In Richard Bass's Real analysis for graduate students, one can read

Let $X$ be an uncountable set and let $\mathcal A$ be the collection of subsets $A$ of $X$ such that either $A$ or $A^c$ is countable. Define $\mu(A)=0$ if $A$ is countable and $\mu(A)=1$ if $A$ is uncountable. Prove that $A$ is a measure.

I think the statement of the problem is flawed, as it seems $\mu$ isn't even finitely additive. Indeed, if $A$ and $B$ are both uncountable disjoint elements of $\mathcal A$, then $A\cup B$ is not countable, hence $\mu(A \cup B) = 1$, but $\mu(A)+ \mu(B) = 1+1=2$...

I think this issue is resolved if one defines $\mu(A)=\infty$ if $A$ is uncountable.

Did I miss something ? What do you think about my revision ?


Carefully reread the assumptions: either $A$ or $A^c$ must be countable. It follows that you can't have two disjoint uncountable sets $A,B$ in your sigma algebra, since the complement of one would contain the other and therefore $A$ and $A^c$ would be uncountable.

  • $\begingroup$ Yes, I should have seen that, thank you $\endgroup$ – Gabriel Romon Oct 24 '16 at 17:44

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.