7
$\begingroup$

Let $(X,\tau)$ be a topological space, $U \subseteq X$ a non-empty open subset of $X$ and let $W$ be an irreducible non-empty closed subset of $X$. Assuming $W \cap U \neq \emptyset$ why is this intersection irreducible in $U$ and its closure (taken in $X$) equal $W$?

Clearly $W \cap U$ is dense in $W$ because it is an open subset of $W$ and $W$ is irreducible so its closure, taken in $W$ is $W$ itself. I don't see why its closure taken in $X$ is equal $W$. Why is this? Also, why is it irreducible in $U$?

EDIT: OK $W=cl_{X}(W)=cl_{X}(W \cap U)$ I think. Why is it irreducible in $U$ though?

$\endgroup$
  • 2
    $\begingroup$ Dear user, a little nitpick: irreducible is an absolute notion. A tpological space is plain irreducible, not irreducible in...If you take a subspace of a big space, you can then ask if it is irreducible when endowed with the subspace topology. That said, I have answered your question below. $\endgroup$ – Georges Elencwajg Feb 11 '12 at 9:21
  • 1
    $\begingroup$ Also, I added in an edit that $U$ is open (which you certainly meant since in the second section you say that $W\cap U$ is open in $W$). rattle shows in his answer below that the statement needn't be true if $U$ isn't open . $\endgroup$ – Georges Elencwajg Feb 11 '12 at 9:34
7
$\begingroup$

To say that $W\neq \emptyset$ is irreducible means that it is not the union of two closed subsets both $\neq W$.
Passing to complements, it means that two non-empty open subsets of $W$ have non-empty intersection.
But then every open subset $V\subset W$ has the same property (since opens of $V$ are in particular opens of $W$) and is thus irreducible.
This applies in particular to $V=W\cap U$.

$\endgroup$
5
$\begingroup$

Your statement is not true, unless you assume more about $U$. For instance, take $X$ the affine plane with Zariski topology, $W$ a parabola and $U$ any line intersecting the parabola in two points. Then, $W\cap U$ consists of two distinct points of the line $U$, which is not irreducible at all. Their closure in $X$ also does not equal $W$.

$\endgroup$
  • 2
    $\begingroup$ +1: let me point out that rattle's answer is perfectly correct and pertinent, but that it refers to a previous version of the question in which user 6495 had forgotten to say that $U$ is open (this condition has now be added in an edit) $\endgroup$ – Georges Elencwajg Feb 11 '12 at 9:40
1
$\begingroup$

A space $X$ is irreducible iff $U\cap V\ne\varnothing$ whenever $U$ and $V$ are non-empty open subsets of $X$. Thus, $X$ is irreducible iff every non-empty open subset of $X$ is dense in $X$.

Now let $\langle X,\tau\rangle$ be any space, let $W$ be a non-empty, irreducible, closed subset of $X$, and suppose that $U\in\tau$ is such that $U\cap W\ne\varnothing$. (Although you didn’t say so, it’s clear that you intended $U$ to be open.) Since $W$ is irreducible, $U\cap W$ is dense in $W$, and since furthermore $W$ is closed in $X$, $\operatorname{cl}_X(U\cap W)=\operatorname{cl}_W(U\cap W)=W$, as you said.

Now let $V_0$ and $V_1$ be non-empty open subsets of $U\cap W$. Then there are $G_0,G_1\in\tau$ such that $V_0=G_0\cap(U\cap W)$ and $V_1=G_1\cap(U\cap W)$, so $V_0$ and $V_1$ are non-empty open subsets of $W$ as well, and hence $V_0\cap V_1\ne\varnothing$. Thus, $U\cap W$ is irreducible.

$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.