nucleation

CLAIM-194: A gap or coupling criterion cannot tell a boundary a system produces from one imposed on it: a membrane the cell makes and a beam joining two pendulums pass alike, so the block test cannot by itself supply the individuality minimal agency needs

4 min read

Properties11
StatusActive
Version4
Rolecounter
Warrantinterpretive
Tagsindividuation, agency, complex-systems
Linedistributed-agency
GroundsTHEORY-073 — In Friston's 2013 formulation and the applications the record holds, a Markov blanket does not individuate a system: where it falls is fixed by modelling choices made before it is traced (the coupling graph, the internal set and the time at which steady state is assumed), and a blanket read from a fixed graph is the same whether the system produced the boundary or it was imposed, THEORY-066 — A Markov-blanket partition is defined relative to a chosen internal set, so a graph has one around almost any set of nodes, and the formalism alone does not say which set is the system, THEORY-1119 — Barandiaran, Di Paolo and Rohde propose that minimal agency requires individuality, interactional asymmetry and normativity set by the system's own organisation, each necessary and together, they say, sufficient; that individuality without the other two is not agency; and that a system optimizing an externally fixed function is not a model of agency
Objects toCLAIM-157 — Relative to a stated target, a combined form is one system while its joint dynamics are nearly decomposable with the form as a block whose gap beats the Occam factor, and it lasts as long as that block persists; the unit is the gap-isolated block, not any member, CLAIM-174 — A candidate factor is an agent at a level when it is one system there, when a goal-directed description beats a device description by more than the Occam cost, and when its policy would adapt if its actions' consequences changed; it is a willed agent when, further, a will-organization over several regimes describes it better than none, and what passes that is the jurisdiction of a will-organization
ComplementsCLAIM-367 — Whether a training-time regulatory condition belongs to a network depends on which unit is individuated for the training dynamics and whose jurisdiction governs the updates; in conventional training most of that governance falls to the optimizer–model–data system, so a condition of training is not thereby a condition of the network, CLAIM-414 — Matched ordinary trajectories can conceal differences in an interpretation's efficacy, in the allocation of that efficacy, and in the maintenance of its enabling constraints: in two finite constructions, systems that agree under ordinary operation differ in each, and a specified intervention exposes the difference that passive observation of the matched trajectories does not
Corroborated byTHEORY-072 — Organisms differ from dissipative structures such as a candle flame not by far-from-equilibrium self-maintenance, nor by circular self-maintenance alone, but by organisational differentiation: several distinct constraints that produce one another at different time scales
Cited as “complements” byCLAIM-522 — Recovery under interventions on state cannot show where maintenance is realized: an outside controller that measures the constraint capacities and supplies the internal production law as state feedback yields the same trajectory under every perturbation of state, damage and recovery included, even with every capacity observed exactly; only interventions on the production mechanisms, relative to a declared boundary, separate the two, CLAIM-528 — That a mechanism realizes a capacity can mean that the capacity is produced through it over time or that the capacity consists in it at a time, and interventions show only the first: a pathway whose disabling changes the capacity after a delay is a cause, can be intervened on surgically, and can be located relative to a declared boundary, whereas every intervention on a capacity that consists in its parts also changes some part along another path, so that relation is underdetermined by intervention data and is inferred by abduction or from simultaneity and settled parthood; and which of the two a given dependence is depends on the time resolution at which a description treats the parts' response as instantaneous, CLAIM-530 — Where maintenance is realized is a question of constitutive relevance only through its parthood condition: an internal producer and an outside controller supplying the same production law as state feedback meet mutual manipulability's two manipulability conditions identically, since disabling or stimulating either changes the maintained capacities and intervening on the capacities changes the activity of either, so what puts one inside and the other outside is parthood, a boundary the experiments presuppose and do not find, CLAIM-531 — Generative entrenchment does not by itself give organizational closure: an entrenched core, which later interactions depend on so that perturbing it is catastrophic and perturbing them tolerable, can sit at the root of an acyclic dependency order, and where interactions turn over such an order persists only while something outside supplies its core; an organization that is both entrenched and closed has a core containing a cycle of mutual maintenance, each interaction on which is itself core unless its share of the maintenance is redundant

Graph View

The claim

CLAIM-157 makes a combined form one system, relative to a target, when its members form a nearly decomposable block whose gap beats the Occam factor. CLAIM-174 takes “one system there” as its first condition for agency. Both read the boundary off the strength of interaction inside a candidate block against interaction across its edge.

THEORY-073 says of the Markov blanket that “the formalism also gives the same description to a membrane the cell makes and a beam nobody’s states make”. THEORY-066 gives the formal reason: the partition “takes the internal set as an input” and “never outputs the internal set”. The gap criterion is better off than the blanket on the second point, since it compares candidate partitions and does not take one as given. It is not better off on the first. Two pendulums joined by a beam clear the threshold for a target such as the next swing, and so does a cell. The criterion has no term for who made the coupling.

The record holds two accounts that put the line there:

  • THEORY-072: an organism differs from a candle flame by “several distinct constraints that produce one another at different time scales”. That is production, not coupling strength.
  • THEORY-1119: minimal agency needs individuality in which “the system distinguishes itself from its environment, and an observer does not do it for the system”.

So the block test can find a system in the sense of a unit of description, but the individuality THEORY-1119 requires is a further condition. When CLAIM-174 uses the block test as its first condition, it admits jointly coupled forms whose boundary is maintained from outside.

What it does not say

  • That the block criterion is wrong as an answer to QUESTION-035. As a test of when a description should name a unit, the counter leaves it standing. It bears on CLAIM-174’s use of it as the individuality condition of agency.
  • That observer-relativity is the problem. CLAIM-174 already concedes that its tests are relative to level and target. This counter holds even with the target fixed.
  • That production is measurable where coupling is. THEORY-072 is Proposed, and its own promote_when asks for a flame and a protocell to be analysed the same way. Neither has been done.

What would answer it

Either a case showing that, for a natural target, imposed boundaries systematically fail the gap test while produced ones pass it, so the criterion tracks production after all. Or a reply that CLAIM-174’s later conditions (adaptation of the policy, governance over regimes) already exclude externally maintained blocks, so the first condition need not.

Note of 2026-10-21: regulation of exchange

A9 §3 of the exchange of 2026-10-20 separated predictive closure from organizational closure and proposed that “A living or agentive boundary may be distinguished by its regulation of exchange, rather than by the absence of exchange”. That is THEORY-1119’s interactional asymmetry, which A9 does not cite. A8 §1 adds the informational side of this claim’s point: since agents are coupled to their environments, exact closure “cannot straightforwardly serve as a necessary condition for agency”, and since noise is closed, it is not sufficient either. So neither a coupling criterion (this claim) nor a closure criterion (THEORY-820) tells a produced boundary from an imposed one. The two closures are separated in TERM-068 and TERM-070.

Note of 2026-11-01: the mechanists’ discriminator is a coupling notion too

Craver and Bechtel (LIT-1928, NOTE-1491) concede that interlevel experiments show only prima facie that an item is a component. They say the difference from a background condition is whether it is “interactively integrated into the system” (p. 553, n. 10). That is a coupling notion, so it inherits this claim’s point that it has no term for who made the coupling. An outside controller coupled to the system in both directions passes their four experiments exactly as an internal producer does (THEORY-1287, CLAIM-530). Same structure, not shown same mechanism: this claim is about observational statistics, and theirs is about interventions. Their framework draws a mechanism’s boundary by its parthood condition, which takes the boundary as an input, as this claim says a gap criterion must.

No status, title or warrant changes.

Backlinks