Show $(A \cup B)\setminus(A \cap B) = (A\setminus B) \cup (B\setminus A)$.

What I have so far...

This is (A or B) and (A and B)' = (A and B') or (B and A')

(A or B) and (A' or B') = (A and B') or (B and A')

((A or B) and A') or ((A or B) and B') = (A and B') or (B and A')

I feel like I'm just getting farther away. There must be something simple I'm missing.

  • $\begingroup$ Try using venn diagrams to visualise this can help with the proof. $\endgroup$ – Kaladin Oct 13 '15 at 22:37
  • $\begingroup$ I am. It helps, but I feel that it makes me want to jump to make claims which I should be able to make by simply knowing general set algebra. $\endgroup$ – user2684794 Oct 13 '15 at 22:40
  • $\begingroup$ Alright. I figured it out. $\endgroup$ – user2684794 Oct 13 '15 at 22:48

As Simon's answer points out, often you want to construct a so-called "element-chasing" proof to show that $S=T$ by showing $S\subseteq T$ and $T\subseteq S$.

However, I would only use an element-chasing proof if you can't "squeeze out" the equality by using some basic set algebra. Assuming you know about what the complement of a set is, distributivity, etc., see if you can follow the proof below (more steps than usual have been added to emphasize clarity, but let me know if a step or two does not make sense): \begin{align} \text{LHS}&=(A\cup B)\setminus(A\cap B)\\ &= (A\cup B)\cap(A\cap B)^C\tag{definition}\\[0.5em] &= (A\cup B)\cap(A^C\cup B^C)\tag{De Morgan}\\[0.5em] &= [(A\cup B)\cap A^C]\cup[(A\cup B)\cap B^C]\tag{distrib.}\\[0.5em] &= [(A\cap A^C)\cup(B\cap A^C)]\cup[(A\cap B^C)\cup(B\cap B^C)]\tag{distrib.}\\[0.5em] &= [\varnothing\cup(B\cap A^C)]\cup[(A\cap B^C)\cup\varnothing]\tag{$S\cap S^C=\varnothing$}\\[0.5em] &= (B\cap A^C)\cup(A\cap B^C)\tag{simplify}\\[0.5em] &= (A\cap B^C)\cup(B\cap A^C)\tag{commutativity}\\[0.5em] &= (A\setminus B)\cup(B\setminus A)\tag{definition}\\[0.5em] &= \text{RHS}. \end{align} Are all of those steps clear? Can you see how the proof above, despite seeming somewhat long winded, is exactly a good bit simpler than cranking out two proofs (one for each direction)?

  • $\begingroup$ Yup. That's the route I took. $\endgroup$ – user2684794 Oct 13 '15 at 23:32


In general to show that two sets, say $S$ and $T$, are equal you want to show that $S \subset T$ and $T \subset S$. In other words $x \in S \Rightarrow x \in T$ and $x \in T \Rightarrow x \in S$.

Hence what you want to show is that

$$1. \quad x \in (A \cup B)\backslash (A \cap B) \ \Rightarrow x \in (A\backslash B)\cup(B\backslash A)$$ and

$$2. \quad x \in (A\backslash B)\cup(B\backslash A) \ \Rightarrow x \in (A \cup B)\backslash (A \cap B) $$

Try writing arguments for these two statements. A Venn diagram will also help you see what is going on here.

Here's the beginning of a possible argument for statement 1:

If $x \in (A \cup B)\backslash (A \cap B)$ then $x \in A$ or $x \in B$. However if $x \in A$ then $x \not\in B$ as $x \not\in A \cap B$. Similarly if $x \in B$ then ...


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.