---
slug: the-last-colour
passed: 2026-10-14
by: the assay line (research/assay-line/cited-1014b/), Codex fixer directed by claude-assaying-boyle-1014; source record by Codex, checked by Codex; copy findings and dated corrections re-read by the Codex repair worker, 2026-10-14
checked: 125
confirmed: 99
wrong: 25
unverifiable: 0
observed: 0
stale: 1
---

This pass re-read the evidence for all 26 directing-triage items and repaired this layer's authored HTML, markdown descriptions and colour verifier at the f5e5f84f34 baseline, with a verbatim refresh of its published verifier. Seven WRONG findings received dated correction lines, eighteen MINOR findings were repaired in this layer's authored text without page notes, and the stale JSON-LD modification date was moved under the stated exception. The 99 CONFIRMED claims and their source readings below are retained from the checkers' N2-codex.md record, rather than presented as 99 new independent readings by this fixer; they describe the checked baseline and preserve its interpretive qualifications. All eighteen MINOR lines count toward wrong. The five UNVERIFIABLE claims in that earlier record were not included in the directing triage and are not promoted to CONFIRMED here. The repair rechecked all seven dated corrections and the three X1 copy findings below. Those copies are dispositions of the existing X1 claim, not additional counted claims; the two authored copies in other layers remain open pending board note IDs.

## Claims

- MINOR | X1 [opening-last] "same name comes up last, almost everywhere: blue." | https://wals.info/chapter/134 | WALS section 3 puts separate primary terms at stage V, allows later derived categories and reports exceptions. This layer's opening and authored descriptions now state a tendency among six primaries. The old-dek share image is assigned to the landing rebuild; two authored copies in other layers remain open below.
- MINOR | X2 [fixed-ladder] "always has the same few" | https://linguistics.berkeley.edu/kay/univ.col.cats.complt.pdf | Kay, Universality of Color Categorization, p. 3 permits either green or yellow in a four-term system. The absolute wording became common sequences.
- MINOR | X3 [implication-world] "never the reverse" | https://wals.info/chapter/134 | WALS section 4.3 includes a separate blue term with a combined yellow/green term. The implication is now scoped to the classical sequence and called a tendency.
- MINOR | X4 [meta-last] "a word for blue is the last of the basic colours to appear" | https://wals.info/chapter/134 | WALS section 3 distinguishes the six primaries from later derived categories. Search metadata now makes that scope and tendency explicit.
- WRONG | X5 [meta-greeks] "they just hadn't named it." | https://aeon.co/essays/can-we-hope-to-understand-how-the-greeks-saw-their-world | Sassi opens with kuaneos and glaukos as descriptions of blue shades. The description and related blue-absence wording now distinguish Greek meanings from modern English blue.
- MINOR | X6 [og-last] "Blue is the last basic colour word languages get" | https://wals.info/chapter/134 | WALS section 3 supports a qualified primary-category sequence, not a final basic word of every kind. Open Graph wording now states that scope.
- MINOR | X7 [twitter-last] "blue is the last colour languages name" | https://wals.info/chapter/134 | WALS section 3 supports a qualified primary-category sequence. Twitter wording now names the six primaries and the tendency.
- WRONG | X8 [winawer-name] "Jay Winawer" | https://bpb-us-e1.wpmucdn.com/wp.nyu.edu/dist/4/2136/files/2017/01/winawer_witthoft_et_al_PNAS_2007.pdf | The actual byline on printed p. 7780 gives Jonathan Winawer. The page now uses that name.
- MINOR | X9 [median-results] "the Russian goluboy / siniy boundary" | https://bpb-us-e1.wpmucdn.com/wp.nyu.edu/dist/4/2136/files/2017/01/winawer_witthoft_et_al_PNAS_2007.pdf; public/strata/the-last-colour/index.html | Results, Boundaries and Analysis, p. 7781 measures each participant's classification boundary. The page fixes BOUNDARY=4, so its result label and code comment now call this illustrative.
- MINOR | X10 [swatch-equal] "no further apart in colour" | public/strata/the-last-colour/index.html; research/the-last-colour/verify.mjs | Re-evaluating the page functions gives ideal Lab example distances 9.8729537654 within and 10.0653854013 across. The page and verifier now say approximately matched model distances.
- MINOR | X11 [confound-removed] "that confound is removed" | public/strata/the-last-colour/index.html; research/the-last-colour/verify.mjs | The cross example remains slightly farther apart, both in ideal Lab and after integer RGB conversion. Neither computation rules out response-time confounds; that assurance is removed from prose and verifier.
- MINOR | X12 [dynamic-english] "the boundary that helps Russian speakers isn't in your head." | https://bpb-us-e1.wpmucdn.com/wp.nyu.edu/dist/4/2136/files/2017/01/winawer_witthoft_et_al_PNAS_2007.pdf | Results, Boundaries, p. 7781 gives nearly identical English and Russian classification boundaries; the reported group difference concerns category advantage. The dynamic explanation now says this and does not infer a reader's categories.
- WRONG | X13 [dynamic-effect] "measured over 50 people" | https://bpb-us-e1.wpmucdn.com/wp.nyu.edu/dist/4/2136/files/2017/01/winawer_witthoft_et_al_PNAS_2007.pdf | Methods, Participants, p. 7784 reports 26 Russian and 24 English recruits; Analysis, p. 7782 excludes five Russian and three English participants. The 1164 versus 1288 ms comparison has t(20): 21 Russian participants, 42 analyzed overall, 50 recruited.
- WRONG | X14 [companion-dozen] "how a dozen translators rendered it" | git show f5e5f84f34:src/content/strata/wine-dark-sea.md | The companion frontmatter and six-hands description name six versions: Chapman, Pope, Cowper, Buckley, Butcher and Lang jointly, and Butler. This layer now says six translation versions, including its related-page note.
- MINOR | X15 [geiger-scope] "Icelandic sagas" | https://archive.org/download/contributionstoh00geigrich/contributionstoh00geigrich_djvu.txt | Re-read Geiger's English translation, On Colour-Sense, p. 55 footnote: the corpus named is Edda hymns. Changed Icelandic sagas to Eddic poems.
- MINOR | X16 [geiger-chinese] "the Icelandic sagas and early Chinese" | https://archive.org/download/contributionstoh00geigrich/contributionstoh00geigrich_djvu.txt | Geiger also discusses Chinese hiuan on p. 52. This duplicate finding receives the same Eddic-poems repair; the supported Chinese example remains.
- WRONG | X17 [kyaneos-objects] "a dark storm-cloud" | https://www.gutenberg.org/files/53004/53004-h/53004-h.htm | Gladstone, vol. III, pp. 463-464, item h lists a dark cloud of vapour and explicitly excludes a storm-cloud. The palette example now follows that distinction.
- WRONG | X18 [violet-wool] "(as ióeis) to wool / sheep and to iron." | https://www.gutenberg.org/files/53004/53004-h/53004-h.htm | Gladstone, vol. III, pp. 470-471 assigns ioeides to the sea, ioeis to iron at Iliad 23.850, and iodnephes to Helen's wool at Odyssey 4.135 and the sheep at 9.426. The palette now distinguishes all three.
- MINOR | X19 [himba-peer-reviewed] "speed and accuracy of discrimination" | https://eprints.goldsmiths.ac.uk/5673/1/PSY_davidoff-robertson-color-categories_2005.pdf | Re-read the supplied Roberson et al. PDF, abstract p. 2: similarity judgments, short-term memory and long-term learning. The prose names these tasks and limits the Russian comparison to category-related performance.
- MINOR | X20 [ramp-half-percent] "every adjacent step has the same ΔE₀₀ to within 0.5%." | public/strata/the-last-colour/index.html; research/the-last-colour/verify.mjs | Re-evaluating the page functions gives maximum adjacent-step deviation 0.4507187204% in ideal Lab and 3.8732660878% after integer RGB conversion. The 0.5% bound is retained with ideal Lab scope; displayed spacing is approximate.
- MINOR | X21 [dek-showings] "a boundary-crossing pair is no further apart in colour than a within-category one" | public/strata/the-last-colour/index.html; research/the-last-colour/verify.mjs | Same computed example pairs as X10: cross 10.0653854013 exceeds within 9.8729537654 in ideal Lab. The dek now says approximately matched model distances.
- WRONG | X22 [relates-grue] "what over half the world keeps fused with green as 'grue.'" | https://wals.info/chapter/134 | WALS sections 1-2 identify 110 surveyed languages and restrict the grue majority to that sample. The related-page note and sampled-language shorthand now name the World Color Survey.
- MINOR | X23 [relates-full-math] "prove its blue swatches are perceptually equidistant." | public/strata/the-last-colour/index.html; research/the-last-colour/verify.mjs | Reference-pair agreement validates model calculations, not exact perceptual equality on a display. The related-page note now describes approximately equal model distances in ideal Lab.
- MINOR | X24 [jsonld-primary-order] "a basic word for blue arrives last" | https://wals.info/chapter/134 | WALS section 3 distinguishes primary and derived categories with exceptions. JSON-LD now qualifies both the six-primary scope and tendency.
- STALE | X25 [date-modified] ""dateModified":"2026-06-30"" | git log f5e5f84f34 -- public/strata/the-last-colour/index.html; public/strata/the-last-colour/index.html | The 2026-06-30 creation value predates later edits, including 2e60bcb5dd1 on 2026-10-01. dateModified is now 2026-10-14 for this revision; datePublished and the June observation remain.
- MINOR | X26 [added-dial-order] "in the order the survey found" | https://files.eric.ed.gov/fulltext/ED019655.pdf | Berlin and Kay's original report, stage VII, PDF page 27, leaves the remaining categories unordered. The dial instruction and dek now label the within-stage choices illustrative.

- CONFIRMED | "Not because anyone was blind to the sky" [...] "they saw it as well as you do" [...] "but because a colour and a word for it are two different things" | https://aeon.co/essays/can-we-hope-to-understand-how-the-greeks-saw-their-world | [opening-vision] Source excerpt read against the whole claim: "The Greeks were perfectly able to perceive the blue tint, but were not particularly interested in describing the blue tone of sky or sea – at least not in the same way as we are, with our modern sensibility." Location: Discussion of colour saliency
- CONFIRMED | "Blue is not literally the last colour any language ever names" [...] "brown, purple, pink and grey usually come after it." | https://wals.info/chapter/134 | [not-literally-last] Source excerpt read against the whole claim: "Usually, derived categories appear only after stage V, where each primary has received a separate term, but there are many exceptions." Location: Section 3
- CONFIRMED | "Blue is the last of the six primary colour categories to split off and get its own basic word: white and black come first, then red, then green and yellow, and only then blue." | https://files.eric.ed.gov/fulltext/ED019655.pdf; https://wals.info/chapter/134 | [primary-order] Read as the classical primary-category sequence and the qualified statistical tendency stated on the page, not a universal exceptionless law. Berlin and Kay: "At Stage V the focus of blue emerges". WALS records alternatives and the "83%" main trajectory.
- CONFIRMED | "in the largest cross-linguistic survey, more than half the languages use a single word for green-and-blue together." | https://wals.info/chapter/134; https://lclab.berkeley.edu/papers/info-evoln-2015.pdf | [survey-majority] WALS: "Over half the languages in the World Color Survey contain a grue term." The original excerpt alone does not establish largest. Xu et al. call WCS "the largest existing publicly available database of color naming data"; the page concerns cross-language colour naming, not every possible cross-linguistic survey.
- CONFIRMED | "In 1969 Brent Berlin and Paul Kay surveyed colour vocabularies across languages" | https://linguistics.berkeley.edu/kay/univ.col.cats.complt.pdf; https://files.eric.ed.gov/fulltext/ED019655.pdf | [berlin-1969] The named authors and 1969 publication are correct. The supplied URL is Paul Kay's later chapter "Universality of Color Categorization", not the 1969 book; "Basic Color Terms (1969)" is a section heading. Fieldwork predates publication, as the 1967 report proves. Read the page's date as the publication date.
- CONFIRMED | "Berlin & Kay's stages, as standardly summarised: I white/black (better read as light-warm vs. dark-cool) · II +red · III–IV +green and yellow (in either order) · V +blue · VI +brown · VII +purple, pink, orange, grey." | https://linguistics.berkeley.edu/kay/univ.col.cats.complt.pdf; https://files.eric.ed.gov/fulltext/ED019655.pdf | [stages] The 1967 report supports blue at V and the later derived colours, but its green-before-yellow sequence does not support the page's "in either order". Kay's later review does: "if four either black, white, red, and green or black, white, red, and yellow". Thus the page checks out but this record needs the later supporting source.
- CONFIRMED | "The World Color Survey (110 unwritten languages) broadly confirmed the lateness of blue while softening the strict staircase into a set of trajectories." | https://linguistics.berkeley.edu/kay/univ.col.cats.complt.pdf; https://wals.info/chapter/134 | [wcs-size] Kay's review explicitly says "Color naming data were collected from 110 unwritten languages". WALS supplies 110 and the main trajectory, but its excerpt alone does not establish unwritten.
- CONFIRMED | "It is a strong statistical tendency, not a law" [...] "and it has serious critics (see the apparatus)." | https://wals.info/chapter/134; https://home.uchicago.edu/~johnlucy/papersmaterials/1997%20Linguistics%20of%20Color%20v1.pdf | [tendency-critics] WALS: "83%" on the main trajectory, rather than all. Lucy's scanned p. 343 was rendered and OCRed individually; it criticizes the category projection and cites Saunders. The claim of a tendency and methodological dispute is supported.
- CONFIRMED | "Over half of surveyed languages still use one word for green-and-blue." | https://wals.info/chapter/134 | [meta-majority] Source excerpt read against the whole claim: "Over half the languages in the World Color Survey contain a grue term." Location: Section 2, Basic colour term concepts
- CONFIRMED | "blue is the last of the six primary colour categories to get its own basic word (after white/black, red, green and yellow)" | https://files.eric.ed.gov/fulltext/ED019655.pdf; https://wals.info/chapter/134 | [dek-six] Read as the classical primary-category sequence and the qualified statistical tendency stated on the page, not a universal exceptionless law. Berlin and Kay: "At Stage V the focus of blue emerges". WALS records alternatives and the "83%" main trajectory.
- CONFIRMED | "in more than half of the world's surveyed languages it still hasn't split off at all, sharing one word with green ('grue')" | https://wals.info/chapter/134 | [dek-half] Source excerpt read against the whole claim: "Over half the languages in the World Color Survey contain a grue term." Location: Section 2, Basic colour term concepts
- CONFIRMED | "Russian has no single word for "blue."" | https://bpb-us-e1.wpmucdn.com/wp.nyu.edu/dist/4/2136/files/2017/01/winawer_witthoft_et_al_PNAS_2007.pdf | [russian-no-single] Source excerpt read against the whole claim: "There is no single generic word for ‘‘blue’’ in Russian that can be used to describe all of the colors in Fig. 1 (nor to adequately translate the title of this work from English to Russian)." Location: Introduction, p. 7780
- CONFIRMED | "It has two basic words" [...] "голубой (goluboy, the lighter blues) and синий (siniy, the darker blues)" [...] "and the boundary between them sits right in the middle of what English calls one colour." | https://bpb-us-e1.wpmucdn.com/wp.nyu.edu/dist/4/2136/files/2017/01/winawer_witthoft_et_al_PNAS_2007.pdf | [russian-terms] Source excerpt read against the whole claim: "Unlike English, Russian makes an obligatory distinction between lighter blues (‘‘goluboy’’) and darker blues (‘‘siniy’’)." Location: Abstract, p. 7780
- CONFIRMED | "This is their task. A square appears on top; two squares appear below; one of them is exactly the same colour as the top one." | https://bpb-us-e1.wpmucdn.com/wp.nyu.edu/dist/4/2136/files/2017/01/winawer_witthoft_et_al_PNAS_2007.pdf | [original-task] Source excerpt read against the whole claim: "In each color discrimination trial, sub- jects were shown a triad of color squares. One of the colors presented on the bottom was physically identical to the top color square (Fig. 1). The task was to indicate which of the bottom squares matched the top square by pressing a key on the right or left side of the keyboard." Location: Methods, Color Discrimination Task, p. 7784
- CONFIRMED | "If you're an English speaker the two bars above will likely be about even, because you have one word for this whole range. That flatness is the point, not a failure." | https://bpb-us-e1.wpmucdn.com/wp.nyu.edu/dist/4/2136/files/2017/01/winawer_witthoft_et_al_PNAS_2007.pdf | [english-prediction] The paper says "English speakers did not show a category advantage". Read as a prediction of no systematic advantage, not a guarantee that a noisy 16-trial sample produces equal bars; the surrounding page expressly disclaims an experiment.
- CONFIRMED | "Winawer et al. (2007, PNAS) ran this with 50 speakers." | https://bpb-us-e1.wpmucdn.com/wp.nyu.edu/dist/4/2136/files/2017/01/winawer_witthoft_et_al_PNAS_2007.pdf | [fifty-speakers] The paper reports "Twenty-six native Russian speakers" and "24 native English speakers" recruited: 26 + 24 = 50. This is correct as a recruitment/run count. Five Russian and three English participants were excluded from analysis, addressed separately in dynamic-effect.
- CONFIRMED | "Russian speakers were reliably ~124 ms faster when the two blues straddled the goluboy/siniy boundary than when both fell inside one name (1,164 vs. 1,288 ms on the hard trials)." | https://bpb-us-e1.wpmucdn.com/wp.nyu.edu/dist/4/2136/files/2017/01/winawer_witthoft_et_al_PNAS_2007.pdf | [124ms] The near-colour, no-interference comparison is "1,164 vs. 1,288 msec". Recomputed 1288 - 1164 = 124. This is a within-Russian category advantage, not Russian-versus-English speed.
- CONFIRMED | "English speakers showed no such advantage." | https://bpb-us-e1.wpmucdn.com/wp.nyu.edu/dist/4/2136/files/2017/01/winawer_witthoft_et_al_PNAS_2007.pdf | [english-no-advantage] Source excerpt read against the whole claim: "English speakers did not show a category advantage" Location: Abstract, p. 7780
- CONFIRMED | "the Russian advantage vanished when speakers held an 8-digit number in mind (verbal interference) but survived a spatial-memory load." | https://bpb-us-e1.wpmucdn.com/wp.nyu.edu/dist/4/2136/files/2017/01/winawer_witthoft_et_al_PNAS_2007.pdf | [interference-result] Read Methods and Results, including "eight-digit number series" and "The category advantage found for Russian speakers was disrupted by verbal, but not spatial, interference." The claim preserves both interventions.
- CONFIRMED | "when speakers held an 8-digit number in mind (verbal interference)" | https://bpb-us-e1.wpmucdn.com/wp.nyu.edu/dist/4/2136/files/2017/01/winawer_witthoft_et_al_PNAS_2007.pdf | [eight-digits] Read Methods and Results, including "eight-digit number series" and "The category advantage found for Russian speakers was disrupted by verbal, but not spatial, interference." The claim preserves both interventions.
- CONFIRMED | "The word is recruited live, online" [...] "it isn't rewiring the eye." | https://bpb-us-e1.wpmucdn.com/wp.nyu.edu/dist/4/2136/files/2017/01/winawer_witthoft_et_al_PNAS_2007.pdf | [online-language] The record excerpt "The exact nature of this interaction cannot be determined from these data" omits the directly relevant preceding argument. The paper says "Our present results do not offer support for this possibility" of permanent changes and concludes that language representations "play an online role". The page's contrast with a permanently changed eye is supported; the precise neural mechanism is not claimed.
- CONFIRMED | "exactly as the original study removed it." | https://bpb-us-e1.wpmucdn.com/wp.nyu.edu/dist/4/2136/files/2017/01/winawer_witthoft_et_al_PNAS_2007.pdf | [original-control] The article used identical physical stimuli for both language groups, grouped trials around each subject's own boundary, and found the effect only in the Russian group. This supports controlling stimulus differences for the between-language conclusion. Its method was not this page's nine-chip CIEDE2000 construction; the source record's isolated two-step excerpt misses that reasoning.
- CONFIRMED | "in the World Color Survey, more than half use a single word for green-and-blue together ("grue"), and never give blue its own basic term." | https://wals.info/chapter/134 | [dynamic-grue] WALS confirms "Over half the languages in the World Color Survey contain a grue term." Read never in the present descriptive sense of no separate basic term in those systems, not a prediction about their future evolution. WALS is a contemporary sample, not a census of all language histories.
- CONFIRMED | "stage V, the last of the six primary colour categories. Everything before it (white/black, red, green, yellow) had to come first; that order almost never reverses." | https://files.eric.ed.gov/fulltext/ED019655.pdf; https://wals.info/chapter/134 | [dynamic-blue] Read as the classical primary-category sequence and the qualified statistical tendency stated on the page, not a universal exceptionless law. Berlin and Kay: "At Stage V the focus of blue emerges". WALS records alternatives and the "83%" main trajectory.
- CONFIRMED | "Past blue, languages add brown and then purple, pink, orange and grey" [...] "which is why "blue is literally last" is wrong. Blue is the last of the primaries, not the last word of all." | https://wals.info/chapter/134 | [dynamic-after] Source excerpt read against the whole claim: "Usually, derived categories appear only after stage V, where each primary has received a separate term, but there are many exceptions." Location: Section 3
- CONFIRMED | "The oldest text in the European tradition" | https://discovery.ucl.ac.uk/1462583/2/Lambrou%2520Thesis.pdf | [oldest-european] Lambrou's thesis says the Iliad and Odyssey "are the earliest extant European literary compositions". The page's European tradition context concerns surviving literature, not the oldest archaeological inscription.
- CONFIRMED | "The oldest text in the European tradition has almost no blue in it." | https://aeon.co/essays/can-we-hope-to-understand-how-the-greeks-saw-their-world | [homer-blue] Sassi identifies kuaneos and glaukos but says the blue area lacks an exact modern English equivalent. This supports a category mismatch, not absence of words for any blue shades. The separately recorded metadata assertion is WRONG.
- CONFIRMED | "In 1858 William Gladstone" | https://aeon.co/essays/can-we-hope-to-understand-how-the-greeks-saw-their-world | [gladstone-1858] Source excerpt read against the whole claim: "Another explanation for the apparent oddness of Greek perception came from the eminent politician and Hellenist William Gladstone, who devoted a chapter of his Studies on Homer and the Homeric Age (1858) to ‘perceptions and use of colour’." Location: Gladstone discussion
- CONFIRMED | "four-times British Prime Minister" | https://www.gov.uk/government/history/past-prime-ministers/william-ewart-gladstone | [gladstone-pm] Source excerpt read against the whole claim: "He served as Prime Minister for 4 separate periods." Location: Interesting facts
- CONFIRMED | "went through the Iliad and the Odyssey counting colour words and found the palette strange beyond the missing blue: it was organised around light and dark far more than hue, and the few colour words land on objects in ways that read as nonsense to us." | https://www.gutenberg.org/files/53004/53004-h/53004-h.htm | [gladstone-counting] Read Gladstone pp. 476–488. He explicitly describes the "vast predominance" of white and black and concludes "Homeric colours are really the modes and forms of light". The page attributes this brightness-centred reading to Gladstone; it is not a new numerical census.
- CONFIRMED | "Gladstone's deepest observation was structural: Homer's colour sense runs mostly on a brightness axis, not a hue wheel." | https://www.gutenberg.org/files/53004/53004-h/53004-h.htm | [brightness-structural] Read Gladstone pp. 476–488. He explicitly describes the "vast predominance" of white and black and concludes "Homeric colours are really the modes and forms of light". The page attributes this brightness-centred reading to Gladstone; it is not a new numerical census.
- CONFIRMED | "Black (mélas) and white (leukós) dominate by far; a true hue word like red is comparatively rare; and a dedicated word for blue is effectively absent." | https://www.gutenberg.org/files/53004/53004-h/53004-h.htm | [black-white-dominance] Read Gladstone pp. 476–488. He explicitly describes the "vast predominance" of white and black and concludes "Homeric colours are really the modes and forms of light". The page attributes this brightness-centred reading to Gladstone; it is not a new numerical census.
- CONFIRMED | "The point is the axis: brightness, not hue." | https://www.gutenberg.org/files/53004/53004-h/53004-h.htm | [homer-axis] Read Gladstone pp. 476–488. He explicitly describes the "vast predominance" of white and black and concludes "Homeric colours are really the modes and forms of light". The page attributes this brightness-centred reading to Gladstone; it is not a new numerical census.
- CONFIRMED | ""The ancient Greeks were colour-blind to blue" [...] "they physically couldn't see it." This is false" | https://aeon.co/essays/can-we-hope-to-understand-how-the-greeks-saw-their-world | [myth-denial] Source excerpt read against the whole claim: "The Greeks were perfectly able to perceive the blue tint, but were not particularly interested in describing the blue tone of sky or sea – at least not in the same way as we are, with our modern sensibility." Location: Discussion of colour saliency
- CONFIRMED | "Human trichromatic colour vision is far older than any colour vocabulary." | https://www.ebi.ac.uk/europepmc/webservices/rest/PMC4270479/fullTextXML; https://aeon.co/essays/can-we-hope-to-understand-how-the-greeks-saw-their-world | [old-trichromacy] The PMC URL gave a challenge to curl, so the same paper was read in full-text XML at Europe PMC. It says "Trichromatic color vision in the human lineage was fully developed by 30 My ago". Together with Sassi this supports ordinary ancient Greek trichromacy, not an assertion that every individual has identical colour vision.
- CONFIRMED | "Homer's Greeks had the same three cone types and the same retinas you do; they saw the sea's blue perfectly well." | https://www.ebi.ac.uk/europepmc/webservices/rest/PMC4270479/fullTextXML; https://aeon.co/essays/can-we-hope-to-understand-how-the-greeks-saw-their-world | [greek-retinas] The PMC URL gave a challenge to curl, so the same paper was read in full-text XML at Europe PMC. It says "Trichromatic color vision in the human lineage was fully developed by 30 My ago". Together with Sassi this supports ordinary ancient Greek trichromacy, not an assertion that every individual has identical colour vision.
- CONFIRMED | "What they lacked was a word that carved blue out as its own basic category. The gap is lexical and cultural, not a defect of the eye." | https://aeon.co/essays/can-we-hope-to-understand-how-the-greeks-saw-their-world | [lexical-gap] Sassi identifies kuaneos and glaukos but says the blue area lacks an exact modern English equivalent. This supports a category mismatch, not absence of words for any blue shades. The separately recorded metadata assertion is WRONG.
- CONFIRMED | "also guessed that vision itself had evolved within historical time." | https://archive.org/download/contributionstoh00geigrich/contributionstoh00geigrich_djvu.txt | [geiger-guess] Read the 1867 lecture, including its explicit claim that "the sensibility to yellow was awakened before that to green". This is a historical claim about what Geiger proposed, not an endorsement of that proposal.
- CONFIRMED | "That guess is rejected by modern science." | https://aeon.co/essays/can-we-hope-to-understand-how-the-greeks-saw-their-world | [geiger-rejected] Source excerpt read against the whole claim: "Today, no one thinks that there has been a stage in the history of humanity when some colours were ‘not yet’ being perceived." Location: Response to Gladstone
- CONFIRMED | "Homer's word for the sea" [...] "the famous "wine-dark sea" (οἶνοψ πόντος)." | https://www.gutenberg.org/files/53004/53004-h/53004-h.htm | [oinops-use] Gladstone p. 472 lists the sea and oxen with Iliad and Odyssey loci. The supplied excerpt is present after normalizing whitespace around HTML italics and footnotes.
- CONFIRMED | "Literally "wine-looking," from oînos (wine) + óps (eye/face)." | https://en.wiktionary.org/wiki/%CE%BF%E1%BC%B6%CE%BD%CE%BF%CF%88 | [oinops-etymology] Source excerpt read against the whole claim: "From οἶνος ( oînos , “ wine ” ) +‎ -οψ ( -ops ) (connected to ὄψ ( óps , “ eye; face ” ) )." Location: Ancient Greek, Etymology
- CONFIRMED | "He uses the very same word for oxen" [...] "wine-dark cattle." | https://www.gutenberg.org/files/53004/53004-h/53004-h.htm | [oinops-oxen] Gladstone p. 472 lists the sea and oxen with Iliad and Odyssey loci. The supplied excerpt is present after normalizing whitespace around HTML italics and footnotes.
- CONFIRMED | "A word that fits both the sea and an ox is plainly not tracking hue." | https://www.gutenberg.org/files/53004/53004-h/53004-h.htm; https://aeon.co/essays/can-we-hope-to-understand-how-the-greeks-saw-their-world | [oinops-not-hue] The excerpt only gives two referents, but the next paragraphs explicitly conclude "vaguely indicates a dark hue, but cannot be referred to any one of the known principal colours". Sassi independently reads the epithet as shine and movement. Treat the page as this attributed interpretation, not a proof that sea and cattle can never share any hue.
- CONFIRMED | "spanning green, yellow-green and pale." | https://www.gutenberg.org/files/53004/53004-h/53004-h.htm; https://aeon.co/essays/can-we-hope-to-understand-how-the-greeks-saw-their-world | [chloro-range] Gladstone pp. 467–468 names fear/paleness, freshly pulled twigs, and honey. Sassi adds "a vivid yellow, like honey". Together these support the page's range and actual referents.
- CONFIRMED | "Homer applies it to fresh twigs, to honey ("green honey"), and to a face blanched with fear" [...] "a brightness-and-freshness word more than a fixed green." | https://www.gutenberg.org/files/53004/53004-h/53004-h.htm; https://aeon.co/essays/can-we-hope-to-understand-how-the-greeks-saw-their-world | [chloro-objects] Gladstone pp. 467–468 names fear/paleness, freshly pulled twigs, and honey. Sassi adds "a vivid yellow, like honey". Together these support the page's range and actual referents.
- CONFIRMED | "In later Greek this becomes "cyan," blue" | https://www.egyptianblue.info/4-production-and-use/; https://aeon.co/essays/can-we-hope-to-understand-how-the-greeks-saw-their-world | [kyaneos-modern] The chosen excerpt about Egyptian "ḫśbd iryt" is irrelevant to Greek cyan. The same source's next sentences give "κύανος" as dark blue/lapis and the root of Latin cyanus and English cyan. Sassi also gives later Greek "dark blue, kuanoun". The broad later-blue development is supported; cyan here is an English descendant/gloss, not an exact ancient spelling.
- CONFIRMED | "The word that would one day carry blue is, in Homer, doing the work of "dark."" | https://www.gutenberg.org/files/53004/53004-h/53004-h.htm | [kyaneos-role] Gladstone p. 465: "With perhaps this exception, the word ‘dark’ will cover all the uses". Sassi recognizes a dark blue merging into black. The page follows the explicit Gladstone interpretation, not an assertion that the Greek word lacks all blue referents.
- CONFIRMED | "Later "purple," but in Homer it marks things dark, deep or churning: blood, a swelling wave, a brooding cloud, the rainbow. Closer to "darkly agitated" than to any one hue." | https://www.gutenberg.org/files/53004/53004-h/53004-h.htm; https://aeon.co/essays/can-we-hope-to-understand-how-the-greeks-saw-their-world | [porphyreos] Read Gladstone's actual object list and interpretation, with Sassi's discussion of "brightness and movement". The page's agitated/dark interpretation is supported as a reading; it is not a measured fixed hue.
- CONFIRMED | ""Violet-like"" [...] "and Homer applies it to the sea" | https://www.gutenberg.org/files/53004/53004-h/53004-h.htm | [violet-sea-use] Source excerpt read against the whole claim: "There is certainly a great temptation, when we find in Homer the ἰοειδέα πόντον , to give him credit for the full meaning of this very beautiful epithet, which he uses thrice for the sea (Il. ix. 298, Od. v. 55, xi. 106), and never in any other connection." Location: Vol. 3, pp. 470-471
- CONFIRMED | "A "violet sea" beside a "wine-dark sea": the sea takes whatever dark epithet the line wants." | https://www.gutenberg.org/files/53004/53004-h/53004-h.htm; https://aeon.co/essays/can-we-hope-to-understand-how-the-greeks-saw-their-world | [violet-interpretation] Read Gladstone's actual object list and interpretation, with Sassi's discussion of "brightness and movement". The page's agitated/dark interpretation is supported as a reading; it is not a measured fixed hue.
- CONFIRMED | "The two words that dominate Homer's colour vocabulary by a wide margin. This is Gladstone's core point: the palette is organised first and foremost around light and dark" [...] "brightness" [...] "with true hue words comparatively rare and blue absent." | https://www.gutenberg.org/files/53004/53004-h/53004-h.htm | [palette-dominance] Read Gladstone pp. 476–488. He explicitly describes the "vast predominance" of white and black and concludes "Homeric colours are really the modes and forms of light". The page attributes this brightness-centred reading to Gladstone; it is not a new numerical census.
- CONFIRMED | "The Egyptians had a word for blue" [...] "ḫsbḏ (the term for lapis lazuli and dark blue)" | https://www.egyptianblue.info/4-production-and-use/; https://www.aic-publishing.org/ojs/index.php/JAIC/article/download/122/115/230 | [egypt-word] The production review gives "ḫśbd iryt, meaning artifcial lapis lazuli, or artifcial blue". The transliterations are explicitly approximate. The material-to-colour word is independently supported by the AIC review: "lapis lazuli was called hsbd" and the term extended to blue.
- CONFIRMED | "around the third millennium BCE they manufactured blue:" | https://www.nature.com/articles/s40494-019-0296-4 | [egypt-date] Opened the source marked UNREADABLE by Node with curl. The pigment study says "early 3rd millennium BCE" and identifies "calcium-copper tetrasilicate" and "CaCuSi4O10". The chemical identity and date are supported.
- CONFIRMED | "Egyptian blue (calcium copper silicate, CaCuSi₄O₁₀)" | https://www.nature.com/articles/s40494-019-0296-4 | [egypt-formula] Opened the source marked UNREADABLE by Node with curl. The pigment study says "early 3rd millennium BCE" and identifies "calcium-copper tetrasilicate" and "CaCuSi4O10". The chemical identity and date are supported.
- CONFIRMED | "whose name ḫsbḏ-ỉrjt translates roughly as "artificial lapis lazuli."" | https://www.egyptianblue.info/4-production-and-use/; https://www.aic-publishing.org/ojs/index.php/JAIC/article/download/122/115/230 | [egypt-name-translation] The production review gives "ḫśbd iryt, meaning artifcial lapis lazuli, or artifcial blue". The transliterations are explicitly approximate. The material-to-colour word is independently supported by the AIC review: "lapis lazuli was called hsbd" and the term extended to blue.
- CONFIRMED | "Why is genuinely unsettled" [...] "there are several live explanations and they pull against each other." | https://home.uchicago.edu/~johnlucy/papersmaterials/1997%20Linguistics%20of%20Color%20v1.pdf; https://tedlab.mit.edu/tedlab_website/researchpapers/gibson2017color.pdf | [unsettled-explanations] Read the scanned Lucy p. 343 by single-page OCR. It both names Saunders and objects that colour-chart elicitation rules out meaning differences. Gibson et al. offers communicative need as an alternative account. The record's one footnote is not the whole evidentiary basis.
- CONFIRMED | "Colour words tend to attach to things people need to name and can point to reliably." | https://tedlab.mit.edu/tedlab_website/researchpapers/gibson2017color.pdf | [words-objects] Gibson et al. directly reports "objects tend to have warm rather than cool colors" and studies colour usefulness for natural and artificial objects. Read the page's causal discussion as a proposed explanation, which its heading and objection explicitly frame as unsettled; this does not establish the separate easiest-dye claim.
- CONFIRMED | "Red is everywhere you'd want a word" [...] "blood, fire, ochre, ripe fruit" | https://tedlab.mit.edu/tedlab_website/researchpapers/gibson2017color.pdf | [red-salience] Gibson et al. directly reports "objects tend to have warm rather than cool colors" and studies colour usefulness for natural and artificial objects. Read the page's causal discussion as a proposed explanation, which its heading and objection explicitly frame as unsettled; this does not establish the separate easiest-dye claim.
- CONFIRMED | "Blue is comparatively scarce as a discrete, manipulable thing" | https://tedlab.mit.edu/tedlab_website/researchpapers/gibson2017color.pdf | [blue-manipulable] Gibson et al. directly reports "objects tend to have warm rather than cool colors" and studies colour usefulness for natural and artificial objects. Read the page's causal discussion as a proposed explanation, which its heading and objection explicitly frame as unsettled; this does not establish the separate easiest-dye claim.
- CONFIRMED | "few foods are blue, many "blue" flowers are really violet" | https://www.ebi.ac.uk/europepmc/webservices/rest/PMC8068391/fullTextXML; https://www.rhs.org.uk/weeds/bluebells-as-weeds; https://www.rhs.org.uk/ponds/pond-plants | [blue-food-flowers] The review supports few blue blossoms and fruits. The quoted excerpt does not support the extra violet-flower clause by itself. RHS descriptions provide concrete violet-blue examples, including bluebells, Iris versicolor, and Iris reticulata. No numerical prevalence is asserted on the page.
- CONFIRMED | "stable blue pigments and dyes are genuinely hard to source and make." | https://www.ebi.ac.uk/europepmc/webservices/rest/PMC8068391/fullTextXML | [blue-pigment-scarcity] Read the full review, not only the food-additive comparison. It says producing true blue compounds typically requires "multiple biosynthetic steps and heavy energy investment" and identifies stability problems. This supports the qualified general difficulty.
- CONFIRMED | "No easy colorant, few canonical blue objects, late word." | https://tedlab.mit.edu/tedlab_website/researchpapers/gibson2017color.pdf | [colorant-late] Gibson et al. directly reports "objects tend to have warm rather than cool colors" and studies colour usefulness for natural and artificial objects. Read the page's causal discussion as a proposed explanation, which its heading and objection explicitly frame as unsettled; this does not establish the separate easiest-dye claim.
- CONFIRMED | "the sky and the sea are blue and utterly everywhere." | https://aeon.co/essays/can-we-hope-to-understand-how-the-greeks-saw-their-world; https://tedlab.mit.edu/tedlab_website/researchpapers/gibson2017color.pdf | [sky-sea] These are ordinary physical descriptions within an explicitly labelled objection to the hypothesis, not a claim that the sea has one invariant hue. Sassi specifically says sea colour changes with "different hours of the day and with different weather". The record's warm-object excerpt does not itself support hourly variation.
- CONFIRMED | "the sky is diffuse, untouchable and ever-changing, and the sea's colour shifts hourly, so neither is the kind of fixed object you can hold, trade, or dye cloth to match." | https://aeon.co/essays/can-we-hope-to-understand-how-the-greeks-saw-their-world; https://tedlab.mit.edu/tedlab_website/researchpapers/gibson2017color.pdf | [sky-sea-variable] These are ordinary physical descriptions within an explicitly labelled objection to the hypothesis, not a claim that the sea has one invariant hue. Sassi specifically says sea colour changes with "different hours of the day and with different weather". The record's warm-object excerpt does not itself support hourly variation.
- CONFIRMED | "Berlin, Kay and their successors read the ladder as evidence of something universal in how human colour categories form" [...] "possibly anchored in the visual system's own salience structure." | https://wals.info/chapter/134; https://linguistics.berkeley.edu/kay/univ.col.cats.complt.pdf | [universal-explanation] The excerpt merely lists primaries. WALS section 2 supplies the missing explanation: "the widely accepted theory of colour opponency". Kay's review sets out the universal-category account and its revisions.
- CONFIRMED | "Barbara Saunders, John Lucy and other relativists" [...] "argue the whole result is shaped by its method: a Munsell colour chart and an English-trained notion of a "basic colour term" can quietly presuppose the very categories they then "discover."" | https://home.uchicago.edu/~johnlucy/papersmaterials/1997%20Linguistics%20of%20Color%20v1.pdf; https://tedlab.mit.edu/tedlab_website/researchpapers/gibson2017color.pdf | [critics-method] Read the scanned Lucy p. 343 by single-page OCR. It both names Saunders and objects that colour-chart elicitation rules out meaning differences. Gibson et al. offers communicative need as an alternative account. The record's one footnote is not the whole evidentiary basis.
- CONFIRMED | "The existence of many grue languages, which never lexicalise blue separately at all" | https://wals.info/chapter/134 | [never-grue] WALS confirms "Over half the languages in the World Color Survey contain a grue term." Read never in the present descriptive sense of no separate basic term in those systems, not a prediction about their future evolution. WALS is a contemporary sample, not a census of all language histories.
- CONFIRMED | "robust cross-linguistic regularities exist and are well-supported, but the strict universal staircase is contested and has been softened into trajectories" [...] "not settled law." | https://wals.info/chapter/134; https://home.uchicago.edu/~johnlucy/papersmaterials/1997%20Linguistics%20of%20Color%20v1.pdf | [regularities-contested] WALS: "83%" on the main trajectory, rather than all. Lucy's scanned p. 343 was rendered and OCRed individually; it criticizes the category projection and cites Saunders. The claim of a tendency and methodological dispute is supported.
- CONFIRMED | "once a language draws a boundary, that boundary measurably speeds its speakers' discriminations near it" [...] "an online, verbally-mediated nudge." | https://bpb-us-e1.wpmucdn.com/wp.nyu.edu/dist/4/2136/files/2017/01/winawer_witthoft_et_al_PNAS_2007.pdf | [boundary-speed] Source excerpt read against the whole claim: "Russian speakers showed a category advantage without interference [1,164 vs. 1,288 msec; t (20) ⫽ 2.59; P ⫽ 0.0176]" Location: Detailed Analyses, p. 7782
- CONFIRMED | "it is a nudge in speed and accuracy, never an inability to see." | https://languagelog.ldc.upenn.edu/nll/?p=18237 | [speed-not-inability] Language Log quotes Roberson on "speed of identification" and sometimes errors; the page's surrounding comparison concerns the literature, not a new accuracy effect estimated by the 124 ms statistic.
- CONFIRMED | "The viral "the Himba people can't see blue / can't tell blue from green"" [...] "usually the TV clip of a blue square nobody can pick out of green ones" [...] "is a media fabrication." | https://languagelog.ldc.upenn.edu/nll/?p=18237 | [himba-fabrication] Read the full Language Log post and all three quoted correspondents. It calls the portrayal a "journalistic fabrication", identifies a BBC demonstration and Korean-English stimuli, prints Davidoff's "there is nothing published", and reports no documented complete failure by Himba participants. No group is read in this cultural/language comparison, not as a denial of clinical colour-vision deficiencies.
- CONFIRMED | "The researcher who set up that on-camera demo (Serge Caparos) says it was a demonstration built for the BBC, not an experiment" | https://languagelog.ldc.upenn.edu/nll/?p=18237 | [caparos-demo] Read the full Language Log post and all three quoted correspondents. It calls the portrayal a "journalistic fabrication", identifies a BBC demonstration and Korean-English stimuli, prints Davidoff's "there is nothing published", and reports no documented complete failure by Himba participants. No group is read in this cultural/language comparison, not as a denial of clinical colour-vision deficiencies.
- CONFIRMED | "using stimuli from a different study" | https://languagelog.ldc.upenn.edu/nll/?p=18237 | [other-stimuli] Read the full Language Log post and all three quoted correspondents. It calls the portrayal a "journalistic fabrication", identifies a BBC demonstration and Korean-English stimuli, prints Davidoff's "there is nothing published", and reports no documented complete failure by Himba participants. No group is read in this cultural/language comparison, not as a denial of clinical colour-vision deficiencies.
- CONFIRMED | "Jules Davidoff says "there is nothing published" about it." | https://languagelog.ldc.upenn.edu/nll/?p=18237 | [davidoff-quotation] Read the full Language Log post and all three quoted correspondents. It calls the portrayal a "journalistic fabrication", identifies a BBC demonstration and Korean-English stimuli, prints Davidoff's "there is nothing published", and reports no documented complete failure by Himba participants. No group is read in this cultural/language comparison, not as a denial of clinical colour-vision deficiencies.
- CONFIRMED | "No group is perceptually unable to see a colour." | https://languagelog.ldc.upenn.edu/nll/?p=18237 | [no-group-unable] Read the full Language Log post and all three quoted correspondents. It calls the portrayal a "journalistic fabrication", identifies a BBC demonstration and Korean-English stimuli, prints Davidoff's "there is nothing published", and reports no documented complete failure by Himba participants. No group is read in this cultural/language comparison, not as a denial of clinical colour-vision deficiencies.
- CONFIRMED | "blue is the last of the six primary colour categories to acquire its own basic term, and in over half of the World Color Survey / WALS sample it still hasn't" [...] "those languages use one word for green-and-blue ("grue")." | https://files.eric.ed.gov/fulltext/ED019655.pdf; https://wals.info/chapter/134 | [apparatus-boundary] Read as the classical primary-category sequence and the qualified statistical tendency stated on the page, not a universal exceptionless law. Berlin and Kay: "At Stage V the focus of blue emerges". WALS records alternatives and the "83%" main trajectory.
- CONFIRMED | "Brown, purple, pink, orange and grey typically come after blue." | https://wals.info/chapter/134 | [apparatus-later] Source excerpt read against the whole claim: "Usually, derived categories appear only after stage V, where each primary has received a separate term, but there are many exceptions." Location: Section 3
- CONFIRMED | "Egyptian transliterations (ḫsbḏ, ḫsbḏ-ỉrjt) are conventional approximations." | https://www.egyptianblue.info/4-production-and-use/; https://www.aic-publishing.org/ojs/index.php/JAIC/article/download/122/115/230 | [egypt-transliteration] The production review gives "ḫśbd iryt, meaning artifcial lapis lazuli, or artifcial blue". The transliterations are explicitly approximate. The material-to-colour word is independently supported by the AIC review: "lapis lazuli was called hsbd" and the term extended to blue.
- CONFIRMED | "English 'blue' is a single word for what Russian splits into goluboy and siniy" | https://bpb-us-e1.wpmucdn.com/wp.nyu.edu/dist/4/2136/files/2017/01/winawer_witthoft_et_al_PNAS_2007.pdf | [relates-russian] Source excerpt read against the whole claim: "Unlike English, Russian makes an obligatory distinction between lighter blues (‘‘goluboy’’) and darker blues (‘‘siniy’’)." Location: Abstract, p. 7780
- CONFIRMED | "Berlin, B. & Kay, P. (1969), Basic Color Terms: Their Universality and Evolution , Univ. of California Press" | https://linguistics.berkeley.edu/kay/univ.col.cats.complt.pdf | [citation-berlin] Kay's references give the authors, title, 1969, and University of California Press exactly. The linked PDF is a later review, not the book itself.
- CONFIRMED | "Kay, P., Berlin, B., Maffi, L., Merrifield, W.R. & Cook, R. (2009), The World Color Survey , CSLI" | https://linguistics.berkeley.edu/kay/; https://linguistics.berkeley.edu/kay/univ.col.cats.complt.pdf | [citation-wcs] Read Paul Kay's official publication list. It gives the named authors and the 2009 CSLI book, and Kay and Maffi's 1999 American Anthropologist paper. The latter is confirmed as 101(4) by the reference list in Kay's review.
- CONFIRMED | "Kay, P. & Maffi, L. (1999), "Color Appearance and the Emergence and Evolution of Basic Color Lexicons," American Anthropologist 101(4)" | https://linguistics.berkeley.edu/kay/; https://linguistics.berkeley.edu/kay/univ.col.cats.complt.pdf | [citation-kay-maffi] Read Paul Kay's official publication list. It gives the named authors and the 2009 CSLI book, and Kay and Maffi's 1999 American Anthropologist paper. The latter is confirmed as 101(4) by the reference list in Kay's review.
- CONFIRMED | "WALS Online, ch. 134 (Green and Blue)" | https://wals.info/chapter/134 | [citation-wals] Source excerpt read against the whole claim: "Chapter Green and Blue by Paul Kay and Luisa Maffi" Location: Chapter title and byline
- CONFIRMED | "Saunders, B. (2000), "Revisiting Basic Color Terms," J. Royal Anthropological Institute 6(1)" | https://api.crossref.org/works/10.1111/1467-9655.00005 | [citation-saunders] The Node NOT FOUND result is an ordering artifact: every quoted JSON fragment is literally present, but the source record arranges the fields in a different order. Positive controls: exact author, title, published date, and container fields all match. Parsed metadata verifies the citation; there is no bibliographic error.
- CONFIRMED | "Lucy, J. (1997), "The linguistics of 'color'"" | https://api.crossref.org/works/10.1017/CBO9780511519819.015 | [citation-lucy] The Node NOT FOUND result is an ordering artifact: every quoted JSON fragment is literally present, but the source record arranges the fields in a different order. Positive controls: exact author, title, published date, and container fields all match. Parsed metadata verifies the citation; there is no bibliographic error.
- CONFIRMED | "Winawer, Witthoft, Frank, Wu, Wade & Boroditsky (2007), "Russian blues reveal effects of language on color discrimination," PNAS 104(19):7780–7785" | https://bpb-us-e1.wpmucdn.com/wp.nyu.edu/dist/4/2136/files/2017/01/winawer_witthoft_et_al_PNAS_2007.pdf | [citation-winawer] The PDF and Crossref agree on all six surnames, 2007, PNAS 104(19), pages 7780–7785, and DOI 10.1073/pnas.0701644104. DOI MISMATCH is a parser error: it mistakes the middle initial C. in Michael C. Frank for his first initial. The record's extraction moves Boroditsky before Jonathan and after Lera; use the actual byline instead.
- CONFIRMED | "Roberson, Davidoff, Davies & Shapiro (2005), "Color categories: evidence for the cultural relativity hypothesis," Cognitive Psychology 50(4)" | https://api.crossref.org/works/10.1016/j.cogpsych.2004.10.001 | [citation-roberson] The Node NOT FOUND result is an ordering artifact: every quoted JSON fragment is literally present, but the source record arranges the fields in a different order. Positive controls: exact author, title, published date, and container fields all match. Parsed metadata verifies the citation; there is no bibliographic error.
- CONFIRMED | "Gladstone, W.E. (1858), Studies on Homer and the Homeric Age , vol. 3" | https://www.gutenberg.org/files/53004/53004-h/53004-h.htm | [citation-gladstone] The title page gives "W. E. GLADSTONE", "VOL. III.", and "M.DCCC.LVIII."; 1858 is the correct conversion.
- CONFIRMED | "Geiger, L. (lectures of the 1860s, pub. 1871), Zur Entwickelungsgeschichte der Menschheit" | https://play.google.com/store/books/details/Lazarus_Geiger_Zur_entwickelungsgeschichte_der_Men?id=Pvq4BhvPt1oC; https://archive.org/download/contributionstoh00geigrich/contributionstoh00geigrich_djvu.txt | [citation-geiger] The Google Books record gives Lazarus Geiger, the title, "Jan 1871", and J.G. Cotta. The English translation dates the colour lecture "September 24, 1867"; the historical lecture/publication distinction is correct.
- CONFIRMED | "Sassi, M.M. (2017), "The sea was never blue," Aeon" | https://aeon.co/essays/can-we-hope-to-understand-how-the-greeks-saw-their-world | [citation-sassi] The article title and byline are present; raw metadata gives "datePublished":"2017-07-31T10:00:24.000Z". The 2017 citation is supported.
- CONFIRMED | "Liberman, M., Language Log on the Himba/BBC demonstration" | https://languagelog.ldc.upenn.edu/nll/?p=18237 | [citation-liberman] The post gives "Himba color perception", "March 17, 2015", and "Filed by Mark Liberman". The source and attribution are correct.
- CONFIRMED | "Sharma, Wu & Dalal (2005), "The CIEDE2000 color-difference formula," Color Res. Appl. 30(1):21–30" | https://hajim.rochester.edu/ece/sites/gsharma/ciede2000/ | [citation-sharma] The author's official page gives G. Sharma, W. Wu, E. N. Dalal, Color Research and Application 30(1), 21–30, February 2005. The page uses a shortened main title; the full title adds Implementation Notes, Supplementary Test Data, and Mathematical Observations.
- CONFIRMED | "Egyptian blue ." | https://en.wikipedia.org/wiki/Egyptian_blue | [citation-egyptian-blue] The linked article names Egyptian blue and says "It is considered to be the first synthetic pigment." Read first as the conventional earliest-known manufactured pigment claim, not a proof that no earlier unknown example could exist.
- CONFIRMED | "the first synthetic pigment in history" | https://en.wikipedia.org/wiki/Egyptian_blue | [egypt-first-pigment] The linked article names Egyptian blue and says "It is considered to be the first synthetic pigment." Read first as the conventional earliest-known manufactured pigment claim, not a proof that no earlier unknown example could exist.
- CONFIRMED | "The ancestor of "chlorophyll,"" | https://en.wikipedia.org/wiki/Chlorophyll | [chlorophyll-derivation] The retrieved article gives Greek "χλωρός" and "φύλλον" in the etymology, confirming the stated root.
- CONFIRMED | "Hebrew ḥesed has no single English word" | https://biblehub.com/hebrew/2617.htm | [hesed-english-equivalent] The lexicon lists "Lovingkindness, mercy, steadfast love, loyalty, faithfulness" and multiple contextual renderings. The claim concerns absence of one full semantic equivalent, not inability to translate the word in a given sentence.
- CONFIRMED | "in the largest cross-linguistic survey" | https://wals.info/chapter/134; https://lclab.berkeley.edu/papers/info-evoln-2015.pdf | [largest-survey] WALS: "Over half the languages in the World Color Survey contain a grue term." The original excerpt alone does not establish largest. Xu et al. call WCS "the largest existing publicly available database of color naming data"; the page concerns cross-language colour naming, not every possible cross-linguistic survey.
- CONFIRMED | "The World Color Survey (110 unwritten languages)" | https://linguistics.berkeley.edu/kay/univ.col.cats.complt.pdf; https://wals.info/chapter/134 | [unwritten-sample] Kay's review explicitly says "Color naming data were collected from 110 unwritten languages". WALS supplies 110 and the main trajectory, but its excerpt alone does not establish unwritten.
- CONFIRMED | "a dichromat collapses red and green" | https://www.ebi.ac.uk/europepmc/webservices/rest/search?query=EXT_ID:2487095%20AND%20SRC:MED&resultType=core&format=json | [dog-dichromat] The PubMed URL returned a challenge to curl and was opened through Europe PMC metadata for PMID 2487095. The abstract reports two cone classes, peaks at 429 and 555 nm, and "dogs have dichromatic color vision". In the dog context this supports red-green confusion, not a claim about every type of human dichromacy.

## Findings

- [fixed] X1, this layer's authored text: WALS section 3 puts separate primary terms at stage V, allows later derived categories and reports exceptions. This layer's opening and authored descriptions now state a tendency among six primaries. The share image and two authored copies in other layers have separate dispositions below. MINOR: recorded here only; no page correction line.
- [fixed] X2: Kay, Universality of Color Categorization, p. 3 permits either green or yellow in a four-term system. The absolute wording became common sequences. MINOR: recorded here only; no page correction line.
- [fixed] X3: WALS section 4.3 includes a separate blue term with a combined yellow/green term. The implication is now scoped to the classical sequence and called a tendency. MINOR: recorded here only; no page correction line.
- [fixed] X4: WALS section 3 distinguishes the six primaries from later derived categories. Search metadata now makes that scope and tendency explicit. MINOR: recorded here only; no page correction line.
- [fixed] X5: Sassi opens with kuaneos and glaukos as descriptions of blue shades. The description and related blue-absence wording now distinguish Greek meanings from modern English blue. Added its dated Corrected line to the page and mirrored it at the end of the markdown body.
- [fixed] X6: WALS section 3 supports a qualified primary-category sequence, not a final basic word of every kind. Open Graph wording now states that scope. MINOR: recorded here only; no page correction line.
- [fixed] X7: WALS section 3 supports a qualified primary-category sequence. Twitter wording now names the six primaries and the tendency. MINOR: recorded here only; no page correction line.
- [fixed] X8: The actual byline on printed p. 7780 gives Jonathan Winawer. The page now uses that name. Added its dated Corrected line to the page and mirrored it at the end of the markdown body.
- [fixed] X9: Results, Boundaries and Analysis, p. 7781 measures each participant's classification boundary. The page fixes BOUNDARY=4, so its result label and code comment now call this illustrative. MINOR: recorded here only; no page correction line.
- [fixed] X10: Re-evaluating the page functions gives ideal Lab example distances 9.8729537654 within and 10.0653854013 across. The page and verifier now say approximately matched model distances. MINOR: recorded here only; no page correction line.
- [fixed] X11: The cross example remains slightly farther apart, both in ideal Lab and after integer RGB conversion. Neither computation rules out response-time confounds; that assurance is removed from prose and verifier. MINOR: recorded here only; no page correction line.
- [fixed] X12: Results, Boundaries, p. 7781 gives nearly identical English and Russian classification boundaries; the reported group difference concerns category advantage. The dynamic explanation now says this and does not infer a reader's categories. MINOR: recorded here only; no page correction line.
- [fixed] X13: Methods, Participants, p. 7784 reports 26 Russian and 24 English recruits; Analysis, p. 7782 excludes five Russian and three English participants. The 1164 versus 1288 ms comparison has t(20): 21 Russian participants, 42 analyzed overall, 50 recruited. Added its dated Corrected line to the page and mirrored it at the end of the markdown body.
- [fixed] X14: The companion frontmatter and six-hands description name six versions: Chapman, Pope, Cowper, Buckley, Butcher and Lang jointly, and Butler. This layer now says six translation versions, including its related-page note. Added its dated Corrected line to the page and mirrored it at the end of the markdown body.
- [fixed] X15: Re-read Geiger's English translation, On Colour-Sense, p. 55 footnote: the corpus named is Edda hymns. Changed Icelandic sagas to Eddic poems. MINOR: recorded here only; no page correction line.
- [fixed] X16: Geiger also discusses Chinese hiuan on p. 52. This duplicate finding receives the same Eddic-poems repair; the supported Chinese example remains. MINOR: recorded here only; no page correction line.
- [fixed] X17: Gladstone, vol. III, pp. 463-464, item h lists a dark cloud of vapour and explicitly excludes a storm-cloud. The palette example now follows that distinction. Added its dated Corrected line to the page and mirrored it at the end of the markdown body.
- [fixed] X18: Gladstone, vol. III, pp. 470-471 assigns ioeides to the sea, ioeis to iron at Iliad 23.850, and iodnephes to Helen's wool at Odyssey 4.135 and the sheep at 9.426. The palette now distinguishes all three. Added its dated Corrected line to the page and mirrored it at the end of the markdown body.
- [fixed] X19: Re-read the supplied Roberson et al. PDF, abstract p. 2: similarity judgments, short-term memory and long-term learning. The prose names these tasks and limits the Russian comparison to category-related performance. MINOR: recorded here only; no page correction line.
- [fixed] X20: Re-evaluating the page functions gives maximum adjacent-step deviation 0.4507187204% in ideal Lab and 3.8732660878% after integer RGB conversion. The 0.5% bound is retained with ideal Lab scope; displayed spacing is approximate. MINOR: recorded here only; no page correction line.
- [fixed] X21: Same computed example pairs as X10: cross 10.0653854013 exceeds within 9.8729537654 in ideal Lab. The dek now says approximately matched model distances. MINOR: recorded here only; no page correction line.
- [fixed] X22: WALS sections 1-2 identify 110 surveyed languages and restrict the grue majority to that sample. The related-page note and sampled-language shorthand now name the World Color Survey. Added its dated Corrected line to the page and mirrored it at the end of the markdown body.
- [fixed] X23: Reference-pair agreement validates model calculations, not exact perceptual equality on a display. The related-page note now describes approximately equal model distances in ideal Lab. MINOR: recorded here only; no page correction line.
- [fixed] X24: WALS section 3 distinguishes primary and derived categories with exceptions. JSON-LD now qualifies both the six-primary scope and tendency. MINOR: recorded here only; no page correction line.
- [dated] X25: Set JSON-LD dateModified to 2026-10-14. No Updated line is required by the explicit dateModified exception; the seven Corrected lines date this revision. Kept datePublished and the historical June verification observation.
- [fixed] X26: Berlin and Kay's original report, stage VII, PDF page 27, leaves the remaining categories unordered. The dial instruction and dek now label the within-stage choices illustrative. MINOR: recorded here only; no page correction line.
- [fixed] X1 share-card copy, public/og/the-last-colour.png: the image repeats the old markdown dek opening. Landing disposition: share card rebuilt at landing from the corrected dek. The matching public/og-thumbs/the-last-colour.webp is left for the landing refresh. Neither image was changed or regenerated by this repair.
- [fixed] X1 copy in Do Bulls Hate Red's relation note (src/content/strata/do-bulls-hate-red.md): it now says languages tend to name blue last among the six primary colour categories (WALS chapter 134), corrected on 2026-10-18 with no line on the page (MINOR); board note 5c9c5a is closed.
- [fixed] X1 copy in The Thirty Sayings's relation note (src/content/strata/the-thirty-sayings.md): it now says blue tends to be the last of the primary colour terms, corrected on 2026-10-18 with no line on the page (MINOR); board note 5c9c5a is closed.