[HN Gopher] What do we mean by "the foundations of mathematics"?
       ___________________________________________________________________
        
       What do we mean by "the foundations of mathematics"?
        
       Author : JoelMcCracken
       Score  : 54 points
       Date   : 2023-11-01 17:52 UTC (5 hours ago)
        
 (HTM) web link (lawrencecpaulson.github.io)
 (TXT) w3m dump (lawrencecpaulson.github.io)
        
       | daoboy wrote:
       | A good opportunity to break out my favorite Bertrand Russell
       | quote:
       | 
       | "Everything is vague to a degree you do not realize till you have
       | tried to make it precise."
        
       | hiAndrewQuinn wrote:
       | Category theory is a pretty good one. But I think a lot of people
       | would get a surprising amount of value from reading Halmos's slim
       | _Naive Set Theory_, to see just how much work we can create using
       | sets alone.
        
       | btilly wrote:
       | We can't define "foundations of mathematics" without defining
       | mathematics. My favorite stab at this is from Thurston's On Proof
       | and Progress in Mathematics. Which may be read in full at
       | https://www.math.toronto.edu/mccann/199/thurston.pdf: Could the
       | difficulty in giving a good direct definition of mathematics be
       | an essential one, indicating that mathematics has an essential
       | recursive quality? Along these lines we might say that
       | mathematics is the smallest subject satisfying the following:
       | 
       | - Mathematics includes the natural numbers and plane and solid
       | geometry.
       | 
       | - Mathematics is that which mathematicians study.
       | 
       | - Mathematicians are those humans who advance human understanding
       | of mathematics.
       | 
       | In other words, as mathematics advances, we incorporate it into
       | our thinking. As our thinking becomes more sophisticated, we
       | generate new mathematical concepts and new mathematical
       | structures: the subject matter of mathematics changes to reflect
       | how we think.
       | 
       | The foundations of mathematics then refers to the essential core
       | that we have based our mathematical thinking on. Which is
       | something that most of us only wish to think about when our
       | thinking has run into problems and we need a better foundation.
        
         | vlovich123 wrote:
         | Is mathematics just the study of all possible languages that
         | are constructed from strict internally consistent rules?
        
           | saalweachter wrote:
           | Mathematics is the study of things you can convince people
           | are true without physical evidence.
        
           | btilly wrote:
           | No, because some possible languages are far more interesting
           | than others.
        
             | bmacho wrote:
             | I don't think that's a problem: if there are uninteresting
             | languages, then you must _study_ them, then _work hard_ to
             | _prove_ that they are uninteresting first. That is,
             | mathematics indeed does study uninteresting languages.
             | 
             | .. mathematics is the same for me as GP said: you have a
             | set of symbols, a set of rules, now derive true statements.
             | (Or at least this is a good view, but other views can be
             | also good.)
        
         | Someone wrote:
         | > Mathematics includes the natural numbers and plane and solid
         | geometry.
         | 
         | That, to me, feels similar to Euclid's fifth postulate, in the
         | sense that it isn't something you'd want to include in a
         | definition.
         | 
         | Maybe, the similarity doesn't end there, and we have multiple
         | mathematics, just as pure mathematicians don't talk about
         | _algebra_ , but about _an algebra_ , with some of them being
         | more interesting than others, and it not always being easy to
         | show that two algebras are different from each other.
         | 
         | For example, we have mathematics without the law of the
         | excluded middle and with it, ones without the axiom of choice
         | and with it, etc.
        
       | westurner wrote:
       | https://en.wikipedia.org/wiki/Zermelo%E2%80%93Fraenkel_set_t... :
       | 
       | > _Today, Zermelo-Fraenkel set theory [ZFC], with the
       | historically controversial axiom of choice (AC) included, is the
       | standard form of axiomatic set theory and as such is the most
       | common_ foundation of mathematics.
       | 
       | Foundation of mathematics:
       | https://en.wikipedia.org/wiki/Foundations_of_mathematics
       | 
       | Implementation of mathematics in set theory:
       | https://en.wikipedia.org/wiki/Implementation_of_mathematics_... :
       | 
       | > _The implementation of a number of basic mathematical concepts
       | is carried out in parallel in ZFC (the dominant set theory) and
       | in NFU, the version of Quine 's New Foundations shown to be
       | consistent by R. B. Jensen in 1969 (here understood to include at
       | least axioms of Infinity and Choice)._
       | 
       | > _What is said here applies also to two families of set
       | theories: on the one hand, a range of theories including Zermelo
       | set theory near the lower end of the scale and going up to ZFC
       | extended with large cardinal hypotheses such as "there is a
       | measurable cardinal"; and on the other hand a hierarchy of
       | extensions of NFU which is surveyed in the New Foundations
       | article. These correspond to different general views of what the
       | set-theoretical universe is like_
       | 
       | IEEE-754 specifies that float64s have +-infinity and specify
       | ZeroDivisionError. Symbolic CAS with MPFR needn't be limited to
       | float64s.
       | 
       | HoTT in CoQ: Coq-HoTT: https://github.com/HoTT/Coq-HoTT
       | 
       | What is the relation between Coq-HoTT & Homotopy Type Theory and
       | Set Theory with e.g. ZFC?
       | 
       | Homotopy Type Theory:
       | https://en.wikipedia.org/wiki/Homotopy_type_theory :
       | 
       | > _" Cartesian Cubical Computational Type Theory: Constructive
       | Reasoning with Paths and Equalities" (2018)_
       | 
       | https://scholar.google.com/scholar?cites=9763697449338277760...
       | and sorted by date:
       | https://scholar.google.com/scholar?hl=en&as_sdt=5,43&sciodt=...
       | 
       | What does Mathlib have for SetTheory, ZFC, NFU, and HoTT?
       | 
       | leanprover-community/mathlib4// Mathlib/SetTheory:
       | https://github.com/leanprover-community/mathlib4/tree/master...
        
       | nyc111 wrote:
       | Mathematics is built on definitions. There are definitions on the
       | foundations of mathematics. Axioms are definitions that are kept
       | constant by consensus.
       | 
       | Euclid starts with definitions. If you don't like Euclid's
       | definitions you make your own definitions. You create your own
       | geometry etc.
        
       | smokel wrote:
       | For those who like to read biographies of the brilliant minds,
       | instead of the actual mathematics they produced, I can highly
       | recommend "Journey to the Edge of Reason: The Life of Kurt Godel"
       | (2021) by Stephen Budiansky, which is nice.
        
       | wisnesky wrote:
       | A foundation for mathematics is any formalism sufficient to prove
       | the results typically taken as axioms in practical mathematics.
       | For example, in ZFC you can define numbers as sets in many
       | different ways and prove that 1+1=2 for each of them - other
       | foundations include higher order logic, topos theory, other set
       | theories, etc.
        
         | wubrr wrote:
         | Is it also possible to define set theory in terms of
         | arithmetic?
        
           | wisnesky wrote:
           | Yes, at least with a strong enough arithmetic (such as
           | Peano's), but that is usually more complicated; for example,
           | you might have to create a Godel numbering or some other
           | "deep embedding" to represent each set as a number. There's
           | also so-called "reverse mathematics", which tries to
           | determine the weakest axiom system capable of establishing a
           | particular result.
        
             | wubrr wrote:
             | Interesting, thanks!
        
           | Kranar wrote:
           | You can certainly define some set theories using arithmetic
           | but arithmetic as formalized by first-order Peano arithmetic
           | can not define a set theory powerful enough to prove results
           | about uncountable sets, such as large cardinal axioms or even
           | properties of real numbers.
        
             | wubrr wrote:
             | Any links/examples of why this isn't possible? My thinking
             | is - if all existing set theory and set theoretic proofs
             | are expressed in finite set-theoretic
             | statements/expressions - can't we 'encode' these
             | expressions within the system of arithmetic?
        
               | Kranar wrote:
               | Do I have links off the top of my head that directly
               | address this, no unfortunately I do not. I do have links
               | to different pieces that you can use to understand this,
               | however. First is the notion of the consistency strength
               | of a formal system which basically is the ability of one
               | formal system to prove the consistency of another formal
               | system [1].
               | 
               | ZFC is strictly stronger than PA in the sense in that ZFC
               | can prove the consistency of PA [2].
               | 
               | From this, however, it follows that PA can not prove the
               | consistency of ZFC. If PA could prove the consistency of
               | ZFC then ZFC would be able to prove the consistency of
               | itself, which we know from Godel's incompleteness theorem
               | would mean that ZFC is inconsistent [3].
               | 
               | So this gives one concrete proof that ZFC can produce
               | that PA can not produce, namely the consistency of PA
               | itself.
               | 
               | [1] https://en.wikipedia.org/wiki/Equiconsistency#Consist
               | ency_st...
               | 
               | [2] https://mathoverflow.net/questions/66121/is-pa-
               | consistent-do...
               | 
               | [3] https://en.wikipedia.org/wiki/G%C3%B6del%27s_incomple
               | teness_...
        
         | nitwit005 wrote:
         | You're getting at the problem with saying "foundation". If you
         | can reach the same conclusions from a large number of starting
         | points, there isn't any special set of ideas on which
         | everything rests.
        
           | epgui wrote:
           | Why does the foundation need to be unique? It seems intuitive
           | to me that they must not be unique.
        
       | tromp wrote:
       | Jeremy Avigad made available a 54-page draft chapter on
       | mathematical logic and foundations [1]. Quoting from the
       | abstract:
       | 
       | > The formal foundation of a proof assistant is the formal
       | deductive system with respect to which the correctness of the
       | proof assistant's implementation is judged. The foundation
       | specifies a language for defining objects and saying things about
       | them, and it determines the means that can be used to prove that
       | such statements are true. This chapter presents some general
       | background in logic that is helpful for thinking and reasoning
       | about formal systems, and it describes the three main families of
       | foundations that are commonly used in proof assistants today.
       | 
       | [1] https://arxiv.org/abs/2009.09541
        
         | 1equalsequals1 wrote:
         | Thank you friend
        
       | cjfd wrote:
       | I have a project to write a proof assistant that is as simple as
       | possible while still usable. A kind of turing tar pit but for
       | proofs instead of programs. Some early attempts did not quite
       | work out. I try to get it off the ground using the calculus of
       | constructions (not the calculus of inductive constructions)
       | together with a small set of axioms. As axioms I aim, at the
       | moment, to use a hilbert epsilon operator to contstruct objects
       | from theorems that state their existence, proof irrelevance and
       | an infinity axiom that states that an infinite set exists. Also I
       | do use an infinite hierarchy of Types a bit like the universes of
       | coq. I am still debating with myself whether I also should
       | provide the luxury of not having to specify universe levels and
       | just keep constraints between them like in coq.
       | 
       | The article proposes the calculus of inductive constructions but
       | I think that is just too big and complicated to be a good
       | candidate for a foundation of mathematics. I think it makes more
       | sense to use a more concise foundation and maybe attempt to prove
       | that the calculus of inductive constructions is a conservative
       | extension of that.
        
         | bvssvni wrote:
         | Is your goal to have as few axioms as possible, or as few
         | syntactic constructions as possible?
        
         | zarzavat wrote:
         | The calculus of inductive constructions is not very big? It
         | seems simpler than ZFC to me. The version that Coq uses is
         | quite large, but that's the difference between theory and
         | practice: Coq is designed to be maximally usable not minimally
         | simple.
        
         | lacker wrote:
         | Once you have an infinite hierarchy of types, including
         | universe levels, can the proof assistant really be "as simple
         | as possible"?
         | 
         | I feel like a simple-as-possible proof assistant would be more
         | like, proving things about programs written in a minimal Lisp
         | dialect, where the proof validator is itself written in that
         | Lisp dialect.
         | 
         | That said, I personally would rather have an extremely usable
         | proof assistant, rather than a simple-as-possible one. In many
         | ways those things point in opposite directions; consider the
         | usability and simplicity of Python vs the lambda calculus.
        
           | blueberry87 wrote:
           | What are your other options? A type-based proof system
           | implies dependent types, implies universe levels if you want
           | it to be consistent.
        
             | lacker wrote:
             | You would have to not use a type-based proof system. I
             | think those are not going to be the "as simple as possible"
             | ones. Consider metamath for example.
        
       | aaimnr wrote:
       | Reminds me of Wilfrid Sellars: "The aim of philosophy, abstractly
       | formulated, is to understand how things in the broadest possible
       | sense of the term hang together in the broadest possible sense of
       | the term".
       | 
       | "Hanging together" doesn't impose as many assumptions as "having
       | specific foundation". It's a very distinct philosophical stance
       | to assume that any kind of human knowledge has a specific
       | foundation, and a naive one at that.
        
       | wannabebarista wrote:
       | For those interested in recent systematic discussions of
       | foundations, I find Penelope Maddy's recent work compelling. For
       | instance, see the paper "What do we want a foundation to do?" [0]
       | 
       | [0]
       | https://sites.socsci.uci.edu/~pjmaddy/bio/What%20do%20we%20w...
        
       | practal wrote:
       | > That the solution could lie in something as simple as Zermelo's
       | separation axiom and the conception of the cumulative hierarchy
       | of sets was seemingly not anticipated. It was a miracle.
       | 
       | If you are a Platonist (I am), then it is not really a miracle.
       | Mathematical objects are real, and logic is a tool to study them.
       | The paradoxes can only exist, because math is real: a paradox
       | shows that something is not real, and if there were no real
       | mathematical objects to begin with, paradoxes would be
       | meaningless. As it stands, a paradox is a tool to distinguish
       | reality from that which is not real.
       | 
       | Have fun making any kind of sense if you are a formalist or
       | intuitionist (hint: you can't).
       | 
       | It is indeed cool that something as simple as set theory covers
       | so much. But it doesn't cover everything. The Banach-Tarski
       | Paradox shows that sets are not what we imagine them to be (but
       | they are still real): we need to add the property of
       | measurability, which acts then as an additional axiom, to get
       | closer to that. And in order to solve BB(748), ZFC is not enough
       | [0, theorem 7]. So, one set theory is not enough. Depending on
       | which properties we ascribe to sets, or in other words, which
       | axioms we add to set theory, we also get different set theories.
       | 
       | > I give up.
       | 
       | Set theory is cool, but it is not special. In the end it
       | describes a fragment of the (real) mathematical universe, but
       | there is more to this universe than just set theory. And
       | depending on which properties you pick for your sets, you get
       | different fragments.
       | 
       | It is interesting that Larry says that Automath is not a
       | Foundation, and that neither is Isabelle, which he created. Maybe
       | I am putting words in his mouth here, but that is how I interpret
       | his "Isabelle/Pure is in the same spirit". I think he is right.
       | BUT, Isabelle is SO close to a foundation. All you have to do is
       | to drop the types :-D. All of them. That's what I did, and you
       | then basically get Abstraction Logic [1]. Abstraction Logic makes
       | no assumptions whatsoever about your mathematical universe,
       | except that it is non-empty, and it better have more than 1
       | object in it. It doesn't assume that sets exist, or types, or
       | anything else. It doesn't force you to define natural numbers or
       | other mathematical objects as sets. You can define any operator
       | you want, and with each choice of axioms you are exposed to the
       | risk of doing nonsense and creating paradoxes. And therefore I
       | believe it is the right foundation for mathematics, and I think
       | it is the only right one. Yes, there is a single logical
       | foundation. Sounds too simple? So did set theory. Yes, I think
       | Abstraction Logic is a miracle, but on the other hand, it simply
       | manifests logic as it is.
       | 
       | [0] https://www.scottaaronson.com/papers/bb.pdf, page 12, theorem
       | 7
       | 
       | [1] http://abstractionlogic.com
        
         | drdeca wrote:
         | > Abstraction Logic makes no assumptions whatsoever about your
         | mathematical universe, except that it is non-empty, and it
         | better have more than 1 object in it. It doesn't assume that
         | sets exist, or types, or anything else.
         | 
         | This sounds kind of similar to something I've thought about,
         | but which makes approximately one fewer assumption, while also
         | remaining closer to set theory.
         | 
         | Namely, the theory of first order logic (without equality),
         | with one binary relation symbol, and no axioms (unless you
         | count the laws of inference of FOL, which you shouldn't).
         | 
         | If you name this binary relation symbol \in_0 , and are willing
         | to use a "definition schema" to define, for each natural number
         | in your meta-theory, binary relations \in_n and =_n (any
         | sentence using these \in_n and =_n can be expanded out to a
         | longer sentence in FOL using just quantifiers and \in_0), then,
         | in the meta theory, you can prove a fair number of nice things
         | about "for all natural numbers n, it is provable in the object
         | theory that [some statement which uses relations \in_n and/or
         | =_n]".
         | 
         | For example, you can show that, for all n, you can show in the
         | object-theory that =_n is transitive and reflexive, and that if
         | x =_n y, then x =_{n+1} y as well, and that for all A, B, if A
         | \subseteq_n B , then A \subseteq_{n+1} B, and you can define
         | notions of one object being a power set of another object at
         | level n (and any two power sets of an object at level n, will
         | be equal at level n), and can show that, for all n, it is
         | provable in the object theory that if A \subseteq_n B , and if
         | PA is a powerset of A at level n, and PB is a powerset of B at
         | level n, then PA \subseteq_n PB , and all sorts of similar
         | things.
         | 
         | I'm a little skeptical of the "something between 1st order
         | logic and 2nd order logic" claim, but I haven't looked closely.
         | 
         | Are you familiar with the metamath project? aiui, metamath is
         | based in string rewrite rules, and is therefore fully general
         | in the way it sounds like you are going for?
        
       | lacker wrote:
       | 100 years ago everyone was worrying about whether the foundations
       | of mathematics were consistent. We figured out some details that
       | we had been taking for granted, like the axiom of choice, how to
       | prevent Russell's paradox, and the inevitable difference between
       | "true" and "provable" (ie Godel's incompleteness). Now that part
       | is pretty much solved, we are fairly confident that the
       | foundations of mathematics are consistent.
       | 
       | IMHO the problem now is that the foundations of mathematics are
       | _impractical_. Most mathematicians don 't actually do mathematics
       | that can be tied back to the foundational axioms. It seems like
       | it should be possible, in theory, but it takes too much work. So
       | most mathematicians prefer communicating their proofs in PDFs
       | rather than in a more rigorous format.
        
         | bvssvni wrote:
         | One thing I would like point out with Godel's incompleteness
         | theorems, is that there are different notions of provability.
         | Godel uses the notion of "provability" you get from Provability
         | Logic, which is a modal logic where you introduce terms from
         | e.g. Sequent Calculus.
         | 
         | Recently I found another notion of "provability" where Lob's
         | axiom, required in Provability Logic, is absurd. It turns out
         | that this notion fits better with Intuitionistic Propositional
         | Logic than Provability Logic. This allows integrating the meta-
         | language into object-language. This is pretty recent, so I
         | think we still have much to learn about the foundations of
         | mathematics.
        
           | layer8 wrote:
           | > Recently I found another notion of "provability"
           | 
           | Any pointers to that alternative notion?
        
             | bvssvni wrote:
             | I'm implementing it in this project:
             | https://crates.io/crates/hooo
        
           | drdeca wrote:
           | This makes it kind of sound as if Lob's axiom is used in the
           | proof of Godel's incompleteness theorems, but Godel was
           | working in (I think) PA, not in Provability Logic?
           | 
           | I guess you just meant "the notion of provability is the same
           | as the one that would later be described in Provability
           | logic" ?
           | 
           | I viewed the page you linked, but I don't see anywhere where
           | you describe the alternate notion of "provability" you have
           | in mind.
           | 
           | "Provability" in the sense of Godel's incompleteness
           | theorems, is just, "There exists a proof of it in [the chosen
           | system of axioms + rules of inference]", yes? I don't see how
           | any other notion would be fitting of the name "provability".
           | "provable" , "possible to prove", "a proof exists", all the
           | same thing?
           | 
           | Oh, I guess, if you want to use "there exists" in a
           | specifically constructive/intuitionistic way? (so that to
           | prove "[]P" you would have to show that you _can produce_ a
           | proof of P?)
           | 
           | I think I've seen other people try to do away with the need
           | for a meta-language / make it the same as the object-
           | language, and, I think this generally ends up being
           | inconsistent in the attempts I've seen?
           | 
           | edit: I see you have written 'Self quality a ~~ a is
           | equivalent to ~a, which is called a "qubit".'
           | 
           | I don't know quite what you mean by this, but, _please_ do
           | not call this a qubit, unless you literally mean something
           | whose value is a vector in a 2d Hilbert space.
        
             | bvssvni wrote:
             | > I guess you just meant "the notion of provability is the
             | same as the one that would later be described in
             | Provability logic" ?
             | 
             | yes
             | 
             | > I viewed the page you linked, but I don't see anywhere
             | where you describe the alternate notion of "provability"
             | you have in mind.
             | 
             | See section "HOOO EP"
             | 
             | >Oh, I guess, if you want to use "there exists" in a
             | specifically constructive/intuitionistic way? (so that to
             | prove "[]P" you would have to show that you can produce a
             | proof of P?)
             | 
             | This would be `|- p` implies `#p`, which is the N axiom in
             | modal logic (used by Provability Logic).
             | 
             | In Provability Logic, you can't prove `#false => false`. If
             | you can prove this, then Lob's axiom implies `false`, hence
             | absurd. `#p => p` for all `p` is the T axiom in modal
             | logic. In IPL, if have `true |- false`, then naturally you
             | can prove `false`. So, you can only prove `#false => false`
             | if you already have an inconsistent theory.
             | 
             | Provability Logic is biased toward languages that assume
             | that the theory is consistent. However, if you want to
             | treat provability from a perspective of any possible
             | theory, then favoring consistency at meta-level is
             | subjective.
             | 
             | > I think I've seen other people try to do away with the
             | need for a meta-language / make it the same as the object-
             | language, and, I think this generally ends up being
             | inconsistent in the attempts I've seen?
             | 
             | I have given this some thought before and I think it is
             | based on fix-point results of predicates of one argument in
             | Provability Logic. For example, in Godel's proof, he needs
             | to encode a predicate without arguments in order to create
             | the Godel sentence. In a language without such fix-points,
             | this might not be a problem.
             | 
             | > I don't know quite what you mean by this, but, _please_
             | do not call this a qubit, unless you literally mean
             | something whose value is a vector in a 2d Hilbert space.
             | 
             | The name "qubit" comes from the classical model, where you
             | generate a random truth table using the input bit vector as
             | seed. So, the proposition is in super-position of all
             | propositions and hence behaves like a "qubit" in a
             | classical approximation of a quantum circuit.
        
           | deltasepsilon wrote:
           | I'm attempting to understand you, so:
           | 
           | Are you saying that Lob's axiom, which states that the
           | provability of "the provability of p implies p" implies the
           | provability of p, necessarily prejudices some implicit
           | assumption of consistency to the meta-language?
           | 
           | How so, and/or, what are the axioms, or derived properties,
           | of this new notion of provability you have uncovered?
        
         | libraryofbabel wrote:
         | > IMHO the problem now is that the foundations of mathematics
         | are impractical. Most mathematicians don't actually do
         | mathematics that can be tied back to the foundational axioms.
         | 
         | I suppose my followup to that is, _why_ is this a problem? Does
         | it lead to bad mathematics? (For what definition of "bad"?)
         | Would we be able to solve unsolved problems if we tied things
         | back to foundations more? Mathematicians have never really done
         | much mathematics that ties back to the foundational axioms --
         | why change? Rigor is not really much of an end in itself -
         | surely people do mathematics because it is useful or beautiful,
         | or both?
        
           | jimwhite42 wrote:
           | And you can apply it, when you do, it really works.
        
           | lacker wrote:
           | I think AI for mathematics will help us solve unsolved
           | problems. There should be a virtuous cycle during its
           | development.
           | 
           | 1. More formalized training data makes the math AI better
           | 
           | 2. Better math AI helps us formalize more known mathematics
           | 
           | But right now math AI is not very useful. It isn't at the
           | point where ChatGPT is for programming, where the AI is
           | really helpful for a lot of programming tasks even if it
           | isn't superhuman yet. So the virtuous cycle hasn't started
           | yet. But I think we will get there.
        
         | jimwhite42 wrote:
         | > It seems like it should be possible, in theory, but it takes
         | too much work.
         | 
         | The definition of a valid mathematical proof that I've heard
         | mathematicians use is if it convinces other mathematicians. I
         | think there's integrity in depth in mathematical proofs for
         | lots of reasons, tying back to an axiomatic basis is a lot of
         | extra work for likely no benefit.
         | 
         | On formal mathematics, almost no mathematicians pay attention
         | to this afaik. It's something that lives in philosophy
         | departments, not mathematics.
        
           | bvssvni wrote:
           | I think the most exciting work in mathematics today is in the
           | formal foundations. However, I can also understand
           | mathematicians who are thinking like this:
           | 
           | 1. I only need normal congruence
           | 
           | 2. I only need perfect information games
           | 
           | Under problems that are solvable using these two assumptions,
           | there is little benefit in tying proofs back to an axiomatic
           | basis. Once you drop one of these two assumptions, proofs get
           | much harder and a solid foundation gets more important.
        
       | bvssvni wrote:
       | The foundations of mathematics are all about language design.
       | 
       | To answer this question, one must say something about which
       | language a foundation of mathematics is using. For example, Set
       | Theory is formalized in First Order Logic. However, First Order
       | Logic uses Propositional Logic, so to build an alternative
       | foundation to Set Theory, you might consider starting with
       | Propositional Logic and extending it another way.
       | 
       | First Order Logic extends Propositional Logic with predicates.
       | This might seem like a good design at first, until you try to
       | reason about uniqueness. In Set Theory, one requires an equality
       | operator in addition to set membership, in order to be able to
       | reason about uniqueness, at all. This equality operator is ugly,
       | because you have to rebuild objects that are isomorphic but using
       | different encodings.
       | 
       | Predicates causes problems because they are unconstrained. For
       | easier formalizing of advanced theories, I suggested Avatar
       | Logic, which replaces predicates with binary relations, avatars
       | and roles. You can try it here: https://crates.io/crates/avalog
       | 
       | Also, most theories assume congruence for all predicates, which
       | is bad for e.g. foundations of randomness.
       | 
       | The next crisis in "the foundations of mathematics" will be
       | "tautological congruence". Luckily, this is already being worked
       | on, by extending Intuitionistic Propositional with exponential
       | propositions. This theory is known as "HOOO EP" and is
       | demonstrated here: https://crates.io/crates/hooo
        
       | tutfbhuf wrote:
       | I'm a bit sad the he doesn't mention univalent foundations, which
       | is based on [?]-groupoids instead of sets.
        
       | graycat wrote:
       | (1) Common K-12 math.
       | 
       | (2) A text on set theory, especially _axiomatic_ set theory.
       | 
       | (3) Abstract algebra, i.e., the natural numbers, the rationals,
       | the reals, the complex numbers, the basic properties, and some on
       | groups, rings, fields, and vector spaces.
       | 
       | That's enough for the _foundations_.
        
       ___________________________________________________________________
       (page generated 2023-11-01 23:02 UTC)