Definition of Galois Cohomology

164 Views Asked by At

For Galois cohomology, one uses the cohomology constructed for profinite groups instead of the usual group cohomology. In other words, one also uses/takes into account the Krull topology we have on the Galois groups.

Is there a significant reason why we are using this definition of Galois cohomology?

For infinite Galois theory, I know that the topology is needed to construct a generalization of the fundamental theorem for finite Galois theory. But is there also a similar reason why we want to consider the topology when talking about cohomology for Galois groups?

Let $G$ be a group, and $M$ be a $G$-module. Then in terms of group cohomology, we have $H^{0}(G,M):=M^{G}$, and $H^{1}(G,M)=\{\text{1-cocycles}\}/\{\text{1-coboundaries}\}$.

When considering profinite groups $G$, we consider discrete $G$-modules $M$. These are $G$-modules such that the group action is continuous with respect to the discrete topology. Then the cohomology groups are given by $H_{c}^{0}(G,M):=M^{G}$, and $H_{c}^{1}(G,M)=\{\text{cont. 1-cocycles}\}/\{\text{cont. 1-coboundaries}\}$. In general, $H_{c}^{1}(G,M)$ does not have to be equal to $H^{1}(G,M)$.

Now a profinite group (for instance a Galois group), is obiously also just a group. Meaning that we could also just consider the usual group cohomology, i.e. $H^{1}(G,M)$ instead of $H^{1}_{c}(G,M)$. Or in other words, just skip the continuity condition.

So I wonder, why do we want also want to take into account the continuity condition. Wouldn't it be a lot easier if we just use the basic group cohomology? Or are there any significant results for which we need the extra continuity condition.