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

Let $X$ be a locally convex space whose topology is defined by a family of seminorms $\mathcal{P}$. Let $f$ be a linear functional on $X$. Then, I am trying to prove that the following statements are equivalent.

  1. $f$ is continuous.
  2. There are seminorms $p_1,...,p_n\in \mathcal{P}$ and positive scalars $\alpha_1,...,\alpha_n$ such that $$|f(x)|\leq \sum_{k=1}^n\alpha_kp_k(x).$$

I am stuck in proving $1\Rightarrow 2$. Applying the definition, and that $f$ is continuous at $0$, I get that there exist seminorms $p_1,...,p_n$ and $\epsilon_1,...,\epsilon_n>0$, such that whenever $$x\in \bigcap_{k=1}^n\{y:p_k(y)<\epsilon_k\}\Rightarrow |f(x)|<1.$$ But I am not able to conjure up positive $\alpha_1,..,\alpha_n$ such that above inequality holds. A hint will be very appreciated. Thanks.

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

Since $f$ is continuous at $0$, the preimage $U=f^{-1}[(-1,1)]$ is open. By linearity, $0\in U$, so $U$ contains an open neighborhood of $0$. Since sets of the form $$V=\{x\in X: p_1(x),\ldots,p_n(x)<\epsilon\}$$ form a local basis for $0$, we have some such set $V\subseteq U$. Thus we have some $\epsilon>0$ and $p_1,\ldots,p_n\in \mathcal P$ such that $$\sum_{i=1}^n p_i(x)<\epsilon\implies p_1(x),\ldots,p_n(x)<\epsilon\implies |f(x)|<1$$ and for any $x\in X$ we have $$\sum_{i=1}^n p_i\left(\frac{\epsilon}{2\sum_{i=1}^n p_i(x)}x\right)=\frac{\epsilon}{2\sum_{i=1}^n p_i(x)}\sum_{i=1}^n p_i\left(x\right)=\epsilon/2$$ thus $$\frac{\epsilon}{2\sum_{i=1}^n p_i(x)}|f(x)|=\left|f\left(\frac{\epsilon}{2\sum_{i=1}^n p_i(x)}x\right)\right|<1$$ which gives us $$f(x)\leq \frac{2\sum_{i=1}^n p_i(x)}{\epsilon}=\sum_{i=1}^n\frac{2}{\epsilon}p_i(x).$$

Edit: If $\sum\limits_{i=1}^n p_i(x)=0$ for some $x\in V$, then for any $\lambda\in\mathbb R$ we have $$\sum\limits_{i=1}^n p_i(\lambda x)=|\lambda| \sum\limits_{i=1}^n p_i(x)=0$$ thus $\lambda x\in V$, so $|\lambda||f(x)|=|f(\lambda x)|<1$. Since $\lambda$ is arbitrary, this means $|f(x)|=0$.

share|cite|improve this answer
Ohk, so the key step was to choose $p_i$'s in such a manner that $\sum_ip_i(x)<\epsilon$. Thanks for the help. – Nirakar Neo Jul 3 '12 at 4:10
just a small doubt - what if $\sum_{i=1}^np_i(x)=0$? how do we manage that? – Nirakar Neo Jul 3 '12 at 14:39
@NirakarNeo Good point. Editing to fix. – Alex Becker Jul 4 '12 at 0:44

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.