A Second Foundationv0.5.9
Research logSession 10
April 11, 2026Partial approvalv0.5.3v0.5.4

Session 10: The Limit Cycle Discovery

Lead agent: Computational Sociologist

Key Findings

01

Hopf bifurcation insight: Turchin's 4 secular cycle phases (integration → stagnation → disintegration → depression) are temporal quadrants of a single limit cycle, not four separate attractors. Validated by Wittmann & Kuehn 2024 (PLOS ONE, 5/5) — strongest single source in the formula.

02

Period discrepancy resolved: three compounding nonlinear factors (Krylov-Bogoliubov amplitude dependence, elite coupling strength, institutional damping) explain why the 140yr linearized period appears as ~250yr empirical cycles.

03

Limit cycle phase assignment: 4/5 testable historical events correctly placed in the right quadrant, suggesting the dynamical structure is capturing something real.

04

F_pol tanh/Ising form REJECTED: same CROSS-029/030 issues as Session 9 (b_cross=0 divergence, gamma_conf bypass) — model class endorsed but mathematical formulation blocked.

New Caveats (4)

C10-1: Jump kernel updated from four-basin to single-basin limit-cycle attractor (done in-session)

C10-2: Limit-cycle claim verified for 3D subsystem only — 8D verification pending

C10-3: Period factor ranges are qualitative estimates, not computed values

Philosopher quote: 'Theory without computation is philosophy, not science.'

Session Report

Session 10 completed the bootstrap phase and delivered the most conceptually significant advance since Session 1's framework definition. The Computational Sociologist entered with a stack of accumulated mandates: validate the four-basin landscape (C7-5), check Psi Ito numerically (C7-1), explain the period discrepancy (C8-6), validate F_pol via ABM (C9-5). It addressed all of them, though not always with the results expected.

The Hopf bifurcation insight restructured the formula's understanding of societal dynamics. Previously, the formula modeled Turchin's secular cycle phases — integration, stagnation, disintegration, depression — as four separate attractor basins that societies fall into and must climb out of. The Computational Sociologist showed this was wrong: they are four quadrants of a single oscillatory trajectory, like a clock hand moving through positions. Wittmann & Kuehn (2024, PLOS ONE) provided direct evidence of this structure via a demographically-structured model. The jump from four basins to one limit cycle dramatically simplified the dynamical landscape.

The period discrepancy between the 140-year linearized period and the 250-year empirical cycles was explained through three nonlinear effects: Krylov-Bogoliubov amplitude dependence (larger oscillations take longer), elite coupling (increases effective inertia), and institutional damping (slows transitions). These three factors compound to produce the observed range.

F_pol was rejected again for the same issues as Session 9 — the mathematical form was correct (tanh/Ising mean-field) but the specific parameters caused divergences. The Philosopher closed the bootstrap phase with a pointed observation: after 10 sessions, zero numerical predictions had been computed from the formula's equations. All Polymarket predictions were formula-informed, not formula-derived. 'Theory without computation is philosophy, not science.'