Question: Prove that the group of order 3 is cyclic.


Let H be a group of order 3. By definition of group, there can be only one identity element in the group H.

So, $H=\left \{ e,x,y \right \}$.

By definition of cyclic group,

we have that the elements x and y

$x=g^{n} \exists n \in \mathbb{Z}$

$y=g^{n} \exists n \in \mathbb{Z}$

In particular, n is positive for if it were not, a contradiction would arise from having more than one identity element.

Any hints or assistance is appreciated.

Thank in advance.

  • $\begingroup$ In third and fourth line of your attempt of solution you seem to assume the thesis ("$H$ is cyclic") as a hypothesis, making your argument circular. $\endgroup$ – user228113 Jun 2 '16 at 10:19
  • $\begingroup$ What is $g$ here? $\endgroup$ – Servaes Jun 2 '16 at 10:19

You're trying to PROVE that $G$ is cyclic, so you cannot (yet) assert that $x = g^n$. Instead, consider $x \cdot x$. It must be either $x, y,$ or $e$. If it's $x$, then you have $$ x^2 = x\\ x^2 (x^{-1}) = x x^{-1}\\ x = e $$ which is a contradiction, because $x$ and $e$ are distinct elements of the group.

If $x^2 = e$, then $x$ has order 2, but 2 does not divide 3, so this contradicts Lagrange's theorem.

Finally, we conclude that $x^2 = y$, and thus the group is cyclic, generated by the element $g = x$.

{Alternative if you don't like Lagrange yet:}

In the case where we suppose that $x^2 = e$:

The elements $xe, xx,$ and $xy$ must all be distinct for if two were the same, then multiplying by $x^{-1}$ on the left would show that two of $e, x, y$ were the same, which is impossible.

Since $xe = x$ and we're assuming $x^2 = e$, we must have $$ xy = y. $$ multiplying on the right by $y^{-1}$ gives $x = e$, a contradiction. So $x^2 = e$ is also impossible.


Actually, any group of prime order is cyclic.

Indeed, let $p=\lvert G\rvert$ and $a\in G$, $a\ne e$. By Lagrange's theorem, the subgroup $\langle\mkern1.5mu a\mkern1.5mu \rangle$ generated by $a$ has order a divisor of $\lvert G\rvert=p$, and is $>1$ since $a\ne e$, whence $\lvert\langle\mkern1.5mu a\mkern1.5mu \rangle\rvert=p$, i.e. $\;\langle\mkern1.5mu a\mkern1.5mu \rangle=G$.

  • $\begingroup$ elegant proof!. $\endgroup$ – Cloud JR Aug 12 '18 at 11:53

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.