in certain situations. If you tell me that it is false that A is false, then certainly A is true. Anything else is really mystic mamboojamboo and not clear thinking. But that's just my opinion, and I am not gonna force it onto you, because this is not something I can prove in a proof-assistant, but just an opinion, and as such a social construct.
I am not really dogmatic about this. You might be able to use intuitionism for things that cannot be done via classical reasoning, for example extract programs from a proof. I have yet to see an example where it is not simpler and more straightforward to just prove an executable program to be equivalent to the specification using classical reasoning.
My point of view is the following: If you are not able to (eventually) make your case within a proof assistant, then what you are trying to tell me is not math.
The truth or falsity of that statement in a model is independent of who observes that model.
The only “social construct” you’ve described is which model to use by default — literally, a notation convention, nothing semantic.
> But, there is something deeply satisfying about thinking that contradictions are (or should be) impossible in our universe, and so this "proof" seems solid.
Our universe either does or doesn’t, depending on which model best represents it — we’re only debating what assumption to make about an unknown.
But I guess the point is that almost no-one does this. I would guess that if someone tried to formally verify every published paper out there (or even every textbook), they would uncover a large number of gaps. Very few of those gaps would be unfixable, and very few results would turn out to be incorrect, but the possibility exists.
Apart from that, with the help of computers, it can be made absolutely precise and clear which statements follow from which axioms, and in that sense it is not a social construct at all. It also is much less cumbersome than it used to be, and will continue to improve quickly.
I can sit down and prove something using a tool like Isabelle, and I will be as sure of its "truth" as I can possibly be, and it really doesn't matter what other people, mathematicians or not, think about it. That's the beauty of it.
Of course, you could say my belief in Isabelle is also a social construct. Except it is not, I know exactly how Isabelle works. There could be issues with Isabelle, but these issues adding up to make my proof wrong are very unlikely, especially in addition to my independent understanding of the proof.
But of course, it is much nicer if others can see the same truth that I do, and for this, computer-assisted proof is actually great, because it allows to understand and trust in the high-level structure of a proof without having to verify every little gritty low-level detail.