PREJUTDICE AND ETHNIC STEREOTYPES IN RURAL HONG KONG E. N. Anderson, Jr. University of California, Riverside Most studies of prejudice have covered situations in which the dominant group is European or of European ancestry. Anthropology is filled with references to mutual antagonisms between groups, but rarely have these been made the subject of study, except within wider contexts of general community studies. This situation is soon to be remedied, if the plans of LeVine and his associates bear fruit (Current Anthro- pology 1966:515-516). But their work is directed mainly toward inves- tigating nonliterate or recently literate groups in areas only recently controlled by central governments. Thus much of the continent of Asia is excluded. There is a certain amount of available material that deals with group rivalries and prejudices in India; Carstairs' The Twice-Born (1958), for example, deals with intercaste and village-tribe conflicts. Conflicts of native peoples and immigrant Chinese have often been studied in southeast Asia (e.g., Skinner 1957, 1958; Coughlin 1960; Willmott 1960). But for China itself, information is scattered and limited. The purpose of the present paper is simply to report certain observations on this subject, in hopes that they will prove useful to those more know- ledgeable than myself in the field of ethnocentrism. The research on which this paper was based was carried out at Castle Peak Bay, New Territories, British Crown Colony of Hong Kong, in the year 1965-66.1 Primary focus of research was on the boat-dwelling people of the bay; during most of the study period, my family and I lived on a small boat. During this period we had excellent opportunities to observe the behavior reported in this paper. Castle Peak Bay is twenty miles by road from the city of Kowloon, and is cut off from it by high, barren hills and wild coasts. The bay is a farming and fishing area, with a few factories of recent origin. Several 90 towns and villages are in the area; the metropolis of the region is Tun Mun San Hui, usually called simply San Hui. It is of postwar origin-- a young, heterogeneous town, where local villagers from nearby settle- ments live together with refugees from Communist China and commuters from Kowloon. There are perhaps 20,000 people living around the bay and in the neighboring agricultural valley at its head. Traditional villages, some of them reached only by trail and boat, are nearby. The area pre- sents a microcosm of Hong Kong Colony. Most of the ethnic groups found among Hong Kong's 3,500,000 people are represented. City-dwelling and relatively sophisticated Chinese meet village peasants and fishermen. In the old villages settlement and even family names have persisted for hundreds of years. On the other hand, San Hui is very new and hetero- geneous, extremely unlike the more typical rural towns reported by Kulp (1925), Chen (1939), Freedman (1958, 1966) and others. The Chinese of Hong Kong divide humanity into groups on the basis of language and place of origin. At Castle Peak Bay, two basic categories may be distinguished: Chinese (tsung4 kuok2 ian5,2 "Chinese," defined by use of one or another Chinese language), and foreign devils (faan4 kuai6 lou6, literally "foreign ghost fellows;" usually shortened to kuai6 lou6, "ghost fellows"). The latter category includes not only Westerners, but also Japanese and other non-Chinese Orientals. Each of these categories is subdivided. The first is divided on the basis of language. Several so-called "dialects" exist in the Castle Peak area. Most of these are actually distinct languages within the Chinese family. The following dialect groups and language groups are well enough known in Castle Peak to have associated stereotypes. The first is the land dwelling Cantonese, speaking Standard Cantonese (McCoy 1965). This group calls itself Pun6 tei3, "original people of the area." This is the majority group in the area, comprising at least 10,000 persons in or near San Hui, the bay towns, and the agri- cultural valley. 91 The second group is the boat dwelling Cantonese, who speak a dialect which is close to Standard Cantonese but is nevertheless dis- tinct. They call themselves soi2 s"ng5 ian5, "people on the water." Others know them as taan5 kal ("Tanka"), a hatred term of uncertain origin. Like the land dwellers, they have been in the area for a very long time, certainly several hundred years. Their numbers have been very greatly increased by refugees in the last ten years. There are perhaps six or seven thousand people in the bay who fall into this cate- gory on the basis of dialect and origin, but only about a third of these live on boats at present. The estimate of population may be too high, as the number of boats fluctuates widely, making accurate censusing im- possible. A third group is the Hakka (hak2 kal). These hill farmers living in small villages or groups of houses speak a completely different language within the Chinese family. An uncertain but large number, about one thousand, farm the hill slopes around the bay. Fourth is the Chiuchow (tshiu4 tsau4; Mandarin Ch'ao-chou. They call themselves Teochiu, by which name they are well known in Southeast Asian literature). They are emigrants from Chiuchow district of Kwangtung province. The language is a variant of Southern Min (South Fukienese) and is extremely different from Cantonese. A small number, perhaps a few dozen, live in the bay area. The Hoklo (hok2 lou6) is a fifth group. They speak a language related to Chiuchow and are boat people. They are concentrated around Tai Po in the eastern New Territories. Only a very few live at Castle Peak Bay and near Tai 0. Sixth is the Toisanese. Toisan (thoi6 saan4) is a district southwest of Hong Kong in Kwangtung province. Its dialect is theoreti- cally comprehensible to speakers of Standard Cantonese, but is quite different therefrom. I know of none permanently resident in the bay area. The seventh group is the Shanghainese. Refugees from Shanghai, speaking still another different language from within the Chinese family, 92 have become numerous in the urban districts of the Colony. I know of none resident in the bay area, but they are well known to residents there. The category of "foreign devils" is divided on the basis of nationality. It includes British, Americans, Indians (known as mo2 1o4 kuai6, tones uncertain here: "more ghosts," Portuguese moro, moor, Negro) and Japanese. Of these, only British reside in significant num- bers in the Castle Peak area, but the other groups are known from tourists and others. ON STUDYING PREJUDICE The present study was made primarily by means of observation and other usual anthropological methods. Some use was made of the lin- guistic-type methods advocated by Metzger and Williams (1966), Frake (1962), and others. Unfortunately the use that could be made of direct questioning--by interviews, questionnaires, or schedules of the Metzger- Williams type--was minimal. Although useful in obtaining the basic categories, direct questions did not produce accurate responses in the field of prejudice. Informants were too polite or too secretive to give their opinions of other groups freely, except where specially close re- lationships were maintained. Chance remarks, overheard conversations, spontaneous statements, and above all observed behavior proved far more useful in arriving at the data. Observed behavior allowed a certain amount of quantification in a rough way, as will appear below. Accuracy of the following data can be more or less guaranteed, however. Agreement among honest informants was very high. Information was perfectly consis- tent with observed behavior, and we had exceptionally fine opportunities to study the latter. We ourselves belonged to a despised minority, we were living with another, the boat people, and we were not British, and therefore not feared to any extent. We constituted a family, man, woman and child, and could observe prejudice behavior toward a whole family in various different settings. As we became known in the bay, where we were very well received, prejudice against us tended to disappear in areas 93 where we were well known. We could then contrast the situation when we visited places in which we were unknown. SHOWING PREJUDICE Castle Peak, like many other heterogeneous areas of the Colony, was an area where intolerance along ethnic group lines was pronounced. In older and more established areas, such intolerance is less marked, but village feuds and other local prejudices are much stronger than at Castle Peak. In urban areas, political and class conflicts may obscure ethnic ones. Castle Peak is almost free of the former and does not have the latter in structured form, although the mutual distaste of rich and poor is pronounced. The means by which intolerance is shown are well recognized and frequently invoked. Violence has been stopped by the Colony's Govern- ment, but is well remembered by many boat people: "Rough kids from the tongs used to catch fishermen and beat them up." Boat people still do not allow women and children to go far from the waterfront, and usually guard them even there. Wars between Cantonese and Hakka were also common at one time. Informants told of fights with groups of Hoklo. Segregation is also becoming a thing of the past, but in Imperial China, the boat people were not allowed to attend schools, live on land, or marry land people (Chen 1935; Ho 1965). Much of this feeling and its attendant segregation persists. Boat people will not send their children to pre- dominantly land people schools: "Our children get teased." Hakka were forced onto marginal land in the hills and not permitted to settle in the valleys, and even with the disappearance of violence, economic forces are invoked to maintain this (Grant 1960). But rigid segregation is breaking down; land-boat marriages are known, mixed schools common, public vehicles open to all. A certain amount of impromptu segregating of whites is found; "foreign devils" are directed to less used parts of restaurants, and otherwise kept from overmuch contact. But such Western- Chinese segregation as exists in Hong Kong is usually initiated from the other direction. 94 At present, rock-throwing and threats are observed on occasion. We have seen rare instances of land Cantonese children so treating boat people, and have seen some half dozen cases of Cantonese so treating Westerners (including ourselves). But most expressions of intolerance are now verbal or otherwise relatively innocuous. Chief among these is the fascinating and extremely widespread practice of uaan6 pan5 tshat3, literally "making a stupid prick." (Polite rendering: "make a fool of one," but obviously much is lost in translation this way.) The basic principle of this art is to show that the victim is incompetent in the language and culture of the tormentor, and/or that his own language is silly and inferior. This is often done by wordplay. Thus, Toisanese are teased with a phrase iu2 khei3 maa7 ko2 hoi6, "I want to ride a horse on the sea," because it sounds lilueh like the Toisanese for "fuck your mother's cunt," the strongest insult in Cantonese. Similarly, the story is told of a Shanghainese who wanted to commit suicide with his wife, and cried out in his accented Cantonese "I want to seize my wife and drop into the sea and die!" "Drop into the sea and die," in Cantonese, is "tit2 hoi6 sei6," but the poor man lapsed into broad Shanghainese dialect at the critical point, and said "tiu2 hai6 sei6," "fuck her cunt to death." Hai means sea in Shanghainese, but--alas-- not in Cantonese. Shanghainese are said to become well aware of this story. Nonsense manifestations of the same wordplay are heard. Thus at a Japanese film, children make mocking noises like "keikokeiko" in a quick chatter; English, perceived as full of "s" sounds, is parodied with phrases like "asikasekaso." A simpler and more childish method of uaan6 pan5 tshat3 is to shout at him insults that he does not understand. English speakers are of course particularly susceptible, since neither the standard Cantonese courses nor the Chinese friends of the Westerners go into much detail on the infinitely rich and varied vocabulary of abuse found in that as in so many other languages. Even if the listener can understand the insult, 95 if he cannot reciprocate he has been made a stupid prick. Parents direct their children to run after persons of despised groups, mocking them. The persons might reciprocate against adults, but cannot do much to the children without occasioning universal disapproval. Usually the insults in this case are phrases like taan5 kal lou6 (the ethnophaulism for the boat people), "foreign devils," hak2 kal lou6 ("Hakka fellows," said in a mocking way), and the like. But occasionally stronger phrases are heard. Other ways of uaan6 pan5 tshat3 are less verbal, but still in- volve a verbal component. Begging is one of these. Begging may have a purely economic motive; but when a person who is not in real need begs, he is frequently doing it to insult rather than to obtain anything, with the implication being "I think you are so stupid that you will give me things even though I don't need them." This is primarily a children's game. Cheating is perhaps an adult equivalent. Price markups come under a slightly different category. Paying an especially high price for something stamps one as a fool, to be sure, but there is another element in it, which is the use of differential prices for persons of different degrees of closeness to the seller. These prices are set on a surprisingly consistent scale. An item that costs a regular customer $1.00 will cost a stranger about $1.50, and if the stranger is of a dif- ferent ethnic group $2.00. If the stranger is a supposedly rich person, the price may go to $3.00. A very neat measure of one's acceptance in a community is provided by the slow but steady fall of prices. Thus, we began by paying two or three times the standard price, and noted the de- cline of prices to parity or something fairly near it. Prices form an excellent measure of hostility (see Ward 1955, and King 1955, for other observations on anti-boat-people prejudice and its expression through pricing). In addition to the above recognized methods, there are of course the usual gamut of unfriendly and discourteous acts of various kinds. Hostility may be shown by the mask-like, frozen facial expression 96 that characterizes the "inscrutable Oriental" in the west, but is a per- fectly scrutable expression of distaste in Hong Kong. Cross-group friendship is found, but is not common. Association is usually with persons of one's own ethnic group. Measuring is complicated by the lack of real accord between Cantonese and English words for such relations, but I know of no more than four or five individuals on the waterfront who had become accepted as friends by another ethnic group. The Chiu- chow clerk of the local Fish Marketing Office was known and liked by most of the boat people, because his job put him in contact with them; he was the most conspicuous case we observed. Interethnic marriage is also rare, and "Would you let your daughter marry one?" is as effective a shibboleth in Castle Peak as in America. From the more than 100 boat people households we knew in some detail, we knew of one interethnic marriage (a shore man and a boat woman; they lived on land). We knew of some half dozen other cases of this kind, and of one in which a land woman had married a boat man and gone to live on a boat, an arrangement that lasted for only a short time. This was the only such case known to informants at Castle Peak. On the other hand, interethnic contact was very frequent in the Castle Peak area, and was usually friendly. Close personal friend- ship was relatively rare, but also was open and expressed antagonism. Voluntary associations often cut across ethnic lines; economic trans- actions and arrangements regularly did so. A community of interdepen- dent groups could hardly be otherwise. Boat people supplied most of the custom of waterfront stores; Hakka brought vegetables to stores and markets. Relations good enough to sustain trade and communication were necessary and were maintained. We heard of one case in which a fish dealer told a fisherman "I can beat you dirty Tanka like dogs!" This was exceptional, and resulted in a lawsuit. The attitude it expresses is all too typical, but not by any means universal, and the expression of the attitude in such violent terms is now rare enough to excite comment. 97 ETHNIC STEREOTYPES Inseparable from the behavior of one group toward another are the ostensible reasons for such behavior. In Castle Peak, the di- mensions of inferiority and superiority are well known, and each ethnic group is stereotyped in regard to these dimensions. The stereotypes can be viewed as predictions about behavior; they usually have some basis in fact, but are of course vastly exaggerated and distorted. Each group has its own configuration of bad qualities, but certain specially dis- valued characteristics are held common to most. In general, a given stereotype is the common property of all groups except, of course, the one victimized thereby. Though group A and group B are hostile to each other, they unite in being hostile to, and holding a single stereotype of, group C.3 It will be best to give stereotypes in order of the groups themselves. Translations of informants' statements are given in quotes; fuller discussion of key words is found below. The Cantonese proper are viewed as overbearing, grasping, corrupt and mean. The boat people of Cantonese speech are almost invariably called "stupid" and "uneducated;" they are also regarded as dirty. They are considered to be non-Chinese in origin and are even said to speak a non-Chinese language, have a non-Chinese physique (a falsehood that has crept into the literature on occasion, e.g., Lo 1955), and even to have tails and six toes (cf. Ward 1965). The Hakka are "ignorant" peasants. "The women work while the men sit and sing." (The women are indeed hard workers. The men once had to spend much of their time on guard against attacks, and left gardening and other such work to the women. Now, of course, both sexes work.) The Hakka are also said to be clannish and to dislike outsiders. The Chiuchow are "dishonest" and "rascals." "Most of the people in the tongs are Chiuchow." They are also "ignorant" back-country pea- sants, like the Hakka and Toisanese. 98 The Hoklo are "always fighting" and "when one gets in a fight the others all help him." They are hard to deal with because of clannish- ness and meanness. They are also "ignorant." The Toisanese are ridiculous rustics who can't speak proper Cantonese; they are perhaps the most "ignorant" of the peasants. The Shanghainese are slick city people, big-time criminals, secretive and dishonest. Unscrupulous in business and extremely wily, they all get rich fast. Moreover, they are "very fat from eating their greasy food." Non-Chinese are lumped under the general heading of "foreign devils" and as such regarded as having an inferior and inadequate cul- ture. The Chinese, of course, are not alone in this attitude, which seems to animate certain American foreign policies. But the concept of ignorance" (see below) makes alien cultures especially unfortunate in the eyes of some Cantonese. This came out most strongly in regard to our child-rearing patterns. The older women would pounce on our daughter and try to use their own methods on her; differences between our child- rearing methods (ordinary American middle class) and their own were not encouraged. We got a few angry lectures to the effect that we were guilty of neglect and cruelty because we did not have a Chinese servant to do the child-rearing. These people had the best intentions in the world; much of their fear derived, we found, from the belief that the British hired Chinese servants because they (the British) were completely incapable of handling themselves or their children. As our daughter con- tinued to flourish without servants, these lectures ceased. Other exam- ples of this opinion of alien cultures were found in the universal negative reaction to Western foods, and, on one memorable occasion, in an attempt to teach a Japanese friend how to use chopsticks. It was assumed that the unfortunate person, being non-Chinese, would not know how to use them. Another stereotype about aliens is that they are all rich, which is not an entirely unreasonable assumption, considering the Westerners to whom the Castle Peak residents are exposed. Americans are 99 the richest of all, British next. America is often known by the term kaml saan4, "golden mountains," a term originally applied to San Fran- cisco because of the 1849 gold rush. We found that many informants in- terpreted the term literally, believing that America was composed of mountains of solid gold. We were occasionally asked if our two sea chests were full of gold. Two groups of Occidentals are not considered rich: the missionaries and the White Russian refugees. We have heard missionaries described in pitying terms as outcasts, failures who became too poor to live among other Westerners, and had to go to live among the poorest Chinese. As such, they were figures of fun as well as of pity. The White Russians are genuinely poor; a few of them still trickle out of Communist China. A special degree of dislike is reserved for Asian foreign devils. Indians are hated because they are supposed to be succeeding better than the Chinese on the Chinese home ground, and because the British are said to favor them over native Chinese. Japanese are still disliked because of World War II, during which the Hong Kong area suf- fered terribly. From the above data, certain concepts emerge as specially important. Of these none is more often invoked than "ignorance." Of several words clustered around this general idea, the important ones for stereotypes are m5 man4 fa2, "uncivilized," and tshou4. This latter word parallels our "rough" and "coarse," in idiomatic extension as well as in the literal sense, but has rather more force. Applied to a person, it denotes a rough, uncouth, uneducated and barbarous individual. Ignorance and moral shortcoming are equated here. The theories of Confucius and Mencius that dominated Chinese scholastic thinking for so long are the source of this equation; traditional Chinese education perfected the man both morally and intellectually (see Meng Tzu and other Confucian writings). Another dimension of prejudice is related to economic success. The rich men of the village are considered grasping, mean, and dishonest; 100 their riches were attributed by gossip to all manner of illegal deals. In San Hui, speculation ran high as to how much the village head had paid in bribes for his position, and how much he would get out of being in said position. This stereotype of the rich is closely related to the picture of the Cantonese and Shanghainese. On the other hand, the poor are invariably described as tshou4; poverty and ignorance are virtually inseparable. Hence all the relatively poorer groups are stereotyped with that label. (Cf. Lin 1948 for much material on atti- tudes toward wealth in rural and town China.) VARIATIONS IN DEGREE OF INTOLERANCE The Cantonese of Hong Kong are extremely variable in many aspects of behavior, including the expression of intolerance. There is a range from almost hysterical behavior to an apparently total lack of hostility, with the range being governed by both individual and social factors. Measuring the hostility itself would be difficult indeed, but observing the expressions of it is easy (sometimes all too easy) and even quantifiable in a very rough sort of way. The use of comparative prices to evaluate feelings has already been mentioned. Other forms of unfriendly behavior are also usable. The best in this connection is children's teasing. Children are usually eager to shout the appropriate ethnophaulism at any member of a minority group, and frequently they are encouraged or supported by their parents. By noting the number of children doing this and the degree of intensity at which they shout, one can discover the degree of intolerance present in a given area with a sur- prising degree of accuracy. Social unrest of other types is also per- haps correlated, to judge from the opinions of policemen and other informants. The policemen had their crime statistics to back them up. As noted before, the Castle Peak Bay area is extremely hetero- geneous. It is not a community in any sense but the physical. Rather, it is a cluster of smaller units with varying degrees of social cohesion. These units range from the boat-dwelling fishermen to the San Hui 101 townsmen, and are perhaps more usefully called "communities" than is the larger aggregate. They vary widely in degree of overt hostility toward members of minority groups. At one extreme stand certain extremely tra- ditional groups: recent refugees from rural parts of Communist China, and inhabitants of very isolated villages near Castle Peak. These groups almost never show overt hostility based on prejudice. Even in the large but isolated town of Tai 0, on an island near Castle Peak, we were almost never called "foreign devil" (two occasions during a total of seven full days spent there), and land Cantonese and boat Cantonese dwell together with virtually no overt hostility. Both police and boat informants at Castle Peak Bay spoke of Tai 0 as an idyllic place compared with the Bay itself. Similarly, although recent refugees from the Communist mainland were somewhat shy toward us at first, they quickly received us with warmth and friendliness, and never expressed hostility along ethnic lines. (This is doubly impressive in view of the fact that we were known to all as Americans, and many of the refugees had lived their entire lives ex- posed to anti-American propaganda.) At the other extreme stood a specialized community: the lower- level workers in Castle Peak Mental Hospital, near the bay. These persons, janitors, telephone operators, ward caretakers, etc., continually expressed extreme hostility toward local peasant farmers, British, and others. During several months, we had occasion to walk through the compound where they lived several times a day. During this time there was no change in their attitude toward us, which began and ended as one of rejection. We, like the local farmers, were the targets of continual insults and occa- sional rocks. Another group markedly anti-outsider consisted of fishermen who for one reason or another had failed at fishing. They had beached their boats in the typhoon shelter and lived by casual shore labor. Since many of them had failed because of ostracism from the fishing-fish dealing community, there was obvious selection here. They were the people to whom no one would advance money, often because of dishonesty or unre- liability. 102 Other groups in the bay area fell between these extremes. The boat people were usually more tolerant than the shore people, but were often quite hostile toward such groups as, for example, the Hoklo. In general, the groups who expressed least overt hostility were the most traditional, and those expressing most hostility were the ones in process of change. Thus the boat people changing over to land labor, especially when the change was forced, were notably more hostile than boat people still fishing--except in the case of the recent refugees, who maintained an extremely traditional way of life afloat or ashore. The hospital workers were semi-Westernized. They frequently wore Western clothes, had Western furniture and equipment, and spoke a few words of English. They had abandoned many Pqpects of traditional Cantonese life. They appeared to feel comfortable ii. zither world. (It is perhaps rele- vant to remark here that the Western-eaucated and successfully Westernized groups we encountered were as free from overt expressions of hostility as the traditional villagers were.) The continuum from traditional-and-tolerant to deculturated-and- hostile is apparently not a universal one. Informants reported, and our limited observations confirmed, that the very conservative and traditional villages north of us such as the walled towns of Kam Tin were as extreme in hostility as the hospital-worker community. The historical position of these villages, the oldest continuous settlements in the Colony, is said to be accountable. The villagers are said to feel superior to others and to look down on them. This, of course, may well be another stereotype. Similarly, the extreme tolerance of Tai 0 may be explained more easily by the fact that boat and land Cantonese have lived there for a long time and adjusted to each other than by any mystical quality of traditionality. Within each of the various communities, there is a range of in- dividual behavior. The limits of the range are different. At Tai 0, people range from withdrawn and cool to very friendly, while at the hos- pital staff quarters, the range is from a casual sort of meanness and trick-playing to near hysteria. But the range itself is always apparent. 103 A typical example of individual difference is seen in our closest neigh- bors. Our boat was moored to that of Kwok Maai-tak and his family. Another boat, containing still another unrelated family, was also moored there. The Kwok family was from Tai 0 and exhibited Tai 0 tolerance to a marked degree. They never under any circumstances exhibited hostility toward us or toward the often objectionable and intolerant land people who visited their boat for business purposes. On the other hand, the other family did so repeatedly. They never adjusted to our presence or to the visits of land people to the Kwoks' boat. This difference in be- havior was an obvious function of other behavioral differences. The Kwoks were widely known as honest, respectable people, while the other family was disliked by most boat people of the area, toward whom they often showed a good deal of hostility similar to that shown to outsiders. The family was, in short, generally unpleasant. They were apparently neurotic by Western standards and were regarded as deviants by the boat people. The range between these two families was quite pronounced, but far from atypical. These differences tend to indicate that social control of in- tolerant behavior is chiefly operative at the family level. Striking differences within a family are unknown to us. Observing the means by which control is maintained is not difficult. Thus, Kwok frequently punished his four-year-old daughter verbally, and his wife sometimes did so physically, when she copied the hostile behavior of the neighboring boat. On the other hand, on that boat the younger children were frequent targets of apparently reasonless aggression themselves. They were also encouraged to express hostility (told what insults to call at whom, for example). Similar child-rearing practices were quite evident at the hos- pital quarters, along the waterfront, and elsewhere. Thus at the hospital quarters, when adults were present and children were teasing an outsider, the adults would watch with laughter and approval, and sometimes add their own remarks. At Tai 0, on the other hand, children who teased others were often ordered to scatter by any adult present. 104 FINAL NOTE The above material cries for analysis by someone more familiar with the literature of pluralism, ethnocentrism and social control than is the present author. Since interest is growing so rapidly in such fields, I have elected to present the information with a minimum of discussion, so that it may be used by persons with more direct interest therein. Two last reservations should be noted. Both are amply evident from the foregoing information, but should be clearly stated here as well. First, ethnic hostility and intolerance is only a pronounced form of a generalized hostility against outsiders, even those from another village. As such, much of the material here presented shows striking similarities with the data discussed by Banfield (1958), Foster (1965) and others, and is no doubt amenable to consideration fn the same theoretical terms. Second, tolerant and friendly behavior is as marked as its opposite. The rather negativistic theme of this paper must not be taken as a negativist evaluation of some nebulous unit called "Castle Peak Bay society." I wish to stress again the fact that variation is extreme, and runs from hostility to the most tolerant and open and spontaneous humanity. SUMMARY This paper constitutes a brief description of ethnocentrism in the area of Castle Peak Bay, rural New Territories, Hong Kong Colony. The 20,000 residents of the area are divided into approximately eight ethnic groups, and other groups are well known in the area. Each group has its stereotype, held by members of all the other groups, and inter- group hostility is common. Overt expressions thereof included violence at one time, but are now chiefly confined to verbal aggression; the most ela- borate form is known as "making a stupid prick," and is characterized by an attempt to make the victim appear both insulted and ridiculous. There is great variation in the expression of hostility. Both individual families and social groups differ widely in the degree to which they aggress against outsiders. 105 NOTES 'Research was supported by a fellowship and grant from the National Institute of Mental Health, to which I am greatly indebted. I am also indebted to the many people of Castle Peak Bay who served as in- formants, to Joan Moore for suggestions and comments on this paper, and especially to my wife Marja, who shared the research for this paper, and to Choi Kwok-tai and Chow Hung-fai, field assistants and chief informants. 2The transcription system used here for Cantonese has been re- cently developed by Dr. John McCoy, Cornell University,' and is not yet published. I am most grateful to Dr. McCoy for his help with Cantonese linguistics and for providing me with a description of this new, scientific transcription. I have given common names familiar in the literature when these exist, following them with accurate transcriptions. For words and phrases not well known in English, I have used Dr. McCoy's system exclusively. 3It should be noted that this is also true of behavior. That is, prejudice behavior is not varied according to which group is being attacked. The behavior is the same no matter what the "in-group" or the "out-group" is. In fact, behavior toward non-ethnic segregates of humanity, the "poor," the "city stranger," etc., is again the same. Thus the Boat People use the same techniques of hazing on Whites such as ourselves and on local Chinese non- boat people. No doubt a certain amount of local variation in stereotypes and in prejudice behavior does exist, but there seems to be little or none between ethnic groups in the Colony. REFERENCES Banfield, Edward C. 1958 The Moral Basis of a Backward Society. Glencoe: The Free Press. Carstairs, Morris 1958 The Twice-Born. London: Hogarth. Chen Su-ching 1935 The origin of the Tanka. Nankai Social and Economic Quarterly 8:250-273. Chen Ta 1939 Emigrant Communities in South China. New York: Institute of Pacific Relations. Coughlin, Richard J. 1960 Double Identity: The Chinese in Modern Thailand. Hong Kong: Hong Kong University Press. Foster, George 1965 Peasant society and the image of limited good. American Anthropologist 67:293-315. Frake, Charles 1962 The ethnographic study of cognitive systems. Anthropology 106 and Human Behavior 72-85. Washington: Society of Washington. Anthropological Freedman, 1958 Maurice Lineage Organization in Southeastern China. Monographs in Social Anthropology No. 18. London: University of London. 1966 Chinese Lineage and Society: Fukien and Kwangtung. London: University of London. Grant, Charles J. 1960 Soils and Agriculture of Hong Kong. Hong Kong: Hong Kong Goverment. Ho Ke-en 1965 A study of the Boat People. Journal of Oriental Studies V:1-40. (In Chinese, English summary.) King, Frank H. H. 1955 Pricing policy in a Chinese fishing village. Journal of Oriental Studies I:215-227. Kulp, Daniel Harrison, II 1925 Country Life in South China: The Sociology of Familism. New York: Columbia University. Lin YZueh-hwa 1948 The Golden Wing. London: Kegan Paul Lo Hsiang-lin 1955 Pai Yiueh Ylan-liu yiu Wen-hua. Hong Kong: China Printing Co. (In Chinese.) McCoy, John 1965 The dialects of Hong Kong Boat People: Kau Sai. Journal of the Hong Kong Branch of the Royal Asiatic Society V:l-19. Metzger, Duane G. and Gerald E. Williams 1966 Some procedures and results in the study of native categories: Tzeltal "Firewood." American Anthropologist 68:389-407. Skinner, G. William 1957 Chinese Society in Thailand. Ithaca: Cornell University Press. 1958 Leadership and Power in the Chinese Community in Thailand. Ithaca: Cornell University Press. Ward, Barbara E. 1955 A Hong Kong fishing village. Journal of Oriental Studies I:195-214. 1965 Varieties of the conscious model: The fishermen of South China. Association of Social Anthropologists Monograph 1, The Relevance of Models for Social Anthropology: 113-138. Willmott, Donald Earl 1960 The Chinese of Sarawak. Ithaca: Cornell University Press. 107