The Strong Force as Geometric Necessity
Octonions, Colour Confinement, and the Algebra of the Nucleus

Octonions, Colour Confinement, and the Algebra of the Nucleus

Martin Scholl — Independent Researcher  ·  This is the fourth paper in the series “It Is All One.” In [1] we showed that cosmological redshift follows from exponential metric contraction  ·  In [2] we showed that the quantum leap is a quaternion rotation, and E = hf is a geometric identity. In [3] we showed that the Pauli exclusion principle is a theorem of quaternion algebra, and reproduced sixty spectral lines of hydrogen and helium from first principles. Here we climb one rung higher on the algebraic ladder—from quaternions to octonions—and show that the strong force, colour confinement, and the internal structure of the proton are geometric consequences of the same algebra that gave us quantum mechanics.

Independent Researcher April 2026 This is the fourth paper in the series “It Is All One.” In [1] we showed that cosmological redshift follows from exponential metric contraction. In [2] we showed that the quantum leap is a quaternion rotation, and E = hf is a geometric identity. In [3] we showed that the Pauli exclusion principle is a theorem of quaternion algebra, and reproduced sixty spectral lines of hydrogen and helium from first principles. Here we climb one rung higher on the algebraic ladder—from quaternions to octonions—and show that the strong force, colour confinement, and the internal structure of the proton are geometric consequences of the same algebra that gave us quantum mechanics. Keywords: octonions, quaternions, Cayley-Dickson construction, strong force, colour confinement, SU(3), gluons, proton structure, division algebras, unified field theory

1Introduction

“The existing formalism is exactly right, but its interpretation is upside down.” — This paper

That bet paid off: sixty spectral lines, two atoms, all from one identity. But it also left a conspicuous hole. Quaternions have four components—one real and three imaginary. They account for the electron’s energy and its three angular-momentum degrees of freedom. They do not account for what happens inside the nucleus. The proton is not a point. It has internal structure: three quarks, eight gluons, a colour charge that is nothing like electric charge. The quaternion framework that works so beautifully for the electron is silent about the quarks. This silence is not a failure. It is a signpost. Quaternions are four-dimensional. The next division algebra is eight-dimensional. It is called the octonions, and it was discovered by John Graves in 1843—the same year Hamilton carved the quaternion equation into Brougham Bridge—and independently by Arthur Cayley in 1845. The octonions are strange: they are non-commutative (like quaternions) and also non-associative. That is, (ab)c ≠ a(bc) in general. For a century and a half, they were a mathematical curiosity with no known application in physics.

If this is correct, then the strong force is not a separate force that nature bolted onto electromagnetism as an afterthought. It is the same algebra, one rung higher. It is all one.

2The Cayley–Dickson Ladder: From Counting to the Nucleus

Mathematics has exactly four division algebras—systems of numbers where you can add, subtract, multiply, and divide without ever getting zero from nonzero factors. They are:

Each rung is built from the one below by the Cayley–Dickson construction. The recipe is simple: take two copies of the algebra you have, and define a new multiplication rule that couples them. Concretely:

21 From Real to Complex

Take two real numbers (a, b) and define:

(1)(a, b) × (c, d) = (ac − db, da + bc)

This is just the complex number a + bi. Multiplication is commutative and associative. You lose the ability to say which number is “bigger”—there is no ordering on the complex plane—but you gain something extraordinary: every polynomial equation has a solution (the fundamental theorem of algebra). The plane is algebraically complete.

22 From Complex to Quaternion

Take two complex numbers (a, b) and define:

(2)(a, b) × (c, d) = (ac − d*b, da + bc*)

where * denotes complex conjugation. This gives the quaternion a + bi + cj + dk. You lose commutativity: ij = k but ji = −k. In exchange, you gain the ability to represent rotations in three-dimensional space. This is the algebra we used in [2] and [3] to derive the quantum leap and the Pauli exclusion principle.

23 From Quaternion to Octonion

Take two quaternions (Q₁, Q₂) and define:

(3)(Q₁, Q₂) × (Q₃, Q₄) = (Q₁Q₃ − Q₄*Q₂, Q₄Q₁ + Q₂Q₃*)

This gives the octonion, an eight-component object:

(4)O = a₀ + a₁e₁ + a₂e₂ + a₃e₃ + a₄e₄ + a₅e₅ + a₆e₆ + a₇e₇

But notice what the Cayley–Dickson construction actually does: it takes two quaternions and welds them together. The first quaternion Q₁ = a₀ + a₁e₁ + a₂e₂ + a₃e₃ lives in the subspace {1, e₁, e₂, e₃}. The second quaternion Q₂ = a₄ + a₅e₅′ + a₆e₆′ + a₇e₇′ lives in the subspace {e₄, e₅, e₆, e₇}. The octonion is not “a quaternion plus four extras.” It is two quaternions, nested. You lose associativity: (e₁e₂)e₄ ≠ e₁(e₂e₄) in general. This sounds catastrophic. How can you do physics with a multiplication that depends on the order you group things? The answer is: you do the physics of the strong force. As we will show, the non-associativity is not a bug—it is the mechanism of colour confinement.

24 And Then It Stops

You can apply the Cayley–Dickson construction one more time and get the sixteen-dimensional sedenions. But they have zero divisors—two nonzero sedenions whose product is zero—so they are not a division algebra. Hurwitz proved this is not a limitation of the construction; it is a theorem about mathematics itself. There are exactly four division algebras, period. Here is the physical claim: each division algebra corresponds to a regime of physics. The first three rows are not controversial—they are standard physics, repackaged. The fourth row is the claim of this paper. Note the structure: the octonion row contains two quaternions side by side. The first is the familiar spatial quaternion. The second is a colour quaternion with its own scalar and its own three-vector.

3Quaternions Inside Octonions: The Embedding Theorem

Before we do anything with the strong force, we must prove a purely mathematical fact: every quaternion is an octonion with four components set to zero. If our quaternion calculations in [2] and [3] are correct—and they are, to 10 parts per million over sixty spectral lines—then the octonion representation of those same calculations must reduce to the quaternion result identically. No approximation. No correction. Identically. This is easy to verify. An octonion has eight components:

O = a₀ + a₁e₁ + a₂e₂ + a₃e₃ + a₄e₄ + a₅e₅ + a₆e₆ + a₇e₇

A quaternion has four:

Q = w + xi + yj + zk

The embedding is: i → e₁, j → e₂, k → e₃, and a₄ = a₅ = a₆ = a₇ = 0 (5) Under this embedding, the quaternion multiplication table is exactly preserved:

e₁e₂ = e₃ (just as ij = k)

e₂e₁ = −e₃ (just as ji = −k)

e₁² = e₂² = e₃² = −1 (just as i² = j² = k² = −1)

This is the key insight. The electron does not “know” it is secretly an octonion—it has no colour charge, no confinement energy. Its second quaternion is zero, and zero times anything is zero. The strong force is invisible to it. But the quark—which does have colour charge—needs the second quaternion. The quark is the full octonion. Think of it this way. A shadow on the wall is a two-dimensional projection of a three-dimensional object. The shadow is perfectly real—it has a definite shape, you can measure it. But it contains less information than the object that casts it. The first quaternion is the shadow. The full octonion is the object. For the electron, the shadow is enough because its colour quaternion is zero. For the quark, you need both.

4Two Nested Quaternions: The Architecture of the Octonion

“The octonion is not a quaternion with four extras bolted on. It is two quaternions, face to face, coupled by a multiplication rule that makes the whole greater than the sum of its parts.” — This paper The Cayley–Dickson construction builds the octonion from two quaternions. This is not a notational convenience—it is the architecture. Let us spell it out.

41 The Spatial Quaternion: What the Electron Uses

The first quaternion lives in the subspace {1, e₁, e₂, e₃}. We have already identified its physical meaning in [2]:

(6a)Q_spatial = E + L_x·e₁ + L_y·e₂ + S_z·e₃

The real part (1-direction) is the energy—a scalar, invariant under spatial rotations. The three imaginary parts (e₁, e₂, e₃) are the angular momenta—a vector that rotates. The relationship between scalar and vector is the relationship between energy and spin. The real part is what you measure when you stop the rotation. The imaginary parts are the rotation itself. For the electron, this is the whole story. Q_spatial is the complete description. The second quaternion is zero.

42 The Colour Quaternion: What the Quark Adds

The second quaternion lives in the subspace {e₄, e₅, e₆, e₇}. Now here is the structural insight that the Cayley–Dickson construction hands us for free: this second quaternion has exactly the same architecture as the first. It has a scalar part and a three-vector part. The unit e₄ is to (e₅, e₆, e₇) what 1 is to (e₁, e₂, e₃).

(6b)Q_colour = E_conf·e₄ + r·e₅ + g·e₆ + b·e₇

The “real part” of the colour quaternion—the e₄ component—is the colour scalar: the quantity that is invariant under colour rotations, just as energy is invariant under spatial rotations. What is this colour scalar physically? It is the confinement energy. The proton’s mass is 938.272 MeV/c². The three quark masses add up to about 9 MeV—less than 1% of the total. The remaining 99% comes from the energy of the colour field itself: the kinetic energy of quarks bouncing inside the proton, the gluon field energy, the quantum fluctuations of the colour vacuum. This is the confinement energy E_conf, and it lives in the e₄ component—the scalar of the colour quaternion. The three imaginary parts—the e₅, e₆, e₇ components—are the three colour charges: red, green, blue. They form a three-vector in colour space, just as (L_x, L_y, S_z) form a three-vector in physical space. They are the rotation. The colour scalar E_conf is what remains when the colour rotation stops—the energy locked inside the proton by the act of confinement.

43 The Full Octonion: Two Quaternions Coupled

The complete octonion for a quark is the sum of both quaternions:

(6)Q_quark = Q_spatial + Q_colour

= E + L_x·e₁ + L_y·e₂ + S_z·e₃ + E_conf·e₄ + r·e₅ + g·e₆ + b·e₇

Note carefully what the plus sign in Eq. (6) does — and does not do. As a vector space, the octonion is a direct sum: 𝕆 = ℍ ⊕ ℍ, two orthogonal subspaces sharing no dimension. Addition is therefore inert: forming Q_spatial + Q_colour changes neither term. No component of the one leaks into the other; project at any time and you recover each quaternion exactly as it was. The sum is coexistence, not interaction. The interaction lives entirely in the multiplication — the Cayley–Dickson cross terms, the gluons, the non-associativity. Sum: the two universes stand side by side, untouched. Product: they talk. Now look at the symmetry. Each quaternion has: The architecture is identical. The scalar is the invariant—the thing that doesn’t change under rotations. The three-vector is the thing that rotates. In space, the rotation is physical rotation (spin, orbital motion). In colour space, the rotation is colour rotation (the SU(3) gauge transformations of the standard model). The Cayley–Dickson construction does not invent this symmetry—it enforces it. The second quaternion must have a scalar and a three-vector, because that is what a quaternion is. This is why the strong force has three colour charges, not two or four. Three is the number of imaginary units in a quaternion. The colour quaternion inherits the structure of the spatial quaternion, including its dimensionality. Three colours, three spatial directions. The same number, for the same algebraic reason.

5Eight Gluons, Eight Units: The Colour Algebra

The standard model describes the strong force using the symmetry group SU(3)—the group of 3×3 unitary matrices with determinant 1. This group has eight generators, called the Gell-Mann matrices λ₁ through λ₈. Each generator corresponds to one gluon. This is not a coincidence we are about to explain; it is a fact of the standard model that we are about to derive from octonion algebra. An octonion has seven imaginary units: e₁ through e₇. But we said there are eight gluons. Where does the eighth come from? The answer lies in the automorphism group of the octonions.

51 The Octonion Multiplication Table: Seven Cyclic Triples

The multiplication of the seven imaginary octonion units is defined by seven cyclic triples—index cycles that specify which three basis elements multiply to give the identity, and in which order. These seven triples are the octonion algebra itself; they are not an external diagram imposed on the algebra, but the multiplication table written in its most compact form. The seven cyclic triples of the octonion multiplication table are: Notice: Line 1 is the spatial quaternion subalgebra {e₁, e₂, e₃}—the same multiplication table as {i, j, k} from Hamilton. The other six lines all couple spatial units (e₁, e₂, e₃) to colour units (e₄, e₅, e₆, e₇). They are the cross-couplings between the two quaternions. It is these cross-couplings that create the non-associativity, and it is the non-associativity that creates confinement.

52 From Seven Units to Eight Gluons

The automorphism group of the octonions—the group of transformations that preserve the multiplication table—is the exceptional Lie group G₂, which has dimension 14. Inside G₂ sits SU(3), the symmetry group of the strong force, as a subgroup. How does SU(3) emerge? Not by fixing the whole spatial quaternion—the subgroup of G₂ that preserves a quaternion subalgebra is SO(4), a different group. The theorem (Cartan 1914; Baez [9], §4.2) is sharper: the subgroup of G₂ that fixes a single imaginary unit is SU(3). The natural unit to fix is e₄—the colour scalar, the confinement-energy axis. Nail down e₄, and the automorphisms that survive are exactly the colour rotations of {e₅, e₆, e₇}. And this is the same move that creates electromagnetism one rung up: complex quantum mechanics is quaternion mechanics with one axis nailed down—the Compton shutter e₃—and the symmetry that survives that nailing is U(1), the electromagnetic phase (companion note on Schrödinger). The gauge group of each rung is the stabilizer of that rung’s shutter axis: U(1) stabilizes the Compton shutter, SU(3) stabilizes the colour shutter, and the weak SU(2) is the double cover of the automorphism group of the quaternion rung itself.

(7)Aut(𝕆) = G₂ ⊃ Stab(e₄) = SU(3)

SU(3) has eight generators, and the count needs no numerology from the seven units: it is a fact about the stabilizer. The fourteen dimensions of G₂ split as 14 = 8 + 6—eight generators that fix e₄ (these are the gluons, the Lie algebra of SU(3)) and six coset directions that would move e₄ itself, the transformations the colour shutter forbids. In the Gell-Mann basis the eight are the familiar six off-diagonal colour-changing generators plus the two diagonal ones, λ₃ and λ₈. Let us count carefully:

• Eight gluons because the colour shutter’s stabilizer is eight-dimensional: dim SU(3) = 8

• dim G₂ = 14 = 8 + 6: the six discarded directions are those that would rotate e₄ away

• Total: 8 gluon states = 8 generators of SU(3), counted by the group, not by the units

53 Do It Yourself: The Octonion Multiplication Table

You can verify every claim in this section with nothing more than the multiplication table below. Each entry gives the product e_a × e_b. Read the row label as the left factor and the column label as the right factor. Table 1: The Octonion Multiplication Table (e_a × e_b) Check the upper-left 3×3 block (spatial quaternion): e₁e₂ = +e₃, e₂e₁ = −e₃, and all diagonal entries are −1. This is exactly Hamilton’s multiplication table. The spatial quaternion sits inside the octonion, undisturbed. Now check the non-associativity. Pick three units that span both quaternions, say e₁, e₂, e₄:

(8a)(e₁ × e₂) × e₄ = e₃ × e₄ = e₇

(8b)e₁ × (e₂ × e₄) = e₁ × e₆ = −e₇

The two results differ by a sign! This is the non-associativity at work. It arises only when a product mixes units from both quaternions—when space talks to colour. Within each quaternion alone, associativity holds. The non-associativity is the coupling between the two quaternions, and it is the geometric origin of colour confinement.

6Confinement as Geometric Necessity

“No one has ever seen a free quark. We claim this is not an empirical accident but a consequence of the algebra—stated here as a precise conjecture, with its one unproven premise in plain sight.” — This paper The deepest mystery of the strong force is confinement. Quarks are never observed alone. Pull two quarks apart and the energy stored in the gluon field between them grows linearly with distance—until there is enough energy to create a new quark–antiquark pair from the vacuum. You end up with two hadrons, not one free quark. Every experiment ever performed confirms this. But the standard model does not derive it—it is put in as an assumption, supported by lattice QCD numerical simulations.

61 Observability Requires Associativity

What does it mean to observe a particle? It means to measure its properties—energy, momentum, spin—and get definite, reproducible numbers. In the algebraic framework, a measurement is a product of the particle’s state with a detector’s state. For the measurement to be unambiguous, it must not matter how we group the factors:

(9)(Detector × Particle) × Environment = Detector × (Particle × Environment)

This is associativity. If the algebra is non-associative, the measurement result depends on how we group the factors—which means the measurement is ambiguous. The particle cannot be isolated and measured independently of its context. For electrons (spatial quaternion only), associativity holds. We can isolate an electron, put it in a trap, measure its charge and magnetic moment to twelve decimal places. No problem. For quarks (both quaternions active), associativity fails. A single quark cannot be measured independently of its colour environment. This is not a practical limitation—it is algebraic.

62 Why Three Quarks Restore Associativity

But hadrons—protons, neutrons—are observed. How does a system of three quarks become associative when each quark alone is not? The answer is colour neutrality. A proton contains three quarks, one red, one green, one blue. In the octonion algebra, the colour-vector components of the three quarks are: Quark 1: r₁·e₅ (red) Quark 2: g₂·e₆ (green) Quark 3: b₃·e₇ (blue) The composite colour-vector is:

(10)C_proton = r₁·e₅ + g₂·e₆ + b₃·e₇

For colour neutrality (r = g = b in magnitude), the colour vector transforms under SU(3) as a singlet—the trivial representation. A singlet is invariant under all SU(3) transformations. And SU(3) is the associative part of the octonion automorphism group—it is precisely the subgroup that preserves the spatial quaternion {1, e₁, e₂, e₃}. In other words: when the colour charges sum to a singlet, the non-associative cross-couplings between the two quaternions cancel. The proton’s effective algebra collapses back to the spatial quaternion—four-dimensional, associative, observable. The colour quaternion is still there, but it is locked into a configuration that makes it invisible from outside. This is exactly what confinement is. The quarks are not “trapped” by a confining potential as though they were balls on springs. They are algebraically invisible as individuals because the cross-coupling between the two quaternions is non-associative, and non-associative objects cannot produce unambiguous measurement results. Only colour-neutral combinations—where the non-associativity cancels—can be observed.

63 Mesons: The Two-Body Case

Mesons consist of a quark and an antiquark: q̅q. The antiquark carries anti-colour: if the quark is red (e₅), the antiquark is anti-red (−e₅). Their colour vectors cancel:

(11)C_meson = r·e₅ + (−r)·e₅ = 0

Zero colour vector. The cross-coupling vanishes. The meson is associative—hence observable. It is unstable (the quark and antiquark can annihilate), but while it exists, it behaves as a purely spatial quaternion. You can measure its mass, spin, parity, charge. Every meson in the particle data tables is colour-neutral, exactly as the algebra requires. But notice: the colour scalar e₄ component (the confinement energy) does not cancel in the meson—it adds. The meson’s mass comes from E_conf, just as the proton’s does. The colour vector vanishes (colour-neutral), but the colour scalar persists (mass).

7The Wells Within Wells, Revisited

In [3], we introduced the image of wells within wells. With the two-quaternion framework, we can now be precise about this hierarchy: The pattern: as you go deeper into the well, the second quaternion wakes up. The electron lives entirely in Q_spatial. The quark lives in both. The transition from one to two active quaternions is the transition from electromagnetic to strong-force physics—from eV to GeV, from the Bohr radius to the proton radius, from hydrogen spectral lines to deep inelastic scattering.

71 A Universe on Every Quark

Let us say plainly what the two-quaternion architecture means. Our universe, in this series, is a quaternion: one scalar clock and three orthogonal axes. The colour quaternion is the same structure again — its own scalar, E_conf on e₄, the inner clock whose stabilizer is the gauge group SU(3) (Section 5.2), and its own three-vector, r, g, b — and it stands orthogonal to ours on all four axes. Orthogonal on every axis is the only definition of another universe that algebra offers: no rotation of our four axes, however long applied, ever produces a component on theirs. The quark does not carry extra properties bolted to a particle. It carries a second universe, of the same make as the first, opened on top of it. The proton and neutron then explain themselves as plays staged entirely within that inner universe — three actors, colour choreography, a closed curtain — of which our universe receives only the box-office total. And the orthogonality must be said with full force: one quaternion does not possess the dimensions of the other. The octonion is their sum, and the sum influences neither term — addition in a direct sum cannot move so much as a milligram between the blocks. This is not a formal nicety; it is measured physics twice over. First: the electron’s spectrum is reproduced to 10 parts per million by the spatial quaternion alone (Section 3), and adding a colour quaternion — zero or otherwise — alters no spatial component, which is why the embedding theorem holds identically rather than approximately. Second: when SLAC finally looked inside the proton, the struck quark scattered exactly like a pure Dirac point — ordinary spin-1/2, ordinary charge, the Callan–Gross relation — despite carrying a whole second universe on its back. As long as the universes merely coexist (addition, projection, observation of one block), each behaves as if the other were not there. Only when they act on each other — multiplication, gluon exchange, the triple products of the companion paper’s closure angle — do the cross terms wake. Identity lives in the sum; interaction lives in the product; confinement lives in the product of products. This inner universe is not parallel; it is coupled. The Cayley–Dickson rule welds the two quaternions into one octonion, and the weld itself is physical: the gluons are the cross-couplings, and the non-associativity that confines is precisely the two universes talking to each other. There is no road into the inner universe — orthogonal means no direction points there — only a door, and the door is energy: 1474 MeV, the wavelength that fits inside the proton (Section 9). SLAC did not travel anywhere in 1968. It paid the door price. From outside, the inner universe shows exactly one number. Its vector part cancels in every observable state — that is confinement — but its scalar survives: the confinement energy, the mass. Ninety-nine percent of the proton’s mass is the budget of its inner universe. Weigh anything — this page, the Earth — and what the scale reads is, overwhelmingly, the summed ledgers of the inner universes, one on every nucleon. Mass is the shadow the inner universe casts on ours. And the inner universe obeys the outer law. Its cage, at R ≈ 0.2 fm, sits at T = 1.8 × 10¹² K on the same family line T·R = ħc/2πk_B = 0.3644 mm·K that carries the Hubble horizon’s whisper and the chemistry of a warm afternoon (companion note on the metrics of the living spaces). Doubling the algebra did not double the law. That is why the title of this series survives the discovery that every proton contains a universe: it is all one — twice.

72 Three Horizons, One Answer

The quark cannot leave the hadron. We cannot leave, or even see past, the cosmological horizon — the exit cone closes (Paper 1 of this series). And a particle cannot hold still under the microscope: localize it below its Compton cell and it refuses, at any price, to be pinned. Three prisons, on three rungs of the ladder, carrying three unrelated names in the textbooks — confinement, causal horizon, uncertainty. They are one design. The proof of the common design is what happens when you press. Pull the quark: the string holds its 0.9 GeV per femtometre, and within a third of a femtometre the invested energy exceeds the price of a quark–antiquark pair — the string snaps into twins (Section 10). You do not get a free quark; you get a larger population. Squeeze the electron: localization below its Compton cell costs more than 2mₑc² = 1.022 MeV, and the vacuum answers with electron–positron twins — you do not get a pinned electron; you get a crowd. Hold station against a horizon: the price of refusing to fall is a thermal bath, the Unruh and Gibbons–Hawking pairs. Every horizon in the theater, pressed hard enough, answers escape with birth. The wall is not a wall. It is a maternity ward. The family law prices all three cages on one line. The hadron cage melts at 1.8 × 10¹² K — deconfinement, the quark–gluon plasma, where quarks do briefly walk, because the whole neighbourhood has become cage. The Compton cell melts near 10⁹ K — the pair plasma, where electrons lose their individuality to the crowd. And the doors of the ladder — the 1474 MeV door of Section 9, the 1.022 MeV pair door, the 80 GeV weak door of the companion paper — are these same barriers read from the other side: from below, a door you may pay to open; from inside, a horizon that answers every payment with twins. Doors and horizons are one geometry, approached from two directions.

8A Worked Example: The Proton

Let us write down the proton explicitly. A proton contains two up quarks and one down quark, in a colour-neutral (singlet) state. We assign colours: up₁ = red, up₂ = green, down = blue.

81 Three Quark Octonions

Each quark is an octonion with both quaternions active. Using Eq. (6):

Q_u1 = E_u + L₁·e₁ + L₂·e₂ + S₃·e₃ + E_conf·e₄ + r·e₅ + 0·e₆ + 0·e₇

Q_u2 = E_u + L₁′·e₁ + L₂′·e₂ + S₃′·e₃ + E_conf′·e₄ + 0·e₅ + g·e₆ + 0·e₇

Q_d = E_d + L₁″·e₁ + L₂″·e₂ + S₃″·e₃ + E_conf″·e₄ + 0·e₅ + 0·e₆ + b·e₇

82 The Colour Singlet

The colour-vector part of the proton’s wave function is the antisymmetric combination:

(12)ψ_colour = (1/√6) × (rgb − rbg + gbr − grb + brg − bgr)

In octonion language, this antisymmetrization over e₅, e₆, e₇ ensures that the composite transforms as a singlet under SU(3). The non-associative cross-couplings between Q_spatial and Q_colour cancel pairwise—just as the k-components cancel in parahelium [3].

83 What the Proton Looks Like from Outside

From outside—from the electron’s perspective—the proton is:

Q_proton = (2E_u + E_d) + (total L)·e₁ + ... + (E_conf,total)·e₄ + 0·e₅ + 0·e₆ + 0·e₇

The colour vector vanishes (e₅ = e₆ = e₇ = 0). But the colour scalar does not—E_conf persists in the e₄ component. This is the proton’s mass! The 938 MeV that the electron “sees” is overwhelmingly confinement energy, sitting in the scalar of the colour quaternion. The electron orbits this mass without knowing where it comes from. It computes hydrogen spectral lines using Q_spatial alone, and gets every line right to 10 ppm. The colour vector is hidden (confinement). The colour scalar is visible (mass). The proton shows its weight but hides its colour. This is the algebraic meaning of colour confinement: the vector part of the colour quaternion cancels to zero, but the scalar part survives as the proton’s mass.

84 The Mass Puzzle

Can we compute the proton’s mass from this framework? Not yet—not with pencil and paper. The proton mass arises from the dynamics of three interacting octonions confined to a colour-neutral state. In the standard model, this requires lattice QCD on supercomputers. But the framework tells us where to look: the mass is E_conf, the scalar of the colour quaternion. It is the “energy” of the colour space, in precisely the same sense that E is the energy of physical space. Computing it requires solving the non-associative dynamics of Q_colour coupled to Q_spatial—the full octonion problem. We leave this to future work, but note that lattice QCD (which does compute the proton mass to 2%) is, in our framework, a numerical solution of exactly this octonion dynamics.

9The Experiment: Proton–Electron Scattering Across Energy Scales

Everything in this paper has been algebra. Let us now make it experimental. Consider the simplest possible collision: a single electron hitting a single proton. By changing the electron’s energy, we can watch the proton’s algebraic nature change before our eyes—from one active quaternion to two—using nothing but a particle accelerator and a detector.

91 Regime I: eV — The Quaternion Handshake

At 13.6 eV, the electron’s de Broglie wavelength is about 53,000 fm. The proton radius is 0.84 fm. The ratio is 63,000 to 1. The electron sees a point—a spatial quaternion with E = 938.3 MeV, spin-1/2, charge +1. The colour quaternion is invisible. The result is hydrogen: twenty-six lines, accurate to 10 parts per million. Q_spatial meets Q_spatial. Both are associative. The physics is governed entirely by {1, e₁, e₂, e₃}—four dimensions, four degrees of freedom.

λ_dB ≈ 53,000 fm ≫ r_p = 0.84 fm → Only Q_spatial visible → ℍ × ℍ

92 Regime II: keV to MeV — Still One Quaternion

Increase the energy. The wavelength shrinks, but slowly: Even at 10 MeV, the electron’s wavelength is 140 times the proton radius. The scattering follows the Mott formula—a single-quaternion calculation—and gives the right answer. The colour quaternion remains hidden. Q_spatial is the whole story.

93 Regime III: GeV — The Second Quaternion Revealed

Now set the electron energy to 1.47 GeV. At this energy, the de Broglie wavelength equals the proton radius:

(13)λ_dB = 2πℏc / pc = 2π × 197.3 / 1474 = 0.84 fm = r_p

This is the threshold. Below it, the electron sees one quaternion. Above it, the electron’s wavelength penetrates the proton and resolves the second quaternion—the colour quaternion. In 1968, SLAC fired electrons at 4–20 GeV into protons: The scattering pattern showed Bjorken scaling: the electron was hitting point-like objects inside the proton—objects with both spatial and colour quantum numbers. It found: Two up quarks and one down quark, carrying about 54% of the proton’s momentum. These are full octonions—both quaternions active, colour vector nonzero (e₅, e₆, e₇). Gluons, carrying about 41% of the proton’s momentum. In the two-quaternion framework, these are the cross-couplings between Q_spatial and Q_colour—the non-associative products that mix the two quaternions. They are not objects in the usual sense. They are the algebra coupling the two quaternions to each other. Forty-one percent of the proton’s momentum is carried not by things but by the relationship between the two quaternions. Sea quarks (virtual quark–antiquark pairs), carrying about 5%.

94 The Punchline

The electron never changes. It is the same spatial quaternion at 13.6 eV and at 20 GeV. What changes is the resolution. Below 1.47 GeV, only Q_spatial is visible. Above 1.47 GeV, Q_colour is revealed. The algebra does not change—the octonion was always there. What changes is whether the experiment can see one quaternion or two. SLAC did not discover quarks. SLAC discovered the proton’s second quaternion.

95 The Number: The Melting Temperature of Matter

A paper that only re-derives known structure owes its referee one number of its own. Here it is. The colour cage is a living space on the family line of this series (T·R = ħc/2πk_B, the law that binds every rung from the Hubble horizon to the Planck scale; companion notes on the metrics of the living spaces and the radii of the worlds). Its curvature radius wears three sizes connected by two geometric gears: the thermal cell R = ħc/(2πk_B·T_c); the seven-sphere radius ħc/Λ, larger by exactly π because Λ = 2k_B·T_c (310 = 2 × 155 MeV, to lattice precision — one confined quark weighs two melting temperatures); and the charge radius, larger again by the tetrahedral factor 4/3 of the companion paper on neutron decay. Chain the gears: r_p = (4π/3)·R, i.e. r_p·k_B·T_c = (2/3)·ħc. Now run it backwards, from the measured proton to the temperature at which this paper’s cage opens: k_B·T_c = (2/3) × ħc / r_p = (2/3) × 197.327 / 0.841 = 156.4 MeV = 1.81 × 10¹² K The deconfinement temperature of matter, from the proton charge radius and nothing else. It lands twice. Lattice QCD — supercomputer years — places the crossover at 155–158 MeV. Better, because it is an experiment: the chemical freeze-out temperature measured in heavy-ion collisions at RHIC and the LHC, the temperature at which the quark–gluon plasma actually re-hadronizes, is 156.5 ± 1.5 MeV. The prediction sits 0.1 MeV from the measured value. (Citation to be verified in the literature pass; the freeze-out determination is the statistical-hadronization analysis of Andronic, Braun-Munzinger, Redlich and Stachel.) What makes this the paper’s number rather than another re-derivation: the Standard Model treats r_p and T_c as unrelated outputs of two separate hard calculations — one from lattice spectroscopy, one from lattice thermodynamics. Here they are the same number in two units, because the cage and its melting point are one geometry read twice. The claim is falsifiable by any improvement of either measurement: if r_p·k_B·T_c drifts from (2/3)ħc = 131.6 MeV·fm, the chain breaks. And the caveat is owed in print: the two gears are independently motivated — Λ = 2k_B·T_c is an observed identity, the 4/3 is the companion paper’s tetrahedral factor — but the chain as a whole is a discovered consistency, not yet a derivation from the algebra. It is stated here so the referee reads it from us first. A related consistency for free: the proton’s mass is six melting temperatures, m_p ≈ 3Λ = 6·k_B·T_c = 930 MeV, within 0.9% of 938.

10The Collision Ledger: Pair Creation as Octonion Arithmetic

“The vacuum can only give birth to twins—a thing and its exact mirror. Because zero, on every axis, equals zero.” — This paper In the preceding sections we described the proton’s internal structure and showed how deep inelastic scattering reveals the otherworldly companion—the colour quaternion hiding inside every quark. But we treated the collision qualitatively. Now we do the bookkeeping. We write the electron and the quark as explicit algebraic objects—a quaternion and an octonion—and track every component through the collision. Eight dimensions in, eight dimensions out. No exceptions. A note on conventions before the arithmetic. This section uses the ledger convention: each octonion component is a slot for a conserved quantity—charge on e₁, spin on e₂, colour on e₅ through e₇. Elsewhere in this series the axes play a dynamical role: rotation planes, with energy as the shutter rate and charge as the winding of the circle fiber (companion notes on the field equations and on Schrödinger). Both dictionaries are used deliberately. The ledger convention organizes conservation—this section’s eight-axis balance—while the dynamical convention organizes motion. The map between them, which windings fill which slots, is part of the State Octonion programme.

101 The Electron as Quaternion

The electron has no colour. Its otherworldly companion is zero on all four axes. It is a pure spatial quaternion:

(10)Q_e = E_e + q_e·e₁ + s_e·e₂ + L_e·e₃

For a 1.5 GeV electron aimed at a stationary proton:

Q_e = 1500 + (−1)·e₁ + (−½)·e₂ + 0·e₃

The real part is the energy in MeV. The e₁ component is the electric charge (−1 for the electron). The e₂ component is the spin projection (−½). The e₃ component is the orbital angular momentum (zero for a beam particle). The colour components e₄ through e₇ are all zero—absent, not merely small. The electron has no access to the otherworldly companion.

102 The Quark as Octonion

The up quark sitting inside the proton is the full eight-dimensional object. Take the red up quark as our example:

(11)O_q = E_q + q_q·e₁ + s_q·e₂ + L_q·e₃ + E_conf·e₄ + r·e₅ + g·e₆ + b·e₇

O_q = 313 + (+⅔)·e₁ + (+½)·e₂ + 0·e₃ + 310·e₄ + 1·e₅ + 0·e₆ + 0·e₇

The energy, 313 MeV, is the constituent quark mass—one-third of the proton. The charge is +2/3. The spin is +1/2 (opposite to the electron’s, as they approach). The confinement energy on e₄ is 310 MeV—the quark’s share of the colour field that holds the proton together. And on e₅ sits the colour charge: red = 1. Note what distinguishes this up quark from the other up quark in the proton: the second up quark sits on e₆ (green) instead of e₅ (red), and has opposite spin. They have the same mass, the same electric charge, the same flavour—but they are not the same octonion, because they live on different imaginary axes. This is the Pauli exclusion principle, now stated in eight dimensions instead of four: no two identical fermions may occupy the same octonion.

103 The Total Before: Eight Conservation Laws

The total state before the collision is the component-wise sum of the electron (quaternion) and the struck quark (octonion). Since the electron has zero colour components, it contributes nothing to e₄ through e₇: These eight numbers are the eight conservation laws. Every one must be identical in the total after the collision. Energy is conserved. Charge is conserved. Spin is conserved (or traded for orbital angular momentum within the spatial quaternion). Confinement energy is conserved as part of the total energy budget. And colour is conserved absolutely—no process can create or destroy net colour.

104 The Collision: Breaking the Eggshell

At 1.5 GeV, the electron’s de Broglie wavelength is 0.83 fm—just under the proton radius of 0.84 fm. The eggshell cracks. The electron punches through the proton’s exterior and hits one quark. It does not hit the proton as a whole; it hits one red up quark. The electron transfers energy and momentum via a virtual photon. The photon is a quaternion object—it carries energy (real axis) and angular momentum (e₂, e₃) but no charge (e₁ = 0) and no colour (e₄ through e₇ = 0). The photon is the messenger of the spatial quaternion. It literally cannot touch the colour components. After the photon exchange, the electron limps away with less energy. Say it retains 500 MeV and its spin has flipped from −1/2 to +1/2. The remaining 1000 MeV has been dumped into the quark system. The struck quark recoils violently. It begins to fly away from the two spectator quarks. As it does, the colour field—the confinement energy on e₄—stretches like a rubber band. But this rubber band has a property unique to the strong force: its tension does not decrease with distance. It stays constant, at approximately 1 GeV per femtometre. Energy pours from the kinetic (real axis) into the confinement (e₄ axis). The e₄ component grows and grows.

105 The Birth of Twins: Pair Creation as Rotation

When the confinement energy on e₄ exceeds the threshold for creating a quark–antiquark pair (roughly 300 MeV for up/down quarks), the field snaps. Energy that was sitting on the real axis, accumulated on e₄, now rotates into the imaginary colour directions. A quark appears on +e₅ and an antiquark appears on −e₅. The vacuum has given birth to twins. This is not a metaphor. In our algebra, pair creation is a rotation from the scalar (real or e₄) axis into a conjugate pair of imaginary directions. The rotation must produce equal and opposite imaginary components—otherwise the net colour would change, violating conservation on e₅. The only way to add something to the universe without changing any conserved quantity is to add a thing and its exact mirror: +1 and −1, +1/2 and −1/2, +1/3 and −1/3. On every imaginary axis, the twins sum to zero.

106 The Four Particles After

Let us write the four outgoing particles explicitly. We take the simplest case: the colour field produces a down quark and a down antiquark. The antiquark (anti-red) pairs with the departing up quark (red) to form a colour-neutral meson (π⁺ = u + d̄). The new down quark (red) fills the colour hole in the remnant baryon. Now we check conservation on every axis: Every axis balances. The algebra closes. The twins—the d and d̄—contribute zero to every conserved quantity: their charges cancel (+1/3 and −1/3), their spins cancel (+1/2 and −1/2), their colours cancel (+1 and −1 on e₅). They came from zero and they sum to zero. The vacuum gave birth to a thing and its exact negative, and the universe noticed nothing.

107 Why the Twins Must Be Conjugate

This is a theorem, not an observation. Consider what would happen if the vacuum produced two quarks of the same colour—say both red. Then e₅ after = 1 + 1 + 1 = 3, but e₅ before = 1. The books would not balance. Colour conservation forbids it. The same argument applies to charge, spin, and every other imaginary axis. The pair creation rule is therefore: The vacuum can only create an octonion O and its conjugate −O. (12) Because O + (−O) = 0 on every imaginary axis. This is the octonion version of the familiar rule that particle–antiparticle pairs are always produced together. But now we see why: the octonion has eight components, each independently conserved. The only way to change the system without changing any conserved quantity is to add zero in disguise—a pair of conjugates. Annihilation is the reverse: the conjugates collapse back onto the real axis, and out comes a photon—pure energy, pure scalar, zero on every imaginary axis.

108 What the Quark Is Afterward

The struck quark is not the same octonion after the collision. Its energy has changed (the real part increased by the photon’s energy). Its spin may have flipped (the e₂ component changed). Its confinement energy is wildly different—the colour field was torn apart and rebuilt. The quark’s identity—its charge on e₁ and its colour on e₅—is unchanged, because the photon cannot touch e₁ (it carries no charge) and cannot touch e₅ (it has no colour). But the quark is now bound in a meson instead of a baryon. Its life is completely different. Same charge, same colour—new address, new roommate, and a much shorter lease. The meson will decay in 10⁻⁸ seconds. The proton, meanwhile, has lost one quark and gained another from the vacuum. The red hole was filled by the new d quark. The remnant—now containing u(green), d(blue), d(red)—is a neutron. The proton has become a neutron plus a pion. Conservation of charge confirms: proton (+1) = neutron (0) + π⁺ (+1).

109 The Philosophical Implication

The otherworldly companion—the colour quaternion—is not visible. It is not directly measurable. No detector has ever registered a colour charge. And yet the algebra demands its existence: without three colour directions, two identical up quarks cannot coexist inside one proton without violating Pauli exclusion. Without a confinement energy scalar, the proton’s mass has no explanation. Without the conjugate rule on all eight axes, pair creation has no mechanism. The otherworldly companion is there because the algebra requires it—not because we chose to put it there, and not because a detector saw it. We allowed the algebra to lead, and it led us to a hidden quaternion that governs the strongest force in nature. The lesson is not about quarks. The lesson is about the courage to follow mathematics beyond what is directly visible, and to trust that what the algebra demands, nature provides.

11Discussion

111 What We Have Shown

This paper makes seven claims: First, the four normed division algebras (ℝ, ℂ, ℍ, 𝕆) correspond to four regimes of physics: classical mechanics, wave mechanics, quantum mechanics, and the strong force. Second, the octonion is not a quaternion with extras. It is two quaternions—a spatial quaternion Q_spatial = (E, L_x, L_y, S_z) and a colour quaternion Q_colour = (E_conf, r, g, b)—coupled by the Cayley–Dickson multiplication rule. The colour quaternion has the same scalar-plus-three-vector architecture as the spatial quaternion. The scalar (e₄) is the confinement energy. The three-vector (e₅, e₆, e₇) is the colour charge. Third, the symmetry group of the strong force, SU(3), arises as the stabilizer of the single colour unit e₄ inside the octonion automorphism group G₂ — not of the spatial quaternion, whose stabilizer is SO(4) (Eq. 7). The eight gluons correspond to the eight generators of SU(3). Fourth, colour confinement is a geometric argument—a conjecture resting on one open premise (that observability requires associative products), stated plainly rather than assumed silently. The non-associativity of the octonion algebra arises from the cross-coupling between the two quaternions. Colour-neutral combinations cancel this cross-coupling, restoring associativity and hence observability. Fifth, the proton’s mass is the scalar of the colour quaternion—the confinement energy E_conf in the e₄ component. The colour vector cancels (confinement), but the colour scalar survives (mass). This is why 99% of the proton’s mass is binding energy, not quark mass. Sixth, the transition from one active quaternion to two is experimentally observable at E ≈ 1.47 GeV—the threshold of deep inelastic scattering. SLAC discovered the proton’s second quaternion in 1968. Seventh, pair creation is an octonion theorem. The vacuum can only produce conjugate pairs—an octonion O and its negative −O—because each of the eight components is independently conserved, and the only way to add something to the universe without changing any conserved quantity is to add zero in disguise.

112 What We Have Not Shown

We have not computed the proton mass from first principles. In the standard model, this requires lattice QCD. In our framework, it requires solving the dynamics of Q_colour coupled to Q_spatial for three quarks in a colour-neutral state. We have not addressed the electroweak force. The weak interaction involves SU(2), which is the automorphism group of the quaternions. This suggests that the weak force lives at the quaternion level. We defer this to a future paper. We have not explained why there are three generations of quarks and leptons. The three division algebras ℂ, ℍ, 𝕆 applied over each other may produce three copies of the fermion spectrum. The number three—three colours, three generations, three spatial dimensions, three imaginary quaternion units—appears too often to be coincidental.

113 Falsifiability

A theory that explains everything explains nothing. What would falsify this framework:

1Discovery of a fifth division algebra (Hurwitz says impossible).

2Observation of a free quark (would break the associativity argument).

3A hadron that is not a colour singlet.

4A symmetry group of the strong force other than SU(3).

None of these have happened in fifty years of experimental particle physics.

12Conclusion

“Hamilton needed three days to carve his equation into a bridge. Graves needed three months to find the octonions. Nature needed 13.8 billion years to build a physicist who could read both.” — This paper We have climbed the Cayley–Dickson ladder from quaternions to octonions and found, at the top, not a foreign algebra but a familiar one—doubled. The octonion is two quaternions, face to face. The first governs space: energy, spin, angular momentum. The second governs colour: confinement energy, red, green, blue. The electron uses only the first. The quark uses both. The proton hides its second quaternion behind colour neutrality, showing only its mass—the scalar of the colour quaternion—to the outside world. The pattern is now clear: ℝ → classical | ℂ → waves | ℍ → quantum | ℍ×ℍ → strong force And the ladder stops at four, because mathematics has exactly four division algebras. The strong force is the last rung—not a new algebra, but the old algebra doubled. In Paper 1, we showed that cosmological redshift is geometry. In Paper 2, that E = hf is geometry. In Paper 3, that the Pauli exclusion is geometry. In this paper, that colour confinement is geometry—the geometry of two quaternions coupled inside an octonion. It is all one.

References

[1] M. Scholl, “Exponential Infall Cosmology: Gravitational Metric Contraction as the Mechanism of Cosmological Redshift,” unpublished manuscript (2026). [2] M. Scholl, “The Quantum Leap as Quaternion Rotation: E = hf as Geometric Identity,” unpublished manuscript (2026). [3] M. Scholl, “The Pauli Exclusion Principle as Logical Consequence of Hamilton’s Quaternion Algebra,” unpublished manuscript (2026). [4] W. R. Hamilton, “On a new species of imaginary quantities connected with a theory of quaternions,” Proceedings of the Royal Irish Academy, vol. 2, pp. 424–434 (1844). [5] J. T. Graves, Letter to W. R. Hamilton, 26 December 1843. [6] A. Cayley, “On Jacobi’s elliptic functions, in reply to the Rev. Brice Bronwin; and on quaternions,” Philosophical Magazine, vol. 26, pp. 210–213 (1845). [7] A. Hurwitz, “Über die Composition der quadratischen Formen von beliebig vielen Variabeln,” Nachrichten von der Gesellschaft der Wissenschaften zu Göttingen, pp. 309–316 (1898). [8] É. Cartan, “Les groupes réels simples, finis et continus,” Annales Scientifiques de l’École Normale Supérieure, vol. 31, pp. 263–355 (1914). [9] J. C. Baez, “The Octonions,” Bulletin of the American Mathematical Society, vol. 39, no. 2, pp. 145–205 (2002). arXiv:math/0105155. [10] M. Günaydin and F. Gürsey, “Quark structure and octonions,” Journal of Mathematical Physics, vol. 14, no. 11, pp. 1651–1667 (1973). [11] G. M. Dixon, “Division Algebras: Octonions, Quaternions, Complex Numbers and the Algebraic Design of Physics,” Kluwer Academic Publishers (1994). [12] C. Furey, “Standard Model Physics from an Algebra?” Ph.D. thesis, University of Waterloo (2015). arXiv:1611.09182. [13] C. Furey, “Three generations, two unbroken gauge symmetries, and one eight-dimensional algebra,” Physics Letters B, vol. 785, pp. 84–89 (2018). [14] E. D. Bloom et al., “High-Energy Inelastic e–p Scattering at 6° and 10°,” Physical Review Letters, vol. 23, no. 16, pp. 930–934 (1969). [15] M. Breidenbach et al., “Observed Behavior of Highly Inelastic Electron–Proton Scattering,” Physical Review Letters, vol. 23, no. 16, pp. 935–939 (1969). [16] Particle Data Group, “Review of Particle Physics,” Physical Review D, vol. 110, 030001, 2024.

Symbols & Terms