The anomaly line bundle of the self-dual field theory (Q393698)

The main reason for the study of gravitational anomalies comes from the fact that we expect low energy field theory limits of quantum gravity to be free of gravitational anomalies, so the study of the latter provides interesting information about what type of quantum field theories can arise in this limit. The main result of the paper is a determination of the anomaly bundle of the self-dual field theory from first principles. There was mentioned by the author that even if the anomaly bundle of a quantum field theory is topologically trivial, there can still exist a ``geometric'' anomaly, associated to the fact that it does not admit any natural trivialization. This anomaly is described by the curvature and holonomies of a natural connection living on the anomaly bundle. The missing piece in our knowledge about the anomalies of the self-dual field theory is the set of holonomies of this connection: the global gravitational anomaly. The author sketched how the determination of the topological anomaly achieved in this paper can be used to compute the global gravitational anomaly. He obtained a formula for the global gravitational anomaly of a pair of self-dual fields, but not in a form very suitable to check anomaly cancellation. The paper contains 5 Sections. The formalism of Belov-Moore is reviewed in Section 2.1. The author's action for a pair of self-dual fields is defined in Section 2.2. In Section 3, the author performs the path integration. The latter decomposes into a sum over topological sectors (Section 3.1) and an integration over the non-zero modes (Section 3.2). The author then showes how to recover the local anomaly from his formula for the partition function (Section 3.3) and discusses informally the problems arising with actions for a single self-dual field on arbitrary Riemannian manifolds (Section 3.4). Section 4 is a pedagogical introduction to modular geometry, factors of automorphy and their relations to line bundles. The author also discusses recent and less well-known results providing a classification of topological line bundles over certain modular varieties. In Section 5, the author characterizes the anomaly bundles of a pair of self-dual fields (Section 5.1) and of a single self-dual field (Section 5.2). The author also discusses to which extent the information gained on bundles over the finite dimensional modular varieties apply to their pullback (Section 5.3). In Section 5.4 the author sketches how mentioned previously results allow to extract a formula for the global gravitational anomaly of a pair of self-dual fields.