Can a simply-connected topological space be written as the union of the images of all possible loops at a given basepoint?

Let $X$ be a simply-connected topological space, fix $x\in X$, and let $\{\gamma_i\}_{i\in I}$ be the collection of possible loops $\gamma: [0,1] \to X$ with $0\mapsto x$ and $1\mapsto x$ (over some indexing set $I$).

Insofar as the concept of "collection of all possible loops" makes sense, is it accurate to say that $X$ can be represented as the union of the images of each of the loops in the collection?

Not sure how you would notate this, but maybe, $\bigcup_{i \in I} \text{Im} \gamma_i$ would work.

Geometrically this seems true, since you could "catch" any point in $X$ just by defining a loop that goes through that point, and since $X$ is simply connected, all the loops are homotopic (and nullhomotopic), so you can catch all the points with loops in the collection.

But, this is just my geometric intuition: are there any counterexamples to this, or issues with this line of argument?

  • 7
    $\begingroup$ This is true for any path-connected space because for every point $x' \in X$, there is a loop based at $x$ which passes through $x'$. $\endgroup$ May 3 at 22:23

1 Answer 1


Lemma. Let $X$ be a topological space and $x_0\in X$ a point. Then $X$ is path connected if and only if for any $x\in X$ there is a loop at $x_0$ passing through $x$.

Proof. "$\Leftarrow$" an exercise.

"$\Rightarrow$" Since $X$ is path connected then there is a path $\lambda:[0,1]\to X$ such that $\lambda(0)=x_0$ and $\lambda(1)=x$. Then $\lambda *\overline{\lambda}$ is the loop we are looking for, where $\overline{\lambda}(t)=\lambda(1-t)$ and "$*$" stands for path composition. $\Box$

Given a topological space $X$ and a fixed point $x_0\in X$ write down

$$C([0,1],X,x_0)=\{\lambda:[0,1]\to X\ |\ \lambda(0)=\lambda(1)=x_0\text{ and }\lambda\text{ is continuous}\}$$

Then the lemma above leads quite easily to the following:

Corollary. A topological space $X$ is path connected if and only if $$X=\bigcup_{\lambda\in C([0,1],X,x_0)}\text{Im}\lambda$$

And so being simply connected is too strong, path connected is enough.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

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