The Problem We Must Solve
In the previous lesson, we learned how to integrate simple functions.
If:
$$\phi=\sum_{i=1}^{n}a_i\mathbf{1}_{A_i}$$
then:
$$\int\phi,d\mu=\sum_{i=1}^{n}a_i\mu(A_i)$$
This is straightforward because the function only takes finitely many values.
However, most functions are not simple.
Consider:
$$f(x)=x^2$$
on:
$$[0,1]$$
This function takes infinitely many values.
How should we define:
$$\int f,d\mu$$
in a way that is consistent with our integration of simple functions?
This was the central problem solved by Lebesgue.
Lebesgue’s Fundamental Idea
Instead of trying to integrate a complicated function directly, approximate it by simpler functions.
Specifically:
Approximate measurable functions using increasing sequences of simple functions. We then define the integral of the original function as the limit of the integrals of those approximations.
This idea is the heart of Lebesgue integration.
Approximating from Below
Suppose:
$$f:X\to[0,\infty]$$
is a nonnegative measurable function.
We construct simple functions:
$$\phi_1,\phi_2,\phi_3,\ldots$$
such that:
$$0\le\phi_1\le\phi_2\le\phi_3\le\cdots\le f$$
and:
$$\phi_n(x)\to f(x)$$
for every:
$$x$$
The sequence climbs upward toward the function.
Visual Intuition
Imagine the graph of:
$$f(x)=x^2$$
on:
$$[0,1]$$
Instead of using rectangles that extend above and below the curve (as in Riemann integration), we build a staircase that always remains underneath.
The staircase becomes finer and finer.
Eventually it approaches the curve perfectly.
The area under these staircases converges to the true integral.
Definition of the Lebesgue Integral
Let:
$$f:X\to[0,\infty]$$
be measurable.
The Lebesgue integral is defined as:
This means:
Take all simple functions lying below:
$$f$$
Integrate each one.
Then take the largest possible value.
That supremum is the integral.
Why the Supremum?
Every simple function below:
$$f$$
provides a lower estimate of the true area.
Different approximations give different estimates.
The supremum captures the best possible lower approximation.
This guarantees that the integral depends only on:
$$f$$
and not on any particular approximation sequence.
Example 1: Constant Function
Suppose:
$$f(x)=5$$
on a measurable set:
$$A$$
with:
$$\mu(A)=3$$
Then:
Exactly as expected.
Example 2: Indicator Function
Suppose:
$$f=\mathbf{1}_A$$
Then:
This agrees with our earlier definition.
Thus the Lebesgue integral extends the integration of simple functions without changing it.
Example 3: A Simple Function
Suppose:
where:
$$\mu(A)=3$$
and
$$\mu(B)=5$$
Then:
Again, nothing new occurs.
The new machinery reproduces the old result.
Integrability
Not every measurable function has a finite integral.
A measurable function is called integrable if:
$$
\int |f|,d\mu
<
\infty
$$
This condition becomes extremely important later.
Integrable functions form one of the most important spaces in analysis:
$$L^1$$
Why Absolute Values Appear
Consider:
$$f(x)=1000$$
and
$$g(x)=-1000$$
Large positive and negative values can cancel.
Without absolute values, a function might appear harmless even though its positive and negative parts are enormous.
The condition:
$$
\int |f|,d\mu
<
\infty
$$
ensures true control over the function.
Integrating Functions with Negative Values
So far we have assumed:
$$f\ge0$$
What about functions that can be negative?
We split:
$$f=f^+-f^-$$
where:
$$f^+=\max(f,0)$$
and:
$$f^-=\max(-f,0)$$
These are called the positive and negative parts.
Then:
provided both sides are finite.
Example
Consider:
$$f(x)=x$$
on:
$$[-1,1]$$
The positive and negative contributions balance.
Thus:
Lebesgue integration agrees with ordinary calculus here.
Basic Properties
The Lebesgue integral satisfies several crucial properties.
Linearity
If:
$$a,b\in\mathbb{R}$$
then:
Positivity
If:
$$f\ge0$$
then:
$$
\int f,d\mu
\ge0
$$
Monotonicity
If:
$$f\le g$$
then:
$$
\int f,d\mu
\le
\int g,d\mu
$$
These properties become the foundation of modern analysis.
Why Lebesgue’s Definition Is Revolutionary
Riemann integration asks:
How do we partition the domain?
Lebesgue integration asks:
How are the values of the function distributed?
This shift may seem small, but it changes everything.
Lebesgue integration can handle:
- discontinuous functions
- probability distributions
- stochastic processes
- Fourier analysis
- functional analysis
- quantum mechanics
far more effectively than Riemann integration.
The First Appearance of Spaces
Once integration exists, we can measure the size of functions.
For example:
$$
\int |f|,d\mu
$$
measures total magnitude.
Later we will study:
$$
\int |f|^p,d\mu
$$
which leads to:
$$L^1,;L^2,;L^p$$
spaces.
These spaces eventually become the natural habitat of Hilbert spaces, operator theory, and noncommutative geometry.
Why Alain Connes Cares About Integration
Classical measure theory gives:
$$
\int f,d\mu
$$
for measurable functions.
Connes asks:
What if functions are replaced by operators?
Can we still integrate?
Can we still define size?
Can we still define geometry?
The answer leads to traces, spectral invariants, and noncommutative integration.
Much of Connes’ work can be viewed as extending the Lebesgue integral beyond classical spaces.
Key Concepts Learned
By the end of this lesson you should understand:
- Nonnegative measurable functions are approximated by simple functions.
- The Lebesgue integral is defined via a supremum over simple-function approximations.
- Indicator and simple functions fit naturally into the theory.
- General functions are handled using positive and negative parts.
- Integrability means:
- The integral is linear, positive, and monotone.
- Lebesgue integration generalizes and vastly extends Riemann integration.
Looking Ahead
In the next lesson:
Lesson 10: Properties of the Lebesgue Integral
we will develop the powerful machinery that makes Lebesgue integration useful, including linearity, continuity properties, comparison principles, and the first major convergence phenomena that distinguish measure theory from ordinary calculus.
Leave a Reply