Why it is a question
The owner, at U14, after the scale-separation synthesis (CLAIM-478): “does this elucidate other work we should account for? maybe in the neighborhood of dynamics? bifurcation/catastrophe theory? far from equilibrium dynamics? pattern formation?”
A14 separated routes by which structure appears, each with a held or registered source:
- Bifurcation. In ẋ = μx − x³ the set of stable states changes at μ = 0. A14: “Possibility formation: Which organizations become dynamically admissible and stable? Realization selection: Which of those organizations does the system actually occupy?”
- Instability and pattern. A homogeneous state stable at k = 0 can be unstable at a finite wavelength, so “Interactions need not merely constrain the configurations of preexisting parts. They can generate the very spatial distinctions that later allow parts to be recognized” (Turing, LIT-535; Cross and Hohenberg, LIT-527).
- Nucleation. “In classical nucleation, the initial state remains locally stable, and a sufficiently large finite fluctuation is required to cross an activation barrier”, which differs from instability in onset, waiting times and the role of seeds (Langer, LIT-1664; Vekilov, LIT-919, read in NOTE-708).
- Hysteresis and loss. In the cusp, which form is present can depend on history, and a stable state can disappear (Thom, named). Tipping can be bifurcation-, noise- or rate-induced, and in the last “A persistent organization may be destroyed not because its equilibrium ceases to exist, but because it cannot reorganize fast enough” (Ashwin et al., LIT-1654).
- Localized structures. Fronts pinned to a pattern’s phase give a bounded patch that grows by cells while keeping its organization, “a persistent dynamical individual without being a fixed collection of particles” (Knobloch, LIT-1642).
And it separated three achievements: “The appearance of a new mode is one event; its stabilization into a persistent organized pattern is another”. A linear instability says which perturbations grow, not which pattern survives. That is settled by nonlinear saturation, wavelength selection, defects and boundaries. A third achievement is the constitutive role: “An attracting pattern does not automatically become an independent individual; it may simply be a useful macro-description.”
Why it is open here
The record’s individuation claims are about forms already present. CLAIM-157 says when a combined form is one system and that “it lasts as long as that block persists”. CLAIM-405 says the verdict is fixed once its criterion is. QUESTION-035 asks how long a form lasts, not how it begins. A14’s routes bear on both ends:
- A transient can leave a regime whose every eigenvalue is stable. Non-normal growth does so (THEORY-724; Trefethen et al., LIT-932, read).
- A form can be lost while its frozen-parameter attractor survives (rate-induced tipping).
- A form can be present without a gap, by exact closure (CLAIM-394) or renormalization.
Each of these is a way for a form’s persistence to come apart from what an instantaneous criterion reads. The record reads Cross and Hohenberg only in part: NOTE-418 is Skimmed, and its technical core on amplitude and phase equations, defects and pattern selection, the part that answers “which pattern survives”, was not read.
What would count as an answer
For a stated system with a form the record individuates, the transition by which the form arose, the mechanism that stabilized it, and the change that would end it, each shown separately. Then a case in which two of the three come apart, a form created but not stabilized or stabilized without any constitutive role, which would show that they are separate achievements and not one described three ways.
Note of 2026-10-27 (later): two readings
- Toner and Tu 1995 is a worked case of creation and stabilization coming apart (NOTE-1228). The αv − β|v|²v terms of their equation (1) are the normal form ẋ = μx − x³: a local speed appears when α changes sign. In two dimensions that bifurcation alone gives no persistent order; the linearized theory loses long-range order for d ≤ 2, as the equilibrium XY model does (p. 4). A separate mechanism, the convective nonlinearity, stabilizes it (pp. 5–9; THEORY-1095, Proposed). The paper says nothing about a constitutive role; its “birds” stay in scare quotes, which bears out A14’s caution.
- Ashwin et al. 2012 corrects the question’s gloss of rate-induced tipping (NOTE-1215; THEORY-1079). The attractor keeps its existence, stability and basin at every frozen parameter (§3.2.2). What bifurcates is “the ability to track the QSE” (p. 11), and the critical rate depends on the distance to the threshold as much as on relaxation. But “destroyed” is too strong: just below the critical rate the climate model “returns to the QSE after a transient” (fig. 7a), and in §3.3 the equilibrium is unique and globally stable. A form is lost for good only where something lies past the threshold. And rate can preserve a form: “B-tipping may be suppressed for fast enough variation of parameters” (p. 19).
Note of 2026-10-27 (U15–U23): three routes, and composition
A23 §V sorted how interconnection produces structure into routes that bear on this question’s first half:
- Suppression. Dissipative coupling damps relative differences and leaves collective modes, which is consensus, synchronization and slow hydrodynamic modes (CLAIM-481).
- Amplification. An instability selects and grows differences, which is Turing patterns, bifurcation and nucleation across a barrier. These are the routes this question already lists.
- Frustration. Constraints that are satisfiable pairwise but not around a cycle leave a residual the system has to distribute. Cycle holonomy in gauge-coupled phase models (Torres-Hugas et al., LIT-1668, unread) and the cycle-space residual of ½‖Dx − b‖² are the exchange’s cases (CLAIM-471).
A23: “We should not mistake consensus for the general mechanism of individuation.” Frustration is a route the question did not name. A form can be stabilized by organizing differences that cannot all be reconciled, not only by eliminating them.
Composition is a fourth way a form gets a role, and it bears on the question’s “constitutive role”. A composite whose interior relaxes fast presents its boundary with a constraint that no part has (CLAIM-480). That is a role acquired at the interface, without the composite being an individual.
Note of 2026-10-27 (evening): the readings of the U15–U23 works
Torres-Hugas et al. (NOTE-1218; THEORY-1087), read, narrow the frustration route the earlier note took from A23.
- In the paper, frustration is a structural fact, the residual left after optimizing over states. It does not stabilize or organize a form.
- In their motif, the nonuniform locked state is forced by unequal degrees, not by holonomy.
- The remote-synchronized pattern grows smoothly out of complete synchrony, so no transition creates it.
- It is unmade when a transverse eigenvalue crosses zero at αc ≃ 1.0164, before the pentagon’s mismatch reaches its maximum, π, at α = π/3.
It is still a stated system with the unmaking transition named, apart from the geometry of its constraints. That is half of what an answer here asks for. It is not a case of frustration making a form.
Note of 2026-10-31: Simondon’s metastability as his organization-studies readers use it
This question’s nucleation route, “from a locally stable state” across “a finite barrier”, is the physical sense of metastability. Simondon’s individuation, read today through Iliadis 2013 and two organization papers, is the resolution of a metastable system. The connection looks direct, and in these readings it is not.
- Their “metastable” is not this question’s. Bencherki and Iliadis gloss metastable systems as those “where energy is not homogeneously distributed” (LIT-1922, ms. p. 7). Bencherki, Brummans and Vézy call crystals and bureaucracies “organized relatively homogeneously, and … therefore relatively stable”, and sand dunes and start-ups “less stable” (LIT-1902, p. 1355). In nucleation the textbook metastable state is a homogeneous supersaturated phase, and the crystal is what resolves it. The readings make metastability a matter of heterogeneity, and they drop the barrier, the waiting time and the seed, the features this question uses to tell nucleation from instability. Iliadis keeps only “Information passes from a state of ‘metastability to stability’” (LIT-1911, p. 11). So these readings are not evidence for the nucleation route or illustrations of it. Simondon’s own crystal, the paradigm case of a germ seeding a metastable solution, is in LIT-1648, Unavailable. A reading of it would settle whether his concept is the dynamical one.
- What does bear: the form keeps a residue. “an individuation process does not ‘use up’ or ‘consume’ the preindividual it integrates” (LIT-1902, p. 1355). An individual “does not exhaust the tensions that allowed its constitution” (Simondon 1989, in LIT-1901, p. 782). On this account a form is made while leaving an unresolved remainder, which can start the next individuation. That is the frustration route the note of 2026-10-27 (U15–U23) added, “organizing differences that cannot all be reconciled”, in other words. The organizational versions say so in their own terms. “Coexisting meanings of the same action indicate that several individuations are ongoing”, against Weick’s reduction of equivocality (LIT-1902, p. 1358; THEORY-1278). Massumi’s micropolitics aims at “a maximally inclusive co-patterning of contrasting tendencies”, an “affirmation of discord” and not its resolution (LIT-1910, p. 64; THEORY-1283). A23: “We should not mistake consensus for the general mechanism of individuation.” Same structure only: none of these readings has a dynamics, and Torres-Hugas et al. found that frustration in their model does not make a form (the evening note above).
- “Pass away” has a name there and no dynamics. Disindividuation is “the experience of needing to reorganize in order to deal with a preindividual element that resists incorporation” (LIT-1902, p. 1356), and “moments of disindividuation may constitute important inflection points that call for creativity” (p. 1365). The readings’ “dense” fields, which “cannot be reorganized easily when they meet other fields” (p. 1356), suggest a deep basin or hysteresis of the kind CASE-111 models. Nothing in the readings says what the transition is, so these are names for this question’s second clause, not answers to it.
The question is unchanged and stays Open.
Note of 2026-11-01: stabilization and entrenchment kept apart, from evolutionary transitions
Griesemer and Shavit (LIT-188, §2; NOTE-173, re-read 2026-11-01) separate two of this question’s three achievements for social groups, without a dynamical model. A connected collection can dissolve, persist on outside scaffolding, or “stabilize the network of interactions” through self-organizing processes. Only then, in the paper’s words, “stabilization of the network affords the opportunity for evolution to entrench particular patterns of interactions”, making some of them “core”, the ones whose perturbation is “catastrophic to the whole system” (§2). Stabilization is one event. Coming to be core, which is a constitutive role in the sense this question asks about, is another and later one. The paper says the result is “highly contingent, even if contingently irreversible”, and that whether a given conexus will get there is “very difficult to predict”. It names no dynamical route among those listed here. Its entrenchment is an evolutionary process of building dependents on what has stabilized, which none of the four routes names. CLAIM-531 adds that entrenchment does not by itself give closure.
No status, title or warrant changes.