What Is an Integral? A Plain Explanation

Two questions, one answer

An integral answers two questions that turn out to be the same question in disguise. The first: given a rate of change, what was the original quantity? If you know a car's velocity at every instant, integrating it recovers distance travelled. The second: what is the area between a curve and the x-axis? Both questions are answered by the same operation, and the link between them — that area-so-far is exactly the quantity whose rate of change is the curve's height — is one of the more surprising facts in mathematics.

That link is called the fundamental theorem of calculus. It is the reason integration and differentiation are taught together: each one undoes the other.

The antiderivative: running differentiation backward

The most direct way to think about an integral is as a reverse operation. Differentiating x³ gives 3x². Integrating 3x² should therefore give back x³ — and it does. Finding an antiderivative means asking 'what, when differentiated, gives this?'

Why every indefinite integral carries a +C

Differentiation destroys constants. The derivative of x³, x³ + 5 and x³ − 200 is the same 3x² in every case, because a constant's rate of change is zero. Run the process backward and that information cannot be recovered — 3x² could have come from any of infinitely many functions, all differing by a constant.

The +C stands in for that unknown constant. It is not a formality tacked on by convention; it is an honest acknowledgement that the reverse operation has thrown away one piece of information the forward operation destroyed. An indefinite integral is really a whole family of functions, all with identical slope at every point, stacked vertically on top of one another.

The definite integral: area, with a sign

Add two limits to an integral and the question changes from 'what function has this derivative' to 'what is the signed area between the curve and the x-axis, from one point to another'. The fundamental theorem of calculus connects the two: find any antiderivative F, then the area is F(b) − F(a). The +C is irrelevant here — it appears in both F(b) and F(a) and cancels exactly.

Signed matters: area below the x-axis counts as negative. A curve that dips below zero and back up over the interval will have some of its area cancel rather than add, which is exactly right if the integral represents something like net displacement rather than total distance travelled.

Where the area idea comes from: adding infinitely many slices

The area interpretation is built by approximation. Slice the region under a curve into thin rectangles, each with height equal to the curve at that point and a small width. Add up all the rectangle areas and you get an estimate — too high if the curve is rising across each slice, too low if it is falling.

• Make the rectangles narrower and there are more of them, but each one fits the curve more closely.
• Let the width shrink toward zero and the number of rectangles grow without bound — this is the same limiting process that defines a derivative, run on area instead of slope.
• What the sum approaches in that limit is defined to be the integral. This construction is called a Riemann sum, and it is what a definite integral literally means before the fundamental theorem gives a shortcut for computing it.

The practical value of the fundamental theorem is exactly that it replaces this infinite sum with two evaluations of an antiderivative — nobody adds up rectangles by hand once the theorem is available.

What the sign and size of an integral tell you

A positive definite integral means the curve sat mostly above the x-axis over that interval; a negative one means it sat mostly below. A large magnitude means either a wide interval or a curve that stood far from the axis — the integral does not distinguish between the two on its own.

If the original quantity being integrated is a rate — velocity, for instance — then the definite integral is a net change: net displacement, not total distance, since a negative velocity subtracts rather than adds.

Why some expressions resist integration

Every function that is built from ordinary pieces has a derivative you can always find by mechanically applying rules. Integration has no such guarantee. Some integrals — e^(x²) is a famous example — have no antiderivative expressible in elementary functions at all, even though the function itself is perfectly ordinary.

More commonly, an antiderivative does exist but finding it needs a technique beyond the basic rules — integration by parts for a genuine product of two functions of x, or a substitution that is not simply linear. Recognising that a problem needs one of these techniques, rather than guessing, is itself part of learning to integrate.

Questions about what is an integral