nucleation

CLAIM-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

4 min read

Properties9
StatusProposed
Rolecounter
Warrantdemonstrated
Tagsphilosophy-of-science, individuation, metaphysics
Linepragmatic-transport
Rests onCLAIM-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-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, CLAIM-046 — Different hidden-state models can produce identical observations, so pragmatic identity and fidelity should be defined on observable responses before any equivalence of internal states is assumed
Objects toCLAIM-139 — A communicative kind's identity is grounded in its relational organization rather than in any expression or encoding; statistical methods recover aspects of that organization from restricted observations, translation transports it, and fidelity is which independently identifiable, decision-relevant relational distinctions persist, CLAIM-142 — Two communicative realizations can agree on every probe in a restricted family without being structurally identical, so observational equivalence is relative to the probes, and only independently justified probes evaluated on held-out cases are evidence that structure was preserved
Objected byCLAIM-217 — For a communicative kind, the standard of sufficiency is the uptake of competent participants who know the standings and norms in play: a defeasible standard, overridden only where the act's other constituents diverge from the judges' verdicts, so a probe family is tested against it, and a kind outruns the families coarser than that standard and the judges only through those constituents
UsesTERM-040 — communicative kind, as a repeatable relational organization that numerically distinct events realize, distinct from a token and from a structural position, TERM-014 — communicative object, as an observational equivalence class

Graph View

The objection

The record says a kind can outrun the probes that observe it. TERM-040 says a kind is “Not an observational class” (TERM-014), and quotes A218 Ch7’s core claim for it: “A structural kind may be real while remaining indistinguishable under an insufficient family of empirical probes.” CLAIM-135: “A kind can be real and still unseparated by an insufficient family of probes.”

Yet the defeat conditions take human judges as the ground truth for kinds. CLAIM-139: “Human judgements of shared communicative kind across translations, recastings and media are predicted as well by …”. CLAIM-137: two realizations “still reliably judged different acts”. A judge’s verdict is a response of an interpreter to a realization, so the judges are a probe family like any other.

The dilemma.

  • If a kind can outrun every probe family, it can outrun the judges. Judges may call two realizations of one kind different (they attend to surface), or two of different kinds the same (their probing is insufficient). Then predicting their verdicts is neither necessary nor sufficient for tracking kinds, and the verdicts can neither confirm nor defeat what CLAIM-139 says about kinds, unless a further premise says the judges’ family is sufficient. CLAIM-142’s rule of evidence forbids assuming that of any family.
  • If some family, the judges’ or another, cannot be outrun, then in every test the kind’s extension is the observational class relative to that family (TERM-014). A realist can say that coinciding in extension is not being identical. But the difference then lies only in the ontological clause, which CLAIM-139’s “What it does not say” grants the programme does not test, and TERM-040’s distinction does no work in any test.

Either way the dilemma reaches the epistemic and fidelity clauses that CLAIM-139 says are tested: “So the empirical programme tests the epistemic, transport and fidelity clauses, not the ontological one.”

Two defeat conditions misfire.

  • CLAIM-142 is defeated if “a finite probe family fixed in advance separates every pair of realizations that judges distinguish, on held-out realizations”. Let the family contain one probe: ask judges whether the two realizations are the same act. That family is finite, can be fixed in advance, and separates every pair the judges distinguish, held out or not. Nothing in the defeat condition excludes it, so the condition is met trivially.
  • CLAIM-139’s third disjunct is “no relational structure beyond the observed response profiles is needed to explain those judgements”. CLAIM-046’s rule is that identity and fidelity “should be defined on observable responses”, so every model the programme fits is a model of response profiles, and a relational structure estimated from them predicts nothing a model of the same profiles cannot. CLAIM-150 grants that an ontology of intrinsically individuated objects “can reproduce the same predictions”. Read as a claim about prediction, then, the disjunct is always met and the thesis is always defeated. Read as a claim about explanation, no data decide it.

What it does not say

  • It does not say kinds are unreal, or that the realist reading is wrong.
  • It is not CLAIM-150’s point that the data underdetermine the ontology. That point concerns the ontological clause. This one concerns the clauses the programme claims to test.
  • It does not say CLAIM-142’s relativity of observational equivalence to probes is wrong. It uses it.

What would answer it

  • A statement of what, besides the judges’ verdicts, fixes a kind’s extension in the tests, with the judges placed explicitly as one probe family among others and their verdicts excluded from the families that CLAIM-142’s defeat condition quantifies over.
  • CLAIM-139’s third disjunct rewritten as an observable comparison: for instance, a relational model against a model of the response profiles alone, compared on transfer to held-out realizations, media or languages.

Where it stands

On 2026-10-10 the record replied. CLAIM-217 answers the first horn: no claim of the record says a kind outruns every family, only an insufficient one, and for a communicative kind the standard of sufficiency is competent participants’ uptake, so the judges are the criterion a probe family is tested against, not one probe among others. TERM-040 version 2 corrects the gloss the second horn used: an equivalence class relative to a family cannot be unseparated by that same family, but can be by a coarser one. CLAIM-142 version 2 excludes the trivial probe from its condition. CLAIM-139’s third disjunct is replaced only in the revised thesis CLAIM-204, so the objection stays open until the owner decides on it.

Note of 2026-10-29: the owner decided

CLAIM-204 superseded CLAIM-139 on the owner’s delegated ruling of 2026-10-29, so the thesis the record holds open no longer has the third disjunct this objection showed always met or undecidable. The objection stays pointed at CLAIM-139, where it holds, and its status stays its credence (ADR-059).

Backlinks