Universal Hermitian Projective Calculus for ℂ ​ H 2 \mathbb{C}\mathrm{H}^{2} Pair, Phase, Contact, and Kernel Invariants in Complex Hyperbolic Two-Space

Karl J. Kreder III · 2026

Read Universal Hermitian Projective Calculus for ℂ ​ H 2 \mathbb{C}\mathrm{H}^{2} Pair, Phase, Contact, and Kernel Invariants in Complex Hyperbolic Two-Space by Karl J. Kreder III (2026) 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-2606-26129.txt.

Introduction

Karl Kreder builds a single algebraic calculus for the geometry of complex hyperbolic two-space. That space can be seen as a ball with a special metric, a symmetric space, or a projective domain of negative lines. Its boundary carries a contact structure rather than an ordinary sphere. Real projective geometry has a tidy invariant language, built on quadrance and spread. The Hermitian setting refuses to collapse into one such pair.

Invariants multiply instead, and Kreder catalogs them. Hermitian pair moduli, determinant quadrance, polar spread, and triple-product phase each capture part of the picture. Cross-ratio laws, boundary contact formulas, and kernel identities on the normalized ball join the list. Everything grows from one datum: a complex space with a Hermitian form of signature two-one. The calculus gives workers in this geometry one toolkit instead of scattered formulas.

Catalog record

CREATORKarl J. Kreder III
YEAR2026
COLLECTIONPAPERS
FORMATTXT
FILE SIZE139 KB
LICENSEhttp://creativecommons.org/publicdomain/zero/1.0/
RIGHTSCC0
SOURCEarXiv

Source: https://arxiv.org/abs/2606.26129

Browse the full library · About the Tower of Babel

Full text

The opening of the work, from the library's hosted text:

SU(2) , these are the irreducibles of highest weights 1 , 2 , and 0 , respectively. Proof. The positive frame E_{x} is a two-dimensional Hermitian vector space, so the simple part of its unitary group is SU(E_{x})\cong SU(2) . Its defining action on E_{x} is the fundamental two-dimensional irreducible representation. The complexified adjoint action is the conjugation action on \mathfrak{su}(E_{x})_{\mathbb{C}}=\mathfrak{sl}(E_{x}), which is the three-dimensional irreducible representation of SU(2) . Finally, scalar frame-invariant quantities transform through the trivial one-dimensional representation. These three representations are canonical once the interior point x and hence the positive frame E_{x} have been chosen. ∎ Proposition 2.5 (Cartan tangent split at an interior point) . Let x=[z]\in\mathbb{C}\mathrm{H}^{2} and let V=L_{x}\oplus E_{x} be the local splitting above. The Lie algebra of SU(2,1) admits the Cartan decomposition \mathfrak{su}(2,1)=\mathfrak{k}_{x}\oplus\mathfrak{p}_{x}, where \mathfrak{k}_{x}\cong\mathfrak{s}\bigl(\mathfrak{u}(L_{x})\oplus\mathfrak{u}(E_{x})\bigr)\cong\mathfrak{u}(2), and \mathfrak{p}_{x}\cong\operatorname{Hom}_{\mathbb{C}}(L_{x},E_{x}) as a real vector space. Consequently T_{x}\mathbb{C}\mathrm{H}^{2}\cong\operatorname{Hom}_{\mathbb{C}}(L_{x},E_{x}),\qquad\dim_{\mathbb{C}}T_{x}\mathbb{C}\mathrm{H}^{2}=2,\quad\dim_{\mathbb{R}}T_{x}\mathbb{C}\mathrm{H}^{2}=4. Proof. Choose an h -orthonormal basis adapted to V=L_{x}\oplus E_{x} , so that the Hermitian matrix of h is J=\begin{pmatrix}-1&0\\ 0&I_{2}\end{pmatrix}. The Lie algebra \mathfrak{su}(2,1) consists of trace-zero complex matrices X satisfying X^{\ast}J+JX=0. Writing X in block form relative to L_{x}\oplus E_{x} , this condition is equivalent to X=\begin{pmatrix}i\alpha&b^{\ast}\\ b&A\end{pmatrix},\qquad\alpha\in\mathbb{R},\quad A\in\mathfrak{u}(2),\quad i\alpha+\operatorname{tr}A=0, with b\in\mathbb{C}^{2} . The block-diagonal part b=0 is the stabilizer of the line L_{x} ; it is \mathfrak{s}\bigl(\mathfrak{u}(L_{x})\oplus\mathfrak{u}(E_{x})\bigr)\cong\mathfrak{u}(2). The off-diagonal part is determined by b\in E_{x} , equivalently by a complex linear map L_{x}\to E_{x} after choosing a nonzero vector in L_{x} . Thus the complement is naturally identified, as a real vector space, with \operatorname{Hom}_{\mathbb{C}}(L_{x},E_{x}) . This complement is the tangent representation of the symmetric space at x , giving the stated complex and real dimensions. ∎ \mathbb{C}\mathrm{H}^{2}=\mathbb{P}(V_{-}) \partial\mathbb{C}\mathrm{H}^{2}=\mathbb{P}(V_{0}) [z] [\xi] L_{p} \mathcal{C}_{p} [p]\in\mathbb{P}(V_{+}) mixed data: interior point, boundary point, positive pole Figure 1: Schematic mixed configuration. A negative line [z] is an interior point, a null line [\xi] is a boundary point, and a positive pole [p] determines both the complex geodesic L_{p} and its boundary chain \mathcal{C}_{p} . The equations h(z,p)=0 and h(\xi,p)=0 are routed through different strata. 3 Uniform algebraic layers and stratum domains The calculus uses the complex model as its main geometric picture, but the algebraic expressions are deliberately organized so that finite Hermitian models use the same formulas. This requires separating four layers. Starred-field projective algebra. Let K be a field with involution \lambda\mapsto\lambda^{\ast} , and let h be a Hermitian form over a rank-three K -space. The scalar norm is N(\lambda)=\lambda\lambda^{\ast}. Pair numerators, pair denominators, two-point Gram determinants, triple products, and cross-ratio numerators and denominators are defined at this level. They are polynomial or rational Hermitian expressions with explicit projective scale weights. Polarity-stratified algebra. Over a general field with involution there may be no order and therefore no literal negative/positive inequality. The replacement is a scale-stable null predicate together with a chosen anisotropic or positive-pole predicate. This is the correct algebraic setting for boundary-chain incidence h(x,p)=0 and for finite Hermitian unital incidence. Analytic complex \mathbb{C}\mathrm{H}^{2} . When K=\mathbb{C} and h has signature (2,1) , the chosen predicates become the familiar strata V_{-},\qquad V_{0},\qquad V_{+}. This layer adds distance readouts, Cartan angular readouts, the Heisenberg boundary chart, CR/contact forms, and analytic kernel formulas. Finite Hermitian specialization. For a quadratic finite-field extension K^{\prime}/K , the involution is the Frobenius over K , and the norm is xx^{\ast} . The same projective Hermitian formulas give finite null points, positive-pole chains, norm-fiber row charts, and the rank-two Hermitian unital incidence package. Proposition 3.1 (Layer descent) . Every invariant in the starred-field projective layer specializes to the complex analytic and finite Hermitian layers after adding the appropriate extra structure. The complex layer adds ordered sign strata and CR/contact analysis; the finite layer replaces ordered signs by scale-stable null/anisotropic polarity strata and finite norm-fiber incidence. Proof. The underlying expressions are built from h , the involution, scalar multiplication, and projective rescaling. Their scale weights are algebraic: non-null pair quantities divide by the matching diagonal weight, while triple phase data quotient by norm scalars. Adding a complex signature or a finite Frobenius norm supplies extra interpretations of the same expressions, but it does not change their projective descent laws. ∎ Definition 3.2 (Field-uniform data) . Field-uniform CH2 data consist of the following four pieces. \begin{array}[]{ll}\text{A. starred projective data}&(K,\ast,V,h),\quad h(\lambda z,\mu w)=\lambda\mu^{\ast}h(z,w),\\ \text{B. projective strata}&\text{scale-stable null, non-null, and pole predicates},\\ \text{C. analytic readouts}&\text{ordered signs, CR/contact forms, kernels, and limits over }\mathbb{C},\\ \text{D. finite readouts}&\text{Frobenius norm fibers and finite Hermitian incidence.}\end{array} Only A and B are part of the algebraic core. C and D are specializations: the analytic layer adds inequalities and differential/contact readouts, while the finite layer replaces inequalities by polarity strata and norm-fiber counts. Definition 3.3 (Shared-determinant rank-pair group) . Let G_{3,2}=U(3)\times_{\det,U(1),\det}U(2)=\{(A,B)\in U(3)\times U(2):\det A=\det B\}. This is the compact unitary group obtained by coupling a rank-three unitary carrier and a rank-two unitary frame by a shared determinant. It is an auxiliary determinant-locking group attached to the rank pair (3,2) , not the holomorphic isometry group of \mathbb{C}\mathrm{H}^{2} . Theorem 3.4 (Shared-determinant quotient) . There is a natural isomorphism of compact Lie groups G_{3,2}\cong\frac{SU(3)\times SU(2)\times U(1)}{\mu_{6}}, where \mu_{6}=\{\lambda\in U(1):\lambda^{6}=1\} is embedded by \lambda\longmapsto(\lambda^{-2}I_{3},\lambda^{-3}I_{2},\lambda). Proof. Define \Phi:SU(3)\times SU(2)\times U(1)\longrightarrow U(3)\times U(2) by \Phi(U,V,\lambda)=(\lambda^{2}U,\lambda^{3}V). For the first factor, \det(\lambda^{2}U)=(\lambda^{2})^{3}\det U=\lambda^{6}, and for the second factor, \det(\lambda^{3}V)=(\lambda^{3})^{2}\det V=\lambda^{6}. Thus \Phi lands in G_{3,2} . The map is onto. Given (A,B)\in G_{3,2} , write \delta=\det A=\det B\in U(1). Choose \lambda\in U(1) with \lambda^{6}=\delta . Then U=\lambda^{-2}A,\qquad V=\lambda^{-3}B satisfy \det U=\lambda^{-6}\det A=1,\qquad\det V=\lambda^{-6}\det B=1. Hence U\in SU(3) , V\in SU(2) , and \Phi(U,V,\lambda)=(A,B). It remains to compute the kernel. If \Phi(U,V,\lambda)=(I_{3},I_{2}), then U=\lambda^{-2}I_{3},\qquad V=\lambda^{-3}I_{2}. The determinant conditions U\in SU(3) and V\in SU(2) give 1=\det U=(\lambda^{-2})^{3}=\lambda^{-6},\qquad 1=\det V=(\lambda^{-3})^{2}=\lambda^{-6}. Thus \lambda^{6}=1 . Conversely, if \lambda^{6}=1 , then (\lambda^{-2}I_{3},\lambda^{-3}I_{2},\lambda) lies in SU(3)\times SU(2)\times U(1) and maps to (I_{3},I_{2}) . Therefore \ker\Phi=\{(\lambda^{-2}I_{3},\lambda^{-3}I_{2},\lambda):\lambda^{6}=1\}\cong\mu_{6}. The first isomorphism theorem gives the displayed quotient. ∎ Definition 3.5 (Central residue coordinates) . Let A_{1}\in\mathbb{Z}/6\mathbb{Z} denote the exponent of the shared \mu_{6} -coordinate in the quotient presentation of G_{3,2} . Let Y_{2}\in\mathbb{Z}/2\mathbb{Z} denote the exponent of the central \mu_{2} -coordinate of the rank-two frame projection. The common sixth-root lattice embeds the frame coordinate by Y_{2}\longmapsto 3Y_{2}\in\mathbb{Z}/6\mathbb{Z}. The associated central residue is Q_{3,2}=\frac{A_{1}}{6}+\frac{Y_{2}}{2}=\frac{A_{1}+3Y_{2}}{6}\quad\in\quad\frac{1}{6}\mathbb{Z}/\mathbb{Z}. Proposition 3.6 (Rank-pair residue lattice) . Let t\in\mathbb{Z}/3\mathbb{Z} be the rank-three central exponent and let w\in\mathbb{Z}/2\mathbb{Z} be the rank-two central exponent. In the common \mu_{6} -lattice define Y_{6}\equiv-2t+3w\pmod{6}. If k=\pm 1 denotes the two frame orientations, represented in the same lattice by 3k , then the four first central residues are Q_{3,2}(t,w,k)=\frac{Y_{6}+3k}{6}. For the rank-three singlet class t=0 and the first rank-two class w=1 , this gives 0,\qquad-1. For the rank-three fundamental class t=1 and w=1 , this gives \frac{2}{3},\qquad-\frac{1}{3}. Proof. The center of SU(3) is \mu_{3} , so its primitive exponent is represented in the common \mu_{6} -lattice by multiplication by 2 . The center of SU(2) is \mu_{2} , so its primitive exponent is represented by multiplication by 3 . The sign convention in the quotient embedding gives the combined central coordinate Y_{6}\equiv-2t+3w\pmod{6}. The two frame orientations differ by the two representatives k=\pm 1 , which contribute 3k on the same sixth-root lattice. Therefore Q_{3,2}=\frac{Y_{6}+3k}{6}. For t=0,w=1 , one has Y_{6}\equiv 3\pmod{6} . Using the representative Y_{6}=-3 gives k=+1:\quad\frac{-3+3}{6}=0,\qquad k=-1:\quad\frac{-3-3}{6}=-1. For t=1,w=1 , one has Y_{6}\equiv-2+3=1\pmod{6}

Read the complete text