On Homogeneous CR Manifolds of Arbitrary Order of Levi Nondegeneracy
Stefano Marini, Costantino Medori · 2025
Read On Homogeneous CR Manifolds of Arbitrary Order of Levi Nondegeneracy by Stefano Marini, Costantino Medori (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2506-00897.txt.
Introduction
Stefano Marini and Costantino Medori present in 2025 a family of manifolds that are the same at every point. These manifolds stay nondegenerate to any set order, a trait that does not depend on how one charts them. They build the family from algebras tied to a classic group. A close study follows of the Levi forms and their kernels, plus the local model equation. For each order, the work gives clear examples where this higher trait can be checked in one uniform frame.
Catalog record
| CREATOR | Stefano Marini, Costantino Medori |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 51 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2506-00897
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
On Homogeneous CR Manifolds of Arbitrary Order of Levi Nondegeneracy Thanks: The Authors were partially supported by PRIN 2022MWPMAB — “Interactions between Geometric Structures and Function Theories” CUP: H53D23002040006; and and by the group G.N.S.A.G.A. of I.N.d.A.M. Stefano Marini, Costantino Medori Abstract . This paper present homogeneous CR hypersurfaces satisfying the C R CR -invariant property of being k k -nondegenerate for an arbitrary integer k ≥ 1 k\geq 1 . The construction of such homogeneous manifolds is based on C R CR algebras defined by irreducible representations of 𝔰 𝔲 ( 2 ) \mathfrak{su}(2) . An explicit study of the iterated Levi forms with their respective kernels, along with the local model equation, is given. 2020 Mathematics Subject Classification 32V05, 32V40, 32V35, 32V30, 53C30, 53C10. Introduction In Cauchy-Riemann geometry , a well-known problem is finding equivalence classes under the relation of CR diffeomorphisms. One approach to solving this problem is to find sufficiently many CR invariants that characterize each class, called the equivalence problem. This problem has been classically solved for CR manifolds that are Levi nondegenerate real hypersurfaces and a model for the homogeneous case is given by \mathbf{SU}{(1+p,q+1)}/\mathbf{H}\simeq\left\{\ [z_{0},\ldots,z_{n+1}]\,|\,\sum_{i=0}^{p}|z^{i}|^{2}{=}\sum_{i=p+1}^{n+1}|z^{i}|^{2}\ \right\}\subset\mathbb{CP}^{n+1} (see [ 10 , 42 , 43 , 44 ] ). In particular, in the context of (locally) homogeneous manifolds, this problem can be studied using the tool of CR algebras [ 33 ] , i.e., a pair (\mathfrak{g}^{\tau},\mathfrak{f}) , where \mathfrak{g}^{\tau} is a real Lie algebra encoding the group of CR diffeomorphisms, and \mathfrak{f}\subset\mathfrak{g}=\mathbb{C}\otimes\mathfrak{g}^{\tau} is a complex Lie subalgebra encoding the CR structure. In [ 31 ] , a classification of all semisimple Levi-Tanaka algebras was obtained. These are the Levi nondegenerate CR algebras (\mathfrak{g}^{\tau},\mathfrak{f}) , where \mathfrak{g} is semisimple, and \mathfrak{g}^{\tau} inherits a \mathbb{Z} -gradation from \mathfrak{g} that is compatible with the CR structure. Several examples of homogeneous CR manifolds are associated with CR algebras (\mathfrak{g}^{\tau},\mathfrak{f}) that are k- nondegenerate for some k{>}1 , i.e., Levi degenerate but also satisfying a weaker condition of nondegeneracy according to their Levi form (see [ 5 , Def.11.1.8 p.317] ). In the case of a point p in a generic analytic CR hypersurface, this notion is equivalent to holomorphic nondegeneracy at p , that is the absence of a germ at p of a holomorphic vector field X tangent to M (i.e., with a representative tangent to M in a neighborhood of p ) such that X\mid_{M}\nequiv 0 (see [ 5 , Th. 11.5.1, p. 330] ). This, in turn, implies the finiteness of the dimension of the CR automorphism group at p . The k- nondegeneracy is widely used in the literature as a key invariant in solving the equivalence problem or studying the group of CR automorphisms, see [ 20 , 34 , 22 , 40 , 36 , 21 , 17 ] . In particular [ 12 , 13 ] the Authors pointed that it would be interesting to construct homogeneous k- nondegenerate CR manifolds for arbitrarily large k . Quoting verbatim: “We do not know how to construct k -nondegenerate homogeneous manifolds with k\geq 5 (should these exist)”. In this paper we address this question. In Section 1 , we recall some basic definitions in CR geometry. In Section 2 , we translate these definitions into the context of homogeneous CR manifolds; in particular, we review some tools defined for non-semisimple homogeneous CR manifolds. Finally, in Section 3 , we define and characterize an example of a homogeneous CR manifold that is k -nondegenerate for arbitrary k\geq 1 , providing a local equation in Section 4 (Theorem 4.7 ). This last result should be compared, for k=1 , with [ 11 , Theorem 1.1] , and for k=2 , with [ 39 , Table 1, Type VII] (see also [ 7 , (19), p. 37] ). For arbitrary k , we also refer to [ 23 , Example 1] . 1. CR manifolds Let M be a (2n+d) -dimensional CR manifold, i.e. a differential manifold equipped with a CR structure given by a rank n complex distribution \textsf{H}^{0,1}\textsf{M}\subset\mathbb{C}\otimes\textsf{T}\textsf{M} such that (1) \textsf{H}^{0,1}\textsf{M}\cap\overline{\textsf{H}^{0,1}\textsf{M}}=\{0\} (2) [\Gamma(\textsf{H}^{0,1}\textsf{M}),\Gamma(\textsf{H}^{0,1}\textsf{M})]\subset\Gamma(\textsf{H}^{0,1}\textsf{M}) . We call (\textsf{M},\textsf{H}^{0,1}\textsf{M}) a Cauchy–Riemann manifold of CR-dimension n and CR-codimension d . When d=1 , we refer to it as being of hypersurface type . A CR manifold can also be given by a rank 2n real subbundle \textsf{H}_{\mathbb{R}}\textsf{M} of T M which is the real distribution underlying the CR structure, i.e., \textsf{H}_{\mathbb{R}}\textsf{M}=\textrm{Re}\left(\textsf{H}^{0,1}\textsf{M}\oplus\textsf{H}^{1,0}\textsf{M}\right) and a smooth complex structure \textrm{J}_{\textsf{M}} on the fibers of \textsf{H}_{\mathbb{R}}\textsf{M} , defined by \textsf{H}^{0,1}\textsf{M}{=}\{X+i\,\textrm{J}_{\textsf{M}}{X}\mid X\in\Gamma(\textsf{H}_{\mathbb{R}}\textsf{M})\}, that it squares to minus identity and satisfies \textsf{N}_{\textrm{J}_{\textsf{M}}}(X,Y){=}[\textrm{J}_{\textsf{M}}X,\textrm{J}_{\textsf{M}}Y]-[X,Y]-\textrm{J}_{\textsf{M}}([\textrm{J}_{\textsf{M}}X,Y]+[X,\textrm{J}_{\textsf{M}}Y])=0 for any two sections X,Y\in\Gamma(\textsf{H}_{\mathbb{R}}\textsf{M}) , taking the name of partial complex structure of M . We label by \mathcal{L}^{1} its Levi form , which it is a sesquilinear form on \textsf{H}^{0,1}\textsf{M} valued in \mathbb{C}\textsf{T}\textsf{M}/(\textsf{H}^{0,1}\textsf{M}\oplus\overline{\textsf{H}^{0,1}\textsf{M})} given at point p in M by (1) \mathcal{L}^{1}(X_{\textsf{p}},Y_{\textsf{p}})=\frac{1}{2i}\left[X,\overline{Y}\right]_{\textsf{p}}\pmod{\textsf{H}^{0,1}_{\textsf{p}}\textsf{M}\oplus\overline{\textsf{H}^{0,1}_{\textsf{p}}\textsf{M}}} for all X,Y\in\Gamma(\textsf{H}^{0,1}\textsf{M}) . The kernel at {\textsf{p}}\in\textsf{M} of \mathcal{L}^{1} is given by \mathcal{F}_{\textsf{p}}^{1}=\left\{v\in\textsf{H}^{0,1}_{\textsf{p}}\textsf{M}\,\left|\,\mathcal{L}^{1}(v,w)=0\,\,\forall\,w\in\textsf{H}^{0,1}_{\textsf{p}}\textsf{M}\right.\right\}. When \mathcal{F}_{\textsf{p}}^{1} is of constant rank at every point then \mathcal{F}^{1}=\bigcup_{\textsf{p}}\mathcal{F}_{\textsf{p}}^{1} is a vector distribution named Levi Kernel . A natural generalization of ( 1 ) then is the k- order Levi form (2) \mathcal{L}^{k}:\textsf{H}^{0,1}\textsf{M}\times\left(\bigodot^{k}\textsf{H}^{0,1}\textsf{M}\right)\to\mathbb{C}\textsf{T}\textsf{M}/(\textsf{H}^{0,1}\textsf{M}\oplus\overline{\textsf{H}^{0,1}\textsf{M}}) inductively defined at a point p of M as \mathcal{L}^{k}\left(X_{\textsf{p}},Y^{1}_{\textsf{p}},\ldots,Y^{k}_{\textsf{p}}\right)=\frac{1}{2i}\left[\cdots\left[\left[X,\overline{Y^{1}}\right],\cdots\right],\overline{Y^{k}}\right]_{\textsf{p}}\pmod{\textsf{H}^{0,1}_{\textsf{p}}\textsf{M}\oplus\overline{\textsf{H}^{0,1}_{\textsf{p}}\textsf{M}}} with X,Y^{1},\ldots,Y^{k}\in\Gamma\left(\textsf{H}^{0,1}\textsf{M}\right) . The left kernels of ( 2 ) at \textsf{p}\in\textsf{M} is given by (3) \displaystyle\mathcal{F}_{\textsf{p}}^{k}=\{v\in\textsf{H}^{0,1}_{\textsf{p}}\textsf{M}\,|\,\mathcal{L}^{k}(v,w_{1},\ldots,w_{k})=0\,\,\forall\,w_{1},\ldots,w_{k}\in\textsf{H}^{0,1}_{\textsf{p}}\textsf{M}\}. Again when the \mathcal{F}_{\textsf{p}} are of constant rank at every points the vector distributions \mathcal{F}^{k}=\bigcup_{{\textsf{p}}\in M}\mathcal{F}_{\textsf{p}}^{k} give rise to a nested sequence of complex subbundle of \textsf{H}^{0,1}\textsf{M} (see [ 14 ] ) (4) \textsf{H}^{0,1}\textsf{M}\supseteq\mathcal{F}^{1}\supseteq\mathcal{F}^{2}\dots\supseteq\mathcal{F}^{k-1}\supseteq\mathcal{F}^{k}\supseteq\dots Definition 1.1 (Def.11.1.8 [ 5 ] ) . A CR manifold M is k- nondegenerate , at {\textsf{p}}\in\textsf{M} , if there exist a positive integer k such that \mathcal{F}_{\textsf{p}}^{(k-1)}\supsetneqq\mathcal{F}_{\textsf{p}}^{(k)}{=}\{0\}, otherwise we say that it is finitely degenerate . 2. Homogeneous CR manifolds and CR algebras From now on we will denote by \mathfrak{g}^{\tau} a real form of a complex Lie algebra \mathfrak{g} , i.e., a subalgebra of \mathfrak{g} such that \mathfrak{g}^{\tau}\otimes\mathbb{C}=\mathfrak{g} . To \mathfrak{g}^{\tau} , we can associate an anti- \mathbb{C} linear involution \tau , that makes \mathfrak{g}^{\tau} the subalgebra of the \tau -fixed elements of \mathfrak{g} . Let \mathbf{G}^{\tau} be a Lie group with \mathrm{Lie}\,\mathbf{G}^{\tau}=\mathfrak{g}^{\tau} acting transitively on a CR manifold M . Fix a point \textsf{p}\in\textsf{M} , and let \piup\colon\mathbf{G}^{\tau}\ni\textsf{a}\mapsto\textsf{a}\cdot\textsf{p}\in\textsf{M} be the natural projection. The differential at p defines a map \piup_{*}\colon\mathfrak{g}^{\tau}\to\textsf{T}_{\textsf{p}}\textsf{M}, which maps the Lie algebra \mathfrak{g}^{\tau} of \mathbf{G}^{\tau} onto the tangent space to M at p . By the formal integrability of the partial complex structure of M , the complexification of the pullback of the CR structure at p is a complex Lie subalgebra \mathfrak{f}=(\piup^{*})^{\mathbb{C}}(\textsf{H}^{0,1}_{\textsf{p}}\textsf{M}) of the complexification \mathfrak{g} of \mathfrak{g}^{\tau} . Conversely, the assignment of a complex Lie subalgebra \mathfrak{f} of \mathfrak{g} yields a formally integrable \mathbf{G}^{\tau} -equivariant partial complex structure on a \mathbf{G}^{\tau} -homogeneous space M by requiring that \textsf{H}_{\textsf{p}}^{0,1}\textsf{M}=\piup^{\mathbb{C}}_{*}(\mathfrak{f}) (see [ 30 ] ). These lead to the following definition: Definition 2.1 . A CR algebra is a pair (\mathfrak{g}^{\tau},\mathfrak{f}) , consisting of a real Lie algebra \mathfrak{g}^{\tau} and a complex Lie subalgebra \mathfrak{f} of its complexification \mathfrak{g}, such that the quotient \mathfrak{g}^{\tau}/(\mathfrak{g}^{\tau}\cap\mathfrak{f}) is a finite-dimensional real vector space. We call the intersection \mathfrak{f}\cap\mathfrak{g}^{\tau} its isotropy subalgebra . The CR -dimension and codimension of