Mathematics Stack Exchange is a question and answer site for people studying math at any level and professionals in related fields. 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

Let $G$ be a connected topological group. And $H$ be a discrete subgroup.

Theorem : $\pi_1(g/H)=H$

This is the content in the book Algebraic Topology -Greenberg and Harper

I want to know the proof. In this book, take an open set $U$ around $1$ such that $H\cap U = \emptyset $. And take open set $V$ in $U$ such that $gh^{-1} \in U$ for all $g$, $h \in V$

I cannot understand why this argument is helpful to prove the theorem. 
share|cite|improve this question
I think you need that $G$ is simply connected. – Jason DeVito Nov 13 '12 at 2:26
Yes. Here $G$ is simply connected and $H$ is normal. – HK Lee Nov 13 '12 at 2:56
up vote 2 down vote accepted

There are two steps to the proof:

  1. Show that the projection $p\colon G\to G/H$ is the universal cover of $G/H$.
  2. Show that the group of deck transformations of $p$ is isomorphic to $H$. Since $\pi_1(G/H)$ is isomorphic to the group of deck transformations of $p$, this will prove the theorem.

Assume for the moment we know (1) to be true. Then statement (2) is not so hard to prove. For each $h\in H$, let $L_h\colon G\to G$ be the map $L_h(g) =hg$. Note that $L_{h_1}\circ L_{h_2} = L_{h_1h_2}$, so the set $\{L_h: h\in H\}$ forms a group isomorphic to $H$ under composition. Every $L_h$ is a deck transformation of $p$. On the other hand, suppose $\varphi$ is a deck transformation of $p$. Then $\varphi(1)\in H$, so $L_{\varphi(1)^{-1}}\circ \varphi$ is a deck transformation of $p$ that fixes $1\in G$. The only deck transformation that fixes a point is the identity, so $L_{\varphi(1)^{-1}}\circ \varphi = Id$, and hence $\varphi = L_{\varphi(1)}$. This proves that the group of deck transformations of $p$ is exactly $\{L_h : h\in H\}\cong H$.

Now we have to show (1), i.e., that $p$ is a covering map. Let $U$ and $V$ be as you stated in your question.

Claim: For all $h\in H$ such that $h\neq 1$, the sets $V$ and $hV$ are disjoint.

Indeed, if this were not the case, then there would exist $v,w\in V$ and $1\neq h\in H$ such that $v = hw$. Now we use your construction of $V$ to conclude that $h = vw^{-1}\in U$. Thus $h\in U\cap H = \{1\}$, a contradiction of the assumption that $h\neq 1$. This proves the claim.

Fix $g\in G$, and let $W = p(Vg)$. Note that $W$ contains $p(g)$, and is an open set in $G/H$, since $p^{-1}(W) = \bigsqcup_{h\in H} hVg$ is open (this is the definition of the quotient topology). Moreover, the restrictions $p|_{hVg}\colon hVg\to W$ are homeomorphisms for each $h\in H$, again by the definition of the quotient topology. This is exactly what needs to happen for $p$ to be a covering map.

share|cite|improve this answer

A handy way to see this is to consider the long exact sequence $...\to\pi_1(G)\to\pi_1(G/H)\to\pi_0(H)\to\pi_0(G)\to...$ you find $\pi_1(G)=0$ and $\pi_0(G)=0$ because $G$ is simply connected and $\pi_0(H)=H$ because $H$ is discrete. This just give $\pi_1(G/H)=H$.

share|cite|improve this answer
Thanks for the answer. I edited the post to format it nicely. Check out for a quick reference on mathjax formatting (essentially the same as LATEX) in the future. – TravisJ Apr 21 '15 at 0:46

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.