Feeling, Speech, and Nonrepresentational Ethics 105
6   Feeling, Speech, and Nonrepresentational Ethics
Rawski's statement that imperial funerary ritual "constituted a structure in which individual feelings could be expressed in purposeful deviations from the norm" (1988:248) must be interpreted in the context of a ganqing ethics of emotion. In many Western contexts the statement would be understood as a matter of making ritual space for the expression of the spontaneous, authentic, innermost feelings of a grieving individual. However, in Rawski's analysis the emperor's "expression of feeling" is more about guanxi propriety than about authentic individual expression. By displaying greater grief for his parents than his concubines, the emperor publicly affirmed a hierarchy of guanxi in which parents were more important than wives. Since Confucian ideology held that the emperor is to a minister as a father is to a son, this hierarchy also implied the emperor's own preeminence. The emperor's embodiment of filiality both affirmed his right to the throne (as the chief mourner of his imperial parents) and reinforced the hegemony of a statist Confucian ideology. In my invocation of Rawski's analysis, I mean to suggest that Fengjia residents also manipulated ganqing within the ritual sequences organized by specialists. In Fengjia too, the ethical dynamics of this manipulation were more about relationships than about emotional sincerity.
In this chapter I wish to extend the contrast between a representational and a guanxi ethics of expression as far as possible. Such a contrast enables a more accurate translation because it unmasks hidden assumptions that individuals of different backgrounds bring to a communicative exchange. In the West, an ethic of accurate representation entails both emotional "sincerity" (accurately representing inner feelings in outward expression) and "honest" speech (accurate verbal representation). This chapter examines various Chinese contexts (in Fengjia and elsewhere) in which a Western ethic of accurate representation is not taken for granted—hence the title "Nonrepresentational Ethics." In
this last chapter of part I, I push my interpretive framework as broadly as possible, including examples from past and present, rural and urban places, and popular and official life. However, I have chosen the phrase "Nonrepresentational Ethics" to prevent this broad approach from suggesting that I hold an essentialist view of Chinese ethics. By emphasizing the lack of a set of Western cultural assumptions, I bring these examples together in terms of the similar problems they pose to Western translators, not because of an essential similarity among the Chinese contexts themselves. As Gadamer (1975, 1976) argues, translation requires unpacking one's own assumptions as much as describing foreign ones.
In their examination titled "The Cultural Construction of Emotion in Rural Chinese Social Life" (1990:180), Sulamith Heins Potter and Jack M. Potter also argue for significant divergence between American and rural Chinese assumptions about the ethics of emotional expression. However, they suggest that emotion is simply irrelevant to relationships in rural China, a view quite the opposite of my own, where ganqing is a central component of guanxi. Thus, I begin my examination of ethical problematics with a critique of the Potters' analysis.
Potter and Potter never say what Chinese terms they are translating as "emotion" or "to feel." As Russell and Yik (1996) point out, many Chinese words might be translated as emotion or feeling, including ganqing, qinggan, and qingxu. If one further considers that the Potters were translating a rural Cantonese dialect and I a rural Mandarin one, one might assume that part of our divergence lies in the difficulties of translation. However, though linguistic problems may contribute to our differences, the range of examples they give to support their argument clearly overlaps with the materializations of ganqing I have described in previous chapters. Moreover, the examples they describe are explicable within my own framework. Thus, I proceed with the caveat that the Potters may not have been referring specifically to ganqing.
Potter and Potter emphasize that in the United States personal emotions "legitimate" social relationships. A marriage, for example, would be illegitimate if not based on feelings of love. These emotions are supposed to be spontaneous and sincere; to feign love for the purpose of gaining a spouse would both be immoral and leave a flimsy basis for one's marriage. In contrast to writers who emphasize the importance of spontaneous emotion in the West,1 Potter and Potter claim that the Western need for "sincerity" eliminates true spontaneity. Instead, they argue, there exists a complex system of psychological repression and denial that allows one to maintain an "emotionally appropriate self"
106    Practices of Guanxi Production
(1990:181). In contrast, they contend, for rural Chinese emotion is irrelevant "to the creation or to the perpetuation of social institutions of any kind . . . [E] motions are not thought of as significant in social relationships . . . (and) are thought of as lacking the power to create, maintain, injure, or destroy social relationships" (1990:183). Since rural Chinese people believe "how I feel doesn't matter" (1990:183), they can be truly spontaneous, and give vent to anger and other unsocial emotions in almost any context. The Potters back their theory with observations of child rearing, in which crying children are neither scolded nor comforted. Supposedly Chinese children thereby learn that emotional outbursts are inconsequential. In addition to not controlling unsociable emotions, rural Chinese people are further said to have no special use for positive ones. When discussing family relationships, Potter and Potter claim that rural Chinese never speak of love. Rather, they emphasize work. Daughters, for example, work incredibly hard before they marry in order to affirm their relationships with their parents. They don't talk about how much they love their parents. In marriages, rural Chinese at best would speak of "good feelings." They quote one villager as saying "We Chinese show our good feelings for one another in our work, not with words" (1990:194), and conclude:
From an outsider's point of view, the social world of the Chinese villager is characterized by an insistent emphasis on work, drudgery, and production. But work is the symbolic medium for the expression of social connection, and affirms relationship in the most fundamental terms the villagers know. (1990:195)
To back up their contention that emotions are irrelevant, Potter and Potter present several examples of rural people displaying negative emotions. They include public grieving over family tragedies and the past and present misfortunes of friends and relatives. They also include displays of anger, such as women screaming at the brigade officials responsible for implementing birth control policy. From these examples they conclude:
The behavior of villagers who show these strong feelings is consistent with the idea that the expression of most feeling is not a significant act, in and of itself, rather than with the idea that affect is in some way intrinsically dangerous and to be concealed. The villagers would be perfectly capable of self-restraint, if they believed that open expressiveness were damaging to their social position. (1990:185)
Feeling, Speech, and Nonrepresentational Ethics 107
The Potters further conclude that brigade officials tolerate outbursts of anger from villagers because they too see it as irrelevant.
I believe that Potter and Potter have missed entirely the significance of these displays of negative emotions. A public embodiment of grief and sorrow creates a collective ganqing with all of those who feel the same grief, including the victim and his or her close friends and relatives. In so doing, a collective subject, and a series of guanxi among the members of that subject, are formed. Likewise, women who yell at birth control officials embody sympathy for all of the members of those families who are hurt by a specific instance of implementation of the birth control policy. The significance of such yelling is not that these women believe emotions don't matter and therefore that yelling will not hurt their guanxi with the official; rather, it is the ganqing and guanxi that are created between the yeller and the people on whose behalf the yelling is done. That officials tolerate such abuse perhaps reflects that they too realize the yelling is more a matter of creating solidarity with a third party than personal abuse directed at them. Potter and Potter find rural people "vividly expressive" and consider that evidence of their seeing emotion as irrelevant. In opposition, I see the vividness of ganqing as evidence of its centrality to social relationships. The statement "How I feel doesn't matter" is better interpreted as a complaint about individual powerless-ness rather than a general statement about the irrelevance of emotions.
Secondly, that rural Cantonese affirmed relationships through work rather than words merely shows the link of material obligation (embodied by working for others) and ganqing in guanxi. It has nothing to do with the "irrelevance of emotion." In one of the dowry parties that I attended in Fengjia, the parents of the bride sat together in one corner of the room with tears in their eyes. They were not interested in talking to me or anybody else. The daughter was also not interested in talking. She kept doing chores—dusting and fetching water. Both parents and daughter were creating and participating in the mutual ganqing of a guanxi they both wanted to continue. That ganqing was embodied in facial expressions and work rather than words made it neither less moving nor irrelevant to social relationships.
Finally, I should point out that in 1988-90 Fengjia children were not, as a rule, left to cry out their temper tantrums. If anything, I felt that Fengjia children received more attention than their American counterparts, though I did not carry out systematic research on this point.
In discussing the importance of sincere, personal emotions in America, Potter and Potter have at least pointed at the type of assump-
io8    Practices of Guanxi Production
Feeling, Speech, and Nonrepresentational Ethics 109
tions one must abandon to interpret the ethics of ganqing in Fengjia. \ Ganqing is not primarily an individual matter. Rather it is a type of feeling that must be conceived of more socially than psychologically (i.e., that is held to exist between and among people as much as within individuals' heads).2 Furthermore, sincerity—at least a notion of sincerity that requires one's words and facial expressions to accurately represent the "inner" feelings of one's heart—is usually absent from ganqing. To be "sincere" with one's ganqing is to be serious about and to live up to the obligations incurred in the guanxi that ganqing involved. Both in and out of Fengjia, I heard Chinese people of various backgrounds criticize crass manipulations of ganqing in pursuit of material gain. However, even these criticisms had little to do with the "insincere" representation of inner, individualized feelings. What was wrong about that sort of ganqing was the inauthenticity of the guanxi created. In the venal sort of gift giving that these manipulations involved, the value of the gift given was negotiated to be exactly equal to the value of the favor extracted from the person to whom one gives the gift. As a result, the material obligation involved in the guanxi was erased soon after it was established. The guanxi and ganqing involved were intentionally shortlived.
In their critique of the Potters, Arthur and Joan Kleinman (1991) problematize the orientalism inherent in the Potters' presentation. By discounting exceptions, asserting homogeneity, and denying shared human qualities, the Potters risk "narrowing the humanity of the other and thereby of ourselves" (Kleinman and Kleinman 1991:286). Certainly Americans can emphasize the social import of emotions while Chinese at times discuss authenticity; yet the timing, methods, and contexts usually differ. Holland and Kipnis (1994), for example, discuss the "sociocentrism" inherent in American cultural models of embarrassment, while Leo Ou-Fan Lee (1973) provides a detailed discussion of the history of emotional authenticity in Chinese literature and philosophy. Lee describes how the problem of meshing sentiment (qing, as in ganqing) and propriety has been debated by Confucians, Buddhists, and Taoists since the Han dynasty. During the May Fourth era, a generation of Chinese writers influenced both by these earlier debates and by Western Romantics like Byron and Goethe created an iconoclastic literature that rejected the propriety of li entirely in favor of impulse, sincerity, ■ sentimentality, and individual, subjective experience. However, as Lee points out, neither the early sentimentalists nor the May Fourth romantics held a dominant position. The former were contained by the hege-
mony of Confucian ideology while the latter, though quite influential for a brief period, were abruptly stifled or transformed by the communists' demands for a revolutionary literature.
In my own consumption of popular culture in Dengist China—including television, movies, short stories, and novels—I was often struck by an intense sentimentality, which I found first maudlin and later moving. Though this sentimentality perhaps reflected the influence of May Fourth romantics, I usually felt that it too had more to do with guanxi propriety than with emotional authenticity. By embodying powerful emotions, fictional characters demonstrated the strength of their commitment to socially legitimate guanxi. Popular stories of romance, like Qiong Yao novels and the television series Blushing Grass (Hanxiu Cao) (both imported from Taiwan), were full of passionate characters embroiled in complex situations that separated them from the objects of their attraction. What struck me was that these characters were never fickle. They endured decades or even lifetimes of separation without ever falling in love a second time or even reducing the strength of their initial passion. Whatever the complications of their situations, the monogamous commitment implied by their unwavering passion made their ganqing legitimate. Fickle characters were most often evil ones. In urban areas, where perhaps a majority of young people selected their own spouses, romance could be important (Jankowiak 1993). However, though such romance could be read as the epitome of emotional spontaneity, the tendency to construct romantic impulses monogamously indicated a practice of propriety as well. In Jankowiak's study of Dengist Huhhot, for example, romantic inclinations often involved lifelong attraction to a single love. Men and women who failed to marry their "first and true love" found their later marriages to be devoid of passion (1993:209).
While at Nanjing University, a friend of mine took a course designed for foreign students on women in modern Chinese short stories. On the first day of class the teacher (a Chinese man) gave the following introduction. He said that when teaching Chinese literature to Westerners, he has found it necessary to emphasize that characters, especially women, were morally evaluated primarily on their emotional reactions. He said that though Chinese readers take this for granted, Western readers often ignore this point and consequently find Chinese literature too sappy. I met many Chinese who felt that Americans were cold and unfeeling and who could not imagine living in American society. Though I don't necessarily agree with this assessment of Americans, I do think it reveals the
no    Practices of Guanxi Production
Feeling, Speech, and Nonrepresentational Ethics m
lack of understanding of how to embody ganqing on the part of many Americans in China.
My own shortcomings in this area can serve as an example. Once in Nanjing I attempted to arrange a Chinese language tutor for an American friend of mine. After several rebuffs, I explained my problem to a Chinese friend. He said my approach had been too direct. He said that I couldn't just explain the tutoring needs of my friend and suggest a price, especially to someone I didn't know well. If the potential tutor accepted the arrangement, she would appear as someone who was interested only in money, who had no ganqing, and thus acted outsidejhe boun^^^uaw^pw^ň^y. He said I should work through intermediaries so that everyone could embody ganqing by doing a favor for someone that they knew well. The price could be suggested by one of the intermediaries as an aside along the way. Unappreciative of a concern for ganqing that would lead me to approach the problem as a matter of favors between friends, I appeared cold and aloof.3 As with imperial funerals, the ethics of ganqing were caught up in guanxi propriety rather than problems of authenticity. Indeed, at the linguistic level, describing someone as having ganqing means that that person acts out of a concern for (specific) other people rather than the selfish yearnings of an individualized heart.
The importance and propriety of ganqing in many Chinese contexts are also gendered differently than in the West. Campbell (1987:225) discusses how Western patriarchal ideologies relegate human feeling and personal relationships to the realm of the feminine and politically unimportant. Most Chinese would see them as central, and Chinese men don't allow women to dominate these arenas. Though the professor of literature cited above suggested that women especially were judged on their ganqing, I was struck by the sentimentality of Chinese men. In both rural and urban settings I often hgardXhinese men speaking of their friendship and ganqing for other men, and saw them do favors for, exchange unexpected gifts with, and hold hands with each other. \ None of these actions implied homosexuality; indeed, most reflected the utmost of guanxi propriety Susan Brownell (1995:213-32) likewise discusses displays of affection between Chinese women without fear of lesbianism. If anything, especially in rural settings, it was exchanges of ganqing between men and women, even husbands and wives, that were beyond the boundaries of propriety and kept fully out of public view.
Nonrepresentational Speech Ethics
The above section described how a Chinese ethics of ganqing was more concerned with guanxi propriety than with accurate emotional representation; the ethical evaluation of emotional responses began from a consideration of the implications of those emotions for individuals' guanxi rather than from a consideration of their "sincerity" (accurate representation of their inner selves). This section suggests that a similar dynamic underlies the moral evaluations of speech acts in Fengjia. In situations in which middle-class Americans would emphasize "truthful" representation, Fengjia residents tended to base their moral evaluations on guanxi propriety or other forms of social pragmatics. The parallels between an ethics of speech and that of emoting reflects the fact that both language and feelings are means of communication that can be used to manipulate guanxi.
Consider the following common scene at the village's clinic. A four-year-old child was brought in with a fever. As soon as he entered the clinic door, he started crying. His mother began repeating "There won't be a shot, there won't be a shot [budazhen]" (children were almost always given a shot in this situation). After a brief examination, the village doctor carried the child into the back room for a shot. The child wailed the whole way, and the doctor and mother attempted to drown out his wails with a chorus of "There won't be a shot." This chorus continued until the shot was actually being administered at which point the doctor switched to his own monotone wail of "Aiiiii..." and then to "No more shots" (budazhenle) as he pulled out the needle.
I was struck by this scene because in my experience middle-class American parents would not handle the situation in the same way. They would consider this method of hushing the child to be "lying" and thus to be a "bad example" to set for the child. For example, Dr. Spock's manual, the classic how-to book of middle-class child-raising, suggests:
The best way to get your child ready for each immunization is to be as honest and simple in your explanations as possible. Tell him that the shot will hurt, but that it will protect him from a sickness that would hurt much more than the shot. (Spock and Rothenberg 1992:239)
In Fengjia when I asked several parents if the budazhen method of hushing the child could be considered "dishonest" (bulaoshi), they all said no. This led me to a consideration of the word laoshi (usually
112    Practices of Guanxi Production
Feeling, Speech, and Nonrepresentational Ethics 113
translated as "honest"). Fengjia residents often complimented people as being very laoshi. When I asked about desirable qualities for a spouse, laoshi also often came up. After hearing several stories about laoshi people, I realized that laoshi almost always had an element of self-sacrifice. For example, if there was a very distasteful job to do, and several people got out of it by giving one excuse or another, the person who finally did it was more laoshi than the others. Being "honest" or "dishonest" was more than a matter of representation; it involved the purpose for which one used language. False representations were only "dishonest" if they were done for selfish purposes. Since, at least from the point of view of adults, giving children shots and reducing their anxiety about these shots were for the children's own good, "lying" about imminent shots could not be called "dishonest."
The New Peasant Encyclopedia of Family Happiness (Xinnong Lejia Baike), an advice manual on handling family relations that went through three printings in the 1980s, gives a plethora of advice on what to say in what sort of situation to improve one's intrafamily guanxi. Of language in general, it says:
Speech is a tool for the exchange of ganqing. One sentence can make someone cry, another can make someone laugh. Let speech be the medium for transmitting ganqing to your interlocutor. [Yuyan shi jiaoliuganqingdegongju. Yijuhua neng baren shuoku, yijuhua neng baren shuoxiao. Rangyuyan zuomeijie, ba ganqing chuandaogei duifang.] (1985:100)
Here, too, a guanxi ethics for speech acts is directly invoked.
Nonrepresentational ethics have been prevalent in official speech acts of imperial and modern Chinese governments as well. Of state historiography in imperial China, Kenneth DeWoskin writes, "the historicity of events themselves is only incidental, a concern secondary to the moral purport and effecting power of the written words" (1991:46-47). During the Republican era, Lin Yutang wrote:
In order to understand Chinese politics, one should understand Chinese literature. Perhaps one should here avoid the word literature (wenhsueh) and speak of belles-lettres (wenchang). This worship of belles-lettres as such has become a veritable mania in the nation. This is clearest in modern public statements, whether of a student body, a commercial concern, or a political party. In issuing such public statements, the first thought is how to make them nice-sounding, how to word them beautifully. And the first thought of a newspaper reader is whether such statements read nicely or not. Such statements almost
always say nothing, but almost always say it beautifully. A palpable lie is praised if it is told in good form. (1977 [19361:223)
Whatever its roots in the rhetorics of earlier Chinese governments, the CCP's practice of propaganda is also typical of Leninist party-states. Katherine Verdery suggests that socialist regimes in Eastern Europe treated language and discourse as "the ultimate means of production. Through discourse even more than through practice, their rulers (hoped) to constitute consciousness, social objects, social life itself" (1991:430, emphasis in original). Paying little attention to the accuracy with which their rhetoric represented lived realities, socialist cadres' manipulation of language in both China and Eastern Europe at times approached the ludicrous. As Vaclav Havel sarcastically described it:
Life in the system is . . . thoroughly permeated with hypocrisy and lies: government by bureaucracy is called popular government; the working class is enslaved in the name of the working class; the complete degradation of the individual is presented as his or her ultimate liberation; depriving people of information is called making it available; the use of power to manipulate is called the public control of power, and the arbitrary abuse of power is called observing the legal code; the repression of culture is called its development... the lack of free expression becomes the highest form of freedom; farcical elections become the highest form of democracy; banning independent thought becomes the most scientific of world views. (1988:30-31, cited in McCormick 1993:1)
Noncadre publics have occasionally attacked official languages for their lack of representational accuracy. Such attacks have been an effective strategy in the undoing of socialisms in Eastern Europe. Vaclav Havel's naming of the lies of socialism was a conscious and important element in his revolutionary strategy (McCormick 1993). Solidarity's resistance to the socialist state in Poland, as Kristi Evans describes it, similarly involved espousing representational accuracy.
Solidarity's own ideology of language was based on the idea that language should directly and clearly reflect an externally constituted reality. This ideology, which is widespread in the Western world, is what Silverstein (1979) terms a reference-and-prediction ideology, in which the ideal operation of language is reference to objective reality. (1992:761)
At times, struggles within the CCP similarly hinged on philosophies of language. The Maoist aphorism "Seek truth from facts" {Shishiqiu-shi) became a common battle cry for heterodox campaigners within the
114    Practices of Guanxi Production
Feeling, Speech, and Nonrepresentational Ethics 115
party. In the late 1920s, Mao used it to oppose orthodox Marxist theories that denied the peasantry potential as a revolutionary class. During the Lu Shan Plenum of 1959, Peng Dehuai used it in opposition to politically correct but apparently distorted accounts of commune productivity in the Great Leap Forward. Deng Xiaoping used it to counter the "whatevers" orthodoxy of the Gang of Four.4 However, unless already encoded in a new party orthodoxy, these battles rarely reached the public at large. CCP propaganda continually presented itself to the public as language with beneficial social effects. Rather than confronting such language straight on with questions of accuracy, the Chinese public has for the most part tended to either subvert it with humor or manipulate it for personal gain.5
Certainly these events are rather distant from Fengjia. However, party debates over "truth" are far from irrelevant. The CCP has often concerned itself with the ethics of local speech acts. During the Cultural Revolution, the expression of "revolutionary spirit" was far more important than representational objectivity. Under Deng Xiaoping, "seeking truth from facts" took a back seat to promoting "stability and unity" (anding tuanjie) as an ethical priority in speech. The following "editorial" heard in Fengjia on a radio broadcast in April of 1989 was typical of Dengist exhortations.
Comrade Deng Xiaoping has recently emphasized the importance of stability and unity. In our daily life in villages and work units, some people seem to voice an opinion (yijian) on every subject no matter how disruptive it might
be----Don't be a person with too many "opinions." Don't do things that aren't
useful to unity; don't say things that aren't useful to unity (Buliyu tuanjiedeshi buzuo; buliyu tuanjiedehua bushuo).
This propaganda both explicitly urged people to concern themselves with the social effects of their speech acts and, like most Chinese propaganda, justified itself by presuming to be a speech act with "beneficial" social effects. The point here is not to excuse CCP propaganda as either socialist necessity or a manifestation of deep-rooted "Chinese culture." Indeed, the self-serving nature of CCP propaganda made it "dishonest" (bulaoshi) by Fengjia standards regardless of its representational accuracy. Rather, the point is that the CCP state, like its predecessors, did not promote representational truth as an ethical priority.
In contrast, as Evans's (1992) citation of Silverstein (1979) suggests, "reference-and-prediction ideologies" of language have been "widespread in the Western world." Consequently, Western sociolinguists
have long felt it necessary to emphasize that speech acts pragmatically in the social world and that representational models of language are hence limited (e.g., Gumperz 1982:9-29 and Levinson 1983). This fe-tishization of "truth" as representation with moral obligation, though beyond the scope of this project, is perhaps a more interesting historical problem than the sort of honesty lived in Fengjia. Certainly Foucault's (1973, 1978, 1980:194-229) explorations of the rise of confession as a technique of power/knowledge and the linkages of "biopower" with our "will to knowledge" are suggestive. As Foucault describes, the rise of confessional techniques in psychoanalytic practice (especially relevant to emotional representation) and the Cartesian separation of language and thought from the material world both have emphasized representational rather than pragmatic ethics and practices of speech acts. In China, on the other hand, ethics of ganqing and speech have been influenced by recent increases in the importance of practices of guanxi production as well as by statist prescriptions for socialist speech. In brief, the forces that have molded Western taken-for-granted notions of truth have not been as influential in China, and vice versa.
This chapter has emphasized the interpretive problematics of moving between representational and pragmatic ethics of speech and feeling. In conclusion, I would again caution against essentialization. Raised to the level of generalities, equations like the West is to representation as China is to social pragmatics become absurd. In the hands of orientalists they can become tools of imperialism. In 1894, for example, the missionary Arthur Smith portrayed "the absence of sincerity" as a Chinese racial characteristic curable only through the introduction of Christianity (Smith 1894). There have been many different versions of, and counterdiscourses to, representational and guanxi ethics of feeling and speech. Fang Lizhi's (1990) championing of the objective language of science in opposition to the hegemony of Marxism-Leninism-Mao Zedong thought is matched by American Marxist, feminist, and cultural studies scholars' insistence that no "scientific" investigation has ever been or could ever be apolitical. At a more mundane level Eve Sweetser (1987) describes the importance of "white lies" and "social lies" in American speech ethics. In both China and the United States, doctors avoid detailed discussions of cancer prognoses with their patients. In fact, it would be difficult even to imagine a place entirely devoid of either representational or pragmatic speech. What is significant is difference in the degree to which and the matter and contexts in which the two sorts of speech are moralized.