If we have a sequence of bounded functions $f_{n}$ converging almost everywhere to another bounded function $f$ in a finite measure space such that
$$ |f_{n}(t)| \leq c$$ for some constant $c$. Then
$$ \int |f_{n}(t) - f(t)|^{2} d\mu$$ goes to zero.
Since we are in a finite measure space, the constants are integrable and hence DCT applies. However, it only gives the result that
$$ \int f_{n}(t)d\mu \rightarrow \int f(t)d\mu$$ and $$ \int |f_{n}(t) - f(t)| d\mu \rightarrow 0$$. How do we prove the convergence in $L^{2}$ norm?
Here is what I tried:
$$ \int |f_{n}(t) - f(t)|^{2} d\mu = \int (f_{n}-f)(\overline{f_{n}}-\overline{f})$$ $$ = \int |f_{n}|^{2} - \int f\overline{f_{n}} - \int f_{n}\overline{f} + \int |f|^{2} $$ which goes to zero by DCT (as $f_{n}\overline{f} \rightarrow |f|^{2}$ and so on). Am I right in all this or am I missing something? I am apprehensive because I intuitively feel that pointwise convergence should not imply $L^{2}$ convergence.