Abstract
Classical results on the classification of reflections in an arithmetic subgroup $\Gamma$ imply that if the graded algebra of modular forms $M_*(\Gamma)$ is freely generated, then $\Gamma$ must be an arithmetic subgroup of either the orthogonal group $\operatorname{O}^+(2,n)$ or the unitary group $\operatorname{U}(1,n)$. Vinberg and Schwarzman showed that in the orthogonal case, if $n > 10$, then it is never free. In this paper, we investigate the remaining unitary case and prove that, up to scaling, there are only finitely many isometry classes of Hermitian lattices of signature $(1, n)$ with $n > 2$ over imaginary quadratic fields with odd discriminant that admit a free algebra of modular forms. In particular, when $n > 99$ (except over $\mathbb{Q}(\sqrt{-3})$, where we require $n > 154$), the graded algebra $M_*(\Gamma)$ is never free for any arithmetic subgroup $\Gamma < \operatorname{U}(1,n)$, thereby partially confirming a conjecture by Wang and Williams. As a byproduct, we also establish a finiteness result for reflective modular forms.
Introduction: The Geometry of Arithmetic Varieties
In the intersection of number theory and algebraic geometry, researchers seek to classify arithmetic varieties—complex manifolds formed by the quotient of a symmetric domain by a discrete symmetry group (an arithmetic subgroup $\Gamma$). A primary goal is identifying varieties with "simple" structures, specifically those that resemble weighted projective spaces. This geometric simplicity is intrinsically linked to the algebraic structure of the associated graded algebra of modular forms $M_*(\Gamma)$; the variety is simple precisely when this algebra is "freely generated," behaving like a simple polynomial ring.
Previous research by Vinberg and Schwarzman established that for orthogonal groups, this simplicity disappears for dimensions $n > 10$. This paper addresses the remaining major case: unitary groups $\operatorname{U}(1, n)$. The authors investigate a conjecture by Wang and Williams, which posits that such free algebras are rare and should eventually disappear as the dimension of the complex ball increases.
Methodology and Technical Framework
The authors employ a "local-to-global" strategy that bridges geometric group theory and modular form theory. The investigation is structured around several key technical milestones, as outlined in the paper's organization.
The core technical engine is a generalization of Prasad’s volume formula. While the original formula calculates the "covolume" (a measure of the quotient space's size) for principal arithmetic subgroups, the authors extend this to any subgroup presented as the stabilizer of a Hermitian lattice. This generalized formula allows for the estimation of the Hirzebruch-Mumford volume, which relates the discrete geometry of the arithmetic group to the continuous geometry of the complex ball.
To determine the limits of freeness, the authors utilize a specific criterion (Theorem 4.2.3) linking algebraic structure to a volume inequality. If the volume of the variety is "too large" relative to its ramification (the way the space folds over itself), the algebra cannot be free. This theoretical framework is supported by detailed p-adic analysis at finite places and computer-based calculations to verify numerical bounds for non-freeness thresholds.
Key Insights: The Volume-Algebra Link
The most significant insight of this work is the quantitative relationship between the "size" of a variety and its algebraic complexity. As the dimension $n$ increases, the "space" available for modular forms to serve as simple polynomial generators effectively vanishes.
Furthermore, the paper establishes that "freeness" is a globally finite phenomenon. There are only a limited number of Hermitian lattices—the underlying grids for these groups—that can ever produce these simple algebras. As a byproduct, the authors also establish a finiteness result for reflective modular forms, which are forms whose zeros lie on the mirrors of the group's reflections. This provides a unitary analogue to famous results in the orthogonal case.
Results and Future Outlook
The study yields several definitive results:
- The $n > 99$ Threshold: For arithmetic subgroups associated with imaginary quadratic fields of odd discriminant, the algebra of modular forms is never free if $n > 99$. For the specific case of Eisenstein integers ($\mathbb{Q}(\sqrt{-3})$), the limit is $n > 154$.
- Global Finiteness: Up to scaling, there are only finitely many isometry classes of Hermitian lattices that admit a free algebra of modular forms.
- Covolume Formula: The paper provides an explicit, computable formula for the covolume of $SU(L)$, which will likely serve as a standard tool for future researchers calculating birational invariants of ball quotients.
Despite these breakthroughs, the authors note that the current bounds may not be sharp. While non-freeness is proven for $n > 99$, conjectures suggest the limit may be as low as $n > 5$. Additionally, while the logic extends to even discriminants (such as the Gaussian integers), a full explicit classification for those cases remains a subject for future work.
Summary updated