3
$\begingroup$

What are the complex rational homogenous spaces $G/P$ ($G$ a semi-simple complex Lie group, $P$ a parabolic subgroup) such that the set of real points $(G/P)(\mathbb R)$ is a (compact) riemannian symmetric space?

This is certainly well-known by the experts, but I'm not one of them...

Thanks for any help!

$\endgroup$
2
  • $\begingroup$ Could you clarify: what do mean by "is"? $(G/P)(\mathbf{R})$ is a set, a topological space, a real analytic manifold... A compact riemannian symmetric space is a metric space, a Riemannian manifold, a topological space, a real analytic manifold... $\endgroup$
    – YCor
    Aug 25, 2014 at 10:41
  • $\begingroup$ My confusion is what are "real points"? There is no ${\mathbb R}$-structure on $G/P$, a priori, unless $G$ and $P$ are defined over $\mathbb R$. $\endgroup$ Aug 25, 2014 at 10:52

1 Answer 1

2
$\begingroup$

Compact Riemannian symmetric spaces admitting a Lie group of diffeomorphisms $G$ properly containing the isometry group are essentialy (up to covers) the symmetric $R$-spaces, which are of the form $G/P$. This is a celebrated theorem of Nagano [Nag, Theorem 3.1].

The list of these spaces is e.g. in the appendix of Isothermic submanifolds of symmetric $R$-spaces (pdf).

[Nag] Tadashi Nagano, Transformation groups on compact symmetric spaces, Trans. Amer. Math. Soc. 118 (1965), 428–453. MR0182937

$\endgroup$
6
  • $\begingroup$ Why the downvote? It's not entirely clear to me what the OP means by $(G/P)(\mathbb{R})$, but the Nagano's theorem should bring him closer to answer in any case. $\endgroup$ Aug 26, 2014 at 7:53
  • $\begingroup$ Vit Tucek's answer is a precise and relevant answer to my confusing question! Thanks a lot. $\endgroup$
    – Lucien
    Aug 26, 2014 at 13:25
  • $\begingroup$ A question concerning the terminology: to what refers the "R" in "R-space"? $\endgroup$
    – Lucien
    Aug 28, 2014 at 13:00
  • $\begingroup$ I have no idea. And it's pretty hard to google. :) $\endgroup$ Aug 28, 2014 at 14:19
  • $\begingroup$ R is for "racine". Terminology coined by Jacques Tits, I believe. $\endgroup$ Nov 23, 2014 at 0:53

Your Answer

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

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