---
slug: eskimo-words-for-snow
passed: 2026-10-11
by: claude-vibrant-lamport-79znpp, two independent Claude checkers on oversight/claims-pass.md (source first, every claim the page makes), an adjudicator who re-read each finding at its source and a skeptic who tried to make the page right; every HIGH and every lone WRONG re-read by the instance at the source before fixing; fixes made by a fixer agent and read against the sources by a separate fix-checker
checked: 61
confirmed: 35
wrong: 21
unverifiable: 3
observed: 2
stale: 0
---
# Claims pass: The Count That Snowballed

Reader-side check, run by both checkers and the adjudicator before the fixes: `curl https://artwaste.land/checks/research/eskimo-words-for-snow/verify.mjs` into an empty directory returned 200, 7,470 bytes, byte-identical to research/eskimo-words-for-snow/verify.mjs; `node verify.mjs` exited 0 with 20 ok and 0 failed, and the figures it printed matched the page. The page's own run instruction named a repository path that returns 404 on the site (F16). A green run settled little: the verifier holds only constants typed in by the page's author, reads no source and checks no quotation, and it passed while the NYT and Boas quotations, Whorf's "no word-count", Woodbury's "17" and the doubling timeline were wrong. The self-reading probe does not apply: verify.mjs reads no file (no import, require, readFile, fetch or argv) and the page prints no transcript of its output.

Of 61 claims, 35 are confirmed, 10 WRONG, 11 MINOR (counted as wrong), 3 unverifiable and 2 observed. Of 21 merged findings, 8 WRONG and 10 MINOR stand, and 3 were refuted and declined (F19 to F21). The headline corrections: the stepper's doubling chain, each writer copying the last and rounding up, is in neither Martin nor Pullum, who record numbers made up in no order; the NYT 1984 quotation was invented; Whorf's 5 was tagged grounded, though he cited no data and his article's drawing reads "ESKIMO - THREE WORDS"; the 17 lexemes are the page's selection, not Woodbury's count; the 50 came from a play and the 200 from 1984 TV forecasts; the 50, 200 and 400 rungs quoted sentences no source gives; and by the page's own measure English does not have more snow roots. Seventeen findings are fixed, the eight WRONG ones named in a Corrected 2026-10-11 line on the page, and the rewritten verify.mjs passes 25 of 25 from the repository. F15, the shared placard sentence that the check "works the answer out again from scratch", is held by note 8d17ac for a fix in scripts/check-placard.mjs.

## Claims
- CONFIRMED | "Everyone “knows” the Eskimos have fifty words for snow — or a hundred, or four hundred." | Pullum 1989 p. 278 (https://theread.me/files/Pullum%20-%20The%20great%20Eskimo%20vocabulary%20hoax%201989.pdf); Pullum 1991 p. 161 (https://pdf.stafforini.com/pullum-1991-great-eskimo-vocabulary.pdf) | "it is "fifty"" ... ""one hundred" (in a New York Times editorial on February 9)" ... "numbers as high as four hundred (repeat, 400)"
- CONFIRMED | In 1911 the anthropologist Franz Boas mentioned, in passing, that the Eskimo language has a few separate roots for snow where English uses one word plus modifiers. | Boas 1911 pp. 25-26, https://archive.org/download/handbookofameric00boas/handbookofameric00boas_djvu.txt; Pullum 1989 p. 276, https://theread.me/files/Pullum%20-%20The%20great%20Eskimo%20vocabulary%20hoax%201989.pdf | Boas: "Another example of the same kind, the words for snow in Eskimo, may be given." Pullum: "Boas' point is simply that English expresses these notions by phrases involving the root snow"
- MINOR | It was a grammar footnote ... Then the footnote left home. | Boas 1911 pp. 25-26, https://archive.org/download/handbookofameric00boas/handbookofameric00boas_djvu.txt; Kaplan, https://www.uaf.edu/anlc/research-and-resources/resources/archives/inuit_snow_terms.php | The passage is running text of the Introduction, pp. 25-26. Kaplan: "it was not in the section of the introduction called 'Influence of Environment on Language' ... but rather in a less enlightening section called 'Limitation on the Number of Phonetic Groups Expressing Ideas.'"; F13. It is a passing example in the main text, not a footnote.
- CONFIRMED | no more a boast than noting that English has separate words for rain, fog, dew and ice | Boas 1911 p. 25, https://archive.org/download/handbookofameric00boas/handbookofameric00boas_djvu.txt | "the idea of water is expressed in a great variety of forms ... (lake) ... (river and brook) ... still other terms express water in the form of rain, dew, wave, and foam"; The page's own analogy, not attributed to Boas; his English example was water.
- CONFIRMED | "Boas counted four roots in 1911 as a dry grammar example" / "Boas, 1911: four roots — aput (snow on the ground), qana (falling snow), piqsirpoq (drifting snow), qimuqsuq (a snowdrift). Verbatim from the Handbook" | Boas 1911 pp. 25-26, https://archive.org/download/handbookofameric00boas/handbookofameric00boas_djvu.txt; Pullum 1989 p. 276, https://theread.me/files/Pullum%20-%20The%20great%20Eskimo%20vocabulary%20hoax%201989.pdf; Martin 1986 p. 418 (page image), https://web.archive.org/web/2010id_/http://www.ldc.upenn.edu/myl/llog/LauraMartinEskimoSnowWords.pdf | Boas: "one word, aput, expressing snow on the ground; another one, qana, falling snow; a third one, piqsirpoq, drifting snow; and a fourth one, qimuqsuq, a snowdrift." Pullum: "the apparently distinct roots aput ... qana ... piqsirpoq ... qimuqsuq". Martin: "a brief citation of four lexically unrelated words for snow in Eskimo"; Boas wrote 'word'/'term'; 'roots' follows Pullum. The four glosses are verbatim.
- MINOR | “…one word, aput, expressing SNOW ON THE GROUND; another, qana, FALLING SNOW; a third, piqsirpoq, DRIFTING SNOW; and a fourth, qimuqsuq, A SNOWDRIFT.” | Boas 1911 pp. 25-26, https://archive.org/download/handbookofameric00boas/handbookofameric00boas_djvu.txt | "Here we find one word, aput, expressing snow on the ground; another one, qana, falling snow; a third one, piqsirpoq, drifting snow; and a fourth one, qimuqsuq, a snowdrift."; F11. 'one' is dropped three times without marking; the meaning is unchanged.
- CONFIRMED | He made no quantitative or perceptual claim. / a grammar example, no claim of “more” words or different perception | Boas 1911 p. 26, https://archive.org/download/handbookofameric00boas/handbookofameric00boas_djvu.txt; Martin 1986 p. 418 (page image), https://web.archive.org/web/2010id_/http://www.ldc.upenn.edu/myl/llog/LauraMartinEskimoSnowWords.pdf | Boas: "It seems fairly evident that the selection of such simple terms must to a certain extent depend upon the chief interests of a people". Martin: "Although for Boas the example illustrated a similarity between English and "Eskimo," Whorf reorients it to contrast them"
- CONFIRMED | Franz Boas, Handbook of American Indian Languages (1911), introduction. | Boas 1911 (archive.org handbookofameric00boas), https://archive.org/download/handbookofameric00boas/handbookofameric00boas_djvu.txt | page header "26 BUREAU OF AMERICAN ETHNOLOGY [bull. 40"; contents "Limitation of the number of phonetic groups expressing ideas 24"
- CONFIRMED | Boas’s real point was the modest, true one: every language carves the world finely where its speakers need the distinctions | Boas 1911 p. 26, https://archive.org/download/handbookofameric00boas/handbookofameric00boas_djvu.txt | "where it is necessary to distinguish a certain phenomenon in many aspects, which in the life of the people play each an entirely independent role, many independent words may develop, while in other cases modifications of a single term may suffice."
- CONFIRMED | “We have the same word for falling snow, snow on the ground, snow packed hard like ice, slushy snow, wind-driven flying snow… To an Eskimo, this all-inclusive word would be almost unthinkable.” Quoted verbatim from “Science and Linguistics.” | The Technology Review April 1940 p. 231, https://archive.org/download/MIT-Technology-Review-1940-04/MIT-Technology-Review-1940-04_djvu.txt; Pullum 1989 pp. 276-277, https://theread.me/files/Pullum%20-%20The%20great%20Eskimo%20vocabulary%20hoax%201989.pdf | "We have the same word for falling snow, snow on the ground, snow packed hard like ice, slushy snow, wind-driven flying snow — whatever the situation may be. To an Eskimo, this all-inclusive word would be almost unthinkable; he would say that ..."; The ellipsis marks the cut correctly.
- CONFIRMED | listed five snow-conditions English “lumps” (falling / on-ground / packed-hard / slushy / wind-driven) | The Technology Review April 1940 p. 231, https://archive.org/download/MIT-Technology-Review-1940-04/MIT-Technology-Review-1940-04_djvu.txt | "We have the same word for falling snow, snow on the ground, snow packed hard like ice, slushy snow, wind-driven flying snow"; F20 refuted. 'lumps' is the page's scare quote, not Whorf's word ('lump' does not occur; the positive control was found), but it restates his claim accurately.
- CONFIRMED | Whorf added a moral Boas never made / added the linguistic-relativity moral Boas never made | Martin 1986 p. 418 (page image), https://web.archive.org/web/2010id_/http://www.ldc.upenn.edu/myl/llog/LauraMartinEskimoSnowWords.pdf; The Technology Review April 1940 p. 231, https://archive.org/download/MIT-Technology-Review-1940-04/MIT-Technology-Review-1940-04_djvu.txt | Martin: "Although for Boas the example illustrated a similarity between English and "Eskimo," Whorf reorients it to contrast them". Whorf: "he would say that falling snow, slushy snow, and so on, are sensuously and operationally different, different things to contend with"
- WRONG | He gave no Eskimo word-count himself. | The Technology Review 42(6), April 1940, p. 230, Fig. 2, https://archive.org/download/MIT-Technology-Review-1940-04/page/n16_w1200.jpg; Martin 1986 p. 419 (page image), https://web.archive.org/web/2010id_/http://www.ldc.upenn.edu/myl/llog/LauraMartinEskimoSnowWords.pdf | Fig. 2: "ENGLISH - ONE WORD (SNOW) / ESKIMO - THREE WORDS". Martin: Brown's "three Eskimo words for snow" was "an assertion apparently based solely on a drawing in Whorf's paper."; F3. Martin's 'failure to cite specific data, numbers' concerns Whorf's text; the article's own figure prints a count.
- WRONG | "1940 — Benjamin Lee Whorf · claimed 5" [tag 'grounded'; legend 'grounded in a real source'] / "The two grounded rungs — Boas’s 4 and Whorf’s 5 — carry real quotes from real sources." | Martin 1986 p. 418 (page image), https://web.archive.org/web/2010id_/http://www.ldc.upenn.edu/myl/llog/LauraMartinEskimoSnowWords.pdf; Pullum 1989 p. 277, https://theread.me/files/Pullum%20-%20The%20great%20Eskimo%20vocabulary%20hoax%201989.pdf; Kaplan, https://www.uaf.edu/anlc/research-and-resources/resources/archives/inuit_snow_terms.php | Martin: "Of particular significance is Whorf's failure to cite specific data, numbers, or sources. His English glosses suggest as many as five words ... his apparently casual revisions of numbers and glosses are but the first mistreatments to which the original data have been subjected." Pullum: "Whorf's statement has illicitly inflated Boas' four terms to at least seven". Kaplan: "he clearly suggests that Eskimo languages have five or more snow words."; F3. The quote is real. The figure is not grounded in data, and it is 'five or more', not five.
- CONFIRMED | Benjamin Lee Whorf, “Science and Linguistics,” MIT Technology Review 42 (1940) | The Technology Review April 1940 (archive.org MIT-Technology-Review-1940-04), https://archive.org/download/MIT-Technology-Review-1940-04/MIT-Technology-Review-1940-04_djvu.txt; Pullum 1989 references, https://theread.me/files/Pullum%20-%20The%20great%20Eskimo%20vocabulary%20hoax%201989.pdf | "SCIENCE AND LINGUISTICS ... By Benjamin Lee Whorf"; Pullum: "Technology Review (MIT) 42, 6 (April), 229-231,247-248"; In 1940 the title was 'The Technology Review', MIT's magazine.
- CONFIRMED | The linguist Laura Martin traced this in 1986 | https://api.crossref.org/works/10.1525/aa.1986.88.2.02a00080; Krupnik & Müller-Wille 2010 p. 390, https://gwern.net/doc/psychology/linguistics/2010-krupnik.pdf | Crossref: American Anthropologist 88(2) 418-423, 1986-06. Krupnik: "In 1986, Laura Martin, linguistic anthropologist and specialist in Mayan languages, produced a short essay"; 'Linguist' is fair for a linguistic anthropologist.
- CONFIRMED | Geoffrey Pullum made it famous in 1991 as “The Great Eskimo Vocabulary Hoax” | Kaplan, https://www.uaf.edu/anlc/research-and-resources/resources/archives/inuit_snow_terms.php; Pullum 1989 PDF, https://theread.me/files/Pullum%20-%20The%20great%20Eskimo%20vocabulary%20hoax%201989.pdf; Pullum 1991 p. 160, https://pdf.stafforini.com/pullum-1991-great-eskimo-vocabulary.pdf | Kaplan: "Linguist Geoffrey Pullum in 1991 published The Great Eskimo Vocabulary Hoax"; Pullum 1989: "Natural Language and Linguistic Theory 7 (1989) 275-281"; F19 refuted. The essay first appeared in 1989; the 1991 book is what made it widely known.
- WRONG | the press doubled the figure for decades with no source behind it / each writer copied the last and rounded up / Each retelling rounds up. / The number simply kept doubling. | Pullum 1989 pp. 277-278, https://theread.me/files/Pullum%20-%20The%20great%20Eskimo%20vocabulary%20hoax%201989.pdf; Martin 1986 p. 420 (page image), https://web.archive.org/web/2010id_/http://www.ldc.upenn.edu/myl/llog/LauraMartinEskimoSnowWords.pdf | Pullum: "arbitrary numbers are just made up as the writer thinks appropriate for the readership" ... "nine, forty-eight, a hundred, two hundred, who cares?" Martin: "Consider a diverse random sample of recent references" (fifty 1978; nine, one hundred, two hundred 1984); F1. The broad growth with no source stands (Martin: "The tendency to inflate the numbers"); the doubling chain does not.
- WRONG | "The “100” reached the New York Times (9 Feb 1984); the rest cite each other." / "untraceable — cites the last teller" / "Cites the previous teller, not any data." | Pullum 1989 p. 278, https://theread.me/files/Pullum%20-%20The%20great%20Eskimo%20vocabulary%20hoax%201989.pdf; Pullum 1991 p. 161, https://pdf.stafforini.com/pullum-1991-great-eskimo-vocabulary.pdf; Martin 1986 p. 420 (page image), https://web.archive.org/web/2010id_/http://www.ldc.upenn.edu/myl/llog/LauraMartinEskimoSnowWords.pdf | Pullum: "arbitrary numbers are just made up"; the 400's author admitted "to having no source for the number whatsoever"; Martin records the tellers (a play, a trivia book, the Times citing Whorf, TV forecasts) with no citation chain; F1. No source shows any popular figure citing a previous teller; the one that cited anyone, the Times, cited Whorf.
- CONFIRMED | The “100” reached the New York Times (9 Feb 1984) | NYT, Feb. 9, 1984, Section A, Page 30, https://web.archive.org/web/20210413175932id_/http://www.nytimes.com/1984/02/09/opinion/topics-well-used.html | "Benjamin Lee Whorf, the linguist, once reported on a tribe that distinguishes 100 types of snow - and has 100 synonyms (like tipsiq and tuva) to match."
- MINOR | The New York Times printed “a hundred” in a 1984 editorial | NYT Feb 9 1984 (Wayback copy), https://web.archive.org/web/20210413175932id_/http://www.nytimes.com/1984/02/09/opinion/topics-well-used.html | "distinguishes 100 types of snow - and has 100 synonyms"; F2. The date and figure are right; the paper printed '100' and credited Whorf, not the words 'a hundred'.
- WRONG | “…the Eskimos have a hundred words for snow.” (New York Times editorial (9 Feb 1984)) | NYT Feb 9 1984, https://web.archive.org/web/20210413175932id_/http://www.nytimes.com/1984/02/09/opinion/topics-well-used.html; Martin 1986 p. 420 (page image), https://web.archive.org/web/2010id_/http://www.ldc.upenn.edu/myl/llog/LauraMartinEskimoSnowWords.pdf | "It's enough to make us envy the Eskimos. Benjamin Lee Whorf, the linguist, once reported on a tribe that distinguishes 100 types of snow - and has 100 synonyms (like tipsiq and tuva) to match."; F2. The quoted sentence is not in the editorial.
- CONFIRMED | "The number reached print — with no grammar behind it." | NYT Feb 9 1984, https://web.archive.org/web/20210413175932id_/http://www.nytimes.com/1984/02/09/opinion/topics-well-used.html; Kaplan, https://www.uaf.edu/anlc/research-and-resources/resources/archives/inuit_snow_terms.php | NYT credits only "Benjamin Lee Whorf, the linguist"; Kaplan: articles "often pull numbers out of thin air, nine in one case, one hundred in another"
- CONFIRMED | The rung Pullum (1991) singles out. | Pullum 1989 p. 278 (reprinted 1991), https://theread.me/files/Pullum%20-%20The%20great%20Eskimo%20vocabulary%20hoax%201989.pdf | "the New York Times returned to the topic four years to the day after committing itself to the figure of one hundred"
- WRONG | no number above five is traceable to any linguistic source. / the figure had no basis in any grammar, and never did. | Krupnik & Müller-Wille, in SIKU: Knowing Our Ice (2010) pp. 388-389 and Table 16.2, https://gwern.net/doc/psychology/linguistics/2010-krupnik.pdf | "Several contemporary Inuit/Eskimo snow lexicons (see Table 16.2) range from 20 to 35 terms ... More extensive lists feature up to 60–80 terms ... The largest known list, compiled from the Nunavik Inuktitut dictionary (Schneider 1985), has more than 100 terms, including numerous derivative forms."; F9 (PARTLY). True in context: the popular tellers cited no linguistic source (Kaplan: "all without reference to primary sources of information"). False as worded, and 'never did' is contradicted by SIKU and by the page's own 17.
- CONFIRMED | by 1978 · 50 | Martin 1986 p. 420 (page image), https://web.archive.org/web/2010id_/http://www.ldc.upenn.edu/myl/llog/LauraMartinEskimoSnowWords.pdf; Pullum 1989 p. 278, https://theread.me/files/Pullum%20-%20The%20great%20Eskimo%20vocabulary%20hoax%201989.pdf | Martin: ""fifty" in Lanford Wilson's 1978 play The Fifth of July"
- WRONG | by 1978 · Popular press & textbooks ... "No linguistic source. The first big jump — documented as citation drift by Laura Martin (1986)." | Martin 1986 pp. 418-420 (page images), https://web.archive.org/web/2010id_/http://www.ldc.upenn.edu/myl/llog/LauraMartinEskimoSnowWords.pdf; Pullum 1989 p. 278, https://theread.me/files/Pullum%20-%20The%20great%20Eskimo%20vocabulary%20hoax%201989.pdf | Martin: "Consider a diverse random sample of recent references: ... "fifty" in Lanford Wilson's 1978 play The Fifth of July". Pullum: "In Lanford Wilson's 1978 play The Fifth of July it is "fifty"."; F5. A play, not the press or textbooks; Martin calls it neither a first jump nor citation drift.
- UNVERIFIABLE | “…about fifty words for snow.” | not kept: searched Martin 1986 pp. 418-421 (https://web.archive.org/web/2010id_/http://www.ldc.upenn.edu/myl/llog/LauraMartinEskimoSnowWords.pdf), Pullum 1989 in full (https://theread.me/files/Pullum%20-%20The%20great%20Eskimo%20vocabulary%20hoax%201989.pdf), Pullum 1991 pp. 160-171 (https://pdf.stafforini.com/pullum-1991-great-eskimo-vocabulary.pdf), and WebSearch for the line in Wilson's Fifth of July (it found pages describing the play's '50', e.g. https://www.sfu.ca/~ramccall/weekten/tsld018.htm, but no script text) | Martin and Pullum quote only "fifty"; F8. Set as a quotation with no source.
- MINOR | “…two hundred words for snow.” | Martin 1986 p. 420 (page image), https://web.archive.org/web/2010id_/http://www.ldc.upenn.edu/myl/llog/LauraMartinEskimoSnowWords.pdf | "the inevitable local television references to "two hundred words" during winter snow forecasts"; F8. Martin's quoted fragment is 'two hundred words'; the page extends it and attributes it to the press.
- WRONG | 1980s–90s · 200 · Popular press ... Cites the previous teller, not any data. | Martin 1986 p. 420 (page image), https://web.archive.org/web/2010id_/http://www.ldc.upenn.edu/myl/llog/LauraMartinEskimoSnowWords.pdf; Pullum 1989 p. 278, https://theread.me/files/Pullum%20-%20The%20great%20Eskimo%20vocabulary%20hoax%201989.pdf | Martin: "(e.g., WEWS-Cleveland, 1984)"; Pullum: "From 1984 alone ... "two hundred" (in a Cleveland TV weather forecast)"; F6. Local TV in 1984, the same year as the Times's 100.
- UNVERIFIABLE | “…four hundred words for snow.” | not kept: Pullum 1991 pp. 160-161 (https://pdf.stafforini.com/pullum-1991-great-eskimo-vocabulary.pdf) gives the figure without wording; nothing else found | "numbers as high as four hundred (repeat, 400)"; F8.
- MINOR | by 2000s · 400 · Popular press | Pullum 1991 p. 161 (archive.org greateskimovocab0000pull scan), https://pdf.stafforini.com/pullum-1991-great-eskimo-vocabulary.pdf | "Professor Martin has seen in writing numbers as high as four hundred (repeat, 400) ... The four hundred figure came from a piece by a would-be author who admitted (under questioning by a magazine fact-checker) to having no source for the number whatsoever."; F7. In writing by 1991, not first in the 2000s.
- CONFIRMED | The high-water mark Pullum traces. | Pullum 1991 p. 161, https://pdf.stafforini.com/pullum-1991-great-eskimo-vocabulary.pdf | "numbers as high as four hundred (repeat, 400) ... The four hundred figure came from a piece by a would-be author"; F7. Checker B found no 400 in the 1989 essay; it is in the 1991 chapter's new preface.
- CONFIRMED | True roots: West Greenlandic has 2 (qanik, aput) / qanik 'snow in the air / snowflake', aput 'snow on the ground' | Martin 1986 p. 422 n. 2 (page image), https://web.archive.org/web/2010id_/http://www.ldc.upenn.edu/myl/llog/LauraMartinEskimoSnowWords.pdf; Kaplan, https://www.uaf.edu/anlc/research-and-resources/resources/archives/inuit_snow_terms.php; Pullum 1989 p. 280, https://theread.me/files/Pullum%20-%20The%20great%20Eskimo%20vocabulary%20hoax%201989.pdf | Martin: "In West Greenlandic, these roots are qanik 'snow in the air; snowflake' and aput 'snow (on the ground)' (Schultz-Lorentzen 1927; cf. Boas's data)."; This is Martin's footnote, quoted by Kaplan; research/eskimo-words-for-snow/sources.md:66-68 presents it as Kaplan's own words.
- CONFIRMED | Proto-Eskimo is reconstructed with 3 stems (*qaniɣ, *aniɣu, *apun) ... West Greenlandic keeps two of them. | Kaplan, https://www.uaf.edu/anlc/research-and-resources/resources/archives/inuit_snow_terms.php | "The Comparative Eskimo Dictionary lists three reconstructed Proto-Eskimo noun stems ...: *qaniɣ 'falling snow', *aniɣu 'fallen snow', and *apun 'snow on the ground' are the three basic roots found in all Eskimo languages and dialects, except, unfortunately, West Greenlandic, which lacks *aniɣu"
- CONFIRMED | Lawrence Kaplan / Alaska Native Language Center, “Inuit Snow Terms: How Many and What Does It Mean?” | Kaplan, ANLC page, https://www.uaf.edu/anlc/research-and-resources/resources/archives/inuit_snow_terms.php | "Inuit Snow Terms: How Many and What Does It Mean? by Larry Kaplan"
- WRONG | Strip every suffix away and ask how many distinct roots mean “snow”: just two. By this measure English (snow, sleet, slush, hail…) has more. | Martin 1986 p. 422 n. 2 (page image), https://web.archive.org/web/2010id_/http://www.ldc.upenn.edu/myl/llog/LauraMartinEskimoSnowWords.pdf; Kaplan, https://www.uaf.edu/anlc/research-and-resources/resources/archives/inuit_snow_terms.php | Martin: "Thus, Eskimo has about as much differentiation as English does for 'snow' at the monolexemic level: snow and flake." Kaplan: "If we disqualify 'flake', that leaves 'snow' as the solitary English snow word (by Martin's criteria), and Eskimo languages with three times that many!"; F10.
- MINOR | Because Inuit and Yup'ik are polysynthetic, "how many words?" has no single answer: two roots, seventeen lexemes, or thousands of word-forms | Kaplan, https://www.uaf.edu/anlc/research-and-resources/resources/archives/inuit_snow_terms.php; Martin 1986 p. 422 n. 2 (page image), https://web.archive.org/web/2010id_/http://www.ldc.upenn.edu/myl/llog/LauraMartinEskimoSnowWords.pdf | Kaplan: "the three basic roots found in all Eskimo languages and dialects, except, unfortunately, West Greenlandic"; Martin: "no reason to posit more than two distinct roots ... in any Eskimo language"; F12 (and F4 for 'seventeen'). The dek drops the body's 'West Greenlandic'; Yup'ik has three.
- WRONG | "Anthony Woodbury’s Central Alaskan Yup’ik “citizen’s guide” counts about 17 snow/ice lexemes — the cards above are his, verbatim" / "a careful linguist’s tally lands near 17" | Woodbury 1991, https://web.archive.org/web/20190824230617id_/http://www.princeton.edu/~browning/snow.html; Pullum 1991 p. 171, https://pdf.stafforini.com/pullum-1991-great-eskimo-vocabulary.pdf; Kaplan, https://www.uaf.edu/anlc/research-and-resources/resources/archives/inuit_snow_terms.php | Woodbury: "Perhaps somewhat arbitrarily I have counted fifteen of them ... But the most arbitrary decision of all is left to the discretion of the reader-the decision of how to count the lexemes themselves." Pullum: "about a dozen words (even a couple of dozen if you are fairly liberal about what you count)"; F4. Woodbury gives 15 meanings and about 32 lexemes; the 17 are the page's selection, inside his one-to-two-dozen range.
- CONFIRMED | the 17 lexeme cards (qanuk 'snowflake' ... cellallir- 'to snow heavily') in Woodbury's four categories: Snow particles / Fallen snow / Snow formations / Meteorological events | Woodbury 1991 (Wayback copy), https://web.archive.org/web/20190824230617id_/http://www.princeton.edu/~browning/snow.html | "A. Snow particles ... B. Fallen snow ... C. Snow formations ... D. Meterological events"; all 17 forms are in his list; Every form is attested in Woodbury.
- MINOR | Lexeme glosses billed as verbatim: utvak 'snow carved in a block', navcaq 'snow cornice, about to collapse', natquik 'drifting snow' | Woodbury 1991 (Wayback copy), https://web.archive.org/web/20190824230617id_/http://www.princeton.edu/~browning/snow.html | "utvak 'snow carved in block'"; "navcaq [NSU] 'snow cornice, snow (formation) about to collapse'"; "natquik 'drifting snow/etc'"; F17.
- CONFIRMED | notes each noun lexeme has ~280 inflected forms and each verb over 1000 / (a lower bound) | Woodbury 1991, https://web.archive.org/web/20190824230617id_/http://www.princeton.edu/~browning/snow.html; Pullum 1991 p. 169, https://pdf.stafforini.com/pullum-1991-great-eskimo-vocabulary.pdf | Woodbury: "each single noun lexeme may have about 280 distinct inflected forms, while each verb lexeme may have over 1000!" Pullum: "For each noun stem there are about 280 different inflectional forms"; F21 refuted.
- CONFIRMED | Anthony C. Woodbury, “Counting Eskimo Words for Snow: A Citizen’s Guide” (1991). / Central Alaskan Yup’ik | Woodbury 1991 (Wayback copy), https://web.archive.org/web/20190824230617id_/http://www.princeton.edu/~browning/snow.html | "Counting Eskimo words for snow: A citizen's guide ... Anthony C. Woodbury, University of Texas at Austin, July 1991 ... Central Alaskan Yupik (or just Yup'ik Eskimo)"
- CONFIRMED | The 17 lexemes alone generate 6,200+ word-forms (15 nouns × 280 + 2 verbs × 1000); word-forms exceed roots by >1000× | research/eskimo-words-for-snow/verify.mjs; recreated: node verify.mjs (published copy https://artwaste.land/checks/research/eskimo-words-for-snow/verify.mjs, empty directory) | "ok  word-form lower bound computes to 6,200"; "ok  word-forms exceed true roots by >1000×"
- CONFIRMED | Laura Martin, “‘Eskimo Words for Snow’: A Case Study in the Genesis and Decay of an Anthropological Example,” American Anthropologist 88(2):418–423 (1986) | https://api.crossref.org/works/10.1525/aa.1986.88.2.02a00080 | title "“Eskimo Words for Snow”: A Case Study in the Genesis and Decay of an Anthropological Example", American Anthropologist, 88, 2, 418-423, 1986-06
- CONFIRMED | Geoffrey K. Pullum, The Great Eskimo Vocabulary Hoax and Other Irreverent Essays (Univ. of Chicago Press, 1991) | https://archive.org/metadata/greateskimovocab0000pull | "The great Eskimo vocabulary hoax, and other irreverent essays on the study of language", 1991, "Chicago : University of Chicago Press"; The subtitle 'on the Study of Language' is cut.
- CONFIRMED | Igor Krupnik et al., SIKU: Knowing Our Ice (2010), on the genuine richness of Yupik sea-ice vocabulary | https://api.crossref.org/works/10.1007/978-90-481-8587-0; Krupnik & Müller-Wille 2010 p. 391, https://gwern.net/doc/psychology/linguistics/2010-krupnik.pdf | "SIKU: Knowing Our Ice" / "Documenting Inuit Sea Ice Knowledge and Use", Springer Netherlands 2010, eds Krupnik, Aporta, Gearheard, Laidler, Kielsen Holm; "Most of the 20-some indigenous sea ice vocabularies collected for the SIKU project feature 60–80 (some of them up to 110–120) terms"
- CONFIRMED | Inuit hunters, living on it, carve snow and ice finer still / the smug correction ... overshoots | Krupnik & Müller-Wille 2010 p. 391, https://gwern.net/doc/psychology/linguistics/2010-krupnik.pdf; Kaplan, https://www.uaf.edu/anlc/research-and-resources/resources/archives/inuit_snow_terms.php | Krupnik: "local snow lexicons commonly feature several dozen terms for types of snow and snow-related phenomena"; Kaplan: "it still appears that Inupiaq at least has an extensive vocabulary for snow and ice"; "Pullum is not above overstating the case just a bit"
- CONFIRMED | "In Inuit and Yup’ik languages, what looks like a “word” is built up from a root by stacking suffixes — so a single root spawns hundreds of fully grammatical word-forms" | Woodbury 1991, https://web.archive.org/web/20190824230617id_/http://www.princeton.edu/~browning/snow.html; Kaplan, https://www.uaf.edu/anlc/research-and-resources/resources/archives/inuit_snow_terms.php | Woodbury: "about 280 distinct inflected forms"; Kaplan: "the number of words in these languages is practically infinite due to highly productive patterns of suffixation"
- CONFIRMED | English skiers and glaciologists have powder, hardpack, firn, névé, graupel, rime and sleet (and the English list snow ... crust) | Woodbury 1991 appendix, https://web.archive.org/web/20190824230617id_/http://www.princeton.edu/~browning/snow.html; NSIDC cryosphere glossary https://nsidc.org/learn/cryosphere-glossary/ (firn, neve, graupel, rime); Merriam-Webster, https://www.merriam-webster.com/dictionary/hardpack (and /firn, /neve, /graupel, /rime, /sleet) | Woodbury's English list: "blizzard, dusting, flurry, frost, hail, hardpack ... powder, sleet"; NSIDC firn: "called névé during the first year"; graupel: "snowflakes that become rounded pellets due to riming"
- CONFIRMED | Comparable to English. So far, no hoax. | Pullum 1991 p. 170, https://pdf.stafforini.com/pullum-1991-great-eskimo-vocabulary.pdf | "So the list is still short, not remarkably different in size from the list in English"; Krupnik 2010 p. 391 disputes Pullum here; the page follows Pullum and Woodbury.
- CONFIRMED | relates note: Gladstone thought Homer couldn't see blue; the lack was in the words, not the eye | Gladstone, Studies on Homer and the Homeric Age vol. III (1858), https://archive.org/download/studiesonhomerho03glad/studiesonhomerho03glad_djvu.txt | "Here Homer had before him the most perfect example of blue. Yet he never once so describes the sky." ... "I conclude, then, that the organ of colour and its impressions were but partially developed among the Greeks of the heroic age."; The second clause is the page's own verdict.
- MINOR | relates note: "a language with no single word for a thing — Villon's antan, 'last year's snow'" | Littré, https://www.littre.org/definition/antan; src/content/strata/the-snows-of-yesteryear.md:4 | "antan (an-tan) s. m. L'année qui précède celle qui court."; sibling: "Villon's antan, which in 1461 meant, exactly, last year"; F18. 'last year's snow' renders neiges d'antan, not antan.
- UNVERIFIABLE | nearby note: the 'Dickens coined boredom' legend | not kept: https://www.etymonline.com/word/boredom returned 403; OED paywalled; WebSearch gave Merriam-Webster 1853 and no earlier dated use (checker A) | (no primary text reached); This is the boredom-word layer's relates note; it belongs to that layer's pass.
- MINOR | Every figure and quote on this page is re-asserted from its source in research/eskimo-words-for-snow/verify.mjs / Every figure and quote re-asserted from its source. | research/eskimo-words-for-snow/verify.mjs; recreated: read in full; grep import/require/readFile/fetch/argv → 0; node verify.mjs | No quotation or source access; literals only, e.g. eq(flat.length, 17, "Woodbury's guide lists 17 snow/ice lexemes"); "PASS — 20 ok, 0 failed"; F14.
- MINOR | This layer's check works the answer out again from scratch, independently of this page. ... It can settle the finding | research/eskimo-words-for-snow/verify.mjs; scripts/check-placard.mjs | Inputs are the page's own claims typed as constants; the only computations are 15*280 + 2*1000 and ordering checks; F15. Shared placard wording.
- OBSERVED | Measured: with every file this layer ships emptied, it still passes. | recreated: curl https://artwaste.land/checks/research/eskimo-words-for-snow/verify.mjs into an empty directory; node verify.mjs | "PASS — 20 ok, 0 failed", exit 0; verify.mjs reads no file
- CONFIRMED | It is published, and it runs on its own: one file, and Node. | recreated: curl https://artwaste.land/checks/research/eskimo-words-for-snow/verify.mjs; node verify.mjs | HTTP 200, 7,470 bytes, text/javascript, byte-identical to the repo file; exit 0, 20 ok
- OBSERVED | Last verified 24 June 2026, when it was written. / (`verify.mjs`, 20/20) | git log -- research/eskimo-words-for-snow/verify.mjs (c98aea94db, 2026-06-24 18:39:56 +0000, adds it); memory/log.d/2026-06-24T1840Z-the-count-that-snowballed.md:45; recreated: node verify.mjs, published copy in an empty directory, 2026-10-11 | Log: "`node research/eskimo-words-for-snow/verify.mjs` → PASS 20/20"; today: "PASS — 20 ok, 0 failed"
- MINOR | (run it: node research/eskimo-words-for-snow/verify.mjs) | curl -sS -o /dev/null -w '%{http_code}' https://artwaste.land/research/eskimo-words-for-snow/verify.mjs | 404 (the working copy is at /checks/research/eskimo-words-for-snow/verify.mjs, linked from the placard); F16.
- CONFIRMED | "datePublished":"2026-06-24","dateModified":"2026-06-24" | git log -- public/strata/eskimo-words-for-snow/index.html src/content/strata/eskimo-words-for-snow.md; git diff c98aea94db HEAD (checker A) | First commit 2026-06-24; later commits change only dashes, the injected placard, nav and footer, and og:image:alt; no claim text changed

## Findings
- [fixed] F1 (WRONG): The dek, all four meta descriptions and the stepper said the press doubled the figure for decades, each writer copying the last and rounding up, a chain neither Martin nor Pullum describes (they record numbers made up in no order, including nine in 1984 and about four dozen in 1988); the descriptions now say later writers "made up numbers as high as four hundred, with no source", the stepper is labelled "step through documented figures (a selection)" with the legend "no source given", the md dek lists the documented figures by teller and year, and verify.mjs's non-decreasing assertions are replaced by checks that the selection is in date order and that the full record is not a rising chain.
- [fixed] F2 (WRONG): The NYT rung quoted "…the Eskimos have a hundred words for snow.", a sentence not in the 9 Feb 1984 editorial; the rung now quotes the editorial's own sentence ("Benjamin Lee Whorf, the linguist, once reported on a tribe that distinguishes 100 types of snow - and has 100 synonyms (like tipsiq and tuva) to match."), and the body says the Times printed "100 types of snow" and credited the figure to Whorf, who never gave it.
- [fixed] F3 (WRONG): The page tagged Whorf's 5 "grounded" and said he gave no Eskimo word-count, though Martin notes he cited no data or source and his article's Fig. 2 reads "ESKIMO - THREE WORDS"; the Whorf rung is now tagged "in print · no source" and shown as 5+, the body says "Only Boas’s 4 rests on data", and the check box says he cited no data and no source and that the article's drawing is labelled "ESKIMO - THREE WORDS".
- [fixed] F4 (WRONG): The page said Woodbury's guide "counts about 17 snow/ice lexemes" and that the cards were "his, verbatim", but Woodbury sorts over thirty lexemes into fifteen meanings and leaves the count to the reader; the check box now says so, quotes the "about a dozen words (even a couple of dozen ...)" Pullum reports him endorsing, and calls the 17 cards "a selection from his list: one per meaning, plus two synonyms for snow on the ground", and the lexeme mode and verify.mjs label them the same way.
- [fixed] F5 (WRONG): The 1978 rung gave the fifty to "Popular press & textbooks" as "The first big jump" documented as citation drift; it now gives it to Lanford Wilson's play The Fifth of July, "One item in what Laura Martin (1986) called “a diverse random sample of recent references”".
- [fixed] F6 (WRONG): The 200 rung was dated "1980s–90s" and given to the popular press as citing the previous teller; it is now dated 1984 and given to "Local TV weather forecasts (e.g. WEWS-Cleveland)", "The same year as the Times’s hundred and a trivia book’s nine (Martin 1986)".
- [fixed] F7 (MINOR): The 400 rung was dated "by 2000s" and given to the popular press, though Pullum reports it in writing by 1991 from a would-be author's piece; it is now dated "by 1991" and given to "A would-be author’s piece (reported by Pullum 1991)".
- [fixed] F8 (WRONG): The 50, 200 and 400 rungs set sentences in quotation marks that no source gives; they now show only what the sources give: "fifty" as Martin and Pullum report it, "two hundred words" as Martin quotes the forecasts, and the figure four hundred with the note that Pullum gives no wording.
- [fixed] F9 (MINOR): The check box said "no number above five is traceable to any linguistic source" and the body that the figure "had no basis in any grammar, and never did", which SIKU's dictionary-based lists and the page's own Woodbury list contradict; the bullet now says "none of these popular figures came from a linguistic source" and the body "no grammar lay behind it".
- [fixed] F10 (WRONG): The roots mode said English (snow, sleet, slush, hail) has more snow roots than West Greenlandic's two, reversing Martin and Kaplan; it now counts roots for snow itself, "two in West Greenlandic, three in every other Eskimo language", and says "By the same measure English has one or two: snow, and arguably flake."
- [fixed] F11 (MINOR): The Boas quotation silently dropped "one" three times; the rung now reads "another one, qana ... a third one, piqsirpoq ... a fourth one, qimuqsuq" as in the Handbook.
- [fixed] F12 (MINOR): The dek said "two roots" for Inuit and Yup'ik generally, though Kaplan gives three outside West Greenlandic; the md dek now says "two or three roots, one to two dozen lexemes", and the check box says the answer spans "2 or 3, a dozen or two, or thousands".
- [fixed] F13 (MINOR): The page called Boas's example "a grammar footnote", though it is running text in his introduction; it now reads "one example in a grammar handbook’s introduction" and "Then the example left home."
- [fixed] F14 (MINOR): The page and dek said every figure and quote is "re-asserted from its source" in verify.mjs, which reads no source and checks no quotation; the check note now says verify.mjs "restates the page’s figures and checks their arithmetic and the shape of the timeline; it does not read the sources or check the quotations", and the md dek now ends "The arithmetic re-checked; every source named."
- [open note:8d17ac] F15 (MINOR): The placard says the check "works the answer out again from scratch" and "can settle the finding", but for this layer its inputs are the page's claims typed as constants, so it cannot settle them and stays green with several of them false. Held because the wording is generated by scripts/check-placard.mjs, a shared template, and a class fix for verifiers that restate typed constants belongs in the placard generator.
- [fixed] F16 (MINOR): The in-text instruction "node research/eskimo-words-for-snow/verify.mjs" named a repository path that returns 404 on the site; the check note now links /checks/research/eskimo-words-for-snow/verify.mjs and says to run "node verify.mjs".
- [fixed] F17 (MINOR): Three Woodbury glosses billed as verbatim were reworded; they now read "drifting snow/etc.", "snow carved in block" and "snow cornice, snow (formation) about to collapse", and the page no longer calls the cards verbatim.
- [fixed] F18 (MINOR): The relates note glossed antan as "last year's snow", though antan means "last year" and the site's own Villon layer says so; the note in src/content/strata/eskimo-words-for-snow.md now reads "Villon's antan, 'last year', in 'the snows of last year'".
- [declined] F19: Checker A said the essay and its title first appeared in 1989 (NLLT) and the 1991 book reprinted it. Declined, refuted by the adjudicator: the sentence is about when Pullum made the story famous, the page cites the 1991 book, and Kaplan dates it the same way.
- [declined] F20: Checker B said the quotation marks around "lumps" suggest Whorf's word, and "lump" does not occur in his article. Declined, refuted: the scare-quoted verb restates Whorf's "the same word" accurately, his words are quoted verbatim separately, and the point is one of style.
- [declined] F21: Checker B said Woodbury hedges ("may have about 280") while the page states it as fact and calls a product built on "about 280" a lower bound. Declined, refuted: "~" carries Woodbury's "about", and 6,200 counts inflection only, so it is a fair floor.