# Dimension theory (algebra)

In mathematics, **dimension theory** is a branch of commutative algebra studying the notion of the dimension of a commutative ring, and by extension that of a scheme.

The theory is much simpler for an affine ring; i.e., an integral domain that is a finitely generated algebra over a field. By Noether's normalization lemma, the Krull dimension of such a ring is the transcendence degree over the base field and the theory runs in parallel with the counterpart in algebraic geometry; cf. Dimension of an algebraic variety. The general theory tends to be less geometrical; in particular, very little works/is known for non-noetherian rings. (Kaplansky's commutative rings gives a good account of the non-noetherian case.) Today, a standard approach is essentially that of Bourbaki and EGA, which makes essential use of graded modules and, among other things, emphasizes the role of multiplicities, the generalization of the degree of a projective variety. In this approach, Krull's principal ideal theorem appears as a corollary.

Throughout the article, denotes Krull dimension of a ring and the height of a prime ideal (i.e., the Krull dimension of the localization at that prime ideal.)

## Basic results

Let *R* be a noetherian ring or valuation ring. Then

If *R* is noetherian, this follows from the fundamental theorem below (in particular, Krull's principal ideal theorem.) But it is also a consequence of the more precise result. For any prime ideal in *R*,

This can be shown within basic ring theory (cf. Kaplansky, commutative rings). By the way, it says in particular that in each fiber of , one cannot have a chain of primes ideals of length .

Since an artinian ring (e.g., a field) has dimension zero, by induction, one gets the formula: for an artinian ring *R*,

## Fundamental theorem

Let be a noetherian local ring and *I* a -primary ideal (i.e., it sits between some power of and ). Let be the Poincaré series of the associated graded ring . That is,

where refers to the length of a module (over an artinian ring ). If generate *I*, then their image in have degree 1 and generate as -algebra. By the Hilbert–Serre theorem, *F* is a rational function with exactly one pole at of order, say, *d*. It also says (contained in the proof) that . Since

we find that, for *n* large, the coefficient of in is of the form

That is to say, is a polynomial in *n* of degree when *n* is large. *P* is called the Hilbert polynomial of .

We set . We also set to be the minimum number of elements of *R* that can generate a -primary ideal of *R*. Our ambition is to prove the **fundamental theorem**:

Since we can take *s* to be , we already have from the above. Next we prove by induction on . Let be a chain of prime ideals in *R*. Let and *x* a nonzero nonunit element in *D*. Since *x* is not a zero-divisor, we have the exact sequence

The degree bound of the Hilbert-Samuel polynomial now implies that . (This essentially follows from the Artin-Rees lemma; see Hilbert-Samuel function for the statement and the proof.) In , the chain becomes a chain of length and so, by inductive hypothesis and again by the degree estimate,

The claim follows. It now remains to show More precisely, we shall show:

(Notice: is then -primary.) The proof is omitted. It appears, for example, in Atiyah–MacDonald. But it can also be supplied privately; the idea is to use prime avoidance.

## Consequences of the fundamental theorem

Let be a noetherian local ring and put . Then

- , since a basis of lifts to a generating set of by Nakayama. If the equality holds, then
*R*is called a regular local ring. - , since .

(Krull's principal ideal theorem) The height of the ideal generated by elements in a noetherian ring *R* is at most *s*. Conversely, a prime ideal of height *s* can be generated by *s* elements.

Proof: Let be a prime ideal minimal over such an ideal. Then . The converse was shown in the course of the proof of the fundamental theorem.

If is a morphism of noetherian local rings, then

The equality holds if is flat or more generally if it has the going-down property. (Here, is thought of as a special fiber.)

Proof: Let generate a -primary ideal and be such that their images generate a -primary ideal. Then for some *s*. Raising both sides to higher powers, we see some power of is contained in ; i.e., the latter ideal is -primary; thus, . The equality is a straightforward application of the going-down property.

If *R* is a noetherian local ring, then

Proof: If are a chain of prime ideals in *R*, then are a chain of prime ideals in while is not a maximal ideal. Thus, . For the reverse inequality, let be a maximal ideal of and . Since is a principal ideal domain, we get by the previous inequality. Since is arbitrary, this implies .

## Regular rings

Let *R* be a noetherian ring. The projective dimension of a finite *R*-module *M* is the shortest length of any projective resolution of *R* (possibly infinite) and is denoted by . We set ; it is called the global dimension of *R*.

Assume *R* is local with residue field *k*.

Proof: We claim: for any finite *R*-module *M*,

By dimension shifting (cf. the proof of Theorem of Serre below), it is enough to prove this for . But then, by the local criterion for flatness, Now,

completing the proof.

Proof: If , then *M* is *R*-free and thus is -free. Next suppose . Then we have: when is the kernel of some surjection from a free module to *M*. Thus, by induction, it is enough to consider the case . Then there is a projective resolution:

which gives:

But tensoring with *M* we see the first term vanishes. Hence, is at most 1.

Proof:^{[2]} If *R* is regular, we can write , a regular system of parameters. An exact sequence , some *f* in the maximal ideal, of finite modules, , gives us:

But *f* here is zero since it kills *k*. Thus, and consequently . Using this, we get:

The proof of the converse is by induction on . We begin with the inductive step. Set , among a system of parameters. To show *R* is regular, it is enough to show is regular. But, since , by inductive hypothesis and the preceding lemma with ,

The basic step remains. Suppose . We claim if it is finite. (This would imply that *R* is a semisimple ring; i.e., a field.) If that is not the case, then there is some finite module with and thus in fact we can find *M* with . By Nakayama's lemma, there is a surjection such that is an isomorphism. Denoting by *K* the kernel we have:

Since , *K* is free. Since , the maximal ideal is an associated prime of *R*; i.e., for some *s* in *R*. Since , . Since *K* is not zero, this implies , which is absurd. The proof is complete.

## Depths

Let *R* be a ring and *M* a module over it. A sequence of elements in is called a regular sequence if is not a zero-divisor on and is not a zero divisor on for each .

Assume *R* is local with maximal ideal *m*. Then the depth of *M* is the supremum of any maximal regular sequence in *m*. It is easy to show (by induction, for example) that . If the equality holds, *R* is called the Cohen–Macaulay ring.

The Auslander–Buchsbaum formula relates depth and projective dimension.

## References

- Part II of {{#invoke:citation/CS1|citation

|CitationClass=citation }}.

- Chapter 10 of {{#invoke:citation/CS1|citation

|CitationClass=citation }}.

- Kaplansky, Irving,
*Commutative rings*, Allyn and Bacon, 1970. - {{#invoke:citation/CS1|citation

|CitationClass=book }}