I am studying set theory. In most of the references on set theory such as the Jech's book, they start with the axioms of ZFC and the first-order language of set theory and eventually they start to talk about certain 'objects' called sets. I see a certain logic jump. I want to understand how is formally defined the sentence '$x$ is a set' from the first-order language. I understand that this sentence means that we can find some well-formed formula $\varphi(u)$ with a free variable $u$ where the sentence $(\exists x)\varphi(x)$ is true in ZFC. Am I right?

  • 2
    $\begingroup$ It is not possible with with "standard" $\mathsf{ZFC}$ because we have no predicate $\text{Set}(x)$. Every element of the domain is a set. $\endgroup$ Mar 2 '18 at 8:37
  • 1
    $\begingroup$ There are versions of set-tehory with atoms (or urelements) i.e. objects that are not set. In this case, we may have a predicate $\text {Ur}$ such that $\text {Ur}(x)$ holds iff $x$ is not a set. $\endgroup$ Mar 2 '18 at 8:39
  • $\begingroup$ @MauroALLEGRANZA As far as I see, in Jech's book, it starts with the axioms but no domain is introduced, then eventually they talk about certain objects sets. It is always necessary to have a domain? $\endgroup$ Mar 2 '18 at 9:15
  • 1
    $\begingroup$ Every "formal" theory pressuposes implicitly that it "speak about" something. Every set theory "speaks about" sets. The "domain" is the technical term of the formal semantics relative to a theory used to formalize the informal notion of "speaking about something". $\endgroup$ Mar 2 '18 at 9:20
  • $\begingroup$ It might be useful to think of the analogous question for more concrete theories; e.g. how do we say "is a natural number" in the context of Peano arithmetic (where everything considered is a natural number)? $\endgroup$ Mar 2 '18 at 21:19

When you work in standard set theory, everything is a set.

In other words, formulas in the language of set theory do cannot even be applied to things that are not sets.

So if you want a formula $\varphi(x)$ which is true when $x$ is a set and false if $x$ is not a set, this means that you want a $\varphi(x)$ that is always true, period. You can use any propositional tautology you want for that, such as $x\in x\to x\in x$, but that won't make you any wiser about sets.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.