On the distribution of shapes of totally real multiquadratic number fields

Anuj Jakhar, Anwesh Ray · 2026

Read On the distribution of shapes of totally real multiquadratic number fields by Anuj Jakhar, Anwesh Ray (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2603-11443.txt.

Introduction

Jakhar and Ray study how the shapes of totally real multiquadratic number fields are spread out. A field's shape is the class of its integer lattice under rotations, reflections, and scaling. These shapes live in a shape space that carries a natural measure from Haar measure. The pair look at fields of degree a power of two where two stays unramified. They show the spread is ruled by that measure cut down to one torus orbit. The result settles a conjecture of Haidar.

Catalog record

CREATORAnuj Jakhar, Anwesh Ray
YEAR2026
COLLECTIONPAPERS
FORMATTXT
FILE SIZE82858
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2603.11443

Browse the full library · About the Tower of Babel

Full text

The opening of the work, from the library's hosted text:

On the distribution of shapes of totally real multiquadratic number fields Anuj Jakhar, Anwesh Ray Abstract . The shape of a number field K K of degree m m is defined as the equivalence class of the lattice of integers under linear operations generated by rotations, reflections, and positive scalar dilations. It may be viewed as a point in the space of shapes 𝒮 m − 1 = GL m − 1 ​ ( ℤ ) \ GL m − 1 ​ ( ℝ ) / GO m − 1 ⁡ ( ℝ ) . \mathcal{S}_{m-1}=\mathrm{GL}_{m-1}(\mathbb{Z})\backslash\mathrm{GL}_{m-1}(\mathbb{R})/\operatorname{GO}_{m-1}(\mathbb{R}). The double quotient space is equipped with a natural measure μ \mu which is induced from the Haar measure on GL m − 1 ​ ( ℝ ) \mathrm{GL}_{m-1}(\mathbb{R}) . We study the distribution of shapes of totally real multiquadratic number fields of degree m := 2 n m:=2^{n} in which 2 2 is unramified. We show that the distribution is governed by the restriction of μ \mu to a certain torus orbit in 𝒮 m − 1 \mathcal{S}_{m-1} . Our result resolves a conjecture of Haidar. 2020 Mathematics Subject Classification 11R29, 11R45, 11R56 1. Introduction 1.1. Motivation and historical context A classical approach to studying the arithmetic of a number field is to translate its algebraic properties into geometric ones. If we consider a number field K of degree n=[K:\mathbb{Q}] , a standard tool for this translation is the Minkowski embedding: J:K\hookrightarrow K_{\mathbb{R}}:=K\otimes_{\mathbb{Q}}\mathbb{R}\cong\mathbb{R}^{n}. This embedding realizes \mathcal{O}_{K} as a full-rank Euclidean lattice \Lambda_{K}=J(\mathcal{O}_{K}) inside K_{\mathbb{R}} . Much of classical algebraic number theory relies on a single invariant to understand this lattice: the absolute discriminant \Delta_{K} . Since the square root of the discriminant computes the covolume of \Lambda_{K} , it effectively measures the “size” of the ring. Yet, a single volume measurement completely ignores the angles between generators of the lattice. Two fields can have identical discriminants but different lattice structures, with different angles and varying lengths of basis vectors. To capture these exact geometric proportions, we turn to a much sharper invariant known as the shape of the number field. To define the shape rigorously, we view K_{\mathbb{R}} as a Euclidean space equipped with the inner product induced by the algebraic trace form. Since \Lambda_{K} always contains the distinguished vector J(1)=(1,\dots,1) , this introduces a rigid linear constraint common to all number fields. It is therefore standard to project the lattice orthogonally onto the trace-zero hyperplane. The shape of K is defined as the equivalence class of this projected lattice of rank n-1 , considered modulo the group \operatorname{GO}_{n-1}(\mathbb{R}) of orthogonal similitudes (the group generated by rotations, reflections, and uniform scalar dilations). The space of all such shapes is identified with the double coset space \mathcal{S}_{n-1}:=\operatorname{GL}_{n-1}(\mathbb{Z})\backslash\operatorname{GL}_{n-1}(\mathbb{R})/\operatorname{GO}_{n-1}(\mathbb{R}), which comes equipped with a natural measure \mu derived from the Haar measure on \operatorname{GL}_{n-1}(\mathbb{R}) . A central pursuit in arithmetic statistics is understanding how these shapes are distributed as K varies over a family of number fields of a fixed degree, ordered by absolute discriminant. For generic families, namely, fields whose Galois closure has Galois group isomorphic to S_{n} , it is expected that the shapes equidistribute across the entire space \mathcal{S}_{n-1} with respect to the measure \mu . This principle of generic equidistribution has been beautifully verified in low degrees. Terr [ 17 ] established that the shapes of cubic fields are equidistributed in the modular surface \mathcal{S}_{2} . Subsequently, Bhargava and Harron [ 1 ] extended this phenomenon to S_{4} -quartic and S_{5} -quintic fields, utilizing parameterizations of low-degree rings (cf. [ 2 , 3 ] ). These methods do not apply to non-generic families of number fields, i.e., when the Galois group of the Galois closure is a proper subgroup of S_{n} . In such families, underlying arithmetic constraints force the shapes to collapse into lower-dimensional subvarieties of \mathcal{S}_{n-1} . The first major breakthrough in this direction was Harron’s study of pure cubic fields [ 10 ] . He proved that their shapes do not fill the space \mathcal{S}_{2} , but instead restrict strictly to two specific one-dimensional geodesics, governed entirely by a tame-wild ramification dichotomy at the prime 3 . Holmes [ 12 ] later generalized this phenomenon to pure prime-degree fields \mathbb{Q}(\sqrt[p]{m}) , proving that their shapes lie on affine subspaces of dimension (p-1)/2 . Similar results have been proven for cyclic prime-degree fields [ 16 ] , Galois quartic fields [ 7 ] , and complex cubic fields with a fixed quadratic resolvent [ 11 ] . Most recently, investigations into non-generic fields of composite degree have revealed even richer geometric structures. Studies on pure quartic fields [ 5 ] , pure sextic fields [ 13 ] , and octic Kummer extensions [ 14 ] have demonstrated that the space of shapes for these families decomposes into finite unions of translated torus orbits. In this paper we study the non-generic family of totally real multiquadratic number fields of degree 2^{n} . Such a field can be written K_{n}=\mathbb{Q}(\sqrt{D_{2^{0}}},\dots,\sqrt{D_{2^{n-1}}}), where the D_{2^{k}} are distinct squarefree positive integers satisfying natural non-degeneracy conditions that ensure [K_{n}:\mathbb{Q}]=2^{n} . K_{n} . The Galois group of K_{n}/\mathbb{Q} is the elementary abelian 2 –group \mathbb{F}_{2}^{n} , and the field contains exactly \ell=2^{n}-1 quadratic subfields. The distribution of shapes in multiquadratic families has only recently begun to be understood. In the biquadratic case this problem was first studied by Harron and Harron [ 8 ] , who described the space of possible shapes of biquadratic fields. Subsequently, Haidar [ 9 ] investigated the triquadratic case and formulated a precise equidistribution conjecture for totally real multiquadratic fields in which the prime 2 is unramified. In particular, he conjectured that the shapes of such fields, when ordered by discriminant, become equidistributed inside the subspace of the full shape space that they occupy. While we explicitly determine the shapes for all totally real multiquadratic fields regardless of their ramification at 2 , the main density result of this paper focuses on the unramified case, proving Haidar’s Conjecture 1.1 for all n . Our approach combines an arithmetic parametrization of multiquadratic fields with analytic counting methods. A key input is the classification of multiquadratic integral bases due to Chatelain ( [ 4 ] , [ 15 , Theorem 2.1] ). 1.2. Main Results Let n\geq 2 and set \ell=2^{n}-1 . We consider totally real multiquadratic extensions K_{n}=\mathbb{Q}(\sqrt{a_{1}},\dots,\sqrt{a_{n}}) of degree 2^{n} . Associated to such a field are \ell quadratic subfields whose radicands may be arranged into a tuple (D_{1},\dots,D_{\ell}) obtained by taking squarefree parts of products of the generators a_{i} . As in [ 9 ] , the shapes of such fields can be described by \ell-1 real parameters \lambda_{2},\dots,\lambda_{\ell} which determine the similarity class of the projected Minkowski lattice. We denote by \mathcal{S}_{K_{n}}\subset\mathcal{S}_{\ell} the subset of the space of shapes consisting of those arising from totally real multiquadratic extensions. Let R_{2},\dots,R_{\ell} be fixed positive real numbers satisfying 1\leq R_{2}\leq R_{3}\leq\dots\leq R_{\ell}. Following [ 9 ] , we define the region W(R_{2},\dots,R_{\ell})=\left\{\operatorname{sh}(\lambda_{2},\dots,\lambda_{\ell})\in\mathcal{S}_{K_{n}}:\lambda_{2}<\dots<\lambda_{\ell}\leq R_{\ell},\;R_{j}\leq\lambda_{j},\quad 2\leq j\leq\ell-1\right\}. For X>0 we write \mathcal{F}(X,R_{2},\dots,R_{\ell})=\#\left\{K_{n}:\Delta_{K_{n}}\leq X,\;\operatorname{sh}(K_{n})\in W(R_{2},\dots,R_{\ell}),\;2\text{ unramified in }K_{n}\right\}. Our first result gives an asymptotic formula for this counting function. The main term involves an explicit constant together with an iterated integral F(R_{2},\dots,R_{\ell}):=\int_{R_{\ell-1}}^{R_{\ell}}\!\!\cdots\int_{R_{2}}^{a_{3}}\frac{da_{2}}{a_{2}}\cdots\frac{da_{\ell}}{a_{\ell}}. This is

Read the complete text