Lagrange Theorem: Statement, Proof & Examples

What Is The Lagrange Theorem?

The Lagrange Theorem, in its most searched sense, is a result in group theory: the order of every subgroup of a finite group divides the order of the group. Here the order of a group is simply the count of its elements, and a subgroup is a smaller set inside the group that is itself a group under the same operation. If $H$ is a subgroup of a finite group $G$, then $\lvert H \rvert$ divides $\lvert G \rvert$ exactly, with no remainder.

The name is shared. In calculus, "Lagrange's theorem" almost always means Lagrange's Mean Value Theorem, a cornerstone of senior-secondary and first-year calculus. Both results are named after the same mathematician, Joseph-Louis Lagrange, which is why one phrase points to two different places in mathematics.

Table: The two theorems that share the name "Lagrange Theorem".

The group theory version comes first below, because that is what most people reach when they type the bare phrase. If you came for the calculus result, jump to the mean value section.

What Does The Lagrange Theorem State In Group Theory?

For a finite group $G$ with a subgroup $H$, the order of $H$ divides the order of $G$. The exact relationship is:

$$\lvert G \rvert = [G : H] \times \lvert H \rvert$$

The number $[G : H]$ is called the index of $H$ in $G$. It counts how many distinct cosets of $H$ fit inside $G$, and because it is a whole number, $\lvert H \rvert$ must divide $\lvert G \rvert$.

Read the statement slowly. It does not say every divisor of $\lvert G \rvert$ gives a subgroup. It says the reverse: every subgroup you can find has an order that is one of those divisors. That direction matters, and the difference between the two is where most errors live.

What Is A Coset, And Why Does It Matter?

A coset is what you get when you shift a subgroup by a single element of the group. For a subgroup $H$ and an element $g$ in $G$, the left coset is the set $gH = {, gh : h \in H ,}$. Cosets are the engine of the whole proof, so they are worth seeing once with numbers.

Take the group $\mathbb{Z}_6 = {0,1,2,3,4,5}$ under addition modulo 6, and the subgroup $H = {0, 3}$. Shifting $H$ by each element gives:

$$0 + H = {0, 3}, \qquad 1 + H = {1, 4}, \qquad 2 + H = {2, 5}$$

After that the cosets repeat: $3 + H = {3, 0}$ is just $H$ again. So there are three distinct cosets, each holding exactly two elements, and together they cover all six elements of $\mathbb{Z}_6$ without overlap.

Two facts about cosets do all the work. Every coset has the same number of elements as $H$, and any two cosets are either identical or completely separate. Together they mean the cosets slice the group into equal, non-overlapping blocks.

How Do You Prove The Lagrange Theorem?

The proof is short once cosets are in hand. The cosets of $H$ partition $G$ into disjoint blocks, and each block has exactly $\lvert H \rvert$ elements.

Suppose there are $[G:H]$ of these blocks. Since they cover $G$ with no overlap, adding up their sizes gives the order of $G$:

$$\lvert G \rvert = \underbrace{\lvert H \rvert + \lvert H \rvert + \cdots + \lvert H \rvert}_{[G:H] \text{ blocks}} = [G:H] \times \lvert H \rvert$$

Because $[G:H]$ is a positive whole number, $\lvert H \rvert$ divides $\lvert G \rvert$. That is the entire theorem. Every coset being the same size as $H$ is the one fact that makes the counting exact, and it holds because shifting by $g$ is a reversible relabelling of $H$.

What Are The Corollaries Of The Lagrange Theorem?

Several results that look harder than the theorem itself fall straight out of it. Each one comes from applying the divides-the-order idea to a well-chosen subgroup.

• The order of every element divides the order of the group. The powers of any element $a$ form a subgroup, so the smallest $n$ with $a^n = e$ must divide $\lvert G \rvert$.

• Raising to the group's order returns the identity. For every element $a$ in a finite group, $a^{\lvert G \rvert} = e$.

• Every group of prime order is cyclic. If $\lvert G \rvert$ is a prime $p$, the only divisors are $1$ and $p$, so $G$ has no subgroups except the trivial one and itself, and any non-identity element generates the whole group.

• Fermat's little theorem follows. Applying the corollary above to the group of nonzero remainders modulo a prime $p$ gives $a^{p-1} \equiv 1 \pmod{p}$ whenever $a$ is not a multiple of $p$.

That last point is worth a pause. A theorem about counting group elements quietly proves a result in number theory that predates group theory by more than a century.

How Do You Use The Lagrange Theorem? A Worked Example

Example 1: Find every subgroup of $\mathbb{Z}_6$ and check the theorem.

The group $\mathbb{Z}_6$ has order 6, so any subgroup must have an order that divides 6, which means an order of 1, 2, 3, or 6. Checking each possibility gives four subgroups and no others.

Table: Every subgroup of the group $\mathbb{Z}_6$, and how each order divides 6.

There is no subgroup of order 4 or order 5, and the theorem tells you so before you search: neither 4 nor 5 divides 6.

Final answer: the subgroup orders are 1, 2, 3, and 6, each a divisor of 6, exactly as the theorem requires.

What Is Lagrange's Mean Value Theorem?

Lagrange's Mean Value Theorem is the calculus result that shares the name. It states that if a function $f$ is continuous on a closed interval $[a, b]$ and differentiable on the open interval $(a, b)$, then there is at least one point $c$ inside $(a, b)$ where the instantaneous rate of change equals the average rate of change:

$$f'(c) = \frac{f(b) - f(a)}{b - a}$$

The two conditions are not optional. The function must be continuous across the closed interval $[a, b]$, endpoints included, and differentiable at every point of the open interval $(a, b)$. Drop either one and the guarantee can fail. Here $f$ maps an interval of real numbers to real outputs, so this is a statement about ordinary functions of a single variable.

Geometrically, the fraction on the right is the slope of the secant line through the two endpoints of the curve. The derivative $f'(c)$ is the slope of the tangent at the point $c$. So the theorem promises a point where the tangent runs parallel to the secant.

Example 2: Apply the theorem to $f(x) = x^2$ on $[1, 3]$.

The function $x^2$ is continuous and differentiable everywhere, so both conditions hold. First find the average rate of change across the interval:

$$\frac{f(3) - f(1)}{3 - 1} = \frac{9 - 1}{2} = 4$$

Now set the derivative equal to that value. Since $f'(x) = 2x$:

$$2c = 4 \implies c = 2$$

The point $c = 2$ lies inside $(1, 3)$, so the theorem is satisfied.

Final answer: at $c = 2$, the tangent to $y = x^2$ is parallel to the secant through $(1, 1)$ and $(3, 9)$.

How Does Lagrange's Mean Value Theorem Relate To Rolle's Theorem?

Rolle's theorem is the special case where the two endpoints sit at the same height. If $f$ is continuous on $[a, b]$, differentiable on $(a, b)$, and $f(a) = f(b)$, then there is a point $c$ where $f'(c) = 0$, a spot where the tangent is flat.

The mean value theorem is the tilted version of the same idea. When $f(a) = f(b)$, the secant is horizontal, its slope is zero, and $f'(c) = 0$ recovers Rolle's theorem exactly. Going the other way, you can prove the mean value theorem from Rolle's by subtracting the secant line from $f$, which levels the endpoints so Rolle's theorem applies. One theorem is the flat case; the other is the same picture on a slope.

Why Does The Lagrange Theorem Matter?

Both versions earn their place by turning a vague question into a hard constraint. Neither was invented to lengthen a course.

• Group theory: it fences in what is possible. Before checking a single case, the theorem rules out subgroup sizes that cannot exist. A group of order 15 can only have subgroups of order 1, 3, 5, or 15, which is why so many classification arguments start by listing divisors.

• Group theory: it builds bigger results. The corollaries above, including Fermat's little theorem, all lean on the same counting fact, which is why Lagrange's theorem is usually the first serious theorem in an abstract algebra course.

• Calculus: it links average to instantaneous. The mean value theorem is the bridge that lets you conclude something about a function's behaviour at a single instant from its behaviour over a whole interval, which is the logic behind error bounds and much of the theory that follows.

The common thread is guarantee. Each version promises that a certain value or structure must exist, without asking you to hunt for it by hand. That is the kind of certainty mathematics is built to provide.

Who Shaped The Lagrange Theorem?

The counting fact came before the field it belongs to. When Lagrange first met the idea, the word "group" did not yet mean what it means now, and he was chasing a completely different problem: why no formula solves the general fifth-degree equation.

Two other names sit close to this story:

• Michel Rolle (1652–1719, France) gave the flat-endpoint theorem that Lagrange's Mean Value Theorem generalises. He spent years doubting the new calculus, then supplied one of its most-used results.

• Augustin-Louis Cauchy (1789–1857, France) proved a partial converse for groups: if a prime divides the order of a group, the group must contain an element of that prime order. It answers, in one important case, the question Lagrange's theorem deliberately leaves open.

Where Is The Lagrange Theorem Used In The Real World?

The two versions surface in strikingly different places, yet both show up in everyday technology.

• Cryptography. The security of widely used public-key systems rests on group orders and the arithmetic that Lagrange's theorem and Fermat's little theorem describe.

• Crystallography and chemistry. The symmetries of a crystal or a molecule form a group, and Lagrange's theorem constrains which smaller symmetry patterns can sit inside the whole.

• Puzzles and error-correcting codes. The reachable states of a Rubik's cube form a group, and the same subgroup counting governs the codes that keep data intact on scratched discs and noisy channels.

• Speed enforcement. Average-speed cameras use the mean value theorem directly: if your average speed between two points beat the limit, then at some instant your actual speed did too.

• Physics and engineering. The mean value theorem underpins the error estimates that tell engineers how far a numerical approximation can stray from the true answer.

One name, two theorems, and a reach that runs from secret codes to snowflakes to speed traps. Mathematics keeps proving to be the shared language beneath fields that look unrelated.

What Are The Most Common Lagrange Theorem Mistakes?

Three errors account for most lost marks here, verified against the Judson abstract algebra text on Mathematics LibreTexts, and against mean value theorem reviews on GeeksforGeeks and Testbook.

Assuming the converse of the group theory version.

Where it slips in:

A student reasons that because 6 divides 12, a group of order 12 must have a subgroup of order 6.

Don't do this:

Do not read the theorem backwards. It says every subgroup's order divides the group's order, not that every divisor produces a subgroup.

The correct way:

Treat divisibility as a necessary condition, not a guarantee. The alternating group $A_4$ has order 12 but no subgroup of order 6, which is the standard counterexample. A partial converse exists only for prime divisors, through Cauchy's theorem.

Ignoring the hypotheses of the mean value theorem.

Where it slips in:

A student applies $f'(c) = \frac{f(b)-f(a)}{b-a}$ to a function with a corner or a break inside the interval, then trusts the answer.

Don't do this:

Do not skip the two conditions. If $f$ is not continuous on all of $[a, b]$ or not differentiable across $(a, b)$, the guaranteed point $c$ need not exist.

The correct way:

Check continuity on the closed interval and differentiability on the open interval first. Only then solve $f'(c) = \frac{f(b)-f(a)}{b-a}$ for $c$, and confirm your $c$ actually lands inside $(a, b)$.

Confusing the two theorems that share the name.

Where it slips in:

A calculus question asks about "Lagrange's theorem," and a student answers with the group theory statement, or the reverse.

Don't do this:

Do not assume the phrase is unambiguous. The bare name points to two results in different fields.

The correct way:

Let the context choose. Subgroups, orders, and cosets signal the group theory version; intervals, continuity, and derivatives signal the mean value version.

Practice Problems On The Lagrange Theorem

Answers follow each problem.

1. A finite group has order 20. List every possible order a subgroup of it could have.
   (Answer: the divisors of 20, so 1, 2, 4, 5, 10, and 20.)

2. Can a group of order 7 have a subgroup of order 3?
   (Answer: no, because 3 does not divide 7; a group of prime order has only the trivial subgroup and itself.)

3. In $\mathbb{Z}_8$, the subgroup $H = {0, 4}$ has order 2. How many distinct cosets does it have?
   (Answer: the index is $\lvert G \rvert / \lvert H \rvert = 8 / 2 = 4$ cosets.)

4. Apply the mean value theorem to $f(x) = x^2$ on $[2, 4]$ and find $c$.
   (Answer: the average rate is $(16 - 4)/2 = 6$, so $2c = 6$ and $c = 3$, which lies in $(2, 4)$.)

5. Apply the mean value theorem to $f(x) = x^3 - x$ on $[0, 2]$ and find $c$.
   (Answer: the average rate is $(6 - 0)/2 = 3$, so $3c^2 - 1 = 3$, giving $c = \tfrac{2}{\sqrt{3}} \approx 1.155$, which lies in $(0, 2)$.)

6. Does the mean value theorem apply to $f(x) = \lvert x \rvert$ on $[-1, 1]$?
   (Answer: no, because $f$ is not differentiable at $x = 0$, so a required hypothesis fails.)

Where Should You Go Next After The Lagrange Theorem?

The theorem opens onto two different roads, and both are worth walking.

1. Derivative. The mean value theorem is a statement about a derivative, so a firm grip on what $f'(c)$ measures makes the whole result feel obvious.

2. Relations And Functions. Both versions rest on the idea of a function or a structured set, the foundation the theorem is built on.

3. Calculus. The mean value theorem sits near the start of a long chain of calculus results, and it is a good place to see where that chain begins.

If your child is ready to take these ideas further with a live trainer, the Bhanzu algebra program teaches the reasoning behind theorems like this one, starting from why they are true rather than how to memorise them.