MathOverflow is a question and answer site for professional mathematicians. 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


Oops---I found the answer to the first question of mine here on Wikipedia---this is really classic material. I'll leave the question open for a bit, in case someone tells me something interesting for my second question.

(A version of the above was mentioned by Michael at about the same time as my "oops", so I accepted his answer)

Sorry for asking a basic and naive question---if this is textbook material somewhere, please let me know, so that I may close this question.

My question is:

When is a metric space $(X,d)$ isometrically embeddable into some Banach space?


Can one say something "stronger", if we know that the closure of $X$ is actually a convex cone?

(By "stronger" one of the things I mean is whether we can actually obtain an embedding that is easy to compute...)

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

Arbitrarily fix $y_0\in X$. Then, with every $y\in X$ you can associate a bounded continuous function from X to R defined by $$f_y(x)=d(y,x)-d(y_0,x).$$ It is easy to show that $$\max_x |f_y(x)-f_z(x)|=d(y,z),$$ with the maximum assumed if $x=y$ or $x=z$. Hence $X$ is isometrically embedded in the Banach space $C_b(X)$.

share|cite|improve this answer
Ach, I'm having another of those "obvious after seeing the answer" moments! However, this observation surely must be a named theorem somewhere? – Suvrit Aug 22 '11 at 21:44
Your answer answers my question and I will have to ask a more refined question to get at what I am trying! – Suvrit Aug 22 '11 at 21:52
This is also a classical way to construct the completion of a metric space (taking its closure in the above embedding). – Mark Aug 23 '11 at 12:58

One useful embedding is into the "free space" (or "Arens-Eels space") over $X$, which is the predual of the space $Lip(X)$ of Lipschitz functions from the (pointed) metric space $(X,0)$ into the real line that map zero to zero. The free space of $X$ is just the closed span in $Lip(X)^*$ of the pointwise evaluations functionals. This millennium Godefroy and Kalton used this embedding to good effect in their paper Lipschitz-free Banach spaces, Studia Math. 159 (2003), 121–141. The embedding is described and discussed in the book N. Weaver, Lipschitz algebras, World Scientific Publishing Co. Inc., River Edge, NJ, 1999.

share|cite|improve this answer
Thanks for the nice references; Weaver seems to be advancing a viewpoint of metric spaces that I like. – Suvrit Aug 22 '11 at 21:51

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.