Mathematics Stack Exchange is a question and answer site for people studying math at any level and professionals in related fields. Join them; it only takes a minute:

Sign up
Here's how it works:
  1. Anybody can ask a question
  2. Anybody can answer
  3. The best answers are voted up and rise to the top

This is from Atiyah Macdonald, Chapter 1, Execrise 20.

If $A$ is a ring and $X=Spec(A)$, then the irreducible components of $X$ are the closed sets $V(p)$, where $p$ is a minimal prime ideal of $A$.

It is not difficult to prove the closed sets $V(p)$ are irreducible. However it isn't clear to me how to prove the second part. It is clear that when closed sets of the form $V(q),q\in X$ then $V(p)$ did provide a maximal irreducible closed set. Neverthelss it is not clear to me why every closed set in $X$ must be of this form(they only provide a basis of closed sets). So theoretically there is the possiblity of some closed set $U$ that contains $V(p)$, and in particular contains some prime ideal $k$ such that $p\not\subset k$. Then what can we do to prove this cannot hold and $k$ must contain $p$?

One way to address this is to take the closure of $U\cap X_{p}=K$. Since the $V(q),q\in X$ form a basis there is some $V(l)\subset \overline K$. Then $V_{l}\cup V_{p}\subset U$, while $V_{l}\cup V_{p}=V_{l\cap p}$. But by definition $l\cap p=p$; so such $V_{l}$ could not exist, and $K$ must be empty.

I am wondering if this approach is long-winded, as I found almost every other proof used the fact every irreducible component of $X$ must be of the form $V_{r(p)}$, where $r(p)$ denotes the radical of $p$. Why this is true, and how it might help to simplify the above proof?

share|cite|improve this question
up vote 3 down vote accepted

Let $I$ be an ideal of $A$. We denote by $rad(I)$ the set $= \{x \in A\colon x^n \in I$ for some integer $n > 0$, where $n$ depends on $x\}$. It is easy to see that $rad(I)$ is an ideal and $V(I) = V(rad(I))$.

Let $E$ be a subset of $X = Spec(A)$. We denote by $\mathfrak{I}(E)$ the ideal $\bigcap_{p \in E} p$.

Lemma 0 Let $I$ be an ideal of $A$. Then $\mathfrak{I}(V(I)) = rad(I)$.

Proof: This is an immediate consequence of proposition 1.8 of Atiyah-MacDonald(it states that $\mathfrak{I}(X) = rad(0)$.

Lemma 1 Let $E$ be a subset of $X$. Then $V(\mathfrak{I}(E))$ is the closure of $E$.

Proof: Let $\bar E$ be the closure of $E$. Since $E \subset V(\mathfrak{I}(E))$, $\bar E \subset V(\mathfrak{I}(E))$.

Conversely suppose $E \subset V(I)$ for some ideal $I$. Then $I \subset \mathfrak{I}(E)$. Hence $V(\mathfrak{I}(E)) \subset V(I)$. Hence $V(\mathfrak{I}(E)) \subset \bar E$. QED

Lemma 2 $X = Spec(A)$ is irreducible if and only if $A/rad(0)$ is an integral domain.

Proof: Since $X = Spec(A/rad(0))$, we can assume $rad(0) = 0$.

Suppose $X$ is reducible. There exist proper closed subsets $F_1, F_2$ of $X$ such that $X = F_1 \cup F_2$. Hence $0 = \mathfrak{I}(X) = \mathfrak{I}(F_1) \cap \mathfrak{I}(F_2)$. By Lemma 1, $V(\mathfrak{I}(F_1)) = F_1$. Since $F_1 \neq X$, $\mathfrak{I}(F_1) \neq 0$. Similarly $\mathfrak{I}(F_2) \neq 0$. Hence $A$ is not an integral domain.

Conversely suppose $A$ is not an integral domain. There exist elements $f, g$ of $A$ such that $f \neq 0, g \neq 0$, $fg = 0$. Then, by Lemma 0, $V(f) \neq X$, $V(g) \neq X$ and $X = V(f) \cup V(g)$. Hence $X$ is reducible. QED

Lemma 3 Let $I$ be an ideal of $A$. Then $V(I)$ is irreducible if and only if $rad(I)$ is a prime ideal.

Proof: This follows immediately from Lemma 2.

By Lemma 3, an irreducible closed subset of $X$ is of the form $V(q)$, where $q$ is a prime ideal of $A$. Hence $V(p)$ is a maximal irreducible closed subset of $X$, if $p$ is a minimal prime ideal of $A$.

share|cite|improve this answer
Sorry for my low level, but the book did not define an irreducible closed subset must be of this form. You already used $V_{p}\cup V_{q}=V_{p\cap q}$? I see. – Bombyx mori Sep 17 '12 at 2:01
@Makoto Your Lemma 1 is the full Nullstellensatz, which is not covered yet in chapter 1 in Atiyah - Macdonald. – user38268 Sep 17 '12 at 3:20
@BenjaLim I added the proof of Lemma 1. – Makoto Kato Sep 17 '12 at 3:33
@MakotoKato Sorry I was mistaken, it is not lemma 1 that is the full Nullstellensatz but the fact before that $I(V(J)) = \sqrt{J}$. By the way in that line you have two $I$'s which make it confusing. On one hand $I$ in the radical means an ideal, while $I$ on the left is the ideal of all functions vanishing on $V(J)$.... – user38268 Sep 17 '12 at 3:35
@user32240 Since $\bar E$ is closed, it is of the form $V(I)$. – Makoto Kato Sep 17 '12 at 5:44

Your Answer


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.