Measure is equal to Lebesgue measure

lebesgue-measuremeasure-theoryreal-analysis

Let $\mathscr{M} \subseteq \mathscr{P}(\mathbb{R})$ be the $\sigma$-algebra of Lebesgue measurable subsets of $\mathbb{R}$, and suppose that $\mu :
\mathscr{M} \rightarrow [0, \infty]$
is a measure on $(\mathbb{R}, \mathscr{M})$ with the property that $\mu(I) = \ell(I)$ for all intervals $I$. Prove that $\mu = m$, where $m$ denotes Lebesgue measure.

The question also provides two useful hints: show separately the inequalities $m \leq \mu$ and $\mu \leq m$, and to use the characterisation of Lebesgue measurable sets proven in an earlier question, namely that $E \subseteq \mathbb{R}$ is measurable if and only if for every $\epsilon>0$ there is an open set $U \supseteq E$ such that $m(U \setminus E) \leq \epsilon$.

I'm kind of stuck and can't seem to make any progress, so any help would be great! Thanks.

Best Answer

Any open set in $\mathbb R$ is a countable disjoint union of open intervals. Hence $\mu (U)=m(U)$ for any open set $U$. Now $\mu (E) \leq \mu (U)=m(U) \leq m(E)+\epsilon$. Let $\epsilon \to 0$ to get $\mu (E) \leq m(E)$. The reverse inequality is proved by a similar argument.