Fermat’s Little Theorem

This article presents Fermat’s little theorem from a group theory perspective. This article assumes rudimentary knowledge of groups. In particular, the reader is expected to know the meaning of groups and inverses of elements.

Modular Arithmetic from Group Perspective

The operator is both associative and commutative. We only need to check if every element is invertible. If an element is not invertible then the pigeonhole principle yields that there exist such that and since , from elementary number theory so but this is impossible as so every element is invertible.

Fermat’s Little Theorem

We will attempt to provide a proof of Fermat’s Theorem using group theory but this requires some sophisticated tools that we will develop before proving this theorem. First, we introduce the notion of cosets and use it to deduce Lagrange’s theorem. Then we will employ cyclic subgroups to complete the proof. A subgroup is simply a subset of a group that follows the group axioms.

We can prove the first hypothesis by establishing a one-to-one correspondence between and . Observe that if then by the cancellation law, and clearly this is a surjection.

To prove the second hypothesis, suppose two cosets are not equal, then for some and our selected , then if , then and thus if then for some so . Repeating the argument in the reverse direction, we get . Therefore, .

From this lemma and the fact that each element must be in as , we can state Lagrange’s theorem.

It is easy to verify that this is indeed a subgroup. We are almost ready to prove Fermat’s theorem.

Consider the subgroup . Suppose it has order , then . By Lagrange’s theorem, and so we have

Now using the above corollary to Lagrange’s theorem, one can make out an algebraic proof of Fermat’s Little Theorem.

Suppose , then this is trivial. Now suppose , then we can treat as a part of the multiplicative group of integers not congruent to zero modulo . This group has an order of , whence by the corollary to Lagrange’s theorem. Finally since , we can multiply on both sides without changing the result to get

A point to remark is that we have developed tools of group theory motivated to prove Fermat’s theorem, some of these results can be generalised but this would take considerable time and hence it is not pursued here.

No comment found.

Add a comment

You must log in to post a comment.