Mathematics Stack Exchange is a question and answer site for people studying math at any level and professionals in related fields. It's 100% free, no registration required.

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 was studying and I got the next doubt:

We suppose that $(X,\|\cdot\|)$ is a Banach space and $C$ it is a convex closed subset of X.

We say that $x\in C$ it is an exposed point of $C$ if $\exists x^*\in X^*$ such that $x$ it is the only one point of $C$ that $x^*(x)=\displaystyle\sup_C x^*$.

We say that $x\in C$ it is an strongly exposed point of $C$ if $\exists x^*\in X^*$ such that for every sequence $(x_n)\subset C$ that verifies $x^*(x_n)\to \displaystyle\sup_C x^*$ then $\|x_n-x\|\to 0$.

It's true that every strongly exposed point it is an exposed point. So I try to view if every exposed point it is an strongly exposed point (it would be false!).

Suppose that $x\in C$ it is an exposed point, then there is $x^*\in X^*$ such that $x^*(x)=\displaystyle\sup_C x^*$. Let $(x_n)\in C$ such that $x^*(x_n)\to \displaystyle\sup_C x^*$. $x_n$ must converge ($x^*$ is continuous), so we suppose that $x_n\to y\in C$ ($C$ is closed). Then, by continuity $x^*(x_n)\to x^*(y)=\displaystyle\sup_C x^*$.

So by the uniqueness, $x=y$ and then $\|x_n-x\|=\|x_n-y\|\to 0$. Then $x$ it is an strongly exposed point.

The question is: what is wrong in my reasoning? There is a counterexample of Lindenstrauss for example (7.73 in the a book of Fabian: Functional Analysis and Infinite-Dimensional Geometry).

Many thanks beforehand.

share|cite|improve this question
You probably mean example 8.73, at least if it is the same edition of the book - see here. – Martin Sleziak May 4 '12 at 12:57
Ah, I refer to 7.73 of another book of Fabian: Banach Space (the basis for...). It's similar. There are counterexamples, so my reasoning is wrong, but I don't know where and it has me baffled. – Shrek May 4 '12 at 13:44
Your problem is with the line "$x_n$ must converge". You have that $x^*(x_n) \to x^*(x)$ for one fixed $x^*$, not for all $x^*$. Even if you had $x^*(x_n) \to x^*(x)$ for all $x^* \in X^*$ (which you do not have), this would be weak convergence and not convergence in the norm topology. – Tom Cooney May 4 '12 at 15:02
Amm, thanks. That was the fail. – Shrek May 4 '12 at 22:35

You have to prove $x_n$ converges to $x$ in norm. You have assumed that $x_n$ converges to $y$. This assumption is wrong. First of all, from the hypothesis (i.e., $x^*(x_n)$ converges to $x^*(x)$) we could only conclude $x_n$ converges to $x$ w. r. to the weak topology on $X$, provided $x_n$ is a bounded (norm) sequence.

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.