Found on 2026-10-10 by a search the owner asked for: a documented case in which the speaker’s intention is held fixed, or nearly, and the act verdicts of informed participants change with a relation among the parties (QUESTION-073). Two research reports read it, and they read its structure differently. This entry reconciles them.
The case
Labov, Cohen, Robins and Lewis recorded group sessions of “sounding”, ritual insult, among the Thunderbirds and the Jets in Harlem. The report is A Study of the Non-Standard English of Negro and Puerto Rican Speakers in New York City, Vol. II (1968), §4.2.3. Pages below are the report’s own, read in the OCR text of the ERIC scan. The quotations were checked against that text on 2026-10-10.
The rule. A sound and a personal insult can have the same form, a pejorative statement about the hearer’s relative. Members tell them apart by uptake: “a personal insult is answered by a denial, excuse or mitigation, whereas a sound or ritual insult is answered by another sound” (p. 103). Rule (1) makes a remark a sound when B believes that A believes P is not true, and B believes that A believes B knows it. Labov sums this up (p. 107) as “it is an AB event that P is not true”, made before an audience C. Then: “If the proposition P were deleted, the ritual insult would become a personal insult. Your family is poor! is not a ritual insult, but a personal one” (p. 107).
Boot and David (Thunderbirds), pp. 100–102. David answers Boot’s sounds with a taunt about Boot’s stepfather’s stutter. Labov: “The fact that this is a personal insult and not a ritual insult is shown by the fact that Boot answers it.” Boot replies about David’s father’s grey hair, and Labov says the reply “is related to A-5 in exactly the way that one sound is related to another” (p. 101). Both are taken as personal. David denies (“Because he old, he’s old, that’s why!”), is brought to tears, and Ricky objects: “Aw man, cut it out.” Labov adds (p. 107): “Boot’s stepfather does stutter; David’s father is old and has gray hair—and all the Thunderbirds know this.” Then Boot “now returns to ritual sounding: his next insults are not intended as allegations of fact, but David continues to take them as personal” (p. 101), and “group support” ebbs from him (p. 102).
The Jets’ chair, p. 111. “When a sound becomes too ordinary — too possible — we can then observe a sudden switch in the pattern of response to that appropriate for a personal insult. This can happen by accident”. A: “I went in Junior house ‘n’ sat in a chair that caved in.” B: “You’s a damn liar”. Labov: it is “the only instance in the Jet sounding session where a statement is denied”.
The hypothetical question, pp. 98–99. Clarence Robins, of the research team, asks Money: “What would you say if Boot said ‘Your father look like Punjie’?” The group laughs, and Boot says: “He’s sounding on you, Money!” Robins: “No, no, if Boot said it…”. Labov: “other staff members join in and try to make it clear that we are only asking what Money would say if”. His gloss is that the question “is quickly transformed into actual sounding.” Here the speaker’s side states its intention in the transcript, and the group takes the remark as a sound anyway.
Standing decides it among the Crow and Hidatsa. A companion datum, from a kin institution. Lowie (1912, p. 204) on the Crow i’watkucé, the joking-relative: one who had done wrong might be twitted “and publicly disgrace[d]”, “yet such action must not be resented”. He gives the non-verbal form of the test too. A man finding his wagon wheels reversed “would resent such action, but as soon as he finds out who has played the joke he cannot get angry”. Lowie (1917, p. 42) on the Hidatsa: “it was not the function of his fellow-clansmen but of his maku’tsati to reprove him or make fun of him”. The same reproof is licensed or an offence by who gives it. CASE-054 files the institution for another purpose.
Two readings of its structure, reconciled
One report reads the case as “different structure, different act”: sounds form chains (S-1 e S-2 e S-3), personal insults form insult–denial dyads, and rule (1) adds a relation that personal insults lack. The other reads it as “same structure, different act”, on Labov’s own words: Boot’s reply relates to David’s “in exactly the way that one sound is related to another”.
Both are right, about different relations.
- The exchange relations are the same. Each turn answers the last in kind, targets the other’s father, and is made before the group. Labov states that correspondence himself.
- One further relation differs: whether the untruth of P is shared knowledge, an AB-event, before the audience. Under it the two exchanges are not isomorphic. Stepfathers who stutter and chairs that cave in are possible. Mothers who look like Howdy Doody are not.
So the case does not show that no structure fixes the act. It shows that the structure shared by the turns does not, and that a specified further relation does. Labov specified it from outside the exchange, from what the group knows.
What it can show
For CLAIM-151. The documented, non-institutional case the claim’s version 2 asks for. Whether a remark is a sound or a personal insult follows a relation among A, B and the audience C. It does not follow the words, which can be the same, or only the speaker’s state. In the Boot exchange Labov attributes a ritual aim to Boot throughout the turn that David takes as personal. In the chair exchange Labov calls the switch an accident. In the Robins exchange the speaker’s side states its aim, and the group’s verdict differs. Labov states the act condition itself in relational terms, over three parties. This is the “frame the parties sustain together” that QUESTION-073 asks about, in a field linguist’s recorded data. The Crow and Hidatsa norm adds the acts with standing conditions, reproof among them, that the claim’s defeat condition names. There, who may say it decides between licensed reproof and offence.
For CLAIM-138. Its point at the level of acts, on documented material. Take the exchange relations alone, and Boot’s and David’s turns correspond as one sound to another while the acts differ. Add the independently specified relation, and structure and act line up. So relational structure fixes the act only once the relations that count are chosen, which is the remedy CLAIM-138 says concedes non-structural input. The chosen relation is epistemic, shared knowledge of untruth. It is not extensional. That is the reading on which NOTE-038’s C4 says the ontic structural realist escapes Newman. So the case shows both what the objection costs and the way out. It does not settle which way CLAIM-136’s owner should go.
For CLAIM-221. The first case the record holds that can be read against the intention-based rival on documented data, not on CASE-018’s invented variation.
What it cannot show
The condition met only partly is that the intention is held fixed and documented independently. In the Boot and chair exchanges the ritual aim is Labov’s attribution (“not intended as allegations of fact”, “by accident”), inferred from the sequence. It is not the speaker’s avowal. Boot’s one apparent avowal, “Yeah, I only be playin’, I only be playin’” (p. 102), comes as mitigation of a charge of David’s, and what it refers to is unclear in the OCR text. Only the Robins exchange has the speaker’s side stating its intention, and there “He’s sounding on you” is ambiguous between Robins and the hypothetical Boot. It is also a use/mention case: a question about a sound, taken as a sound.
The strongest reply. Rule (1) is stated in nested beliefs about beliefs, which is Gricean in shape. The intention-based theorist reads it as the condition under which a speaker can rationally intend to sound, and a hearer recognise that intention. On that reading Boot did sound, and David misread him: the act was a sound and the denial is mistaken uptake. The Robins exchange gets the same answer, that the group misclassified a question. The chair case resists the reply best. The group raises no contradiction, and Labov treats the outcome as a personal insult, not as a misread sound. But the speaker’s view is not recorded there.
The category is contested. Kochman (1983) argues that personal insults lie inside sounding, so Labov’s boundary is one analyst’s (abstract only read). That cuts both ways: competent analysts disagree about the act category of the same exchanges.
What would close the gap. An elicited test on these transcripts. Hold the speaker’s stated intention fixed and vary only whether P’s untruth is shared knowledge, as a change of situation, do(S), not an instruction to the rater, do(I) (CLAIM-130). For instance, one group of raters takes the audience role with the group’s history in hand: they have heard, in the session, that Boot’s stepfather stutters. Another meets the line cold. Labov predicts outsiders “would have to suspend judgment” (p. 107). Use the chair line and a matched impossible line (“sat in a chair made of Dog Yummies”), Boot’s stutter turn and a matched ritual turn. These could serve as items for the discriminating study, CASE-055, as an arm with no rendering in which only the audience’s relation changes. Raters who are told the speaker meant it ritually and still call it personal would meet the rival. Raters who follow the stated intention would support it.
Unverified. The 1972 book chapter, the canonical revised text, was not read (lending only). Its wording and pages are unverified, and the quotations stand on the 1968 report.
Sources
- W. Labov, P. Cohen, C. Robins and J. Lewis, A Study of the Non-Standard English of Negro and Puerto Rican Speakers in New York City, Vol. II: The Use of Language in the Speech Community, Cooperative Research Project 3288, Columbia University, 1968. ERIC ED028424; https://archive.org/details/ERIC_ED028424 and https://files.eric.ed.gov/fulltext/ED028424.pdf. Read directly in the OCR text, §4.2.3–4.2.4, pp. 88–111. The key quotations were checked against it on 2026-10-10.
- W. Labov, “Rules for ritual insults”, in Language in the Inner City (University of Pennsylvania Press, 1972), ISBN 0812210514. Not read. The chapter’s page range (commonly cited as pp. 297–353) and its reprint in Sudnow (ed.), Studies in Social Interaction (Free Press, 1972), are unverified.
- T. Kochman, “The boundary between play and nonplay in black verbal dueling”, Language in Society 12(3) (1983), 329–337, DOI 10.1017/S0047404500009970. Abstract only.
- R. H. Lowie, Social Life of the Crow Indians, AMNH Anthropological Papers 9(2) (1912), https://archive.org/details/b33444833. Read directly, pp. 204–205.
- R. H. Lowie, Notes on the Social Organization and Customs of the Mandan, Hidatsa, and Crow Indians, AMNH Anthropological Papers 21(1) (1917), https://archive.org/details/customsofmandan00lowirich. Read directly, p. 42.
- Considered and ranked lower. A. Gorman and C. H. Jordan, “‘I know you’re kidding’”, Personal Relationships 22 (2015), 173–187, DOI 10.1111/pere.12071 (the search read the abstract only): a staged tease with closeness manipulated, but targets also rated it “better intentioned”, which fits the rival. M. Sap et al., “The Risk of Racial Bias in Hate Speech Detection”, ACL 2019, DOI 10.18653/v1/P19-1163 (the search read §4 and Appendix A.5): annotators primed by instruction, which is do(I). Sap et al. is an NLP paper and would need an anthology check before any filing.