What Is Rolle's Theorem?
Rolle's theorem is the simplest exact statement about where derivatives must vanish: if f is continuous on [a, b], differentiable on (a, b), and takes equal values at the endpoints, f(a) = f(b), then
some c in (a, b) satisfies f′(c) = 0
A smooth curve that starts and ends at the same height must turn around somewhere, and at the turnaround the tangent is horizontal. Named for Michel Rolle (1691), the theorem is the seed from which the Mean Value Theorem grows by tilting, and through it most of differential calculus's global structure. This calculator verifies all three hypotheses, computes f′ symbolically, solves f′(c) = 0 on the interval, and draws each horizontal tangent on the graph.
How to Use the Rolle's Theorem Calculator
Enter f and an interval whose endpoint values match. The steps report the continuity scan, the endpoint comparison f(a) versus f(b), the symbolic derivative, and every interior solution of f′(c) = 0 with its function value. If the endpoint values differ, the tool says so and points you to the Mean Value Theorem, which handles tilted chords with slope (f(b) − f(a))/(b − a) instead of 0. A convenient way to manufacture valid intervals: take a and b to be two roots of f, as the parabola chip does, since equal values of zero certainly match.
Worked Example
Take f(x) = x³ − 3x on [−√3, √3]. Both endpoints give f = 0, the function is a polynomial (continuous and differentiable everywhere), so Rolle's theorem applies. The derivative is
f′(x) = 3x² − 3 = 0 ⇒ x = ±1
Both lie inside the interval: the theorem promised at least one flat point and this curve delivers two, the local maximum at (−1, 2) and the local minimum at (1, −2). The graph shows both horizontal tangents. The sine chip gives the cleanest single instance: sin(0) = sin(π) = 0 forces a flat point, and it lands exactly at the crest x = π/2.
Why the Theorem Is True
The proof is a two-line argument built on the Extreme Value Theorem. A continuous f on [a, b] attains a maximum and a minimum. If both extremes occur at the endpoints, then (since f(a) = f(b)) the max equals the min and f is constant, making f′ = 0 everywhere. Otherwise some extreme value occurs at an interior point c, and Fermat's theorem says an interior extremum of a differentiable function has f′(c) = 0. Every hypothesis earns its keep: continuity powers the extreme value step, differentiability powers Fermat's step, and equal endpoints force the extremum inward. The classic counterexample for differentiability is |x| on [−1, 1]: equal endpoints, continuous, yet no horizontal tangent anywhere, because the would-be turnaround at 0 is a corner where f′ does not exist.
Rolle's Theorem as a Root-Counting Tool
Read contrapositively, the theorem bounds how many roots a function can have: between any two roots of f lies at least one root of f′. So if f′ never vanishes on an interval, f has at most one root there; if f′ has k roots, f has at most k + 1. This is how one proves x³ + x − 1 = 0 has exactly one real solution: existence from the Intermediate Value Theorem, uniqueness because f′ = 3x² + 1 > 0 always. Applied repeatedly, the argument shows a degree-n polynomial has at most n real roots, and it underlies the theory of Sturm sequences that computer algebra systems use to count roots exactly.
Common Mistakes to Avoid
- Forgetting to check f(a) = f(b). It is the theorem's signature hypothesis. Unequal endpoints call for the Mean Value Theorem, whose promised slope is the chord's, not zero.
- Ignoring differentiability failures inside. |x| and x^(2/3) have corner or cusp turnarounds where no derivative exists; Rolle's conclusion genuinely fails for them.
- Expecting exactly one c. The cubic example has two; a wiggly function can have many. "At least one" is the full strength of the claim.
- Applying it on intervals containing poles. f(x) = 1/x² has equal values at ±1, but the discontinuity at 0 voids everything; indeed f′ never vanishes.
- Confusing c with a root of f. c is where the derivative vanishes, a flat point of the graph, generally not where f itself is zero.
Real-World Applications
Rolle's theorem operationalizes a physical truism: whatever returns to its starting state passed through a moment of zero rate. A projectile back at launch height had zero vertical velocity at its apex; a pendulum returning to release angle was momentarily still at the far swing; a company whose revenue matches last year's had at least one instant of zero growth in between. In numerical analysis the theorem is the engine of error formulas: the Lagrange remainder for Taylor polynomials and the error terms of interpolation and quadrature rules are all proved by applying Rolle's theorem repeatedly to cleverly built auxiliary functions. Root-counting via the contrapositive keeps polynomial solvers honest, and in statistics the same logic bounds the modes of distributions between their antimodes. It is a small theorem that does outsized bookkeeping across analysis.
Frequently Asked Questions
What are the three hypotheses of Rolle's theorem?
Continuity on the closed interval [a, b], differentiability on the open interval (a, b), and equal endpoint values f(a) = f(b). All three are essential; each has a standard counterexample showing the conclusion fail without it.
How is Rolle's theorem different from the Mean Value Theorem?
Rolle's is the level special case: equal endpoints force a zero-slope tangent. The MVT tilts the picture, guaranteeing a tangent parallel to the chord of any endpoints. Proving the MVT amounts to subtracting the chord from f and applying Rolle's theorem, so the two are equivalent in strength.
Why must differentiability hold on the open interval only?
The guaranteed point c is strictly interior, so derivatives are needed only there. Endpoint corners are harmless: the function can arrive at a and b as sharply as it likes, provided it is continuous up to them.
Can Rolle's theorem produce the location of c?
The theorem itself only asserts existence; locating c means solving f′(c) = 0, which this calculator does numerically after computing the symbolic derivative. For polynomials of low degree the equation can be solved exactly by algebra.
What does the |x| counterexample teach?
That smoothness is not decoration. |x| on [−1, 1] meets every hypothesis except differentiability at one point, and the conclusion fails completely: the graph turns around at a corner, where there is no tangent to be horizontal.
How does the theorem bound the number of roots?
Between consecutive roots of f sits a root of f′. Counting: f can have at most one more root than f′. Since a degree-n polynomial's derivative has degree n − 1, induction gives the familiar "at most n real roots," with Rolle's theorem doing the inductive step.
Is the c from Rolle's theorem always an extremum?
The proof finds c at an interior extremum, but f′ can also vanish at non-extremal flat points (like x³ at 0) that happen to exist. The theorem guarantees at least one zero of f′; classifying each zero is the job of the first or second derivative test.
Does a constant function satisfy the theorem?
Trivially and maximally: every interior point has f′(c) = 0. The proof's dichotomy actually treats this case separately, and it explains the theorem's sharpness: equality of endpoints cannot force more than flatness somewhere, and a constant is flat everywhere.
What role does Rolle's theorem play in Taylor error formulas?
The Lagrange remainder is proved by building an auxiliary function that vanishes at n + 2 points and applying Rolle's theorem n + 1 times, once per differentiation, until a single point ξ carries the entire error in the (n+1)st derivative. Quadrature error bounds follow the same script.
Who was Rolle, and did he like calculus?
Michel Rolle (1652-1719), a French algebraist, published the result for polynomials in 1691, ironically while being one of calculus's sharpest early critics; he called infinitesimal reasoning a "collection of ingenious fallacies." The theorem now bearing his name became a cornerstone of the rigorous calculus he doubted.