Mathematical remarks on the cohomology of gauge groups and anomalies

Abstract

Anomalies can be viewed as arising from the cohomology of the Lie algebra of the group of gauge transformations and also from the topological cohomology of the group of connections modulo gauge transformations. We show how these two approaches are unified by the transgression map. We discuss the geometry behind the current commutator anomaly and the Faddeev- Mickelsson anomaly using the recent notion of a gerbe. Some anomalies (notably 3-cocycles) do not have such a geometric origin. We discuss one example and a conjecture on how these may be related to geometric anomalies.