Why is the $\epsilon-\delta$ definition of continuity not the other way around

calculuscontinuityreal-analysis

The intuition often given for continuity is that as $x$ gets arbitrarily close to a point $c$, $f(x)$ gets close to $f(c)$. Just looking at this picture, why do we not arbitrarily bound $|x-c|$ and see if a corresponding bound can be placed on $|f(x) – f(c)|$? What is the reasoning behind doing this the other way around?

Worded differently, why do we not have for $\delta > 0$, there exists some $\epsilon > 0$ such that
$$|x-c| < \delta \Rightarrow |f(x) – f(c)|< \epsilon$$

Best Answer

This definition would not be very useful. For example, any bounded function would be continuous. Take $f\colon\Bbb R\to\Bbb R$ where $f(x)=1$ if $x\in\Bbb Q$ and $f(x)=0$ if $x\notin\Bbb Q$. For any $c\in\Bbb R$ and any $\delta>0$, we have $|x-c|<\delta\implies |f(x)-f(c)|\le 1$. So $f$ is continuous at every point.

Related Question