---
name: unit-distance-cm-fields
type: reference
title: "CM Fields, Class Towers & the Algebraic Toolkit"
description: "The algebraic number theory foundation: CM fields, imaginary quadratic fields, complex multiplication, class groups, genus theory, unramified extensions, and Lemma 2.2."
tags: [cm-fields, imaginary-quadratic, complex-multiplication, class-groups, genus-theory, unramified-extensions, pro-p-towers, lemma-2-2]
timestamp: 2026-07-20
---

# CM Fields, Class Towers & the Algebraic Toolkit

## Overview

The unit distance proof bridges combinatorial geometry and algebraic number theory. This page documents the algebraic toolkit used in the construction: CM fields, class groups, genus theory, unramified extensions, and the critical Lemma 2.2 that generates unit-norm elements.

## 1. Number Fields and Extensions

### 1.1 Basic Definitions

A **number field** $K$ is a finite extension of $\mathbb{Q}$. Key invariants:

- **Degree:** $[K:\mathbb{Q}]$
- **Discriminant:** $\Delta_K$ (measures ramification)
- **Ring of integers:** $\mathcal{O}_K$
- **Unit group:** $\mathcal{O}_K^\times$ (Dirichlet's unit theorem)

### 1.2 Places and Embeddings

A number field $K$ of degree $n$ has $n$ embeddings into $\mathbb{C}$:
- $r_1$ real embeddings: $\sigma_i: K \hookrightarrow \mathbb{R}$
- $r_2$ pairs of complex conjugate embeddings: $\sigma_j, \bar{\sigma}_j: K \hookrightarrow \mathbb{C}$

with $r_1 + 2r_2 = n$.

### 1.3 Ramification

A prime $p$ **ramifies** in $K$ if $p$ divides the discriminant $\Delta_K$. In the ring of integers:

$$p\mathcal{O}_K = \mathfrak{p}_1^{e_1} \cdots \mathfrak{p}_g^{e_g}$$

where $e_i > 1$ for at least one $i$. The prime is **unramified** if all $e_i = 1$.

## 2. CM Fields

### 2.1 Definition

A **CM field** (Complex Multiplication field) is a totally imaginary quadratic extension of a totally real number field:

$$F = F^+(\sqrt{-\alpha})$$

where:
- $F^+$ is a totally real field (all embeddings land in $\mathbb{R}$)
- $\alpha \in F^+$ is totally positive
- $F$ is totally imaginary (no real embeddings)

### 2.2 CM Involution

The **CM involution** is complex conjugation $\sigma: F \to F$, which:
- Fixes the totally real subfield: $\sigma|_{F^+} = \text{id}$
- Acts non-trivially on $F$: $\sigma(\sqrt{-\alpha}) = -\sqrt{-\alpha}$
- Is an automorphism of order 2: $\sigma^2 = \text{id}$

### 2.3 Embedding Structure

A CM field $F$ of degree $2n$ has:
- $r_1 = 0$ real embeddings
- $r_2 = n$ pairs of complex conjugate embeddings

This means all embeddings land in $\mathbb{C} \setminus \mathbb{R}$, which is crucial for unit norm generation.

### 2.4 Examples

| Field | Degree | Totally Real Subfield | CM Involution |
|---|---|---|---|
| $\mathbb{Q}(i)$ | 2 | $\mathbb{Q}$ | Complex conjugation |
| $\mathbb{Q}(\sqrt{-2})$ | 2 | $\mathbb{Q}$ | Complex conjugation |
| $\mathbb{Q}(\sqrt{-2}, \sqrt{3})$ | 4 | $\mathbb{Q}(\sqrt{3})$ | Complex conjugation |
| $\mathbb{Q}(\sqrt{-2}, \sqrt{3}, \sqrt{5})$ | 8 | $\mathbb{Q}(\sqrt{3}, \sqrt{5})$ | Complex conjugation |
| $\mathbb{Q}(\sqrt{-2}, \sqrt{3}, \sqrt{5}, \sqrt{7})$ | 16 | $\mathbb{Q}(\sqrt{3}, \sqrt{5}, \sqrt{7})$ | Complex conjugation |

The last field is the H16 base field — see [[unit-distance-h16-breakthrough]].

## 3. Imaginary Quadratic Fields

### 3.1 Definition

An **imaginary quadratic field** is $F = \mathbb{Q}(\sqrt{-D})$ where $D > 0$ is a squarefree positive integer.

- Degree: $[F:\mathbb{Q}] = 2$
- Discriminant: $\Delta_F = -D$ (if $D \equiv 3 \pmod{4}$) or $-4D$ (otherwise)
- Real embeddings: $r_1 = 0$
- Complex embeddings: $r_2 = 1$

### 3.2 Class Groups

The **class group** $\text{Cl}(F)$ measures the failure of unique factorization in $\mathcal{O}_F$. Its order $h(F)$ is the **class number**.

For imaginary quadratic fields:
- $h(F) = 1$ for $D = 1, 2, 3, 7, 11, 19, 43, 67, 163$ (Heegner-Baker-Stark)
- $h(F)$ grows roughly as $\sqrt{D}$ (Gauss class number formula)

### 3.3 2-Class Rank

The **2-class rank** $d$ is the rank of the 2-part of the class group:

$$d = \text{rk}_2(\text{Cl}(F))$$

By **genus theory** for imaginary quadratic fields:

$$d = \ell - 1$$

where $\ell$ is the number of distinct prime factors of $D$. This linear growth is the key limitation that H16 overcomes.

## 4. Multi-Quadratic Fields

### 4.1 Construction

A **multi-quadratic field** is the compositum of $N$ quadratic extensions:

$$F = \mathbb{Q}(\sqrt{d_1}, \sqrt{d_2}, \ldots, \sqrt{d_N})$$

where $d_i$ are squarefree integers.

### 4.2 Degree and Galois Group

$$[F:\mathbb{Q}] = 2^N$$

$$\text{Gal}(F/\mathbb{Q}) \cong (\mathbb{Z}/2\mathbb{Z})^N$$

Each automorphism independently chooses the sign of each square root.

### 4.3 2-Class Rank

By genus theory for multi-quadratic fields:

$$d = 2^N - 1$$

This is the **exponential growth** that makes multi-quadratic fields superior for the unit distance construction. Compare:

| $N$ | Degree | $d$ | Growth |
|---|---|---|---|
| 1 | 2 | 1 | Linear |
| 2 | 4 | 3 | Linear |
| 3 | 8 | 7 | Linear |
| 4 | 16 | 15 | Linear |
| 5 | 32 | 31 | Linear |

Wait — the growth is $2^N - 1$, which is exponential in $N$. But for fixed $N$, it's linear in the number of quadratic extensions. The key is that $N$ is fixed (e.g., $N = 4$ for H16), and the 2-class rank is determined by $N$, not by the number of ramified primes.

### 4.4 Discriminant

For a multi-quadratic field ramified at primes $p_1, \ldots, p_\ell$:

$$\log H = \frac{1}{2} \sum_{i=1}^{\ell} \log p_i$$

This grows **linearly** with $\ell$, while $d = 2^N - 1$ is **independent** of $\ell$ (as long as $\ell \ge N$). This is the exponential-vs-linear divergence that powers H16.

## 5. Complex Multiplication

### 5.1 Definition

**Complex multiplication** (CM) refers to endomorphisms of abelian varieties that go beyond the standard multiplication-by-$n$ map. For elliptic curves, CM means the endomorphism ring is larger than $\mathbb{Z}$.

### 5.2 CM Fields and Class Groups

The CM theory connects:
- The class group of a CM field $F$ to the class group of its totally real subfield $F^+$
- The **relative class number** $h^-(F) = h(F) / h(F^+)$
- The **CM type** — a choice of $n$ embeddings out of $2n$ total

### 5.3 Role in Unit Distance

CM provides:
1. **Conjugate prime pairs** — complex conjugation pairs prime ideals $(P, \bar{P})$
2. **Unit norm generation** — Lemma 2.2 uses these pairs to generate elements with $|u| = 1$
3. **Global involution** — the CM involution is defined globally on $F$, not just locally

## 6. Class Groups and Genus Theory

### 6.1 Class Groups

The **class group** $\text{Cl}(K)$ of a number field $K$ is the group of fractional ideals modulo principal ideals:

$$\text{Cl}(K) = I_K / P_K$$

where $I_K$ is the group of fractional ideals and $P_K$ is the group of principal fractional ideals.

### 6.2 Genus Theory

**Genus theory** computes the 2-class rank of quadratic fields using quadratic reciprocity.

For an imaginary quadratic field $F = \mathbb{Q}(\sqrt{-D})$ with $D = p_1 \cdots p_\ell$:

$$d = \text{rk}_2(\text{Cl}(F)) = \ell - 1$$

This is because:
- Each prime $p_i$ ramifies in $F$
- The genus characters are $\left(\frac{p_i}{\cdot}\right)$ for $i = 1, \ldots, \ell$
- One relation exists: the product of all genus characters is trivial
- Thus $d = \ell - 1$

### 6.3 Application to Multi-Quadratic Fields

For multi-quadratic fields, genus theory generalizes:

$$d = 2^N - 1$$

where $N$ is the number of quadratic extensions. This counts the independent genus characters that are non-trivial.

## 7. Unramified Extensions

### 7.1 Definition

An **unramified extension** $L/K$ is a finite extension where no prime of $K$ ramifies in $L$:

$$e(\mathfrak{P}|\mathfrak{p}) = 1 \quad \text{for all primes } \mathfrak{P} \text{ of } L, \mathfrak{p} \text{ of } K$$

### 7.2 Hilbert Class Field

The **Hilbert class field** $H(K)$ is the maximal unramified abelian extension of $K$. It satisfies:

$$\text{Gal}(H(K)/K) \cong \text{Cl}(K)$$

### 7.3 Unramified Pro-p Extensions

The **maximal unramified pro-$p$ extension** $K^{(p)}$ is the compositum of all unramified extensions of $p$-power degree. Its Galois group is the pro-$p$ completion of $\text{Cl}(K)$.

### 7.4 The Tower

The unramified pro-$p$ tower is:

$$K = K_0 \subset K_1 \subset K_2 \subset \cdots$$

where $K_{i+1}$ is the Hilbert class field of $K_i$ (restricted to $p$-power degree).

The [[unit-distance-golod-shafarevich|Golod-Shafarevich inequality]] determines whether this tower is infinite.

## 8. Pro-p Towers and the Frattini Quotient

### 8.1 Pro-p Groups

A **pro-$p$ group** is an inverse limit of finite $p$-groups. Key examples:
- $\mathbb{Z}_p$ (p-adic integers)
- Free pro-$p$ groups
- Galois groups of unramified $p$-extensions

### 8.2 The Frattini Subgroup

For a pro-$p$ group $G$:

$$\Phi(G) = G^p [G, G]$$

where $G^p$ is the closed subgroup generated by $p$-th powers and $[G, G]$ is the commutator subgroup.

### 8.3 The Frattini Quotient

$$\overline{G} = G / \Phi(G)$$

is an elementary abelian $p$-group (vector space over $\mathbb{F}_p$).

- **Rank:** $d = \dim_{\mathbb{F}_p} \overline{G}$
- **Generator rank:** $d$ is the minimum number of generators for $G$
- **Relation rank:** $r$ is the number of independent relations among generators

### 8.4 Role in the Construction

The Frattini quotient encodes:
1. The 2-class rank $d$ of the base field
2. The capacity for split primes (via the GS inequality)
3. The structure of the unramified tower

## 9. Lemma 2.2: Unit Norm Generation

### 9.1 Statement

The critical lemma from `unit-distance-remarks.pdf` (Page 4):

> **Lemma 2.2.** Let $K$ be a number field embedded in $\mathbb{C}$. Let $k_1, \ldots, k_s$ be positive integers. Let $Q := \prod_{j=1}^s (P_j \overline{P}_j)^{k_j} \subseteq \mathcal{O}_K$ be an ideal. Let $U := \{u \in Q^{-2} : |u| = 1\}$. Then:
>
> $$|U| \ge \frac{\prod_{j=1}^s (k_j + 1)}{h(K)}$$

### 9.2 Interpretation

The lemma says:
- Given $s$ conjugate prime pairs $(P_j, \bar{P}_j)$
- With exponents $k_j$ on each pair
- The number of unit-norm elements in the ideal $Q^{-2}$ is at least $\prod (k_j + 1) / h(K)$

### 9.3 Why CM is Required

The lemma requires:
1. **Conjugate pairs** $(P_j, \bar{P}_j)$ — provided by complex conjugation
2. **Unit norm condition** $|u| = 1$ — satisfied because $|P_j| = |\bar{P}_j|$ in the complex embedding
3. **CM involution** — ensures the norm is balanced across conjugate pairs

Without CM, there are no conjugate pairs, and the lemma cannot be applied.

### 9.4 The Entropy Formula

For uniform exponents $k_j = k$ and $h(K) \le H$:

$$\log |U| \ge s \log(k+1) - \log H$$

The entropy per degree is:

$$\gamma = \frac{\log |U|}{[K:\mathbb{Q}]} \ge \frac{s \log(k+1) - \log H}{[K:\mathbb{Q}]}$$

This entropy drives the exponent $\delta$ in the lower bound.

### 9.5 Multi-Quadratic Amplification

In a multi-quadratic CM field of degree $2^N$:
- Each rational split prime contributes $2^{N-1}$ conjugate pairs (not just 1)
- The entropy becomes $s \cdot 2^{N-1} \log(k+1) - \log H$
- This is the **Galois multiplier** that powers H16

## 10. The Algebraic Toolkit in Action

### 10.1 The Construction Chain

```mermaid
graph TD
    A["Base Field F<br/>(CM field)"] --> B["Class Group<br/>2-class rank d"]
    B --> C["Golod-Shafarevich<br/>Infinite tower?"]
    C --> D["Unramified Tower<br/>F = F₀ ⊂ F₁ ⊂ F₂ ⊂ ..."]
    D --> E["Split Primes<br/>Killing Frobenius"]
    E --> F["Lemma 2.2<br/>Unit-norm elements"]
    F --> G["Point Configurations<br/>Many unit distances"]
    G --> H["Lower Bound<br/>u(n) = Ω(n^{1+δ})"]
```

### 10.2 Parameter Summary for H16

| Parameter | Value | Source |
|---|---|---|
| Base field $F$ | $\mathbb{Q}(\sqrt{-2}, \sqrt{3}, \sqrt{5}, \sqrt{7})$ | CM field construction |
| Degree $[F:\mathbb{Q}]$ | 16 | $2^4$ |
| Totally real subfield $F^+$ | $\mathbb{Q}(\sqrt{3}, \sqrt{5}, \sqrt{7})$ | CM structure |
| 2-class rank $d$ | 15 | Genus theory: $2^4 - 1$ |
| Unit rank $r_1 + r_2 - 1$ | 7 | $0 + 8 - 1$ |
| Relation rank $r$ | 22 | $d + r_1 + r_2 - 1$ |
| Split primes $t$ | 17 | Legendre symbol sieve |
| Unramified split primes | $\{59, 131, 251, \ldots, 2411\}$ | $(-2/q) = (3/q) = (5/q) = (7/q) = 1$ |
| GS bound $d^2/4$ | 56.25 | $15^2/4$ |
| GS total $r + 2t$ | 56 | $22 + 34$ |
| GS margin | 0.25 | $56.25 - 56$ |
| Exponent $\delta$ | 0.019603 | Closed-form computation |

## 11. Glossary

| Term | Definition |
|---|---|
| **CM field** | Totally imaginary quadratic extension of a totally real field |
| **CM involution** | Complex conjugation automorphism of a CM field |
| **Class group** | Group of fractional ideals modulo principal ideals |
| **2-class rank** | Rank of the 2-part of the class group |
| **Genus theory** | Method to compute 2-class rank using quadratic reciprocity |
| **Unramified extension** | Extension where no prime ramifies |
| **Pro-p tower** | Inverse limit of unramified p-power degree extensions |
| **Frattini quotient** | Elementary abelian p-quotient of a pro-p group |
| **Relation rank** | Dimension of the space of relations among generators |
| **Lemma 2.2** | Lower bound on unit-norm elements from conjugate prime pairs |
| **Split prime** | Prime that factors completely in the base field |

## References

- Neukirch, *Algebraic Number Theory* (1999)
- Washington, *Introduction to Cyclotomic Fields* (1982)
- Lang, *Algebraic Number Theory* (1994)
- Cox, *Primes of the Form x² + ny²* (1989)
- OpenAI, `unit-distance-remarks.pdf` (Lemma 2.2)
- OpenAI, `unit-distance-cot.pdf` (Chain of thought)