Since Berlin and Kay's now-classic study that suggested underlying universals in the way colors are categorized across cultures and languages, there have been hundreds of studies of color categorization in numerous languages and cultures all over the world. With few exceptions (e.g., data from one Korean speaker in Berlin & Kay's original survey; Caskey-Sirmons & Hickerson, 1977; Kim, 1985), however, there has been little organized study of the Korean color system to date. This is especially unfortunate because Korean has an extremely rich color naming system, including not only indigenous Korean terms, but also terms borrowed from Chinese, Japanese, and more recently, English.
        This paper, based on data from the author's doctoral dissertation (Tyson, 1994), reports results from an extensive study of color naming and color categorization in Korean. The major data for the study were elicited from forty-four native speakers of Korean living in Seoul ranging in age from 18-85 using Munsell color samples as stimulus materials. In a three-part interview procedure based on the work of MacLaury (1986, 1991, 1992), each person was asked to: (1) name 330 different loose chips to determine that person's individual vocabulary of color terms; (2) choose a focus for each term from an array of the same color chips; and (3) "map" the range of each term on the array. These data were supplemented by questionnaires, informal interviews, and other ethnographically-oriented information. It was found that individual speakers often varied greatly both in the specific terms they used and in the number of categories they seemed to be using, with age, sex, education, occupation, and attitude toward the topic among the possible influences. Younger speakers (ages 18-24) as a group, however, were found to be much more consistent than older speakers (ages 47-85) in both the terms and categories they used. Another specific finding was that many of these Korean speakers seemed to have distinct color categories in addition to the eleven categories that Berlin and Kay (1969) suggested were "basic" in Korean, including LIGHT BLUE, DARK BLUE, YELLOW-GREEN and several others, which is similar to the situation that Stanlaw (1987) found in Japanese.

        In contrast to the relativistic notion that different languages divide the color spectrum into arbitrary categories (e.g., Gleason, 1961, p. 4), Berlin and Kay (1969) proposed in their now-classic study of color naming by speakers of ninety-eight different languages that there are only a limited number of possible types of color terminologies in the world, which could be categorized according to "basic color terms." Although the boundaries of color categories were found to be inconsistent even for the same speaker at different times in Berlin and Kay's study, the "focal colors" (or the best examples of color categories) were found to be remarkably consistent across speakers and languages. A basic color term was defined by Berlin and Kay as one which was monomorphemic, generally known, commonly used, not a recent loanword, and not restricted to a small number of objects. Berlin and Kay suggested that the color categories that basic color terms could name were those referred to in English as black, white, red, yellow, green, blue, brown, purple, pink, orange, and grey. Some languages, like English, have basic terms for all of these categories, while others, like Dani, a language spoken in New Guinea, use as few as two--black and white (Rosch, 1972).
        Moreover, Berlin and Kay suggested that the "evolution" of basic color terms in a language follows a partially fixed order, so that if a language has terms for fewer than eleven basic color categories, "there are strict limitations on which categories it may encode" (p. 2). They proposed that all languages could be placed on the following implicational universal hierarchy, with the Roman numerals indicating they "evolutionary" stage of the language:

I. BLACK and WHITE
II. BLACK, WHITE, and RED
IIIa. BLACK, WHITE, RED, and GREEN, or
IIIb. BLACK, WHITE, RED, and YELLOW, or
IV. BLACK, WHITE, RED, YELLOW, and GREEN
V. BLACK, WHITE, RED, YELLOW, GREEN, and BLUE
VI. BLACK, WHITE, RED, YELLOW, GREEN, BLUE, and BROWN
VII. BLACK, WHITE, RED, YELLOW, GREEN, BLUE, BROWN, and any or all of ORANGE, PINK, PURPLE, and GREY

2. Research Methodology and Objectives

        The major data for this study were elicited from forty-four native speakers of Korean living in Seoul ranging in age from 18 to 85 using Munsell² color samples as stimulus materials. In a three-part formal interview procedure based on the work of MacLaury (1986, 1991a, 1991b, 1992), each person was asked to: (1) name 330 different loose color chips to determine that person's individual vocabulary of color terms; (2) choose a focus for each term from an array of the same color chips; and (3) "map" the range of each term on the array.³ These data were supplemented by questionnaires, informal interviews, and other ethnographically-oriented information.
        Several studies have attempted to establish lists of presumed "basic" Korean color terms. Some researchers (A. I. Kim, 1985; P. H. Kim, 1993) have simply relied on their own intuitions as native speakers to suggest what these terms might be, while others (Caskey-Sirmons & Hickerson, 1977), following the example of Berlin and Kay (1969), have based their suggestions on the results of saliency listing tasks in which native Korean speakers were asked to make lists of all the color terms that came to mind in a short time. Terms that occurred highest on the list were assumed to be the most salient, and therefore, the most "basic."
        This paper focuses mainly on the data from the first section of the formal interviews described above, that is, the chip naming section, to attempt to provide deeper insights into the color naming system in Korean. The use of this more sophisticated procedure (cf. Berlin, Kay, & Merrifield, 1991; Kay, Berlin, & Merrifield, 1991; McLaury, 1986) shows that native speaker intuition and saliency tests are not sensitive enough indicators of the "basicness" of terms in the complex Korean color system. Instead, it was found that individual speakers in the study varied greatly both in the specific terms they used and in the number of categories they seemed to be using. In addition, some speakers appeared to have distinct color categories in addition to the eleven "basic" categories suggested by some earlier studies.
 

3. Previous Studies of Korean Color Naming

        Despite the fact that Korean is known to have an extremely rich color naming system, including not only indigenous Korean terms, but also a large number of terms borrowed from Chinese, Japanese, and more recently, English, there has been relatively little organized research in Korean color naming and categorization. However, a few studies have atttempted to identify "basic" color terms and/or categories in Korean. First Berlin and Kay (1969) classified Korean as a Stage VII language. Based on data obtained through a saliency listing task from a single Korean consultant living in the San Francisco area, they suggested that Korean had twelve basic terms for eleven different color categories (pp. 96 & 124):

1. hayahta⁴  'white'
2. kkamahta  'black'
3. ppalkahta  'red'
4. phalahta  'green'
4a. noksayk  'green'
5. nolahta  'yellow'
6. chengsayk⁵  'blue'
7. kalsayk  'brown'
8. casayk  'purple'
9. punhongsayk pink
10. tungsayk  'orange'
11. hoysayk  'grey'

1. kamansayk  'black'
2. hayansayk  'white'
3. palkansayk  'red'
4. nolansayk  'yellow'
5. choloksayk  'green'
6. phalansayk  'blue'
7. kalsayk  'brown'
8. polasayk  'purple'
9. punhongsayk  'pink'
10. cwuhongsayk  'orange'
11. hoysayk  'grey'

        A. I. Kim (1985) challenged Berlin and Kay's classification of Korean as a Stage VII language consisting of terms for eleven basic colors. He argued that the Korean color system is actually composed of only five basic color terms (i.e., the five indigenous Korean terms kkamaht, hayahta, ppalkahta, nolahta, and phalahta), and therefore, Korean should be classified as a Stage IV language. Kim based his argument on certain characteristics of the Korean color system that Berlin and Kay had overlooked, including the inflected nature of the indigenous terms as opposed to the uninflected nature of the Chinese loans listed by Berlin and Kay for the categories BLUE, BROWN, PURPLE, PINK, ORANGE, and GREY and the phenomenon of a GRUE (or GREEN + BLUE) category. Kim also discussed two kinds of phonetic symbolism common in the use of the indigenous terms (i.e., vowel and consonant), which he saw as a challenge to Berlin and Kay's suggestion of a correlation between the number of color terms in a language and cultural complexity. Kim argued that the five basic terms in Korean "allow a greater diversity of color change by applying...morphological rules which are in existence in that language" (p. 134), which may have allowed the Korean color system to meet the needs of increasing cultural complexity without resorting to adding new basic terms.
        In an article comparing certain elements of the Korean and Japanese color systems, P. H. Kim (1993) suggested the following list of Korean terms for Berlin and Kay's eleven basic color categories (p. 53). This list represents only Kim's own native intuitions:

1. kkamang  'black'
2. hayang  'white'
3. ppalkang  'red'
4. nolang  'yellow'
5. cholok  'green'
6. phalang  'blue'
7. kalsayk  'brown'
8. cinhongsayk  'purple'
9. punhongsayk  'pink'
10. cwuhwangsayk  'orange'
11. hoysayk  'grey'

4. Results of the Chip-naming Procedure

        For comparison purposes, the forty-four Korean speakers who participated in the formal color interviews described above were divided into two groups: (1) a "college student" group consisting of twenty-one consultants between the ages of 18 and 24 and (2) an "older" group consisting of twenty-three consultants ranging in age from 47 to 85. These two groups were subdivided by sex. The results of the first part of the interview procedure, that is, the chip-naming experiment, are presented in this section.
 

4.1 College Students

        The college students (ages 18-24) tended to use a very high number of different terms in naming the 330 loose chips, with the females often using thirty to forty or even more different terms. Males in this group also used more terms than their older counterparts, with two using twenty-nine terms each, for instance. The consultants in this age group also were more consistent in the term and/or category names they used, making a fairly clear distinction between categories as a group.
        For instance, they consistently used forms of indigenous Korean color terms to name the most basic categories (i.e., BLACK, WHITE, RED, YELLOW, and BLUE), and they agreed strongly on which Sino-Korean constructions were appropriate for naming secondary colors (e.g., chenglok 'blue-green' and not nokcheng 'green-blue'). Most of the Sino-Korean constructions they used were those that also appear in dictionaries (e.g., cwuhwang, punhong, cacwu, hwangtho, etc.), which was not always the case for older consultants. In addition, all of them made a strong distinction between the GREEN and BLUE categories, using nok and/or cholok for 'green' and usually phalan for 'blue'. No one in this age group used only one term to name all of GREEN and BLUE, nor did any of them use the terms phalan or cheng 'blue' consistently to name chips in the GREEN area. Finally, although they used object names as color terms very often, none of them used object names consistently to name large areas of the array. Instead, when they used these terms, they were more likely to identify a certain chip as being "special" in some way. For example, when one consultant was shown a bright yellow chip, she said, "When young girls wear this color in the spring, we call it pyengalisayk ('chick color')." But she named only one chip in the YELLOW area of the array with that term, using nolansayk, as expected, for all of the others.
 

4.2 Older Consultants

        The older consultants (ages 47-85) were far less consistent in their naming as a group than the younger consultants were. They tended to use fewer different terms in naming the 330 chips than the younger group did--usually about twenty to thirty different terms, and sometimes even fewer. Like the younger group, however, females generally used more than the males; one 55-year-old male, for instance, used only eighteen different terms, the least among all of the consultants. Correspondingly, the area of color space covered by individual terms tended to be larger for the older group. The consultants in this group were also generally less consistent in the term names they used for individual chips, as well as the names they seemed to be using for various categories. In fact, in four cases, twice for females and twice for males, all of the consultants of the same sex actually gave a different name to the same chip. For example, the eleven male consultants named a chip in the reddish-brown area of the array, respectively, pamsayk 'chestnut color', hwangsayk 'yellow', hulksayk 'earth color', cwuhongsayk 'orange', oleyncisayk 'orange', pyektolsayk 'brick color', kamsayk 'persimmon color', kotongsayk 'old copper color', thosayk 'earth color', ppalkangsayk 'red', and kalsayk 'brown'.
        Although these consultants also most often used indigenous Korean terms for the five most basic categories, they were more likely than the younger consultants to use Chinese loanwords for basic categories (i.e., huksayk for BLACK, payksayk for WHITE, hongsayk, ceksayk, or cwusayk for RED, hwangsayk for YELLOW, and chengsayk for GREEN/BLUE (or GRUE)). For example, the oldest consultant, an 85-year-old female, used only hongsayk to name red chips, and a 78-year-old male used Chinese loanwords not only for all of the white, red, and yellow chips, but was also the only one of the forty-four consultants to use casayk (the term suggested by Berlin and Kay, 1969) as his basic term for the category PURPLE.
        In both of these cases, it is likely that education played a role. The woman, a graduate of a college in North Korea, and the man, a graduate of a technical college in Japan, must both be considered extremely well-educated for Koreans in their age group. They received their entire education during the Japanese colonial period when use of the (Japanese form of) Chinese loans would have been mandatory. Ages also is no doubt another important factor, however. One 71-year-old male, for instance, also used a large number of Chinese loans for the most basic categories although he had no formal education at all. On the other hand, the 51-year-old daughter of the older woman mentioned above, also a college graduate, used no Chinese loans in place of indigenous Korean color terms, but did use three English loanwords, oleyncisayk 'orange', peycisayk 'beige', and aipolisayk 'ivory', to name a number of chips. She also used the term solasayk 'light blue', a Japanese loanword, to name an extensive area of the array (i.e., fifty-seven chips in all). She would have been a child at the end of the Japanese colonial period in 1945, and therefore, received virtually all of her education in Korean, along with mandatory classes in English at school.
        In addition, the older consultants made a much weaker distinction between the GREEN and BLUE categories than the younger consultants did, strongly suggesting the remnants of a recent GRUE category in Korean. Their naming ranges show that virtually all of them at the very least mixed terms together that the younger consultants divided very distinctly into the two categories. Several, in fact, seem to have named a definite GRUE category, with four consultants using the term phalangsayk and one using the term chengsayk to name an area of the array that includes many of the greens and blues. One 80-year-old male, in fact, used the term choloksayk ‘green’ to name only one chip; he named everything else that the younger consultants usually named with that term or with noksayk 'green' with forms of the verb phalahta--phalansayk and phulunsayk (presumably 'grue'). Other consultants mixed phalansayk and chengsayk with the terms choloksayk and/or noksayk in a way that suggested that they too were using those terms as GRUE. Finally, the 78-year-old male who received three years of education in Japan used the Japanese loanword konsayk 'dark blue' to name most of the chips ranging from the dark green to the green and blue chips. These naming patterns support the claim that speakers of a language can maintain a composite category, such as GRUE, even after adding terms for other less basic categories, since all of these consultants also used terms for (one or more) of the categories BROWN, PURPLE, ORANGE, PINK, and GREY.
        Furthermore, the older consultants were more likely to use object names to name any color--even for almost entire basic categories. A 47-year-old female, for instance, used the term kyecasayk 'mustard color' to name most of the yellow chips without using the term nolansayk, as all of the other consultants did, even once. Similarly, three other female consultants named all or most of the PURPLE category with the name of a tree, otongsayk 'paulownia tree color', or of a flower, tolacisayk 'Chinese bellflower color'. In addition, five of the older females name a large number of chips with names of vegetables, fruits, and nuts. Particularly popular terms were paychwusayk 'cabbage color' and unhaynsayk 'gingko nut color' for light green chips and swupaksayk 'watermelon color' for dark green chips. Perhaps not surprisingly, this tendency to make heavy use of vegetable names as color terms appears to be closely related to sex, occupation, and educational level; all of the consultants just mentioned were females who identified themselves as housewives (and in one case also as a farmer), and none had a high level of education. Two had finished middle school, one elementary school, and the other two had had no formal education whatsoever. Although males, younger consultants, and more highly-educated older females also used vegetable names as color terms, none of them did so nearly as frequently as the consultants just mentioned.
 

4.3 Female/Male Differences in Naming

        Overall, the females in both age groups used more terms for more categories and agreed more often on term names for specific chips, as well as names and numbers of categories, than did their male counterparts. In other words, they seemed to focus more strongly on the differences between colors and categories than males did, resulting generally in smaller and more clearly defined naming ranges for individual terms. One younger female, for example, broke up the red area of the array, which was usually named with just one term by males, into four distinct ranges. She named five chips near the usual focus for red with the indigenous term ppalkangsayk, five slightly brighter chips tahongsayk 'very red', three adjacent chips closer to red-purple yencisayk '(cheek) rouge color', and three chips closer to pink hongmaysayk 'red plum blossom color'.
        In addition, nearly all of the female consultants in the older group consistently named a rather large area in the light green and/or light blue areas of the array with one of two terms--oksayk 'jade' or the Japanese loanword solasayk 'light blue'. None of the men named a large area with either of those terms. In fact, the term oksayk was used by only a few males to name a very few chips, and solasayk was not used even once by any male in either group. College-age females did use the term oksayk occasionally, but not to name as extensive a range of chips as the older females. One younger female, for example, named only one chip with that term, saying, "Well, my grandmother would call that oksayk." Another college-age female avoided use of the terms oksayk and solasayk entirely, but named an area of the array ranging from light green to light blue-green with the English loanword eymeylaltusayk 'emerald color'. It seems that the two terms oksayk and solasayk, then, are examples of Korean color terms that currently are mainly used by and strongly associated with females of a certain age.
        Many males, especially those in the older group, often began the interviews by saying something like, "I really don't know much about color" or "I don't know many color words." They often suggested that I ask women instead because, as one 58-year-old male put it, "Women have experience with [the colors associated with] hanpok ('traditional Korean clothes')." Many women, in both the older and younger groups, seemed to share this opinion and, in fact, often had a much easier time naming colors during the interviews and also seemed to enjoy the interviews more. Males in both age groups left more chips unnamed than the females, especially in the brown and purple areas of the array, and generally named fewer distinct categories than females of the same age. One 65-year-old male consultant left eighty-one chips unnamed, nearly one-fourth of the entire array. As expected, he left many chips in the brown and purple areas unnamed, but it is especially interesting that the largest area of the array he could not name was the light green and light blue area that the older females called oksayk 'jade color' or solasayk 'light blue'.
        Finally, males also made more "errors" in naming than females did; that is, they more often named chips with unexpected or very unusual terms. For example, one younger male consultant named several purple chips with the term yentwusayk 'yellow-green' and several light green chips with the term polasayk 'purple', even though he used both terms a number of times with their expected meanings. A 65-year-old male named a number of light purple chips with the term oksayk 'jade color'. When I asked him to choose a focus for that color in the second stage of the interview, however, he admitted that he did not know the meaning of that term. He had simply guessed that it might be appropriate to name the light purple chips he did not have another name for because he knew the term was used as a color word. It seems that for many of the male consultants, and for those in the older group especially, color was a field of expertise considered most appropriate for women. This attitude resulted in more difficulty and less interest in naming colors, more chips left unnamed or carelessly named, fewer terms used in naming, and fewer and larger categories overall.
 

4.4 Discussion of the Chip-naming Results

        In the chip-naming section of the interviews, the forty-four consultants used a very large number of different terms and used multiple terms to name chips that presumably fall into the same category. They most commonly used indigenous Korean color terms to name the most basic categories, BLACK, WHITE, RED, YELLOW, and BLUE (or GRUE), and mostly Sino-Korean terms to name the other basic categories. Younger consultants made a very strong distinction between GRUE and BLUE, although overlaps in terms used to name those two categories were common among the older consultants, with a few apparently maintaining a true GRUE category. Rather than making use of the phonetic symbolism potential of the indigenous terms to name various shades of color (as might be suggested by A. I. Kim, 1985), the consultants much more commonly used different terms entirely (e.g., cwuhwang, cwuhong, kywul, and/or oleynci for different shades of ORANGE), combinations of Chinese characters (e.g., chenglok 'blue-green', hwangkal 'yellow-brown', or hoypayk 'grey-white'), and adjectives and other modifiers, including yen-, yenhan-, yethun-, yelpun-, or yelin- for 'light' colors, cin-, cinhan-, cithun-, thakhan-, or etwuwun- for 'dark' colors, etc.
        The main results of the chip-naming procedure for each of the four groups are quantified in Tables 1 and 2. Table 1 shows the color terms for chips that at least half of the consultants in the respective groups agreed upon. These figures clearly show some of the major trends that have already been discussed, as well as some possible differences in term preference for the same category by consultants of different age and sex. Males in general, for instance, showed a preference for the term nok over cholok for 'green', and older males preferred the term cheng over phalan for 'blue'. Table 2 shows the naming of apparent categories for each of the four groups at a 70% or higher level of agreement. These figures begin to suggest the types and number of categories that might be used in Korean for speakers of different age and sex. Once again, for example, in most cases it appears that younger speakers tend to agree more strongly on which chips belong in a specific category as well as what the names of those chips should be. Also, younger females appear to be making the strongest distinction between categories, agreeing on the largest number of categories, while older males seem to agree on the fewest number of categories.
        The terms used during the chip-naming experiment suggest that there are definitely commonly used Korean terms for each of Berlin and Kay's eleven basic categories, although consultants in different age groups sometimes preferred different terms, and even members of the same group often used a variety of terms for chips presumably in the same category. Also, some consultants did not use terms for all eleven of the categories. Several older males, in particular, seemed to lack basic terms for BROWN, PURPLE, and/or ORANGE. Also, as a group, older females did not seem to agree strongly on any single term (or even any combination of terms) for naming either the GREEN or BLUE category. On the other hand, Table 1 shows that many consultants in both age groups used some terms that do not name one of Berlin and Kay's eleven basic categories in such a way as to suggest the possiblity of additional categories for some speakers of Korean. Table 2, perhaps, suggests roughly what those categories might be: LIGHT BLUE, YELLOW-GREEN, and PURPLE-RED for younger consultants and older females, LIGHT PINK for younger consultants, DARK BLUE for females regardless of age, DARK YELLOW and BLUE-GREEN for younger females only, and LIGHT GREEN/BLUE for older females only.
 
 

Table 1.  Color term naming by group--50% agreement or higher (for at least one chip on the array)
 

Color Term	College Females	College Males	Older Females	Older Males
ppalkansayk 'red'	+	+	+	+
nolansayk 'yellow'	+	+	+	+
punhongsayk 'pink'	+	+	+	+
hoysayk 'grey'	+	+	+	+
polasayk 'purple'	+	+	+	-
cwuhwangsayk 'orange'	+	+	+	-
yentwusayk 'yellow-green'*	+	+	+	-
cacwusayk 'purple-red'*	+	+	+	-
hanulsayk 'light blue (sky)'*	+	+	+	-
salsayk 'light pink (flesh)'*	+	+	+	-
noksayk 'green'	+	+	-	+
choloksayk '(grass) green'	+	+	-	+
phalansayk 'blue (green)'	+	+	-	-
kalsayk 'brown'	+	+	-	-
huynsayk 'white'	+	-	+	-
hwangthosayk 'yellow earth (ocher)'*	+	-	-	-
chengloksayk 'blue-green'*	+	-	-	-
tahongsayk 'very (red)'*	+	-	-	-
patasayk 'ocean color'*	+	-	-	-
phulsayk 'grass color'*	+	-	-	-
sswuksayk 'mugwart color'*	+	-	-	-
kotongsayk 'brown (old copper)'	-	+	-	-
kemceng 'black'	-	-	+	-
namsayk 'dark blue'*	-	-	+	-
oksayk 'jade color'*	-	-	+	-
pukchengsayk 'dark green'*	-	-	+	-
kwukpangsayk 'military defense color'*	-	-	+	-
chengsayk 'blue (grue)'	-	-	-	+
Total number of terms6 =	21	15	16	7

_______________
        *Indicates a term that does not name one of Berlin & Kay's (1969) eleven basic categories.
 
 

Table 2. Apparent category naming by group--70% agreement or higher (for at least one chip on the array)
 

Color Category	College Females	College Males	Older Females	Older Males
BLACK	+	+	+	+
WHITE	+	+	+	+
RED	+	+	+	+
YELLOW	+	+	+	+
GREEN	+	+	+	+
BLUE	+	+	+	+
PURPLE	+	+	+	+
PINK	+	+	+	+
GREY	+	+	+	+
ORANGE	+	+	+	-
LIGHT BLUE (SKY)*	+	+	+	-
YELLOW-GREEN*	+	+	+	-
PURPLE-RED (MAROON)*	+	+	+	-
BROWN	+	+	-	-
LIGHT PINK (FLESH)*	+	+	-	-
DARK BLUE*	-	+	-	-
DARK YELLOW (OCHER)*	+	-	-	-
BLUE-GREEN*	+	-	-	-
LIGHT GREEN/BLUE (JADE)*	-	-	+	-
Total number of apparent categories =	18	15	15	9

_______________
        *Indicates a "category" not recognized by Berlin and Kay (1969) as basic.
 

5. Conclusion

        This study, above all, has highlighted the extremely rich and complex color naming system available in Korean. While failing to define absolute "basic" color terms in the original Berlin and Kay (1969) sense, it has identified some of the most common color terms (and possible color categories) used by Korean speakers at the present time and shows that there is a certain degree of consistency in their use. In that sense, of course, these data support the idea that color naming is definitely not an arbitrary and totally individual pursuit. There was no indication that Korea contradicts the notion that there are certain universals underlying color naming practices in the world's languages (cf. Kay, Berlin, & Merrifield, 1991). It was also found, however, that individual consultants in the study often varied greatly both in the specific terms and number of terms that they used as well as in the number of categories they seemed to be naming. This was true even for members of the same family. This finding lends strong support to the view that color naming is not only "constrained by physiology" but also "guided by social values" (MacLaury, 1991b, p. 34), allowing for a great deal of individual variation in color naming and categorization.
        This study suggests that the following, at least, were other possible (and not doubt interrelated) influences on the color naming systems of the Korean speakers interviewed: (1) Age: Younger speakers, for example, generally used more terms and categories overall and were much more consistent in their choice of terms than were their older counterparts. (2) Sex: Female consultants tended to use more terms to name more categories than males in the same age group, while older women agreed very strongly on a LIGHT GREEN/BLUE category named with the term oksayk 'jade color', for instance. (3) Education: Highly-educated people who were educated during the Japanes colonial period (1910-1945) were more likely to use Chinese loanwords for the most basic categories, and the older female consultants who identified themselves as poorly-educated tended to use a very large number of vegetable names and other object names as color terms. Similarly, it seems likely that the relatively consistent preference for indigenous Korean terms for naming the most basic categories by the younger consultants was at least partially a result of more recent educational policies. (4) Occupation: Older females who identified their occupation as “housewife” (as well as one who gave her occupation as "housewife and farmer") sometimes used the names of vegetables and food to name relatively large sections of the array. They often named dark green chips swupaksayk 'watermelon color' and lighter green chips paychwusayk 'cabbage color', for example, and one named the entire YELLOW category kyecasayk 'mustard color'. These terms were rarely used by other consultants. (5) Attitude toward the topic: Male consultants in general, and males in the older group in particular, seemed to have less interest in color than the females did. They left more chips totally unnamed, and sometimes complained that they did not know about color words simply because of their sex; perhaps color is a field more closely related to "femininity" for many Koreans.
        Finally, another specific finding of this study was that many of the consultants made extensive use of terms that seemed to name distinct categories in addition to the eleven that have usually been accepted as "basic" in Korean (cf. Berlin & Kay, 1969; Caskey-Sirmons & Hickerson, 1977; P. H. Kim, 1993), with some speakers (especially older males) possibly using as few as seven or eight and others (especially younger females) possibly using as many as seventeen or eighteen. Although many of these categories have not been substantiated in prior studies of other languages, MacLaury (1986) states: "Kay and MacDaniel...suggest that [a language] might allow additional basic categories, such as chartreuse, maroon, and turquoise, which so far are unobserved; there is no specification of an ultimate limit upon the number of basic categories with which humans can distinguish color" (p. 39). Three of these possible additional categories in Korean include LIGHT BLUE, DARK BLUE, and YELLOW-GREEE, which are categories Stanlaw (1987) suggests Japanese also "has or is in the process of developing" (p. 188). Other possible categories include PURPLE-RED (or MAROON), DARK YELLOW (or OCHER), LIGHT PINK (or FLESH), BLUE-GREEN, and LIGHT GREEN/BLUE (or JADE).
 

Notes

   ¹This study is based on data collected in Seoul for my doctoral dissertation (Tyson, 1994) submitted to the University of Arizona in December 1994. I thank the members of my dissertation committee, Rudolph Troike, Muriel Saville-Troike, and Jane Hill, for their valuable comments and suggestions on the dissertation which have carried over into this paper. I am also especially grateful to Robert MacLaury for kindly lending me the Munsell color materials used in this study.
   ²The Munsell system represents color as a "solid" forming a "lopsided and partially flattened globe" (D'Andrade & Egan, 1974, p. 52). Areas in the solid are identified by numbers corresponding to the circular dimension of hue, the vertical dimension of value (or brightness), and the horizontal dimension of chroma (or saturation). D'Andrade and Egan (1974) and Stanlaw (1987, pp. 492-494) provide explanations and black and white drawings of the Munsell color solid, and Webster's Third New International Dictionary (1981) describes the system in detail (p. 448) and provides color drawings on two color plates (located between p. 448 & p. 449). MacLaury (1986, p. 20, 1992, p. 138) outlines the history of the use of Munsell colors by ethnographers studying color categorization in various languages.
   ³See Tyson (1994) for a more complete description of the interview techniques and consultants (pp. 315-317), and complete data from each of the three sections of the interviws for all forty-four consultants (pp. 321-433).
   ⁴The Yale system of Romanization is used throughout this paper (see Martin & Lee, 1986).
   ⁵Berlin and Kay (1969) actually listed the nonexistent term *changsayk as the Korean term for 'blue' (pp. 40, 96, & 124). I have assumed, as did A. I. Kim (1985), that the intended term is chengsayk 'blue; green (or grue)'.
   ⁶The figures in Table 1 are slightly distorted by the fact that not all groups agreed on common terms for black and white although, of course, every individual used at least one term for each.
 

References

Berlin, B. and E. A. Berlin. 1975. Aguarun Color Categories. American Ethnologist 2:1.61-87.

Berlin, B. and P. Kay. 1969. Basic Color Terms: Their Universality and Evolution. Berkeley and Los Angeles: University of California Press.

Berlin, B., P. Kay, and W. R. Merrifield. 1991. The World Color Survey. Dallas: Academic Publications of the Summer Institute of Linguistics.

Caskey-Sirmons, L. A., and N. P. Hickerson, 1977. Semantic Shift and Bilinguals: Variation in the Color Terms of Five Languages. Anthropological Linguistics 19:8.358-367.

D’Andrade, R. D. and M. Egan. 1974. The Colors of Emotion. American Ethnologist 1.49-63.

Gleason, H. A., Jr. 1961. An Introduction to Descriptive Linguistics. New York: Holt, Rinehart and Winston.

Kay, P. 1975. Synchronic Variability and Diachronic Change in Basic Color Terms. Language and Society 4.257-270.

Kay, P., B. Berlin and W. R. Merrifield. 1991. Biocultural Implications of the Systems of Color Naming. Journal of Linguistic Anthropology 1:1.12-25.

Kim, A. I. 1985. Korean Color Terms: An Aspect of Semantic Fields and Related Phenomena. Anthropological Linguistics 27:4.425-436.

Kim, P. H. 1993. <Pulkun Cawi> Ko (Thoughts on "Red Whites"). Language Research 15.45-60.

MacLaury, R. E. 1986. Color in Mesoamerica, Vol. I: A Theory of Composite Categorization. Doctoral dissertation, University of California, Berkeley.

MacLaury, R. E. 1987a. Coextensive Semantic Ranges: Different Names for Distinct Vantages of One Category. Papers from the Annual Meeting of the Chicago Linguistics Society 23: Part I.268-282.

MacLaury, R. E. 1987b. Color-Category Evolution and Shuswap Yellow-with-Green. American Anthropologist 89:1.107-124.

MacLaury, R. E. 1991a. Exotic Color Categories: Linguistic Relativity to What Extent? Journal of Linguistic Anthropology 1:1.26-51.

MacLaury, R. E. 1991b. Social and Cognitive Motivations of Change: Measuring Variability in Color Semantics. Language, 67:1.34-62.

MacLaury, R. E. 1992. From Brightness to Hue: An Explanatory Model of Color-Category Evolution. Current Anthropology, 33:2.137-186.

Martin, S. E. and Y.-S. C. Lee. 1986. Beginning Korean. Rutland, VT and Tokyo: Charles E. Tuttle Co.

Rosche, E. H. 1972. Probabilities, Sampling, and Ethnographic Method: The Case of Dani Colour Names. Man 7.448-466.

Shim, R. J. 1994. Englishized Korean: Structure, Status, and Attitudes. World Englishes. 13:2.225-244.

Stanlaw, J. M. 1987. Color, Culture, and Contact: English Loanwords and Problems of Color Nomenclature in Modern Japanese. Doctoral dissertation, University of Illinois at Urbana-Champaign.

Tyson, R. E. 1994. Korean Color Naming and Korean-English Language Contact: A Study in Linguistic Variation and Semantic Change. Doctoral dissertation, University of Arizona, Tucson.

Webster's Third New International Dictionary of the English Language: Unabridged. 1981. Springfield, MA: Merriam-Webster.