Primitives (Antiderivatives)
BasisPrerequisites
Differentiation takes a function and produces its rate of change. Primitives reverse the process: you are given a rate of change and asked to reconstruct the original function. This inverse problem is the starting point for integration theory, and it arises naturally whenever you compute distance from velocity, displacement from acceleration, or a quantity from its known rate of growth.
Definition
Definition. Let be defined on an interval . A differentiable function is called a primitive (or antiderivative) of on if
Examples.
- is a primitive of on , since .
- is a primitive of on .
- is a primitive of on .
- The function (the sign function) has no primitive on any interval containing , because a derivative cannot have a jump discontinuity (by Darboux’s theorem).
Primitives differ by a constant
Once you have one primitive , you get infinitely many: is also a primitive for every constant . The following theorem says these are the only primitives.
Theorem. If and are both primitives of on an interval , then for some constant .
Proof. Set . Then for every ,
Apply Lagrange’s Mean Value Theorem to on any sub-interval : there exists with
Since were arbitrary, takes the same value at every two points of , so is constant on .
Why the interval hypothesis matters. On a disconnected domain (a union of two disjoint open intervals, say), a function with zero derivative need not be globally constant — it can take different constant values on each connected component. Primitives on disconnected domains are unique only up to one constant per component. This checkpoint assumes is an interval (connected).
The indefinite integral
The notation for the family of all primitives of is the indefinite integral:
where is any one primitive of and is an arbitrary constant. The symbol is read “the indefinite integral of … with respect to ”.
The is not optional decoration: it records the fact that there is a whole family of primitives, all differing by a constant. Dropping it would mean claiming a specific function rather than the general family.
Indefinite vs. definite integral
These two uses of the integral sign refer to fundamentally different objects:
| Indefinite integral | Definite integral | |
|---|---|---|
| Result | A family of functions | A real number |
| Depends on | The integrand and the variable | The integrand and the bounds |
| Arbitrary constant | Yes — is undetermined | No — the value is fixed |
| Requires | A primitive to exist | Riemann integrability of |
The Newton–Leibniz formula (proved in the next checkpoint) connects them: if is continuous and is any primitive of , then .
Basic table of primitives
The following primitives follow directly from differentiating the right-hand side. Each holds on an interval where the right-hand side is defined.
| Notes | ||
|---|---|---|
| () | All , if | |
| ; one constant per sign of | ||
| () | special case of with |
You can verify each entry by differentiating the right-hand side. For instance, for , and .
Linearity of the indefinite integral
Because differentiation is linear, so is taking primitives.
Theorem (Linearity). If is a primitive of and is a primitive of on , then for any constants ,
Proof. .
Example. Compute .
By linearity and the table:
Example. Compute for .
Simplify the integrand first:
Then by linearity:
Summary
- A primitive (antiderivative) of on an interval is a differentiable function with on .
- Any two primitives of on differ by a constant: if , then for some . This follows from Lagrange’s MVT applied to .
- The indefinite integral denotes the entire family of primitives; the is essential.
- The indefinite integral is a family of functions; the definite integral is a number. The Newton–Leibniz formula connects them.
- Linearity: (up to the constant).
- The standard table of primitives — powers, , , , , inverse trig — is built by reversing known differentiation formulas.