Strongly invariant differential operators on parabolic geometries modelled on Gr(3,3)
Listen
Radio episode about this paper
Transcript
Introduction to the show: ident: Mathematics. The day in mathematics: new theorems, new proofs, and what they open.
Ada: We're looking at the paper: "Strongly invariant differential operators on parabolic geometries modelled on Gr(3,3)".
Emmy: Here's the gist of the paper. We consider the curved geometries modelled on the homogeneous space G/P, where G = SL(6, R) acts transitively on the Grassmannian Gr(3, 3) of three-dimensional subspaces in R6, and P is the corresponding isotropic subgroup The paper classifies strongly invariant operators between sections of vector bundles induced on such geometries by irreducible P-modules Basic Concepts and Notation Grassmannian manifolds are homogeneous spaces Gr(m, n) = G/P, where G = SL(m + n, R) and P is its parabolic subgroup The Lie algebra p of P has got sl(m, R) × sl(n, R) as the simple part of the Levi subalgebra of p We shall deal with normal parabolic geometries M modelled on the homogeneous space G/P, where G = SL(6, R) acts transitively on the Grassmannian Gr(3, 3) of three-dimensional subspaces in R6 and P is the corresponding isotropic subgroup Irreducible representations of g are labeled by highest weights λ = (λ1λ2λ3λ4λ5λ6) with λi ∈ Z; i = 1,..., 6; λi ≥ λi+1; i = 1,..., 5, modulo the equivalence relation: λ'≃ λ if and only if λ'i = λi + c; i = 1..., 6, for any fixed integer c ∈ Z Often we shall fix a representative of the equivalence relation by the additional condition λ6 = 0 The differences λi − λi+1 correspond to the coefficients in the alternative description of the highest weights as linear combinations of the fundamental weights Thus, ρ = (543210) is the lowest weight in the interior of the dominant Weyl chamber of (g) The Lie algebra g is 1-graded, it decomposes under the adjoint action of g0 into irreducible components as g = g−1 ⊕ g0 ⊕ g1 Classification of Homomorphism between Verma Modules The order of a homomorphism Φ: V (α) → V (β) is defined as ϕ(β) − ϕ(α) (and this clearly corresponds to the order of the corresponding differential operators The Verma modules are U(g)-modules Clearly, nontrivial homomorphisms between two Verma modules may exist only if the actions of the center Z(g) of U(g) on them coincide The action of Z(g) is called the infinitesimal character and it is always shared just by Verma modules for the weights α on one orbit of the affine Weyl group action (and this is the reason why the notation α = λ + ρ is so useful The infinitesimal character of V (α) is called regular if the affine orbit contains only one gdominant inside the dominant Weyl chamber, and this happens if and only if all components αi, i = 1..., 6 are mutually different The case of 2-singular infinitesimal characters The 2-singular infinitesimal characters are parametrized by g-dominant weights in the intersection of two walls of the dominant Weyl chamber for g Hence there are six different cases parametrized by dominant weights with two repeating components For each such case, there is a unique homomorphism Ψ (up to a multiple) Theorem 1 gives the complete list of homomorphisms Ψ: V (α) → V (β) between Verma modules with 2-singular infinitesimal characters The two 1st order analogues of the Dirac operators are: 0 • 1 • −4 × 0 • → 0 • → -3 × 0 •, and the analogue of the Laplace operator between suitable densities has got order three: 0 • −4 × 0 • → -2 ×.
Ada: The case of regular infinitesimal character Theorem 3 gives the list of all homomorphisms between Verma modules with a fixed regular infinitesimal character (abcdef) is given in the following diagram The structure of the scheme is independent of the chosen infinitesimal character, but the order of operators is changing In the lowest regular infinitesimal character, the standard homomorphisms correspond to the relevant restrictions and projections of the exterior differential on exterior forms, further there are 6 nonstandard operators of order 4 (dotted arrows), one nonstandard operator of order 7, and one nonstandard operator between functions and top forms (of order 9) The translation principle Suppose W is a G-module (and P-module by restriction), and E a P-module Then U(g) ⊗ E∗ ⊗ W∗, as a g-module, can be viewed in two different ways: X(x ⊗ e ⊗ w) = Xx ⊗ e ⊗ w or X(x ⊗ e ⊗ w) = Xx otimes e × w The identity on I 1e × w extends to an isomorphism between these two modules: (1) V (E × W) = V (E) ⊗ W∗ This simple proposition implies again that all first and second order linear invariant differential operators extend canonically from the Klein’s models to the entire category of the corresponding Cartan geometries Semiholonomic Verma modules and morphisms The main tool consists in understanding the semi-holonomic jets of sections of the associated bundles E corresponding to a P-module E and the universal differential operator σ 7→ ¯j kσ on sections σ of the bundles E, provided by the iterated Cartan connection fundamental derivative The analogous construction to the Verma modules then provides the so-called semi-holonomic universal enveloping algebra U¯(g) and the topological dual to the infinite semi-holonomic jets V¯ (E) = U¯ ⊗U(p) E∗ Every nontrivial homomorphism Φ: V¯ (F) → V¯ (E) obviously provides an invariant differential operator E → F, of the same order Proposition 4 (
10, 13: ) states that a homomorphism V (F) → V (E) of order at most two always lifts to a homomorphism V¯ (F) → V¯ (E) The proof of Theorem 5 We start again in the lowest infinitesimal character (543210).
Emmy: All standard homomorphisms (included in the BGG complex) are of the first order, hence we can lift them To treat nonstandard ones, we are going to use again one-way translations from 1-singular cases The next theorem resolves the two nonstandard homomorphisms of order 4 in the lowest possible infinitesimal character and this proves the very last claim in Theorem 5 Theorem 6 The homomorphisms (A) (521430) → (543210) and (B) (210543) → (430521) cannot be lifted to semi-holonomic case The proof of Theorem 6 (A) In terms of highest weights, we deal with the homomorphism V (E) → V (F), where E = (0, −2, −22, 2, 0) and F = (000000) Suppose now that there is a lift of the homomorphism (A) to semi-holonomic Verma modules corresponding to a singular vector w ⊗ e, where e is a basis vector for the trivial module F, and w has the weight (0, −2, −22, 2, 0), and it is a (noncommutative) polynomial in variables yij ∈ g−1 Due to its weight, the polynomial depends only on yij, 3 ≤ i ≤ 4, 1 ≤ j ≤ 2 Then the same formula also defines a singular vector for a semi-holonomic lift of a nonstandard homomorphisms in g′, because the actions of elements in g′0 and g′1 on w⊗e coincide with the actions of elements in g0 and g1 on w ⊗e The proof of Theorem 5 The proof of the statement for higher level infinitesimal characters is finished by standard translation procedure using the defining representation W = (100000) and Proposition 1 The next theorem resolves the two nonstandard homomorphisms of order 4 in the lowest possible infinitesimal character and this proves the very last claim in Theorem 5 References
1: B.
Ada: Boe, D.
Paper summary: Ada: To recap where we are: we've seen that this paper is classifying strongly invariant differential operators between sections of vector bundles induced on curved geometries modeled on Gr(three three) <ref:2412.20369#pg1,operators between sections of vector bundles induced on>. The main point is that they are classifying the homomorphisms between semi-holonomic Verma modules using irreducible P-modules.
Emmy: They claim that these operators automatically extend to all curved geometries, and this classification corresponds exactly to the classification of homomorphisms of semi-holonomic Verma modules, which is what they summarize in their recent survey article.
Kolya: So what's the big picture takeaway for someone not deep in representation theory? Is it just confirming that the structure holds up under curvature?
Ada: It’s more than confirmation; it’s a statement that these operators are intrinsically linked to the geometry's underlying algebraic structure, specifically through those P-modules. If you understand how those modules work, you understand the operators on any space modeled on Gr(three three) <ref:2412.20369#pg1>.
Emmy: The significance is that they use representation theory—the highest weights and their equivalence relations—to precisely determine which differential operators can exist in these curved settings. They move the classification from a purely geometric problem to one rooted in algebra.
Kolya: That seems like a shift. So instead of just finding an operator and checking if it works, you are classifying the allowed operators based on representation theory first?
Ada: Precisely. You start with the algebraic side—the P-modules—and that determines the possible differential operators on any curved space modeled on Gr(three three) <ref:2412.20369#pg1>. The paper is about that precise mapping.
Conclusion: Ada: So, looking at the title "Strongly invariant differential operators on parabolic geometries modelled on Gr(three three)" it tells you a lot about the focus: they are focusing on a very specific type of geometry and operator symmetry <ref:2412.20369#pg1,Strongly invariant differential operators on parabolic geometries modelled on Gr(3, 3>.
Emmy: It emphasizes that these operators aren't just any operators; they must be strongly invariant, which means they have to respect the full structure of the geometry and the induced modules from P-modules.
Kolya: What does this mean in simple terms for someone outside this field? Does it suggest anything about how we view symmetry in physics or mathematics generally?
Ada: It suggests that deep symmetries are not accidental; they come from deep algebraic structures, like the way groups act on spaces, and these symmetries manifest as constraints on the allowed mathematical tools you can use.
Emmy: The implication is that understanding these specific classification results gives us a concrete way to understand how algebraic representations translate into tangible geometric objects—the differential operators themselves.
Kolya: So it’s like finding the precise rules for how symmetry dictates what kinds of physics we can even write down in a given background?
Ada: That’s a good way to put it. It moves the discussion from just describing what's there to understanding why only these specific things can be there.
JAN SLOVAK, VLADIMćIR SOUCEK
Masaryk University · Charles University
math.DG
Submitted: 2024-12-29
Updated: 2026-11-10
Comments: 24 pages, new section and 2 ancillary files with Python code added
License: http://creativecommons.org/licenses/by-nc-sa/4.0/
Importance score: 83/100
The gist: We consider the curved geometries modelled on the homogeneous space G/P, where G = SL(6, R) acts transitively on the Grassmannian Gr(3, 3) of three-dimensional subspaces in R6…
Key concepts
- Homogeneous Space G/P
- This is the geometric setting where the problem is studied. Gr(3,3) is a specific example of such a space, representing three-dimensional subspaces in R6. The group G acts transitively on this space, meaning all points look geometrically the same relative to each other.
- Verma Modules
- These are algebraic modules used to study differential operators. They are constructed from representations of the Lie algebra g and help describe how sections of vector bundles behave under certain differential operations. The order of a homomorphism between them corresponds to the order of the resulting differential operator.
- |2|-singular Infinitesimal Characters
- These specific characters arise when analyzing homomorphisms between Verma modules. They are parametrized by weights with two repeating components, leading to unique homomorphisms up to a multiple. These cases yield specific low-order differential operators, such as those related to the Dirac and Laplace operators.
- Semi-holonomic Verma Modules
- These modules are used for a more general approach involving jets of sections. They allow the construction of a universal enveloping algebra, U¯(g), which provides a framework to lift homomorphisms from simpler Verma modules to those in the full category, enabling the study of nonstandard operators.
Terminology
Summary
We consider the curved geometries modelled on the homogeneous space G/P, where G = SL(6, R) acts transitively on the Grassmannian Gr(3, 3) of three-dimensional subspaces in R6, and P is the corresponding isotropic subgroup
The paper classifies strongly invariant operators between sections of vector bundles induced on such geometries by irreducible P-modules
Basic Concepts and Notation
Grassmannian manifolds are homogeneous spaces Gr(m, n) = G/P, where G = SL(m + n, R) and P is its parabolic subgroup The Lie algebra p of P has got sl(m, R) × sl(n, R) as the simple part of the Levi subalgebra of p We shall deal with normal parabolic geometries M modelled on the homogeneous space G/P, where G = SL(6, R) acts transitively on the Grassmannian Gr(3, 3) of three-dimensional subspaces in R6 and P is the corresponding isotropic subgroup Irreducible representations of g are labeled by highest weights λ = (λ1λ2λ3λ4λ5λ6) with λi ∈ Z; i = 1,..., 6; λi ≥ λi+1; i = 1,. p. 1 Let g be the Lie algebra of G. Irreducible representations of g are labeled by highest weights
λ = (λ1λ2λ3λ4λ5λ6) with λi ∈ Z; i = 1, .Where the paper says it
Classification of Homomorphism between Verma Modules
The order of a homomorphism Φ: V (α) → V (β) is defined as ϕ(β) − ϕ(α) (and this clearly corresponds to the order of the corresponding differential operators The Verma modules are U(g)-modules Clearly, nontrivial homomorphisms between two Verma modules may exist only if the actions of the center Z(g) of U(g) on them coincide The action of Z(g) is called the infinitesimal character and it is always shared just by Verma modules for the weights α on one orbit of the affine Weyl group action (and this is the reason why the notation α = λ + ρ is so useful The infinitesimal character of V (α) is called regular if the affine orbit contains only one gdominant inside the dominant Weyl chamber, and this happens if and only if all components αi, i = 1..., 6 are mutually different
The case of |2|-singular infinitesimal characters
The 2-singular infinitesimal characters are parametrized by g-dominant weights in the intersection of two walls of the dominant Weyl chamber for g Hence there are six different cases parametrized by dominant weights with two repeating components For each such case, there is a unique homomorphism Ψ (up to a multiple) Theorem 1 gives the complete list of homomorphisms Ψ: V (α) → V (β) between Verma modules with 2-singular infinitesimal characters The two 1st order analogues of the Dirac operators are: 0 • 1 • −4 × 0 • → 0 • → -3 × 0 •, and the analogue of the Laplace operator between suitable densities has got order three: 0 • −4 × 0 • → -2 ×.
The case of regular infinitesimal character
Theorem 3 gives the list of all homomorphisms between Verma modules with a fixed regular infinitesimal character (abcdef) is given in the following diagram The structure of the scheme is independent of the chosen infinitesimal character, but the order of operators is changing In the lowest regular infinitesimal character, the standard homomorphisms correspond to the relevant restrictions and projections of the exterior differential on exterior forms, further there are 6 nonstandard operators of order 4 (dotted arrows), one nonstandard operator of order 7, and one nonstandard operator between functions and top forms (of order 9)
The translation principle
Suppose W is a G-module (and P-module by restriction), and E a P-module Then U(g) ⊗ E∗ ⊗ W∗, as a g-module, can be viewed in two different ways: X(x ⊗ e ⊗ w) = Xx ⊗ e ⊗ w or X(x ⊗ e ⊗ w) = Xx otimes e × w The identity on I 1e × w extends to an isomorphism between these two modules: (1) V (E × W) = V (E) ⊗ W∗ This simple proposition implies again that all first and second order linear invariant differential operators extend canonically from the Klein’s models to the entire category of the corresponding Cartan geometries
Semiholonomic Verma modules and morphisms
The main tool consists in understanding the semi-holonomic jets of sections of the associated bundles E corresponding to a P-module E and the universal differential operator σ 7→ ¯j kσ on sections σ of the bundles E, provided by the iterated Cartan connection fundamental derivative The analogous construction to the Verma modules then provides the so-called semi-holonomic universal enveloping algebra U¯(g) and the topological dual to the infinite semi-holonomic jets V¯ (E) = U¯ ⊗U(p) E∗ Every nontrivial homomorphism Φ: V¯ (F) → V¯ (E) obviously provides an invariant differential operator E → F, of the same order Proposition 4 ([10, 13]) states that a homomorphism V (F) → V (E) of order at most two always lifts to a homomorphism V¯ (F) → V¯ (E)
The proof of Theorem 5
We start again in the lowest infinitesimal character (543210). All standard homomorphisms (included in the BGG complex) are of the first order, hence we can lift them To treat nonstandard ones, we are going to use again one-way translations from 1-singular cases The next theorem resolves the two nonstandard homomorphisms of order 4 in the lowest possible infinitesimal character and this proves the very last claim in Theorem 5
Theorem 6
The homomorphisms (A) (521430) → (543210) and (B) (210543) → (430521) cannot be lifted to semi-holonomic case
The proof of Theorem 6
(A) In terms of highest weights, we deal with the homomorphism V (E) → V (F), where E = (0, −2, −22, 2, 0) and F = (000000) Suppose now that there is a lift of the homomorphism (A) to semi-holonomic Verma modules corresponding to a singular vector w ⊗ e, where e is a basis vector for the trivial module F, and w has the weight (0, −2, −22, 2, 0), and it is a (noncommutative) polynomial in variables yij ∈ g−1 Due to its weight, the polynomial depends only on yij, 3 ≤ i ≤ 4, 1 ≤ j ≤ 2 Then the same formula also defines a singular vector for a semi-holonomic lift of a nonstandard homomorphisms in g′, because the actions of elements in g′0 and g′1 on w⊗e coincide with the actions of elements in g0 and g1 on w ⊗e
The proof of the statement for higher level infinitesimal characters is finished by standard translation procedure using the defining representation W = (100000) and Proposition 1 The next theorem resolves the two nonstandard homomorphisms of order 4 in the lowest possible infinitesimal character and this proves the very last claim in Theorem 5
References
[1] B. Boe, D. Collinwood: Intertwinning operators between holomorphically induced modules, Pacific Jour. p. 0 We consider the curved geometries modelled on the homogeneous space G/P, where
G = SL(6, R) acts transitively on the Grassmannian Gr(3, 3) of three-dimensional subspaces in
R6
, and P is the corresponding isotropic subgroup. We classify the strongly invariant operators
between sections of vector bundles induced on such geometries by irreducible P-modules, i.Where the paper says it
[2] B.Boe, M. Hunziker: Kostant modules in category OS, Comm. p. 3 The case of |2|-singular infinitesimal characters. The results found by Enright and
Shelton, [2], tell that the category of Verma modules with a given |2|-singular infinitesimal character
is equivalent to the category of Verma modules with regular infinitesimal character for the projective
one-dimensional space Gr(1, 1).Where the paper says it
[3] D.M.J. Calderbank, T. Diemer: Differential invariants and curved Bernstein-Gelfand-Gelfand sequences, J. p. 0 They correspond to the well
known patterns of (generalized) Verma module homomorphisms, a story going back to Gelfand,
Kostant and others, many decades back. There are several approaches how to extend and understand these patterns for the curved Cartan
geometries, cf.Where the paper says it
[4] A. Cap, J. Slov´ak: Parabolic geometries. p. 0 20369v1 [math. DG] 29 Dec 2024
STRONGLY INVARIANT DIFFERENTIAL OPERATORS ON PARABOLIC
GEOMETRIES MODELLED ON Gr(3, 3)
JAN SLOVAK AND VLADIM ´ ´IR SOUCEK ˇ
Abstract.Where the paper says it
[5] A. Cap, J. Slov´ak, V. Souˇcek: Bernstein-Gelfand-Gelfand sequences, Ann. p. 0 In particular, there is
the complete understanding of patterns of all invariant linear differential operators between the
homogeneous vector bundles for the flat Klein geometry models. They correspond to the well
known patterns of (generalized) Verma module homomorphisms, a story going back to Gelfand,
Kostant and others, many decades back.Where the paper says it
[6] A. Cap, V. Souˇcek: Curved Casimir operators and the BGG machinery, SIGMA Symmetry Integrability Geom. p. 9 The homogeneous bundles determined by P-modules extend to functors of the associated bundles
to the Cartan bundles G, but the question which of the invariant operators extend to natural
differential operators between such bundles is much more intriguing. On one hand, the standard
operators in the schemes with regular infinitesimal characters always admit such extensions and
they are available by direct constructions, known as the BGG machinery, cf.Where the paper says it
[7] A. Cap, V. Souˇcek: Subcomplexes in curved BGG-sequences, Math. p. 0 arXiv:2412. 20369v1 [math.Where the paper says it
[8] M.G. Eastwood: Notes on Conformal Differential Geometry, Supp. p. 6 STRONGLY INVARIANT DIFFERENTIAL OPERATORS ON Gr(3, 3) 7
of intertwining differential operators on line bundles with symbols given by
det( ∂
∂xij
)
n
, for all
n ∈ N. They are the analogues of powers of Laplacians in conformal geometry (see [12]).Where the paper says it
[9] M.G. Eastwood: Variations on the de Rham complex, Notices Amer. Math Soc., 46, 11 (1999), 1368–1376
[10] M. Eastwood, J. Slov´ak: Semiholonomic Verma modules, Jour. p. 9 Moreover, notice that we are always using a g-module with the composition
series of length at most three, and so the injection and projection operators in the translations are
always of order at most two! Semiholonomic Verma modules and morphisms
While the invariant differential operators between homogenous bundles over Klein geometries
G/P can be completely understood via the purely algebraic and functorial setup mentioned above,
the curved Cartan geometries are not that easy.Where the paper says it
[11] T.Enright, B. p. 3 4 JAN SLOVAK AND VLADIM ´ ´IR SOUCEK ˇ
The singular infinitesimal characters can be nicely treated using equivalence of categories due to
Enright-Shelton, [11, 2], providing a reduction of singular cases to the regular cases for lower rank
groups. The case of |2|-singular infinitesimal characters.Where the paper says it
[12] R. Howe, Soo Teck Lee: Spherical harmonics on Grassmannians, Colloq. Math., 118, 1 (2010), 349-364
[13] J. Slov´ak, V. Souˇcek: Semiholonomic jets and induced modules in Cartan geometry calculus, Arch. p. 9 Moreover, notice that we are always using a g-module with the composition
series of length at most three, and so the injection and projection operators in the translations are
always of order at most two! Semiholonomic Verma modules and morphisms
While the invariant differential operators between homogenous bundles over Klein geometries
G/P can be completely understood via the purely algebraic and functorial setup mentioned above,
the curved Cartan geometries are not that easy. 20369v1 [math. DG] 29 Dec 2024
STRONGLY INVARIANT DIFFERENTIAL OPERATORS ON PARABOLIC
GEOMETRIES MODELLED ON Gr(3, 3)
JAN SLOVAK AND VLADIM ´ ´IR SOUCEK ˇ
Abstract.Where the paper says it
Where the paper says it
[14] D. A. Vogan: Representation of Real Reductive Lie Groups, Progress in Mathematics, Vol.
Green — the paper says it, and the page is printed. Red — no passage carries it, so treat that sentence as unverified. The same check runs on your own text: check a summary against its source, $5.
Improvements for AI systems
-
Improved AI systems can classify strongly invariant differential operators between sections of vector bundles induced on parabolic geometries modeled on Gr(3, 3) by classifying homomorphisms of semi-holonomic Verma modules. This allows for a complete understanding of patterns of all invariant linear differential operators between the homogeneous vector bundles for the flat Klein geometry models.
-
Improved AI systems can predict whether a homomorphism between Verma modules lifts to a homomorphism between the corresponding semi-holonomic Verma modules based on the order and type of singularity of the infinitesimal characters. This capability is crucial for determining which nonstandard operators extend canonically from Klein's models to curved Cartan geometries.
-
Improved AI systems can perform translations of entire schemes of equisingular weights using g-modules with composition series of length at most three, ensuring that all projections and injections in the twisting procedure lift to morphisms between semi-holonomic Verma modules. This enables the recovery of all nonstandard operators from standard ones via a systematic algebraic procedure.
Abstract
We consider the curved geometries modelled on the homogeneous space G/P, where G=SL(6, R) acts transitively on the Grassmannian Gr(3,3) of three-dimensional subspaces in R 6, and P is the corresponding isotropic subgroup. We classify the strongly invariant operators between sections of vector bundles induced on such geometries by irreducible P-modules, i.e., those obtained via homomorphisms of semi-holonomic Verma modules.
Related papers
- Scalar curvature bounds and tangent cones for K"ahler--Ricci shrinkers
- Almost-splitting of spacetimes under timelike sectional curvature bounds
- The anisotropic Weyl law and Grassmannian equidistribution
- Hyperpolarity of polar foliations of codimension two
- A sharp pinching theorem for closed minimal Legendrian submanifolds in the unit sphere
- Projective Rigidity of U(k) -Radial Ricci-Flat K"ahler Metrics