nucleation

CLAIM-150: An ontology of intrinsically individuated objects with relational properties can reproduce the categorical and operational identifications, and the predictions, of a relational one, so identification underdetermines constitution

6 min read

Properties12
StatusActive
Version2
Roledisclaimed
Warrantinterpretive
Tagsmetaphysics, philosophy-of-science
Linepragmatic-transport
GroundsTHEORY-032 — The Yoneda lemma determines an object only up to isomorphism, from its whole hom-functor, among objects the category already has; as a formal version of "an object is determined by its relations" it presupposes the relata, LIT-222 — No categorial support for radical ontic structural realism
SupportsCLAIM-206 — If a communicative kind can outrun every probe, human same-kind judgements, themselves one probe family, can neither confirm nor defeat the October thesis; if it cannot, the kind does in the tests only what an observational class does, which the record's term for a kind says it is not
UndercutsARG-023 — The Yoneda lemma shows that relations constitute what a communicative act is, ARG-038 — A character's classificatory sufficiency shows that relations constitute what a communicative act is
ComplementsCLAIM-151 — Which communicative act an utterance performs is partly constituted by its relations to participant roles, norms and expected uptake, so a transformation that changes those relations can change the act and not only its realization, CLAIM-137 — A communicative kind is identified by an invariant that does not depend on any one realization, as a group character identifies a representation up to isomorphism, once one symmetry group is identified across the realizations by its action on the situation and not by labels of a medium, so that invariant has classificatory priority over any particular realization, CLAIM-135 — Recovering latent structure from relational data is three achievements, reconstruction, identification up to stated ambiguities, and constitution of a communicative kind, and identifiability theory reaches the second at most, CLAIM-450 — Three standards for a communicative distinction come apart: statistical distinguishability of an interpreter's response profiles, measured locally by the Fisher metric and never raised by a parameter-independent garbling; usability by a bounded receiver, which a transformation can raise; and pragmatic efficacy in the act; and where response models are singular, as Watanabe's theory treats, a missing response can come from an absent attempt or an unresponsive receiver alike, and only intervention or later dynamics separate them
UsesTERM-038 — descriptive, causal and constitutive relations of a communicative act
Rendered intodocs/arguments-by-conclusion.md

Graph View

The claim

Three things can be said of a communicative act’s relations, and the first two do not give the third.

  • Categorical identification. In a specified category, an object’s full hom-functor fixes it up to isomorphism. That is the Yoneda lemma.
  • Operational identification. A chosen family of probes, interventions or predictive models tells some realizations apart from others.
  • Constitution. Certain relations partly make the act the kind of act it is (CLAIM-151).

A191 §4 states the split: “Yoneda establishes the first. Our experiments investigate the second. Moderate communicative structural realism asserts the third. Neither of the first two logically entails the third.” A187 had made the first half of the point at U52: “Yoneda establishes the first. It does not independently establish the second.” It pairs with the distinction between descriptive, causal and constitutive relations (TERM-038), which A191 §11 draws and A203 §16 repeats: “Causal intervention results can support a constitutive account, but they do not by themselves prove metaphysical constitution.”

Characters. A198 §VI separates three senses of “essential”: “Representational independence … Classificatory sufficiency … Ontological constitution”. “Character theory can establish the first, and under suitable mathematical assumptions the second. Your non-eliminative OSR makes a claim about the third. It would be a mistake to infer the third automatically from the first two.” So the owner’s character analogy, at its strongest, gives classificatory priority (CLAIM-137) and not constitution.

Predictive success. A203 §35 separates an ontological, a mathematical and an empirical thesis: “These claims support one another, but they are not logically equivalent. A successful predictive model does not prove OSR. A mathematical character does not establish metaphysical constitution.” A214 §13 gives the reason: “A traditional ontology of intrinsically individuated objects plus relational properties might reproduce the same statistical predictions.” So the data underdetermine the ontology. A214 then splits the argument into two parts. The “Conceptual argument” is that some acts “cannot be adequately specified without reference to constitutive social and normative relationships”. The “Empirical argument” is that relational models capture stable, discriminating, predictive properties of those kinds. Success in the second would support a real-patterns reading of communicative kinds and make a non-eliminative structuralism attractive. Only the first bears on constitution. A218 §13: “The central ontological thesis must be defended through argument concerning social acts and supported by robust explanatory evidence, not derived automatically from algebraic or statistical models.”

The stronger form. The record’s reading of the lemma goes further than the exchange’s (THEORY-032, from LIT-221 read in NOTE-195). The Yoneda lemma fixes an object only up to isomorphism, among objects the category already has. y(c) is indexed by every object of C, and the conclusion is an isomorphism between two of them. So as a formal version of “an object is determined by its relations” it presupposes the relata. It cannot establish constitution, and it gives elimination of the relata no support at all. Lam and Wüthrich reach the matching conclusion for category theory generally (LIT-222; THEORY-038): morphisms relate objects the category presupposes. That fits the non-eliminative choice the owner made at U54.

What it does not say

  • That the ontological thesis is false or idle. It says only that the mathematics and the data underdetermine it.
  • That the three senses are unrelated. A198 §VI: “we can formulate a coherent metaphysical position in which they reinforce one another.”
  • What would establish constitution. A214’s conceptual argument is named, not given. The question of when relational structure warrants treating a category as a real kind is QUESTION-028.

A187 §8 proposed a narrower disclaimer for filing, “Observational relational identity does not entail radical OSR”. It is merged here rather than filed separately: the manuscript’s relational account does not commit it to ontic structural realism, and Yoneda does not supply the commitment.

Note of 2026-10-14: a proved instance in networks

THEORY-243, filed from LIT-961 and LIT-968 (read in NOTE-809 and NOTE-782), is a proved instance of this claim’s premise that operational identification does not give the third sense. In a constructed three-unit network, interchange interventions on a fixed set of variables are satisfied equally by alignments related by a change of basis, so success under those interventions does not even fix which direction mediates; it fixes it only relative to a chosen decomposition into units. That is weaker than constitution and more basic. It is shown exactly for one constructed network.

Note of 2026-10-27: the guard restated, and its form for specifications

The exchange of 2026-10-27 restated this claim without citing it, and then cited it as “Nucleation’s strongest existing philosophical safeguard: identification is not constitution” (A10 §5). Its forms there: A7’s “Demonstrable” against “Philosophical hypothesis” for luria; A10 §5’s three levels “that should never be conflated”, formal (which configurations satisfy a specification), dynamical (which realized constraints affect a system) and ontological (whether the organization constitutes an individual), with “Success at one level provides potential evidence for another, not an automatic entailment”; and A9 §1’s form for specifications, “Mathematical satisfaction is not physical production.” The three claims filed from the exchange that go past the guard (CLAIM-477, CLAIM-476, CLAIM-472) each name it among what they do not say. Nothing here changes.

Note of 2026-10-27: the edge says what the claim uses

The status review of this date (ADR-057, ADR-058) removed THEORY-038 from grounds. It is cited for the categorial point it shares with THEORY-032. That point, and Lam and Wüthrich’s matching conclusion, are carried by THEORY-032 and LIT-222, both already grounds. The body citation stays.

Note of 2026-10-28: split under ADR-061

The old title named three routes that do not show constitution: the Yoneda lemma, the invariance of a character, and predictive or interventional success. The first two are inferences between entries, and they are now arguments to CLAIM-151, both Rejected: ARG-023 from THEORY-032 and LIT-222, and ARG-038 from CLAIM-137. The third has no premise entry, since no experiment has been run, and it stays here. This claim keeps its reason, which is about the subject matter: an ontology of intrinsically individuated objects can reproduce the identifications and the predictions, so they underdetermine constitution. It undercuts both arguments, and it is still the guard CLAIM-206 rests on and CLAIM-472, CLAIM-476 and CLAIM-477 name. It keeps THEORY-032 and LIT-222 among its grounds, since the lemma presupposing its relata is why an object ontology reproduces the categorical identification.

Note of 2026-11-01: a proved instance, for mechanisms

Baumgartner & Gebharter (2016) prove a case of this claim’s pattern (THEORY-1288). If a whole supervenes on its parts and constitution is not causal, then every constitutive model of a mechanism has a pure common-cause model that fits all the manipulation data equally well, even under ideal conditions. Interventional evidence underdetermines constitution, and only a further inference settles it. No status, title or warrant changes.

Backlinks