Quadratic iteration reveals a no-go theorem in octonions, suggesting new physics insights.
If non-associativity is to be treated as physical content rather than as an algebraic nuisance, the natural first candidate for a fundamental law is quadratic iteration on the octonions, ψₙ₊₁ = ψₙ · ψₙ + c — the direct generalization of the Mandelbrot map to the largest normed division algebra — with standard physics arising as its associative (quaternionic) truncation. The first result of this paper is a no-go theorem: by the Cayley–Hamilton identity for quadratic algebras, the orbit of this map is confined to the three-dimensional linear span of {1, ψ₀, c}, hence to an associative subalgebra, in every Cayley–Dickson algebra. The dynamics is never genuinely octonionic and all associators formed along the orbit vanish identically. This closes, with a theorem valid for every seed, a question opened by Griffin and Joshi in 1992–93: they proved the degeneracy for the critical orbit, repaired the map by adding an associator term c(ψa) − (cψ)a = −[c, ψ, a] with strength parameter, studied the resulting structural transitions numerically, and proved that their generalized Mandelbrot set remains insensitive to the associator. The second result identifies the algebraic source of that insensitivity and shows it is removable. The repair term is the torsion operator Tx,y = [x, · , y] = [R_y, L_x], whose kernel is the quaternion subalgebra ⟨x, y⟩; in the Griffin–Joshi parametrization the pair is locked to (c, a), which places the drive c inside the kernel, so the torsion term vanishes identically along the critical orbit (Theorem 4.3). Unlocking the pair — taking (a, b) independent of c — yields the completed law ψₙ₊₁ = ψₙ² + [a, ψₙ, b] + c. Via the Schafer derivation decomposition Ta,b = ⅓(Da,b − ad_[a,b]), the operator is proved antisymmetric, purely imaginary in output, of kernel exactly ⟨a, b⟩, and of minimal polynomial T³ = −ω²T with a single frequency ω = ‖[a, b]‖. The completed dynamics admits an exact skew-product reduction (real driver, scaled visible block, hidden block rotated at ω with norm multiplier √(4r² + ω²)), an exact associative truncation, and a closed confinement law: every orbit lies in the T-closure span{1, Tʲψ₀, Tʲc}j≤2, of dimension 7 generically and exactly 4 for the critical orbit — on which mutual orbit associators are nonzero. The payoff is the object the 1992–93 program sought: for ω > 2 it is a theorem, and for ω < 1 a verified computation, that membership of c in the completed Mandelbrot set depends on the direction of Im c, not only its magnitude. A genuinely octonionic Mandelbrot set exists one parameter away from the Griffin–Joshi map. Physical identifications — the real channel as universal (gravitational) coupling, the torsion-active complement as a dark sector distinguished only by ω, chirality as the orientation of the torsion pair — are stated as explicit conjectures, separated from the theorems. All structural claims are verified in an accompanying numerical and exact-arithmetic suite. The two companion Python scripts verify every numerically checkable claim in the paper; all reported figures reproduce deterministically under the canonical seeds fixed in the scripts (seed 7).
No takes yet. Share an insight, caveat, or question.
Ryan Sossai (2026) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: