0

Isotropy and homogeneity are important notions in cosmology. Isotropy at a point $p$ of the manifold $M$ means for two vectors $V,W$ in the tangent space $T_pM$, there is an isometry of $M$ that takes $V$ to be parallel to $W$. Homogeneity of the manifold means for any two points $p,q$ of $M$, there is an isometry that takes $p$ to $q$. I found in different places the statement that isotropy everywhere implies homogeneity. But how to prove this statement?

Intuitively, by isotropy we can always bring two points closer, for example, by flipping $p$ around a point $p^{'}$ between $p,q$. I am not sure if this line of thought can lead to a rigorous proof, as less technical as possible.

Edit: As pointed out by Janisch and Desteny, we can choose $r$ the mid point of the geodesic connecting the two points $p,q$. The isometry that flips the tangent vector of the geodesic at $r$ toward $p$ will take $p$ to $q$.

  • 1
    Checkout this answer https://math.stackexchange.com/a/448454/631742 – Maximilian Janisch Jan 28 '25 at 09:21
  • 2
    As pointed out, this was already anwered. However, a quick visualization: We can assume the manifold is connected. If you have separate points p,q you can connect them with a (unit speed) geodesic. Now take the middle point on the geodesic. By the isotropy condition you can pivot the manifold around this middle point by 180 degree via an isometry, so that the geodesic (to be precise its image) is mapped onto itself, but in the other direction. Because the isometry maps geodesics to geodesics p will be mapped to q and vice versa. – Desteny Jan 28 '25 at 14:24
  • 1
    CONTINUATION: i.e. your idea is correct, if you take p' to be the middle point of a geodesic connecting the two points. – Desteny Jan 28 '25 at 14:26
  • @Desteny But the statement is isotropy everywhere implies homogeneity; you used isotropy at just one single point (the midpoint) how did that work? – Sanjana Mar 19 '25 at 15:21
  • 1
    @Sanjana Yes, for this particular choice of two points (the endpoints of the geodesic) I used that the manifold is isotropic in the midpoint. But to prove homogeneity, we need that this works for all choice of two points. Therefore, we need that the manifold is isotropic in every midpoint of a geodesic between two given points, which is clearly every point. – Desteny Mar 19 '25 at 16:43
  • @Desteny Thank you for your answer. – liyiontheway Mar 20 '25 at 02:00

0 Answers0