Why it is a question
A218’s RQ7, “Ontological interpretation”, asks it in these words. It has been asked four times before.
- A187 §8 proposed it for filing: “When is a communicative structure a real pattern?” A187 §4 had offered an answer in outline: “cross-context predictive and explanatory success provides grounds for treating certain patterns as more than arbitrary descriptions.” TV Tropes was its example.
- A191 §7 gave four criteria: robust identification, cross-realization generalization, causal relevance, and nonredundant explanatory value (“A structural representation improves prediction or explanation beyond simpler models based on surface features, topic, or generic contextual correlations”). It added: “None of these independently establishes a metaphysical ontology.” The criteria restate the manuscript’s success criterion (CLAIM-078), that the theory succeeds if its measures predict changes of communicative identity that text similarity misses.
- A203’s crystallized argument (§X, its last paragraph) ends on it: “whether their predictive robustness warrants treating them as real, effective communicative kinds”.
- A214 §13: success “would support a real-patterns interpretation of communicative kinds”, but “would not deductively establish that relations are metaphysically prior to all objects”.
None of these says what would settle it. The exchange’s guard (CLAIM-150) says only that predictive success does not establish constitution.
It refines QUESTION-021 and reopens it. QUESTION-021 asked whether there is a frame-independent act that translations approximate or whether the act is constituted by its frame-relative realizations, and A24 refined it away into QUESTION-001. This question takes a third answer for granted, a repeatable communicative kind (TERM-040), and asks when such a kind is real.
What the record holds toward an answer
The record already holds the real-patterns readings the exchange cites as new, and they locate the difficulty.
- Compressibility over-admits. Dennett’s criterion (LIT-220) makes a pattern real “if there is a description of the data that is more efficient than the bit map”. THEORY-036 reads it as admitting almost every non-random pattern. A criterion of explanatory compression alone would make every stable TV Tropes page a real kind.
- Projectibility does the work. Ladyman 2024 (LIT-219) replaces compressibility with indispensability to projectible generalizations, relative to perspective and scale. Ladyman and Lorenzetti’s effective ontic structural realism (LIT-184) is the read statement of “effective” kinds. THEORY-036 holds that these added clauses do the excluding, and that the text that would confirm it, Every Thing Must Go (LIT-301), is unread.
On that reading, the question’s work is done by its qualifiers, “independently generalizable” and “intervention-relevant”. Those are the projectibility clause in the record’s terms. What is missing is a statement of them that a communicative category could fail.
What would count as an answer
A criterion, stated before the data, that a candidate communicative category could fail while still being compressible: for instance, generalization to realizations, media or speakers not used to define it, under probes chosen independently of the category, together with predicted responses to interventions on the relations said to constitute it. An answer would also say which of these is necessary. A further demonstration that some category predicts well within the data used to define it would not count.
What it is not
Not the question of which relations constitute an act and which only accompany it (QUESTION-030). A category could be a real effective kind with its constitutive relations still unknown, and the converse.
Note of 2026-10-11: material in the record that may bear on this
A synthesis review of 2026-10-11 found these entries bearing on this
question without being linked to it. Each is listed for what its title
says; none has yet been assessed as an answer, and none is linked by
answers.
- THEORY-057: Emotion categories are populations of variable instances grouped by the perceiver’s concepts, not natural…
- THEORY-036: Dennett’s real-pattern criterion, compressibility against the bit map, admits almost every non-random…
Note of 2026-10-12: edge for a prose citation
THEORY-036, which this question already rests on in prose, now carries an answered_by edge.
Note of 2026-10-14: the natural-kinds entry, read
Bird and Tobin’s SEP entry (LIT-244, read in NOTE-769) sharpens the question and does not answer it. THEORY-273, filed from it, holds that inductive and explanatory usefulness is necessary but not sufficient for a natural kind: it holds of classes that share any natural property, and it cannot tell a natural kind from a conventional or pragmatic one. So the question’s qualifiers cannot be enough by themselves on the mainstream view. The entry supplies two candidate criteria. One is Slater’s stable property cluster, which would make the qualifiers sufficient if stability is judged before the data. The other is Hacking’s test for constructed categories, that real instances are “found through empirical investigation”. Neither is assessed here as an answer.
Note of 2026-10-21: re-raised, with one criterion new
A27 of the exchange of 2026-10-20, asked at U27 whether the record is better placed on this question, offered five “candidate criteria” for an “operationally warranted effective kind”. Four are A191’s (above). The fifth is “independence from self-confirming classification effects”, or a causal account of them, which is NOTE-769’s point about Hacking’s interactive kinds: a category used by speakers can change what they do. A27 called the result “close to a defensible conditional answer”. It is not one on this question’s terms. A27 declines to say which criteria are necessary (“not five individually necessary conditions”), and its conclusion, a domain-relative kind and not a natural kind, is THEORY-036 and THEORY-273, which it did not cite. The fifth criterion is filed as CLAIM-402, which bears on this question without settling it.
Espeland and Sauder, read the same day (NOTE-1050), state the effect the fifth criterion guards against, sociologically and in passing. Their footnote 10, citing Porter on “Hispanics”: “As a category acquires constituencies who use it, those encompassed by the category develop shared interests and identities that make the category more meaningful and ‘real.’” Their conclusion: “the census produces ‘Hispanics,’ and the LSAT defines ‘merit.’” It is reported, not studied, and their study is of rankings, where every school is under the same measure (THEORY-862). The question stays Open.