Consequences of the Gauss-Jacobi identities and cubic reciprocity
ProvedQRH.gauss_jacobi_consequencesLet a, b, n be squarefree primary elements prime to S, with a and b coprime in the sense that their only common divisor is the unit. Then the interface axioms imply the four conversion identities of Lemma 4.1 of the source: first, conj(alpha n) gamma_2(n) gamma_1(n) = mu(n) G(cl n); second, mu(n) gamma_{-1}(n) = chi_n(-1) G(cl n)^{-1} conj(alpha n) gamma_2(n); third, the quotient identity chi_a(-1) conj(G(cl a)) G(cl b) R(cl a, cl b) = G(cl b inv(cl a)) in the ray class group; and fourth, the multiplicativity of the cubic Gauss-sum coefficients up to a sextic twist: a_xi(ab) = a_xi(a) a_xi(b) chi_b(a)^4.
import Mathlib import Definitions.Def_QRH_arith import Definitions.Def_QRH_sums import Definitions.Def_QRH_lfun
universe u
namespace QRH
/-- The consequences of the Gauss-Jacobi identities and cubic
reciprocity (Lemma 4.1 of the source), for squarefree primary elements
prime to S. The four conjuncts are the conversion identities between
Möbius coefficients and cubic Gauss sums (equations 4.7 and 4.8): the
Möbius factor combines with the Poisson γ₋₁ into the cubic coefficient,
and back; the quotient identity for the ray class factor G (equation
4.9); and the multiplicativity of the cubic Gauss-sum coefficients a_ξ
up to a sextic twist (equation 4.10). Each follows from the interface
axioms by complex algebra: the Gauss sums and the phase have modulus one,
Möbius squares to one on squarefree elements, and the reciprocity
bicharacter is real. -/
theorem gauss_jacobi_consequences (W : QRHWorld.{u}) (a b n : W.E)
(hsn : W.IsSqf n) (hgn : W.Sgood n)
(hsa : W.IsSqf a) (hga : W.Sgood a)
(hsb : W.IsSqf b) (hgb : W.Sgood b)
(hcop : ∀ d : W.E, d ∈ W.divs a → d ∈ W.divs b → d = W.one) :
star (W.alpha n) * W.gamma 2 n * W.gamma 1 n
= W.mu n * W.Gcl (W.cl n)
∧ W.mu n * W.gamma (-1) n
= W.chiNeg n * (W.Gcl (W.cl n))⁻¹ * star (W.alpha n)
* W.gamma 2 n
∧ W.chiNeg a * star (W.Gcl (W.cl a)) * W.Gcl (W.cl b)
* W.Rcl (W.cl a) (W.cl b)
= W.Gcl (W.mulC (W.cl b) (W.invC (W.cl a)))
∧ axi W (W.mul a b) = axi W a * axi W b * W.chi b a ^ 4 := by
sorry
end QRH
Read-back
What the Lean code literally says, in plain math · glm-5-latest-short
Read-back (self-check authored by the same model that produced the statement): the formal theorem asserts, for squarefree primary a, b, n prime to S with a and b sharing no divisor but the unit, the conjunction of four complex equalities in the world W: (1) star(alpha n) * gamma 2 n * gamma 1 n = mu n * G(cl n); (2) mu n * gamma(-1) n = chiNeg n * (G(cl n))^{-1} * star(alpha n) * gamma 2 n; (3) chiNeg a * star(G(cl a)) * G(cl b) * Rcl(cl a)(cl b) = G(mulC (cl b) (invC (cl a))); (4) axi W (mul a b) = axi W a * axi W b * chi b a^4, where axi W x = star(W.alpha x) * W.gamma 2 x * W.xi x. Each conjunct is derivable from the modulus-one fields (alpha_abs, gamma_abs, Gcl_abs, chiNeg_abs, Rcl_abs), the Gauss-Jacobi fields (gamma_1_2, gamma_1_m1, gamma_2_cube), Mobius squaring (mu_sqf), the cocycle identity (Gcl_mul), the inverse identity (Gcl_invC), and the multiplicativity fields (alpha_mul, xi_mul, gamma_2_mul, chi_mul_den, chi_mul_num, recip_cubic).