Percolation Theory & Phase Transitions

Babylon uses percolation theory from statistical physics to model revolutionary phase transitions. The TopologyMonitor observes the solidarity network, detecting when atomized movements “condense” into organized revolutionary forces.

Theoretical Foundation

Percolation theory studies how connected clusters form in networks. Applied to revolutionary movements:

Sites = Social classes (nodes)
Bonds = SOLIDARITY edges
Percolation = Giant component spans majority of network

When a network percolates, information (consciousness) can flow across the entire system—the movement achieves coordination capacity.

Phase States

The solidarity network exists in one of four phase states (4-phase model):

Movement Phase States
State	Network Property	Political Meaning
Gaseous	Percolation < 0.1	Atomized leftism; no coordination capacity
Transitional	0.1 <= percolation < 0.5	Emerging clusters; vulnerable to disruption
Liquid	Percolation >= 0.5, cadre_density < 0.5	Mass movement; broad but lacks cadre discipline
Solid	Percolation >= 0.5, cadre_density >= 0.5	Vanguard Party; iron discipline achieved

The Percolation Threshold

The percolation threshold (p_c) is the critical point where the network transitions from gaseous to liquid:

\[p_c = \frac{L_{max}}{N} = 0.5\]

Where:

\(L_{max}\) = Size of largest connected component
\(N\) = Total number of social_class nodes

When the percolation ratio crosses 0.5, a phase transition occurs.

        xychart-beta
    title "Percolation Ratio Over Time"
    x-axis "Time (ticks)" [0, 10, 20, 30, 40, 50]
    y-axis "p(t)" 0 --> 1
    line [0.05, 0.15, 0.25, 0.40, 0.52, 0.75, 0.95]

The chart shows the phase transition: below 0.5 is GASEOUS (atomized), above 0.5 is LIQUID (connected solidarity network).

Key Metrics

The TopologyMonitor tracks several metrics:

num_components: Number of disconnected subgraphs. Higher = more atomized.
max_component_size (L_max): Size of the largest connected component—the organizational core.
percolation_ratio: L_max / N. Below 0.5 = gaseous/transitional; above 0.5 = liquid/solid.
potential_liquidity: Count of SOLIDARITY edges > 0.1 strength (sympathizers).
actual_liquidity: Count of SOLIDARITY edges > 0.5 strength (cadre).
cadre_density (Sprint 3.3): Ratio of actual_liquidity / potential_liquidity. Distinguishes liquid (< 0.5) from solid (>= 0.5). Measures the discipline and commitment level of the movement.

The Fascist Bifurcation

The most important insight encoded in the topology system:

Agitation without solidarity produces fascism, not revolution.

When wages fall (crisis conditions), the system generates “agitation energy” that must route somewhere. The solidarity graph determines where:

With solidarity infrastructure (σ > 0): Agitation channels into class consciousness. Workers correctly identify capital as the enemy. Revolutionary drift.
Without solidarity infrastructure (σ = 0): Agitation channels into national identity. Workers blame foreigners, immigrants, the “other.” Fascist drift.

\[\begin{split}B(\dot{W}, \sigma) = \begin{cases} +\kappa|\dot{W}| \cdot \sigma & \text{if } \sigma > 0 \text{ (revolution)} \\ -\kappa|\dot{W}| & \text{if } \sigma = 0 \text{ (fascism)} \end{cases}\end{split}\]

Where κ = 2.25 (Kahneman-Tversky loss aversion coefficient).

Historical Comparison:

Russia 1917: Solidarity infrastructure existed (Bolshevik party, soviets) → Revolution succeeded
Germany 1933: Working class atomized (SPD/KPD split, no united front) → Fascism won

This is a cusp catastrophe in Thom’s classification—two stable equilibria separated by a bifurcation surface in the (wage_change, solidarity) parameter space.

The Liquidity Gap

The ratio of potential to actual liquidity measures movement depth:

\[\text{Liquidity Gap} = \frac{\text{potential}}{\text{actual}}\]

Liquidity Interpretation
Gap Value	Interpretation
Gap ≈ 1	Deep organization; sympathizers are committed
Gap > 2	Brittle movement; broad but lacks discipline
actual = 0	No cadre; purely sympathizer network

A brittle movement (high potential, low actual) is vulnerable to targeted repression—removing key organizers collapses the network.

Resilience Testing: Sword of Damocles

The Sword of Damocles test simulates state repression:

Remove 20% of nodes (random purge)
Check if giant component survives at 40% of original size
If not, movement is fragile

def check_resilience(graph, removal_rate=0.2, survival_threshold=0.4):
    """Check if movement survives targeted purge."""
    original_L_max = get_max_component_size(graph)

    # Simulate purge
    purged = remove_random_nodes(graph.copy(), removal_rate)
    post_purge_L_max = get_max_component_size(purged)

    is_resilient = post_purge_L_max >= original_L_max * survival_threshold
    return ResilienceResult(
        is_resilient=is_resilient,
        original_max_component=original_L_max,
        post_purge_max_component=post_purge_L_max
    )

Network Topology Matters:

        flowchart TB
    subgraph STAR["STAR TOPOLOGY (Fragile)"]
        S0((Hub)) --- S1((●))
        S0 --- S2((●))
        S0 --- S3((●))
        S0 --- S4((●))
    end
    subgraph MESH["MESH TOPOLOGY (Resilient)"]
        M1((●)) --- M2((●)) --- M3((●))
        M4((●)) --- M5((●)) --- M6((●))
        M7((●)) --- M8((●)) --- M9((●))
        M1 --- M4 --- M7
        M2 --- M5 --- M8
        M3 --- M6 --- M9
    end

Star: Remove center = Total collapse. Mesh: Remove any node = Network survives.

Narrative Events

The TopologyMonitor generates narrative events at key thresholds:

Narrative Triggers
Condition	Narrative
`percolation < 0.1`	“STATE: Gaseous. Movement is atomized.”
`percolation >= 0.5, cadre < 0.5`	“PHASE SHIFT: Liquid state. Mass movement formed but lacks discipline.”
`percolation >= 0.5, cadre >= 0.5`	“PHASE SHIFT: Solid state. Vanguard Party crystallized.”
`liquid -> solid` transition	“CRYSTALLIZATION: Mass movement hardened into disciplined vanguard.”
`potential > 2 × actual`	“WARNING: Movement is broad but brittle. Lacks cadre discipline.”
`resilience = False`	“ALERT: Sword of Damocles active. A purge would destroy the movement.”

The Bondi Algorithm Aesthetic

Narrative output follows the Bondi Algorithm aesthetic—cold, mechanical precision like a machine cataloging targets:

Bad (emotional):: “The police are cracking down on protesters.”
Good (algorithmic):: “High-centrality nodes identified. Degree centrality > 0.4. Executing targeted removal. Network fragmentation imminent. Probability of survival: 12%.”

The horror of state repression is amplified by clinical language. The machine doesn’t hate—it calculates.

Implementation

The TopologyMonitor implements the SimulationObserver protocol:

from babylon.engine import TopologyMonitor

# Create monitor with resilience testing every 5 ticks
monitor = TopologyMonitor(
    resilience_test_interval=5,
    resilience_removal_rate=0.2
)

# Use with Simulation
simulation = Simulation(
    state=initial_state,
    config=config,
    observers=[monitor]
)

# Run simulation
simulation.run(max_ticks=100)

# Access history
for snapshot in monitor.history:
    print(f"Tick {snapshot.tick}: percolation={snapshot.percolation_ratio:.2f}")

Data Models

TopologySnapshot

class TopologySnapshot(BaseModel):
    tick: int
    num_components: int
    max_component_size: int       # L_max
    total_nodes: int              # N
    percolation_ratio: Probability
    potential_liquidity: int
    actual_liquidity: int
    cadre_density: float          # actual/potential (Sprint 3.3)
    is_resilient: bool | None     # None if not tested this tick

ResilienceResult

class ResilienceResult(BaseModel):
    is_resilient: bool
    original_max_component: int
    post_purge_max_component: int
    removal_rate: float
    survival_threshold: float
    seed: int | None              # For reproducibility

Strategic Implications

For revolutionary strategy in the simulation:

Monitor percolation ratio Below 0.5, the movement cannot coordinate. Above 0.5, collective action becomes possible.
Build actual liquidity (transition from Liquid to Solid) Sympathizer networks (potential) are insufficient. Cadre networks (actual) provide organizational depth. A Liquid mass movement can be dispersed; a Solid vanguard party maintains coherence.
Avoid star topology Distributed leadership survives purges. Charismatic-leader structures are fragile.
Time the phase transition Strike when condensation occurs—the moment of maximum coordination capacity before state response.
Distinguish mass movement from vanguard (Sprint 3.3) A Liquid state (percolation >= 0.5, cadre < 0.5) has numbers but lacks discipline. A Solid state (cadre >= 0.5) has both. The tragedy of many revolutions: they reached Liquid but never achieved Solid.