The $\mathcal{D}$-Geometric Hilbert Scheme -- Part I: Involutivity and Stability

The Big Picture

Imagine trying to catalog every possible way water can flow. The equations governing fluid dynamics are partial differential equations (PDEs), and there are infinitely many solutions, each describing a different flow pattern. Now imagine building a map of all possible equation systems themselves: a geometric space where each point represents a different type of PDE, and nearby points represent equations that behave similarly. That’s moduli theory applied to PDEs, and it has remained out of reach until now.

PDEs are the backbone of physics. They describe how fields propagate, how particles interact, how spacetime curves. Yet despite centuries of study, mathematicians still lack a unified geometric framework for classifying systems of PDEs the way they can classify curves or surfaces. The problem is that PDEs carry hidden consistency requirements. Differentiate the equations repeatedly and new constraints emerge, making naive geometric approaches collapse.

Jacob Kryczka and Artan Sheshmani have constructed the first rigorous moduli space for a broad class of PDE systems using tools from algebraic geometry, and proved that their new stability condition recovers one of the deepest results in gauge theory.

By treating PDE systems as geometric objects called D-ideal sheaves and introducing Spencer stability, this work constructs a moduli space for differential equations that unifies classical stability theories in geometry.

How It Works

The starting point is a shift in perspective. Instead of studying a PDE by seeking its solutions, Kryczka and Sheshmani encode the equation itself as a geometric object: a D-ideal sheaf living inside the jet bundle of a manifold. Jet bundles package a function together with all its derivatives at each point, recording not just the value of a quantity but how fast it’s changing, how fast that rate is changing, and so on, to arbitrary depth. A PDE system carves out a structured subspace of this jet bundle, and the “D” refers to the ring of differential operators that captures how derivatives interact and compose.

The property they require is involutivity, a stringent compatibility condition that guarantees no hidden contradictions emerge when you differentiate repeatedly. An involutive system is maximally self-consistent: you can always extend it to higher orders without hitting new obstructions. This is the algebraic counterpart of the classical Cartan-Kähler theorem from differential geometry.

To build a moduli space, you need two things: a notion of equivalence (which PDEs count as “the same”?) and a boundedness condition (ensuring only finitely many equivalence classes share a given invariant). For classical vector bundles, geometric objects that attach a vector space to every point of a manifold, the answer to the second question is slope stability. Kryczka and Sheshmani introduce the analogous concept for D-ideal sheaves:

• Spencer cohomology encodes obstructions and deformations of the PDE system, standing in for ordinary bundle cohomology
• Spencer slope is defined via a new “D-Hilbert polynomial” measuring how the PDE ideal grows at high differential order
• Spencer (semi-)stability requires that every D-ideal subsheaf has Spencer slope no greater than the ambient ideal

From here, the authors define the D-Hilbert functor and D-Quot functor, direct analogs of Grothendieck’s classical constructions that now parametrize families of involutive PDE systems rather than algebraic subvarieties. The core technical result is that these functors are representable: a genuine geometric space exists whose points correspond bijectively to Spencer-stable D-ideal sheaves of fixed numerical type. Getting there requires showing that Spencer stability implies boundedness, meaning stable PDE systems with a given Hilbert polynomial form a bounded family. This step demands careful control of Castelnuovo-Mumford regularity for differential modules.

Why It Matters

The payoff is a concrete theorem. For flat connections on compact Kähler manifolds, geometric structures central to both mathematics and theoretical physics, Kryczka and Sheshmani prove that Spencer polystability of the associated PDE ideal is equivalent to the existence of a Hermitian–Yang–Mills metric. This refines the Donaldson–Uhlenbeck–Yau (DUY) correspondence, one of the landmark results connecting algebraic geometry to gauge theory. The DUY theorem says slope-stable holomorphic bundles admit Hermitian-Yang-Mills connections; this paper shows that the stability of the equations themselves, not just the bundle, is the fundamental condition.

Spencer cohomology subsumes de Rham, Dolbeault, and foliated cohomologies as special cases of the same formalism applied to different PDEs. The authors conjecture that K-stability, which governs the existence of Kähler-Einstein metrics and is a central open problem in complex geometry, should also emerge as a Spencer-stability condition for the Monge-Ampère equation. If that holds, Spencer stability would unify two of the deepest stability theories in modern geometry under a single concept.

The paper also points toward derived algebraic geometry. A forthcoming companion will construct derived enhancements of these moduli spaces (D-geometric Hilbert dg-schemes), capturing higher-order obstructions and homotopical invariants while connecting to Bridgeland stability conditions and categorical approaches to PDEs.

Spencer stability is a new fundamental concept that unifies gauge-theoretic and algebro-geometric stability in a single framework. The geometry of differential equations turns out to be far richer and more structured than previously known.