In his recent work on the divisor function in arithmetic progressions to smooth moduli, A. Irving proves the following rather amusing lemma (see Lemma 4.5 in his paper):

LemmaLet be an odd prime number, let be an integer and let be a -tuple of elements of . For any subset of , denote

and for any , let

denote the multiplicity of among the .

Then if none of the is zero, there exists some for which isodd.

I will explain two proofs of this result, first Irving’s, and then one that I came up with. I’m tempted to guess that there is also a proof using some graph theory, but I didn’t succeed in crafting one yet.

**Irving’s proof.** This is very elegant. Let be a primitive -th root of unity. We proceed by contraposition, hence assume that all multiplicities are even. Now consider the element

of the cyclotomic field . By expanding and using the assumption we see that

In particular, the norm (from to ) of is an even integer, but because is odd, the norm of is known to be odd for all . Hence some factor must have , as desired.

**A second proof.** When I heard of Irving’s Lemma, I didn’t have his paper at hand (or internet), so I tried to come up with a proof. Here’s the one I found, which is a bit longer but maybe easier to find by trial and error.

First we note that

is even. In particular, since is odd, there is at least some with *even*.

Now we argue by induction on . For , the result is immediate: there are two potential sums and , and so if , there is some odd multiplicity.

Now assume that and that the result holds for all -tuples. Let be a -tuple, with no equal to zero, and which has all multiplicities even. We wish to derive a contradiction. For this, let . For any , we have

by counting separately those with sum which contain or not.

Now take such that is odd, which exists by induction. Our assumptions imply that is also odd. Then, iterating, we deduce that is odd for all integers . But the map is surjective onto , since is non-zero. Hence our assumption would imply that all multiplicities are *odd*, which we have seen is not the case… Hence we have a contradiction.

Dear professor Kowalski,

Thank you for this nice lemma, I am fond of this kind of results.

It seems to me (if I am not wrong) that the argument of the second proof holds

also whenever the problem is considered in any finite abelian group of odd cardinality,

working one coset after the other (and may be even in non abelian groups ?!).

I was wondering if the first proof does also hold ?

Best regards from Paris,

Eric