Mathematics Stack Exchange is a question and answer site for people studying math at any level and professionals in related fields. It's 100% free, no registration required.

Sign up
Here's how it works:
  1. Anybody can ask a question
  2. Anybody can answer
  3. The best answers are voted up and rise to the top

Let $G$ be a group and $f,g: G \rightarrow GL(V)$ be two faithful representations over some field $K$ with $f:x\mapsto f(x)$ and $g:x \mapsto f(x^{-1})$.

I would like to find out if $f$ and $g$ are isomorphic. So I need an ismomorphism $\alpha: V \rightarrow V$ with $$\alpha \circ f(x) \circ \alpha^{-1} = f(x^{-1}). $$ What could such an isomorphism be? Does it depend on $K$?

Thank you for hints.

Later edit. As it is not possible to solve this in general: Is there an answer to the example $G=C_p$ cyclic group, $K$ some field of characteristic zero, $f,g$ irreducible representations of degree $> 1$?

share|cite|improve this question
Are $f,g$ faithful? If not, I doubt it. – Easy Apr 20 '13 at 14:49
yes, they are both faithful – user73521 Apr 20 '13 at 14:55
Faithfulness has very little to do with this. Abelianness (module the kernel of $f$) of $G$, on the other hand, is sort of important. – Mariano Suárez-Alvarez Apr 20 '13 at 16:43
I don't understand the question. Is it how to decide when $f$ and $g$ are isomorphic? To classify the set of representations $f$ for which $f(x^{-1})$ is an isomorphic representation? – Stephen Apr 20 '13 at 17:09
I would like to decide if $f$ and $g$ are isomorphic or not. The only information I have is that $g(x)=f(x^{-1}). Do you mean, that both cases can occur and this depends on the group, the field,... ? – user73521 Apr 20 '13 at 17:16
up vote 3 down vote accepted

It depends. There are two separate issues here. Firstly, if $f$ is a faithful representation such that $g(x)=f(x^{-1})=f(x)^{-1}$ is also a representation, then the group $G$ is necessarily abelian:

$$f(x)^{-1}f(y)^{-1}=g(x)g(y)=g(xy)=f(xy)^{-1}=(f(x) f(y))^{-1}$$ implies that $f(x)$ and $f(y)$ commute and hence that $G$ is abelian.

Secondly, some representations (for instance, the inclusion of $\{\pm 1 \}$ in the multiplicative group of your field) have the property that $f$ and $g$ are isomorphic and others do not---for instance, the defining representation of the group of third roots of unity (working over the complex numbers) is not isomorphic to its dual $g$.

In response to the edited question, calculate the character. When working with subfields of the complex numbers, the representation will be self-dual precisely when the character is real-valued.

share|cite|improve this answer
Thank you for your help – user73521 Apr 20 '13 at 17:29
You're welcome! – Stephen Apr 20 '13 at 17:32

If $f$ is faithful and $g$ is a homomorphism of groups, then $G$ is abelian. It follows that $V$ is a direct sum of representations of degree $1$, and it is easy to see that it is enough to suppose that $V$ is of degree $1$, for the general case follows from this.

Can you do that case?

share|cite|improve this answer
The problem here is, that the field must not be $\mathbb{C}$, so that the representations can have degrees other than 1. – user73521 Apr 20 '13 at 17:03
The point of this answer is that, using its observations, you can easily construct examples where the tworepresentations are not isomorphic. – Mariano Suárez-Alvarez Apr 20 '13 at 17:11
If I try to calculate e.j. $G=C_p$ cyclic, $K$ some field of characteristic zero and $f,g$ have degree bigger than one and be irreducible. Are they always not isomorphic then, or do I need to know how $f$ and $g$ looks exactly? – user73521 Apr 20 '13 at 17:21
Well, using the observations of my answer you can easily construct $f$s which are isomorphic to the corresponding $g$s, so they are not always not isomorphic! – Mariano Suárez-Alvarez Apr 20 '13 at 17:23
Thanks for helping – user73521 Apr 20 '13 at 17:33

Your Answer


By posting your answer, you agree to the privacy policy and terms of service.

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