Consider the following question:
Let $A$ be a normed space containing a closed subset $B\subseteq A$ and a dense subset $D\subseteq A$. Is $B \cap D$ necessarily a dense subset of $B$?

My conclusion is that it need not hold in general. To see this, take $A=\mathbb{R}$ (with the standard metric), $B=\{\pi\}$ and $D=\mathbb{Q}$. Then $B\cap D = \{\pi\} \cap \mathbb{Q} = \emptyset$, the closure of which is $\emptyset$. Therefore $B\cap D$ is not dense in $B$.

I am particularly interested in understanding what happens if we add the assumption that $A$ is a (unital) C*-algebra, $B$ is a sub-C*-algebra of $A$ (hence closed) and $D$ is a dense $*$-subalgebra of $A$. Is the answer affirmative then? Can anyone come up with a counterexample?


Take $A = L^{\infty}[0,1]$ with pointwise multiplication and let $B = C[0,1]$ be the closed $C^{\ast}$-subalgebra of continuous functions. The $\ast$-subalgebra $D \subset A$ consisting of the simple functions (finite linear combinations of characteristic functions of measurable sets) is dense in $L^{\infty}[0,1]$ but $D \cap B = \{\text{constant functions}\}$ is as far from dense in $B$ as it gets. Note that all $*$-algebras are commutative and unital here.

  • $\begingroup$ That's a nice example! $\endgroup$ – Nate Eldredge Oct 17 '11 at 2:23

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.