Encyclopedia of Color Science and Technology

2016 Edition
| Editors: Ming Ronnier Luo

Ancient Color Categories

Reference work entry
DOI: https://doi.org/10.1007/978-1-4419-8071-7_75



Color terms or partitions of color denotata evidenced in ancient language artifacts.

Words and Hues, Languages and Time: An Overview

The sources for understanding the earliest color terms and categories are from the lands to the east and south of the Mediterranean Sea. Evidence of color categories from proto-cuneiform, Sumerian, Egyptian, and Akkadian in Mesopotamia and Egypt (from the end of the fourth millennium BC onwards) is followed (during the second millennium) by Greek in the West and Chinese in the East. Linguistic terms relating to color are present in all these languages.

What is known about the earliest color categories is derived from artifacts and texts. The use of color goes back at least 100,000 years, but the origins of color vocabulary lie in the period since roughly 8000 BC (=10,000 years ago), and the earliest texts (from ca. 3200 BC) appear millennia later. By comparison with the languages discussed here, virtually all other languages are much younger (e.g., Hebrew, Latin), or contemporary (e.g., Eblaitic, Hittite, Ugaritic), but linguistically related to the languages discussed here.

Vocabularies in the earliest preserved languages offer representative and definitive evidence concerning the origins of color categorization and its linguistic expression, as well as allowing evaluations of different steps in the process of abstraction and the early linguistic partitioning of perceptual color space. (For linguistic and historical details, see Refs. [1, 2, 3, 4, 5, 6, 7].)

Color Terminology

Black and White, Bright and Dark

The earliest color lexicons from languages of the Middle East and the eastern Mediterranean (proto-cuneiform, Sumerian, Akkadian, Egyptian, and early Greek) have words signifying “dark” and “light” as well as terms denoting something closer to “black” and “white.” Yet like most of the more specific terms for “black” in most ancient languages, even classical Greek melas had a semantic range including “black” and “dark” [8] that encompassed some regions described in English as “brown.” In general, virtually all of the linguistic glosses for “black” and “white” had category central exemplars or “foci,” but some also included a broader “light” and “dark.” Despite some overlap, words similar in meaning to English “black” and “white” are different from glosses such as “light,” “shining,” “gleaming,” “sparkling,” and “dark” and “gloomy,” respectively, as the latter glosses appear to not have category foci.


In all ancient languages, there is a very clear tendency for a division of the reddish continuum of color experiences into several different hues. While the color typically labeled “red” in modern English language usage was among the earliest distinguished in art and written artifacts, the concept of a category of “red” as a distinct linguistic unit was not dominant in the second and third millennia BC. In addition, there is clear evidence, in both earlier Greek and earlier Egyptian, of a tendency for a “red” and “white” opposition [3, 6, 9]. Moreover, in Akkadian and Greek, the word for a generic “red” was frequently used as a synonym for “colorful” or “colored” – and there is evidence suggesting this may be the case for Egyptian “red” as well.

Green, Green-Blue, and Green-Yellow

The Egyptian (wādj) and Akkadian (warqu) color terms that denote what today is referred to as “green” in English are derived from the same linguistic root but do not invariably denote the same color, since the Mesopotamian terms sig and warqu probably included “yellow” as well. The Akkadian warqu denoted both “green” and “yellow” appearances and was used to describe both vegetation and gold. The Akkadian warqu definitely did not mean “green-blue” or “green-yellow” [2]. Although the Egyptian wādj is related to the Akkadian warqu, the category centroid, or focus, of the Egyptian term was in “green” and neither “green-yellow” nor “green-blue” – and certainly not “yellow” [4]. By comparison, the Akkadian color term ḫašmānum has been associated with “blue-green” (as well as “light blue”) [10].

There was no generic word for our “green” in Greek, although xlōros eventually came to mean something like “green,” but the earliest use of color terms in Greek was not specific; green was divided and not dominated by a “green-blue.”

Qīng (“dark,” “green,” or “blue”) has not yet been found in the earliest Chinese inscriptions. Since the first millennium BC, qīng was used for “dark,” “blue,” and “black”; only slightly later, the word , today’s “green,” also appeared, so that to some extent “green” has since been divided into “light” and “dark” (green) [7].


Some of the color words preserved in the earliest Semitic languages (e.g., uqnu, “lapis lazuli” or “dark blue”) are loanwords for materials from other unknown older languages. Other terms – later shared in different languages – were possibly words (but certainly not the corresponding category “abstractions”) corresponding to “red” and “green” which may have existed in the early Neolithic of the Near East, perhaps 10,000–12,000 years ago, prior to the documentation of language [10].

In Akkadian (uqnu) and Egyptian (xsbdj), terms for lapis lazuli designated “dark blue.” In Greek, a term (kyaneos) for blue appearances is derived from the Akkadian. In Egyptian, turquoise (mfkāt) denoted “light blue.” Akkadian used several terms for “light blue” (including ḫašmanum, possibly from the Egyptian word for amethyst, ḥsmn, which was not used as a color word in Egyptian). Chinese lán is a term for “blue” colors but appears quite late (in comparison to, e.g., “red,” “white,” “black,” “yellow”). As a category, the modern English term “blue” evolved to ultimately eclipse the distinction (still preserved in Russian) between light and dark blue. Through the second millennium BC, color terms are mostly rooted in materials – most of which were later eclipsed with abstract words.


Early evidence of terms glossing “yellow” is less common but documented. In Egyptian, the word for “gold” (nb.w) was occasionally used to represent yellow. The linguistic usage of “gold” (nb.w) for “yellow” is not common in Egyptian; in Egyptian painting, however, the color yellow was frequently used to depict what was intended to represent gold where required and also the sun on occasion [11]. The usage of gold in Akkadian texts with the meaning of “yellow” is rarer than in Egyptian, and the form was a simile. In contrast to this earliest material, the later Greek words based on xrusos, “gold,” were frequently used to designate a color which was most probably “yellow”; significantly, zanthos “yellow” is also documented.

Table 1 offers an impression of what can be identified in the way of colors in these earliest languages.
Ancient Color Categories, Table 1

Summary of identified color terms from the earliest known languages










Time period

4th millennium

3rd millennium

3rd–2nd millennia

3rd–2nd millennia

2nd millennium

1st millennium

2nd millennium

1st millennium

Color terms:


BAR, ?U4










mi, gíg, ĝi6







“Bright red”

si/u4, NE6


























“Dark blue”








“Light green”




Other “Blues”


jrtjw, mfkāt

ḫašmānum, pelum, tukiltum





Other “Reds”


ša4, su9, sa5

rwdj, mss, tjms, mroš

ruššu, ḫuššu


jiàng, hóng



“White,” “light,” “bright”


ara, bar, ḫáda, dalla, kára, kug, píriĝrín, še-er, tán, zalag

ḥdj, tjḥnt


míng, qĭ

míng, qĭ




dara, ge, gíg/gege, kúkku, mi, šúš/šú


ṣallamu, tarku


Material Color

One of the greatest obstacles to understanding the nature of the earliest origins of ancient color terminology (in the Ancient Near East) and the origins of abstraction (in Greek and Chinese) is appreciating that in the earliest usage the ancients did not classify the world according to modern terminological divisions of the diversity of visible light, but rather that they initially used their perceptions of precious materials to express many of the colors they perceived. Thus, their base was the colors of the materials which they then applied to other domains. This eventually created the basis for abstract color terminology. However, the origins render the discussion complicated. (For references for this section, see [10, 11].)

Figure 1 approximates the relationships between color appearances and color terms.
Ancient Color Categories, Fig. 1

Examples of physical materials related to ancient color terminology displayed in congruent regions of a Munsell Color Chart

A good example of color abstraction is “gold.” In several languages, the word for “gold” implies color associations mostly from “yellow” portions of color space but can also imply “orange” (and even “red”) areas of color space. In contemporary terms, the color “gold” is generally considered similar to colors in the yellow range. By comparison, ancient users of color lexicons were unlikely to associate “gold” with “yellow” because ancient languages tended to strictly associate the color term “gold” with the physical materials of gold metals. This suggests variation in the underpinnings of modern and ancient concepts associated with terms denoting golden color appearances. Thus, color and material were related – and led to a linguistic partitioning of color in a fashion differing from our own “modern” understanding (as reflected in, e.g., English).

The earliest documented use of red (in the form of ochre) dates back 100,000 years – and thus long before the earliest documented written sources. Red and yellow ochre, along with black soot, is easily recognizable in the Paleolithic cave paintings from the Upper Paleolithic, from ca. 35,000 years ago. Green and blue are strikingly absent until tens of thousands of years later.

Stones with material properties producing green hues obtained prominence beginning approximately 10,000 years ago. Blue lapis lazuli appears gradually in South and Western Asia starting in the sixth millennium; red carnelian and blue-green turquoise are seen slightly later, as is jade in Europe and China. Gold and silver appear since the fifth and fourth millennia (respectively) in Europe and the Near East.

Many kinds of precious materials contributed to the concept of “shining” and “gleaming.” Significantly, the word for “white” or “bright” in Sumerian (babbar) and in earlier Egyptian (ḥdj) was the word for silver. In Greek, silver played a role in a word for “shining” (argos) – with yet another etymology. In his list of Indo-European etymologies, Shields seems to suggest that the abstract terms “bright,” “gleaming,” etc., are the basis for many words that eventually became abstract color words. However, it may be that the evidence suggests the opposite. That is, precious materials (rather than abstract “brightness”) are more likely to be linked to the origins of abstract color words. For example, Sumerian šuba means “agate” or “precious stone,” but also means “shining.” And Sumerian kug is the term for the metal silver and “bright” and “white.” Another Egyptian word for “gleaming” or “dazzling” can be related to tjḥn.t, a word later used for faïence, but probably originally referring to a form of naturally generated glass. In the ancient languages, there are many more words for appearance properties of “gleaming,” “shining,” “brilliant,” “bright,” etc., than can be fully inventoried in the present survey. In general, in ancient languages, ideas of shining and color originated from associations with precious materials although many did not lead to color words.

Yet examples of terms for precious materials that did have color meaning were nevertheless abundant: Egyptian ḥdj means silver – and has been identified as the basic color term for “white” in early Egyptian. The Akkadian sāmu refers to “carnelian” but is what is interpreted as the basic color term for “red.” Egyptian/Akkadian wādj/warqu are probably derived (through metathesis) from a widely used (Neolithic?) designation for jade (or a “greenstone”) that later became the basic color term “green” (in English; grün in German, etc.; see below). Egyptian xsbdj and Akkadian uqnu meant lapis lazuli but were used for “dark blue” – and the latter is related to the Greek kyaneos for “blue”; Egyptian mfkāt was turquoise but used for “light blue.” Egyptian nb.w was “gold” but used for “yellow.” Egyptian ḥsmn meant only amethyst in Egyptian, but Akkadian ḫašmānum designated not only a stone but also a “blue-green” or “light blue” and so on.

Of these materials, several eventually led to abstract color words, but usually only in the languages into which they were imported. This process seems to have begun in the second millennium BC, but only began to have systematic effects from the first millennium BC onwards.

Linguistic Issues in Ancient Color Naming

Black and White, Dark and Light

Significantly, a word for “darkness” shared in Sumerian (kuku) and Egyptian (kk.w) means that the word must have been of great antiquity, since it fed into two distinct language families. Yet these glosses for “darkness” were not related to the words used for “black” in either language; the origin of “black” in both languages lay elsewhere. The Chinese term for black also meant dark but was otherwise used as an adjectival modifier. The Chinese term for white bái did not signify “bright” or “clear” (which were míng and ).


It is significant that material, not color, properties of several terms are what was apparently salient throughout the history of the languages. In medieval Coptic, the early Egyptian for “white,” ḥdj, is replaced by ūbaš (of which the etymological meaning is to “shine”). The salient meaning of ḥdj lasting through Coptic is not “white,” but the material “silver.” In Chinese, the term for “red” is not salient in the sense described by Berlin and Kay [12], since the common Chinese word in the first two millennia of the language is chì and not hóng (which later replaced chì). In Mycenaean Greek, the main word for “red” is po-ni-ko-ro (later phonikos, a loanword referring to the Phoenicians who furnished the red dye) rather than e-ru-to (which gave rise to the later basic color term erythros).

Etymologies, Materials, and Loanwords

Egyptian and Akkadian “green” are most likely the same word and probably at the root of English “green” (sharing the radicals r and q/k/g/ĝ). An argument could be made for the diffusion of “red,” where Akkadian ruššu is probably the same as Greek erythros and Italian rosso. Thus, these are ancient concepts that have moved between languages.

As loanwords, precious materials also play an important role. In English, lapis lazuli has contributed the words “azure” (derived from Persian lazuward for lapis lazuli) and “cyan” (originally a loanword imported into Akkadian as uqnu for lapis lazuli and subsequently to Greek kynaeos).

Precious Materials, Loanwords, Abstraction, and Grammar

The classical Greek term kyaneos is derived from ko-wa-no, which was probably the Mycenaean Greek word for glass paste. The word ko-wa-no itself was derived (via Ugaritic or Hittite) from uqnu, the Akkadian for lapis lazuli (itself a loanword in Akkadian). In the Aegean ko-wa-no, the Asian uqnu was used to designate the artificial material glass (as opposed to the semiprecious stone lapis lazuli). Thus, this process of linguistic exchange apparently gave birth to the abstract color term kyaneos – but only in Greek of the first millennium BC. Earlier, its primary role was that of designating a precious material which gave birth to the color term. In Akkadian, the nouns lapis lazuli and cornelian appear regularly together in the same texts, along with ḫurṣāsu and ḫurāṣānû, “gold” and “golden.” In Mycenaean Greek, this same word appears as ku-ru-so “gold,” and later as Greek xrusos, where it is used as a color word – even though Greek had a word for “yellow” which can be traced back to the second millennium.

In Egyptian, the first reference to the “sky” as having a color is in Coptic, the latest stage of the language, in the first millennium AD. Prior to this, the sky was described as being turquoise or lapis lazuli (rather than having a color itself). Although vegetation was known to be “green,” the word for green is frequently associated with a classifier signifying a stone. Lapis lazuli was treated as a precious stone in these societies, only gold and lapis lazuli had prices higher than silver (which served as money). The statues of the gods in the temples were not made of granite, but rather of gold, silver, lapis lazuli, turquoise, ivory, etc. These materials were the origins of color words.

Although often considered “abstract,” the adjectives for colors in Sumerian were never attached to the nam abstract determinative. Given the difficulties of understanding adjectives in Egyptian, Schenkel used verbs as his criterion, i.e., ḥdj “to be white,” km “to be black,” dšr “to be red,” and wādj “to be green.” It is probably true that these words were verbs, but in some cases, they are also used as adjectives. The noun xsbdj “lapis lazuli” “dark blue” was a metaphor, a simile, and an adjective, but not a verb. The same is true for mfkāt “turquoise” used for “light blue.”

In Akkadian, color words are all adjectives. However, in the case of uqnu, the principal meaning is the noun “lapis lazuli”; an identical adjective means “lapis lazuli color(ed)”; uqnâtu is an adjectival form with the same meaning. The same is true of sāmtu “cornelian” and sāmtu “redness” and sāmu “red.”

In Chinese and Greek, color words are largely adjectives. Some of the Greek terms can be traced back to Near Eastern materials. Of the languages discussed here, Chinese is the only language with a completely abstract color vocabulary, where hues and terms match, more or less. Although silk appears as a component in the writing of some color words in Chinese, even as a component, jade played an even more marginal role. Thus Chinese color terms never bore primary relations to precious, natural object categories; the evolution of Chinese color vocabulary differed fundamentally from that of the West – but the results are similar to those in Greek, implying diffusion.

The Mediterranean languages usually had a word for “color” as a phenomenal experience; however, the words were not restricted to a single meaning in terms of hue. In Egyptian, the word jwn is probably derived from the designation for a “vein” of ore (meaning the material color which was visible), but it also meant “character” of a person in the figurative sense (what was hidden under the surface). By contrast, although later Chinese has such a word, in the earliest Chinese, no term for color as such has been discovered; in later Chinese, the suffix –si is frequently attached to color words.

Theory of Ancient Color Term Emergence

Debate regarding the sequence of emergence in early color terminology is constrained by limited artifact sources and historical evidence of ancient color term sequences. Moreover, color term emergence theories (see also Encyclopedia of Color Science and Technology entry on “ Berlin and Kay Theory”) that describe color language hierarchies found in contemporary and very recent language data cannot easily be applied to ancient color term emergence as they are based primarily on modern observations of modern phenomena and therefore do not take account of the relevant ancient data discussed here. This makes emergence theories of modern color lexicons problematic as models of both ancient partitions of gradient color continua and ancient color category acquisition sequences.

Warburton [10] argues that the earliest acquisition sequence was not an issue related to individual specific languages, but rather a shared phenomenon based on shared terminology, concepts, and materials. In this scheme, “ochre” (as dark red) will have been the earliest color used (in the Paleolithic, 100,000 years ago), followed by “greenstones” (in the earliest Neolithic, from 10,000 years ago), and “gold,” “silver,” lapis lazuli,” “carnelian,” and “black” (in the following late Neolithic and earliest Bronze Age) and then by turquoise (before 4,500 years ago).

Of extreme importance is that in ancient Egypt, the sun is occasionally called “red” and painted as such. This does not imply a “red-yellow,” but rather the use of a single term to designate different colors, as with the Mesopotamia “green” used for yellow and green. The reasons behind this do not lie in perception so much as expression (based on gold, bronze, sun, vegetation, etc.).

Loanwords and common etymologies play important roles in emergence of ancient color terms in that, as detailed above, many of the earliest words are related to designations of materials. Terms for “white” and “black” do not stand at the origins, and the red, blue, and green ranges were partitioned before being lumped together. Only later did the further color words begin to crystallize (in the last two millennia). The concept of “blue-green” is not documented in the earliest languages; it appears only recently in Chinese and does not universally evoke color significance among informants [7].

Color in Art and Language

Of particular interest for the understanding of the evolution of color terminology is the fact that for Egyptian, Akkadian, and Sumerian (which all date to before the middle of the second millennium BC), it is agreed that the color terminology did not match the colors that were used in the contemporary art. In contrast to this, already for Mycenaean Greek and classical Greek (thus, from the second half of the second millennium BC onwards), Blakolmer is persuaded that “it is hardly surprising that in the use of color, language and representative art are two sides of the same coin” [6]. Thus, in comparing Bronze Age paintings with Roman-era painting, for Blakolmer, the use of color in painting reflects the development of a vocabulary in the Greek language. By comparison, in Egyptian art, there are many colors used in painting that are not accounted for by any kind of vocabulary [11].


Precious materials played an important role in the development of abstract color terminology, usually in the form of loanwords during the period well after the earliest recorded languages. Warburton suggests that the materials dominated and that the concept of abstract color terminology never developed in the ancient Near East; yet the flow of loanwords meant that abstract terminology may have already been embryonic in the Aegean by ca. 1400 BC [11].

There is a clear difference between the descriptive uses of color in literature and economic texts. Although the sources for various languages are not equally balanced in terms of genre, it is significant that there is little evidence of color from the earliest administrative texts in proto-cuneiform whereas color terminology appears in the administrative texts written later in all other languages (where it also appears in other genres of text).

The gradual emergence of color as a means of categorization was thus a historical development that contributed to the emergence of abstraction in relatively recent historical times. In the art of the Paleolithic and Neolithic societies, few colors appear, whereas there is an abundance of color in the Near Eastern Bronze Age – well beyond what appears in evidence from language artifacts. By comparison, in the Greek world, the use of color in art and language marched hand in hand. It is clear that the Greeks acquired color concepts and terminology from the Near East through the diffusion of materials, ideas, and terminology. These in turn influenced the color lexicons of more recent languages.



  1. 1.
    Englund, R.K., Nissen, H.J.: Die Lexikalischen Listen der Archaischen Texte aus Uruk. Gebr Mann, Berlin (1993)Google Scholar
  2. 2.
    Landsberger, B.: Über Farben im Sumerisch-Akkadischen. J. Cuneif. Stud. 21, 139–173 (1967)CrossRefGoogle Scholar
  3. 3.
    Schenkel, W.: Die Farben in ägyptischer Kunst und Sprache. Z. Ägyptische Sprache 88, 131–147 (1963)Google Scholar
  4. 4.
    Schenkel, W.: Color terms in ancient Egyptian and Coptic. In: MacLaury, R.E., Paramei, G.V., Dedrick, D. (eds.) Anthropology of Color: Interdisciplinary Multilevel Modelling, pp. 211–228. Benjamins, Amsterdam (2007)CrossRefGoogle Scholar
  5. 5.
    Bulakh, M.: Basic color terms from proto-semitic to old ethiopic. In: MacLaury, R.E., Paramei, G.V., Dedrick, D. (eds.) Anthropology of Color: Interdisciplinary Multilevel Modelling, pp. 247–261. Benjamins, Amsterdam (2007)CrossRefGoogle Scholar
  6. 6.
    Blakolmer, F.: Zum Charakter der frühägäischen Farben: Linear B und Homer. In: Blakolmer, F. (ed.) Österreichische Forschungen zur Ägäischen Bronzezeit 1998: Akten der Tagung am Institut für Klassisiche Archäologie der Universität Wien 2.–3. Mai 1998, pp. 225–239. Wiener Forschungen zur Archäologie, Vienna (2000)Google Scholar
  7. 7.
    Xu, W.: A Study of Chinese Colour Terminology. Lincom, Munich (2007)Google Scholar
  8. 8.
    Liddel, H.G., Scott, R., Jones, H.S.: A Greek-English Lexicon. Oxford University Press, Oxford (1958)Google Scholar
  9. 9.
    Baines, J.: Color terminology and color classification: ancient Egyptian color terminology and polychromy. Am. Anthropol. 87, 240–262 (1985)CrossRefGoogle Scholar
  10. 10.
    Warburton, D.A.: The theoretical implications of ancient Egyptian colour vocabulary for anthropological and cognitive theory. Lingua Aegyptia 16, 213–259 (2008)Google Scholar
  11. 11.
    Warburton, D.A.: Colors in bronze age Egyptian art and language. In: Brinkmann, V., Primavesi, O., Hollein, M. (eds.) Circumlitio: The Polychromy of Antique and Mediaeval Sculpture, pp. 170–187. Liebighaus, Frankfurt (2010)Google Scholar
  12. 12.
    Berlin, B., Kay, P.: Basic Color Terms: Their Universality and Evolution. University of California Press, Berkeley (1969)Google Scholar

Copyright information

© Springer Science+Business Media New York 2016

Authors and Affiliations

  1. 1.TopoiBerlinGermany