<p>The emergence of native anthropology has resulted in a fundamental reconsideration of the past anthropological practice in which stronger people study and describe weaker people under colonial conditions. No longer are "natives" in the formerly "primitive" world objects of research who silently acquiesce to their representations by the former colonizers - "indigenes" who never talk back. They are now active agents of research - "literati," if you will - who do talk about their own culture in their own words from their own viewpoints. In the "world system of anthropology," however, in which the United States, Great Britain, and France occupy the center, the natives' discourse tends to be devalued due to their peripheral status in the system. In this article, I dare define the Japanese as "natives," despite their own colonial past, in order to examine the problems involved in the writing of their own culture in English. The first section introduces a new concept, the "ethnographic triad," which consists of the writer, the described, and the reader. The status of natives in anthropology may best be understood in terms of the shift from the described to the reader and then to the writer. Focusing on the reader, I distinguish four major categories within the ethnographic readership: (1) people who belong to the same linguistic and cultural community as the writer; (2) natives who have been studied and described; (3) native anthropologists; and (4) people who are neither describers nor the described. The first category ordinarily constitutes the assumed readership of ethnography, but when the Japanese write in English, that does not apply. Instead, their writings are intended for the non-Japanese whose cultural backgrounds are different from that of the Japanese. This fact explains why the same ethnography (e.g., a series of works by Emiko Ohnuki-Tierney) is appreciated differently in Japan and elsewhere. More generally, it shows the critical importance of the writer's ability to grasp the reader's mind in ethnographic writing. That argument is illustrated in the second section by my example of Ruth Benedict's classical work on Japan, The Chrysanthemum and the Sword. In writing about the Japanese, Benedict was clearly aware that her target readership, or more broadly, audience, was American - her fellow citizens. Her task was, therefore, to make the Japanese intelligible to the Americans, even though her depictions of Japan might not make much sense to the Japanese. In this regard, Benedict chose to be an "Orientalist" in Edward Said's terminology. A most notable example of Benedict's "Orientalism" is her likening of the Japanese emperor to the Stars and Stripes: both are sacred and inviolable. Such explanations are very useful in helping Americans understand Japan, but whether or not they make sense to the people described is a completely different issue. In the third section, I further discuss the importance of the assumed readership in ethnographic writing by citing three examples from my own experiences as a professor at an American university. The first example concerns the question of how to explain miso soup for Americans unfamiliar with Japanese food. After trials and errors, I found that we could best convey the cultural meanings of miso soup by likening it to apple pie, for they both symbolize mother's cooking and, as the expression "as American as apple pie" shows, the entire nation. We must remember, though, that this comparison is of little use to the Japanese, as well as people in other countries, who know very little about American food. The second example concerns the question of how to explain Japan's place in the global community. In the United States, there is a generalized Asian stereotype, which makes Japan</p><p>(View PDF for the rest of the abstract.)</p>