Take the 2-minute tour ×
Mathematics Stack Exchange is a question and answer site for people studying math at any level and professionals in related fields. It's 100% free, no registration required.

Let $X$ be a topological space and $A$ a subset of $X$. On $X\times\{0,1\}$ define the partition composed of the pairs $\{(a,0),(a,1)\}$ for $a\in A$, and of the singletons $\{(x,i)\}$ if $x\in X\setminus A$ and $i \in \{0,1\}$.

Let R be the equivalence relation defined by this partition, let Y be the quotient space $[X \times \{0,1\}]/R$ and let $p:X \times \{0,1\} \to Y$ be the quotient map.

(a) Prove that there exists a continuous map $f:Y \to X$ such that $f \circ p(x,i)=x$ for every $x\in X$ and $i \in \{0,1\}$.

(b) Prove that Y is Hausdorff if and only if X is Hausdorff and A a closed subset of X.

So far I have done part (a), but am struggling with part (b). I am thinking I will have to use the fact that if a quotient space $Z/R$ is Hausdorff then the graph of the equivalence relation in $Z \times Z$ is closed. However, I am not positive. If anyone could even just help me get started somewhere on this question I would be grateful, thank you.

share|improve this question

3 Answers 3

The approach that you suggest is indeed reasonable. Let $Z=X\times\{0,1\}$, where $\{0,1\}$ has the discrete topology. Let $E$ be the equivalence relation, considered as a subset of $Z\times Z$. Now consider a point $p=\big\langle\langle x,i\rangle,\langle y,j\rangle\big\rangle\in E$; what does $p$ actually look like? Clearly we must have $x=y$. If $x=y\in A$, $i$ and $j$ can both be either $0$ or $1$, but if $x=y\in X\setminus A$, then we must have $i=j$. Thus,

$$p\in E\quad\text{ iff }\quad x=y\,\text{ and }\,(x\in A\text{ or }i=j)\;.$$

More usefully,

$$p\in(Z\times Z)\setminus E\quad\text{ iff }\quad x\ne y\,\text{ or }\,(x\in X\setminus A\text{ and }i\ne j)\;:$$

you can use this and the definition of the product topology on $Z\times Z$ to show that $(Z\times Z)\setminus E$ is open (and hence $E$ is closed) iff $X$ is Hausdorff and $A$ is closed.

share|improve this answer

Hints: It actually shouldn't be too difficult to just go through the proof by hand without relying on other theorems.

  • ($\Rightarrow$) If $X$ is not Hausdorff, then there are distinct points $x , y \in X$ which cannot be separated. Show that $p ( x , 0 ) , p ( y , 0 )$ cannot be separated in $Y$. If $A \subseteq X$ is not closed, let $x \in \overline{A} \setminus A$. Show that the points $p ( x , 0 ) , p ( x , 1 )$ (which are distinct) cannot be separated in $Y$.
  • ($\Leftarrow$) This should just be working through the cases. First note that if $x , y \in X$ are distinct, then they can be separated by open sets in $X$, and it is not too hard to separate $p ( x , i ) , p ( y , j )$ for any $i,j \in \{ 0 , 1 \}$. So the real difficulty will be in separating $p ( x , 0 ) , p ( x , 1 )$ for $x \in X$. But note that we only have to worry about this when $p ( x , 0 ) \neq p ( x , 1 )$. When does this happen? (Obvious choices for open sets should jump out at you.)
share|improve this answer

Hint: Show that $p$ is a perfect map, that is a closed continuous surjection such that $p^{-1}(y)$ is compact for each $y\in Y$. These maps preserve many properties of $X$ among which is also Hausdorff'ness.

Here is a proof that perfect maps preserve $T_2$: If $y$ and $z$ are distinct points in $Y$, then $f^{-1}(y)$ and $f^{-1}(z)$ are disjoint compact sets which by $T_2$ can be seperated by disjoint open neighborhoods. $U,V$. The disjoint open neighborhoods around $y$ and $z$ can be obtained by taking the complements of $p(X\backslash U)$ and $p(X\backslash V)$ in $Y$.

I'll leave it to you to show that $p$ is perfect, which should be very easy.

Edit: The closedness of the graph $R$ is not sufficient, it is only necessary condition for $Y$ to be Hausdorff. If the quotient map $p$ is open, then closedness $R$ is enough. Another approach would be to show the compactness of $R$, because this makes $p$ a perfect map, see also Product of quotient map a quotient map when domain is compact Hausdorff?

share|improve this answer

Your Answer

 
discard

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.