Monsky–Washnitzer cohomology

Monsky–Washnitzer cohomology is a p-adic cohomology theory for non-singular affine algebraic varieties over finite fields, developed to give a cohomological framework compatible with p-adic methods. It provides a p-adic analogue of de Rham cohomology that admits a Frobenius action and can be used to compute zeta functions of varieties over finite fields in the spirit of the Weil conjectures and works by Bernard Dwork, Alexander Grothendieck, and later contributors. The theory played a central role in bridging explicit arithmetic computations exemplified by Dwork cohomology and structural frameworks later formalized in Rigid cohomology and p-adic Hodge theory.

History and motivation

Monsky–Washnitzer cohomology originated in the late 1960s as an answer to computational and conceptual problems raised by Bernard Dwork’s analytic proof of the rationality of the zeta function for varieties over finite fields. Influenced by the work of Alexander Grothendieck on algebraic de Rham cohomology and the search for p-adic counterparts to classical theories used in the proof of the Weil conjectures by Pierre Deligne, Paul Monsky and Gerald Washnitzer formulated a rigidified de Rham complex for liftings of affine varieties to characteristic zero. The construction was motivated by applications to explicit point-counting algorithms related to the Hasse–Weil zeta function and computational aspects pursued by researchers in Number theory and Algebraic geometry at institutions such as Harvard University and University of Chicago.

Definition and construction

Given a smooth affine variety over a finite field of characteristic p, Monsky–Washnitzer cohomology begins by choosing a lifting of the coordinate algebra to a p-adic, complete, weakly complete, or dagger algebra over a complete discrete valuation ring such as the ring of integers of a finite extension of Qp. The construction uses completions akin to those in Tate algebra theory and formations analogous to the de Rham complex on smooth schemes over Zp. One forms a differential graded algebra of overconvergent differential forms on the lifted algebra and takes its cohomology; the result is independent of many choices up to canonical isomorphism. A Frobenius endomorphism is induced by lifting an absolute Frobenius morphism, connecting the theory to operators studied by Bernard Dwork and enabling trace formulas related to Lefschetz trace formula computations.

Properties and comparison theorems

Monsky–Washnitzer cohomology satisfies finite-dimensionality over suitable p-adic coefficient fields, Poincaré duality for properly paired situations, and long exact sequences in Mayer–Vietoris style similar to those of classical de Rham cohomology. Comparison theorems relate it to crystalline cohomology for proper smooth liftable schemes as developed by Jean-Marc Fontaine and Pierre Berthelot, and to algebraic de Rham cohomology for varieties lifted to characteristic zero as in the work of Alexander Grothendieck and Gerritzen. The Frobenius action on Monsky–Washnitzer cohomology yields information about the Hasse–Weil zeta function via traces, paralleling methods by Bernard Dwork; this connects to the theory of slopes and Newton polygons studied by Katz and others. Independence of the choice of lifting is a significant structural property proven using methods that echo ideas from Grothendieck’s theory of schemes and comparisons with crystalline cohomology.

Examples and computations

Concrete computations via Monsky–Washnitzer cohomology have been carried out for affine curves, hypersurfaces, and toric varieties, yielding explicit zeta function data used in algorithmic point counting and cryptographic applications related to Elliptic curve cryptography and Hyperelliptic curve cryptography. Classical examples include computations for affine lines and conics, where the cohomology groups reproduce the expected Betti-like counts after accounting for Frobenius eigenvalues as in Dwork’s calculations. Implementations in computational packages have drawn on methods developed in the tradition of researchers at Institute for Advanced Study and universities such as University of California, Berkeley and Princeton University, and informed later algorithmic approaches analogous to those used in computing zeta functions via p-adic cohomology techniques.

Relation to rigid cohomology and p-adic Hodge theory

Monsky–Washnitzer cohomology served as a precursor and building block for Rigid cohomology formulated by Pierre Berthelot, which extends the construction to non-affine and singular varieties using rigid analytic geometry à la John Tate. Rigid cohomology generalizes Monsky–Washnitzer’s dagger algebras and overconvergent forms, providing functoriality and comparison results with crystalline cohomology and étale cohomology that are central in p-adic Hodge theory. Through comparison isomorphisms and the action of Frobenius and monodromy operators, Monsky–Washnitzer methods contribute to the study of Fontaine’s period rings and the classification of p-adic representations as in the work of Jean-Marc Fontaine and collaborators. These connections embed Monsky–Washnitzer cohomology in the broader tapestry linking Arithmetic geometry, Galois representations, and the study of zeta and L-functions pursued by mathematicians at institutions such as IHÉS and Collège de France.

Category:Cohomology theories