3
$\begingroup$

When letting an element be an arbitrary member of a set $A$ do I always have to account for $A =\emptyset$?

E.g., if I wanted to prove (an incorrect) theorem $$ \forall A,B,C,D : A \times B \subset C \times D \implies (A \subset C \wedge B \subset D).$$

I would let $x$ be an arbitrary element of $A$ but I cannot proceed with the proof without considering the case when A is $\emptyset$. Do I always have to do this whenever I pick an arbitrary element of a set?

$\endgroup$
1
  • $\begingroup$ One actually should ,but It is usually not done because of simple reasons: often the statement is simply void or most of the time it is simply uninteresting. $\endgroup$ – b00n heT Jul 24 '16 at 9:43
3
$\begingroup$

You have to account for it. However, this is often automatically done, because any statement in the form $$[\forall x\in A, P(X)]\equiv [\forall x,(P(x)\vee\neg x\in A)]$$ is always true when $A=\emptyset$.

You must be careful when you make an existence claim, because any satement in the form $$[\exists x\in A,\ P(x)]\equiv [\exists x,\ (x\in A\wedge P(x))]$$ is always false when $A=\emptyset$.

$\endgroup$
2
$\begingroup$

Generally, yes. Usually, the theorem is easy to prove when the set is empty, so just start off by handling the case where $A$ is empty, then assume $A$ is nonempty so you can pick $x\in A$.

EDIT- Though there are some cases where you don't have to handle the two cases. For instance, the definition of $A\subset B$ is $x\in A \implies x\in B$; to prove $A \subset B$, you simply show that if we have some $x\in A$, then it follows that $x\in B$ (regardless of whether $A$ happens to be empty). Generally, just ask yourself: if you don't handle the possibility that your set is empty, is your proof incomplete? Have you failed to prove your result in the case that $A=\emptyset$?

Your example statement would fall into the second category.

$\endgroup$
1
$\begingroup$

Yes, you do have to do that everytime.

It could even be a method to prove that a given set $E$ is empty :

Let $x$ be in $E$... and then you get to a contradiction. And you conclude that $E=\emptyset$.

$\endgroup$

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.