A result of Helgason's is that any selfintersecting geodesic in a Riemannian globally symmetric space is simple and closed. To what extent does this generalise to Riemannian naturally reductive homogeneous spaces with the canonical connection?
Every geodesic loop on a compact homogeneous space is a closed geodesic. Let c be such a loop c(L)=c(0). Take (n1) Killing vector fields $X_i$ whose value at p are a basis of $c'(0)^\perp$. Then $X_i$ restricted to c is a Jacobi field and hence $<X_i(c(t),c'(t)>$ is linear. Since X has bounded length, $<X_i(c(t),c'(t)>=0$ and hence $c'(L)=c'(0)$. I think this may be also true if M is not compact.
For a homogeneous metric there is no reason for a geodesic loop at a point to be necessarily a closed geodesic. Counterexamples can probably be constructed already using metrics on the 3sphere constructed by rescaling the round metric along the fiber of the Hopf fibration.

$\begingroup$ Thanks for the answer, Katz. However, what I really want to know is whether every such homogeneous space, which is not symmetric, admits a selfintersecting geodesic. $\endgroup$ – Oliver Jones Dec 10 '15 at 8:24

$\begingroup$ No, not every homogeneous space will have a selfintersecting geodesic. Take for example your favorite simply connected space of nonpositive curvature. You can certainly pick one that's not symmetric. $\endgroup$ – Mikhail Katz Dec 10 '15 at 8:52

$\begingroup$ Sorry, I should have said that the space must have positive curvature. Are there any examples in this class? $\endgroup$ – Oliver Jones Dec 10 '15 at 9:16
It turns out that Helgason's result does extend to naturally reductive homogeneous spaces. One reference for this is Wolfang Ziller's paper "Closed geodesics on homogeneous spaces", Math. Z. 152, 1976.