[Paper Review] A note on o-minimal flows and the Ax--Lindemann--Weierstrass theorem for abelian varieties over $\mathbb{C}$

Ya’acov Peterzil, Sergei Starchenko|arXiv (Cornell University)|Nov 15, 2016

Mathematical Dynamics and Fractals3 citations

TL;DR

This paper presents an elementary proof of the Ax-Lindemann-Weierstrass theorem for semi-abelian varieties over ℂ using o-minimal structures, avoiding the Pila-Wilkie counting theorem. The key contribution is a simplified, model-theoretic approach that establishes that the Zariski closure of the image of an algebraic subvariety under the exponential map is a translate of an algebraic subgroup, leveraging definable sets and linear algebra in o-minimal expansions of the real field.

ABSTRACT

In this short note we present an elementary proof of the Ax-Lindemann-Weierstrass theorem for abelian and semi-abelian varieties. The proof uses ideas of Pila, Ullmo, Yafaev, Zannier and is based on basic properties of sets definable in o-minimal structures. It does not use the Pila-Wilkie counting theorem.

Motivation & Objective

• To provide a new, elementary proof of the Ax-Lindemann-Weierstrass theorem for semi-abelian varieties over ℂ.
• To demonstrate that the Pila-Wilkie counting theorem is unnecessary for proving ALW in the semi-abelian setting.
• To clarify the geometric structure of preimages under the exponential map using o-minimal definability and linear algebra.
• To generalize the proof technique to include abelian and semi-abelian varieties via definable large domains and Lie algebra decompositions.

Proposed method

• Utilizes o-minimal structures, particularly the real analytic structure ℝ_an, to ensure definability of preimages of algebraic varieties under the exponential map.
• Applies the fact that definable discrete subsets in o-minimal structures are finite, to control the behavior of lattice points and closures.
• Employs the concept of a definable large domain F for the uniformizing map π: V → G, ensuring π|F is definable and F + Λ = V.
• Uses linear algebra in the Lie algebra T_G to show that the preimage of the Zariski closure of exp_G(X) lies in a finite union of translates of a complex linear subspace.
• Applies Fact 4.4: if a complex analytic set Y' is contained in a real linear subspace U, then Y' ⊆ iU, to deduce that the real and imaginary parts of the Lie algebra must align.
• Establishes that the Zariski closure Z of exp_G(X) is a product Z_V × (p·B), where B is an algebraic subgroup and Z_V is an algebraic subvariety of the domain V.

Experimental results

Research questions

• RQ1 Can the Ax-Lindemann-Weierstrass theorem for semi-abelian varieties be proven without invoking the Pila-Wilkie counting theorem?
• RQ2 What role do o-minimal structures play in simplifying the proof of ALW in the semi-abelian case?
• RQ3 How does the definability of the preimage of an algebraic variety under the exponential map constrain the structure of its Zariski closure?
• RQ4 Under what conditions does the preimage of a Zariski closed set in a semi-abelian variety become a finite union of translates of linear subspaces in the Lie algebra?

Key findings

• The Zariski closure Z of exp_G(X) for an irreducible algebraic subvariety X ⊆ T_G is a translate of an algebraic subgroup of G, proving the geometric form of the Ax-Lindemann-Weierstrass theorem.
• The proof avoids the Pila-Wilkie counting theorem by relying solely on o-minimal definability and linear algebraic properties of Lie algebras.
• For any definable connected real analytic submanifold X ⊆ T_G, the preimage of Z = Zcl(exp_G(X)) in T_G is contained in a finite union of translates of a real linear subspace M, which is shown to be a complex linear subspace.
• The intersection (V + T_B + iℝ^k) ∩ (V + T_B + ℝ^k) equals V + T_B, which implies that the real and imaginary parts of the Lie algebra must align with the subgroup structure.
• The final structure of Z is shown to be Z = Z_V × (p·B), where Z_V is an algebraic subvariety of V and p·B is a translate of an algebraic subgroup B of G.