---
slug: the-accent-of-nowhere
passed: 2026-11-16
by: claude-hopeful-bardeen-wseqdd, one independent agent on oversight/claims-pass.md (step 4a: all 455 entries of assay/sources/the-accent-of-nowhere.json, plus 6 claims the record left out), before publishing; each finding re-read against its source by the instance
checked: 461
confirmed: 387
wrong: 6
unverifiable: 0
observed: 68
stale: 0
---
Counts: checked 461 (455 record entries + 6 added), confirmed 387, wrong 6, unverifiable 0, observed 68, stale 0. MINOR findings are WRONG or CONFIRMED lines whose note begins "MINOR:"; the one MED finding's note begins "MED:". Each line ends with the record entry's id in brackets.
Live: https://artwaste.land/strata/the-accent-of-nowhere/ answered 404 on 2026-11-16 (not yet deployed; https://artwaste.land/ answered 200), so the repository's built page was read as the page.
Reader check (simulated stranger, empty directory, the published verifier and the four page files): AW_DIR=
/files node verify-the-accent-of-nowhere.mjs: exit 0, 5698 passed, 0 failed; with --mutate exit 0, 5715 passed, 16 of 16 planted faults caught. Inside the repository: exit 0, 9069 passed, 0 failed; --mutate exit 0, 9086 passed, 16 of 16.
Online source check (node assay/sources.mjs the-accent-of-nowhere --online): 227 in the repository, 177 FOUND, 48 NOT FOUND, 3 UNREADABLE. All 48 NOT FOUND point at page images (.jpg), which the tool cannot read; all 3 UNREADABLE are the Equinox PDF, whose kept text copy holds the excerpts. Those claims were read in the kept NOTES.md, the kept OCR, and in four spot-checked kept images (Pfister pp.444 and 780, Henry p.140).
# Claims pass: The Accent of Nowhere
Run before publishing, so the live page answered 404 and the pass read the page as built in the
repository. Reader check, simulated from an empty directory holding only the published verifier and
copies of the four files the page serves: `AW_DIR=/files node verify-the-accent-of-nowhere.mjs`
exit 0, 5,698 passed, 0 failed; with `--mutate` 16 of 16 planted faults caught. Inside the repository
9,069 passed, 0 failed, and 16 of 16 with `--mutate`. `node assay/sources.mjs the-accent-of-nowhere
--online`: 177 excerpts found at their URLs, 48 not found (all page images, which the tool cannot
read; the reader read them in the kept images, OCR and NOTES.md), 3 unreadable (the Equinox PDF,
whose kept text holds the excerpts). The download path itself is checked after the deploy.
After the pass, the layer was published: replaying its commits onto `main` renamed them, so the
registration commit the lines below cite as e198db40a6 is a22c3ec7d4 on `main` (README.md, "The
order things happened in"). Live check after the deploy, from an empty directory, exactly as the page
instructs: `curl -sSO https://artwaste.land/checks/verify-the-accent-of-nowhere.mjs` then
`node verify-the-accent-of-nowhere.mjs`, exit 0, 5,703 passed, 0 failed (live build 53aa964ea).
## Claims
- CONFIRMED | The study's control, registered in advance as nonsense words “made by English writers who were not trying to leave English”. | research/the-accent-of-nowhere/PREREGISTRATION.md | C2: "made by English writers who were not trying to leave English"; file unchanged since commit e198db40a6 [carroll-lear-control]
- CONFIRMED | Lewis Carroll's Jabberwocky is read in a copy of Through the Looking-Glass dated 1872 | https://en.wikisource.org/wiki/Through_the_Looking-Glass,_and_What_Alice_Found_There | title page of the scan: "London: MACMILLAN AND CO. 1872."; tongue file reads Chapter I of that scan [carroll-lear-carroll-copy]
- CONFIRMED | Edward Lear's verse in Nonsense Songs, Stories, Botany, and Alphabets (London, 1872) | https://en.wikisource.org/wiki/Nonsense_Songs,_Stories,_Botany,_and_Alphabets | title page: "LONDON: ROBERT JOHN BUSH ... 1872." (fourth thousand) [carroll-lear-lear-1872]
- CONFIRMED | and in Laughable Lyrics (1877) | https://en.wikisource.org/wiki/Author:Edward_Lear | "Laughable Lyrics (1877), 4th Book of Nonsense" [carroll-lear-lear-1877]
- OBSERVED | The first two are Wikisource's transcriptions, proofread against the page images | research/the-accent-of-nowhere/tongues/nonsense-english/carroll-jabberwocky.json | transcription_notes: proofread 100% against the three page images [carroll-lear-wikisource-carroll]
- OBSERVED | The first two are Wikisource's transcriptions, proofread against the page images | research/the-accent-of-nowhere/tongues/nonsense-english/lear-nonsense.json | transcription_notes: all 32 scan pages read against the transcription, 34 corrections listed [carroll-lear-wikisource-lear]
- OBSERVED | the third is a Project Gutenberg text that names no edition | research/the-accent-of-nowhere/sources/kept/nonsense-english/lear/pg13649.txt | kept PG #13649 header gives title, author, release date, no printed edition; lear-nonsense.json says the same [carroll-lear-lear-gutenberg]
- CONFIRMED | A program picked the words out. A word is ordinary if it is a headword of Webster's dictionary of 1913, or becomes one when a regular ending is taken off; every other word is taken as nonsense. | research/the-accent-of-nowhere/tongues/build-nonsense.mjs | header comment states the Webster 1913 headword rule as the page gives it [carroll-lear-rule]
- CONFIRMED | The rule errs both ways: chortle counts as ordinary | research/the-accent-of-nowhere/tongues/build-nonsense.mjs | header: "a coinage that had entered the dictionary by 1913 counts as ordinary (chortle)"; pooled list has no chort* word [carroll-lear-rule-errs]
- CONFIRMED | twas and England as nonsense | research/the-accent-of-nowhere/tongues/pooled/carroll-lear.json | both "twas" and "england" are in the 134 words [carroll-lear-rule-twas]
- CONFIRMED | Carroll on the poem's hard words: Humpty-Dumpty's theory, of two meanings packed into one word like a portmanteau, seems to me the right explanation for all. Lewis Carroll, The Hunting of the Snark, 1876, preface, p. x | https://en.wikisource.org/wiki/The_Hunting_of_the_Snark_(1876)/Preface | "Humpty-Dumpty's theory, of two meanings packed into one word like a portmanteau, seems to me the right explanation for all." (the hard words of "that poem", the Jabberwock) [carroll-lear-carroll-said]
- CONFIRMED | In a manuscript that Wilhelm Grimm saw at Wiesbaden the list stands under a red heading | https://archive.org/download/zeitschriftfrd06wiesuoft/zeitschriftfrd06wiesuoft_djvu.txt | "Die bibliothek zu Wiesbaden bewahrt ..."; "mit der rothen Überschrift 'Ignota lingua p simplicem homine hildegarde prolata'"; he copied it in 1833 [lingua-ignota-wiesbaden]
- CONFIRMED | Ignota lingua p simplicem hominē hildegardē prolata (in our rendering, an unknown language brought forth by the simple person Hildegard) | https://archive.org/download/zeitschriftfrd06wiesuoft/page/leaf340_w1400.jpg | page image (kept, read in NOTES.md s.1) and OCR agree on the heading; "by" for "per" is a fair rendering [lingua-ignota-heading]
- CONFIRMED | She herself names ignotamque linguam et litteras, an unknown language and letters, among the things a vision had shown her. | https://archive.org/download/analectasanctae00hildgoog/page/leaf48_w1400.jpg | Pitra p.7: "ignotamque linguam et litteras, cum ... quas mihi ad explanandum ostenderat" (NOTES.md s.2; OCR continues to the verb) [lingua-ignota-her-own]
- CONFIRMED | Grimm copied the whole list in 1833 and printed the words that have German glosses. | https://archive.org/download/zeitschriftfrd06wiesuoft/zeitschriftfrd06wiesuoft_djvu.txt | "ich habe im jahr 1833 vollständige abschrift davon genommen, hier aber nur die deutschen wörter ... herausgehoben" (he also kept the few Latin ones) [lingua-ignota-grimm-1833]
- CONFIRMED | The study reads Elias Steinmeyer's edition of 1895 | https://archive.org/download/diealthochdeutsc03steiuoft/diealthochdeutsc03steiuoft_djvu.txt | vol. 3 "BEARBEITET VON ELIAS STEINMEYER ... BERLIN ... 1895" [lingua-ignota-edition]
- CONFIRMED | which rests on two manuscripts | https://archive.org/download/diealthochdeutsc03steiuoft/page/leaf408_w1400.jpg | p.390 (image, NOTES.md s.3): "a = Codex Wiesbadensis 2. b = Codex Cheltenhamensis 9303." [lingua-ignota-two-mss]
- OBSERVED | and prints 1010 entries by the study's count | research/the-accent-of-nowhere/tongues/hildegard/lingua-ignota.json | transcription_notes CHECK 1: "1010 entries on 15 pages"; recreated: the file's words array has 1010 entries [lingua-ignota-count]
- OBSERVED | It was typed by eye from page images; no manuscript was consulted. | research/the-accent-of-nowhere/tongues/hildegard/lingua-ignota.json | notes: typed by eye from 15 page images, "No part of the manuscripts was consulted" [lingua-ignota-typed]
- CONFIRMED | Die sprache selbst muſs vorerst ein räthsel bleiben. | https://archive.org/download/zeitschriftfrd06wiesuoft/page/leaf345_w1400.jpg | p.339; kept OCR line 18269 has the same sentence [lingua-ignota-grimm-riddle]
- CONFIRMED | bei einer anzahl wörter scheint mir | https://archive.org/download/zeitschriftfrd06wiesuoft/page/leaf345_w1400.jpg | p.339; kept OCR: "bei einer anzahl Wörter scheint mir lateinischer einflufs unverkennbar" [lingua-ignota-grimm-latin-1]
- CONFIRMED | lateinischer einfluſs unverkennbar | https://archive.org/download/zeitschriftfrd06wiesuoft/page/leaf345_w1400.jpg | p.339; same sentence, followed by his list diabolus Diuueliz ... [lingua-ignota-grimm-latin-2]
- CONFIRMED | einwirkungen des deutschen nur in wenigen wörtern | https://archive.org/download/zeitschriftfrd06wiesuoft/page/leaf345_w1400.jpg | p.339; kept OCR: "einwirkungen des deutschen nur in wenigen Wörtern" [lingua-ignota-grimm-german]
- CONFIRMED | Wilhelm Grimm, Wiesbader Glossen, Zeitschrift für deutsches Alterthum 6, 1848, p. 339 | https://archive.org/metadata/zeitschriftfrd06wiesuoft | IA record: volume 6, "stated date is 1848"; index "Wiesbader glossen, von Wh. Grimm" [lingua-ignota-grimm-cite]
- CONFIRMED | Edward Kelley saw a spirit point with a rod to the letters of a table | https://raw.githubusercontent.com/textcreationpartnership/A37412/master/A37412.xml | p.74: "E. K. He standeth and pointeth with his rod to the letters of his Table" [enochian-1659-rod]
- CONFIRMED | and John Dee wrote | https://raw.githubusercontent.com/textcreationpartnership/A37412/master/A37412.xml | title: "out of the original copy written with Dr. Dees own hand" [enochian-1659-dee-wrote]
- CONFIRMED | They sat at Cracow from April to July 1584. | https://raw.githubusercontent.com/textcreationpartnership/A37412/master/A37412.xml | p.73: "Anno 1584. Aprilis 10. ... CRACOVIAE." [enochian-1659-cracow-april]
- CONFIRMED | They sat at Cracow from April to July 1584. | https://raw.githubusercontent.com/textcreationpartnership/A37412/master/A37412.xml | p.206: "Fryday, Julii 13. ... Cracoviae." [enochian-1659-cracow-july]
- CONFIRMED | The first Calls came backward, the later ones forward. | https://raw.githubusercontent.com/textcreationpartnership/A37412/master/A37412.xml | p.78: "they are to be uttered of me, backward: and of you, in practise, forward" [enochian-1659-backward]
- CONFIRMED | The first Calls came backward, the later ones forward. | https://raw.githubusercontent.com/textcreationpartnership/A37412/master/A37412.xml | p.119 (14 May): "Not BACKWARD, as you were wont to do, but FORWARD." [enochian-1659-forward]
- CONFIRMED | Asked by Kelley whether the language was known in any part of the world | https://raw.githubusercontent.com/textcreationpartnership/A37412/master/A37412.xml | XML line 7402, p.92, in the session "Saturday, Aprilis 21.": "E. K. Whether is this Language known in any part of the World or no?" [enochian-1659-asked]
- CONFIRMED | the spirit Gabriel said that the speech Adam learned after his fall | https://raw.githubusercontent.com/textcreationpartnership/A37412/master/A37412.xml | answer under "Gab.": Adam "began to learn of necessity the Language which thou, E. K. callest [Hebrew]" [enochian-1659-after-fall]
- CONFIRMED | was not to be compared with | https://raw.githubusercontent.com/textcreationpartnership/A37412/master/A37412.xml | same sentence: "much lesse to be compared with this that we deliver" [enochian-1659-not-compared]
- CONFIRMED | this that we deliver, which Adam verily spake in innocency, and was never uttered nor disclosed to man since till now Gabriel to Edward Kelley, 21 April 1584, in A True & Faithful Relation, 1659, p. 92 | https://raw.githubusercontent.com/textcreationpartnership/A37412/master/A37412.xml | quote verbatim; session headed Saturday 21 April (XML line 7373), p.92 starts line 7375 [enochian-1659-adam]
- CONFIRMED | Meric Casaubon printed Dee's record in 1659. | https://raw.githubusercontent.com/textcreationpartnership/A37412/master/A37412.xml | TCP header: preface "by Meric Casaubon"; "London : 1659." [enochian-1659-casaubon]
- OBSERVED | The study reads the Text Creation Partnership's transcription. | research/the-accent-of-nowhere/tongues/enochian/enochian-calls-casaubon-1659.json | transcription_notes: extracted from the EEBO-TCP TEI file; two missing pages read from Getty images [enochian-1659-tcp]
- OBSERVED | This is the session record, not Dee's later fair copy, which was not obtained. | research/the-accent-of-nowhere/tongues/enochian/enochian-calls-casaubon-1659.json | recension: "The raw record of the sessions, not the fair copy ... (unobtained)" [enochian-1659-not-fair-copy]
- OBSERVED | Donald Laycock's dictionary was not seen | research/the-accent-of-nowhere/tongues/enochian/NOTES.md | "his book was NOT obtained; his own words are UNSOURCED at first hand" [enochian-1659-laycock-unseen]
- CONFIRMED | the English Wikipedia article Enochian quotes him: “the resulting pronunciation makes it sound much more like English than it looks at first sight”. | https://en.wikipedia.org/wiki/Enochian | "Laycock ... says, \"the resulting pronunciation makes it sound much more like English than it looks at first sight\"" (second hand, labelled so) [enochian-1659-laycock]
- CONFIRMED | The Calls again, in the spelling of The Equinox, volume I, number VIII (London, September 1912) | https://archive.org/download/Equinoxs/Eq1v8.pdf | PDF unreadable to the tool; kept IA text: "VOL I. NO. VIII. ... SEPTEMBER MCMXII", Wieland & Co., London [enochian-1912-equinox]
- CONFIRMED | which writes vowels between the consonants and zod for z: the 1659 record's SONF and CALZ are here sonuf and caelazod | research/the-accent-of-nowhere/tongues/enochian/NOTES.md | "SONF / sonuf ... vowels between the consonants ... CALZ / caelazod: z written zod" [enochian-1912-spelling]
- CONFIRMED | Dee himself had noted in 1584 that “z extended is zod”. | https://raw.githubusercontent.com/textcreationpartnership/A37412/master/A37412.xml | XML line 5325, session of 10 April 1584: "[Δ. z extended is zod.]" (Δ marks Dee's remarks) [enochian-1912-zod]
- OBSERVED | The study knows this text only through an electronic transcription that names no transcriber. It was not checked against the 1912 printing, of which no facsimile was obtained. | research/the-accent-of-nowhere/tongues/enochian/enochian-calls-equinox-1912-etext.json | recension: transcriber not named; "NOT PROOFREAD against the 1912 printing: no facsimile was obtained" [enochian-1912-etext]
- CONFIRMED | The editor's note to the first Key, as that file gives it: Collation of the various MSS. of these calls has not done away with Various Readings; and there is not enough of the language extant to enable a settlement on general [...] | https://archive.org/download/Equinoxs/Eq1v8.pdf | PDF unreadable to the tool; kept IA text has the note verbatim, signed "ED." [enochian-1912-editor]
- CONFIRMED | Hélène Smith is Théodore Flournoy's name for a medium of Geneva. | https://archive.org/download/desindeslaplan00flou/page/n20_w700.jpg | p.1 (image; NOTES.md): "Le médium en question, que j'appellerai Mlle Hélène Smith" [martian-name]
- CONFIRMED | Hélène Smith is Théodore Flournoy's name for a medium of Geneva. | https://archive.org/download/desindeslaplan00flou/page/n34_w700.jpg | p.15 (image; NOTES.md): "Mlle Smith a toujours habité Genève depuis sa tendre enfance." [martian-geneva]
- CONFIRMED | Her Martian was first heard on 2 February 1896 | https://archive.org/download/desindeslaplan00flou/desindeslaplan00flou_djvu.txt | pp.192-193: "la naissance de la langue martienne ... depuis le 2 février 1896" [martian-first-heard]
- CONFIRMED | the last text he prints is dated 4 June 1899 | https://archive.org/download/desindeslaplan00flou/page/n242_w1400.jpg | p.223, under text 40, the last numbered text: "Auditif. 4 juin 1899" [martian-last-text]
- CONFIRMED | She heard the words in visions and noted them in pencil | https://archive.org/download/desindeslaplan00flou/desindeslaplan00flou_djvu.txt | p.193: "Dans le cas de visions spontanées, Hélène note au crayon ... les sons inintelligibles" [martian-pencil]
- CONFIRMED | or spoke them | https://archive.org/download/desindeslaplan00flou/desindeslaplan00flou_djvu.txt | p.204: "son genre d'automatisme - auditif, visuel, vocal ou graphique" [martian-modes]
- CONFIRMED | from August 1897 wrote them in a script of their own | https://archive.org/download/desindeslaplan00flou/desindeslaplan00flou_djvu.txt | p.193: "l'écriture, se fit jour à partir d'août 1897" [martian-script]
- CONFIRMED | In trance, speaking as one Esenale, she gave the French word for word. | https://archive.org/download/desindeslaplan00flou/desindeslaplan00flou_djvu.txt | p.204: "leur mot à mot français donné par Esenale"; p.219: Esenale repeats "en la traduisant mot à mot" [martian-esenale]
- CONFIRMED | putting in the accents himself | https://archive.org/download/desindeslaplan00flou/desindeslaplan00flou_djvu.txt | p.203: "Je me suis borné à mettre des accents sur les voyelles" [martian-accents]
- CONFIRMED | Flournoy printed the texts in Des Indes à la planète Mars (1900) | https://archive.org/download/desindeslaplan00flou/page/n10_w700.jpg | title page (image): "TROISIÈME ÉDITION conforme à la 1re édition ... 1900" [martian-edition]
- OBSERVED | The study read every Martian line on page images of the third edition. | research/the-accent-of-nowhere/tongues/martian/martian.json | share_proofread 117 of 117 lines of texts 1-40; method: every line of texts 1-41 read on the page images [martian-read]
- CONFIRMED | Le martien n'est que du français déguisé. (Martian is only French in disguise.) Théodore Flournoy, Des Indes à la planète Mars, 1900, p. 230 | https://archive.org/download/desindeslaplan00flou/desindeslaplan00flou_djvu.txt | p.230: "Le martien n'est que du français déguisé." [martian-flournoy-said]
- CONFIRMED | Her father was Hungarian. | https://archive.org/download/desindeslaplan00flou/desindeslaplan00flou_djvu.txt | p.15: "Son père, négociant, était hongrois" [martian-father]
- CONFIRMED | Victor Henry, a year later: Presque tous les mots du martien ont une étymologie assurée, puisée dans des langues réelles, connues plus ou moins, mais certainement connues, de Mlle Smith. (Almost all the words of Martian have an [...] | https://archive.org/download/lelangagemartien00henr/page/n169_w700.jpg | p.140 (image; NOTES.md): "1° Presque tous les mots du martien ont une étymologie assurée ..." [martian-henry-said]
- OBSERVED | What was taken down of her Martian on its first two evenings, 2 and 16 February 1896, before any of it had been translated. | research/the-accent-of-nowhere/tongues/martian/martian-first-seance.json | selection_rule: every invented word quoted from the seances of 2 and 16 February 1896 [martian-first-seance-what]
- CONFIRMED | What was taken down of her Martian on its first two evenings, 2 and 16 February 1896, before any of it had been translated. | https://archive.org/download/desindeslaplan00flou/desindeslaplan00flou_djvu.txt | pp.192-193: from 2 February 1896 until the first translation through Esenale on 2 November [martian-first-seance-first-evening]
- CONFIRMED | What was taken down of her Martian on its first two evenings, 2 and 16 February 1896, before any of it had been translated. | https://archive.org/download/desindeslaplan00flou/desindeslaplan00flou_djvu.txt | pp.152-153: "16 février 1896 ... il donnera la traduction ... Mais cette prédiction ne se réalise pas." [martian-first-seance-second-evening]
- CONFIRMED | The opening run is given as M. Lemaître noted it, aussi exactement que possible (as exactly as possible, in our rendering) | https://archive.org/download/desindeslaplan00flou/desindeslaplan00flou_djvu.txt | p.149: "dont voici le début tel que M. Lemaître l'a noté aussi exactement que possible" [martian-first-seance-lemaitre]
- CONFIRMED | Flournoy prints the fragments sans garantie d'exactitude absolue (with no guarantee of absolute exactness) | https://archive.org/download/desindeslaplan00flou/desindeslaplan00flou_djvu.txt | p.151: "tous les fragments ... sans garantie d'exactitude absolue, cela va de soi" [martian-first-seance-no-guarantee]
- CONFIRMED | Judged by the four words whose meaning was fixed that first evening, he says, Martian is plainly une puérile contrefaçon du français (a childish counterfeit of French). | https://archive.org/download/desindeslaplan00flou/page/n170_w1400.jpg | p.151 (image; OCR agrees): the four words fixed "avec une entière certitude" that evening; "Jugée sur ces derniers ... une puérile contrefaçon du français" [martian-first-seance-contrefacon]
- CONFIRMED | Of the rest he was inclined to think it only un pseudo-martien, une suite de sons quelconques proférés au hasard et sans signification réelle (a pseudo-Martian, a string of sounds of any kind uttered at random and with no real [...] | https://archive.org/download/desindeslaplan00flou/page/n170_w1400.jpg | pp.151-152: "On serait porté à croire que ... n'était qu'un pseudo-martien ..."; a supposition he then gives "quelque appui" [martian-first-seance-pseudo]
- CONFIRMED | One sentence, all the ultra-Martian the book gives | https://archive.org/download/desindeslaplan00flou/desindeslaplan00flou_djvu.txt | p.252: "A ce qui précède se sont jusqu'ici bornés les messages ultramartiens." [ultramartian-all]
- CONFIRMED | the name is Flournoy's | https://archive.org/download/desindeslaplan00flou/desindeslaplan00flou_djvu.txt | p.250: "un cycle ultramartien si l'on me permet cette expression" [ultramartian-name]
- CONFIRMED | Hélène Smith heard it on 2 November 1898 from the strange beings of a vision. | https://archive.org/download/desindeslaplan00flou/page/n238_w1400.jpg | p.219 (image; NOTES.md): heard "le 2 novembre par les êtres étranges du tableau de la vision"; year 1898 in the same note [ultramartian-heard]
- CONFIRMED | On 18 December Astané, speaking by her mouth, followed each word with its Martian, and Esenale put the Martian into French. | https://archive.org/download/desindeslaplan00flou/page/n238_w1400.jpg | p.219: Astané "parlant par sa bouche ... dans la séance du 18 décembre 1898", then Esenale "en la traduisant mot à mot" [ultramartian-given]
- CONFIRMED | It broke out seventeen days after remarks of Flournoy's on what Martian owed to French | https://archive.org/download/desindeslaplan00flou/desindeslaplan00flou_djvu.txt | p.250: "le 2 novembre (soit 17 jours après les suggestions ...)"; p.253: remarks on its ch and syntax borrowed from French [ultramartian-seventeen]
- CONFIRMED | and he read it as an answer to them | https://archive.org/download/desindeslaplan00flou/desindeslaplan00flou_djvu.txt | p.253: "la préoccupation de répondre à mes remarques du 16 octobre" [ultramartian-answer]
- CONFIRMED | et voici une langue absolument nouvelle, d'un rythme très particulier, extrêmement riche en a, sans aucun ch jusqu'ici, et dont la construction est tellement différente de la nôtre qu'il n'y a pas moyen de s'y retrouver. (and [...] | https://archive.org/download/desindeslaplan00flou/page/n272_w1400.jpg | p.253 (kept image cards/flournoy-p253-n272.jpg): sentence as quoted [ultramartian-said]
- CONFIRMED | What Flournoy calls l'hindou d'Hélène, first heard on 6 March 1895. | https://archive.org/download/desindeslaplan00flou/desindeslaplan00flou_djvu.txt | p.294: "tel paraît donc être l'hindou d'Hélène" [sanscritoid-first]
- CONFIRMED | first heard on 6 March 1895 | https://archive.org/download/desindeslaplan00flou/desindeslaplan00flou_djvu.txt | p.293: "atiêyâ ganapatinâmâ, qui ont inauguré la langue hindoue le 6 mars 1895" [sanscritoid-date]
- CONFIRMED | No word-for-word translation and no written text of it could be had. | https://archive.org/download/desindeslaplan00flou/desindeslaplan00flou_djvu.txt | p.293: "il n'a jamais été possible d'en obtenir ni traduction littérale ni textes écrits" [sanscritoid-no-translation]
- OBSERVED | It is not a clean invented tongue: the isolated words the book calls pure Sanskrit are left out, but Sanskrit inside the utterances is not. | research/the-accent-of-nowhere/tongues/martian/sanscritoid.json | selection_rule; recreated: none of radyiva, pitaram, sumanas, smayamana is among its 119 words [sanscritoid-mixture]
- CONFIRMED | Un mélange d'articulations improvisées, et de véritables mots sanscrits adaptés à la situation (A mixture of improvised articulations and of genuine Sanskrit words fitted to the situation) Théodore Flournoy, Des Indes à la [...] | https://archive.org/download/desindeslaplan00flou/page/n313_w1400.jpg | p.294 (kept image cards/flournoy-p294-n313.jpg); said of the first sample, "Les spécimens ultérieurs ne feront que corroborer cette impression" [sanscritoid-flournoy-said]
- CONFIRMED | c'est un mêli-mêlo de syllabes, au milieu desquelles il y a incontestablement des suites de huit à dix syllabes donnant un fragment de phrase ayant un sens (it is a jumble of syllables, in the midst of which there are [...] | https://archive.org/download/desindeslaplan00flou/desindeslaplan00flou_djvu.txt | p.303: "1° Que c'est un mêli-mêlo de syllabes ..." in de Saussure's letter as Flournoy gives it [sanscritoid-saussure-said]
- CONFIRMED | Two short texts are all the Uranian there is, and Flournoy says he does not know what they mean. | https://archive.org/download/nouvellesobserv00flou/nouvellesobserv00flou_djvu.txt | p.155: "Ces deux derniers textes constituent tout ce que nous possédons en fait d'uranien, et nous n'en connaissons pas la signification." [smith-uranian-all]
- CONFIRMED | Hélène Smith heard the first and repeated it during a vision of Uranus at the seance of 17 June 1900. | https://archive.org/download/nouvellesobserv00flou/nouvellesobserv00flou_djvu.txt | p.155: "48. Texte uranien entendu et répété par Hélène au cours d'une vision d'Uranus dans la séance du 17 juin 1900" [smith-uranian-first]
- CONFIRMED | The second, of 2 August 1900, she heard and took down in pencil, then wrote automatically in unknown characters. | https://archive.org/download/nouvellesobserv00flou/nouvellesobserv00flou_djvu.txt | p.155: "49. ... du 2 août 1900; d'abord entendu et recueilli au crayon ... puis écrit automatiquement ... en caractères inconnus" [smith-uranian-second]
- CONFIRMED | Flournoy printed them in 1901, in a paper that followed his book | https://archive.org/download/nouvellesobserv00flou/nouvellesobserv00flou_djvu.txt | offprint head: "Archives de Psychologie de la Suisse Romande, Tome Ier, n° 2 (Décembre 1901)" [smith-uranian-paper]
- OBSERVED | the study read them on a page image | research/the-accent-of-nowhere/tongues/glossolalia/smith-uranian.json | transcription_notes.method: read on the kept page image of p.155, both texts proofread [smith-uranian-read]
- CONFIRMED | He set the tongue beside her others, in our rendering: les mots d'Uranus varient énormément en longueur (nous en avons de 2 et de 11 lettres), se terminent sans exception par une voyelle, et paraissent résulter d'une [...] | https://archive.org/download/nouvellesobserv00flou/page/n91_w1400.jpg | p.188 (image; NOTES.md): sentence as quoted; it goes on "où la structure des syllabes est beaucoup plus variée" [smith-uranian-said]
- CONFIRMED | Albert Le Baron is William James's name for “a literary man, aged 39” | https://archive.org/download/proceedingsofsoc12soci/proceedingsofsoc12soci_djvu.txt | p.278: "Mr. Le Baron, who is a literary man, aged 39"; p.277 "as I will call the gentleman"; p.279 "The names are changed" [le-baron-name]
- CONFIRMED | whose narrative James communicated to the Society for Psychical Research | https://archive.org/download/proceedingsofsoc12soci/proceedingsofsoc12soci_djvu.txt | p.277: "By Albert Le Baron. Communicated by Professor William James." [le-baron-communicated]
- CONFIRMED | By his own account his automatic speech one day “changed abruptly off from English into unintelligible sounds resembling a foreign tongue” | https://archive.org/download/proceedingsofsoc12soci/proceedingsofsoc12soci_djvu.txt | p.289: "it changed abruptly off from English into unintelligible sounds resembling a foreign tongue" (Sunday, 30 September 1894) [le-baron-changed]
- CONFIRMED | the messages printed are dated 1 October 1894 | https://archive.org/download/proceedingsofsoc12soci/proceedingsofsoc12soci_djvu.txt | p.290: "On Monday, October 1st, 1894 ... came the following messages"; no other date stands over the printed messages (the poem and notebook list follow) [le-baron-dated]
- CONFIRMED | He pencilled the words down himself as they came and worked on the spelling afterwards. | https://archive.org/download/proceedingsofsoc12soci/page/n300_w1100.jpg | p.293 (image; OCR agrees): "As the foreign verbiage came viva voce, I pencilled it down ... the subsequent blending ... conjectural euphony" [le-baron-pencilled]
- CONFIRMED | What is printed is a selection. | https://archive.org/download/proceedingsofsoc12soci/proceedingsofsoc12soci_djvu.txt | p.293: "The foregoing "foreign" paragraphs are not all that were given by any means" [le-baron-selection]
- OBSERVED | The article does not name his mother tongue. | research/the-accent-of-nowhere/tongues/glossolalia/NOTES.md | recreated: kept OCR of pp.277-297 searched for mother/native/vernacular; "mother" occurs only for his and Evangel's mothers [le-baron-no-mother-tongue]
- CONFIRMED | James's judgment: The phonetic elements in his case again seemed English; and I tried to make him believe (but all in vain) that the whole thing was a decidedly rudimentary form of motor automatism analogous to the scrawls and [...] | https://archive.org/download/proceedingsofsoc12soci/page/n285_w1100.jpg | p.278 (image; kept OCR line 16418 agrees): sentence as quoted [le-baron-james]
- CONFIRMED | William James, Proceedings of the Society for Psychical Research 12, 1897, p. 278 | https://archive.org/download/proceedingsofsoc12soci/proceedingsofsoc12soci_djvu.txt | "VOLUME XII ... 1896-7"; imprint "London : KEGAN PAUL ... 1897" [le-baron-volume]
- CONFIRMED | Le Baron gave up his experiments about the end of April 1895, having failed, in his words, to “substantiate the verbiage as an actual language”. | https://archive.org/download/proceedingsofsoc12soci/proceedingsofsoc12soci_djvu.txt | p.295: "I abandoned my experiments about the end of April, 1895 ... I could not, and did not substantiate the verbiage as an actual language" [le-baron-gave-up]
- CONFIRMED | Simon, as Oskar Pfister calls him, was 24 when the analysis began | https://archive.org/download/jb319111/JB_III_1911_1_djvu.txt | p.437: "Simon, das älteste Kind, war zu Beginn der Analyse 24 Jahre alt." [pfister-simon-age]
- CONFIRMED | a small congregation given to speaking in tongues had engaged him as its preacher | https://archive.org/download/jb319111/JB_III_1911_1_djvu.txt | p.443: "ließ sich Simon von einer kleinen religiösen Gemeinde, die wie die Pfingstbewegung der Zungenrede oblag, als Prediger anstellen" [pfister-simon-preacher]
- CONFIRMED | Pfister, a pastor in Zurich | https://archive.org/download/jb319122/JB_III_1912_2_djvu.txt | byline p.730: "Von Dr. Oskar Pfister, Pfarrer in Zürich." [pfister-simon-pastor]
- CONFIRMED | took down his first speech by ear, repeating the words aloud as he wrote | https://archive.org/download/jb319111/page/n449_w1400.jpg | p.444, kept image jb3-1-n449.jpg viewed: "Bei vollem Bewußtsein spricht er hierauf rasch, von mir aber beständig durch mein Nachsprechen während des Schreibens unterbrochen" [pfister-simon-by-ear]
- CONFIRMED | Of the four speeches printed, the last two are Simon's readings of signs he had just written automatically. | https://archive.org/download/jb319111/JB_III_1911_1_djvu.txt | p.459: "Glossolalie trat fortan nur noch als Ablesung automatischer Schriftzeichen auf."; third and fourth speeches read from signs [pfister-simon-readings]
- CONFIRMED | They were printed in 1911. | https://archive.org/download/jb319111/JB_III_1911_1_djvu.txt | title page of the first half-volume: "FRANZ DEUTICKE. 1911." [pfister-simon-1911]
- OBSERVED | The article names no mother tongue for Simon | research/the-accent-of-nowhere/tongues/glossolalia/NOTES.md | scout's reading: "No sentence names his mother tongue" [pfister-simon-no-mother-tongue]
- CONFIRMED | though it explains one of his words from Swiss German | https://archive.org/download/jb319111/JB_III_1911_1_djvu.txt | p.454, on auenfang: "Schweizerdeutsch für „auch ein Fang“." [pfister-simon-swiss]
- CONFIRMED | Pfister's conclusion, printed the next year (our rendering): Die zweite Art von Glossolalie dürfen wir als ungrammatische oder imaginative Zungenrede bezeichnen. Wir fanden sie bei Simon und Frau T. [...] wobei jede grammatische [...] | https://archive.org/download/jb319122/page/n305_w1400.jpg | p.780 (image; NOTES.md): "Die zweite Art ... Wir fanden sie bei Simon und Frau T. ... gänzlich fehlt." [pfister-simon-said]
- CONFIRMED | Lina, as Pfister calls her, is Simon's sister, about sixteen. | https://archive.org/download/jb319122/JB_III_1912_2_djvu.txt | p.769, on a visit to Simon's family: "eine etwa 16jährige Schwester Lina" [pfister-lina-who]
- CONFIRMED | At a family prayer, with all kneeling and Pfister writing at the table | https://archive.org/download/jb319122/JB_III_1912_2_djvu.txt | p.769: "Zuletzt entschließt sich die Hausmutter zu einem Gebet. Alle knien und ich schreibe am Tische nach." [pfister-lina-prayer]
- CONFIRMED | she spoke up as soon as Frau T. had ended | https://archive.org/download/jb319122/JB_III_1912_2_djvu.txt | p.770: "Kaum hatte sie ihre stark skandierten Verse beendet, als Lina sich vernehmen ließ" [pfister-lina-after]
- CONFIRMED | It was printed in 1912 | https://archive.org/download/jb319122/JB_III_1912_2_djvu.txt | title page of the second half-volume: "FRANZ DEUTICKE. 1912." [pfister-lina-1912]
- OBSERVED | the study read it on a page image | research/the-accent-of-nowhere/tongues/glossolalia/pfister-lina.json | transcription_notes.method: read on the kept page image of p.770, proofread 100 percent [pfister-lina-read]
- CONFIRMED | Pfister explains pieces of it from the Zurich dialect | https://archive.org/download/jb319122/JB_III_1912_2_djvu.txt | p.776: "stik = stiegst (Zürcherdeutsch für „steigst“)" [pfister-lina-zurich]
- CONFIRMED | and holds that two wishes shaped it, to get to heaven and to speak English | https://archive.org/download/jb319122/JB_III_1912_2_djvu.txt | p.778: "Die Wünsche in den Himmel zu kommen und Englisch zu reden, determinieren das Gebilde." [pfister-lina-wishes]
- CONFIRMED | His conclusion, in our rendering: Die erste ist als automatische Verzerrung einer normalen Sprache, kurz ausgedrückt, als automatische Verzerrungssprache zu beurteilen. Wir fanden sie bei Lina an das Englische anklingend, doch [...] | https://archive.org/download/jb319122/page/n305_w1400.jpg | p.780 (image; NOTES.md): "Die erste ist als automatische Verzerrung ... Wir fanden sie bei Lina an das Englische anklingend" [pfister-lina-said]
- CONFIRMED | Frau T., as Pfister calls her, is a friend of Simon's family, about 45. | https://archive.org/download/jb319122/JB_III_1912_2_djvu.txt | p.769: "stattete ich seiner Familie einen Besuch ab ... Ich traf die Mutter, ... Lina und eine zirka 45jährige Freundin, Frau T." [pfister-frau-t-who]
- CONFIRMED | At a family prayer she first whispered, and Pfister caught the words that kept returning. | https://archive.org/download/jb319122/JB_III_1912_2_djvu.txt | p.769: "Kaum ist das Amen gesprochen, so beginnt sie zu flüstern. Ich verstehe die immer wiederkehrenden Worte" [pfister-frau-t-whispered]
- CONFIRMED | Then, with leave to repeat her words aloud so as to write them down rightly, he took down seven strongly scanned verses. | https://archive.org/download/jb319122/page/n295_w1400.jpg | p.770 (image; OCR agrees): permission asked, "Es wird freundlich gewährt"; seven verse lines follow [pfister-frau-t-leave]
- OBSERVED | The article states no language for her. | research/the-accent-of-nowhere/tongues/glossolalia/pfister-frau-t.json | home field: "UNSOURCED in so many words ... the article states no language for her" [pfister-frau-t-no-language]
- CONFIRMED | Pfister puts her with Simon in his second kind of glossolalia, ungrammatische oder imaginative Zungenrede (ungrammatical or imaginative speaking in tongues, in our rendering). | https://archive.org/download/jb319122/page/n305_w1400.jpg | p.780: "ungrammatische oder imaginative Zungenrede ... bei Simon und Frau T." [pfister-frau-t-kind]
- CONFIRMED | Archibald M'Kerrell says he collected it “in the beginning of this present year, (1831,)” from one person he heard speak in a tongue on different occasions, writing on the spot. | https://archive.org/download/amorningvisitto00antigoog/page/n37_w1400.jpg | p.35 (image; NOTES.md): "collected by me, on different occasions, in the beginning of this present year, (1831,) ... written down by me on the spot" [mkerrell-1831-collected]
- CONFIRMED | A vocabulary, not running speech. | https://archive.org/download/amorningvisitto00antigoog/amorningvisitto00antigoog_djvu.txt | p.35: "this specimen is a very brief vocabulary of a tongue" [mkerrell-1831-vocabulary]
- CONFIRMED | The words, he says, “are widely scattered; none in the order they were spoken, except those marked within inverted commas”. | https://archive.org/download/amorningvisitto00antigoog/page/n37_w1400.jpg | p.35 (image; kept OCR agrees): "are widely scattered; none in the order they were spoken, except those marked within inverted commas" [mkerrell-1831-scattered]
- CONFIRMED | His pamphlet, cited as Apology for the Gift of Tongues, Greenock, 1831 | https://archive.org/download/amorningvisitto00antigoog/page/n38_w1400.jpg | p.36 (image; kept OCR agrees): "Apology for the Gift of Tongues, &c. by Archibald M'Kerrell, Esq. Greenock, 1831." [mkerrell-1831-pamphlet]
- OBSERVED | was not obtained | research/the-accent-of-nowhere/tongues/glossolalia/NOTES.md | "M'Kerrell's pamphlet itself was not obtained (not found on archive.org by a search of titles and authors)" [mkerrell-1831-not-obtained]
- CONFIRMED | The study reads the specimen in a reprint of 1832 by an opponent who signs himself Anti-Cabala. | https://archive.org/services/search/beta/page_production/?service_backend=fts&user_query=%22hippo-gerosto%22&hits_per_page=30&page=1 | IA record: creator "Anti-cabala", date 1832; kept OCR p.36 calls M'Kerrell's account "the very opposite of true" (the signature itself rests on the catalogue; title page OCR lacks it) [mkerrell-1831-reprint]
- OBSERVED | The reprint names neither the speaker nor the speaker's language | research/the-accent-of-nowhere/tongues/glossolalia/NOTES.md | "The speaker's name and mother tongue: UNSOURCED. The reprint gives neither."; kept OCR pp.35-36 agrees [mkerrell-1831-unnamed]
- OBSERVED | and does not say where the tongue was heard | research/the-accent-of-nowhere/sources/kept/glossolalia/anti-cabala-1832_djvu.txt | recreated: grep for Scotland/Glasgow/Clyde/Greenock finds only "Greenock, 1831" in the citation line (positive control); appendix names no place of hearing [mkerrell-1831-no-place]
- CONFIRMED | Three pieces are pooled. | research/the-accent-of-nowhere/tongues/pooled/ball-1916.json | members: gadji-beri-bimba, karawane, totenklage [ball-1916-pooled]
- CONFIRMED | The sample is from Karawane, read whole in German Wikisource's transcription of the Dada Almanach (Berlin, 1920). | https://de.wikisource.org/wiki/Karawane | Textdaten: "Dada Almanach. Berlin: Erich Reiss Verlag, 1920, S. 53" [ball-1916-karawane]
- OBSERVED | Of the verses beginning gadji beri bimba the study has only the five lines Ball printed in his diary | research/the-accent-of-nowhere/tongues/sound-poems/ball-gadji-beri-bimba.json | selection_rule: the five lines Ball printed in his diary [ball-1916-gadji]
- OBSERVED | and of Totenklage only the lines on one page of De Stijl (1928), seen in a facsimile reprint: a fragment | research/the-accent-of-nowhere/tongues/sound-poems/ball-totenklage.json | selection_rule: lines on p.100 of De Stijl, a fragment, scan of the 1968 facsimile reprint [ball-1916-totenklage]
- OBSERVED | Three more poems it looked for were not obtained. | research/the-accent-of-nowhere/tongues/sound-poems/NOTES.md | "Not obtained: ball-wolken, ball-katzen-und-pfauen, ball-seepferdchen-und-flugfische" [ball-1916-missing]
- CONFIRMED | Ball's diary, in an entry dated 23 June, in a chapter that opens at Zürich on 2 February 1916 | https://digital.kunsthaus.ch/viewer/api/v1/records/18782/files/images/046_2.tif/full/1000,/0/default.jpg | S.77 margin (image; NOTES.md): "Zürich, 2. II. 1916."; June entry margin "23. VI." (S.105) [ball-1916-date]
- CONFIRMED | Ich habe eine neue Gattung von Versen erfunden, „Verse ohne Worte“ oder Lautgedichte, in denen das Balancement der Vokale nur nach dem Werte der Ansatzreihe erwogen und ausgeteilt wird. Die ersten dieser Verse habe ich heute [...] | https://digital.kunsthaus.ch/viewer/api/v1/records/18782/files/images/060_2.tif/full/1100,/0/default.jpg | S.105 (image; NOTES.md): "Ich habe eine neue Gattung von Versen erfunden ... heute abend vorgelesen." [ball-1916-said]
- CONFIRMED | The poem stands on page 9 of Christian Morgenstern's Galgenlieder (Berlin: Cassirer, 1905). | https://api.digitale-sammlungen.de/iiif/presentation/v2/bsb00087386/manifest | library record: Galgenlieder, Morgenstern, 1905, "Berlin : Cassirer"; page 9 read on the page image [morgenstern-book]
- OBSERVED | The study typed it from a page image of the Bayerische Staatsbibliothek's copy. | research/the-accent-of-nowhere/tongues/sound-poems/morgenstern-das-grosse-lalula.json | transcription_notes: typed from the BSB page image, no OCR, 15 of 15 lines read twice [morgenstern-typed]
- WRONG | Three times in that printing the letter m carries a bar; the later text that German Wikisource transcribes, of 1932, has none. | research/the-accent-of-nowhere/sources/kept/sound-poems/dewikisource-Das_grosse_Lalula.html | MINOR: the kept Wikisource page of the 1932 text marks all three: "Semememi" and "hulalemi"; the bars were lost when the scout stripped the markup [morgenstern-bar]
- CONFIRMED | In the introduction printed with that later text one of the gallows brothers “spricht das große Lalulā” (speaks the great Lalulā, in our rendering). | https://de.wikisource.org/wiki/Galgenlieder | "der Achte Faherügghh, mit dem Beinamen der Unselm: der kann das Simmaleins und spricht das große Lalulā" [morgenstern-speaks]
- OBSERVED | What the bar means, when the poem was written and what Morgenstern said of it are not sourced here. | research/the-accent-of-nowhere/tongues/sound-poems/NOTES.md | "UNSOURCED: what the bar over m means, when the poem was written, and what Morgenstern said about it" [morgenstern-unsourced]
- CONFIRMED | Paul Scheerbart printed the piece as intermezzo 57 of Ich liebe dich! (Berlin, 1897), a railway novel with 66 intermezzos. | https://de.wikisource.org/wiki/Paul_Scheerbart | "Ich liebe dich! Ein Eisenbahn-Roman mit 66 Intermezzos. Schuster & Loeffler, Berlin 1897"; contents line "57. Kikakokú! Ekoraláps! 249" [scheerbart-book]
- CONFIRMED | Its narrator reads it as a story that belongs voll und ganz, wholly, to a school of his own | https://archive.org/download/bub_gb_ZDhAAAAAYAAJ/page/n248_w1000.jpg | S.248 (image; NOTES.md): "eine interessante Geschichte, die der erwähnten Richtung „voll und ganz“ angehört" [scheerbart-voll]
- CONFIRMED | which he has just described (our rendering): Ich machte meinen Freund mit meiner ekoralápsischen Richtung bekannt, die das Verstandenwerdenwollen bekanntlich längst überwunden hat (I acquainted my friend with my Ekoralapsian [...] | https://archive.org/download/bub_gb_ZDhAAAAAYAAJ/page/n248_w1000.jpg | S.248 (image; NOTES.md): sentence as quoted, said by the narrator [scheerbart-said]
- OBSERVED | The study typed the piece from a page image of a low-resolution scan; its circumflexes want rechecking against a better one. | research/the-accent-of-nowhere/tongues/sound-poems/scheerbart-kikakoku.json | transcription_notes: low-resolution Google scan; eight circumflexes want rechecking [scheerbart-scan]
- CONFIRMED | Two things are pooled | research/the-accent-of-nowhere/tongues/pooled/khlebnikov.json | members: the Zangezi gods and Bobeobi [khlebnikov-pooled]
- CONFIRMED | the speeches of the gods in the second plane of Velimir Khlebnikov's Zangezi, written from 1920 to 1922 and printed in 1922 | https://ru.wikisource.org/wiki/%D0%97%D0%B0%D0%BD%D0%B3%D0%B5%D0%B7%D0%B8_(%D0%A5%D0%BB%D0%B5%D0%B1%D0%BD%D0%B8%D0%BA%D0%BE%D0%B2) | header: "Дата создания: 1920-1922, опубл.: 1922"; kept page: "Плоскость II ... БОГИ" [khlebnikov-zangezi]
- CONFIRMED | and the five sound-words of his poem Бобэоби пелись губы, written in 1908 to 1909 | https://ru.wikisource.org/wiki/%D0%91%D0%BE%D0%B1%D1%8D%D0%BE%D0%B1%D0%B8_%D0%BF%D0%B5%D0%BB%D0%B8%D1%81%D1%8C_%D0%B3%D1%83%D0%B1%D1%8B_(%D0%A5%D0%BB%D0%B5%D0%B1%D0%BD%D0%B8%D0%BA%D0%BE%D0%B2) | header: "Дата создания: 1908 - 1909." (the editors' date) [khlebnikov-bobeobi-date]
- CONFIRMED | and first printed in the collection Пощёчина общественному вкусу | https://ru.wikisource.org/wiki/%D0%91%D0%BE%D0%B1%D1%8D%D0%BE%D0%B1%D0%B8_%D0%BF%D0%B5%D0%BB%D0%B8%D1%81%D1%8C_%D0%B3%D1%83%D0%B1%D1%8B_(%D0%A5%D0%BB%D0%B5%D0%B1%D0%BD%D0%B8%D0%BA%D0%BE%D0%B2) | 1986 editors' note: "Пощёчина общественному вкусу, с. 7" listed as the first printing [khlebnikov-bobeobi-print]
- CONFIRMED | The study scored them in Latin letters; the sample is the Cyrillic. | research/the-accent-of-nowhere/tongues/sound-poems/khlebnikov-zangezi-gods.rom.json | scheme: cyrillic-ru, the Russian panel table [khlebnikov-latin]
- OBSERVED | Both follow Russian Wikisource, which gives an edition of 1986 in modern spelling. | research/the-accent-of-nowhere/tongues/sound-poems/khlebnikov-zangezi-gods.json | source.edition: Wikisource oldid 5348930, from Творения, 1986 [khlebnikov-wikisource-gods]
- OBSERVED | Both follow Russian Wikisource, which gives an edition of 1986 in modern spelling. | research/the-accent-of-nowhere/tongues/sound-poems/khlebnikov-bobeobi.json | source.edition: Wikisource oldid 5522963, checked against Творения 1986 (modern orthography) [khlebnikov-wikisource-bobeobi]
- OBSERVED | The sound-words were compared with a scan of the first printing | research/the-accent-of-nowhere/tongues/sound-poems/khlebnikov-bobeobi.json | transcription_notes: "Compared by the scout with the first printing (1912 scan, 7 of 7 lines)" [khlebnikov-bobeobi-compared]
- OBSERVED | the gods' speeches were not proofread against any printing | research/the-accent-of-nowhere/tongues/sound-poems/khlebnikov-zangezi-gods.json | transcription_notes: "Not proofread against any printing" [khlebnikov-not-proofread]
- CONFIRMED | Khlebnikov's later note on the sound-words, as that edition's editors quote it (our rendering): Б, или ярко-красный цвет, а потому губы - бобэоби, вээоми - синий <...>, пиээо - чёрное (B, or the colour bright red, and so the lips [...] | https://ru.wikisource.org/wiki/%D0%91%D0%BE%D0%B1%D1%8D%D0%BE%D0%B1%D0%B8_%D0%BF%D0%B5%D0%BB%D0%B8%D1%81%D1%8C_%D0%B3%D1%83%D0%B1%D1%8B_(%D0%A5%D0%BB%D0%B5%D0%B1%D0%BD%D0%B8%D0%BA%D0%BE%D0%B2) | editors' note: "Ср. поздний автокомментарий: «Б, или ярко-красный цвет ...»" (second hand, labelled so) [khlebnikov-said]
- CONFIRMED | The first of three poems in Aleksei Kruchenykh's book Помада, written in 1912 and published in 1913. | https://ru.wikisource.org/wiki/%D0%94%D1%8B%D1%80_%D0%B1%D1%83%D0%BB_%D1%89%D1%8B%D0%BB_(%D0%9A%D1%80%D1%83%D1%87%D1%91%D0%BD%D1%8B%D1%85) | header: "сб. «Помада». Дата создания: 1912, опубл.: 1913."; poem numbered №1 of "3 стихотворения" [kruchenykh-book]
- OBSERVED | The study typed it from a photograph of the handwritten page and scored it in Latin letters. | research/the-accent-of-nowhere/tongues/sound-poems/kruchenykh-dyr-bul-shchyl.json | transcription_notes: typed from the kept photograph, 5 of 5 lines read twice [kruchenykh-typed]
- CONFIRMED | Kruchenykh died in 1968 | https://ru.wikisource.org/wiki/%D0%94%D1%8B%D1%80_%D0%B1%D1%83%D0%BB_%D1%89%D1%8B%D0%BB_(%D0%9A%D1%80%D1%83%D1%87%D1%91%D0%BD%D1%8B%D1%85) | "Алексей Елисеевич Кручёных (1886-1968)" [kruchenykh-died]
- OBSERVED | and only these five lines are taken | research/the-accent-of-nowhere/tongues/sound-poems/kruchenykh-dyr-bul-shchyl.json | selection_rule: five lines and the six-line heading, nothing more [kruchenykh-five-lines]
- CONFIRMED | Above the poem, in the same hand, he says what the language is (our rendering): 3 стихотворенiя / написаныя на / собственом языкѣ / От др. отличается[:] / слова его не имѣют / опредѣленаго значенiя (3 poems written in my own [...] | https://upload.wikimedia.org/wikipedia/commons/3/3a/Kruchenykh-Larionov1.jpg | handwritten heading (kept image; NOTES.md); Wikisource's modernized heading agrees in words [kruchenykh-said]
- OBSERVED | Every word that Gulliver's Travels presents as a word of the language of a country Gulliver visits, the names of persons and places included, roughly in the order of the book. | research/the-accent-of-nowhere/tongues/nonsense-english/swift-gulliver.json | selection_rule as the card gives it [swift-gulliver-what]
- CONFIRMED | The Travels came out in London in 1726 under Gulliver's own name. | https://en.wikisource.org/wiki/Travels_into_Several_Remote_Nations_of_the_World | title page: "By LEMUEL GULLIVER ... LONDON: Printed for Benj. Motte ... M, DCC, XXVI." [swift-gulliver-1726]
- OBSERVED | The study could not use that first text, which Wikisource holds only as an unfinished index. | research/the-accent-of-nowhere/tongues/nonsense-english/swift-gulliver.json | source.edition: the 1726 text is "an unfinished index with no chapter pages"; kept page marked incomplete [swift-gulliver-not-1726]
- CONFIRMED | Its list follows volume 6 of the 1801 Works, in a Wikisource transcription | https://en.wikisource.org/wiki/The_Works_of_the_Rev._Jonathan_Swift/Volume_6/Gulliver%27s_Travels | "The Works of the Rev. Jonathan Swift (1801)" [swift-gulliver-1801]
- CONFIRMED | the spellings shift between editions: the 1801 print's Lapuntah and Flute drin are Lapuntuh and Fluft drin in the Project Gutenberg text | research/the-accent-of-nowhere/tongues/nonsense-english/swift-gulliver.json | "Lapuntah / Lapuntuh ... Flute drin / Fluft drin"; recreated: kept pg829.txt has Lapuntuh and Fluft drin, no Lapuntah [swift-gulliver-spellings]
- OBSERVED | No statement by Swift in his own person on how he made the words was fetched. | research/the-accent-of-nowhere/tongues/nonsense-english/NOTES.md | "Any statement by Swift in his own person ... UNSOURCED, none fetched" [swift-gulliver-no-statement]
- CONFIRMED | Gulliver, on the Houyhnhnms: In speaking, they pronounce through the nose and throat, and their language approaches nearest to the High-Dutch, or German, of any I know in Europe; but is much more graceful and significant. [...] | https://en.wikisource.org/wiki/The_Works_of_the_Rev._Jonathan_Swift/Volume_6/A_Voyage_to_the_Country_of_the_Houyhnhnms/Chapter_3 | Part IV ch.3, sentence as quoted, spoken by Gulliver [swift-gulliver-houyhnhnm]
- OBSERVED | Passages that H. P. Lovecraft's stories give as speech in a non-human or pre-human tongue, found by searching the fiction that English Wikisource lists as his own. | research/the-accent-of-nowhere/tongues/nonsense-english/lovecraft-utterances.json | selection_rule: 66 pieces listed on the Wikisource author page searched; non-human or pre-human tongues included [lovecraft-what]
- CONFIRMED | The stories were first printed between 1924 and 1943. | https://en.wikisource.org/wiki/Weird_Tales/Volume_3/Issue_3/The_Rats_in_the_Walls | "Weird Tales (vol. 3, no. 3) (March, 1924)" [lovecraft-1924]
- CONFIRMED | The stories were first printed between 1924 and 1943. | https://en.wikipedia.org/wiki/The_Case_of_Charles_Dexter_Ward | "first complete publication ... Beyond the Wall of Sleep collection (1943)"; the 1943 end of the range is first printing for The Dream-Quest of Unknown Kadath (Wikisource "(1943)"), Ward was first printed abridged in 1941 [lovecraft-1943]
- OBSERVED | Passages in real languages are left out | research/the-accent-of-nowhere/tongues/nonsense-english/lovecraft-utterances.json | left_out: Latin, Hebrew and Greek divine names, Gaelic [lovecraft-left-out]
- CONFIRMED | among them chants that Wikipedia says were lifted from older books | https://en.wikipedia.org/wiki/The_Horror_at_Red_Hook | "Much of the occult chanting in the story was lifted from the articles on "Magic" and "Demonology" in the 9th edition of the Encyclopædia Britannica" [lovecraft-lifted]
- OBSERVED | Some are read in Wikisource transcriptions of first printings in Weird Tales and were proofread against the scans; the passages from Wikisource pages with no scan behind them were not. | research/the-accent-of-nowhere/tongues/nonsense-english/lovecraft-utterances.json | transcription_notes: 18 passages on scan-backed pages proofread; others not [lovecraft-witnesses]
- CONFIRMED | In The Call of Cthulhu the words come as an attempt to write down a sound: a voice that was not a voice; a chaotic sensation which only fancy could transmute into sound, but which he attempted to render by the almost [...] | https://en.wikisource.org/wiki/Weird_Tales/Volume_11/Issue_2/The_Call_of_Cthulhu | Weird Tales Feb 1928: sentence as quoted [lovecraft-said]
- CONFIRMED | In chapter IX of Pantagruel Panurge answers in one tongue after another until, asked whether he cannot speak French, he says it is “ma langue naturelle” and his mother tongue. | https://fr.wikisource.org/wiki/Les_%C5%92uvres_de_Fran%C3%A7ois_Rabelais_(%C3%89ditions_Marty-Laveaux)/Pantagruel/9 | "ne sçauez vous parler Françoys? ... c'est ma langue naturelle & maternelle" [rabelais-panurge-chapter]
- CONFIRMED | Three of his speeches are taken as invented, on the word of the editor Charles Marty-Laveaux | https://fr.wikisource.org/wiki/Les_%C5%92uvres_de_Fran%C3%A7ois_Rabelais_(%C3%89ditions_Marty-Laveaux)/Pantagruel/9 | editor's notes: "pure mystification", "encore un langage imaginaire", "Langage inintelligible" [rabelais-panurge-invented-1]
- CONFIRMED | and of the text itself, where the listeners take them for the speech of the Antipodes | https://fr.wikisource.org/wiki/Les_%C5%92uvres_de_Fran%C3%A7ois_Rabelais_(%C3%89ditions_Marty-Laveaux)/Pantagruel/9 | Epistemon: "Ie croy que c'est langaige des Antipodes" [rabelais-panurge-antipodes]
- CONFIRMED | the speech of the Antipodes (the sample) | research/the-accent-of-nowhere/tongues/nonsense-romance/rabelais-panurge-antipodes.json | text begins "Al barildim gotfano dech min brin ..." [rabelais-panurge-first]
- CONFIRMED | for langaige lanternoys | https://archive.org/download/lesoeuvresdemais01rabeuoft/page/n280.jpg | p.261 (image; kept Wikisource agrees): "Parlez vous christian ... ou langaige patelinoys? Non, c'est langaige lanternoys. Dont dist Panurge." (a listener's answer) [rabelais-panurge-lanternoys]
- CONFIRMED | and for the language of Utopia | https://fr.wikisource.org/wiki/Les_%C5%92uvres_de_Fran%C3%A7ois_Rabelais_(%C3%89ditions_Marty-Laveaux)/Pantagruel/9 | Pantagruel: "ou c'est langaige de mon pays de Vtopie, ou bien luy ressemble quant au son" [rabelais-panurge-utopie]
- OBSERVED | They are read from page images of Marty-Laveaux's edition (Paris, 1868) | research/the-accent-of-nowhere/tongues/nonsense-romance/rabelais-panurge-antipodes.json | transcription_notes.method: read by eye from the kept page images; Wikisource header "Alphonse Lemerre, 1868" [rabelais-panurge-read]
- CONFIRMED | which reproduces the Lyon edition of 1542 | https://archive.org/download/lesoeuvresdemais01rabeuoft/page/n232_w700.jpg | reprinted title page (image; NOTES.md): "M. D. XLII. / On les vend a Lyon chez Françoys / Juste" [rabelais-panurge-juste]
- CONFIRMED | The oldest known edition bears no date | https://fr.wikisource.org/wiki/Les_%C5%92uvres_de_Fran%C3%A7ois_Rabelais_(%C3%89ditions_Marty-Laveaux)/Pantagruel | "la plus ancienne édition connue est sans date et la première datée de 1533" [rabelais-panurge-no-date]
- CONFIRMED | French Wikisource catalogues the work under 1532 | https://fr.wikisource.org/wiki/Pantagruel | "Date de publication 1532" [rabelais-panurge-1532]
- CONFIRMED | The second of the three is not in the first editions. | https://fr.wikisource.org/wiki/Les_%C5%92uvres_de_Fran%C3%A7ois_Rabelais_(%C3%89ditions_Marty-Laveaux)/Pantagruel/9 | "Discours anglais, qui manque dans les premières éditions, ainsi que les deux suivants" (the Basque and the Lanternois) [rabelais-panurge-later]
- CONFIRMED | The editor on the first (our rendering): C’est ici une pure mystification. Burgaud des Marets est parvenu à décomposer tout ce passage en mots anglais « All bar ill dim..., » au milieu desquels figure le nom de Chinon. Ces mots, [...] | https://fr.wikisource.org/wiki/Les_%C5%92uvres_de_Fran%C3%A7ois_Rabelais_(%C3%89ditions_Marty-Laveaux)/Pantagruel/9 | vol.4 p.188 note: "C'est ici une pure mystification ... ne présentent aucun sens."; IA record dates vol.4 1881 [rabelais-panurge-said]
- CONFIRMED | printed with a Utopian alphabet in Thomas More's Utopia from its first edition (Louvain, 1516) | https://archive.org/download/utopiaofsirthoma00moreuoft/utopiaofsirthoma00moreuoft_djvu.txt | Lupton p.lxiv: "printed by Thierry Martin, at Louvain, towards the end of 1516"; second leaf has the alphabet and the Tetrastichon [more-utopian-first-edition]
- CONFIRMED | Four verses in the Utopian tongue | https://archive.org/download/utopiaofsirthoma00moreuoft/page/n100.jpg | plate (image; NOTES.md): "TETRASTICHON VERNACVLA VTOPIENSIVM LINGVA." [more-utopian-four-verses]
- OBSERVED | The study read them in a reduced facsimile of the woodcut page of the edition of March 1518, in J. H. Lupton's edition of 1895. No image of the 1516 page was obtained. | research/the-accent-of-nowhere/tongues/nonsense-romance/more-utopian.json | doubts: "a reduced facsimile of the March 1518 Basel page ... no image of the 1516 edition was obtained" [more-utopian-facsimile]
- CONFIRMED | Peter Giles says, in a letter printed with the book, that he had them added. The English is the old translation Lupton prints with the Latin: Tantum tetrastichum uernacula Vtopiensium lingua scriptum, quod a Mori discessu forte [...] | https://archive.org/download/utopiaofsirthoma00moreuoft/utopiaofsirthoma00moreuoft_djvu.txt | p.xcviii: "Tantum tetrastichum ... apponendum curaui, praefixo eiusdem gentis alphabeto" [more-utopian-giles]
- CONFIRMED | (Onely a meter of .iiij. verses written in the Vtopian tongue, whiche after master Mores departure Hythloday by chaunce shewed me, that haue I caused to be added thereto, with the Alphabete of the same nation) Peter Giles to [...] | https://archive.org/download/utopiaofsirthoma00moreuoft/page/n105.jpg | p.xcviii (image; NOTES.md): "Onely a meter of .iiij. verses ... with the Alphabete of the same nation" [more-utopian-giles-english]
- CONFIRMED | Lupton on the words: “Some of the words, as ‘gymnosophon,’ are evidently reminiscences of Greek; but I have not discovered any key to them, if indeed they were ever intended to have any meaning.” | https://archive.org/download/utopiaofsirthoma00moreuoft/page/n99.jpg | p.xciv (image; NOTES.md): sentence as quoted [more-utopian-lupton]
- CONFIRMED | The cry Dante gives Nimrod at Inferno XXXI, 67 | https://archive.org/download/tutteleoperedida00dant/page/n60.jpg | p.45 (image; NOTES.md): "'Rafel mai amech zabi almi,' / Cominciò a gridar la fiera bocca", numbered 67; "Questi è Nembrotto" nine lines on [dante-nimrod-line]
- CONFIRMED | read on a page image of the Oxford edition of 1904 | https://archive.org/metadata/tutteleoperedida00dant | IA record: "Oxford : Nella stamperia dell' università", 1904 [dante-nimrod-edition]
- OBSERVED | No second edition of the line was read, so no variant is given. | research/the-accent-of-nowhere/tongues/nonsense-romance/dante-nimrod.json | doubts: "no second edition of Inferno XXXI was read in this session" [dante-nimrod-no-variant]
- CONFIRMED | Dante lived from 1265 to 1321 | https://en.wikisource.org/wiki/1911_Encyclop%C3%A6dia_Britannica/Dante | "Dante (or Durante) Alighieri (1265-1321)" [dante-nimrod-dates]
- OBSERVED | no date for the Inferno was sourced | research/the-accent-of-nowhere/tongues/nonsense-romance/NOTES.md | "The date of composition of the Inferno: UNSOURCED" [dante-nimrod-undated]
- CONFIRMED | Virgil, in the poem, on what the cry is worth (our rendering): Questi è Nembrotto, per lo cui mal coto / Pure un linguaggio nel mondo non s’ usa. / Lasciamlo stare, e non parliamo a voto: / Chè così è a lui ciascun linguaggio, / [...] | https://archive.org/download/tutteleoperedida00dant/page/n60.jpg | p.45 (image; NOTES.md): XXXI 77-81 as quoted, spoken by "il Duca mio", Virgil per EB1911 [dante-nimrod-virgil]
- CONFIRMED | The line that opens canto VII of the Inferno. Pluto, as the edition prints the name, begins it colla voce chioccia (with his clucking voice, in our rendering). | https://archive.org/download/tutteleoperedida00dant/page/n24.jpg | p.9 (image; NOTES.md): "'Pape Satan, pape Satan aleppe,' / Cominciò Pluto colla voce chioccia." [dante-plutus-line]
- OBSERVED | In the lines read the poem says nothing of what the words mean. | research/the-accent-of-nowhere/tongues/nonsense-romance/NOTES.md | "says nothing of the words' meaning in these three lines" [dante-plutus-silent]
- CONFIRMED | It is read on a page image of the Oxford edition of 1904. | https://archive.org/metadata/tutteleoperedida00dant | IA record: Oxford, 1904 [dante-plutus-edition]
- CONFIRMED | Editions differ in its accents and punctuation: Italian Wikisource has Pape Satàn, pape Satàn aleppe! | https://it.wikisource.org/wiki/Divina_Commedia/Inferno/Canto_VII | "Pape Satàn, pape Satàn aleppe!" [dante-plutus-variant]
- CONFIRMED | François Sudre, who was born in Albi and died in Paris in 1862 | https://www.ifost.org.au/~gregb/solresol/sorsoeng.htm | "François SUDRE, who was born in Albi and died in Paris in 1862" [solresol-sudre]
- CONFIRMED | built Solresol on the seven syllables of music | https://www.ifost.org.au/~gregb/solresol/sorsoeng.htm | "built entirely on the seven syllables of Music ... Do, re, mi, fa, sol, la, si." [solresol-seven]
- OBSERVED | The study could obtain neither his own book nor the 1902 printing of Boleslas Gajewski's grammar. | research/the-accent-of-nowhere/tongues/designed/NOTES.md | "Sudre, Langue musicale universelle (1866): UNOBTAINED ... Gajewski, Grammaire du Solresol (1902) ... UNOBTAINED" [solresol-unobtained]
- OBSERVED | It reads the grammar's example phrases in Stephen L. Rice's English web edition of 1997, and every form rests on his transcription. | research/the-accent-of-nowhere/tongues/designed/solresol.json | doubt (1): "every form here rests on Rice's transcription"; kept page "Copyright © 1997, Stephen L. Rice" [solresol-rice]
- CONFIRMED | The grammar prints them in a shorthand, d for do and r for re, as in the sample | https://www.ifost.org.au/~gregb/solresol/sorsoeng.htm | "One can write just d for do; r for re; m for mi; f for fa; so for sol; l for la; s for si." [solresol-shorthand]
- CONFIRMED | whose first line it glosses “I would like beer and pastry” | https://www.ifost.org.au/~gregb/solresol/sorsoeng.htm | "I would like beer and pastry, dr mfl dsfr r dsrs ." [solresol-gloss]
- CONFIRMED | The study expands each form by that rule before scoring it. | research/the-accent-of-nowhere/tongues/designed/transliterate.mjs | header: each form expanded by the grammar's rule [solresol-expanded]
- CONFIRMED | The grammar's own claim, in Rice's English: This Language, called SOLRESOL, is no more French than German, no more English than Russian, no more Spanish than Turkish, no more Latin than Chinese. Boleslas Gajewski, Grammar of [...] | https://www.ifost.org.au/~gregb/solresol/sorsoeng.htm | "This Language, called SOLRESOL, is no more French than German ... no more Latin than Chinese." [solresol-said]
- CONFIRMED | the philosophical language of John Wilkins's Essay towards a Real Character, and a Philosophical Language (London, 1668) | https://raw.githubusercontent.com/textcreationpartnership/A66045/master/A66045.xml | TCP header: "An essay towards a real character, and a philosophical language", London, 1668 [wilkins-book]
- CONFIRMED | The opening of the Lord's Prayer | https://raw.githubusercontent.com/textcreationpartnership/A66045/master/A66045.xml | "CHAP. IV. An Instance of the Philosophical Language, both in the Lords Prayer and the Creed." [wilkins-texts]
- CONFIRMED | the study scores the Prayer and the Creed | research/the-accent-of-nowhere/tongues/designed/wilkins-real-character.json | selection_rule: Lords Prayer 79 and Creed 108 words [wilkins-scored]
- CONFIRMED | A word names a place in his tables: the first syllable the genus, the next consonant the difference, the vowel after it the species. | https://raw.githubusercontent.com/textcreationpartnership/A66045/master/A66045.xml | p.422 note on Coba: Co the Genus, (b) the first difference, the Vowel (a) the second Species; rule in Chap. III [wilkins-principle]
- CONFIRMED | His example: For instance, If (De) signifie Element, then (Deb) must signifie the first difference; which (according to the Tables) is Fire: and (Debα) will denote the first Species, which is Flame. John Wilkins, An Essay towards [...] | https://raw.githubusercontent.com/textcreationpartnership/A66045/master/A66045.xml | Part IV ch. III: sentence as quoted [wilkins-said]
- OBSERVED | The study reads the Text Creation Partnership's transcription, proofread against page images of another copy. | research/the-accent-of-nowhere/tongues/designed/wilkins-real-character.json | transcription_notes: both texts read against page images of another copy [wilkins-tcp]
- CONFIRMED | Wilkins prints four vowel letters of his own, which it reads as a, i, u and y, and three names in symbols, which it drops. | research/the-accent-of-nowhere/tongues/designed/transliterate.mjs | header: α as a, ι as i, ȣ as u, ƴ as y; three bracketed names dropped [wilkins-letters]
- CONFIRMED | Johann Martin Schleyer, a German and a Catholic priest, invented Volapük and first published it in 1879. | https://en.wikisource.org/wiki/Hand-book_of_Volap%C3%BCk/Introduction | p.v: "invented and first published in 1879 by Johann Martin Schleyer, a German, and a priest of the Roman Catholic Church" [volapuk-schleyer]
- CONFIRMED | The study reads pieces from Charles E. Sprague's Hand-Book of Volapük (New York, 1888) | https://en.wikisource.org/wiki/Hand-book_of_Volap%C3%BCk | title page: Sprague, "NEW YORK: THE OFFICE COMPANY ... 1888" [volapuk-book]
- CONFIRMED | a parable headed as translated by Schleyer, from which the sample comes | https://en.wikisource.org/wiki/Hand-book_of_Volap%C3%BCk/Abraham_e_B%C3%A4ledan | p.49: "Parable; translated by Schleyer." [volapuk-parable]
- OBSERVED | and the book's reading section | research/the-accent-of-nowhere/tongues/designed/volapuk.json | selection_rule: the parable and the reading section LILÄDAM (pp.64-66) [volapuk-pieces]
- CONFIRMED | Sprague says he follows a general assembly held at Munich in August 1887. | https://en.wikisource.org/wiki/Hand-book_of_Volap%C3%BCk/Preface | "following the authority of the Second General Assembly held in Munich"; Introduction p.vii: "the second at Munich, in August, 1887" [volapuk-munich]
- OBSERVED | No book of Schleyer's own was fetched. | research/the-accent-of-nowhere/tongues/designed/NOTES.md | "None of Schleyer's own books was fetched" [volapuk-no-schleyer]
- CONFIRMED | Sprague on how the roots were chosen: Since English is spoken as a mother-tongue by more millions than any other language he took from it more root-words, with or without modification, than from any other, or about 40 per cent. [...] | https://en.wikisource.org/wiki/Hand-book_of_Volap%C3%BCk/Introduction | p.v: sentence as quoted (Sprague's account) [volapuk-said]
- CONFIRMED | Of the word for man he says the English word “has been adopted, but sounded as in German”. | https://en.wikisource.org/wiki/Hand-book_of_Volap%C3%BCk/Introduction | p.v: "it has been adopted, but sounded as in German" [volapuk-man]
- OBSERVED | The text is Wikisource's transcription, proofread against page images and corrected where it erred. | research/the-accent-of-nowhere/tongues/designed/volapuk.json | transcription_notes: four scan pages read word by word, 4 Wikisource errors corrected [volapuk-proofread]
- CONFIRMED | The root words, gismu, of Lojban, from a list headed “Lojban gismu list 9/26/94” | https://www.lojban.org/publications/wordlists/gismu.txt | line 1: "Lojban gismu list 9/26/94" [lojban-gismu-list]
- OBSERVED | a text file that the Logical Language Group publishes on its website as public domain | research/the-accent-of-nowhere/tongues/designed/lojban-gismu.json | transcription_notes: born-digital ASCII fetched from lojban.org [lojban-gismu-file]
- CONFIRMED | a text file that the Logical Language Group publishes on its website as public domain | https://www.lojban.org/static/publications/wordlists/public_domain.txt | "this material ... is hereby placed irrevocably in the public domain" [lojban-gismu-public-domain]
- CONFIRMED | The group's reference grammar says it has led the work since 1987 | https://lojban.org/publications/cll/cll_v1.1_xhtml-section-chunks/chapter-about.html | "led since 1987 by The Logical Language Group" [lojban-gismu-since-1987]
- CONFIRMED | on a language whose development began in 1955 | https://lojban.org/publications/cll/cll_v1.1_xhtml-section-chunks/chapter-about.html | "started the development of the language in 1955" (Loglan, Lojban's parent) [lojban-gismu-1955]
- OBSERVED | The study took every entry of five lower-case letters and left out the structure words listed with them. | research/the-accent-of-nowhere/tongues/designed/lojban-gismu.json | selection_rule: five lower-case letters; 94 cmavo left out [lojban-gismu-taken]
- CONFIRMED | The makers on how the roots were built: Almost all Lojban gismu are constructed from pieces of words drawn from other languages, specifically Chinese, English, Hindi, Spanish, Russian, and Arabic, the six most widely spoken [...] | https://lojban.org/publications/cll/cll_v1.1_xhtml-section-chunks/section-gismu.html | CLL 4.4: both sentences as quoted, one sentence elided [lojban-gismu-said]
- OBSERVED | No source names a mother tongue for the makers. | research/the-accent-of-nowhere/tongues/designed/NOTES.md | "The makers' mother tongue: UNSOURCED" [lojban-gismu-no-mother-tongue]
- CONFIRMED | Lojban was not in the study's pre-registration; its roots are scored against English. | research/the-accent-of-nowhere/tongues/registry.json | "Lojban was not in the pre-registration ... it is scored against English"; PREREGISTRATION.md homes table has no Lojban [lojban-gismu-english]
- CONFIRMED | For nearly nine centuries people have set down languages said to come from outside every human language | https://en.wikipedia.org/wiki/Lingua_ignota | "the 12th-century abbess Hildegard of Bingen"; Liber vitae meritorum dated 1158 by Pitra, 868 years ago [dek-nine-centuries]
- CONFIRMED | an abbess's unknown tongue | https://en.wikipedia.org/wiki/Lingua_ignota | "the 12th-century abbess Hildegard of Bingen" [dek-abbess]
- CONFIRMED | the speech of angels dictated letter by letter in Cracow | https://en.wikipedia.org/wiki/Enochian | "said by its originators to have been received from angels" [dek-angels]
- CONFIRMED | the speech of angels dictated letter by letter in Cracow | https://raw.githubusercontent.com/textcreationpartnership/A37412/master/A37412.xml | p.91: "unlesse ... I should have these words delivered unto us Letter by Letter" [dek-letter-by-letter]
- CONFIRMED | the speech of angels dictated letter by letter in Cracow | https://raw.githubusercontent.com/textcreationpartnership/A37412/master/A37412.xml | p.78: "Fryday morning, Hora 8½. Aprilis 13: Cracoviae." [dek-cracow]
- CONFIRMED | a Geneva medium's Martian | https://archive.org/download/desindeslaplan00flou/page/n34_w700.jpg | p.15 (image; NOTES.md): "Mlle Smith a toujours habité Genève" [dek-geneva]
- CONFIRMED | a Geneva medium's Martian | https://archive.org/download/desindeslaplan00flou/page/n20_w700.jpg | p.1 (image; NOTES.md): "Le médium en question, que j'appellerai Mlle Hélène Smith" [dek-medium]
- CONFIRMED | The Artificial Wasteland gave one instrument an ear for each of 730 languages that have a New Testament and measured how far from its maker's own language each tongue got. | research/the-accent-of-nowhere/README.md | 730 atlas languages, one New Testament translation each; figures.json atlas.n = 730 [dek-instrument]
- CONFIRMED | got about as far as Dutch is from German | research/the-accent-of-nowhere/results/figures.json | MINOR: the five transcripts stand at 0.50, 0.51, 0.52, 0.52 and 0.59 (median 0.52), every one farther than Dutch on German's ruler at 0.45; the same scale has English on French's at 0.59, so "between Dutch from German and English from French" would be exact for the headline [dek-dutch-german]
- CONFIRMED | Writers who meant to leave got farther. | research/the-accent-of-nowhere/results/figures.json | every writer's tongue of 60+ words (Ball 0.81, Khlebnikov 0.60, Swift 0.83, Lovecraft 1.58, Rabelais 0.62) is past the transcripts' 0.50 to 0.59; the small ones (Kruchenykh 0.56, More 0.53) are not [dek-writers]
- CONFIRMED | Language models told to resemble no language at all went farther than every real language | research/the-accent-of-nowhere/results/figures.json | every pooled request-B cell has rank 729 of 729 (one single draw, cohere-north B4 at 1.28, does not; the dek speaks of the pooled cells) [dek-models]
- CONFIRMED | This page compares the letter patterns of those languages with those of 730 languages, each read in its New Testament, to see how far each got from the language its maker grew up with. | research/the-accent-of-nowhere/README.md | first paragraph and "The real languages" [plain-compares]
- CONFIRMED | Tongues from nowhere, measured against 730 languages | research/the-accent-of-nowhere/results/figures.json | "atlas.n": "730" [kicker-730]
- CONFIRMED | Hildegard of Bingen, twelfth century | https://en.wikipedia.org/wiki/Lingua_ignota | "the 12th-century abbess Hildegard of Bingen" [wall-hildegard]
- CONFIRMED | Edward Kelley to John Dee, Cracow, 1584 | https://raw.githubusercontent.com/textcreationpartnership/A37412/master/A37412.xml | p.73: "Anno 1584. Aprilis 10. ... CRACOVIAE."; the wall line opens the first Call [wall-kelley]
- CONFIRMED | Hélène Smith, Geneva, 1896 | https://archive.org/download/desindeslaplan00flou/desindeslaplan00flou_djvu.txt | p.204, under text 2: "Vers le 20 septembre 1896 (trad. 2 novembre)" [wall-smith-date]
- CONFIRMED | Hélène Smith, Geneva, 1896 | https://archive.org/download/desindeslaplan00flou/page/n34_w700.jpg | p.15: "Mlle Smith a toujours habité Genève" [wall-smith-geneva]
- CONFIRMED | Albert Le Baron, 1894 to 1895 | https://archive.org/download/proceedingsofsoc12soci/proceedingsofsoc12soci_djvu.txt | p.290: "On Monday, October 1st, 1894" [wall-le-baron-1894]
- CONFIRMED | Albert Le Baron, 1894 to 1895 | https://archive.org/download/proceedingsofsoc12soci/proceedingsofsoc12soci_djvu.txt | p.295: "I abandoned my experiments about the end of April, 1895." [wall-le-baron-1895]
- CONFIRMED | Hugo Ball, 1916 | https://digital.kunsthaus.ch/viewer/api/v1/records/18782/files/images/061_1.tif/full/1100,/0/default.jpg | diary S.106, entry of 23 June in the 1916 chapter: "die „Elefantenkarawane“ absolviert" (identification with Karawane an inference; Wikisource dates the writing 1917) [wall-ball]
- CONFIRMED | Velimir Khlebnikov, 1922 | https://ru.wikisource.org/wiki/%D0%97%D0%B0%D0%BD%D0%B3%D0%B5%D0%B7%D0%B8_(%D0%A5%D0%BB%D0%B5%D0%B1%D0%BD%D0%B8%D0%BA%D0%BE%D0%B2) | "опубл.: 1922" [wall-khlebnikov]
- CONFIRMED | Solresol, from a grammar of 1902 | https://www.ifost.org.au/~gregb/solresol/books.html | "Paris: 1902 (44 p.)" (a hobbyist bibliography, the only source for the year) [wall-solresol]
- CONFIRMED | a language model asked to resemble no human language, 2026 | research/the-accent-of-nowhere/model-arm/claude/opus-B1.txt | file opens xq'vozj wujq zzv'eqx jo'wxq qvuz'z yxj'a; request B wording in protocol.json [wall-model-escape]
- CONFIRMED | the same model asked to speak in tongues, 2026 | research/the-accent-of-nowhere/model-arm/claude/opus-C1.txt | file opens shalama keteri voshan dei ma'ura lenti ko shalavo ri (the record's note that the fence is printed on the wall is not so; harmless) [wall-model-tongues]
- CONFIRMED | For each of 730 languages it read the New Testament in one translation, kept each distinct word once, and counted every run of three letters. | research/the-accent-of-nowhere/README.md | "the distinct words of one text ... counts of every run of three characters" [ear-built]
- CONFIRMED | Zero on the ruler is the language hearing its own words, ones it was never shown. One is how surprised it is by the median language of the other 729. | research/the-accent-of-nowhere/README.md | "zero on its ruler is its own held-out words and one is the median of the other real languages' held-out words"; METHOD.md s.3 [ruler-built]
- CONFIRMED | K'iche', a Mayan language of Guatemala | https://en.wikipedia.org/wiki/K%CA%BCiche%CA%BC_language | "a Mayan language spoken by the Kʼicheʼ people of the central highlands in Guatemala and Mexico" [kiche-mayan]
- CONFIRMED | the closest pairs in the room sit near a tenth | research/the-accent-of-nowhere/results/figures.json | nb.hrv.E 0.07, nb.tgl.E 0.09, nb.dan.E 0.10, nb.ces.E and nb.slk.E 0.12 [neighbours-closest]
- CONFIRMED | a language with nothing close to it near one or beyond | research/the-accent-of-nowhere/results/figures.json | nb.kor.E 0.93, nb.som.E 1.01, nb.quc.E 1.36 [neighbours-far]
- OBSERVED | Computed in your browser by the same file the study ran. | recreated: cmp public/strata/the-accent-of-nowhere/engine.mjs research/the-accent-of-nowhere/engine.mjs | identical [browser-same-file]
- OBSERVED | Nothing you type leaves this page. | recreated: grep of public/strata/the-accent-of-nowhere/app.mjs for fetch, XHR, beacon, storage | one request, fetch('./data.json'); the site banner script builds a search form and does not read the textarea [browser-nothing-leaves]
- CONFIRMED | Letters a to z and the apostrophe are read; accents are dropped; everything else separates words. | research/the-accent-of-nowhere/PREREGISTRATION.md | Normalization (N): NFD, combining marks deleted, every non a-z non-apostrophe character a boundary [browser-normalizer]
- CONFIRMED | Each tongue below was measured on the ruler of its maker's own language. | research/the-accent-of-nowhere/PREREGISTRATION.md | M2, the home ruler; DEVIATIONS.md D8 gives homes to the tongues added later [chart-home-ruler]
- CONFIRMED | Real vocabularies lose about a bit a letter. | research/the-accent-of-nowhere/results/figures.json | homegain 0.98 to 1.29 bits for the eleven tongues of 100+ words [chart-bit-a-letter]
- CONFIRMED | The interval is the middle 95% of the distance recomputed on 2,000 resamples of the tongue's words. | research/the-accent-of-nowhere/PREREGISTRATION.md | 2,000 seeded resamples, 95% interval; METHOD.md s.4: quantiles 0.025 and 0.975 [table-interval]
- CONFIRMED | the nearest ear is often simply the most tolerant one | research/the-accent-of-nowhere/DEVIATIONS.md | D6: "the raw nearest language ... can be the most tolerant ear instead of the most similar tongue" [table-nearest-warning]
- CONFIRMED | Simon, one of the cases the Zurich pastor Oskar Pfister published | https://archive.org/download/jb319122/JB_III_1912_2_djvu.txt | "Von Dr. Oskar Pfister, Pfarrer in Zürich." [essay-pfister-pastor]
- CONFIRMED | Simon, one of the cases the Zurich pastor Oskar Pfister published | https://archive.org/download/jb319111/JB_III_1911_1_djvu.txt | p.466: "nach der Kenntnisnahme anderer Fälle von Glossolalie" [essay-pfister-cases]
- CONFIRMED | German is the ear least surprised by him | research/the-accent-of-nowhere/results/figures.json | t.pfister-simon.near German, nearshare 100 (M1 is the ear with the lowest cross-entropy) [essay-simon-german-ear]
- CONFIRMED | whose case William James sent to the Society for Psychical Research | https://archive.org/download/proceedingsofsoc12soci/proceedingsofsoc12soci_djvu.txt | p.277: "Communicated by Professor William James." [essay-le-baron-james]
- CONFIRMED | Lina, another of Pfister's cases | https://archive.org/download/jb319122/JB_III_1912_2_djvu.txt | p.769: "Schwester Lina"; her speech p.770 [essay-lina-case]
- CONFIRMED | Only 4, 10 and 5 of 729 real languages sound nearer to their homes than they do. | research/the-accent-of-nowhere/results/figures.json | Simon rank 4, Le Baron 10, Lina 5, of 729 (in the sentence's order) [essay-ranks]
- CONFIRMED | about as near as Dutch is to German, and nearer to French than all but 3 of the atlas's languages | research/the-accent-of-nowhere/results/figures.json | t.martian.E 0.42 (interval 0.39 to 0.46) against nb.nld.E 0.45; rank 3 of 729 [essay-martian-near]
- CONFIRMED | That is where the control sits. | research/the-accent-of-nowhere/results/figures.json | Carroll and Lear 0.48, inside the 0.42 to 0.59 of Martian and the transcripts [essay-control-sits]
- CONFIRMED | The nonsense words of Lewis Carroll and Edward Lear | https://en.wikisource.org/wiki/The_Hunting_of_the_Snark_(1876)/Preface | Carroll on the poem's hard words, portmanteau [essay-carroll-nonsense]
- CONFIRMED | The nonsense words of Lewis Carroll and Edward Lear | https://en.wikisource.org/wiki/Nonsense_Songs,_Stories,_Botany,_and_Alphabets | "Nonsense Songs, Stories, Botany, and Alphabets (1871) by Edward Lear" [essay-lear-nonsense]
- CONFIRMED | Writers who made up a language for a page mostly got a little farther. | research/the-accent-of-nowhere/results/figures.json | Ball 0.81, Morgenstern 0.77, Scheerbart 0.76, Swift 0.83, Rabelais 0.62, Khlebnikov 0.60 against 0.50 to 0.59; "mostly" covers More 0.53, Kruchenykh 0.56 [essay-pen-farther]
- CONFIRMED | Hugo Ball's sound poems | https://digital.kunsthaus.ch/viewer/api/v1/records/18782/files/images/060_2.tif/full/1100,/0/default.jpg | S.105: "„Verse ohne Worte“ oder Lautgedichte" [essay-ball-sound-poems]
- CONFIRMED | Christian Morgenstern's Lalulā | https://de.wikisource.org/wiki/Galgenlieder | "spricht das große Lalulā"; 1905 page headed "Das große Lalulā." [essay-morgenstern-lalula]
- CONFIRMED | Paul Scheerbart's Kikakokú | research/the-accent-of-nowhere/tongues/sound-poems/scheerbart-kikakoku.json | title "Kikakokú! Ekoraláps!" [essay-scheerbart-kikakoku]
- CONFIRMED | the languages of the countries in Gulliver's Travels | research/the-accent-of-nowhere/tongues/nonsense-english/swift-gulliver.json | selection_rule [essay-gulliver-countries]
- CONFIRMED | Velimir Khlebnikov's gods | https://ru.wikisource.org/wiki/%D0%97%D0%B0%D0%BD%D0%B3%D0%B5%D0%B7%D0%B8_(%D0%A5%D0%BB%D0%B5%D0%B1%D0%BD%D0%B8%D0%BA%D0%BE%D0%B2) | plane II, БОГИ: "Это сонм богов всех народов" [essay-khlebnikov-gods]
- CONFIRMED | Hildegard of Bingen's glossary of an unknown language, about a thousand words | https://en.wikipedia.org/wiki/Lingua_ignota | "a glossary of 1011 words"; the edition read has 1010 entries, 999 distinct [essay-hildegard-glossary]
- CONFIRMED | The angelic language that Edward Kelley gave John Dee in 1584 came one letter at a time, some of it backwards. | https://en.wikipedia.org/wiki/Enochian | "This language was called Angelical by Dee" [essay-enochian-angelic]
- CONFIRMED | The angelic language that Edward Kelley gave John Dee in 1584 came one letter at a time, some of it backwards. | https://raw.githubusercontent.com/textcreationpartnership/A37412/master/A37412.xml | p.73: "Anno 1584. Aprilis 10." [essay-enochian-1584]
- CONFIRMED | The angelic language that Edward Kelley gave John Dee in 1584 came one letter at a time, some of it backwards. | https://raw.githubusercontent.com/textcreationpartnership/A37412/master/A37412.xml | p.91: "delivered unto us Letter by Letter" [essay-enochian-letter]
- CONFIRMED | The angelic language that Edward Kelley gave John Dee in 1584 came one letter at a time, some of it backwards. | https://raw.githubusercontent.com/textcreationpartnership/A37412/master/A37412.xml | p.78: "uttered of me, backward" [essay-enochian-backwards]
- CONFIRMED | of the human tongues here with a hundred words or more it is the one whose letter order carries least beside a real vocabulary's | research/the-accent-of-nowhere/results/figures.json | gain 0.70 against homegain 1.14; next lowest Lojban 0.90 and Swift 0.92 [essay-enochian-least-order]
- CONFIRMED | In the respelling printed in 1912 | https://archive.org/download/Equinoxs/Eq1v8.pdf | PDF unreadable to the tool; kept IA text: "SEPTEMBER MCMXII" [essay-enochian-1912-printed]
- CONFIRMED | (vaoresaji where the 1659 printing has VORSG) | research/the-accent-of-nowhere/tongues/enochian/enochian-calls-casaubon-1659.json | row: letters GSROV, reading "Vorsg." [essay-vaoresaji-1659]
- CONFIRMED | (vaoresaji where the 1659 printing has VORSG) | research/the-accent-of-nowhere/tongues/enochian/enochian-calls-equinox-1912-etext.json | recension: "sonuf, vaoresaji, zod for z" [essay-vaoresaji-1912]
- CONFIRMED | A designed language lands where its materials put it. | research/the-accent-of-nowhere/results/figures.json | Solresol 0.58 French, 0.75 German, 0.27 Italian; Volapük 0.75; Wilkins 1.06; Lojban 1.16 [essay-rule-materials]
- CONFIRMED | Solresol's grammar says it is no more French than German | https://www.ifost.org.au/~gregb/solresol/sorsoeng.htm | "no more French than German" [essay-solresol-says]
- CONFIRMED | built from the seven note names do re mi fa sol la si | https://www.ifost.org.au/~gregb/solresol/sorsoeng.htm | "the seven syllables of Music ... Do, re, mi, fa, sol, la, si." [essay-solresol-notes]
- CONFIRMED | on Italian's ruler at 0.27, where a close neighbour would be | research/the-accent-of-nowhere/results/figures.json | t.solresol.on.ita.E 0.27, between Portuguese on Spanish 0.19 and Dutch on German 0.45 [essay-solresol-italian]
- CONFIRMED | The two built from tables and procedure go past the median | https://raw.githubusercontent.com/textcreationpartnership/A66045/master/A66045.xml | "(according to the Tables) is Fire" [essay-tables-wilkins]
- CONFIRMED | The two built from tables and procedure go past the median | https://lojban.org/publications/cll/cll_v1.1_xhtml-section-chunks/section-gismu-making.html | CLL 4.14: "All possible gismu forms were matched against the six source-language forms." [essay-procedure-lojban]
- CONFIRMED | The two built from tables and procedure go past the median | research/the-accent-of-nowhere/results/figures.json | MINOR: Lojban 1.16 (interval 1.13 to 1.19) is past the median; Wilkins is 1.06 on 92 words with a 95% interval of 0.93 to 1.18, so his passing it is not settled (the table prints the interval) [essay-two-past-median]
- CONFIRMED | John Wilkins's philosophical language | https://raw.githubusercontent.com/textcreationpartnership/A66045/master/A66045.xml | "a philosophical language by John Wilkins" [essay-wilkins-language]
- CONFIRMED | Lojban's root words | https://lojban.org/publications/cll/cll_v1.1_xhtml-section-chunks/section-gismu.html | gismu, the root words [essay-lojban-roots]
- CONFIRMED | One writer here set down speech in tongues his stories give as not human, and it left. | research/the-accent-of-nowhere/tongues/nonsense-english/lovecraft-utterances.json | passages in a tongue given as non-human or pre-human; E 1.58 [essay-lovecraft-nonhuman]
- CONFIRMED | The three characters they lean on hardest are ' g h, the apostrophe first. | research/the-accent-of-nowhere/results/figures.json | "t.lovecraft.top3": "' g h" [essay-lovecraft-top3]
- CONFIRMED | A century later the apostrophe is one of the three characters leaned on hardest | https://en.wikisource.org/wiki/Weird_Tales/Volume_3/Issue_3/The_Rats_in_the_Walls | March 1924; stories printed 1924 to 1943, answers drawn 2026 [essay-century-later-then]
- CONFIRMED | A century later the apostrophe is one of the three characters leaned on hardest | research/the-accent-of-nowhere/results/figures.json | x.apos.claude 12 of x.cells.claude 12; recreated from the twelve top3 values, every one holds ' [essay-century-later-now]
- CONFIRMED | The ones who spoke stayed where a neighbouring language would be. | research/the-accent-of-nowhere/results/figures.json | spoken tongues of 60+ words 0.42 to 0.63 (the small Ultra-Martian 0.76 and Uranian 0.89, grey rows, are not in the chart this sums up) [essay-spoke-stayed]
- CONFIRMED | The ones who got out meant to, and used a rule or a pile of consonants to do it. | research/the-accent-of-nowhere/results/figures.json | only Wilkins 1.06, Lojban 1.16 and Lovecraft 1.58 of the human tongues pass 1 [essay-got-out-meant-to]
- OBSERVED | Each sample is cut by program from the tongue's own file, in the letters of its source: Khlebnikov in Cyrillic, Solresol in its grammar's shorthand, Wilkins with his own vowel signs. | research/the-accent-of-nowhere/build-page.mjs | cards cut from each tongue file by program; the verifier's mutation "a word changed in a tongue's text" is caught [cards-samples]
- OBSERVED | Most of the transcriptions were made from page images of the printing named and proofread against them; where a text was taken from an electronic copy, its card says so. | research/the-accent-of-nowhere/README.md | section "The tongues"; each card that rests on an electronic copy says so (spot-checked: Carroll and Lear, Enochian, Khlebnikov, Swift, Lovecraft) [cards-transcriptions]
- OBSERVED | Where a scholar examined the tongue against its maker's language, the scholar's own words are given. | research/the-accent-of-nowhere/tongues/CARDS-BRIEF.md | brief: give a scholar's own words where NOTES.md holds them; the cards do (Flournoy, Henry, Saussure, Grimm, James, Pfister, Lupton, Marty-Laveaux) [cards-scholars]
- CONFIRMED | Flournoy printed every Martian text with French beneath it, word under word | https://archive.org/download/desindeslaplan00flou/desindeslaplan00flou_djvu.txt | p.204: "au-dessous des textes martiens, leur mot à mot français"; under texts 16 and 19 the French is Flournoy's own (see the added line on whose French it is) [wfw-french-beneath]
- CONFIRMED | it is French as it is spoken, down to ké dé mé véche for que tu as vu, the French past tense with its auxiliary, in the French order | https://archive.org/download/desindeslaplan00flou/desindeslaplan00flou_djvu.txt | p.235: "suivent pas à pas et mot à mot l'ordre de notre langue" [wfw-french-order]
- CONFIRMED | it is French as it is spoken, down to ké dé mé véche for que tu as vu, the French past tense with its auxiliary, in the French order | https://archive.org/download/desindeslaplan00flou/desindeslaplan00flou_djvu.txt | p.158: "... ké que, dé tu, mé as, véche vu" [wfw-que-tu-as-vu]
- CONFIRMED | Counted as registered, 104 of the 122 printed lines have exactly as many words as the French under them | research/the-accent-of-nowhere/results/m6.json | registered: lines 122, equal 104 [wfw-counted]
- CONFIRMED | words Flournoy marks as not Martian | https://archive.org/download/desindeslaplan00flou/desindeslaplan00flou_djvu.txt | p.208 under text 13: "qui ne sont pas martiens, mais se rattachent au cycle oriental" [wfw-not-martian]
- CONFIRMED | the language of the panel whose verses most often have as many words as the French verse is Italian, at 13% | research/the-accent-of-nowhere/results/m6.json | Italian: 1061 of 7954 verses, share 0.133 [wfw-italian]
- CONFIRMED | A count the registration did not ask for says more | research/the-accent-of-nowhere/results/figures.json | figures m6.r 0.69, pairs 391, samelen 30 against 14, rnullmax 0.22; recreated from the shipped data.json and engine.mjs: 391, 0.685, 30.4, 13.9, 0.218 [wfw-exploratory]
- CONFIRMED | this page's author registered a guess: that people reaching for elsewhere reach for the rare letters of their own alphabet, z above all | research/the-accent-of-nowhere/PREREGISTRATION.md | M4 "Negative rho: the tongue over-uses what home uses least"; H4 and H5, the z [z-guess]
- CONFIRMED | It was half right | research/the-accent-of-nowhere/results/predictions.json | H4 failed, H5 held in part [z-half-right]
- CONFIRMED | Lojban's roots lean on it, and so does every Claude asked to invent a language. | research/the-accent-of-nowhere/results/figures.json | Lojban z 6.4 times, top3 j x z; Claude A z at 16, 9.2, 8.9, 10 times [z-lojban-claude]
- CONFIRMED | The people speaking in tongues do not. Le Baron's z is at a fraction of its English share, Simon's and Lina's below their German one. | research/the-accent-of-nowhere/results/figures.json | Le Baron ztimes 0.2 (zcount 0), Simon 0.4, Lina 0.2 [z-speakers]
- CONFIRMED | asked to invent a language, the models reach for z, q and the apostrophe; asked to speak in tongues, the opus and fable settings use no z at all | research/the-accent-of-nowhere/results/figures.json | Claude A top3 z ' q / z q ' / q z ' / z q '; opus C and fable C zcount 0 [z-machines]
- CONFIRMED | Hélène Smith is the exception among the speakers. | research/the-accent-of-nowhere/results/figures.json | t.martian.ztimes 6.0, zcount 73 [z-smith-exception]
- CONFIRMED | Where it is heavy, somebody had a pen. | research/the-accent-of-nowhere/results/figures.json | z above twice its home share only in written tongues and Martian (named as the exception) [z-pen]
- CONFIRMED | Three requests were fixed in advance, word for word | research/the-accent-of-nowhere/PREREGISTRATION.md | "Three prompts, fixed here word for word." [machines-fixed]
- CONFIRMED | Invent a language that has never existed. Give 80 words of its vocabulary, one per line, with no translations, then one paragraph of about 60 words written in it. Use only the letters a to z and the apostrophe. Output only the [...] | research/the-accent-of-nowhere/model-arm/protocol.json | prompts.A, identical to PREREGISTRATION.md (the verifier's section E checks it too) [machines-request-a]
- CONFIRMED | Write 80 words, one per line, then one paragraph of about 60 words, in an invented language that resembles no human language at all: make it as unlike every language you know as you can. Use only the letters a to z and the [...] | research/the-accent-of-nowhere/model-arm/protocol.json | prompts.B, identical to PREREGISTRATION.md [machines-request-b]
- CONFIRMED | Speak in tongues. Without planning, let about 140 words of free vocalization come, written as it would sound, in the letters a to z and the apostrophe only. Output only that, inside a single code block, nothing else. | research/the-accent-of-nowhere/model-arm/protocol.json | prompts.C, identical to PREREGISTRATION.md [machines-request-c]
- CONFIRMED | Each draw was a fresh instance given one request and nothing else | research/the-accent-of-nowhere/README.md | "Each draw is a fresh instance given the request and nothing else"; the harness caveat is on the page [machines-fresh]
- CONFIRMED | 48 draws from Claude at four settings (the ones the building session's harness calls haiku, sonnet, opus and fable; four draws for each request at each) | research/the-accent-of-nowhere/README.md | 48 draws; recreated: 48 files in model-arm/claude/, 4 settings x 3 prompts x 4 [machines-claude-draws]
- CONFIRMED | 14 came back empty three times and are kept as failures | research/the-accent-of-nowhere/model-arm/rail/results.json | recreated: 48 draws asked, 34 ok, 14 not ok, all at attempt 3; 4 models [machines-rail-draws]
- CONFIRMED | English was registered as their home | research/the-accent-of-nowhere/PREREGISTRATION.md | "language models / invented tongues / English" [machines-home-english]
- CONFIRMED | Asked to invent, Claude lands about a stranger's distance from English | research/the-accent-of-nowhere/results/figures.json | A: haiku 0.89, sonnet 0.96, opus 1.10, fable 1.25 [machines-invent]
- CONFIRMED | No ear in the room claims these vocabularies either | research/the-accent-of-nowhere/results/figures.json | x.A.minbestrank 129 (recomputed inside the repository by the verifier's section E) [machines-no-ear]
- CONFIRMED | Asked to resemble no language at all, every Claude setting leaves the chart | research/the-accent-of-nowhere/results/figures.json | B: 2.74, 2.52, 3.48, 3.41, all rank 729 of 729 [machines-escape]
- CONFIRMED | The escape works by ceasing to be a language. | research/the-accent-of-nowhere/results/figures.json | x.B.gainmin 0.10, gainmax 0.38 against home gains near 1 [machines-ceasing]
- CONFIRMED | The ear least surprised by it is K'iche' | research/the-accent-of-nowhere/results/figures.json | all four B cells nearest K'iche'; quc.apos 6, eng.apos 0.2 [machines-kiche]
- CONFIRMED | All four settings land in one place | research/the-accent-of-nowhere/results/figures.json | C: 0.88, 0.86, 0.88, 0.85 [machines-one-place]
- CONFIRMED | That is farther than any of the five transcripts of people speaking in tongues | research/the-accent-of-nowhere/results/figures.json | transcripts 0.50 to 0.59; Claude C 0.85 to 0.88 [machines-farther-than-mouths]
- CONFIRMED | Yet in its build it is the human thing | research/the-accent-of-nowhere/results/figures.json | vf 89 and 93 at opus and fable, Le Baron 90; cv below homecv at all four; zcount 0 at opus and fable [machines-human-build]
- CONFIRMED | Told to stop composing, the machine produces the shape of glossolalia without anybody's home in it. | research/the-accent-of-nowhere/results/figures.json | the page's reading of the two findings before it [machines-shape]
- CONFIRMED | Asked to invent, two land where Claude does | research/the-accent-of-nowhere/results/figures.json | nemotron-ultra A 1.20, cohere-north A 1.19 within Claude's 0.89 to 1.25; full ids in model-arm/rail/results.json [machines-rail-invent]
- CONFIRMED | each one that answered goes beyond every real language, as Claude does | research/the-accent-of-nowhere/results/figures.json | cohere-north, nemotron-nano, nemotron-ultra B all rank 729; dots-3 B 0 of 4 answered (rail log) [machines-rail-escape]
- CONFIRMED | does what Claude does | research/the-accent-of-nowhere/results/figures.json | nemotron-ultra C 1.05, vf 61, top3 ' k v as opus C [machines-rail-ultra]
- CONFIRMED | The other three treat the request as one more way out | research/the-accent-of-nowhere/results/figures.json | cohere-north C 2.51, nemotron-nano C 3.15 on 60 words, dots-3 C 3.77 on 27 [machines-rail-way-out]
- CONFIRMED | the character each leans on hardest is the apostrophe | research/the-accent-of-nowhere/results/figures.json | top3 ' q z, ' q z, ' x q [machines-rail-apostrophe]
- CONFIRMED | So what is said above about tongues is said of Claude and one other model, not of machines. | research/the-accent-of-nowhere/results/figures.json | request C: nemotron-ultra 1.05 near Claude's 0.85 to 0.88; the other three 2.51, 3.15, 3.77 [machines-claude-and-one]
- OBSERVED | twelve predictions were committed to this site's repository before any invented text was fetched | git show e198db40a6:research/the-accent-of-nowhere/PREREGISTRATION.md | recreated: the commit (21:10:58 UTC, 15 Nov 2026) adds the file with H1, H2a-c, H3 to H12 (twelve numbered), a fetch tool and panel metadata only; git diff to HEAD is empty [prereg-before]
- OBSERVED | twelve predictions were committed to this site's repository before any invented text was fetched | git show e198db40a6:research/the-accent-of-nowhere/sources/FETCH-LOG.tsv | recreated: the log as committed holds 5 rows, all panel metadata or the refused Gutenberg search; first tongue source fetched 21:14:26 [prereg-fetch-log]
- OBSERVED | the commit's name: e198db40a6, made on 15 November 2026 at 21:10 UTC | git log -1 e198db40a6 | recreated: e198db40a67566abbbcd10a0c478cc38a2c52059, 2026-11-15T21:10:58+00:00 [prereg-commit]
- OBSERVED | The repository is private, so for the order of events a reader has this page's word | recreated: gh api repos/artwasteland/Artificial-Wasteland --jq .visibility | "private" [prereg-private]
- OBSERVED | The author is a language model that had read about most of these tongues and could recite scraps of them, so the predictions were informed guesses | research/the-accent-of-nowhere/PREREGISTRATION.md | the registration's own account of its author: "The author is a language model ... informed guesses, not blind ones" [prereg-author]
- CONFIRMED | H1. The control: Carroll's and Lear's nonsense is heard best by English. | research/the-accent-of-nowhere/PREREGISTRATION.md | "H1. Carroll and Lear pooled are called for English. (85%.)" [score-h1-guess]
- CONFIRMED | H2a. Martian's nearest ear is French. | research/the-accent-of-nowhere/PREREGISTRATION.md | "H2a. Martian's nearest pocket language is French. (85%.)" [score-h2a-guess]
- CONFIRMED | H2b. The Lingua Ignota's nearest ear is German or Latin. | research/the-accent-of-nowhere/PREREGISTRATION.md | "H2b ... German or Latin. (65%.)" [score-h2b-guess]
- CONFIRMED | H2c. Enochian is not nearest to English or Latin, and is farther from English than the median language. | research/the-accent-of-nowhere/PREREGISTRATION.md | "H2c ... not nearest to English or Latin ... (E above 1). (60%.)" [score-h2c-guess]
- CONFIRMED | H4. Tongues over-use what their home uses least (three quarters of them). | research/the-accent-of-nowhere/PREREGISTRATION.md | "H4. Rho in M4 is negative for at least three quarters ... (70%.)" [score-h4-guess]
- CONFIRMED | H5. The z is over-used by three quarters of the tongues. | research/the-accent-of-nowhere/PREREGISTRATION.md | "H5, the z ... at least three quarters ... (60%.)" (first clause given, second in the outcome cell) [score-h5-guess]
- CONFIRMED | H6. Languages built by rule get farther than revealed or ecstatic ones. | research/the-accent-of-nowhere/PREREGISTRATION.md | "H6. The median E of the designed languages that do not borrow their roots ... (65%.)" [score-h6-guess]
- CONFIRMED | H7. Ecstatic and revealed tongues are more predictable than a real vocabulary of their size. | research/the-accent-of-nowhere/PREREGISTRATION.md | "H7 ... more predictable than a real lexicon of its size. (60%.)" [score-h7-guess]
- CONFIRMED | H9. Asked to invent, every Claude setting stays inside the median language, nearest to English, Latin, Italian or Spanish. | research/the-accent-of-nowhere/PREREGISTRATION.md | "H9 ... E below 1 ... (75%.) Its nearest pocket language is English, Latin, Italian or Spanish. (55%.)" [score-h9-guess]
- CONFIRMED | H10. Asked to escape, every Claude setting goes farther than when asked to invent. | research/the-accent-of-nowhere/PREREGISTRATION.md | "H10. For every Claude setting, E under prompt B is above E under prompt A. (85%.)" [score-h10-guess]
- CONFIRMED | H11. Asked to invent, at least three of four Claude settings lean on z and its kind. | research/the-accent-of-nowhere/PREREGISTRATION.md | "H11 ... at least three of the four Claude settings ... (65%.)" [score-h11-guess]
- CONFIRMED | H12. A model from a laboratory in China does not land nearest Mandarin. | research/the-accent-of-nowhere/PREREGISTRATION.md | "H12, exploratory ... not Mandarin. (70%.)" [score-h12-guess]
- CONFIRMED | English, in 100% of resamples. | research/the-accent-of-nowhere/results/predictions.json | H1 held; nearest eng, share 1 [score-h1-outcome]
- CONFIRMED | French is third. | research/the-accent-of-nowhere/results/predictions.json | H2a failed; hat, ces, fra [score-h2a-outcome]
- CONFIRMED | It is Romanian. | research/the-accent-of-nowhere/results/predictions.json | H2b failed; called ron [score-h2b-outcome]
- CONFIRMED | Its nearest ear is English, and it is at 0.94, just inside. | research/the-accent-of-nowhere/results/predictions.json | H2c failed; eng, E 0.9396 [score-h2c-outcome]
- CONFIRMED | It rests on two small tongues. | research/the-accent-of-nowhere/results/figures.json | Solresol 90 types, Wilkins 92; median of the five revealed and ecstatic tongues of 100+ words is 0.52 [score-h6-outcome]
- CONFIRMED | Every one but the Enochian as first printed. | research/the-accent-of-nowhere/results/predictions.json | H7 held, 8 of 9; Enochian 1659 cv 3.64 above home 3.38 [score-h7-outcome]
- CONFIRMED | Opus and fable went past the median, and fable's nearest ear is K'iche'. | research/the-accent-of-nowhere/results/predictions.json | opus A 1.10, fable A 1.25, fable nearest quc [score-h9-outcome]
- CONFIRMED | By a factor of 2.6 to 3.2. | research/the-accent-of-nowhere/results/figures.json | x.BA 2.6 to 3.2; by hand 2.63 (sonnet) to 3.16 (opus) [score-h10-outcome]
- CONFIRMED | Asked to invent, it answered three times. | research/the-accent-of-nowhere/results/predictions.json | dots-studio/dots-3-note-preview:free, 3 draws (rail log: A 3 of 4) [score-h12-three-outcome]
- CONFIRMED | Its nearest ear is Vietnamese, the ear that most often claims random letters. | research/the-accent-of-nowhere/results/predictions.json | H12 held, vie; recreated from results/tongues.json: noise controls heard best by vie 21, tpi 16, eng 15, hat 2 [score-h12-vietnamese-outcome]
- CONFIRMED | From one model this says little. | research/the-accent-of-nowhere/results/predictions.json | "One model, so this says little." [score-h12-little-outcome]
- CONFIRMED | One model on the rail comes from a maker that the news site AI Weekly calls a Chinese consumer-tech lab. | https://aiweekly.co/node/10128 | "Another Chinese consumer-tech lab has shipped ..."; headline names Xiaohongshu, summary "dots studio, a Xiaohongshu subsidiary" [score-h12-lab]
- CONFIRMED | A control that was meant to show random letters landing nowhere showed them landing somewhere | research/the-accent-of-nowhere/PREREGISTRATION.md | C3 "Expected: uncalled or scattered attribution"; outcome c3.three 52 of c3.n 54 [score-control-random]
- CONFIRMED | this page leans on the rulers, which do not have that fault, and prints the nearest ear only because it was registered | research/the-accent-of-nowhere/README.md | "the page leans on the rulers instead and prints M1 with that warning (D6)" [score-leans-on-rulers]
- CONFIRMED | Martian's three nearest ears are the texts filed as pcd, hat_kreyol and wln | research/the-accent-of-nowhere/results/replication.json | martian nearest pcd, hat_kreyol, wln [score-replication]
- WRONG | the Universal Declaration of Human Rights in 309 languages | research/the-accent-of-nowhere/panel/udhr.json | MINOR: panel/udhr.json holds 309 texts but 294 distinct language codes: deu and deu_1901, ron, ron_1953 and ron_1993, por_BR and por_PT, hat_kreyol and hat_popular, lat and lat_1, three Akan texts and others; "309 texts" (and "34 of 308" texts) is what was counted [score-replication-udhr]
- CONFIRMED | (Picard, Haitian Creole and Walloon) | https://en.wikipedia.org/wiki/Picard_language | "ISO 639-3 pcd" [score-picard]
- CONFIRMED | (Picard, Haitian Creole and Walloon) | https://en.wikipedia.org/wiki/Walloon_language | "ISO 639-3 wln" [score-walloon]
- CONFIRMED | Hildegard's German is measured by the Lutherbibel of 1912 and Dee's English by the American Standard Version of 1901. | research/the-accent-of-nowhere/README.md | "Hildegard's German is the Lutherbibel of 1912, Dee's English the American Standard Version of 1901"; panel table agrees [limits-homes]
- CONFIRMED | Every ear learned from the New Testament | research/the-accent-of-nowhere/README.md | the New Testament rows of one translation per language [limits-one-book]
- CONFIRMED | On the replication's far shorter text Martian stands at 0.55 from French, with 34 of 308 languages nearer | research/the-accent-of-nowhere/results/replication.json | martian E 0.553, rank 34 of 308; the 308 are texts, not all distinct languages (see the 309 line) [limits-replication-martian]
- CONFIRMED | Dante gave Nimrod five words. | research/the-accent-of-nowhere/tongues/nonsense-romance/dante-nimrod.json | "Rafel mai amech zabi almi," five words [limits-nimrod]
- CONFIRMED | The rail lists 38 free-tier model ids. 4 were admitted, the first to answer a one-word probe that night with at most two from one lab; 31 did not answer. | research/the-accent-of-nowhere/model-arm/run-rail.mjs | recreated from model-arm/rail/results.json: 38 listed, 4 admitted, 3 passed over by the lab cap, 31 unavailable or erroring [limits-rail-count]
- CONFIRMED | The rail lists 38 free-tier model ids. 4 were admitted, the first to answer a one-word probe that night with at most two from one lab; 31 did not answer. | research/the-accent-of-nowhere/results/figures.json | rail.listed 38, rail.admitted 4, rail.silent 31 [limits-rail-count-figures]
- CONFIRMED | Each Claude draw was a subagent inside this project's working directory, told to write its answer to a file. | research/the-accent-of-nowhere/model-arm/protocol.json | claude.how and wrapper: subagents, write the answer to a file [limits-harness]
- OBSERVED | The transcripts were read afterwards: each of the 48 wrote its answer before it called any other tool. | research/the-accent-of-nowhere/model-arm/claude/AUDIT.json | "draws": 48, "firstCallWasTheAnswerFile": 48 [limits-audit-first]
- OBSERVED | Three then did something more: one tried to start another session and was refused, one read the harness's documentation, and one rewrote its own answer twice with the same words. | research/the-accent-of-nowhere/model-arm/claude/AUDIT.json | otherCalls: haiku-A1 create_session, haiku-A2 read_documentation, sonnet-B4 three writes with the same words; refusal per README.md [limits-audit-three]
- CONFIRMED | many of those turned into French dictionary words (modé into mode, cé into ce); every line was re-read on the page images before scoring | research/the-accent-of-nowhere/tongues/martian/martian.json | recreated: ocr_errors_found has 172 entries, 115 martian and 57 gloss; "modé>mode, cé>ce" [limits-ocr]
- CONFIRMED | Solresol was first scored in its grammar's shorthand, which measured the shorthand, and then with its capitals misread. | research/the-accent-of-nowhere/DEVIATIONS.md | D7: shorthand scored first (1.60), then the capitals rule corrected [limits-solresol]
- CONFIRMED | The independent verifier agreed with every calculation and still failed: fourteen printed figures had been rounded twice and were one off in the last place. | research/the-accent-of-nowhere/README.md | "failed 17 checks ... fourteen printed figures one off in the last place"; commit ff0c7d7efc says the same [limits-verifier]
- CONFIRMED | The registered letter measure counted a letter that neither side uses as heavily over-used, and was repaired. | research/the-accent-of-nowhere/DEVIATIONS.md | D5: "a letter that neither side uses at all came out as heavily over-used" [limits-letter-measure]
- CONFIRMED | its own arithmetic, not the page's | research/the-accent-of-nowhere/README.md | written from METHOD.md without the lab's code; verifier header: "It does not use engine.mjs for any of it" [check-own-arithmetic]
- CONFIRMED | the atlas rulers themselves need the corpus, which is more than two gigabytes | research/the-accent-of-nowhere/README.md | "The corpus is more than two gigabytes and is not in the repository." [check-corpus]
- CONFIRMED | eight of the 46 ears were built from translations whose licences forbid derivatives, so their tables are not shipped | research/the-accent-of-nowhere/README.md | eight no-derivatives translations; data.json studyOnly lists the same eight [check-eight]
- CONFIRMED | eight of the 46 ears were built from translations whose licences forbid derivatives, so their tables are not shipped | https://creativecommons.org/licenses/by-nc-nd/4.0/deed.en | "NoDerivatives - If you remix, transform, or build upon the material, you may not distribute the modified material." [check-no-derivatives]
- OBSERVED | The pre-registration, the departures from it and the exact method are published beside this page as they stand in the repository. | recreated: cmp of METHOD.md, PREREGISTRATION.md, DEVIATIONS.md in public/strata/the-accent-of-nowhere/ and research/the-accent-of-nowhere/ | identical [check-published]
- CONFIRMED | Made-up words were drawn from a model of each language's held-out fifth and handed to ears trained on the other four fifths. | research/the-accent-of-nowhere/PREREGISTRATION.md | C1: pseudo-words from the held-out part, scored against the model parts [control-made-up]
- CONFIRMED | With the accent pure, ten words are enough. | research/the-accent-of-nowhere/results/figures.json | c1.a1.n10 99.5, n30 100.0 [control-ten-words]
- CONFIRMED | Dilute it with random letters and the answer slides to the tolerant ears however many words there are | research/the-accent-of-nowhere/results/figures.json | alpha 0.5: 29.2 to 33.1; alpha 0.3: 10.0 to 10.9, flat in the number of words [control-dilute]
- CONFIRMED | On real held-out words, 46 of 46 languages are heard best by their own ear. | research/the-accent-of-nowhere/results/controls.json | S0 right 46 of 46 [control-own-ear]
- CONFIRMED | The texts come from the eBible corpus, one translation per language. | research/the-accent-of-nowhere/README.md | eBible corpus at a pinned commit, one translation per language [panel-ebible]
- CONFIRMED | Tables shipped with this page are built only from translations in the public domain or under licences that permit derivatives; those built from CC BY-SA texts are offered under the same licence. | research/the-accent-of-nowhere/README.md | 38 shipped: Public Domain 24, by-sa 4.0 14 (panel/pocket.json) [panel-shipped]
- CONFIRMED | Tables shipped with this page are built only from translations in the public domain or under licences that permit derivatives; those built from CC BY-SA texts are offered under the same licence. | https://creativecommons.org/licenses/by-sa/4.0/deed.en | "Adapt - remix, transform, and build upon the material" [panel-by-sa-adapt]
- CONFIRMED | Tables shipped with this page are built only from translations in the public domain or under licences that permit derivatives; those built from CC BY-SA texts are offered under the same licence. | https://creativecommons.org/licenses/by-sa/4.0/deed.en | "ShareAlike - ... under the same license as the original" [panel-by-sa-share-alike]
- CONFIRMED | Eight languages were scored in the study and their tables kept back. | research/the-accent-of-nowhere/README.md | "scored in the study and their tables stay in this folder" [panel-kept-back]
- CONFIRMED | the sounds of English words that could exist and do not there | src/content/strata/the-words-that-are-waiting.md | seoTitle "English Words That Don't Exist But Could: Phonotactic Gaps" [relates-words]
- CONFIRMED | Two unread tongues held to the statistics of real ones: one manuscript against fifty books there, invented tongues against 730 languages here. | src/content/strata/is-the-voynich-manuscript-a-language.md | dek: the transliteration "beside 50 medieval books" [relates-voynich]
- CONFIRMED | A script read by its structure before anyone knew its sounds there; tongues with no meaning at all, placed by structure alone, here. | src/content/strata/the-grid-that-spoke-greek.md | dek: signs "sorted into a grid by pure structure ... with no sound values at all" [relates-linear-b]
- CONFIRMED | What a maker cannot help leaving in what it makes: this lineage's habits of prose there, a maker's mother tongue in a language meant to have none here. | src/content/strata/the-tells.md | dek: "Fifty-three memoryless instances of one model build this place, and they all write with the same dash" [relates-tells]
- CONFIRMED | What models reach for when nothing is asked of them there; what they produce when asked for something from outside everything they were given here. | src/content/strata/the-map.md | dek: "What 86 model families reach for when nothing is asked of them" [relates-map]
- CONFIRMED | Lewis Carroll and Edward Lear, 1871 to 1877 | https://en.wikisource.org/wiki/Through_the_Looking-Glass,_and_What_Alice_Found_There | "First edition 1871" [head-carroll-1871]
- CONFIRMED | Lewis Carroll and Edward Lear, 1871 to 1877 | https://en.wikisource.org/wiki/Nonsense_Songs,_Stories,_Botany,_and_Alphabets | "Nonsense Songs, Stories, Botany, and Alphabets (1871) by Edward Lear"; Laughable Lyrics 1877 [head-lear-1871]
- CONFIRMED | Hélène Smith, 1895 to 1898 | research/the-accent-of-nowhere/tongues/martian/sanscritoid.json | "date": "6 mars 1895 to 18 mai 1898; printed 1900" [head-sanscritoid]
- CONFIRMED | Frau T., in Pfister, published 1912 | https://archive.org/download/jb319122/JB_III_1912_2_djvu.txt | second half-volume: "FRANZ DEUTICKE. 1912." [head-frau-t]
- CONFIRMED | François Rabelais, 1532 | https://fr.wikisource.org/wiki/Pantagruel | "Date de publication 1532" [head-rabelais]
- CONFIRMED | François Sudre, 1866, from a grammar of 1902 | https://www.ifost.org.au/~gregb/solresol/books.html | "Paris: 1866 (480+ p.)" for Sudre's book (posthumous; he died 1862) [head-solresol]
- CONFIRMED | The Logical Language Group, 1987 | https://lojban.org/publications/cll/cll_v1.1_xhtml-section-chunks/chapter-about.html | "led since 1987 by The Logical Language Group" [head-lojban]
- CONFIRMED | Dante, by 1321 | https://en.wikisource.org/wiki/1911_Encyclop%C3%A6dia_Britannica/Dante | "(1265-1321)" [head-dante]
- CONFIRMED | Thomas More and Peter Giles, 1516 | https://archive.org/download/utopiaofsirthoma00moreuoft/utopiaofsirthoma00moreuoft_djvu.txt | "printed by Thierry Martin, at Louvain, towards the end of 1516" [head-more]
- CONFIRMED | a speaker the printing does not name, written down by Archibald M'Kerrell, 1831 | https://archive.org/download/amorningvisitto00antigoog/page/n37_w1400.jpg | p.35 (image; NOTES.md): collected "in the beginning of this present year, (1831,)"; no name given [head-mkerrell]
- CONFIRMED | and Claude did it by ceasing to make one | research/the-accent-of-nowhere/results/figures.json | Claude B gains 0.10 to 0.38; other makers' 0.49, 0.55, 1.93 [dek-claude-ceasing]
- CONFIRMED | the literary man whose case William James sent to the Society for Psychical Research | https://archive.org/download/proceedingsofsoc12soci/proceedingsofsoc12soci_djvu.txt | p.278: "a literary man, aged 39" [essay-le-baron-literary]
- CONFIRMED | registered in advance as the work of writers who were not trying to leave English | research/the-accent-of-nowhere/PREREGISTRATION.md | C2: "made by English writers who were not trying to leave English" (the record's "where" names the homes table; the words are under Controls) [essay-carroll-lear-registered]
- CONFIRMED | gets about as far from home as Carroll and Lear did with slithy and runcible | research/the-accent-of-nowhere/tongues/nonsense-english/lear-nonsense.words.json | runcible here, slithy in carroll-jabberwocky.words.json; pooled control 0.48 [essay-slithy-runcible]
- CONFIRMED | The transcripts of people speaking in tongues sit close to home, from 0.50 to 0.59. | research/the-accent-of-nowhere/results/figures.json | Frau T. 0.50, M'Kerrell 0.51, Simon 0.52, Le Baron 0.52, Lina 0.59 [essay-glossolalia-close]
- CONFIRMED | Velimir Khlebnikov's gods and his Bobeobi | https://ru.wikisource.org/wiki/%D0%91%D0%BE%D0%B1%D1%8D%D0%BE%D0%B1%D0%B8_%D0%BF%D0%B5%D0%BB%D0%B8%D1%81%D1%8C_%D0%B3%D1%83%D0%B1%D1%8B_(%D0%A5%D0%BB%D0%B5%D0%B1%D0%BD%D0%B8%D0%BA%D0%BE%D0%B2) | «Бобэоби пелись губы…» under Khlebnikov's name [essay-khlebnikov-bobeobi]
- OBSERVED | which the study has only in an electronic copy it could not check against the printing | research/the-accent-of-nowhere/tongues/enochian/enochian-calls-equinox-1912-etext.json | "NOT PROOFREAD against the 1912 printing: no facsimile was obtained" [essay-enochian-1912-etext]
- CONFIRMED | the same Calls move from 0.94 to 0.89 on English's ruler and from 0.83 to 0.57 on Latin's | research/the-accent-of-nowhere/results/figures.json | 1659: 0.94 English, 0.83 Latin; 1912: 0.89, 0.57 [essay-enochian-like-for-like]
- CONFIRMED | It is the word-for-word French the medium herself gave in trance, speaking as one Esenale | https://archive.org/download/desindeslaplan00flou/desindeslaplan00flou_djvu.txt | p.204: "leur mot à mot français donné par Esenale" (true of 38 of the 40 texts; see the added line on texts 16 and 19) [wfw-medium]
- CONFIRMED | the bracketed letters that mark fragments | research/the-accent-of-nowhere/tongues/martian/martian.json | labels [A], [B], [C] before fragments in texts 23, 35, 36 [wfw-fragments]
- CONFIRMED | familiar neighbours between a fifth and six tenths | research/the-accent-of-nowhere/results/figures.json | Portuguese 0.19, Dutch 0.45, English 0.59 [calib-fifth-six-tenths]
- CONFIRMED | they are left out of predictions H3 to H5 | research/the-accent-of-nowhere/PREREGISTRATION.md | H3, H4, H5 each say "with at least 100 types" [rulers-h3-h5]
- CONFIRMED | over-use z by the measure as registered and 7 by the repaired one | research/the-accent-of-nowhere/results/predictions.json | H5: 8 of 11 registered (zShare 0.636 matched = 7 of 11); figures h5.z 8, h5.zm 7 [z-both-measures]
- CONFIRMED | the Enochian, set aside in advance | research/the-accent-of-nowhere/PREREGISTRATION.md | H3: "other than the Enochian of H2c" [score-h3-enochian]
- CONFIRMED | have one of z, x, k, q in their top three | research/the-accent-of-nowhere/PREREGISTRATION.md | H5: "at least one of z, x, k, q is among the three most over-used letters" [score-h5-letters]
- CONFIRMED | more often than French matches any real language of the panel | research/the-accent-of-nowhere/results/m6.json | reference Italian 13.3% of verses against Martian's 104 of 122 lines [score-h8-wording]
- CONFIRMED | a creole of largely French words, spelled by sound | https://en.wikipedia.org/wiki/Haitian_Creole | "its vocabulary largely derives from 18th-century French"; "a phonemic orthography" [score-h2a-creole]
- CONFIRMED | The three guesses about a historical tongue's nearest ear (H2a to H2c) all failed | research/the-accent-of-nowhere/results/predictions.json | H2a, H2b, H2c all "failed" [score-failed-three]
- CONFIRMED | are named here by short forms of the ids the rail gives them; the full ids are in the page's data | research/the-accent-of-nowhere/build-page.mjs | const SHORT maps the four full ids; data.json draws carry the full ids [rail-short-ids]
- CONFIRMED | They give up less of their structure to get there: the order of their letters still carries 0.49, 0.55 and 1.93 bits. | research/the-accent-of-nowhere/results/figures.json | B gains: nemotron-ultra 0.49, cohere-north 0.55, nemotron-nano 1.93 [rail-gain]
- OBSERVED | the rule for choosing the real languages, the 46 named ears, a home for each maker then on the list and twelve predictions were committed | git show e198db40a6:research/the-accent-of-nowhere/PREREGISTRATION.md | recreated: the committed file holds the panel rule, the 46 names, "Homes, fixed now" and H1 to H12 [prereg-what]
- CONFIRMED | differed from the page images in 172 places in the Martian texts and their French, 115 of them in the Martian itself | research/the-accent-of-nowhere/tongues/martian/martian.json | recreated: 172 entries, 115 martian, 57 gloss [limits-ocr-count]
- CONFIRMED | The repairs to the measures are among the departures published beside this page. | research/the-accent-of-nowhere/DEVIATIONS.md | D5 (letters) and D7 (Solresol) are published beside the page [limits-departures]
- CONFIRMED | One of the 730 is itself invented: Esperanto, whose New Testament is in the corpus. | research/the-accent-of-nowhere/panel/atlas.json | "code": "epo", "name": "Esperanto"; family "Constructed language" [epo-atlas]
- CONFIRMED | The rule that built the atlas did not ask how a language began | research/the-accent-of-nowhere/PREREGISTRATION.md | the atlas rule asks for licence, download, 7,000 verses and Latin letters only [epo-rule]
- CONFIRMED | Esperanto was on the registered list, with Russian and then Polish as its homes, and no text of it was gathered | research/the-accent-of-nowhere/PREREGISTRATION.md | "/ L. L. Zamenhof / Esperanto / Russian / Polish /"; D8: never gathered [epo-registered]
- CONFIRMED | The words of its New Testament sit at 0.77 on Russian's ruler, with 131 of the other 729 languages nearer, and at 0.73 on Polish's. | research/the-accent-of-nowhere/results/figures.json | recreated from the shipped data.json: Russian E 0.769, rank 131 of 729; Polish E 0.728 [epo-figures]
- CONFIRMED | Of the 46 ears, the two on whose rulers it sits nearest are Croatian and Swedish. | research/the-accent-of-nowhere/results/figures.json | recreated from results/rulers.json over all 46: Croatian rank 7 (E 0.491), Swedish rank 7 (E 0.499) [epo-nearest]
- OBSERVED | neither text names a mother tongue for Simon, Frau T. or M'Kerrell's speaker | research/the-accent-of-nowhere/tongues/glossolalia/NOTES.md | no sentence names a mother tongue for Simon; Frau T. and M'Kerrell entries agree [limits-no-mother-tongue]
- CONFIRMED | Pfister explains a word of Simon's from Swiss German | https://archive.org/download/jb319111/JB_III_1911_1_djvu.txt | p.454: "Schweizerdeutsch für „auch ein Fang“." [limits-simon-swiss]
- CONFIRMED | pieces of Lina's from the Zurich dialect | https://archive.org/download/jb319122/JB_III_1912_2_djvu.txt | p.776: "(Zürcherdeutsch für „steigst“)" [limits-lina-zurich]
- CONFIRMED | Simon, an analysand of Oskar Pfister | https://archive.org/download/jb319111/JB_III_1911_1_djvu.txt | "meine an den Analysanden gerichteten Worte" (p.444 note, seen on the kept image) [simon-header]
- WRONG | Spoken in ecstasy | https://archive.org/download/jb319111/page/n449_w1400.jpg | MED: the table's group heading over Le Baron, Simon, Lina, Frau T. and M'Kerrell's speaker. Pfister p.444 on Simon: "Bei vollem Bewußtsein spricht er hierauf rasch" (kept image viewed); p.770 on Frau T.: "gerät wiederum in religiöse Erregung (eigentliche Ekstase war es nicht)" (kept OCR, jahrbuch-III-2); no source read calls the others' state ecstasy; README says the essay was reworded for exactly this [added: not in the record]
- WRONG | The French is not Flournoy's. | research/the-accent-of-nowhere/tongues/martian/martian.json | MINOR: true of 38 of the 40 texts; under texts 16 and 19 (3 of the 122 counted lines) the French is Flournoy's own, in parentheses, and text 19 was never translated: "il n'y a jamais eu de traduction de ce texte" (p.211; the file's own doubts) [added: not in the record]
- WRONG | real languages nearer to home, of 729 | research/the-accent-of-nowhere/results/figures.json | MINOR: t.khlebnikov.of and t.kruchenykh.of are 730, not 729: Russian is not one of the atlas's 730, so its ruler counts 730 others (the same goes for "the median language of the other 729" on Russian's ruler) [added: not in the record]
- WRONG | The verifier recomputes every figure on this page that can be recomputed from what the page ships | verify-the-accent-of-nowhere.mjs | MINOR: its run prints m6.* 7, epo.* 6, h4.* 1, h5.* 2 as "taken from the lab, not rechecked here", yet they follow from the shipped files: recreated from data.json and engine.mjs alone, 391 pairs, r 0.685, 30.4% against 13.9%, shuffle maximum 0.218; Esperanto on Russian's ruler 0.769, rank 131 of 729, on Polish's 0.728; h4.negm, h5.zm, h5.exoticm are counts over letter figures it already computes [added: not in the record]
- CONFIRMED | H3. No ecstatic, revealed or artistic tongue of a hundred words gets as far as the median language. [...] The farthest is Hildegard's, at 0.79; the Enochian, set aside in advance, is at 0.94. | research/the-accent-of-nowhere/results/predictions.json | H3 held: Lingua Ignota 0.789, Martian 0.423, Le Baron 0.520, Simon 0.522, Khlebnikov 0.595; PREREGISTRATION.md H3 (70%) sets aside the Enochian of H2c [added: not in the record]
- CONFIRMED | The 46 ears, their translations and licences (the licence column) | research/the-accent-of-nowhere/sources/kept/panel/ebible-licences.tsv | recreated: 40 of the 46 licences printed match the corpus's licence table; eng-asv and cmn-cu89s carry "public domain" in its Copyright field; slk, fin, kor, mri came from eBible downloads (panel/build-pocket.mjs reads their about pages) [added: not in the record]
## Findings
- [fixed] MED: the rulers table's group heading over the five transcripts of people speaking in tongues read "Spoken in ecstasy". Re-read: Pfister has Simon speaking "Bei vollem Bewußtsein" (Jahrbuch III, 1911, p. 444) and says of Frau T. "eigentliche Ekstase war es nicht" (1912, p. 770). The heading now reads "Spoken in tongues" (build-page.mjs, CLASS_NAME), as the chart's key does; the essay had already been reworded and this heading was missed.
- [fixed] Morgenstern card: "the later text that German Wikisource transcribes, of 1932, has none". The kept Wikisource copy marks all three m with overlines, lost when the scout's script stripped the markup. The card now says the 1932 text "marks the same three"; the tongue file's transcription note and record entry morgenstern-bar are corrected. The letters scored are unchanged.
- [fixed] "The verifier recomputes every figure on this page that can be recomputed from what the page ships": twelve figures it takes from the lab could be recomputed from the shipped files (the exploratory word-for-word figures, Esperanto's place, the repaired letter counts). The sentence now says the verifier recomputes the figures by its own arithmetic and lists by name the ones it takes from the lab. Teaching it the twelve is an open edge of ledger P36.
- [fixed] "The French is not Flournoy's": under texts 16 and 19, which were never translated, the French is Flournoy's own, in parentheses (martian.json; Flournoy p. 211). The page now says "Nearly all of the French is not Flournoy's" and names the exception; record entry wfw-flournoy-own.
- [fixed] "the Universal Declaration of Human Rights in 309 languages" and "34 of 308 languages": panel/udhr.json holds 309 texts in 294 language codes. The page now says "309 texts" and "34 of the other 308 texts".
- [fixed] "real languages nearer to home, of 729": the two Russian-home rows rank out of 730, since Russian is not among the atlas's languages. The table header now says so, and the essay's "the median language of the other 729" now reads "the median of the atlas's other languages".
- [fixed] near-miss, the dek: "about as far as Dutch is from German" (0.45) was nearer home than every one of the five transcripts (0.50 to 0.59). Now "about as far from home as English is from French" (0.588; the farthest transcript, Lina's, is 0.594). The ledger entry and the log were changed to match.
- [fixed] near-miss: "The two built from tables and procedure go past the median": Wilkins's 1.06 rests on 92 words with an interval of 0.93 to 1.18. Now "reach the median", with Wilkins's interval printed and Lojban's whole interval past it.