MathOverflow is a question and answer site for professional mathematicians. Join them; it only takes a minute:

Sign up
Here's how it works:
  1. Anybody can ask a question
  2. Anybody can answer
  3. The best answers are voted up and rise to the top

Let $H$ and $K$ be finitely generated subgroups of a free group $F$, and suppose that $H$ has finite index in $F$. Is it true that $rank(H \cap K)-1 \leq (rank(H)-1)(rank(K)-1)$?

share|cite|improve this question
(actually, I missed the "finite-index" part, so ignore my deleted comment). – Andy Putman Nov 3 '13 at 19:32
@mary seva, it looks like your question is about to be closed (and/or migrated) as 'not research level'. In fact, I think it is graduate-student level, and hence acceptable on MO. But I suspect the formulation, which reads like a homework problem, has irritated the voters to close. Could you tell us how this problem arises in your research? – HJRW Nov 3 '13 at 19:39
While this is certainly not a great question, why does it get THAT many downvotes? – Stefan Kohl Nov 3 '13 at 19:50
@StefanKohl I am guessing because it looks like homework. – Igor Rivin Nov 3 '13 at 20:00

Yes, this statement is true and for a long time was known as the Hanna Neumann Conjecture. It was proved in 2011 by Igor Mineyev.

share|cite|improve this answer
@StephanKohl The first response was knee-jerk, the second was a result of some thought. There is no way to retract a close vote, as far as I know. – Igor Rivin Nov 3 '13 at 20:29
Igor, it was asked about the case, where one of the subgroups is of finite index. This is indeed an exercise on Schreier formula. – Anton Klyachko Nov 3 '13 at 21:29
@WillJagy What do you mean "wanders off"? The question was asked just three hours ago. – Todd Trimble Nov 3 '13 at 21:49
@WillJagy Yes, I do. A "normal person" (not a fanatic like you or me (-:) might post, go about their usual day, and maybe check back in the evening when there's a spare moment. Personally I think getting back within 24 hours is fine, but I feel some impatience if it takes more than 48 hours. – Todd Trimble Nov 4 '13 at 0:41
By the way, Friedman also gave a correct proof of the Hanna Neumann Conjecture (after a couple of incorrect attempts). – HJRW Nov 4 '13 at 15:00

Your Answer


By posting your answer, you agree to the privacy policy and terms of service.

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