Take the 2-minute tour ×
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.

I got stuck on the following:

Let $G$ be a finite cyclic group. Then it is a well-known fact, that one can compute its Tate-cohomology groups from the complex $$\cdots\xrightarrow{\;\tau \;-\;\text{Id} \;}M \xrightarrow{\;\;Tr_{G}\;\;} M \xrightarrow{\;\tau \;-\;\text{Id} \;} M \xrightarrow{\;\;Tr_G\;\;} M \xrightarrow{\;\tau \;-\;\text{Id} \;}\cdots $$ where $\tau$ is multiplication by some generator of $G$ and $Tr_G(x)= \sum_{g \in G} gx$.

Can you tell me why this defines an acyclic resolution of our module? I do not see why this complex necessarily has to compute the cohomology groups.

share|improve this question
add comment

1 Answer 1

up vote 4 down vote accepted

It doesn't define an acyclic resolution (at least, unless $M$ is acyclic). But one can show that the cohomology groups that arise from that complex are isomorphic to the usual ones as follows:

The following is from Chapter VIII, $\S$4 of Serre's Local Fields (I changed the notation a bit to fit yours though):

Define a cochain complex $K$ as follows: $K^i=\mathbb{Z}[G]$ for all $i$, $d:K^i\to K^{i+1}$ is $(\tau - \text{Id})$ if $i$ is even and $Tr_G$ if $i$ is odd. For each $G$-module $A$, put $K(A)=K\otimes_{\mathbb{Z}[G]}A$. Then $K^i(A)=A$ for all $i$, with the induced maps being the same. An exact sequence $0\to A\to B\to C\to 0$ gives rise to an exact sequence of complexes $0\to K(A)\to K(B)\to K(C)\to 0$ whence to an exact cohomology sequence, and, in particular, to a coboundary operator $\delta$.

Proposition 6. The cohomological functor $\{H^q(K(-)),\delta\}$ is isomorphic to the functor $\{H^q(G,-),\delta\}$.

First of all it is clear that $\widehat{H}{}^0(G,A)=H^0(K(A))$, $\widehat{H}{}^{-1}(G,A)=H^{-1}(K(A))$, and that the coboundary operator $\delta$ relating $H^0$ to $H^{-1}$ is the same. Hence $$H^q(K(A))=0$$ for $q=0,-1$ when $A$ is relatively projective, thence for all $q$ (as the $H^q(K(A))$ depend only on the parity of $q$). That suffices to give the isomorphism.

The last bit is using that a $G$-module is relatively projective if and only if it is relatively injective when $G$ is finite, and an abstract characterization of derived functors (of which group cohomology is an example) - see the last sentence of the section here.

share|improve this answer
add comment

Your Answer

 
discard

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.