ArticlesV ConclusionIn the past, Jiro Kawakita argued that there were only four truly original civilizations in Asia: East Asia (China), South Asia (India), West Asia (Islam), and Tibet [Kawakita 1987], but Tibet is now a stateless civilization. This political situation is itself the causal factor of poor dialogue in Tibetan studies. It is directly linked to the whole history of how researchers turned to substitutive fields in the first place and then branched out into several different substitutive fields. The details of the history have already been made clear in this paper. In order to avoid misunderstandings, I shall add that there are aspects worth assessing positively in such an indirect method. For example, the fact that anthropologists hold a major share in modern historical research is indicative how such substitutive anthropology has contributed to academia in a way that transcends the framework of anthropology. On reflection, this contribution is a powerful testimony to those outside anthropology of the usefulness of anthropology in the humanities. The central argument of this paper is that this diversification must be understood through the framework of the attempt of substitutive ethnography to respond to the problematic situation in ethnography whereby field surveys were forbidden.I will conclude my analysis here, adding that the situation I have described was the case up until the 1990s. This situation is undergoing major changes thanks to the opening of China and the advance of Chinese political studies. For example, at the present time, it is possible to tentatively label the new trends in modern Tibetan studies as “Beijing-centrism” and “Amdo-centrism.” Based on the framework of political studies on modern China, Beijing-centrism uses Chinese documents to analyze the situation in modern Tibet from an ethnic-issue perspective, and it is for the most part positioned as a sub-field of Chinese political studies. The flourishing of Beijing-centrism indicates that the era of anthropology’s monopoly over modern Tibetan studies is drawing to an end. By contrast, Amdo-centrism, which seems to have become the mainstream of anthropological Tibetan studies, is ethnographic research conducted in the Amdo region (northeast Tibet). Amdo is not as strictly controlled as the Tibetan Autonomous Region, and so many anthropologists have entered the region in recent years.24 While space constraints do not allow me to discuss it in this paper, the trend of analyzing the refugee community within the framework of refugee studies continues to be ever present. Regarding the Tibetan Autonomous Region, while it is still difficult for foreign researchers to conduct surveys there, in recent years, a number of collaborative research projects have been underway, and channels to studying in the Autonomous Region are steadily being opened.25 Thus, research on modern Tibet is currently undergoing a major boom, and research results based on field surveys are being released. Aside from Beijing-centrism, most of these research trends are led by anthropologists, but I am concerned with the lack of interactive dialogue even within anthropology.26 That being said, this dialogue must not simply be a process of emphasizing the expansion and commonality of Tibetan society. Regarding this point, we can learn lessons from the experience of Chinese substitutive ethnography. Harrell produced a disquisition that may be viewed as self-criticism at how he in the past headed to Taiwan to conduct a substitutive field study and wrote it up as “Chinese ethnography” [HARRELL 1999]. However, he avoided the patchy “Taiwan is not China” style discourse. The issue is not the search for commonality between China and Taiwan itself, but is rather the way such a search is premised on the existence of a holistic Chinese society (anthropological holism). If such holism becomes the premise, then no matter to what extent particular characteristics are discovered in Taiwan and no matter to what extent Taiwan was affected by over half a century of Japanese colonial rule, such phenomena end up being viewed as a mere variation in the results of such a holistic social structure or cultural characteristics adapting themselves to individual environments. There is a danger here that the research findings in individual 023