3

In building up the theory of Lebesgue integration, at some point in the exposition we get something like:

For a non-negative measurable function $f$, the integral is defined as:

$$\int_E f \, d\mu = \sup\left\{\,\int_E s\, d\mu : 0 \le s \le f,\ s\ \text{simple}\,\right\}.$$

Now the fact that $f$ is measurable is usually made as an assumption, from the expositions that I have seen. However, I don't see why $f$ needs to be measurable. The set is well-defined and non-empty for any non-negative function, why is $f$ assumed to be measurable? Seems extraneous.

Davide Giraudo
  • 172,925
iMath
  • 31

2 Answers2

5

The definition makes sense whether $f$ is measurable or not. In fact for nonmeasurable $f$ it defines the lower integral of $f$. The lower integral does not enjoy many of the properties of the integral, though.

Umberto P.
  • 52,165
0

I could be missing something but this is based on the definition of integrals of simple functions and then the way you define integral of simple functions is like $\int (\sum a_i1_{A_i}) := \sum a_i \mu(A_i)$

If $s = \sum a_i1_{A_i}$ is not measurable then how are you going to do $\mu(A_i)$ ?

Or do you mean even if each simple $s$ is measurable, the $f$ in question actually need not be measurable?

BCLC
  • 13,459