國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
appendix.
1.3 Organization of Thesis
After introducing our motivation and research purpose in this study, we will review previous studies on color terms in Chapter Ⅱ. Chapter Ⅲ will present theoretical background, including theories of metaphor and metonymy, cultural factors in metaphor (Kövecses 2005) and lexicalization (Packard 2000). Afterwards, in Chapter Ⅳ we will display the analysis of the present study. Finally, Chapter Ⅴ concludes the results and suggests further research in the future.
‧ 國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
9
CHAPTER Ⅱ
PREVIOUS STUDIES ON COLOR TERMS
Usages of color terms are widespread and indispensable in our languages. In this chapter, we will review some previous studies on color terms. Studies of color
universals will be presented in Section 2.1 including representative work by Berlin and Kay (1969), studies of semantic universals in the domain of color vocabulary (Williams et el. 1970, Derrig 1978, Kay and McDaniel 1978), standpoints of using environmental reference in studying color terms (Wierzbicka 1996, Goddard 1998, Lucy 1997). In Section 2.2 we will display research of color terms in Taiwanese Mandarin such as diachronic studies (Cheng 1991, 2002, Wu 2011), examination of cognitive mechanisms behind color terms (Chen 1994, Liu 2001, Lin 2009) and traditional viewpoints of colors (Zeng 2002, Huang 2003, Liang 2005, Xie 2011).
Afterwards, research of color terms in Taiwanese Hakka and Taiwanese Southern Min is also mentioned in Section 2.3 (e.g., Cheng 2002, Liang 2005, Liu 2002). Finally, we will make some remarks on these previous studies and draw connections with our forthcoming research.
‧ 國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
2.1 Studies of Color Universals
The doctrine of the Sapir-Whorf hypothesis emphasizes the relativity of semantic structures instead of the role of linguistic universals. Nevertheless, studies of color terms (Berlin and Kay 1969, McDaniel 1972, 1974) hold the belief that “all languages share a universal system of basic color categorization” and that “these universals are inherent in the human perception of color” as in Kay and McDaniel (1978: 610). Kay and McDaniel (1978) present the lexical categorization of color as a paradigmatic example of the existence of biologically based semantic universals.
The representative work by Berlin and Kay (1969) investigates 98 languages via the use of standardized color chips. They find that color categorizations are not random and that all languages seem to share similarity regarding the foci of basic color terms. The eight criteria that Berlin and Kay (1969: 5-6) stipulate for defining the concept of basic color terms are listed below. The first four criteria are the most essential whereas the last four subsidiary criteria are for handling a few doubtful cases:
1. It is monolexemic; that is, its meaning is not predictable from the meaning of its parts. This criterion eliminates [bluish, lemon-colored, salmon-colored, and the color of the rust on my aunt’s old Chevrolet] and perhaps also [blue-green].
Its signification is not included in that of any other color term. This criterion
‧ 國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
eliminates [crimson] and [scarlet], which are both kinds of red for most speakers of English.
2. Its application must not be restricted to a narrow class of objects. This criterion eliminates [blond] which may be predicated only of hair, complexion, and furniture.
3. It must be psychologically salient for informants.
4. The doubtful form should have the same distributional potential as the previously established basic terms. For example, in English, allowing the suffix –ish, for example, reddish, whitish, and greenish are English words, but *anguaish and
*chartreuse(e)ish are not.
5. Color terms that are also the name of an object characteristically having that color are suspect, for example, gold, silver, and ash.
6. Recent foreign loanwords may be suspect.
7. In cases where lexemic status is difficult to assess, morphological complexity is given some weight as a secondary criterion. The English term blue-green might be eliminated by this criterion.
Berlin and Kay (1969: 4) contend that “the referents for the basic color terms of all languages appear to be drawn from a set of eleven universal perceptual categories, and these categories become encoded in the history of a given language in a partially
‧
fixed order.” They delineate the evolutionary stages of basic color terms as displayed in the following:
Figure 1. Berlin and Kay (1969: 4): Order of Color Evolution
Furthermore, from their investigation of a wide variety of languages, Berlin and Kay (1969: 104) discover that “[c]olor lexicons with few terms tend to occur in association with relatively simple cultures and simple technologies, while color lexicons with many terms tend to occur in association with complex cultures and complex technologies.” Namely, the number of color lexicons is in direct proportion to the complexity of culture development and technology evolution.
Berlin and Kay (1969) discover that in some languages colors are divided into two categories, ‘brilliant’ such as white, most red and yellow and ‘dull’ like black and
most greens. In their evolutionary stages of colors sequence, there are nine languages belonging to stage one7 and containing only two basic color terms, black and white,
7 The nine languages containing only two basic color terms are Dugum Dani (New Guinea), Lower Valley Hitigima (New Guinea), Jalé (New Guinea), Murry Island (New Guinea), Ngombe (Congo), Paliyan (South India), Pyramid-Wodo (New Guinea), Upper Pyramid (New Guinea) and Torres Straits Tribes (New Guinea) (Berlin and Kay 1969: 46).
Stage 7
‧ 國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
which include the concepts of other associated colors or objects. Also, in revising the basic color category encoding sequence, Witkowsky and Brown (1978) define the stage one as ‘macro-white’ and ‘macro-black’. For these very aboriginal languages, it seems that their worldview is simply divided into a clear-cut black and white.
In addition to Berlin and Kay (1969), a lot of researches indicate that semantic universals exist in the domain of color vocabulary. For instance, Williams et al. (1970) support the hypothesis of cross-cultural generality in the connotative meanings of basic color terms through the semantic differential ratings of six languages of college students, American Caucasians, American Negroes, Germans, Danes, Hong Kong Chinese and Asiatic Indians. They conduct the rank-order displacement of ten English color terms along the Evaluation, Potency and Activity dimension. White is generally viewed as positive, weak and less active while black is usually considered to be negative, strong and passive. Also, they find that all groups rank white as the most favorable and brown and black as the least favorable. These colors are often used to designate major racial groups. For example, Caucasians are white and Negroes are black. Nevertheless, although color names retain evaluative connotations of racial groups, the authors discourage such inadequate usages because these terms fail to manifest mutual respect and cooperative harmony among miscellaneous races.
Derrig (1978: 89) examines color terms of four “typologically, genetically,
‧ 國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
ecologically and areally diverse languages”—Chinese, Mongolian, Yucatecan Maya, and Zulu. She finds that the extensional senses of color terms share “the continuum of cross-cultural salience”. Moreover, through presenting evidence of the existence of biologically based semantic universals, Kay and McDaniel (1978: 610) verify the claim that “these universals are inherent in the human perception of color”.
Despite the fact that some studies support color universals on the basis of neurophysiological evidence, some researchers hold another point of view. For
example, Wierzbicka (1996) questions Berlin and Kay’s (1996) theory of evolutionary sequence. She maintains that the use of color charts or Munsell’s color chips are not suitable for showing the meaning of color terms. Wierzbicka (1996: 334) maintains that “[h]uman conceptualization of colour, which is reflected in language, may be constrained by the neurophysiology of vision, but it can be neither described nor explained in terms of neurophysiology. To describe it, we need to take recourse to human conceptual universals.” She proposes that we can associate our visual
categories with certain easily accessible models such as fire, the sun, vegetation and the sky. It is these ‘shared concepts’ of fire, sun or sky that can function as cognitive anchors for color naming because only cognitive concepts rather than
neurophysiological perceptions are intelligible and communicable to others.
Goddard (1998) also holds the same standpoints as Wierzbicka (1996). He
‧ 國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
contends that meanings are communicable, color words are signs, but neural
responses are not communicable and are not signs. Also, since speakers do not know the facts about neurophysiology and therefore we cannot use these scientific facts to measure the semantic competence of speakers. Goddard (1998) claims that we should use visual and environmental things as common reference points in the study of color meanings. Take black and white for example. The most obvious distinction in all colors is the light vs. dark distinction. The most significant environmental prototypes of this distinction are the night and day because “the cycle of day and night is a recurrent and universal (or near-universal) human experience” as claimed by Goddard (1998: 126). The representative colors for day and night are white and black,
respectively.
Similar to the standpoints of Wierzbicka (1996) and Goddard (1998), Lucy (1997) also recognizes that adequate knowledge of color terms cannot be restricted within color chips and color labeling and that color exists in human beings’ interpretation of light. Lucy (1997: 341) contends that “the communicatively relevant encodings of visual experience do not lie ‘in there’ in the biology but out in socially anchored linguistic systems.” Therefore, when studying semantic universals of color terms, we should take their corresponding reference into consideration.
‧ 國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
2.2 Research of Color Terms in Taiwanese Mandarin
The pioneer study of color terms in Mandarin Chinese is done by Berlin and Kay
(1969) and they determine that Chinese is at stage five and the basic color terms are
hēi (黑) ‘black’, bái (白) ‘white’, hóng (紅) ‘red’, lǚ (綠) ‘green’, lán (藍) ‘blue’ and huáng (黃) ‘yellow’.
Cheng (1991, 2002) examines Chinese color terms through the diachronic research of È rShíWŭShǐ (二十五史) ‘twenty-five books of Chinese history’. She traces the origin of basic color terms, outlines evolutionary stages from the ancient times to the present and verifies the universal tendency proposed by Berlin and Kay (1969) with Chinese data. Besides, she compares the semantic structure of color terms of different periods in Chinese history and sketches a profile of the chronology of color terms as displayed in the following figure 2:
Figure 2. Cheng (2002: 325): A Profile of the Chronology of Color Terms
‧ 國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
Like Cheng (1991, 2002), Wu (2011) also investigates the evolution of Chinese basic color terms over the past 4000 years through counting word frequency based on works of Chinese literature. He discusses what stages or periods of color terms in history should be categorized under Berlin and Kay’s (1969) work and divides the evolution into five phases and eight periods.
Cheng (1991: 67) points out that ShuōWénJiěZì (說文解字) offers the origins of the Chinese characters hēi, wū (烏) and bái as in the following:
Table 1. Origins of Chinese Characters hēi, wū and bái
Character Explanation in ShuōWénJiěZì Cheng (1991)
白 bái 西方色也，陰用事，物色白，從入，合二，二陰數。 a phonetic
loan character
黑 hēi 北方色也，火所熏支色也從炎，上出囪。 associative
compound
烏 wū 孝鳥也。 a pictographic
character
We see that in Mandarin Chinese wū does not refer to the color black but stands for a kind of bird, wūyā (烏鴉) ‘crow’. It is the character hēi that represents the color black in Mandarin. Cheng (1991: 63) finds that the terms for black are slightly different in Chinese dialects. Most dialects use the character hēi as the color black.
However, NánChāng (南昌) is a transitional area where both hēi and wū are used. In some southern regions, people only use the character wū to denote the color black.
These areas are MéiXiàn (梅縣), XiàMén (廈門), CháoZhōu (潮州) and FúZhōu (福
‧ 國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
州), where people speak Hakka or Southern Min. This may be the reason why wū
serves as the color black instead of hēi in TH and TSM. In addition, Huang (2003) finds that hēi is only used for rhetoric purposes or proper names in DōngShì (東勢) Hakka such as Hed Liung Gong (黑龍江). He and Zeng (2006) hold the same viewpoints in their study of color terms in SìXiàn Hakka. From the above-mentioned studies, we can tentatively conclude that the terms vu (烏) and oo (烏) represent the color black in TH and TSM, respectively.
Chen (1994) investigates the linguistic universals and linguistic relativities of color-word metaphors in Mandarin Chinese and in English. He proposes a model which is based on Danesi’s (1993) idea and Fearing’s (1954) process of
conceptualization. He hypothesizes that those lowly-abstracted color-word metaphors incline to be linguistic universals. On the other hand, if color-word metaphors are highly-abstracted, they are more prone to linguistic relativities.8
8 Chen (1994) divides color terms into five types: Perception-based Type (感知類), Prototype (典型 類), Sensational Type (情緒類), Symbols or signals (符號類) and Slang and idioms (俚語類).
However, we consider Chen’s (1994) model to be erroneous because the five types are not classified in the same dimension or according to the same criteria.
‧
Some studies (e.g., Liu 2001, Lin 2009, Wu 2007) examine cognitive
mechanisms of color terms. To begin with, Liu (2001) investigates the six color terms in Mandarin and proposes a model for the semantic extensions of the six basic color words as in the following Table 2. She divides color terms into Concrete/
Perception-based Type, Metonymic Extension, Metaphorical Extension and Arbitrary Extension.
Table 2. Liu’s (2001) Categorization of Mandarin Six Basic Color Words
Category Some Examples
1. Concrete/
Perception-based type
hēifă (黑髮) ‘black hair’、báilà (白蠟) ‘white wax’、
hóngdòu (紅豆) ‘red beans’、lántiān (藍天) ‘bluesky’
2. Abstract/ Conceptualization type 2a. Metonymic Extension
bànhēiliăn (扮黑臉) ‘to wear a black face; serious’、
báishǒu (白首) ‘white hair; old’、miànhóngěrchì (面紅
耳赤) ‘the face and ears turning red; angry’2b. Metaphorical Extension
hēidào (黑道) ‘the underground’、qīngbái (清白)
‘innocent’、hóngrén (紅人) ‘famous people’
2c. Arbitrary Extension
hóngchén (紅塵) ‘the mundane world’、
huángniú (黃牛) ‘a yellow ox; a scalper of tickets’
Lin (2009) adopts the same division of data analysis as Liu’s (2001) and
compares six basic color terms in Mandarin and in English, black, white, red, yellow,
green and blue. One interesting observation she draws is the corresponding three
models between the two languages as presented in Table 3.
‧
Table 3. Lin (2009): Corresponding Models Between English and Chinese
Model Explanation Data Example: Chinese-English 1. Associative jealousy’ — see red ‘angry’qīngyăn (青眼) ‘with black eye;
to look with favor’
— a black eye ‘disgraced’
Wu (2007) specifically focuses on the Chinese color term hēi. She divides the
meaning of the color term hēi into two main types: the prototypical meaning such as
hēifă (黑髮) ‘black hair’ and the non-prototypical extended meaning. The
non-prototypical extended meaning includes schematic-related extension like hēiyè (黑夜) ‘night’, metaphorical extension such as hēidào (黑道) ‘the underground, gangsters’, metonymic extension like hēiliăn (黑臉) ‘black face; strict person’ and history-based and borrowed-translation extension like hēiànshídài (黑暗時代) ‘Dark Ages’.
Zeng (2002) analyzes the relationship between five fundamental colors (五正色) and YīnYángWŭXíng (陰陽五行) ‘Yin Yang Five Elements’. He finds that Yin Yang Five Elements deeply influence the development and interpretation of the five
‧ 國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
fundamental colors, hēi, bái, chì (赤) ‘red’, qīng (青) ‘green or blue’, and huáng.
Zeng (2002) discovers that the symbolic meanings of color terms are more prominent than the needs of visual aesthetics for Chinese people in early historical periods.
These extensional meanings of color terms also reflect people’s philosophy of life about the natural world and the life cycle.
From the traditional viewpoints (Huang 2003, Liang 2005, Zeng 2002, Xie 2011), in Yin Yang Five Elements, the color black in Chinese is situated in the north and belongs to winter, during which creatures become dormant, plants wither, and people easily have negative emotions. Therefore, hēi has always been regarded as
inauspicious, disastrous, evil and negative in the Chinese community. The color white in Chinese is located in the north-east which is the position of death in Chinese
FēngShuǐ (風水). Consequently bái has been connected with Chinese funerals and the funeral clothes are white.
2.3 Research of Color Terms in Taiwanese Hakka and Taiwanese Southern Min
The research of color terms in TH (Huang 2003, Liang 2005, Cheng 1991, 2002, He and Zeng 2006) and in TSM (Cheng 1991, 2002, Liu 2002) has always been relatively little in comparison with those abundant studies of Chinese color terms.
Cheng (1991, 2002) determines that the color category of TH or TSM is similar to
‧ 國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
that of ShuōWénJiěZì (說文解字) in Hàn period. There are five basic color terms. For TH, they are vu (烏) ‘black’, pag (白) ‘white’, fung (紅) ‘red’, vong (黃) ‘yellow’ and
qiang (青) ‘grue category of blue and green’. For TSM, these basic color terms are oo
(烏) ‘black’, h (白) ‘white’, âng (紅) ‘red’, n g (黃)9 ‘yellow’ and tshenn (青) ‘grue category of blue and green’. Cheng (1991) believes that studies of synchronic
variation in dialects can contribute to our understanding of the development of Chinese basic color terms.
Liang (2005) makes a simple comparison of basic color terms in TM, TH and English and thereby observes economic activities and unique culture in Hakka groups.
He and Zeng (2006) investigate TH of SìXiàn (四縣) dialects and find that the surroundings are often embedded in color terms or color-related proverbs. They believe that we can perceive the racial characteristics and customs of Hakka people through the color rhetoric in SìXiàn Hakka.
Liu (2002) discusses color terms of ABB form in TSM such as oo-sim-sim (烏 BB) and analyzes the different meanings of ABB through prototype theory, semantic fields and semantic extensions. For example, h-phau-phau (白 BB) is used to describe the fair skin while h-bông-bông (白 BB) illustrates the scenery of natural beauty. In addition, Liu (2002) contends that the development of semantic extensions
9 Notice that in TSM n g (黃) ‘yellow’ is for colloquial usages and hông (黃) ‘yellow’ is for rhetorical usages.
‧ 國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
is unidirectional from the concrete to the abstract. For instance, oo-ma-ma (烏 BB) describes the lack of lights, which derives from the concrete color of oo to the abstract notions of dirtiness and the lack of lights. Also, h-siak-siak (白 BB) illustrates the luster of objects, which originates from the concrete color of h to the abstract notions of cleanliness and the luster of objects.
2.4 Remarks
In investigating the four diverse languages, Derrig (1978) does not address so much on the cross-linguistic similarities among the extensional senses of color words.
Nevertheless, in reviewing Derrig’s (1978) work, Cheng (1991, 2002) deduces one conclusion. Namely, the earlier stages (black, white, red) in the implicational order are more productive in creating metaphorical uses than those at the later stages (yellow, green, blue). In the study of basic color terms in Mandarin Chinese, Cheng (1991, 2002) and Liu (2001) both mention that their studies agree with this conclusion. We want to observe whether the color terms at the earlier stages, namely, black and white, in TH and TSM also contain abundant metaphorical usages.
Liu (2001) proposes a model for the semantic extensions of the six basic color words in Chinese and classifies the meaning extensions into concrete
perception-based type, metonymic extension, metaphorical extension and arbitrary
‧ 國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
extensions. Some studies like Wu (2007) or Lin (2009) adopt her model to analyze color terms in Chinese and in English. However, we have some doubts about their explanations of metonymic extension and metaphorical extension.
First of all, we take some examples from Liu’s (2001) category of ‘metaphorical extension’ for illustration. For the phrases hēidào (黑道) ‘the underground’, hēijīn (黑 金) ‘underhand payment’ and hēishì (黑市) ‘black market’, Liu (2001: 36) gives the
following explanation: “Therefore, hēi in hēidào, hēijīn and hēishì indicates such an
‘illegal’ property.” For the phrase báikāishuǐ (白開水) ‘plain water’, Liu (2001: 46)
explains in the following: “As for example (42) báikāishuǐ, bái indicates ‘plain’
because plain water does not contain any flavor.” From Liu’s (2001) ambiguous explanations, it seems to us that she refers to the color terms themselves, i.e., hēi or
bái, as the metaphorical extension. Maybe Liu (2001) does not clarify her explanation
clearly. However, we would like to clarify and contend that it is the whole chunks or compounds, e.g., hēidào, hēijīn, hēishì and báikāishuǐ, that are the manifestations of metaphor. Only when the color character collocates with its partner in the whole
compound can the metaphorical extension be derived (e.g., hēi with dào, bái with
kāishuǐ). Likewise, this clarification also applies to the category of metonymic
extension. For example, the whole phrase jŭbáiqí (舉白旗) ‘hold a white flag’ refers‧ 國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
to the act of surrender through the metonymy ACTION FOR RESULT.10 The color character, bái, in this phrase still refers to the perception color, white, without denoting any metonymic meaning extension.
In addition, we think Wu (2007) makes the same obscure explanation as Liu (2001) does with regard to the categories of metonymic extension and metaphorical extension as displayed in the following: “The second category is ‘Metaphorical Extension’ in which the concept of Hēi ‘black’ represents some abstract concepts in virtue of mapping properties across two different domains.” and “The third category is ‘Metonymic Extension’ which means that Hēi ‘black’ here has its primary extension with the metonymic extension of linguistic expression” (Wu 2007: 38). Wu (2007) seemingly thinks of the color term hēi itself as the realization of metaphorical or metonymic extension. Therefore, we should reiterate our contention. From our viewpoints, it is the whole compounds that are the manifestations of metaphorical or metonymic extension instead of the color terms hēi and bái themselves.
All in all, at this moment we conclude that metonymic extension or metaphorical extension for color words happen only when they collocate with their modified components. We surmise that there are only two types for the meanings of the color terms. One type refers to the meaning of the physiologically visual color. The other
10 We consider jŭbáiqí (舉白旗) to be the metonymy ACTION FOR RESULT (cf. Kövecses and
10 We consider jŭbáiqí (舉白旗) to be the metonymy ACTION FOR RESULT (cf. Kövecses and