10.08.2013 Views

A Brief Introduction to Classical and Adelic Algebraic ... - William Stein

A Brief Introduction to Classical and Adelic Algebraic ... - William Stein

A Brief Introduction to Classical and Adelic Algebraic ... - William Stein

SHOW MORE
SHOW LESS

Create successful ePaper yourself

Turn your PDF publications into a flip-book with our unique Google optimized e-Paper software.

78 CHAPTER 12. DIRICHLET’S UNIT THEOREM<br />

Koch wrote that:<br />

... important parts of mathematics were influenced by Dirichlet. His<br />

proofs characteristically started with surprisingly simple observations,<br />

followed by extremely sharp analysis of the remaining problem.<br />

I think Koch’s observation nicely describes the proof we will give of Theorem 12.1.2.<br />

The following proposition explains how <strong>to</strong> think about units in terms of the<br />

norm.<br />

Proposition 12.1.4. An element a ∈ OK is a unit if <strong>and</strong> only if Norm K/Q(a) =<br />

±1.<br />

Proof. Write Norm = Norm K/Q. If a is a unit, then a −1 is also a unit, <strong>and</strong> 1 =<br />

Norm(a)Norm(a −1 ). Since both Norm(a) <strong>and</strong> Norm(a −1 ) are integers, it follows<br />

that Norm(a) = ±1. Conversely, if a ∈ OK <strong>and</strong> Norm(a) = ±1, then the equation<br />

aa −1 = 1 = ± Norm(a) implies that a −1 = ± Norm(a)/a. But Norm(a) is the<br />

product of the images of a in C by all embeddings of K in<strong>to</strong> C, so Norm(a)/a is<br />

also a product of images of a in C, hence a product of algebraic integers, hence an<br />

algebraic integer. Thus a −1 ∈ OK, which proves that a is a unit.<br />

Let r be the number of real <strong>and</strong> s the number of complex conjugate embeddings<br />

of K in<strong>to</strong> C, so n = [K : Q] = r + 2s. Define a map<br />

by<br />

ϕ : UK → R r+s<br />

ϕ(a) = (log |σ1(a)|, . . .,log |σr+s(a)|).<br />

Lemma 12.1.5. The image of ϕ lies in the hyperplane<br />

H = {(x1, . . .,xr+s) ∈ R r+s : x1 + · · · + xr + 2xr+1 + · · · + 2xr+s = 0}. (12.1.1)<br />

Proof. If a ∈ UK, then by Proposition 12.1.4,<br />

<br />

r<br />

<br />

s<br />

|σi(a)| ·<br />

i=1<br />

i=r+1<br />

Taking logs of both sides proves the lemma.<br />

Lemma 12.1.6. The kernel of ϕ is finite.<br />

Proof. We have<br />

|σi(a)| 2<br />

<br />

= 1.<br />

Ker(ϕ) ⊂ {a ∈ OK : |σi(a)| = 1 for all i = 1, . . .,r + 2s}<br />

⊂ σ(OK) ∩ X,<br />

where X is the bounded subset of R r+2s of elements all of whose coordinates have<br />

absolute value at most 1. Since σ(OK) is a lattice (see Proposition 5.2.4), the<br />

intersection σ(OK) ∩ X is finite, so Ker(ϕ) is finite.

Hooray! Your file is uploaded and ready to be published.

Saved successfully!

Ooh no, something went wrong!