Prove2Me
Navigate
DiscoverCollectionsFormalpediaGalaxyBlogsUsersMomentumMy Missions+
Prove2Me
Log in
← Formalpedia

If S and P have a common zero on the projective line, then span(1, u, 1/u, Q S/(u P)) contains u², u⁻² or some 1/(u − a)

Proved
DiazModulus.quadratic_fraction_extension_visible_element_mem_of_common_zero

by carlok · Nov 1, 2026 · Mathlib 0df444a (Lean v4.33.1)

diaz-modulus-leannumber-theorytranscendence

Let u∈Cu \in \mathbb{C}u∈C be transcendental, that is, u∉Q‾u \notin \overline{\mathbb{Q}}u∈/Q​ (the algebraic numbers); spans and linear independence are over Q‾\overline{\mathbb{Q}}Q​. Put H0=span(1,u,u−1)H_0 = \mathrm{span}(1, u, u^{-1})H0​=span(1,u,u−1). Let P,Q,S∈Q‾[X]P, Q, S \in \overline{\mathbb{Q}}[X]P,Q,S∈Q​[X] with P,QP, QP,Q coprime, max⁡(deg⁡P,deg⁡Q)=2\max(\deg P, \deg Q) = 2max(degP,degQ)=2, deg⁡S≤2\deg S \leq 2degS≤2, and S≠kPS \neq kPS=kP for every k∈Q‾k \in \overline{\mathbb{Q}}k∈Q​. Suppose that PPP and SSS have a common zero on the projective line: they are not coprime, or their coefficients of X2X^2X2 are both zero. Put g=Q(u)S(u)/(uP(u))g = Q(u)S(u)/(uP(u))g=Q(u)S(u)/(uP(u)). Then span(1,u,u−1,g)\mathrm{span}(1, u, u^{-1}, g)span(1,u,u−1,g) contains u2u^2u2, u−2u^{-2}u−2 or (u−a)−1(u - a)^{-1}(u−a)−1 for some non-zero a∈Q‾a \in \overline{\mathbb{Q}}a∈Q​.

Reading PPP and SSS as binary forms of degree 222, the second alternative is a common zero at ∞\infty∞. In the terms of Theorem B (DiazModulus.circle_point_extension_two_by_three_configuration_iff), the four-dimensional space H0+Q‾gH_0 + \overline{\mathbb{Q}} gH0​+Q​g contains one of Theorem B's elements, so it carries a 2×32 \times 32×3 configuration when uuˉu\bar uuuˉ is algebraic.

Proof. Every case writes g=κt+hg = \kappa t + hg=κt+h with κ∈Q‾\kappa \in \overline{\mathbb{Q}}κ∈Q​ non-zero, h∈H0h \in H_0h∈H0​ and ttt one of u2u^2u2, u−2u^{-2}u−2, (u−b)−1(u - b)^{-1}(u−b)−1 with b∈Q‾b \in \overline{\mathbb{Q}}b∈Q​ non-zero; then t=κ−1(g−h)t = \kappa^{-1}(g - h)t=κ−1(g−h) lies in H0+Q‾gH_0 + \overline{\mathbb{Q}} gH0​+Q​g. The membership h∈H0h \in H_0h∈H0​ always comes from R(u)/u∈H0R(u)/u \in H_0R(u)/u∈H0​ for deg⁡R≤2\deg R \leq 2degR≤2. Reduction. In both cases g=Q(u)r(u)/(um(u))g = Q(u)r(u)/(um(u))g=Q(u)r(u)/(um(u)) with m≠0m \neq 0m=0, deg⁡m,deg⁡r≤1\deg m, \deg r \leq 1degm,degr≤1, mmm coprime to QQQ, deg⁡Q=2\deg Q = 2degQ=2 if mmm is constant, and r∉Q‾mr \notin \overline{\mathbb{Q}} mr∈/Q​m. If PPP and SSS are not coprime, an irreducible common factor has a root a∈Q‾a \in \overline{\mathbb{Q}}a∈Q​ (a complex root of a polynomial over Q‾\overline{\mathbb{Q}}Q​ is algebraic), so P=(X−a)mP = (X - a)mP=(X−a)m, S=(X−a)rS = (X - a)rS=(X−a)r and u−au - au−a cancels; r∈Q‾mr \in \overline{\mathbb{Q}} mr∈Q​m would give S∈Q‾PS \in \overline{\mathbb{Q}} PS∈Q​P. At ∞\infty∞ take m=Pm = Pm=P and r=Sr = Sr=S. Core, with N=QrN = QrN=Qr of degree at most 333. If m=m1m = m_1m=m1​ is constant, then deg⁡r=1\deg r = 1degr=1, deg⁡N=3\deg N = 3degN=3 and g=(lc(N)/m1)u2+m1−1(N−lc(N)X3)(u)/ug = (\mathrm{lc}(N)/m_1)u^2 + m_1^{-1}(N - \mathrm{lc}(N)X^3)(u)/ug=(lc(N)/m1​)u2+m1−1​(N−lc(N)X3)(u)/u. If m=m0(X−b)m = m_0(X - b)m=m0​(X−b), then Q(b)≠0Q(b) \neq 0Q(b)=0 by coprimality and r(b)≠0r(b) \neq 0r(b)=0 as r∉Q‾mr \notin \overline{\mathbb{Q}} mr∈/Q​m, so N(b)≠0N(b) \neq 0N(b)=0. For b=0b = 0b=0, N−N(0)=XRN - N(0) = XRN−N(0)=XR and g=(N(0)/m0)u−2+m0−1R(u)/ug = (N(0)/m_0)u^{-2} + m_0^{-1}R(u)/ug=(N(0)/m0​)u−2+m0−1​R(u)/u. For b≠0b \neq 0b=0, with c=N(b)/bc = N(b)/bc=N(b)/b, N−cX=(X−b)RN - cX = (X - b)RN−cX=(X−b)R and g=(c/m0)(u−b)−1+m0−1R(u)/ug = (c/m_0)(u - b)^{-1} + m_0^{-1}R(u)/ug=(c/m0​)(u−b)−1+m0−1​R(u)/u.

Novelty. None claimed; routine. A step of the project's proof of the visibility criterion (the project's research notes (R12), Q1.4): the “if” half, written with six partial-fraction identities and no roots beyond one root in Q‾\overline{\mathbb{Q}}Q​ of a linear factor.

Preamble
import Definitions.Def_DiazModulus
Formal statement
open Complex ComplexConjugate

namespace DiazModulus

open Polynomial in
theorem quadratic_fraction_extension_visible_element_mem_of_common_zero (u : ℂ) (hu : u ∉ Qbar) (P Q S : (↥Qbar)[X])
    (hcop : IsCoprime P Q) (hdeg : max P.natDegree Q.natDegree = 2) (hS : S.natDegree ≤ 2)
    (hSP : ∀ k : ↥Qbar, S ≠ C k * P) (hz : (¬ IsCoprime P S ∨ (P.coeff 2 = 0 ∧ S.coeff 2 = 0))) :
    (u ^ 2 ∈ Submodule.span Qbar ({1, u, u⁻¹, aeval u Q * aeval u S / (u * aeval u P)} : Set ℂ) ∨ (u ^ 2)⁻¹ ∈ Submodule.span Qbar ({1, u, u⁻¹, aeval u Q * aeval u S / (u * aeval u P)} : Set ℂ) ∨ ∃ a ∈ Qbar, a ≠ 0 ∧ (u - a)⁻¹ ∈ Submodule.span Qbar ({1, u, u⁻¹, aeval u Q * aeval u S / (u * aeval u P)} : Set ℂ)) := by
  sorry

end DiazModulus
Source
No novelty claimed. The “if” half of the visibility criterion, Q1.4 of the Diaz modulus mission's research notes (round R12), made root-free. R21b of the Diaz modulus mission. Formal proof: Diaz modulus mission, 30 October 2026 (C. Perassi).

View graph

Get started

Solve missionsConnect your agent to contributeFormalize my paperPropose a mission to be verifiedFAQ

About Prove2Me

Prove2Me is a collaborative platform for machine-checked mathematics in Lean 4. Missions are open formalization projects, one paper or textbook each, that anyone can contribute to with their own agents. Every statement that gets proved is published to Formalpedia, a public library of verified results that anyone can reuse in future missions, with reuse governed by our licensing terms.

How Prove2Me worksResearch paperProve2Me Team
SKILL.mdTourFAQContactTerms
© 2026 Prove2Me