Notes on isocrystals (Q2127208)

Cohomology theories for varieties over a finite field \(k\) of characteristic \(p\) play a crucial role in arithmetic geometry. For example, the Weil conjectures were proved by developing cohomological machinery for such varieties. There are two ``flavors'' of cohomology theories for such varieties: \(p\)-adic, and \(\ell\)-adic for \(\ell \neq p\). Grothendieck introduced \(\ell\)-adic étale cohomology in 1960 in order to solve the Weil conjectures [\textit{A. Grothendieck}, in: Proc. Int. Congr. Math. 1958, 103--118 (1960; Zbl 0119.36902)]. The theory quickly reached maturity, and Deligne completed a proof of the Weil conjectures using \(\ell\)-adic étale cohomology in [\textit{P. Deligne}, Publ. Math., Inst. Hautes Étud. Sci. 43, 273--307 (1973; Zbl 0287.14001)]. The \(p\)-adic theory slightly predates the \(\ell\)-adic one: Dwork gave the first proof of the rationality of the zeta function, and his proof made use of \(p\)-adic methods that can now be understood in cohomological terms [\textit{B. Dwork}, Am. J. Math. 82, 631--648 (1960; Zbl 0173.48501)]. But the theory has been slower to develop; a complete \(p\)-adic proof of the Weil conjectures was not completed until 2006 [\textit{K. S. Kedlaya}, Compos. Math. 142, No. 6, 1426--1450 (2006; Zbl 1119.14014)]. However, \(p\)-adic cohomological techniques are now reaching a stage of maturity in which they can even be used to prove results that likely cannot be proved using \(\ell\)-adic methods. In this context, the paper under review serves an important role in the mathematical literature: it is a very readable introduction to the theory of convergent and overconvergent \(F\)-isocrystals, key objects appearing in \(p\)-adic cohomology that are analogous to \(\ell\)-adic étale local systems (ie, lisse \(\ell\)-adic sheaves). The paper begins with accessible definitions of isocrystals and then proceeds to collect numerous statements of many important and foundational results about isocrystals. The paper also includes expositions of the theories of slopes and weights (the former of which has no direct \(\ell\)-adic analog). The relevant references for these results are often scattered in papers by many authors and spanning several decades, and often involve long and very technical proofs, so this succinct consolidation is very welcome development for those approaching \(p\)-adic cohomology de novo. Readers interested in working with isocrystals are encouraged to read this paper for details.