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

I have numbered this question as (1) because I will be posting series of questions where I don't understand. I hope its allowed. I want to prove the following :

If $X$ is a topological vector space then : If $A\subset X $ then $\bar A = \cap(A+V)$, where $V$ runs through all nbd of $0$.

I tried like this : $x\in \bar A\subset A+V$ $\implies x\in A+V $ for all V. The other way , $x\in A+V$ $\implies$ $x\in A$ which means $x\in \bar A$.

Is my argument correct? If it fails what should I take care of? Thank you very much.

share|cite|improve this question
The implication $x \in A + V \Rightarrow x \in A$ doesn't hold. – J. J. Jun 14 '12 at 10:02
@J.J : How can i argue to give a reasonable answer ? – Theorem Jun 14 '12 at 10:11
up vote 3 down vote accepted

For the other direction it's helpful to remember that $x \in \bar{A}$ if and only if every neighbourhood $U$ of $x$ intersects $A$. So let $U$ be an arbitrary neighbourhood of $0$, whence $x + U$ is an arbitrary neighbourhood of $x$. If $x \in \bigcap (A + V)$, then in particular $x \in A - U$ so that we can write $x = a - u$ for some $a \in A$, $u \in U$. Thus $a \in x + U$ and we see that $x + U$ intersects $A$.

share|cite|improve this answer
Can you tell me why should $x\in A-U$ ? – Theorem Jun 14 '12 at 10:31
@Theorem: $-U$ is a neighbourhood of $0$ and hence $A-U$ is included in the intersection. – J. J. Jun 14 '12 at 10:47
I am getting confused with complements and addition :(. – Theorem Jun 14 '12 at 10:52

We assume that $x\in A+V$ for all neighborhood $V$ of $0$. Let $W$ a neighborhood of $x$. We have to show that $W\cap A\neq\emptyset$. We can write $W=x+W'$, where $W'$ is a neighborhood of $0$. $x\in A+W'$ hence $x=a+w'$ where $a\in A$ and $w'\in W'$. We have $x-w'=a\in A$ and $x-w'\in W$ if we assume $W$ symmetric (without loss of generality, just substituting $W$ by $W\cap (-W)$.

share|cite|improve this answer

The main ingredients:

  1. Show that $$\bigcap_{V\text{ a nbhd of }0}(A+V)=\bigcap_{V\text{ a symmetric}\atop \text{nbhd of }0}(A+V)\;.$$

  2. Show that if $V$ is a symmetric nbhd of $0$, then $x\in A+V$ iff $A\cap(x+V)\ne\varnothing$.

  3. Show that $x\in\operatorname{cl}A$ iff $A\cap(x+V)\ne\varnothing$ for every nbhd of $0$ iff $A\cap(x+V)\ne\varnothing$ for every symmetric nbhd of $0$.

share|cite|improve this answer

We need to recall the following:

Let $X$ be a topological space , $A\subseteq X$ and $x\in X.$ Then $x\in \overline{A}$ iff for any nbd $U$ of $x$, $U\cap A \ne \varnothing$. Also, if $X$ is a TVS with topology $\tau$, then it is known that $a+G$ and $tG$ belong to $\tau$ whenever $a\in X$, $G\in \tau$ and $t\ne 0$. We call this, the invariant property. We now prove that $$ \overline{A} = \bigcap_{V\in \mathcal{V}}(A+V). $$ where $\mathcal{V}$ denotes the collection of all nbs of 0.

First, let $x\in \overline{A}$ and let $V\in \mathcal{V}$. Then $-V$ is nbd of 0. Hence, $x+(-V)=x-V$ is nbd of $x$. Thus, $(x-V)\cap A \ne \varnothing$. Choose $y\in x-V$ and $y\in A$. Then it follows that $x\in A+V$. Hence, $x\in A+V$ for every $V\in \mathcal{V}$. This implies that $x\in \displaystyle{\bigcap_{V\in \mathcal{V}}(A+V)}$.

Next, let $x\in \displaystyle{\bigcap_{V\in \mathcal{V}}(A+V)}$. We want to show that $x\in \overline{A}$. Let $U$ be any nbd of $x$. Then $-x+U$ is nbd of 0 and so is $-(x+U)=x-U$. Hence, $x-U\in \mathcal{V}$. Thus, $x\in A+(x-U)$. Hence, there exist $a\in A$ and $u\in U$ such that $x=a+(x-u)$. This implies that $a=u$. Hence, $U\cap A \ne \varnothing$. Thus, $x\in \overline{A}$. This completes the proof.


share|cite|improve this answer

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.