Weak nullstellensatz for arbitrary fields: Difference between revisions

From Commalg
No edit summary
Line 26: Line 26:
* We use [[Steinitz theorem]] to show that we can find a subfield <math>k(T)</math> of <math>K</math>, which is the field of fractions of a subset <math>T</math> of <math>k</math>, such that <math>K</math> is algebraic over <math>k(T)</math>. In our case, since <math>K</math> is finitely generated over <math>k</math>, it is also finitely generated over <math>k(T)</math>, so in fact <math>K</math> is a finite field extension of <math>k(T)</math>. {{further|[[Finitely generated and integral implies finite]]}}
* We use [[Steinitz theorem]] to show that we can find a subfield <math>k(T)</math> of <math>K</math>, which is the field of fractions of a subset <math>T</math> of <math>k</math>, such that <math>K</math> is algebraic over <math>k(T)</math>. In our case, since <math>K</math> is finitely generated over <math>k</math>, it is also finitely generated over <math>k(T)</math>, so in fact <math>K</math> is a finite field extension of <math>k(T)</math>. {{further|[[Finitely generated and integral implies finite]]}}
* We use [[Artin-Tate lemma]] and the fact that fields are Noetherian, to deduce that <math>k(T)</math> is finitely generated as a <math>k</math>-algebra (here <math>A = k, B = k(T), C = K</math>)
* We use [[Artin-Tate lemma]] and the fact that fields are Noetherian, to deduce that <math>k(T)</math> is finitely generated as a <math>k</math>-algebra (here <math>A = k, B = k(T), C = K</math>)
* We now use the fact that if <math>T</math> is nonempty, <math>k(T)</math> can never be finitely generated over <math>k</math>. Thus, <math>T</math> is empty, forcing <math>K</math> to be a finite field extension of <math>k</math> (This uses the fact that the polynomial ring is a unique factorization domain with infinitely many irreducibles).
* We now use the fact that if <math>T</math> is nonempty, <math>k(T)</math> can never be finitely generated over <math>k</math>. Thus, <math>T</math> is empty, forcing <math>K</math> to be a finite field extension of <math>k</math> (This uses the fact that the polynomial ring is a [[unique factorization]] domain with infinitely many irreducibles).


==Proof using Noether normalization theorem==
==Proof using Noether normalization theorem==

Revision as of 22:21, 8 February 2008

This fact is an application of the following pivotal fact/result/idea: Artin-Tate lemma
View other applications of Artin-Tate lemma OR Read a survey article on applying Artin-Tate lemma

This fact is an application of the following pivotal fact/result/idea: Noether normalization theorem
View other applications of Noether normalization theorem OR Read a survey article on applying Noether normalization theorem

Statement

Here are two equivalent formulations:

  • Suppose is a field and is a field extension of , such that is finitely generated as a -algebra. Then, is algebraic over , and in fact, is a finite field extension of .
  • Suppose is a maximal ideal in a polynomial ring in finitely many variables over . Then the quotient field for is a finite field extension of

Applications

Proof using Artin-Tate lemma

Facts used

Proof outline

  • We use Steinitz theorem to show that we can find a subfield of , which is the field of fractions of a subset of , such that is algebraic over . In our case, since is finitely generated over , it is also finitely generated over , so in fact is a finite field extension of . Further information: Finitely generated and integral implies finite
  • We use Artin-Tate lemma and the fact that fields are Noetherian, to deduce that is finitely generated as a -algebra (here )
  • We now use the fact that if is nonempty, can never be finitely generated over . Thus, is empty, forcing to be a finite field extension of (This uses the fact that the polynomial ring is a unique factorization domain with infinitely many irreducibles).

Proof using Noether normalization theorem

Facts used

Proof outline

Suppose is a finitely generated algebra over , that happens to be a field. Then, by the Noether normalization theorem, there exists a polynomial algebra inside such that is finite over this polynomial algebra.

But being a field, must at any rate contain the field of fractions of , and this yields a contradiction if .

Alternatively, we can observe that the injective map from to being a finite morphism, must give a surjective map on spectra, but the spectrum of has one element, and so can surject to the spectrum of a polynomial ring only if .

This yields that is finite-dimensional over .