Let $V$ be a vector space with inner product $(\phi,\psi)$ antilinear in the second argument - not necessarily a Hilbert space. Let $\Phi$ be an antilinear functional on $V$.

What are the precise (necessary and sufficient) conditions that must be imposed on $\Phi$ that will guarantee the existence of a sequence or net of vectors $\phi_k\in V$ such that $$\Phi(\psi)=\lim_{k} (\phi_k,\psi) ~~~\forall~ \psi\in V?$$ The answer is given by the Riesz representation theorem when $V$ is a Hilbert space, but I am interested in the general case.

  • 2
    $\begingroup$ Do you want that limit to just hold pointwise, or do you want something stronger? If you just want it pointwise, the Riesz representation theorem has nothing to do with this--you can find such a net for any $\Phi$ (because you can find a $\phi$ that works for any finite set of $\psi$s). On the other hand, if you restrict to sequences, $\Phi$ does have to be bounded if $V$ is complete, but this is not obvious (it follows from the Banach-Steinhaus theorem). $\endgroup$ Aug 24, 2015 at 15:18
  • $\begingroup$ @EricWofsey: The limit is a limit of numbers, hence there is no weaker or stronger version. How to construct the net? $\endgroup$ Aug 24, 2015 at 16:08
  • 2
    $\begingroup$ I think Eric is asking whether you wanted some kind of uniformity with respect to $\psi$ in that limit. (It can be interpreted as a limit of functions of $\psi$ rather than just a limit of numbers.) $\endgroup$ Aug 24, 2015 at 22:56
  • $\begingroup$ @QiaochuYuan: I see. Although asking for more seems interesting, the pointwise construction is enough for me. $\endgroup$ Aug 25, 2015 at 9:14

1 Answer 1


Such a net exists for any $\Phi$ (in fact, there is a canonical such net). First, note that if $F\subseteq V$ is a finite-dimensional subspace, then there is a unique $\phi_F\in F$ such that $\Phi(\psi)=(\phi_F,\psi)$ for all $\psi\in F$. The collection of such $F$ form a directed set under inclusion, and the net $(\phi_F)$ will have the desired property.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge that you have read and understand our privacy policy and code of conduct.

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