A recognition principle for the existence of descent data
Abstract
Suppose is a faithfully flat ring map. The theory of twisted forms lets one compute, given an -module , how many isomorphism classes of -modules satisfy . This is really a uniqueness problem. But this theory does not help one to solve the corresponding existence problem: given an -module , does there exists {\em some} -module such that ? In this paper we work out (as a special case of a general theorem about existence of coalgebra structures over a comonad) a criterion for the existence of such an -module , under some reasonable hypotheses on the map .
Keywords
Cite
@article{arxiv.1303.3670,
title = {A recognition principle for the existence of descent data},
author = {A. Salch},
journal= {arXiv preprint arXiv:1303.3670},
year = {2015}
}
Comments
This version corrects an error in the applications section in the previously posted arxiv version, and also proves the main result in greater generality, as suggested by an anonymous referee. It also contains some example computations which did not appear in the published version