-
Notifications
You must be signed in to change notification settings - Fork 0
Expand file tree
/
Copy pathCommRingLec06.tex
More file actions
executable file
·118 lines (104 loc) · 5.6 KB
/
Copy pathCommRingLec06.tex
File metadata and controls
executable file
·118 lines (104 loc) · 5.6 KB
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
\stepcounter{lecture}
\setcounter{lecture}{6}
\sektion{Lecture 6}
\renewcommand\P{\mathcal{P}}
\subsektion{\S 4 Noetherian Rings and Noetherian Induction}
Recall the following facts about rings and modules with chain conditions.
\begin{enumerate}
\item A module is noetherian if and only if every submodule is finitely generated.
\item A module is noetherian and artinian if and only if it has a (finite) composition
series (quotients are simple). The Jordan-H\"older theorem tells us that the quotients
are unique up to permutation.
\item If $N\subseteq M$, $M$ is noetherian (resp.\ artinian) if and only if both
$N$ and $M/N$ are.
\item If $R$ is noetherian, then $M$ is noetherian if and only if it is finitely
generated.
\end{enumerate}
\begin{theorem}[I.\ S.\ Cohen]\label{lec06T:cohen}
A ring $R$ is noetherian if and only if all prime ideals are finitely generated.
\end{theorem}
For the proof, we need the following lemma and proposition.
\begin{lemma}[Oka's Lemma]
Let $I\< R$ and $b\in R$. If $I+(b)$ and $I:b$ are finitely generated, then so is $I$.
\end{lemma}
\begin{proof}
Since $I+(b)$ is finitely generated, it can be written as $I_0+(b)$ for some finitely
generated ideal $I_0\subseteq I$. Then $I = I_0+b(I:b)$ is finitely generated.
\end{proof}
\begin{proposition}\label{lec06P:maxisprime}
If $I\< R$ is maximal with respect to \emph{not} being finitely generated (i.e.\ every
$J\supsetneq I$ is f.g.), then $I$ is prime.
\end{proposition}
\begin{proof}
Clearly $I\neq R$. Suppose $I$ is not prime, so there are $a,b\not\in I$, with $ab\in
I$. Then $I+(b)$ and $I:b$ ($\ni a$) are finitely generated. By Oka's Lemma, $I$ is
finitely generated, which is a contradiction.
\end{proof}
\begin{proof}[Proof of Theorem \ref{lec06T:cohen}]
Assume all primes in $R$ are finitely generated, and that $\F=\{$non-f.g.\
ideals$\}\neq \varnothing$. Since the union of non-finitely-generated ideals is not
finitely generated, Zorn's Lemma gives us a maximal element, which is prime by
Proposition \ref{lec06P:maxisprime}, so it is finitely generated by assumption.
Contradiction.
\end{proof}
\[\xymatrix@!0 @R=3pc @C=10pc{
\text{ \emph{all} ideals f.g.} \ar@{=>}@/_2.2ex/[r] \ar@{<=>}[d]
& \text{primes f.g.} \ar@{=>}[l]_{\text{Cohen}}
\ar@{==>}[r]^(.43){\text{Krull's PIT}}
\ar@{:>}[d]
& \text{DCC on primes}\\
\text{ACC on \emph{all} ideals} \ar@{=>}@/^2.2ex/[r]^{\text{Obvious}} & \text{ACC on primes}
}\]
% Noetherian induction is this idea: let $R$ be noetherian, and we want to prove some
% statement for $R$. Assume it is false, and look at the collection of counterexamples,
% and get a maximal counterexample (by the maximum principle). Get a contradiction by hook
% or by crook.
\begin{theorem}[Noetherian Induction Principle]
Let $R$ be a noetherian ring, let $\P$ be a property, and let $\F$ be a family of
ideals $R$. Suppose the inductive step: if all ideals in $\F$ strictly larger than
$I\in \F$ satisfy $\P$, then $I$ satisfies $\P$. Then all ideals in
$\F$ satisfy $\P$.
\end{theorem}
\begin{proof}
Assume $\F_\text{crim} = \{J\in \F|J\text{ does not satisfy }\P\}\neq \varnothing$.
Since $R$ is noetherian, $\F_\text{crim}$ has a maximal member $I$. By maximality, all
ideals in $\F$ strictly containing $I$ satisfy $\P$, so $I$ also does by the inductive
step.
\end{proof}
\begin{definition}
An element $r$ is \emph{irreducible} if it cannot be written as a product of two
non-units. A (proper) ideal $I$ is \emph{irreducible} if it cannot be written as the
intersection of two strictly larger ideals. (irreducible ideals are primary!
\anton{ref this result})
\end{definition}
\begin{example} For a noetherian ring $R$, we can prove the following results by
checking the inductive step in each case. For the first two, take $\F$ to be the set of
all ideals.
\begin{enumerate}
\item Every ideal is a finite intersection of irreducible ideals.
Assume every ideal strictly containing $I$ is a finite intersection of irreducible
ideals. If $I=R$, it is the empty intersection. If $I$ is irreducible, then we're
done. Otherwise, $I = J_1\cap J_2$ for strictly larger ideals $J_1$ and $J_2$. By
assumption, $J_i$ is a finite intersection of irreducible ideals, so $I$ is also.
\item ($R\neq 0$) Every ideal in $R$ contains a finite product of primes.
Assume any ideal larger than $I$ contains a finite product of primes. If $I$ is
prime or $R$, we're done. Otherwise, there are ideals $J_1$ and $J_2$ which contain
$I$ such that $J_1J_2\subseteq I$. Since each $J_i$ contains a finite product of
primes, so does $I$.
\item ($R$ a domain) Every $r\not\in (0)\cup U(R)$ is a finite product of
irreducible elements.
Here we let $\F$ be non-zero, non-$R$, principal ideals, and let $\P$ be the
statement that the generator is a finite product of irreducible elements (note that
this is independent of the choice of generator since $R$ is a domain). \anton{finish
... easy}
\item If $J=\sqrt J$, then $J$ is a \emph{finite} intersection of primes.
(Kaplansky's Theorems 87 and 88)
Take $\F$ to be the set of radical ideals.\anton{finish}
As a corollary, for any $I\< R$ in a noetherian ring, there are finitely many
minimal primes over $I$.
\end{enumerate}
\vspace*{-1.7\baselineskip}
\end{example}
Note that the examples where $\F$ is not the set of all ideals illustrate that to
apply noetherian induction, you only need the ideals \emph{in $\F$} to satisfy the
ascending chain condition.