I Approach my topic—the development of the modern Japanese political cartoon—with some trepidation. Humor is a fragile product that can easily be damaged by academic scrutiny. As Evelyn Waugh once remarked, analyzing humor is like dissecting a frog—much is learned but in the end the frog is dead. Waugh was right. Most analyses of humor cannot be read for amusement. On the other hand, why should they be? If Shakespeare scholars are not expected to write in iambic pentameter, why should students of humor be expected to keep their readers in stitches? As the editor of the International Journal of Humor Studies recently told a reporter, “We are not in the business of being funny” (New York Times, 19 December 2000).

Since its appearance in 1983, Benedict Anderson's Imagined Communities: Reflections on the Origin and Spread of Nationalism has taken on almost talismanic status, ritually invoked for its quasi-magical efficacy in thousands of scholarly discussions of nationalism and ethnicity. Anderson's book has given focus to a line of argument in which not only nationalism but also ethnicity are seen as reflexes of the Euroamerican colonial enterprise. Anderson proposes what we might call an “appendency theory” of ethnicity, holding that ethnicity is a phenomenon that is secondary to, contingent upon, and necessarily later than the modern nation, and that therefore it could not possibly have existed prior to the colonial conquest and the creation of nation-states. In this view, ethnicity itself is a recent product of modern history—of the colonial or postcolonial era, the age of the modern nation state and the capitalist world system.

Located on the high altitude slopes in China's mountainous southwestern hinterlands, the Nuosu (Yi) have long been characterized by ethnologists in China as a slave society. In this essay I explore the appropriateness of the label with reference to the distinction, associated with the classicist Moses Finley in his studies of Greco-Roman society, between slave societies and societies with slaves (Finley 1983, 79–83). But my main purpose in focusing on Nuosu slavery, as we know it from the first half of the twentieth century, is to gain a better understanding of Nuosu identity. It makes sense to me that the process by which outsiders, that is, slaves brought in as captives, became insiders reveals a great deal about the constructs of the Nuosu ideology of common origins and identity and about how this transformation from alien to native worked in practice among people historically “remarkably resistant to acculturation,”with boundaries that remain to this day sharp, clear, and relatively impenetrable (Harrell 1995, 101, 104).

These words were expressed by Gangzhen and Cizhu, two peasant women born in the early twentieth century who lived in two villages in Jiangyong County of Hunan Province in southern China. When confronted with the loss of their husbands, these women relied on local tools familiar to women in this area—nüge and nüshu—to articulate their thoughts and emotions. Nüge (literally, “female song”), which includes bridal laments and folk narratives, was widely practiced among peasant women in southern China, especially prior to the Liberation of 1949. Nüshu (“female writing”), in contrast, was used extensively by peasant women in Jiangyong. Moreover, it is one of the very few writing systems in the world that is mostly illegible to men. This male-illegible script,3 which is semiphonetic compared to the ideographs of Chinese hanzi,4 was used for centuries among hanzi-illitetate women, but it remained basically undocumented and unknown to the outside world until the 1980s, just as it was becoming extinct. Prior to Liberation, Jiangyong women had used nüshu to write sisterhood letters, biographic laments, wedding literature, folk stories, and other narratives in verse form. Combined with nüge, they documented peasant women's experiences and gave voices to their existence.

The marketplace and the temple: traditionally, Western scholars have been nervous and ambivalent about bringing these two spheres into too close a proximity. On one side, historians of religions trained in the lineage of Joachim Wach or Mircea Eliade have long warned that “reductionism” is the cardinal sin in the study of religion, which is claimed to be a sui generis, or irreducible phenomenon. Hence, “economic reductionism” or “vulgar economism” would be the most heinous crime of all (Wach 1958; Eliade 1958, xi). On the other side, those trained in Marxist and neo-Marxist traditions have typically wanted to explain religious myths and rituals either as ideological screens masking deeper material forces or as symbolic expressions of misrecognized social interests. As Pierre Bourdieu—one of the most sophisticated recent representatives of this tradition—explains his own method, he wishes to “utilize the economic model to extend materialist critique into the realm of religion and to uncover the specific interests of the protagonists of the religious game, priests, prophets and sorcerers” (1990, 107).