Take the 2-minute tour ×
Mathematics Stack Exchange is a question and answer site for people studying math at any level and professionals in related fields. It's 100% free, no registration required.

Milne defines the conductor of an abelian extension $L/K$ to be the smallest modulus $\mathfrak{m}$ s.t. the Artin map factors as

$$\psi_{L/K}:I_K^{\mathfrak{m}}\to \textrm{Cl}_\mathfrak{m}(K)\to \textrm{Gal}(L/K)$$

I'm trying to understand the class field theoretic proof of the Kronecker-Weber theorem. For the proof to work, one needs to show that given some conductor $\mathfrak{m}$ and its ray class field $L_\mathfrak{m}$, then the abelian extensions of $K$ contained in $L_\mathfrak{m}$ are precisely the fields s.t. their conductor divides $\mathfrak{m}$. Hence, one needs to prove (denoting the conductor of $L/K$ by $\mathfrak{f}(L/K)$) that

$$\mathfrak{f}(L/K)\mid \mathfrak{m}\Rightarrow L\subset L_\mathfrak{m}$$

Anyone know how to prove this? I've been looking at a few other books, but they seem to define things differently, so the proofs did not translate to Milne's definition, which is what I'm currently trying to understand.

EDIT: This question is about trying to fill in the details in Remark 3.8 on page 155 in his notes: http://www.jmilne.org/math/CourseNotes/CFT.pdf

share|improve this question
add comment

1 Answer

up vote 4 down vote accepted

I've never read Milne's notes, so I don't know what you are giving yourself at this point.

One way to argue is as follows: If the conductor $\mathfrak f$ of $L$ over $K$ divides $\mathfrak m$, then all the primes that split completely in $L_{\mathfrak m}$ also split completely in $L$. Consider now the compositum of $L$ and $L_{\mathfrak m}$; one sees that the primes that split completely in this extension are precisely the primes that split completely in $L_{\mathfrak m}$.

But a standard consequence of Cebotarev density is: if $E_1$ and $E_2$ are two Galois extensions of the number field $K$ (in some fixed algebraic closure $\overline{K}$) and the sets of primes that splits completely in each of them coincide (perhaps away from a finite number of bad primes), then $E_1$ and $E_2$ are equal.

This implies that $L_{\mathfrak m} L = L_{\mathfrak m}$, i.e. that $L \subset L_{\mathfrak m}$, as desired.

[In the context of abelian extensions there are other proofs that avoid Cebotarev; e.g. once you have the whole class field theory apparatus, the inclusion you ask about follows immediately. But the above argument is a stanard and important one, which gets at the heart of how splitting behaviour can pin down a field --- after all, $L_{\mathfrak m}$ is defined a priori by the splitting behaviour of primes, and it is the above result that shows that this splitting behaviour suffices to pin down $L_{\mathfrak m}$ uniquely.]

share|improve this answer
    
Thanks. The fact that $(p,L/K)=1\Rightarrow (p,L_\mathfrak{m}/K)=1$ is an obvious consequence of the definition, yet I had never thought about how useful this is. I don't mind that the proof uses Chebotarev as the Chebotarev part is a very standard argument. –  pki Nov 6 '11 at 0:25
add comment

Your Answer

 
discard

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.