🔢 mathematics

Near-Trinions: Complete Classification of Unital Three-Dimensional Real Associative Algebras

This paper provides a complete classification of unital three-dimensional real associative algebras, termed "near-trinions," into exactly six isomorphism classes by stratifying them according to their radical dimension and utilizing tools such as the Wedderburn-Artin theorem, Peirce decomposition, and module-theoretic invariants.

Original authors: Joel A. Shelton

Published 2026-06-12

📖 5 min read🧠 Deep dive

CC BY 4.0

Original authors: Joel A. Shelton

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 an architect trying to build a specific type of house: a three-dimensional, self-contained, rule-abiding structure made of real numbers.

For a long time, mathematicians knew that if you wanted a "perfect" house (where you can divide by anything without breaking the rules), you could only build them in sizes 1, 2, 4, or 8. There is no perfect 3-story house. This was a famous discovery by a mathematician named Frobenius in 1878.

But this paper asks a different, slightly more relaxed question: What if we don't need a "perfect" house? What if we just want any 3-story structure that follows the basic rules of multiplication?

The author, Joel A. Shelton, calls these structures "Near-Trinions" (a play on the Latin word for "three"). The paper's main discovery is that there are exactly six unique ways to build these 3-story structures. No more, no less.

Here is the breakdown of the paper's journey, explained simply:

1. The Problem: The Missing Three

Think of the number system as a ladder.

Step 1: Real numbers (1D).
Step 2: Complex numbers (2D).
Step 4: Quaternions (4D).
Step 8: Octonions (8D).

The ladder skips 3. You can't build a "perfect" 3D division algebra. But the author says, "Okay, let's stop trying to be perfect. Let's just build any 3D algebra, even if it has some weak spots (like zero divisors)."

2. The Method: Sorting by "Rot"

To find all possible 3D structures, the author uses a clever sorting method based on the Radical.

The Metaphor: Imagine every building has a "foundation" or a "core." Sometimes this core is solid (no rot). Sometimes it has a little bit of rot (a small weak spot). Sometimes it has a lot of rot.
The Strategy: The author groups all possible buildings by how much "rot" (mathematically called the Jacobson radical) they have.
- Group A (No Rot): The building is perfectly solid.
- Group B (One Unit of Rot): The building has a small weak spot.
- Group C (Two Units of Rot): The building is mostly weak, with only a tiny solid tip at the top.

(Note: You can't have 3 units of rot, because then the whole building would collapse into nothing, and it wouldn't be a valid algebra anymore.)

3. The Six Unique Buildings

By sorting through these groups, the author finds exactly six distinct blueprints. Here is what they are, using simple analogies:

Group A: The Solid Buildings (No Rot)

These are the "perfect" ones, just split into smaller rooms.

The Triple Room (NT1): Imagine a house with three separate, identical rooms that never talk to each other. It's just three copies of the real numbers sitting side-by-side.
The Twin Suite (NT2): Imagine a house with one real room and one complex room (which has an "imaginary" side). They are separate but share the same foundation.

Group B: The One-Rot Buildings (The Split)

Here, the building has a small weak spot. The author discovers that the location of this weak spot changes everything.
3. The Stacked Room (NT3): The weak spot is tucked neatly inside one of the main rooms. The whole building still behaves nicely (it's commutative, meaning order doesn't matter when you multiply).
4. The Sliding Door (NT6): The weak spot is stuck in the corner between two rooms. This creates a weird asymmetry. If you walk through the door one way, you get a result; if you go the other way, you get zero. This is the only non-commutative building in the list (order matters here!).

Why not a Complex Room? The author proves that you cannot build a 3D house where the solid part is the "Complex" numbers and the weak part is attached to it. The math simply breaks down (a negative number cannot equal a positive square).

Group C: The Two-Rot Buildings (The Local Ones)

Here, the building is almost entirely weak, with only a tiny solid tip (the number 1) at the very top.
5. The Flat Floor (NT4): The weak part is so squishy that if you multiply any two weak pieces together, they vanish instantly. It's like a floor made of foam that collapses on contact.
6. The Staircase (NT5): The weak part is slightly stronger. You can multiply two weak pieces to get a third weak piece, but if you multiply three, they vanish. It's like a staircase where you can go up two steps, but the third step leads to the void.

4. How to Tell Them Apart

The paper provides a "ID Card" for each of these six buildings so you can never confuse them. The ID cards check:

How much rot is there? (Separates the three main groups).
Is the building symmetrical? (Tells apart the "Stacked Room" from the "Sliding Door").
How deep is the collapse? (Tells apart the "Flat Floor" from the "Staircase").

Summary

The paper is a complete catalog. It says: "If you want to build a 3D algebra using real numbers, you have exactly six choices. Here are their blueprints, here is how they work, and here is why no other choices are possible."

It doesn't just list them; it explains why the list looks the way it does, proving that the "missing" 3D perfect algebra forces the other six to exist in very specific, constrained shapes.

Technical Summary: Near-Trinions: Complete Classification of Unital Three-Dimensional Real Associative Algebras

Problem Statement
While Frobenius' theorem (1878) definitively establishes that no three-dimensional real division algebras exist, the classification of general three-dimensional real associative algebras remains a distinct problem. The author, Joel A. Shelton, addresses the existence and classification of these objects, termed "near-trinions." The central question is to determine the complete set of isomorphism classes for finite-dimensional, associative, unital $\mathbb{R}$ -algebras of dimension three.

Methodology
The paper employs a structural approach based on the stratification of algebras by the dimension of their Jacobson radical, $d = \dim_{\mathbb{R}} \text{rad}(A)$ . Since the algebra is unital and three-dimensional, $d$ is restricted to the set $\{0, 1, 2\}$ (as $d=3$ would imply the algebra is nilpotent, contradicting the existence of a unit). The classification proceeds through three strata:

Semisimple Stratum ( $d=0$ ): The Wedderburn–Artin theorem is applied. Since the algebra is semisimple, it decomposes into a direct product of matrix rings over division algebras. Frobenius' theorem restricts the division algebras over $\mathbb{R}$ to $\mathbb{R}$ , $\mathbb{C}$ , and $\mathbb{H}$ . Dimensional constraints ( $n^2 \dim D = 3$ ) eliminate $\mathbb{H}$ and $M_2(\mathbb{R})$ , leaving only specific combinations of $\mathbb{R}$ and $\mathbb{C}$ .
One-Dimensional Radical Stratum ( $d=1$ ): The quotient $A/\text{rad}(A)$ $A / rad (A)$ is a two-dimensional semisimple algebra, isomorphic to either $\mathbb{R}^2$ $R^{2}$ or $\mathbb{C}$ $C$ .
- The case $A/\text{rad}(A) \cong \mathbb{C}$ is proven impossible via a characteristic sign obstruction: if the radical generator $\eta$ were acted upon by the imaginary unit $i$ of the quotient, one would derive $i \cdot \eta = c\eta$ for $c \in \mathbb{R}$ , leading to $c^2 = -1$ , which has no real solution.
- The case $A/\text{rad}(A) \cong \mathbb{R}^2$ is analyzed using Peirce decomposition relative to the orthogonal idempotents of the quotient. The radical generator is located either in a diagonal corner (yielding a commutative algebra) or an off-diagonal corner (yielding a non-commutative algebra).
Two-Dimensional Radical Stratum ( $d=2$ ): The quotient is isomorphic to $\mathbb{R}$ , making the algebra local. The classification splits based on whether the square of the radical vanishes ( $\text{rad}(A)^2 = 0$ ) or is non-zero.

Key Results
The paper establishes that there are exactly six isomorphism classes of near-trinions. These are presented with canonical representatives, explicit multiplication tables, and a set of module-theoretic invariants sufficient to distinguish every pair.

The six classes are:

$d=0$ (Semisimple):
- (NT1) $\mathbb{R}^3 = \mathbb{R} \times \mathbb{R} \times \mathbb{R}$
- (NT2) $\mathbb{R} \times \mathbb{C}$
$d=1$ (One-dimensional radical):
- (NT3) $\mathbb{R} \times \mathbb{R}[\varepsilon]/(\varepsilon^2)$ (Commutative; radical in diagonal Peirce corner)
- (NT6) $T_2(\mathbb{R})$ (Non-commutative; radical in off-diagonal Peirce corner)
$d=2$ (Two-dimensional radical):
- (NT4) $\mathbb{R}[\varepsilon_1, \varepsilon_2]/(\varepsilon_1^2, \varepsilon_2^2, \varepsilon_1\varepsilon_2)$ (Radical square is zero)
- (NT5) $\mathbb{R}[x]/(x^3)$ (Radical square is non-zero; Loewy length 3)

Distinguishing Invariants
The paper demonstrates irredundancy (no two classes are isomorphic) using the following invariants:

Radical dimension ( $d$ ): Separates the three strata.
Commutativity: Distinguishes NT3 from NT6 within the $d=1$ stratum.
Loewy length ( $\ell\ell(A)$ ) and $\text{rad}(A)^2$ : Distinguishes NT4 from NT5 within the $d=2$ stratum.
Idempotent count: Distinguishes NT1 from NT2 within the $d=0$ stratum.

Significance and Contribution
The paper notes that while the classification of low-dimensional algebras has been addressed previously (specifically by Scheffers in 1891 and Kobayashi et al. in modern treatments), the contribution of this work lies in its methodological framework and explanatory depth.

Narrative Coherence: The "near-trinion" framing connects the classification to the historical context of the non-existence of 3D division algebras.
Structural Explanation: Unlike coordinate-based approaches that vary structure constants to enumerate classes, this paper explains why the classification takes its specific form. It elucidates why the radical dimension is restricted to $\{0, 1, 2\}$ , why the $d=1$ case bifurcates strictly on commutativity, and why a residue field of $\mathbb{C}$ is structurally impossible in this dimension over $\mathbb{R}$ .
Comparison with Prior Work: The paper explicitly maps its six classes to the entries in the classification by Kobayashi et al., confirming agreement while providing a distinct derivation via Jacobson radical stratification and Peirce decomposition.

Limitations and Future Directions
The classification is specific to finite-dimensional, associative, unital algebras over $\mathbb{R}$ . The author notes that the results do not extend to infinite dimensions (where Nakayama's Lemma and radical bounds may fail) or non-associative algebras (Lie, Jordan, octonions). A proposed direction for subsequent work is the classification over finite fields, where the sign obstruction used in the $d=1$ case (the unsolvability of $c^2 = -1$ ) may not hold, potentially altering the number of isomorphism classes.