New methods for \((\varphi, \Gamma)\)-modules (Q290401)

The last few years (as of 2018) have seen the introduction of new ideas in \(p\)-adic Hodge theory. These ideas are Scholze's theory of perfectoid spaces (see [\textit{P. Scholze}, Publ. Math., Inst. Hautes Étud. Sci. 116, 245--313 (2012; Zbl 1263.14022)]), Kedlaya-Liu's approach to relative \(p\)-adic Hodge theory ([the author and \textit{R. Liu}, Relative \(p\)-adic Hodge theory: foundations. Paris: Société Mathématique de France (SMF) (2015; Zbl 1370.14025)]), and Fargues-Fontaine's theory of ``the curve'' in \(p\)-adic Hodge theory ([\textit{L. Fargues} and \textit{J.-M. Fontaine}, ``Courbes et fibrés vectoriels en théorie de Hodge \(p\)-adique'', Preprint]). These three theories are closely related. The purpose of the paper under review is to give new proofs, based on these new ideas, of two important results in the theory of \((\varphi,\Gamma)\)-modules. These results are (1) Fontaine-Wintenberger's isomorphism between the Galois groups of certain \(p\)-adic fields and the Galois groups of certain fields of characteristic \(p\), and (2) Cherbonnier-Colmez' theorem on the overconvergence of cyclotomic \((\varphi,\Gamma)\)-modules. Fontaine-Wintenberger's isomorphisms are based on their theory of the field of norms, which depends heavily on ramification theory. Cherbonnier-Colmez' theorem likewise uses a lot of ramification theory in order to ``decomplete'' \((\varphi,\Gamma)\)-modules. The author's approach avoids a lot of ramification theory. It uses instead the theory of Witt vectors over valuation rings, and can be generalized more easily and naturally to geometric situations.