Factorial-prefactor matching for the finite Tet sum
Proved
QuantumSixJ.sum_prefactor_match
by igor.kh · Nov 13, 2026 · Mathlib 0df444a (Lean v4.33.1)
Let a,b,c,d,j,l be integers satisfying horizontal and vertical admissibility, with j>0. Write R for the three-term residual of the normalized Appendix Tet sum, P for its Appendix factorial prefactor, Dj=[2][j][j+1][j+2], and R for the sum of cleared raw neighboring residuals defined by the certificate bundle. Then
R=DjPR.
This identity relates independently defined finite sums, including absent-neighbor cases; it does not assert that either residual vanishes.
Independent mathematical read-back
Read-back: QuantumSixJ.sum_prefactor_match
For every six integers a,b,c,d,j,l, assume that each of the four triples (a,d,j), (b,c,j), (a,b,l), (c,d,l) consists of nonnegative integers, satisfies all three triangle inequalities, and has even total, and additionally assume j>0. The theorem asserts the rational-function identity
with every quantity specified as follows within this paragraph. All scalars belong to K=Q(q), where q is the image of the polynomial indeterminate, and [u]=(qu−q−u)/(q−q−1) for every integer u; write F(u)=0 for u<0 and F(u)=∏v=1u[v] otherwise, so F(0)=1. All inverses and divisions are total field operations, with 0−1=0 and x/0=0; integer powers are field powers, every label subtraction is integer subtraction, and every half-sum below uses Euclidean integer division by 2. Let T(u,v,w) denote exactly the nonnegativity, three triangle inequalities, and even-total condition just stated, and put Hr=T(a,d,r)∧T(b,c,r) and Vs=T(a,b,s)∧T(c,d,s). Define
Here the bare symbol F is an integer parameter and F(⋅) continues to mean the signed-index factorial; the maximum is converted to a natural number before adding 3. Set
The sum over S consists of the natural numbers 0,…,n−1 cast to integers; the stated hypotheses imply that the total defining n is positive, but the underlying length expression truncates a negative total to zero before adding 3. There is no hypothesis that j−2 or j+2 is admissible, that either neighbor exists, or that any of the displayed summands or their denominator factorials are nonzero. In particular j=1 is included, with negative lower channel and hence m−=Uj−2,l=0; inadmissible upper or lower channels use the displayed zero branches. The assumption j>0 excludes j=0, but permits the external labels or l to be zero whenever the four triple conditions hold. No complex specialization, numerical parameter, typeclass assumption, or additional existentially chosen object is part of the statement.
Preamble
import Definitions.Def_QuantumSixJSumCertificate
open scoped BigOperators QuantumSixJ
open QuantumSixJ.Certificate
Formal statement
namespace QuantumSixJ
theorem sum_prefactor_match (a b c d j l : ℤ)
(hj : hAdmissible a b c d j) (hl : vAdmissible a b c d l) (hpos : 0 < j) :
mainResidual a b c d j l =
(prefactor a b c d j l / clearingFactor j) * totalResidual a b c d j l
:= by sorry
end QuantumSixJ
Source
I. Khavkine, Recurrence relation for the 6j-symbol of su_q(2) as a symmetric eigenvalue problem, https://arxiv.org/abs/1009.2261, Appendix finite Tet formula and Eq. (24); auxiliary factorial-shift and telescoping-certificate calculation for those formulas.