🔢 mathematics

Implication semilattice of 990 quasigroup equational laws

This paper analyzes 990 equational laws on quasigroups originally considered by Ernst Schröder, determining all 114 equivalence classes of their conjunctions and the complete implication structure between them, which includes a five-element non-distributive lattice.

Original authors: Bruno Le Floch

Published 2026-04-01

📖 5 min read🧠 Deep dive

CC BY 4.0

Original authors: Bruno Le Floch

Original paper licensed under CC BY 4.0 (http://creativecommons.org/licenses/by/4.0/). ✨ This is an AI-generated explanation of the paper below. It is not written or endorsed by the authors. For technical accuracy, refer to the original paper. Read full disclaimer

Imagine you are a master chef trying to understand the fundamental rules of cooking. You have three basic tools: a Knife (cutting), a Spoon (mixing), and a Fire (heating). In the world of mathematics, these are like the three operations in a "Quasigroup": multiplication, right-division, and left-division.

Most of the time, we assume cooking rules are simple and predictable. If you chop an onion, then mix it, then heat it, the order matters, but the result is consistent. But what if the rules of your kitchen were weird? What if chopping an onion after mixing it changed the flavor in a way that mixing after chopping didn't?

This paper is a massive, 135-year-old detective story about finding the hidden rules of such a "weird kitchen."

The Backstory: Schröder's Lost List

In the 1890s, a mathematician named Ernst Schröder was trying to prove a famous philosopher (Peirce) wrong. Peirce claimed that all logical structures (lattices) were "distributive"—a fancy way of saying they follow a nice, predictable pattern like $A \times (B + C) = (A \times B) + (A \times C)$ .

Schröder thought, "I bet I can find a kitchen where this rule breaks." To do this, he wrote down 990 different recipes (equations). Each recipe was a potential rule for how these three tools (operations) interact. He wrote them down, but he only had time to check a few. He left the rest as a giant, unsorted pile of possibilities.

The Mission: Sorting the Pile

The author of this paper, Bruno Le Floch, decided to finish Schröder's homework. He took those 990 recipes and asked two main questions:

Which recipes are actually the same?
Imagine you have a recipe that says "Add salt, then pepper" and another that says "Add pepper, then salt." In some kitchens, these are different. In others, they might result in the exact same taste. Bruno used powerful computer programs (like digital taste-testers) to group the 990 recipes into 47 unique "flavor profiles." If two recipes always produce the same result, they are in the same group.
Which recipes force other recipes to be true?
This is the "Implication" part. If you follow Recipe A, does it automatically mean you are also following Recipe B?
- Analogy: If a rule says "You must wear a hat," it implies "You must wear something on your head." You don't need a separate rule for the second one; the first one forces it.
- Bruno mapped out every single connection. He built a giant family tree (or a map) showing which rules are the "parents" and which are the "children."

The Big Discovery: The "Bad" Lattice

The most exciting part of the paper is the discovery of a specific, tiny corner of this map.

Schröder was looking for a structure that was not distributive (not predictable). He found a small group of rules that, when combined, created a structure that looked like a five-pointed star rather than a neat grid.

Think of it like this:

Distributive Logic: Like a standard Lego set. If you have a red brick and a blue brick, you can snap them together in any order, and the final tower looks the same.
Schröder's Discovery: Like a set of magnetic blocks that repel each other in specific ways. If you try to build a tower, the order in which you snap them together changes the final shape. The rules are so specific that they create a "knot" in the logic that cannot be untangled into a simple grid.

The Four-Recipe Mystery

The paper highlights one specific variety (a specific type of weird kitchen) that is the hardest to describe. It requires four specific rules to define it. You can't get there with just one or two rules.

The author writes these four rules down in a complex equation, but the takeaway is simple: This is the most complex, stubborn little corner of the mathematical universe that Schröder was looking for. It's the "smoking gun" that proves not all logical structures are simple and predictable.

Why Should You Care?

You might think, "Who cares about 990 equations about weird math?"

But this is about classification. Just as biologists classify animals into species, families, and orders, mathematicians classify logical systems.

This paper is the field guide to a specific jungle of mathematical rules.
It tells us exactly how many different "species" of these weird math-kitchens exist (114 of them).
It shows us which ones are simple (like a group of friends who all agree on everything) and which ones are chaotic (like a room full of people shouting different instructions).

In a Nutshell

Bruno Le Floch took a dusty, unfinished list of 990 mathematical riddles from the 1890s, used modern computers to solve them, and organized them into a neat map. He found that while most of these rules lead to predictable, boring math, there is a tiny, fascinating corner where the rules break down in a beautiful, non-distributive way.

He didn't just finish Schröder's work; he turned a pile of scribbles into a library of logical possibilities, proving that even in a world of abstract rules, there is still room for surprise and complexity.

1. Problem Statement

The paper addresses a historical mathematical problem initiated by Ernst Schröder in the 1890s. Schröder sought to disprove a claim by Charles Sanders Peirce that all lattices are distributive. To do this, Schröder constructed a specific lattice structure based on the logical implications between 990 equational laws defined on quasigroups.

Quasigroups: Algebraic structures $(Q, *, //, \backslash)$ consisting of a set with a binary multiplication operation $*$ and two division operations ( $//$ and $\backslash$ ) that are inverses of multiplication. Unlike groups, quasigroups are not necessarily associative.
The Equations: The 990 laws are of the form $x?y?z = \alpha?\beta?\gamma$ , where the operations $?$ are chosen from $\{*, //, \backslash\}$ , variables are permuted, and associativity is varied.
The Goal: Schröder identified a small, non-distributive 5-element lattice within these laws but left the full structure of the implications and the classification of the resulting algebraic varieties incomplete. The author's goal is to complete this quest by determining the entire implication semilattice of these 990 laws and classifying the quasigroup varieties they define.

2. Methodology

The author employs a combination of historical reconstruction and modern automated reasoning tools, building upon the Equational Theories Project (ETP).

Formalization:
- The 990 equations are generated systematically based on Schröder's criteria, excluding tautologies and redundant permutations.
- The author defines "good subsets" of equations (those closed under logical implication). These subsets form a lattice where:
  - Product ($AB$): Set-theoretic intersection ( $A \cap B$ ).
  - Sum ( $A+B$ ): The smallest logically closed subset containing $A \cup B$ (the set of all logical consequences).
  - Implication ( $A \implies B$ ): Defined as $B \subset A$ .
Automated Theorem Proving (ATP):
- Prover9: Used to prove positive implications (i.e., to show that a set of equations $A$ logically implies an equation $E$ ).
- Mace4: Used to find finite countermodels (specifically, finite quasigroups) to disprove implications.
- Strategy: For any implication $A \implies E$ , the system first attempts a proof with Prover9. If that fails, it searches for a finite countermodel with Mace4. A crucial finding of this work is that all non-implications admit a finite counterexample, simplifying the search space compared to previous projects that required infinite countermodels.
Data Processing:
- Python scripts were used to interface with ATP tools, generate the 990 equations, identify equivalence classes, and assemble conjunctions of equations to find all logically closed subsets.

3. Key Contributions

Complete Classification of Equivalence Classes: The 990 equations are grouped into 47 equivalence classes. Two equations are equivalent if they imply each other. The paper lists representatives for these 47 classes.
Full Implication Semilattice: The author determines all logical implications between these 47 representatives. This allows for the construction of the full lattice structure of the 990 laws.
Enumeration of Varieties: The paper identifies and lists 114 distinct quasigroup varieties (classes of quasigroups defined by subsets of these laws). These are categorized by the minimum number of equations required to define them:
- 1 variety defined by 0 equations (all quasigroups).
- 47 varieties defined by 1 equation.
- 50 varieties defined by 2 equations (but not 1).
- 15 varieties defined by 3 equations (but not 2).
- 1 variety defined by 4 equations (but not 3).
Reconstruction of Schröder's Lattice: The paper explicitly locates Schröder's original 5-element non-distributive lattice within this larger structure, confirming the failure of the distributive law in this context.

4. Key Results

The "Long" Classes: The most significant result is the identification of the 114 logically closed subsets (varieties).
The Four-Equation Variety: One specific variety requires at least four equations to be defined. It is characterized by the conjunction of:
$\begin{aligned} x * (y // z) &= y * (x // z) \\ x * (y \backslash z) &= z * (y \backslash x) \\ x // (y \backslash z) &= y // (x \backslash z) \\ x \backslash (y * z) &= y \backslash (x * z) \end{aligned}$
This variety is highlighted as a potential candidate for further mathematical characterization.
Equivalence to Known Structures: Several single-equation varieties are mapped to known algebraic concepts, such as:
- Commutativity: $x * y = y * x$ .
- Groups: Associative quasigroups.
- Boolean Groups: Quasigroups where operations coincide with addition in a Boolean group.
- Semi-symmetry: Varieties related to twisted automorphisms.
Data Availability: The author provides supplementary files containing the full list of equations, representatives, implication graphs, and the JSON data for the 114 varieties, making the results reproducible and usable for further research.

5. Significance

Historical Completion: This work resolves a problem left open for over 130 years, providing the first complete map of the logical landscape of Schröder's 990 quasigroup laws.
Methodological Benchmark: It demonstrates the power of combining automated theorem proving with systematic enumeration to solve complex algebraic classification problems. The finding that finite countermodels suffice for all non-implications in this specific domain is a notable simplification for future algebraic investigations.
Algebraic Insight: By cataloging 114 distinct varieties, the paper provides a rich taxonomy of non-associative algebraic structures. It bridges the gap between abstract lattice theory and concrete quasigroup varieties, offering new candidates for studying non-distributive lattices and non-associative algebras.
Resource for the Community: The release of the data files (JSON, Python scripts) and the explicit lists of equations serves as a foundational resource for researchers in universal algebra, logic, and automated reasoning.

In summary, Bruno Le Floch has successfully completed Schröder's unfinished work, transforming a historical curiosity into a comprehensive, computationally verified classification of quasigroup varieties, thereby establishing a new reference point in the study of equational theories.