-1

How to prove using Lagrange's theorem $a^{p-1}\equiv 1\bmod p$.

Here, exponentiation (repeated multiplication) is of concern.

So, first want to show this property in action. Below shown for two values of $p=5,7$ with operation of exponentiation, the desired property, i.e. for $4$-th power $\equiv 1\bmod 5$, and $6$-th power $\equiv 1\bmod 7$, for all elements of the group. I.e., the last column follows the rule.

Say, $$\begin{array}{|c|c|c|c|c|} \hline &a^1&a^2&a^3&a^4\\ \hline 1 & 1& 1& 1&1\\ \hline 2 & 2 & 4& 3& 1\\ \hline 3 & 3 & 4& 2& 1\\ \hline 4 & 4 & 1& 4& 1\\ \hline \end{array}$$

The same pattern occurs for $p=7$, as shown in the below table : $$\begin{array}{|c|c|c|c|c|c|c|} \hline &a^1&a^2&a^3&a^4&a^5&a^6\\ \hline 1 & 1& 1& 1&1&1&1\\ \hline 2 & 2 & 4& 1& 2&4&1\\ \hline 3 & 3 & 2& 6& 4&5&1\\ \hline 4 & 4 & 2& 1& 4&2&1\\ \hline 5 & 5 & 4& 6 & 2 &3&1\\ \hline 6 & 6 & 1& 6 & 1 &6&1\\ \hline \end{array}$$

Doubt 1: But, none of the above is a group, so can the Lagrange theorem apply?

It says: order of element divides order of group.$- \tag{1}$

There is no group (wrt, exponential operation), and no order hence comes into view.

So, cannot say that can use property (1) to derive the given property from that.

Edit : expect a simple answer, so given tag accordingly.

Edit 1: as per suggestion by @Cpc - to see instead the group properties in multiplication table, which seems is due to the reason that basically exponentiation is repeated multiplication.

So, rather than checking if $<\mathbb {Z^*}, \wedge >$ is a group, need check that $<\mathbb{Z^*}, \times > $ is a group table or not.

$$\begin{array}{|c|c|c|c|c|} \hline \times \bmod 5 &1&2&3&4\\ \hline 1 & 1& 2& 3&4\\ \hline 2 & 2 & 4& 1& 3\\ \hline 3 & 3 & 1& 4& 2\\ \hline 4 & 4 & 3& 2& 1\\ \hline \end{array}$$

$$\begin{array}{|c|c|c|c|c|c|c|} \hline \times \bmod 7 &1&2&3&4&5&6\\ \hline 1 & 1& 2& 3&4&5&6\\ \hline 2 & 2 & 4& 6& 1&3&5\\ \hline 3 & 3 & 6& 2& 5&1&4\\ \hline 4 & 4 & 1& 5& 2&6&3\\ \hline 5 & 5 & 3& 1 & 6 &4&2\\ \hline 6 & 6 & 4& 4 & 3 &2&1\\ \hline \end{array}$$

Edit 2: Also, is there any other such problem in multiplication (similar to exponentiation operation here) where instead need check addition table (similar to addition operation here) for group properties?

jiten
  • 4,960
  • 1
    Hint: Consider the the cyclic group generated by $a$ (multiplicatively), which is a subgroup of $(\Bbb Z/p\Bbb Z)^*$. – Milten Jun 28 '22 at 11:48
  • Because of Lagramge´s theorem the orders of the elements of the multiplicative groups $(\mathbb{Z}/p\mathbb{Z})^*$ must divide the order of these groups which are $p-1$ for $p$ prime – Peter Melech Jun 28 '22 at 11:52
  • @PeterMelech In first table, why last row doesn't have entire group? It has only two elements. Similarly, in second row; only two elements. Single element in last column; the mainstay for the problem; also is issue. In second table, even more issues. – jiten Jun 28 '22 at 12:24
  • Please first search for answers before posting questions to avoid duplication. – Bill Dubuque Jun 28 '22 at 13:35
  • 2
    @jitenLet $G$ be a finite group of order $|G|$ . Then for any $a\in G$ , $a^{|G|}=e $. (Direct consequence of Lagrange's theorem)

    Let $U(\Bbb{Z}_n) $ denotes the group of units of the ring $(\Bbb{Z_n}, +, •) $

    $$U(\Bbb{Z}_n) ={a\in \Bbb{Z_n} : a\le n \text{ and } (a, n) =1}$$( see here )

    $|U(\Bbb{Z}_p)|=\varphi(p)=p-1$

    Now use the first result.

    – SoG Jun 28 '22 at 13:36

2 Answers2

2

You're working in the group of invertible elements modulo $n$, denoted e.g. $(\mathbb{Z}_n^*,\times)$.

  • The set is $\{i \in \mathbb{Z}_n : \gcd(i,n)=1\}$, and
  • The binary operation is multiplication modulo $n$.

So yes, Lagrange's Theorem applies here. When $n$ is a prime, the order of the group is $n-1$ (noting that $\gcd(0,n) \neq 1$).

When $n$ is prime, you're rediscovering Fermat's Little Theorem, otherwise we have Euler's Theorem.

  • Your first link states about multiplicative inverse, by use of Bezout lemma, for $a$ in $ax\equiv 1$. Want to add that though exponentiation is repeated multiplication, but still $a^x \neq ax$. – jiten Jun 28 '22 at 12:09
  • Might be not able to understand why it is a group, and that is the core issue then. – jiten Jun 28 '22 at 12:13
  • 1
    I've always just thought of it as the extended Euclidean algorithm. The main thing is to avoid conflating $\mathbb{Z}_n$ under addition with $\mathbb{Z}_n^*$ under multiplication. They do not have the same number of elements, nor the same binary operation. – Rebecca J. Stones Jun 28 '22 at 12:32
  • Also, is there any other such problem in multiplication (similar to exponentiation operation here) where instead need check addition table (similar to addition operation here) for group properties? – jiten Jun 28 '22 at 13:33
  • 1
    Please strive not to post more (dupe) answers to dupes of FAQs, cf. recent site policy announcement here. – Bill Dubuque Jun 28 '22 at 13:33
1

See you actually start with a ring, $\Bbb Z_p$. Or, actually, it's even a field. And for any field, the nonzero elements form a (multiplicative) group, also denoted $\Bbb F^*$.

The group is then$\Bbb Z_p^×\cong\Bbb Z_{p-1}$, the group of units (of the ring).

So in your examples we get $\Bbb Z_5^×\cong\Bbb Z_4$ and $\Bbb Z_7^×\cong\Bbb Z_6$.

When $p$ is prime the group of units has order $p-1$.

And by Lagrange anything raised to the power $n=|G|$ is equal to $e$.

  • Seems stuck as why it is a group. If you can show the group property in any of the table(s) given, then would be highly thankful. – jiten Jun 28 '22 at 12:18
  • It's a group because every element has an inverse. The other requirements for a group are already in place. – suckling pig Jun 28 '22 at 12:20
  • In first table, why last row doesn't have entire group? It has only two elements. Similarly, in second row; only two elements. Single element in last column; the mainstay for the problem; also is issue. In second table, even more issues. – jiten Jun 28 '22 at 12:27
  • If you change to a table of multiplication for the nonzero elements of the field, you will see that you have a group. – suckling pig Jun 28 '22 at 12:28
  • Thanks, kindly see the edited OP. – jiten Jun 28 '22 at 12:36
  • Please strive not to post more (dupe) answers to dupes of FAQs, cf. recent site policy announcement here. – Bill Dubuque Jun 28 '22 at 13:33