--- slug: the-banned-question passed: 2026-10-11 by: claude-vibrant-lamport-79znpp, two independent Claude checkers on oversight/claims-pass.md (source first, every claim the page makes), an adjudicator who re-read each finding at its source and a skeptic who tried to make the page right; every HIGH and every lone WRONG re-read by the instance at the source before fixing; fixes made by a fixer agent and read against the sources by a separate fix-checker checked: 61 confirmed: 39 wrong: 18 unverifiable: 2 observed: 2 stale: 0 --- # Claims pass: The Question They Banned Reader-side check, run before the fixes from an empty directory by both checkers and the adjudicator with the same result: `curl -sS -o verify-the-banned-question.mjs https://artwaste.land/checks/research/the-banned-question/verify-the-banned-question.mjs` returned HTTP 200, 7,837 bytes, byte-identical to research/the-banned-question/verify-the-banned-question.mjs, and `node verify-the-banned-question.mjs` printed "30/30 PASS", exit 0 (B's run of the /checks/manifest.json recipe gave the same). Every figure it printed matched the page. The self-reading probe (step 6) does not apply in its strict form: the verifier reads no file and the page prints no transcript of its output. B ran it anyway on a scratch copy of the page with "917 participants" changed to "971 participants", and the verifier still printed 30/30 PASS, as a C0 check must; the placard says it cannot notice a figure that drifted on the page. After the fixes the repository verifier gives 30/30, exit 0, and a scratch copy with the old cry model pasted back fails 4 of 30. Five claim lines are WRONG and 13 MINOR; the MINOR lines are counted as wrong. Of 20 merged findings, 17 were fixed. The four WRONG findings: Mampe's 0.45 and 0.60 are positions as a fraction of each cry's length, not seconds, and the cry model was rebuilt to peak there (F1); Ramachandran and Hubbard report 95% of people, not 95–98% of English and Tamil speakers (F2); Köhler's 1929 book already prints maluma and reports no test on people (F3); the ban's motive is contested, not the fight against a flood of unfalsifiable papers the page described (F4). The fifth WRONG line, the dead link to the 1866 statutes, is F5, which the director classed MINOR. The other MINOR findings include hedges dropped from Mampe, a hymn said to force an exact 0.1 Hz breath, a naive pooled z-test, 72% applied to words Ćwiek did not test, a check box that claimed to regenerate every number, origin cards without their Hypothesis tag and a stale dateModified. The page now carries a dated "Corrected 2026-10-11" line for F1 to F4. Three findings were declined after re-reading the sources: founded in 1864 (F18), 8 March 1866 as the date of the bylaw (F19) and the historians' correction of the legend (F20). No finding is held by a note. ## Claims - CONFIRMED | On 8 March 1866 the Linguistic Society of Paris wrote into its bylaws that it would hear no paper on the origin of language | https://www.slp-paris.com/statuts1866.html ; Quintili 2016, https://art.torvergata.it/retrieve/e291c0d6-9694-cddb-e053-3a05fe0aa144/03%20THS%202016%20Paolo%20Quintili.pdf | "Approuvés par décision ministérielle du 8 mars 1866." / Quintili: "The famous prohibition of the Société de Linguistique de Paris, issued on March 8th, 1866, which forbade its members to submit research papers about the origin of language"; 8 March 1866 is the date of ministerial approval, and the page body says so. 'Bylaws' is the usual English (Graffi: 'the bylaw, Article II, adopted in 1866'). B's MINOR is refuted (F19). - CONFIRMED | «La Société n'admet aucune communication concernant, soit l'origine du langage, soit la création d'une langue universelle.» (Article 2, in full) | https://www.slp-paris.com/histoire.html ; https://www.slp-paris.com/statuts1866.html ; Mémoires SLP t. 1 (1868), https://archive.org/download/mmoiresdelasoci01parigoog/mmoiresdelasoci01parigoog_djvu.txt | histoire.html: "Célèbres pour leur article 2 : « La Société n’admet aucune communication concernant, soit l’origine du langage, soit la création d’une langue universelle. »"; Word for word. The statuts1866 page omits the comma after 'langage'; the 1868 Mémoires (per A) have it. - CONFIRMED | "The Society will accept no communication concerning either the origin of language or the creation of a universal language." | https://www.slp-paris.com/statuts1866.html | "La Société n’admet aucune communication concernant, soit l’origine du langage soit la création d’une langue universelle." - CONFIRMED | Approved by ministerial decision · 8 March 1866 | https://www.slp-paris.com/statuts1866.html | "Statuts de 1866 ... Approuvés par décision ministérielle du 8 mars 1866." - CONFIRMED | Read it twice, because two things are banned, not one. | https://www.slp-paris.com/statuts1866.html | "soit l’origine du langage soit la création d’une langue universelle" - CONFIRMED | The Société de Linguistique de Paris, founded in 1864 and given official sanction in 1866 | https://api.crossref.org/works/10.2143/bsl.91.1.2002511 ; https://cths.fr/an/societe.php?id=515 ; https://www.slp-paris.com/histoire.html | Bergounioux 1996: "Aux origines de la Société de Linguistique de Paris (1864-1876)"; CTHS: "Année de création : 1863" ... "président fondateur 1864-1865"; SLP: "apparaît vers 1863" ... "Statuts de 1866"; Sources give 1863, 1864 and 1865; 1864 is defensible. B's MINOR is refuted (F18). - UNVERIFIABLE | Both were barred for the same reason: they filled the meeting hall with confident papers that nobody could ever check. | https://www.slp-paris.com/statuts1866.html ; https://www.slp-paris.com/histoire.html ; Quintili 2016, https://art.torvergata.it/retrieve/e291c0d6-9694-cddb-e053-3a05fe0aa144/03%20THS%202016%20Paolo%20Quintili.pdf ; Graffi, https://docs.univr.it/documenti/OccorrenzaIns/matdid/matdid134765.pdf | The statutes give no reason. histoire.html: "Elle est à cette époque un cercle de discussion où se retrouvent quelques américanistes" ... "Ils visent à distinguer la S.L.P. des cercles positivistes et républicains"; No record was found of origin papers filling the society's meetings, and the motive is contested (F4). - WRONG | the young Société de Linguistique defining itself as a rigorous science against the more speculative Société d'Anthropologie of Paul Broca next door — and it was aimed squarely at the flood of unfalsifiable origin papers | https://www.slp-paris.com/histoire.html (Bergounioux 1997, BSL 92/1) ; Quintili 2016, https://art.torvergata.it/retrieve/e291c0d6-9694-cddb-e053-3a05fe0aa144/03%20THS%202016%20Paolo%20Quintili.pdf ; Graffi, https://docs.univr.it/documenti/OccorrenzaIns/matdid/matdid134765.pdf | "Leur intention est de créer, face à la Société d’Anthropologie de Paris fondée par Broca en 1859 et inspirée par les travaux d’inspiration matérialiste de Chavée, une association concurrente, proche des cercles monarchistes et catholiques"; Quintili: "It was a political and ideological choice"; PARTLY (F4). The rivalry and the political moment are right. The society's historian frames the opposition as ideological; Graffi names Botha (2003) as the proponent of the lack-of-evidence reading and disagrees with it. No source exists for the 'flood'. - CONFIRMED | Société d'Anthropologie of Paul Broca | https://www.slp-paris.com/histoire.html | "la Société d’Anthropologie de Paris fondée par Broca en 1859" - CONFIRMED | The ban was a house rule of one society in Paris, not a law, not a global taboo. | https://www.slp-paris.com/statuts1866.html | "Statuts de 1866 ... Article 2 . La Société n’admet aucune communication" - CONFIRMED | Five years later Darwin published a whole theory of language origin / Darwin's Descent of Man, with its full musical-origin theory, appeared in 1871, five years later | https://darwin-online.org.uk/EditorialIntroductions/Freeman_TheDescentofMan.html ; https://archive.org/download/descentofmansele01darw/descentofmansele01darw_djvu.txt | "The first issue, of 2,500 copies, was published on February 24" [1871]; The 1871 text itself cites "'On the Origin of Language,' by H. Wedgwood, 1866". - CONFIRMED | Charles Darwin 1871: Before articulate speech, our ancestors courted and challenged one another with musical cries, as songbirds and gibbons do; words later grew over that singing. | https://archive.org/download/descentofmansele01darw/descentofmansele01darw_djvu.txt (1871 first edition, vol. 1) | "primeval man, or rather some early progenitor of man, probably used his voice largely, as does one of the gibbon-apes at the present day, in producing true musical cadences, that is in singing ; we may conclude from a widely -spread analogy that this power would have been especially exerted during the courtship of the sexes, ... and serving as a challenge to their rivals. The imitation by articulate sounds of musical cries might have given rise to words expressive of various complex emotions." - CONFIRMED | Otto Jespersen 1922 · Language: The first "words" were long, sung, emotional, half-musical outbursts — poetry before prose. / Jespersen was writing on origins into the 1920s. | https://archive.org/download/languageitsnatur00jespuoft/languageitsnatur00jespuoft_djvu.txt | "First published in January 1922"; "Language, then, began with half-musical unanalyzed expressions for individual beings and solitary events."; "poetic language is on the whole older than prosaic language" - CONFIRMED | Robin Dunbar 1996 · vocal grooming: As groups grew, vocal "grooming at a distance" — chatter, then gossip — replaced the hands. | https://api.crossref.org/works/10.1017/s0140525x00032325 (Dunbar 1993 BBS, abstract via both checkers) ; Open Library search, https://openlibrary.org (Grooming, Gossip and the Evolution of Language, 1996, Faber and Faber) | "Maintaining stability of human-sized groups by grooming alone would make intolerable time demands ... language uniquely fulfills this requirement ... In human conversations about 60% of time is spent gossiping"; Uncontested; both checkers confirmed it, and I confirmed the BBS record on Crossref. - CONFIRMED | Steven Brown 2000 · musilanguage: two branches off one ancestral "musilanguage" ... a shared stem, not one born from the other | https://api.crossref.org/works/10.7551/mitpress/5190.003.0022 ; https://en.wikipedia.org/wiki/Musilanguage | Crossref: "The 'Musilanguage' Model of Music Evolution", The Origins of Music, pp. 271-300; Wikipedia: the shared features are not "due to one system emerging from the other"; Crossref dates the book 22 Nov 1999; it is conventionally cited as 2000. - CONFIRMED | Dean Falk 2004 · "putting the baby down": Mothers who could not carry infants constantly soothed them with melodic "motherese" | https://api.crossref.org/works/10.1017/s0140525x04000111 ; Europe PMC abstract of doi:10.1017/s0140525x04000111 | "It is hypothesized that hominin mothers adopted new foraging strategies that entailed maternal silencing, reassuring, and controlling of the behaviors of physically removed infants"; BBS 27:491-503 (2004) - CONFIRMED | Steven Mithen 2005 · The Singing Neanderthals: "Hmmmmm" — Holistic, manipulative, multi-modal, musical and mimetic ... later broken into words | https://acampbell.uk/bookreviews/r/mithen2.html ; https://www.cambridge.org/core/product/B929887B5DDB8D30E7871909E7E48B0D | "To describe this he uses an acronym, \"Hmmmmm\" (Holistic, manipulative, multi-modal, musical and mimetic)."; "a capacity for segmentation ... broke the continuous line of Hmmmmm, creating discrete replicated units"; Both sources are secondary; uncontested. - CONFIRMED | Historians of linguistics have made exactly this correction — the ban is real, its legend is inflated. | https://docs.univr.it/documenti/OccorrenzaIns/matdid/matdid134765.pdf (Graffi, citing Auroux 1989) | "It carried on throughout the 19th century, despite the «ban» by the Linguistic Society of Paris in 1866 ... (as is clearly shown by Auroux 1989): indeed, the greatest number of works about language origin dates from the 19th century."; Both checkers marked this UNVERIFIABLE; adjudication found the historians (F20, REFUTED). Graffi also records a gap of about a century starting late in the 19th century. - WRONG | In 1929 the Gestalt psychologist Wolfgang Köhler showed people two shapes ... takete and baluma (later maluma), and asked which was which. Almost everyone agreed. | https://archive.org/download/gestaltpsycholog00khle/page/n257_w800.jpg (Köhler, Gestalt Psychology, Liveright, 1929, p. 242) | "If, looking at these two figures (18 and 19), the reader is asked to choose which he would rather call “takete” or “maluma,” he will probably be able to decide with ease."; F3. The 1929 book says 'maluma' and reports no test on people. - WRONG | Ramachandran & Hubbard ... reported that 95–98% of English and Tamil speakers give the same answer. | https://www.neurohumanitiestudies.eu/archivio/RamachandranHubbard_Synaesthesia.pdf (JCS 8(12):3-34, 2001), p. 19 | "95% of people pick the left as kiki and the right as bouba, even though they have never seen these stimuli before."; F2. The paper has no 98% and no English/Tamil bouba/kiki sample. - CONFIRMED | Ramachandran & Hubbard revived it in 2001 with bouba and kiki | https://www.ebi.ac.uk/europepmc/webservices/rest/PMC8591387/fullTextXML | "The phenomenon was popularized in the twenty-first century by Ramachandran & Hubbard [42] with bouba and kiki." - CONFIRMED | Ramachandran & Hubbard, "Synaesthesia — A Window Into Perception, Thought and Language," J. Consciousness Studies (2001), tracing to Köhler, Gestalt Psychology (1929) | https://www.neurohumanitiestudies.eu/archivio/RamachandranHubbard_Synaesthesia.pdf | "consider stimuli like those shown in figure 7, originally developed by Köhler (1929; 1947)" - CONFIRMED | a direct, checkable dent in Saussure's doctrine that the sign is arbitrary | https://www.ebi.ac.uk/europepmc/webservices/rest/PMC8591387/fullTextXML ; https://fr.wikisource.org (Cours de linguistique générale, Première partie) | "dominated by the dogma of arbitrariness, according to which the forms of words do not resemble their meanings"; Saussure: "le signe linguistique est arbitraire" - CONFIRMED | Ćwiek and colleagues (2022) ran the test on 917 participants speaking 25 languages from nine families across ten writing systems. | https://www.ebi.ac.uk/europepmc/webservices/rest/PMC8591387/fullTextXML | "The remaining sample contained data from 917 participants of 25 languages from nine different language families ... The 25 languages included 10 different scripts." - CONFIRMED | Overall, about 72% chose the "congruent" pairing (a 95% credible interval of roughly 56–82%). | https://www.ebi.ac.uk/europepmc/webservices/rest/PMC8591387/fullTextXML | "The estimated posterior mean proportion of bouba/kiki matches across languages was 72%, with a 95% credible interval ranging from 56% to 82%." - CONFIRMED | Roman-script speakers landed near 75%, non-Roman near 63% — and crucially, both sit far above the 50% a coin would give | https://www.ebi.ac.uk/europepmc/webservices/rest/PMC8591387/fullTextXML | "Languages that predominantly use the Roman script had numerically higher bouba/kiki matches (descriptive average: 75%) than languages that use other scripts (63%)." / "Participants who used a script other than the Roman alphabet were, on average, far above chance (63%)" / "compared to a conservative baseline of one-half = 50%"; B's MINOR on the '50% a coin' baseline is refuted: 50% is the paper's own conservative baseline, and the paper itself says 'far above chance'. - CONFIRMED | z = (0.720 − 0.500) / √(0.5·0.5 / 917) = 13.32 (arithmetic) | public/strata/the-banned-question/index.html (cohort line, zProp(0.72, 917, 0.5)) ; recreated: node -e (0.72-0.5)/Math.sqrt(0.25/917) | z 13.324083458159512 - MINOR | 72% of 917 vs. chance gives z ≈ 13.3, far beyond any threshold (as inference) | https://www.ebi.ac.uk/europepmc/webservices/rest/PMC8591387/fullTextXML | "The logit coefficient of the intercept was above zero (+0.93, s.e. = 0.31)" / "The estimated posterior probability of the logit intercept being above zero (=chance level) was very high ( p ( β 0 > 0 ) = 0.99 )."; F9. A pooled z-test ignores the variation between languages that the authors modelled. - MINOR | The expected pairing — round ↔ bouba/maluma/lomo, spiky ↔ kiki/takete/kipi — is the one 72% of people across 25 languages choose | https://www.ebi.ac.uk/europepmc/webservices/rest/PMC8591387/fullTextXML | "Participants viewed the rounded and spiky shapes as they listened to the spoken words bouba and kiki in two successive trials."; F10. The 72% is for bouba/kiki only. - CONFIRMED | The effect is weaker in some writing systems, differs for people blind from birth | https://www.ebi.ac.uk/europepmc/webservices/rest/PMC8591387/fullTextXML ; Europe PMC abstracts of Fryer et al. 2014 (doi:10.1016/j.cognition.2014.03.015) and Bottini et al. 2019 (doi:10.1016/j.cognition.2019.01.006) | Fryer: "in a sample of people with a range of visual impairments, from congenital total blindness to partial sight, the effect was significantly less pronounced." Bottini: "Experiment 2, however, showed a partially different pattern of SSC in sighted and blind"; The evidence is mixed; Ćwiek notes the effect 'has also been found in congenitally blind individuals'. 'Differs' is fair. - CONFIRMED | Ćwiek et al., "The bouba/kiki effect is robust across cultures and writing systems," Phil. Trans. R. Soc. B 377:20200390 (2022) | https://api.crossref.org/works/10.1098/rstb.2020.0390 | title 'The bouba/kiki effect is robust across cultures and writing systems'; Philosophical Transactions of the Royal Society B; volume 377; published-print 2022-01-03; authors Ćwiek, Fuchs, Draxler, Asu ... - CONFIRMED | Mampe and colleagues (2009) recorded the cries of 60 newborns — 30 from French-speaking homes, 30 from German — all just a few days old | Mampe et al. 2009 full text, https://web.archive.org/web/20230314222504id_/https://www.cell.com/current-biology/fulltext/S0960-9822(09)01824-7 | "Here, we analyzed the crying patterns of 30 French and 30 German newborns"; "during the first 2–5 days of life" - MINOR | French babies cried with a rising melody; German babies with a falling one, mirroring the prosody of the language they had been hearing through the womb wall / prosody ... is learned before words, in the womb | Mampe et al. 2009 full text, https://web.archive.org/web/20230314222504id_/https://www.cell.com/current-biology/fulltext/S0960-9822(09)01824-7 | "The French group preferentially produced cries with a rising melody contour, whereas the German group preferentially produced falling contours." / "probably learned these characteristics of their mother tongue by listening to it prenatally (although we cannot completely exclude early postnatal learning during the first 2–5 days of life)"; F6. The page drops 'preferentially' and 'probably'. - WRONG | The average German cry peaked in pitch and loudness early, around 0.45 s into the cry; the French cry peaked later, around 0.6 s. / pinned to the reported peak-timing marks / on the exact timing marks from a 60-baby study | Mampe et al. 2009 full text, https://web.archive.org/web/20230314222504id_/https://www.cell.com/current-biology/fulltext/S0960-9822(09)01824-7 ; https://languagelog.ldc.upenn.edu/nll/?p=1869 | "after normalizing arc duration to 1 s, the normalized time [t norm (F0 max )] corresponding to the maximum pitch (F0 max ) of each melody arc was determined" / "a marked difference in the median values of t norm (F0 max ) ... (French group, 0.60 s; German group, 0.45 s)"; F1. These are positions as a fraction of each cry's length, not seconds. The model's French peak falls at t/D = 0.50006, the paper's dividing line. - MINOR | German · peak ≈ 0.45 s · net falling / French · peak ≈ 0.60 s · net rising; the French curve ... ends higher than it began | Mampe et al. 2009 full text, https://web.archive.org/web/20230314222504id_/https://www.cell.com/current-biology/fulltext/S0960-9822(09)01824-7 | "Only simple cries containing single rising-and-then-falling melody arcs were analyzed." / "t norm (F0 max ) values < 0.5 s represent “falling contours”; t norm (F0 max ) values > 0.5 s represent “rising contours”"; Part of F1. The paper's Discussion glosses 'rising (low to high)', but its definition is where the peak sits; the model's net-direction property is not reported. - CONFIRMED | It is one study, n=60, and it drew methodological pushback (Language Log among others). | https://languagelog.ldc.upenn.edu/nll/?p=1869 (Mark Liberman, 'Native wails', Nov 2009) | "This was apparently done by hand on narrow-band spectrograms, and again was apparently not done \"blind\""; "the babies were recorded in different places, apparently under different circumstances, by different experimenters" - CONFIRMED | Mampe, Friederici, Christophe & Wermke, "Newborns' Cry Melody Is Shaped by Their Native Language," Current Biology 19:1994–1997 (2009) | https://api.crossref.org/works/10.1016/j.cub.2009.09.064 | "Newborns' Cry Melody Is Shaped by Their Native Language"; Mampe, Friederici, Christophe, Wermke; Current Biology 19, 1994-1997; 2009-12 - CONFIRMED | Vickhoff and colleagues (2013) wired up a choir of 18-year-old singers (11 in the group analysis, of 15 recorded; 5 more in case studies) | https://www.ebi.ac.uk/europepmc/webservices/rest/PMC3705176/fullTextXML | "Of the 15 subjects who participate in the study, four recordings are not included in the analysis due to technical reasons (discontinuous registration). The sample size is this n = 11 for the group study." / "we examine five subjects (who did not participate in the first study) in a separate case study" / "15 healthy 18 year old subjects of mixed gender"; A's doubt that the 5 were additional people is answered by the paper: they were. - MINOR | Vickhoff has a small n (~11–18) | https://www.ebi.ac.uk/europepmc/webservices/rest/PMC3705176/fullTextXML | "The sample size is this n = 11 for the group study."; "15 healthy 18 year old subjects"; F8. 18 is the singers' age, not a count. - MINOR | The headline became "choir members' hearts beat as one." | https://www.wtkr.com/2013/07/09/study-says-choir-singers-hearts-beat-as-one ; https://www.npr.org/sections/health-shots/2013/07/09/200390454/when-choirs-sing-many-hearts-beat-as-one |