Short answer
Russian (goluboy and siniy), Greek (ghalazio and ble) and Italian (azzurro and blu, with celeste as a third for Tuscan speakers) are the documented cases; Turkish lacivert and Japanese mizu are argued over. In every case the line is drawn by lightness, not hue. English draws the same line with secondary words: in the Colourwise lexicon blue covers OKLCH lightness 0.35–0.75, navy stops at 0.32 and sky blue starts at 0.72.
Every pair below separates a lighter blue from a darker one. English has the same distinction without a basic word for either side. In the lexicon the blue region covers OKLCH lightness 0.35–0.75; navy covers 0.15–0.32 and shares 0% of its sRGB grid colours with blue; sky blue covers 0.72–0.88 and shares 5%. So English speakers already treat very dark and very light blues as needing their own names. What differs in Russian is the status of those names: goluboy is not a kind of siniy in the way navy is a kind of blue, and a speaker cannot use one general word and leave the lightness unstated.
English does the same thing elsewhere: pink is a basic term for light red. The question the two-blue languages raise is why English stopped there.
For Russian, a 1994 naming study counted goluboy as a twelfth basic term, and Berlin and Kay had already listed Russian as a possible exception to their limit of eleven. For Greek, counts of words in texts and four naming studies with different samples, lighting and instructions converged on twelve basic terms, with ghalazio among the most frequent colour words in the language. For Italian, an unconstrained naming study of 367 colours found azzurro and blu both behaving as basic terms and dividing the blue area by lightness, and noted that an earlier design which allowed only blu had biased the result. In each language the evidence is frequency, agreement between speakers and consistency of use, not a dictionary's say-so.
Turkish lacivert, dark blue, is used often and consistently, and by those measures a 1998 study found twelve Turkish basic terms. But when students were asked directly, 85.5% said lacivert is a kind of mavi, the general blue word — which breaks the rule that a basic term must not be included in another. The study's authors called eleven the safest conclusion, and a later field test called lacivert a candidate. Japanese shows the movement in the other direction. In 1987 ten observers gave the same eleven basic terms as English speakers; in 2017, 98% of 57 informants used mizu for light blues, and the researchers proposed it as a twelfth. A second blue can become established within a generation.
It would be convenient if siniy, ble, blu and lacivert all meant one thing. They do not. The Greek study compared the pairs and found that although all are divided mainly by lightness, the best examples of the terms sit in different places: Russian siniy is on average darker than Greek ble, and Turkish lacivert darker again, while Turkish mavi is somewhat darker than the cross-language blue landmark and Russian goluboy about level with it. Italian adds regional variation: azzurro covers medium and light blue in Verona but only a darker medium blue in Alghero, where celeste takes the lighter range. A translation that maps 'the dark blue word' in one of these languages onto 'the dark blue word' in another will be off by a visible amount.
The two-blue languages are where the definition of a basic colour term comes under most strain. Lacivert passes the tests of salience and agreement and fails the test of not being a kind of something else. Goluboy has been argued to be basic for cultural reasons — its symbolic weight in Russian — and not because perception demands a division there. Celeste is basic by one study's measures for Tuscan speakers and not elsewhere in Italy. Colourwise would expect English navy to score much as lacivert does: frequent, consistent, and understood as a kind of blue. The practical reading is that 'basic' is a matter of degree measured several ways, and that a language's count of basic terms is a conclusion from a particular study, not a fixed property.
| Language | Term | What it names | Basic colour term? | English lexicon landmarks |
|---|---|---|---|---|
| Russian | голубой (goluboy) | The lighter blues. | Basic. Counted as a twelfth basic term in Davies and Corbett's 1994 naming study, as reported by Rätsep (2012), and by Paramei (2005), who argues it became basic for cultural reasons. Its status is why Russian breaks the proposed limit of eleven. | sky blue, blue |
| Russian | синий (siniy) | The darker blues. | Basic. Treated as a basic term throughout the literature reviewed by Paramei (2005). | blue, navy |
| Modern Greek | γαλάζιο (ghalazio) | The lighter blues. | Basic. Androulaki and colleagues (2006): the twelfth basic term. | sky blue, blue |
| Modern Greek | μπλε (ble) | The darker blues. | Basic. Androulaki and colleagues (2006): one of twelve basic terms. | blue, navy |
| Italian | celeste | Light blue; literally 'of the sky'. | Contested. A third basic blue for Tuscan speakers according to Del Viva and colleagues (2022, 31 participants); the same method gave two basic blues in Verona and in Alghero. | sky blue |
| Italian | azzurro | Light to medium blue; how light depends on the region. | Basic. Paggetti, Menegaz and Paramei (2016): behaves as a basic term beside blu. | sky blue, blue |
| Italian | blu | Dark blue. | Basic. Paggetti, Menegaz and Paramei (2016). | blue, navy |
| Turkish | mavi | Blue. | Basic. Özgen and Davies (1998), as reported by Rätsep (2012). | blue |
| Turkish | lacivert | Dark blue. | Contested. Özgen and Davies (1998) found it meets the salience and consensus tests for a twelfth basic term but fails the test that a basic term must not be included in another; Rätsep (2012) calls it a candidate, not a settled basic term. | navy |
| Japanese | 水 (mizu) | Light blue; the word means 'water'. | Emerging. Kuriki and colleagues (2017): used by 98% of 57 informants and proposed as a twelfth basic term. Uchikawa and Boynton (1987) did not find it basic. | sky blue |
| Japanese | 青 (ao) | Blue; also said of some things English calls green. | Basic. Uchikawa and Boynton (1987) and Kuriki and colleagues (2017) for modern Japanese; Tobioka (2020) for the eighth century. | blue, green |
| Japanese | 紺 (kon) | Dark blue; the word is glossed 'indigo' in the cited study. | Not basic. Kuriki and colleagues (2017). | navy |
| Term | Lightness range | Hue range | Status | Share of its region inside blue |
|---|---|---|---|---|
| Sky blue | 0.72–0.88 | 220°–252° | Secondary | 5% |
| Blue | 0.35–0.75 | 240°–275° | Basic | — |
| Royal blue | 0.32–0.62 | 260°–272° | Secondary | 92% |
| Cobalt | 0.36–0.52 | 252°–266° | Secondary | 100% |
| Navy | 0.15–0.32 | 250°–272° | Secondary | 0% |
| Midnight blue | 0.12–0.32 | 262°–285° | Secondary | 0% |
Why: English 'blue' was translated with the general-looking word, which in these languages names the darker blues only.
Fix: Choose the term from the swatch, not from the English word: goluboy, azzurro or celeste for a light blue.
Why: The interface assumed one basic blue.
Fix: Offer a light and a dark blue as separate named choices, as you would offer pink apart from red in English.
Each statement is labelled by kind — established fact, a standard’s requirement, observed market data, a convention, or Colourwise’s own interpretation or analysis — with the strength of the evidence behind it.
FactStrong evidence
Modern Greek has twelve basic colour terms including two for blue, whose ranges differ mainly in lightness; the division resembles the Russian and Turkish ones, but the best examples of the terms differ between the three languages.
FactStrong evidence
In an unconstrained naming study of 367 colours, Italian azzurro and blu both performed as basic colour terms, dividing the blue area along the lightness dimension.
Source: Color naming in Italian language (Color Research & Application 41: 402–415, 2016)
FactModerate evidence· Tuscany, Verona and Alghero, Italy
Among 31 Tuscan-born participants, blu, azzurro and celeste were three distinct categories separated mainly by lightness; the same method gave two basic blues in Verona (azzurro and blu) and in Alghero (celeste and blu).
Caveat: Small regional samples; the pattern may not hold for all speakers in each region.
Source: Florence “blues” are clothed in triple basic terms (i-Perception 13(5), 2022); Diatopic variation in the referential meaning of the “Italian blues” (in Progress in Colour Studies, pp. 83–106, 2018)
FactModerate evidence
Turkish lacivert met the salience and consensus measures for a twelfth basic term in a 1998 study, but 85.5% of the students asked regarded it as a kind of mavi.
Caveat: The 1998 study's figures are taken from a later paper that reports and quotes them, because the original could not be read directly.
Source: Turkish color terms: tests of Berlin and Kay's theory of color universals and linguistic relativity (Linguistics 36(5): 919 onwards, 1998); Emphasis on the blue: Turkish basic colour terms (Rätsep; Studia uralo-altaica 49: 447–457, 2012)
FactStrong evidence
In a 2017 study 98% of 57 Japanese informants used mizu for light blue, and it was proposed as a twelfth basic term; a 1987 study of ten observers had found eleven.
Source: The modern Japanese color lexicon (Journal of Vision 17(3): 1, 2017); Categorical color perception of Japanese observers: comparison with that of Americans (Vision Research 27: 1825–1833, 1987)
Colourwise analysisLimited evidence
In the Colourwise lexicon the navy region shares 0% of its sRGB grid colours with the blue region and the sky blue region shares 5%.
Based on: Grid colours tested against the lexicon's editorial regions for blue, navy and sky blue, which are drawn around cited anchors.
Caveat: Reflects how the regions were drawn; English speakers readily call navy and sky blue 'blue', which the regions do not capture.
Source: CSS Color Module Level 4; Color survey results, and the rgb.txt list of the 954 most common colour names
Reviewed 1 October 2026. Colourwise summarises its sources in its own words and does not reproduce standards text or proprietary colour data. Spotted an error? Tell us.