Suppose $p$ is an odd prime. Show that $x^4 \equiv-1$ (mod $p$) has a solution if and only if $p \equiv1$ (mod $8$).

I have proven one similar result '$x^2 \equiv-1$ (mod $p$) has a solution if and only if $p \equiv1$ (mod $4$)'. I try to mimick the proof but i fail. The following is my attempt:

Suppose $x^4 \equiv-1$ (mod $p$) has a solution, say $a$. Then $a^4 \equiv -1$ (mod $p$). The congruence implies that $\gcd(a^4,p)=1 \Rightarrow \gcd(a,p)=1$. By Fermat's little theorem, $a^{p-1} \equiv 1$ (mod $p$). Note that $1 \equiv a^{p-1} \equiv a^{4{\frac{p-1}{4}}} \equiv (-1)^{\frac{p-1}{4}}$ (mod $p$).

But I don know whether $\frac{p-1}{4}$ is an integer or not. If it is not an integer, then the congruence does not hold.

Can anyone guide me?

EDIT: Since $a^4 \equiv -1$ (mod $p$), we have $a^8 \equiv 1$ (mod $p$) $\Rightarrow $ $ord_p(a)=8 \Rightarrow 8 | \phi(p) \Rightarrow p \equiv 1$ (mod $p$)

Suppose $p \equiv 1$ (mod $8$). Then there exists a primitive root for $p$, say $r$. Then we have $r^{p-1} \equiv (r^{4})^{\frac{p-1}{4}} \equiv1$ (mod $p$). Hence there is a solution to $x^4 \equiv-1$ (mod $p$).

Is my proof in edit correct?

  • $\begingroup$ You have already proved half of it on your own. You've proved that if $x^2=-1$ has a solution in $(\mathbb{Z}/p\mathbb{Z})^{\times}$ then $\frac{p-1}{4}=2k$, so $p \equiv 1 \pmod{8}$. Now can you construct a solution for $x^4 = -1$ if $p \equiv 1 \pmod{8}$ like you did earlier with $x^2 = -1$? $\endgroup$ – user66733 Sep 23 '13 at 8:16


Note that $x^4\equiv -1\mod p$ has a solution if and only if there is an element $x$ of order $8$ in $(\mathbb{Z}/p\mathbb{Z})^\times$.

| cite | improve this answer | |

By Legendre, since $x^4 \equiv -1 \pmod p$, we have $$\Big(\frac{-1}{p}\Big) = \Big(\frac{x^4}{p}\Big)$$.

Since $p \equiv 1 \pmod 8$, $p = 8k+1$. By Legendre,

$$\Big(\frac{-1}{p}\Big)= (-1)^{(p-1)/2}=(-1)^{((8k+1)-1)/2}= 1$$.

Thus, we have $\Big(\dfrac{x^4}{p}\Big) = x^{4\cdot (p-1)/2} \pmod p=x^{2\cdot(p-1)} \pmod p=1$

Now, by Fermat's Little Theorem, this is clearly a solution.

I'll leave it to you to prove this the other way around, but that should do 'er.

| cite | improve this answer | |
  • $\begingroup$ Not sure what this is proving. Aren't you using both sides of the equivalence? $\endgroup$ – darij grinberg Sep 18 '15 at 22:42

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.