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

Suppose $G$ and $H$ are discrete groups. If $\rho_G$ and $\rho_H$ are reps of $G$ and $H$ on $V_G$ and $V_H$, respectively, then we get a rep of $G\times H$ on $V_G\otimes V_H$ by sending $(g,h)$ to $\rho_G(g)\otimes\rho_H(h)$. This induces a map $$ R(G) \otimes R(H) \to R(G\times H) $$ where $R(\cdot)$ denotes the representation ring. I know, by character theory, that this is an isomorphism for $G$ and $H$ finite.

I am guessing this result is not true in general-- what is a simple counterexample?

Are there nice conditions on $G$ and $H$ under which the map above is an isomorphism?


Because of some of the issues with the correct definition of representations rings, I will rephrase my question to precisely what I want to know:

Is it always the case that any finite dimensional representation (over $\mathbb C$) of $G \times H$ is isomorphic to $\bigoplus_j V_j \otimes W_j$ where $V_j$ is a rep of $G$ and $W_j$ is a rep of $H$?

share|cite|improve this question
This might depend on how you define $R(G)$ without semisimplicity. The main result (that the tensor product of an irrep of $G$ with an irrep of $H$ is always an irrep of $G\times H$, and that every irrep of $G\times H$ can be written in this form) does hold for any groups $G$ and $H$, even infinite ones. See . – darij grinberg Jun 7 '13 at 15:03
@darijgrinberg I want $R(G)$ to be the ring of all representations of $G$ (not e.g. just the ring generated by irreps). Thanks for the link-- I did not know that that result holds in general. – Eric O. Korman Jun 7 '13 at 15:21
@Eric: darij's point is that there are two ways to define $R(G)$ even as an additive group in the absence of semisimplicity; you can either impose $[A] = [B] + [C]$ whenever $A \cong B \oplus C$ or you can impose $[A] = [B] + [C]$ whenever there is a short exact sequence $0 \to B \to A \to C \to 0$ (the split Grothendieck group and the Grothendieck group respectively). The first option gives you a bigger group in general. – Qiaochu Yuan Jun 7 '13 at 18:46
Silly question, is it true for the analog of my example? Is every finite dimensional $\mathbb{C}[x,y]$ module of the form $\bigoplus_j V_j \otimes W_j$ for finite dimensional $\mathbb{C}[x]$ (or y) modules $V_j,W_j$? I kind of thought they were weirder than that – Jack Schmidt Jun 7 '13 at 20:40
(Take $G=H=\mathbb{Z}$ and handle inverses, which I wouldn't expect to change too much.) – Jack Schmidt Jun 7 '13 at 20:41
up vote 1 down vote accepted

I asked this on mathoverflow and got a counterexample: .

share|cite|improve this answer
The same counterexample as mentioned here :P – Jack Schmidt Jun 12 '13 at 2:37

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.