A proof of Fermat's little theorem

Question

I'm looking for a hint regarding the following proof, I've played with a few approaches, mainly rewriting the conditions as congruences and equations and toying with the algebra, but am having trouble connecting the dots. Thanks! -

Consider $n = p_1p_2 \dotsc p_s$ with $p$ primes, such that $\forall_i \ \ (p_i-1) \ | \ (n-1)$.

Show that $a^n \equiv a \ (\text{mod}\ n)$ for all $a$ ($1 \leq a \leq n -1$)

I think that hint should take you all the way there, but I'll post a complete solution w/ details tomo if you haven't managed. - Try and beat me to it :P — mdave16, Mar 08 '17 at 23:53
$,p\mid a^n-a,$ is clear if $,p\mid a,,$ else apply little Fermat to show $,p\mid a^{n-1}-1.,$Then since each $p_i$ divides $,a^n-a,$ so too does their lcm = product. The converse holds too, see here. — Bill Dubuque, Mar 08 '17 at 23:54
I think these numbers are called Carmichael numbers. For example, $561 = 3 \times 11 \times 17$, and $560$ is a multiple of $2,10,16$. These numbers actually have nicer properties than the above. For example, the third Carmichael number is $1729$, which rings a bell. — Sarvesh Ravichandran Iyer, Mar 09 '17 at 00:03
Thank you for the comments. I think I've worked it out, will post my answer in a couple of hours. — normally distributed, Mar 09 '17 at 07:20

normally distributed · Accepted Answer · 2017-03-09T16:53:56.417

So this is what I arrived at:

By FLT $\rightarrow$ $\forall_i \forall{a} \ (p_i \mid a^{p_i -1} -1)$. Because we know that $(p_i - 1) \mid (n-1)$, we have that $p_i \mid a^{n-1} - 1$. Since the relation holds for all $p_i$, then it also holds for their product $n$. Therefore, $n \mid a^{n-1} - 1$.

This gives $a^{n-1} \equiv 1 \ (\text{mod} \ n)$. Multiplying by $a$ on both sides, we get $a^{n} \equiv a \ (\text{mod} \ n)$, as we wanted.

A proof of Fermat's little theorem

1 Answers1