Colour and Colour Terminology Author(s): N. B. McNeill Source: Journal of Linguistics, Vol. 8, No. 1 (Feb., 1972), pp. 21-33 Published by: Cambridge University Press Stable URL: http://www.jstor.org/stable/4175133 . Accessed: 10/08/2011 16:58 Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at . http://www.jstor.org/page/info/about/policies/terms.jsp JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship. For more information about JSTOR, please contact support@jstor.org. Cambridge University Press is collaborating with JSTOR to digitize, preserve and extend access to Journal of Linguistics. http://www.jstor.org JL8 (1972) i-i99 Printed in Great Britain Colour and colour terminology N. B. McNEILL PsycholinguisticLaboratories,The Universityof Chicago (Received 30 May I97I) The continuous gradation of colour which exists in nature is represented in languageby a series of discrete categories.1Athough there is no such thing as a naturaldivision of the spectrum, every languagehas colour words by which its speakerscategorizeand structurethe colour continuum. The number of colour wordsand the mannerin which differentlanguagesclassifythe colourcontinuum differ. Bassa, a languageof Liberia, has only two terms for classifyingcolours; hui and ziza (Gleason, I955: 5). Hui corresponds roughly to the cool end of the spectrum (black, violet, blue, and green) and ziza correspondsto the warm end of the spectrum (white, yellow, orange and red); in Bambara,one of the languagesof the Congo area,there are three fundamentalcolour words: dyema, blema and fima (Zahan, 195I: 52). Dyema includes white, beige, and natural (cotton) colour; blema denotes reddish, brownish shades; and finally fima includesdarkgreen,indigo and black.MaerzandPaul (I930) list over3000 colour names in English, but generallyit is consideredto have eight basic names: black, white, red, orange, yellow, green, blue and violet. The question of how many colours there are and how names are related to colourexperienceshas intriguedmany scholars;much researchhas been done on the question of the discrepancybetween colour perceptionand colour identification. Woodworthsummarizedthe past researchon this topic in 'The puzzle of color vocabularies'(i910) and drew two conclusions,one that there is an invariable sequence in the emergence of basic colour terms: 'Color nomenclature begins, almost always,with red, and spreadsto the other colors in spectralorder, usually, however, skipping such transitionalcolors as orange, blue-green and violet' (327) and two, that the emergence of a colour term is based on the functionalimportanceof the colour and its frequencyof usage: 'Wherethe need for designatinga certaincolor, or range of colors, is infrequent,the fluid condition of color designationwill be adequate'(328). By 'fluidcondition'Woodworth means the state where the colour of an objectis said to be like the colour of some other object but the basis of referenceis not yet fixed so that it varies from one comparisonto another.If on the other hand such a need is frequent,'fixedusage begins to emergewhich centersaboutsome one comparisonand all objectswhich [i] Thanks are due to Dr Eric P. Hamp and Dr Richard G. Stern of The University of Chicago for their advice and encouragement, and to my husband, David McNeill, for his criticisms during the preparation of this paper. The research work leading to this paper was supported in part by the United States Office of Education and the Early Education Research Center of The University of Chicago, 21 JOURNAL OF LINGUISTICS have a certaincolor are groupedtogether and consideredto have the color of one particularobject'(328). The name of the objectthen graduallybecomesseparated from the object and becomes the 'name' of a colour. The first of these conclusions has been developed by Berlin and Kay in their recent publicationon colour terms.2Among their findings are the following: i. An exhaustiveand universalinventory of i i basic colour categoriesexists for all humansfrom which the i i or fewer basic colourterms in any languageare always drawn. 2. The eleven basic colour categoriesare white,black,red,green,yellow, blue, brown,purple,pink, orangeand grey. 3. There is a temporal-evolutionarysequence, as follows: I Swhite II red LblackILyellow III* IV green yellow green V VI bluepurple ~ j VII rw1lpink orange Lgray (* At stage III either green or yellow emerges, and at stage IV whichever did not emerge at stage III emerges.) 4. A given languageat a given point of time can be in one and only one of the seven evolutionarystages and in any stage must have passed through all prior stages in the describedorder. The conclusionsI wish to draware (i) that in contrastto Berlinand Kay there is no universal sequence in which colour terms arise, because this order is determinedby the function of colour terms in a culture (the second of Woodworth'sconclusions);(2) that there is a kind of universalityin the humanperception of colour based on the physiologyof vision and it has at times played a role in the emergenceof colour terms. In dealingwith this problem,the view will be takenthat a set of colourwords is a system of structuringthe perceptionof nature. There are two majorfactors [i] Berlin & Kay, I969. The language types covered range from Jal, a Danian language of New Guinea which contains colour terms for only black and white, to languages like English which contains the full complement of eleven colour categories. Of the 98 languages, data were collected first hand on 20 languages from informants who resided in the San Francisco Bay Area. The materials on the other 78 languages presented have been drawn from published sources and personal communications with linguists and ethnographers who have specialized knowledge of the languages in question. In collecting their own data Berlin and Kay used a set of 320 Munsell colour chips (320 colour chips of 4o hues and 8 degrees of brightness, and 9 chips of neutral hues (white, black and greys) ). First the basic colour words of the languages in question were elicited from the informants. Then each subject was instructed to map both the focal point and the outer boundary of each of his colour terms on the array of colour stimuli. When the responses were studied and compared, a surprising uniformity occurred, as described above. 22 COLOUR AND COLOUR TERMINOLOGY involved in the evolution and use of colour terms: (i) the naturalresourcesand the availabilityof coloursin the externalworld, and (2) the physiologyof human vision. The interplay of these two factors can be shown for all languages, but for this study the linguistic communities of Navaho, Pukapukaand Ainu are selected as the focal languages.When need arises, various other languageswill be introduced. I. COLOUR TERMS AND NATURAL RESOURCES (i) The Navaho Berlin and Kay list the Navaho colour terms as follows: lagai lidzin lichi dotl'ish litso 'white' 'black' 'red' 'blue-green' 'yellow' (Since colournamesin variouslanguagesare not easilytranslatable,non-English colour names should ideally be expressedin terms of some non-linguisticsystem of notation; e.g., Munsell colour numbersor spectralwavelengths.However, for convenienceEnglish colourwords will be used with the understandingthat what is intended is the spectral loci of the range to which the English colour word refers and not the English colour name as such. When the colour term itself is intended, it will be italicized. This convention is used for both English and non-English colour terms.) Berlin and Kay then place Navaho in Stage IV, according to their evolutionary scheme, along with such other languages as Batak(Sumatra)and Eskimo(Canada).However, a study of the religiousrites of the Navaho will make clear the arbitrarinessof such placement. The above colour names were originally the names of the minerals and other earth substances commonlyused in the religiousceremonies.The name and focus of each colour categoryis thereforeuniquely that of the Navaho, and not the result of a universalevolution of colour terms. Among the Navaho rituals one of the most importantis the rite of divination. When a personbecomes ill and is not certainwhich of the many things he should or should not have done is responsiblefor the sickness, the cause is determined by the rite of divination.Besidesthe function of diagnosis,divinationis used also to prescribethe ritual treatmentof the disordersuch as the precise form of the ceremony,the time when it should be performed,and the selection of the right practitionerto performthe rite (Kluckhohn & Leighton, I946). The form most frequentlyused in the rite of divinationis 'hand-trembling'. The rite of hand-tremblingis of particularinterest from the point of view of colour study. Jet beads (black), white shells (white), turquoise (blue-green), pollen (yellow) and white clay (white) play a conspicuous role in this ritual. 23 JOURNAL OF LINGUISTICS According to Kluckhohn, this ritual involves the following prayer: 'Black Gila Monster, I want you to tell me what is wrong with this patient. Do not hide anything from me. I am giving you a jet bead to tell me what the illness of this patientis' (I48). The prayeris then repeatedfor each finger,substitutinga different colour of Gila Monster and correspondingbead for each one - a Blue Gila Monster and a turquoise bead for the index finger, a Yellow Gila Monster and a Haliotis shell bead for the middle and ring fingers, a White Gila Monster and a white shell bead for the little finger. Turquoise and white shells are seen in other contexts also. In every dwelling, buckskinpouches are preparedin which are found bits of turquoise and white shells along with pollen, herbs and tiny carved images of animals. In Navaho drypainting,charcoal,local minerals,and red and yellow ochre which are ground and placed in bark receptaclescompose the pigments. An interesting relationshipis found between the colour names and the substances used in the above ceremonials.In Navaho, white shell bead is yo'lagai (yo' meaning'bead'),turquoiseis dotl'ishor tl'if (H. Landar,personalcommunication), black ochre is 1ez or Iiiin (as le4in it designatesjet or black coal), red ochre is cih (ci.h also means to redden or dye red), and finally, yellow ochre is 1e-coi(Haile, I951). In other words the five colour names listed earlierand the names of the earth substancesused for ceremonialsare identical except for the omission of 1iin the case of red. Even in this case lichi and ci h can be considered identicalif we accept Durbin's view that the morpheme{i- (and its variousforms le, lee-, la-) is relatedto leesh,leezhmeaningdirt or earth(Durbin, 1970, in press). (Durbin states that the differencein phonetic form among those morphemesis due to phonetic assimilationto the following sounds.) These observationssuggest that the Navaho colour terms were the common names of availablenaturalresources,and support the conclusionthat the emergence of colour terms is culture specific. That this is the case is reflectedin the results obtained by Landar, Ervin and Horowitz in their psychologicalexperiment on Navaho colour categories (I960). They report that when the Navaho monolingualswere shown colour chips from the blue-green range, the answer was often 'bluish-greenish,the color of turquoise'. They report also that where English monolinguals distinguish two categories of blue and green, Navaho monolingualsproduced only one category of turquoise. This suggests not only that colour categoriesare structureddifferentlyfrom cultureto culture, but also that the names of colour categoriesand their foci differsince these have much to do with the availability of the colours and their function and frequency in practicallife. (2) The Pukapukans A similar relationshipbetween colour terms and natural resources appears in Pukapuka,the languageof three remote isles northeastof Samoa. 24 COLOUR AND COLOUR TERMINOLOGY Berlin and Kay provide the following four terms as the basic Pukapukan colour terms: kena uli kula yengayengaor yenga 'white' 'black' 'red' 'blue' or 'yellow' or mixture of two colours and categorizePukapukanas accordinglybelonging to Stage III on their evolutionaryscale. However,a study of the life of the Pukapukansprovidesan explanation for some of these colour names. Pukapukanslive in isolated conditionsuntouched by the culturaland technological developmentsmade in other parts of the world; they subsist on fish and plants (McGregor, I935). Among Pukapukans'most valued staple vegetable food is talo which is an indigenous plant of the South Pacific Islands and grows starchytubers. A full-growntuber is as largeand as heavy as a largesweet potato, and just underneaththe outer barkof this tuber is an inner layer, which may be rose (kula,ula), white (kina),darkgreenishor purplish(uli) in colour, depending on the variety (MacCaughey& Emerson, I9I3; Beaglehole& Beaglehole,I938). The names in parenthesesrepresent the variety of talo, and as we notice the names of talo and the colour names in Pukapukaare representedby the same terms. I have not been ableto find the sourceforyengayenga.However,according to MacCaugheyand Emerson, 'the interiorof the raw tuber is usually light grey with a slight bluish tinge, but in some varietiesmay be yellow, orange, or even purple' (227). The dual meaning of blue and yellow for yengayengamay, therefore, have something to do with the colour of the interiorof the raw talo tuber. (3) Japanese Since i86o when synthetic dyes based on such materialsas coal-tar,petroleum and others were invented, and the so-called basic colours were imported to Japan, Japanesehas had the following eleven basic colour words as Berlin and Kay state: shiro 'white' kuro 'black' 'red' aka midori 'green' ki 'yellow' ao 'blue' cha 'brown' murasaki 'purple' momo 'pink' daidai 'orange' hai, nezumi 'grey' 25 JOURNAL OF LINGUISTICS However, Berlin and Kay do not mention the traditionalsystem of colour names that has existed for centuriesand is still in use in many partsof Japan.There are three such basic colour names, aside from black (kuro)and white (shiro),and all threewere derivedfromindigenousplantsfromwhichnaturaldyes wereextracted (Uemura& Yamazaki,I943; Uemura,I943). colournames plants (source of dyes) akane(red) hanada(blue) kariyasu(yellow) akane(roots) hanada(entire plant) kariyasu(entire plant) The resulting colours have a characteristicfocus and range of shades which are uniquely Japanese. An examinationof these colours shows that what ancient Japaneseclassified as akane would be called orangeand what they classifiedas hanadawould be called turquoiseby English speakers.In other words, to ancient Japaneseorange is the representativeexample of the red range which centres around red in English, and turquoise is the representativeexample of the blue range which centres around English blue. A given wave of light has the same length to all humans, but the same wavelengthmay be categorizedas different colours in different cultures. Conversely, different wavelengths may become representativecolours of the same spectral range. Returning to the JapaneseEnglish examplesabove, the representativecolours of the red, yellow, and blue rangein English have wavelengthsof 66o mg, 585 my, and 485 mp, respectively (Woodworth & Schlosberg, I955: 387-397); and I estimate the representative colours of the red, yellow and blue range in ancient Japaneseto have had wavelengths of 6io mp, 570 mu and 490 m,u,respectively(based on informationin Uemura, I943, and Wada, 1954). This reflectsthe extent to which the available naturalresourcesaffectthe resulting relationshipbetween the colour names and their referents. The colour names themselves were originally names of actual objects which gradually became abstract when the usage was frequent and importantenough in the community,as is the case here. The tendency to create abstract colour names out of concrete objects is a common phenomenon of English; emerald,azure (lazuli), topaz from precious stones; rose,pink, lilac, mauve,lavenderfrom flowers;pomegranate,lemon,olive, cherry,plum, cerisefrom fruits; salmonfromfish; chartreuseand champagnefrom liquor. Among the most striking is the term crimsonwhich was derived from a worm. According to Maerz and Paul, crimson came from medieval Latin and for a long time it referredonly to a bluish-reddye obtainedfrom kermesinus, the Kermes insect. Kermes also was the source of the originalscarlet dye. The term scarlet came from Persian saqalat, siqalat, suglat, a red cloth dyed with Kermes. 'Kermes', a red dye, is produced by the female insect kermococcus. According to Laudermilk (I949), at some point in its development, after a moulting process, the female kermococcusturns into a sac in which the young 26 COLOUR AND COLOUR TERMINOLOGY (eggs) are the contents. Before the eggs are hatched, the insects are killed by vapoursof vinegarand a brilliantred dye is extractedfrom the fully grown eggs. The term 'kermes' meaning 'worm' exists in many languageswith variations: krmib(Sanskrit),kirm (Persian), kirmis(Lithuanian),crgivi(Old Slavic), cruim (Old Irish) (Meillet & Ernout, I967: 724). Some of what Berlinand Kay call BASIC colourterms such as orangeandpurple were also derivedfrom actualobjects,purplefrom a shell as discussedbelow and orangefrom a fruit. We should also note the fact that where need exists, a wide variety of colour names emerge. In English we have, for the reddish colour, red, rubicund,russet,auburn,crimson,carmine,scarlet, vermilion,maroonand many others each of which represents a colour. In languages such as Bassa all the colours mentioned above can be described by ziza. In other words, from the point of view of anotherlanguage,each languagemay be arbitraryin classifying colours; what is expressed by a single colour word in one language may be characterizedby a series of colour words in another. This is a common phenomenon in linguistic developmentin general, and it seems importantat this point to stress the fact that the principlesof classification are differentfrom languageto language.In Eskimo, for example, there are four differentwords for snow: aput for 'snow on the ground', qanafor 'fallingsnow', for 'a snowdrift'(Boas, I9II). English piqsirpoqfor 'driftingsnow', and qimuqsuq has only the one term, snow, which describes the four states. The automobile gives rise to the American English convertible,club coupe',sedan, hardtop,twodoor,four-door,sportscar, station-wagon,runabout,limousineand many others (Brown, I958). Some other languageshave only one term for these. In Hopi, all flying things except birds have a single name. In English, there is aviator,butterfly, and aeroplane;in Hopi, there is just masa'ytaka(Whorf, I956). According to Boas, this tendency of a languageto express a varietyof ideas by a single term is called 'holophrasis'. There areexamplesof holophrasisin the case of colour terms. Hanunoocolour terms provide a good example. In Hanunoo,the languageof a tribe on Mindoro Island in the Philippines,mararameans 'exhibitingredness' and malatuymeans 'exhibiting greenness' in some contexts, while in other contexts mararameans 'dryness or desiccation' and malatuy means 'wetness or succulence' (Conklin, I964). Purple,woadand safflowerin the ancientworld and ao in Japanesepresent the case where a single colour name designatesa wide varietyof differenthues. (i) Purple Originally,purple derivedfrom shells (Purpura)found on the coast of the eastern MediterraneanSea. The animals gatheredinto shoals in spring time; abrasion produced a milky white fluid from which purple dye was obtained. When the shells were broken,the white substanceoozed out. Upon exposureto the air and 227 JOURNAL OF LINGUISTICS light this substance passed through a series of colours: first citron-yellow,then greenish yellow, then green, and finally, purple or scarlet. The juice obtained from Murex brandaris,a kind of Purpura,changed photochemicallyinto a deep blue-violet, but that of Murex trunculus, another kind of Purpura, gave a scarletredhue (Forbes,1964: II4-I22; Gipper,I964). Mixingshellsin various ratios and stopping the photochemical process at different points produced yellow, blue, green, red and violet. Accordingto OED, in the middle agespurple appliedvaguelyto variousshades of red but now it appliesto 'mixtureof red and blue in various proportions'. The purple dye industry goes back into preclassical period. However, its heyday was reached during the classical period, and the Greeks appliedthe term -rropcpVpEoS to cover all these hues. Woad According to Forbes (I964: Iio), woad is an ancient source of blue which was known to both the Egyptians and the people of Mesopotamia.Woad is made from the leaves of Isatis tinctoria,a herbaceousbiennialwhose essentialconstituent, indigotin, is the same as that of indigo. When the Isatis leavesturnedyellow, they were groundto a smooth paste which was then formedinto oval balls. The balls turned darkblue (almost black) on the outside if they were exposed to the sun. If, on the other hand, they were stored in a closed place, they took on a yellowish hue which became particularlypronounced when the weather was rainy. Before the dyer used the material,these balls were ground to powder, moistened and fermented for several weeks. The dye then gave a strong and permanentblue. All three hues present in the differentstages were called by the same name, woad. (2) (3) Safflower The dye called carthameor safflowerwas made from Carthamustinctorius, a thistle-like herb, which yields large orange coloured flower heads. The flower heads are picked by hand and pressed into cakes. This flower contained two colouring substances: saffloweryellow which is a weak yellow and soluble, and carthamicacid which is red and insoluble. Different shades rangingfrom red to yellow could thus be attained, and both were called by the same colour name, saffloweror carthame(Forbes, I964: I22-I23). (4) Ao in Japanese Ao (blue) was derived from an indigenous plant, 'ai'. The colour obtainedfrom this plant ranges from indigo to bluish green. Midori (green) is also obtained from ai by adding a small amount of kariyasu (a yellow vegetable dye). In Japaneseall these colours are called ao (Uemura & Yamazaki,I943). The above studies show that the developmentof colour names and their foci 28 COLOUR AND COLOUR TERMINOLOGY are specific to the culture in which the name developed and have much to do with the available natural resources and the range of colours they provide. Therefore, they cannot be explainedin terms of a universalorder. II. COLOUR TERMS AND THE PHYSIOLOGY OF HUMAN VISION In the preceding section, I have shown how colour terms are related to the available natural resources. A similar close relationshipexists between colour terms and human colour perception and, as we will see, certain limitations inherent in human colour perceptioncreate an interestingtwist to the development of colour words. It is generallybelieved that the fact that a certaincolour name does not exist in a culturedoes not mean that its people are incapableof perceivingthat colour. Colour discriminationis probably the same for all human populations with healthy vision. To normal human vision, nature appearsin various degrees of brightness- ranging from black through a variety of shades of grey to white and in various coloured hues. Among the millions of hues that exist in nature, red, green, blue and yellow are consideredprimalcolours. Of these primalhues some are compatiblewith each other and may be sensed in the same place at the same time; others are incompatibleand will never be simultaneouslypresent in the colour of an object. For example, red can appear only in a mixture with yellow producing orange or in a mixture with blue producing violet, magenta, mauve, purple, etc.; it never appears in a mixture with green (Linskz, 1952: 52-I44, I964). There is no such hue as a greenish red or a reddish green, since in this mixture if red predominates,we obtain only unsaturatedshades of red, and if green predominates,we get unsaturatedshades of that same green. Any of the otherthree primarycolourswill also mix with only two of the otherthree blue with red or green, yellow with red or green and green with yellow or blue. However, we never sense a combinationof red and green or yellow and blue. In other words, red and green, blue and yellow have a strong relationshipas pairs of opposite colours. The same pairingof red and green, blue and yellow is reflectedin the human perception of colour in such phenomenaas induced colours, after-images,and colour blindness. Various experiments have shown that when one member of the pair is presented, the eye tends to call up the other member. For example, when one stares for a while at a red surfaceand then switches quickly to a white one, one will see green instead of white. A similareffect is found between yellow and blue. If we fixate our eyes on the sun and then switch to a white surface,we see a circle of 'robin's egg blue' (Goethe, I840; Krech, Crutchfield& Livson, I969: II4-I47). Combinationsof red-greenand blue-yellow are seen also in colour blindness. Among the symptoms of colour blindness, deuteranopiainvolves an inability to differentiatered and green and protanopiainvolves the same inability plus a 29 JOURNAL OF LINGUISTICS weakness in perceiving the red end of the spectrum-.Another form of colour blindness, tritanopia, involves an inab-ilityto perceive the difference between blue and yellow. All this shows that red and green on the one hand and blue and yellow on the other are physiologicallylinked in human perception.These pairs of colours are complementarycolours, and as Arnheim (I969) states, they represent'combinationsthat add up to the completewhole of whitenessor, when subtractedfrom each other, producethe complete nothingnessof black'. The innate tendency of human vision to combine red and green, or blue and yellow, is directly reflectedin the colour terms of some contemporarylanguages in the world. Clearexamplesare found in relativelyunacculturatedand isolated linguistic communities. Ainu, the languageof the indigenouscaucasoidpeople of Japan,has four basic colour terms (Chiri, I953). kunne 'black' retar 'white' hu 'red-green' siwnin 'blue-yellow' The terms siwnin and hu are of particularinterest to us. Accordingto Chiri, siwninmeans both 'yellow' and 'blue' as the following list shows: siwninus siwninrit siwninsame siwninonkerax siwninmarewrew siwninarake 'blue mildew' 'blue line' (veins as on the forehead) 'blue shark' 'yellow phlegm' 'yellow butterfly' 'jaundice' This phenomenon of categorizingblue and yellow together found in the Ainu language is seen in other linguistic communities also. According to Le Coeur (I956), in Daza, a Nilo-Saharanlanguageof east Nigeria, zedo means 'blue' and zede means 'yellow'. Also, zede to some informantssignified both bright yellow and violet. A word meaning both 'blue' and 'yellow' appears in a number of contemporarySlaviclanguages.In Serbo-Croatian,plavi meaning'blue' becomes 'blond' as it applies to human hair, and in Russian,polovyj refers to both blue and yellow (Vasmer, I955: 395). The same meaning is also shared by plowy in Polish(Bruckner, I957: 422) andplavyin Czech(Machek,I957: 372). Miklosich lists a Proto-Slavicterm polvu meaningboth 'blue' and 'yellow' from which the above terms in Slavoniclanguageswere derived. The languageof the Mechopdo Indians of Northern Californiahas a term epotiwhich means 'sky-blue','purple' and 'blue with a yellow tinge', and epotimpapaga means 'the yolk of an egg' (Gatschet, I879). In Chinese and Japanesethe character refersto the blue of the sky and the sea and also to a sallow complexion, in particularof old age 30 COLOUR AND COLOUR TERMINOLOGY (Morohashi, I960). In Latin, fiavus, meaning 'yellow' and 'blond' corresponds to bldo 'blue' in Old High German, bla 'yellow' in Medieval German and blau 'blue' in Modern German (Andre, I949; Walde & Hofmann, I938: 5I3). The close relationshipbetween red and green is also seen in many languagesof the world. In the case of red and green there is an added link between the two coloursin that they often representthe coloursof two differentstagesof the same plants or fruits. In the Ainu languagehu means both 'red' and 'green', and accordingto Chiri, hu meaning 'fresh' or 'green' appearsin such words as hu-ham'green leaves', hu-kina'greengrass',and hu-ni 'raw tree'. However,in other contexts hu means 'red' (cf. hu-turex'red fruit'). In Chineseand Japanese,the character meaning 'green' consists of E 'fresh' and Jf 'red' and indicates the colour of young (immature)plants and fruits (Morohashi, I960: 835-840). The fact that languagesas widely differentas Serbo-Croatian,the languageof the Mechopdo Indians, Ainu and Chinese show the same phenomena would lead us to believe that creatingone word meaningboth 'yellow' and 'blue', 'red' and 'green' is a reflectionof an innate characterof human colour perceptionand hence may be called a universal tendency of all humans. However, with the advent of general technological development, which includes the invention of chemicaldyes and the resultingnecessity to differentiatea large numberof hues of variousbrightnessand saturation,this universaltendency has left only a faint trace in the developmentof colour words. The invention of chemical dyes is very recent, historically,dating only from the nineteenth century. It is easy to overlookjust how highly disrupting the availabilityof these dyes has been for the traditionalcolourterminologies,which have been evolving for many centuries.Not only has there been a great need for new colour terms, but traditional terms have been recruited to denote new, artificially-madecolours. The original referents of these colour terms, which were the colours naturallyavailableto the culture, have generallybeen relegated to peripheralstatus, and in their place has appeareda vast new set of colours. Berlin and Kay state, 'It appearsnow that, although differentlanguagesencode in their vocabulariesdifferent number of basic color categories, a total universal inventory of exactly eleven basic color terms exists from which the eleven or fewer basic color terms of any given language are always drawn' (2), and it may indeed be that if we askcontemporaryspeakersof Japaneseto show us examplesof aka 'red', Germanspeakersexamplesof rot, Hopi speakersexamples of pala 'red', we will find a high degree of agreementin their responses.Even so, this universalityof categorieswhich Berlinand Kay believe they have demonstratedis itself an artificialresult of the invention of man-madedyes only a century ago. Because of the ready availabilityof these colours, the precision and cheapnessof their manufacture,plus a host of other economicfactors,man-made 3' JOURNAL OF LINGUISTICS chemicaldyes are in use virtuallyeverywhere.They thereforeprovide a basis for universal colour terms. The Berlin and Kay set of colours is thus the latest exampleof how abstractcolourtermsareinfluencedby the availabilityof concrete sources.Their theory, in other words, has not overturnedthe theory of linguistic relativity as they claimed. In the developmentof colour words as in any other linguistic creation, necessity and functional importance are the determining factors, and as Boas states, such developments seem to depend on the chief interest of a people. REFERENCES Andre, J. (1949). 18tudesur les termes de couleur dans la langue latine. Paris: Klincksieck. Arnheim, R. (I969). Art and visual perception. Berkeley: University of California Press. Beaglehole, E. & Beaglehole, P. (I938). Ethnology of Pukapuka (Bernice P. Bishop Museum Bulletin No. I50). Honolulu: Bernice P. Bishop Museum. Berlin, B. & Kay, P. (I969). Basic color terms: their universality and evolution. Berkeley: University of California Press. Boas, F. (9 i9I). On grammatical categories. In Hymes, D. (ed.), Language in culture and society. New York: Harper & Row. Brown, R. W. (1958). Words and things. New York: The Free Press. Bruckner, A. (I957). Slownik Etymologiczny Jfzyka Polskiego. Warszawa: Wiedza Powszechna. Chiri, M. (1953). Bunrui Ainu-go Jiten - Daiikkan Shokubutsu-hen (Dictionary of Ainu, Vol. i: Plants). Tokyo: Nihon Jomin Bunka Kenkyusho. Conklin, H. C. (1955). Hanuno6 color categories. In Hymes, D. (ed.), Language in culture and society. I89-I 92. New York: Harper & Row. Durbin, M. (I970). Review of Berlin & Kay, I969. To appear in Semiotica. Ernout, A. & Meillet, A. (I967). Dictionnaire etymologique de la langue latine: histoire des mots. 4th Ed. Paris: Klincksieck. Forbes, R. J. (I964). Studies in ancient technology. (Second Revised Edition), Vol. 4. Leiden: E. J. Brill. 99-127. Gatschet, A. S. (I879). Adjectives of color in Indian languages. The American naturalist 13, 8: 475-485. Gipper, H. (I964). Purpur. Glotta 42, 39-69. Gleason, H. A. (I955). An introduction to descriptive linguistics. New York: Holt, Rinehart & Winston. Goethe, J. W. von (I840). Theory of colours. (Eastlake, C. L.). London: John Murray. Haile, B. (I95i). A stem vocabulary of the Navaho language, Vol. z. St Michaels: St Michaels Press. Kluckhohm, C. & Leighton, D. (1946). The Navaho, Cambridge: Harvard University Press. Krech, D., Crutchfield, R. S. & Livson, N. (I969). Elements of psychology. New York: Alfred A. Knopf. Landar, H. J., Ervin, S. M. & Horowitz, A. E. (I960). Navaho color categories. Lg 36, 368-382. Laudermilk, J. (I949). The bug with a crimson past. Natural History (March 1949), I I4-I i 8. Le Coeur, C. (1956). Grammaire et textes Teda-Daza. Memoires de l'institut franfais d'Afriquenoire46, 259-394. Linskz, A. (I964). An essay on color vision and clinical color vision tests. New York: Grune & Stratton. Linskz, A. (1I952). Physiology of the eye, Vol. 2: Vision. New York: Grune & Stratton. 32 COLOUR AND COLOUR TERMINOLOGY MacCaughey, V. & Emerson, J. S. (I9I3). The Kalo in Hawaii. The Hawaiian forester and agriculturalist. 225-231, 280-288, 315-323, 349-358. McGregor, G. (I935). Notes on the ethnology of Pukapuka. Occasional papers (Bernice P. Bishop Museum of Polynesian Ethnology and Natural History), 9, 6: 3-52. Honolulu: Bernice P. Bishop Museum. Machek, V. (I957). Etymologicky slovnik -jazyka c'eskehoa slovenskeho. Praha. Maerz, A. & Paul, M. R. (I930). Dictionary of color. New York: McGraw Hill. Miklosich, F. (i886). Etymologisches W6rterbuch der Slavischen Sprachen. Wien: Wilhelm Braumruller. Morohashi, T. (i960). Daikanwa Jiten (Chinese-Japanese Dictionary), Vol. 9. Tokyo: Daishukan. Murray, J. A. H. (I909). A new English dictionary on historical principles. Oxford: The Clarendon Press. Uemura, R. (I1943). Manyosenshoku no Kenkyu (Studies in the Manyo Dyes). Tokyo: Kobunsha. Uemura, R. & Yamazaki, K. (I943). Nihon Shikimei Taikan (Dictionary of Japanese Colour Terms). Tokyo: Kodori Shorin. Vasmer, M. (I955). Russisches Etymologisches W6rterbuch, Vol. 2. Heidelberg: Winter. Wada, S. (954). Shikimei Daijiten (Dictionary of Colour). Tokyo: Sogensha. Walde, A. & Hofmann, J. B. (I938). Lateinisches Etymologisches W6rterbuch. Heidelberg: Winter. Whorf, B. L. (I956). Language, thought and reality. Cambridge: MIT Press. Woodworth, R. S. (I9g0). The puzzle of color vocabularies. The psychological bulletin, 7, I0. 325-334Woodworth, R. S. & Schlosberg, H. (I955). Experimental psychology. New York: Holt. Zahan, D. (195i). Les couleurs chez les Bambara du Soudan fran9ais. Notes Africaines (Bulletin d'Information et de Correspondance de l'Institut franQaisd'Afrique Noire), 50, 52-56. c 33