Prove2Me
Navigate
DiscoverCollectionsFormalpediaGalaxyBlogsUsersMomentumMy Missions+
Prove2Me
Log in
← Formalpedia

Functional equation of the zeta function of a smooth projective geometrically irreducible variety over a finite field: Z(X,1/(qnT))=c (qnT)deg⁡B−deg⁡A Z(X,T)Z(X,1/(q^nT))=c\,(q^nT)^{\deg B-\deg A}\,Z(X,T)Z(X,1/(qnT))=c(qnT)degB−degAZ(X,T) for Z=A/BZ=A/BZ=A/B

Open
FiniteFields.zeta_functional_equation

by Idealson · Oct 20, 2026 · Mathlib 0df444a (Lean v4.33.1)

finite-fieldsfunctional-equationweil-conjectureszeta-functions

Let k=Fqk=\mathbb{F}_qk=Fq​ be a finite field of characteristic ppp, and let XXX be a smooth projective geometrically irreducible scheme of dimension nnn over kkk: the structure morphism fff is smooth of relative dimension nnn and geometrically irreducible, and XXX admits a closed immersion into some PkN\mathbb{P}^N_kPkN​ over kkk. For m≥1m\ge1m≥1 let NmN_mNm​ be the number of Fqm\mathbb{F}_{q^m}Fqm​-points of XXX (morphisms Spec⁡Fqm→X\operatorname{Spec}\mathbb{F}_{q^m}\to XSpecFqm​→X over Spec⁡k\operatorname{Spec}kSpeck), and let

Z(X,T)=exp⁡(∑m≥1Nmm Tm)∈Q[[T]].Z(X,T)=\exp\Big(\sum_{m\ge1}\frac{N_m}{m}\,T^m\Big)\in\mathbb{Q}[[T]].Z(X,T)=exp(m≥1∑​mNm​​Tm)∈Q[[T]].

For a polynomial AAA of degree ddd write Arev(T)=TdA(1/T)A^{\mathrm{rev}}(T)=T^{d}A(1/T)Arev(T)=TdA(1/T) for its reciprocal polynomial.

Theorem (functional equation). There exist A,B∈Z[T]A,B\in\mathbb{Z}[T]A,B∈Z[T] and a nonzero rational number ccc such that

  1. A(0)=1A(0)=1A(0)=1 and B(0)=1B(0)=1B(0)=1;
Z(X,T)⋅B(T)=A(T)in Q[[T]];Z(X,T)\cdot B(T)=A(T)\qquad\text{in }\mathbb{Q}[[T]];Z(X,T)⋅B(T)=A(T)in Q[[T]];
Arev(qnT)⋅B(T)=c⋅A(T)⋅Brev(qnT)in Q[T].A^{\mathrm{rev}}(q^nT)\cdot B(T)=c\cdot A(T)\cdot B^{\mathrm{rev}}(q^nT)\qquad\text{in }\mathbb{Q}[T].Arev(qnT)⋅B(T)=c⋅A(T)⋅Brev(qnT)in Q[T].

Since Arev(qnT)=(qnT)deg⁡AA(1/(qnT))A^{\mathrm{rev}}(q^nT)=(q^nT)^{\deg A}A(1/(q^nT))Arev(qnT)=(qnT)degAA(1/(qnT)), condition 3 says that, as rational functions,

Z(X,1qnT)=c (qnT)deg⁡B−deg⁡A Z(X,T).Z\Big(X,\frac{1}{q^nT}\Big)=c\,(q^nT)^{\deg B-\deg A}\,Z(X,T).Z(X,qnT1​)=c(qnT)degB−degAZ(X,T).

Equivalently, the divisor of zeros and poles of Z(X,T)Z(X,T)Z(X,T) is invariant under a↦qn/aa\mapsto q^n/aa↦qn/a on inverse roots. It follows from Grothendieck's functional equation Z(X,1/(qnT))=±qnE/2TEZ(X,T)Z(X,1/(q^nT))=\pm q^{nE/2}T^{E}Z(X,T)Z(X,1/(qnT))=±qnE/2TEZ(X,T), where EEE is the ℓ\ellℓ-adic Euler characteristic of XkˉX_{\bar k}Xkˉ​, which is a consequence of Poincaré duality and the Lefschetz trace formula in ℓ\ellℓ-adic cohomology.

Formalization Note. The hypotheses are exactly those of WeilConjectures.rationality_functional_equation: SmoothOfRelativeDimension n f, GeometricallyIrreducible f and the platform's IsProjectiveOver k f; Z(X,T)Z(X,T)Z(X,T) is the platform's zetaFunction p f and qqq is Nat.card k. The statement is root-free: it uses only Z[T]\mathbb{Z}[T]Z[T], Q[T]\mathbb{Q}[T]Q[T] and Q[[T]]\mathbb{Q}[[T]]Q[[T]]. Here ArevA^{\mathrm{rev}}Arev is Mathlib's Polynomial.reverse (equal to Tdeg⁡AA(1/T)T^{\deg A}A(1/T)TdegAA(1/T) because A(0)≠0A(0)\neq0A(0)=0), and Arev(qnT)A^{\mathrm{rev}}(q^nT)Arev(qnT) is A.reverse.comp (C (q ^ n) * X). Polynomials over Z\mathbb{Z}Z are mapped to Q[T]\mathbb{Q}[T]Q[T] by Int.castRingHom ℚ and viewed in Q[[T]]\mathbb{Q}[[T]]Q[[T]] by the canonical coercion. The presentation A/BA/BA/B is not required to be reduced, and the decomposition of ZZZ into the factors PiP_iPi​ of the Weil conjectures is not part of the statement. Together with the integral presentation FiniteFields.zetaFunction_integral_presentation, this statement implies the conclusion of WeilConjectures.rationality_functional_equation for n≥2n\ge2n≥2.

Preamble
import Mathlib
import Definitions.Def_WeilConjectures_projectiveSpace
import Definitions.Def_WeilConjectures_zetaFunction

open AlgebraicGeometry CategoryTheory Polynomial
open WeilConjectures
Formal statement
theorem FiniteFields.zeta_functional_equation (p : ℕ) [Fact p.Prime] (k : Type) [Field k]
    [Finite k] [CharP k p] (n : ℕ) (X : Scheme.{0}) (f : X ⟶ Spec (.of k))
    [SmoothOfRelativeDimension n f] [GeometricallyIrreducible f]
    (hproj : IsProjectiveOver k f) :
    ∃ A B : ℤ[X], A.coeff 0 = 1 ∧ B.coeff 0 = 1 ∧
      zetaFunction p f * ((B.map (Int.castRingHom ℚ) : ℚ[X]) : PowerSeries ℚ) =
        ((A.map (Int.castRingHom ℚ) : ℚ[X]) : PowerSeries ℚ) ∧
      ∃ c : ℚ, c ≠ 0 ∧
        ((A.reverse.comp (C ((Nat.card k : ℤ) ^ n) * Polynomial.X)) * B).map
            (Int.castRingHom ℚ) =
          C c * (A * (B.reverse.comp (C ((Nat.card k : ℤ) ^ n) * Polynomial.X))).map
            (Int.castRingHom ℚ) := by sorry
Source
A. Grothendieck, *Formule de Lefschetz et rationalité des fonctions L*, Séminaire Bourbaki 17 (1964/65), exposé 279; SGA 5, *Cohomologie ℓ-adique et fonctions L*, Lecture Notes in Math. 589 (1977); P. Deligne, *Cohomologie étale* (SGA 4½), Lecture Notes in Math. 569 (1977), « Rapport sur la formule des traces »; R. Hartshorne, *Algebraic Geometry*, Appendix C (the Weil conjectures: Poincaré duality implies the functional equation). Root-free form and its equivalence with the symmetry of the divisor of ZZZ: `Bridge/FunctionalEquationSplit.lean` of the WeilConjecture project (`zetaDivisorSymm_of_funEq`).

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