Bourbaki

Exercise 2, § 1

Tag 03HB, permanent. Cite it as [Bourbaki, Tag 03HB].

Theory of Sets, II, 123

Show that ∅≠{x}\emptyset \neq \{x\} is a theorem. Deduce that (∃x)(∃y)(x≠y)(\exists x)(\exists y)(x \neq y) is a theorem.

Solution

This solution was written by nemotron-3-ultra-free and judged by a machine. It is not Bourbaki's and it has not been checked by a person.

By Theorem 1 of § 1, the relation (∀x)(x∉∅)(\forall x)(x \notin \emptyset) is a theorem; in particular, for any term xx we have x∉∅x \notin \emptyset. Also by Theorem 1, the relation ∅⊂X\emptyset \subset X is a theorem for every set XX; hence ∅⊂{x}\emptyset \subset \{x\} is a theorem.

By Definition 2 of § 1, the relation z∈{x}z \in \{x\} is equivalent to z=xz = x. Since equality is reflexive, x=xx = x holds, and therefore x∈{x}x \in \{x\}.

Now assume that ∅={x}\emptyset = \{x\} were a theorem. Because ∅⊂∅\emptyset \subset \emptyset is a theorem by Proposition 1 of § 1, we may substitute ∅\emptyset by {x}\{x\} in this relation to obtain {x}⊂∅\{x\} \subset \emptyset. The definition of inclusion (Definition 1 of § 1) gives (∀z)(z∈{x}⇒z∈∅)(\forall z)(z \in \{x\} \Rightarrow z \in \emptyset). Applying this to z=xz = x yields x∈{x}⇒x∈∅x \in \{x\} \Rightarrow x \in \emptyset. Since x∈{x}x \in \{x\} is true, we would have x∈∅x \in \emptyset, contradicting x∉∅x \notin \emptyset from Theorem 1. Therefore ¬(∅={x})\lnot(\emptyset = \{x\}) is a theorem, i.e., ∅≠{x}\emptyset \neq \{x\} is a theorem.

Since ∅≠{x}\emptyset \neq \{x\} is a theorem (with xx free), we may instantiate xx with the term ∅\emptyset. The existence of {∅}\{\emptyset\} is guaranteed by Axiom A2 of § 1, which asserts (∀x)(∀y)Coll⁡z(z=x or z=y)(\forall x)(\forall y)\operatorname{Coll}_z(z = x \text{ or } z = y); taking x=y=∅x = y = \emptyset gives Coll⁡z(z=∅)\operatorname{Coll}_z(z = \emptyset), so the set {∅}=Ez(z=∅)\{\emptyset\} = \mathscr{E}_z(z = \emptyset) exists. Thus ∅≠{∅}\emptyset \neq \{\emptyset\} is a theorem. By the rule of existential introduction (Chapter I, § 4, no. 1), we obtain (∃y)(∅≠y)(\exists y)(\emptyset \neq y), and then (∃x)(∃y)(x≠y)(\exists x)(\exists y)(x \neq y). Hence (∃x)(∃y)(x≠y)(\exists x)(\exists y)(x \neq y) is a theorem.

Cited by

Cites

Nothing.

Transcribed from the printed volume. Not translated.