The Pattern
Page 6 of Constitution as Colimit — early-stage research program.
The three axioms are not specific to physics. They describe the constitutional process at every scale. The same process that produces quarks from sub-quark structure produces cells from molecules, ecosystems from organisms, and societies from individuals. What changes between scales is the signal space and the measurement apparatus, not the process.
This page is about what happens when that process fails.
Three failure modes structural
The framework identifies three ways an emergent system can fail. The first two (fragmentation and condensation) have formal categorical definitions. The third (rate-exceeds-modulation) is currently a conjectured mode awaiting formalization. Proving that these are exhaustive — that no fourth mode exists — remains an open problem.
Fragmentation. The system loses connections until the network breaks into disconnected pieces. Hub nodes — the ones that hold the system together — are removed or degraded. The network fragments into isolated clusters that cannot sustain the whole.
An ecosystem that loses keystone species. A supply chain that loses critical suppliers. A democracy that loses the institutions connecting citizens to governance.
Condensation. The system collapses toward a few controlling nodes. Signal flow concentrates until one entity absorbs most of the network’s function. The system still looks connected, but the connections all point inward to the same center.
A market that consolidates into monopoly. A cancer network where a few driver genes dominate signal flow. A political system where power concentrates in fewer and fewer hands.
Rate-exceeds-modulation. The rate of change in the system exceeds its capacity to adapt. The system is not fragmented (still connected) and not condensed (still distributed). But the environment is changing faster than the system’s internal processes can respond. The system becomes incoherent — its parts are still connected, but they are no longer coordinated.
An ecosystem subjected to rapid climate change faster than species can migrate or adapt. A financial system subjected to algorithmic trading faster than regulatory frameworks can modulate. An institutional framework disrupted faster than its capacity to absorb the disruption.
If these are exhaustive — if there is no fourth way a colimit can fail — then the taxonomy is complete. This is a conjecture, not yet a theorem.
Empirical results
The framework was tested against three real-world networks from different domains. The analysis code is reproducible from the GitHub repository.
The food web structural
A food web is a constitutional system. A predator-prey relationship IS a constitutional event — the predator uses the prey’s output to maintain its own constitution. Neither is complete without the other. The relationship, not the species, is the constitutional unit.
Data: The published paper uses three food web datasets from the KONECT repository (Florida Bay dry/wet seasons, Maayan food web). Preliminary analyses also examined the Chesapeake Bay food web (N=33) and Little Rock Lake food web (Martinez 1991, N=183, 2,494 edges).
Method: Targeted hub removal (Albert et al. 2000) vs. random node removal. The framework predicts that hub removal should produce a fragmentation signature distinct from random removal — measured via Gini coefficient trajectories of betweenness centrality under progressive node removal.
Result: The food web’s fragmentation trajectory under targeted hub removal is significantly different from configuration model nulls (p < 0.01) — networks with the same degree sequence but randomized connections. The structural signature survives degree-sequence control. This is the strongest empirical result: the food web’s topology actively resists what its own degree sequence would predict.
Caveat: Power-law betweenness centrality in food webs is a known phenomenon in network science (Dunne et al. 2002). The framework retrodicts this pattern, not predicts it. A niche model null (Williams & Martinez 2000) that preserves trophic structure would be needed to distinguish constitutional organization from trophic organization. This test has not been performed.
The cancer network structural
Data: Cancer driver gene interactions (COSMIC Tier 1 via STRING v12, 411 nodes, 574 edges).
Prediction: Constitutional condensation should produce betweenness centrality more concentrated than comparable random networks.
Result: Negative. The cancer condensation trajectory is not distinguishable from a configuration model null (p = 0.58). The concentrated topology is a degree-sequence artifact — the network was built by querying known cancer driver genes, which are by construction high-degree hub nodes. The seed-gene query creates the concentrated topology, not the biology.
A rigorous test would use paired tumor/normal tissue-specific interactome networks across TCGA cancer types, not seed-gene networks. This test has not been performed. The cancer result is an honest negative — the framework’s condensation prediction is not supported by the current data.
The connectome structural
Data: The published paper uses the C. elegans connectome (White et al. 1986, 297 neurons, 2,148 connections). Preliminary analyses also examined human fMRI connectome data (Power et al. 2011 parcellation, 264 ROIs).
Result: The connectome shows structural baseline concentration and greater robustness than configuration model nulls. The C. elegans result is positive — the neural network’s topology differs from same-degree random networks in the direction the framework predicts.
The bridge
If the failure modes are structural — built into the mathematics of how parts constitute wholes — then they should appear wherever constitution appears. The framework predicts the same three modes at every scale, with the same topological signatures.
The political work on this site documents what looks like one of these modes. The Briefing assembles the public record of a congressional representative — votes, FEC filings, financial disclosures — and the pattern that emerges resembles condensation. Campaign funding concentrates into fewer, larger sources. Legislative attention concentrates toward the industries that fund it. The district’s needs — manufacturing, agriculture, healthcare — receive no enacted legislation.
This is a conceptual analogy, not a demonstrated result. No network analysis — no betweenness centrality, no Gini coefficient, no null model comparison — has been performed on the political data. The FEC filing patterns are consistent with the condensation concept, but the formal tools have not been applied. Applying them is future work.
The framework provides the formal language. The journalism provides the observations. Whether they connect formally, not just conceptually, remains to be demonstrated.
The formal version
The sustainability criterion. A constitutional diagram is sustainable if and only if it produces a valid, non-trivial colimit whose cocone does not factor through a single node. The three failure modes are the three ways this criterion can fail:
- Fragmentation: The diagram becomes disconnected. No single colimit exists for a disconnected diagram — it decomposes into independent components, each with its own colimit. The whole is lost.
- Condensation: The cocone factors through a single node. The colimit exists but is trivial — one node absorbs all signal flow. The system is formally connected but functionally collapsed.
- Rate-exceeds-modulation: The morphisms in the diagram change faster than the colimit can be recomputed. The colimit that was valid at time t is invalid at time t+Δt because the underlying diagram has changed. The system is connected and distributed but incoherent — its colimit is perpetually out of date.
Null model methodology. The food web result uses a configuration model null — a random network with the exact same degree sequence as the real network. This controls for the possibility that the observed topology is an artifact of degree distribution. The food web’s structural signature survives this control (p < 0.01 slope difference), establishing that the constitutional organization is genuinely structural. The simpler Erdos-Renyi null (same N and edge density, random degree distribution) is insufficient — it produces the opposite conclusion.
Cross-domain applicability. The four sustainability conditions — structural (compatible signals), topological (non-degenerate colimit), dynamical (rate within modulation capacity), and reachability (the current configuration is reachable from the system’s history) — apply identically at every grain. At the quantum grain, the signal is electromagnetic. At the ecological grain, the signal is biomass transfer. At the political grain, the signal is money and legislative attention. The conditions do not change. The materials do.
All analysis code, datasets, and null model comparisons are available in the GitHub repository.
Sources
Food web data -- KONECT repository (Florida Bay, Maayan); Martinez, N.D. (1991), “Artifacts or attributes? Effects of resolution on the Little Rock Lake food web,” Ecological Monographs 61(4), 367–392. Hub removal methodology -- Albert, R., Jeong, H. & Barabási, A.-L. (2000), “Error and attack tolerance of complex networks,” Nature 406, 378–382. Power-law methodology -- Clauset, A., Shalizi, C.R. & Newman, M.E.J. (2009), “Power-law distributions in empirical data,” SIAM Review 51(4), 661–703. Cancer network data -- COSMIC Cancer Gene Census: Tate, J.G. et al. (2019), Nucleic Acids Res. 47(D1), D941–D947; STRING database: Szklarczyk, D. et al. (2023), Nucleic Acids Res. 51(D1), D638–D646. Connectome -- White, J.G. et al. (1986), “The structure of the nervous system of the nematode Caenorhabditis elegans,” Phil. Trans. R. Soc. B 314, 1–340; Power, J.D. et al. (2011), Neuron 72(4), 665–678. Niche model -- Williams, R.J. & Martinez, N.D. (2000), Nature 404, 180–183. Food web network robustness -- Dunne, J.A., Williams, R.J. & Martinez, N.D. (2002), Ecology Letters 5(4), 558–567.