If every Jordan curve $J \subset D$ is null-homotopic in a domain $D \subset \mathbb{R}^2$, then is it clear that every closed curve $C \subset D$ is also null-homotopic in $D$?
Apparently the set $\mathbb{R}^2 \setminus C$ is an union of countably many Jordan domains $D_i$, but it's not clear to me that the result for every $\partial D_i$ is enough. What is the right approach?
EDIT: Before you mofos shut down my question with silly down votes I ask to check out the following link http://faculty.up.edu/wootton/Complex/Chapter8.pdf
There this question is answered with seemingly simple compactness argument. It would be nice if someone of you can explain why those argument are enough?
