Mathematics Stack Exchange is a question and answer site for people studying math at any level and professionals in related fields. 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

A set of lecture notes I'm reading on Halasz's theorem makes the following statement in a proof, which I can't quite follow - I was hoping someone might be able to clear up what I'm missing:

Therefore, $Re(\sum_p p^{-s} = \log \log x + O_M(1)$, where $O_M$ denotes that the constant involved in the big O notation depends on $M$, and $s:= 1 + \frac{1}{\log x} +it$. Recall that $Re(\sum_p p^{-s}) = \log |\zeta(s)| +O(1)$; hence $|\zeta(s)| \geq c_M\log x$, some $c_M$.

I hopefully won't need to give any more context about the proof itself as I don't think it's relevant, but I can if needs be. The zeta referred to is the usual zeta function, and $x$ is usually some real $\geq 1$.

What I can't understand is the $\geq$ in the final line; it appears both in my handwritten notes and the lecturer's typed notes, so I'm fairly sure it's meant to be there. However, the deduction which seems to be being made is to equate $ \log \log x + O_M(1)= \log |\zeta(s)| +O(1)$ and then deduce the inequality; but wouldnt this give us (taking exponents) $|\zeta(s)|=\log x \cdot \exp(O_M(1))$? So then if anything surely that would give us an upper bound for $|\zeta|$, not a lower bound?

If it helps, I can try to provide more information to clarify. The lecture notes are online at courtesy of Ben Green (who I hope doesn't mind me linking to his page), namely on the penultimate page of chapter 2, though I'm not sure whether the notes are accessible outside the Cambridge domain.

Many thanks in advance for your help.

share|cite|improve this question
I am very eager to read Green's note on Halasz's theorem, but the Link you provided does not work. – m.b Aug 29 '14 at 18:39
up vote 0 down vote accepted

Taking exponents you can just as easily arrive at $\log x = |\zeta(s)| \cdot \exp(O_M(1))$

which should provide the lower bound on $|\zeta(s)|$, I think

share|cite|improve this answer
Ah ok, yes that was silly of me. I wasn't really thinking of the big-O being a "2-sided" bound but I guess in this sense it is - thanks very much. – Paul Jun 5 '12 at 16:32

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.