The mechanism
How freedom is lost: the lock-in sequence
One of the more consistent findings is that the suppression of political freedom is not random. It follows a reproducible sequence that appears across systems separated by thousands of years and tens of thousands of kilometres.
L → I → A → E↓ → D↓ → R↓
Reading left to right: surplus resources (L) fund institutionalisation (I: the construction of durable offices, rules, records, and enforcement routines), which in turn expands administrative capacity (A). Once that apparatus is sufficiently developed, it begins to constrain the freedom to leave (E). As exit becomes harder, the freedom to disobey (D) follows, and finally the ability to replace rulers (R) is foreclosed.
This sequence maps onto the history of state formation across the ancient world. The Mauryan Empire, the Tang Dynasty, the Roman Republic’s transformation into the Principate, the Ottoman Empire under the Hamidian autocracy: all follow recognisable variants of this pattern.
Source note: lock-in sequence
The lock-in sequence is documented in Paper 2 (draft complete), which applies logistic regression and survival analysis to the full 401-system dataset. The sequence is derived empirically from transition probabilities between states, not imposed as a theoretical prior. The D⊂0; odds ratio (6.46) is from a logistic regression with D⊂0; as predictor and lock-in status as outcome. The 5.03× exit rate is from a Cox proportional hazards model. Both are reported with 95% confidence intervals in the paper.
Notation used in this project
S = sovereignty index (0–1): concentrated coercive power of the ruling apparatus.
A = administrative capacity (0–1): depth of state penetration through records, taxes, education, surveillance.
D = disobedience freedom (0–1): practical freedom to refuse compliance or organise opposition without severe sanction.
D₀ = the initial level of D at the start of a trajectory or lock-in process. A high D₀ gives citizens more structural room to resist before the window closes.
θ = the empirically estimated critical threshold, here D = 0.45. Systems that fall below this level show sharply lower probabilities of recovery. It is a statistical regularity in the data, not a theoretical assumption.
L = surplus resource legibility (surpluses that are identifiable and extractable by the state). I = institutionalisation. E = exit freedom. R = replacement freedom.
D₀ as a key moderator
Not all systems lock in equally. The key variable is the initial level of disobedience freedom, D₀, at the point when the lock-in process begins. Systems starting above the θ = 0.45 threshold show a 5× higher exit rate from the lock-in pathway. They have more room to organise, protest, and resist before the window closes.
This matters practically. Political freedom is not just an outcome but a structural resource. Societies that begin with high D₀ are substantially harder to lock in. This is consistent with why authoritarian consolidations so often target civil society, press freedom, and the right to organise first: not because these things are immediately threatening, but because they represent the structural D₀ that makes resistance possible later.
Societies that begin with high disobedience freedom are substantially harder to lock in. High D₀ is the structural resource that makes later resistance possible.
The appointment attractor
A related finding from a separate model (a continuous-time Markov chain applied to succession mechanisms, Paper 3) shows what happens to leadership replacement as lock-in progresses. As administrative capacity rises and political competition falls, governance systems converge on an appointment attractor: a stable state in which power is transferred by the incumbent to a chosen successor. The equilibrium probability of this state reaches π = 0.595 in high-surplus systems.
This appears in Bronze Age palaces, mediaeval courts, twentieth-century single-party states, and contemporary hybrid regimes. The mechanism is structural, not cultural.
