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

A step in the proof of Proposition 3.3 (p.6) in this paper argues the following:

Let $M^n$ be a topological sphere with a Riemannian metric $g$. Now assume further $(M^n,g)$ is locally symmetric, then it must be the standard round sphere by some results from a paper by Borel. I just can't follow Borel's paper, it might be something like 'For the sphere in the form $G/H$ with $G$ the isometric group, even-dim sphere has Euler number 2, which after some arguments can determine G and H'. But I am not sure how it really works or what happens for odd-dim sphere. Maybe the whole question follow from the classification theory for symmetric space of compact type, which I know almost zero.

Now I was wondering if a geometric way possible, or if anyone knows how exactly Borel's aurgment works.

share|cite|improve this question
Being simply connected, it must be globally symmetric, a first step. – Ben McKay Sep 21 '13 at 7:06
up vote 0 down vote accepted

Perhaps the following is not the quickest way, but it works. A locally symmetric simply-connected manifold is symmetric, and hence homogeneous. Homogeneous simply-connected rational cohomology spheres are all classified. References are listed in Section 2.4 in the thesis (‎) of Oliver Bletz-Siebert. These results give a list of pairs $(G,H)$ such that $G/H$ is homeomorphic to a sphere. One can go down the list, compare it with the classification of symmetric spaces, and see that the only possibility is $SO(n+1)/SO(n)$, the round sphere.

share|cite|improve this answer
Excellent! It works. – Bo_Y Sep 23 '13 at 20:07

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.