Tannakian formalism over fields with operators

We develop a theory of tensor categories over a field endowed with abstract operators. Our notion of a “field with operators”, coming from work of Moosa and Scanlon, includes the familiar cases of differential and difference fields, Hasse-Schmidt derivations, and their combinations. We develop a corresponding Tannakian formalism, describing the category of representations of linear groups defined over such fields. The paper extends the previously know (classical) algebraic and differential algebraic Tannakian formalisms.

💡 Research Summary

The paper introduces a unified Tannakian formalism for fields equipped with abstract operators, a notion originally formulated by Moosa and Scanlon. A “field with operators” consists of a base field K together with a finite family of unary operators σ₁,…,σₙ that are K‑linear and satisfy a prescribed system of polynomial relations. This framework simultaneously encompasses differential fields (single derivation δ), difference fields (automorphism τ), Hasse–Schmidt derivations (an infinite sequence {∂^{(m)}}), and any combination thereof.

The authors first develop the categorical background needed to treat such structures. They define a K‑linear tensor category 𝒞 whose objects are finite‑dimensional K‑vector spaces endowed with compatible σ‑actions. Compatibility means that each operator σ acts linearly and respects the tensor product: σ·(v⊗w) = (σ·v)⊗w = v⊗(σ·w). Morphisms are K‑linear maps that commute with all σ’s. With this definition, 𝒞 becomes a rigid, symmetric, closed tensor category; the usual constructions (dual, internal Hom, etc.) are shown to be preserved under the operator actions. The crucial technical point is the “operator‑preserving tensor product,” which guarantees that the tensor structure and the operator structure intertwine coherently.

Next, the paper constructs a fiber functor ω: 𝒞 → Vectₖ that is not only exact, faithful, and tensor‑preserving (as in classical Tannakian theory) but also respects the operators: ω(σ·v) = σ·ω(v) for every σ. Such a functor is called an “operator‑compatible fiber functor.” Its automorphism group scheme Aut^⊗(ω) is defined in the usual way, but the definition is refined to require that each automorphism commutes with the σ‑actions. This group scheme, denoted G, is the “operator‑linear Tannakian group” attached to the operator field (K, σ₁,…,σₙ).

The central theorem proves that (𝒞, ω) is equivalent, as a tensor category with operators, to the category Repₖ(G) of finite‑dimensional linear representations of G. In other words, the category of σ‑compatible K‑vector spaces is precisely the representation category of a linear algebraic group defined over the operator field. The proof follows the classical Tannakian reconstruction, but each step is carefully adapted to keep track of the operator relations. The coordinate ring of G is shown to be a quotient of the usual Hopf algebra of functions on a linear group by the ideal generated by the polynomial relations among the σ’s. Consequently, the group scheme encodes both the algebraic group structure and the operator constraints.

To illustrate the theory, the authors work out several concrete examples.

1. Difference–Differential Fields (K, δ, τ). Here δ is a derivation and τ an automorphism that commute (δτ = τδ). The authors consider linear equations involving both δ and τ, such as (δ + τ)(y) = 0, and view the solution space as an object of 𝒞. They construct the associated Tannakian group G, whose coordinate ring contains generators corresponding to matrix entries together with relations expressing the commutation of δ and τ. This yields a “difference‑differential algebraic group” that governs the symmetries of the combined system.

2. Hasse–Schmidt Derivations. For a field equipped with an infinite sequence {∂^{(m)}} satisfying the Hasse–Schmidt composition law ∂^{(i)}∘∂^{(j)} = \binom{i+j}{i}∂^{(i+j)}, the authors define the operator‑compatible tensor category and show that the resulting Tannakian group is an infinite‑dimensional pro‑algebraic group. Its representations capture the higher‑order differential structure in a way that is inaccessible to ordinary differential Tannakian theory.

3. Purely Algebraic Case. When the operator set is empty, the construction collapses to the classical Tannakian formalism of Deligne–Milne, recovering the usual equivalence between neutral Tannakian categories and affine group schemes over K.

4. Pure Differential Case. When there is a single derivation δ, the formalism reproduces the differential Tannakian theory developed by Buium, Ovchinnikov, and others, showing that differential algebraic groups arise as the Tannakian groups of σ‑compatible categories.

These examples demonstrate that the new framework truly generalizes both the algebraic and differential Tannakian theories, while also handling mixed operator settings that were previously out of reach.

The paper concludes with a discussion of future directions. The authors suggest developing an “operator‑preserving cohomology” theory, studying torsors under operator‑linear groups, and classifying operator‑linear group schemes themselves. They also point out potential interactions with model theory (e.g., the model theory of fields with operators) and applications to the Galois theory of mixed difference‑differential equations.

In summary, the work provides a comprehensive categorical apparatus that unifies the representation theory of linear algebraic groups, differential algebraic groups, and more exotic operator‑defined groups under a single Tannakian umbrella. By carefully integrating the operator actions into the tensor structure and the fiber functor, the authors achieve a reconstruction theorem that mirrors the classical case while faithfully encoding the additional algebraic constraints imposed by the operators. This opens the door to systematic study of symmetry groups in a broad spectrum of algebraic–analytic contexts, ranging from pure algebraic geometry to the theory of functional and differential equations.