Taiwan as an island country is an immigrant society where interethnic marriages have been common....This interesting 2010 paper Geneticizing Ethnicity: A study on the “Taiwan Bio-Bank" (Tsai Yu-yueh) has some useful history of how Taiwan's colonial governments classified the people who live here (which is not necessarily the way they themselves thought about it).
I put that quote from the abstract there because it is striking how the author is struggling for inclusive language, which sadly negates the historical experience of Taiwan's indigenes. Another way to look at is to say that Taiwan is a settler and colonial society... but to use that language is to recognize that the Han are latecomers to whom Taiwan does not belong.
The chart above is taken from the paper, and shows how in the Qing Era, the Manchu state organized Taiwan into 3 groups of settlers and 2 groups of indigenous people. The Japanese inherited these categories but consolidated them into the Fujian and Guangdong Han -- though each province sent both Hakkas and Hoklos to Taiwan -- and, until 1935, the Raw and Cooked aborigines. In 1935 the nomenclature changed and the Cooked Aborigines became the Plains Aborigines, while the Raw Aborigines became the Mountain Aborigines.
The Japanese retained these categories because the presence of aboriginal peoples in Taiwan highlighted the need for a "civilizing mission" (colonization) and helped define the difference between "modern" Japan (no savages here!) and "primitive" Taiwan (savages abound!)
The KMT inherited that division, but then repositioned its census categories. The Plains Aborigine category was abolished "because it was believed that they had been assimilated into Han Chinese society", say the authors. Because so many Hakkas had lost the ability to speak Hakka by 1945, they were characterized in that first census as Hoklos.
I suspect the Plains Aborigines were deleted because by abolishing that category, on the practical level, the Pingpu people could then make no claims on land, especially KMT land, which after all was seized from the Japanese, who had seized it from the Plains Aborigines in many cases. Further, once the Pingpu were gone as a recognized people, there is no prior claimant to the plains, meaning that there is no visible reminder that Taiwan does not belong to the Han. The mountain peoples can then be dismissed as "ethnic minorities", exactly the strategy followed by the Han chauvinists running China. Recognizing this, Pingpu activists have fought a decades-long struggle for recognition and land rights.
In 1954 the KMT switched to a nine-group system. After the ending of the security state in the 1990s, the oft-quoted four group system of Hoklo, Hakka, Mainlander, and Aborigine grew. The paper attributes it to a 1993 proposal by DPP politician Yeh Chu-lan. The categories had been in common use for quite some time before that, though. Note that the KMT kept track of "home province" of mainlanders since that was used in certain government applications where jobs or other opportunities were distributed by province of origin, a way to screen out Taiwanese.
The reality is that these groups are fluid and flexible, and people often change their identities over time. As the paper notes:
In fact, during the past few centuries, many people in Taiwan have changed their ethnic identities for one reason or another. Take a recent case for example. A survey conducted by the Council of Hakka Affairs in 2004 showed that when questionnaires about one’s ethnic identity were provided with multiple-choice answers, subjects tended to disclose their Hakka identity more easily, increasing the number of those who identify themselves as Hakka (Xingzhengyuan kejia weiyuanhui 2008).Another issue is blending:
Xu (2002) investigated interethnic marriages for three generations of the “four great ethnic groups.” The rate of interethnic marriages among those married before 1961 was 12.8%; for those who married between 1961 and 1981, the figure was 21.5%; it grew to 28.2% among those married after 1981 (see Table 1). As the following tables show, the rate of interethnic marriages in the third generation of Hoklo was 15%, 63.4% in Hakka, 82% in Mainlanders, and 38.2% in aborigines (Tables 2, 3, 4, and 5). Although the rate of interethnic marriages among Hoklo remained low, those of the other three groups were very high.And of course, the group "Aborigines" lumps together disparate peoples, and makes the differences and conflicts between them disappear, repositioning them as an identifiable Other. It is interesting to imagine a modern anthropologist landing in Taiwan in 1400. How many different groups would she find it useful to define?
Looking at Lumley's old piece in The Anthropology of Chinese Society entitled "Subethnic Rivalry in the Ch'ing Period", it is easy to see why the settler population was classified by place of origin rather than what we would call ethnicity. During the 19th century Chinese settlers in Taiwan not only saw themselves through the Hoklo-Hakka lens, but also through a Changchou-Chuanchou lens (and others) based on their place of origin. These groups venerated different deities and had other differences, and extensive political and commercial rivalries.
The Hakkas spoke four different but mutually intelligible versions of Hakka, and did not usually have ethnic/place of origin conflicts amongst themselves, since they were generally scraping out a living in harder areas and were less wealthy, and were surrounded by indigenous and Hoklo communities that feuded with them.
The Hoklo, by contrast, did confront rivalries amongst themselves. For example, the famous Lungshan Temple in Taipei served as a military command center in the 19th century for Hoklo settler groups feuding with incoming settlers from Tungan, and then later, for feuds between Chuanchou and Changchou settlers. Throughout Taiwan there were temples that served as such military command centers for both Hakkas and Hoklos in these feuds.
Needless to say, these place-of-origin distinctions have disappeared. Yet, had they been maintained by the census, by temple practice, and by marriage and feuding practices, we might be discussing ethnicity and origins in very different ways today. Instead, the KMT switched to a "provincial" level of definition for Taiwan's myriad peoples because that definition was more useful in generating a distinct Mainlander identity as the basis for KMT colonial power and in populating the bureaucracy and military with its people.
The groups we used today hardly begin to reflect the ethnic diversity of modern Taiwan, where numerous children are born to foreign mothers. That is why so many people are turning to the useful rubric of "Taiwanese" to swallow up all this immigrant and settler diversity.
Discussing the difference between the four great ethnic groups, some sociologists state that none of the groups is a concrete reality. Systems of ethnic categorization amount to ideologies. What we should be asking is when, why, and how this classification became so important.Lurking behind these categories is the hazy idea of "blood" and genetic origins, which the author argues are dangerous misconceptions: these groups are not identifiable by genes though many believe they are. That is why the current government needs to move forward on the idea of a national citizenship by birth and immigration, and reject the "blood"-based citizenship idea of the KMT.
[Taiwan] Don't miss the comments below! And check out my blog and its sidebars for events, links to previous posts and picture posts, and scores of links to other Taiwan blogs and forums!