Just a quick question about the the following theorem

In a locally convex space $X$, every weakly bounded set is originally bounded and viceversa.

Proof: Since every weak neighborhood of $0$ in $X$ is an original neighborhood of $0$, it is obvious from the definition of "bounded" that every originally bounded subset of $X$ is weakly bounded. The converse is the non trivial part of the theorem. Suppose $E \subset X$ is weakly bounded and $U$ is an original neighborhood of $0$ in $X$. Since $X$ is locally convex, there's a convex, balanced, original neighborhood $V$ of $0$ in $X$ such that $\overline{V} \subset U$. Let $K \subset X^*$ be the polar of $V$: $$ K = \left\{\Lambda \in X^* : |\Lambda x | \leq 1 \;\text{for all} \; x \in V \right\} \;\;\;\; (1) $$ We claim that $$ \overline{V} = \left\{x \in X : |\Lambda x| \leq 1 \; \text{for all} \; \Lambda \in K \right\} \;\;\; (2) $$ It is clear that $V$ is a subset of the right side of (2) and hence so is $\overline{V}$, since the right side of (2) is closed. Suppose $x_0 \in X$ but $x_0 \notin \overline{V}$. Theorem 3.7 (with $\overline{V}$ in place of $B$) then shows $\Lambda x_0 > 1$ for some $\Lambda \in K$. This proves (2)

It is clear to me the author is somehow using theorem 3.7 to reach a contradiction. The statement is below for convenience

Suppose $B$ is a convex, balanced, closed set in a locally convex space $X$, $x_0 \in X$, but $x_0 \notin B$. Then there exists $\Lambda \in X^*$ such that $|\Lambda x | \leq 1$ for all $x \in B$, but $\Lambda x_0 > 1$

How exactly is such theorem used to reach the conclusion that $\Lambda x_0 > 1$ for some $\Lambda \in K$? I'm sure it has something to do using the definitions of $K$ and $\overline{V}$ but I can't manage to put all the pieces together.


Let's rewrite the conclusion of Theorem 3.7 with $B$ replaced by $\overline V$:

Suppose $x_0\in X$, but $x_0\notin\overline V$. Then there exists $\Lambda\in X^*$ such that $|\Lambda x|\leq 1$ for all $x\in\overline V$ but $\Lambda x_0>1$.

For the given $\Lambda$, if $x\in V$, then $x\in \overline V$, so that so by construction (i.e. by the conclusion of the theorem) we have $|\Lambda x|\leq 1$. Since $x\in V$ was arbitrary, we have $\Lambda\in K$.

  • 1
    $\begingroup$ The constructions the author uses sometimes to prove the theorems always amaze me. $\endgroup$ Jan 10 '19 at 16:21
  • $\begingroup$ Are you using Rudin's Functional Analysis? $\endgroup$
    – Aweygan
    Jan 10 '19 at 16:22
  • $\begingroup$ Yes, indeed I'm. $\endgroup$ Jan 10 '19 at 16:27

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.