MathOverflow is a question and answer site for professional mathematicians. 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

More generally, suppose $S$ is a subspace of a Hilbert space $H$ that contains an orthonormal basis of $H$ (For example- the Schwartz space inside $L^2(\mathbb{R}^n)$). If $A:S \rightarrow S$ is symmetric, is $A$ necessarily essentially self-adjoint? That is, does $A$ have a unique self-adjoint extension?

This seems like it would be a standard theorem if it were true, and my inability to find such a statement on, say, Wikipedia, suggests that it is probably false.

This seems elementary- but my first attempts at a proof have not been successful. Maybe somebody knows of a good counterexample?

share|cite|improve this question
Note that by the Gram-Schmidt, every dense subspace $S$ contains an orthonormal basis. Thus here the operator $A$ is just symmetric, densely defined, with $A(S)\subset A$, and the standard counterexamples work (e.g. $i d/dx$ on the half-line). Or am I missing something? – Pietro Majer Nov 5 '12 at 7:50
up vote 5 down vote accepted

One needs additional assumptions to obtain essential self-adjointness. In particular, it is sufficient that S contains a dense set of analytic vectors. The whole subject is explained very well in Vol.2 of "Methods of Modern Mathematical Physics" by Reed and Simon.

share|cite|improve this answer
Thanks! I'll check it out. – Alex Zorn Nov 5 '12 at 15:44

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.